Aquaculture Sep2012

This article was downloaded by: [Department Of Fisheries] On: 25 September 2012, At: 23:59 Publisher: Taylor &
Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
North American Journal of Aquaculture

Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/unaj20
Parental Male Effects on Landlocked Fall Chinook Salmon Progeny Survival

Matthew M. Wipf
a a b
& Michael E. Barnes
South Dakota Game, Fish, and Parks, McNenny State Fish Hatchery, 19619 Trout Loop, Spearfish, South Dakota, 57783, USA
b
South Dakota Game, Fish, and Parks, McNenny State Fish Hatchery, 19619 Trout Loop, Spearfish, South Dakota, 57783, USA Version of record first published: 22 Aug 2012.
To cite this article: Matthew M. Wipf & Michael E. Barnes (2012): Parental Male Effects on Landlocked Fall Chinook Salmon Progeny Survival, North American Journal of Aquaculture, 74:4, 443-448 To link to this article: http://dx.doi.org/10.1080/15222055.2012.681105
PLEASE SCROLL DOWN FOR ARTICLE Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.
North American Journal of Aquaculture 74:443448, 2012 C American Fisheries Society 2012 ISSN: 1522-2055 print / 1548-8454 online DOI: 10.1080/15222055.2012.681105
ARTICLE
Parental Male Effects on Landlocked Fall Chinook Salmon Progeny Survival

Matthew M. Wipf*1 and Michael E. Barnes
South Dakota Game, Fish, and Parks, McNenny State Fish Hatchery, 19619 Trout Loop, Spearsh, South Dakota 57783, USA
Downloaded by [Department Of Fisheries] at 23:59 25 September 2012
Abstract
The Lake Oahe, South Dakota, population of landlocked fall-run Chinook salmon Oncorhynchus tshawytscha is maintained entirely by hatchery propagation and exhibits relatively poor egg survival during hatchery incubation. This study was undertaken to determine the inuence of male gametes on embryo survival. Eggs from an individual female were subdivided and subsequently fertilized with milt from four discrete males. This was repeated with three additional females using the milt from the same four males. This entire procedure was then replicated three times, using four new females and four new males each time, for a total of 16 males and 16 females. The eggs from each unique cross were then incubated discretely. There was no signicant effect of spawning males on subsequent embryo survival to the eyed stage of egg development. Swim-up fry length and weight were also not signicantly affected by male parentage. In contrast, there was a signicant maternal effect on eyed egg survival, and swim-up fry length and weight, which varied signicantly among progeny from individual females. These results suggest that the relatively poor survival exhibited by Lake Oahe landlocked fall Chinook salmon eggs during hatchery incubation is largely a function of initial egg quality from spawning females.
Eggs from landlocked fall-run Chinook salmon Oncorhynchus tshawytscha from Lake Oahe, South Dakota exhibit poor, and extremely variable, survival during hatchery incubation (Barnes et al. 1999b, 2000, 2001). This high egg mortality has been attributed to poor egg quality (Barnes et al. 2001), and research has primarily focused on possible maternal effects (Barnes et al. 2003b). The possible effects of spawning male contributions on egg survival have historically been controlled by the use of pooled milt (a spawning ratio of multiple males to one female; Barnes et al. 2003a). In salmonids the reported effects of male contributions on subsequent embryo survival have generally been variable (Smoker 2000, 2004). Spawning male age and size can inuence sperm potency (Wedekind et al. 2007), suggesting that the use of pooled milt may be benecial (Billard et al. 1996). In addition, the use of pooled milt can increase the speed and
efciency of spawning procedures (Billard et al. 1996; Withler and Beacham 1994). Anderson (1994) and Alcock (2005) both suggested that offspring quality can be improved by using milt from multiple males during spawning, with subsequent increases in overall offspring tness. The use of pooled milt was recommended for spawning of many sh species by Piper et al. (1982). However, the use of pooled milt has produced mixed results in a variety of salmonids, such as Chinook salmon (Withler 1988; Withler and Beacham 1994), rainbow trout O. mykiss (Gile and Ferguson 1995; Babiak et al. 1998), pink salmon O. gorbuscha (Gharrett and Shirley 1985), coho salmon O. kisutch (Fleming and Gross 1994), and sockeye salmon O. nerka (Foote et al. 1997). The potentially negative genetic consequences of using pooled milt during spawning procedures led Simmons and Kotiaho (2002) and Campton (2004) to recommend spawning
*Corresponding author: matt.wipf@state.sd.us 1 Present address: South Dakota Game, Fish, and Parks, McNenny State Fish Hatchery, 19619 Trout Loop, Spearsh, South Dakota 57783, USA. Received December 5, 2011; accepted March 25, 2012
443
444
WIPF AND BARNES
ratios of one male to one female. Likewise, Withler (1988) stated that in small populations of sh, eggs and milt should be added at a one female to one male ratio so as to combat genetic homogeneity and reduce the chances of genetic bottlenecking. There has been no prior research examining paternal inuences on landlocked fall Chinook salmon embryo characteristics. Thus, the objective of the current study was to determine the inuence of male gametes on the subsequent embryo survival to the eyed egg stage, as well as the fry lengths and weights at complete yolk sac absorption. METHODS Spawning occurred during October, 2008 at Whitlocks Spawning Station, Lake Oahe, using randomly selected ripe sh that had ascended the sh ladder. After each sh was euthanized using carbon dioxide, milt was manually stripped from 16 males (Piper et al. 1982), split into 16 discrete tubes per male, and then placed on ice. Eggs were pneumatically spawned from ripe females with compressed oxygen at a pressure of approximately 0.56 kg/cm2 injected via a needle into the coelomic cavity to express the eggs into a plastic pan (33 38 cm, 13 cm deep). The entire egg contribution from an individual female was divided into four similar-sized subgroups, each placed into a separate container (15 15 cm, 5 cm deep). Each of these subgroups was fertilized by milt from one of four discrete males. Using the milt from the same four males, this process was repeated with the subdivided egg contributions from three additional females. This entire procedure was then replicated three times, using four new females and four new males each time, for a total of 16 males and 16 females in the study. No spawn containing noticeably overripe eggs was used (Barnes et al. 2000; 2003a). Weights (g) and total lengths (mm) were recorded from all of the sh after spawning. After placement of the milt from discrete males into the eggs from discrete females, sperm was activated using 16 C lake water for 1 min. The fertilized eggs were then washed with lake water, put into discrete, lake water-lled plastic bags, and transported 4 h to McNenny State Fish Hatchery, rural Spearsh, South Dakota (Barnes et al. 1999a). Upon arrival at McNenny,
the eggs were disinfected in a 100-mg/L active iodine solution for 10 min and inventoried using the water displacement method (Piper et. al. 1982). Eggs were then placed into discrete vertical-ow upwelling incubator trays and incubated in 11 C well water (total hardness as CaCO3 , 360 mg/L; alkalinity as CaCO3 , 210 mg/L; pH, 7.6; total dissolved solids, 390 mg/L). At the eyed egg stage (incubation day 32), all dead eggs were removed, eyed eggs were reinventoried, and returned to discrete incubator trays. If survival in the egg group was extremely low (<1%) the remaining eggs and possible subsequent fry were considered too few to be representative, the egg group was not retrayed. Percent survival to the eyed stage was calculated: 100[1 (egg mortality/initial number of eggs)]. At complete yolk sac absorption, fry lengths and weights were recorded. Because percentage data are not normally distributed (Ott 1984), egg survival data were arcsine-transformed prior to analysis of variance (ANOVA). Percent survival data were analyzed using a randomized complete block design, the eggs from each female acting as a block. Two-way ANOVA for incomplete blocks was conducted on swim-up fry length and weight data. Analysis was completed using the General Linear Model, Univariate Analysis function of the SPSS, version 9.0, statistical analysis program (SPSS, Inc., Chicago, Illinois). Signicance was set at = 0.05.
RESULTS AND DISCUSSION Mean spawning female lengths (702 mm) and postspawn weights (3,027 g) were not signicantly different among the replicates (Table 1) nor were male lengths (731 mm) and weights (3,946 g). There was no signicant effect of male parentage on embryo survival (Table 2). However, spawning females did have a signicant effect on egg survival. There was no signicant interaction between spawning males and females on egg survival. Fry lengths and weights at complete yolk sac absorption were also signicantly different among the female contributions, but showed no paternal inuence (Table 3). Although mean survival to the eyed egg stage in this study was very low and only ranged
TABLE 1. Mean postspawning length (mm) and weight (g) of male and female fall Chinook salmon broodsh from Lake Oahe, South Dakota.
Mean SE by group Gender and size Female Length (mm) Weight (g) Male Length (mm) Weight (g)
a
1 721 3 2,965 129 731 28 3,880 469
2 713 18 3,229 340 755 11 4,166 272
3 696 13 3,111 146 725 23 3,795 361
4 678 16 2,803 218 713 13

a
Overall 702 8 3,027 108 731 10 3,946 202
Weights not obtained due to mechanical error.
MALE EFFECTS ON PROGENY SURVIVAL
445
TABLE 2. Percent survival to the eyed egg stage of development of progeny from 16 discrete pairs of autonomous male and female fall Chinook salmon broodsh in each of four different groups and overall means for each male and female.
Female Male 1 2 3 4 Mean SE 1 65.5 81.9 77.4 78.7 75.9 0.8 5 15.1 16.2 11.4 7.4 12.5 1.1 9 0.5 0.9 5.0 3.6 2.5 1.4 13 4.4 6.6 4.6 11.5 6.8 1.3 2 Group 1 55.8 52.1 49.6 51.1 52.2 0.4 Group 2 5 6 7 8 Mean SE 6 33.3 14.9 42.7 35.2 31.5 2.1 Group 3 9 10 11 12 Mean SE 10 10.5 7.6 10.5 12.5 10.3 0.6 Group 4 13 14 15 16 Mean SE 14 59.3 46.9 47.3 42.4 49.0 1.0 15 3.2 3.4 7.6 3.7 4.5 1.0 16 0.2 0.6 0.2 1.0 0.5 0.6 16.8 14.4 21.8 19.0 6.9 5.8 5.6 5.0 11 0.7 8.0 9.7 11.5 7.5 1.7 12 6.8 3.4 2.2 2.1 3.6 1.2 4.6 5.0 6.8 7.4 2.3 1.5 1.5 2.0 7 49.7 39.9 43.2 41.0 43.5 0.7 8 52.4 50.4 29.8 44.3 44.2 1.5 37.6 30.3 31.8 32.0 2.8 3.2 2.7 3.0 19.7 10.9 4.1 8.7 10.9 2.0 31.9 20.3 20.3 27.8 25.1 1.2 43.2 41.3 37.8 41.6 3.2 3.3 5.3 4.7 3 4 Mean SE
from 4.6% to 43.2%, these values are within the normal range for this salmon population (Barnes et al. 1997). Egg survival is a function of both fertilization and embryo development. This study did not differentiate between these two components but rather looked at their combined effect on survival to the eyed stage of development. It is possible that paternal effects may occur at either fertilization or during development, and that these effects may be masked by just focusing on overall egg survival. The results of this study indicate minimal paternal effects on progeny survival. The differences in reproductive success could be attributed almost entirely to the spawning female. Similarly Nagler et al. (2000) found that female rainbow trout had a signicant effect on embryo survival and that male contributions to embryo survival were negligible. Both Smith and Fretwel (1974) and Einum and Fleming (2000) found that increases in embryo survival and growth were due to maternal tness. Maternal effects are obvious with respect to egg size, which can be an
indicator of maternal tness and a predictor of embryo survivability (Einum et al. 2002; Mousseau and Fox, 1998). However, previous studies have stated embryo survival and development can be independently related to parental donations whether paternal (Aas et al. 1991) or maternal (Springate et al. 1984). The use of pooled milt was historically used in Pacic salmon hatcheries (Billard et al. 1996) and is the established protocol for salmon spawning in South Dakota (Barnes et al. 2003a). Withler and Beacham (1994) suggest using pooled milt to maximize spawning efciency and potential fertilization success. Spawning time is particularly important in large production hatcheries or remote facilities where personnel travel times must be considered. However, using pooled milt may reduce the effective number of breeders and therefore reduce genetic diversity and effective population sizes (Babiak 1998; Withler 1988; Campton 2004; Wedekind et al. (2007)). Although not directly addressed in this study, using pooled milt may be inuencing the reproductive success of Lake Oahe
446
WIPF AND BARNES
TABLE 3. Lengths (mm) and weights (g) of fry progeny from 16 discrete pairs of autonomous male and female fall Chinook salmon broodsh in each of four different groups and overall means for each male and female. Dashes indicate unsuccessful pairing.
Female Male 1 2 3 4 Mean SE 1 33.3 33.8 33.7 30.3 32.8 0.8 5 33.2 33.1 34.1 32.7 33.3 0.3 9 13 32.8 33.3 33.6 32.6 33.1 0.2 1 0.29 0.30 0.28 0.28 0.29 0.01 5 0.29 0.29 0.29 0.28 0.29 0.01 2 3 4 31.1 30.9 32.4 31.7 31.5 0.3 8 31.5 32.0 32.3 32.4 32.0 0.2 12 33.2 31.3 32.8 32.7 32.5 0.4 16 4 0.25 0.26 0.26 0.26 0.26 0.01 8 0.24 0.24 0.24 0.25 0.24 0.01 Mean SE 32.3 32.4 32.3 31.8 0.5 0.6 0.6 0.6
5 6 7 8 Mean SE
9 10 11 12 Mean SE
Lengths (mm) for group 1 33.0 31.8 32.8 31.9 32.2 30.9 32.9 32.4 32.7 0.2 31.7 0.3 Lengths (mm) for group 2 6 7 34.0 31.5 34.9 32.0 34.4 33.6 35.5 32.9 34.7 0.3 32.5 0.5 Lengths (mm) for group 3 10 11 33.5 34.1 33.7 33.9 33.8 0.1 Lengths (mm) for group 4 14 15 32.7 34.3 32.3 33.7 33.2 33.5 32.1 33.6 32.6 0.2 33.8 0.2 Weights (g) for group 1 2 3 0.25 0.25 0.25 0.24 0.25 0.26 0.25 0.24 0.25 0.01 0.25 0.01 Weights (g) for group 2 6 7 0.27 0.25 0.29 0.25 0.28 0.26 0.29 0.26 0.29 0.01 0.26 0.01
32.6 33.0 33.6 33.4
0.6 0.7 0.5 0.7
33.3 32.7 33.2 33.3
0.2 1.4 0.4 0.6
13 14 15 16 Mean SE
33.3 33.1 33.4 32.8
0.5 0.4 0.1 0.5
1 2 3 4 Mean SE
0.26 0.26 0.26 0.26
0.01 0.01 0.01 0.01
5 6 7 8 Mean SE
0.26 0.27 0.27 0.27
0.01 0.01 0.01 0.01
MALE EFFECTS ON PROGENY SURVIVAL TABLE 3. Continued.
447
9 9 10 11 12 Mean SE 13 0.25 0.25 0.25 0.25 0.25 0.01
Weights (g) for group 3 10 11 0.26 0.26 0.27 0.26 0.26 0.01
12 0.24 0.25 0.25 0.25 0.25 0.01 16 0.25 0.25 0.25 0.25 0.01 0.01 0.01 0.01
13 14 15 16 Mean SE
Weights (g) for group 4 14 15 0.23 0.29 0.22 0.29 0.23 0.28 0.22 0.29 0.22 0.01 0.29 0.01
0.26 0.25 0.25 0.26
0.02 0.02 0.02 0.02
salmon population during hatchery rearing by decreasing the genetic health of the spawning population. The relatively small size and genetic isolation (Barnes et al. 2000) of this unique population, coupled with the use of pooled milt, may lead to potential inbreeding, which has been linked to poor embryo survival (Klug et al. 2007). The genetic status of Lake Oahe salmon is unknown, and there is no genetic spawning protocol for this population. Because there was no signicant paternal effect on egg survival during pairwise spawning, this may be an effective technique to negate possible homogeneity, potentially improve genetic diversity, and possibly increase embryo survival in Lake Oahe fall Chinook salmon, unless there are overriding spawning efciency and timing concerns. REFERENCES
Aas, G. H., T. Refstie, and B. Gjerde. 1991. Evaluation of milt quality of Atlantic salmon. Aquaculture 95:125132. Alcock, J. 2005. Animal behaviour: an evolutionary approach, 8th edition. Sinauer, Sunderland, Massachusetts. Anderson, M. 1994. Sexual selection. Princeton University Press, Princeton, New Jersey. Babiak, I., J. Glogowski, M. Luczynski, K. Goryczko, S. Dobesz, and H. Kuzminski. 1998. The effect of individual male potency on fertilization ability of fresh and cryopreserved milt of rainbow trout, Oncorhynchus mykiss (Walbaum). Aquaculture Research 29:337340. Barnes, M. E., R. J. Cordes, and W. A. Sayler. 1997. Use of formalin during incubation of eyed eggs of inland fall Chinook salmon. Progressive FishCulturist 59:303306. Barnes, M. E., R. P. Hanten, R. J. Cordes, W. A. Sayler, and J. Carreiro. 2000. Reproductive performance of inland fall Chinook salmon. North American Journal of Aquaculture 62:203211. Barnes, M. E., R. P. Hanten, J. J. P. Lott, and M. Gabel. 2001. Environmental inuences on landlocked fall Chinook salmon reproductive characteristics. North American Journal of Aquaculture 63:5865. Barnes, M. E., J. P. Lott, W. A. Sayler, and R. J., Cordes. 1999b. Practical observations on the use of eggs from electroshocked females during spawning of inland fall Chinook salmon. North American Journal of Aquaculture 61:162166.
Barnes, M. E., W. A. Sayler, and R. J. Cordes, 1999a. Transportation inuences on inland fall Chinook salmon egg survival. North American Journal of Aquaculture 61:2733. Barnes, M. E., W. Sayler, J. C. Cordes, and R. P. Hanten. 2003a. Potential indicators of egg viability in landlocked fall Chinook salmon spawn with or without the presence of overripe eggs. North American Journal of Aquaculture 65:4955. Barnes, M. E., M. H. Zehfus, J. A. Schumacher, K. S. Stock, F. Farrokhi, and R. L. Nutter. 2003b. Thiamine inuences on landlocked fall Chinook salmon reproductive characteristics. Prairie Naturalist 35:113116. Billard, R., J. Jorgen, and O. T. Jensen. 1996. Gamete removal, fertilization and incubation. Pages 291364 in W. Pennel and B. A. Batron, editors. Developments in aquaculture and sheies science volume 29: principles of salmonid culture. Elsevier, Amsterdam, The Netherlands. Campton, D. E. 2004. Sperm competition in salmon hatcheries: the need to institutionalize genetically benign spawning protocols. Transactions of the American Fisheries Society 133:12771289. Einum, S., and I. A. Fleming. 2000. Highly fecund mothers sacrice offspring survival to maximize tness. Nature 405:565567. Einum, S., A. P. Hendry, and I. A. Fleming. 2002. Egg-size evolution in aquatic environments: does oxygen availability constrain size? Proceedings of the Royal Society London Biological Sciences 269:23252330. Fleming, I. A., and M. R. Gross. 1994. Breeding competition in a Pacic salmon (coho: Oncorhynchus kisutch): measures of natural and sexual selection. Evolution 48:637657. Foote, C. J., C. S. Brown, and C. C. Wood. 1997. Spawning success of males using alternative mating tactics in sockeye salmon, Oncorhynchus nerka. Canadian Journal of Fisheries and Aquatic Sciences 54:17851795. Gharrett, A. J., and S. M. Shirley. 1985. A genetic examination of spawning methodology in a salmon hatchery. Aquaculture 47:245256. Gile, R. S., and M. M. Ferguson. 1995. Factors affecting male potency in pooled gamete crosses of rainbow trout, Oncorhynchus mykiss. Environmental Biology of Fishes 42:267275. Klug, W. S., M. R. Cummings, and C. A. Spencer. 2007. Essentials of genetics, 6th edition. Pearson Prentice Hall, Upper Saddle River, New Jersey. Mousseau, T. A., and C. W. Fox. 1998. Maternal effects as adaptations. Oxford University Press. New York. Nagler, J. J., J. E. Parsons, and J. G. Cloud. 2000. Single pair mating indicates maternal effects on embryo survival in rainbow trout, Oncorhynchus mykiss. Aquaculture 184:177183. Ott, L. 1984. An introduction to statistical methods and data analysis. PWS Publishers, Boston.
448
WIPF AND BARNES Springate, J. R. C., N. R. Bromage, J. A. K. Elliot, and D. L. Hudson. 1984. The timing of ovulation and stripping and their effects on the rates of fertilization and survival to eying, hatch and swim-up in the rainbow trout (Salmo gairdneri R.) Aquaculture 43:313322. Wedekind, C., G. Rudolfsen, A. Jacob, D. Urbach, and R. M uller. 2007. The genetic consequences of hatchery-induced sperm competition in a salmonid. Biological Conservation 137:180188. Withler, R. E. 1988. Genetic consequences of fertilizing Chinook salmon (Oncorhynchus tshawytscha) eggs with pooled milt. Aquaculture 68: 1525. Withler, R. E., and T. D. Beacham. 1994. Genetic consequences of the simultaneous or sequential addition of semen from multiple males during hatchery spawning of Chinook salmon (Oncorhynchus tshawytscha). Aquaculture 126:1123.
Piper, R. G., I. B. McElwain, L. E. Orme, J. P. McCaren, L. G. Fowler, and J. R. Leonard. 1982. Fish hatchery management. U.S. Fish and Wildlife Service, Washington, D.C. Simmons, L. W., and J. S. Kotiaho. 2002. Evolution of ejaculates: patterns of phenotypic and genotypic variation and condition dependence in sperm competition traits. Evolution 56:16221631. Smith, C. C., and S. D. Fretwell. 1974. The optimal balance between size and number of offspring. American Naturalist 108:499506. Smoker, W. W., A. J. Gharrett, M. S. Stekoll, and S. G. Taylor. 2000. Genetic variation of fecundity and egg size in anadromous pink salmon Oncorhynchus gorbuscha Walbaum. Alaska Fishery Research Bulletin 7:4450. Smoker, W. W., I. A. Wang, A. J. Gharrett, and J. J. Hard. 2004. Embryo survival and smolt to adult survival in second-generation outbred coho salmon. Journal of Fish Biology 65:254262.
This article was downloaded by: [Department Of Fisheries] On: 26 September 2012, At: 00:00 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Koi Goldfish Hybrid Females Produce Triploid Progeny when Backcrossed to Koi Males
Boris Gomelsky , Kyle J. Schneider & Debbie A. Plouffe
a a a b
Aquaculture Research Center, Kentucky State University, 103 Athletic Road, Frankfort, Kentucky, 40601, USA
b
AquaBounty Canada, Inc., 718 Bay Fortune, Souris, Prince Edward Island, C0A 2B0, Canada
Version of record first published: 04 Sep 2012.
To cite this article: Boris Gomelsky, Kyle J. Schneider & Debbie A. Plouffe (2012): Koi Goldfish Hybrid Females Produce Triploid Progeny when Backcrossed to Koi Males, North American Journal of Aquaculture, 74:4, 449-452 To link to this article: http://dx.doi.org/10.1080/15222055.2012.676014
COMMUNICATION
Koi Goldsh Hybrid Females Produce Triploid Progeny when Backcrossed to Koi Males
Boris Gomelsky* and Kyle J. Schneider
Aquaculture Research Center, Kentucky State University, 103 Athletic Road, Frankfort, Kentucky 40601, USA
Debbie A. Plouffe
AquaBounty Canada, Inc., 718 Bay Fortune, Souris, Prince Edward Island C0A 2B0, Canada
Abstract
Hybrids of koi (an ornamental variant of the common carp Cyprinus carpio) and goldsh Carassius auratus auratus were produced by articial spawning. All 3-year-old F1 hybrid males examined were sterile, whereas some F1 hybrid females were fertile and produced eggs after hormonal injection. Backcross progeny were obtained by using intact koi sperm to inseminate eggs from F1 hybrid females; gynogenetic progeny were obtained by inseminating eggs from F1 hybrid females with koi sperm that was genetically inactivated by ultraviolet irradiation. Flow cytometric analysis of DNA content indicated that the backcross progeny were triploid, while the gynogenetic progeny, pure koi, pure goldsh, and F1 hybrids were all diploid. The triploidy of backcross progeny obtained without application of any treatment to the eggs demonstrates that the koi goldsh hybrid females produce diploid eggs.
The common carp Cyprinus carpio and goldsh Carassius auratus auratus are nonnative cyprinid species that naturally reproduce in North America (Panek 1987; Schoeld et al. 2005). Ornamental forms of common carp (i.e., koi) and goldsh are popular decorative sh in many countries throughout the world, including the United States. Goldsh are also raised in the United States for use as bait sh and as forage in sh hatcheries (Schoeld et al. 2005). Common carp naturally hybridize with goldsh in their native range and in many areas where both species have been introduced (Trautman 1957; Bardach et al. 1972; Taylor and Mahon 1977; Pullan and Smith 1987; Schoeld et al. 2005). Cases of hybridization between koi and goldsh have also been described in the aquarium sh and koi hobbyist literature (Hemdal 2003; Muha 2007). Hybrids of common carp or koi and goldsh have been produced articially for different types of studies (e.g., Suzuki 1962; Hedrick et al. 2006). The morphometric and meristic characteristics of common carp (or koi)
*Corresponding author: boris.gomelsky@kysu.edu Received November 11, 2011; accepted February 18, 2012
goldsh hybrids have been widely reported (Taylor and Mahon 1977; Pulan and Smith 1987; Schoeld et al. 2005). In contrast to many interspecies sh hybrids, the hybrids of common carp and goldsh are not sterile, and the backcrossing of hybrids with parental species has been reported (Trautman 1957; Bardach et al. 1972). Previous studies on hybridization of common carp with two other subspecies of C. auratus (Japanese crucian carp C. auratus cuvieri and silver crucian carp C. auratus gibelio) revealed that F1 hybrid females produced triploid progeny when backcrossed to males of parental species (Ojima et al. 1975; Cherfas et al. 1994). Triploidy of backcross hybrids resulted from the diploidy of eggs produced by hybrid females. The present study was performed to investigate whether koi goldsh hybrid females have the same feature. For this purpose, the ploidy of backcross progeny obtained from hybrid females was determined. METHODS Fish spawning and rearing were conducted at the Aquaculture Research Center, Kentucky State University, Frankfort. Koi goldsh hybrids were produced by articial spawning in spring 2008. To induce nal oocyte maturation and ovulation in females and spermiation in males, broodsh were given intramuscular injections of carp pituitary extract (Sigma Chemical, St. Louis, Missouri) at a concentration of 3 mg/kg of body weight. Males received a single injection approximately 16 h before stripping of sperm. Females received two injections (10% and 90% of the total dose) 12 h apart. After injections, broodsh were kept in tanks at a water temperature of 21.5 C; ovulation was observed 1112 h after the resolving injection. Eggs and sperm were collected from broodsh by hand stripping. A mixture of eggs from six koi females of different colors (whitered, whiteyellow, and
449
450
GOMELSKY ET AL.
solid white) was fertilized with a mixture of sperm taken from two solid-red goldsh males. Eggs were articially inseminated in plastic bowls according to standard techniques for common carp breeding (Horvath et al. 2002) and were treated with a solution of water : pasteurized cows milk (volumetric ratio = 8:1) to remove adhesiveness. Embryos were incubated in McDonald hatching jars. The resulting F1 hybrids were raised in 20-m3 outdoor tanks during their rst summer and then were transferred to earthen ponds, where they were reared for an additional 2.5 years. The functional fertility of 3-year-old hybrids was investigated in spring 2011. For this purpose, 10 selected hybrid females and 20 hybrid males were injected with carp pituitary extract by using the dosages and methodology described above for koi and goldsh breeders. After hormonal injection, hybrid males did not release any sperm. To determine whether the testes of hybrid males contained any viable spermatozoa, we sacriced the sh, removed their testes, and thoroughly dissected and washed the testes with a saline solution (0.85% NaCl). The resulting suspension was used for insemination of koi eggs. Five koi goldsh hybrid females produced ovulated eggs after hormonal injection and were used in crosses. Two separate groups of backcross progeny were obtained by crossing F1 hybrid females with koi males. Backcross progeny group 1 was obtained by using sperm from a single koi male to inseminate a mixture of eggs collected from four hybrid females. Backcross progeny group 2 was produced by crossing a single hybrid female with a single koi male. Gynogenetic progeny were also obtained by using genetically inactivated (by ultraviolet [UV] irradiation) koi sperm to inseminate eggs from the same F1 hybrid female that was used to produce backcross progeny group 2. Sperm was irradiated with a UV crosslinker (FB-UVXL-1000; Fisher Scientic) at 4,000 J/m2. This dosage of UV irradiation was chosen based on previously performed experiments involving the induction of gynogenesis in koi (Alsaqu 2011). Before irradiation, the sperm was diluted (1:9) in a saline solution (0.85% NaCl). The backcross progeny groups and the gynogenetic progeny were obtained without ap-
plication of any treatment to the eggs. After transition to active feeding, the larvae were stocked in 20-m3 outdoor tanks for rearing. Ploidy of 4-month-old backcross progeny (mean total length SD = 13.8 0.85 cm) and gynogenetic progeny (12.8 0.75 cm) was determined by ow cytometric analysis of DNA content. The analyses were performed at AquaBounty Canada, Inc., Prince Edward Island, by using a Becton Dickinson (BD) FACSCalibur ow cytometer. The ploidy of some koi, goldsh, and F1 hybrids was also analyzed for comparison. Instrument quality control for DNA quantitation was performed using CellQuest Pro software and DNA QC particles (BD, catalog number 349523) to assess resolution and linearity. Blood samples were drawn from each sh by caudal venipuncture and were collected in 3.0-mL Vacutainer tubes containing lithium heparin (BD, catalog number 366667). For each sample, two drops of heparinized blood from a syringe were collected in 500 L of sheath uid (BioSure, catalog number 1019). An 80-L quantity of the bloodsheath uid mixture was stained in 500 L of propidium iodide solution containing a detergent for lysing (BioSure, catalog number 1021) along with 40 L of chicken red blood cells (BioSure, catalog number 1005) as an internal staining control. Samples were incubated in propidium iodide solution in the dark for 10 min prior to analysis. For each sample analyzed, 10,000 events were recorded and the relative DNA content was determined as the ratio of sample uorescence peak intensity to internal standard (i.e., chicken red blood cells) uorescence peak intensity.
RESULTS The koi goldsh hybrids exhibited dark coloration that is typical of wild-type common carp and goldsh. After eggs from female koi were mixed with the suspension of dissected testes from F1 hybrid males, examination of eggs under a dissecting microscope indicated that none of the eggs was fertilized. The results of ploidy analysis for backcross progeny, gynogenetic progeny, the parental species, and the F1 hybrids are
TABLE 1. Ploidy of koi, goldsh, F1 hybrids, and backcross and gynogenetic progeny groups as determined by ow cytometric analysis. The relative DNA content was calculated as the ratio of sample uorescence peak intensity to internal standard uorescence peak intensity.
Relative DNA content Group Koi Goldsh F1 ( koi goldsh) Backcross progeny group 1 ( F1 koi) Backcross progeny group 2 ( F1 koi) Gynogenetic progeny ( F1 koi [inactivated sperm]) Number of sh analyzed 4 2 10 15 12 12 Mean (SD) 2.56 (0.04) 2.52 2.56 (0.05) 3.75 (0.08) 3.84 (0.07) 2.51 (0.04) Range 2.532.62 2.512.52 2.492.60 3.613.84 3.733.96 2.432.59 Fish ploidy 2n 2n 2n 3n 3n 2n
COMMUNICATION
451
presented in Table 1. The mean relative DNA content was similar for koi, goldsh, F1 hybrids, and gynogenetic progeny and varied from 2.51 to 2.56. The mean relative DNA content was 3.75 in backcross progeny group 1 and 3.84 in backcross progeny group 2. The ratio of DNA content in backcross progeny to the DNA content in the other groups was approximately 1.5, indicating that the backcross progeny were triploid (3n) sh, whereas the gynogenetic progeny, parental species, and F1 hybrids were all diploid (2n).
The koi goldsh hybrids had a color pattern that is typical of wild-type common carp and goldsh, which indicates that the alleles causing melanin depigmentation in parental forms are not expressed in the hybrids. The analysis of color inheritance in hybrids will be presented in a separate publication.
ACKNOWLEDGMENTS Support for this study was provided by Kentuckys Regional University Trust Fund to the Aquaculture Program as Kentucky State Universitys Program of Distinction.
DISCUSSION The results of the present study reveal the functional genetic sterility of F1 males produced by the hybridization of koi and goldsh. These males developed testes, but their gonads did not contain spermatozoa with the capability of fertilizing eggs. Sterility of goldsh common carp hybrid males was also reported by Yamaha et al. (2003). Flow cytometric analysis indicated that the DNA content in koi and goldsh was similar. These results concur with data reported by Ohno et al. (1967), who indicated that common carp and goldsh had comparable amounts of DNA comprising 5052% of the typical DNA content in placental mammals. Similar mean values of relative DNA content were observed in F1 hybrids, whereas backcross progeny appeared to be triploid. The triploidy of backcross progeny obtained from hybrid females (without application of any treatment to eggs) indicates that these females produced diploid eggs. Diploidy of eggs from hybrid females resulted in diploid gynogenetic progeny, which were produced by use of genetically inactivated spermatozoa chromosomes but without application of any treatment to the eggs. Hybrid females obtained by crossing common carp males with females of two other C. auratus subspecies (Japanese crucian carp and silver crucian carp) have previously been shown to produce diploid eggs (Ojima et al. 1975; Cherfas et al. 1994). The results of the present study demonstrate that the same phenomenon is observed in hybrids obtained from two popular ornamental shes, the koi and goldsh. Specically, the hybrid females producing diploid eggs were obtained by crossing koi females with goldsh males. The ability of interspecies F1 hybrids to produce diploid eggs has been described for several other sh taxa in addition to Carassius and Cyprinus; these include hybrids of the brown trout Salmo trutta and Atlantic salmon S. salar (Johnson and Wright 1986; Galbreath and Thorgaard 1995) and hybrids of the pumpkinseed Lepomis gibbosus and green sunsh L. cyanellus (Dawley et al. 1985; Dawley 1987). Cherfas et al. (1994) and Shimizu et al. (2000) showed that generation of unreduced diploid eggs by hybrid females results from the occurrence of premeiotic endomitosis (i.e., doubling of chromosomes without cytokinesis) in early oogenesis; the resulting tetraploid oocytes undergo two normal, consecutive meiotic divisions.
REFERENCES
Alsaqu, A. S. 2011. Application of microsatellite DNA markers in studies on induced gynogenesis in ornamental (koi) carp. Masters thesis. Kentucky State University, Frankfort. Bardach, J. E., J. H. Ryther, and W. O. McLarney. 1972. Aquaculture: the farming and husbandry of freshwater and marine organisms. Wiley, New York. Cherfas, N. B., B. I. Gomelsky, O. V. Emelyanova, and A. V. Recoubratsky. 1994. Induced diploid gynogenesis and polyploidy in crucian carp, Carassius auratus gibelio (Bloch), common carp, Cyprinus carpio L., hybrids. Aquaculture and Fisheries Management 25:943954. Dawley, R. M. 1987. Hybridization and polyploidy in a community of three sunsh species (Pisces: Centrarchidae). Copeia 1987:326335. Dawley, R. M., J. M. Graham, and R. J. Schultz. 1985. Triploid progeny of pumpkinseed green sunsh hybrids. Journal of Heredity 76:251257. Galbreath, P. F., and G. H. Thorgaard. 1995. Sexual maturation and fertility of diploid and triploid Atlantic salmon brown trout hybrids. Aquaculture 137:299311. Hedrick, R. P., T. B. Waltzek, and T. S. McDowell. 2006. Susceptibility of koi carp, common carp, goldsh and goldsh common carp hybrids to cyprinid herpesvirus-2 and herpesvirus-3. Journal of Aquatic Animal Health 18:2634. Hemdal, J. F. 2003. Aquarium sh breeding. Barrons, Hauppauge, New York. Horvath, L., G. Tamas, and C. Seagrave. 2002. Carp and pond sh culture, 2nd edition. Blackwell Scientic Publications, Oxford, UK. Johnson, K. R., and J. E. Wright. 1986. Female brown trout Atlantic salmon produce gynogens and triploids when backcrossed to male Atlantic salmon. Aquaculture 57:345358. Muha, L. 2007. Spring ponds part 2: the good, the bad and the ugly. Tropical Fish Hobbyist (April):6467. Ohno, S., J. Muramoto, L. Christian, and N. B. Atkin. 1967. Diploid-tetraploid relationship among old world members of the sh family Cyprinidae. Chromosoma 23:19. Ojima, Y., M. Hayashi, and K. Ueno. 1975. Triploidy appeared in the backcross offspring from funa-carp crossing. Proceedings of Japanese Academy of Science 51:702706. Panek, F. M. 1987. Biology and ecology of carp. Pages 115 in E. L. Cooper, editor. Carp in North America. American Fisheries Society, Bethesda, Maryland. Pullan, S., and P. J. Smith. 1987. Identication of hybrids between koi (Cyprinus carpio) and goldsh (Carassius auratus). New Zealand Journal of Marine and Freshwater Research 21:4146. Schoeld, P. J., J. D. Williams, L. G. Nico, and M. R. Thomas. 2005. Foreign nonindigenous carps and minnows (Cyprinidae) in the United Statesa guide to their identication, distribution, and biology. U.S. Geological Survey, Scientic Investigations Report 20055041. Shimizu, Y., N. Shibata, M. Sakaizumi, and M. Yamashita. 2000. Production of diploid eggs through premeiotic endomitosis in the hybrid medaka
452
GOMELSKY ET AL. Trautman, M. B. 1957. Fishes of Ohio with illustrated keys. Ohio State University Press, Columbus. Yamaha, E., M. Murakami, K. Hada, S. Otani, T. Fujimoto, M. Tanaka, S. Sakao, S. Sato, and K. Arai. 2003. Recovery of fertility in male hybrids of a cross between goldsh and common carp by transplantation of PGC (primordial germ cell)-containing graft. Genetica 119:121 131.
between Oryzias latipes and O. curvinotus. Zoological Science 17:951 958. Suzuki, R. 1962. On the behavior of carp-goldsh hybrids. Japanese Journal of Ichthyology 10:1315. Taylor, J., and R. Mahon. 1977. Hybridization of Cyprinus carpio and Carassius auratus, the rst two exotic species in the lower Laurentian Great Lakes. Environmental Biology of Fishes 1:205208.

Demonstration of Blue Crab Culture in Inland LowSalinity Waters of West Alabama

Luke A. Roy , Gregory N. Whitis & William C. Walton
a a a a
Department of Fisheries and Allied Aquacultures, Auburn University, 203 Swingle Hall, Auburn, Alabama, 368495419, USA Version of record first published: 04 Sep 2012.
To cite this article: Luke A. Roy, Gregory N. Whitis & William C. Walton (2012): Demonstration of Blue Crab Culture in Inland Low-Salinity Waters of West Alabama, North American Journal of Aquaculture, 74:4, 453-456 To link to this article: http://dx.doi.org/10.1080/15222055.2012.676006
COMMUNICATION
Demonstration of Blue Crab Culture in Inland Low-Salinity Waters of West Alabama

Luke A. Roy,* Gregory N. Whitis, and William C. Walton
Department of Fisheries and Allied Aquacultures, Auburn University, 203 Swingle Hall, Auburn, Alabama 368495419, USA
Abstract
The decline in natural stocks of blue crabs Callinectes sapidus in U.S. coastal waters has created more interest in the aquaculture of this species. The 2010 Deepwater Horizon oil spill threatened the health of blue crab populations in the Gulf of Mexico and spurred interest in alternative sources of blue crab harvest. Inland lowsalinity waters (ILSW) of west Alabama have a proven potential for the culture of euryhaline crustaceans. This study sought to evaluate the potential for blue crab culture in ILSW for commercial aquaculture and as a potential nursery for restoration efforts. A 21-d bioassay was conducted using four different pond waters of differing ionic composition and salinity (2.05.7). No signicant differences in survival, weight gain (%), and nal weight were observed among different pond waters. An on-farm demonstration was conducted in a 1.5-acre low-salinity pond (5.0) at a shrimp farm in west Alabama. Crabs grew from 2.0 to 202.0 g in less than 120 d when offered a varied diet of commercial shrimp feed, catsh offal, and fresh whole catsh. Results from these trials suggest that there is excellent biological potential for culture of blue crabs in ILSW of west Alabama.
The shery for blue crab Callinectes sapidus is the most valuable shery in several states, including North Carolina, Maryland, Virginia, and Mississippi, having an estimated value approaching US$200 million/year (Eggleston et al. 2009). The decline in capture harvests of blue crab has increased interest in the aquaculture of this species, particularly in the four states previously mentioned. Along the Gulf Coast, the 2010 BP Deepwater Horizon Oil Spill posed an additional threat to natural populations of blue crab and other commercial species, and spurred further interest in culture of this species within the region. The Alabama inland shrimp industry is concentrated about 150 mi north of Mobile and the Gulf of Mexico, where inland artesian groundwater salinity ranges from 2 to 10 (Roy et al. 2010). Inland low-salinity water (ILSW) suitable for
*Corresponding author: royluke@auburn.edu Received October 4, 2011; accepted February 14, 2012
mariculture has been identied in several counties in west Alabama (Boyd et al. 2009). Inland shrimp farmers have been growing the Pacic white shrimp Litopenaeus vannamei in west Alabama since 1999 and are currently producing 250,000 lb of shrimp annually, which has a farm gate value of over $1 million (D. Teichert-Coddington, Alabama Inland Shrimp Producers Association, personal communication). Farmers in west Alabama have also successfully cultured a number of other marine species in ILSW, including Gulf killish Fundulus grandis (Phelps et al. 2010; S. McNulty, Aquatic Innovations LLC, personal communication) and pinsh Lagodon rhomboides (McNulty, personal communication). Elsewhere in the United States, ILSW has also been used to culture red drum Sciaenops ocellatus (Miranda and Sonski 1987). Thus, the viability of ILSW has been well established for rearing euryhaline marine species that are capable of effective osmotic and ionic regulation in low-salinity environments. In order to effectively grow marine species in ILSW of west Alabama, farmers must supplement potassium and magnesium fertilizers to their pond water. The ILSW in west Alabama is naturally decient in these two ions (Saoud et al. 2003). Deciencies in potassium and magnesium, as well as high sodium : potassium ratios, have resulted in reduced survival and osmoregulatory capacity of marine species cultured in ILSW (Fielder et al. 2001; Saoud et al. 2003; Roy and Davis 2010). West Alabama shrimp farmers typically counteract low levels of potassium and magnesium with fertilizers rich in these two ions, including muriate of potash (agricultural grade potassium chloride) and K-Mag (potassium magnesium sulfate; Davis et al. 2004). While blue crabs are excellent osmoregulators capable of tolerating and thriving in extremely low salinities (Henry 1988; Henry 2005), the effects of ILSW with altered ionic levels and ratios differing from dilute seawater has not yet been examined in regards to blue crab culture.
453
454
ROY ET AL.
TABLE 1. Pond water ionic composition for 21-d bioassays conducted with blue crabs using four different low-salinity shrimp pond waters.
Ions K Mga Na Caa Na:K Salinity () Total hardnessa

a
Control 53.99 646.62 1,629.15 121.40 30.18 4.60 768.02
TC S3 69.52 52.03 778.98 47.07 11.20 5.80 99.10
TC S4 30.94 47.07 819.47 101.58 26.49 5.40 148.65
DO 1 73.03 201.67 2,459.07 205.63 33.67 2.10 407.30
DO 7 16.40 282.43 2,023.87 141.22 123.39 2.10 423.65
Crab pond 35.45 180.86 2,003.63 178.38 56.53 5.00 359.23
Expressed as mg/L calcium carbonate.
The overall goal of this study was to assess the aquaculture potential of the blue crab as an alternative crop for existing shrimp producers, and to test the potential of ILSW of west Alabama as a nursery for restoration efforts. To meet this objective, growth and survival of blue crabs reared in low-salinity pond waters with a range of salinities and ionic proles was evaluated in a 21-d bioassay and commercial pond demonstration study. METHODS A 21-d bioassay was conducted in a static aquarium system located at the Alabama Fish Farming Center (AFFC) in Greensboro, Alabama. Two months prior to starting the bioassay, pond water was collected from four ponds located at two different lowsalinity shrimp farms (DO and TC) in Greene County, Alabama. Three of the four ponds (TC S3, TC S4, DO 1) had received potassium and magnesium fertilizers (muriate of potash and KMag) at rates commonly applied annually to shrimp ponds prior to the production season to raise aqueous potassium and magnesium to levels optimal for shrimp culture (Roy et al. 2010). One of the ponds (DO 7) had not yet been fertilized and had levels of potassium and magnesium that would be considered suboptimal for shrimp culture (Roy et al. 2010). Eighty gallons of water were collected from each pond using a bilge pump and transported to the AFFC in hauling tanks. Water from each pond was transferred to four randomly selected replicate 20-gal aquariums. Additionally, to serve as a control, four replicate aquariums were lled with dechlorinated tap water. The salinity of the control treatment was raised to 4.6 using reconstituted seawater (Crystal Sea Salt, Baltimore, Maryland). The aquarium system consisted of 20 aquariums, each equipped with an air stone and supplied with air from a regenerative blower. Water samples were taken at the beginning of the experiment for ion analysis (Table 1). Sodium and potassium were measured using a Cole Parmer digital ame photometer (Vernon Hills, Illinois; model 265500) according to Roy et al. (2007). Magnesium and calcium were measured according to Eaton et al. (2005). Blue crabs were obtained from the Gulf Coast Research Laboratory crab hatchery in Ocean Springs, Mississippi. Crabs were transported from Mississippi to the AFFC in coolers and stocked into the aquaria system, one crab per aquarium. The initial weight of the crabs was 2.04 0.53 g (mean SD).
Throughout the experiment, the crabs were fed daily in the mornings and evenings. The diet consisted of llets of fresh channel catsh Ictalurus punctatus or shrimp. Feed was placed in the aquaria, and crabs were allowed to feed for 60 min. Leftover feed was removed from the tank to limit ammonia and nitrite buildup in the tanks. Aquaria were checked for molts 23 times/d, and throughout the 21-d trial each crab molted twice with the exception of one crab that only molted once. Once detected, all molts were removed immediately from the experimental tanks. Dissolved oxygen, pH, and temperature were measured daily, while salinity was measured weekly. Total ammonia nitrogen was analyzed weekly according to Nesslers method (APHA et al. 1989), whereas nitrite was measured weekly according to Parsons et al. (1985). At the end of the bioassay, crabs were harvested and individually weighed to determine growth and survival. Statistical analyses were performed using SAS (version 9.2; SAS Institute, Cary, North Carolina). All data were analyzed using one-way ANOVA and StudentNewmanKeuls multiplerange test to determine if signicant differences (P < 0.05) existed among treatment means (Steel and Torrie 1980). On July 7, 2010 (the same day that the 21-d bioassay began), 4,187 blue crabs were stocked into a 1.5-acre pond (5.0) at a low-salinity shrimp farm in Greene County, Alabama. The initial weight of the crabs at stocking was 2.04 0.53 g (mean SD). The crabs were transported from Mississippi as previously described and allowed to acclimate to temperature for 1 h prior
FIGURE 1. Growth of blue crabs in a 1.5-acre low-salinity (5.0) pond fertilized with potassium and magnesium fertilizers in Greene County, Alabama, from July 2010 to June 2011. [Figure available in color online.]
COMMUNICATION
455
TABLE 2. Growth performance of blue crabs in static 20-gal aquaria with pond water from four different shrimp ponds for 21 d. Values represent the mean SD. No signicant differences were observed among treatments.
Variable
TC S3
TC S4
DO 1
DO 7 1.02 2.16 0.0 100 5.78 13.55 67.05 529.42 1.59 3.80
Control
PSEa
P-value
Initial weight (g) 2.09 Survival (%) 100.0 Final weight (g) 12.93 Weight gain (%) 516.64 Growth (g/week) 3.61
a
0.19 1.77 0.0 100.0 2.09 9.85 45.86 461.60 0.64 2.70
0.25 2.47 0.0 100.0 0.70 14.30 45.51 490.91 0.17 3.94
0.16 1.71 0.0 100.0 0.61 9.53 28.95 438.40 0.16 2.60
0.37 0.255 0.2620 0.0 5.06 1.790 0.2518 222.86 54.390 0.7490 1.59 0.526 0.2656
PSE = pooled SE.
to stocking in the pond. The demonstration pond was fertilized (Table 1) several weeks prior to stocking crabs with muriate of potash and K-Mag at application rates typically used by west Alabama inland shrimp farmers (Roy et al. 2010). Crabs were fed a variety of different feeds during the trial, including oating catsh feed, sinking shrimp feed, whole shrimp, whole catsh, and catsh offal from a local sh processing plant. Whole shrimp and whole catsh were fed following die-offs in the farmers shrimp and catsh ponds. Crabs were sampled monthly (collected by trapping) until October when water temperatures began to drop (Figure 1). From November to March, no crabs were trapped, presumably due to colder water temperatures. Although crabs were up to harvest size by the end of October, the crabs were not harvested in 2010 due to decreased water temperatures. Crabs were sampled again in April 2011 and then harvested in June and July by using two crab pots (18 18 18 in). The traps were baited with frozen shrimp. A catsh seine was utilized on two occasions in unsuccessful attempts to harvest the crabs before the traps were utilized. After the traps no longer successfully captured crabs, the pond was completely drained and the remaining crabs were counted.
RESULTS AND DISCUSSION After 21 d, there were no differences in survival, nal weight, or percent weight gain in crabs reared in the four different lowsalinity pond waters or the control (Table 2). Water quality parameters remained acceptable throughout the course of the experiment (Table 3). No mortalities were observed in any of
the treatments. Mean nal weight and weight gain (%) ranged from 9.53 to 14.30 g and 438.40529.42%, respectively. The lowest mean nal weights and weight gains were observed in the control water and TC S4, although the differences were not signicant. Crabs grew 2.63.93 g/week, growth rates of less than 3 g/week being observed in the control and TC S4. The sodium : potassium ratio (Na:K) in seawater is approximately 28:1 (Roy et al. 2010), whereas inland saline waters of west Alabama are typically low in potassium and have high Na:K ratios, typically greater than 100:1 (Saoud et al. 2003; McNevin et al. 2004; Roy et al. 2010). Past experiments with penaeid shrimp in west Alabama revealed that as Na:K ratios were reduced closer to 28:1, by means of potassium fertilizers, survival and growth improved (Roy et al. 2006, 2007, 2010; Roy and Davis 2010). The Na:K ratio of DO 7 was 123.39:1, which in our experience with shrimp would result in poor growth and survival. The Na:K ratio of the other treatments ranged from 33.67:111.20:1, which would be ideal for Pacic white shrimp culture in west Alabama. However, the blue crabs that were reared in DO 7 pond water actually had the highest weight gain and second-highest nal individual weight in the 21-d bioassay, albeit not statistically signicant. This suggests that blue crabs might be more tolerant of high Na:K ratios than Pacic white shrimp and that west Alabama farmers might be able to reduce their fertilizer application rates and still successfully grow blue crabs. The pond demonstration yielded a total of 1,100 crabs, for 26.3% survival at harvest with an average weight of 212.19 46.51 g (mean SD). The average carapace width at harvest was 151.7 mm and ranged from 125 to 175 mm. In a pond study
TABLE 3. Water quality parameters for a 21-d bioassay with blue crabs in static 20-gal aquaria with pond water from four different low-salinity shrimp ponds. Values represent the mean SD.
Pond Control DO 1 DO 7 TC S3 TC S4
Dissolved oxygen (mg/L) 7.34 7.30 7.32 7.35 7.31 0.07 0.04 0.02 0.05 0.05
Temperature ( C) 26.9 26.8 26.9 26.8 26.8 0.12 0.11 0.02 0.09 0.17 7.9 8.2 8.1 8.5 8.6
pH 0.21 0.12 0.10 0.04 0.02
Salinity () 4.6 5.8 5.4 2.1 2.1 0.18 0.13 0.59 0.06 0.08
Total ammonia nitrogen (mg/L) 4.13 1.92 2.29 2.04 2.17 1.78 0.35 0.42 0.64 0.24
Nitrite nitrogen (mg/L) 1.77 0.42 1.60 0.36 0.78 0.98 0.26 1.54 0.19 0.46
456
ROY ET AL. Eggleston, D. B., G. Plaia, and H. Daniels. 2009. Blue crab stock enhancement: further progress in freshwater pond rearing. Final Report to the North Carolina Blue Crab Fishery Grant Program, 07-STOK-01, Raleigh. Fielder, D. A., W. J. Bardsley, and G. L. Geoff. 2001. Survival and growth of Australian snapper, Pagrus auratus, in saline groundwater from inland New South Wales, Australia. Aquaculture 201:7390. Henry, R. P. 1988. Subcellular distribution of carbonic anhydrase activity in the gills of the blue crab, Callinectes sapidus. Journal of Experimental Zoology 245:18. Henry, R. P. 2005. Critical salinity, sensitivity, and commitment of salinitymediated carbonic anhydrase induction in the gills of two euryhaline species of decapod crustaceans. Journal of Experimental Zoology 303:4556. McNevin, A. A., C. E. Boyd, O. Silapajarn, and K. Silapajarn. 2004. Ionic supplementation of pond waters for inland culture of marine shrimp. Journal of the World Aquaculture Society 35:460467. Miranda, L. E., and A. J. Sonski. 1987. Survival of red drum ngerlings in fresh water: dissolved solids and thermal minima. Proceedings of the Annual Conference of Southeastern Association of Fish and Wildlife Agencies 39(1985):228237. Parsons, T. R., Y. Maita, and C. M. Lalli. 1985. A manual of chemical and biological methods for seawater analysis. Pergamom Press, New York. Phelps, R. P., W. H. Daniels, N. Sansing, and T. W. Brown. 2010. Production of Gulf killish in the Black Belt region of Alabama using saline groundwater. North American Journal of Aquaculture 72:219224. Roy, L. A., and D. A. Davis. 2010. Requirements for the culture of the Pacic white shrimp, Litopenaeus vannamei, in low salinity waters: water modication and nutritional strategies for improving production. Pages 6178 in L. E. Cruz-Su arez, D. Ricque-Marie, M. Tapia-Salazar, M. G. Nieto-L opez, D. A. Villareal-Cavazos, and J. Gamboa-Delgado, editors. Avances en nutrici on acu cola X. Memorias del decimo simposio internacional de nutrici on acu cola. [Advances in aquatic nutrition X. Memoirs of the 10th international symposium on aquatic nutrition]. Universidad Aut onoma de Nuevo Le on, Monterrey, Nuevo Le on, Mexico. Roy, L. A., D. A. Davis, and I. P. Saoud. 2006. Effect of lecithin and cholesterol supplemented to practical diets for Litopenaeus vannamei reared in low salinity water. Aquaculture 257:446452. Roy, L. A., D. A. Davis, I. P. Saoud, C. A. Boyd, H. J. Pine, C. E. Boyd. 2010. Shrimp culture in inland low salinity waters. Reviews in Aquaculture 2:191208. Roy, L. A., D. A. Davis, I. P. Saoud, and R. P. Henry. 2007. Effects of varying levels of aqueous potassium and magnesium on survival, growth, and respiration of the Pacic white shrimp, Litopenaeus vannamei, reared in low salinity waters. Aquaculture 262:461469. Saoud, I. P., D. A. Davis, and D. B. Rouse. 2003. Suitability studies of inland well waters for Litopenaeus vannamei culture. Aquaculture 217:373383. Steel, R. G. D., and J. H. Torrie. 1980. Principles and procedures of statistics: a biometrical approach. McGraw-Hill, New York.
conducted in North Carolina, Eggleston et al. (2009) reported survivals of blue crab ranging from 2.9% to 18.8%. Eggleston et al. (2009) also reported that hatchery-reared crabs reached market size in 77 d when reared in ponds ranging in salinity from 0.4 to 0.5. Although the crabs in our study were not harvested before winter as low water temperatures made harvest impractical, the crabs reached market size by the end of 3 months. Taken together, the 21-d bioassay and pond demonstration suggest that blue crabs can grow well in ILSW of west Alabama. Techniques utilized by the inland shrimp industry to raise levels of potassium and magnesium in pond waters appear to also be effective for the culture of blue crabs. In addition to aquaculture potential and given the blue crabs ability to effectively acclimate to low and high salinities, inland saline waters of west Alabama might also be valuable as a nursery for blue crab restoration efforts. ACKNOWLEDGMENTS The authors would like to extend their thanks to those who have taken the time to critically review this manuscript as well as those who helped in supporting this research. This research was supported by the AFFC and Odom Farms, Eutaw, Alabama. Mention of a trademark or proprietary product does not constitute an endorsement of the product by Auburn University and does not imply its approval to the exclusion of other products that may also be suitable. REFERENCES
APHA (American Public Health Association), American Water Works Association, and Water Pollution Control Association. 1989. Standard Methods for the examination of water and waste water, 17th edition. APHA, Washington, D.C. Boyd, C. A., P. L. Chaney, C. E. Boyd, and D. B. Rouse. 2009. Distribution of ground water suitable for use in saline-water aquaculture in central and west-central Alabama. Journal of Applied Aquaculture 21:228240. Davis, D. A., T. M. Samocha, and C. E. Boyd. 2004. Acclimating Pacic white shrimp, Litopenaeus vannamei, to inland, low salinity waters. Southern Regional Aquaculture Center, Publication 2601, Stoneville, Mississippi Eaton A. D., L. S. Clesceri, E. W. Rice, and A. E. Greenberg. 2005. Standard methods for the examination of water and waste water, 21st edition. American Public Health Association, Washington, D.C.

Effects of Diets with 28% or 32% Protein and Meat and Bone Meal or Corn Gluten Feed on Performance of Golden Shiners in Pools
R. T. Lochmann & H. Phillips
a a a
Aquaculture/Fisheries Center of Excellence, The University of Arkansas at Pine Bluff, 1200 North University Drive, Mail Slot 4912, Pine Bluff, Arkansas, 71601, USA Version of record first published: 04 Sep 2012.
To cite this article: R. T. Lochmann & H. Phillips (2012): Effects of Diets with 28% or 32% Protein and Meat and Bone Meal or Corn Gluten Feed on Performance of Golden Shiners in Pools, North American Journal of Aquaculture, 74:4, 457-462 To link to this article: http://dx.doi.org/10.1080/15222055.2012.676009
ARTICLE
Effects of Diets with 28% or 32% Protein and Meat and Bone Meal or Corn Gluten Feed on Performance of Golden Shiners in Pools
R. T. Lochmann* and H. Phillips
Aquaculture/Fisheries Center of Excellence, The University of Arkansas at Pine Bluff, 1200 North University Drive, Mail Slot 4912, Pine Bluff, Arkansas 71601, USA
Abstract
Alternative diets made without marine proteins and with more plant ingredients are being used in channel catsh Ictalurus punctatus with the goal of maintaining protability. The potential to use similar diets for baitsh may be greater because they consume natural foods throughout production. We conducted a feeding trial in outdoor pools with golden shiners Notemigonus crysoleucas using four practical diets with 28% or 32% protein in formulas with porcine meat and bone meal (MBM) or corn gluten feed (CGF) and no animal protein. Groups of 200 sh with a total initial weight of 37.7 0.9 g (mean SE) were stocked into each of four 4.1-m3static pools per treatment and fed daily to apparent satiation for 8 weeks. Diet effects were assessed by measuring growth, survival, feed conversion, condition index, and body composition. Chlorophyll a and zooplankton were sampled to gauge natural productivity. Individual weight and total length of golden shiners were greater in sh fed diets with 32% protein than in those fed diets with 28% protein and in sh fed diets with CGF than in those fed diets with MBM. However, the differences were not commercially relevant as all sh would grade into the small crappie minnow category. Mean individual weight gain (based on group initial and nal weights), feed intake, feed conversion, and survival were similar in golden shiners fed diets with 28% or 32% protein and MBM or CGF. However, golden shiners fed diets with MBM had more body fat, higher Fultons K index, and higher relative weight than those fed diets with CGF. These traits may indicate greater robustness, which is more important in baitsh than rapid growth. The 28%-protein diet with MBM was the least expensive option that increased body fat and condition of golden shiners. Natural productivity was positively correlated with sh growth, enhancing the potential to use less expensive diets in golden shiners in outdoor systems.
The production of golden shiner Notemigonus crysoleucas for bait occurs primarily in Arkansas, and the total farm gate value of baitsh sold in the USA in 2007 was approximately US$21.5 million (NASS 2009). Baitsh production has declined slightly since 2002 but not as dramatically as food sh such as channel catsh Ictalurus punctatus. In contrast to food sh, diet cost is a lower percentage of variable costs in baitsh production. Nevertheless, the increased cost of diets over the past few years has reduced the protability of baitsh production. Attempts to cope with higher diet costs by reducing the feeding rate or frequency also reduces yield and prots (Pounds et al. 1992). Therefore, other solutions are needed to restore the protability of the baitsh industry. Alternative diets with mostly or only
*Corresponding author: rlochmann@uaex.edu Received October 21, 2011; accepted February 14, 2012
plant ingredients have been tested recently in channel catsh (Robinson and Li 2008, Li et al. 2010, 2011) with the goal of maintaining protability while using lower-cost diets. The potential to use similar diets for baitsh may be higher because they consume natural foods throughout production (Lochmann and Phillips 1996). One way to reduce diet cost while maintaining nutritional quality is to use animal protein sources other than marine sh meal. Porcine meat, bone, and blood meal (MBM) is a byproduct of the rendering industry that has a good balance of most essential amino acids (NRC 2011), is available in large quantities year-round, and costs less than half the price of sh meal. Meat and bone meal has been used successfully in diets for
457
458
LOCHMANN AND PHILLIPS
channel catsh (Hedrick et al. 2005), tilapia Oreochromis spp. (Nguyen et al. 2009), and gibel carp Carassius auratus gibelio (Yang et al. 2004; Hu et al. 2008). However, no studies have specically tested MBM as a diet ingredient for golden shiners. Plant protein sources often cost less than animal protein sources. However, plant ingredients may also contain more antinutritional factors (trypsin-inhibitor, phytic acid, ber, etc.) than animal products, and nutrients in plants may be less bioavailable than those in animal ingredients (Francis et al. 2001). Many plant feedstuffs also have inferior amino acid proles relative to animal feedstuffs (NRC 2011). Ingredients with incomplete or imbalanced amino acid contents are likely to reduce growth and feed conversion efciency and can compromise sh health (Glencross et al. 2007). However, it is possible to formulate all-plant diets with a mixture of ingredients that minimize the potential negative effects of plant ingredients while decreasing diet cost and maintaining sh performance (Wu et al. 1999; Li et al. 2003; Lochmann et al. 2004; Sink et al. 2010). Corn gluten feed (CGF) is one of the coproducts of corn that remains after starch and other components are removed for ethanol or corn syrup production (Schroeder 2010). Traditionally, CGF has been used in livestock and pet foods, but supplies are increasing due to the growth of the biofuels industry. Corn gluten feed has been tested as a feed ingredient for Nile tilapia Oreochromis niloticus (Wu et al. 1995) and channel catsh (Robinson et al. 2001) with positive results, but CGF has not been tested in baitsh. We conducted a feeding trial in outdoor pools with golden shiners using four practical diets with 28% or 32% protein in formulas with MBM or no animal protein and CGF. In addition to different ingredients, two different total dietary protein levels were also tested. Most commercially available diets for baitsh contain 28% or 32% protein, and performance of golden shiners is similar over this range of protein (Lochmann and Phillips 2009). However, diets with 32% protein generally cost more than those with 28% protein. Diet effects were assessed by measuring growth, survival, feed conversion, condition index, and body composition.
METHODS Feeding trial.An 8-week feeding trial was conducted with golden shiners in plastic-lined, 4.1-m3 outdoor pools at the University of Arkansas at Pine Bluff. The pools were lled with reservoir water (4,062 L) and maintained as static systems during the trial to encourage plankton blooms. Groups of 200 sh with a total initial weight of 37.7 0.9 g (mean SE) and a calculated individual weight of 0.19 g per sh ( = 0.4 lbs/1000 size) were stocked into each of four pools per treatment. Four experimental diets were formulated (Table 1) to contain either 28% or 32% protein using different ingredients. The traditional diets contained 5% MBM, while the alternative diets contained 20% CGF. The diet with 32% protein and MBM was the control because it is most similar to traditional commercial
formulas, which contain a small amount of animal protein. We did not use a commercial diet as a control because the ingredient composition would be proprietary, limiting our ability to explain diet effects. All diets were designed to meet or exceed the nutrient requirements of channel catsh (NRC 2011), which are similar to those of golden shiners (Lochmann and Phillips 2009). Analyzed proximate composition was provided by the manufacturer (ARKAT, Dumas, Arkansas) except for crude ber, which was analyzed at the University of Arkansas at Pine Bluff according to the ANKOM lter bag technique (AOAC 2005). The gross energy of the diets was estimated from the analyzed protein, lipid, and carbohydrate content (nitrogen-free extract) of the diet (Serrano et al. 1992), because digestible energy values for individual diet ingredients are not available for golden shiners. At the time of manufacture, the diets cost $409/ton (28-MBM), $399/ton (28-CGM), $431/ton (32-MBM), and $421/ton (32-CGF). Diets were extruded as 0.625-cm oating pellets, then crumbled in a blender and screened to retain particles of 12 mm in diameter prior to feeding. The sh were fed twice on weekdays and once on weekends. Preweighed feed equivalent to 68% of body weight daily (with larger amounts used early in the study when the sh were smaller) was offered to the sh until they fed to apparent satiation, which occurred in 30 min or less. Due to the small particle size, it was not possible to recover feed that was offered but was not consumed. The remaining feed from the preweighed ration was weighed and subtracted from the total to estimate feed consumption. Subsamples of 50 sh per pool were counted and weighed every 2 weeks to track growth. Temperature and dissolved oxygen (DO) were measured in the morning and afternoon daily in each pool using a DO meter (YSI 55 DO, Yellow Springs Instruments, Yellow Springs, Ohio). Chlorophyll a content (corrected for pheophytin a) was determined in each pool at weeks 2.5 and 5. Chlorophyll a analysis followed the American Public Health Association (APHA; 2005), except that ethanol was used as a solvent (Nusch 1980). Major zooplankton groups (rotifers, copepod nauplii, and adult copepods) were also determined in each pool at week 3. Six, 1-L water samples were obtained with a tube sampler that encompassed the entire water column. The samples were concentrated by straining them through a70-m Wisconsin plankton net and then preserving the resulting concentrate in 70% isopropyl alcohol. Samples were identied and quantied using a Sedgwick-Rafter cell and a microscope. The pH (pH meter, Denver Instruments, Colorado), total ammonia nitrogen (salicylatecyanurate method), and nitrite were measured in each pool at weeks 2.5 and 5 with a Hach DR/890 colorimeter test laboratory (Hach company, Loveland, Colorado). Un-ionized ammonia was calculated from the results of the total ammonia nitrogen and pH measurements. Water quality data are summarized in Table 2, and all parameters were within acceptable limits for golden shiners (Stone et al. 1997). At the end of the trial all sh were counted and weighed in groups. One group of 50 sh per pool was retained for measurements of lengths and weights of individual sh to
ALTERNATIVE DIETS FOR GOLDEN SHINERS
459
TABLE 1. Ingredient (%, as-fed) and analyzed composition of diets with 28% or 32% protein with or without animal protein fed to golden shiners in outdoor pools for 8 weeks. Traditional diets contained porcine meat, bone, and blood meal (MBM), while alternative diets contained corn gluten feed (CGF).
Ingredienta Meat, bone, blood meal, pork (65%) Soybean meal (48%) Cottonseed meal (41%) Corn gluten feed Corn Wheat middlings Lysine-HCI Dicalcium phosphate C-free vitamin premixb Trace mineral premixb Poultry fatc Crude proteind Lipidd Ashd Fiberd Nitrogen-free extracte Moisture Energy: proteinf
a b
Diet 1 (28-MBM) 5.0 32.6 10.0 29.6 20.0 0.1 0.6 0.1 0.1 2.0 30.0 6.0 6.8 4.7 49.9 5.0 52.0
Diet 2 (28-CGF) 34.2 10.0 20.0 20.0 12.4 0.2 1.0 0.1 0.1 2.0 29.4 6.8 6.3 5.6 48.4 6.8 52.9
Diet 3 (32-MBM) 5.0 44.4 10.0 27.8 10.0 0.6 0.1 0.1 2.0 33.9 5.4 6.9 4.3 46.1 6.8 45.4
Diet 4 (32-CGF) 46.5 10.0 20.0 20.1 0.1 1.1 0.1 0.1 2.0 33.1 5.7 6.3 5.2 45.7 8.2 46.2
All diets were supplemented with 0.02% Stay-C 35 (DSM Nutritional Products, Basel, Switzerland), which provided an active vitamin C level 50mg/kg in nished diets. Met requirements for channel catsh (NRC 2011) and presumably adequate for golden shiners (Chen et al. 2003, 2004). c Sprayed on nished diets. d Analyzed composition (percent, dry basis). e Nitrogen-free extract (NFE), calculated as NFE = 100 (protein + lipid + moisture + ash + ber), is an estimate of soluble carbohydrate. f Kilojoules of gross energy per gram of protein. Estimated energy content of the diets was based on values of 16.7, 16.7, and 37.7 kJ/g for carbohydrate, protein, and lipid, respectively (Serrano et al. 1992).
TABLE 2. Water quality parameters in a feeding trial with golden shiners fed diets with 28% or 32% protein with or without animal protein in pools for 8 weeks. Diet did not have signicant effects on water quality parameters, so data (except for zooplankton) was analyzed by RM ANOVA with time as the independent variable. Zooplankton was identied and quantied only once.
Parameter Temperature (morning, C) Temperature (afternoon, C) Dissolved oxygen (morning, mg/L) Dissolved oxygen (afternoon, mg/L) Chlorophyll a (g/L) pH Ammonia (mg/L) Nitrite (mg/L) Rotifers/L Copepods/L Nauplii/L
Mean SE 28.6 0.1 32.3 0.1 6.5 0.0 9.0 0.0 33.8 8.6 0.03 0.03 434.2 16.8 5.4
P-value (time) <0.0001 (uctuated) <0.0001 (uctuated) <0.0001 (decreased) <0.0001 (increased)
7.1 0.0003 (increased) 0.1 <0.0001 (increased) 0.01 0.03 (increased) 0.01 0.02 (increased) 282.0 5.7 3.3
determine condition factor. A single pooled sample of ve individuals from each group of 50 was homogenized and used for proximate analysis. Protein was analyzed using the Kjeldahl procedure, and dry matter and ash content were determined according to standard methods (AOAC 1995). Total lipids were extracted and quantied gravimetrically (Folch et al. 1957). Statistical Analysis.The experimental design was a 2 2 factorial. Percentage data were arcsine square root transformed prior to statistical analysis. Performance data and body composition were analyzed with a two-way analysis of variance (ANOVA) in Statview (SAS Institute, Cary, North Carolina) with protein amount (28% or 32%) and protein source (CGM or MBM) as the independent variables. Water quality data were initially analyzed with a two-way ANOVA, but there were no diet effects (P > 0.05). Therefore, data (except for zooplankton abundance) were combined by sampling date and analyzed using repeated measures ANOVA in Statview. Treatment means were considered different at P < 0.05, and specic differences in means were identied using Fishers least signicant difference test. Correlations between weight gain and individual types of plankton were determined with a z-test and results were considered signicant at P < 0.05. Water quality data were also
460
LOCHMANN AND PHILLIPS
TABLE 3. Mean individual nal weight, length, weight gain, Fultons K index, relative weight, feed conversion, and survival of golden shiners fed diets with 28% or 32% protein with or without animal protein for 8 weeks. Traditional diets contained porcine meat, bone, and blood meal (MBM), while alternative diets contained corn gluten feed (CGF). Individual nal weight, length, Fultons K index, and relative weight are based on measurements of 50 individual sh per replicate (200 per treatment). Mean individual weight gain, feed conversion, and survival are means of four replicates consisting of 99200 sh each.
Diet and statistics 28-MBM 28-CGF 32-MBM 32-CGF Pooled SE P (Protein amount) P (Protein type) P (Protein amount type)
a b
Mean individual nal weight (g) 1.3 1.5 1.4 1.7 0.4
Mean individual nal length (cm) 4.9 5.2 5.1 5.4 0.3
Fultons K index (g) 1.05 1.00 1.03 0.96 0.15 Two-way ANOVA 0.16
Relative weight (Wr) 127.6 118.6 122.8 113.6 19.1
Mean individual weight gain (g)a 0.80 0.56 0.70 0.95 0.16 0.39 0.98 0.15
Feed conversionb 1.9 2.5 2.6 2.0 0.4 0.82 0.92 0.12
Survival (%) 94.0 95.1 83.6 97.4 6.0 0.51 0.24 0.31
0.001 <0.0001 0.07 (32>28) (32>28) <0.0001 <0.0001 0.006 0.001 (CGF>MBM) (CGF>MBM) (MBM>CGF) (MBM>CGF) 0.57 0.71 0.79 0.97
Weight gain was calculated as follows: Final group weight/nal number of sh initial group weight/initial number of sh. Dry feed weight/sh weight gain.
analyzed using descriptive statistics to obtain overall means and standard errors for each parameter for the entire trial. RESULTS Growth, Feed Utilization, and Inuence of Natural Productivity Mean individual nal weight and length of golden shiners were greater in sh fed diets with 32% protein than 28% protein and in sh fed diets with CGM than MBM (Table 3). Fultons K index and relative weight were similar in sh fed diets with 28% or 32% protein but greater in sh fed diets with MBM than sh fed diets with CGF (Table 3). Apparent mean SE individual feed intake was 1.4 0.1, 1.5 0.0, 1.8 0.5, and 2.0 0.7 g for diets 14, respectively, and was not different among diets. However, feed intake was probably overestimated slightly due to the difculty in observing sh consuming small, sinking particles in water with low transparency. Mean individual weight gain, feed conversion, and survival were similar in golden shiners fed diets with 28% or 32% protein and MBM or CGM (Table 3). Survival in one pool fed the 32-MBM diet was only 49.5%, which greatly increased the variability in that treatment. The mortality did not appear to be diet related, as survival in the other pools fed the 32-MBM treatment was 94%. The water quality was good, similar among pools, and did not explain the mortality. There were also no signs of disease observed in sh in the pool. Although the cause of poor survival is unknown, excluding the pool from statistical analysis did not produce any additional diet effects. There were no interactive effects between protein amount and source on general
performance criteria. Although chlorophyll a concentration and zooplankton abundance did not differ by diet, nal weight gain of golden shiners was positively correlated with the abundance of rotifers (correlation = 0.891, P < 0.0001), copepods (correlation = 0.848, P < 0.0001), and nauplii (correlation = 0.588, P = 0.02). The correlation between chlorophyll a and weight gain was weaker (0.353, p = .0468) but also signicant. Whole-body protein, lipid, dry matter, and ash were similar in sh fed diets with 28% or 32% protein (Table 4). Whole-body protein, dry matter, and ash also did not differ in sh fed diets with MBM or CGF (Table 4). Whole-body lipid was higher
TABLE 4. Mean proximate composition (%, wet) of whole golden shiners fed diets with 28% or 32% protein with or without animal protein for 8 weeks. Each mean represents four pooled samples consisting of ve individual sh each. MBM = meat and bone meal; CGF = corn gluten feed.
Diet and statistics 28-MBM 28-CGF 32-MBM 32 -CGF Pooled SE P (Protein amount) P (Protein type) P (Protein amount type)
Protein
Lipid
Dry matter 27.5 26.4 26.4 27.3 0.9 0.87 0.90 0.28
Ash 2.3 2.3 2.1 2.7 0.2 0.44 0.13 0.11
13.9 9.9 14.5 8.5 13.8 9.2 13.2 8.5 0.4 0.4 Two-way ANOVA 0.06 0.47 0.96 0.03 (MBM>CGF) 0.09 0.40
ALTERNATIVE DIETS FOR GOLDEN SHINERS
461
in sh fed diets with MBM than in those fed diets with CGF (Table 4). There were no interactive effects of protein amount and source on body composition.
DISCUSSION Weight and length data from individual sh revealed more diet differences than weight gain data obtained from initial and nal group weights of groups divided by the number in the group. Higher-protein diets often support better growth of small sh, but there are no documented benets of using 32%-protein diets over 28%-protein diets in golden shiners, especially in outdoor systems (Lochmann and Phillips 2009). Data from individual sh may reect true biological responses to diet more accurately than bulk responses to pooled data, because the latter obscure individual variability. The greater length and weight of golden shiners fed diets with CGF was unexpected, as CGF is low in lysine and other essential amino acids (Schroeder 2010). The diets with CGF also had more ber than diets with MBM, which reduces the digestible energy in the diet (Robinson et al. 2001). Digestible energy values are unknown for specic feed ingredients for golden shiners, but digestible energy is always lower than gross energy due to the loss of fecal energy from indigestible components (NRC 2011). Therefore, the slightly higher gross energy-to-protein ratios calculated for the diets with CGF in this study do not conict with other studies showing that CGF reduces the overall available energy from the diet. Goldsh Carassius auratus fed diets formulated with kaolin to dilute energy density compensated for the lower dietary energy by increasing their feed intake (Rozin and Mayer 1961). In rainbow trout Oncorhynchus mykiss, feed intake was inversely related to digestible energy content of diets with different protein concentrations and sources (Morales et al. 1994). Feed consumption by small golden shiners in pools is difcult to quantify, but there were no apparent differences in feed form, otation, or feed intake among diets in this study. There were also no differences in natural productivity in the pools among treatments, so there is no obvious explanation for the enhanced growth of golden shiners fed diets with CGF. However, most of the sh in this study would grade into the small crappie minnow category (1416 grader size), which all have the same retail price (J. Anderson, Anderson Minnow Farm, personal communication). Therefore, the differences in individual weights and lengths would have little commercial signicance. Perhaps more importantly, golden shiners fed diets with CGF also had less body fat than those fed diets with MBM. A similar effect occurred in channel catsh fed diets with CGF (Robinson et al. 2001), resulting in higher carcass yield and potentially higher product quality. As mentioned previously, this effect is likely due to the lower available energy content of the CGF diets because they contain more indigestible ber than the MBM diets. In contrast to food sh, baitsh must withstand
multiple stressors after harvest, such as repeated grading, transport, and retail display, when feed is often withheld. Golden shiners fed diets with MBM (and presumably more available energy than CGF) had more body fat, higher relative weight, and higher condition factor, which might improve their resilience postharvest compared to leaner sh. Hardiness is valued over fast growth in the baitsh industry, where sh must be maintained in good condition once they reach market size until they are sold and reach their nal destination. Aside from the prepared diets, the plankton clearly contributed to the weight gain of golden shiners. Pools with more natural productivity had larger sh than other pools of sh fed the same prepared diet, and phytoplankton (as indicated by chlorophyll a concentrations) as well as zooplankton (rotifers, copepods, and nauplii) abundance correlated positively with sh weight gain. This conrms past ndings of the large contribution of natural foods to golden shiner growth even when nutritionally complete diets are fed (Lochmann and Phillips 1996). Although natural productivity is difcult to predict and control, the additional nutrients supplied by plankton may allow greater use of less expensive alternative diet formulas in pondraised golden shiners without reduced sh performance. For instance, zooplankton contain all essential amino acids (Aragao et al. 2004; Mitra et al. 2007), which are often suboptimal in alternative protein sources. Live foods also contain substantial amounts of essential fatty acids, vitamins, and minerals. In addition, live foods can enhance digestion in young sh by contributing exogenous sources of enzymes such as proteases, lipases, and amylases (Mitra et al. 2007) and by stimulating benecial gut microora (Conceicao et al. 2010). Therefore, good performance of pond-reared baitsh using diets with marginal concentrations of essential nutrients is likely due to coconsumption of natural foods throughout the production cycle. Partial budget analysis could not be conducted in this study because there were no differences in yield among treatments. All diets supported similar weight gain, feed conversion efciency, and survival of golden shiner. However, the 28-MBM diet also increased whole-body lipid, relative weight, and condition index of golden shiner. The 28-MBM diet cost $10 per ton more than the 28-CGF diet, but the 28-MBM diet enhanced more desirable production traits in golden shiners at a lower cost than either diet with 32% protein. Additional information is needed to verify whether or not sh with more body fat and higher condition index are more resilient to postharvest events, and to assess the economic impact of these factors on the baitsh industry.
ACKNOWLEDGMENTS We thank Anderson Minnow Farm for donating sh for this study. Sam Harrison, Emily Goodwin, and Ruguang Chen provided technical assistance. We thank Nathan Stone, Peter Perschbacher, and Hugh Thomforde for reviewing this manuscript.
462 REFERENCES
LOCHMANN AND PHILLIPS Lochmann, R., N. Stone, and H. Phillips. 2004. Evaluation of 36%-protein diets with or without animal protein for rearing tank-hatched golden shiner Notemigonus crysoleucas fry in ponds. North American Journal of Aquaculture 66:271277. Mitra, G., P. K., Mukhopadhyay, and S. Ayyappan, 2007. Biochemical composition of zooplankton community grown in freshwater earthen ponds: nutritional implication in nursery rearing of sh larvae and early juveniles. Aquaculture 272:346360. Morales, A. E., G. Cardenete, M. de la Higuera, and A. Sanz. 1994. Effects of dietary protein source on growth, feed conversion and energy utilization in rainbow trout, Oncorhynchus mykiss. Aquaculture 124:117126. NASS (National Agricultural Statistics Service). 2009. Census of agriculture (2007), United States, summary and state data, volume 1, geographic area series part 51. U.S. Department of Agriculture, NASS AC-07-A-51, Washington, D.C. Nguyen, T. N., D. A. Davis, and I. P. Saoud. 2009. Evaluation of alternative protein sources to replace sh meal in practical diets for juvenile tilapia, Oreochromis spp. Journal of the World Aquaculture Society 40:113121. NRC (National Research Council). 2011. Nutrient requirements of sh and shrimp. National Academy Press, Washington, D.C. Nusch, E. 1980. Comparison of different methods for chlorophyll and phaeopigment determination. Archiv f ur Hydrobiologie Beiheft Ergebnisse der Limnologie 14:1436. Pounds, G. L., C. R. Engle, and L. W. Dorman. 1992. Economic effects of intensication of baitsh production. Journal of the World Aquaculture Society 23:6476. Robinson, E. H., and M. H. Li. 2008. Replacement of soybean meal in channel catsh, Ictalurus punctatus, diets with cottonseed meal and distillers dried grains with solubles. Journal of the World Aquaculture Society 39:521527. Robinson, E. H., M. H. Li, and B. B. Manning. 2001. Evaluation of corn gluten feed as a dietary ingredient for pond-raised channel catsh Ictulurus punctatus. Journal of the World Aquaculture Society 32:6871. Rozin, P., and J. Mayer. 1961. Regulation of food intake in the goldsh. American Journal of Physiology 201:968974. Schroeder, J. W. 2010. Corn gluten feed. Composition, storage, handling, feeding and value. North Dakota State University, Extension Publication AS-1127, Fargo. Serrano, J. A., G. R. Nematipour, and D. M. Gatlin III. 1992. Dietary protein requirement of the red drum (Sciaenops ocellatus) and relative use of dietary carbohydrate and lipid. Aquaculture 101:283291. Sink, T. D., R. Lochmann, and N. Kinsey. 2010. Growth and survival of channel catsh Ictalurus punctatus fry fed diets with 36 or 45% total protein and allplant or animal-protein sources. Journal of the World Aquaculture Society 40:124129. Stone, N., E. Park, L. Dorman, and H. Thomforde. 1997. Baitsh culture in Arkansas: golden shiners, goldsh, and fathead minnows. University of Arkansas at Pine Bluff Cooperative Extension Program, Publication MP386, Pine Bluff. Wu, Y. V., R. Rosati, D. J. Sessa, and P. Brown. 1995. Utilization of corn gluten feed by Nile tilapia. Progressive Fish-Culturist 57:305309. Wu, Y. V., K. W. Tudor, P. B. Brown, and R. R. Rosati. 1999. Substitution of plant proteins or meat and bone meal for sh meal in diets of Nile tilapia. North American Journal of Aquaculture 61:5863. Yang, Y., S. Xie, Y. Cui, W. Lei, Y. Zhu, Y. Yang, and Y. Yu. 2004. Effect of replacement of dietary sh meal by meat and bone meal and poultry byproduct meal on growth and feed utilization of gibel carp, Carassius auratus gibelio. Aquaculture Nutrition 10:289294.
AOAC (Association of Ofcial Analytical Chemists). 1995. Ofcial methods of analysis, 16th edition. AOAC International, Arlington, Virginia. AOAC (Association of Ofcial Analytical Chemists). 2005. Ofcial methods of analysis, 18th edition. AOAC International, Gaithersburg, Maryland. APHA (American Public Health Association). 2005. Standard methods for the examination of water and wastewater, 21st edition. American Public Health Association, Washington, D. C. Aragao, C., L. E. Conceicao, M. T. Dinis, and H. Fyhn. 2004. Amino acid pools of rotifers and Artemia under different conditions: nutritional implications for sh larvae. Aquaculture 234:429445. Chen, R., R. Lochmann, A. Goodwin, K. Praveen, K. Dabrowski, and K.-J. Lee. 2003. Alternative complement activity and resistance to heat stress in golden shiners (Notemigonus crysoleucas) are increased by dietary vitamin C levels in excess of requirements for prevention of deciency signs. Journal of Nutrition 133:22812286. Chen, R., R. Lochmann, A. Goodwin, K. Praveen, K. Dabrowski, and K.-J. Lee. 2004. Effects of dietary vitamins C and E on alternative complement activity, hematology, tissue composition, vitamin concentrations and response to heat stress in juvenile golden shiner (Notemigonus crysoleucas). Aquaculture 242:553569. Conceicao, L. E. C., M. Yufera, P. Makridis, S. Morais, and M. T. Dinis. 2010. Live feeds for early stages of sh rearing. Aquaculture Research 41:613640. Folch, J., M. Lees, and G. H. Sloane-Stanley. 1957. A simple method for the isolation and purication of total lipids from animal tissues. Journal of Biological Chemistry 226:497509. Francis, G., H. P. S. Makkar, and K. Becker. 2001. Antinutritional factors in plant-derived alternate sh ingredients and their effects in sh. Aquaculture 199:197227. Glencross, G. D., M. Booth, and G. L. Allan. 2007. A feed is only as good as its ingredientsa review of ingredient evaluation strategies for aquaculture feeds. Aquaculture Nutrition 13:1734. Hedrick, R. L., T. J. Popma, and D. A. Davis. 2005. Pond production and fatty acid proles of the llets of channel catsh reared on diets with different protein sources. North American Journal of Aquaculture 67:304311. Hu, M., Y. Wang, Q. Wang, M. Zhao, B. Xiong, X. Qian, Y. Zhao, and Z. Luo. 2008. Replacement of sh meal by rendered animal protein ingredients with lysine and methionine supplementation to practical diets for gibel carp, Carassius auratus gibelio. Aquaculture 275:260265. Li, M. H., E. H. Robinson, D. F. Oberle, and P. M. Lucas. 2010. Effects of various corn distillers by-products on growth, feed efciency, and body composition of channel catsh, Ictalurus punctatus. Aquaculture Nutrition 16:188193. Li, M. H., E. H. Robinson, B. G. Bosworth, D. F. Oberle, and P. M. Lucas. 2011. Use of corn gluten feed and cottonseed meal to replace soybean meal and corn in diets for pond-raised channel catsh. North American Journal of Aquaculture 73:153158. Li, M. H., D. J. Wise, B. B. Manning, and E. H. Robinson. 2003. Effect of dietary total protein and animal protein on growth and feed efciency of juvenile channel catsh Ictalurus punctatus and their response to Edwardsiella ictaluri challenge. Journal of the World Aquaculture Society 34:223228. Lochmann, R., and H. Phillips. 1996. Stable isotopic evaluation of the relative assimilation of natural and articial foods by golden shiners Notemigonus crysoleucas in ponds. Journal of the World Aquaculture Society 27: 168177. Lochmann, R., and H. Phillips. 2009. Nutrition and feeding of baitsh. University of Arkansas at Pine Bluff, Cooperative Extension Program ETB256-PD2-09RV, Pine Bluff.

