Available online at www.sciencedirect.

com

Theriogenology 74 (2010) 989 1001 www.theriojournal.com

Deep intra-uterine articial inseminations using cryopreserved spermatozoa in beluga (Delphinapterus leucas)
T.R. Robecka,*, K.J. Steinmana, G.A. Montanoa,b, E. Katsumatac, S. Osbornd, L. Daltond, J.L. Dunne, T. Schmittf, T. Reidarsonf, J.K. OBriena,g
a

SeaWorld and Busch Gardens Reproductive Research Center, SeaWorld Parks and Entertainment Corporation, San Diego, CA 92109, USA b Department of Animal Science, Texas A&M University, College Station, Texas 77840, USA c Kamogawa SeaWorld, Kamogawa, Chiba 296-0041, Japan d SeaWorld San Antonio, 10500 SeaWorld Drive, San Antonio, TX 78251, USA e Mystic Aquarium and Institute for Exploration, Mystic, CT 06355 USA f SeaWorld San Diego, 500 SeaWorld Drive, San Diego, CA 92109, USA g Faculty of Veterinary Science, University of Sydney, NSW 2006, Australia Received 20 November 2009; received in revised form 24 April 2010; accepted 25 April 2010

Abstract Articial insemination (AI) with liquid-stored spermatozoa and sperm cryopreservation using directional freezing (DF) have been successful in the beluga. This study built on this foundation to develop a deep intra-uterine AI technique with frozen-thawed semen in beluga. Forty-two ejaculates from one male were cryopreserved using DF technology and subsequently used for 10 insemination attempts with seven females. Percentage pre- and post-thaw progressive motility and viability were (mean SD) 73.0 12.2, 38.4 8.8, 88.0 0.1, and 59.3 15.7%, respectively. A series of GnRH injections (3 x 250 g, IV, 1.5 to 2 h apart) were used to induce ovulation, once a growing follicle 2.5 cm in diameter was visualized via trans-abdominal ultrasonography. Articial insemination was performed at 30.1 3.8 h post-initial GnRH injection with semen deposited in the uterine horn, 92.6 16.2 cm beyond the genital opening using a exible endoscope. The external cervical os (cEOS) was located beyond a series of 5 to 10 vaginal rings, 44.8 9.3 cm from the external genital opening. The internal bifurcation of the uterus was 27 6.8 cm beyond the cEOS. Ovulation occurred at 8.5 7.6 h post-AI. Two of 10 inseminations (20%) resulted in pregnancy. The rst pregnancy resulted in twins; both calves were born 442 d after AI, with one surviving. The second pregnancy is ongoing. These ndings represent the rst successful application of AI using frozen-thawed semen in beluga, and are important examples of how assisted reproductive technologies can provide tools for the global management of threatened species. 2010 Elsevier Inc. All rights reserved.
Keywords: Articial insemination; Conservation biology; Directional semen freezing; Genome resource banking; Ovulation induction

1. Introduction Conservation of wild populations requires a multidiscipline approach aimed at describing both the

* Corresponding author: Tel.: 1 619 225 3177; Fax: 1 210 225 3178. E-mail address: todd.robeck@seaworld.com (T.R. Robeck). 0093-691X/$ see front matter 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.theriogenology.2010.04.028

population biology of a species and the reproductive biology of their ex-situ representatives. Beluga are currently listed as near-threatened by the International Union for the Conservation of Nature (IUCN); however, the Cook Inlet sub-population is listed as critically endangered [1]. Ongoing environmental changes in the arctic due to global warming have already increased anthropogenic beluga habitat encroachment and may