Evaluation of Hydrogen Peroxide and Temperature to Control Mortality Caused by Saprolegniasis and to Increase Hatching Success of Largemouth Bass Eggs
Michael D. Matthews , Joshua C. Sakmar & Nick Trippel
a a a b
Florida Fish and Wildlife Conservation Commission, Florida Bass Conservation Center, 3583 County Road 788, Webster Florida, 33597, USA
b
Florida Fish and Wildlife Conservation Commission, Eustis Fisheries Research Laboratory, 601 West Woodward Avenue, Eustis, Florida, 32726, USA Version of record first published: 04 Sep 2012.
To cite this article: Michael D. Matthews, Joshua C. Sakmar & Nick Trippel (2012): Evaluation of Hydrogen Peroxide and Temperature to Control Mortality Caused by Saprolegniasis and to Increase Hatching Success of Largemouth Bass Eggs, North American Journal of Aquaculture, 74:4, 463-467 To link to this article: http://dx.doi.org/10.1080/15222055.2012.676608
ARTICLE
Evaluation of Hydrogen Peroxide and Temperature to Control Mortality Caused by Saprolegniasis and to Increase Hatching Success of Largemouth Bass Eggs
Michael D. Matthews* and Joshua C. Sakmar
Florida Fish and Wildlife Conservation Commission, Florida Bass Conservation Center, 3583 County Road 788, Webster Florida 33597, USA
Nick Trippel
Florida Fish and Wildlife Conservation Commission, Eustis Fisheries Research Laboratory, 601 West Woodward Avenue, Eustis, Florida 32726, USA
Abstract
We evaluated the use of hydrogen peroxide (H2 O2 ), temperature, or both, to control mortality presumptively caused by fungus Saprolegnia in Florida largemouth bass Micropterus salmoides oridanus eggs in a ow-through hatchery system. Four treatmentsambient water (AW; control), ambient water and H2 O2 (AWHP), heated water (HW), and both heated water and H2 O2 (HWHP)were tested on each of 30 replicate spawns. Four 8 cm 5 cm sections of spawning mat, one for each treatment, were cut from each of the 30 selected spawns. Egg counts on each cut mat section were recorded. Water temperature in all AW and HW trial tanks ranged from 17.9 C to 19.2 C and from 22.1 C to 23.6 C, respectively, during treatment. The water temperature difference between treatments averaged 4.4 C. Hydrogen peroxide trial tanks received two, 100 mg/L H2 O2 treatments 8 h apart, until hatch. The 7.5 L/min incoming water ow was not reduced during treatment. The addition of HW, H2 O2 , or both, signicantly increased mean percent hatch (79, 79, and 91%, respectively) over that of the untreated controls (49%). Signicant differences were found in mean percent hatch levels between AW and the other three treatment groups (P < 0.05). The combination of higher incubation temperature and H2 O2 administration also resulted in signicantly greater hatch than did either higher incubation temperature or increased H2 O2 administration alone; percent hatch did not differ signicantly between eggs that either were only incubated at the higher temperature or received only H2 O2 administration. Increasing incubation water temperature to 2223 C instead of using only water at ambient temperature (1819 C) or adding 100 mg/L H2 O2 twice daily in a ow-through system signicantly increased hatching percentage of Florida largemouth bass eggs.
Fungi from the family Saprolegniaceae commonly infest sh and sh eggs at temperatures below 25.5 C (Tucker and Robinson 1990). Eight genera reportedly cause infestions, but only Saprolegnia sp., Achlya sp., and Aphanomyces sp. are of concern in freshwater aquaculture (Woo and Bruno 1999). Outbreaks of Saprolegnia regularly cause chronic losses in sh populations and can cause rapid increases in egg mortality. Koeypudsa et al. (2005) reported that temperatures suitable for
growth of Saprolegnia ranged from 5 C to 30 C, but they observed no growth at temperatures above 30 C. Marking et al. (1994) tested 21 different antifungal chemicals on rainbow trout Oncorhynchus mykiss eggs and determined that salt (sodium chloride), formalin, and H2 O2 were the most suitable and effective at controlling fungal infestations. Rach et al. (2005) showed that H2 O2 concentrations of 1,000 mg/L and formalin concentrations of 1,667 mg/L effectively increased
*Corresponding author: michael.matthews@myfwc.com Received May 3, 2011; accepted March 11, 2012
463
464
MATTHEWS ET AL.
survival of lake trout Salvelinus namaycush eggs up to the eyed egg stage. Egg hatch increased in walleye Sander vitreus, white sucker Catostomus commersonii, and paddlesh Polyodon spathula when treated with 283, 565, and 1,130 mg/L H2 O2 for 15 min every other day (Gaikowski et al. 2003). Barnes and Gaikowski (2004) found that a daily 15-min treatment with 1,000 mg/L H2 O2 was suitable for use on Chinook salmon O. tshawytscha eggs. Saprolegnia control with H2 O2 has been proven equally effective in warmwater species. In three trials conducted by Small and Wolters (2003), povidoneiodine treatment at 100 mg/L, followed by 15 min of treatment with H2 O2 at 250 mg/L, increased the hatching success of channel catsh Ictalurus punctatus by 26%. Hatching success of channel catsh eggs treated for 15 min with 250 mg/L H2 O2 was 30% more than for povidone iodine-treated eggs and formalin-treated eggs. Treatment of egg masses with 70 mg/L hydrogen peroxide in a ow-through trough increased survival by 48% over nontreated controls. Mitchell et al. (2009) increased channel catsh hatch rates by using 125, 250, and 500 mg/L H2 O2 , but cautioned against using concentrations above 500 mg/L in hatching troughs. According to Mitchell et al. (2010), H2 O2 was as effective as copper sulfate pentahydrate, diquat bromide, and formalin at limiting fungal outbreaks on catsh eggs. Use of hydrogen peroxide to control fungi due to Saprolegniaceae has been approved for use on warmwater nsh eggs at 7501,000 mg/L for 15 min/d in a continuous-ow system on consecutive or alternate days until hatch (USFDA 2007). Despite good raceway sanitation practices and ozone-treated recycled water supply, fungal outbreaks still caused high mortality of bass eggs in incubation raceways at the Florida Bass Conservation Commission (FBCC). Van West (2006) reported that ozone was effective in fungal control in water supplies, but could not be used to treat affected sh. Saprolegnia-induced mortality on Florida largemouth bass Micropterus salmoides oridanus eggs has accounted for unacceptable losses in total egg production in years past at the FBCC. Hydrogen peroxide treatments of 750 and 1,000 mg/L controlled fungus on eggs at the FBCC, but both concentrations caused moderate to high levels of mortality on newly hatched fry. Rach et al. (1998) noted that high concentrations of H2 O2 (3 g/L or higher) reduced hatching success of warmwater species. The FBCC produces all of the Phase I (total length >25 mm but <75 mm) Florida largemouth bass in Florida. Wild-caught broodsh are spawned indoors in 25-m-long concrete raceways on Spawntex mats as substrate. Spring spawning of largemouth bass occurred naturally from February to March at 1822 C. Fall largemouth bass spawned at water temperatures between 19 C and 24 C in late September through October after temperature and photoperiod manipulation. High levels of fungal-induced egg mortality documented at the FBCC extended spawning seasons longer than needed. Rather than testing different concentrations of H2 O2 as reviewed in past studies, we tested elevated temperature, H2 O2 , or both combined to reduce egg loss caused by fungus. Increasing temperature may be a strategy
to reduce fungal-related mortality by decreasing the duration of egg incubation. McCormick (1978) demonstrated that increasing temperature on white bass Morone chrysops eggs decreased hatching time. Hardy (1978) reported similar results for largemouth bass, nding hatch at 10 C after 1321 d; at 17.7 C, hatch occurred in 34 d, and eggs kept at 22.2 C hatched in 2 d. We found no studies that used the temperature and H2 O2 combination against fungal growth. Our objective was to nd an effective yet practical way in a large-scale indoor production setting to reduce largemouth bass egg loss to fungal growth, using temperature manipulation, H2 O2 , or a combination of both. METHODS During the 2010 spring spawning season at the FBCC, an experiment was designed to test fungal control using elevated temperature and H2 O2 on largemouth bass eggs. Three 25-mlong raceways were stocked with 20 pairs of male and female Florida largemouth bass. Each raceway received 20 Spawntex mats for spawning substrate. Spawns were collected each morning with mats immediately replaced. A total of 30 spawns were randomly selected from the 109 spawns collected during the 20-d spawning period. The four treatments were heated water (HW), H2 O2 and ambient water (AWHP), heated water and H2 O2 (HWHP), and ambient water (AW), which was used for the control. Four 8 cm 5 cm sections, one for each treatment, were cut from each of the 30 randomly selected mats. Each mat section needed to include approximately 100 eggs. Eggs were visually counted, noting the initial number of live and dead eggs. All spawning mats were collected less than 6 h after completed spawning. Each of four 378-L trial tanks, 1.8 m 0.5 m 0.5 m in size, was assigned one of the four treatments. Each trial tank was separated into three sections with screens, thus allowing for three simultaneous replicates per tank. This setup was repeated 10 times over the 20-d spawning period to achieve 30 test replicates per treatment. To avoid temperature shock, the eggs on all mat sections were counted and moved at ambient water temperatures. Tanks assigned to the AW treatment received no antifungal agents or any other changes from the conditioned 1819 C well water supply and were used as the control. Heated water and HWHP tanks were maintained at 2223 C. Boilers supplied continuous heated water for this experiment. Eggs (<6 h old) were tempered for 2 h to the increased temperature. Hydrogen peroxide (35% hydrogen peroxide, w/w; Eka Chemicals Inc., Marietta, Georgia, trade name 35% PEROX-AID, specic gravity 1.1317) was added to the inlet of the tank twice daily at 100 mg/L to tanks labeled AWHP and HWHP at 0800 hours and again at 1600 hours until hatch. Calculation for a 100 mg/L concentration of 35% H2 O2 was as follows (see manufacturers label): [(target conc., mg/L)/396,100 mg/L PEROX-AID)] [(tank vol, L) (1,000 mL/L)] = mL of H2 O2
CONTROLLING SAPROLEGNIASIS ON LARGEMOUTH BASS EGGS
465
Mat sections were suspended by hanging each section from a wooden dowel in the respective trial tanks. All test tanks had a constant 7.5 L/min ow rate and received conditioned well water from one source. Constant aeration was applied to each tank for homogeneous mixing of tank water and H2 O2 if applicable. Dose verication was recorded in ve heated and ve ambient tanks that received H2 O2 treatments using the following equation (USFWS AADAP 2008): H2 O2 , mg/L = [(a M 85.025)/V ] 1,000 where a = mL of potassium permanganate (KMnO4 ), M = 0.02 molar concentration of KMnO4 , 85.025 is a constant, V = volume of sample in mL, and 1,000 is a constant. Concentrations of H2 O2 between 85 mg/L and 115 mg/L were considered acceptable. Water samples were taken to check for residual H2 O2 before the second dose of H2 O2 was added to these same 10 tanks. Water temperatures were recorded daily in each tank until hatch. Daily dissolved oxygen and pH levels were monitored on ve HW and ve AW treatment tanks. Temperature and dissolved oxygen were measured with a YSI model 550A (Yellow Springs, Ohio), and pH was measured with an Oakton pHTestr 30 (Vernon Hill, Illinois). All dissolved oxygen readings were taken before H2 O2 administration if applicable. Total ammonia nitrogen, total alkalinity, and total hardness were monitored daily in the source water. We used the salicylate method to determine total ammonia nitrogen (Hach, Loveland, Colorado); total alkalinity and total hardness were measured by titration using the LaMotte AQ-2 test kit (Chestertown, Maryland). Fry were collected after hatch with a siphon and counted. Percent hatch was calculated from the number of fry collected divided by the number of initial eggs (minus dead eggs). Fungal coverage on each mat section was ranked on a 010 scale, 0 indicating no fungal growth and 10 denoting complete mat coverage. All evaluations were completed by the same person throughout the entire study. This score was compared to percent hatch to view graphically how fungal growth affects hatch rate. A natural logarithmic transformation was used to meet the assumptions of equal variances. A generalized linear mixed model (SAS PROC GLIMMIX) was investigated to determine whether the independent variable treatment assignment explained the variation in the probability of dependent variable percent hatch. The adjusted least-squares means (Scheff es adjustment) were calculated to identify differences in the probability of hatch between treatments; differences were declared statistically signicant if P < 0.05. RESULTS The H2 O2 concentrations determined during dose verication ranged from 86 to 96 mg/L. Hydrogen peroxide was not detected in water samples taken immediately before administration of the second daily H2 O2 dose. Oxygen levels never fell below 8 mg/L, and pH ranged from 7.7 to 8.1. Total
ammonia nitrogen remained negligible throughout the study, and total hardness and alkalinity consistently remained at 340 and 360 mg/L as CaCO3 , respectively. The mean SE numbers of largemouth bass eggs per mat section were 94 6.5, 91 5.8, 86 5.2, and 87 5.4 for AW, AWHP, HW, and HWHP treatments, respectively. The mean number of dead eggs per mat ranged from 2 to 4 on day zero for all treatment groups. The initial egg viability mean SE were 97 0.7, 97 0.6, 96 1.2, and 96 0.8%, for AW, AWHP, HW, and HWHP treatments, respectively. Treatment explained a signicant amount of the variation in egg hatch, Type III test of xed effects (GLIMMIX: F = 52.70, df = 4, P < 0.0001). All HW- and HWHP-treated eggs hatched in 4048 h while all AW- and AWHP-treated eggs hatched in 84 96 h. Mean percent hatches in the AWHP, HW, and HWHP treatment groups (79, 79, and 91%) were signicantly greater than the 49% in the AW group (Figure 1). The addition of HW, H2 O2 , or both, decreased fungus levels observed on hatching mats relative to AW controls. The mean SE of observed fungus levels for each treatment were 6.5 0.4, 3.5 0.3, 2.9 0.4, and 1.1 0.3 for AW, AWHP, HW, and HWHP, respectively (Figure 1). Signicant differences in percent hatch were identied between all treatment comparisons (P < 0.05) except for HW versus AWHP. The odds ratios determined to describe the probability of egg hatch indicated that eggs treated with HP or HW were four times more likely to hatch than the untreated (AW) eggs and eight times more likely to hatch when HP and HW were used in combination. The combination treatment (HWHP) signicantly increased the probability of egg hatch compared with that of eggs administered HW or HP alone. the eggs assigned to the HWHP being about twice as likely to hatch as the single-treatment eggs.
DISCUSSION This experiment focused on Florida largemouth bass eggs treated with 2223 C heated water, 100 mg/L H2 O2 , or both. In lieu of evaluating different H2 O2 concentrations, we extended treatment duration and number of daily H2 O2 treatments (two) at a concentration 7.510 times less than that approved for use in warm water. The 7.5 L/min continuous ow rate created a turnover of the complete 378-L tank volume in 50 min. The increased time of the ever-decreasing H2 O2 concentration decreased fungal-induced mortality at ambient temperatures. Published data on largemouth bass eggs treated with H2 O2 in ow-through systems are sparse, possibly because bass spawning generally occurs in outdoor ponds, not indoors in raceways. We compared our results with a published study on warmwater channel catsh eggs that used one daily H2 O2 treatment added at the tank inlet (trial 1: H2 O2 additions of 0, 100, 200, and 300 mg/L; trial 2: H2 O2 additions of 0, 35, 70, and 100 mg/L) at a continuous ow rate similar to our study. The results showed that the use of 100 mg/L and 70 mg/L H2 O2 in the rst and second trial signicantly improved hatching success compared
466
MATTHEWS ET AL.
FIGURE 1. Mean (+ SE, n = 30) hatching success (%) versus fungus level (scored 0 = no growth to 10 = total fungal coverage) for largemouth bass eggs treated twice daily until hatch with 100 mg/L hydrogen peroxide at a constant 7.5 L/min ow rate with continuous aeration. All treatments were added at the inlet of the tanks. Different letters (a, b, and c) denote a signicant difference between the two treatments.
with no treatment (0 mg/L H2 O2 ), the 70 mg/L H2 O2 treatment yielding the higher success between the two trials (Small and Wolters 2003). Our results similarly showed improved hatching success with the use of lower concentrations of H2 O2 . Mitchell et al. (2009) also found that H2 O2, applied at 125, 250, and 500 mg/L for 6 d, signicantly reduced fungal growth on channel catsh eggs under continuous ow rates of two exchanges per hour. The 100 mg/L concentration increased hatching success, and preliminary observations provided evidence that mortality of newly hatched bass fry did not increase. Further investigation is required to determine lethal concentrations of H2 O2 on newly hatched bass fry to verify our observations. However, target Animal Safety Study CAP-97-00048-08 reported zero mean percent mortality of largemouth bass fry after 60- and 180-min treatments with 121 and 103 mg/L H2 O2 (USFDA 2007). Small and Wolters (2003) noticed that using H2 O2 concentrations of 500 mg/L on channel catsh eggs in a 15-min bath caused poor hatching success; 200 and 300 mg/L concentrations caused premature hatching in a ow-through system. We found no evidence of premature hatching of bass eggs at 100 mg/L when we compared the hatch times of AW to AWHP and of HW to HWHP. Heated water without H2 O2 was also effective at decreasing fungal mortality by reducing duration until hatch. The duration of largemouth bass egg incubation is reduced by nearly 2 d by increasing water temperatures from 17.7 C to 22.2 C (Hardy 1978). We recorded similar results in our study; eggs in all tanks receiving HW or HWHP treatments hatched 3648 h sooner than those receiving AW and AWHP treatments. Kelly (1968) stated
that temperatures between 12.7 C and 23.9 C did not directly cause mortality in largemouth bass eggs and reported generally poor egg survival of nonacclimated eggs at temperatures of 10, 26.7, and 29.4 C. The poor survival was hypothesized to be due to critical periods in early embryo development when the embryo may be more sensitive to rapid temperature changes. Acclimation was cited as another reason for poor egg survival, but eggs not acclimated to temperatures between 12.7 C and 23.9 C showed no difference in survival compared with acclimated eggs (Kelly 1968). White bass eggs incubated at 26 C hatched in 24 h, but incubation at 14 C took 108 h (McCormick 1978). Implementing heated water in incubation tanks can have the added value of maintaining a stabilized water temperature, thus adding the benet of consistency in incubation duration. The negative aspect of heating the incubation water is that if the broodstock are spawned at a water temperature different from the incubation temperature, the eggs may need to be tempered to incubation temperature, which can be very time-consuming. Past experience found the lack of tempering can lead to high egg mortality, in contrast to Kellys (1968) ndings. Heated water decreased incubation duration and bass eggs hatched in less time. The H2 O2 increased hatching percentage by visibly reduced fungal growth. Results from this study support both of these statements and show that, in combination, heated water and H2 O2 have a synergistic effect on hatching success. We did not directly compare the 100 mg/L concentration with a 750 or 1,000 mg/L H2 O2 concentration and are not claiming that the lower dose improves hatching success over that at the currently approved H2 O2 concentrations. Our results show that the 100 mg/L H2 O2 dose signicantly improves hatch rate over
CONTROLLING SAPROLEGNIASIS ON LARGEMOUTH BASS EGGS
467
no treatment. The proposed benet of the lower H2 O2 concentration allows spawnings from multiple days to be treated in the same tank without the increased fry mortality noted at the approved dosage rate (USFDA 2007). From a hatchery management perspective, raceway space limitations mandate that multiple day collections of spawns be placed together. Adding approved concentrations of H2 O2 (7501,000 mg/L, continuous water ow) to the raceways decreased egg mortality on mats hung on days 1 and 2, but caused unacceptable fry losses from eggs collected on days 1 and 2 when mats from a third day were added and treated accordingly. Due to observed increases in fry mortality, 7501,000 mg/L H2 O2 concentrations limited use to prehatch conditions only. Without treatment, the mats from day 3 had high levels of fungal-induced egg mortality. Hydrogen peroxide concentrations of 100 mg/L are effective at reducing fungal growth and have the added benet of not causing mortality in the newly hatched fry. In fall 2010 we tested the 100 mg/L H2 O2 concentration administered twice daily in 4,920-L tanks stocked with three consecutive days of spawns (up to 40 spawns/raceway). This resulted in zero spawns lost to fungus from 205 spawns treated. No increased fry mortality was observed. Water temperatures averaged 23 C. This study revealed some options for production managers faced with site-specic restrictions. Heating water or the use of chemicals may not be practical or allowed at some facilities, but implementing just one of the three treatments signicantly increased hatch rates. Another benet from this treatment method resulted from not needing to cut water ow to the treatment tank. This completely removed the risk of egg loss to lower water levels and to anoxic conditions that occur when using other chemical treatments when water ow is not promptly restored. Higher egg survival equates to requiring fewer spawns, less broodstock, less hatchery space, and less time spent on bass production. We recommend that approval status for lower levels of H2 O2 be investigated for fungus control on warmwater eggs. Potentially this could be completed with some additional efcacy and with target animal safety studies that show improved egg hatch and fry survival rates at reduced H2 O2 concentrations and increased treatment duration. This should also provide the data to justify that the proposed two treatments at 100 mg/L would have less chemical discharge and be no less safe than one treatment at 1,000 mg/L. Potential benets from this will be lower amounts of H2 O2 needed, lower risk of fry mortality, and lower concentrations of chemical in hatchery efuent. ACKNOWLEDGMENTS We thank Rick Stout and the Florida Bass Conservation Center staff for their assistance with mat collection. We also thank Bill Pouder and Jon Fury for their comments on this manuscript. We also acknowledge Dr. Jesse Trushenski for her insightful
review of this manuscript. Finally, we thank Mark Gaikowski for his help with data analysis. REFERENCES
Barnes, M. E., and M. P. Gaikowski. 2004. Use of hydrogen peroxide during incubation of landlocked fall Chinook salmon eggs in vertical-ow incubators. North American Journal of Aquaculture 66:2934. Gaikowski, M. P., J. J. Rach, M. Drobish, J. Hamilton, T. Harder, L. A. Lee, C. Moen, and A. Moore. 2003. Efcacy of hydrogen peroxide in controlling mortality associated with saprolegniasis on walleye, white sucker, and paddlesh eggs. North American Journal of Aquaculture 65:349355. Hardy, J. D. Jr. 1978. Aphredoderidae through Rachycentridae. U.S. Fish and Wildlife Service FWS/OBS-78/12:volume 3. Kelly, J. W. 1968. Effects of incubation temperature on survival of largemouth bass eggs. Progressive Fish-Culturist 30:159163. Koeypudsa, W., P. Phadee, J. Tangtrongpiros, and K. Hatai. 2005. Inuence of pH, temperature, and sodium chloride concentration on growth rate of Saprolegnia sp. Journal of Scientic Research at Chulalongkorn University 30:123130. Marking, L. L., J. J. Rach, and T. M. Schreier. 1994. Evaluation of antifungal agents for sh culture. Progressive Fish-Culturist 56:225231. McCormick, J. H. 1978. Effects of temperature on hatching success and survival in the white bass. Progerssive Fish-Culturist 40:133137. Mitchell, A. J., A. A. Radomski, D. L. Straus, and R. Carter. 2009. The effect of hydrogen peroxide on the hatch rate and Saprolegnia sp. infestation of channel catsh eggs. North American Journal of Aquaculture 71: 276280. Mitchell, A. J., D. L. Straus, B. Farmer, and R. Carter. 2010. Comparison of percent hatch and fungal infestation in channel catsh eggs after copper sulde, diquat bromide, formalin, and hydrogen peroxide treatment. North American Journal of Aquaculture 72:201206. Rach, J. J., M. P. Gaikowski, G. E. Howe, and T. M. Schreier. 1998. Evaluation of the toxicity and efcacy of hydrogen peroxide treatments on eggs of warmand coolwater shes. Aquaculture 165:1125. Rach, J. J., S. Redman, D. Bast, and M. P. Gaikowski. 2005. Efcacy of hydrogen peroxide versus formalin treatments to control mortality associated with saprolegniasis on lake trout eggs. North American Journal of Aquaculture 67:148154. Small, B. C., and W. R. Wolters. 2003. Hydrogen peroxide treatment during egg incubation improves channel catsh hatching success. North American Journal of Aquaculture 65:314317. Tucker, C. S., and E. H. Robinson. 1990. Channel catsh farming handbook. Van Nostrand Reinhold, New York. USFDA (U.S. Food and Drug Administration). 2007. Freedom of information summary: original new animal dug application. NADA 141255 35% PEROX-AID hydrogen peroxide liquid solution. USFDA, Washington, D.C. Available: http://www.fda.gov/downloads/AnimalVeterinary/Products/ ApprovedAnimalDrugProducts / FOIADrugSummaries / UCM051418 . pdf. (June 2012). USFWS AADAP (U.S. Fish and Wildlife Service, Aquatic Animal Drug Approval Partnership Program) 2008. Procedures to verify concentrations of hydrogen peroxide (35% PEROX-AID R ) solutions. USFWS AADAP, Standard Operating Procedure Miscellaneous 240.1, Bozeman, Montana. Available: http://www.fws.gov/sheries/aadap/National%20Aquaculture%20Drug%20 Research%20Forum/SOPs/SOP%20240-1%20analysis%20of%20H2O2.pdf. (June 2012). Van West, P. 2006. Saprolegnia parasitica, an oomycete pathogen with a shy appetite: new challenges for an old problem. Mycologist 20:99104. Woo, P. T. K., and D. W. Bruno. 1999. Fish diseases and disorders, volume 3. Viral, bacterial, and fungal infections. CAB International, New York.

New Programmable Electrofishing Device for Use in Aquaculture

Fulgencio Soto , Manuel Jimnez , Antonio Mateo , Jos A. Villarejo , Esther de Jdar & Jacinto Jimnez
a a a a a a a
Departamento de Tecnologa Electrnica, Universidad Politcnica de Cartagena, Campus Muralla del Mar, s/n 30202 Cartagena, Spain Version of record first published: 04 Sep 2012.
To cite this article: Fulgencio Soto, Manuel Jimnez, Antonio Mateo, Jos A. Villarejo, Esther de Jdar & Jacinto Jimnez (2012): New Programmable Electrofishing Device for Use in Aquaculture, North American Journal of Aquaculture, 74:4, 468-476 To link to this article: http://dx.doi.org/10.1080/15222055.2012.690828
ARTICLE
New Programmable Electroshing Device for Use in Aquaculture

Fulgencio Soto, Manuel Jim enez,* Antonio Mateo, Jos e A. Villarejo, Esther de J odar, and Jacinto Jim enez
Departamento de Tecnolog a Electr onica, Universidad Polit ecnica de Cartagena, Campus Muralla del Mar, s/n 30202 Cartagena, Spain
Abstract
The new equipment described here was developed in response to a shortcoming in the technology available for electroshing marine species. The system supplies a need that exists in commercial operations, and also allows research into the effect of the waveform on the quality of the meat. The equipment is capable of generating any kind of waveform, identifying the environment the electrodes are in, and logging the voltage and current values that are applied to kill the sh. Although it has been designed in the context of electroshing for tuna, it is perfectly compatible with electroshing in freshwater. The system is composed of two interacting parts: a software section and a hardware section. The software is capable of determining the relationship of each of the electrical wave parameters with the nal quality of the meat. The electrical hardware main constituent is a full bridge with an inductor-capacitor (LC) lter. The effectiveness of two different types of control for this kind of converter is compared: one being a feedback control, and the other a feedforward control. This equipment has been tested under natural conditions using three different waveforms, selected on the basis of our previous tuna electrostunning experience. These are: high-frequency pulsed DC, low-frequency pulsed DC, and low-frequency, decreasing exponential AC. Of the tested waveforms, low-frequency, decreasing exponential AC showed the best results as far as esh quality was concerned. The system showed, in all cases, a high degree of reliability and safety.
Throughout the world, various species of tuna are farmed in cages. Since 1995, the farming of red tuna Thunnus thynnus has expanded in the Mediterranean Sea (Miyake et al. 2003). The main objective is to obtain the best possible quality in order to achieve the highest possible return. The price of red tuna meat can vary between US$80 and $280 per kilogram depending on its quality, and this is strongly inuenced by the slaughtering method. Usually, farmed tuna have been slaughtered by shooting them underwater using a power head (Lupara), shooting to the head from above the water, coring or spiking, or a blow on the head with a club (Kavatic et al. 2003; EFSA 2009). Those techniques cause a high degree of stress to the tuna, leading to a loss in the quality of the meat (De la G andara 2003; Gregory 2008). To avoid these negative effects, consideration has been given to the
use of electrical discharge for slaughter (Van de Vis et al. 2003; Gregory 2008). The use of electricity for the slaughter of freshwater sh species has been known for nearly 100 years since the rst ofcial record in 1917, the year in which the rst electroshing system was patented. The basic method followed in freshwater electroshing is shown in Figure 1. The electrodes of the source of power are introduced into the water, a voltage is applied, and the electrical eld generated stuns or kills the sh. This method cannot be used in seawater with tuna or other marine species because of the high level of conductivity of the water. One cubic meter of freshwater is equivalent to an electrical resistance of 100 , whereas 1 m3 of seawater is equivalent to an electrical resistance of only 0.2 . Under such conditions, to generate a potential gradient capable of stunning
*Corresponding author: manuel.jimenez@upct.es Received January 11, 2012; accepted April 24, 2012
468
NEW PROGRAMMABLE ELECTROFISHING DEVICE
469
sh are very similar. The most important ones are hemorrhages and bloodspots, injuries to the nerves and internal tissues (McMichael 1993), and even more serious damage such as spinal injuries (Dalbey et al. 1996) (see Figure 2). Owing to the shortcomings in the technology available for electroshing of marine species our research group (Electronic Engineering and Systems Division) has been working for the last few years in the development of marine electroshing devices. After some preliminary experiences carried out with simple selfmade equipment (Soto et al. 2006), the aim of developing a exible and programmable electroshing device was established. METHODS
FIGURE 1. Basic eld of effect on sh of electroshing method in freshwater. [Figure available in color online.]
or slaughtering tuna in seawater would require an unreasonable amount of electric power (Kolz 1989; Lines et al. 2004). Because of this, the electricity is applied by means of a harpoon that acts as an electrode inside the sh (Garc a 2002). Many studies regarding the use of electricity in freshwater can be found in the literature (McMichael 1993; Sharber et al. 1994; Reynolds 1996; Lines 2003, 2005). However, experience with electroshing in seawater is much more limited and focused on small-sized species (Roth 2004; Lambooij 2008; Nordgreen et al. 2008). The use of electricity for slaughtering large, farmed red tuna is almost nonexistent. The causes of injury in sh induced by electroshing are similar in freshwater and seawater, with the most important factor being the waveform and frequency used. Despite the differences in method and in the size and weight of the sh (rainbow trout Oncorhynchus mykiss or salmon, in comparison with red tuna), the injuries caused by the application of the electrical discharge within the body of the
Experimental Conditions Laboratory conditions.The electroshing device was rst tested in the laboratory using the equipment described below. Various preliminary tests were performed simulating eld conditions such as water, sh or air presence at the output of the system (harpoon). The AC power supply was provided by a 220-V (root mean square), 50-Hz transformer bank connected to the AC main supplies. Two rheostats were used as load: the rst one (110 , 3 A) was used to simulate the sh, and the second one (11 , 3 A) acted as the water. The test in air was simulated with no load. Tektronix P5200 high-voltage differential probes were employed to measure voltage. Current measurements were performed using A6302 current probes connected to AM 503B ampliers, both from Tektronix. To capture waveforms, a Yokogawa DL1640 color oscilloscope was selected, which provided four 200-MS/s channels. Field conditions.The tests described were performed at a tuna farm located in the southeast of Spain (coast of Region of Murcia). Each cage of this farm was 30 m deep and 50 m long. At the beginning of the harvest time each cage contains about 1,000
FIGURE 2. Spinal injuries caused by the application of the electrical discharge in (a) rainbow trout (source: Lamarque 1990) and (b) red tuna using our equipment. [Figure available in color online.]
470
SOTO ET AL.
FIGURE 3. Example of the slaughtering method used in large tuna tanks for commercial aquaculture.
specimens. The tests were carried on from November 2008 to January 2009. At this time, the average water temperature was 16.2 C and conductivity was 55 mS/cm. The slaughtering process is as follows. Two scuba divers enter the cage (see Figure 3). One carries a harpoon gun whose
harpoon is connected by a wire to the power converter, and the other carries an underwater switch. The diver carrying the harpoon gun selects a tuna for slaughter and res at it. The diver controls the power converter and hence the timing of the discharge with the underwater switch. Although this process could be automated, the scuba divers feel safer if they have control of the process. In order to electroslaughter the tuna, a potential gradient has to be generated between the inside of its body and the water. When the harpoon penetrates the tuna (see Figure 4), the current will seek pathways of lower impedance to the water, and so the impact area will inuence the results. The best point of impact is located in the lateral body wall, similar to that shown in Figure 4. The possible routes that the current could follow inside the tuna are shown in Figure 4. Despite the importance of the point of impact (location and depth), it is very difcult to control these variables. Depth is controlled with the harpoon limiter; however, the point of impact will depend on the divers experience. After the discharge to the tuna, the sh is hoisted on deck. This is done with a crane using a rope that the scuba divers tie to the tunas tail. On deck, the tuna is bled and a metallic rod is inserted through the vertebrae to rapidly kill the sh. This process is also useful in detecting serious spinal injuries, since when such injuries occur, there is a
FIGURE 4.
Harpoon details and current pathways for electroslaughtering tuna.
471
misalignment of successive vertebrae that blocks the path of the rod. The test protocol followed to evaluate each waveform is described below. The tests began using a high voltage (or current) to achieve a fast electroslaughtering in order to avoid unsafe situations for the scuba divers. If an optimal (fast and safe) electroslaughtering was not achieved with the rst discharge (which may be a sign of a badly stuck harpoon), the same voltage and current values were used for the next tuna. If the waveform continued to be ineffective (i.e., it did not slaughter the tuna outright) in the subsequent tests, it was discarded and a new one tested. Otherwise (waveform effective), depending on the scuba divers comments and the on-deck spinal rod examination, the voltage was either maintained if no spinal injuries were observed or decreased so as to reduce them. In some cases, the efciency of the tests was so low that it was not possible to kill the tuna, in which case that waveform was quickly rejected. In other cases, the waveform was rejected because spinal injuries were too serious and this could involve nancial loss for the company. The size of the tuna tested and the number of samples for each test day (not for each waveform type) were dependent on customer orders. The waveforms that we rst tested with the TEFSYS (Tuna ElectroFishing System) equipment were selected on the basis of our previous tuna electrostunning experience (Soto et al. 2006). The waveforms used and their main parameters are shown in Figure 5. These are high-frequency pulsed DC (HF-PDC), lowfrequency pulsed DC (LF-PDC), and low-frequency, decreasing exponential AC (LF-DEAC). The rst two have been widely used in freshwater electroshing (Lamarque 1990; Dalbey 1996; Ainslie 1998) and previously used by us with ad hoc systems specically designed for each waveform. The third waveform (LF-DEAC), not tested so far, was selected to match the good results previously obtained with lowfrequency waveforms and the regulated AC (Soto et al. 2006). We also added an exponential decreasing component in order to achieve a high initial value to stun the tuna followed by an attenuation with the aim of reducing the damage to the sh. To assess the waveforms, the following procedure was performed. Besides the rst examination carried out on deck, denitive results were obtained and recorded at the company factory. There, the tuna were lleted in four parts; these were inspected by an expert who rated the meat quality and the spinal injuries. The meat quality score (Core) varies from 0 to 5 (5 is the best). The meat quality index is inuenced not only by the color of the meat but also by the presence of burns, blood clots, and bloodspots. The spinal injury score (RC) varies also from 0 to 5 (0 indicates no spinal injuries). Spinal injuries have also been observed to affect the nal quality, since the esh located around an injury is affected. The company established two quality thresholds, one for the Core (3.5) and a second one for the RC (<2). A waveform was accepted if both thresholds were met. A total of 83 tuna were sacriced: 23 by means of HF-PDC, 22 by using LF-PDC, and 28 by using LF-DEAC. The number
of tests is not very large owing to their high cost (e.g., the price of an average-sized tuna can reach $3,000). For each tested waveform the following statistics were calculated: average weight, weight SD, and mean and SD for the meat quality score (Core) and spinal injury score (RC). Furthermore, the two most representative indicators of the suitability of the waveform were obtained as well, these being the percentage of tuna with meat quality score (Core) greater than 3.5 and the percentage of tuna with spinal injury score (RC) less or equal to 2.
Description of the Equipment The multifunctional and programmable electronic equipment, TEFSYS, for use in sh farming that is described here has been developed to meet the need for versatility. It can be used in research and, in a complementary form, in commercial sh farming for both marine and freshwater species, although its design has been oriented towards use in seawater. The system allows all the electrical variables used in the process to be recorded in detail. Furthermore, this equipment is versatile enough to generate a wide range of waveforms (note that the traditional use of simple waveforms has been dictated by the lack of electroshing equipment with the necessary versatility). The equipment has two differentiated but interacting parts: a software section and a hardware section. The software allows the user to select the waveform that will be applied to the sh through the electrodes, and this is displayed on the user interface. Sensors allow the software to identify the physical environment (air, water, or sh) in which the electrodes are situated, and activation of the electrical discharge is disabled if the electrodes are in air or, in cases when they are integrated in a harpoon, if this harpoon is not piercing a sh. The user interface is shown in Figure 6. At the top are the controls that congure the waveform, and the form is drawn in real time in the graph at the right. At the bottom are two differentiated areas. To the left is a set of indicators that show the state of the discharge switch of the equipment. Also shown is the physical environment (water, air, or sh) surrounding the harpoon for seawater electroshing or the physical environment (water or air) surrounding the electrodes for freshwater electroshing. To the right of the gure, the values of applied voltage and current are displayed during the electrical discharge. These values are stored for later retrieval and analysis. With the use of this software and the recording of currents and voltages used during each capture, it is possible to determine the relationship of each of the parameters of the electrical wave with the nal quality of the meat. The electrical hardware is shown in Figure 7. Its main constituent is a full bridge with a LC lter. In addition, an isolation transformer protects the equipment of the boat. The converter acts as a power amplier that reproduces the waveform signal applied. In the output, the two electrodes that apply electrical discharges to the sh can be integrated in a harpoon or can be
472
SOTO ET AL.
Waveform
100 90 80 70 60 50 40 30 20 10 0
Parameters
Waveform 1: High frequency pulsed direct current (HF-PDC) Frequency = 1 kHz Vpp = 100 V Duty cycle = 50%
0,0 2,0m 4,0m 6,0m 8,0m 10,0m time 12,0m 14,0m 16,0m 18,0m 20,0m
100 90 80 70 60 50 40 30 20 10 0
Waveform 2: Low frequency pulsed direct current (LF-PDC) Frequency = 10 Hz Vpp = 100 V Duty cycle = 50%
0,0 20,0m 40,0m 60,0m 80,0m 100,0m time 120,0m 140,0m 160,0m 180,0m 200,0m
100 80 60 40 20 0 -20 -40 -60 -80 -100
Waveform 3: Low frequency decreasing exponential alternate current (LF-DAC) Frequency = 20 Hz Vpp = 200 V
0,0 200,0m 400,0m 600,0m 800,0m 1,0 1,1 1,2 1,3 1,4 1,5 1,6 1,7 1,8 1,9 2,0 time
FIGURE 5. Waveforms used to test programable electroshing device.
physically independent, depending on the mode of operation that is required. If this equipment is used for applications in seawater, such as for the slaughter of tuna, it will be subjected to abrupt variations of load. The harpoon is red at the tuna and the discharge is applied. Sometimes the discharge starts before the electrical harpoon is inside the tuna, and this can cause changes in the load impedance of the equipment from about 5 (impedance of the wires) to 50 (impedance of the wires plus the esh of the tuna). Further, the harpoon may subsequently become detached from the tuna, or water may be introduced through the
wound, causing a steep change in the impedance of the equipment. Hence, the converter may be subjected to very high and unpredictable variations in load of up to 10 fold. Nevertheless, the output signal must remain as close as possible to the reference value under any situation. Because of this, the control of the converter must be able to maintain the output waveform selected, despite wide variations in the load. The effectiveness of two different types of control for this kind of converter was compared: one being a feedback control, and the other being a feedforward control by which the load current is measured. To perform these tests two control boards
473
FIGURE 6.
User interface of the new programmable electroshing device. [Figure available in color online.]
were designed, simulated, and implemented for feedback and feedforward control, respectively. The main electrical features of the converter to be controlled are shown in Table 1. RESULTS Laboratory Results After calculation and simulation of both control schemes, it was necessary to implement and test them rst at the laboratory. To test the equipment, a reference signal was applied and was replayed by the system analyzed with both control types and
TABLE 1. Characteristics of the power converter to be controlled.
abrupt load variations. The selected reference waveforms were a positive offset sinusoidal current and a low-frequency pulsed DC. The watertuna transition was the object of study for each of the described controls and waveforms. This is the critical moment of the electroshing process, when the signal applied to the tuna must be exactly equal to the chosen reference value; otherwise, the good or bad quality of the meat obtained might not be due to the waveform but might result from a power surge. The results are shown in Figure 8. With a positive offset sinusoidal current or a low-frequency pulsed DC and feedback control (Figure 8a, c) large power surges can be seen when the load switches from the water to the tuna. With the same waveforms and feedforward control (Figure 8b, d) large power surges do not occur when the load switches from water to tuna. Field Results The statistical values for the waveforms tested as described in the Methods are summarized in Table 2. In the tests performed with the HF-PDC waveform, the meat quality was good but all the sh tested suffered serious spinal injuries. With the LF-PDC waveform the meat quality was optimal for most of the samples.
Parameter Input voltage Switch mode Switching frequency Output voltage Dynamic response
Characteristic 150 V Unipolar 20 kHz 100 V, + 100 V [0 Hz, 1.5 kHz]
474
SOTO ET AL.
FIGURE 7.
Electrical hardware circuit of the new programmable electroshing device.
However, in this case the spinal injuries were lower than those registered with HF-PDC. Finally, in the tests performed with LF-DEAC, the meat quality was the best, but the percentage of spinal injuries in the sh was quite high.
TABLE 2. Results obtained for each tested waveform. HF-PDC = highfrequency pulsed DC, LF-PDC = low-frequency pulsed DC, LF-DEAC = low-frequency, decreasing exponential AC.
Waveform type Metric assessed Number of tuna tested Average weight (kg) Weight SD (kg) Tuna with meat quality score (Core) > 3.5 (%) Mean meat quality score (Core) Meat quality score (Core) SD Tuna with spinal injury score (RC) 2 (%) Mean spinal injury score (RC) Spinal injury score (RC) SD HF- PDC LF-PDC LF-DEAC 23 49.9 11.5 87.0 4.1 0.33 0.0 3.7 0.58 22 178.3 61.3 86.4 4.5 0.62 81.8 1.8 0.89 28 223.8 86.8 92.9 4.5 0.47 25.0 3.1 0.97
DISCUSSION In the feedback control large power surges can be seen when the load switches from water to tuna, which is damaging to the sh. In the harpoon it is very important to have an output signal as similar as possible to the reference signal, since the electrical discharge is very often applied before the harpoon has been driven into the sh. In order to nd out the inuence of the waveform on the tuna meat, the applied waveform must have exactly the selected parameters; otherwise, the results could be attributable to voltage peaks rather than to the waveform itself. Even though this control functions to keep the output voltage similar to the reference whatever the load, it is not able to avoid power surges when the harpoon switches from water to tuna. As mentioned previously, this may be an important drawback in experimental assessments of the inuence of waveform on the quality of tuna meat. By using the feedforward control, we have managed to remove power surges at the switch from water to tuna. The output voltage practically equals the reference signal. With regard to the electroslaughtering eld tests, the effects of each waveform in the Core and RC are summarized in Table 2. The results obtained with HF-PDC and LF-PDC are very similar to those previously reported with our previous ad hoc systems (Soto et al. 2006). The HF-PDC waveform produced spinal injuries above the quality threshold (RC 2) in all specimens with a mean value of meat quality worse than the other two waveforms (low frequency). For the LF-PDC waveform spinal injuries were minimal with a good meat quality (Core
475
FIGURE 8. Laboratory results. (a) A positive offset sinusoidal current and control feedback. (b) A positive offset sinusoidal current and control feedforward. (c) A low-frequency pulsed DC and feedback control. (d) A low-frequency pulsed DC and feedforward control. [Figure available in color online.]
mean = 4.5). If we compare these results with those obtained for other freshwater species, they are very similar (Lamarque 1990; McMichael 1993; Dalbey 1996; Ainslie 1998; Snyder 2003). In view of these results, we discarded HF-PDC for electroslaughtering sh. There are no previous studies in freshwater or seawater for the use of LF-DEAC, which makes it interesting to compare it with the other low frequency waveform tested, LF-PDC. Although the use of waveforms LF-PDC and LF-DEAC provided an identical meat quality average value (4.5), the LFDEAC stands out because the number of samples exceeding the quality threshold (Core > 3.5) is higher with a lower SD value. Therefore, the meat quality resulting from LF-DEAC is the best of the three waveforms tested. However, it presents a problem: high spinal injuries. These are due to the high value of peak-topeak voltage (200 Vpp ) applied at the beginning. This voltage
was set to assure the initial stun of the tuna. In the future we intend to use the same waveform with a lower initial voltage to further improve the results. CONCLUSION A exible and programmable equipment for electroshing in both seawater and freshwater has been presented. The system supplies a need that exists in commercial operations, and also allows research into the effect of the waveform on the quality of the meat. The equipment is capable of generating any kind of waveform, identifying the environment the electrodes are, in and logging the voltage and current values that are applied to kill the sh. Although the equipment has been designed in the context of electroslaughtering tuna, it is perfectly compatible with freshwater shing; for this, the harpoon used for tuna would be replaced with electrodes.
476
SOTO ET AL. Garcia, editors. Domestication of the bluen tuna, Thunnus thynnus thynnus. Cahiers Options M editerran eennes, Cartagena, Spain. Kolz, A. L. 1989. A power transfer theory for electroshing. U.S. Fish and Wildlife Service Fish and Wildlife Technical Report 22. Lamarque, P., 1990. Electrophysiology of sh in electrics eld. Pages 433 in I. G. Coux and P. Lamarque, editors. Fishing with electricity: applications in freshwater sheries management. Fishing News Books, Oxford, UK. Lambooij, B., M. A. Gerritzen, H. Reimert, D. Burggraaf, G. Andr e, and H. Van De Vis. 2008. Evaluation of electrical stunning of sea bass (Dicentrarchus labrax) in seawater and killing by chilling: welfare aspects, product quality and possibilities for implementation. Aquaculture Research 39:5058. Lines, J., and S. Kestin. 2004. Electrical stunning of sh: the relationship between the electric eld strength and water conductivity. Aquaculture 241:219 234. Lines, J., and S. Kestin. 2005. Electric stunning of trout: power reduction using a two-stage stun. Aquacultural Engineering 32:483491. Lines, J. A., D. H. Robb, S. C. Kestin, S. C. Crook, and T. Benson. 2003. Electric stunning: a humane slaughter method for trout. Aquacultural Engineering 28:141154. McMichael, G. A. 1993. Examination of electroshing injury and short-term mortality in hatchery rainbow trout. North American Journal of Fisheries Management 13:229233. Miyake, P. M., J. M. de la Serna, A. Di Natale, A. Farrugia, I. Katavic, N. Miyabe, and V. Ticina. 2003. General review of bluen tuna farming in the Mediterranean area. Collective Volume of Scientic Papers International Commission for the Conservation of Atlantic Tunas 55:114124. Nordgreen, A. H., E. Slinde, D. Mller, and B. Roth. 2008. Effect of various electric eld strengths and current durations on stunning and spinal injuries of Atlantic herring. Journal of Aquatic Animal Health 20:110115. Reynolds, J. B. 1996. Electroshing. Pages 221253 in B. R. Murphy and D. W. Willis, editors. Fisheries techniques. American Fisheries Society, Bethesda, Maryland. Roth, B., D. Moeller, and B. Slinde. 2004. Ability of electric eld strength, frequency, and current duration to stun farmed Atlantic salmon and pollock and relations to observed injuries using sinusoidal and square wave alternating current. North American Journal of Aquaculture 66:208216. Ryan, M. J., and R. D. Lorenz. 1995. A high-performance sine wave inverter controller with capacitor current feedback and back-EMF decoupling. Pages 507513 in Power electronics specialists conference, 1995 PESC 95 record., 26th annual IEEE, volume 1. Institute of Electrical and Electronics Engineering, Atlanta. Sharber, N. G., S. W. Carothers, J. P. Sharber, J. C. De Vos, Jr., and D. A. House. 1994. Reducing electroshing-induced injury of rainbow trout. North American Journal of Fisheries Management 14:340346. Snyder, D. E., 2003. Electroshing and its harmful effects on sh. U.S. Geological Survey, Information and Technology Report USGS/BRD/ITR20030002, Denver. Soto, F., J. A. Villarejo, A. Mateo, J. Roca-Dorda, F. De la G andara, and A. Garc a. 2006. Preliminary experiences in the development of bluen tuna Thunnus thynnus (L., 1758) electroslaughtering techniques in rearing cages. Aquacultural Engineering 34:8391. Van de Vis, H., S. Kestin, D. Robb, J. Oehlenschl ager, B. Lambooij, W. M unkner, H. Kuhlmann, K. Kloosterboer, M. Tejada, A. Huidobro, H. Ottera, B. Roth, N. K. Sorensen, L. Akse, H. Byrne, and P. Nesvadba. 2003. Is humane slaughter of sh possible for industry? Aquaculture Research 34:211220.
The control is the crucial aspect of the design, since it must achieve an output signal that closely follows the reference value, even under marked load variations. The feedforward control proved to be more suitable for this application than the feedback control, since the former responds more rapidly to abrupt load variations. Furthermore, with the feedforward control large power surges do not occur, which is one of the most important aspects of this feature. Hence, we have achieved a robust and easily operated control at a low development and manufacturing cost. The control multiplier proposed by (Ryan et al. 1995) has been suppressed simply by replacing the proportional controller by a proportionalintegral one. The equipment has been tested in eld conditions using different waveforms that show, in all cases, a high degree of reliability and safety for the scuba divers. Of the tested waveforms, LF-DEAC showed the best results as far as meat quality is concerned, although spinal injuries were higher than expected due to the high voltage level. In the future, we intend to conduct more tests to rene the parameters of the LF-DEAC waveform and evaluate new signals. ACKNOWLEDGMENTS This work was supported by the Council of Science, Technology, Industry and Commerce of the Region of Murcia. REFERENCES
Ainslie, B. J., J. R. Post, and A. J. Paul. 1998. Effects of pulsed and continuous DC electroshing on juvenile rainbow trout. North American Journal of Fisheries Management 18:905918. Dalbey, S. R., T. E. McMahon, and W. Fredenberg. 1996. Effect of electroshing pulse shape and electroshing-induced spinal injury on long-term growth and survival of wild rainbow trout. North American Journal of Fisheries Management 16:560569. De la G andara, F. 2003. Son las tareas propias de un cultivo intensivo de peces mediterr aneos compatibles con su bienestar? [Are the tasks of Mediterranean sh farming compatible with their welfare?] Pages 17 in I Congreso internacional sobre bienestar animal [Animal welfare congress, 2003 AWC record, volume 1]. Spanish Department of Agriculture, Murcia. EFSA (European Food and Safety Authority). 2009. Species-specic welfare aspects of the main systems of stunning and killing of farmed tuna. EFSA Journal [online serial] 1072. DOI: 10.2903/j.efsa.2009.1072. Garc a, M. 2002. Procedure for electrostunning and/or electrosacrice, for industrial application to marine inchthyological species in oating cages. U.S. patent 6,453,596. September 24, 2002. Gregory, N. G. 2008. Animal welfare at markets and during transport and slaughter. Meat Science 80:211. Kavatic, I., V. Ticina, and V. Franicevic. 2003. Bluen tuna (Thunnus thynnus L.) farming on the Croatian coast of the Adriatic Seapresent stage and future plans, volume 60. Pages 101105 in C. R. Bridges, H. Gordin, and A.

A Comparative Analysis of the Economic Risk of Hybrid Striped Bass Fingerling Production in Ponds and Indoor Tanks
Carole Engle & Pratikshya Sapkota
a a a
Aquaculture/Fisheries Center, University of Arkansas at Pine Bluff, 1200 North University Drive, Mail Slot 4912, Pine Bluff, Arkansas, 71601, USA Version of record first published: 04 Sep 2012.
To cite this article: Carole Engle & Pratikshya Sapkota (2012): A Comparative Analysis of the Economic Risk of Hybrid Striped Bass Fingerling Production in Ponds and Indoor Tanks, North American Journal of Aquaculture, 74:4, 477-484 To link to this article: http://dx.doi.org/10.1080/15222055.2012.685214
ARTICLE
A Comparative Analysis of the Economic Risk of Hybrid Striped Bass Fingerling Production in Ponds and Indoor Tanks
Carole Engle* and Pratikshya Sapkota
Aquaculture/Fisheries Center, University of Arkansas at Pine Bluff, 1200 North University Drive, Mail Slot 4912, Pine Bluff, Arkansas 71601, USA
Abstract
A risk analysis was conducted to compare the economic risk of producing hybrid striped bass ( white bass Morone chrysops striped bass M. saxatilis) ngerlings in ponds and indoor tanks. Cost budgets developed previously for three pond sizes (0.4, 1.2, and 2.4 ha) and three tank sizes (945, 2,457, and 5,670 L) for each of six scales of production (50,000, 100,000, 250,000, 500,000, 1,000,000, and 2,000,000 ngerlings/year) were used as the initial starting point for the development of a spreadsheet-based risk analysis. Probability distributions and cumulative frequency distributions of break-even prices above total costs were calculated for each scenario. Pond production was less risky than tank production, and larger pond sizes had lower risk. Survival rates contributed the most to the economic risk of hybrid striped bass ngerling production. Research is needed to improve survival rates for hybrid striped bass ngerling production in ponds.
Hybrid striped bass (HSB; white bass Morone chrysops striped bass M. saxatilis), commonly known as sunshine bass, ngerling production is characterized by uctuations in yield that result primarily from varying survival rates in both ponds and tanks. Risk has been dened as uncertain consequences (Hardaker et al. 1997), as a situation with more than one possible outcome (Levy 2006), some of which might be unfavorable (Kay et al. 2008), and as all the possible outcomes and the probabilities of their occurring (Olson 2004). Thus, risk exists when the values of key parameters uctuate (Engle 2010); uctuating yields, survival rates, prices, and costs all result in risk to the farmer. Improved understanding of how various risk factors affect the economics of hybrid striped bass ngerling production may provide guidance for management decisions. The survival rates of hybrid striped bass ngerlings in commercial ponds have been reported to range from 0% to 70% (Mike Freeze, Keo Fish Farm, personal communication). In indoor tanks, the survival rates of fry through 15 d posthatch
(dph) have ranged from 3% to 94%, depending upon the types of feed used (Ludwig 1994a, b; Ludwig et al. 2008; Ludwig and Lochmann 2000, 2009), rotifer concentration (Ludwig and Lochmann 2000), tank stocking density (Ludwig and Lochmann 2007), and feeding frequency (Ludwig 2003). The coefcients of variation for survival ranged from 2% to 104% across these same studies. However, the majority of studies on hybrid striped bass ngerling production in tanks have focused only on production through 1421 dph, with little work on culture through the phase-I size (approximately 1 g [3045 d old]). Varying survival rates and prices of inputs create economic risk that is reected in uctuating break-even prices. Eklund et al. (2012) demonstrated that the cost of producing hybrid striped bass ngerlings would be reduced by 714% in ponds and 511% in tanks for each 5% improvement in survival. However, the static cost analysis developed by Eklund et al. (2012) did not allow for explicit evaluation of the effects of the risk of uctuations in survival and other parameters.
*Corresponding author: cengle@uaex.edu Received December 9, 2011; accepted April 9, 2012
477
478
ENGLE AND SAPKOTA
A variety of types of spreadsheet-based risk analysis have been used in the aquaculture economics literature. These studies have used software add-ons (Crystal Ball, Oracle USA, Inc., Redwood City, California; @RISK software, Palisade Corporation, Ithaca, New York) to commercially available spreadsheet software. Several studies have examined risk levels and the probability of losses due to alternative management choices. For example, Valderrama and Engle (2001) used survey data to measure the levels of nancial risk at Honduran shrimp farms, nding that risk levels were higher on small whiteleg shrimp Penaeus vannamei farms and for farms using intensive production systems. In Mexico, Martinez and Seijo (2001) measured the risks associated with various water exchange and aeration rates in shrimp production, identifying the price of postlarvae, growth, survival rates, and the harvest-size shrimp price as the major contributors to risk. Partial aeration was the risk-neutral choice among three aeration regimes for cage culture of channel catsh Ictalurus punctatus in the United States, although none of the aeration methods was preferable to the others according to rst- or second-degree stochastic dominance criteria (Nelson et al. 2001). Seijo (2004) showed that alternative harvest scheduling of shrimp can reduce risk. Lower probabilities of nancial losses were found with Nile tilapia Oreochromis niloticus raised in monoculture with pelleted feeds rather than rice bran or in the culture of sharptooth catsh Clarias gariepinus (Neira et al. 2009). There are relatively few studies that have examined risk in the production of intermediate aquaculture products, such as ngerlings or stocker sh. Engle and Valderrama (2002) showed that economic risk was lower at catsh ngerling stocking densities that resulted in higher yields. Pomerleau and Engle (2003) showed that stocking catsh ngerlings at a medium density resulted in the lowest risk of the break-even costs of stocker production being above market prices. The survival rate of ngerlings was identied as a critical factor in costs and risk. Other analyses have examined the feasibility of new species, investments, production systems, or types of farm organization from the perspective of risk. For example, Medley et al. (1994) showed that the economic feasibility of producing Australian red claw crawsh Cherax quadricarinatus was sensitive to various risk factors, including the cost of the juveniles stocked, the production of larger crawsh, price, and the length of the growing season. Net present values for shrimp farming investments in Florida were found to be negative, even with price premiums (Clark et al. 2010). In Rwanda, sh farming was shown to entail more risk than farming traditional crops such as cabbages and beans (Hishamunda et al. 1998). The Hishamunda et al. (1998) study also looked at the form of organization and found that cooperatives had a higher probability of failure than individually managed farms. Zucker and Anderson (1999) used a spreadsheet-based risk modeling approach to identify the production and marketing targets most likely to result in economic success for newly developed technologies in indoor, land-based
summer ounder Paralichthys dentatus systems. Their ndings showed that such facilities should be located near a source of saltwater and that medium-sized sh products should be sold mostly to premium market outlets. Much of the work done on the economics of hybrid striped bass production has studied costs at the food sh stage. Enterprise budgets have been developed (Dunning and Daniels 2001), the costs of pond and tank production compared (Gempesaw et al. 1992b), and the costs of two-and three-phase systems (DAbramo et al. 2002, 2004) and economies of scale measured in ponds (Gempesaw et al. 1992a). The costs of various efuent treatment options (Wui and Engle 2004; Sydorovich and Daniels 2011) and the economic effects of restrictions on access to black carp Mylopharyngodon piceus have been evaluated (Wui and Engle 2007). Eklund et al. (2012) developed a comprehensive cost analysis of hybrid striped bass ngerling production in ponds and indoor tank systems. Three pond sizes (0.4, 1.2, and 2.4 ha) and three sizes of tanks (945, 2,457, and 5,670 L) were modeled at six scales of production (50,000, 100,000, 250,000, 500,000, 1,000,000, and 2,000,000). The cost per thousand ngerlings produced in ponds was estimated to range from US$86 to $107, compared with $260 to $385 per thousand ngerlings in tanks. Pond production was the lowest-cost production system. Economies of scale were identied in pond production and, to some degree, in tank production; costs per thousand ngerlings decreased with increasing numbers of production cycles per year. The overall goal of this project was to compare the economic risk of producing hybrid striped bass ngerlings in various sizes of ponds and tanks and at various scales of production. The specic objectives were to (1) estimate probability density functions of the break-even prices of hybrid striped bass ngerling production in various sizes of ponds (0.4, 1.2, and 2.4 ha) and tanks (945, 2,457, and 5,670 L) for production scales of 50,000, 100,000, 250,000, 500,000, 1,000,000, or 2,000,000 ngerlings per year; (2) identify the parameters that contribute the most to the economic risk of hybrid striped bass ngerling production in ponds and tanks; and (3) compare the economic risks of hybrid striped bass ngerling production in ponds and tanks.
METHODS The Eklund et al. (2012) budgets for 15 pond and 17 tank production scenarios were used as the starting point for development of a risk analysis. Data were compiled for the following key parameters: (1) feed prices from 1990 to 2009 (Hanson and Sites 2010); (2) the federal funds rate (Federal Reserve Board; www.federalreserve.gov) as a measure of the variability in interest rates; (3) wage rates (Bureau of Labor Statistics; www.bls.gov); (4) electric and gasoline/diesel rates (Department of Energy; www.eia.doe.gov); and (5) the prices of fertilizer, salt, and lime (U.S. Department of Agriculture, Economic Research Service; www.ers.usda.gov).
ECONOMIC RISK OF HYBRID STRIPED BASS FINGERLING PRODUCTION
479
TABLE 1.
Assumptions for the risk analysis of hybrid striped bass ngerling production in tanks and ponds; n.a. = not applicable.
Variable Survival rate Feeds Labor %
Unit
Distribution Triangular Normal Uniform Uniform Uniform Uniform Normal
Data set used Ludwig (2004, 2006) Hanson and Sites (2010) BLSc BLSc BLSc Department of Energye Department of Energye Department of Agriculturef Department of Agriculturef Department of Agriculturef Department of Energye Eklund et al. (2012) Eklund et al. (2012) Eklund et al. (2012) Eklund et al. (2012) Eklund et al. (2012) Eklund et al. (2012) Eklund et al. (2012) Federal Reserve Boardg Eklund et al. (2012) Eklund et al. (2012)
Statistical test used for normality Observation ShapiroWilk n.a. n.a. n.a. None Kolmogorov Smirnov D test None ShapiroWilk ShapiroWilk None n.a. n.a. n.a. n.a.
Tanks 0, 35, 70a $3.94 0.17 n.a. 6.9, 8.0d 635, 736.1d 0.11, 0.16d
Ponds 0, 35, 70a $3.938 0.169b 412.5, 478.20d n.a. n.a. 240, 352.63d 325 73.625b 510 2.5b 2,373.8 346.2b
$/kg $/ha $/h (fry) $/L (rotifers) $/L, $/ha $/ha $/ha $/metric ton $/kg $/ha $/L, $/ha $/tank rental $/box $/ha $/ha $/ha $/ha % $/million 0.5 kg
Electricity
Gas and diesel
Fertilizer Lime Salt Pumping Repairs and maintenance Oxygen Shipping bags Bird control Telephone Ofce supplies Legal, accounting Interest rates
Normal Normal normal Uniform Triangular Triangular Triangular Triangular Triangular Triangular Triangular Normal
0.21 0.03b
0.21 0.03b 527.5, 755.1d
0.02, 0.06, 0.03a 18.7, 31.2, 25a 45, 75, 60a
125, 485, 242.5a 338, 562, 450a 45, 75, 60a 11.7, 31.2, 15.6a
n.a.
37.5, 85, 42.5a 24.7, 27.5, 27.5a
ShapiroWilk
42.3, 70.5, 47.0a 10 5b 380 16.3 21.8 0.9
42.3, 70.5, 47.0a 10 5b
Rotifer cysts Rotifer diet

a
Normal Normal
n.a. n.a.
Minimum, maximum, and likeliest values used for the triangular distributions. Mean SD used for the normal distributions. c Bureau of Labor Statistics (www.bls.gov). d Minimum and maximum used for the uniform distributions. e www.eia.doe.gov. f www.ers.usda.gov. g www.federalreserve.gov
b
480
ENGLE AND SAPKOTA
A spreadsheet-based risk analysis was developed in which distributions estimated from data sets of uctuating values for various parameters were substituted for single-point values. Previous risk analyses primarily used triangular distributions for parameters with uctuating values. Nelson et al. (2001) used triangular distributions for all risky parameter values, Neira et al. (2009) for all but the feed conversion ratio (FCR) and growth rate, and Valderrama and Engle (2001) for all but yield and shrimp price. Pomerleau and Engle (2003) used normal distributions for FCR, yield, survival, and sh length but a uniform distribution for feed price. In the present study, the data for interest rates and the prices of feed, lime, and salt were determined to be distributed normally by the ShapiroWilk test (SAS version 9.2), and fuel and diesel prices were found to be distributed normally by the KolmogorovSmirnov test using the EasyFit Statistical Program (version 5.4, MathWave Technologies, Dnepropetrovsk, Ukraine). Minimum wage legislation in the United States establishes a price oor for wages. A uniform distribution was used that established minimum ($412/ha) and maximum ($478/ha) rates and assumed that all values within the range were equally likely. The maximum rate was established based upon the annual average rate of increase in wage rates projected forward for 5 years. In the same vein, electric rates are not distributed normally because the price oors are set by public utility commissions. A uniform distribution was established for electric rates, with the minimum rate being set as current charges and the maximum rate by projecting forward the annual average rates of increase for 5 years. Table 1 lists the distributions and the values used for each parameter. Monte Carlo simulations with 1,000 iterations were conducted with Crystal Ball to generate probability distributions of the break-even prices (BEPs) of hybrid striped bass ngerling production for each of the 32 pond/tank scenarios. The probability of a BEP above total costs being higher than the mean price in the budget was calculated for each scenario. Cumulative distribution functions of BEPs above total costs were developed and presented graphically to show whether rst-degree stochastic dominance was present. First-degree stochastic dominance is present if every point on a cumulative distribution function is preferable to the corresponding point on another cumulative distribution function (Levy 2006). RESULTS Of the production and cost parameters associated with the production of hybrid striped bass ngerlings, survival rate was the most variable, followed by the federal funds rate and fertilizer prices (Figure 1). Gasoline and diesel prices were the fourth most variable parameters and wage rates the fth, while less variation was observed in the prices of lime, electricity, and salt. Probability distributions of the break-even prices above total costs for hybrid striped bass ngerling production were devel-
FIGURE 1. Coefcients of variation for the key parameters in the spreadsheetbased risk analysis.
oped for each of the 32 production scenarios; examples of two (1 million ngerlings produced in 1.2-ha ponds and in 2,457-L tanks) are provided in Figure 2, using the convention of expressing frequency of occurrence as probability values. All probability distributions were positively skewed, indicating that there are proportionately more observations of very high costs of production than of very low costs of production for both pond and tank scenarios. The tank scenarios, however, had greater probabilities of higher-than-average costs than the pond scenarios and reect the overall higher break-even prices for tanks.
FIGURE 2. Probability distributions of break-even prices above total costs for 1 million hybrid striped bass ngerlings in (a) a 1.2-ha pond and (b) a 2,457-L tank. The values on the y-axis are estimated probabilities of occurrence.
ECONOMIC RISK OF HYBRID STRIPED BASS FINGERLING PRODUCTION
481
TABLE 2. Certainty levels of break-even prices (BEPs; $/1,000 ngerlings) being higher than the most efcient (lowest-cost) pond sizeproduction scale combination; n.a. = not applicable.
Pond production scenario 50,000 100,000 250,000 500,000 1,000,000 2,000,0000 50,000 100,000 250,000 500,000 1,000,000 2,000,0000 50,000 100,000 250,000 500,000 1,000,000 2,000,0000
BEP 0.4-ha ponds 107 93 89 88 88 86 1.2-ha ponds n.a. 93 89 88 88 86 2.4-ha ponds n.a. n.a. 89 88 88 86
Certainty level (%) 51 64 68 72 66 68 n.a. 52 55 57 56 59 n.a. n.a. 54 50 52 54
FIGURE 3. Cumulative distribution functions of break-even prices above total costs for 1 million hybrid striped bass ngerlings/year in (a) ponds and (b) tanks of different sizes. The values on the y-axis are estimated probabilities of occurrence.
Pond size affected the economic risk of producing hybrid striped bass ngerlings (Figure 3a). The cumulative distribution functions of BEPs above total costs demonstrate that 1.2-ha and 2.4-ha ponds exhibit rst-order stochastic dominance over 0.4-ha ponds (lines to the left reect lower production costs for every given probability), indicating higher economic risk in 0.4ha ponds when holding production constant. Thus, the probabilities of costs being higher than the most efcient (lowest-cost) pond sizescale combination were higher for the 0.4-ha ponds (5172%) than for the larger pond sizes (5259% for the 1.2-ha ponds and 5054% for the 2.4-ha ponds) (Table 2). The cumulative distribution functions for the different sizes of tanks were more similar to each other, but again the larger tank sizes tended to dominate when production was held constant (Figure 3b). The scale of production had little effect on the economic risk of producing hybrid striped bass ngerlings in ponds (Figure 4a), but some scale effects were found in tank production (Figure 4b). The cumulative distribution functions were nearly identical across production scales for ponds, indicating similar amounts of risk. With tanks, however, the cumulative distribution functions showed rst-degree stochastic dominance of several production scales over others: (1) the highest scales of production (100,0002 million) dominated the 50,000 scale; (2) the 100,000, 500,000, 1 million, and 2 million scales dominated
the 250,000 scale; and (3) the 500,000, 1 million, and 2 million scales dominated the 100,000 scale (Figure 4b). Thus, larger scales of production generally entailed less risk for a given tank size (with the exception of the 100,000 scale dominating the 250,000 scale). Since tank size and production scale affected the risk of producing hybrid striped bass ngerlings in tanks, the probabilities of production costs being higher than in the most efcient scenario generally increased with production scale in the small (945-L) tanks and generally decreased with production scale in the larger tanks (Table 3). For the lowest production scale, the probability of higher costs was lower for the smallest tank size. Tank production of hybrid striped bass ngerlings entailed higher levels of risk than did pond production (Figure 5). Pond production dominated tank production by rst-degree stochastic dominance for all scales of production; the probability of tank production of hybrid striped bass ngerlings costing less than pond production was 0%. Survival rate was the greatest source of economic risk for hybrid striped bass ngerling production in ponds (9496%) and tanks (9698%) (Tables 4 and 5). The price of lime and interest rates were the next most important factors contributing to the risk of raising hybrid striped bass ngerlings in ponds, whereas electric rates and the price of rotifer cysts were the next
482
ENGLE AND SAPKOTA
FIGURE 5. Comparative economic risks of hybrid striped bass ngerlings produced in ponds and tanks. The values on the y-axis are estimated probabilities of occurrence
FIGURE 4. Cumulative distribution functions of break-even prices above total costs for different scales of production in (a) a 1.2-ha pond and (b) a 2,457-L tank. The values on the y-axis are estimated probabilities of occurrence.
most important in the case of tanks, although these contributed substantially less to risk than did the survival rate. DISCUSSION Increasing pond size reduced the economic risk of hybrid striped bass ngerling production. Larger (1.2-ha and 2.4-ha) ponds entailed less risk than did 0.4-ha ponds at each scale of production. The economies of larger pond sizes are well documented in pond aquaculture production (Keenum and Waldrop 1988; Engle 2003, 2007, 2010); smaller ponds are associated with higher per-unit xed costs of production, which result in
greater nancial risk. However, farmers must also consider that larger ponds can be more difcult to manage because more hectares are tied up in each management decision. There have been few studies on the risk of tank aquaculture production and few that have compared the costs and risk associated with varying sizes of tanks. In this study, tank size had a small though measurable effect on risk. Recirculating systems include several component parts that must be sized proportionately (Losordo et al. 1999). Components of sizes that are optimal for the varying tank sizes are not always available commercially. When several individual components have excess capacity, the system operates less efciently, increasing overall nancial risk (Dunning et al. 1998). This is why the 100,000 production scale demonstrated slightly less risk than the 250,000 scale. For the component sizes available on the market, the 100,000 scale used these components relatively more efciently. Contracting an engineer to construct all component parts would be prohibitively expensive for a hybrid striped bass hatchery seeking to add a recirculating system as just one part of a larger farming operation. Increasing the scale of tank production had less effect on risk, although some effects were found, particularly with regard to low scales of production. Overall, pond production entailed less risk than tank production, primarily due to the increased costs associated with tank
TABLE 3. Certainty levels of break-even prices (BEPs; $/1,000 ngerlings) being higher than the most efcient (lowest-cost) tank sizeproduction scale combination; n.a. = not applicable.
Certainty level (%) by tank size Production scale 50,000 100,000 250,000 500,000 1,000,000 2,000,000 BEP 337 305 275 260 269 269 945 L 53 57 59 62 57 61 2,457 L 64 49 66 57 53 55 5,670 L n.a. 66 51 50 50 52
ECONOMIC RISK OF HYBRID STRIPED BASS FINGERLING PRODUCTION TABLE 4. Percent contribution of various parameters to the risk of producing hybrid striped bass ngerlings in different sizes of ponds at a scale of 1 million ngerlings per cycle.
483
Parameter Survival Lime Interest rate (real estate) Legal, accounting Interest rate (equipment) Electricity
0.4 ha 96 1 1 <1 <1 <1
1.2 ha 94 2 <1 <1 <1 <1
2.4 ha 96 1 <1 <1 <1 <1
CONCLUSIONS The economic risk of producing hybrid striped bass ngerlings was found to be higher in tanks than in ponds, indicating that pond production is preferable to tank production. Larger pond sizes reduced risk, but farmers must also consider the practical management limits to increasing pond size to produce hybrid striped bass ngerlings. Survival rate was the greatest contributing factor to the economic risk of hybrid striped bass ngerling production. There is great need for research to identify ways to improve survival through to the phase-I size in ponds. ACKNOWLEDGMENTS This study was funded in part by USDAARS Specic Cooperative Agreement 58-6225-8-036. Peter Wui, Anita Kelly, and Ganesh Karunakaran provided helpful comments on the manuscript. REFERENCES
Clark, J. L., R. N. Weldon, C. M. Adams, and F. F. Wirth. 2010. Risk assessment of a shrimp aquaculture investment in Florida. Aquaculture Economics and Management 14:332357. DAbramo, L. R., C. L. Ohs, and T. R. Hanson. 2002. Production and economic analysis of two-phase and three-phase culture of sunshine bass in earthen ponds. North American Journal of Aquaculture 64:103112. DAbramo, L. R., C. L. Ohs, and T. R. Hanson. 2004. Effect of stocking weight and stocking density on production of hybrid striped bass (sunshine) in earthen ponds in the second phase of a 2-phase system. Journal of the World Aquaculture Society 35:3345. Dunning, R., and H. Daniels. 2001. Hybrid striped bass production in ponds: enterprise budget. Southern Regional Aquaculture Center, Publication 3000, Stoneville, Mississippi. Dunning, R. D., R. M. Losordo, and A. O. Hobbs. 1998. The economics of recirculating tank systems: a spreadsheet for individual analysis. Southern Regional Aquaculture Center, Publication 456, Stoneville, Mississippi. Eklund, P., C. Engle, and J. Ludwig. 2012. Comparative cost analysis of hybrid striped bass ngerling production in ponds and tanks. North American Journal of Aquaculture 74:3953. Engle, C. R. 2003. The evolution of farm management, production efciencies, and current challenges to catsh production in the United States. Aquaculture Economics and Management 7:6784. Engle, C. R. 2007. Species-specic public policy for sustainable development: the U.S. catsh industry. Pages 313332 in P.-S. Leung, C.-S. Lee, and P. OBryen, editors. Species and system selection for sustainable aquaculture. Blackwell Scientic Publications, Oxford, UK. Engle, C. R. 2010. Aquaculture economics and nancing: management and analysis. Wiley-Blackwell Scientic Publications, Ames, Iowa. Engle, C. R. and D. Valderrama. 2001. Effect of stocking density on production characteristics, costs, and risk of producing ngerling channel catsh. North American Journal of Aquaculture 63:201207. Engle, C. R. and D. Valderrama. 2002. Production characteristics, costs, and risk of producing channel catsh, Ictalurus punctatus, ngerlings on farms with thinning. Journal of Applied Aquaculture 12:5164. Gempesaw, C. M., J. R. Bacon, and F. F. Wirth. 1992a. Economics of pond size for hybrid striped bass growout. Journal of the World Aquaculture Society 23:3848. Gempesaw, C. M., D. Lipton, V. Varma, and J. R. Bacon. 1992b. A comparative economic analysis of hybrid striped bass production in ponds and tanks. Pages 3141 in N. Stone and V. Pennington, editors. Proceedings of the National
production. This supports ndings of Eklund et al. (2012) that pond production is economically preferable to tank production. As with the Eklund et al. (2012), study results may differ for palmetto bass ( striped bass Morone saxatilis white bass M. chrysops). Tuncer et al. (1990) showed that palmetto bass do not require the costly rotifer feeding stage. However, sunshine bass are preferred by industry due to the difculties of inducing striped bass females to spawn (Hodson et al. 1999) and the longer time to reach sexual maturity. Varying survival rates contributed the most to the economic risk of producing hybrid striped bass ngerlings. Pomerleau and Engle (2003), in a study of catsh stocker production, also found that survival-related risks can be critical factors in the overall cost and risk of producing intermediate aquaculture products. Engle and Valderrama (2001) found lower economic risks in catsh ngerling production with management strategies that resulted in higher yields. Higher yields are affected indirectly by the survival rate because higher densities are associated with higher yields. Valderrama and Engle (2001) also found that the yield risk of shrimp was the primary source of risk. Martinez and Seijo (2001) explicitly identied survival rate as a partial source of risk in shrimp production in Mexico. Additional attention to the effects of survival on production risk is warranted in the aquaculture economics literature. The hybrid striped bass literature has few reports of survival through to the phase-I size, in spite of the number of studies comparing the survival of larval stages to 1421 dph.
TABLE 5. Percent contribution of various parameters to the risk of producing hybrid striped bass ngerlings in different sizes of tanks at a scale of 1 million ngerlings per cycle.
Parameter Survival Interest rate (equipment) Feed (rotifer cyst) Feed (48% protein) Salt Electricity
945 L 97 <1 1 <1 <1 <1
2,457 L 98 <1 <1 <1 <1 <1
5,670 L 96 <1 <1 <1 <1 2
484
ENGLE AND SAPKOTA Ludwig, G. M., S. D. Rawles, and S. E. Lochmann. 2008. Effects of rotifer enrichment on sunshine bass Morone chrysops M. saxatilis larvae growth and survival and fatty acid composition. Journal of the World Aquaculture Society 39:158173. Martinez, J. A., and J. C. Seijo. 2001. Economics of risk and uncertainty of alternative water exchange and aeration rates in semi-intensive shrimp culture systems. Aquaculture Economics and Management 5: 129145. Medley, P. B., R. G. Nelson, U. Hatch, D. B. Rouse, and G. F. Pinto. 1994. Economic feasibility and risk analysis of Australian red claw craysh Cherax quadricarinatus aquaculture in the southeastern United States. Journal of the World Aquaculture Society 25:135146. Neira, I., C. R. Engle, and C. Nguji. 2009. Economic and risk analysis of tilapia production in Kenya. Journal of Applied Aquaculture 21:110. Nelson, R. G., S. A. Duarte, and M. Amasser. 2001. Financial risk analysis of three aeration regimes in catsh cage culture. Aquaculture Economics and Management 5:171178. Olson, K. D. 2004. Farm management: principles and strategies. Iowa State Press, Ames. Pomerleau, S., and C. R. Engle. 2003. Production of stocker-size channel catsh: effect of stocking density on production characteristics, costs, and economic risk. North American Journal of Aquaculture 65:112119. Seijo, J. C. 2004. Risk of exceeding bioeconomic limit reference points in shrimp aquaculture systems. Aquaculture Economics and Management 8: 201212. Sydorovych, O., and H. Daniels. 2011. Economic analysis of alternative efuent treatment options for pond production of hybrid striped bass in Aurora, North Carolina. Aquaculture Economics and Management 15:4670. Tuncer, H., R. M. Harrell, and E. D. Houde. 1990. Acceptance and consumption of food by striped bass and hybrid larvae. Journal of the World Aquaculture Society 21:225234. Valderrama, D., and C. R. Engle. 2001. Risk analysis of shrimp farming in Honduras. Aquaculture Economics and Management 5:4968. Wui, Y. S., and C. R. Engle. 2004. A mixed integer programming analysis of efuent treatment options proposed for pond production of hybrid striped bass, Morone chrysops M. saxatilis. Journal of Applied Aquaculture 15:121 158. Wui, Y.-S., and C. R. Engle. 2007. The economic impact of restricting use of black carp for snail control on hybrid striped bass farms. North American Journal of Aquaculture 69:127138. Zucker, D. A., and J. L. Anderson. 1999. A dynamic, stochastic model of a land-based summer ounder Paralichthys dentatus aquaculture rm. Journal of the World Aquaculture Society 30:219235.
Aquaculture Extension Workshop. Arkansas Cooperative Extension Program, University of Arkansas at Pine Bluff, Pine Bluff. Hanson, T. R., and D. Sites. 2010. Catsh production data. Mississippi State University, Stoneville. Hardaker, J. B., R. B. M. Huirne, and J. R. Anderson. 1997. Coping with risk in agriculture. CABI Publishing, New York. Hishamunda, N., C. M. Jolly, and U. Hatch. 1998. Evaluating and managing risk in small-scale sh farming in a developing economy: an application to Rwanda. Aquaculture Economics and Management 2:3138. Hodson, R. J., R. W. Clark, M. S. Hopper, A. S. McGinty, G. M. Weber, and C. V. Sullivan. 1999. Pages 2332 in T. Tamaru, C. S. Tamaru, J. McVey, and K. Ikuta, editors. Reproduction of domesticated striped bass: commercial mass production of ngerlings. University of Hawaii Sea Grant College Program, United States and Japan Cooperation Program in Natural Resources, Report 28, Honolulu. Kay, R. D., W. M. Edwards, and P. A. Duffy. 2008. Farm management. McGrawHill, New York. Keenum, M. E. and J. E. Waldrop. 1988. Economic analysis of farm-raised catsh production in Mississippi. Mississippi Agriculture and Forestry Experiment Station Technical Bulletin 155. Levy, H. 2006. Stochastic dominance: investment decision making under uncertainty. Springer, New York. Losordo, T. M., M. P. Masser, and J. E. Rakocy. 1999. Recirculating aquaculture tank production systems: a review of component options. Southern Regional Aquaculture Center, Publication 453, Stoneville, Mississippi. Ludwig, G. M. 1994a. Tank culture of sunshine bass Morone chrysops M. saxatilis fry with freshwater rotifers Brachionus calyciorus and salmon starter meal as rst food sources. Journal of the World Aquaculture Society 25:337341. Ludwig, G. M. 1994b. Rearing sunshine bass fry on tank-cultured freshwater rotifers. Aquaculture Magazine (September/October):3945. Ludwig, G. M. 2003. Tank culture of larval sunshine bass, Morone chrysops (Ranesque) M. saxatilis (Walbaum) at three feeding levels. Aquaculture Research 34:12771285. Ludwig, G. M., and S. Lochmann. 2000. Culture of sunshine bass Morone chrysops M. saxatilis, fry in tanks with zooplankton cropped from ponds with a drum lter. Journal of Applied Aquaculture 10:1126. Ludwig, G. M., and S. E. Lochmann. 2007. Effect of tank stocking density on larval sunshine bass growth and survival to the ngerling stage. North American Journal of Aquaculture 69:407412. Ludwig, G. M., and S. E. Lochmann. 2009. Tank culture of sunshine bass ngerlings without using rotifers. North American Journal of Aquaculture 71:224228.

An Ultrastructure Study of Diet-Related Changes in Epithelial Tissue of Hybrid Striped Bass Larvae
Margie L. Gallagher
a a
College of Human Ecology, East Carolina University, RW 238 Rivers Building, Greenville, North Carolina, 27858-4353, USA Version of record first published: 04 Sep 2012.
To cite this article: Margie L. Gallagher (2012): An Ultrastructure Study of Diet-Related Changes in Epithelial Tissue of Hybrid Striped Bass Larvae, North American Journal of Aquaculture, 74:4, 489-493 To link to this article: http://dx.doi.org/10.1080/15222055.2012.676021
COMMUNICATION
Margie L. Gallagher*
College of Human Ecology, East Carolina University, RW 238 Rivers Building, Greenville, North Carolina 27858-4353, USA
Abstract
This study characterized the ultrastructure of normal epidermal and gastrointestinal epithelial cells of larval hybrid striped bass (white bass Morone chrysops striped bass M. saxatilis) that were fed live prey (brine shrimp Artemia spp. nauplii) or an articial dry diet. Scanning electron microscopy of larvae that were given live feed revealed normal epidermal cells with microridge structures and proliferation of neuromasts along the lateral line. Cell junctions had double microridge structures. Transmission electron microscopy of the gut showed keratinized epithelial cells with underlying mucous cells in the foregut, while the midgut was characterized by columnar epithelial cells with extensive pinocytotic activity. The hindgut had columnar epithelial cells with centrally located nuclei, regularly spaced microvilli, and numerous tubular mitochondria surrounded by rough endoplasmic reticulum. Larvae that received live feed reached metamorphosis by 27 d posthatch (dph), and clearly identiable gastric glands were present. Larvae that were given the articial diet did not develop; both epidermal and gastrointestinal cells showed apparent osmotic stress by 10 dph, and the condition worsened through 24 dph. Stress was indicated by loss of microridges in epidermal cells and eventual necrosis. Increased keratinization and reduced pinocytotic activity were observed in midgut cells, whereas hindgut epithelial cells showed dissociation of the rough endoplasmic reticulum and reduced numbers of mitochondria. However, tight cell junctions, desmosomes, and double microridges at cell junctions persisted.
striped bass) that were offered even sophisticated articial diets was due to rapid gut evacuation or a lack of appropriate digestive enzyme activity (or both) in the premetamorphic larvae. Digestive enzymes are known to be active in larval striped bass from the time of hatch (Baragi and Lovell 1986); the notable exception is pepsin, which is not produced until the gastric glands of the stomach become differentiated. Feed technology development efforts have sought to address this issue. The purpose of the present study was to characterize the normal structure of the epithelial cells of premetamorphic larval hybrid striped bass and the changes associated with the use of dry articial diets. Epithelial cells of the gastrointestinal (GI) tract mucosa and epithelial cells of the epidermis (which are continuous with those of the GI tract) were examined. METHODS Hybrid striped bass (age = 4 d posthatch [dph]) were obtained from Keo Fish Farms (Keo, Arkansas) and immediately placed into one of nine 40-L cylindrical tanks with rounded bottoms; the tanks were tted with center standpipes and were supplied with gentle aeration via one air stone, similar to the system described by Denson and Smith (1997). Larvae were given one of three feed types: (1) newly hatched live brine shrimp Artemia spp. nauplii, (2) a dry feed consisting of 10% sh silage (prepared according to Gallagher 1994) and 90% commercially prepared larval feed, or (3) no feed. Ten larval sh were sampled from each tank at 4, 10, 20, and 27 dph. As suggested by Perez-Dominguez and Dahm (2011), larval status was measured based on the striped bass development stages described by Rogers et al. (1982). The stages are based primarily on yolk sac disappearance; eye, GI, and n development; and swimming and feeding behavior. Six larvae were prepared for transmission electron microscopy (TEM) and scanning electron microscopy (SEM). Whole larvae were xed in 1% formalin and 3% glutaraldehyde. Samples were
High mortality associated with the period of rst feeding to metamorphosis in striped bass Morone saxatilis and striped bass hybrids remains problematic for the culture of these sh despite advances in microencapsulation technologies and culture methods (Langdon and Barrows 2011). Striped bass and hybrids have a short yolk sac period, with larval stages that possess a very rudimentary digestive tract (i.e., lacking a functional stomach or gastric glands) and that must undergo complex metamorphosis of the digestive system to survive. Ohs et al. (1998) and Ohs (1995) suggested that the poor survival of striped bass and white bass Morone chrysops striped bass hybrids (hereafter, hybrid
*E-mail: gallagherm@ecu.edu Received January 4, 2012; accepted February 17, 2012
489
490
GALLAGHER
washed in sodium cacodylate, postxed with 1% osmium tetraoxide, and dehydrated in alcohol. For TEM, specimens were dissected into foregut, midgut, and hindgut sections and embedded in Epon 812 resin. Specimen blocks were trimmed, and silver sections were prepared for examination by TEM using a Phillips CM-10 high-resolution scope. For SEM, whole, xed larvae were mounted on aluminum stubs and were sputter coated with platinum for observation by using an International Scientic Instruments ISI-40 or a Quanta 200 scanning electron microscope. To be noted in this study, ultrastructure details had to occur repeatedly in all larval samples. RESULTS AND DISCUSSION Larval sh that were fed live brine shrimp nauplii developed normally and progressed through the morphological stages ah described by Rogers et al. (1982). The larvae achieved transformation (stage h; differentiation of rays in dorsal and pectoral ns; initiation of lateral line scales) by 27 dph, exhibiting clearly identiable gastric glands with numerous tubular mitochondria surrounded by rough endoplasmic reticulum (RER) and dense secretion granules (Figure 1) similar to those described by Cataldi et al. (2002) for 12-dph Adriatic sturgeon Acipenser naccarii. However, larvae that were starved or that were fed the dry diet did not develop morphologically past stage d (active pelagic feeding, well-developed mouth, and initial n differentiation) despite the fact that food could be seen in the GI tracts of sh that were given the dry feed. Starved larvae began to die at 6 dph, and all had died by 19 dph; larvae that received dry feed did not begin to die until 10 dph, but all had died by 24 dph.
Epithelial Cells of the Epidermis Hybrid striped bass larvae at 4 dph were at stage c (oil globule and yolk nearly absorbed; swimming pelagically) when they arrived at the laboratory. The epidermis, including the mouth, exhibited typical microridge structure as described by Ostrander (2000), and neuromasts were present as noted by Y ufera (2011). Bereiter-Hahn et al. (1979) described the microridge structures found on the surface of sh epidermis cells over 30 years ago and proposed a link between microridges and microvilli of the intestinal mucosa, which is also epithelial tissue that is continuous with the epidermis. The function of microridges remains unclear; certainly, they provide structural support, as the primary components are actin and alpha-actinin. However, microridges also increase surface area and may facilitate the exchange of gases and ions (Rulifson et al. 1986). Sperry and Wassersug (2005) proposed that microridges hold protective mucous secretions to the epidermis, thus providing a barrier between the sh and the surrounding environment. By 10 dph, larvae that received live feed had progressed to stage e (yolk sac absorbed; pectoral ns and teeth [covered with epithelial cells, Figure 2] visible). The epidermis cells were smaller (<10 m) and continued to exhibit typical microridge structures (Figure 3A). There was a continuous double microridge structure at the cell junctions; tight junction complexes and numerous cellular invaginations were present. Larvae that were given dry feed or that were starved had advanced only to stage d, and their cells demonstrated apparent osmotic stress as evidenced by poorly dened microridges and the looseness of the cell membrane (Figure 3B, C). However, the continuous double microridge structure persisted in these sh, even when other microridge structures were scarce or nonexistent (Figure 3B inset). Starved larvae died by day 19. All sh exhibited the formation of neuromasts (Figure 3) along the lateral line.
FIGURE 1. Cross section of gastric glands of a hybrid striped bass larva (at 27 d posthatch) that was fed live brine shrimp nauplii; secretion granules (sg) and microtubule structures (arrows) are apparent. Numerous mitochondria are also present (scale bar = 1 m). [Figure available in color online.]
FIGURE 2. Epithelial cells of the mouth of a hybrid striped bass larva (at 27 d posthatch) that received a diet of live brine shrimp nauplii; teeth can be seen emerging from the covering of epithelial cells with microridges (arrows; scale bar = 10 m).
COMMUNICATION
491
FIGURE 4. Epithelial cells along the lateral line of a hybrid striped bass larva (at 20 d posthatch) that was fed a dry articial diet. Note the persistence of neuromasts (arrow), even when many cells appear to be necrotic (scale bar = 10 m). [Figure available in color online.]
By 20 dph, larvae that were given the live feed had progressed to stage f or g (differentiation of ns, rudimentary spines, and gas bladder present) and the neuromasts had proliferated. Larvae that received dry feed remained at stage d, and numerous epidermal cells exhibited necrosis, although neuromasts persisted (Figure 4). Epithelial Cells of the Gastrointestinal Tract At 10 and 20 dph, the foregut epithelial cells of larvae that were given live feed appeared to be keratinized, with underlying mucous-secreting (goblet) cells (Figure 5A). As was noted by Lazo et al. (2011), mucous cells are typical of even undifferentiated esophageal tissue in many sh species and are thought to serve as a lubricant since sh do not have salivary glands. Although in some species the goblet cells are often found scattered throughout the intestine at the onset of exogenous feeding (Lazo et al. 2011), goblet cells were not observed beyond the foregut in these hybrid striped bass larvae. Instead, the midgut was characterized by cells with sparse microvillus structures and many vacuoles that appeared to be pinocytotic in nature (Figure 5B). The posterior gut was characterized by columnar epithelial cells, with numerous, regularly spaced microvilli, numerous mitochondria, and extensive RER typical of intestinal cells (Figure 5B). Mitochondria in all cells contained tubular rather than lamellar cristae and were similar to mitochondria that are characteristic of endocrine cells (Bozzola and Russell 1992). The mitochondria were surrounded closely by dense RER (Figure 5B inset). In the developing larvae before stomach differentiation, an intestinal-type cell could often be observed immediately posterior to a pinocytotic-type cell as in Figure 5B. A few enteroendocrine cells were also observed in this section of the gut. At 10 dph, larvae that were fed the dry diet had the same basic features of the gut as did larvae that were given live feed. In 20-dph larvae that received the dry diet, the midgut cells
FIGURE 3. Epithelial cells along the lateral line, with neuromasts present, in hybrid striped bass larvae (at 10 d posthatch): (A) a larva that was given live brine shrimp nauplii as feed, (B) a larva that was fed a dry articial diet, and (C) a starved larva. Note the stressed epidermal cells with sparse microridges (arrows; scale bar = 10 m). Inset in (B) depicts a transmission electron micrograph (21,000 ) of the epidermis, showing the double microridge structure at the cellular junction ( ), which persisted even in stressed cells. [Figure available in color online.]
492
GALLAGHER
FIGURE 6. Transmission electron micrographs of gastrointestinal cells in hybrid striped bass larvae (at 20 d posthatch) that received dry articial feed: (A) pinocytotic cells, showing a marked decline in activity, few or no mitochondria, numerous free ribosomes ( ), the presence of keratin-like brils (black arrow), the occurrence of tight junctions between cells, and the persistence of desmosomes (white arrow; scale bar = 1 m); and (B) microvillus cells, exhibiting much fewer mitochondria, rough endoplasmic reticulum that was rounded or nonexistent, and the presence of free ribosomes ( ; scale bar = 1 m).
showed a high degree of keratinization and pinocytotic activity had ceased (Figure 6A). Intestinal-type cells did not show normal RER, and mitochondria were sparse. Ribosomes appeared to be free or associated in spherical clusters (Figure 6B). Conclusions In the limited number of larval sh that were observed, both SEM and TEM revealed several aspects of the epithelial cells of the epidermis and gut during the premetamorphosis period in larvae that were given a live feed or a dry articial diet. Microridges and neuromasts were evident by 4 dph. Microridges of the epidermis were continuous with the gut, as revealed by scanning electron micrographs of the mouth. Three cell types were evident in the gut, including highly keratinized cells in the foregut, pinocytotic cells in the midgut, and intestine-like cells immediately posterior to the pinocytotic cells. Pinocytotic cells, which presumably develop into gastric glands of the
FIGURE 5. Epithelial cells in the intestinal tracts of hybrid striped bass larvae (at 20 d posthatch) that received a diet of live brine shrimp nauplii: (A) foregut, characterized by highly keratinized (arrow) cells with underlying mucous cells (mu; scale bar = 10 m); and (B) midgut, characterized by columnar epithelial cells with centrally located nuclei; the anterior occurrence of pinocytotic cells with sparsely spaced microvilli, extensive tight junctions (black arrow), and desmosomes (white arrows); and (immediately posterior to pinocytotic cells) intestinal absorptive-type cells with slender microvilli (scale bar = 1 m) and large numbers of tubular mitochondria surrounded by extensive rough endoplasmic reticulum (RER). Inset in (B) shows a transmission electron micrograph (6,600 ) of tubular mitochondria (mi) and RER.
COMMUNICATION
493
stomach at metamorphosis, could be seen lying immediately next to intestine-like cells in the premetamorphic larvae. The lack of development in larvae that were fed the dry feed may be due to a lack of digestive enzymes, particularly pepsin, and a lack of GI development, as has been suggested by others (Dabrowski and Culver 1991; Ohs 1995; Langdon and Barrows 2011). In addition, I hypothesize that the lack of development in the hybrid striped bass larvae may have been related to osmotic stress due to the ingestion of dry feed with a high osmotic load or (as expressed by Ohs et al. 1998) a high density. In larvae that received the dry articial diet, both the epidermal and gut epithelial cells exhibited osmotic stress, as indicated by the loss of microridge structure, dissociation of the RER, and eventual cell necrosis. The GI tract is known to be involved in osmoregulation in sh (Allen et al. 2009); however, more studies are needed to conrm this idea. I also observed that in larvae receiving the dry diet, the neuromasts and double microridges at cell junctions persisted even under stressful conditions, thus demonstrating the signicance of these structures for the integrity of larval epithelial cells. REFERENCES
Allen, P. A., J. J. Check Jr., and D. K ultz. 2009. Mechanisms of seawater acclimations in a primitive, anadromous sh, the green sturgeon. Journal of Comparative Physiology Part B 179:903920. Baragi, V., and R. T. Lovell. 1986. Digestive enzyme activities in striped bass from rst feeding through larva development. Transactions of the American Fisheries Society 15:478484. Bereiter-Hahn, J., M. Osborn, K. Weber, and M. Voth. 1979. Filament organization and formation of microridges at the surface of sh epidermis. Journal of Ultrastructure Research 69:316330. Bozzola, J. J., and L. D. Russell. 1992. Electron microscopy. Jones and Bartlett, Boston. Cataldi, E., C. Albano, D. Boglione, L. Dini, G. Monaco, P. Bronzi, and S. Cataudella. 2002. Acipenser naccarii: ne structure of the alimentary canal
with references to its ontogenesis. Journal of Applied Ichthyology 18:329 337. Dabrowski, K., and D. Culver. 1991. The physiology of larval sh. Digestive tract and formulation of starter diets. Aquaculture Magazine 17:4961. Denson, M. R., and T. I. J. Smith. 1997. Tank culture of larval sunshine bass. Progressive Fish-Culturist 59:5963. Gallagher, M. L. 1994. Preliminary evaluation of sh silage as a weaning diet for hybrid striped bass. Bulletin of the Aquaculture Association of Canada 1:2628. Langdon, C., and R. Barrows. 2011. Microparticulate diets: technology. Pages 335351 in G. J. Holt, editor. Larval sh nutrition. Wiley, West Sussex, UK. Lazo, J. P., M. J. Darias, and E. Gisbert. 2011. Ontogeny of the digestive tract. Pages 546 in G. J. Holt, editor. Larval sh nutrition. Wiley, West Sussex, UK. Ohs, C. L. 1995. The development and evaluation of a spray-dried articial diet for larval culture of freshwater prawn (Macrobrachium rosenbergii), hybrid striped bass (Morone saxatilis M. chrysops) and striped bass (M. saxatilis). Masters thesis. Mississippi State University, Mississippi State. Ohs, C. L., L. R. DAbramo, R. K. Buddington, H. R. Robinette, J. M. Roethke. 1998. Evaluation of a spray-dried articial diet for larval culture of freshwater prawn, Macrobrachium rosenbergii, and striped bass, Morone saxatilis. Aquaculture Nutrition 4:7382. Ostrander, G. K. 2000. The laboratory sh. Academic Press, London. Perez-Dominguez, R., and R. Dahm. 2011. Methods for assessing embryonic and larval growth in sh. Pages 373402 in G. J. Holt, editor. Larval sh nutrition. Wiley, West Sussex, UK. Rogers, B. A., D. T. Westin, and S. B. Saila. 1982. Development of techniques and methodology for the laboratory culture of striped bass, Morone saxatilis. U.S. Environmental Protection Agency, Report PB-820217795, Washington, D.C. Rulifson, R. A., J. E. Cooper, and G. Colombo. 1986. Development of feed and starved striped bass (Morone saxatilis) larvae from the Roanoke River, North Carolina. North Carolina Department of Natural Resources and Community Development, Albemarle-Pamlico Estuary System Project 9012, Raleigh. Sperry, D. G., and R. J. Wassersug. 2005. Proposed function for microridges on epithelial cells. Anatomical Record 185:253258. Y ufera, M. 2011. Feeding behavior in larval sh. Pages 285305 in G. J. Holt, editor. Larval sh nutrition. Wiley, West Sussex, UK.

Margie L. Gallagher
a a
College of Human Ecology, East Carolina University, RW 238 Rivers Building, Greenville, North Carolina, 27858-4353, USA Version of record first published: 04 Sep 2012.
To cite this article: Margie L. Gallagher (2012): An Ultrastructure Study of Diet-Related Changes in Epithelial Tissue of Hybrid Striped Bass Larvae, North American Journal of Aquaculture, 74:4, 489-493 To link to this article: http://dx.doi.org/10.1080/15222055.2012.676021
COMMUNICATION
Margie L. Gallagher*
College of Human Ecology, East Carolina University, RW 238 Rivers Building, Greenville, North Carolina 27858-4353, USA
Abstract
This study characterized the ultrastructure of normal epidermal and gastrointestinal epithelial cells of larval hybrid striped bass (white bass Morone chrysops striped bass M. saxatilis) that were fed live prey (brine shrimp Artemia spp. nauplii) or an articial dry diet. Scanning electron microscopy of larvae that were given live feed revealed normal epidermal cells with microridge structures and proliferation of neuromasts along the lateral line. Cell junctions had double microridge structures. Transmission electron microscopy of the gut showed keratinized epithelial cells with underlying mucous cells in the foregut, while the midgut was characterized by columnar epithelial cells with extensive pinocytotic activity. The hindgut had columnar epithelial cells with centrally located nuclei, regularly spaced microvilli, and numerous tubular mitochondria surrounded by rough endoplasmic reticulum. Larvae that received live feed reached metamorphosis by 27 d posthatch (dph), and clearly identiable gastric glands were present. Larvae that were given the articial diet did not develop; both epidermal and gastrointestinal cells showed apparent osmotic stress by 10 dph, and the condition worsened through 24 dph. Stress was indicated by loss of microridges in epidermal cells and eventual necrosis. Increased keratinization and reduced pinocytotic activity were observed in midgut cells, whereas hindgut epithelial cells showed dissociation of the rough endoplasmic reticulum and reduced numbers of mitochondria. However, tight cell junctions, desmosomes, and double microridges at cell junctions persisted.
striped bass) that were offered even sophisticated articial diets was due to rapid gut evacuation or a lack of appropriate digestive enzyme activity (or both) in the premetamorphic larvae. Digestive enzymes are known to be active in larval striped bass from the time of hatch (Baragi and Lovell 1986); the notable exception is pepsin, which is not produced until the gastric glands of the stomach become differentiated. Feed technology development efforts have sought to address this issue. The purpose of the present study was to characterize the normal structure of the epithelial cells of premetamorphic larval hybrid striped bass and the changes associated with the use of dry articial diets. Epithelial cells of the gastrointestinal (GI) tract mucosa and epithelial cells of the epidermis (which are continuous with those of the GI tract) were examined. METHODS Hybrid striped bass (age = 4 d posthatch [dph]) were obtained from Keo Fish Farms (Keo, Arkansas) and immediately placed into one of nine 40-L cylindrical tanks with rounded bottoms; the tanks were tted with center standpipes and were supplied with gentle aeration via one air stone, similar to the system described by Denson and Smith (1997). Larvae were given one of three feed types: (1) newly hatched live brine shrimp Artemia spp. nauplii, (2) a dry feed consisting of 10% sh silage (prepared according to Gallagher 1994) and 90% commercially prepared larval feed, or (3) no feed. Ten larval sh were sampled from each tank at 4, 10, 20, and 27 dph. As suggested by Perez-Dominguez and Dahm (2011), larval status was measured based on the striped bass development stages described by Rogers et al. (1982). The stages are based primarily on yolk sac disappearance; eye, GI, and n development; and swimming and feeding behavior. Six larvae were prepared for transmission electron microscopy (TEM) and scanning electron microscopy (SEM). Whole larvae were xed in 1% formalin and 3% glutaraldehyde. Samples were
High mortality associated with the period of rst feeding to metamorphosis in striped bass Morone saxatilis and striped bass hybrids remains problematic for the culture of these sh despite advances in microencapsulation technologies and culture methods (Langdon and Barrows 2011). Striped bass and hybrids have a short yolk sac period, with larval stages that possess a very rudimentary digestive tract (i.e., lacking a functional stomach or gastric glands) and that must undergo complex metamorphosis of the digestive system to survive. Ohs et al. (1998) and Ohs (1995) suggested that the poor survival of striped bass and white bass Morone chrysops striped bass hybrids (hereafter, hybrid
*E-mail: gallagherm@ecu.edu Received January 4, 2012; accepted February 17, 2012
489
490
GALLAGHER
washed in sodium cacodylate, postxed with 1% osmium tetraoxide, and dehydrated in alcohol. For TEM, specimens were dissected into foregut, midgut, and hindgut sections and embedded in Epon 812 resin. Specimen blocks were trimmed, and silver sections were prepared for examination by TEM using a Phillips CM-10 high-resolution scope. For SEM, whole, xed larvae were mounted on aluminum stubs and were sputter coated with platinum for observation by using an International Scientic Instruments ISI-40 or a Quanta 200 scanning electron microscope. To be noted in this study, ultrastructure details had to occur repeatedly in all larval samples. RESULTS AND DISCUSSION Larval sh that were fed live brine shrimp nauplii developed normally and progressed through the morphological stages ah described by Rogers et al. (1982). The larvae achieved transformation (stage h; differentiation of rays in dorsal and pectoral ns; initiation of lateral line scales) by 27 dph, exhibiting clearly identiable gastric glands with numerous tubular mitochondria surrounded by rough endoplasmic reticulum (RER) and dense secretion granules (Figure 1) similar to those described by Cataldi et al. (2002) for 12-dph Adriatic sturgeon Acipenser naccarii. However, larvae that were starved or that were fed the dry diet did not develop morphologically past stage d (active pelagic feeding, well-developed mouth, and initial n differentiation) despite the fact that food could be seen in the GI tracts of sh that were given the dry feed. Starved larvae began to die at 6 dph, and all had died by 19 dph; larvae that received dry feed did not begin to die until 10 dph, but all had died by 24 dph.
Epithelial Cells of the Epidermis Hybrid striped bass larvae at 4 dph were at stage c (oil globule and yolk nearly absorbed; swimming pelagically) when they arrived at the laboratory. The epidermis, including the mouth, exhibited typical microridge structure as described by Ostrander (2000), and neuromasts were present as noted by Y ufera (2011). Bereiter-Hahn et al. (1979) described the microridge structures found on the surface of sh epidermis cells over 30 years ago and proposed a link between microridges and microvilli of the intestinal mucosa, which is also epithelial tissue that is continuous with the epidermis. The function of microridges remains unclear; certainly, they provide structural support, as the primary components are actin and alpha-actinin. However, microridges also increase surface area and may facilitate the exchange of gases and ions (Rulifson et al. 1986). Sperry and Wassersug (2005) proposed that microridges hold protective mucous secretions to the epidermis, thus providing a barrier between the sh and the surrounding environment. By 10 dph, larvae that received live feed had progressed to stage e (yolk sac absorbed; pectoral ns and teeth [covered with epithelial cells, Figure 2] visible). The epidermis cells were smaller (<10 m) and continued to exhibit typical microridge structures (Figure 3A). There was a continuous double microridge structure at the cell junctions; tight junction complexes and numerous cellular invaginations were present. Larvae that were given dry feed or that were starved had advanced only to stage d, and their cells demonstrated apparent osmotic stress as evidenced by poorly dened microridges and the looseness of the cell membrane (Figure 3B, C). However, the continuous double microridge structure persisted in these sh, even when other microridge structures were scarce or nonexistent (Figure 3B inset). Starved larvae died by day 19. All sh exhibited the formation of neuromasts (Figure 3) along the lateral line.
FIGURE 1. Cross section of gastric glands of a hybrid striped bass larva (at 27 d posthatch) that was fed live brine shrimp nauplii; secretion granules (sg) and microtubule structures (arrows) are apparent. Numerous mitochondria are also present (scale bar = 1 m). [Figure available in color online.]
FIGURE 2. Epithelial cells of the mouth of a hybrid striped bass larva (at 27 d posthatch) that received a diet of live brine shrimp nauplii; teeth can be seen emerging from the covering of epithelial cells with microridges (arrows; scale bar = 10 m).
COMMUNICATION
491
FIGURE 4. Epithelial cells along the lateral line of a hybrid striped bass larva (at 20 d posthatch) that was fed a dry articial diet. Note the persistence of neuromasts (arrow), even when many cells appear to be necrotic (scale bar = 10 m). [Figure available in color online.]
By 20 dph, larvae that were given the live feed had progressed to stage f or g (differentiation of ns, rudimentary spines, and gas bladder present) and the neuromasts had proliferated. Larvae that received dry feed remained at stage d, and numerous epidermal cells exhibited necrosis, although neuromasts persisted (Figure 4). Epithelial Cells of the Gastrointestinal Tract At 10 and 20 dph, the foregut epithelial cells of larvae that were given live feed appeared to be keratinized, with underlying mucous-secreting (goblet) cells (Figure 5A). As was noted by Lazo et al. (2011), mucous cells are typical of even undifferentiated esophageal tissue in many sh species and are thought to serve as a lubricant since sh do not have salivary glands. Although in some species the goblet cells are often found scattered throughout the intestine at the onset of exogenous feeding (Lazo et al. 2011), goblet cells were not observed beyond the foregut in these hybrid striped bass larvae. Instead, the midgut was characterized by cells with sparse microvillus structures and many vacuoles that appeared to be pinocytotic in nature (Figure 5B). The posterior gut was characterized by columnar epithelial cells, with numerous, regularly spaced microvilli, numerous mitochondria, and extensive RER typical of intestinal cells (Figure 5B). Mitochondria in all cells contained tubular rather than lamellar cristae and were similar to mitochondria that are characteristic of endocrine cells (Bozzola and Russell 1992). The mitochondria were surrounded closely by dense RER (Figure 5B inset). In the developing larvae before stomach differentiation, an intestinal-type cell could often be observed immediately posterior to a pinocytotic-type cell as in Figure 5B. A few enteroendocrine cells were also observed in this section of the gut. At 10 dph, larvae that were fed the dry diet had the same basic features of the gut as did larvae that were given live feed. In 20-dph larvae that received the dry diet, the midgut cells
FIGURE 3. Epithelial cells along the lateral line, with neuromasts present, in hybrid striped bass larvae (at 10 d posthatch): (A) a larva that was given live brine shrimp nauplii as feed, (B) a larva that was fed a dry articial diet, and (C) a starved larva. Note the stressed epidermal cells with sparse microridges (arrows; scale bar = 10 m). Inset in (B) depicts a transmission electron micrograph (21,000 ) of the epidermis, showing the double microridge structure at the cellular junction ( ), which persisted even in stressed cells. [Figure available in color online.]
492
GALLAGHER
FIGURE 6. Transmission electron micrographs of gastrointestinal cells in hybrid striped bass larvae (at 20 d posthatch) that received dry articial feed: (A) pinocytotic cells, showing a marked decline in activity, few or no mitochondria, numerous free ribosomes ( ), the presence of keratin-like brils (black arrow), the occurrence of tight junctions between cells, and the persistence of desmosomes (white arrow; scale bar = 1 m); and (B) microvillus cells, exhibiting much fewer mitochondria, rough endoplasmic reticulum that was rounded or nonexistent, and the presence of free ribosomes ( ; scale bar = 1 m).
showed a high degree of keratinization and pinocytotic activity had ceased (Figure 6A). Intestinal-type cells did not show normal RER, and mitochondria were sparse. Ribosomes appeared to be free or associated in spherical clusters (Figure 6B). Conclusions In the limited number of larval sh that were observed, both SEM and TEM revealed several aspects of the epithelial cells of the epidermis and gut during the premetamorphosis period in larvae that were given a live feed or a dry articial diet. Microridges and neuromasts were evident by 4 dph. Microridges of the epidermis were continuous with the gut, as revealed by scanning electron micrographs of the mouth. Three cell types were evident in the gut, including highly keratinized cells in the foregut, pinocytotic cells in the midgut, and intestine-like cells immediately posterior to the pinocytotic cells. Pinocytotic cells, which presumably develop into gastric glands of the
FIGURE 5. Epithelial cells in the intestinal tracts of hybrid striped bass larvae (at 20 d posthatch) that received a diet of live brine shrimp nauplii: (A) foregut, characterized by highly keratinized (arrow) cells with underlying mucous cells (mu; scale bar = 10 m); and (B) midgut, characterized by columnar epithelial cells with centrally located nuclei; the anterior occurrence of pinocytotic cells with sparsely spaced microvilli, extensive tight junctions (black arrow), and desmosomes (white arrows); and (immediately posterior to pinocytotic cells) intestinal absorptive-type cells with slender microvilli (scale bar = 1 m) and large numbers of tubular mitochondria surrounded by extensive rough endoplasmic reticulum (RER). Inset in (B) shows a transmission electron micrograph (6,600 ) of tubular mitochondria (mi) and RER.
COMMUNICATION
493
stomach at metamorphosis, could be seen lying immediately next to intestine-like cells in the premetamorphic larvae. The lack of development in larvae that were fed the dry feed may be due to a lack of digestive enzymes, particularly pepsin, and a lack of GI development, as has been suggested by others (Dabrowski and Culver 1991; Ohs 1995; Langdon and Barrows 2011). In addition, I hypothesize that the lack of development in the hybrid striped bass larvae may have been related to osmotic stress due to the ingestion of dry feed with a high osmotic load or (as expressed by Ohs et al. 1998) a high density. In larvae that received the dry articial diet, both the epidermal and gut epithelial cells exhibited osmotic stress, as indicated by the loss of microridge structure, dissociation of the RER, and eventual cell necrosis. The GI tract is known to be involved in osmoregulation in sh (Allen et al. 2009); however, more studies are needed to conrm this idea. I also observed that in larvae receiving the dry diet, the neuromasts and double microridges at cell junctions persisted even under stressful conditions, thus demonstrating the signicance of these structures for the integrity of larval epithelial cells. REFERENCES
Allen, P. A., J. J. Check Jr., and D. K ultz. 2009. Mechanisms of seawater acclimations in a primitive, anadromous sh, the green sturgeon. Journal of Comparative Physiology Part B 179:903920. Baragi, V., and R. T. Lovell. 1986. Digestive enzyme activities in striped bass from rst feeding through larva development. Transactions of the American Fisheries Society 15:478484. Bereiter-Hahn, J., M. Osborn, K. Weber, and M. Voth. 1979. Filament organization and formation of microridges at the surface of sh epidermis. Journal of Ultrastructure Research 69:316330. Bozzola, J. J., and L. D. Russell. 1992. Electron microscopy. Jones and Bartlett, Boston. Cataldi, E., C. Albano, D. Boglione, L. Dini, G. Monaco, P. Bronzi, and S. Cataudella. 2002. Acipenser naccarii: ne structure of the alimentary canal
with references to its ontogenesis. Journal of Applied Ichthyology 18:329 337. Dabrowski, K., and D. Culver. 1991. The physiology of larval sh. Digestive tract and formulation of starter diets. Aquaculture Magazine 17:4961. Denson, M. R., and T. I. J. Smith. 1997. Tank culture of larval sunshine bass. Progressive Fish-Culturist 59:5963. Gallagher, M. L. 1994. Preliminary evaluation of sh silage as a weaning diet for hybrid striped bass. Bulletin of the Aquaculture Association of Canada 1:2628. Langdon, C., and R. Barrows. 2011. Microparticulate diets: technology. Pages 335351 in G. J. Holt, editor. Larval sh nutrition. Wiley, West Sussex, UK. Lazo, J. P., M. J. Darias, and E. Gisbert. 2011. Ontogeny of the digestive tract. Pages 546 in G. J. Holt, editor. Larval sh nutrition. Wiley, West Sussex, UK. Ohs, C. L. 1995. The development and evaluation of a spray-dried articial diet for larval culture of freshwater prawn (Macrobrachium rosenbergii), hybrid striped bass (Morone saxatilis M. chrysops) and striped bass (M. saxatilis). Masters thesis. Mississippi State University, Mississippi State. Ohs, C. L., L. R. DAbramo, R. K. Buddington, H. R. Robinette, J. M. Roethke. 1998. Evaluation of a spray-dried articial diet for larval culture of freshwater prawn, Macrobrachium rosenbergii, and striped bass, Morone saxatilis. Aquaculture Nutrition 4:7382. Ostrander, G. K. 2000. The laboratory sh. Academic Press, London. Perez-Dominguez, R., and R. Dahm. 2011. Methods for assessing embryonic and larval growth in sh. Pages 373402 in G. J. Holt, editor. Larval sh nutrition. Wiley, West Sussex, UK. Rogers, B. A., D. T. Westin, and S. B. Saila. 1982. Development of techniques and methodology for the laboratory culture of striped bass, Morone saxatilis. U.S. Environmental Protection Agency, Report PB-820217795, Washington, D.C. Rulifson, R. A., J. E. Cooper, and G. Colombo. 1986. Development of feed and starved striped bass (Morone saxatilis) larvae from the Roanoke River, North Carolina. North Carolina Department of Natural Resources and Community Development, Albemarle-Pamlico Estuary System Project 9012, Raleigh. Sperry, D. G., and R. J. Wassersug. 2005. Proposed function for microridges on epithelial cells. Anatomical Record 185:253258. Y ufera, M. 2011. Feeding behavior in larval sh. Pages 285305 in G. J. Holt, editor. Larval sh nutrition. Wiley, West Sussex, UK.

Effect of Copper Sulfate on Aeromonas hydrophila Infection in Channel Catfish Fingerlings

Julie Bebak , Julio C. Garcia & Ahmed Darwish
a a a b
U.S. Department of Agriculture, Agriculture Research Service, Aquatic Animal Health Research Unit, 990 Wire Road, Auburn, Alabama, 36830, USA
b
U.S. Department of Agriculture, Food Safety and Inspection Service, 620 Central Avenue, Building 2C, Alameda, California, 94501, USA Version of record first published: 10 Sep 2012.
To cite this article: Julie Bebak, Julio C. Garcia & Ahmed Darwish (2012): Effect of Copper Sulfate on Aeromonas hydrophila Infection in Channel Catfish Fingerlings, North American Journal of Aquaculture, 74:4, 494-498 To link to this article: http://dx.doi.org/10.1080/15222055.2012.685212
North American Journal of Aquaculture 74:494498, 2012 American Fisheries Society 2012 ISSN: 1522-2055 print / 1548-8454 online DOI: 10.1080/15222055.2012.685212
ARTICLE
Effect of Copper Sulfate on Aeromonas hydrophila Infection in Channel Catsh Fingerlings

Julie Bebak* and Julio C. Garcia
U.S. Department of Agriculture, Agriculture Research Service, Aquatic Animal Health Research Unit, 990 Wire Road, Auburn, Alabama 36830, USA
Ahmed Darwish
U.S. Department of Agriculture, Food Safety and Inspection Service, 620 Central Avenue, Building 2C, Alameda, California 94501, USA
Abstract
Motile Aeromonas septicemia results from primary or secondary infection with bacteria from Aeromonas spp., including Aeromonas hydrophila. Since 2009, an emerging strain of A. hydrophila has been associated, as a primary pathogen, with signicant morbidity and mortality in the U.S. catsh industry. Two 2 2 factorial experiments with ve replicates were conducted to evaluate whether copper sulfate pentahydrate (CuSO4 ) at a concentration of 1% of total alkalinity (total alkalinity = 98 mg/L as CaCO3 ; total hardness = 60 mg/L as CaCO3 ; pH = 7.4) can reduce mortality of channel catsh Ictalurus punctatus after their exposure to this emerging strain of A. hydrophila. In experiment 1, ngerling channel catsh received an 18-h continuous bath exposure to CuSO4 after A. hydrophila challenge. Survival in the treatments challenged with A. hydrophila, both when exposed or unexposed to CuSO4 , was signicantly lower than survival in sham-exposed controls. Fish exposed to A. hydrophila and treated with copper sulfate had the lowest percent survival, at 18% (SE, 7.0), and survival was signicantly different from the treatment in which sh were exposed to A. hydrophila but not treated with copper sulfate. In experiment 2, sh received a 4-h pretreatment with CuSO4 before exposure to A. hydrophila plus a 4-h treatment the next day. In experiment 2, when sh were exposed to A. hydrophila but not CuSO4 , survival was 80.0% (SE, 5.5). For sh exposed to A. hydrophila and to CuSO4 , survival was 50.0% (SE, 3.2). The percent mortality in the treatment exposed to A. hydrophila and to CuSO4 was signcantly different from all of the other treatments. This study demonstrated that, under these experimental conditions, CuSO4 application reduced survival when used as a treatment for infection of ngerling channel catsh with this strain of A. hydrophila.
Motile Aeromonas septicemia (MAS) affects cultured and wild warmwater sh worldwide. The MAS results from infection with bacteria from Aeromonas spp., including Aeromonas hydrophila, which are ubiquitous in the environment. Species that cause MAS typically act as secondary pathogens and can cause signicant losses in the presence of predisposing stressors. However, these bacteria can also be primary pathogens (Plumb 1999). Since 2009, A. hydrophila has been responsible for high rates of mostly acute, but also chronic, morbidity and mortality in the Alabama catsh industry. Losses of tens of millions
of pounds of sh have been reported (Hemstreet 2011). This emerging strain of A. hydrophila can apparently cause significant morbidity and mortality without any predisposing factors. A hemorrhagic septicemia is associated with this infection that includes external and internal hemorrhagic lesions in skin and in viscera (data of L. Khoo and colleagues [paper presented at the 35th Annual Eastern Fish Health Workshop, 2010]). In the USA, there are three antibiotics that may be added to sh feed to treat A. hydrophila infection in catsh Ictalurus spp. (USFWS AADAP 2011). First, oxytetracycline dihydrate
*Corresponding author: jbebakwilliams@gmail.com Received September 30, 2011; accepted April 9, 2012
494
EFFECT OF COPPER SULFATE ON AEROMONAS HYDROPHILA INFECTION
495
(Terramycin 200 for Fish, Phibro Animal Health, Ridgeeld Park, New Jersey) is labeled for use under the previous name for A. hydrophila (Aeromonas liquefaciens). Second, recognizing that U.S. Food and Drug Administration Guidelines (US FDA 2001) must be followed, a veterinarian may choose to use Sulfadimethoxine/Ormetoprim (Romet TC, PHARMAQ AS, Overhalla, Norway) as an extra-label drug. Third, from 2009 through 2011 catsh farmers could participate in the Investigational New Animal Drug Approval process to use orfenicol (Aquaor; Schering-Plough Animal Health Corporation, Summit, New Jersey) medicated feed. A few studies that report the effect of application of copper sulfate pentahydrate (CuSO4 ) as a treatment for external and systemic bacterial infections have been published. Grifn and Mitchell (2007) reported that channel catsh Ictalurus punctatus ngerlings exposed to copper sulfate for 24 h before challenge with Edwardsiella ictaluri experienced signicantly reduced mortality. Darwish et al. (2012) found that catsh ngerlings that received a 4-h CuSO4 treatment given 5.5 h after exposure to Flavobacterium columnare were signicantly more likely to survive the infection. In contrast, Thomas-Jinu and Goodwin (2004) found that applying an indenite bath treatment of CuSO4 starting 20 h after exposure to F. columnare was not effective at reducing mortality in channel catsh ngerlings. The purpose of this study was to evaluate whether CuSO4 has an effect on mortality from exposure to this emerging strain of A. hydrophila. Experiment 1 (Exp 1) was conducted to determine whether an 18-h continuous bath exposure to CuSO4 results in decreased mortality from A. hydrophila infection. Experiment 2 (Exp 2) was conducted to determine whether a 4-h pretreatment with CuSO4 before exposure to A. hydrophila plus a 4-h treatment the next day would reduce mortality from A. hydrophila infection.
METHODS Culture conditions and sh.Well water (total alkalinity = 98 mg/L as CaCO3 [Method 8203, Hach Company, Loveland, Colorado]; total hardness = 60 mg/L as CaCO3 [Method 8213, Hach Company]; pH = 7.4; total Cu <0.02 mg/L) was used for both experiments. Mean tank water volume was 42.2 L (SE, 0.4). When water ows were turned on in the tanks, the mean SE ow rate was 0.11 0.01 L/min. An aquarium heater was placed in each tank to maintain water temperature at 28.5 C. Water temperatures were measured in each tank twice daily. Channel catsh ngerlings (U.S. Department of Agriculture Agricultural Research Service Industry Pool strain) with a mean body weight of 10.6 g (SE, 0.4) and 27.3 g (SE, 1.3) were used for Exp 1 and Exp 2, respectively. Fish were fed a commercial catsh diet (50% protein:16% fat; AquaMax Fingerling Diet, PMI Nutrition International, Brentwood, Missouri) once a day to satiation except for the 24-h period preceding A. hydrophila challenge and when CuSO4 was present in the tanks.
Bacterial culture and challenge conditions.An A. hydrophila isolate (ML-1048-K) cultured from the kidney of a channel catsh that died in 2010 during an outbreak in an Alabama catsh pond was used for the experimental challenges. History, clinical signs, and the API 20E Identication System (Biom erieux, Marcy l Etoile, France) were used to identify this isolate as a member of this virulent strain of A. hydrophila. The isolate, which was stored at 80 C until needed, was grown on sheep blood agar for 24 h at 28 C. Colonies were then used to inoculate 1 L of tryptic soy broth (TSB), which was incubated, shaking vigorously on an automated shaker at 130 rpm, for 24 h at 28 C. To enhance virulence after storage, ve channel catsh ngerlings were injected intracoelomically with 100 L of bacterial broth. The bacterium was reisolated from a dead sh, transferred into TSB and incubated, shaking vigorously, for 24 h at 28 C, until it reached an optical density of about 1.0 ( = 540 nm; Bio-Rad Smart Spec 3000, Hercules, California). The A. hydrophila challenge conditions were the same for Exp 1 and Exp 2. Well water (1 L) was mixed with 100 mL of A. hydrophila broth (optical density = 1.0). Ten channel catsh ngerlings were challenged for 1.5 h in buckets supplemented with vigorous aeration. For the sham-exposed controls, sh were immersed in well water with 100 mL of TSB added. At the end of the challenge, sh were removed from the buckets and returned to their tanks without transferring excess water. Colony forming units (CFU)/mL used for the two experiments were determined from plate counts of 10-fold serial dilutions of the bacteria used to challenge sh. The challenge concentration was 8.2 107 CFU/mL in Exp 1 and was 1.3 106 CFU/mL in Exp 2. Copper sulfate. Copper sulfate (CuSO4 5H2 0; SigmaAldrich, St. Louis, Missouri) was used at a treatment concentration of 1.0% of total alkalinity (Wise et al. 2004), or 0.98 mg/L CuSO4 . Copper (Cu) is 25.4% of the CuSO4 5H2 0 molecule, so this concentration resulted in a targeted Cu concentration of 0.25 mg/L. The CuSO4 was dissolved in distilled water and added to the tanks. Samples for the measurement of the total Cu in the water were collected; the 14.5-mL water sample was preserved with 150 L of concentrated nitric acid. For dissolved Cu (i.e., cupric ion, Cu + 2), the 14.5 mL of sample was ltered through a 0.20-m lter into a 15-mL polypropylene tube and 150 L of concentrated nitric acid was added. Copper concentrations were analyzed at Arkansas Analytical (Little Rock, Arkansas) with a 730-ES Simultaneous ICP-AES Analyzer (Agilent Technologies, Santa Clara, California) following USEPA (1991). The analytical sensitivity of the method was 0.02 ppm Cu. Experimental design.Both experiments were set up as the same 2 2 factorial design, with two levels of exposure to A. hydrophila and two levels of exposure to CuSO4 . The sh in treatment 1 (Trt 1) were not exposed to A. hydrophila or CuSO4 , the sh in treatment 2 (Trt 2) were exposed to A. hydrophila but not CuSO4 , the sh in treatment 3 (Trt 3) were not exposed to A. hydrophila but were exposed to CuSO4 . Treatment 4 (Trt 4) sh were exposed to both A. hydrophila and CuSO4 .
496
BEBAK ET AL. TABLE 1. Mean percent survival of channel catsh in treatments for experiment 1 and experiment 2. Treatment 1 (Trt 1) was not exposed to A. hydrophila or copper sulfate, treatment 2 (Trt 2) was exposed to A. hydrophila but not copper sulfate, treatment 3 (Trt 3) was not exposed to A. hydrophila but was exposed to copper sulfate, and treatment 4 (Trt 4) was exposed to A. hydrophila and copper sulfate.
There were ve replicate tanks for each of the four treatments and a starting density of 10 sh per tank. The sh were acclimated for at least 96 h after stocking into experimental tanks. Tank position was randomly assigned. Each experiment was 14 d in length. Dead sh were removed and counted each day. The sh in suitable condition were necropsied and cultured for bacteria. In Exp 1, at the end of the 1.5-h A. hydrophila challenge, sh were returned to their tanks. Water ow was turned off 1 h later and 0.98 mg/L CuSO4 (i.e., 0.25 mg/L Cu) was added to each tank in Trt 3 and Trt 4. Distilled water was added to each tank in Trt 1 and Trt 2. About 2 min after CuSO4 was added, water samples for Cu analysis were taken from three tanks, then again at 2 h and 18 h later. After 18-h exposure, water was replaced in all the tanks. Samples for Cu analysis were taken after replacing the water. In Exp 2, tanks in Trt 3 and Trt 4 were exposed to a 0.98 mg/L CuSO4 bath treatment for 4 h before they were exposed to A. hydrophila. Only water was added to Trt 1 and Trt 2 tanks. Tanks in all treatments were ushed at the end of the 4-h exposure. Water samples for Cu analysis were taken 2 min after copper was added (Time 0), at the end of the 4-h bath treatment, and then after ushing the tanks. Another 4-h immersion treatment with CuSO4 or distilled water was carried out 17 h after the rst CuSO4 treatment. Water samples for Cu analysis were taken 2 min after CuSO4 was added and then after ushing the tanks. Data Analysis.Regression analysis with arcsine squareroot-transformed proportion survival data was used to determine whether there were statistically signicant responses to treatment for Exp 1 and Exp 2 (StataCorp 2005). Statistical signicance was assigned as P < 0.05. RESULTS Experiment 1 The water temperature averaged 29.0 C (SE, 0.06) throughout the experiment. Kidneys from a total of 58 out of 59 (98.3%) dead sh were cultured for bacteria. Deaths occurred within the rst 24 h in the Aeromonasexposed tanks, both in Trt 2 and Trt 4. All sh that died in Trt 2 or Trt 4 had clinical signs of a bacterial septicemia. There were no deaths in the rst 24 h in Trt 3 tanks; the four sh that died in this treatment were from three tanks and died between 48 and 72 h after the experiment began. During the 14-d experiment, mortality occurred in all 10 of the tanks in the two treatments exposed to A. hydrophila. No mortality occurred in any of the ve Trt 1 (i.e., sham-exposed) tanks, with a resulting mean survival of 100.0% (Table 1). Aeromonas hydrophila was recovered from every dead sh sampled in Trt 2 and Trt 4 but not from sham-challenged sh in Trt 3. Comparison of the treatments in a regression analysis showed that percent survival was signicantly different between treatments (P < 0.001). Treatment 4, which was exposed to A. hydrophila and treated with Cu, had the lowest percent
% Survival (SE) Treatment Trt 1 Trt 2 Trt 3 Trt 4 Exposure Sham control A. hydrophila Copper sulfate A. hydrophila + copper sulfate Experiment 1 100 (0.0) 72.0 (6.0) 92.0 (4.0) 18.0 (7.0) Experiment 2 100 (0.0) 80.0 (5.5) 100 (0.0) 50.0 (3.2)
survival, at 18.0% (SE, 7.0) (Table 1). Survival in this treatment was signicantly different from all other treatments (P < 0.001). When compared to Trt 1, the sham-challenged tanks, survival in Trt 4 and Trt 2 were both signicantly different, at P < 0.001. Survival in Trt 3, not exposed to A. hydrophila but treated with CuSO4 , was not signicantly different (P = 0.08) than Trt 1. After CuSO4 was added to the tanks, mean copper concentrations were 0.264 mg/L (SE, 0.003) and 0.256 mg/L (SE, 0.005) for total Cu and dissolved Cu, respectively (Table 2). Eighteen hours later, mean copper concentrations were 0.183 mg/L (SE, 0.012) and 0.171 mg/L (SE, 0.010) for total Cu and dissolved Cu, respectively. After replacing the water, dissolved Cu concentration was <0.02 mg/L. Experiment 2 The water temperature averaged 28.5 C (SE, 0.06) throughout the experiment. Kidneys from 22 out of 35 (62.9%) dead sh were cultured for bacteria. For both treatments (Trt 1 and Trt 3) in which sh were not exposed to A. hydrophila, there was 100% survival during the 14-d experiment (Table 1). In the treatment where sh
TABLE 2. Copper concentrations (Cu; mg/L) for experiment 1 and experiment 2.
Time Time 0 2h 18 h Postush Trt 1, Time 0 Trt 1, 4 h Trt 1, Postush Trt 2, Time 0 Trt 2, Postush
Mean total Cu (SE) Mean dissolved Cu (SE) Experiment 1 0.264 (0.003) 0.183 (0.012) Experiment 2 0.257 (0.010) 0.256 (0.005) 0.215 (0.008) 0.171 (0.010) <0.02 (0.0) 0.238 (0.008) 0.148 (0.006) <0.02 (0.0) 0.237 (0.007) 0.030 (0.003)
0.249 (0.006)
EFFECT OF COPPER SULFATE ON AEROMONAS HYDROPHILA INFECTION
497
were exposed to A. hydrophila but not CuSO4 , survival was 80.0% (SE, 5.5). In the treatment where sh were exposed to A. hydrophila and to CuSO4 , survival was 50.0% (SE, 3.2). The percent survival in these treatments were both signicantly different than the sham-exposed control (Trt 1) at P < 0.001. The percent mortality in the treatment exposed to A. hydrophila and to CuSO4 was signicantly lower than all of the other treatments (P < 0.001). Aeromonas hydrophila was isolated from every dead sh that was cultured. In Exp 2, just after the CuSO4 treatment was added to the tank, total Cu was 0.257 mg/L (SE, 0.010) and dissolved Cu was 0.238 mg/L (SE, 0.008) (Table 2). After tanks were ushed 4 h later, copper concentrations returned to baseline levels of <0.020 mg/L. The next day, the second treatment resulted in concentrations of 0.249 mg/L (SE, 0.006) and 0.237 mg/L (SE, 0.007) for total Cu and dissolved Cu, respectively. After ushing the tanks, the copper concentration was an average of 0.030 mg/L (SE, 0.003) dissolved Cu.
DISCUSSION Under the conditions used for this study, CuSO4 application at 1% of alkalinity was not benecial when used as a treatment for infection of ngerling channel catsh with the strain of A. hydrophila that was rst isolated in the Alabama catsh industry in 2009. When sh infected with A. hydrophila were treated with an 18-h CuSO4 immersion (Exp 1) mean percent survival was signicantly lower than for those infected with A. hydrophila but not treated with CuSO4 . A 4-h pretreatment with CuSO4 before bacterial challenge plus another 4-h immersion exposure the next day (Exp 2) also was not benecial. In Exp 1, the survival in the treatment that was exposed to CuSO4 but not A. hydrophila was 92%, indicating that there was some toxic effect of CuSO4 , which was controlled for with the experimental design that included ve replicates per treatment. The free Cu concentration stayed below the recommended concentration of 1% of alkalinity to avoid toxic effects. Straus and Tucker (1993) reported a 96-h LC50 of 0.762 mg/L (95% condence interval, 0.600.95) for water at a temperature of 17.0 0.5 C, a total alkalinity of 76 mg/L CaCO3, and 83 mg/L total hardness. The total alkalinity of the water in this experiment was slightly higher, at 90 mg/L CaCO3 , but the total hardness was lower, at 60 mg/L. Also, the mean water temperature was 29.0 C, 12 C higher than Straus and Tuckers (1993) conditions. These differences in total hardness and water temperature may have contributed to the slightly increased toxicity seen in Exp 1. In Exp 1, the mean survival in the treatment exposed to both A. hydrophila and CuSO4 was the lowest (18.0%) of the four treatments, indicating an enhanced toxic effect of CuSO4 application when A. hydrophila is also present. This survival was signicantly lower (P < 0.001) than the mean survival in the treatment exposed to A. hydrophila but not exposed to CuSO4 . In Exp 2, the same pattern was observed; the lowest
mean survival (50.0%) occurred in sh exposed to both A. hydrophila and CuSO4 . In addition, mean percent survival was signicantly different in this treatment compared to the treatment exposed to A. hydrophila only (80.0%) at a P < 0.001. The results in both experiments suggest that CuSO4 may increase mortality when used as a treatment for A. hydrophila infection of ngerling channel catsh. These results contrast with Darwish et al. (in press) who did nd that two applications of CuSO4 at 1% of alkalinity after exposure to a mixed infection of A. hydrophila and F. columnare resulted in signicantly increased survival of sunshine bass (female white bass Morone chrysops male striped bass M. saxatilis). In this case, the A. hydrophila isolate used was from an outbreak in sunshine bass. The pathogenesis of the bacterial infection may contribute to the toxicity of the copper. In the present study, and with this virulent strain of A. hydrophila, the severe bacterial septicemia that occurs as a result of infection with this isolate may make the sh more susceptible to the toxic effects of the copper. Our results are more consistent with previous reports regarding the detrimental effects of exposure of sh to copper sulfate. In salmonids, for example, Knittel (1981) and Baker et al. (1983) also reported that copper may increase susceptibility to mortality from bacterial infections. R abago-Castro et al. (2006) found that at concentrations commonly used in aquaculture, prophylactic, intermittent exposure of healthy channel catsh to CuSO4 caused growth suppression, which has also been shown in salmonids (Kamunde et al. 2002). This work does not support the use of CuSO4 as a treatment for this emerging strain of A. hydrophila and indeed suggests that its use will increase mortality. The diversity of effects that vary according to sh species and pathogen demonstrate that much work needs to be done to determine the circumstances, if any, under which CuSO4 may have useful benecial effects. ACKNOWLEDGMENTS This work was supported by the U.S. Department of Agriculture, Agricultural Research Service, Current Research Information Systems Project Number 6420-32000-024-00D. This research was conducted in compliance with all relevant federal guidelines and institutional policies. The mention of trade names or commercial products in this publication is solely for the purpose of providing specic information and does not imply recommendation or endorsement by the U.S. Department of Agriculture.
REFERENCES
Baker, R. J., M. D. Knittel, and J. L. Fryer. 1983. Susceptibility of Chinook salmon, Oncorhynchus tshawytscha (Walbaum), and rainbow trout, Salmo gairdneri Richardson, to infection with Vibrio anguillarum following sublethal copper exposure. Journal of Fish Diseases 6:267275. Darwish, A. M., J. Bebak, and K. K. Schrader. In press. Preliminary assessment of Aquaor, copper sulfate and potassium permanganate for control of
498
BEBAK ET AL. sulfate on growth, condition and feeding indices in channel catsh (Ictalurus puntatus). Aquaculture 253:343349. StataCorp. 2005. Stata statistical software: release 9. StataCorp, College Station, Texas. Straus, D. L., and C. S. Tucker. 1993. Acute toxicity of copper sulfate and chelated copper to channel catsh Ictalurus punctatus. Journal of the World Aquaculture Society 24:390395. Thomas-Jinu, S., and A. E. Goodwin. 2004. Acute columnaris infection in channel catsh, Ictalurus punctatus (Ranesque): efcacy of practical treatments for warmwater aquaculture ponds. Journal of Fish Diseases 27: 2328. US FDA (U.S. Food and Drug Administration). 2001. Compliance policy guide section 615.115. Extra-label use of medicated feeds for minor species. US FDA, Washington, D.C. USEPA (U.S. Environmental Protection Agency). 1991. Methods for chemical analysis of water and wastewater, method 200.7, revision 4.4. USEPA, Cincinnati, Ohio. USFWS AADAP (U.S. Fish and Wildlife Service Aquatic Animal Drug Approval Partnership). 2011. Approved drugs for use in aquaculture, 1st edition. USFWS AADAP, Bozeman, Montana. Wise, D. J., A. C. Camus, T. E. Schwedler, and J. S. Terhune. 2004. Health management. Pages 444502 in C. S. Tucker and J. A. Hargreaves, editors. Biology and culture of channel catsh. Elsevier, New York.
Aeromonas hydrophila and Flavobacterium columnare infection in sunshine bass, Morone chrysops female Morone saxatilis male. Journal of Fish Diseases. DOI: 10.1111/j.1365-2761.2012.01393.x. Darwish, A. M., D. Mitchell, and D. L. Straus. 2012. Evaluation of a 4-h static copper sulphate treatment against experimental infection of Flavobacterium columnare in channel catsh (Ictalurus punctatus). Journal of Aquaculture Research 43:688695. Grifn, B. R., and A. J. Mitchell. 2007. Susceptibility of channel catsh, Ictalurus punctatus (Ranesque), to Edwardsiella ictaluri challenge following copper sulfate exposure. Journal of Fish Diseases 30:581585. Hemstreet, B. 2011. Aeromonas status report2010. Fish Farming News (winter):2. Kamunde, C., M. Grosell, D. Higgs, and C. M. Wood. 2002. Copper metabolism in actively growing rainbow trout (Oncorhynchus mykiss): interactions between dietary and waterborne copper uptake. Journal of Experimental Biology 205:279290. Knittel, M. D. 1981. Susceptibility of steelhead trout Salmo gairdneri Richardson to redmouth infection Yersinia ruckeri following exposure to copper. Journal of Fish Diseases 4:3340. Plumb, J. A. 1999. Health maintenance and principal microbial diseases of cultured shes. Iowa State University Press, Ames. R abago-Castro, J. L., J. G. Sanchez, R. P erez-Casta neda, and A. Gonz alezGonz alez. 2006. Effects of the prophylactic use of Romet-30 and copper