990

T.R. Robeck et al. / Theriogenology 74 (2010) 989 1001

have adverse effects on food sources. Since these environmental changes are not likely to subside, and until further population data can be collected, the IUCN cautions that beluga are a conservation-dependent species [1]. Currently, 37 beluga are collectively managed by seven facilities within North America [2]. Although the rst successful zoologically born beluga occurred in 1981 [3], the fractured ex situ population has made establishment of sustainable beluga populations problematic. Whereas problems facing wild populations of beluga differ from those in aquaria, it is clear that without sound management practices and increased research efforts, ex situ populations are also threatened. Thus, collaborative multi-institutional management and research projects have been conducted to improve the reproductive success of those members of this species in human care, while concurrently improving our knowledge of the reproductive physiology of the species in general [37]. These efforts have helped to improve neonatal survival, and have identied inequalities in founder representation, with only half of the adult founders (10/18, 56%) having successfully reproduced. Efforts to increase genetic representation of all founder animals are required for long-term viability of the population. Once developed, assisted reproductive technologies (ART) provide management tools, including long-term semen cryopreservation/storage and AI, that aid in maintaining genetic diversication of ex situ populations. Although rapid advances in applying these technologies to other cetacean species have occurred [8 13], variations in beluga reproductive biology and difculties in storing spermatozoa have slowed progress in this species [14,15]. Steinman et al [16] demonstrated with endocrinological and ultrasonographic evidence that beluga were facultative induced ovulators and then developed an ovulation induction protocol using GnRH. Anatomical information of the beluga reproductive tract, important for developing an insemination methodology, has only been described from post-mortem samples [17]. The objective of this research was to develop and implement methodology for AI, utilizing frozenthawed spermatozoa in the beluga. Initial attempts to develop a genome resource bank with semen collected from wild animals harvested during native subsistence hunts was not successful due to the fragility of beluga spermatozoa [15,18]. Once regular access to semen collected voluntarily from a trained captive beluga was available, a short-term liquid storage method was developed, followed by the rst successful AI with liquid-

stored semen [14]. However, to effectively manage global beluga populations and develop a genome resource bank, methods for sperm cryopreservation were necessary. Through the use of a novel cryopreservation methodology, directional freezing, combined with a unique extender-cryoprotectant combination, a method which produced satisfactory post-thaw in vitro sperm characteristics, was developed [15]. Demonstrating the fertility of this frozen-thawed semen through the application of AI was the next logical step. 2. Materials and methods 2.1. Reagents and media All chemicals were of analytical grade and cell culture tested, where possible, by the manufacturer. Unless otherwise stated, all media components were purchased from Sigma-Aldrich (Sigma, St Louis, MO, USA) and were prepared with tissue culture grade water (Sigma or Millipore, Billerica, MA, USA). Diluents containing egg yolk (free range eggs) were prepared by ultracentrifugation for 1.5 h at 10,000 g. The supernatant was ltered (0.22 m) and frozen at 80 C for a maximum of 18 mo. 2.2. Animals The seven females used in the AI trials (aged 6 30 y) were housed at four institutions: Kamogawa Sea World (KSW: Kamogawa, Chiba, Japan); Mystic Aquarium and Institute for Exploration (MA-IFE: Mystic, CT, USA); SeaWorld San Antonio (SWSA, San Antonio, TX, USA); and SeaWorld San Diego (SWSD, San Diego, CA, USA; Table 1). The animals were housed in a variety of indoor (KSW) or outdoor pools (SWSA, SWSD, MA-IFE), ranging in size from 258 to 1741 m3. For all AI trials, semen was collected from a wild born, proven adult beluga (aged 21 y in 2005, weighing 820 kg) from Feb 2005 to April 2009 (Table 1). The male was housed at SWSA from February 2005 to January 2008, at which time he was moved to SWSD. The diet consisted of frozen-thawed whole sh (herring, Clupea harengus, and/or Columbia River smelt, Thaleichthys pacicus) fed at approximately 4 to 5% of body weight per day. 2.3. Endocrine monitoring Voluntary urine samples were collected from two of the seven female beluga, as previously described for bottlenose dolphins (Robeck et al 2005). Briey, the animals were trained to lie in dorsal recumbency in shallow water ( 1.5 m) with their ukes and peduncle

T.R. Robeck et al. / Theriogenology 74 (2010) 989 1001 Table 1 Description of belugas used and samples collected during the study. Animal 1 2 3 4 5 6 7 1
a