Embryonic and Early Larval Development in HatcheryReared Common Snook

Carlos Yanes-Roca Kevan L. Main
a b c b a b
, Nicole R. Rhody , Michael Nystrom , Matthew L. Wittenrich &
Institute of Aquaculture, Stirling University, Stirling, Scotland, FK9 4LN, UK Mote Marine Laboratory, 1600 Ken Thompson Parkway, Sarasota, Florida, 34236, USA
Florida Institute of Technology, Department of Biological Sciences, 150 West University Boulevard, Melbourne, Florida, 32901, USA Version of record first published: 10 Sep 2012.
To cite this article: Carlos Yanes-Roca, Nicole R. Rhody, Michael Nystrom, Matthew L. Wittenrich & Kevan L. Main (2012): Embryonic and Early Larval Development in Hatchery-Reared Common Snook, North American Journal of Aquaculture, 74:4, 499-511 To link to this article: http://dx.doi.org/10.1080/15222055.2012.676013
ARTICLE
Embryonic and Early Larval Development in Hatchery-Reared Common Snook

Carlos Yanes-Roca*
Institute of Aquaculture, Stirling University, Stirling, Scotland FK9 4LN, UK; and Mote Marine Laboratory, 1600 Ken Thompson Parkway, Sarasota, Florida 34236, USA
Nicole R. Rhody and Michael Nystrom

Mote Marine Laboratory, 1600 Ken Thompson Parkway, Sarasota, Florida 34236, USA
Matthew L. Wittenrich
Florida Institute of Technology, Department of Biological Sciences, 150 West University Boulevard, Melbourne, Florida 32901, USA
Kevan L. Main
Mote Marine Laboratory, 1600 Ken Thompson Parkway, Sarasota, Florida 34236, USA
Abstract
To gain an improved understanding of the early life history of common snook Centropomus undecimalis and rene hatchery production techniques for this species, a combination of digital photography and histological techniques were used to document the embryonic and early larval development of hatchery-reared individuals. Embryo development from fertilization to hatching took 15 h at 28 C. Larvae at 2 d posthatch showed fully pigmented eyes, and histological sections of the digestive tract revealed the presence of cellular structures indicative of a functional gut. This suggests that common snook larvae have the mechanical ability to detect, capture, and digest prey at 2 d posthatch.
The common snook Centropomus undecimalis is a diadromous, stenothermic, euryhaline, estuarine-dependent species found in the tropical and subtropical western Atlantic Ocean and Gulf of Mexico from about 34 N to about 25 S latitude (Howells et al. 1990). Snook Centropomus spp. are protandric hermaphrodites: some males develop into females between 1 and 7 years of age, having a maximum 20-year lifespan. The spawning of common snook has been studied for the last 55 years, but despite the importance of common snook as a popular game sh, the description of its reproductive biology is incomplete. Common snook in Florida have a daily spawning cycle in which spawning episodes occur during the late afternoon and the early evening hours during the lunar phases and during all tidal stages (Taylor et al. 1998).
*Corresponding author: cyanes@mote.org Received November 10, 2011; accepted February 14, 2012
The identication of critical embryonic and larval stages, such as eye and gut formation, rst feeding, and swim bladder ination, is essential for a better understanding of any sh species. This understanding may help to improve common snook larval rearing techniques and therefore increase larval survival rates. Although less developed than in adults, the larval digestive tract is functional when feeding is initiated (Govoni et al. 1986). Additionally, the digestive tract develops as the larvae grow. This facilitates changes in the rates of ingestion, digestion, and the assimilation efciency inuence resulting in improved larval growth (Sarasquete et al. 1995). The development of the digestive system tract, as well as the possible abnormalities and deciencies that result from the absence or inadequacy of food, has been studied in several teleosts (Cousin and Baudin-Laurencin
499
500
YANES-ROCA ET AL.
1985; Avila and Juario 1987; Eckman 1987; Ferrais et al. 1987; Deplano et al. 1991; Boulhic and Gabaudan 1992) and is dened by the development of key digestive system structures. Eye development for most sh species is critical for their survival, especially once exogenous feeding starts (Mani-Ponset et al. 1996), mainly due to their visual feeding nature (Blaxter 1986; Batty 1987). Prey capture, orientation, schooling, and eluding predators are other basic activities that rely on vision (Paul 1983; Blaxter 1986; Porter and Theilacker 1999). The role of vision in feeding has been investigated using varied light intensities. Signicant differences in the consumption of prey were found among a variety of sh species including cisco Coregonus artedi (Jonh and Hasler 1956), sole Solea solea, and plaice Pleuronectes platessa (Blaxter 1968), with most variations in foraging corresponding to periods of dusk and dawn. The decrease in rate of feeding corresponds with dusk and dawn periods. Some species can feed in the dark, e.g., cisco (Jonh and Hasler 1956), especially when food is present in high concentrations. For example, sole feed in the dark for most of their larval life and others, like plaice, only at later stages around metamorphosis (Blaxter 1968). Kawamura (1984), Pankhurst (1996), and Roo et al. (1999) reported major changes in the visual system in the lecithotrophic phase as preparation for onset feeding on sparids. At the same time, Roo et al. (1999) studied the relationship between gut and eye development in red porgy Pagrus pagrus, demonstrating that visual capability was developed before the onset of exogenous feeding. No common snook eggs, embryos, or early larvae have been documented from wild collections. This is a void in information on the life history and early feeding habits that generally forms the building blocks of aquaculture protocols. The objective of this study was to describe the physiological, embryonic, and larval developmental features of the common snook. METHODS All the samples were collected at the Mote Aquaculture Research Park located in Sarasota, Florida, from May to August 2007. The eggs were stripped from wild females from the eld and fertilized with milt from wild males collected at the same time using a 92-m seine net deployed from a research vessel. Once fertilized, the eggs were transported to the main facilities and stocked in the rearing tanks. Egg sampling procedures.One hour after fertilization, common snook embryos were stocked in a 50-L tank at a constant temperature of 28 C and salinity of 35. No aeration was used, dissolved oxygen was constant at 8 mg/L. The eggs were kept in the dark, simulating night time light characteristics. A sample of 10 embryos was collected every hour until hatching occurred. Once collected, samples were placed under a compound light microscope (Olympus 3500 ) with a dark eld. The egg diameter was measured and embryonic development was observed and documented using an attached 35 mm camera
(Nikon 500). This sampling was repeated three times during three separate spawning events over the course of three nights. Larval sampling procedures.Larvae were sampled from a 3,300-L production tank, where water temperature was maintained at 28 C, salinity at 35 and dissolved oxygen at 10 mg/L. Five larvae were randomly collected daily from day 0 to day 3 and every two days from day 4 to day 14 after hatching. Once collected larvae were placed under a light microscope (Olympus 4000), which had a digital camera mounted (Sony 600), larval pictures were taken under dark eld conditions. Standard length (SL) and myomere height from the specimens were recorded using a calibrated microscope reticule. This sampling regime was repeated for seven different spawning events, collecting a total of 280 larvae, between day 0 and day 14. Another 10 larvae were collected from day 1 to day 3 after hatching and xed for transmission electron microscope (TEM) and scanning electron microscope (SEM) work. Electron microscopy specimen xation and preparation. The larvae collected for the TEM were xed in Karnovskys xative at 4 C for 2 hours and placed in cacodylate buffer rinse (pH 7.5) at 4 C until further processing. The tissues were then transferred to the Electron Microscopy Laboratory, Institute of Aquaculture, at the University of Stirling, Stirling, Scotland, where nal processing was completed. Once at the processing unit, the specimens were processed as described in Table 1. Block sectioning was carried out using a glass knife for semithin sections (0.5 m) to be observed under the light microscope. For the TEM, block sectioning was carried out with a diamond knife for ultrathin sections (50 nm). Ultrathin sections were mounted on copper Formavar carbon-coated grids and metal stained with uranyl acetate and lead citrate and stained with a metal stain; this step was carried out following a metal stain protocol described below. A drop of saturated uranyle acetate was placed on a piece of dental wax and the grid was oated over the drop for 30 min in the dark. The grid was rinsed with 50% alcohol and then with distilled water and oated on a drop of lead citrate for 25 min. The grid was thoroughly rinsed, dried on lter paper, and immediately removed to avoid dust accumulation on the grid. Sectioned samples were viewed and
TABLE 1. Mean SD standard length (mm) and myomere height (mm) of common snook during the rst 14 d after hatching (n = 35).
Day 0 2 4 6 8 10 12 14
Standard length 1.785 0.289 2.311 0.434 2.266 0.360 2.520 0.515 2.556 0.454 3.137 0.495 3.580 0.454 4.433 0.780
Myomere height 0.165 0.015 0.211 0.024 0.211 0.059 0.246 0.036 0.277 0.040 0.368 0.047 0.401 0.055 0.549 0.025
EMBRYONIC AND EARLY LARVAL DEVELOPMENT
501
photographed at the Institute of Aquaculture on the TEM (FEI Tecnai E2 Biotwin). All the handling during the course of this project followed the animal welfare rules and regulations from the United States of America Animal Welfare Institute (Washington, D.C.). RESULTS Embryonic Development The development of common snook embryos at 28 C took 15 h from fertilization to hatching. The fertilized eggs were spherical, with a homogenous yolk, a smooth chorion, and a single oil droplet. Egg diameters ranged from 0.65 mm to 0.72 mm (mean = 0.69 mm, n = 300) and the oil droplet diameter ranged from 0.15 to 0.30 mm (mean = 0.26 mm, n = 300). Fertilization occurred in the eld; therefore, no embryonic development was documented prior to the blastodisc stage (rst hour after fertilization). We estimated that the blastoderm separated from the yolk approximately 30 min after fertilization. Shortly after that, the blastoderm was then divided in two blastomeres. The following is a description of development at approximate 1-h intervals: One hour and thirteen minutes after fertilization, the second segmentation or four-cell stage occurred. The second cleavage furrow developed on two blastomeres at a right angle to the rst cleavage plane. It deepened until each blastomere divided into two cells of the same size. The oil droplet was larger and gathered toward the vegetal pole (Figure 1, item a). One hour and thirty-two minutes after fertilization another division occurred. This time, the blastoderm divided in 16 blastomeres (Figure 1, item b), also called the 16-cell stage, where the fourth cleavage plane, which parallels the second, divides the two rows of four blastomeres into four rows of four blastomeres (Figure 1, item b). Two hours after fertilization, another division occurred and the blastoderm divided into 32 blastomeres (Figure 1, item c), the 32-cell stage, where the fth cleavage plane divided the marginal 12 blastomeres meridionally into 24, and the central four blastomeres horizontally into eight thereby forming two layers, an outer and an inner layer, in the central region. The number of marginal cells is 14 (Figure 1, item c). Three hours after fertilization, the blastomeres were still distinct, but the rapid division was too advanced to observe the number of blastomeres; this is the late morula stage (Figure 1, item d). At this time, the planes of the sixth and the later cleavages were difcult to precisely trace. The blastomeres (256512) had different cleavage planes depending on their positions within the dome-shaped blastoderm and were arranged in three layers. The peripheral blastoderms (2124) were attened in shape. The cells were arranged in three to four layers but still easily dissociated from each other (Figure 1, item d). Four hours and ten minutes after fertilization, the blastomeres were no longer distinguishable and the blastocoel, or segmentation cavity, began to form. This was the blastula stage (Figure 1,
FIGURE 1. Embryonic development of common snook from fertilization to hatch (all times are postfertization): a = 1 h, 13 min; b = 1 h, 32 min; c = 2 h; d = 3 h; e = 4 h, 10 min; f and g = 5 h, 12 min; h = 6 h, 12 min; i = 7 h, 15 min; j = 8 h, 10 min; k = 9 h, 21 min; l = 10 h, 12 min; m = 11 h, 12 min; n = 12 h, 30 min; o = 13 h, 5 min; p = 14 h, 22 min. [Figure available in color online.]
item e). Projection of the underside of the blastoderm (central cells) into the yolk sphere was observed. In this stage, some blastomeres began to cleave asynchronously and to migrate. Several rows of periblast nuclei were visible around the blastoderm (Figure 1, item e). Five hours and twelve minutes after fertilization, the blastoderm covered more than half the yolk (Figure 1, item f) and the blastocoel were completely formed (Figure 1, item f). This was the mid-gastrula stage, where a streak was visible in the midline of the embryonic shield projecting into the germ ring area (Figure 1, item g). Six hours and twelve minutes post fertilization, the blastoderm covered three-fourths of the yolk sphere (Figure 1, item h) and the embryonic shield became more clearly visible as a narrow streak. The enveloping layer extends uniformly over the yolk sphere through this stage (Figure 1, item h). Seven hours and fteen minutes after fertilization, the blastoderm covered more than four-fths of the yolk surface, leaving a small area around the vegetal pole exposed (Figure 1, item i). This was the neurula stage, and the head could be clearly recognized anteriorly in the distinct embryonic body. A beak-like mass of cells was seen in front of the head, the embryo completes an arc of 150 . The brain and nerve cord in the arrow-shaped embryonic body developed as a solid rod of cells. A solid optic bud appeared on each side of the cephalic end. The beak-like cell mass was still visible and the blastopore was closed (Figure 1, item i).
502
YANES-ROCA ET AL.
Eight hours and ten minutes after fertilization, melanophores were visible on the embryo for rst time (Figure 1, item j) and the optic vesicles were clearly outlined. This is the somite stage, where a pair of otic (auditory) vesicles appeared at the posterior region of the head. Depressions began to form at the dorsal surface of the eye vesicles (Figure 1, item j). Nine hours and twenty-one minutes after fertilization, the embryo was strongly pigmented (Figure 1, item k), especially dorsolaterally. Melanophores on the yolk concentrated on the ventral surface on each side of the embryo. Some melanophores were also apparent on the oil droplet (Figure 1, item k). Ten hours and twelve minutes after fertilization (Figure 1, item l), the optic vesicles differentiated to form the optic cup and the lenses began to form. The small otic vesicles appeared, but they lacked otoliths. The regions of the brain were well dened, and the neural fold was seen as a median line along the body (Figure 1, item l). Eleven hours and twelve minutes after fertilization (Figure 1, item m), the tubular heart appeared under the head from the posterior end of the midbrain to the anterior end to the hindbrain. The body cavity extended further toward the posterior end of the eye vesicles. The melanophores on the oil droplet were larger and more distinct; those on the head expanded to outline the olfactory lobes and optic vesicles. The tail separated from the yolk (Figure 1, item m). Twelve hours and thirty minutes after fertilization (Figure 1, item n), the melanophores were much larger and fewer, forming aggregations. The anterior portion of the heart, which exhibited a slow pulsation, extended upward to the anterior end of the forebrain. Cuverian ducts and the vitello-caudal vein were still incomplete. The embryonic body encircles nearly three-fourths of the yolk sphere (Figure 1, item n). Thirteen hours and ve minutes after fertilization (Figure 1, item o), the blood began to circulate and the heart beat was faster and more constant. At the same time, the embryo started moving with quick, short movements mainly produced by the tail. Melanophores were more compact and they accumulated in four places along the embryos body (Figure 1, item o). Fourteen hours and twenty-two minutes after fertilization, the rst larvae hatched (Figure 1, item p). At this time, the melanophores accumulations could be observed clearly. Five groups spread along the newly fresh larvae. There were two on the head region, one over the olfactory region, and another around the eyes. The other two were situated in the midlateral and dorsolateral area, and another one in the postanal area (Figure 1, item p). Fifteen hours after fertilization massive hatching took place; 90% of the eggs hatched at that time. Larval Development Common snook larvae hatched 15 h after fertilization at 28 C (Figure 1, item p), measuring 1.71 mm in mean SL, ranging from 1.38 to 1.84 mm (n = 35). The body of newly hatched yolk sac larvae was elongated with an oval shape and a mean size of 0.91 mm in length, ranging from 0.85 to 1.02 mm (n = 300), and
with a mean width of 0.58, ranging between 0.41 and 0.72 mm. A single oil droplet was present, located on the yolk sacs front side under the head. A transparent voluminous n fold covered most of the body (Figure 1, item p). The eyes and mouth were not formed, and no eye pigmentation was present. The larvae were concentrated on the surface, oating and moving mainly in circles due to their limited n development. Yolk Sac Stage On day 0 (24 h after fertilization), common snook larvae (Figures 2, 3, and 4, item a) measured 2.25 mm (mean SL), ranging from 1.90 to 2.45 mm (n = 35), and had a mean myomere height (MH ) of 0.164 mm, ranging from 0.152 to 0.171 mm (n = 35). The yolk sac was reduced in size with a mean length of 0.51 mm, ranging from 0.31 to 0.63 mm, and a mean width of 0.35 mm, ranging from 0.29 to 0.45 mm. Eyes were starting to form, and some pigmentation was observed. The mouth was not formed, although some denition was observed. The pectoral ns were starting to develop. In terms of larval behavior, larvae were mainly at the surface, although some larvae were distributed in the water column but close to the surface. Swimming movements were more directional. Day 1 larvae (Figure 3) had a mean SL of 2.07 mm, ranging from 1.92 to 2.17 mm, and an average MH of 0.202 mm, ranging from 0.180 to 2.071 mm (n = 35). The yolk sac, although reduced to half the size from the previous day, was still present with a mean length of 0.16 mm, ranging from 0.12 to 0.20 mm. The eyes were starting to gain denition and the retina can be now differentiated with eye pigmentation observed but not fully completed. The optic tectum or primary optic center of the brain is large (Figure 5A) as in most of sh, which rely on their sense of vision. Lenses were fully complete and visible (Figure 5A) and the internal layer organization is well established, although layer thickness is still low. The cornea and the cartilaginous ring were present, as well as the optic chiasma right below the infundibulum. The retina layer organization can be observed (Figure 5B) and the main layers were visible, such as the outer limiting layer and the outer nuclear layer, where the columnar nuclear bodies were present although undeveloped. Also, the outer plexiform layer and inner nuclear layer can be observed, yet most of layers were not fully developed with some main organelles missing (Figure 5C), where the outer limiting layer and the outer nuclear layer were still lacking organelles denition. Some pigments in the epithelium were present, although still low in levels (Figure 5D). The alimentary canal of 1-d-old larvae shows some differentiation along its length. Cilia, which help to circulate the contents, can be observed in the lumen (Figure 6BC), and at the same time some irregular small microvilli appear along the digestive wall. Other organelles observed include the mitochondrion and nucleus of the epithelial cells, although these organelles were present in low numbers. Other structures, such as the nonvillous region and the terminal web, were clearly dened although they were lacking in thickness (Figure 6C).
503
FIGURE 2.
Common snook larval growth during the rst 14 d after hatching.
On day 2 (Figures 3 and 4, item b), SL has increased to a mean 2.31 mm, ranging from 2.12 to 2.71 mm, and with an average MH of 0.211 mm, ranging from 0.185 to 0.282 mm (n = 35). The yolk sac has been nearly absorbed reducing its size to a mean length of 0.15 mm, ranging from 0.11 to 0.20 mm. The oil droplet was still present although reduced in size (Figures 3 and 4, item b). Eyes were formed and pigmented (Figure 5EF) and the cornea has gained in thickness and was tight against the lens. The retina layer was more dened and each layer was thicker. The optical nerve was fully formed and connected to the main nervous system (Figure 5EJ). For the rst time, the clear layer of pigment epithelium cells was present and the other layers, such as the outer nuclear layer, were gaining in complexity (Figure 5FG) with the development of organelles such as the cones and rods (Figure 5H) nearly completed (Figure 5I). The mouth was formed and opened (Figures 3 and 4, item b) with the main cartilages, such as the Meckels car-
tilage, the hyposynplectic cartilage, and the basihyal cartilage, already present (Figure 5E). Also the tongue can be observed (Figures 4, item b, and 5E). The digestive system was straight and long, extending past the posterior margin of the yolk sac and into the ventral n fold, and although undeveloped, it has some food inside. The digestive system walls were dened, no cilia was observed in the lumen, and the microvilli layer was now well established along the walls (Figure 6D). Also the microvilli were long and compact (Figure 6EF). The epithelium cell structure was forming and the main organelles were observed, such as the mitochondrion, the clear and dark apical cells, and the epithelium nucleus (Figure 6EF); however, organelle numbers were low. Pectoral ns were well developed and functional. Slight caudal n denition could be observed. Larvae show a photopositive reaction gathering at those areas with more light. The distribution of larvae was more diverse, having larvae distributed in the water column and at
504
YANES-ROCA ET AL.
FIGURE 3. Common snook larval development from day 0 to day 14 after hatching. Length is shown in millimeters.
505
FIGURE 4. Common snook organ development from day 0 to day 14 after hatching: a = day 0, b = day 2, c = day 3, d = day 4, e = day 6, f = day 8, g = day 10, h = day 12, j = day 14. Abbreviations are as follows: A = anus, ADF = anterior dorsal n, AF = anal n, AMI = anterior median intestine, C = ceratohyal, CF = caudal n, CM = cerebellum, CN = constriction, E = eye, FF = n fold, H = heart, HS = hyposynplectic cartilage, IN = intestine, L = liver, MC = Merkels cartilage, MO = medula oblongata, N = notochord, OD = oil droplet, OL = optic lobe of brain, OT = optic tectum, P = pigments, PH = posterior part of hind, PM = premaxilla, R = rectal area, SB = swim bladder, T = teeth, V = intestine-recto valve, and YS = yolk sac. [Figure available in color online.]
the surface. Burst movements towards prey were commonly observed. On day 3 (Figures 3 and 4, item c), snook larvae had a mean SL of 2.26 mm, ranging from 2.15 to 3.1 mm, and a mean MH of 0.214 mm, ranging from 0.20 to 0.35 mm (n = 35). The yolk sac was almost completely absorbed and the oil droplet was still present, although severely reduced. The eyes had gained pigmentation and were developed. The retina layer was well structured due to the development of most of the layer organelles with the cornea fully formed (Figure 5KL). All the different layers were clearly differentiated, such as the outer ganglion layer, the inner nuclear layer and the inner ganglion layer (Figures 5L, 6A). The pigment epithelium cell layer was fully formed with photoreceptor inclusions (Figure 6A),
the bodies of the photoreceptors were also now well dened (Figures 5L, 6A). The maxillar jaws were developing and mouth gap was increasing. The alimentary canal was no longer a long straight tube; some structure could be observed especially in the anal region, which, by this stage, was well developed. Structural epithelium organelles, such as the nucleus, mitochondrion, and dark vesicles, were all present and in high numbers (Figure 6H I). The microvilli layer had gained length and was more compact (Figure 6H). Other important organelles were also present, such as the Golgi apparatus, rough endoplasmatic reticulum, desmosome of apical junction and pinocytotic vesicle (Figure 6I). Food was observed in the gut, mainly rotifers and algae. The swim bladder was formed and showing signs of ination. Swimming speed had increased, as well as larvae motility.
506
YANES-ROCA ET AL.
FIGURE 5. Optical SEM and TEM cross sections of common snook larva from day 1 to day 3 posthatch (dph): (A) semithin cross section from the head of a 1-dph common snook larva (scale bar = 100 m; B = buccal cavity, C = cornea, CR = cartilaginous ring, F = infundibulum, INL = inner nuclear layer, L = lens, OC = optic chiasma, and OT = optic tectum); (B) cross section of the eye of a common snook larva at 1 dph (scale bar = 15 m; CNB = columnar nuclear bodies [nuclei of photoreceptors], OLM = outer limiting membrane, ONL = outer nuclear layer, and OPL = outer plexiform layer); (C) cross section of the eye of a common snook larva at 1 dph (scale bar = 10 m); (D) cross section of the eye of a common snook larva at 1 dph (scale bar = 5 m; PE = pigment epithelium); (E) semithin cross section from the head of a 2-dph common snook larva (scale bar = 90 m; BC = basihyal cartilage, M = Meckels cartilage, ON = optic nerve, and T = tongue); (F) semithin cross section from the head of a 2-dph common snook larva (scale bar = 40 m); (G) cross section of the a eye of common snook larva at 2 dph (scale bar = 5 m); (H) electron micrograph of a cross section of the eye of a common snook larva at 2 dph (scale bar = 2 m; C = cones, PEC = pigment epithelium cell granule, and R = rods); (I) cross section of the eye of a common snook larva at 2 dph (scale bar = 1 m; OD = oil droplet, PRES = photoreceptor outer segment); (J) cross section of the eye of a common snook larva at 2 dph (scale bar = 10 m); (K) semithin cross section from the head of a 3-dph common snook larva (scale bar = 100 m); and (L) cross section of the eye of a common snook larva at 3 dph (scale bar = 5 m; B = bodies of photoreceptors and IGL = inner ganglion layer). [Figure available in color online.]
507
FIGURE 6. Optical SEM and TEM cross sections of the eye and digestive system of common snook larva from 1 to 3 dph (see Figure 5 for abbreviations not dened here): (A) cross section of the eye of a common snook larva at 3 dph (scale = 10 m); (B) cross section of the antero-medial intestine of a common snook at 1 dph (scale bar = 2 m; C = cilia, L = lumen, M = mitochondrion, MV = microvilli, and N = nucleus); (C) cross section of the rectal area of a common snook at 1 dph (scale bar = 2 m; NVR = nonvillous region and TW = terminal web); (D) semithin cross section of the antero-medial intestine of a common snook larva at 2 dph (scale bar = 70 m; CA = clear apical cell in epithelium, DA = dark apical cell in epithelial, DSW = digestive system walls, and N = nucleus of columnar cell); (E) cross section of the antero-medial intestine of a common snook larva at 2 dph (scale bar = 5 m; LV = light vesicle); (F) cross section of the rectal area of a common snook larva at 2 dph (scale bar = 1 m); (G) cross section of the antero-medial intestine of a common snook larva at 3 dph (scale bar = 100 m; DD = dark droplet, F = food particles, and N = nucleus of enterocyte); (H) cross section of the antero-medial intestine of a common snook larva at 3 dph (scale bar = 5 m); (I) cross section of the rectal area of a common snook at 3 dph (scale bar = 2 m; D = desmosome of the apical junction, GA = Golgi apparatus, and RER = rough endoplasmic reticulum); and (J) cross section of the antero-medial intestine of a common snook larva at 3 dph (scale bar = 1 m). [Figure available in color online.]
508
YANES-ROCA ET AL.
Preexion Stage Day 4 larvae (Figures 3 and 4, item d) have a mean SL of 2.35 mm, ranging from 2.22 to 3.23 mm, and a mean MH of 0.211 mm, ranging from 0.20 to 0.28 mm (n = 35). By day 4, the yolk sac has been totally absorbed and the oil droplet was not present anymore (Figure 3 and 4, item d). The medulla oblongata can be observed as well as the cerebellum, which was fully formed. The eyes were fully pigmented, the mouth was fully functional with more jaw development, and teeth formation can also be observed. Swim bladder, positioned posterior to the pectoral n base and above the stomach, was developed and inated. The gut gained in thickness and became partitioned. Rotifers were observed in the gut. At this stage larvae were scattered throughout the water column, although no larvae could be seen on the tank bottom. There were many active swimmers, which spent most of the time seeking prey. On day 6 (Figures 3 and 4, item e), the mean SL of snook larvae was 2.51 mm, ranging from 2.1 to 3.41 mm, with a mean MH of 0.245 mm, ranging from 0.214 to 0.351 mm (n = 35). Rows of teeth were present at the same time that mouth gap had increased and the gas bladder was inated. The gut was well partitioned and food was commonly observed inside with 90% of larvae observed having full stomachs. The n fold around the larvae was no longer present and the ns were starting to take shape, especially the caudal and dorsal ns, and the pectoral ns were fully functional but still developing. At day 8, snook larvae (Figures 3 and 4, item f) had a mean SL of 2.55 mm, ranging from 2.41 to 3.7 mm, with a mean MH of 0.276 mm, ranging from 0.21 to 0.34 mm (n = 35). At this stage, the snook larvae have increased body depth, and increased head size in relation to eye size. The notochord was fully formed and ends at the caudal n. There was an increase in volume of the digestive system, and denition of the different organs was more apparent. The caudal n started developing into the shape of an adult caudal n, going from a more rounded initial shape towards a forked n shape. Flexion Stage Ten days after hatching (Figures 3 and 4, item g) common snook larvae have a mean SL of 3.136 mm, ranging from 3 to 4.2 mm, and the mean MH was 0.381 mm, ranging from 0.291 to 0.41 mm (n = 35). At this stage, notochord exion had started, larvae had increased head size, and both maxillar and premaxillar bones were gaining in thickness and strength. Teeth were present now in both jaws bones. The digestive system was well differentiated, with all the organs gaining in volume and structure. Also, 90% of the observed stomachs were full, rotifers were observed all along the digestive system. Fin shape denition continued to develop, especially the dorsal and caudal ns (Figures 3, 4g). Common snook larvae at day 12 (Figures 3 and 4, item h) had a mean SL of 3.57 mm, ranging from 3.41 to 3.84 mm, and an average MH of 0.401 mm, ranging from 0.266 to 0.467 mm (n =
35). The development of the dorsal and anal n bases could be observed. At the same time, notochord exion was nearly complete. Larval body depth continued to increase and the head size was still proportionally larger in width than the rest of the body. By day 14, common snook larvae (Figures 3 and 4, item i) had a mean SL of 4.43 mm, ranging from 3.57 to 5.71 mm, and their mean MH was 0.41 mm, ranging from 0.36 to 0.46 mm (n = 35). By this stage, the lower dorsal n was developing faster than the upper one and nine rays could be observed. Similar development happened to the anal n, although rays were not as developed and notochord exion was completed. During the rst 14 d after hatching, development of common snook was rapid in terms of SL (Figure 2). We observed three stages: the rst stage or the yolk sac stage occurred over a 2-d period where the SL increased from 1.78 to 2.311 mm. The next stage was the preexion stage, which occurred over a 4-d period, where the increase in SL was less than in the yolk sac stage, going from an average SL of 2.262.56 mm (Figure 2). The last stage or the exion stage occurred over a 6-d period and the average SL went from 2.56 mm to 4.43 mm, increasing exponentially each day. The mean SL and MH measurements during the rst 14 d of growth can be seen in Table 1. DISCUSSION Common snook Embryonic Development The developmental features of common snook described in this paper are typical of most teleost species with planktonic eggs. No common snook eggs have been described from the wild, although it is known that, immediately after spawning, eggs are taken towards the open sea by the outgoing tide and it is assumed that the following incoming tide brings them back into the estuarine and mangrove environment. Sampling efforts to nd wild common snook eggs have been carried out, but the quest has been so far unsuccessful. Many factors may have resulted in this unsuccessful outcome, such as sampling gear, location, time, and the inability to recognize common snook eggs once collected. The search for buoyant planktonic embryos in the wild will provide useful information for the development of incubation protocols in the laboratory. This is important since the inuence of the environmental conditions inuences early development and physiology of the offspring (Blaxter 1992). The processes of cleavage, formation of layers, and morphogenesis of teleost eggs during incubation have been described in a number of standard textbooks (such as Rudnick 1955, Waddington 1956, and Smith 1957), with Oppenheimer (1947) and Devillers (1961) stressing structural changes from the viewpoint of experimental embryology. Most of those descriptions were done for freshwater species, although extensive work has been done for some marine species including silver warehou Seriolella punctata (Grimes and Robertson 1981), gilthead bream Sparus aurata, European ounder Platichthys esus, dab Limanda limanda, Atlantic herring Clupea harengus,
509
Atlantic cod Gadus morhua, plaice, and Atlantic salmon Salmo salar. Common snook eggs are similar to those of many other sh species in their round shape, although in the anchovy Engraulis spp. and bitterling Rhodeus sericeus the eggs are ovoid and in certain gobies (family Gobiidae) the eggs are pear shaped. At the same time, the common snook embryo has only one oil globule, whereas more than one can be found in other marine species (Simpson 1956) and in most sh species that have telolecithal eggs with the yolk more concentrated at the vegetative pole. As with hagsh (family Myxinidae) and elasmobranches, common snook exhibits meroblastic cleavage, even though snook are teleosts. They do not have the holoblastic cleavage that characterizes lampreys (family Petromyzontidae). Other groups, such as bown Amia spp., gar Lepisosteus spp., and sturgeon Acipenser spp., have intermediate features. All the above embryo characteristics, plus the melanophores pattern, egg size, and oil droplet size described in the results, will make the recognition of common snook embryos more accurate and provide a tool for egg quality evaluation in common snook culture. Common snook Morphological Development Common snook larval development was described in 1982 by Lau and Shaand. They primarily focused on larvae from 14 d old and beyond, mainly looking at osteological, cephalic, and n development. Also, Wittenrich et al. (2009) described the osteological development of the feeding apparatus with feeding performance, but other than this no reported common snook development studies have been carried out. Eye Development At hatching, common snook larvae have an unpigmented and nonfunctional visual system; many other species are similar in this respect (e.g., sole, Atlantic mackerel Scomber scombrus, whiting Merlangius merlangus, European pilchard Sardina pilchardus, and Pacic sardine Sardinops sagax caeruleus), while some others have pigmented eyes (e.g., plaice, Atlantic cod, Atlantic herring, and salmonids) (Blaxter and Staines 1970). Blaxter (1986) described the important role that vision has on sh larvae orientation, as a consequence of them being visual feeders. Fish species with a relatively small yolk sac need to develop fast in order to survive. In common snook larvae, just like in sparids such as madai Pagrus major (Kawamura 1984) and New Zealand snapper Pagrus auratus (Pankhurst 1996), the most important changes in the eye structure occur in the lecitotrophic larvae as a preparation for prey capture. One-day-old common snook larvae had all the basic structural elements necessary for visual function, but most of them were incomplete. This was an indication that the eye was about to become functional. Kawamura (1984) found that the visual system of madai is functional at 36 h posthatch when visual cells and pigments are present and nerve optic bers connect with the optic tectum. In common snook, the visual system could not be functional at day 1 posthatch, principally because the pigmenta-
tion pattern, responsible for photon absorption, was very sparse at this stage. Retinas of most sh larvae mainly have greensensitive single cones (Evans and Browman 2004). This is the case of 2-d-old common snook larvae: although the retina was not as fully developed as in the adult stage, all the retina structural layers were complete but not fully functional. Histological observations support the partial functionality of the eye; by this time the pigment cell layer was present and the optic nerves were connected to the optic tectum. Thus, the visual system was completely ready for prey capture. The pure cone retina has been found in the earlier stages of many teleost larvae such as Pacic salmon Oncorhynchus spp. (Ali 1959), Atlantic herring (Blaxter and Jones 1967), and plaice (Blaxter 1968). However, at rst feeding common snook are only equipped with simple cones, as with madai (Kawamura 1984) and New Zealand snapper (Pankhurst 1996), and rods and twin cones appear at metamorphosis (Blaxter and Staines 1970). At day 3 the common snook larval retina has well-developed presumptive cone receptors, the pigmentation layer is fully complete, the structural layers are fully functional and differentiated, and the formation of the rod precursors can be spotted, although they are scarce. Overall, by day 2 posthatching the visual system of common snook larvae is developed sufciently to locate and capture prey, although due to the underdeveloped stage of the rods (which provide better vision in low light [OConnell 1981; Kawamura 1984; Pankhurst 1996]) adequate light conditions are necessary to optimize their ability to capture prey (Huse 1993). Taking the rod development into consideration, light intensity during larval development should be altered accordingly, and to investigate this matter more samples of older larvae should be examined using histology. Digestive System Development At hatching, common snook larvae had a simple undifferentiated straight gut linked to an unstructured mouth and anus, as described in other teleost species (Stroband and Dabrowski 1979; Govoni 1980; Cousin and Baudin-Laurencin 1985; Govoni et al. 1986; Boulhic and Gabaudan 1992; Bisbal and Bengston 1995; Roo et al. 1999; Pe na et al. 2004). It is generally assumed that lipid absorption takes place in the anterior intestine, and based on this assumption and the importance that lipids have over the larval development, all the histological work done in this study was based on the anterior intestine development. During the rst 2 d, the common snook larval digestive system undergoes major changes. The anus and the mouth open, gut cells undergo signicant growth, development of organelles is increased, and the intestinal valve is formed. On day 1, the alimentary canal differentiation is starting to appear, the mouth is starting to be formed, but jaw cartilages are still developing. At the cell level, organelles can be observed but their underdeveloped stage and low numbers render the digestive system unable to function. The alimentary canal epithelium at some parts of the luminal surface showed the presence of microvilli, although
510
YANES-ROCA ET AL.
in low numbers and underdeveloped, at the same time ciliated cells were present in the lumen. Such ciliated cells were not found after day 1, and similar ndings were observed in other species (Govoni et al. 1986; Loewe and Eckman 1988; Calzada et al. 1998). The presence of these ciliated cells in larvae (which did not have any peristaltic movements) was reported by Iwai (1964) and contributes to the circulation of the intestinal contents (mainly yolk). Common snook larvae by day 2 posthatch had absorbed the yolk sac, which was also observed in goldlined seabream Rhabosargus sarba (Ibrahim 2004) and in sea bass Lates calcifer (Walford and Lam 1993). Other species, such as Atlantic cod and sheephead seabream Archosargus probatocephalus (John and Tucker 1987; Kjrsvik et al. 1991), exhausted their yolk sac at 4 d after hatching. The mouth was open and the main jaw cartilages such as the Meckels cartilage or the hyomandibular cartilage were present. These results were also seen in other species, such as rainbow darter Etheostoma caeruleum, white sucker Catostomus commersonii, logperch Percina caprodes and Atlantic cod (McElman and Balon 1981; Paine and Balon 1985; Kjrsvik et al. 1991). The dermal bones, such as the maxillary and the premaxillary are formed later on; however, food was observed in the gut. Morphological and histological observations suggest that 2 d after hatching common snook larvae possess digestive organs enabling digestion, absorption, and metabolization of endogenous food. At the onset of day 2 posthatch, enterocytes are morphologically developed, yet as with cod larvae (Kjrsvik et al. 1991), the digestive mechanisms are immature and their functionality relies on lipid absorption and possibly temporary lipid storage in the anterior part of the gut (Tanaka 1972a; Stroband and Dabrowski 1979). Although no histology of the rectal area was done, the epithelial cells are probably responsible for food protein ingestion and intracellular digestion (Iwai and Tanaka 1968; Tanaka 1972b; Watanabe 1982a, b). Day 3 common snook larvae had continued to develop. Mouth jaw cartilages were gaining in denition and speeding their functionality. The alimentary canal is now more structured with a clear differentiation between the different gut parts. Organelles, such as mitochondrion, rough endoplasmatic reticule, and the golghi apparatus, increased in numbers. The microvilli layer is increased in length and consistency. In conclusion, the alimentary canal of common snook larvae develops from a undifferentiated tube at hatching to a complex tract before the onset of day 2 after hatching (when the yolk sac is exhausted). This fast development is parallel to that of the visual system; synchronization of the formation of these two systems is important for prey capture and predator avoidance. Together with the digestive and visual system development, common snook larvae have a partially developed n structure that allows them to move in the water column and approach prey. Like their feeding habits and behavior, the location of common snook larvae in the planktonic column is unknown. This paper has described the common snook larval development dur-
ing the rst 14 d in laboratory conditions in order to provide information for future studies on wild common snook larvae and for aquaculture purposes. Although there are many theories regarding the diet of common snook larvae, there are no reported studies on this topic; common snook larval diets are one of the major bottlenecks in snook aquaculture. In addition, issues such as prey type and size during the rst 57 d are still poorly understood. Therefore, the collection of wild larvae will provide useful information to identify the optimal prey, improve the rearing protocol, and increase knowledge of the wild common snook larvae ecology.
ACKNOWLEDGMENTS This work was supported by grants from the Institute of Aquaculture at Stirling University, the Florida Fish and Wildlife Conservation Commission, the National Oceanic and Atmospheric Administration funded research consortium, the Science and the Consortium for Ocean Replenishment, and the Mote Scientic Foundation.
REFERENCES
Ali, M. A. 1959. The ocular structure, retimotor and photobehavioral responses of juvenile Pacic salmon. Canadian journal of Zoology 39:511526. Avila, E. M., and J. M. Juario. 1987. Yolk and oil-globule utilization and development morphology of the digestive tract epithelium in larval rabbitsh, Sinatus guttatus. Aquaculture 65:319331. Batty, R. S. 1987. Effect of light intensity on activity and food-searching of larval herring (Clupea harengus): a laboratory study. Marine Biology 94:323327. Blaxter, J. H. S. 1992. The effect of temperature on larval shes. Netherlands Journal of Zoology 42:336357. Blaxter, J. H. S. 1986. Development of sense organs and behaviour of teleost larvae with special reference to feeding and predator avoidance. Transactions of the American Fisheries Society 115:98114. Blaxter, J. H. S. 1968. Light intensity, vision and feeding in young place. Journal of Experimental Marine Biology and Ecology 48:3953. Blaxter, J. H. S., and M. P. Jones. 1967. The development of the retina and retinomotor responses in the herring. Journal of Marine Biology 47:677697. Blaxter, J. H. S., and M. Staines. 1970. Pure cone retina and retinomotor responses in larval teleosts. Journal of the Marine Biology Association 50:449 460. Bisbal, G. A., and D. A. Bengston. 1995. Development of the digestive tract in larval summer ounder. Journal of Fish Biology 47:277291. Boulhic, M., and J. Gabaudan. 1992. Histological study of the organogenesis of the digestive system and swim bladder of the Dover sole, (Solea solea) (Linnaeus 1758). Aquaculture 102:373396. Calzada, A., A. Medina, and M. L. Gonzalez de Canales. 1998. Fine structure of the intestine development in cultured sea bream larvae. Journal of Fish Biology 53:340365. Cousin, J. C. B., and F. Baudin-Laurencin. 1985. Morphogen` ese de lappareil digestif et de la vessie gazeuse du turbot, Scophthalmus maximus. [Morphogenesis of the digestive system and swim bladder of the turbot Scophthalmus maximus.] Aquaculture 47:305319. Deplano, M., J. P. Diaz, M. Connes, R. Kentouri-Divanach, and F. Cavalier. 1991. Appearance of lipid-absorption capacities in larvae of the sea bass, Dicentrarchus labrax during transition to the exotrophic phase. Marine Biology 108:361371. Devillers, C. 1961. Structural and dynamic aspects of the development of the teleostean egg. Advance Morphogenesis 1:379428.
EMBRYONIC AND EARLY LARVAL DEVELOPMENT Eckman, R. 1987. Pathological changes in the midgut epithelium on grayling, Thymallus thymallus L., and larvae reared on different kinds of food, and their relation to mortality and growth. Journal of Fish Diseases 10:9199. Evans, B. I., and H. Browman. 2004. Variation in the development of sh retina. Pages 145166 in J. J. Govoni, editor. The development of form and function in shes and the question of larval adaptation. American Fisheries Society, Symposium 40, Bethesda, Maryland. Ferrais, R. P., J. D. Tan, and M. C. de laCruz. 1987. Development of the digestive tract of milksh, Chanos chanos (Forskal): histology and histochemistry. Aquaculture 61:241257. Govoni, J. J. 1980. Morphological, histological, functional aspects of alimentary canal and associated organ development in larval (Leiostomus xanthurus). Reveu Canadiene 39:6980. Govoni, J. J., G. W. Boenhlert, and Y. Watanabe. 1986. The physiology of digestion in sh larvae. Environmental Biology of Fish 16:5977. Grimes, P. J., and D. A. Robertson. 1981. Egg and larva development of the silver warehou Seriolella punctata (Pisces: Centrolophidae). New Zealand Journal of Marine and Freshwater Research 15:261266. Howells, R. G., A. J. Sonski, P. L. Shaand, and B. D. Hilton. 1990. Lower temperature tolerance of snook, Centropomus undecimalis. Northeast Gulf Science 11:155158. Huse, I. 1993. First feeding larval sensory perception and behavioural programming: implications for systems and procedures in culture. Pages 146152 in B. T. Walther and H. J. Fyhn, editors. Physiology and biochemical aspects of sh development. University of Bergen, Bergen, Norway. Ibrahim, F. S. 2004. Reproductive biology of wild gold lined seabream, Rhabdosargus sarba captive breeding and larval development in the Sultanate of Oman. Doctoral dissertation. Stirling University, Stirling, Scotland. Iwai, T. 1964. Feeding and ciliary conveying mechanisms in larvae of salmonid sh Plecoglossus altivelis. Physiological Ecology 12:3844. Iwai, T., and M. Tanaka. 1968. The comparative study of the digestive tract of the teleost larvae-III. Epithelial cells in the posterior gut of helfbreak larvae. Bulletin of the Japanese Society of Scientic Fisheries 34:4448. John, W., and J. R. Tucker. 1987. Sheephead culture and preliminary evaluation for farming. The progressive sh culturist 49:224228. Jonh, K. R., and A. D. Hasler. 1956. Observations on some factors affecting the hatching of eggs and the survival of young shallow-water cisco Leucichthys artedi Le Suer, in Lake Mendota, Wisconsin. Limnology and Oceanography 1:176194. Kawamura, G. 1984. The visual cell morphology of Pagrus pagrus and its adaptive changes with shift from pelagic and benthic habitats. Bulletin of Japanese Society of Scientic Fisheries 50:19751980. Kjrsvik, E., T. Van derMeer, J. Kryvi, S. Arnnnson, and P. G. Kvenseth. 1991. Early development of the digestive tract of cod larvae, Gadus morhua L., during start feeding and starvation. Journal of sh Biology 38:115. Lau, S. R., and P. L. Shaand. 1982. Larval development of snook Centropomus undecimalis (Pisces: Centropomidae). Copeia 1982:618627. Loewe, H., and R. Eckman 1988. The ontogeny of the alimentary tract of coregonid larvae: normal development. Journal of Fish Biology 33: 841850. Mani-Ponset, L., E. Guyot, J. P. Diaz, and R. Connes. 1996. Utilization of yolk reserves during postembryonic development in three teleostean species: the sea bream, (Sparus aurata), the sea bass, (Dicentrarchus labrax) and the pike-perch (Stizostedion lucioperca). Marine Biology 126:539547. McElman, J. F., and E. K. Balon. 1981. Early ontogeny of the white sucker, Catostomus comersoni, with steps of saltatory development. Journal of Fish Biology 5:191224. OConnell, C. P. 1981. Development of organ systems in the northern anchovy, Engraulis mordax, and other teleosts. American Zoology 21:429446.
511
Oppenheimer, J. M. 1947. Organization of the teleost blastoderm. Quarterly review of Biology 22:105118. Paine, M. D., and E. K. Balon. 1985. Early development of the northern log perch, Percina caprodes semifsciata, according to the theory of the saltatory ontogeny. Environmental Biology of Fishes 11:173190. Pankhurst, N. W. 1996. Changes in visual morphology throughout life history stages of the New Zealand snapper, Pagrus auratus. New Zealand Journal of Marine Freshwater Research 30:7990. Paul, A. J. 1983. Light, temperature, nauplii concentrations, and prey capture by rst feeding pollock larvae (Theragra chalcogramma). Marine Ecology Progress Series 13:175179. Pe na, R., S. Dumas, R. Saldivar-Lucio, G. Garcia, A. Trasvin, and D. HernandezCeballos. 2004. The effects of light intensity on rst feeding of the spotted sand bass Paralabrax maculatofasciatus Steindachner larvae. Aquaculture Research 35:345349. Porter, S. M., and G. H. Theilacker. 1999. The development of the digestive tract and eye in larval walleye pollock, Theragra chalcograma. U.S. National Marine Fisheries Service Fishery Bulletin 97:722729. Roo, F. J., J. Socorro, M. S. Izquierdo, M. J. Caballero, C. M. Hernandez-Cruz, A. Fernandez, and H. Fernadez-Palacios. 1999. Development of red porgy (Pagrus pagrus) visual system in relation with changes in the digestive tract and larval feeding habits. Aquaculture 179:499511. Rudnick, D. 1955. Teleost and birds. Pages 297314 in B. H. Willier, P. A. Weiss, and V. Hamburger, editors. Analysis of development. Saunders, Philadelphia. Sarasquete, M. C., A. Polo, and M. Yufera. 1995. Histology and histochemistry of the development of the digestive system of larval gilthead seabream, Sparus aurata L. Aquaculture 130:7992. Simpson, A. C. 1956. The pelagic phase. Pages 207250 in M. Graham, editor. Sea sheries: their investigation in the United Kingdom. Arnold, London. Smith, S. 1957. Early development hatching. Pages 323359 in M. E. Brown, editor. The physiology of shes. Academic Press, New York. Stroband, H. W. J., and K. R. Dabrowski. 1979. Morphological and physiological aspects of the digestive system and feeding in freshwater sh larvae. Pages 355376 in M. Fontaine, editor. La nutrition des poissons. [Fish nutri tion.] Centre National dEtudes et de Recommandations sur la Nutrition et lAlimentation, Paris. Tanaka, M. 1972a. Studies on the structure and function of the digestive system in teleost larvae -IV changes in the epithelium related to fat absorption in the anterio-medium part of the intestine after feeding. Japanese Journal of Ichthyology 19:1525. Tanaka, M. 1972b. Studies on the structure and function of the digestive system in teleost larvae-V. Epithelial changes in the posterior gut and protein ingestion. Japanese Journal of Ichthyology 19:172180. Taylor, R. G., H. J. Grier, and J. A. Whittington. 1998. Spawning rhythms of common snook in Florida. Journal of Fish Biology 53:502520. Waddington, C. H. 1956. Principles of development. Allen and Unwin, London. Walford, J., and T. J. Lam. 1993. Development of digestive tract and proteolytic enzyme activity in sea bass (Lates calcifer) larvae and juveniles. Aquaculture 109:187205. Watanabe, Y. 1982a. Intercellular digestion of horseradish peroxidase by the intestinal cells of teleost larvae and juveniles. Bulletin of the Japanese Society of Scientic Fisheries 48:3742. Watanabe, Y. 1982b. Ultraestructure of epithelial cells of the antero-median intestine and the rectum in larval and juvenile teleost. Bulletin of the Faculty of Fisheries 33:217228. Wittenrich, M. L., N. R. Rhody, R. G. Turigan, and K. L. Main. 2009. Coupling osteological development of the feeding apparatus with feeding performance in common snook, Centropomus undecimalis, larvae: identifying morphological constrains to feeding. Aquaculture 294:221227.