991

Facilitya KSW MA-IFE SWSA SWSA SWSA SWSD SWSD SWSA/ SWSD

Sex F F F F F F F M

Birth date 1987c 1981c July 24, 2000d 1985c July 25, 1999d 1984c 1979c 1984c

Body length (cm) 395 362 328 337 315 340 350 416

Weight (kg) 865 535 580 619 481 623 571 1004

Reproductive historyb Nulliparous Nulliparous 1 abortion 2 calves 1 calf Nulliparous 1 calf Sired 4 calves

No. AI Attempts 3 1 2 1 1 1 1

Samples collectede P, E2, Ult P, Ult EC, LH, P, Ult EC, LH, P, Ult P, Ult P, Ult P, Ult

b c d e

MA-IFE: Mystic Aquarium and Institute for Exploration; SWSD: SeaWorld San Diego; SWSA: SeaWorld San Antonio; KSW: Kamogawa Sea World. Reproductive history prior to the start of the articial insemination trials. Estimated age of wild caught animals was based on length at capture. Captive born. P: serum progesterone; E2: serum estradiol; Ult: ultrasound evaluations; EC: urinary estrone conjugates; LH: urinary LH.

held at the water surface by a trainer who was standing in the pool. A second trainer would apply rm, steady pressure on the abdomen, directly over the urinary bladder. The animals would urinate in response to the pressure and eventually became conditioned to urinate with only a slight touch in the same location. The urine was aspirated from the genital opening with a 20 mL syringe or caught midstream directly into a sterile collection cup (118 mL, Fisher Scientic, Pittsburg, PA, USA). Urine samples (n 281) were collected daily until 10 d after synchronization, then three times a day during the estimated peri-ovulatory period. Samples were stored in duplicate at 80 oC pending analysis. Non-extracted urine samples were analyzed for determination of total immunoreactive concentrations of estrogen conjugates (EC) by a single antibody (estrone glucuronide), direct enzyme immunoassay (EIA) as previously described in beluga and other cetacean species [8 10,16]. Urinary EC concentrations were indexed against creatinine (Cr) and expressed as ng/mg Cr. Creatinine concentrations were determined by methods previously described [16,19]. In addition, a rapid LH semi-quantitative canine LH (cLH) kit (Witness Synbiotics Corp. Kansas City, MO, USA), previously validated for cetaceans [10], was used to detect urinary or serum LH in response to the ovulation induction protocol. Urinary LH proles were obtained using a single antibody, direct EIA which allowed LH concentration determination within 2.5 h [10,12,16]. The EIA for LH was modied from the double antibody EIA previously developed [20]. Intra-assay variation

was 10% and inter-assay variation was 10.3 and 10.9% at 30 and 60% binding (n 24). Serial dilutions of beluga urine yielded displacement curves that were similar to the standard curve (R2 0.99). The mean recovery of LH added to a pool of beluga urine was 78.4 22.8% (y 0.67x 0.19, R2 0.99). In one of the four animals (Female 1) not trained for urine collection, serum samples were collected daily or every other day after detecting a large ( 2.0 cm in diameter) follicle by trans-abdominal ultrasonography. Serum was obtained after centrifugation at 1000 g for 15 min and stored as 1 mL aliquots at 80 C until analysis. Estradiol (E2) concentrations were determined in Female 1 by a commercial laboratory (Kameda Medical Center, Kamogawa, Chiba, Japan) using a commercially available E2 EIA kit (Cobias, Roche Diagnostics, Indianapolis, IN, USA). The assay had a sensitivity of 5 pg/mL and the E2 antibody major crossreactivity was to estrone (51.5%), 17 -estradiol-3,17 sulfate (41.1%), 17 -ethinyl-estradiol (30.9%), 17 estradiol-17-valerate (29.4%), estriol (7.7%), 2-methoxy-estradiol (5.4%), and 5% for all others tested. Samples for progesterone analysis were collected from all animals after each AI (n 7 animals; n 10 AIs) on Day 0 (Day of insemination), then weekly until Day 42 post-AI for pregnancy determination. Progesterone concentrations were determined using an immunoflourescence assay (IFA) at a commercial laboratory as previously validated for beluga ([35,7] SWSA, SWSD, MA: Quest Diagnostics, Irving, TX, USA; KSW: Kameda Medical Center, Kamogawa, Japan).

992

T.R. Robeck et al. / Theriogenology 74 (2010) 989 1001

2.4. Synchronization of estrus For estrus synchronization, animals were either given 0.044 mg/kg po of altrenogest (Regu-Mate, Intervet Inc., Millsboro, DE, USA) or 0.05 mg/kg po medroxyprogesterone acetate (Provera, Pzer Inc., Tokyo, Japan) once a day for 30 d. The drug was administered by either lling gel capsules then inserting or by directly injecting the hormone into the coelomic cavity of a herring just prior to feeding. 2.5. Ovarian ultrasonography Trans-abdominal ultrasonography was used to help dene the timing of ovulation post-GnRH induction during natural or following altrenogest-induced cycles, and to conrm pregnancy as previously described for bottlenose dolphins, Pacic white-sided dolphins, and killer whales [8 10,21]. Ultrasonographic examinations were performed using either a Sonosite Micromaxx (Kabushiki Kaisha Sonosite Japan, Tokyo, Japan), Aloka 900 (MA, KSW: Corimetrics Medical Systems, Tokyo, Japan), or a GE Logiqbook, (SWSA, SWSD: GE LogiqGE Medical Systems, Milwaukee, WI, USA) using a 3.5 MHz transducer (wide footprint convex linear transducer). Females were examined once on Days 0 and 10 post-altrenogest, then daily to thrice daily from Day 11 to detection of ovulation. Once a preovulatory follicle (POF) was observed, ovaries were visualized ultrasonographically at least once daily. Ovarian follicular diameter, follicular circumference, and the time and side of ovulation were determined as previously described [9]. 2.6. Semen collection The male was trained for voluntary semen collection over a 2 y interval using operant conditioning, as previously described for beluga [10,14,22]. Briey, the animal received a variety of tactile stimuli to elicit voluntary extrusion of the penis from the genital groove. Once a complete erection was obtained, the male was conditioned to ejaculate by manual stimulation of the penis with a gloved hand (Nitrisoft, Nitrile powder-free examination glove; Sintex, Houston, TX, USA), lubricated sparingly with Pre-seed lubricant (INGfertility, Valleyford, WA, USA). Once manual stimulation commenced and the male subjectively appeared in a pre-ejaculatory state, the distal end of the penis was wiped with sterile cotton gauze, or rinsed with a HEPES-TALP medium, to remove residual saltwater. The penis tip was then quickly directed into a modied articial vagina [14], and the ejaculate was