An Injectable, Slow-Release Implantation Method for Exposing Fish to Chemicals over a Period of Weeks
Gerald E. Zaroogian , Ruth E. Gutjahr-Gobell , Doranne Borsay Horowitz , Saro Jayaraman , Mark Cantwell , Clinton O. Chichester & Lesley J. Mills
a a a b a a a a
U.S. Environmental Protection Agency, Office of Research and Development, National Health and Environmental Effects Research Laboratory, Atlantic Ecology Division, 27 Tarzwell Drive, Narragansett, Rhode Island, 02882, USA
b
Department of Biomedical and Pharmaceutical Sciences, University of Rhode Island, Kingston, Rhode Island, 02881, USA Version of record first published: 14 Sep 2012.
To cite this article: Gerald E. Zaroogian, Ruth E. Gutjahr-Gobell, Doranne Borsay Horowitz, Saro Jayaraman, Mark Cantwell, Clinton O. Chichester & Lesley J. Mills (2012): An Injectable, Slow-Release Implantation Method for Exposing Fish to Chemicals over a Period of Weeks, North American Journal of Aquaculture, 74:4, 512-521 To link to this article: http://dx.doi.org/10.1080/15222055.2012.697097
ARTICLE
An Injectable, Slow-Release Implantation Method for Exposing Fish to Chemicals over a Period of Weeks
Gerald E. Zaroogian, Ruth E. Gutjahr-Gobell, Doranne Borsay Horowitz, Saro Jayaraman, and Mark Cantwell
U.S. Environmental Protection Agency, Ofce of Research and Development, National Health and Environmental Effects Research Laboratory, Atlantic Ecology Division, 27 Tarzwell Drive, Narragansett, Rhode Island 02882, USA
Clinton O. Chichester
Department of Biomedical and Pharmaceutical Sciences, University of Rhode Island, Kingston, Rhode Island 02881, USA
Lesley J. Mills*
U.S. Environmental Protection Agency, Ofce of Research and Development, National Health and Environmental Effects Research Laboratory, Atlantic Ecology Division, 27 Tarzwell Drive, Narragansett, Rhode Island 02882, USA
Abstract
A slow-release, injectable implant method was developed for administering test chemicals to cunners Tautogolabrus adspersus. The implant is composed of a matrix of a test chemical homogenized in a mixture of Ethocel (Dow Chemical) and coconut oil. The effectiveness of a subcutaneous implant of this matrix in vivo was determined by tracing plasma concentrations of three separate chemicals (estradiol, ethynylestradiol, and atrazine) over time in treated male cunners. Release from the implant was determined based on the percentage of the implanted concentration of test chemical (plus metabolites) that was detected in sh plasma over a 12-week period after implantation. Circulating estrogen concentrations measured in plasma from two different cunners that received the estradiol implant were almost identical, indicating that there is a reasonably even distribution of test chemical within the Ethocelcoconut oil preparation and that individual variability may be minimal for release of test chemical from the implant. Metabolites of estradiol and atrazine were a major portion of the circulating concentration of these chemicals. Estradiol and atrazine demonstrated metabolic and clearance proles that were very different from those of the xenoestrogen ethynylestradiol. A follow-up in vitro study was conducted to further characterize the release of estradiol from the implant matrix. Results showed a rapid release of estradiol from the matrix bolus during the rst 24 h, followed by a more gradual release over subsequent days. The in vitro tests indicated that measuring in vivo plasma concentrations may not accurately reect the release rate of a chemical from the implant matrix, in part because metabolism and clearance affect the circulating concentrations in vivo.
For years, aquaculturists and aquatic toxicologists have had an interest in nding effective and efcient ways to administer hormones or other chemicals to sh in a sustained, controlled manner. Administration of hormones to sh is used for multiple
purposes in aquaculture, including the manipulation of reproduction, the induction of rapid growth, and the production of monosex populations (Pandian and Sheela 1995; Shelton and Mims 2003). In the past, exogenous hormones and chemicals
*Corresponding author: mills.lesley@epa.gov Received January 12, 2012; accepted May 19, 2012
512
SLOW-RELEASE METHOD FOR EXPOSING FISH TO CHEMICALS
513
have been administered to sh through a variety of methods, including diet (e.g., Gutjahr-Gobell et al. 1999; Patyna et al. 1999; Bayley et al. 2002; Madsen et al. 2003; Keen et al. 2005), water column exposure (e.g., Hunter and Donaldson 1983; Gimeno et al. 1998; Kramer et al. 1998; Metcalfe et al. 2001; Segner et al. 2003; Arslan and Phelps 2004; Lahnsteiner et al. 2006), and oral gavage (Sundararaj and Goswami 1968; Pizza and OConnor 1983). Although exposure through such methods may be desirable when trying to simulate environmental routes of exposure, the administration of hormones or chemicals through injection or surgical implantation may be more practical in circumstances where a direct route of exposure with minimal effort is needed. Incorporation of test chemicals into a vegetable oil vehicle (e.g., olive oil, corn oil, peanut oil, sesame oil, coconut oil, or cocoa butter), cholesterol, a cholesterolcocoa butter matrix, or a cholesterolanimal lard matrix has been used by various researchers for intraperitoneal implantation of chemicals into sh (Leatherland 1985; Pankhurst et al. 1986; Carolsfeld et al. 1988; Crim et al. 1988; Cyr and Eales 1989; Garcia 1989; Bisbal and Specker 1991; Donohoe and Curtis 1996; Black et al. 1998; Mandiki et al. 2004; Sangiao-Alvarellos et al. 2005; Wang et al. 2005). A consideration with intraperitoneal implantation is that the test chemical is placed adjacent to key abdominal organs (e.g., gonads or liver), which could result in an increased, and perhaps undesirable, local exposure of these organs. However, cholesterol and some oils appear to provide a prolonged release. For example, Yamada et al. (1997) reported that after rainbow trout Oncorhynchus mykiss received intraperitoneal implantation with a steroid in coconut oil, an initial increase in plasma steroid concentration was observed 1 d after implantation, followed by a gradual decrease over the subsequent 15 d. They suggested that coconut oil, which solidies when implanted, was responsible for the slow release of steroid into the peritoneal cavity. Wang at al. (2005) reported similar results when using cocoa butter as a carrier for the injection of cortisol into grass carp Ctenopharyngodon idella. Sherwood et al. (1988) found that cholesterol-based pellets yielded sustained releases of gonadotropin-releasing hormone analog (GnRHa) in vitro over a period of 28 d. Crim et al. (1988) found that both intraperitoneal and intramuscular pellet implants composed of plasma gonadotropin and cholesterol (with or without cocoa butter) yielded sustained releases of luteinizing hormone releasing hormone analog in juvenile rainbow trout. Implantation of crystalline exogenous compounds enclosed in silicone or silastic tubing has been used with some success, resulting in biological effects lasting several weeks to months in some cases (Pickering and Duston 1983; Pankhurst et al. 1986; Trudeau et al. 1991; Joakim Larsson et al. 2002; Yamaguchi et al. 2004; Gonc alves et al. 2007). However, this method requires that minor surgery be conducted to implant the capsule into each sh. In addition, chemicals that are hydrophilic do not penetrate easily through the hydrophobic matrix, thus decreasing implant effectiveness (Mylonas and Zohar 2000).
A less-invasive method that has been used with some success is the incorporation of chemicals into coconut oil or into a cocoa buttercholesterol mixture. The mixture may then be injected into sh intramuscularly or subcutaneously (Lee et al. 1986; Scott et al. 1999). Coconut oil has been used successfully in our laboratory (U.S. Environmental Protection Agency [USEPA], National Health and Environmental Effects Research Laboratory [NHEERL]) to implant exogenous estrogens and other chemicals subcutaneously into summer ounder Paralichthys dentatus (Mills et al. 2001). Our research into the biological effects of endocrinedisrupting chemicals on sh necessitated the development of an inexpensive delivery method that could be used on our test species, the cunner Tautogolabrus adspersus. Since laboratory water exposure of cunners was impractical due to the size and spawning behavior of this species, the goal of the present study was to develop and test the utility of a minimally invasive, injectable implant that would continuously release exogenous chemicals into the circulation of treated sh over time. We report here on the suitability and effectiveness of our implant method by tracing plasma concentrations of three separate chemicals (estradiol, ethynylestradiol, and atrazine) that were implanted by subcutaneous injection just below the dorsal n of laboratoryheld cunners. Plasma concentrations of test chemical and major metabolites were measured over time. To better characterize the release of chemical from our implant without the in vivo effects of metabolism and clearance, we also examined release of radio-labeled estradiol from boluses of implant matrix in vitro. METHODS Animals Cunners were collected from the East Passage of Narragansett Bay, Rhode Island, off a large stone pier at the southeastern end of Jamestown (Conanicut Island) during the summer in 19992005. Details of collection methods are provided by Gutjahr-Gobell et al. (2002). Briey, cunners were captured by using modied Gee minnow traps (Memphis Net and Twine, Memphis, Tennessee) and were transported to the laboratory in coolers lled with ambient seawater. Fish were held in large, aerated rectangular (4,400 L) or round (1,000 L) holding tanks that received ow-through Narragansett Bay seawater. Cunners were fed an ad libitum ration of thawed and chopped krill, squid, and mussels; the sh were held over winter in the laboratory at the ambient temperatures and photoperiod of Narragansett Bay. Chemicals Ethocel Standard FP Premium 10 (a formulation of ethyl cellulose) was obtained from Dow Chemical (Midland, Michigan). Rened coconut oil (100% pure), distributed by Spectrum Essentials (Petaluma, California), was purchased locally. 17-estradiol (1,3,5-[10]-estratrien-3,17-diol; Chemical Abstracts Service [CAS] Number 50-28-2) and estrone
514
ZAROOGIAN ET AL.
(1,3,5[10]-estratrien-3-ol-17-one; CAS Number 53-16-7) were obtained from Steraloids (Wilton, New Hampshire). 17ethynylestradiol (17-ethynyl-1,3,5[10]-estratriene-3,17diol; CAS Number 57-63-6), estriol (1,3,5[10]-estratriene3,16,17-triol; CAS Number 50-27-1), and pyridine (highperformance liquid chromatography grade; CAS Number 11086-1) were purchased from Sigma-Aldrich (St. Louis, Missouri). The derivatizing agent bis(trimethylsilyl)triuoroacetamide (BSTFA) was obtained from Supelco (Bellefonte, Pennsylvania). Atrazine (6-chloro-N-ethyl-N -[1-methylethyl]1,3,5-triazine-2,4-diamine; CAS Number 1912-24-9) and its metabolites deethylatrazine (DEA; 2-chloro-4-amino6-isopropylamino-s-triazine), deisopropylatrazine (DIA; 2chloro-4-ethylamino-6-amino-s-triazine), and diaminochlorotriazine (DACT; 2-chloro-4,6-diamino-1,3,5-triazine) were provided to the Reproductive Toxicology Division (USEPA, NHEERL) by Syngenta Crop Protection (Greensboro, North Carolina). The deuterated estradiol (17-estradiol-d4 ) and atrazine (atrazine-ethylamine-d5 ) used as internal standards for analytical chemistry procedures were purchased from Cambridge Isotope Laboratories (Andover, Massachusetts). Preparation of the Injectable Implant Matrix Stock solutions of the test chemicals were prepared in different solvents, depending upon their solubilities, as follows: estradiol was dissolved in a mixture of 0.5 mL of ethanol and 0.5 mL of acetone to make a 100-mg/mL stock solution; ethynylestradiol was dissolved in acetone to prepare a 100-mg/mL stock solution; and atrazine was dissolved in a mixture of 0.5 mL of chloroform and 0.5 mL of ethanol to prepare a 75-mg/mL stock solution. To prepare the slow-release matrix, Ethocel was dissolved in methylene chloride. The solution was then reduced in volume to approximately 1 mL. The desired amount of test chemical stock solution was added to the Ethocelmethylene chloride mixture and was mixed well. The ratio of Ethocel to chemical in the ethynylestradiol and atrazine preparations was approximately 40:1. The 40:1 ratio was chosen based upon prior experimentation with the solubilities of these compounds in the nal implant matrix. Our objective was to maximize the amount of test chemical that was incorporated into the matrix without having precipitation occur during later steps of the implant preparation process. In the preparation of the estradiol implant, the ratio of Ethocel to estradiol was limited to a maximum of approximately 17:1 because any concentration of Ethocel exceeding 50 mg/mL was prone to precipitation during subsequent steps. Slow-release matrix was made in 5-mL batches. Liquid coconut oil (2 mL), used as a carrier for the slow-release matrix, was added to the Ethocelmethylene chloridetest chemical mixture. The mixture was homogenized thoroughly until emulsied. An additional 2-mL quantity of coconut oil was added, and the mixture was homogenized again. Since preliminary experiments showed that residual solvent alone caused irritation and necrosis at the injection site in sh that received
the implant matrix containing no test chemical, the mixture was placed in a light-tight vial and stirred under vacuum on a magnetic stirrer until the solvent had completely dissipated. The implant preparation was then brought up to a nal volume of 5 mL by adding coconut oil and mixing thoroughly. Control implants were prepared with the same amounts of solvent and Ethocel as the highest test chemical concentration but with no chemical added. Just prior to injection, the implant matrix was liquied by gently warming to approximately 35 C. Each sh was lightly anesthetized with tricaine methanesulfonate (MS222). By use of a 1-mL glass syringe with an 18-gauge needle, the appropriate liquied matrix (at 2 L/g of sh wet weight) was implanted subcutaneously into each sh just below the dorsal n. Gentle nger pressure was applied over the injection site for a few seconds until the coconut oil in the bolus solidied, thus preventing any leakage of the matrix. Experimental Design All sh tanks were aerated and received 1820 C owthrough seawater at a ow rate of 1 L/min. To provide submerged cover for the sh, each tank also contained a 20-cm length of 10-cm-diameter polyvinyl chloride pipe. Fish were allowed to acclimate to experimental conditions for 35 d before an experiment started. All sh were fed fresh or thawed mussels ad libitum every day. Further details of the experimental system are described by Gutjahr-Gobell et al. (2002). Each plasma sample was obtained from blood that was drawn from a single sh; individual samples were not pooled for analysis. Percentages of each chemical in the sh plasma samples taken over time were calculated by using the following formula: %Fp = Cp 100, Wf D
where %Fp is the percentage of the nominal concentration of implanted chemical that was detected in the sh plasma, Cp is the amount of chemical (g) measured in 1 mL of sh plasma, Wf is the weight of the sh (g), and D is the nominal concentration of chemical (g/g) administered to the sh. Chemical analysis of boluses was not possible in either the in vivo or in vitro studies because of the high bolus lipid content. Estradiol treatment.A repeat sampling design was used to obtain blood samples during this experiment, meaning the same sh was sampled at six time points. This design required the use of very large cunners that could tolerate repeated blood draws. Because the availability of sufciently large sh was limited, only two sh were used. Each of two large male sh received a single concentration of estradiol (6 g/g of sh wet weight) in the slow-release implant matrix. Fish wet weight was approximately 180 g, and sh length (measured as total length) was approximately 22 cm. Each individual was kept in its own tank an 80-cm-tall, 114-L-capacity, high-density polyethylene barrel (47-cm diameter) with a clear Plexiglas cover. Just before implantation, 0.4 mL of blood was drawn from a caudal vein of each
515
sh by using a 1-mL tuberculin syringe with a 22-gauge needle rinsed with a heparin sodium salt solution (1,000 units/mL; United States Biochemical). These pretreatment samples were used as controls. Subsequent blood samples (0.4 mL) were taken at 3 h after implantation and at 1, 2, 3, 4, and 7 d after implantation. All blood samples were processed as described by GutjahrGobell et al. (2002). Plasma was stored at 80 C until assayed. Ethynylestradiol treatment.Five smaller male cunners were used for the ethynylestradiol study. Wet weight of sh ranged from 23 to 36 g and averaged 30 g (SD = 5). Length was between 12.5 and 14.0 cm and averaged 13.2 cm (SD = 0.6). Four males received implanted ethynylestradiol at 1.2 g/g of sh weight, and one male received a control matrix containing no ethynylestradiol. Fish were housed individually in 38-L (10-gal) glass aquaria with aeration and ow-through seawater. Each sh was sampled twice during the experiment by using methods previously described (Gutjahr-Gobell et al. 2002). Blood samples were drawn from one sh at 1 and 8 d after implantation, from a second sh at 3 and 10 d postimplantation, from a third sh at 5 and 12 d postimplantation, and from a fourth sh at 7 and 14 d postimplantation. Blood from the control sh was sampled at 7 and 14 d after control matrix implantation. All blood samples were processed as described by Gutjahr-Gobell et al. (2002). Plasma samples were stored at 80 C until assayed. Atrazine treatment.Six male cunners were housed individually in 38-L (10-gal) glass aquaria with aeration and owthrough seawater. Wet weight of sh ranged from 32 to 47 g, with a mean of 41 g (SD = 6). Length was between 13.1 and 14.6 cm, with a mean of 14.0 cm (SD = 0.6). Four sh were given atrazine in slow-release matrix at 2.4 g/g of sh weight on day 0 and were sampled at 1, 3, 5, and 7 d postimplantation. A different sh was bled on each sampling day, and as much blood as possible was drawn from each sh. The process of sampling and separating plasma was as described by GutjahrGobell et al. (2002). Slow-release matrix containing no atrazine was implanted into two control sh; one of the control sh was sampled on day 1, and the other control sh was sampled on day 7. Plasma samples were stored at 80 C until assayed. In vitro study.To better characterize the release of chemical from our implant without the inuence of metabolism or clearance, we conducted an in vitro study to quantify the release of estradiol from the implant matrix into Leibovitzs L-15 culture medium (used as a surrogate for sh plasma). The average wet weight of cunners used in our in vivo laboratory exposure experiments was approximately 35 g, which meant that a sh receiving 2 L of matrix per gram of wet weight would receive a bolus of 0.07 mL. We therefore chose a bolus volume of 0.07 mL for the in vitro study based on the 2-L/g dose received by a typical 35-g sh. Slow-release matrix was prepared in the same way as the estradiol implantation matrix described earlier for the in vivo experiment, except that 50 L of tritiated estradiol (0.99 curie/mmol in ethanol) were added to the Ethocel. Only a very
small amount of the total estradiol in each implant matrix was tritiated estradiol. After the solvents had dissipated, the slowrelease matrix was brought up to a nal volume of 5 mL with coconut oil and was homogenized thoroughly. A tuberculin syringe with the tip cut off was used to draw up 0.07 mL of slow-release matrix for each bolus. The syringe and matrix were chilled at 20 C for 48 h to solidify the matrix into a compact bolus. After 48 h, each chilled bolus was expelled from the syringe into a scintillation vial containing 2.0 mL of L-15 medium at 18 C. Scintillation vials were then incubated at 18 C and gently shaken on an orbit shaker (Lab-Line Instruments) until sampled. We prepared matrix at two dosage levels. The rst preparation contained 210 g of estradiol and 3.08 mg of Ethocel per bolus, resulting in an Ethocel: estradiol ratio of approximately 15:1. If implanted into a 35-g sh, this bolus would deliver an estradiol concentration of 6 g/g of sh wet weight. The second preparation contained 42 g of estradiol and 0.7 mg of Ethocel per bolus, resulting in an Ethocel : estradiol ratio of approximately 17:1. If implanted, this latter bolus would deliver an estradiol concentration of 1.2 g/g of sh wet weight. Tritiated estradiol released from boluses was measured at time 0; at 1, 3, and 6 h; and at 1, 2, 3, 4, 7, 9, and 11 d. Three replicates at each dosage concentration were sampled at each time period (n = 66). For sampling, a bolus was removed and 3 mL of Hionic-Fluor scintillation cocktail (PerkinElmer) were added to the remaining L-15 medium. Radioactivity in the medium was counted in disintegrations per minute (dpm) by using a Packard Tri-Carb Liquid Scintillation Analyzer (Model 2500TR; PerkinElmer). A new bolus was used for each sampling. The cumulative percentage of estradiol measured in the L-15 medium at each sampling time was calculated as 100 {[mean dpm of the three 3H-estradiol-spiked replicate samples at time t (d)]/[mean dpm at time 0]}. Analytical Chemistry Estrogens.Ethynylestradiol, estradiol, estrone, and estriol were extracted from the plasma samples before being analyzed by gas chromatography (GC)mass spectrometry (MS). The method used to extract estrogens from sh plasma was adapted from L opez de Alda and Barcel o (2001) and Jeannot et al. (2002), with some modications. Fish plasma samples were ground to a ne powder with anhydrous sodium sulfate. Three milliliters of diethyl ether and known amounts of the internal standard, 17-estradiol-d4 , were added to each sample. Samples were then vortexed (30 s), sonicated (10 min), and centrifuged (1,200 revolutions/min; 10 min). By use of disposable pipettes, the supernatants were carefully transferred into 2-dram screw-top vials. This extraction procedure was repeated two more times. The combined extracts were then volume reduced, solvent exchanged to methanol, and adjusted to a nal volume of 2 mL with a 30:70 solution of water: methanol. The extract mixture was passed through a C-18 cartridge (Sep-Pak Environmental tC18 cartridge WAT036800; Waters,
516
ZAROOGIAN ET AL.
Milford, Massachusetts) that was previously conditioned with water (5 mL) and methanol (5 mL). After the cartridge was dried under full vacuum for 1 min, estrogens were eluted with 8 mL of acetonitrile. The acetonitrile was then volume reduced and taken to dryness under a gentle stream of nitrogen. For derivitization, 25 L each of BSTFA and pyridine were added to each sample, which was then heated at 65 C for 25 min. The sample was allowed to cool and was vortexed; 2 L were then immediately injected into an Agilent 6890 gas chromatograph (Agilent Technologies, Santa Clara, California) that was tted with a mass spectrometer (5973N; Agilent) equipped with a 60-m DB-5ms capillary column (Agilent J&W Scientic). Further details of the analytical method are as described by Hernando et al. (2004). A standard curve was made by the addition of known amounts of ethynylestradiol, estradiol, and estrone that were brought to 1 mL with hexane. Method detection limits were 4.96, 8.50, and 11.4 ng/mL for estradiol, estrone, and ethynylestradiol, respectively. Only a single point calibration was performed for estriol; no method detection limit was determined. Atrazine.Plasma samples were extracted and analyzed for atrazine and the metabolites DACT, DIA, and DEA based on methods described by Brzezicki et al. (2003). Final extract vol-
ume was 0.5 mL. Two microliters of extract were injected into an Agilent 6890 gas chromatograph tted with an Agilent 5973N mass spectrometer that was equipped with a DB-17ms analytical column (Agilent J&W Scientic). Method detection limits were 0.53, 0.49, 0.23, and 0.16 ng/mL for atrazine, DACT, DIA, and DEA, respectively. RESULTS Estradiol Prior to treatment, no measurable estradiol, estrone, or estriol was found in the plasma of the two male cunners used for the estradiol experiment. Concentrations of estradiol and estrone detected in the plasma were summed to determine the total percentage of implanted estrogen that was in circulation at any given time. The concentration of estrogen measured in the cunner plasma after implantation with estradiol at 6 g/g of sh wet weight is shown in Table 1. Chemical analysis of plasma for estrogens revealed that in estradiol-treated sh, estrone made up 1927% of the total circulating estrogen by 3 h after estradiol implantation, and the estrone percentage increased to between 35% and 56% at subsequent sampling times. No estriol was detected in any plasma sample. The total concentration of estrogen
TABLE 1. Concentrations (ng/mL) of the test chemicals and their metabolites detected over time in plasma from male cunners that received the implanted slowrelease matrix. Percentage of the total is indicated in parentheses ( = indicates no sample; ND = no chemical was detected; DACT = diaminochlorotriazine; DIA = deisopropylatrazine; DEA = deethylatrazine).
Time after implantation Chemical or metabolite Fish Aa Estradiol Estrone Estriol Total Fish B Estradiol Estrone Estriol Total Ethynylestradiol Estrone Atrazine DACT DIA DEA Total
a
3h
1d
2d
3d
4d
5d
7d
8d
10 d
12 d
14 d
Estradiol Treatment (6 g/g of sh wet weight) 4,428 (73) 1,672 (27) ND 6,100 4,903 (81) 1,115 (19) ND 6,018 765 (56) 597 (44) ND 1,362 780 (59) 553 (41) ND 1,333 451 (53) 396 (47) ND 847 648 (52) 608 (48) ND 1,256 255 (53) 222 (47) ND 477 396 (65) 212 (35) ND 608 136 (44) 170 (56) ND 306 44.7 (46) 52.8 (54) ND 97.5 77.8 (64) 43.7 (36) ND 121.5 176 ND ND ND ND ND ND ND ND ND 224 ND 172 ND ND ND ND ND 77.3 ND
Ethynylestradiol Treatment (1.2 g/g of sh wet weight) 710 1,455 664 493 99 172 ND ND Atrazine Treatment (2.4 g/g sh wet weight) 17.9 (18) 2.4 (8) ND ND 15.1 (15) 18.1 (62) 6.0 (100) 11.9 (82) 34.6 (35) 4.8 (16) ND ND 30.9 (31) 3.8 (13) ND 2.6 (18) 98.5 29.1 6.0 14.5
Fish A was found dead on day 11.
517
FIGURE 1. Percentage of estradiol plus estrone measured in 1 mL of sh plasma over time relative to the total amount of estradiol implanted in two male cunners.
measured in the plasma of the two males was similar at every sampling time (Table 1). The percentage of estrogen (estradiol and estrone) in plasma at any given time was very low in comparison with the concentration of implanted estradiol; the maximum estrogen percentage (i.e.,%Fp ) of approximately 0.56% was observed at 3 h postimplantation (Figure 1). At 7 d postimplantation, the %Fp had dropped to an average of 0.01%, representing a plasma estrogen concentration of about 110 ng/mL. Plasma estrogen was not detectable by day 10. Since the method detection limit for estradiol was 4.96 ng/mL using GCMS, estrogen may have been present at concentrations below this detection limit in the day-10 samples. Ethynylestradiol No ethynylestradiol was detected in the plasma of cunners with control matrix implants. In sh that received the ethynylestradiol implant, a small concentration of estrone was detected on days 1 and 3 (Table 1); the commercial stock of ethynylestradiol that was used to prepare the implants was analyzed, and estrone was found to be a minor contaminant. The maximum concentration of ethynylestradiol (1,455 ng/mL of plasma) was observed at 3 d postimplantation and represented about 4% of the implanted concentration. On days 1, 5, and 7, circulating ethynylestradiol (%Fp ) was 1.7, 1.8, and 1.4%, respectively, of the implanted concentration (Figure 2). Ethynylestradiol was still detectable in plasma from treated cunners at 14 d after implantation (Table 1). Atrazine No atrazine was detected in the two males that received the control matrix. Atrazine metabolites made up a large percentage (82100%) of the total chemical in circulation (Table 1). At 5 and 7 d after atrazine implantation, only metabolites were still detectable in the plasma of treated sh. By 7 d postimplantation,
FIGURE 2. Percentage of ethynylestradiol plus estrone measured in 1 mL of sh plasma over time relative to the total amount of ethynylestradiol implanted in male cunners.
82% of the total chemical detected was DACT and the remaining 18% was DEA, whereas no atrazine or DIA was detected. The percentage of implanted atrazine that was in circulation at any given time (%Fp ; Figure 3) was calculated based on the total detected concentration of atrazine plus its three major metabolites (DACT, DIA, and DEA). The maximum %Fp was 0.09%, which was measured in plasma sampled at 1 d postimplantation. By day 3, the percentage had dropped to 0.03%; on days 5 and 7,%Fp was further reduced to 0.006% and 0.017%, respectively (Figure 3). In Vitro Study Within the rst 24 h of incubation, there was a rapid release of estradiol from the boluses, averaging 9.2% of the estradiol in the
FIGURE 3. Percentage of atrazine plus its metabolites (diaminochlorotriazine, deisopropylatrazine, and deethylatrazine) measured in 1 mL of sh plasma over time relative to the total amount of atrazine implanted in male cunners.
518
ZAROOGIAN ET AL.
TABLE 2. Mean (SD in parentheses) cumulative percentage of estradiol measured in L-15 medium from slow-release matrix boluses over time during the in vitro study.
Time 1h 3h 6h 1d 2d 3d 5d 7d 9d 11 d
Percentage of 6-g/g bolus 2.5 (0.4) 3.9 (0.3) 5.9 (1.7) 9.2 (2.8) 11.2 (0.6) 10.6 (0.4) 13.3 (0.5) 13.7 (0.9) 15.1 (2.4) 15.2 (0.9)
Percentage of 1.2-g/g bolus 3.7 (0.3) 8.9 (0.8) 11.9 (1.6) 17.5 (2.5) 21.6 (0.7) 24.6 (0.4) 32.4 (5.4) 35.0 (4.5) 31.6 (6.8) 30.0 (1.5)
FIGURE 5. Cumulative percentage of total estrogen (estradiol plus estrone) in the plasma of two male cunners that received the estradiol treatment (6-g/g dosage) during the in vivo study. A logarithmic trend line is plotted for reference.
6.0-g/g boluses and 17.5% of the estradiol in the 1.2-g/g boluses (Table 2). This rate of release slowed appreciably after the rst day to a more-gradual average release (over the next 10 d) of 0.6% per day from the 6.0-g/g boluses and 1.3% per day from the 1.2-g/g boluses. Release was approximately twice as rapid for the 1.2-g/g boluses, in which the Ethocel amount was only about one-fth of that added to the 6.0-g/g boluses (Table 2). A visual comparison of the data points and logarithmic trend lines in Figures 4 and 5 shows that the patterns of estradiol release from the boluses into sh plasma and L-15 medium were quite similar, with a sharp increase occurring during the rst 24 h, followed by a more gradual release for the rest of the observation period. However, much greater concentrations of total estrogen were retrieved from the L-15 mediumapproximately
an order of magnitude greater than the concentrations in the cunner plasma. DISCUSSION A number of implant systems (reviewed by Mylonas and Zohar 2000) have been developed for delivering a sustained dose of hormone to large sh in captivity. These systems may be adaptable to administering other chemicals of interest as well. A currently available example for hormone administration is the commercial product Ovaplant (Western Chemical), which is implanted by use of the commercially available Ralogun pellet injector (Schering-Plough Animal Health). However, the manufacture of products like these requires technical expertise and specialized equipment (Mylonas and Zohar 2000). We were interested in developing an injectable implant that would be inexpensive yet practical to prepare and administer in the laboratory. Ideally, the implant would release a small amount of test chemical into a shs circulation over a period of weeks and could be used for a wide variety of chemicals. In addition, we wanted to accurately deliver the test chemical based on sh weight while causing only minimal trauma to the sh; thus, we surmised that a liquid injectable delivery method was the best choice. Our earlier experiments with juvenile summer ounder indicated that coconut oil was a good matrix carrier because it could be injected as a lukewarm liquid that would quickly solidify (Mills et al. 2001). To ensure that the release of the selected test chemical was slow enough, we opted to mix an ethyl cellulose polymer, Ethocel, into our coconut oil implant. Ethocel is used in pharmaceutical products that require controlled-release dosage formulations (Dow Chemical 2005). The resulting implant preparation of Ethocel and test chemical in a coconut oil carrier was relatively simple to prepare and easy to inject, and it left a bolus that was still discernable when sh were necropsied
FIGURE 4. Cumulative percentage of estradiol released (based on the radioactivity, in disintegrations per minute [dpm], of tritiated estradiol) as measured in 1 mL of L-15 medium relative to the total amount of estradiol in the 6-g/g and 1.2-g/g boluses during the in vitro study. Logarithmic trend lines for each dosage are plotted for reference.
519
at 2 weeks postimplantation. Although we acknowledge that our in vivo experiment results could have been enhanced by increasing the number of study sh, the results presented here show that estradiol, ethynylestradiol, atrazine, and the associated metabolites were detectable in the plasma of treated sh for 714 d after initial implantation. This indicates that our implant is an effective method for the administration of these chemicals for up to 2 weeks. Further in vivo experimentation would be necessary to determine whether, with modication, the implant can deliver chemical for a longer period of time. In Vivo Experiments The similar concentrations of circulating estrogen measured in the two individual cunners that received estradiol at 6 g/g indicate that (1) there is a reasonably even distribution of test chemical within the Ethocelcoconut oil preparation and (2) individual variability is apparently minimal for the release of test chemical from the implant. Because estradiol was implanted at the highest total test concentration (6 g/g) among the three chemicals, we expected that plasma concentrations would be higher for estradiol than for the other chemicals. This was borne out: the average plasma concentration of chemical in the two estradiol-implanted sh was approximately 1,350-ng/mL estrogen at 1 d after implantation compared with 710 ng/mL for the ethynylestradiol-implanted sh and 105 ng/mL for the atrazine-implanted sh (Table 1). Because the ratio of Ethocel to test chemical was lower in the estradiol implant (17:1) than in the ethynylestradiol and atrazine implants (both 40:1), we also expected that the percentage of estradiol detected in circulation might be higher than the detected levels of the other chemicals. However, the percentage of estrogen (estradiol plus its metabolite, estrone) measured in plasma from both of the estradiol-implanted sh on day 1 after implantation was just above 0.1%approximately equal to the atrazine percentage (0.1%) and lower than the ethynylestradiol percentage (1.7%). Rapid metabolism and clearance of estradiol from the plasma of the sh provide a possible explanation. Our data show that at 3 h after estradiol implantation, 1927% of the total circulating estrogen in estradiol-treated male cunners was the metabolite estrone, and the estrone percentage increased to 3556% of the circulating estrogen in plasma at subsequent sampling times. On day 7, estrogen concentrations were still elevated in comparison with normal physiological concentrations in male sh, which are in the picograms per milliliter range (Miura et al. 1999; Amer et al. 2001; Geraudie et al. 2010). Specker and Chandlee (2003) reported that within 1 d, larval and juvenile summer ounder depurated estradiol taken up from water; Zohar (1982) demonstrated that estradiol had a half-life of less than 30 min in the bloodstream of rainbow trout. We might expect the release of estradiol from the Ethocelcoconut oil bolus to be initially rapid as the reservoir of steroid in zones nearest to the surface of the bolus diffuse into the bloodstream. Remaining steroid must diffuse further to reach the surface of the bolus, so that the release rate slows over time until it becomes
lower than the clearance rate. A decreasing rate of release from the bolus, in combination with metabolism and clearance by the sh, may partially explain why estrogen was undetectable in the plasma of treated cunners by day 10 postimplantation. In contrast to the apparently rapid rate of estradiol metabolism and clearance, the xenoestrogen ethynylestradiol appeared to be neither metabolized nor rapidly excreted. Ethynylestradiol reached its peak concentration in cunner plasma (i.e., %Fp = 4%) on day 3 after implantation, indicating that enterohepatic recirculation of this xenoestrogen, as documented in rainbow trout (Schultz et al. 2001), could also be occurring in cunners. Schultz et al. (2001) found that after injection, ethynylestradiol was extensively conjugated and secreted into the bile of treated rainbow trout. When the gall bladder emptied, the stored ethynylestradiol was released into the gut, where most was deconjugated and reabsorbed, effectively redosing the rainbow trout with ethynylestradiol. Our data suggest that a similar phenomenon took place in cunners. The concentration of ethynylestradiol measured in the circulation of cunners more than doubled from day 1 to day 3 postimplantation. In contrast, circulating levels of estradiol and atrazine from implants were characterized by an initial burst within the rst day after implantation, followed by a decline thereafter. This initial burst of chemical is also characteristic of other implant delivery systems (Mylonas and Zohar 2000). Interestingly, estrone, a contaminant in our ethynylestradiol stock, was detectable only in plasma samples taken at 1 and 3 d after implantation, thus lending further credence to the premise that natural estrogens are cleared from the circulation of cunners much faster than the xenoestrogen ethynylestradiol. Metabolism appears to be a major consideration in the case of atrazine implantation. On day 1 postimplantation, approximately 81% of the circulating chemical from the implant was in the form of three atrazine metabolites: DACT, DIA, and DEA. If we had measured only the concentration of atrazine (not metabolites) in the plasma of cunners, none would have been detected after day 3. By including the major metabolites in our chemical analysis, we were able to detect the atrazine metabolites in plasma samples up to day 7 after implantation. Detection of metabolites and of the parent compound is a major advantage of using GCMS to quantify the release of chemical from an implant. In Vitro Study In vitro, maximum concentrations of estrogen measured in the L-15 medium surrounding the boluses were much higher than those measured in sh plasma during the in vivo exposures (9.2% in vitro versus 0.56% in vivo). These results indicate that measurement of in vivo plasma estrogen concentrations does not capture the actual release rate from an implanted bolus because uptake, metabolism, and clearance rates all contribute to the plasma concentrations observed in vivo. Comparing the release rates of estrogen between in vitro preparations, a higher maximum release rate was observed in
520
ZAROOGIAN ET AL.
the 1.2-g/g boluses (17.5%) than in the 6-g/g boluses (9.2%). The higher release rate is likely related to the smaller amount of Ethocel added to the matrix in the 1.2-g/g preparation (0.7 mg of Ethocel/bolus) relative to the 6-g/g preparation (3.08 mg of Ethocel/bolus). Because Ethocel forms a lm (Dow Chemical 2005) that reduces diffusion, increasing the amount of Ethocel in the matrix would be expected to result in a slower release rate of test chemical, just as we observed. The behavior of Ethocel in our implants is in direct contrast to that reported for cellulose. Sherwood et al. (1988) observed higher release rates of GnRHa with increased amounts of cellulose in cellulosecholesterol pellets. Sherwood et al. (1988) hypothesized that the cellulose expanded and became soft upon hydration, providing diffusion channels and promoting degradation of the matrix. However, water-insoluble Ethocel forms a lm that reduces diffusion, and our results indicate that diffusion rates can be manipulated by increasing or decreasing the amount of Ethocel in a matrix preparation. Conclusions We have developed a long-lasting, easily prepared, injectable implant consisting of a test chemical in an Ethocelcoconut oil matrix. This matrix can be used in sh to deliver chemicals for release over a period of 1 week or more. Our implants eliminate the need for a pellet press, pellet injector implant gun, or surgical incisions. Implants are deposited subcutaneously, far away from sensitive organs in the peritoneal cavity; they are soft and pliable, not brittle or crumbly, and are easily injected by using a syringe. We used the implant to conduct sh reproductive studies with endocrine-disrupting chemicals, but this injectable implant may also provide aquaculturists with an inexpensive, simple alternative to current procedures used for administering hormones or other treatments to sh. In cunners, estradiol and atrazine demonstrated much different metabolism and clearance proles than ethynylestradiol. This information suggests that sh plasma concentrations do not always represent the true release rate of a test chemical from the implant matrix due to metabolism, clearance, or recirculation of the chemical in the intact animal. ACKNOWLEDGMENTS We thank Martha Simoneau for revising, editing, proofreading, and formatting this manuscript. We also thank Richard J. Pruell for supplying technical advice during the research; Alycia Collins for providing technical assistance; and Roxanne Johnson, Janet Nye, Joseph LiVolsi, and Bryan Taplin for providing valuable comments on an earlier version of the manuscript. This research would not have been possible without the support and encouragement given by Suzanne Ayvazian. The manuscript has been reviewed and approved for publication by the USEPA, Ofce of Research and Development, NHEERL, Atlantic Ecology Division as Contribution Number AED-11-089. Approval does not signify that the contents of this report necessarily reect the views and policies of the USEPA. Mention of trade
names or commercial products does not constitute endorsement or recommendation for use by the USEPA. REFERENCES
Amer, M. A., T. Miura, C. Miura, and K. Yamauchi. 2001. Involvement of sex steroid hormones in the early stages of spermatogenesis in Japanese huchen (Hucho perryi). Biology of Reproduction 65:10571066. Arslan, T., and R. P. Phelps. 2004. Production of monosex male black crappie, Pomoxis nigromaculatus, populations by multiple androgen immersion. Aquaculture 234:561573. Bayley, M., M. Junge, and E. Baatrup. 2002. Exposure of juvenile guppies to three antiandrogens causes demasculinization and a reduced sperm count in adult males. Aquatic Toxicology 56:227239. Bisbal, G. A., and J. L. Specker. 1991. Cortisol stimulates hypo-osmoregulatory ability in Atlantic salmon, Salmo salar L. Journal of Fish Biology 39: 421432. Black, D. E., R. Gutjahr-Gobell, R. J. Pruell, B. Bergen, and A. E. McElroy. 1998. Effects of a mixture of non-ortho- and mono-ortho-polychlorinated biphenyls on reproduction in Fundulus heteroclitus (Linnaeus). Environmental Toxicology and Chemistry 17:13961404. Brzezicki, J. M., M. E. Andersen, B. K. Cranmer, and J. D. Tessari. 2003. Quantitative identication of atrazine and its chlorinated metabolites in plasma. Journal of Analytical Toxicology 27:569573. Carolsfeld, J., N. M. Sherwood, H. Kreiberg, and S. A. Sower. 1988. Induced sexual maturation of herring using GnRH quick-release cholesterol pellets. Aquaculture 70:169181. Crim, L. W., N. M. Sherwood, and C. E. Wilson. 1988. Sustained hormone release: II. effectiveness of LHRH analog (LHRHa) administration by either single time injection or cholesterol pellet implantation on plasma gonadotropin levels in a bioassay model sh, the juvenile rainbow trout. Aquaculture 74:8795. Cyr, D. G., and J. G. Eales. 1989. Effects of short-term 17-estradiol treatment on the properties of T3 -binding proteins in the plasma of immature rainbow trout, Salmo gairdneri. Journal of Experimental Zoology 252:245251. Donohoe, R. M., and L. R. Curtis. 1996. Estrogenic activity of chlordecone, o,p -DDT and o,p -DDE in juvenile rainbow trout: induction of vitellogenesis and interaction with hepatic estrogen binding sites. Aquatic Toxicology 36: 3152. Dow Chemical. 2005. Dow cellulosics: ETHOCELTM ethylcellulose polymers technical handbook. Dow Chemical, Midland, Michigan. Available: www.dow.com/dowwolff/en/pdfs/192-00818.pdf. (January 2012). Garcia, L. M. B. 1989. Dose-dependent spawning response of mature female sea bass, Lates calcarifer (Bloch), to pelleted luteinizing hormone-releasing hormone analogue (LHRHa). Aquaculture 77:8596. Geraudie, P., M. Gerbron, and C. Minier. 2010. Seasonal variations and alterations of sex steroid levels during the reproductive cycle of male roach (Rutilis rutilis). Marine Environmental Research 69(Supplement 1):5355. Gimeno, S., H. Komen, S. Jobling, J. Sumpter, and T. Bowmer. 1998. Demasculinisation of sexually mature male common carp, Cyprinus carpio, exposed to 4-tert-pentylphenol during spermatogenesis. Aquatic Toxicology 43:93109. Gonc alves, D., J. Alpedrinha, M. Teles, and R. F. Oliveira. 2007. Endocrine control of sexual behavior in sneaker males of the peacock blenny Salaria pavo: effects of castration, aromatase inhibition, testosterone and estradiol. Hormones and Behavior 51:534541. Gutjahr-Gobell, R. E., D. E. Black, L. J. Mills, R. J. Pruell, B. K. Taplin, and S. Jayaraman. 1999. Feeding the mummichog (Fundulus heteroclitus) a diet spiked with non-ortho- and mono-ortho-substituted polychlorinated biphenyls: accumulation and effects. Environmental Toxicology and Chemistry 18:699707. Gutjahr-Gobell, R. E., M. Huber, D. J. Borsay Horowitz, G. E. Zaroogian, and L. J. Mills. 2002. A temperate reef sh, Tautogolabrus adspersus (Walbaum), as a potential model species for laboratory studies evaluating
SLOW-RELEASE METHOD FOR EXPOSING FISH TO CHEMICALS reproductive effects of chemical exposure. Environmental Toxicology and Chemistry 21:380389. Hernando, M. D., M. J. Mezcua, M. J. G omez, O. Malato, A. Ag uera, and A. R. Fern andez-Alba. 2004. Comparative study of analytical methods involving gas chromatographymass spectrometry after derivatization and gas chromatographytandem mass spectrometry for the determination of selected endocrine disrupting compounds in wastewaters. Journal of Chromatography A 1047:129135. Hunter, G. A., and E. M. Donaldson. 1983. Hormonal sex control and its application to sh culture. Pages 223303 in W. S. Hoar, D. J. Randall, and E. M. Donaldson, editors. Fish physiology, volume 9B. Academic Press, New York. Jeannot, R., H. Sabik, E. Sauvard, T. Dagnac, and K. Dohrendorf. 2002. Determination of endocrine-disrupting compounds in environmental samples using gas and liquid chromatography with mass spectrometry. Journal of Chromatography A 974:143159. Joakim Larsson, D. G., T. S. Sperry, and P. Thomas. 2002. Regulation of androgen receptors in Atlantic croaker brains by testosterone and estradiol. General and Comparative Endocrinology 128:224230. Keen, P. L., D. A. Higgs, K. J. Hall, and M. Ikonomou. 2005. Effects of dietary exposure of 4-nonylphenol on growth and smoltication of juvenile coho salmon (Oncorhynchus kisutch). Science of the Total Environment 349: 8194. Kramer, V. J., S. Miles-Richardson, S. L. Pierens, and J. P. Giesy. 1998. Reproductive impairment and induction of alkaline-labile phosphate, a biomarker of estrogen exposure, in fathead minnows (Pimephales promelas) exposed to waterborne 17-estradiol. Aquatic Toxicology 40:335360. Lahnsteiner, F., B. Berger, M. Kletzl, and T. Weismann. 2006. Effect of 17estradiol on gamete quality and maturation in two salmonid species. Aquatic Toxicology 79:124131. Leatherland, J. F. 1985. Effects of 17-estradiol and methyl testosterone on the activity of the thyroid gland in rainbow trout, Salmo gairdneri Richardson. General and Comparative Endocrinology 60:343352. Lee, C. S., C. S. Tamaru, and C. D. Kelley. 1986. Technique for making chronicrelease LHRH-a and 17-methyltestosterone pellets for intramuscular implantation in shes. Aquaculture 59:161168. L opez de Alda, M. J., and D. Barcel o. 2001. Use of solid-phase extraction in various of its modalities for sample preparation in the determination of estrogens and progestogens in sediment and water. Journal of Chromatography A 938:145153. Madsen, L. L., B. Korsgaard, and P. Bjerregaard. 2003. Estrogenic effects in ounder Platichthys esus orally exposed to 4-tert-octylphenol. Aquatic Toxicology 64:393405. Mandiki, S. N. M., M. B. Houbart, I. Babiak, E. Vandeloise, J. N. Gardeur, and P. Kestmont. 2004. Are sex steroids involved in the sexual growth dimorphism in Eurasian perch juveniles? Physiology and Behavior 80:603609. Metcalfe, C. D., T. L. Metcalfe, Y. Kiparissis, B. G. Koenig, C. Khan, R. J. Hughes, T. R. Croley, R. E. March, and T. Potter. 2001. Estrogenic potency of chemicals detected in sewage treatment plant efuents as determined by in vivo assays with Japanese medaka (Oryzias latipes). Environmental Toxicology and Chemistry 20:297308. Mills, L. J., R. E. Gutjahr-Gobell, R. A. Haebler, D. J. Borsay Horowitz, S. Jayaraman, R. J. Pruell, R. A. McKinney, G. R. Gardner, and G. E. Zaroogian. 2001. Effects of estrogenic (o,p-DDT; octylphenol) and anti-androgenic (p,p-DDE) chemicals on indicators of endocrine status in juvenile male summer ounder (Paralichthys dentatus). Aquatic Toxicology 52:157176. Miura, T., C. Miura, T. Ohta, M. R. Nader, T. Todo, and K. Yamauchi. 1999. Estradiol-17 stimulates the renewal of spermatogonial stem cells in males. Biochemical and Biophysical Research Communications 264:230234. Mylonas, C. C., and Y. Zohar. 2000. Use of GnRHa-delivery systems for the control of reproduction in sh. Reviews in Fish Biology and Fisheries 10: 463491.
521
Pandian, T. J., and S. G. Sheela. 1995. Hormonal induction of sex reversal in sh. Aquaculture 138:122. Pankhurst, N. W., N. E. Stacey, and R. E. Peter. 1986. An evaluation of techniques for the administration of 17-estradiol to teleosts. Aquaculture 52:145155. Patyna, P. J., R. A. Davi, T. F. Parkerton, R. P. Brown, and K. R. Cooper. 1999. A proposed multigeneration protocol for Japanese medaka (Oryzias latipes) to evaluate effects of endocrine disruptors. Science of the Total Environment 233:211220. Pickering, A. D., and J. Duston. 1983. Administration of cortisol to brown trout, Salmo trutta L., and its effects on the susceptibility to Saprolegnia infection and furunculosis. Journal of Fish Biology 23:163175. Pizza, J. C., and J. M. OConnor. 1983. PCB dynamics in Hudson River striped bass: II. accumulation from dietary sources. Aquatic Toxicology 3:313327. Sangiao-Alvarellos, S., J. M. M guez, and J. L. Soengas. 2005. Actions of growth hormone on carbohydrate metabolism and osmoregulation of rainbow trout (Oncorhynchus mykiss). General and Comparative Endocrinology 141: 214225. Schultz, I. R., G. Orner, J. L. Merdink, and A. Skillman. 2001. Doseresponse relationships and pharmacokinetics of vitellogenin in rainbow trout after intravascular administration of 17-ethynylestradiol. Aquatic Toxicology 51:305318. Scott, A. P., P. R. Witthames, E. L. M. Vermeirssen, and J. Carolsfeld. 1999. Prolonged-release gonadotrophin-releasing hormone analogue implants enhance oocyte nal maturation and ovulation, and increase plasma concentrations of sulphated C21 steroids in North Sea plaice. Journal of Fish Biology 55:316328. Segner, H., K. Caroll, M. Fenske, C. R. Janssen, G. Maack, D. Pascoe, C. Sch afers, G. F. Vandenbergh, M. Watts, and A. Wenzel. 2003. Identication of endocrine-disrupting effects in aquatic vertebrates and invertebrates: report from the European IDEA project. Ecotoxicology and Environmental Safety 54:302314. Shelton, W. L., and S. D. Mims. 2003. Fabrication of silastic implants for in vivo steroid delivery in sh. North American Journal of Aquaculture 65:158161. Sherwood, N. M., L. W. Crim, J. Carolsfeld, and S. M. Walters. 1988. Sustained hormone release: I. characteristics of in vitro release of gonadotropinreleasing hormone analogue (GnRH-A) from pellets. Aquaculture 74: 7586. Specker, J. L., and M. K. Chandlee. 2003. Methodology for estradiol treatment in marine larval and juvenile sh: uptake and clearance in summer ounder. Aquaculture 217:663672. Sundararaj, B. I., and S. V. Goswami. 1968. Effects of estrogen, progesterone, and testosterone on the pituitary and ovary of catsh, Heteropneustes fossilis (Bloch). Journal of Experimental Zoology 169:211227. Trudeau, V. L., R. E. Peter, and B. D. Sloley. 1991. Testosterone and estradiol potentiate the serum gonadotropin response to gonadotropin-releasing hormone in goldsh. Biology of Reproduction 44:951960. Wang, W. B., A. H. Li, T. Z. Cai, and J. G. Wang. 2005. Effects of intraperitoneal injection of cortisol on non-specic immune functions of Ctenopharyngodon idella. Journal of Fish Biology 67:779793. Yamada, H., R. Satoh, T. Yamashita, A. Kambegawa, and M. Iwata. 1997. Development of a time-resolved uoroimmunoassay (TR-FIA) for testosterone: measurement of serum testosterone concentrations after testosterone treatment in the rainbow trout (Oncorhynchus mykiss). General and Comparative Endocrinology 106:181188. Yamaguchi, S., H. Kagawa, K. Gen, K. Okuzawa, and M. Matsuyama. 2004. Silicone implants for delivery of estradiol-17 and 11-ketotestosterone to red seabream Pagrus major. Aquaculture 239:485496. Zohar, Y. 1982. L evolution de la pulsatilit e et des cycles nycth em erau de la s ecr etion gonadotrope chez la truite arc-en-ciel femelle en relation avec le cycle sexuel annuel et par rapport a lactivit e st eroidog` ene de lovaire. Doctoral dissertation. Universit e Pierre et Marie Curie, Paris.

Characterization of Greater Amberjack Microsatellite Markers in Lesser Amberjacks, Yellowtail Jacks, Almaco Jacks, and Banded Rudderfish
Mark A. Renshaw , Alejandro Buentello & John R. Gold
a a b a
Center for Biosystematics and Biodiversity, Texas A&M University, College Station, Texas, 77843-2258, USA
b
Schillinger Genetics, 4401 Westown Parkway, Suite 225, West Des Moines, Iowa, 50266, USA
Version of record first published: 14 Sep 2012.
To cite this article: Mark A. Renshaw, Alejandro Buentello & John R. Gold (2012): Characterization of Greater Amberjack Microsatellite Markers in Lesser Amberjacks, Yellowtail Jacks, Almaco Jacks, and Banded Rudderfish, North American Journal of Aquaculture, 74:4, 522-529 To link to this article: http://dx.doi.org/10.1080/15222055.2012.686959
TECHNICAL NOTE
Characterization of Greater Amberjack Microsatellite Markers in Lesser Amberjacks, Yellowtail Jacks, Almaco Jacks, and Banded Ruddersh
Mark A. Renshaw*
Center for Biosystematics and Biodiversity, Texas A&M University, College Station, Texas 77843-2258, USA
Alejandro Buentello
Schillinger Genetics, 4401 Westown Parkway, Suite 225, West Des Moines, Iowa 50266, USA
John R. Gold
Center for Biosystematics and Biodiversity, Texas A&M University, College Station, Texas 77843-2258, USA
Abstract
Thirty-one microsatellite markers that were previously isolated from and characterized in greater amberjacks Seriola dumerili were assayed for cross-species amplication in four other members of the carangid genus Seriola: the lesser amberjack S. fasciata, yellowtail jack S. lalandi, almaco jack S. rivoliana, and banded ruddersh S. zonata. The number of markers that consistently amplied and were polymorphic ranged from 16 in yellowtail jacks to 25 in lesser amberjacks. The microsatellites characterized in this study will be useful for a variety of applications, including stock structure assessments of wild sh and parentage assignments of farmed sh.
Five species of the carangid genus Seriola support commercial and recreational sheries along the coast of the mainland United States: the greater amberjack S. dumerili, lesser amberjack S. fasciata, yellowtail jack S. lalandi, almaco jack S. rivoliana, and banded ruddersh S. zonata. Three of the species (greater amberjack, yellowtail jack, and almaco jack) also are important components of commercial aquaculture production. The commercial aquaculture of greater amberjacks and yellowtail jacks is currently being researched or is already in progress in Japan, Australia, New Zealand, and a number of Mediterranean countries, including Italy, Spain, Malta, and Greece (Repulles-Albelda et al. 2008; Hamasaki et al. 2009;
*Corresponding author: mrenshaw@nd.edu Received February 7, 2012; accepted April 16, 2012
Stuart and Drawbridge, in press). In the United States, almaco jacks are produced commercially in Hawaii by using offshore aquaculture cages (Simpson 2011), and the production of yellowtail jacks is currently under research in California by utilizing broodsh that are collected locally from the wild (Stuart and Drawbridge, in press). Nuclear-encoded microsatellites are useful markers for elucidating stock structure of wild sh (Miller et al. 2011) and can contribute to sheries management decisions (Reiss et al. 2009). Microsatellites also are valuable for use in aquaculture programs, as they can serve as tools for parentage assignments (Gold et al. 2010), species authentication (Iguchi et al. 2012), and marker-assisted selection in farmed sh (Liu and Cordes 2004). In this paper, we evaluate the amplication of 31 microsatellites developed from the genomic DNA of greater amberjacks for cross-amplication in lesser amberjacks, yellowtail jacks, almaco jacks, and banded ruddersh.
METHODS Ninety-six sh were assayed in this study: 19 lesser amberjacks (all from Johns Island, South Carolina), 29 yellowtail jacks (all from San Diego, California), 30 almaco jacks (all from Johns Island), and 18 banded ruddersh (7 from Johns Island; 11 from Panama City, Florida). Fin clips were taken and placed in a 95% solution of ethanol (Florida
522
TECHNICAL NOTE
523
samples), sarkosylurea (South Carolina samples), or a 20% solution of DMSO (California samples). Genomic DNA was extracted using a standard phenylchloroform protocol. The PCR primer sequences followed those given by Renshaw et al. (2006, 2007) for microsatellites isolated from the DNA of greater amberjacks. Unlabeled and uorescently labeled primers (6-FAM and HEX) were obtained from Integrated DNA Technologies; primers that were uorescently labeled with NED were obtained from Applied Biosystems, Inc. (ABI). The uorescent label was attached to one of the primers from each microsatellite marker pair (Table 1). The PCR amplications were performed in 10-L reactions by using 1 L of DNA, 1 Colorless GoTaq Flexi Buffer (Promega), 2-mM MgCl2 , 200 m of each dNTP, 5 pmol of each primer (forward and reverse), and 0.5 unit of GoTaq Flexi DNA Polymerase (Promega). Cycling conditions employed (1) an initial denaturation step at 95 C for 3 min; (2) seven cycles with denaturation at 95 C for 30 s, rst annealing temperature (dened below) for 45 s, and extension at 72 C for 1 min; (3) seven cycles with denaturation at 95 C for 30 s, second annealing temperature for 45 s, and extension at 72 C for 1 min; (4) 38 cycles with denaturation at 95 C for 30 s, third annealing temperature for 45 s, and extension at 72 C for 1 min; and (5) a nal extension step at 72 C for 10 min. The rst annealing temperature was the one listed by Renshaw et al. (2006, 2007), the second annealing temperature was 2 C lower than the rst annealing temperature, and the third annealing temperature was 4 C lower than the rst annealing temperature (Table 1). Amplied PCR products were run on an ABI 377 Automated Sequencer. Allele sizes were estimated with the GeneScan 400HD ROX Size Standard (ABI); allele sizes were determined using GeneScan version 3.1.2 and Genotyper version 2.5 (ABI). Genetic variability of each microsatellite marker was measured by the number of alleles, gene diversity (expected heterozygosity HE ), and observed heterozygosity (HO ) as calculated in Genetic Data Analysis (GDA) software (Lewis and Zaykin 2001). Fishers exact tests as implemented in GDA were used to test for signicant departures from HardyWeinberg expectations at individual microsatellites and for departures from genotypic equilibrium at pairs of microsatellites; Bonferroni corrections for multiple tests (Rice 1989) were applied manually to the P-value outputs from GDA. MicroChecker software (Van Oosterhout et al. 2004) was used to test for evidence of null alleles, scoring errors due to stuttering, and scoring errors due to large-allele dropout at each microsatellite marker.
RESULTS AND DISCUSSION Of the 31 microsatellites that were assayed, only Sdu23 did not consistently amplify for at least one of the four species. Summary data for the remaining 30 microsatellites are presented in Table 1; previously published data for greater amberjacks
(Renshaw et al. 2006, 2007) are also included in Table 1 to facilitate comparisons across species. In total, 25 markers were polymorphic for lesser amberjacks, and the number of alleles ranged from 2 to 16; HE ranged from 0.053 to 0.930, while HO ranged from 0.053 to 0.947. After Bonferroni correction for multiple tests (Rice 1989), genotypes at all markers conformed to HardyWeinberg expectations and all pairs of microsatellites were in genotypic equilibrium; analysis with Micro-Checker indicated no evident issues. Sixteen markers were polymorphic for yellowtail jacks, with the number of alleles ranging from 2 to 14. The HE ranged from 0.100 to 0.859, and HO ranged from 0.103 to 1.000. After Bonferroni correction, genotypes at two markers (Sdu29 and Sdu46) deviated signicantly from HardyWeinberg expectations and 17 microsatellite pairs deviated signicantly from genotypic equilibrium (Sdu29Sdu1, Sdu29Sdu2, Sdu29Sdu4, Sdu29 Sdu10, Sdu29Sdu19, Sdu29Sdu21, Sdu29Sdu31, Sdu29 Sdu32, Sdu29Sdu33, Sdu29Sdu37 , Sdu29Sdu39, Sdu29 Sdu40, Sdu29Sdu43, Sdu29Sdu44, Sdu29Sdu46, Sdu46 Sdu4, and Sdu46Sdu19). Analysis with Micro-Checker indicated the possibility of null alleles at Sdu29. Twenty-three markers were polymorphic for almaco jacks, and the number of alleles ranged from 2 to 22; HE ranged from 0.033 to 0.951, while HO ranged from 0.033 to 0.933. After Bonferroni correction, genotypes at all markers conformed to HardyWeinberg expectations and all pairs of microsatellites were in genotypic equilibrium. Analysis with Micro-Checker indicated the possibility of null alleles at Sdu10. Eighteen markers were polymorphic for banded ruddersh. The number of alleles ranged from 2 to 17, HE ranged from 0.203 to 0.954, and HO ranged from 0.111 to 1.000. After Bonferroni correction, genotypes at two markers (Sdu12 and Sdu31) deviated signicantly from HardyWeinberg expectations and one microsatellite pair (Sdu31Sdu43) deviated signicantly from genotypic equilibrium. Analysis with Micro-Checker indicated the possibility of null alleles at both Sdu12 and Sdu31. For sheries managers, the 30 microsatellites characterized in the present study can be used as tools for assessing stock structure and will provide valuable population genetic indices within each species. The markers will also facilitate the improvement of aquaculture programs through a variety of applications. For aquaculture facilities producing larvae that are subsequently grown for future sale, parentage assignments (Gold et al. 2010) can identify broodstock contribution and can be used to maximize mating design and output. Differences in microsatellite allele sizes between species can be used to authenticate the labeling of aquaculture products as they go to market (Iguchi et al. 2012). As part of a larger set of markers, these microsatellites can be employed for the mapping of quantitative trait loci and the marker-assisted selection of future broodsh (Liu and Cordes 2004).
524
Primer sequence (5 3 )a* T Ab Spc 53 51 49 N, NAd HE , H O f (GACA)20 Repeat sequence* Size rangee PHW g MicroCheckerh (GAA)8 50 48 46 (CAA)8 53 51 49 (GACA)8 (GGCA)4 (GACA)6 60 58 56 (TAA)3 CAA(TAA)10 CAA(TAA)5 56 54 52 (GATA)12 50 48 46 Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu 19, 4 29, 7 30, 8 N/A 28, 11 N/A 29, 3 30, 5 18, 3 29, 5 19, 2 N/A 30, 4 N/A 29, 5 N/A 29, 3 30, 5 N/A 29, 10 19, 13 N/A 30, 1 18, 13 23, 5 19, 10 N/A 30, 11 N/A 29, 12 289309 352420 306362 N/A 305385 N/A 135147 129144 141150 145160 208211 N/A 214223 N/A 212227 N/A 284300 281287xx N/A 320360 250295 N/A 202 203257 206236 227267 N/A 215259 N/A 231279 0.605, 0.789 0.808, 0.897 0.403, 0.333 N/A 0.751, 0.714 N/A 0.327, 0.345 0.532, 0.433 0.252, 0.278 0.716, 0.670 0.444, 0.526 N/A 0.715, 0.800 N/A 0.602, 0.690 N/A 0.493, 0.448 0.275, 0.267 N/A 0.808, 0.897 0.899, 0.947 N/A 0.000, 0.000 0.921, 0.889 0.560, 0.478 0.892, 0.842 N/A 0.860, 0.767 N/A 0.844, 0.759 0.0363 0.0103 0.1334 N/A N/A 0.252 N/A N/A 0.4566 0.1278 1.0000 0.918 0.5997 N/A N/A 0.5003 N/A N/A 0.717 N/A N/A 0.6056 0.5084 N/A N/A 0.972 0.8891 N/A N/A 1.0000 0.3231 0.390 0.0309 N/A N/A 0.1031 N/A N/A 0.043 (Continued on next page)
TABLE 1. Summary data for 30 greater amberjack microsatellite markers that were characterized in lesser amberjacks, yellowtail jacks, almaco jacks, and banded ruddersh. The uorescently labeled primer is in bold text, with the appropriate label signied by the number of plus signs (+ = 6-FAM; ++ = HEX; +++ = NED). Asterisks indicate information that was taken from earlier descriptions of these microsatellites (Renshaw et al. 2006 for Sdu1Sdu27 ; Renshaw et al. 2007 for Sdu29Sdu46); all information included for greater amberjacks was published previously (Renshaw et al. 2006, 2007). N/A indicates that the marker failed to amplify consistently.
Microsatellite
Sdu1
CGTTTCCATCGCACTTTT +++ GCTAACACTCACTGGTG
Sdu2
CTATATTCACTCTGTTGCC ++ GTGTAGGAGAGACTGTAAG
Sdu3
CGGTGTATTGTTACTGTGAC + TCGTCTCTGATTGGTTAG
Sdu4
GGAAATAGTTTGGATCACGCTGG +++ GGATGCTCAGTGAAGTTGTGC
Sdu5
GTAAGGATTTGTCATGTAGCC ++ GGAGACGAGTTCTCTTTGC
Sdu6
CCAAAGCAGGTGAAAGTGA + GGTCCATACAACAACTCAG
TABLE 1.
Continued.
Microsatellite (CAA)8 56 54 52
Primer sequence (5 3 )a* T Ab Spc N, NAd HE , H O f
Repeat sequence*
Size rangee
PHW g
MicroCheckerh N/A N/A N/A N
Sdu7
CACTTTCAACTGGAACACC ++ GGTTCTGCTGGCTCATTG
Sdu8
CCAGTCTATGAAACACAACC + CCTGAAGCGATGAAGCGT (GAA)9 56 54 52
Sdu9
CTGTTGTCCTTCCAGAC +++ CCACATCGTCTGAATAGC (GA)4 (GAA)9 53 51 49
Sdu10
CCAAGTCCTCCTGCTACTACCAT + CCTTGTGGATGACCTGTTTG (GAA)18 56 54 52
Sdu11
GCTCTCGTGTGTTACTCAAG + GCAACTGTCAGATCCTCCA (CAA)7
56 54 52
60 48 46 N/A N/A N
Sdu12
CCACAAGTTATCACAAGCCACC ++ GCTTTGTCCCCTGTGTGCTG
(GACA)5 GGCA (GACA)5 GGCA (GACA)7
Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu
19, 1 N/A 30, 3 N/A 26, 3 19, 3 29, 1 30, 2 18, 1 29, 3 19, 8 29, 1 30, 1 N/A 29, 2 19, 2 29, 4 30, 12 18, 9 29, 15 19, 1 29, 1 30, 2 18, 1 29, 2 N/A N/A 30, 7 18, 9 29, 10
337 N/A 346352 N/A 343364 101107 95 101107 95 105111 233249 237 219 N/A 230238 273276 261282 285321 287314 295346 168 168 168171 168 169172 N/A N/A 232260 229277 237313
0.000, 0.000 N/A 0.501, 0.367 N/A 0.361, 0.269 0.152, 0.158 0.000, 0.000 0.033, 0.033 0.000, 0.000 0.068, 0.069 0.528, 0.579 0.000, 0.000 0.000, 0.000 N/A 0.131, 0.138 0.462, 0.474 0.577, 0.552 0.793, 0.633 0.887, 0.944 0.902, 0.966 0.000, 0.000 0.000, 0.000 0.066, 0.067 0.000, 0.000 0.160, 0.172 N/A N/A 0.807, 0.767 0.792, 0.500 0.776, 0.690
1.0000 N/A 0.1678 N/A 0.102 1.0000 1.0000 1.0000 1.0000 1.000 0.8750 1.0000 1.0000 N/A 1.000 1.0000 0.8619 0.0722 0.9606 0.412 1.0000 1.0000 1.0000 1.0000 1.000 N/A N/A 0.0759 0.0013 0.550
525
526
Primer sequence (5 3 )a* T Ab Spc 50 48 46 N, NAd HE , H O f (CAA)11 Repeat sequence* Size rangee PHW g MicroCheckerh (CAGA)16 56 54 52 (GATA)25 56 54 52 (GAA)21 56 54 52 (GATA)13 56 54 52 (GA)14 60 58 56 Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu 19, 3 29, 1 30, 2 18, 3 29, 4 19, 7 29, 2 30, 7 18, 2 29, 10 19, 16 29, 10 30, 22 18, 17 27, 20 19, 6 29, 1 30, 1 18, 3 29, 11 19, 3 29, 1 30, 7 18, 8 29, 8 19, 2 29, 8 30, 4 18, 7 29, 11 108120 102 105108 96105 114126 212236 212216 232256 212216 236272 277369 297389 289385xx 301389 264380 300318 292 295 295304 311341 313317xx 303 290322 324352 266298 299301 319365 311321 309325 311377 0.198, 0.211 0.000, 0.000 0.033, 0.033 0.294, 0.167 0.462, 0.552 0.805, 0.895 0.100, 0.103 0.848, 0.833 0.203, 0.111 0.823, 0.724 0.930, 0.947 0.789, 0.897 0.951, 0.867 0.954, 1.000 0.950, 0.926 0.762, 0.632 0.000, 0.000 0.000, 0.000 0.586, 0.611 0.832, 0.862 0.284, 0.263 0.000, 0.000 0.605, 0.567 0.857, 0.778 0.783, 0.630 0.235, 0.263 0.808, 0.345 0.571, 0.667 0.765, 0.833 0.847, 0.793 1.0000 1.0000 1.0000 0.0397 0.174 0.0441 1.0000 0.0181 0.1691 0.178 0.2709 0.9997 0.0863 1.0000 0.697 0.7447 1.0000 1.0000 0.2134 0.694 0.3553 1.0000 0.8688 0.4006 0.060 1.0000 0.0000 N 0.4981 0.3300 0.487 (Continued on next page)
TABLE 1.
Continued.
Microsatellite
Sdu16
GAGTTGTACTGTGGTAAAC + GGACATTAGAGTCTGTGG
Sdu19
GCATTCTGGCATTAGCAT +++ GGTACTCTAGTTAGCCCTAC
Sdu21
CTCAGGACAATGTTGGTAG + GCTAACAAGTTCACGACAT
Sdu22
CATTCTCCAAGTATGTGACCTC ++ GCTCTATGCGAATACCTCCA
Sdu27
CCTTCTGTCTTGACTCTGC +++ CGATTCATCCAGCTTTAGG
Sdu29
CCTTGCCATACCGATGCCAG + GACTGCTCTGCCTGCTTGTTG
TABLE 1.
Continued.
Microsatellite (CA)14 56 54 52
Primer sequence (5 3 )a* T Ab Spc N, NAd HE , H O f
Repeat sequence*
Size rangee
PHW g
MicroCheckerh N N/A
Sdu31
CACATTTGGACGGATTCTTC + GCTGTTATCCTCCAGTGCT
Sdu32
CCTGTGAGAGCATTTGGTAT ++ GTGCTTGTCTCTTCTGTCAT (CA)17 53 51 49
Sdu33
CCTCTAACAGCCACAATCA ++ GCTCTTCACCTTCCTCATA (GA)13 56 54 52
Sdu34
CCTTGTGTTGTATCTGCTGTAA +++ GGAATAAACCTCGTCTGTCA (GA)20 56 54 52
Sdu36
CTGTTATGAAGCAGTGAAGAGG + GGACCATCCTGCTCTGACA (GA)23
56 54 52
53 51 49 N/A
Sdu37
CCTCTAATGGACTTCAGCG +++ GGTTATTTTGAGAGCCGTC
(CA)16
Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu
19, 4 29, 9 30, 20 18, 7 29, 7 19, 10 29, 9 30, 8 18, 12 29, 21 19, 7 29, 2 30, 5 18, 7 29, 3 19, 4 29, 1 30, 3 N/A 29, 8 19, 3 29, 1 30, 16 18, 4 29, 9 19, 8 29, 14 N/A 18, 12 29, 25
8496 106140 90164 86104 8498 93125 97131 99165 103133 99177 194212 210212 194202 188216 202208 93103 194 9197 N/A 84118 194198 186 192250 198210 200226 164180 173281 N/A 167219 160278
0.745, 0.684 0.781, 0.759 0.941, 0.933 0.578, 0.333 0.738, 0.759 0.851, 0.789 0.653, 0.862 0.811, 0.700 0.921, 0.944 0.948, 0.862 0.802, 0.789 0.290, 0.345 0.671, 0.600 0.838, 0.833 0.296, 0.276 0.681, 0.737 0.000, 0.000 0.532, 0.567 N/A 0.778, 0.517 0.522, 0.474 0.000, 0.000 0.912, 0.933 0.487, 0.500 0.815, 0.690 0.858, 0.842 0.859, 1.000 N/A 0.887, 0.833 0.889, 0.828
0.4228 0.2719 0.2394 0.0028 0.650 0.5653 0.8653 0.1963 0.8969 0.231 0.8488 0.5616 0.4063 0.4069 0.598 0.8497 1.0000 0.5847 N/A 0.028 0.4838 1.0000 0.4456 0.3928 0.132 0.3169 0.2088 N/A 0.5094 0.448
527
528
Primer sequence (5 3 )a* T Ab Spc 56 54 52 N, NAd HE , H O f PHW g (CA)16 Repeat sequence* Size rangee MicroCheckerh N/A N/A N/A N/A 53 51 49 N/A (CA)17 7bp (CA)5 64 62 60 (CA)18 53 51 49 (CA)28 60 58 56 (GA)12 56 54 52 (GA)30 Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu Sfa Sla Sri Szo Sdu 19, 4 29, 2 30, 4 N/A 29, 6 19, 7 29, 3 30, 12 18, 3 28, 9 19, 6 N/A N/A N/A 29, 10 19, 5 29, 10 30, 22 18, 14 29, 11 19, 3 29, 2 30, 1 18, 10 29, 3 19, 2 29, 5 N/A 18, 1 29, 13 145153 137139 137147 N/A 154180 187201 181185 196232 189193 197235 96112 N/A N/A N/A 96130 259271 283315 272340 265305 276298 118124 117119 114 116136 114124 238252 232244 N/A 211 217259 0.642, 0.684 0.242, 0.276 0.653, 0.533 N/A 0.338, 0.345 0.805, 0.737 0.617, 0.655 0.838, 0.900 0.541, 0.778 0.825, 0.893 0.718, 0.842 N/A N/A N/A 0.636, 0.724 0.518, 0.579 0.845, 0.931 0.941, 0.900 0.913, 0.833 0.835, 0.828 0.656, 0.579 0.100, 0.103 0.000, 0.000 0.813, 0.722 0.402, 0.379 0.053, 0.053 0.749, 1.000 N/A 0.000, 0.000 0.817, 0.897 1.0000 1.0000 0.0491 N/A 0.688 0.1863 0.7009 0.3556 0.0975 0.884 0.3794 N/A N/A N/A 0.548 0.8453 0.2253 0.6472 0.1897 0.654 0.5681 1.0000 1.0000 0.6769 0.574 1.0000 0.0003 N/A 1.0000 0.628
TABLE 1.
Continued.
Microsatellite
Sdu39
AGTGGCTTCTGCTGCTGT ++ CGTGTGCGTGCTTGTAAA
Sdu40
CGATGCTTTCAACTCCGACACAC +++ CCATCCTTCATCAGCAACAACATCC
Sdu41
AGCGTGGACAGTTTATGG ++ GTCTGTTTACTGGTCGCA
Sdu43
GGAACATTTGGAGCCATAAGAC CAGAAGAAGAGCGTGGTGGAGAG +++
Sdu44
GGTAATGGGAGGTGTGAGTGT ++ CCTTCTCCTGTTAATCCATCTCC
Sdu46
GCAGTGTGAGCCATACATTAC +++ CTACAGGACAAAAGCCATT
Primer sequences are forward (top) and reverse (bottom). TA is the annealing temperature ( C) used for PCR amplication (see Methods text). c Species (Sp) characterized are the lesser amberjack (Sfa), yellowtail jack (Sla), almaco jack (Sri), banded ruddersh (Szo), and greater amberjack (Sdu). d N is the number of individuals assayed; NA is the number of alleles detected. e Size range (bp) refers to alleles that have been discovered thus far; the superscript xx indicates that alleles were not spaced as anticipated (Sdu4, Sdu21, and Sdu27 ). f HE is expected heterozygosity; HO is observed heterozygosity. g PHW is the probability of deviation from HardyWeinberg expectations; deviations that were signicant after Bonferroni correction (Rice 1989) are shown in bold italics. h Possible issues with loci as indicated by Micro-Checker software (Van Oosterhout et al. 2004): N = evidence for null alleles; = no evident issues. Micro-Checker was not utilized with the previously published data for greater amberjacks.
TECHNICAL NOTE
529
ACKNOWLEDGMENTS We thank the following for sh sampling: D. Player, E. Muhammed, and B. White (South Carolina Department of Natural Resources); K. Gruenthal and M. Drawbridge (HubbsSeaWorld Research Institute); T. Morris (Rancheros del March); and R. Allman, D. DeVries, and B. Walling (National Marine Fisheries Service). Funding was provided by Texas AgriLife Research (Project H-6703) and Texas A&M UniversityNational Council of Science and Technology, Mexico (Project 2010-004). This paper is Number 92 in the series Genetic Studies in Marine Fishes and Contribution Number 205 of the Center for Biosystematics and Biodiversity at Texas A&M University.
REFERENCES
Gold, J. R., M. A. Renshaw, E. Saillant, and R. R. Vega. 2010. Spawning frequency of brood dams and sires in a marine sh stock-enhancement hatchery. Journal of Fish Biology 77:10301040. Hamasaki, K., K. Tsuruoka, K. Teruya, H. Hashimoto, K. Hamada, T. Hotta, and K. Mushiake. 2009. Feeding habits of hatchery-reared larvae of greater amberjack, Seriola dumerili. Aquaculture 288:216 225. Iguchi, J., Y. Takashima, A. Namikoshi, and M. Yamashita. 2012. Species identication method for marine products of Seriola and related species. Fisheries Science 78:197206.
Lewis, P. O., and D. Zaykin. 2001. Genetic Data Analysis: computer program for the analysis of allelic data, version 1.0 (d16c). Available: http://hydrodictyon.eeb.uconn.edu/people/plewis/software.php. (July 2012). Liu, Z. J., and J. F. Cordes. 2004. DNA marker technologies and their applications in aquaculture genetics. Aquaculture 238:137. Miller, P. A., A. J. Fitch, M. Gardner, K. S. Hutson, and G. Mair. 2011. Genetic population structure of yellowtail kingsh (Seriola lalandi) in temperate Australian waters inferred from microsatellite markers and mitochondrial DNA. Aquaculture 319:328336. Reiss, H., G. Hoarau, M. Dickey-Collas, and W. J. Wolff. 2009. Genetic population structure of marine sh: mismatch between biological and sheries management units. Fish and Fisheries 10:361395. Renshaw, M. A., J. C. Patton, C. E. Rexroad III, and J. R. Gold. 2006. PCR primers for trinucleotide and tetranucleotide microsatellites in greater amberjack, Seriola dumerili. Molecular Ecology Notes 6:11621164. Renshaw, M. A., J. C. Patton, C. E. Rexroad III, and J. R. Gold. 2007. Isolation and characterization of dinucleotide microsatellites in greater amberjack, Seriola dumerili. Conservation Genetics 8:10091011. Repulles-Albelda, A., F. E. Montero, A. S. Holzer, K. Ogawa, K. S. Hutson, and J. A. Raga. 2008. Speciation of the Paradeontacylix spp. (Sanguinicolidae) of Seriola dumerili. Two new species of the genus Paradeontacylix from the Mediterranean. Parasitology International 57:405414. Rice, W. R. 1989. Analyzing tables of statistical tests. Evolution 43:223225. Simpson, S. 2011. The blue food revolution. Scientic American 304:5461. Stuart, K. R., and M. A. Drawbridge. In press. Captive spawning and larval rearing of California yellowtail (Seriola lalandi). Aquaculture Research. DOI: 10.1111/j.1365-2109.2011.03077.x. Van Oosterhout, C., W. F. Hutchinson, and P. Shipley. 2004. Micro-Checker: software for identifying and correcting genotyping errors in microsatellite data. Molecular Ecology Notes 4:535538.

Dietary Lipid Levels Affect Growth and Fatty Acid Profiles of Malaysian Mahseer Tor tambroides
Ehsan Ramezani-Fard , Mohd Salleh Kamarudin , Che Roos Saad , Sharr Azni Harmin & Goh Yong Meng
a b a a a a
Department of Aquaculture, Faculty of Agriculture, Universiti Putra Malaysia, Selangor, Serdang, 43400, Malaysia
b
Department of Veterinary Preclinical Sciences, Faculty of Veterinary Medicine, Universiti Putra Malaysia, Selangor, Serdang, 43400, Malaysia Version of record first published: 14 Sep 2012.
To cite this article: Ehsan Ramezani-Fard, Mohd Salleh Kamarudin, Che Roos Saad, Sharr Azni Harmin & Goh Yong Meng (2012): Dietary Lipid Levels Affect Growth and Fatty Acid Profiles of Malaysian Mahseer Tor tambroides , North American Journal of Aquaculture, 74:4, 530-536 To link to this article: http://dx.doi.org/10.1080/15222055.2012.690829
COMMUNICATION
Dietary Lipid Levels Affect Growth and Fatty Acid Proles of Malaysian Mahseer Tor tambroides
Ehsan Ramezani-Fard, Mohd Salleh Kamarudin,* Che Roos Saad, and Sharr Azni Harmin
Department of Aquaculture, Faculty of Agriculture, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia
Goh Yong Meng

Department of Veterinary Preclinical Sciences, Faculty of Veterinary Medicine, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia
Abstract
After protein, the second major essential macronutrient in sh diet is lipid. This study was conducted to determine the optimum level of dietary lipid for the best growth performance of juvenile Malaysian mahseer Tor tambroides. Four isonitrogenous diets containing 40% crude protein were formulated to contain different levels of lipid (5, 10, 15, or 20% on an as-fed basis). Cod liver oil was incorporated into the feed as the main dietary lipid source used to formulate the diets while residual oil coming from other ingredients contributed about 5% of dietary lipid. The experimental diets were labeled as L5, L10, L15, or L20 to denote the levels of dietary lipid. Fish were fed the experimental diets in triplicate groups for 63 d. Growth performance, survival rate, and daily feed intake by Malaysian mahseer signicantly decreased when fed diets in which levels of dietary lipid increased from 5% to 20%. However, the growth performance did not vary signicantly between sh fed the L5 and L10 diets. The increase in dietary lipid signicantly increased the hepatosomatic index but did not inuence the viscerosomatic index. Increasing dietary lipid levels also decreased the lipid content in the whole body composition of sh. The monounsaturated fatty acid (MUFA) and n-3 polyunsaturated fatty acid (PUFA) contents of sh liver signicantly increased with the increase of dietary lipid. The results of this study suggested that 5% dietary lipid is sufcient for the best survival rate and growth performance of juvenile Malaysian mahseer.
The Malaysian mahseer Tor tambroides (also known as Thai mahseer) is one of the most important and highly valued cyprinids throughout the Himalayan region and southeast Asia (Ambak et al. 2007). Since the decline in natural populations of this species owing to overexploitation and destruction of natural
*Corresponding author: msalleh@agri.upm.edu.my Received January 21, 2012; accepted April 29, 2012
habitats (Ingram et al. 2007), there has been great interest in the conservation and culture of Malaysian mahseer (Ramezani-Fard et al. 2011a). Efforts have been made to replenish its natural stock through several conservation programs and to culture the sh to meet the high demand for this species (Kamarudin et al., in press). The rst success in induced spawning of pond-reared Malaysian mahseer broodstock through hormonal treatment was reported by Ingram et al. (2005). After this breakthrough, development of an optimized diet that meets the nutritional requirements of this sh is crucial to ensure the success of Malaysian mahseer conservation and aquaculture goals. Earlier works showed that juveniles of this species required 4050% dietary protein for the optimal growth performance (Ng et al. 2008; Misieng et al. 2011). Lipids, such as essential fatty acids (EFAs), phospholipids, and fat-soluble vitamins, are major essential macronutrients in sh diets and act as a source of energy (Sargent et al. 2002). However, incorporation of traditional marine-derived lipids in aquafeeds has become increasingly costly owing to the recent scarcity and high price of sh oil. Bazaz and Keshavanath (1993) reported a positive effect of increasing dietary lipid on the growth performance of Deccan mahseer T. khudree. An excessive amount of dietary lipid can reduce sh growth by reducing feed consumption and can increase the amount of lipid deposited in the carcass (Wang et al. 2005). Therefore, the optimal dietary lipid level plays a critical role in the increase of growth performance and decrease of feed costs. The present study was carried out to investigate the effects of different levels of dietary lipid on growth performance, survival rate, body composition, and fatty
530
COMMUNICATION
531
acid (FA) proles of muscle and liver of juvenile Malaysian mahseer. METHODS Diet preparation.Four isonitrogenous diets with 40% crude protein and 5, 10, 15, or 20% crude fat (on an as-fed basis) were formulated, and experimental diets were labeled as L5, L10, L15, or L20 (Tables 1, 2). Cod liver oil was incorporated into the feed as the main dietary lipid source used to formulate the diets while residual oil coming from other ingredients (sh meal, soy meal, corn meal) contributed about 5% of dietary lipid. Dry ingredients were thoroughly mixed using a dough mixer. This was followed by adding dietary oils and distilled water to produce a homogenous dough. The moist dough was then screw-pressed through a 2-mm die and the feed pellets formed were oven dried and stored at 20 C. Rearing and sampling.Wild caught juvenile Malaysian mahseer were provided by a local supplier and their authenticity was veried using taxonomical and morphological features (Rainboth 1996; Kottelat 1998). The sh were transferred to the Aquaculture Experimental Station, Universiti Putra Malaysia,
TABLE 1. Feed ingredients and proximate compositions of the test diets fed to juvenile Malaysian mahseer. Four isonitrogenous diets with 40% crude protein and 5, 10, 15, or 20% crude fat (on an as-fed basis) were formulated, and experimental diets were labeled as L5, L10, L15, or L20.
TABLE 2. Fatty acid composition (% of total fatty acid) of the experimental diets fed to juvenile Malaysian mahseer. See Table 1 for explanation of diet labels.
Diets Fatty acid 14:0 16:0 16:1n-7 18:0 18:1n-9 18:2n-6 18:3n-3 20:0 20:1n-9 20:2n-6 20:4n-6 20:5n-3 22:0 22:1n-11 22:5n-3 22:6n-3 SFA MUFA n-3 PUFA n-3 : n-6 ratio L5 1.10 25.76 3.50 6.38 31.87 19.70 0.88 0.70 0.34 0 0 4.43 0.28 0 1.05 4.00 34.22 35.72 10.37 0.53 L10 3.12 22.81 4.44 5.85 26.34 12.03 0.76 0.10 3.85 1.19 0.23 6.29 0.23 3.79 2.25 6.74 32.11 38.42 16.03 1.19 L15 3.49 21.23 5.49 5.19 23.26 6.34 0.63 0.23 6.30 0.94 0.51 7.89 0.21 6.87 2.43 9.01 30.35 41.92 19.95 2.56 L20 4.35 18.35 6.95 4.38 21.40 3.83 0.51 0.31 8.31 0.74 0.64 8.21 0.15 8.74 2.89 10.23 27.54 45.40 21.85 4.19
Diets Diet parameter L5 L10 L15 L20 38.0 13.0 14.8 17.6 14.6 1.0 1.0 39.6 19.6 10.9 19.1 20.1 89.2 and acclimated to laboratory conditions for 2 weeks in 1,000L berglass tanks equipped with a recirculating system that continuously puried water through a series of mechanical and biolter systems. During the acclimation period, sh were on a diet containing 40% crude protein and 5% crude fat and were fed twice per day. After this period, 12 rectangular-shaped glass aquaria (65 L) were each stocked with 10 sh of similar size. The rearing conditions and water quality of this laboratory setting were fully explained by Ramezani-Fard et al. (2011b). In summary, each aquarium was equipped with a recirculating system. Water temperature was between 27.5 C and 29 C, while pH was between 8 and 8.8. Fish were fed twice per day (0900 and 1600 hours) close to apparent satiation. After a 10-d conditioning period, the feeding trial was commenced using sh with a mean initial weight of 2.6 g (SD, 0.2). All treatments were conducted in a controlled rearing condition, each of which was triplicated. Fish in each aquarium were batch-weighed at the start and end of the experiment, as well as every 3 weeks, and the feeding trial was conducted for 9 weeks. Six sh of the initial batch as well as six sh from each treatment (two sh per tank) at the end of the experiment were sacriced, weighed individually, and kept frozen at 80 C for subsequent whole-body proximate analysis. Similarly, an additional six sh were individually weighed, sacriced, and dissected. Their liver and viscera were extracted and weighed. The
Ingredient (g/100 g on as-fed basis) 38.0 38.0 38.0 Fishmeala Soy meal 13.0 13.0 13.0 14.8 14.8 14.8 Caseinb Corn meal 32.2 28.4 23.1 3.8 9.1 Fish oilc 1.0 1.0 1.0 Vitamin premixd 1.0 1.0 1.0 Mineral premixe Proximate analysis (% on as-fed basis) Crude protein 40.4 40.1 39.7 Crude lipid 4.7 9.4 14.8 Ash 10.4 10.8 10.6 32.1 28.6 25.0 Carbohydratesf Gross energy (kJ/g) 17.3 18.0 19.2 Dry matter 87.6 88.9 90.1
a b c
Malaysian sh meal (63% crude protein). Casein from bovine milk (Sigma-Aldrich). Cod liver oil (Seven Seas). d Vitamin premix (g/kg premix): ascorbic acid, 45; myo-inositol, 5; choline chloride, 75; niacin, 4.5; riboavin, 1; pyridoxine, 1; thiamin mononitrate, 0.92; Ca-pantothenate, 3; retinyl acetate, 0.6; cholecalciferol, 0.083; vitamin K menadione, 1.67; -tocopheryl acetate (500 IU/g), 8; biotin, 0.02; folic acid, 0.09; vitamin B12 , 0.001; cellulose, 845.11. e Mineral premix (g/kg premix): KCL, 90; KI, 0.04; CaHPO42H2 O, 500; NaCl, 40; CuSO4 5H2 O, 3; ZnSO4 7H2 O, 4; CoSO4 , 0.02; FeSO4 7H2 O, 20; MnSO4 H2 O, 3; CaCo3 , 215; MgOH, 124; Na2 SeO3 , 0.03; NaF, 1. f Carbohydrates = dry matter (protein + lipid + ash).
532
RAMEZANI-FARD ET AL.
dissected sh were subsequently dressed, and muscle from the area between the lateral and dorsal line was removed. The collected liver and muscle were immediately stored at 80 C for further FA analyses. The following growth performance and feed utilization efciency parameters were also calculated: weight gain (WG), specic growth rate (SGR), feed conversion ratio (FCR), daily feed intake (DFI), and protein efciency ratio (PER). Biochemical analysis.Prior to chemical analysis, the whole body and llets of sh were lyophilized in triplicate groups per sample (all sh from a treatment group were combined and triplicate samples were taken from a group) for 48 h, and the lost moisture was calculated. Proximate composition and FA analysis of samples were described in Ramezani-Fard et al. (2012). In summary, crude protein and crude lipid were determined by the Kjeldahl method and Soxhlet extraction, respectively. The ash content was determined by incinerating the dry sample at 600 C for 4 h, and the gross energy was measured by direct combustion in an adiabatic bomb calorimeter. Lipid from feed, liver, and lyophilized llet was extracted with a chloroform : methanol (2:1, v:v) mixture. Fatty acid methyl esters (FAMEs) were then separated and quantied on a fused silica capillary column (Supelco SP-2330; 30 m 0.25 mm; lm thickness, 0.20 mm) in a gas chromatograph (Agilent 7890N, Agilent Technologies, Santa Clara, California). Fatty acids were identied by comparing the relative retention time with the reference standards (Supelco, 37 components FAME Mix; Supelco, Bellefonte, Pennsylvania) and menhaden oil and expressed as the area percentage of FAMEs. Statistical analysis.All experimental data were analyzed by ANOVA. The mean differences were evaluated using Dun-
cans multiple-range test. Homogeneity of variances was tested using Levenes test, and data identied as nonhomogeneous were subjected to arcsine transformation before statistical analysis. All analysis was carried out with SPSS 15 for Windows (SPSS, Chicago, Illinois) and the difference was considered signicant at P < 0.05. RESULTS Growth performance, survival rate, feed efciency, and body indices of Malaysian mahseer fed different diets are presented in Table 3. Growth of sh signicantly decreased as dietary lipid increased from 5% to 20%. However, the growth performances of sh fed the L5 and L10 diets were not signicantly different. No mortality was observed in sh fed either the L5 or L10 diet. The survival rate was signicantly lower in sh fed the L15 or L20 diet compared with those fed the other diets. The sh fed the L5 or L10 diet had the highest DFI while sh fed the L20 diet had the lowest intake of feed. The best and worst FCR were observed in sh fed the L20 and L10 diets, respectively. The hepatosomatic index (HSI) was signicantly lower in sh fed the L5 or L10 diet than in those fed the L15 or L20 diet. The viscerosomatic indices (VSIs) were not signicantly inuenced by the diets. The whole-body protein, lipid, and moisture contents of Malaysian mahseer were signicantly inuenced by the experimental diets (Table 4). Fish fed the L5 and L10 diets had signicantly higher whole-body lipid than those fed the L15 and L20 diets. Increased moisture content and decreased protein content were also observed in the whole-body composition of sh when sh were fed the L20 diet. There were no signicant
TABLE 3. Growth performance, survival rate, feed utilization efciency, and body indices of juvenile Malaysian mahseer fed the experimental diet for 63 d. See Table 1 for explanation of diet labels. Mean SE (n = 3 except for HSI and VSI where n = 6). Values within the same row with different lowercase letters are signicantly different at P < 0.05.
Diets Parameters Initial body weight (g) Final body weight (g) WGa (g) SGRb (%/d) Survival (%) FCRc DFId (%/d) PERe HSIf VSIg
a b
L5 2.70 0.09 9.19 0.57 z 6.50 0.52 z 1.94 0.08 z 100.0 z 2.08 0.08 y 3.58 0.03 z 1.12 0.04 1.30 0.04 y 6.67 0.34
L10 2.70 0.08 8.29 0.13 zy 5.59 0.06 z 1.78 0.02 z 100.0 z 2.40 0.05 z 3.83 0.06 z 1.08 0.05 1.62 0.06 zy 7.14 0.33
L15 2.86 0.05 7.32 0.14 y 4.46 0.18 y 1.49 0.06 y 77.8 2.8 y 1.68 0.01 x 2.49 0.17 y 1.27 0.14 2.46 0.33 z 6.68 0.56
L20 3.0 0.15 7.20 0.30 y 4.21 0.15 y 1.39 0.01 y 75.0 4.8 y 1.42 0.04 w 1.85 0.04 x 1.34 0.13 2.43 0.58 z 6.23 0.35
P 0.19 0.01 0.00 0.00 0.00 0.00 0.00 0.27 0.02 0.39
Final weight gain. Specic growth rate: [(loge nal mean weight loge initial mean weight) / experimental days] 100. c Feed conversion ratio: dry feed intake (g) / wet weight gain (g). d Daily feed intake: 100 dry feed intake (g) / [(total nal body weight + total initial body weight) / 2] experimental days. e Protein efciency ratio: wet weight gain (g) / protein intake (g). f Hepatosomatic index: 100 liver weight (g) / body weight (g). g Viscerosomatic index: 100 visceral weight (g) / body weight (g).
COMMUNICATION
533
TABLE 4. Whole-body proximate composition, liver lipid, and muscle lipid content (% of wet weight) of juvenile Malaysian mahseer fed the experimental diets for 63 d. See Table 1 for explanation of diet labels. Mean SE, n = 3; values within the same row with different lowercase letters are signicantly different at P < 0.05.
Diets Parameter Moisture Protein Fat Ash Carbohydratea Liver fat Muscle fat
a
Initial 68.4 0.1 17.3 0.1 9.5 0.1 3.5 0.0 0.6 0.1 19.6 1.0 5.5 0.1
L5 68.2 0.1 y 17.5 0.2 z 9.8 0.0 z 3.6 0.1 0.8 0.1 z 20.1 0.6 5.1 0.1
L10 67.6 0.2 y 17.7 0.1 z 9.7 0.1 z 3.6 0.0 1.3 0.2 zy 19.4 0.6 5.1 0.3
L15 68.1 0.3 y 17.4 0.1 z 8.1 0.0 y 3.5 0.1 2.6 0.3 y 19.1 1.0 5.2 0.1
L20 70.7 0.3 z 16.0 0.1 y 7.9 0.0 x 3.6 0.1 2.7 0.4 y 21.0 1.3 5.1 0.0
P 0.00 0.02 0.00 0.42 0.03 0.75 0.64
Carbohydrates = dry matter (protein + lipid + ash).
differences between total lipid contents of liver and muscle of sh fed different diets (Table 4). Palmitic acid (16:0) and oleic acid (18:1n-9) were the most abundant FAs in the muscle of Malaysian mahseer. In addition to these FAs, stearic acid (18:0), linoleic acid (18:2n-6), and docosahexaenoic acid (DHA) (22:6n-3) were the other FAs contributing most to the total FA content of sh muscle (Table 5). The increase of sh oil inclusion in the diet led to a signicant decrease of saturated fatty acid (SFA) as well as an increase of monounsaturated fatty acid (MUFA) in sh mus-
cle. The concentration of eicosapentaenoic acid (EPA; 20:5n-3) signicantly increased with the increase of dietary lipid level. However, the muscle contents of most of long-chain polyunsaturated fatty acids (LC-PUFAs) including 22:6n-3, 20:5n-3, and 20:4n-6 decreased at the end of the experiment compared with their initial amounts. There were no signicant differences between the concentrations of 20:4n-6, 22:5n-3, and 22:6n-3 in the muscle of sh fed different diets. Similar to the muscle, the concentrations of SFAs, MUFAs, and n-3 PUFAs in the sh liver were signicantly affected by the
TABLE 5. Fatty acid composition (% of total fatty acid) of muscle tissue of juvenile Malaysian mahseer at the beginning and end of the experimental period. See Table 1 for explanation of diet labels. Mean SE, n = 3; values within the same row with different lowercase letter are signicantly different at P < 0.05.
Diets Fatty acid 14:0 14:1 16:0 16:1n-7 18:0 18:1n-9 18:2n-6 18:3n-3 20:0 20:1n-9 20:3n-6 20:4n-6 20:5n-3 22:1n-11 22:5n-3 22:6n-3 SFA MUFA n-3 PUFA n-3 : n-6 ratio Initial 3.6 0.21 25.0 5.7 10.2 24.2 7.9 0.18 2.60 1.3 0.54 6.1 3.6 0.21 1.3 7.2 41.4 31.7 12.3 0.85 L5 3.5 0.0 z 0.28 0.08 28.4 0.2 z 3.7 0.1 x 8.5 0.1 32.4 0.6 8.2 0.2 z 0.20 0.02 0.80 0.02 1.6 0.1 x 0.74 0.10 z 3.1 0.4 1.4 0.1 y 1.7 0.1 5.4 0.2 41.3 0.3 z 38.0 0.4 w 8.7 0.3 0.72 0.01 y L10 3.6 0.0 z 0.22 0.03 28.3 0.1 z 6.3 0.1 zy 8.4 0.4 30.6 0.6 7.0 0.0 y 0.22 0.03 0.61 0.02 2.3 0.0 x 0.48 0.05 y 2.6 0.2 1.5 0.0 zy 0.64 0.02 x 1.7 0.1 5.6 0.6 41.0 0.5 z 40.0 0.6 x 8.9 0.7 0.89 0.09 zy L15 3.3 0.0 y 0.2 0.02 25.2 0.3 y 6.0 0.2 y 8.2 0.4 31.9 0.2 6.6 0.2 y 0.35 0.03 0.77 0.07 4.1 0.1 y 0.42 0.03 y 2.5 0.3 1.8 0.1 z 1.72 0.37 y 1.7 0.2 5.6 0.2 37.4 0.4 y 43.8 0.5 y 9.4 0.2 0.99 0.03 z L20 3.3 0.1 y 0.13 0.03 21.4 0.3 x 6.6 0.2 z 8.3 0.3 32.0 0.5 6.6 0.1 y 0.32 0.06 0.75 0.13 5.5 0.4 z 0.39 0.04 y 2.5 0.1 1.8 0.2 z 2.84 0.05 z 1.7 0.2 5.8 0.5 33.8 0.2 x 47.1 0.6 z 9.6 0.6 1.01 0.05 z P 0.00 0.22 0.00 0.00 0.89 0.12 0.00 0.05 0.33 0.00 0.01 0.33 0.03 0.00 0.99 0.94 0.00 0.00 0.57 0.01
534
RAMEZANI-FARD ET AL.
TABLE 6. Fatty acid composition (% of total fatty acid) of liver tissue of juvenile Malaysian mahseer at the beginning and end of the experimental period. See Table 1 for explanation of diet labels. Mean SE, n = 3; values within the same row with different lowercase letters are signicantly different at P < 0.05.
Diets Fatty acid 14:0 16:0 16:1n-7 18:0 18:1n-9 18:2n-6 18:3n-3 20:0 20:1n-9 20:2n-6 20:3n-6 20:4n-6 20:5n-3 22:1n-11 22:5n-3 22:6n-3 SFA MUFA n-3 PUFA n-3 : n-6 ratio Initial 4.1 26.0 5.9 9.1 35.8 6.2 0.17 3.5 0.46 0.74 4.6 0.2 0.68 0.24 2.2 42.8 42.4 3.3 0.29 L5 4.4 0.1 28.6 0.3 z 6.5 0.2 8.5 0.1 z 39.9 0.5 z 4.9 0.2 y 0.07 0.01 w 0.56 0.03 1.4 0.1 w 0.12 0.01 y 0.61 0.02 1.1 0.1 0.66 0.05 x 0.21 0.03 x 2.4 0.1 x 42.0 0.5 z 47.9 0.6 zy 3.4 0.1 w 0.50 0.02 x L10 4.6 0.0 28.1 0.2 z 6.4 0.2 7.4 0.2 y 35.9 0.3 y 6.4 0.1 z 0.23 0.01 x 0.60 0.09 2.8 0.0 x 0.27 0.01 z 0.77 0.07 1.4 0.2 0.82 0.11 yx 1.0 0.1 x 0.51 0.02 y 2.8 0.1 x 40.6 0.3 zy 46.2 0.4 y 4.4 0.2 x 0.50 0.01 x L15 4.8 0.6 26.7 0.7 y 6.6 0.3 7.4 0.2 y 32.3 0.9 x 5.8 0.2 z 0.35 0.01 y 0.83 0.07 5.0 0.1 y 0.13 0.07 y 0.68 0.16 1.4 0.3 1.04 0.06 zy 2.3 0.2 y 0.52 0.05 y 4.2 0.2 y 39.7 1.0 y 46.2 1.1 y 6.1 0.3 y 0.76 0.03 y L20 4.3 0.1 23.4 0.1 x 7.6 0.4 7.1 0.5 y 30.8 0.2 x 4.1 0.3 x 0.48 0.07 z 0.62 0.07 7.5 0.2 z 0.25 0.01 z 0.37 0.03 1.3 0.0 1.14 0.07 z 4.3 0.2 z 1.12 0.06 z 5.6 0.3 z 35.4 0.5 x 50.2 0.7 z 8.4 0.4 z 1.39 0.08 z P 0.68 0.00 0.06 0.03 0.00 0.00 0.00 0.09 0.00 0.03 0.06 0.68 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00
diets (Table 6). The increase of dietary lipid level signicantly decreased the concentration of SFA and 18:1n-9 in the liver. Increased total n-3 PUFA content was observed in the liver of sh, increasing from 3.38 to 8.37 (% of total FA) when sh were fed increasing amounts of dietary lipid. A notable decrease of 20:4n-6 content was observed in the liver of sh fed all the experimental diets compared with the initial amount. However, liver content of 20:4n-6 did not differ signicantly among sh fed diets with increasing lipid levels.
DISCUSSION Earlier studies showed that increasing dietary lipid levels are likely to improve or be ineffective in improving the growth performance of many marine carnivorous species such as white seabass Atractoscion nobilis and Atlantic halibut Hippoglossus hippoglossus (Bright et al. 2005; Martins et al. 2007; L opez et al. 2009). However, a high dietary lipid level can inhibit the growth of some freshwater species such as hybrid tilapia Oreochromis niloticus Oreochromis aureus, silver barb Puntius gonionotus, and grass carp Ctenopharyngodon idella (Chou and Shiau 1996; Du et al. 2005; Mohanta et al. 2008). The reduction of growth performance of Malaysian mahseer with increasing dietary lipid was in agreement with the latter group of farmed freshwater species. As shown in earlier studies (Satpathy et al.
2003; Hamre et al. 2004; Alam et al. 2009a), this study also revealed that FCR decreased with increasing dietary lipid. It is well established that sh adjust their feed intake to meet their digestible energy requirements (Du et al. 2005). Although the feed intake of some carnivorous species such as Atlantic halibut and southern ounder Paralichthys lethostigma is not extremely affected by dietary lipid level (Martins et al. 2007; Alam et al. 2009b), the feed intake of most omnivorous species reduces with increasing dietary lipid level (Pei et al. 2004; Du et al. 2005; Wang et al. 2005). In the present study, DFI decreased dramatically with the increase of dietary lipid level. Considering that the test diets were not isocaloric, sh on high lipid diets may reduce their feed intake for energy regulation purposes. This reduction can lead to reduced growth because of insufcient protein intake (Lee and Kim 2001). An early work by Ng et al. (2008) showed that optimal growth of Malaysian mahseer ngerlings is achieved when they are fed a semipuried diet containing 10% lipid and 1819 kJ/g gross energy. According to their study, the increase of dietary lipid up to 15% does not improve nor inhibit the growth performance of sh. It should be noted that corn oil supplied more than 50% of lipid in their study, while sh oil was the primary source of lipid in the present study. High amounts of dietary sh oil have been hypothesized to reduce growth performance of some warmwater shes such as tilapia and some carps (Alava 1998; Du et al. 2008).
COMMUNICATION
535
Proximate analysis showed that whole-body lipid content of Malaysian mahseer decreased when their dietary lipid was increased from 10% to 15%. De Silva et al. (2001) found similar results in freshwater shortn eel Anguilla australis that were fed excess amounts of dietary lipid in which the muscle lipid content was reduced after a 10-week feeding trial. The present study also showed that the HSI signicantly increased as dietary lipid levels increased; however, such an increase did not inuence the VSI. The HSI in Malaysian mahseer fed either the 15% or 20% lipid diets were higher compared with that in sh fed the 5% lipid diets. The increased HSI, along with the lack of lipid accumulation in the liver, could be associated with the accumulation of carbohydrate molecules. Several authors have noted that tissue FA proles reect the dietary FA intake composition (dos Santos et al. 1993; AliyuPaiko et al. 2010). This is in agreement with increasing levels of MUFAs and n-3 PUFAs in both muscle and liver of Malaysian mahseer fed increasing levels of dietary lipid. The increase of long-chain MUFAs (20:1n-9 and 22:1n-11), which were prevalent in sh oil, was notable in Malaysian mahseer tissues in response to the increase of cod liver oil in the diets. The amounts of n-3 PUFAs, particularly 22:6n-3, increased in the liver of sh fed diets with 15% or 20% lipid compared with those fed the other diets. A notable reduction of 20:4n-6 levels in the tissues of sh fed all the experimental diets compared with its initial amount may be associated with deciency of 18:2n-6 and 18:3n-3, the parental FAs of 20:4n-6, in the diets. This reduction can be prevented by partial replacement of dietary sh oil with vegetable oils containing high amounts of C18 PUFA. The ndings of this study suggest that a high level of n-3 LC-PUFA is not required in the diet of Malaysian mahseer. In agreement with this nding, some freshwater sh species may have a higher requirement for n-6 PUFA over n-3 PUFA in the diet, particularly those species capable of synthesizing LC-PUFAs de novo from 18-carbon precursors such as 18:2n-6 (Mishra and Samantaray 2004; Zuraini et al. 2006). In conclusion, this study demonstrated that a diet with 5% lipid is sufcient to provide the best survival rate and growth performance in juvenile Malaysian mahseer. A dietary lipid level beyond 10% containing more than 3.8% sh oil reduces both survival rate and growth performance of this sh. ACKNOWLEDGMENTS This study was supported through a Malaysian government E-Science grant no. 05-01-04-SF0209. The authors thank Mahdi Ebrahimi for technical assistance. REFERENCES
Alam, M. S., W. O. Watanabe, P. M. Carroll, and T. Rezek. 2009a. Effects of dietary protein and lipid levels on growth performance and body composition of black sea bass Centropristis striata (Linnaeus 1758) during grow-out in a pilot-scale marine recirculating system. Aquaculture Research 40:442449. Alam, M. S., W. O. Watanabe, and H. V. Daniels. 2009b. Effect of different dietary protein and lipid levels on growth performance and body composition of
juvenile southern ounder, Paralichthys lethostigma, reared in a recirculating aquaculture system. Journal of the World Aquaculture Society 40:513521. Alava, V. R. 1998. Effect of salinity, dietary lipid source and level on growth of milksh (Chanos chanos) fry. Aquaculture 167:229236. Aliyu-Paiko, M., R. Hashim, A. S. Chong, L. Yogarajah, and A. M. El-Sayed. 2010. Inuence of different sources and levels of dietary protein and lipid on the growth, feed efciency, muscle composition and fatty acid prole of snakehead Channa striatus (Bloch, 1793) ngerling. Aquaculture Research 41:13651376. Ambak, M. A., A. H. Ashraf, and S. Budin. 2007. Conservation of the Malaysian mahseer in Nenggiri basin through community action. Pages 217228 in S. S. Siraj, A. Christianus, C. K. Ng, and S. S. De Silva, editors. Proceedings of the international symposium on the mahseer. Malaysian Fisheries Society, Kuala Lumpur. Bazaz, M. M., and P. Keshavanath. 1993. Effect of feeding different levels of sardine oil on growth, muscle composition and digestive enzyme activities of mahseer, Tor khudree. Aquaculture 115:111119. Bright, L. A., S. D. Coyle, and J. H. Tidwell. 2005. Effect of dietary lipid level and protein energy ratio on growth and body composition of largemouth bass Micropterus salmoides. Journal of the World Aquaculture Society 36:129 134. Chou, B. S., and S. Y. Shiau. 1996. Optimal dietary lipid level for growth of juvenile hybrid tilapia, Oreochromis niloticus Oreochromis aureus. Aquaculture 143:185195. De Silva, S. S., R. M. Gunasekera, G. Gooley, and B. A. Ingram. 2001. Growth of Australian shortn eel (Anguilla australis) elvers given different dietary protein and lipid levels. Aquaculture Nutrition 7:5357. dos Santos, J., I. C. Burkow, and M. Jobling. 1993. Patterns of growth and lipid deposition in cod (Gadus morhua L.) fed natural prey and sh-based feeds. Aquaculture 110:173189. Du, Z. Y., P. Clouet, L. M. Huang, P. Degrace, W. H. Zheng, J. G. He, L. X. Tian, and Y. J. Liu. 2008. Utilization of different dietary lipid sources at high level in herbivorous grass carp (Ctenopharyngodon idella): mechanism related to hepatic fatty acid oxidation. Aquaculture Nutrition 14:7792. Du, Z. Y., Y. J. Liu, L. X. Tian, J. T. Wang, Y. Wang, and G. Y. Liang. 2005. Effect of dietary lipid level on growth, feed utilization and body composition by juvenile grass carp (Ctenopharyngodon idella). Aquaculture Nutrition 11:139146. Hamre, K., R. Christiansen, R. Waagb, A. Maage, B. E. Torstensen, B. Lygren, . Lie, E. Wathne, and S. Albrektsen. 2004. Antioxidant vitamins, minerals and lipid levels in diets for Atlantic salmon (Salmo salar, L.): effects on growth performance and llet quality. Aquaculture Nutrition 10:113123. Ingram, B. A., S. Sungan, G. J. Gooley, S. Y. Sim, D. Tinggi, and S. S. De Silva. 2005. Induced spawning, larval development and rearing of two indigenous Malaysian mahseer, Tor tambroides and T. douronensis. Aquaculture Research 36:983995. Ingram, B. A., S. Sungan, D. Tinggi, S. Y. Sim, G. J. Gooley, and S. S. De Silva. 2007. Development in the spawning of Tor tambroides and T. douronensis in captivity. Pages 123126 in S. S. Siraj, A. Christianus, C. K. Ng, and S. S. De Silva, editors. Proceedings of the international symposium on the mahseer. Malaysian Fisheries Society, Kuala Lumpur. Kamarudin, M. S., E. Ramezani-Fard, C. R. Saad, and S. A. Harmin. In press. Effects of dietary sh oil replacement by various vegetable oils on growth performance, body composition and fatty acid prole of juvenile Malaysian mahseer, Tor tambroides. Aquaculture Nutrition. DOI: 10.1111/j.1365-2095.2011.00907.x. Kottelat, M. 1998. Fishes of the Nam Theun and Xe Bangfai basins, Laos, with diagnoses of twenty-two new species (Teleostei: Cyprinidae, Balitoridae, Cobitidae, Coiidae and Odontobutidae). Ichthyological Exploration of Freshwaters 9:1128. Lee, S. M., and K. D. Kim. 2001. Effects of dietary protein and energy levels on the growth, protein utilization and body composition of juvenile masu salmon (Oncorhynchus masou Brevoort). Aquaculture Research 32(Supplement 1):3945.
536
RAMEZANI-FARD ET AL. Ramezani-Fard, E., M. S. Kamarudin, S. A. Harmin, and C. R. Saad. 2012. Dietary saturated and omega-3 fatty acids affect growth and fatty acid proles of Malaysian mahseer. European Journal of Lipid Science and Technology 114:185193. Ramezani-Fard, E., M. S. Kamarudin, S. A. Harmin, C. R. Saad, M. K. Abd Satar, and S. K. Daud. 2011a. Ontogenic development of the mouth and digestive tract in larval Malaysian mahseer, Tor tambroides Bleeker. Journal of Applied Ichthyology 27:920927. Ramezani-Fard, E., M. S. Kamarudin, C. R. Saad, and S. A. Harmin. 2011b. Changes over time in muscle fatty acid composition of Malaysian mahseer, Tor tambroides, fed different dietary lipid percentage. African Journal of Biotechnology 10:1825618265. Sargent, J. R., D. R. Tocher, and J. G. Bell. 2002. The lipids. Pages 181257 in J. E. Halver and R. W. Hardy, editors. Fish nutrition, third edition. Academic Press, San Diego, California. Satpathy, B. B., D. Mukherjee, and A. K. Ray. 2003. Effects of dietary protein and lipid levels on growth, feed conversion and body composition in rohu, Labeo rohita (Hamilton), ngerlings. Aquaculture Nutrition 9:1724. Wang, J. T., Y. J. Liu, L. X. Tian, K. S. Mai, Z. Y. Du, Y. Wang, and H. J. Yang. 2005. Effect of dietary lipid level on growth performance, lipid deposition, hepatic lipogenesis in juvenile cobia (Rachycentron canadum). Aquaculture 249:439447. Zuraini, A., M. N. Somchit, M. H. Solihah, Y. M. Goh, A. K. Arifah, M. S. Zakaria, N. Somchit, M. A. Rajion, Z. A. Zakaria, and A. M. Mat Jais. 2006. Fatty acid and amino acid composition of three local Malaysian Channa spp. sh. Food Chemistry 97:674678.
L opez, L. M., E. Durazo, M. T. Viana, M. Drawbridge, and D. P. Bureau. 2009. Effect of dietary lipid levels on performance, body composition and fatty acid prole of juvenile white seabass, Atractoscion nobilis. Aquaculture 289:101105. Martins, D. A., L. M. P. Valente, and S. P. Lall. 2007. Effects of dietary lipid level on growth and lipid utilization by juvenile Atlantic halibut (Hippoglossus hippoglossus, L.). Aquaculture 263:150158. Mishra, K., and K. Samantaray. 2004. Interacting effects of dietary lipid level and temperature on growth, body composition and fatty acid prole of rohu, Labeo rohita (Hamilton). Aquaculture Nutrition 10:359369. Misieng, J. D., M. S. Kamarudin, and M. Musa. 2011. Optimum dietary protein requirement of Malaysian mahseer (Tor tambroides) ngerling. Pakistan Journal of Biological Sciences 14:232235. Mohanta, K. N., S. N. Mohanty, J. K. Jena, and N. P. Sahu. 2008. Optimal dietary lipid level of silver barb, Puntius gonionotus ngerlings in relation to growth, nutrient retention and digestibility, muscle nucleic acid content and digestive enzyme activity. Aquaculture Nutrition 14:350359. Ng, W. K., N. Abdullah, and S. S. De Silva. 2008. The dietary protein requirement of the Malaysian mahseer, Tor tambroides (Bleeker), and the lack of protein-sparing action by dietary lipid. Aquaculture 284:201206. Pei, Z., S. Xie, W. Lei, X. Zhu, and Y. Yang. 2004. Comparative study on the effect of dietary lipid level on growth and feed utilization for gibel carp (Carassius auratus gibelio) and Chinese longsnout catsh (Leiocassis longirostris G unther). Aquaculture Nutrition 10:209216. Rainboth, W. J. 1996. Fishes of the Cambodian Mekong. Food and Agriculture Organization of the United Nations, Rome.

Evidence of Female Heterogamety in Largemouth Bass, Based on Sex Ratio of Gynogenetic Progeny
Robert P. Glennon , Boris Gomelsky , Kyle J. Schneider , Anita M. Kelly & Alf Haukenes
c a b a b b c
J. M. Malone and Son, Incorporated, 1156 Malone Lake Road, Lonoke, Arkansas, 72086, USA
Aquaculture Research Center, Kentucky State University, 103 Athletic Drive, Frankfort, Kentucky, 40601, USA
c
Aquaculture/Fisheries Center, University of Arkansas at Pine Bluff, Mail Slot 4912, Pine Bluff, Arkansas, 71601, USA Version of record first published: 17 Sep 2012.
To cite this article: Robert P. Glennon, Boris Gomelsky, Kyle J. Schneider, Anita M. Kelly & Alf Haukenes (2012): Evidence of Female Heterogamety in Largemouth Bass, Based on Sex Ratio of Gynogenetic Progeny, North American Journal of Aquaculture, 74:4, 537-540 To link to this article: http://dx.doi.org/10.1080/15222055.2012.700906
COMMUNICATION
Robert P. Glennon
J. M. Malone and Son, Incorporated, 1156 Malone Lake Road, Lonoke, Arkansas 72086, USA

Aquaculture Research Center, Kentucky State University, 103 Athletic Drive, Frankfort, Kentucky 40601, USA
Anita M. Kelly and Alf Haukenes

Aquaculture/Fisheries Center, University of Arkansas at Pine Bluff, Mail Slot 4912, Pine Bluff, Arkansas 71601, USA
Abstract
Meiotic gynogenetic progeny in largemouth bass Micropterus salmoides have been obtained by inseminating largemouth bass eggs with UV-irradiated sperm from white bass Morone chrysops or striped bass Morone saxatilis and suppressing the second meiotic division by hydrostatic pressure. The sex composition of gynogenetic progeny was determined by dissection or ultrasound investigation of 1-year-old sh. Among the 21 sh analyzed, 7 sh (33.3%) were male and 14 sh (66.7%) were female. The presence of males in meiotic gynogenetic progeny suggests the existence of female heterogamety (WZ females, ZZ males) in largemouth bass.
The largemouth bass Micropterus salmoides is the most popular game sh in the United States. Females of this species exhibit faster growth and attain larger size than do the males (Padeld 1951). Manipulating the sex ratio of largemouth bass populations is of great interest because of its potential to produce higher ratios of females or monosex female populations for the purposes of creating larger and faster growing sh for aquaculture and trophy sport shing. Several studies have evaluated hormonal sex reversal with the goal of shifting sex ratios under inuence of androgens or estrogens in largemouth bass (Garrett 1989; Porter 1996; Al-Ablani and Phelps 2001; Arslan et al. 2009). Induced triploidy has also been tested in this species for reproduction control and obtaining potentially larger sh (Garrett et al. 1992; Fries et al. 2002; Neal et al. 2004).
The optimal method for production of monosex sh populations involves crossing previously obtained sex-reversed sh with normal breeding sh (Purdom 1993; Donaldson 1996). The scheme of crosses directed to production of monosex progenies depends on the type of heterogamety in given sh species. In shes, both male (XY males, XX females) and female (WZ females, ZZ males) heterogamety are described; sometimes different types of heterogamety are revealed in closely related species (Dabrowski et al. 2000; Devlin and Nagahama 2002). Until now, there were no data on sex determination mechanism in largemouth bass. One of the basic methods for determining the type of heterogamety in sh is based on sex composition of gynogenetic progenies. In the case of male heterogamety (XY/XX), gynogenetic progenies usually consist of females only; the presence of both females and males indicates female heterogamety (WZ/ZZ; Thorgaard 1983; Van Eenennaam et al. 1999; Devlin and Nagahama 2002). This article presents data on sex composition of a gynogenetic progeny in largemouth bass that suggests the existence of female heterogamety in this species. METHODS Experiments on production of gynogenetic progeny in largemouth bass were conducted at the facilities of J. M. Malone and Son, Inc. (Lonoke, Arkansas) in April 2009. To induce gynogenetic development, largemouth bass eggs were inseminated
*Corresponding author: boris.gomelsky@kysu.edu Received January 19, 2012; accepted June 4, 2012
537
538
GLENNON ET AL.
with UV-irradiated sperm from white bass Morone chrysops or striped bass Morone saxatilis. For production of gynogenetic diploids, the second meiotic division in eggs was suppressed by application of hydrostatic pressure. Ovulation was induced in mature largemouth bass females by a single intramuscular injection of human chorionic gonadotropin (HCG, 4,000 IU/kg), after which they were placed into separate 246-L berglass tanks. The time between hormonal injection and ovation was 3640 h. White bass and striped bass males were each given one intramuscular HCG injection (1,000 IU/kg). The sperm obtained was irradiated based on techniques previously used for white bass sperm irradiation in experiments on induced gynogenesis in black crappie Pomoxis nigromaculatus (Gomelsky et al. 2000). Specically, sperm was irradiated at a dose of 1,000 J/m2 with a FisherBiotech UV microprocessor-controlled Crosslinker (FB-UVXL-1000; Fisher Scientic). For irradiation, 2 mL of sperm diluted 10fold with 0.85% NaCl solution was placed in 6-cm-diameter glass Petri dishes. Parameters of pressure shock for suppression of the second meiotic division in largemouth bass were chosen based on published data from experiments on induced triploidy (Garrett et al. 1992; Neal et al. 2004); pressure shock of 8,000 psi for 1 min was initiated 5 min after the eggs were inseminated with irradiated sperm. The fertilized eggs were placed into a 1L hydrostatic pressure chamber and the desired pressure was achieved using a 20 ton hydraulic jack. In total, ovulated eggs were produced from 12 females. Nine trials on insemination of eggs with irradiated sperm and application of hydrostatic pressure were performed; in some trials eggs from several females were mixed. After pressure treatment, eggs were put into separate 246-L berglass tanks for embryo incubation and hatching of larvae. The resulting swim-up gynogenetic larvae were reared in the same tanks in which embryos had been incubated and fed with live zooplankton supplied from ltered pond water for 30 days after hatching. Two hundred and sixty gynogens survived to the fry stage. Sixty gynogenetic sh were randomly selected and reared in 1 m3 tanks at a separate facility on live foods (minnow and bluegill) for 1 year. The remaining 200 gynogenic sh were moved to a different facility for hormonal sex reversal by androgen treatment; the results of that study will be reported in a separate publication. Twenty-one gynogenetic sh from the untreated group survived to the age of one year. On April 19, 2010, 15 sh were lost due to equipment failure; they were dissected for sex determination by macroscopic observation of gonads and their total length was recorded. The six surviving gynogenetic sh from this group were sexed 3 days later by ultrasound investigation using Tela-Vet 1000 equipped with a 58 MHz linear transducer (Classic Medical, Tequesta, Florida). Ultrasound is a noninvasive method that has been shown to accurately determine the sex in several species of sh (Martin et al. 1983; Reimers et al. 1987; Shields et al. 1993; Blythe et al. 1994; Colombo et al. 2004; Masoudifard et al. 2011). The sh were anesthetized with
quinaldine sulfate and placed in a pan of water. Each sh was held submerged in the water against a styrofoam pad to prevent double imaging while the ultrasound was performed. The probe was placed near the abdomen to determine the sex of each sh. Female sh were identied by the presence of eggs in the ovarian sacs; males had no eggs but showed the presence of testes structures. RESULTS From 15 dissected sh, 7 sh were male and 8 sh were female. Males and females had well-developed maturing testes and ovaries, respectively, with normal gonad morphological structure and color for this species. Mean lengths ( SD) of males and females were 27.09 ( 1.17) cm and 25.44 ( 0.95) cm, respectively. The six remaining 1-year-old sh were identied as females by ultrasound investigation. In the summer of 2011, at 2 years of age, further observations of these sh conrmed their identication as females; when pressed on the abdomen, no sperm was expressed and other traits typical for females were observed (e.g., swollen abdomen). In total, of the 21 gynogenetic sh analyzed, 7 (33.3%) were male and 14 (66.7%) were female. DISCUSSION The use of heterologous sperm, instead of sperm from the same species, is a very effective method for experiments on diploid gynogenesis in sh. If hybrids between two species are nonviable, survivors are exclusively of gynogenetic origin (Chourrout 1987; Mims et al. 1997; Dabrowski et al. 2000; Gomelsky et al. 2000). Preliminary experiments conducted at the Aquaculture Research Center of Kentucky State University in 2008 have shown that white bass spermatozoa were capable to fertilizing largemouth bass eggs, but the hybrids obtained were nonviable and perished before or soon after hatching. Since the present study used only irradiated sperm of Morone species for induction of gynogenetic development in largemouth bass eggs, all obtained larvae could be only of gynogenetic origin. In an earlier study, irradiated sperm of white bass was used for induced gynogenesis in another species of the family Centrarchidae, black crappie (Gomelsky et al. 2000). Usually the presence of males in meiotic gynogenetic progenies suggests the existence of female heterogamety (WZ/ZZ) in given species. Other possible factors such as autosomal and environmental inuences on sex determination can also result in appearance of gynogenetic males (Devlin and Nagahama 2002; Flynn et al. 2006). Males in meiotic gynogenetic progenies were described in plaice Pleuronectes platessa (Purdom and Lincoln 1973), blue tilapia Oreochromis aureus (Penman et al. 1987), common barbel Barbus barbus (Castelli 1994), muskellunge Esox masquinongy (Dabrowski et al. 2000), white sturgeon Acipenser transmontanus (Van Eenennaam et al. 1999), shortnose sturgeon A. brevirostrum (Flynn et al. 2006), and
COMMUNICATION
539
some other sh. The frequency of heterozygotes in a meiotic gynogenetic progeny obtained from a female that is heterozygous for some gene depends on the frequency of crossing over between the gene and centromere during prophase of the rst meiotic division; the proportion of heterozygotes increases with increasing crossing over frequency (Thompson 1983; Thorgaard et al. 1983). Similarly, when meiotic gynogenetic progeny is obtained from heterogametic female (WZ), the resulting sex ratio depends on the frequency of crossing over between a sex-determining gene and centromere. If WW sh are viable, then in the absence of crossing over, the meiotic gynogenetic progeny obtained from WZ female will consist of WW females (so called super-females) and ZZ males with ratio 1:1. Crossing over between the sex-determining gene and the centromere will result in the appearance of WZ females and, with increasing recombination rates, the phenotypic sex ratio will be shifted towards prevalence of females. If all three categories are viable, the proportion of ZZ males should be approximately equal to the proportion of WW superfemales (Van Eenennaam et al. 1999; Devlin and Nagahama 2002). One third (33%) of the meiotic gynogenetic progeny of largemouth bass were males. This is similar to the proportion of males in meiotic gynogenetic progenies observed in plaice (37%; Purdom and Lincoln 1973), muskellunge (40%; Dabrowski et al. 2000), and shortnose sturgeon (35%; Flynn et al. 2006). If WW super-females in largemouth bass are viable, their proportion in gynogenetic progeny should be approximately 33%, the same as proportions of ZZ males and WZ females. The WW super-females are of special interest with regard to possible genetic sex regulation, since crossing them with normal ZZ males should produce all-female progeny (WZ). Crossing of WW super-females with phenotypic WW males obtained by sex reversal will allow for production of WW super-females in mass quantities. The existence of female heterogamety in largemouth bass should be conrmed later by identication of WW sh in test crosses.
REFERENCES
Al-Ablani, S. A., and R. P. Phelps. 2001. Induction of feminization in largemouth bass (Micropterus salmoides) by oral administration of estradiol-17 beta or diethylstilbestrol and associated pathological effects. Journal of Aquaculture in the Tropics 16:185195. Arslan, T., R. P. Phelps, and J. A. Osborne. 2009. Effects of oestradiol17 or 17-methyltestosterone administration on gonadal differentiation of largemouth bass Micropterus salmoides (Lacep` ede). Aquaculture Research 40:18131822.
Blythe, B., L. A. Helfrich, W. E. Beal, B. Bosworth, and G. S. Libey. 1994. Determination of sex and maturational status of striped bass (Morone saxatilis) using ultrasonic imaging. Aquaculture 125:175184. Castelli, M. 1994. Study on sex determination in the common barbel (Barbus barbus L.) (Pisces, Cyprinidae) using gynogenesis. Pages 509519 in A. R. Beaumont, editor. Genetics and evolution of aquatic organisms. Chapman and Hall, London. Chourrout, D. 1987. Genetic manipulations in sh: review of methods. Pages 111126 in K. Tiews, editor. Selection, hybridization, and genetic engineering in aquaculture, volume 2. Heenemann Verlags, Berlin. Colombo, R. E., P. S. Wills, and J. E. Garvey. 2004. Use of ultrasound imaging to determine sex of shovelnose sturgeon. North American Journal of Fisheries Management 24:322326. Dabrowski, K., J. Rinchard, F. Lin, M. A. Garcia-Abiado, and D. Schmidt. 2000. Induction of gynogenesis in muskellunge with irradiated sperm of yellow perch proves diploid muskellunge male homogamety. Journal of Experimental Zoology 287:96105. Devlin, R. H., and Y. Nagahama. 2002. Sex determination and sex differentiation in sh: an overview of genetic, physiological, and environmental inuences. Aquaculture 208:191364. Donaldson, E. M. 1996. Manipulation of reproduction in farmed sh. Animal Reproduction Science 42:381392. Flynn, S. R., M. Matsuoka, M. Reith, D. J. Martin-Robichaud, and T. J. Benfey. 2006. Gynogenesis and sex determination in shortnose sturgeon, Acipencer brevirostrum Lesuere. Aquaculture 253:721727. Fries, L. T., G. L. Kurten, J. Isaac Jr., T. Engelhardt, D. Lyon, and D. G. Smith. 2002. Mass production of polyploid Florida largemouth bass for stocking public waters in Texas. Pages 393399 in D. P. Philipp and M. S. Ridgway, editors. Black bass: ecology, conservation, and management. American Fisheries Society, Symposium 31, Bethesda, Maryland. Garrett, G. P. 1989. Hormonal sex control of largemouth bass. Progressive Fish-Culturist 51:146148. Garrett, G. P., M. C. F. Birkner, and J. R. Gold. 1992. Triploidy induction in largemouth bass, Micropterous salmoides. Journal of Applied Aquaculture 1(3):2734. Gomelsky, B., S. D. Mims, R. J. Onders, W. L. Shelton, K. Dabrowski, and M. A. Garcia-Abiado. 2000. Induced gynogenesis in black crappie. North American Journal of Aquaculture 62:3341. Martin, R. W., J. Myers, S. A. Sower, D. J. Phillips, and C. McAuley. 1983. Ultrasonic imaging, a potential tool for sex determination of live sh. North American Journal of Fisheries Management 3:258264. Masoudifard, M., A. R. Vajhi, M. Moghim, R. M. Nazari, A. R. Naghavi, and M. Sohrabnejad. 2011. High validity sex determination of three years old cultured beluga sturgeon (Huso huso) using ultrasonography. Journal of Applied Ichthyology 27:643647. Mims, S. D., W. L. Shelton, O. Linhart, and C. Wang. 1997. Induced meiotic gynogenesis of paddlesh Polyodon spathula. Journal of the World Aquaculture Society 28:334343. Neal, J. W., D. M. Neal, R. L. Noble, and M. V. McGee. 2004. Articial propagation and induction of triploidy in largemouth bass Micropterus salmoides and ploidy discrimination using erythrocyte length. Journal of the World Aquaculture Society 35:4654. Padeld, J. H., Jr. 1951. Age and growth differentiation between the sexes of the largemouth black bass, Micropterous salmoides (Lacepede). Journal of the Tennessee Academy of Science 26:4254. Penman, D. J., M. S. Shah, J. A. Beardmore, and D. O. F. Skibinski. 1987. Sex ratios of gynogenetic and triploid tilapia. Pages 267276 in K. Tiews, editor. Selection, hybridization, and genetic engineering in aquaculture, volume 2. Heenemann Verlags, Berlin. Porter, M. D. 1996. Effects of methyltestosterone on largemouth bass, Micropterous salmoides. Journal of Applied Aquaculture 6(4):3945. Purdom, C. E. 1993. Genetics and sh breeding. Chapman and Hall, London. Purdom, C. E., and R. F. Lincoln. 1973. Chromosome manipulation in sh. Pages 8389 in J. H. Schr oder, editor. Genetics and mutagenesis of sh. Springer-Verlag, New York.
540
GLENNON ET AL. Thorgaard, G. H. 1983. Chromosome set manipulation and sex control in sh. Pages 405434 in W. S. Hoar, D. J. Randall, and E. M. Donaldson, editors. Fish physiology, volume 9B. Academic Press, New York. Thorgaard, G. H., F. W. Allendorf, and K. L. Knudsen. 1983. Gene-centromere mapping in rainbow trout: high interference over long map distances. Genetics 103:771783. Van Eenennaam, A. L., J. P. Van Eenennaam, J. F. Medrano, and S. I. Doroshov. 1999. Evidence of female heterogametic genetic sex determination in white sturgeon. Journal of Heredity 90:231233.
Reimers, E., P. Landmark, T. Sorsdal, E. Bohmer, and T. Solum. 1987. Determination of salmonids sex, maturation and size: an ultrasound and photocell approach. Aquaculture Magazine 13(November/December): 4144. Shields, R. J., J. Davenport, C. Young, and P. L. Smith. 1993. Oocyte maturation and ovulation in the Atlantic halibut, Hippoglossus hippoglossus (L.), examined using ultrasonography. Aquaculture Research 24:181186. Thompson, D. 1983. The efciency of induced diploid gynogenesis in inbreeding. Aquaculture 33:237244.

Robert P. Glennon , Boris Gomelsky , Kyle J. Schneider , Anita M. Kelly & Alf Haukenes
c a b a b b c
J. M. Malone and Son, Incorporated, 1156 Malone Lake Road, Lonoke, Arkansas, 72086, USA
Aquaculture Research Center, Kentucky State University, 103 Athletic Drive, Frankfort, Kentucky, 40601, USA
c
Aquaculture/Fisheries Center, University of Arkansas at Pine Bluff, Mail Slot 4912, Pine Bluff, Arkansas, 71601, USA Version of record first published: 17 Sep 2012.
To cite this article: Robert P. Glennon, Boris Gomelsky, Kyle J. Schneider, Anita M. Kelly & Alf Haukenes (2012): Evidence of Female Heterogamety in Largemouth Bass, Based on Sex Ratio of Gynogenetic Progeny, North American Journal of Aquaculture, 74:4, 537-540 To link to this article: http://dx.doi.org/10.1080/15222055.2012.700906
COMMUNICATION
Robert P. Glennon
J. M. Malone and Son, Incorporated, 1156 Malone Lake Road, Lonoke, Arkansas 72086, USA

Aquaculture Research Center, Kentucky State University, 103 Athletic Drive, Frankfort, Kentucky 40601, USA
Anita M. Kelly and Alf Haukenes

Aquaculture/Fisheries Center, University of Arkansas at Pine Bluff, Mail Slot 4912, Pine Bluff, Arkansas 71601, USA
Abstract
Meiotic gynogenetic progeny in largemouth bass Micropterus salmoides have been obtained by inseminating largemouth bass eggs with UV-irradiated sperm from white bass Morone chrysops or striped bass Morone saxatilis and suppressing the second meiotic division by hydrostatic pressure. The sex composition of gynogenetic progeny was determined by dissection or ultrasound investigation of 1-year-old sh. Among the 21 sh analyzed, 7 sh (33.3%) were male and 14 sh (66.7%) were female. The presence of males in meiotic gynogenetic progeny suggests the existence of female heterogamety (WZ females, ZZ males) in largemouth bass.
The largemouth bass Micropterus salmoides is the most popular game sh in the United States. Females of this species exhibit faster growth and attain larger size than do the males (Padeld 1951). Manipulating the sex ratio of largemouth bass populations is of great interest because of its potential to produce higher ratios of females or monosex female populations for the purposes of creating larger and faster growing sh for aquaculture and trophy sport shing. Several studies have evaluated hormonal sex reversal with the goal of shifting sex ratios under inuence of androgens or estrogens in largemouth bass (Garrett 1989; Porter 1996; Al-Ablani and Phelps 2001; Arslan et al. 2009). Induced triploidy has also been tested in this species for reproduction control and obtaining potentially larger sh (Garrett et al. 1992; Fries et al. 2002; Neal et al. 2004).
The optimal method for production of monosex sh populations involves crossing previously obtained sex-reversed sh with normal breeding sh (Purdom 1993; Donaldson 1996). The scheme of crosses directed to production of monosex progenies depends on the type of heterogamety in given sh species. In shes, both male (XY males, XX females) and female (WZ females, ZZ males) heterogamety are described; sometimes different types of heterogamety are revealed in closely related species (Dabrowski et al. 2000; Devlin and Nagahama 2002). Until now, there were no data on sex determination mechanism in largemouth bass. One of the basic methods for determining the type of heterogamety in sh is based on sex composition of gynogenetic progenies. In the case of male heterogamety (XY/XX), gynogenetic progenies usually consist of females only; the presence of both females and males indicates female heterogamety (WZ/ZZ; Thorgaard 1983; Van Eenennaam et al. 1999; Devlin and Nagahama 2002). This article presents data on sex composition of a gynogenetic progeny in largemouth bass that suggests the existence of female heterogamety in this species. METHODS Experiments on production of gynogenetic progeny in largemouth bass were conducted at the facilities of J. M. Malone and Son, Inc. (Lonoke, Arkansas) in April 2009. To induce gynogenetic development, largemouth bass eggs were inseminated
*Corresponding author: boris.gomelsky@kysu.edu Received January 19, 2012; accepted June 4, 2012
537
538
GLENNON ET AL.
with UV-irradiated sperm from white bass Morone chrysops or striped bass Morone saxatilis. For production of gynogenetic diploids, the second meiotic division in eggs was suppressed by application of hydrostatic pressure. Ovulation was induced in mature largemouth bass females by a single intramuscular injection of human chorionic gonadotropin (HCG, 4,000 IU/kg), after which they were placed into separate 246-L berglass tanks. The time between hormonal injection and ovation was 3640 h. White bass and striped bass males were each given one intramuscular HCG injection (1,000 IU/kg). The sperm obtained was irradiated based on techniques previously used for white bass sperm irradiation in experiments on induced gynogenesis in black crappie Pomoxis nigromaculatus (Gomelsky et al. 2000). Specically, sperm was irradiated at a dose of 1,000 J/m2 with a FisherBiotech UV microprocessor-controlled Crosslinker (FB-UVXL-1000; Fisher Scientic). For irradiation, 2 mL of sperm diluted 10fold with 0.85% NaCl solution was placed in 6-cm-diameter glass Petri dishes. Parameters of pressure shock for suppression of the second meiotic division in largemouth bass were chosen based on published data from experiments on induced triploidy (Garrett et al. 1992; Neal et al. 2004); pressure shock of 8,000 psi for 1 min was initiated 5 min after the eggs were inseminated with irradiated sperm. The fertilized eggs were placed into a 1L hydrostatic pressure chamber and the desired pressure was achieved using a 20 ton hydraulic jack. In total, ovulated eggs were produced from 12 females. Nine trials on insemination of eggs with irradiated sperm and application of hydrostatic pressure were performed; in some trials eggs from several females were mixed. After pressure treatment, eggs were put into separate 246-L berglass tanks for embryo incubation and hatching of larvae. The resulting swim-up gynogenetic larvae were reared in the same tanks in which embryos had been incubated and fed with live zooplankton supplied from ltered pond water for 30 days after hatching. Two hundred and sixty gynogens survived to the fry stage. Sixty gynogenetic sh were randomly selected and reared in 1 m3 tanks at a separate facility on live foods (minnow and bluegill) for 1 year. The remaining 200 gynogenic sh were moved to a different facility for hormonal sex reversal by androgen treatment; the results of that study will be reported in a separate publication. Twenty-one gynogenetic sh from the untreated group survived to the age of one year. On April 19, 2010, 15 sh were lost due to equipment failure; they were dissected for sex determination by macroscopic observation of gonads and their total length was recorded. The six surviving gynogenetic sh from this group were sexed 3 days later by ultrasound investigation using Tela-Vet 1000 equipped with a 58 MHz linear transducer (Classic Medical, Tequesta, Florida). Ultrasound is a noninvasive method that has been shown to accurately determine the sex in several species of sh (Martin et al. 1983; Reimers et al. 1987; Shields et al. 1993; Blythe et al. 1994; Colombo et al. 2004; Masoudifard et al. 2011). The sh were anesthetized with
quinaldine sulfate and placed in a pan of water. Each sh was held submerged in the water against a styrofoam pad to prevent double imaging while the ultrasound was performed. The probe was placed near the abdomen to determine the sex of each sh. Female sh were identied by the presence of eggs in the ovarian sacs; males had no eggs but showed the presence of testes structures. RESULTS From 15 dissected sh, 7 sh were male and 8 sh were female. Males and females had well-developed maturing testes and ovaries, respectively, with normal gonad morphological structure and color for this species. Mean lengths ( SD) of males and females were 27.09 ( 1.17) cm and 25.44 ( 0.95) cm, respectively. The six remaining 1-year-old sh were identied as females by ultrasound investigation. In the summer of 2011, at 2 years of age, further observations of these sh conrmed their identication as females; when pressed on the abdomen, no sperm was expressed and other traits typical for females were observed (e.g., swollen abdomen). In total, of the 21 gynogenetic sh analyzed, 7 (33.3%) were male and 14 (66.7%) were female. DISCUSSION The use of heterologous sperm, instead of sperm from the same species, is a very effective method for experiments on diploid gynogenesis in sh. If hybrids between two species are nonviable, survivors are exclusively of gynogenetic origin (Chourrout 1987; Mims et al. 1997; Dabrowski et al. 2000; Gomelsky et al. 2000). Preliminary experiments conducted at the Aquaculture Research Center of Kentucky State University in 2008 have shown that white bass spermatozoa were capable to fertilizing largemouth bass eggs, but the hybrids obtained were nonviable and perished before or soon after hatching. Since the present study used only irradiated sperm of Morone species for induction of gynogenetic development in largemouth bass eggs, all obtained larvae could be only of gynogenetic origin. In an earlier study, irradiated sperm of white bass was used for induced gynogenesis in another species of the family Centrarchidae, black crappie (Gomelsky et al. 2000). Usually the presence of males in meiotic gynogenetic progenies suggests the existence of female heterogamety (WZ/ZZ) in given species. Other possible factors such as autosomal and environmental inuences on sex determination can also result in appearance of gynogenetic males (Devlin and Nagahama 2002; Flynn et al. 2006). Males in meiotic gynogenetic progenies were described in plaice Pleuronectes platessa (Purdom and Lincoln 1973), blue tilapia Oreochromis aureus (Penman et al. 1987), common barbel Barbus barbus (Castelli 1994), muskellunge Esox masquinongy (Dabrowski et al. 2000), white sturgeon Acipenser transmontanus (Van Eenennaam et al. 1999), shortnose sturgeon A. brevirostrum (Flynn et al. 2006), and
COMMUNICATION
539
some other sh. The frequency of heterozygotes in a meiotic gynogenetic progeny obtained from a female that is heterozygous for some gene depends on the frequency of crossing over between the gene and centromere during prophase of the rst meiotic division; the proportion of heterozygotes increases with increasing crossing over frequency (Thompson 1983; Thorgaard et al. 1983). Similarly, when meiotic gynogenetic progeny is obtained from heterogametic female (WZ), the resulting sex ratio depends on the frequency of crossing over between a sex-determining gene and centromere. If WW sh are viable, then in the absence of crossing over, the meiotic gynogenetic progeny obtained from WZ female will consist of WW females (so called super-females) and ZZ males with ratio 1:1. Crossing over between the sex-determining gene and the centromere will result in the appearance of WZ females and, with increasing recombination rates, the phenotypic sex ratio will be shifted towards prevalence of females. If all three categories are viable, the proportion of ZZ males should be approximately equal to the proportion of WW superfemales (Van Eenennaam et al. 1999; Devlin and Nagahama 2002). One third (33%) of the meiotic gynogenetic progeny of largemouth bass were males. This is similar to the proportion of males in meiotic gynogenetic progenies observed in plaice (37%; Purdom and Lincoln 1973), muskellunge (40%; Dabrowski et al. 2000), and shortnose sturgeon (35%; Flynn et al. 2006). If WW super-females in largemouth bass are viable, their proportion in gynogenetic progeny should be approximately 33%, the same as proportions of ZZ males and WZ females. The WW super-females are of special interest with regard to possible genetic sex regulation, since crossing them with normal ZZ males should produce all-female progeny (WZ). Crossing of WW super-females with phenotypic WW males obtained by sex reversal will allow for production of WW super-females in mass quantities. The existence of female heterogamety in largemouth bass should be conrmed later by identication of WW sh in test crosses.
REFERENCES
Al-Ablani, S. A., and R. P. Phelps. 2001. Induction of feminization in largemouth bass (Micropterus salmoides) by oral administration of estradiol-17 beta or diethylstilbestrol and associated pathological effects. Journal of Aquaculture in the Tropics 16:185195. Arslan, T., R. P. Phelps, and J. A. Osborne. 2009. Effects of oestradiol17 or 17-methyltestosterone administration on gonadal differentiation of largemouth bass Micropterus salmoides (Lacep` ede). Aquaculture Research 40:18131822.
Blythe, B., L. A. Helfrich, W. E. Beal, B. Bosworth, and G. S. Libey. 1994. Determination of sex and maturational status of striped bass (Morone saxatilis) using ultrasonic imaging. Aquaculture 125:175184. Castelli, M. 1994. Study on sex determination in the common barbel (Barbus barbus L.) (Pisces, Cyprinidae) using gynogenesis. Pages 509519 in A. R. Beaumont, editor. Genetics and evolution of aquatic organisms. Chapman and Hall, London. Chourrout, D. 1987. Genetic manipulations in sh: review of methods. Pages 111126 in K. Tiews, editor. Selection, hybridization, and genetic engineering in aquaculture, volume 2. Heenemann Verlags, Berlin. Colombo, R. E., P. S. Wills, and J. E. Garvey. 2004. Use of ultrasound imaging to determine sex of shovelnose sturgeon. North American Journal of Fisheries Management 24:322326. Dabrowski, K., J. Rinchard, F. Lin, M. A. Garcia-Abiado, and D. Schmidt. 2000. Induction of gynogenesis in muskellunge with irradiated sperm of yellow perch proves diploid muskellunge male homogamety. Journal of Experimental Zoology 287:96105. Devlin, R. H., and Y. Nagahama. 2002. Sex determination and sex differentiation in sh: an overview of genetic, physiological, and environmental inuences. Aquaculture 208:191364. Donaldson, E. M. 1996. Manipulation of reproduction in farmed sh. Animal Reproduction Science 42:381392. Flynn, S. R., M. Matsuoka, M. Reith, D. J. Martin-Robichaud, and T. J. Benfey. 2006. Gynogenesis and sex determination in shortnose sturgeon, Acipencer brevirostrum Lesuere. Aquaculture 253:721727. Fries, L. T., G. L. Kurten, J. Isaac Jr., T. Engelhardt, D. Lyon, and D. G. Smith. 2002. Mass production of polyploid Florida largemouth bass for stocking public waters in Texas. Pages 393399 in D. P. Philipp and M. S. Ridgway, editors. Black bass: ecology, conservation, and management. American Fisheries Society, Symposium 31, Bethesda, Maryland. Garrett, G. P. 1989. Hormonal sex control of largemouth bass. Progressive Fish-Culturist 51:146148. Garrett, G. P., M. C. F. Birkner, and J. R. Gold. 1992. Triploidy induction in largemouth bass, Micropterous salmoides. Journal of Applied Aquaculture 1(3):2734. Gomelsky, B., S. D. Mims, R. J. Onders, W. L. Shelton, K. Dabrowski, and M. A. Garcia-Abiado. 2000. Induced gynogenesis in black crappie. North American Journal of Aquaculture 62:3341. Martin, R. W., J. Myers, S. A. Sower, D. J. Phillips, and C. McAuley. 1983. Ultrasonic imaging, a potential tool for sex determination of live sh. North American Journal of Fisheries Management 3:258264. Masoudifard, M., A. R. Vajhi, M. Moghim, R. M. Nazari, A. R. Naghavi, and M. Sohrabnejad. 2011. High validity sex determination of three years old cultured beluga sturgeon (Huso huso) using ultrasonography. Journal of Applied Ichthyology 27:643647. Mims, S. D., W. L. Shelton, O. Linhart, and C. Wang. 1997. Induced meiotic gynogenesis of paddlesh Polyodon spathula. Journal of the World Aquaculture Society 28:334343. Neal, J. W., D. M. Neal, R. L. Noble, and M. V. McGee. 2004. Articial propagation and induction of triploidy in largemouth bass Micropterus salmoides and ploidy discrimination using erythrocyte length. Journal of the World Aquaculture Society 35:4654. Padeld, J. H., Jr. 1951. Age and growth differentiation between the sexes of the largemouth black bass, Micropterous salmoides (Lacepede). Journal of the Tennessee Academy of Science 26:4254. Penman, D. J., M. S. Shah, J. A. Beardmore, and D. O. F. Skibinski. 1987. Sex ratios of gynogenetic and triploid tilapia. Pages 267276 in K. Tiews, editor. Selection, hybridization, and genetic engineering in aquaculture, volume 2. Heenemann Verlags, Berlin. Porter, M. D. 1996. Effects of methyltestosterone on largemouth bass, Micropterous salmoides. Journal of Applied Aquaculture 6(4):3945. Purdom, C. E. 1993. Genetics and sh breeding. Chapman and Hall, London. Purdom, C. E., and R. F. Lincoln. 1973. Chromosome manipulation in sh. Pages 8389 in J. H. Schr oder, editor. Genetics and mutagenesis of sh. Springer-Verlag, New York.
540
GLENNON ET AL. Thorgaard, G. H. 1983. Chromosome set manipulation and sex control in sh. Pages 405434 in W. S. Hoar, D. J. Randall, and E. M. Donaldson, editors. Fish physiology, volume 9B. Academic Press, New York. Thorgaard, G. H., F. W. Allendorf, and K. L. Knudsen. 1983. Gene-centromere mapping in rainbow trout: high interference over long map distances. Genetics 103:771783. Van Eenennaam, A. L., J. P. Van Eenennaam, J. F. Medrano, and S. I. Doroshov. 1999. Evidence of female heterogametic genetic sex determination in white sturgeon. Journal of Heredity 90:231233.
Reimers, E., P. Landmark, T. Sorsdal, E. Bohmer, and T. Solum. 1987. Determination of salmonids sex, maturation and size: an ultrasound and photocell approach. Aquaculture Magazine 13(November/December): 4144. Shields, R. J., J. Davenport, C. Young, and P. L. Smith. 1993. Oocyte maturation and ovulation in the Atlantic halibut, Hippoglossus hippoglossus (L.), examined using ultrasonography. Aquaculture Research 24:181186. Thompson, D. 1983. The efciency of induced diploid gynogenesis in inbreeding. Aquaculture 33:237244.