collected shortly after further manual stimulation of the penis into a WHIRL-PAK bag (2041 mL, NASCO, Fort Atkinson, WI, USA). 2.7. Evaluation of raw semen Raw ejaculates were held at room temperature (21 C) and processed immediately after collection. Ejaculates were diluted 1:1 (v:v) with Beltsville extender (BF5F [23]) that had been modied for liquid storage of beluga semen (52.3 mM TES, 16.5 mM Tris, 105.4 mM fructose, 105.4 mM glucose, 20% v:v egg yolk, and 50 ug/mL gentamicin sulfate; 360 5 mOsm/kg and pH 7.0 0.1 [14]). The BF5F extender was supplemented with hyaluronic acid (HA; sodium hyaluronate, Bioniche Animal Health, Bogart, GA, USA, nal concentration 1 mg/mL) to prevent sperm head-to-head agglutination [14]. Ejaculate concentration, volume, pH (pH indicator strips; EM Science, Gibbstown, NJ, USA), osmolality (Advanced Instruments Inc., Norwood, MA, USA), as well as viability (plasma membrane integrity), acrosomal status and morphology were determined as previously described for beluga [14]. Motility was evaluated objectively using computer assisted sperm analysis (CASA, HTM-IVOS Version 12.2, Hamilton-Thorne Biosciences, Beverly, MA, USA) as previously described for beluga [14,15]. Briey, ve randomly selected microscopic elds were scanned to determine average pathway velocity (VAP, m/s) and straightline velocity (VSL, m/s). The instrument settings were 30 frames at a frame rate of 60 frames/s, minimum contrast was 40, minimum cell size (pixels) was 5; for progressive cells the VAP was 15 m/s and STR (straightness of sperm movement) was 70%. Spermatozoa with a VAP 5 m/s were considered motile. Only ejaculates free of contamination with saltwater or urine (osmolality 365 mOsm/kg) were used. Viability for fresh and post-thaw samples was analyzed by mixing 10 L of sample with 10 L of a live-dead exclusion stain (eosin-nigrosin; IMV Technologies USA, Maple Grove, MN, USA) for 30 s. A smear was made and allowed to air dry for evaluation within 30 min (100 spermatozoa per sample, 1000x). Spermatozoa were classied as viable (no stain uptake) or nonviable (partial or complete stain uptake). 2.8. Semen cryopreservation Ejaculates (n 42) were cryopreserved as previously described [15]. Semen was diluted with BF5F HA, cooled to 5 C over 1.5 h ( 0.2 C/min) and

T.R. Robeck et al. / Theriogenology 74 (2010) 989 1001

993

resuspended in the cryodiluent BF5F 182 mM trehalose (96 mM, nal trehalose concentration) for a nal concentration of 60 to 100 x 106 spermatozoa/mL. Sperm samples were transferred to 9 mL hollow tubes (IMT International Ltd, Chester, UK) for cryopreservation using a directional solidication (directional freezing) machine (MTG-516, IMT International Ltd). The hollow tube was moved through the rst block (5 C) for 45 s at a constant velocity (3 mm/s) before reaching a distance of 2 mm into the opening of a second block ( 50 C) and held for 60 s for initiation of seeding (rapid induction of ice nucleation from the seeding point throughout the length of the glass tube). The tube was then moved at 3 mm/s across the second block for 5 min before entering the collection chamber ( 100 to 110 C), followed by immediate transfer to liquid nitrogen. Hollow tubes were thawed in air for 90 s then transferred to a 35 C water bath equipped with modications to enable uniform sample thawing over 50 s (Harmony CryoCare Activator, IMT International Ltd). An aliquot (100 L) of the sample was removed and diluted (1:1 over 5 min) with BF5F-HA (warmed to 35 oC) for post-thaw analysis of sperm characteristics, as described for raw ejaculates. Motility parameters of post-thaw ejaculates were evaluated in remote locations without access to a CASA machine and thus, were subjectively analyzed for progressive motility (PM) and kinetic rating (KR, 0 5 scale, 0 no forward movement, 5 rapid forward progressive movement). The remainder of the sample was stored at 21 oC until AI was performed. 2.9. Ovulation induction and AI A total of 10 AIs were performed with frozen-thawed semen in seven females from 2005 to 2009. Separate studies indicated that beluga were facultative induced ovulators, indicating that an exogenous source of GnRH must be administered to ensure ovulation [14,16]. Consequently, females with growing follicles that had reached a preovulatory size ( 2.5 cm in diameter) were given GnRH (Cystorelin, Merial, Duluth, GA, USA; 3 x 250 g iv, q. 1.5 to 2 h) to induce ovulation. The insemination was done 30 h after the rst GnRH injection. One to 2 h prior to each procedure, all females were pre-medicated with valium (Diazepam, Abbott Lab, Chicago, IL, USA; 0.1 to 0.2 mg/kg). The females were removed from the water and placed in lateral recumbency on multiple 10 cm thick closed cell foam pads. The females were placed on their side, kept wet during the procedure and vital signs were monitored throughout. According to previous anatomical studies, beluga