Effects of Two Anesthetics on Survival of Juvenile Culter mongolicus during a Simulated Transport Experiment
Mingli Lin
a a a b
, Qidong Wang
a
a b
, Yuguo Xia
a
a b
, Brian R. Murphy , Zhongjie Li , Jiashou Liu
, Tanglin Zhang & Shaowen Ye
State Key Laboratory of Freshwater Ecology and Biotechnology, Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan, 430072, China
b c
Graduate University of Chinese Academy of Sciences, Beijing, 100039, China
Department of Fish and Wildlife Conservation and Conservation Management Institute, Virginia Polytechnic Institute and State University, Blacksburg, Virginia, 24061, USA Version of record first published: 18 Sep 2012.
To cite this article: Mingli Lin, Qidong Wang, Yuguo Xia, Brian R. Murphy, Zhongjie Li, Jiashou Liu, Tanglin Zhang & Shaowen Ye (2012): Effects of Two Anesthetics on Survival of Juvenile Culter mongolicus during a Simulated Transport Experiment, North American Journal of Aquaculture, 74:4, 541-546 To link to this article: http://dx.doi.org/10.1080/15222055.2012.700905
ARTICLE
Effects of Two Anesthetics on Survival of Juvenile Culter mongolicus during a Simulated Transport Experiment
Mingli Lin, Qidong Wang, and Yuguo Xia
State Key Laboratory of Freshwater Ecology and Biotechnology, Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan 430072, China; and Graduate University of Chinese Academy of Sciences, Beijing 100039, China
Brian R. Murphy
Department of Fish and Wildlife Conservation and Conservation Management Institute, Virginia Polytechnic Institute and State University, Blacksburg, Virginia 24061, USA
Zhongjie Li, Jiashou Liu, Tanglin Zhang, and Shaowen Ye*

State Key Laboratory of Freshwater Ecology and Biotechnology, Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan 430072, China
Abstract
Cultivation of the redtail culter Culter mongolicus has been increasing substantially over the last decade along the Yangtze River basin; such increases in production lead to increased juvenile transportation. However, redtail culter juveniles have high transport mortality rates due to a strong stress response that is exacerbated by the accumulation of toxic metabolic waste. Through a 24-h simulated transport experiment (sampling every 6 h), we assessed effects of tricaine methanesulfonate (MS-222) at 10 mg/L of water, 20 mg/L, and 40 mg/L on redtail culter survival and water quality parameters, and similarly we assessed clove oil at 2 mg/L of water, 5 mg/L, and 10 mg/L. None of the anesthetics signicantly improved water quality during the initial 612 h of the experiment. However, MS-222 treatments at the rst 1224 h of the experiment had signicantly higher dissolved oxygen (DO), ammonia, and pH than the control but failed to decrease un-ionized ammonia content. In contrast, the clove oil treatment signicantly reduced the un-ionized ammonia but failed to improve DO and pH at 1224 h. The improvements in water quality were reected in cumulative mortality, MS-222 and clove oil anesthetic treatments having signicantly lower cumulative mortality than the control at 1224 h. The MS-222 and clove oil slowed water quality deterioration, ensured a better transport environment, and improved juvenile survival during transportation. We recommend 5 mg/L clove oil be used when transporting juvenile redtail culters because that concentration improves sh survival while keeping cost low.
Transport of juvenile shes is an important procedure in aquaculture operations and sheries management. The transportation of juveniles in closed polyethylene bags charged with oxygen is considered a simple and effective method, particularly for air travel, and this practice is widespread throughout the world (Berka 1986; Lim et al. 2003). However, water quality generally deteriorates in this closed system due to sh respiration and excretion, pH and dissolved oxygen decreasing and
*Corresponding author: yeshw@ihb.ac.cn Received December 21, 2011; accepted June 4, 2012
ammonia and un-ionized ammonia increasing (Guo et al. 1995; Pramod et al. 2010). Deterioration of water quality stresses the nervous, immunological, and hormonal systems of juvenile shes, and it could cause mortality if it deteriorates beyond normal tolerance limits of sh (Urbinati et al. 2004; Harmon 2009). Additionally, juveniles may be physically damaged during handling and transport procedures (Murai et al. 1979). Therefore, juvenile sh transport in polyethylene bags usually has high
541
542
LIN ET AL.
mortality, especially during long-distance or high-density transportation (Teo et al. 1989; Gomes et al. 2003). Anesthetics are widely used before and during transport to slow metabolism and reduce stress of sh (Harmon 2009; Pramod et al. 2010). Anesthetics reduce the contractile force of the ventricular myocardium and alter gill haemodynamics (Hill et al. 2002) by depressing neuronal activity (Sp ath and Schweickert 1977; Arnolds et al. 2002) and preventing plasma cortisol elevation (Iversen et al. 2003). These neural and physiological variations reduce sh activity and metabolic rate, thus decreasing oxygen uptake and carbon dioxide and ammonia production (Zhuang et al. 2009). Therefore, water quality does not decline as rapidly during sh transport when some anesthetics are added (Berka 1986; Park et al. 2009). When anesthetics are used at proper doses, physical damage may also be reduced because sh are calmer and less active. Consequently, anesthetics generally reduce sh mortality during transportation (Estudillo and Duray 2003; Pramod et al. 2010). Somewhat paradoxically, anesthetics have also been shown to cause an acute stress response immediately after sedation (Trushenski et al. 2012; Zahl et al. 2012). During this acute response, the stress hormone cortisol may increase, leading to increased levels of glucose and lactate in the blood after sedation. Compared with the levels of these chemicals that occur after a sh is stressed by handling, the amounts released in response to anesthesia are low (Zahl et al. 2012). Levels of cortisol, glucose, and lactate in the blood return to normal within 6 h after sedation (Trushenski et al. 2012). A number of anesthetics, including tricaine methanesulfonate (MS-222), benzocaine, 2-phenoxyethanol, quinaldine sulphate, metomidate, and lidocaine have been used in juvenile sh transportation (Guo et al. 1995; Park et al. 2009; Pramod et al. 2010). Among the broad spectrum of anesthetics, MS-222 appears to be used most frequently (Marking and Meyer 1985; Berka 1986). Clove oil (major constituent eugenol: 2-methoxy-4-[2-propenyl] phenol), is another common sh anesthetic due to its efcacy, affordability, and short withdrawal period (Harper 2003). Although many studies have assessed the effects of anesthetics on ornamental sh transported in polyethylene bags (Teo et al. 1989; Pramod et al. 2010) and on low-density transportation of cultivated juvenile sh (Guo et al. 1995), the effects of anesthetics on juvenile survival and water quality during high-density transport in polyethylene bags are not well studied. Redtail culter (or Mongolian culter) Culter mongolicus is an important piscivorous sh in China, widely distributed in lakes, rivers, reservoirs, and other freshwater bodies (CAS 1976). The species is characterized by high price and high market potential (Zhang 2008). However, its population has declined seriously during recent decades, mainly due to overshing, habitat destruction, and water pollution (Zhang 2005; Ye 2007). To rebuild the population, Chinas Ministry of Agriculture initiated stock enhancement programs in some lakes of the Yangtze River basin. Furthermore, pond and cage farming of redtail culters has developed rapidly, increasing the need for juvenile transportation.
However, high mortality has been observed during transport for redtail culters (Xu et al. 2009). Our objective in this study was to evaluate the effects of MS-222 and clove oil on survival of redtail culter juveniles during transportation in polyethylene bags. Furthermore, we monitored water quality parameters to help dene important mechanisms of anesthetic action.
METHODS Experimental sh.We obtained juvenile redtail culters from Niushan Lake Fish Breeding Center, Wuhan, Hubei Province, China. Juveniles were stocked in a cultivation pond for 1 month prior to the experiment. They were fed a commercial pelleted diet in the pond, but feeding was terminated 1 d before the experiment. Juveniles were 41 d old when the experiment began (mean SE: total length = 46.09 0.91 mm, body weight = 0.75 0.04 g). Anesthetics.Two commercial anesthetics, MS-222 (Sigma Chemicals, St. Louis, Missouri) and chemically pure clove oil (Shanghai Feixiang Chemical Factory, Shanghai, China), were used to lightly sedate the sh. Each anesthetic was tested at three concentrations based on previous experience, MS-222 at 10, 20, and 40 mg/L of water and clove oil at 2, 5, and 10 mg/L. Sodium bicarbonate was used in MS-222 solutions to adjust pH to 7.0. Experimental setup.The experiment was conducted on 31 July 2010. Juveniles scoop-netted after being seined from the pond were batch weighted and transferred to bags. Eighty-four bags used in this experiment (7 treatments 4 time samples 3 replicates). The clear plastic bags were 20 L (40 cm wide, 63 cm high) and were sealed to be airtight after adding 5 L of water,15 L oxygen, the sh, and their anesthetic treatment dose. Transport density was 50 g of sh/bag (average, 67 sh/bag) or 10 g of sh per 1 L of water (13 sh/L of water). This density was based on actual transport practices and a previous experiment to measure oxygen consumption of juveniles. All the bags were put in a vehicle and transported for approximately 1 h to the laboratory of the Institute of Hydrobiology, Chinese Academy of Sciences. Natural light was maintained in the laboratory, and an air conditioner was used to maintain the air temperature at 25 C. The experiment lasted for 24 h, and three bags from each treatment were examined at each 6-h interval to assess water quality. Transport water came from the cultivation pond: temperature = 29 C, dissolved oxygen (DO) = 6 mg/L, pH = 7.8, conductivity = 325 S/cm, ammonia = 0.47 mg/L). During the experiment, DO (YSI Model 85 instrument, YSI Inc., Beijing, China) and pH (YSI pH100) were measured immediately after the bags were opened, and total ammonia was titrated within 6 h after the bags were opened. Total ammonia content was determined by the Nessler reagent spectrophotometric method (Huang et al. 1999). Concentration of un-ionized ammonia was calculated by multiplying the total ammonia by the appropriate conversion factor according to the measured water temperature
EFFECTS OF ANESTHETICS IN TRANSPORT
543
and pH (Emerson et al. 1975). Dead sh were counted and weighed at each sampling period (6, 12, 18, and 24 h). Statistical analysis.Cumulative mortality (CM) was calculated as CMi = (WDi /WTi )100%, where CMi was cumulative mortality, WDi was weight of dead individuals, and WTi was total weight of juveniles, and i was the ith hour. To determine signicant differences between treatments, water quality variables and sh mortality were analyzed using oneway ANOVA, followed by least-signicant difference (LSD) multiple-comparisons analysis. Statistical Package for the Social Sciences (SPSS) Version 15.0 was used as statistical software, and the signicance level for statistical analyses was set at = 0.05. RESULTS The DO level remained high in all treatments through the rst 12 h of the experiment (Figure 1), but DO at 18 h and 24 h was signicantly higher in MS-222 treatments than in both the control and clove oil treatments (P < 0.05). The pH declined sharply during the rst 6 h of the experiment, and no signicant differences were observed between control and anesthetic treatments. However, pH of MS-222 treatments declined significantly more slowly than in the control and clove oil treatments after 12 h (P < 0.05; Figure 2). Total ammonia increased rapidly during the experiment. In the control group, total ammonia increased from 0.47 mg/L at the beginning of experiment to 11.01 mg/L (SE, 0.48) at the end (Figure 3). There were no signicant differences in total ammonia between control and anesthetic treatments before the initial 12 h, but concentration of ammonia in the control was signicantly higher than in anesthetic treatments at 18 h and 24 h
20
7.8 Control MS-10 MS-20 MS-40 CO-2 CO-5 CO-10 a a 7.2 ab b a a b 6.8 0 6 12 18 24 b bc bc c
7.6
pH value
7.4
7.0
Time(hours)
FIGURE 2. Changes of pH values in water used in simulated transportation of redtail culter juveniles. The water was treated with one of three MS-222 treatments or one of three clove oil treatments (the treatment abbreviations are explained in Figure 1). At each time point, values accompanied by different letters are signicantly different.
(P < 0.05). However, the results for total ammonia differed from the results for un-ionized ammonia. The control group had lower un-ionized ammonia than the two anesthetic treatments at 6 h and 12 h (Figure 4), but un-ionized ammonia in the control group increased quickly after 12 h, exceeding the un-ionized ammonia levels for all treatment groups at 24 h. Both MS-222 and clove oil treatments displayed no signicant differences in un-ionized ammonia concentration during the rst 12 h (P > 0.05), but signicantly lower concentrations of un-ionized ammonia were observed in clove oil treatments at 18 h and 24 h (P < 0.05) than in the MS-222 treatments.
14
a a ad ab
12
Dissolved oxygen (mg/L)
15
Ammonia(NH -N, mg/L)
bcd bc c
10
a 10 Control CO-2 CO-5 CO-10 MS-10 MS-20 MS-40 a ab ab abc b bc bc c 0 0 6 12 18 24
Control CO-2 CO-5 CO-10 MS-10 MS-20 MS-40
ab ab ab ab
6 a
b bc c a
b b
0 0 6 12 18 24
Time(hours)
Time(hours)
FIGURE 1. Changes in dissolved oxygen concentration in water used in simulated transportation of redtail culter juveniles. The water was treated with one of three MS-222 treatments (10 mg/L [MS-10], 20 mg/L [MS-20], or 40 mg/L [MS-40]) or one of three clove oil treatments (2 mg/L [CO-2], 5 mg/L [CO-5], or 10 mg/L [CO-10]). At each time point, values accompanied by different letters are signicantly different.
FIGURE 3. Changes in total ammonia in water used in simulated transportation of redtail culter juveniles. The water was treated with one of three MS-222 treatments or one of three clove oil treatments (the treatment abbreviations are explained in Figure 1). At each time point, values accompanied by different letters are signicantly different.
544
.06 Control MS-10 MS-20 MS-40 CO-2 CO-5 CO-10
LIN ET AL.
Un-ionized Ammonia(NH , mg/L)
a a a a ab bc bc bc c a d d b a ab b c a ab ab
.05
the control group at 45.5% (SE, 10.8) was signicantly higher than mortality in all anesthetics treatments (P < 0.05) except clove oil at 10 mg/L. DISCUSSION Our research showed that total ammonia and un-ionized ammonia concentrations increased during simulated transport and pH and DO decreased. Similar results have been observed in other species, including the southern platysh Xiphophorus maculatus, and Indian tiger barb Puntius lamentosus (Amend et al. 1982; Guo et al. 1995; Pramod et al. 2010). As sh respire, they consume oxygen consume and accumulate waste products. Fish excrete carbon dioxide and consume DO through respiration, which decreases water acidity and DO (Harmon 2009). The major pathways for ammonia production are transamination and deamination of adenylates in sh muscle and gill tissue (Randall and Wright 1987). Un-ionized ammonia and ionized ammonia are at a dynamic equilibrium in water. Un-ionized ammonia content increases when pH, temperature, and ionized ammonia concentration increase (Emerson et al. 1975). Water temperature was constant and pH decreased during the experiment. Therefore, the increase in un-ionized ammonia we observed in the control group was due to sharp increases of total ammonia. None of the anesthetics signicantly improved water quality parameters during the 612-h period. However, all levels of anesthetic treatments beneted water quality during the 1224h period, though MS-222 and clove oil had different mechanisms and different impacts on water variables. The MS-222
.04
.03
.02
b b
.01 0 6 12 18 24
Time(hours)
FIGURE 4. Changes in un-ionized ammonia in water used in simulated transportation of redtail culter juveniles. The water was treated with one of three MS-222 treatments or one of three clove oil treatments (the treatment abbreviations are explained in Figure 1). At each time point, values accompanied by different letters are signicantly different.
Mortality of juvenile redtail culters occurred throughout the experiment, and the highest mortality rate was observed during the rst 6 h (Figure 5). At 12 h, mortality did not increase signicantly, and no signicant differences were observed among treatments (P > 0.05). At 18 h, cumulative mortality of the control group was signicantly higher than mortality in all anesthetic treatments (P < 0.05). At 24 h, cumulative mortality in
FIGURE 5. Cumulative mortality of redtail culter juveniles during simulated transportation. The water was treated with one of three MS-222 treatments or one of three clove oil treatments (the treatment abbreviations are explained in Figure 1). At each time point, values accompanied by different letters are signicantly different.
EFFECTS OF ANESTHETICS IN TRANSPORT
545
treatments had signicantly higher DO, ammonia, and pH than the controls during 1224 h, but failed to decrease un-ionized ammonia content. In contrast, the clove oil treatment signicantly reduced the un-ionized ammonia but failed to improve DO and pH during 1224 h. These results have also been observed in southern platysh, where 2-phenoxyethanol, quinaldine sulphate, metomedate, and MS-222 had different effects on water quality (Guo et al. 1995). Juveniles become agitated when transported at high density, which leads to more oxygen consumption and metabolic waste products (Berka 1986). This stress response may explain different effects on water quality variables during different phases. However, the anesthetic mechanisms that affect water metrics and the differences between anesthetics, still are not clearly understood. Generally, mortality increases with time during juvenile transport due to the deterioration of water quality (Pramod et al. 2010). Among the factors affecting sh mortality, DO is considered the most important (Berka 1986). Moreover, high concentration of ammonia in water causes high ammonia levels and pH in sh blood, which damage the red blood cells and gills, affect osmoregulation, and increase the oxygen demand of sh (Lawson 1995). Compared with ionized ammonia, un-ionized ammonia is considered more toxic to sh. Though the maximum safe concentration of un-ionized ammonia is unknown, 0.0125 mg/L is commonly accepted by sh culturists (Meade 1985). Sublethal un-ionized ammonia concentrations are known to cause behavioral, physiological, and histologic changes in sh; high concentrations directly result in mortalities (Evans et al. 2006). Furthermore, water variables can act together; the ability of sh to use oxygen depends on their tolerance to stress, water temperature, pH, and concentrations of carbon dioxide and metabolic products such as ammonia (Berka 1986). Furthermore, un-ionized ammonia toxicity increases when DO is low (Merkens and Downing 1957). Compared with the control group, both MS-222 and clove oil signicantly improved survival of juvenile redtail culters at 6 h and 18 h. This result was similar to that of Pramod et al. (2010), who found that juvenile survival during transportation improved when anesthetics were added. Higher survival of anesthetized sh at and beyond 18 h can be attributed to the better water quality maintained by the actions of the anesthetics. Additionally, we found that mortality was higher in the control group than in anesthetic treatments even at 6 h, despite the fact that no signicant differences in water quality were observed between them. We speculate that this was the result of physical damage avoidance due to stress suppression from anesthetics because clove oil and MS-222 have been shown to reduce sh stress (Iversen et al. 2003; Inoue et al. 2005). Regardless of the mechanisms, our results demonstrate that MS-222 and clove oil are useful in short-term and long-term transport of juvenile redtail culters. In conclusion, our study showed that MS- 222 and clove oil reduced deterioration of water quality, thus improving juve-
nile redtail culter survival during transport. Although MS-222 and clove oil were equally effective, we recommend the use of 5 mg/L clove oil because that anesthetic and dose provide a cost-effective means of improving sh survival during transport in China. However, other countries may restrict the use of clove oil for food sh, so further research into eugenol and other clove oil derivatives is needed. ACKNOWLEDGMENTS We express our thanks to technician Xinnian Chen for assistance in tagging operations. We also thank M. D. Klopfer and two anonymous reviewers for their critical reviews of this manuscript. This research was nancially supported by the National Natural Science Foundation of China (grants 30830025 and 30900182) and National Science and Technology Supporting Program (grant 2012BAD25B08). Participation of B. R. Murphy was supported by the AcornAlcinda Foundation, Lewes, Delaware. REFERENCES
Amend, D. F., T. R. Croy, B. A. Goven, K. A. Johnson, and D. H. McCarthy. 1982. Transportation of sh in closed systems: methods to control ammonia, carbon dioxide, pH, and bacterial growth. Transactions of the American Fisheries Society 111:603611. Arnolds, D. E. W., S. J. Zottoli, C. E. Adams, S. M. Dineen, S. Fevrier, Y. Guo, and A. J. Pascal. 2002. Physiological effects of tricaine on the supramedullary/dorsal neurons of the cunner, Tautogolabrus adspersus. Biological Bulletin 203:188189. Berka, R. 1986. The transport of live sh: a review. EIFAC (European Inland Fisheries Advisory Commission) Technical Paper 48. CAS (Chinese Academy of Sciences). 1976. Fishes of the Yangtze River. CAS, Department of Ichthyology, Institute of Hydrobiology, Science Press, Beijing. Emerson, K., R. C. Russo, R. E. Lund, and R. V. Thurston. 1975. Aqueous ammonia equilibrium calculations: effect of pH and temperature. Journal of the Fisheries Research Board of Canada 32:23792383. Estudillo, C. B., and M. N. Duray. 2003. Transport of hatchery-reared and wild grouper larvae, Epinephelus sp. Aquaculture 219:279290. Evans, J. J., D. J. Pasnik, G. C. Brill, and P. H. Klesius. 2006. Un-ionized ammonia exposure in Nile tilapia: toxicity, stress response, and susceptibility to Streptococcus agalactiae. North American Journal of Aquaculture 68:23 33. Gomes, L. C., R. Roubach, C. A. R. M. Araujo-Lima, A. R. Chippari-Gomes, N. P. Lopes, and E. C. Urbinati. 2003. Effect of sh density during transportation on stress and mortality of juvenile tambaqui Colossoma macropomum. Journal of the World Aquaculture Society 34:7684. Guo, F. C., L. H. Teo, and T. W. Chen. 1995. Effects of anaesthetics on the water parameters in a simulated transport experiment of platysh, Xiphophorus maculatus (G unther). Aquaculture Research 26:265271. Harmon, T. S. 2009. Methods for reducing stressors and maintaining water quality associated with live sh transport in tanks: a review of the basics. Reviews in Aquaculture 1:5866. Harper, C. 2003. Status of clove oil and eugenol for anesthesia of sh. Aquaculture Magazine 29(6):4142. Hill, J. V., W. Davison, and M. E. Forster. 2002. The effects of sh anaesthetics (MS222, metomidate and AQUI-S) on heart ventricle, the cardiac vagus and branchial vessels from Chinook salmon (Oncorhynchus tshawytscha). Fish Physiology and Biochemistry 27:1928. Huang, X., W. Chen, and Q. Cai. 1999. Standard methods for observation and analysis in Chinese ecosystem research network-survey: observation and analysis of lake ecology. Science Press, Beijing.
546
LIN ET AL. Sp ath, M., and W. Schweickert. 1977. The effect of metacaine (MS-222) on the activity of the efferent and afferent nerves in the teleost lateral-line system. Naunyn-Schmiedebergs Archives of Pharmacology 297:916. Teo, L. H., T. W. Chen, and B. H. Lee. 1989. Packaging of the guppy, Poecilia reticulata, for air transport in a closed system. Aquaculture 78:321 332. Trushenski, J. T., J. D. Bowker, B. R. Gause, and B. L. Mulligan. 2012. Chemical and electrical approaches to sedation of hybrid striped bass: induction, recovery, and physiological responses to sedation. Transactions of the American Fisheries Society 141:455467. Urbinati, E. C., J. Sampaio de Abreu, A. C. da Silva Camargo, and M. A. Landinez Parra. 2004. Loading and transport stress of juvenile matrinx a (Brycon cephalus, Characidae) at various densities. Aquaculture 229:389 400. Xu, W., L. W. Geng, Z. H. Jia, C. T. Li, J. S. Yi, and C. B. Zhang. 2009. Preliminary studies on articial propagation of Culter mongolicus mongolicus in Jingpo Lake. Freshwater Fisheries 4:6366. Ye, S. 2007. Studies on sh communities and trophic network model of shallow lakes along the medium reach of the Yangtze River. Doctoral dissertation. Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan. Zahl, I. H., O. Samuelsen, and A. Kiessling. 2012. Anaesthesia of farmed sh: implications for welfare. Fish Physiology and Biochemistry 38:201218. Zhang, T. 2005. Life-history strategies, trophic patterns and community structure in the shes of Lake Biandantang. Doctoral dissertation. Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan. Zhang, X., Z. Ruan, and B. Xiong. 2008. Age and growth characteristics of Culter mongolicus in Poyand Lake. Transactions of Oceanology and Limnology 3:137143. Zhuang, P., B. Xu, L. Z. Zhang, and G. P. Feng, and L. H. Chen. 2009. Effects of MS-222 and clove oil on oxygen consumption rate and ammonia excretion rate of juvenile Chinese sturgeon, Acipenser sinensis. Journal of Fishery Sciences of China 4:612618.
Inoue, L. A. K. A., L. O. B. Afonso, G. K. Iwama, and G. Moraes. 2005. Effects of clove oil on the stress response of matrinx a (Brycon cephalus) subjected to transport. Acta Amazonica 35:289295. Iversen, M., B. Finstad, R. S. McKinley, and R. A. Eliassen. 2003. The efcacy of metomidate, clove oil, Aqui-STM and Benzoak R as anaesthetics in Atlantic salmon (Salmo salar L.) smolts, and their potential stress-reducing capacity. Aquaculture 221:549566. Lawson, T. B. 1995. Fundamentals of aquacultural engineering. Chapman and Hall, New York. Lim, L. C., P. Dhert, and P. Sorgeloos. 2003. Recent developments and improvements in ornamental sh packaging systems for air transport. Aquaculture Research 34:923935. Marking, L. L., and F. P. Meyer. 1985. Are better anesthetics needed in sheries? Fisheries 10(6):25. Meade, J. W. 1985. Allowable ammonia for sh culture. Progressive FishCulturist 47:135145. Merkens, J. C., and K. M. Downing. 1957. The effect of tension of dissolved oxygen on the toxicity of un-ionized ammonia to several species of sh. Annals of Applied Biology 45:521527. Murai, T., J. W. Andrews, and J. W. Muller. 1979. Fingerling American shad: effect of Valium, MS-222, and sodium chloride on handling mortality. Progressive Fish-Culturist 41:2729. Park, I. S., M. O. Park, J. W. Hur, D. S. Kim, Y. J. Chang, Y. J. Kim, J. Y. Park, and S. C. Johnson. 2009. Anesthetic effects of lidocaine-hydrochloride on water parameters in simulated transport experiment of juvenile winter ounder, Pleuronectes americanus. Aquaculture 294:7679. Pramod, P. K., T. P. Sajeevan, A. Ramachandran, S. Thampy, and S. S. Pai. 2010. Effects of two anesthetics on water quality during simulated transport of a tropical ornamental sh, the Indian tiger barb Puntius lamentosus. North American Journal of Aquaculture 72:290297. Randall, D. J., and P. A. Wright. 1987. Ammonia distribution and excretion in sh. Fish Physiology and Biochemistry 3:107120.

Effect of Photoperiod on Growth and Feed Efficiency of Fingerling Walleye

Thomas M. Harder , Gordon G. Gotsch & Robert C. Summerfelt
a b a a b
McGraw Wildlife Foundation, Post Office Box 9, Dundee, Illinois, 60118, USA
Department of Natural Resource Ecology and Management, Iowa State University, Ames, Iowa, 50011-3221, USA Version of record first published: 18 Sep 2012.
To cite this article: Thomas M. Harder, Gordon G. Gotsch & Robert C. Summerfelt (2012): Effect of Photoperiod on Growth and Feed Efficiency of Fingerling Walleye, North American Journal of Aquaculture, 74:4, 547-552 To link to this article: http://dx.doi.org/10.1080/15222055.2012.700907
ARTICLE
Effect of Photoperiod on Growth and Feed Efciency of Fingerling Walleye

Thomas M. Harder* and Gordon G. Gotsch
McGraw Wildlife Foundation, Post Ofce Box 9, Dundee, Illinois 60118, USA
Robert C. Summerfelt
Department of Natural Resource Ecology and Management, Iowa State University, Ames, Iowa 50011-3221, USA
Abstract
We examined the effect of photoperiod on growth and feed efciency of ngerling walleyes Sander vitreus cultured in a water reuse aquaculture system in northern Illinois from fall to late winter. We evaluated three photoperiod regimens over 24-h days: continuous light (24 h), 18-h light, and 12-h light. Twelve, 0.55-m3 tanks (four/treatment) were each stocked with 100 feed-trained walleyes (10.9 kg/m3; 196203 mm total length, 6166 g in weight). Survival over the 131-d culture interval was greater than 98.8% in all photoperiod treatments. All measures of growth were signicantly slower and feeding efciency was lower for sh in the 12 h of light treatment than for the 24-h and 18-h light treatments. Differences in performance between the 24-h and 18-h treatments were not signicant, but all measures of growth and feed efciency were higher for sh in continuous light than in the 18-h group. The ndings support use of continuous, in-tank lighting for intensive culture of ngerlings walleye.
The response of walleyes Sander vitreus to light intensity is of importance in both extensive and intensive culture. Walleyes change from positive phototaxis to negative when they are 32 40 mm. They are attracted to high light intensity (7,800 lx) up to 8 weeks after hatching, but older sh preferred intensities of 24 lx (Bulkowski and Meade 1983). Attraction of fry and small walleye ngerlings to light has been used in pond culture to concentrate sh around in-pond feeders and to aggregate sh for nighttime pond-harvesting operations (Harder and Gotsch 2007). Walleye larviculture has been done with overhead light of 100 lx (Moore 1996) to 680 lx (Colesante 1996). Scherer (1976) found a consistent inverse relationship between overhead light intensity (200, 20, and 2 lx) and vertical position of ngerling walleyes. Tank color and light have been evaluated to determine effects on growth and survival during habituation of pond-cultured juvenile walleyes to dry diets. Tests have included black compared with blue tank walls (Harder and
Summerfelt 1996), uncovered tanks with overhead light of low intensity (<16 lx; Kuipers and Summerfelt 1994), tanks in a lighted room with turbid water (Johnson and Rudacille 2010), tanks in dimly lighted rooms with in-tank lights (Siegwarth and Summerfelt 1992), covered tanks with in-tank lighting (Nagel 1996), and tanks in a dark room with in-tank light (Johnson and Rudacille 2010). Although the visual system of juvenile and adult walleyes is biologically adapted to be sensitive to low light intensity, Clayton et al. (2009) found that very dim overhead light (about 10 lx) with high turbidity (111 NTU) reduced survival, suggesting that although walleyes prefer low light intensity, there is a light threshold below which feeding and survival is negatively affected. Nickum (1986) stated that there has not been a comparison between a scheduled photoperiod and continuous lighting for the culture of walleyes. Longer day length is said to stimulate growth of sh (Boeuf and Le Bail 1999). Continuous light rather
*Corresponding author: tharder@mcgrawwildlife.org Received February 6, 2012; accepted June 1, 2012
547
548
HARDER ET AL.
than a 12-h light : 12-h dark cycle resulted in higher survival of larval walleye. This was attributed to the benecial effect of continuous light in keeping sh dispersed in the water column (Loadman et al. 1986). He recommended a photoperiod of 16 h light: 8 h dark (16 L: 8 D) to simulate summer solstice at midlatitudes in the northern hemisphere (15.23 h light : 8.77 h dark on June 21 in northern Illinois where our study was conducted). The study by Huh et al. (1976) is the only study we found that related photoperiod and growth of juvenile walleyes. Their data showed only a small difference between growth of walleye ngerlings exposed to a photoperiod with either an 8-h or 16-h light interval. The expanding use of indoor, intensive water reuse aquaculture systems (WRAS) for culture of many species of sh, including walleyes (Summerfelt and Penne 2007), provides the opportunity to control environmental conditions to optimize sh growth. These past studies demonstrate that photoperiod has the potential for increasing growth rate and enhancing utilization of feed, but the topic has not been thoroughly studied. The objective of this study was to evaluate growth and measures of feed conversion ratio (FCR) and feed efciency ratio (FER) of juvenile walleyes cultured in a WRAS under three photoperiod treatments over 24-h days: continuous (i.e., 24 h) light (24 L), 18 L, and 12 L (this treatment being slightly longer than the natural photoperiod of 1011 h light for this season at this location).
to measure light intensity underwater, but light intensity at the water surface was 36 lx measured 30 cm below the light. Light duration was controlled for each treatment by a 24-h timer. Feeding schedule.Total daily feed amounts for each tank were dispensed at 5-min intervals within each treatments feeding schedule: 24 L: from 1000 hours one day to 0800 hours the next day (22 h/d); 18 L: from 1000 hours one day to 0200 hours the next (16 h/d); 12 L: from 1000 hours to 2000 hours in the same day (10 h/d). The lights came on at 0800 hours in the 12 L and 18 L treatments and were on continuously in the 24 L treatment. Feed was dispensed with auger feeders controlled by electric timer. Feeding was not done between 0800 and 1000 hours when tank cleaning and water samples were collected. Fish showed a weak startle response to the sudden light-on. Water quality.Water samples were collected between 0800 and 1000 hours from each tank. Daily measurements were taken of temperature and dissolved oxygen (YSI model 550, YSI Inc., Yellow Springs, Ohio). The pH (YSI model 63), nitrite, alkalinity (ALK), and total ammonia nitrogen (TAN; Hach FF-2, Hach Company, Loveland, Colorado) were measured once weekly. Values for carbon dioxide (CO2 ) were calculated: [CO2 ] = ALK 10(6.3-pH) (Summerfelt, S.T. 1996); un-ionized ammonia was calculated as (TAN) (% un-ionized NH3 )/100, where NH3 was at pH and temperature (Piper et al. 1982). Measures of performance. On day 1 of the experiment, 100 sh were stocked into each of 12 tanks, from which a sample of 25 sh were measured to determine initial mean lengths, weights, tank biomass, and feeding rates. Tank means of initial sh total length and weight ranged from 196 to 203 mm and 6166 g. Initial stocked biomass was 6 kg/tank and density was10.9 kg/m3. Thereafter, a sample of 25 sh were weighed and measured at 2-week intervals to obtain interval growth rates and adjust feed amounts to maintain the same 2% of body weight per day. At the conclusion of the 131-d culture interval (March 8, 2011), all sh from each tank were counted, weighed, and measured individually. Growth was expressed as relative growth (weight gain as a percentage of initial weight). Absolute growth rate (mm/d and g/d) was calculated, as was specic (logarithmic) growth rate (SGR as percent per day): SGR = 100 [loge nal weight (loge initial weight/t)], where t represents the number of days between initial and nal weight. Feed conversion ratio (FCR) and feed efciency ratio (FER) were calculated by standard formulas (all weights being grams): FCR = total feed fed/total live weight gain and FER = (live weight gain/total feed fed)100. An ANOVA was used to test differences in performance on the tank means of the growth and food conversion variables among the treatments with = 0.05. When the P-value of
METHODS Culture system.Fish were cultured in a WRAS at the McGraw Wildlife Foundation, Dundee, Illinois (88 17 W, 42 02 N), at an elevation of 247 m (810 ft) msl. The design features of this WRAS were similar to that described by Summerfelt (1996), except that the system at McGraw site included use of liquid oxygen infused into an oxygen cone to supersaturate the water supply to the culture tanks. The culture tanks were maintained at volume of 0.55 m3 and a water depth of 52 cm. The mean water inow (14 L/m) provided an exchange rate of 1.5 exchanges/h. Lighting.Each tank was randomly assigned to a photoperiod treatment, and there were four replicates for each of the three photoperiod treatments. Tanks were covered and located in a dark room, and each tank was equipped with a single submerged light similar in design to that illustrated by Siegwarth and Summerfelt (1992). The submerged lights were constructed with 5-cm, schedule-40 polyvinyl chloride (PVC) pipe with a clear acrylic cap on one end that was positioned 5 cm below the water surface. Light was emitted through the clear cap from a single 150-mA, 6.3-V DC lamp. These small lamps require only 0.945 W of energy per bulb. Power was supplied to the lights from a single battery charger, which had an output of 12 V DC at 2 amp. Except for a small opening under the feeder and another for the water inlet, the tanks were covered with foam board, and overhead uorescent lights were turned off. We were unable
EFFECT OF PHOTOPERIOD ON FINGERLING WALLEYE
549
TABLE 1. Water quality means during 131-d culture interval of walleyes in photoperiod testing, including dissolved oxygen (DO), total ammonia nitrogen (TAN), un-ioinized ammonia (UIA), alkalinity (ALK), and CO2 .
Photoperiod (hours of light/d) 24 18 12
Concentration (mg/L)
pH 7.7 7.6 7.7
DO 8.9 8.2 8.5
NO2 0.6 0.5 0.5
TAN 0.38 0.33 0.30
UIA 0.007 0.006 0.006
ALK 202 200 201
CO2 10.1 9.9 9.2
21.9 21.9 21.8
an ANOVA was signicant, Fishers protected least signicant difference multiple range test was used to determine differences among treatments. Fish and feed.The walleyes used in this study were cultured from gametes obtained from Wolf Lake, Wisconsin, in 2010. Eggs were water hardened on site then disinfected with povidone-iodine (100 mg/L of water for 10 min) before transport to the McGraw Foundation for incubation. They were hatched in May, stocked in a fertilized pond, and pond reared for about 45 d. After harvest from the pond, they were transferred to indoor tanks where they were habituated to a commercial dry diet (Otohime C2, Reed Mariculture, San Jose, California) and subsequently weaned to 1-mm WG 9206 (Nelsons Silver Cup, Tooele, Utah), which was increased to 4-mm WG as sh grew (phase II). At the onset of the present study (phase III), sh were still on 4-mm WG but were converted to the 4-mm, slow sinking, extruded, production steelhead diet (Silver Cup Fish Feed, Tooele, Utah) in December 2010; they remained on this diet until nal inventory on March 8, 2011. The manufactures steelhead diet label indicates typical composition: 46% protein,
16% fat, 1% ber, and 4,325 kcal/kg digestible energy content. The feeding rate was 2% of body weight/d, recalculated every 2 weeks through the 131-d of the study. RESULTS Water Quality Because of the use of supersaturated water, mean dissolved oxygen for all tanks averaged 97% saturation (Table 1). The water supply had an average temperature of 6.9 C (range, 3.8 10.4 C), but it was possible to maintain mean temperature in the culture tanks at 21.821.9 C without heating the water because the tanks and the components of the WRAS were located in a heated room. All other water quality measures were similar between treatments. Survival Survival was approximately 99% in all treatments (Table 2), which is greater than typically reported for intensive culture of advanced ngerlings. At Rathbun Hatchery in southcentral
TABLE 2. Comparative performance of juvenile walleyes cultured in tanks for 131 d with light durations of 24, 18, or 12 h per 24-h day. Within a row, means with the same letters are not signicantly different (P > 0.05).
Photoperiod treatment groups (means SE) Variable Total length (mm) Total weight (g) Relative weighta Length growth (mm/d) Weight growth (g/d) SGRb Growth (% weight gain) FCRc FERd Biomass density (kg/m3) Survival (%) 24 h/d 282.4 218.1 99.5 0.65 1.18 0.96 250.0 1.72 58.3 39.2 99.2 1.2 z 3.2 z 1.2 z 0.01 z 0.03 z 0.02 z 8.1 z 0.07 z 2.4 z 0.6 z 0.5 z 18 h/d 280.8 207.6 96.2 0.63 1.10 0.91 230.8 1.82 54.7 37.2 98.8 2.7 zy 2.6 z 1.8 zy 0.01 z 0.02 z 0.01 z 3.9 z 0.06 z 1.8 z 0.6 z 0.9 z 12 h/d 276.2 189.4 92.2 0.59 0.96 0.84 196.5 2.13 47.0 34.7 99.8 2.3 y 4.7 y 0.2 y 0.02 y 0.03 y 0.02 y 6.9 y 0.03 y 0.6 y 0.7 y 0.5 z
ANOVA P-value 0.04 <0.01 <0.01 0.02 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 0.59
a Relative weight (Wr ) was calculated from the standard weight (Ws ) formula of Anderson and Neumann (1996) as log10 Ws = 5.453 + (3.180 log10 TL), where TL is total length (mm). b SGR is specic growth rate (percent/d) = [ln(nal weight) ln(initial weight)/d]100. c FCR is feed conversion ratio = total feed fed/total live weight gain, where weights are grams. d FER is feed efciency ratio = (live weight gain/total feed fed)100, where weights are grams.
550
HARDER ET AL.
Iowa survival of top-graded phase III walleyes fed in outside tanks and exposed to a natural photoperiod over 90 d averaged 88.9% (range, 8397%) over a 10-year interval (Summerfelt et al. 2011). Growth, Feeding Efciency, and Biomass Relative growth, absolute growth (length and weight), and SGR were not signicantly (P > 0.05) different for sh in the 24 L or 18 L treatments, but sh with continuous light had higher mean values for all measures of growth. All growth measures for both the 24 L and 18 L were signicantly greater than those of the 12 L sh (Table 2). Feed conversion ratio was signicantly lower and feeding efciency ratio signicantly higher (P < 0.01) for the 24 L sh than for the 12 L sh (Table 2). Again, there was no difference between sh in the 24 L and 18 L treatments. Final densities in the three treatments in our study ranged from 34.739.2 kg/m3. The ending tank biomass was highest (39.2 kg/m3, SE = 0.06) for the 24 L treatment. DISCUSSION Temperature treatment means in our study were slightly lower than physiological optimum temperature of 25.3 C at 45 lx, based on measures of metabolic rate (Cai and Summerfelt 1992). Nitrite nitrogen (range, 0.50.6 mg/L) was higher than recommended levels (<0.1 mg/L) for optimal health of coldwater and warmwater sh in intensive culture (Wedemeyer 1996), but nitrite (NO2 ) is in a pH-dependent equilibrium with nitrous acid (HNO2 ), which is the more toxic form. At the pH in our study tanks (7.67.7), all of the total nitrite would be present in the ionized form that is not readily diffusible by the gills (Colt and Tomasso 2001). Water quality measures were similar between treatments and were within the suggested range for optimal sh health (Wedemeyer 1996). Published growth rates for walleye ngerlings vary with sh size; however, smaller sh typically grow faster than larger sh, and temperatures closer to optimum result in faster growth. Growth rates (mm/d) in our study ranged from 0.59 to 0.65 mm/d for sh that were intially196203 mm total length and ending at 276282 mm. Siegwarth and Summerfelt (1992) reported that walleye ngerlings (176216 mm) at 25 C grew at a rate of 0.55 mm/d. Walleyes of larger size (285324 mm) and cultured at lower temperature (20.7 C) had much slower growth (0.31 mm/d; Siegwarth and Summerfelt 1993). Summerfelt (1996) summarized walleye ngerling growth data from several studies and calculated a mean of means growth rate of 0.61 mm/d. Summerfelt et al. (2011) reported phase-III growth of walleye ngerlings from 229 to 250 mm of 1.41.5 mm/d and 1.11.2 g/d. Final densities in the three treatments in our study (34.739.2 kg/m3) were more than twice that of the 16.8 kg/m3 reported for the phase-III culture at the Rathbun Hatchery, Morovia, Iowa, one of the few state hatcheries in the country that use
an intensive culture systems for raising phase-III walleyes (Summerfelt et al. 2011). Phase-III to food-size walleyes in a WRAS were grown to a density of 72.1 kg/m3 (Summerfelt 1996), which suggests that given good water quality, walleyes can be raised to densities comparable to that of many other cultured sh. In regard to research on photoperiod affect, an important point is to be certain that light affects sh growth through better food conversion efciency and not just through stimulated food intake (Boeuf and Le Bail 1999). The present research demonstrates an improvement in growth and feed efciency as photoperiod increases. Fish in our three photoperiod treatments were fed the same daily rate (2%) of the total biomass at 5-min intervals when the light was on (except for 2 h for tank cleaning and water quality measurements). So, there was a substantial difference in the total number of feedings per day among the three treatments ([hours of light 2]60 min/5 min): 264 daily feedings for 24 L, 192 for 18 L, and 120 12 L. If feeding frequency was the treatment rather than photoperiod, the analysis would show the same statistical differences because feeding frequency per day and photoperiod were the same. A previous study, (Phillips et al. 1998) found that growth rates and food conversion did not differ signicantly between feeding frequencies of 9 or 90/d and 3 or 30/d for ngerling walleyes raised in intensive culture with light : dark photoperiod of 16: 8. Thus, the photoperiod effect is the total hours of light for feeding. The signicant difference in FCR among the three treatments, with FCR increasing as photoperiod shortened, substantiates this hypothesis. Natural photoperiod during the interval of our study was even less than that of the shortest photoperiod treatment (12 L); thus, natural light or an articial photoperiod that mimics natural day length during this interval is insufcient to produce the best growth. The ndings and most literature support a hypothesis that continuous light encourages continuous feeding and maximizes growth and feeding efciency. Thus, to obtain maximum fallspring growth benet for culture of juvenile walleyes in a WRAS, it is important to have at least 18-h to 24-h of articial light of low intensity. The basic biological mechanism for enhanced growth with longer photoperiod is not known, but it may result from stimulation of the hypothalamicpituitary axis (Jobling 2010). However, the benet of extended photoperiod has been reported for many unrelated, freshwater and marine shes. First-feeding Arctic char Salvelinus alpinus subjected to 24-h light and continuous feed availability had lower mortality and higher mean nal weights with less size variation within treatments than sh under restricted feeding and ambient photoperiod (Burke et al. 2005). Red seabream Pagrus major reared with 24-h continuous light showed the highest total weight gain and specic growth rate of four different photoperiod regimes tested. Food intake and feeding efciency was also highest for sh in continuous light than for sh in a balanced light : dark treatment of 12:12 h (Biswas et al. 2008).
EFFECT OF PHOTOPERIOD ON FINGERLING WALLEYE
551
Constant photoperiod promotes overall growth regardless of temperature, but constant temperature and photoperiod together produces the best overall growth performance in both male and female yellow perch Perca avescens, and manipulation of both can also inhibit maturation (Shewmon 2005). Petit et al. (2003) reported faster growth of largemouth bass Micropterus salmoides cultured with continuous light than sh held under 12:12 h light : dark cycle. They attributed the faster growth to greater feed consumption of sh exposed to continuous light, even when both groups of sh were fed the same daily ratio. With respect to the cost of energy consumed by using the continuous (24 L) lighting system we used, four tanks with intank lighting would have an expenditure of only $0.0073/d at a charge rate of $0.08/kilowatt-hour. By comparison, if 20-W bulbs were used in overhead lighting of four tanks, the energy cost would be $0.12/d at the same charge rate. Because a 24-h photoperiod is more benecial to walleye growth and feeding efciency than a 12-h photoperiod and the cost of this type of lighting is minimal, a continuous lighting regime should be used in walleye culture. The price of feed varies with the specic attributes of the feed, the quantity, shipping costs, and other variables, but whatever the price, it is generally the single most expensive variable cost item for sh culture. Therefore, it is imperative for the culture of any species to implement cultural technology that will provide for the lowest FCR. In the our study, the FCR was 1.72 for the 24 L sh and 1. 82 for 18 L, both being signicantly lower than for the 2.13 for the 12 L sh (Table 2). This tells us that the quantity of feed per unit weight gain will be 25% greater for sh cultured with light for only 12 h/d than for sh cultured at 24 h/d. Thus, the daily photoperiod becomes an important part of cultural technology for this species that will have substantial impact on production costs. This study further demonstrates the utility of a WRAS for culture of ngerling walleyes for food sh or enhancement stocking. Continuing culture of fall to spring ngerlings under optimum conditions of temperature and light would provide sh at stocking that are far less vulnerable to predators, as well as shorten the time between stocking and when the sh would reach a minimum harvestable length. Larscheid (1995) and Santucci and Wahl (1993) reported that stocking larger walleye ngerlings provides a better economic return than does stocking fry or summer ngerlings. A larger sh is less vulnerable to poststocking mortality from predation. Growing walleyes overwinter in a WRAS for stocking in spring or early summer should provide the stocked sh with enhanced abundance of prey that would increase their growth and shorten the time between stocking and harvest.
boat and nets to collect spawning walleyes. We appreciate the assistance of our summer intern Lucas Brown. REFERENCES
Anderson, R. O., and R. M. Neumann. 1996. Length, weight, and associated structural indices. Pages 447482 in B. R. Murphy and D. W. Willis, editors. Fisheries techniques, 2nd edition. American Fisheries Society, Bethesda, Maryland. Biswas, A. K., M. Seoka, Y. Tanaka, K. Takii, and H. Kumai. 2008. Use of photoperiod manipulation to stimulate the growth performance of juvenile red sea bream (Pagrus major). World Aquaculture 39(2):1215. Boeuf, G., and P. Y. Le Bail. 1999. Does light have an inuence on sh growth? Aquaculture 177:129152. Bulkowski, L., and J. W. Meade. 1983. Changes in phototaxis during early development of walleye. Transactions of the American Fisheries Society 112:445447. Burke, M. G., M. R. Kirk, N. A. MacBeth, D. J. Bevan, and R. D. Moccia. 2005. Inuence of photoperiod and feed delivery on growth and survival of rst-feeding Arctic char. North American Journal of Aquaculture 67: 344350. Cai, Y., and R. C. Summerfelt. 1992. Effects of temperature and size on oxygen consumption and ammonia excretion by walleye. Aquaculture 104:127 138. Clayton, R. D., J. E. Morris, and R. C. Summerfelt. 2009. Effect of turbidity duration on culture of walleye larvae. North American Journal of Aquaculture 71:174177. Colesante, R. T. 1996. Intensive culture of walleye using brine shrimp and formulated diets. Pages 191194 in R. C. Summerfelt, editor. Walleye culture manual. North Central Regional Aquaculture Center Publications, Iowa State University, Ames. Colt, J. E., and J. R. Tomasso. 2001. Hatchery water supply and treatment. Pages 91186 in G. A. Wedemeyer, editor. Fish hatchery management, 2nd edition. American Fisheries Society, Bethesda, Maryland. Harder, T. M., and G. G. Gotsch. 2007. Nighttime lighting and feeding in ponds enhance survival of ngerling walleyes during habituation to manufactured feed. North American Journal of Aquaculture 69:250256. Harder, T. M., and R. C. Summerfelt. 1996. Effects of tank color and size on the success of training walleye ngerlings to formulated feed. Pages 631 636 in G. S. Libey and M. B. Timmons, editors. Successes and failures in commercial recirculating aquaculture: aquacultural engineering society proceedings, volume 2. Northeast Regional Agricultural Engineering Service, NRAES-98, Cornell University, Ithaca, New York. Huh, H. T., H. E. Calbert, and D. A. Stuiber. 1976. Effects of temperature and light on growth of yellow perch and walleye using formulated feed. Transactions of the American Fisheries Society 105:254258. Jobling, M. 2010. Fish culture: the rearing environment. Pages 3360 in N. R. Le Franc ois, M. Jobling, C. Carter, P. U. Blier, and A. Savoie, editors. Finsh aquaculture diversication. CAB International, Oxfordshire, UK. Johnson, J. A., and J. B. Rudacille. 2010. Intensive culture of walleye converted to a dry diet as small ngerlings. Iowa Department of Natural Resources, Federal Aid in Sport Fish Restoration, Project F-160-R, Completion Report, Des Moines. Kuipers, K. L., and R. C. Summerfelt. 1994. Converting pond-reared walleye ngerlings to formulated feeds: effects of diet, temperature, and stocking density. Journal of Applied Aquaculture 4(2):3157. Larscheid, J. G. 1995. Development of an optimal stocking regime for walleyes in East Okoboji Lake, Iowa. Pages 472483 in H. L. Schramm Jr. and R. G. Piper, editors. Uses and effects of cultured shes in aquatic ecosystems. American Fisheries Society, Symposium 15, Bethesda, Maryland. Loadman, N. L., G. E. E. Moodie, and J. A. Mathias. 1986. Signicance of cannibalism in larval walleye (Stizostedion vitreum). Canadian Journal of Fisheries and Aquatic Sciences 43:613618.
ACKNOWLEDGMENTS We thank Steve Newman of the Wisconsin Department of Natural Resources, Escanaba Research Station, for supplying a
552
HARDER ET AL. Scherer, E. 1976. Overhead-light intensity and vertical positioning of the walleye, Stizostedion vitreum vitreum. Journal of the Fisheries Research Board of Canada 33:289292. Shewmon, L. N. 2005. Culture methods for growth enhancement and off-season production of yellow perch Perca avescens. Masters thesis. North Carolina State University, Raleigh. Siegwarth, G. L., and R. C. Summerfelt. 1992. Light and temperature effects on performance of walleye and hybrid walleye ngerlings reared intensively. Progressive Fish-Culturist 54:4953. Siegwarth, G. L., and R. C. Summerfelt. 1993. Performance comparison and growth models for walleyes and walleye sauger hybrids reared for two years in intensive culture. Progressive Fish-Culturist 55:229235. Summerfelt, R. C., J. A. Johnson, and C. P. Clouse. 2011. Culture of walleye, sauger, and hybrid walleye. Pages 451570 in B. A. Barton, editor. Biology, management, and culture of walleye and sauger. American Fisheries Society, Bethesda, Maryland. Summerfelt, R. C., and C. R. Penne. 2007. Nutrient retention by sh in a multispecies recirculating aquaculture facility. International Journal of Recirculating Aquaculture 8:4364. Summerfelt, S. T. 1996. Aquaculture of walleye as a food sh. Pages 215230 in R. C. Summerfelt, editor. Walleye culture manual. North Central Regional Aquaculture Center Publications Ofce, Iowa State University, Ames. Wedemeyer, G. A. 1996. Physiology of sh in intensive culture systems. Chapman and Hall, New York.
Moore, A. A. 1996. Intensive culture of walleye fry on formulated feed. Pages 195197 in R. C. Summerfelt, editor. Walleye culture manual. North Central Regional Aquaculture Center Publications, Iowa State University, Ames. Nagel, T. 1996. Intensive culture of ngerling walleye on formulated feeds. Pages 205207 in R. C. Summerfelt, editor. Walleye culture manual. North Central Regional Aquaculture Center Publications, Iowa State University, Ames. Nickum, J. G. 1986. Walleye. Pages 115126 in R. R. Stickney, editor. Culture of nonsalmonid freshwater shes. CRC Press, Boca Raton, Florida. Petit, G., M. Beauchaud, J. Attia, and B. Buisson. 2003. Food intake and growth of largemouth bass (Micropterus salmoides) held under alternated light/dark cycle (12L:12D) or exposed to continuous light. Aquaculture 228:397 401. Phillips, T. A., R. C. Summerfelt, and R. D. Clayton. 1998. Feeding frequency effects on water quality and growth of walleye ngerlings in intensive culture. Progressive Fish-Culturist 60:18. Piper, R. G., I. B. McElwain, L. E. Orme, J. P. McCraren, L. G. Fowler, and J. R. Leonard. 1982. Fish hatchery management. U.S. Fish and Wildlife Service, Washington, D.C. Santucci, V. J., Jr., and D. H. Wahl. 1993. Factors inuencing survival and growth of stocked walleye (Stizostedion vitreum) in a centrarchid-dominated impoundment. Canadian Journal of Fisheries and Aquatic Sciences 50:1548 1558.

Effects of Winter Feeding on Growth, Body Composition, and Processing Traits of Co-Cultured Blue Catfish, Channel Catfish, and Channel Catfish Blue Catfish Hybrids
Brian G. Bosworth
a a
U.S. Department of Agriculture, Agricultural Research Service, Catfish Genetics Research Unit, Post Office Box 38, Stoneville, Mississippi, 38776, USA Version of record first published: 18 Sep 2012.
To cite this article: Brian G. Bosworth (2012): Effects of Winter Feeding on Growth, Body Composition, and Processing Traits of Co-Cultured Blue Catfish, Channel Catfish, and Channel Catfish Blue Catfish Hybrids, North American Journal of Aquaculture, 74:4, 553-559 To link to this article: http://dx.doi.org/10.1080/15222055.2012.686958
North American Journal of Aquaculture 74:553559, 2012 American Fisheries Society 2012 ISSN: 1522-2055 print / 1548-8454 online DOI: 10.1080/15222055.2012.686958
ARTICLE
Effects of Winter Feeding on Growth, Body Composition, and Processing Traits of Co-Cultured Blue Catsh, Channel Catsh, and Channel Catsh Blue Catsh Hybrids
Brian G. Bosworth*
U.S. Department of Agriculture, Agricultural Research Service, Catsh Genetics Research Unit, Post Ofce Box 38, Stoneville, Mississippi 38776, USA
Abstract
The effects of winter feeding on growth, body composition, and processing yield were compared for co-cultured blue catsh Ictalurus furcatus, channel catsh I. punctatus, and channel catsh blue catsh hybrids. Fish (0.4 0.7 kg) from each genetic group were stocked communally at 5,625 sh/ha in ten 0.04-ha ponds during mid-November. Fish in ve ponds were fed at 2% of initial body weight twice per week; sh in the other ve ponds were not fed. The study was terminated after 14 weeks, and sh were weighed and processed. Analyzed traits were weight change (%), survival, and yields (%) of carcass, shank llet, nugget, head, viscera, skin, intraperitoneal fat, liver, and ovary. Among fed sh, hybrids gained the most weight, channel catsh had intermediate weight gain, and blue catsh gained the least weight. Unfed blue catsh lost more weight than unfed channel catsh and hybrids. Survival was not different among genetic groups or between feeding regimes. Interactions among main effects for processing and body composition traits made generalizations difcult, but carcass yield was consistently higher for blue catsh and hybrids than for channel catsh, higher for females than for males, and higher for fed sh than for unfed sh. Shank llet yield was higher for hybrids than for blue catsh and channel catsh, higher for females than for males, and higher for fed sh than for unfed sh. Nugget yield was higher for blue catsh than for channel catsh and hybrids and was higher for fed sh than for unfed sh. Blue catsh and hybrids had higher intraperitoneal fat yield and lower liver and ovary yield than channel catsh. Fed sh had higher intraperitoneal fat and liver yield than unfed sh. Winter feeding improved growth and llet yield in all groups, but the benets of winter feeding were lower for blue catsh than for channel catsh and hybrid catsh.
Production of channel catsh Ictalurus punctatus is the largest aquaculture enterprise in the United States, representing more than half of aquaculture production by weight. The majority of feeding and growth occurs between April and October, when water temperatures are conducive to catsh feeding activity (Tucker and Hargreaves 2004). During the winter months, many catsh producers withhold feed or provide very little feed because cold water temperatures reduce feeding activity and because increased rainfall often limits the access of feeding equipment to areas around pond levees. In addition, the weight that is lost due to the withholding of feed in the winter months is typically regained fairly quickly through compensatory growth when feeding resumes in the spring (Robinson et al.
*E-mail: brian.bosworth@ars.usda.gov Received November 29, 2011; accepted April 16, 2012
2001). However, studies of channel catsh indicate that winter feeding results in weight gain (Reagan and Robinette 1979; Robinette et al. 1985; Kim and Lovell 1995) or reduced weight loss (Heidinger 1975; Nanninga et al. 2011) in comparison with sh that are unfed. Effects of winter feeding on processing yield (carcass and llet yield) are inconsistent; some reports indicate that winter feeding improves processing yield (Kim and Lovell 1995), while other studies show no effect (Nanninga et al. 2011). Less is known about the effects of winter feeding on blue catsh I . furcatus. Grant and Robinette (1992) reported that winter-fed, market-weight channel catsh gained more weight than blue catsh. Tidwell et al. (1995) reported that winter-fed ngerling channel catsh and blue catsh lost weight at similar
553
554
BOSWORTH
rates, although those authors did not compare weight loss in the fed groups with that in unfed control groups. Production of channel catsh blue catsh hybrids has increased substantially in the last 5 years, but information on the effects of winter feeding on hybrid catsh growth and processing yield is not available. Anecdotal reports indicate that hybrid catsh feed more aggressively than channel catsh at cold water temperatures. The objectives of this study were to determine effects of winter feeding on growth, body composition, and processing traits of co-cultured channel catsh, blue catsh, and channel catsh blue catsh hybrids. Producers and processors can use this information to make decisions about the benets of winter feeding for a particular genetic group of catsh.
METHODS Channel catsh of the U.S. Department of Agriculture (USDA) 103 strain, blue catsh of the D&B strain, and hybrids produced by crossing USDA 103 females with D&B males were used in this study. All sh were approximately 17 months old when the study was initiated, and all were treated similarly (i.e., in terms of stocking density, feeding regime, feed, etc.) prior to the study. Market-weight sh (0.40.7 kg) from each genetic group were stocked communally in ten 0.04-ha ponds during mid-November 2006; each pond was stocked with 75 sh of each group. Fish were stocked communally due to limited pond availability. Previous trials at the Catsh Genetics Research Unit (USDA Agricultural Research Service) have indicated that 5 replicate ponds/treatment are required to obtain adequate statistical power for feeding trials. Therefore, separate culture of the genetic groups would have required 30 pondsmore than were available. From each pond, 15 sh of each genetic group were randomly selected for tagging with a PIT tag (Allex USA, Inc., Dallas, Texas) to allow tracking of individual sh growth. Tagged sh were weighed individually, and gender was recorded at stocking. The additional, untagged sh (60 sh/group for each pond) were counted, weighed as a group, and stocked into each pond to increase the sh density to 5,625 sh/ha. A sample of 50 sh from each genetic group was processed at the time of stocking to provide initial values for processing and body composition traits. Fish in ve ponds were fed a 32% protein, slow-sink pellet (Delta Western, Indianola, Mississippi) at 2% of initial body weight twice per week (Monday and Thursday) regardless of water temperature. Fish in the other ve ponds were not fed. Feeding was terminated after 14 weeks. Feed was withheld for 4 d to allow clearance of feed from the gastrointestinal tracts of the sh. Ponds were then seined, and the tagged and untagged sh were sorted into genetic groups based on visual inspection. Gender and individual weight were recorded for tagged sh. Untagged sh in each pond were sorted by genetic group, counted, and weighed as a group. From each pond, 1015 sh/genetic group were selected for processing; these sh were stunned by
electricity, mechanically deheaded (Baader 166; Baader, Inc., Lubeck, Germany), mechanically skinned (Collum Jet 470-SS; Collum Tool, Greenville, Mississippi), and eviscerated by hand. Carcasses were placed individually in plastic bags and then placed in coolers on ice. The next day, carcasses were lleted by an experienced employee from a catsh processing plant. Gender was recorded, and weights of the gutted carcass, gutted deheaded carcass, gutteddeheadedskinned carcass, shank llet, nugget (rib meat), intraperitoneal fat, and liver were measured from individual sh. Ovary weight was recorded for females. Intraperitoneal fat, liver, and ovary were weighed to the nearest 0.1 g; carcass traits were weighed to the nearest 0.5 g. Weights of head, skin, and viscera were estimated by subtraction of the appropriate carcass weights (for example, weight of viscera = whole weight gutted carcass weight). Survival was determined for each genetic group in fed and unfed treatments. The food conversion ratio (FCR) was estimated for fed sh as follows: (weight of feed fed)/(total weight of sh harvested total weight of sh stocked). Water temperatures for the study period were obtained from a database maintained by the Mississippi State University Extension Service. At the beginning of the study, chlorides in ponds were adjusted to approximately 100 mg/L by the addition of NaCl. Other water quality variables were not measured. Analyzed traits included percent change in weight of tagged individuals, survival, and yields (relative to whole weight) of head, viscera, gutteddeheadedskinned carcass, skin, shank llet, nugget, intraperitoneal fat, liver, and ovary. Traits were analyzed by ANOVA. The model for percent weight change, processing yield traits, and body composition traits included xed effects of gender, feeding regime, and genetic group and relevant interactions; the random effects of pond within feeding regime and pond genetic group gender within feeding regime were also included. Fish weight was used as a covariate for llet yield and visceral fat. Survival was analyzed with a model that included the xed effects of feeding regime and genetic group and the random effects of pond within feeding regime and pond genetic group within feeding regime. The FCR is reported here but was not analyzed since FCR was undened for the unfed treatment. Weight gain of untagged sh was not analyzed because gender data were not recorded for untagged sh and because means for percent weight change were similar to those for individually weighed sh. Statistical analyses were conducted with the MIXED procedure in the Statistical Analysis System version 9.1 (SAS Institute, Inc., Cary, North Carolina), and differences among xed effects were declared signicant at P < 0.05. RESULTS AND DISCUSSION The choice to use a co-culture experimental design was primarily due to the lack of adequate replicate ponds for rearing each genetic group separately. Therefore, the differences observed among genetic groups in response to winter feeding may
WINTER FEEDING EFFECTS ON CATFISH AND CATFISH HYBRIDS
555
be due, at least in part, to the co-culture design used. However, consistencies between the data observed with the co-culture design and data observed during other studies in which sh were cultured as separate groups suggest that the impact of co-culture was not substantial. In addition, the overall effects of the fed versus unfed treatments should be relevant, and comparisons of fed and unfed sh within each genetic group should provide useful information despite the necessity of the co-culture design. Mean water temperature for the study period was 10.8 C (range = 5.120.5 C) and was similar to the average water temperature for the same period as recorded during the previous 7 years (20002006) at this location (10.5 C). Therefore, temperatures for the study period were typical for a winter in the Mississippi Delta, and results of the study were not inuenced by an abnormally warm or cold winter. Survival was not different among genetic groups (survival = 95.1% for blue catsh, 93.4% for channel catsh, and 96.2% for hybrids) and was not affected by feeding regime (survival = 94.5% for fed sh; 95.4% for unfed sh). Grant and Robinette (1992) reported similar high survival rates for market-sized blue catsh and channel catsh (98% and 99%, respectively) that were fed over the winter. Kim and Lovell (1995) and Nanninga et al. (2011) also reported high survival (>90%) and no difference in survival between winter-fed and unfed channel catsh. Fish were visually sorted to genetic group by experienced personnel; misidentication of genetic group is rare for sh of the size used in this study.
Although commercial producers report catsh mortalities associated with winter kill, visceral toxicosis of catsh (VTC), and other diseases during winter months, the results of this study and others suggest that there are no direct benets of winter feeding on catsh survival. However, there has been speculation that winter feeding may indirectly reduce mortalities due to VTC, which is believed to be caused by the ingestion of botulism toxin that is present in sh carcasses decaying on pond bottoms (Gaunt et al. 2007). Winter feeding could reduce the likelihood that catsh will feed on decaying carcasses and may thereby reduce the incidence of VTC (Nanninga et al. 2011). In addition, there is evidence that winter feeding improves survival during the subsequent spring (Kim and Lovell 1995). Initial values for trait means of genetic groups and sexes are presented in Table 1; nal values for trait means of genetic groups, sexes, and feeding regimes are presented in Table 2. Initial values are primarily presented as a reference to reect changes in traits over the course of the study. The main focus of the discussion is related to the data that were collected at the termination of the trial. Hybrid catsh were larger at stocking (672 g) than blue catsh (534 g) and channel catsh (520 g); genders did not differ in initial weight. Percent weight change was affected by genetic group and feeding regime; differential responses of genetic groups to feeding regime resulted in a signicant feeding regime genetic group interaction (Table 2). Gender did not affect percent weight change. Among fed sh, blue catsh had the lowest percent weight gain, channel catsh exhibited
TABLE 1. Least-squares means of initial values for viscera, head, carcass, skin, shank llet, nugget, liver, intraperitoneal (IP) fat, and ovary yields (all values are percentages relative to whole weight) from winter-fed and unfed blue catsh (BC), channel catsh (CC), and channel catsh blue catsh hybrids (HC). For a given effect (genetic group, gender, or interaction) and a given yield characteristic, values with different letters are signicantly different (P < 0.05). Signicant effects are summarized (G = genetic group; S = gender [sex]; G S = genetic group gender interaction).
Effect, group, or statistic Genetic group BC (n = 50) CC (n = 50) HC (n = 50) SE Gender Female (n = 55) Male (n = 95) SE Genetic group gender BC (n = 18) BC (n = 33) CC (n = 17) CC (n = 33) HC (n = 20) HC (n = 29) SE Signicant effects
Viscera 10.3 z 9.7 y 9.9 zy 0.2 10.1 9.8 0.2 10.2 z 10.5 z 10.0 zy 9.4 y 10.3 z 9.6 y 0.3 G, G S
Head 19.5 z 24.1 y 20.4 x 0.3 20.5 z 22.2 y 0.2 19.4 z 19.6 z 22.5 y 25.7 x 19.6 z 21.2 z 0.4 G, S, G S
Carcass 63.5 60.1 63.3 0.3 63.1 z 61.5 y 0.3 63.9 z 63.2 z 61.6 x 58.5 w 63.8 z 62.9 z 0.5 G, S, G S
Skin 6.6 6.2 6.3 0.2 6.3 6.5 0.2 6.6 6.7 5.9 6.4 6.3 6.3 0.3
Shank llet 37.3 z 36.9 z 39.2 y 0.3 38.4 z 37.2 y 0.3 37.5 zy 37.2 z 38.3 y 35.5 x 39.5 w 38.8 yw 0.5 G, S, G S
Nugget 10.9 z 9.1 y 9.2 y 0.2 9.7 9.8 0.1 10.8 11.0 9.1 9.1 9.1 9.3 0.2 G
Liver 1.03 z 1.22 y 1.08 z 0.05 1.12 1.11 0.04 1.02 1.05 1.28 1.17 1.05 1.12 0.07 G
IP fat 3.79 z 2.81 y 4.58 x 0.16 3.70 3.76 0.13 3.70 3.88 2.73 2.89 4.67 4.50 0.22 G
Ovary 0.20 z 1.20 y 0.51 z 0.19
556
BOSWORTH
TABLE 2. Least-squares means of nal values for weight change (%) and for viscera, head, carcass, skin, shank llet, nugget, liver, intraperitoneal (IP) fat, and ovary yields (all yield values are percentages relative to whole weight) from winter-fed and unfed blue catsh (BC), channel catsh (CC), and channel catsh blue catsh hybrids (HC). For a given effect (genetic group, gender, feeding regime, or interaction) and a given yield characteristic, values with different letters are signicantly different (P < 0.05). Signicant effects are summarized (G = genetic group; F = feeding regime; S = gender [sex]; F G = feeding regime genetic group interaction; G S = genetic group gender interaction; F G S = feeding regime genetic group gender interaction).
Effect, group, or statistic
Weight change
Viscera
Head 19.5 z 23.8 y 20.2 x 0.2 20.4 z 21.9 y 0.2 20.5 z 21.8 y 0.2 18.5 z 19.4 zy 22.0 x 24.3 v 18.8 z 20.0 zy 19.8 zy 20.2 zy 23.2 w 25.6 u 20.2 zy 21.7 x 0.4 F, G, S, GS
Carcass 63.3 z 59.9 y 64.2 x 0.2 62.8 z 62.1 y 0.2 62.9 z 62.0 y 0.2
Skin 7.6 z 7.0 y 6.5 x 0.1 6.9 7.1 0.1 6.8 z 7.3 y 0.1
Shank llet 36.2 z 35.6 y 38.3 x 0.2 37.2 z 36.3 y 0.2 37.4 z 36.1 y 0.2 37.3 z 37.0 z 36.9 z 35.5 y 39.3 x 38.3 w 35.3 y 35.3 y 36.2 zy 33.9 v 37.8 zw 37.8 zw 0.4 F, G, S, GS
Nugget 10.6 z 8.7 y 9.2 x 0.1 9.3 z 9.6 y 0.1 9.7 z 9.2 y 0.1
Liver 0.96 z 1.20 y 1.06 z 0.04 1.12 z 1.02 y 0.03 1.15 z 0.99 y 0.04
IP fat 3.2 z 1.5 y 3.1 z 0.1 2.7 2.6 0.1 2.9 z 2.3 y 0.1
Ovary 0.23 z 1.83 y 0.65 z 0.22 0.93 0.88 0.19 0.02 z 2.27 y 0.50 zx 0.44 zx 1.38 w 0.82 x 0.30 G, F G
Genetic group BC (n = 126) 1.3 z 9.7 z CC (n = 118) 5.0 y 9.3 y HC (n = 122) 8.1 x 9.1 y SE 0.7 0.1 Gender Female (n = 149) 4.0 9.8 z Male (n = 217) 4.0 8.9 y SE 0.5 0.1 Feeding regime Fed (n = 191) 11.4 z 9.8 z Unfed (n = 175) 3.4 y 8.9 y SE 0.7 0.1 Feeding regime genetic group gender Fed BC (n = 24) 5.3 z 10.1 z Fed BC (n = 40) 5.6 z 9.7 zx Fed CC (n = 28) 11.3 y 11.0 y Fed CC (n = 33) 13.2 y 8.6 x Fed HC (n = 28) 17.5 x 10.2 z Fed HC (n = 38) 15.6 yx 9.3 x Unfed BC (n = 31) 7.8 w 9.7 zx Unfed BC (n = 43) 7.6 w 9.1 x 1.9 v 9.0 x Unfed CC (n = 25) Unfed CC (n = 32) 2.5 v 8.6 x Unfed HC (n = 25) 0.8 v 8.9 x Unfed HC (n = 31) 0.00 v 7.9 w SE 1.8 0.3 Signicant effects G, F, F, G, S, FG G S, FGS
64.2 z 7.2 z 63.7 z 7.3 z 60.5 y 6.5 y 60.1 y 6.9 zy 64.7 z 6.3 yx 64.5 z 6.3 yx 62.4 x 8.0 w 62.8 x 7.8 w 60.8 y 6.9 zy 58.1 w 7.6 w 64.2 z 6.6 y 63.5 z 6.8 zy 0.4 0.2 F, G, S, F, G, S, G S, GS FGS
11.0 z 1.00 3.4 10.8 z 1.06 3.3 8.4 y 1.44 2.0 9.3 x 1.14 1.7 9.1 x 1.18 3.5 9.7 w 1.07 3.8 10.1 wv 0.92 3.4 10.3 v 0.86 2.8 8.5 y 1.12 1.2 8.6 y 1.08 1.2 8.7 y 1.06 2.7 9.2 x 0.92 2.6 0.2 0.08 0.2 F, G, F, G, S F, G, GS FG
intermediate weight gain, and hybrid catsh had the greatest weight gain (Table 2). Grant and Robinette (1992) also reported a similar pattern of better growth in winter-fed channel catsh (18% increase in body weight) than in winter-fed blue catsh (9% increase in body weight). In the present study, fed hybrid catsh had the highest percent increase in weight, supporting anecdotal observations that hybrids feed more aggressively in cold water than blue catsh or channel catsh. However, the difference between hybrids and channel catsh was not large (4.4%). The differences in growth among genetic groups in response to winter feeding may be associated with inherent biological or behavioral differences among the groups. Feeding activity and
appetite at cold temperatures may simply be reduced in blue catsh relative to channel catsh and hybrid catsh. Another possibility is that because blue catsh tend to school more tightly and to stay higher in the water column than channel catsh, they are less likely to scavenge for sinking feed from the pond bottom. Competition for food in a co-culture situation is another potential cause of growth differences among genetic groups of winter-fed sh. Blue catsh may be less-aggressive feeders than channel catsh or hybrids and may have been outcompeted for the feed. However, Grant and Robinette (1992) also reported slower growth for winter-fed blue catsh than for channel catsh when the two groups were reared separately. In addition, the feeding rate used in the present study (2% of body weight/d
557
twice per week) is considerably higher than that suggested for winter feeding of catsh (Robinson et al. 2001). Therefore, it seems unlikely that competition for food among co-cultured genetic groups resulted in the differential growth response to winter feeding. Another possible factor resulting in the winter growth differences among genetic groups was competition related to sh size. Hybrids grew the fastest and were also larger at stocking, and large sh typically dominate feeding activity. However, channel catsh and blue catsh were similar in size at stocking, but the fed blue catsh had substantially lower growth than the fed channel catsh. In addition, there was no effect of initial weight on percent gain for fed sh within each genetic group; this result indicates that, at least within genetic groups, there were no substantial growth differences related to size-based competition. Hybrid catsh typically grow faster than channel catsh and blue catsh when the three groups are grown separately at warm water temperatures. Taken as a whole, it seems that the differences among groups were primarily attributable to genetic differences rather than to size-based competition, although the effects of stocking weight on growth cannot be determined given the co-culture study design. Unfed blue catsh lost more weight than unfed channel catsh or unfed hybrids. The fairly low weight losses for unfed channel catsh and hybrids were unexpected, as overwinter weight losses in unfed channel catsh are generally reported to be in the 510% range (Lovell and Sirikul 1975; Kim and Lovell 1995). One possible explanation for the relatively low weight loss in unfed hybrids and channel catsh was the presence of signicant numbers of small craysh observed in the ponds at harvest. Ponds were drained and dried prior to stocking and no craysh were noticed, but craysh typically hatch and grow during the winter; therefore, it is possible that catsh in both unfed and fed ponds were consuming craysh. However, because craysh occurred in all ponds, their presence was unlikely to have caused the large differences in growth among genetic groups and between feeding regimes. Although the reported magnitude of benet from winter feeding (i.e., in terms of catsh weight gain) is inconsistent among studies, all studies report greater weight loss in unfed sh than in fed sh. The differences in the magnitude of response to winter feeding may be related to the frequency and duration of feeding. In studies reporting a weight increase in response to winter feeding, sh are typically fed two or more times per week (Reagan and Robinette 1979; Grant and Robinette 1992; Burtle and Newton 1993; Kim and Lovell 1995; present study). In studies that have reported weight loss, sh were fed based on a schedule that required higher water temperatures to permit feeding, thereby reducing the number of times the sh were fed (Tidwell et al. 1995; Nanninga et al. 2011). In several studies that have demonstrated large weight gains in catsh (Reagan and Robinette 1979; Burtle and Newton 1993; Kim and Lovell 1995), the time frame dened as winter included fairly warm
months (October, March, and/or April), which may have allowed for greater feeding activity and higher growth. The FCR of fed sh in this study was 6.2, which was within the range of FCRs (312) that have been reported for other winter feeding trials with channel catsh and blue catsh (Reagan and Robinette 1979; Robinette et al. 1985; Grant and Robinette 1992; Kim and Lovell 1995). The FCRs reported for winter feeding are considerably higher than those typical for the normal feeding season (2.02.5). High FCR (and thus high feed costs relative to sh weight gain) is one of the reasons why many catsh producers do not provide feed during the winter. At current catsh feed prices (US$425.00 per ton) and an FCR of 6.2, the cost of feed to produce 1 kg of sh is approximately $2.90, nearly identical to the current pond-bank catsh price per kilogram. Development of new feeds for winter feeding or the incorporation of winter feeding into new, high-sh-density production systems (e.g., partitioned aquaculture systems; Brune et al. 2004) may result in improved FCRs and better economics for winter feeding. High-density systems may improve winter FCRs because the sh are conned to a small area and may feed more efciently. Groups were co-cultured, and therefore it was not possible to determine the effects of genetic group or gender on FCR. However, the poor growth of winter-fed blue catsh suggests that those sh would have had poorer FCRs than fed channel catsh or hybrids. Grant and Robinette (1992) reported FCRs of 11.59 and 5.89 for winter-fed blue catsh and channel catsh, respectively, supporting the idea that blue catsh will have poorer FCRs than channel catsh during winter feeding. Therefore, I recommend either (1) withholding feed from blue catsh during winter or (2) applying a lower winter feeding rate for blue catsh than for channel catsh or hybrid catsh. Most of the initial and nal values for body composition and processing yield traits were affected by genetic group, gender, and feeding regime; signicant interactions between genetic group and gender and among feeding regime, genetic group, and gender were common (Tables 1, 2). Head yields at the start and end of the study were lowest for blue catsh, intermediate for hybrids, and highest for channel catsh. Male channel catsh had larger heads than females, whereas the difference between male and female hybrids was intermediate and the difference between male and female blue catsh was not significant; thus, a gender genetic group interaction was observed at the start and end of the study. Fed sh had lower head yield than unfed sh. Head size is unlikely to actually increase, but during feed restriction the size of the head (which is primarily bone) remains relatively constant while other body components (fat, muscle, etc.) are mobilized and utilized for energy needs; thus, the head weight expressed as a proportion of whole-body weight increases in unfed sh. The same pattern of gender, genetic group, and gender genetic group interaction effects on head yield has also been reported previously for channel catsh, blue catsh, and hybrid catsh (Bosworth et al. 2004; Jiang et al. 2008). Bosworth and Wolters (2005) observed that after a
558
BOSWORTH
period of feed restriction in the fall, unfed channel catsh had a higher head yield than fed sh, similar to the results of the present study. At the start and end of the study, percent yield of viscera was higher for blue catsh than for channel catsh and hybrids. There was a consistent gender genetic group effect across time due to channel catsh and hybrid females having signicantly more viscera than males, whereas the effects of gender on yield of viscera were minimal in blue catsh. Winter feeding resulted in consistently higher viscera yield across genetic groups. Blue catsh and hybrids had more intraperitoneal fat than channel catsh at the start and end of the study, and fed sh had more intraperitoneal fat than unfed sh. Gender did not affect intraperitoneal fat level. Intraperitoneal fat levels increased as body weight increased in blue catsh and hybrids. Initial and nal liver yields were higher for channel catsh than for blue catsh and hybrids. Females had larger livers than males throughout the study, and fed sh had larger livers than unfed sh. Channel catsh females had larger ovaries than hybrid or blue catsh females at the start and end of the study. A feeding regime genetic group interaction was observed due to the higher ovary yield in fed channel catsh females relative to unfed females, while feeding regime had no effect on ovary yield of hybrid or blue catsh females. The observed differences in ovary yield among blue catsh, channel catsh, and hybrids are similar to those reported previously (Grant and Robinette 1992; Bosworth et al. 2004) and are likely the result of blue catsh maturing at a later age than channel catsh (Graham 1999). Bosworth and Wolters (2005) reported that fall feed restriction of channel catsh resulted in effects on viscera yield similar to those observed in the current study (i.e., higher viscera, intraperitoneal fat, liver, and ovary yields in fed sh than in unfed sh). At the start of the study, the yield of deheadedgutted skinned carcasses (marketed as whole sh by catsh processors) was higher for blue catsh and hybrid catsh than for channel catsh; at the end of the study, the carcass yield was highest for hybrids, intermediate for blue catsh, and lowest for channel catsh. Fed sh had a higher carcass yield than unfed sh, but the genetic group feeding regime interaction was signicant due to the greater effect of feeding regime on carcass yield for blue catsh than for hybrids or channel catsh. Females had higher carcass yields than males, but a signicant genetic group gender interaction was present at both the start and end of the study because the effect of gender on carcass yield was greater for channel catsh than for blue catsh or hybrid catsh. Higher carcass yields for hybrids and blue catsh in comparison with channel catsh have been reported in other studies (Grant and Robinette 1992; Argue et al. 2003; Bosworth et al. 2004; Li et al. 2004; Jiang et al. 2008). The smaller head size in hybrids and blue catsh relative to channel catsh appears to be the main factor inuencing the higher carcass yields for hybrids and blue catsh in comparison with channel catsh. The observed higher carcass yield for females
than for males in the channel catsh and hybrid groups was also reported by Bosworth et al. (2004). Effects of feed restriction on carcass yield have been inconsistent, with some reports showing a response of lower carcass yield in feed-restricted catsh (i.e., as was observed in the present study; Kim and Lovell 1995; Li et al. 2004, 2006) and others showing no effect of feed restriction on carcass yield (Bosworth and Wolters 2005; Nanninga et al. 2011). Initial yield of shank llets, the highest value and highest volume product sold by catsh processors, was greater for hybrid catsh than for blue catsh and channel catsh, which had similar llet yields. At the end of the study, hybrids had the highest shank llet yield, blue catsh had an intermediate llet yield, and channel catsh had the lowest llet yield. Females had a higher llet yield than males at the start and end of the study. Gender genetic group interactions for llet yield were consistent across time and resulted from gender-based differences being large for channel catsh, intermediate for hybrids, and nonsignicant for blue catsh. Fed sh had a higher llet yield than unfed sh when averaged across genetic groups and genders. Differences in llet yield among channel catsh, blue catsh, and hybrids were similar to those reported previously by Bosworth et al. (2004) and Jiang et al. (2008) and indicate that hybrid catsh generally exhibit a llet yield that is superior to the yield obtained from blue catsh or channel catsh. Similar to the results of this study, previous studies have demonstrated that restricting feeding typically reduces llet yield in catsh (Li et al. 2004, 2006; Bosworth and Wolters 2005). Higher llet yields for female channel catsh and female hybrids relative to males have also been reported (Bosworth et al. 2004; Bosworth and Wolters 2005). Nugget yield was higher for blue catsh than for channel catsh and hybrids at the start and end of the study. Fed sh had a higher nugget yield than unfed sh. Males had a higher nugget yield than females at the end of the study, but nugget yield did not differ between genders at the start of the study. Male channel catsh and male hybrids had higher nugget yields than females, while gender did not affect nugget yield in blue catsh, thus resulting in a signicant gender genetic group effect. Bosworth et al. (2004) and Jiang et al. (2008) also reported a higher nugget yield in blue catsh relative to other genetic groups. In summary, winter feeding had positive effects on growth and processing yield of channel catsh, blue catsh, and hybrid catsh. However, the benets of winter feeding on growth and processing yield were greater in channel catsh and hybrids than in blue catsh. Hybrid catsh had the best combined response (i.e., growth and processing yield) to winter feeding. Producers and processors would realize the most benet from winter-fed hybrid catsh relative to winter-fed channel catsh or blue catsh. However, actual economic benets will depend on the price of ngerlings (hybrid ngerlings cost more) and the development of feeding methods that will reduce FCRs for winter feeding. The issue that remains to be
559
addressed is how to efciently and economically feed catsh during winter, when feeding activity is sporadic and not easily observed.
REFERENCES
Argue B. J., Z. Liu, and R. A. Dunham. 2003. Dressout and llet yields of channel catsh, Ictalurus punctatus, blue catsh, Ictalurus furcatus, and their F1, F2 and backcross hybrids. Aquaculture 228:8190. Bosworth, B. G., and W. R. Wolters. 2005. Effects of short-term feed restriction on production, processing, and body shape traits in market-weight channel catsh, Ictalurus punctatus. Aquaculture Research 36:344351. Bosworth, B. G., W. R. Wolters, J. L. Silva, R. S. Chamul, and S. Park. 2004. Comparison of production, meat yield, and meat quality traits of NWAC 103 line channel catsh (Ictalurus punctatus), Norris line channel catsh, and channel catsh female blue catsh male (I. furcatus) F1 hybrids. North American Journal of Aquaculture 66:177183. Brune, D. E., G. Schwartz, A. G. Eversole, J. A. Collier, and T. E. Schwedler. 2004. Partitioned Aquaculture Systems. Southern Regional Aquaculture Center, Publication 4500, Stoneville, Mississippi. Burtle, G. J., and L. G. Newton. 1993. Winter feeding frequency for channel catsh in cages. Progressive Fish-Culturist 55:137139. Gaunt, P. S., S. R. Kalb, and J. R. Barr. 2007. Detection of botulinum type E toxin in channel catsh with visceral toxicosis syndrome using catsh bioassay and endopep mass spectrometry. Journal of Veterinary Diagnostic Investigation 19:349354. Graham, K. 1999. The review of the biology and management of blue catsh. Pages 3749 in E. R. Irwin, W. A. Hubert, C. F. Rabeni, H. L. Schramm Jr., and T. Coon, editors. Catsh 2000: proceedings of the international ictalurid symposium. American Fisheries Society, Symposium 24, Bethesda, Maryland. Grant, J. C., and H. R. Robinette. 1992. Commercially important traits of blue and channel catsh as related to second summer, winter, and third summer growth. Aquaculture 105:3745. Heidinger, R. C. 1975. Growth of hybrid sunshes and channel catsh at low temperatures. Transactions of the American Fisheries Society 104:333 334.
Jiang M., W. H. Daniels, H. J. Pine, and J. A. Chappell. 2008. Production and processing trait comparisons of channel catsh, blue catsh, and their hybrids grown in earthen ponds. Journal of the World Aquaculture Society 39:736745. Kim, M. K., and R. T. Lovell. 1995. Effect of overwinter feeding regime on body weight, body composition and resistance to Edwardsiella ictaluri in channel catsh, Ictalurus punctatus. Aquaculture 134:237246. Li, M. H., E. H. Robinson, B. B. Manning, and B. G. Bosworth. 2004. Effect of dietary protein concentration on production characteristics of pond-raised channel catsh Ictalurus punctatus fed once daily or once every other day to satiation. North American Journal of Aquaculture 66:184190. Li, M. H., E. H. Robinson, D. F. Oberle, and B. G. Bosworth. 2006. Effects of dietary protein concentration and feeding regime on channel catsh, Ictalurus punctatus, production. Journal of the World Aquaculture Society 37: 370377. Lovell, R. T., and B. Sirikul. 1975. Winter feeding of channel catsh. Proceedings of the Annual Conference Southeastern Association of Game and Fish Commissioners 28(1974):208216. Nanninga, A. S., C. R. Engle, N. Stone, and A. E. Goodwin. 2011. Effects of winter feeding of channel catsh on production in multibatch systems. North American Journal of Aquaculture 73:1, 6067. Reagan, R. E., and R. H. Robinette. 1979. Feeding of channel catsh ngerlings in mild and severe winters in Mississippi. Proceedings of the Annual Conference Southeastern Association of Fish and Wildlife Agencies 32(1978):426 428. Robinette, R. H., R. L. Busch, S. H. Newton, C. J. Haskins, S. Davis, and R. R. Stickney.1985. Winter feeding of channel catsh in Mississippi, Arkansas, and Texas. Proceedings of the Annual Conference Southeastern Association of Fish and Wildlife Agencies 36(1982):162171. Robinson, E. H., M. H. Li, and B. B. Manning. 2001. A practical guide to nutrition, feeds, and feeding (second revision). Mississippi Agricultural and Forestry Experiment Station Technical Bulletin 1113. Tidwell, J. H., C. D. Webster, and S. D. Mims. 1995. Effects of two dietary protein levels on body weight and composition of juvenile blue and channel catsh during the winter. Journal of Applied Aquaculture 5: 4554. Tucker, C. S., and J. A. Hargreaves. 2004. Biology and culture of channel catsh. Elsevier, Amsterdam.

Efficacy and Physiological Responses of Grass Carp to Different Sedation Techniques: I. Effects of Various Chemicals on Sedation and Blood Chemistry
Brian R. Gause , Jesse T. Trushenski , John C. Bowzer & James D. Bowker
a a a a b
Fisheries and Illinois Aquaculture Center, Southern Illinois University Carbondale, 1125 Lincoln Drive, Life Science II, Room 173, Carbondale, Illinois, 62901-6511, USA
b
U.S. Fish and Wildlife Service, Aquatic Animal Drug Approval Partnership Program, 4050 Bridger Canyon Road, Bozeman, Montana, 59715, USA Version of record first published: 21 Sep 2012.
To cite this article: Brian R. Gause, Jesse T. Trushenski, John C. Bowzer & James D. Bowker (2012): Efficacy and Physiological Responses of Grass Carp to Different Sedation Techniques: I. Effects of Various Chemicals on Sedation and Blood Chemistry, North American Journal of Aquaculture, 74:4, 560-566 To link to this article: http://dx.doi.org/10.1080/15222055.2012.691013
TECHNICAL NOTE
Efcacy and Physiological Responses of Grass Carp to Different Sedation Techniques: I. Effects of Various Chemicals on Sedation and Blood Chemistry
Brian R. Gause, Jesse T. Trushenski,* and John C. Bowzer
Fisheries and Illinois Aquaculture Center, Southern Illinois University Carbondale, 1125 Lincoln Drive, Life Science II, Room 173, Carbondale, Illinois 62901-6511, USA
James D. Bowker
U.S. Fish and Wildlife Service, Aquatic Animal Drug Approval Partnership Program, 4050 Bridger Canyon Road, Bozeman, Montana 59715, USA
Abstract
Grass carp Ctenopharyngodon idella are commonly used as a low cost, biological control for aquatic vegetation in aquaculture ponds and other private and public waters. In order to minimize the risk of establishing self-sustaining populations in U.S. waters, many states now require grass carp be certied as triploid prior to sale and stocking. To facilitate ploidy testing, grass carp are typically sedated before collecting blood samples. Chemical sedatives such as tricaine methanesulfonate (MS-222) and carbon dioxide (CO2 ) are most commonly used to sedate sh, but there is increasing interest in other chemical sedatives such as benzocaine and eugenol. We evaluated time to induction to Stage IV sedation and recovery, survival, and postsedation blood chemistry of grass carp (301 8 g, mean SE) sedated with MS-222 (150 mg/L), benzocaine (150 mg/L), eugenol (60 mg/L), or CO2 (400 mg/L). Induction times for all sedatives excluding CO2 (14.9 min) were less than 2.4 min (range, 1.52.4 min). Average recovery time after induction was 5.8 min (range, 2.88.3 min) excluding benzocaine, which had a recovery time of 15.4 min. Survival was high and unaffected by sedative option. Plasma cortisol and lactate levels peaked between 0.5 and 1 h postinduction before returning to resting levels at 6 h postinduction. No obvious changes were observed in blood glucose or hematocrit. Each of the sedatives was effective in sedating grass carp, and though changes in blood chemistry indicated that an acute stress response occurred, the response was transient. Although each of the evaluated sedatives would facilitate ploidy testing, some strategies may be more appropriate than others based on FDA approval status and access to the sedative compound, handling time, withdrawal period, and on-site conditions and resources.
Grass carp Ctenopharyngodon idella are commonly used as a low cost, biological control for aquatic vegetation in aquaculture
*Corresponding author: saluski@siu.edu Received October 24, 2011; accepted April 13, 2012
ponds and other private and public waters (Masser 2002). Concerns regarding the establishment of self-sustaining populations of this nonnative species have led to bans on stocking fertile, diploid grass carp in many states (Kelly et al. 2011). Triploid sh, rendered functionally sterile through the interfering effects of ploidy manipulation on gametogenesis (Benfey 1999; Zajicek et al. 2011), may be legally stocked in some states, but triploidy must be veried prior to sale and stocking in those states allowing such sh (Zajicek et al. 2011). A rapid triploidy verication test developed by Wattendorf (1986) requires only a small blood sample for analysis to verify ploidy state. Fish are typically sedated to facilitate blood sampling, but there are a limited number of drugs or chemical sedatives currently available for this purpose that are approved by the U.S. Food and Drug Administration (FDA) or are otherwise made available by the FDA for use. The only drug currently approved by the FDA for the temporary immobilization of sh is tricaine methanesulfonate, most commonly referred to as MS-222. The use of MS-222 is limited to ictalurids, salmonids, esocids, percids, or other laboratory and hatchery shes at water temperatures greater than 10 C. Users of MS-222 must adhere to a 21-d withdrawal period prior to sh being released or slaughtered for consumption. Fish must be fed and kept healthy during this holding period, and in the case of grass carp, must also be maintained in separate holding systems to maintain validity of the ploidy verication tests. Owing to limitations of suitable holding tanks, it is often impractical to hold segregated sh for an extended period of time and doing so probably contributes to additional costs for producers and
560
TECHNICAL NOTE
561
customers alike. Carbon dioxide (CO2 ) is not currently approved by the FDA as a sedative for shes, but is considered a drug of low regulatory priority, meaning that the FDA is unlikely to enforce regulations provided that CO2 is administered according to the stipulations made in the FDAs Enforcement Priorities for Drug Use in Aquaculture documentation (USFDA 2011). Although CO2 has no withdrawal period, creating and maintaining intended sedative concentrations in the eld can be difcult, and is not effective for all shes or environments (e.g., hypercapnia tolerant species, marine environments). A number of crude and puried drugs (e.g., clove, spearmint, and wintergreen oils, quinaldine) are widely used to sedate shes in the eld and laboratory. These compounds, however, are not FDA-approved and can only be used in sh for research purposes, provided that all treated sh are destroyed via incineration, burial, or some other method to ensure they do not enter the food chain. As such, many sheries professionals have desired FDA approval of a chemical sedative that is safe, effective, and affordable, and for which a lengthy withdrawal period is not required. Two drugs that are not currently approved by the FDA, benzocaine (Benzoak; 20% benzocaine; manufacturer: ACD Pharmacueticals AS, Leknes, Norway; U.S. distributor : Frontier Scientic, Logan, Utah) and eugenol (AQUI-S 20E [10% eugenol]; Aqui-S New Zealand, Lower Hutt, New Zealand), may be used as sedatives under an Investigational New Animal Drug (INAD) authorization held by the U.S. Fish and Wildlife Service. Although a 3-d withdrawal period is currently required under the INAD authorization, both of these drugs are being investigated as immediate release sedatives that would allow sh to be released immediately after sedation and could be used on sh intended for human consumption. Although many studies have been conducted to evaluate the effectiveness of chemical sedatives to sedate or anesthetize sh (Gilderhus and Marking 1987; Pirhonen and Schreck 2003; Davis and Grifn 2004), few studies have compared chemical sedatives side by side in terms of their efcacy and effects on shes (Sattari et al. 2009; Trushenski et al. 2012). Although there is considerable interest in the use of drugs or chemicals to sedate sh during some stage of production, the need for an effective sedative is particularly great in triploid grass carp production because of blood testing and ploidy verication requirements. Accordingly, we evaluated the effect of different chemical sedatives (MS-222, benzocaine, eugenol, and CO2 ) on induction and recovery times, and on postsedation survival and blood chemistry of grass carp.
METHODS Sedation procedures.A reference population of triploid grass carp (301 8 g and 30.9 0.3 cm total length, mean SE) was held in an outdoor raceway congured as a partial ow-through system (static raceway, periodically ushed with screened surface water) with supplemental aeration at Keo Fish Farm, Keo, Arkansas. Feed was withheld for 24 h prior to
sampling. Groups of 15 sh were randomly collected from the reference population and transferred into a sedation chamber (114-L cooler) lled with 70 L of culture water (water depth of 10 cm) containing a sedative solution. Sedatives were applied under static conditions as follows: CO2 , 400-mg/L solutions prepared according to the sodium bicarbonatesulfuric acid method described by Post (1979) (analytically veried as 360 mg/L); 150 mg/L benzocaine (Benzoak; 20% benzocaine; manufacturer: ACD Pharmacueticals AS, Leknes, Norway; U.S. distributor: Frontier Scientic, Logan, Utah); 60 mg/L eugenol (AQUI-SE; 50% eugenol; manufacturer: Aqui-S New Zealand, Lower Hutt, New Zealand); and a 150-mg/L solution of MS222 (Finquel; Argent Chemical, Redmond, Washington). Culture water used to prepare all baths was aerated before use, but baths were not aerated after the addition of the chemical sedative or during use. Composite water samples, collected by combining aliquots collected from the sedative baths before and after use, were analyzed in duplicate along with water collected from the holding system for the following: temperature, dissolved oxygen (YSI 550 meter, Yellow Springs Instruments, Yellow Springs, Ohio) conductivity, pH, salinity (Multi-Parameter PCSTestr 35, Eutech/Oakton Instruments, Vernon Hills, Illinois), hardness, alkalinity (digital titrator and reagents, Hach, Loveland, Colorado), and total ammonia nitrogen, nitrite-nitrogen, and nitrate-nitrogen (DR 2800 spectrophotometer and reagents, Hach). All measured water quality characteristics were within ranges appropriate for grass carp (Masser 2002) at the start of the experiment (Table 1). Fish were monitored during sedation to determine induction to Stage IV of anesthesia (Summerfelt and Smith 1990; although we have elected to use the term sedation, anesthesia is the term used by these authors). Stage IV is associated with the total loss of equilibrium, muscle tone, and responsiveness to visual and tactile stimuli, but maintenance of a slow, steady, opercular ventilation rate. After the loss of equilibrium, sedation was veried by slight manual pressure along the trunk and caudal peduncle as a tactile stimulus. Fish were considered sedated when they no longer responded to this stimulus, but the opercular ventilation rate remained steady, albeit reduced, relative to unsedated sh. After induction, blood samples were collected from three sh (t = 0; see below) from the caudal vasculature using heparinized, evacuated, blood collection assemblies (Vacutainer, Becton Dickinson, Franklin Lakes, New Jersey). The remaining 12 sh were then monitored to determine recovery of normal equilibrium and tactile responsiveness. When all sh were able to maintain equilibrium and were responsive to tactile stimulus, the group was considered fully recovered (i.e., recovery time = time for last sh to recover). Since assessment of induction and recovery can be somewhat subjective, bias was minimized by having the same observer apply all stimuli and assess when sh were sedated and recovered. Recovered sh were transferred to a second raceway congured in the same manner as the one housing the reference population; sh in different
562
GAUSE ET AL.
TABLE 1. Water quality characteristics measured during the trial. Values represent means of water samples analyzed in duplicate.
Characteristic Temperature ( C) Dissolved oxygen (mg/L) Total ammonia nitrogen (mg/L) Nitrite-nitrogen (mg/L) Nitrate-nitrogen (mg/L) Alkalinity (mg/L, as CaCO3 ) Hardness (mg/L) Salinity () Conductivity (S/cm) pH
Holding system 16.1 9.62 0.00 0.004 0.75 228 450 0.425 876 8.22
CO2 15.8 8.73 0.02 0.003 0.95 248 488 0.834 1,681 6.32
MS-222 16.0 9.57 0.04 0.003 0.95 208 458 0.432 897 7.24
Benzocaine 15.9 9.66 0.09 0.004 0.9 230 468 0.426 882 8.27
Eugenol 15.8 9.46 0.71a 0.008 1.6 226 440 0.426 880 8.18
a The presence of eugenol results in a yellowgreen color to the water, which can interfere with the Nessler ammonia method used in this trial. Similar results were observed in previous trials using eugenol (Trushenski et al. 2012).
treatment groups were separated by means of raceway dividers crowders positioned at 1-m intervals along the length of the recovery raceway. Sample collection and analysis.Blood samples were then collected from three sh per group at t = 0.5, 1, 2, and 6 h postsedation (three sh per group per time point, individuals were only sampled once). To facilitate handling, all sh were immersed in a bath of metomidate hydrochloride (Aquacalm, Western Chemical, Ferndale, Washington; 510 mg/L for 30 s), a sh sedative known to minimize corticosteroid increase during sampling (Olsen et al. 1995; Davis and Grifn 2004), before sampling; although additional sedation was not necessary for sh sampled at t = 0, these sh were also treated with metomidate hydrochloride to ensure consistent treatment of all sh. Once sh were sedated to a stage where they were easily handled, they were weighed (to the nearest gram) and measured (total length to the nearest 0.5 cm), and a blood sample was collected as described previously. Although metomidate hydrochloride was used to sedate sh before collecting blood samples, all samples were collected within 5 min of capture to minimize the possibility of other confounding responses of handling and blood sample collection as additional stressors. To establish resting blood chemistry characteristics of nonsedated sh, two sh from the reference population were sampled every hour during the course of the experiment (n = 14). After blood collection, sh were returned to a separate area in the recovery raceway and monitored for 24 h. Blood samples were kept on wet ice (<36 h) until analysis. Although this is a somewhat lengthy period of time to hold blood samples prior to analysis, some assays could not be immediately conducted in the eld and samples had to be transported back to the Fisheries and Illinois Aquaculture Center, Carbondale, Illinois. It is possible that levels of metabolically relevant molecules (e.g., glucose and lactate) could have changed slightly during this holding period; however, all samples were treated in the same manner to ensure validity of comparisons among treatments. Hematocrit (Statspin cen-
trifuge, Fisher Scientic, Pittsburgh, Pennsylvania) and glucose (Freestyle Freedom Lite glucose meter, Abbott Laboratories, Abbott Park, Illinois) were determined using aliquots of whole blood, and then the remaining whole blood was centrifuged (3000 g, 4 C, 45 min). Resultant plasma was collected and stored at 80 C until further analysis. Plasma samples were analyzed to determine lactate (Accutrend lactate meter, Roche, Mannheim, Germany), osmolality (Vapro 5520, Wescor, Logan, Utah), and cortisol (EIA 1887, DRG International, Mountainside, New Jersey). Although portable lactate and glucose meters such as those used in this study can slightly underestimate metabolite levels in sh blood relative to laboratory methods, they are considered precise and reliable for use in generating comparative data (Wells and Pankhurst 1999; Venn Beecham et al. 2006). The cortisol kit used has a range of 0800 ng/mL with a sensitivity of 2.5 ng/mL for human samples, and has been validated and used successfully to measure cortisol in samples from a variety of sh species including Nile tilapia Oreochromis niloticus (Delaney et al. 2005), tench Tinca tinca (Owen et al. 2009), common carp Cyprinus carpio (SepiciDinc el et al. 2009), cobia Rachycentron canadum (Trushenski et al. 2010), striped bass Morone saxatilis (Woods et al. 2008), and hybrid striped bass (female white bass M. chrysops male striped bass) (Trushenski et al. 2012). Although multiple sh were sampled from each treatment group at each time point, individuals were group-sedated and housed together after sedation. Therefore, it was determined that individuals did not represent independent observations. Since the experiment lacked true replication, no quantitative statistical analysis was performed and only qualitative comparisons were made from summary statistical values.
RESULTS All sh were successfully induced to Stage IV sedation; however, the observed induction and recovery times varied among sedatives (Figure 1). With the exception of CO2 (induction
TECHNICAL NOTE
563
Benzocaine
E R
CO2
E R
Eugenol
MS-222
10
12
14
16
18
20
Time (min)
I = Induced to Phase IV Sedation E = Maintain Equilibrium R = Recovery; Respond to Tactile Stimulus
FIGURE 1.
Schematic illustrating induction and various stages of recovery of grass carp sedated to Stage IV sedation using various chemical sedatives.
time = 14.9 min), all sedative treatments were successful in inducing Stage IV sedation within approximately 2 min (mean excluding CO2 = 1.8 min; range, 1.52.3 min). Once sedation was achieved, time to regain equilibrium (mean = 4.5 min postinduction; range, 2.58.0 min) was less variable than time to recover tactile responsiveness. With the exception of benzocaine (regained tactile responsiveness = 15.4 min), sh regained tactile responsiveness within 5.8 min (range, 2.88.3 min). All sh ultimately recovered and no sh died during the 24-h postsedation observation period. Physiological responses varied among the sedatives evaluated at different time points following sedation. Plasma lactate varied among treatments at each time point from t = 0 to t = 2 h by as much as 8.5 mmol/L (t = 2 h,) with maximum concentrations observed at t = 0.5 h for some sedatives (eugenol > MS-222) and t = 0 h for the others (CO2 > benzocaine). The range of plasma lactate concentrations at t = 6 was considerably narrower (1.12.7 mmol/L) than that observed at other time points. Plasma osmolality varied at each time point with an overall range of 278361 mOsm/kg (reference population = 296 mOsm/kg). Peak plasma cortisol concentrations occurred at t = 0 for CO2 (164 ng/mL) and at t = 0.5 h for all other sedatives (73166 ng/mL). Cortisol levels in sedated sh appeared to be approaching the levels observed in sh from the reference population (22 ng/mL) by t = 6 h, except for those in sh sedated with MS-222, in which levels increased slightly from t =
2 h to t = 6 h. Hematocrit (range, 2029%; reference population = 22%) and blood glucose (range, 6198 mg/dL; reference population = 79 mg/dL) did not vary much among sedative treatments at any time point. No sh died during the study. During sedation with CO2 , sh were observed piping at the surface (appeared to be gasping for air) and had to be routinely pushed back down into the water to prevent attempts to avoid CO2 narcosis via air breathing. Slight petechial hemorrhaging was observed along the lower ank and opercular area in a few sh before and after sedation, but the occurrence of these hemorrhages did not appear to be related to the sedative used.
DISCUSSION Induction times for three of the four sedatives were considered relatively rapid and would probably be considered acceptable to sheries professionals for sedation of grass carp. Induction time for the CO2 dose used was nearly seven times longer than that for the other chemical sedatives. This lengthy induction time was probably due to the low oxygen demands and metabolic rates of grass carp (Fu et al. 2009) and their ability to avoid CO2 narcosis via air breathing. Furthermore, grass carp used in the current study were held at cooler water temperatures (15.816.1 C) than the water temperature that grass carp tend to prefer (2130 C, Masser 2002). The cooler water temperature associated with the present study probably reduced
564
GAUSE ET AL.
FIGURE 2. Time course of hematological responses (A = plasma cortisol, B = blood glucose, C = hematocrit, D = plasma lactate, and E = plasma osmolality) of grass carp following chemical sedation. Points represent means SE; grey reference bars represent means of values observed for sh sampled from the reference population throughout the course of the experiment.
TECHNICAL NOTE
565
the resting metabolic rate and oxygen demand of our test sh even further. Reduced oxygen demand combined with the relatively high environmental oxygen concentrations (>8 mg/L), may have reduced the need for respiratory gas exchange, opercular ventilation, or gill perfusion rates (Itazawa and Takeda 1978), which may have affected CO2 uptake and the subsequent times to sedation. Additionally, CO2 sedation times may have been slowed because sh were observed piping at the surface and had to be routinely pushed back below the surface of the water to ensure that sh were constantly exposed to the CO2 treated water. Grass carp treated with CO2 and eugenol regained tactile responsiveness and recovered quickly after rst regaining equilibrium (<0.3 min). Grass carp in the benzocaine and MS-222 treatments, on the other hand, required additional time (10.4 and 3.7 min, respectively) to recover from sedation after gaining equilibrium. The chemical similarity of benzocaine (ethyl para-aminobenzoate) and MS-222 (ethyl meta-aminobenzoate; Kiessling et al. 2009) may partially explain why grass carp sedated with these chemicals had similar patterns in which full recovery was preceded by regaining equilibrium by at least several minutes. Recovery results observed for benzocaine and MS-222 in grass carp were considerably longer than that observed in a previous study conducted by Trushenski et al. (2012) using hybrid striped bass, in which hybrid striped bass fully recovered in less than 2 min after regaining equilibrium. The pattern of recovery observed when sh were sedated with CO2 and eugenol, in which sh fully recovered within seconds of regaining equilibrium, was also observed in a study conducted concurrently with ours by Bowzer et al. (2012, this issue), in which grass carp were electrosedated with varying voltages and exposure durations. It is unclear whether the observed differences in recovery times between grass carp and hybrid striped bass were inuenced by differences in resting metabolic rate or other intertaxonomic differences, differential rates of chemical sedative metabolism and excretion during recovery, or some combination of these factors. Physiological responses generally followed the pattern of the generalized stress response (Barton 2002) suggesting that sedation should be considered as a stressor (Zahl et al. 2010). Plasma cortisol in sh sedated with all sedatives except CO2 peaked at 0.5 h postinduction, then dropped steadily over the next 6 h, but was still elevated compared with that in the reference population. Cortisol levels in sh from the CO2 treatment, however, peaked at t = 0, and were probably due to the lengthy time required to induce sedation before sampling at t = 0. Plasma lactate increased rapidly with all sedative options, peaking at 0.5 or 1 h postinduction and steadily decreasing after that. Peak lactate levels were somewhat lower than that observed in hybrid striped bass (Trushenski et al. 2012) and may be due to the decreased oxygen demand observed in grass carp. Grass carp have a resting oxygen demand of 56 mg O2 /kg per hour (Fu et al. 2009) while hybrid striped bass have a resting oxygen demand of 132 mg O2 /kg per hour (Tuncer et al. 1990; Brougher et al. 2005). Another
factor that may have contributed to the differences observed in our study and that reported by Trushenski et al. (2012) was that hybrid striped bass were sedated at higher temperatures (21 C) than were grass carp (16 C). We speculate that the higher water temperature under which hybrid striped bass were tested and their greater overall metabolic demands could have resulted in a more rapid depletion of available oxygen within the tissues, an increase in anaerobic respiration, and ultimately the greater increase in plasma lactate observed in hybrid striped bass (maximum observed value: 17.2 0.2 mmol/L in hybrid striped bass versus 12.4 mmol/L in grass carp). Conversely, the cooler water temperatures under which grass carp were tested in our study, combined with the lower overall metabolic demands of grass carp, may have resulted in lower peak lactate values. Blood glucose and hematocrit values of sedated sh were similar to those of the reference population regardless of sedative used or time of sample collection. Plasma osmolality appeared to vary somewhat among sedatives and over time; however, the observed values did not differ greatly from those observed in the reference sh. The lack of substantial change in these physiological characteristicsglucose, hematocrit, and osmolalitymay suggest a relatively minor and brief acute stress response following sedation. Although differences in the magnitude of blood chemistry responses were observed, the responses to sedation we noted are generally consistent with the results of concurrent work involving sedation of grass carp with pulsed DC electricity (Bowzer et al. 2012). Maximum plasma lactate observed in our study was slightly higher (12.4 mmol/L in CO2 treated sh) than that observed by Bowzer et al. (2012) when sh were sedated with pulsed DC electricity at 150 V for a 10-s exposure (9.4 mmol/L), which was probably due to the prolonged exposure to CO2 and the resultant increase in anaerobic metabolism. Peak plasma cortisol generally occurred 0.5 h postsedation in both studies (CO2 being the exception) with slightly higher peak cortisol occurring in electrosedated grass carp (162288 ng/mL) than in chemically sedated grass carp (73166 ng/mL). Although a noticeable peak in blood glucose concentration was not observed in either study, a slight increase in concentration between 2 and 6 h postsedation was detected. Overall, glucose levels were slightly higher in the electrosedation study (73124 mg/dL) than in the current study (6198 mg/dL) while hematocrit (current study: 2029%; electrosedation study: 2231%) and osmolality (current study: 278361 mOsm/kg; electrosedation study: 301375 mOsm/kg) did not appear to vary greatly from resting levels in either study. In conclusion, each of the sedatives was effective in sedating grass carp with no apparent negative effects on selected blood chemistry characteristics or survival. Use of CO2 as a sedative option for grass carp at the concentration used in this study would not likely be recommended owing to the lengthy time required to reach sedation and the difculty of maintaining steady and effective CO2 concentrations in water. Although benzocaine rapidly induced sedation, the lengthy recovery time should be
566
GAUSE ET AL. Olsen, Y. A., I. E. Einarsdottir, and K. J. Nilssen. 1995. Metomidate anaesthesia in Atlantic salmon, Salmo salar, prevents plasma cortisol increase during stress. Aquaculture 134:155168. Owen, M. A. G., S. J. Davies, and K. A. Sloman. 2009. Light colour inuences the behaviour and stress physiology of captive tench (Tinca tinca). Reviews in Fish Biology and Fisheries 20:375380. Pirhonen, J., and C. B. Schreck. 2003. Effects of anaesthesia with MS-222, clove oil and CO2 on feed intake and plasma cortisol in steelhead trout (Oncorhynchus mykiss). Aquaculture 220:507514. Post, G. 1979. Carbonic acid anesthesia for aquatic organisms. Progressive Fish-Culturist 41:142144. Sattari, A., S. Mirzargar, A. Abrishamifar, R. Lourakzadegan, A. Bahonar, H. E. Mousavi, and A. Niasari. 2009. Comparison of electroanesthesia with chemical anesthesia (MS222 and clove oil) in rainbow trout (Oncorhynchus mykiss) using plasma cortisol and glucose responses as physiological stress indicators. Asian Journal of Animal and Veterinary Advances 4:306313. Sepici-Dinc el, A., A. C a glan Karasun Benli, M. Selvi, R. Sarikaya, D. S ahin, I. Ayhan Ozkul, and F. Erkoc . 2009. Sublethal cyuthrin toxicity to carp (Cyprinus carpio L.) ngerlings: biochemical, hematological, histopathological alterations. Ecotoxicology and Environmental Safety 72:14331439. Summerfelt, R. C., and L. S. Smith. 1990. Anesthesia, surgery, and related techniques. Pages 213272 in C. B. Schreck and P. B. Moyle, editors, Methods for sh biology. American Fisheries Society, Bethesda, Maryland. Trushenski, J. T., J. D. Bowker, B. R. Gause, and B. L. Mulligan. 2012. Chemical and electrical approaches to sedation of hybrid striped bass: induction, recovery, and physiological responses to sedation. Transactions of the American Fisheries Society 141:455467. Trushenski, J. T., M. Schwarz, R. Takeuchi, B. Delbos, and L. A. Sampaio. 2010. Physiological responses of cobia Rachycentron canadum following exposure to low water and air exposure stress challenges. Aquaculture 307:173177. Tuncer, H., R. M. Harrell, and E. D. Houde. 1990. Comparative energetics of striped bass (Morone saxatilis) and hybrid (M. saxatilis M. chrysops) juveniles. Aquaculture 86:387400. USFDA (Food and Drug Administration). 2011. Enforcement priorities for drug use in aquaculture. USFDA, Center for Veterinary Medicine, Program Policy and Procedures Manual 1240.4200, Silver Spring, Maryland. Available: http://www.fda.gov/downloads/AnimalVeterinary/GuidanceCompliance Enforcement/PoliciesProceduresManual/UCM046931.pdf. (June 2012). Venn Beecham, R., B. C. Small, and C. D. Minchew. 2006. Using portable lactate and glucose meters for catsh research: acceptable alternatives to established laboratory methods? North American Journal of Aquaculture 68:291295. Wattendorf, R. J. 1986. Rapid identication of triploid grass carp with a coulter counter and channelizer. Progressive Fish-Culturist 48:125132. Wells, R. M. G., and N. W. Pankhurst. 1999. Evaluation of simple instruments for the measurement of blood glucose and lactate, and plasma protein as stress indicators in sh. Journal of the World Aquaculture Society 30:276284. Woods, L. C., D. D. Theisen, and S. He. 2008. Efcacy of Aqui-S as an anesthetic for market-sized striped bass. North American Journal of Aquaculture 70:219222. Zahl, I. H., A. Kiessling, O. B. Samuelsen, and R. E. Olsen. 2010. Anesthesia induces stress in Atlantic salmon (Salmo salar), Atlantic cod (Gadus morhua), and Atlantic halibut (Hippoglossus hippoglossus). Fish Physiology and Biochemistry 36:719730. Zajicek, P., A. E. Goodwin, and T. Weier. 2011. Triploid grass carp: triploid induction, sterility, reversion, and certication. North American Journal of Fisheries Management 31:614618.
taken into consideration. Eugenol and MS-222 would probably be considered the most effective chemical sedatives (at the doses tested) owing to relatively rapid induction and recovery times. Lastly, since sedating sh can also act as a stressor, consideration should be given to reduce other potential stressors when conducting ploidy testing in grass carp. ACKNOWLEDGMENTS We thank Mike Freeze, Mike Clark, and the staff of Keo Fish Farm for their assistance and for space, sh, and water for the conduct of our experiment. REFERENCES
Barton, B. A. 2002. Stress in shes: a diversity of responses with particular reference to changes in circulating corticosteroids. Integrative and Comparative Biology 42:517525. Benfey, T. J. 1999. The physiology and behavior of triploid shes. Reviews in Fisheries Science 7:3967. Bowzer, J. C., J. T. Trushenski, B. R. Gause, and J. D. Bowker. 2012. Efcacy and physiological responses of grass carp to different sedation techniques: II. Effect of pulsed DC electricity voltage and exposure time on sedation and blood chemistry. North American Journal of Aquaculture 74:567574. Brougher, D. S., L. W. Douglass, and J. H. Soares Jr. 2005. Comparative oxygen consumption and metabolism of striped bass Morone saxatilis and its hybrid M. chryshops M. saxatilis. Journal of the World Aquaculture Society 36:521529. Davis, K. B., and B. R. Grifn. 2004. Physiological responses of hybrid striped bass under sedation by several anesthetics. Aquaculture 233:531548. Delaney, M. A., P. H. Klesius, and R. A. Shelby. 2005. Cortisol response of Nile tilapia, Oreochromis niloticus (L.), to temperature changes. Journal of Applied Aquaculture 16:95104. Fu, S.-J., L.-Q. Zeng, X-.M. Li, X. Pang, S.-D. Cao, J.-L. Peng, and Y.-X. Wang. 2009. The behavioural, digestive, and metabolic characteristics of shes with different foraging strategies. Journal of Experimental Biology 212:22962302. Gilderhus, P. A., and L. L. Marking. 1987. Comparative efcacy of 16 anesthetic chemicals on rainbow trout. North American Journal of Fisheries Management 7:288292. Itazawa, Y., and T. Takeda. 1978. Gas exchange in the carp gills in normoxic and hypoxic conditions. Respiration Physiology 35:263269. Kelly, A. M., C. R. Engle, M. L. Armstrong, M. Freeze, and A. J. Mitchell. 2011. History of introductions and governmental involvement in promoting the use of grass, silver, and bighead carps. Pages 163174 in D. C. Chapman and M. H. Hoff, editors. Invasive Asian carps in North America. American Fisheries Society, Symposium 74, Bethesda, Maryland. Kiessling, A., D. Johansson, I. H. Zahl, and O. B. Samuelsen. 2009. Pharmacokinetics, plasma cortisol, and effectiveness of benzocaine, MS-222 and isoeugenol measured in individual dorsal aorta-cannulated Atlantic salmon (Salmo salar) following bath administration. Aquaculture 286:301308. Masser, M. P. 2002. Using grass carp in aquaculture and private impoundments. Southern Regional Aquaculture Center, Publication 3600, Stoneville, Mississippi.

Efficacy and Physiological Responses of Grass Carp to Different Sedation Techniques: II. Effect of Pulsed DC Electricity Voltage and Exposure Time on Sedation and Blood Chemistry
John C. Bowzer , Jesse T. Trushenski , Brian R. Gause & James D. Bowker
a a a a b
Fisheries and Illinois Aquaculture Center, Southern Illinois University Carbondale, 1125 Lincoln Drive, Life Science II, Room 173, Carbondale, Illinois, 62901-6511, USA
b
U.S. Fish and Wildlife Service, Aquatic Animal Drug Approval Partnership Program, 4050 Bridger Canyon Road, Bozeman, Montana, 59715, USA Version of record first published: 21 Sep 2012.
To cite this article: John C. Bowzer, Jesse T. Trushenski, Brian R. Gause & James D. Bowker (2012): Efficacy and Physiological Responses of Grass Carp to Different Sedation Techniques: II. Effect of Pulsed DC Electricity Voltage and Exposure Time on Sedation and Blood Chemistry, North American Journal of Aquaculture, 74:4, 567-574 To link to this article: http://dx.doi.org/10.1080/15222055.2012.690830
TECHNICAL NOTE
Efcacy and Physiological Responses of Grass Carp to Different Sedation Techniques: II. Effect of Pulsed DC Electricity Voltage and Exposure Time on Sedation and Blood Chemistry
John C. Bowzer, Jesse T. Trushenski,* and Brian R. Gause
Fisheries and Illinois Aquaculture Center, Southern Illinois University Carbondale, 1125 Lincoln Drive, Life Science II, Room 173, Carbondale, Illinois 62901-6511, USA
James D. Bowker
U.S. Fish and Wildlife Service, Aquatic Animal Drug Approval Partnership Program, 4050 Bridger Canyon Road, Bozeman, Montana 59715, USA
Abstract
Owing to the current absence of an approved immediaterelease chemical sedative for use on sh, researchers have been exploring alternative methods that would allow treated sh to be released immediately after sedation, including the use of electrosedation. To address the efcacy of this approach, we evaluated induction and recovery times, survival, and postsedation hematology of grass carp Ctenopharyngodon idella (291 6.7 g, 30.6 0.3 cm TL, mean SE) sedated by exposure to 100, 150, or 200 V of pulsed DC (30 Hz and 25% duty cycle) for 5 or 10 s. Regardless of voltage strength or exposure time, all sh were sedated to Stage IV sedation within 0.75 min and recovered within 1.5 min. Although recovery times for sh exposed to electrosedation for 10 s were longer than those for sh electrosedated for 5 s using 100 and 150 V, the opposite trend was observed among sh sedated using 200 V. Overall, induction and recovery times were short: total time elapsed from induction to full recovery ranged from 1.0 to 2.1 min (mean, 1.6 min). No mortalities were observed 24 h postsedation. Hematological changes observed were consistent with an acute stress response, but these effects were transient and few differences were observed among the electrosedation protocols used. Our results indicate that pulsed DC electrosedation is an effective strategy for quickly and easily sedating grass carp.
Grass carp Ctenopharyngodon idella are an herbivorous species introduced from China to the United States as a biological control for aquatic vegetation in aquaculture ponds and
*Corresponding author: saluski@siu.edu Received October 24, 2011; accepted April 13, 2012
other private and public waters (Masser 2002). As grass carp are a nonnative species, there is concern that these sh may reproduce and establish selfsustaining populations in U.S. waters. This has led ofcials from many states to ban stocking fertile (diploid) individuals (Kelly et al. 2011), whereas triploid grass carp may be allowed in some cases (Masser 2002). Triploidy induction results in functional sterility (Benfey 1999; Zajicek et al. 2011) and is recognized as an effective strategy to control undesired reproduction. A rapid detection method to verify triploidy was developed by Wattendorf (1986), and this simple blood test is used to verify the chromosome number of every grass carp prior to sale and transfer to states with diploid grass carp bans (Zajicek et al. 2011). To facilitate testing, sh are typically sedated so that they can be easily handled during blood sampling. Currently, there are very few practical and effective chemical sedative options available to sh culturists to facilitate sample collection, and none of the sedative products available in the United States are legal (i.e., approved by the U.S. Food and Drug Administration [FDA]) for use on food sh (including sh that may be consumed after stocking in U.S. waters) without adhering to a lengthy withdrawal period (321 d) following exposure (see Gause et al. 2012, this issue). Although tricaine methanesulfonate (MS-222), benzocaine, eugenol, or carbon dioxide (CO2 ) may be used as sh sedatives under certain circumstances, none of these are fully suitable for procedures such as blood sampling and triploidy verication of grass carp because they
567
568
BOWZER ET AL.
require lengthy withdrawal times, are difcult to use, or are not legally available for such use. Given the constraints associated with chemical sedatives and the amount of time and resources required to gain FDA-approval of these compounds for use in sh, sh culturists have been exploring alternative sedation techniques, including electrosedation. Although electroshing has been used as a eld sampling technique for decades by sheries professionals, the technique has recently been modied specically for the purpose of sedating sh (Zydlewski et al. 2008; Trushenski et al. 2012). Electroanesthesia, or more accurately, electrosedation, can immobilize sh via electronarcosis (stunning) or electrotetany (tetanic muscle contraction) caused by electrically induced interference with neurotransmission. Although present electrosedation technology may be somewhat limited in sh with comparatively fragile vertebrae (e.g., salmonids; C. V. Burger, Smith-Root, personal communication), in many species it may offer several advantages over chemical sedatives in terms of withdrawal periods, chemical disposal, and potentially, ease of use. Perhaps more importantly, electrosedation of sh is currently not subject to FDA regulation and can be used legally without having to go through the arduous, multiyear, multimillion dollar process of getting a chemical sedative approved for use in sh. However, it is important to develop use protocols to ensure that sh can be effectively sedated with minimal risk of adverse postsedation outcomes, including mortality. Although preliminary experimentation suggested pulsed-DC electricity is suitable for sedating grass carp (described below), it is unclear whether different waveforms (e.g., different voltage strengths, frequencies) or exposure durations affect induction or recovery times, blood chemistry responses, or overall efcacy. Accordingly, we evaluated electrosedation effects (induction and recovery times, survival, and postsedation blood chemistry) on grass carp (a representative warmwater sh) using three different voltage strengths and two different exposure durations.
approximately 8 cm and equipped with an electrosedation unit (PES Portable Electroanesthesia System, Smith-Root, Vancouver, Washington). These sh were exposed to 100 V of pulsed DC (60 Hz, 25% duty cycle, 5-s exposure). Using this waveform, mean sedation and recovery times were 0.5 and 1.8 min, respectively, and no overt signs of postsedation distress were observed. Accordingly, we determined that this waveform was effective and would serve as the lowest intensity waveform in the subsequent, principal investigation with one modication: given that higher voltages and exposure durations were to be investigated, we decided that the experimental protocols would use a lower frequency of 30 Hz. For the principal investigation, groups of 15 sh were randomly collected from the reference population and transferred into the electrosedation unit congured as described above and lled with fresh culture water from the holding system. Fish groups were exposed to 100, 150, or 200 V of pulsed DC (30 Hz and 25% duty cycle) for 5 or 10 s in a 3 2 factorial design (100 V for 5 s, 100 V for 10 s, 150 V for 5 s, 150 V for 10 s, 200 V for 5 s, 200 V for 10 s). Culture water in the sedation chamber was aerated after sedating each group of sh but was not exchanged over the course of the experiment. A water sample was collected from the holding system at time (t) = 0, and the sample was analyzed in duplicate for the following: temperature and dissolved oxygen (YSI 550 meter, Yellow Springs Instruments, Yellow Springs, Ohio), conductivity, pH, salinity (Multi-Parameter PCSTestr 35, Eutech/Oakton Instruments, Vernon Hills, Illinois), hardness, alkalinity (digital titrator and reagents, Hach, Loveland, Colorado), total ammonia nitrogen, nitrite-nitrogen, and nitrate-nitrogen (DR 2800 spectrophotometer and reagents, Hach). All measured water quality characteristics were within ranges appropriate for grass carp (Masser 2002) (Table 1). Fish were monitored during sedation to determine induction to Stage IV of anesthesia (Summerfelt and Smith 1990; although anesthesia is the term used by Summerfelt and Smith, we have used the term sedation throughout the manuscript to better reect the behavioral responses we observed), which is
TABLE 1. Holding system water quality measured at the beginning of the experiment to examine electrosedation in grass carp.
METHODS Electrosedation procedures.A reference population of triploid grass carp (291 6.7 g, 30.6 0.3 cm total length, mean SE) was held at Keo Fish Farm, Keo, Arkansas, in an outdoor raceway congured as a partial ow-through system (static raceway, periodically ushed with screened surface water) with supplemental aeration. Prior to experimentation, sh were fasted for a minimum of 24 h. To determine an appropriate control waveform from which to derive other waveforms for experimentation in the principal investigation, a preliminary experiment was conducted to determine whether a relatively mild waveform (based on our previous experience with electrosedation) would effectively sedate grass carp to Stage IV of sedation (see below). A group of 15 sh were randomly collected from the reference population and transferred into a 142-L cooler prelled with 70 L of aerated culture water to achieve a depth of
Characteristic Temperature ( C) Dissolved oxygen (mg/L) Total ammonia nitrogen (mg/L) Nitrite-nitrogen (mg/L) Nitrate-nitrogen (mg/L) Alkalinity (mg/L) Hardness (mg/L) Salinity () Conductivity (S/cm) pH
Value 18.6 7.98 0 0.003 0.9 240 374 0.427 877 7.5
TECHNICAL NOTE
569
associated with the total loss of equilibrium, muscle tone, and responsiveness to visual and tactile stimuli, but maintenance of a slow, steady, opercular ventilation rate. After the loss of equilibrium, slight manual pressure was applied along the trunk and caudal peduncle as a tactile stimulus. Fish were considered induced to Stage IV when they no longer responded to this stimulus, but the ventilation rate remained steady albeit reduced relative to unsedated sh. A tremor was observed immediately following electrosedation. Although sh were not responsive during this tremor (and were perhaps momentarily in Stage V or VI of sedation), induction was considered complete after the tremor had ceased and sh resumed ventilating. Immediately after induction, blood samples were collected from three sh using procedures described below. The remaining 12 sh were returned to holding tanks and monitored to determine recovery from sedation (return of normal equilibrium and tactile responsiveness). The group was considered recovered when the last sh recovered, (i.e., recovery time = time for last sh to recover). Assessment of induction and recovery can be somewhat subjective so bias was minimized by having the same observer apply all stimuli and assess when sh were sedated and recovered. Recovered sh were transferred to a second raceway (congured in the same manner as the one housing the reference population) and kept separated by raceway dividers positioned at 1-m intervals along the length of the recovery raceway. Sample collection and analysis.In addition to collecting blood from sh sampled immediately after sedation (t = 0), blood samples were collected from three sh per group at 0.5, 1, 2, and 6 h postsedation (three sh per group per time point, individuals sampled once). Prior to sampling, all sh were immersed in a bath of metomidate hydrochloride (Aquacalm, Western Chemical, Ferndale, Washington; 510 mg/L for 30 s) to facilitate handling. Although additional sedation was not necessary for sh sampled at t = 0, these sh were also exposed to a metomidate hydrochloride bath to ensure consistent treatment of all sh. Once handleable, sh were weighed (to the nearest gram) and measured (total length to the nearest 0.5 cm), and a blood sample was collected from the caudal vasculature using heparinized, evacuated blood collection assemblies (Vacutainer, Becton Dickinson, Franklin Lakes, New Jersey). Although metomidate hydrochloride was selected, in part, because it limits or prevents corticosteroid increase during sampling (Olsen et al. 1995; Davis and Grifn 2004), all blood samples were collected within 5 min of capture to minimize the possibility of other confounding responses of handling and venipuncture. In addition to sh sampled at set time points after sedation, two sh from the reference population were also sampled every hour over the course of the experiment (experiment duration, 7 h; no sh were sampled more than once). After blood collection, all sh were returned to a separate area in the recovery system and monitored for adverse behavior and survival for 24 h. Blood samples were kept on wet ice (<36 h) until analysis for glucose, lactate, cortisol, and osmolality as described
by Gause et al. (2012). Although 36 h might be considered a lengthy period of time to hold blood samples prior to analysis, some assays could not be immediately conducted in the eld and samples had to be transported back to the Fisheries and Illinois Aquaculture Center, Carbondale, Illinois. It is possible that levels of metabolically relevant molecules (e.g., glucose and lactate) could have changed slightly during this holding period; however, all samples were treated in the same manner to ensure validity of comparisons among treatments. Briey, hematocrit (Statspin centrifuge, Fisher Scientic, Pittsburgh, Pennsylvania) and glucose (Freestyle Freedom Lite glucose meter, Abbott Laboratories, Abbott Park, Illinois) were determined using aliquots of whole blood, and then the remaining whole blood was centrifuged (3,000 g, 4 C, 45 min). Resultant plasma was collected and stored at 80 C until further analysis. Plasma samples were analyzed to determine lactate (Accutrend lactate meter, Roche, Mannheim, Germany), osmolality (Vapro 5520, Wescor, Logan, Utah), and cortisol (EIA 1887, DRG International, Mountainside, New Jersey). Although portable lactate and glucose meters such as those used in this study can slightly underestimate metabolite levels in sh blood relative to laboratory methods, they are considered precise and reliable for use in generating comparative data (Wells and Pankhurst 1999; Venn Beecham et al. 2006). Although multiple sh were sampled from each treatment group at each time point, they were group-sedated and cohoused after sedation. Therefore, we determined that individual sh did not represent truly independent observations. Since the experiment lacked replicate experimental units, qualitative comparisons of within-group mean values were assessed rather than using statistical analysis to try to make quantitative comparisons.
RESULTS All sh were successfully induced to Stage IV sedation in less than 1 min (mean = 0.6 min, range = 0.50.7 min), regardless of voltage strength or exposure duration (Figure 1). Recovery times were more variable, but recovery of equilibrium (mean = 0.7 min; range, 0.31.3 min) and tactile responsiveness (mean = 0.9 min; range, 0.51.4 min) were achieved in less than 2 min postsedation. Although a positive relationship was evident between longer exposure durations and increasing recovery times in sh sedated using 100 and 150 V, recovery times were shorter among sh sedated using 200 V. Overall, induction and recovery times were short, total time elapsed from induction to full recovery ranged from 1.0 to 2.1 min (mean = 1.6 min), there was minimal group-to-group induction or recovery variability, and no mortalities were observed 24 h postsedation. Although most blood chemistry parameters did not vary substantially by electrosedation protocol at any single timepoint, hematocrit, blood glucose, and plasma cortisol, lactate, and osmolality varied over time following sedation (Figure 2AE). Plasma cortisol and lactate concentrations initially increased
570
BOWZER ET AL.
100 V for 5 s 100 V for 10 s
150 V for 5 s 150 V for 10 s
200 V for 5 s
200 V for 10 s
1.5 2 2.5
0.5
Time (min)
I = Induced to Phase IV Sedation E = Maintain Equilibrium R = Recovery; Respond to Tactile Stimulus
FIGURE 1. Schematic illustrating induction and various stages of recovery of grass carp electrosedated to Stage IV sedation using various pulsed DC voltage strengths and exposure durations.
after sedation and then decreased over time. The cortisol response was rapid and transient, peaking (162288 ng/mL) at 0.5 h postsedation and returning to resting levels (0100 ng/mL) between 2 and 6 h postsedation. The peak lactate response developed more slowly than cortisol, reaching maximum levels (69 mmol/L) between 0.5 and 2 h postsedation, but dropped below resting levels by 6 h postsedation. Peak levels of blood glucose (98124 mg/dL) observed within the rst 0.5 h postsedation, decreased slightly over the next 0.5 h and then increased slightly from 1 to 6 h postsedation. Hematocrit and plasma osmolality uctuated near resting levels throughout the sampling period. There was no indication that any one combination of voltage strength and exposure duration consistently produced the highest or lowest blood chemistry responses. During electrosedation, sh exhibited opercular aring, n extension, and body rigidity but regained normal posture after resolution of the postsedation tremor. Slight petechial hemorrhaging was observed along the lower ank and opercular area in a few sh during electrosedation. DISCUSSION Pulsed DC, applied at voltage strengths of 100200 V and exposure durations of 5 or 10 s, was effective in sedating grass carp
to Stage IV sedation. Voltage strength and exposure duration had little effect in terms of time to induction and recovery from sedation. In addition, there was little variability among electrosedation protocols on the effects on blood chemistry responses following sedation. Although slight numeric differences were noted for induction and recovery times, the maximum difference observed (0.23 and 1.12 min, respectively) would probably not be considered practically relevant to most sheries professionals. All sh recovered within 2 min of induction, which would probably be considered adequate for sedating grass carp to facilitate procedures such as triploidy verication. The observed changes in blood chemistry were consistent with an acute stress response (Barton 2002), but these physiological responses were resolved or nearly resolved within 6 h of sedation, and no postsedation mortality was observed. Slight epidermal hemorrhaging was observed in some electrosedated sh. However, we concluded that these relatively uncommon lesions were probably unrelated to electrosedation since petechiae were also observed in some sh prior to electrosedation. Based on these data, it would appear that juvenile grass carp are resilient to electrosedation at the range of voltage strengths and exposure durations tested in this experiment, and that the electrosedation protocols tested are reasonably safe with respect to postsedation survival and physiological status of these sh. In the present work, we
TECHNICAL NOTE
571
FIGURE 2. Time course of blood chemistry responses (A = plasma cortisol, B = blood glucose, C = hematocrit, D = plasma lactate, E = plasma osmolality) of grass carp following electrosedation using various pulsed DC voltage strengths and exposure durations. Points represent mean values; grey reference bars represent means of values observed for sh sampled from the reference population throughout the course of the experiment.
572
BOWZER ET AL.
did not examine sh for vertebral or other internal injuries following electrosedation. These types of injuries have been observed following exposure to pulsed DC electrosedation in some (Gaikowski et al. 2001; Zydlewski et al. 2008) but not all shes (Vandergoot et al. 2011). Electrically induced injury and mortality rates are a function of the type and strength of the waveform used, as well as the sh involved (Snyder 2003). In general, short duration exposure to low-intensity, pulsed DC waveforms is considered less risky than longer duration exposure to high intensity, AC waveforms, though the reported effects of these and other factors are sometimes sparse, difcult to compare, and often questionable (Snyder 2003). Regardless, the absence of direct or delayed mortality and overt signs of injury suggest that each of the protocols assessed in the present work are reasonably safe when used to sedate grass carp. Blood chemistry responses observed in this experiment were comparable with those reported in two experiments conducted on hybrid striped bass (white bass Morone chrysops striped bass M. saxatilis) using the same Portable Electroanesthesia System and in the experiment conducted by Gause et al. (2012) in which grass carp were sedated using a variety of chemical sedatives. Trushenski et al. (2012) reported a slightly greater plasma cortisol pulse and greater plasma glucose and lactate pulses, but similar hematocrit and osmolality levels, in hybrid striped bass (510 12 g, 33.7 0.2 cm, mean SE) electrosedated at 100 V, 30 Hz, and 25% duty cycle for 3 s than we observed in grass carp in this experiment. In another experiment, Trushenski and Bowker (in press) used similar electrosedation protocols to sedate smaller hybrid striped bass (211 4 g, 26.1 0.1 cm total length, mean SE), and found that grass carp plasma cortisol levels were lower than or comparable with those observed in the smaller hybrid striped bass at t = 0, 1, 2, and 6 h. However, the plasma cortisol levels observed in grass carp were much lower (150300 ng/mL) at t = 0.5 h than observed in the smaller hybrid striped bass (400650 ng/mL) (Trushenski and Bowker, in press). Responses of lactate, hematocrit, and osmolality noted for grass carp were of a comparable or smaller magnitude than those reported for smaller hybrid striped bass by Trushenski and Bowker (in press), but followed the same basic patterns of acute response and resolution within 6 h of sedation. The somewhat attenuated lactate response observed in grass carp, in comparison with hybrid striped bass, is probably the result of the comparatively lower metabolic rate of grass carp (Tuncer et al. 1990; Brougher et al. 2005; Fu et al. 2009). Increased lactate formation results from anaerobic metabolic activity occurring during periods of limited or no oxygen availability, such as when environmental oxygen availability is limiting or during exhaustive physical activity when respiration is insufcient to meet tissue oxygen demand for aerobic metabolism (Bennett 1978; Burton and Heath 1980). Sedated sh exhibiting reduced ventilation rates may accumulate lactate, particularly if their metabolic rate and oxygen demand is high. Juvenile hybrid striped bass have a considerably higher oxygen demand at rest (132 mg O2 kg1h1) (Tuncer et al. 1990;
Brougher et al. 2005) than some other sh, including grass carp (56 mg O2 kg1h1; Fu et al. 2009); for purposes of comparison, Clarke and Johnston (1999) modeled the metabolic rate of 55 species of sh, and estimated the resting oxygen consumption of a 50-g sh at 15 C to range from 27 to 133 mg O2 kg1h1 depending on species. This may explain why electrosedated hybrid striped bass experience greater postsedation lactate pulses than do electrosedated grass carp. The blood glucose response of grass carp observed in this experiment uctuated near 100 mg/dL compared with the smaller hybrid striped bass in the previous experiment by Trushenski and Bowker (in press); in that study blood glucose displayed a pulse from resting levels of approximately 55 mg/dL to a peak at t = 1 h of 180220 mg/dL. The reduced metabolic rate and lower plasma cortisol levels of grass carp may explain the relatively minor glucose response observed in the current experiment. Cortisol can activate glycogenolysis and gluconeogenesis processes in sh, which cause increases in substrate levels (glucose) in the blood to produce enough energy to meet the demand of the organism (Barton and Iwama 1991; Mart nez-Porchas et al. 2009), and since grass carp have lower metabolic demands and experienced a lower cortisol response to electrosedation than hybrid striped bass (and possibly a lower catecholamine response), a lower glucose response may be expected. This and other research with chemical sedatives and various methods of electrosedation (AC, continuous DC, pulsed DC) or electroshock (Schreck et al. 1976; Mesa and Schreck 1989; Barton and Grosh 1996; Barton and Dwyer 1997) have demonstrated that sh undergo the generalized stress response (Barton 2002) following sedation (Bourne 1984; Bernier and Randall 1998; Davidson et al. 2000; Davis and Grifn 2004; Woods et al. 2008; Feng et al. 2009; Neiffer and Stamper 2009; Sattari et al. 2009; Carter et al. 2011; Trushenski et al. 2012; Gause et al. 2012). Because these effects also occur after exposure to sedatives in the absence of handling further emphasizes that the sedatives themselves act as stressors (Zahl et al. 2010). Differences in the magnitude of physiological responses aside, our present results are broadly consistent with our previous work sedating adult hybrid striped bass (Trushenski et al. 2012) and the majority of published works on the subject. In conclusion, pulsed DC electrosedation is an effective strategy for sedating grass carp quickly and easily for routine handling procedures. Electrosedation offers one distinct advantage over other currently available options: sh can be released immediately after treatment. Like other sedatives, electrosedation induces an acute stress response in sh. Although electrosedated grass carp exhibited responses consistent with the generalized stress response in sh, none of the protocols used elicited responses that were particularly severe in comparison with the others or the reported effects of chemical sedatives (Gause et al. 2012), and sh were observed to recover from these effects within 6 h. Although slight differences in induction and recovery times were associated with different voltage strengths and exposure durations, all of the protocols used yielded sedation
TECHNICAL NOTE
573
patterns that would be considered acceptable for handling grass carp and testing them for triploidy. However, to minimize the incidence of unforeseen injuries or physiological alterations, we recommend that users employ the lowest voltage strength and exposure duration that yields effective electrosedation. ACKNOWLEDGMENTS We thank Smith-Root, Inc. for providing access to a Portable Electroanesthesia System, and Jack Wingate and Mike Holliman for providing training and technical support in using the electrosedation unit. We also thank Mike Freeze, Mike Clark, and the staff of Keo Fish Farm for their assistance and for accommodating our experiment. REFERENCES
Barton, B. A. 2002. Stress in shes: a diversity of responses with particular reference to changes in circulating corticosteroids. Integrative and Comparative Biology 42:517525. Barton, B. A., and W. P. Dwyer. 1997. Physiological stress effects of continuousand pulsed-DC electroshock on juvenile bull trout. Journal of Fish Biology 51:9981008. Barton, B. A., and R. S. Grosh. 1996. Effect of AC electroshock on blood features in juvenile rainbow trout. Journal of Fish Biology 49:13301333. Barton, B. A., and G. K. Iwama. 1991. Physiological changes in sh from stress in aquaculture with emphasis on the response and effects of corticosteroids. Annual Review of Fish Diseases 1:326. Benfey, T. J. 1999. The physiology and behavior of triploid shes. Reviews in Fisheries Science 7:3967. Bennett, A. F. 1978. Activity metabolism of the lower vertebrates. Annual Review of Physiology 40:447469. Bernier, N. J., and D. J. Randall. 1998. Carbon dioxide anaesthesia in rainbow trout: effects of hypercapnic level and stress on induction and recovery from anaesthetic treatment. Journal of Fish Biology 52:621637. Bourne, P. K. 1984. The use of MS 222 (tricaine methanesulphonate) as an anaesthetic for routine blood sampling in three species of marine teleosts. Aquaculture 36:313321. Brougher, D. S., L. W. Douglass, and J. H. Soares Jr. 2005. Comparative oxygen consumption and metabolism of striped bass Morone saxatilis and its hybrid M. chrysops M. saxatilis. Journal of the World Aquaculture Society 36:521529. Burton, D. T., and A. G. Heath. 1980. Ambient oxygen tension (PO2 ) and transition to anaerobic metabolism in three species of freshwater sh. Canadian Journal of Fisheries and Aquatic Sciences 37:12161224. Carter, K. M., C. M. Woodley, and R. S. Brown. 2011. A review of tricaine methanesulfonate for anesthesia of sh. Reviews in Fish Biology and Fisheries 21:5159. Clarke, A., and N. M. Johnston. 1999. Scaling of metabolic rate with body mass and temperature in teleost sh. Journal of Animal Ecology 68:893905. Davidson, G. W., P. S. Davie, G. Young, and R. T. Fowler. 2000. Physiological responses of rainbow trout Oncorhynchus mykiss to crowding and anesthesia with AQUI-STM. Journal of the World Aquaculture Society 31:105114. Davis, K. B., and B. R. Grifn. 2004. Physiological responses of hybrid striped bass under sedation by several anesthetics. Aquaculture 233:531548. Feng, G. P., P. Zhuang, L. Z. Zhang, N. N. Chen, Z. F. Yao, and Y. Y. Men. 2009. Effects of electroanesthesia on behavior and serum iron concentration of juvenile Acipenser baeri. Marine Fisheries 2009(1):article 6. (In Chinese with English abstract.) Available: en.cnki.com.cn/Article en/CJFDTOTALHTYY200901005.htm. (October 2011). Fu, S. J., L. Q. Zeng, X. M. Li, X. Pang, Z. D. Cao, J. L. Peng, and Y. X. Wang. 2009. The behavioural, digestive and metabolic characteristics of shes with
different foraging strategies. Journal of Experimental Biology 212:2296 2302. Gaikowski, M. P., W. H. Gingerich, and S. Gutreuter. 2001. Short-duration electrical immobilization of lake trout. North American Journal of Fisheries Management 21:381392. Gause, B. R., J. T. Trushenski, J. C. Bowzer, and J. D. Bowker. 2012. Efcacy and physiological responses of grass carp to different sedation techniques: I. Effects of various chemicals on sedation and blood chemistry. North American Journal of Aquaculture 74:560566. Kelly, A. M., C. R. Engle, M. L. Armstrong, M. Freeze, and A. J. Mitchell. 2011. History of introductions and governmental involvement in promoting the use of grass, silver, and bighead carps. Pages 163174 in D. C. Chapman and M. H. Hoff, editors. Invasive Asian carps in North America. American Fisheries Society, Symposium 74, Bethesda, Maryland. Mart nez-Porchas, M., L. R. Mart nez-C ordova, and R. Ramos-Enriquez. 2009. Cortisol and glucose: reliable indicators of sh stress? Pan-American Journal of Aquatic Sciences 4:158178. Masser, M. P. 2002. Using grass carp in aquaculture and private impoundments. Southern Regional Aquaculture Center, Publication 3600, Stoneville, Mississippi. Mazeaud, M. M., F. Mazeaud, and E. M. Donaldson. 1977. Primary and secondary effects of stress in sh: some new data with a general review. Transactions of the American Fisheries Society 106:201212. Mesa, M. G., and C. B. Schreck. 1989. Electroshing markrecapture and depletion methodologies evoke behavioral and physiological changes in cutthroat trout. Transactions of the American Fisheries Society 118:644 658. Neiffer, D. L., and M. A. Stamper. 2009. Fish sedation, anesthesia, analgesia, and euthanasia: considerations, methods, and types of drugs. ILAR (Institute for Laboratory Animal Research) Journal 50:343360. Olsen, Y. A., I. E. Einarsdottir, and K. J. Nilssen. 1995. Metomidate anaesthesia in Atlantic salmon, Salmo salar, prevents plasma cortisol increase during stress. Aquaculture 134:155168. Sattari, A., S. S. Mirzargar, A. Abrishamifar, R. Lourakzadegan, A. Bahonar, H. E. Mousavi, and A. Niasari. 2009. Comparison of electroanesthesia with chemical anesthesia (MS222 and clove oil) in rainbow trout (Oncorhynchus mykiss) using plasma cortisol and glucose responses as physiological stress indicators. Asian Journal of Animal and Veterinary Advances 4: 306313. Schreck, C. B., R. A. Whaley, M. L. Bass, O. E. Maughan, and M. Solazzi. 1976. Physiological responses of rainbow trout (Salmo gairdneri) to electroshock. Journal of the Fisheries Research Board of Canada 33:7684. Snyder, D. E. 2003. Electroshing and its harmful effects on sh. U.S. Geological Survey, Information and Technology Report USGS/BRD/ITR20030002, U.S. Government Printing Ofce, Denver. Somero, G. N., and J. J. Childress. 1980. A violation of the metabolism-size scaling paradigm: activities of glycolytic enzymes in muscle increase in larger-size sh. Physiological Zo ology 53:322337. Trushenski, J. T., and J. D. Bowker. In press. Effect of voltage and exposure time on sh response to electrosedation. Journal of Fish and Wildlife Management. DOI: 10.3996/102011-JFWM-060. Trushenski, J. T., J. D. Bowker, B. R. Gause, and B. L. Mulligan. 2012. Chemical and electrical approaches to sedation of hybrid striped bass: induction, recovery, and physiological responses to sedation. Transactions of the American Fisheries Society 141:455467. Tuncer, H., R. M. Harrell, and E. D. Houde. 1990. Comparative energetics of striped bass (Morone saxatilis) and hybrid (M. saxatilis M. chrysops) juveniles. Aquaculture 86:387400. Vandergoot, C. S., K. J. Murchie, S. J. Cooke, J. M. Dettmers, R. A. Bergstedt, and D. G. Fielder. 2011. Evaluation of two forms of electroanesthesia and carbon dioxide for short-term anesthesia in walleye. North American Journal of Fisheries Management 31:914922. Venn Beecham, R., B. C. Small, and C. D. Minchew. 2006. Using portable lactate and glucose meters for catsh research: acceptable alternatives to
574
BOWZER ET AL. Zahl, I. H., A. Kiessling, O. B. Samuelsen, and R. E. Olsen. 2010. Anesthesia induces stress in Atlantic salmon (Salmo salar), Atlantic cod (Gadus morhua) and Atlantic halibut (Hippoglossus hippoglossus). Fish Physiology and Biochemistry 36:719730. Zajicek, P., A. E. Goodwin, and T. Weier. 2011. Triploid grass carp: triploid induction, sterility, reversion, and certication. North American Journal of Fisheries Management 31:614618. Zydlewski, G. B., W. Gale, J. Holmes, J. Johnson, T. Brigham, and W. Thorson. 2008. Use of electroshock for euthanizing and immobilizing adult spring Chinook salmon in a hatchery. North American Journal of Aquaculture 70:415 424.
established laboratory methods? North American Journal of Aquaculture 68:291295. Wattendorf, R. J. 1986. Rapid identication of triploid grass carp with a Coulter Counter and Channelyzer. Progressive Fish-Culturist 48:125132. Wells, R. M. G., and N. W. Pankhurst. 1999. Evaluation of simple instruments for the measurement of blood glucose and lactate, and plasma protein as stress indicators in sh. Journal of the World Aquaculture Society 30: 276284. Woods, L. C., III, D. D. Theisen, and S. He. 2008. Efcacy of Aqui-S as an anesthetic for market-sized striped bass. North American Journal of Aquaculture 70:219222.

Aquaculture Sep2012

Transféré par

Informations du document

Description originale:

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Aquaculture Sep2012

Transféré par

Droits d'auteur :

Formats disponibles

This article was downloaded by: [Department Of Fisheries] On: 25 September 2012, At: 23:59 Publisher: Taylor &

North American Journal of Aquaculture

Parental Male Effects on Landlocked Fall Chinook Salmon Progeny Survival

& Michael E. Barnes

Parental Male Effects on Landlocked Fall Chinook Salmon Progeny Survival

Downloaded by [Department Of Fisheries] at 23:59 25 September 2012

WIPF AND BARNES

Downloaded by [Department Of Fisheries] at 23:59 25 September 2012

1 721 3 2,965 129 731 28 3,880 469

2 713 18 3,229 340 755 11 4,166 272

3 696 13 3,111 146 725 23 3,795 361

4 678 16 2,803 218 713 13

Overall 702 8 3,027 108 731 10 3,946 202

Weights not obtained due to mechanical error.

MALE EFFECTS ON PROGENY SURVIVAL

Downloaded by [Department Of Fisheries] at 23:59 25 September 2012

WIPF AND BARNES

Downloaded by [Department Of Fisheries] at 23:59 25 September 2012

32.6 33.0 33.6 33.4

0.6 0.7 0.5 0.7

33.3 32.7 33.2 33.3

0.2 1.4 0.4 0.6

33.3 33.1 33.4 32.8

0.5 0.4 0.1 0.5

0.26 0.26 0.26 0.26

0.01 0.01 0.01 0.01

0.26 0.27 0.27 0.27

0.01 0.01 0.01 0.01

MALE EFFECTS ON PROGENY SURVIVAL TABLE 3. Continued.

9 9 10 11 12 Mean SE 13 0.25 0.25 0.25 0.25 0.25 0.01

Downloaded by [Department Of Fisheries] at 23:59 25 September 2012

0.26 0.25 0.25 0.26

0.02 0.02 0.02 0.02

Downloaded by [Department Of Fisheries] at 23:59 25 September 2012

North American Journal of Aquaculture

Version of record first published: 04 Sep 2012.

Downloaded by [Department Of Fisheries] at 00:00 26 September 2012

Downloaded by [Department Of Fisheries] at 00:00 26 September 2012

Downloaded by [Department Of Fisheries] at 00:00 26 September 2012

Downloaded by [Department Of Fisheries] at 00:00 26 September 2012

North American Journal of Aquaculture

Demonstration of Blue Crab Culture in Inland LowSalinity Waters of West Alabama

Demonstration of Blue Crab Culture in Inland Low-Salinity Waters of West Alabama

Downloaded by [Department Of Fisheries] at 00:01 26 September 2012

Ions K Mga Na Caa Na:K Salinity () Total hardnessa

Control 53.99 646.62 1,629.15 121.40 30.18 4.60 768.02

TC S3 69.52 52.03 778.98 47.07 11.20 5.80 99.10

TC S4 30.94 47.07 819.47 101.58 26.49 5.40 148.65

DO 1 73.03 201.67 2,459.07 205.63 33.67 2.10 407.30

DO 7 16.40 282.43 2,023.87 141.22 123.39 2.10 423.65

Crab pond 35.45 180.86 2,003.63 178.38 56.53 5.00 359.23

Expressed as mg/L calcium carbonate.

Downloaded by [Department Of Fisheries] at 00:01 26 September 2012

PSE = pooled SE.

Downloaded by [Department Of Fisheries] at 00:01 26 September 2012

pH 0.21 0.12 0.10 0.04 0.02

Downloaded by [Department Of Fisheries] at 00:01 26 September 2012

North American Journal of Aquaculture

Downloaded by [Department Of Fisheries] at 00:01 26 September 2012

LOCHMANN AND PHILLIPS

Downloaded by [Department Of Fisheries] at 00:01 26 September 2012

ALTERNATIVE DIETS FOR GOLDEN SHINERS