have a relatively long vagina (25 to 50 cm) that contains 10 to 15 circumferential vaginal folds or rings that become more pronounced as they approach the cervix [17]. The length of the vagina and its folds, and the depth of the external cervical os (cEOS), uterine bifurcation and semen placement, were all recorded. Semen was deposited using a 9 mm o.d., 200 cm long exible endoscope (Olympus America, Melville, NY, USA) equipped with a custom-made catheter in the working channel (2.7 mm o.d. [7 Fr], 350 cm long, Smiths Medical PM, Inc., Waukesha, WI, USA). 2.10. Pregnancy diagnosis Pregnancy diagnosis was accomplished on unrestrained animals by a combination of serum progesterone monitoring on a weekly basis post-AI, and trans-abdominal ultrasonography (GE Logiqbook) bimonthly from Day 21 post-AI. Fetal measurements, biparietal and cross-sectional thorax at the level of the heart were determined as described for bottlenose dolphins [24]. 2.11. Statistical analysis All statistical analyses were performed using SigmaStat (Version 3.5; SSPS, San Rafael, CA, USA). Students t-tests were used to compare progesterone concentrations post-AI between pregnant and nonpregnant animals. Data are expressed as means SD. 3. Results 3.1. Semen collection and cryopreservation Due to the relatively low ejaculate volume and sperm concentration, a total of 42 ejaculates were required for use during inseminations (Table 2). The mean ejaculate volume, sperm concentration and total spermatozoa were 1.6 0.9 mL, 361.0 128.9 x 106 spermatozoa/mL, and 594.4 362.2 x 106 spermatozoa, respectively. Overall, post-thaw samples cryopreserved in 91 mM trehalose using directional freezing retained 52.6, 95.1, and 67.4% of their raw progressive motility, kinetic rating and viability, respectively. 3.2. Estrus synchronization, ovulation induction and endocrinology In all but one AI attempt (n 9/10), estrus was synchronized using Regu-mate (n 7) or Provera (n 2). The animal (Female 6) which did not receive a progestagen was articially inseminated on a natural cycle after a growing follicle was detected by routine

994 Table 2 Characteristics (means End point Semen characteristics Volume (mL) Osmolality (mOsm/kg) Sperm concentration (x 107/mL) Total spermatozoa per ejaculate (x 107) Sperm characteristics Subjective total motility (%, n 26)c Objective total motility (%, n 16)d Subjective progressive motility (%)c Objective progressive motility (%)d Kinetic rating (0-5)e VAP ( m/s, n 16) VSL ( m/s, n 16) Viability (%)

T.R. Robeck et al. / Theriogenology 74 (2010) 989 1001

SD) of beluga semen from Male 1 used for articial insemination. Raw ejaculate (n 42) 1.6 364.4 36.1 59.4 83.8 88.0 73.0 71.0 4.1 85.8 104.0 88.0 0.9 18.7 12.8 36.2 9.0 8.8 12.2 8.8 0.6 16.6 17.2 0.1 Pre-freeze Post-thaw

6.8 3.8 8.9 4.1 78.8 82.3 65.8 66.5 3.9 78.7 102.7 12.1 13.9 12.9 15.6 0.7 19.6 19.6

15.3a 8.0 16.8 281.9 166.3b 33.9 40.5 38.4 3.9 10.1 8.8 0.2

59.3

15.7

VAP: average pathway velocity; VSL: straight-line velocity. a Final volume of insemination dose. b Total number of spermatozoa per insemination. c Estimated by examining four to ve elds of diluted ( 25 x 106/ml) spermatozoa at 35C. d Determined by computer assisted sperm analysis. e Kinetic rating of spermatozoa graded subjectively: 0 no movement, 5 rapid forward progression.

ultrasonographic examination. All animals placed on a progestagen responded with follicular development. For nine of the 10 inseminations, animals were given intravenous GnRH in an attempt to induce ovulation at 21.3 5.5 d post-progestagen withdrawal. Female 1 was synchronized four times; twice with Provera and twice with Regu-mate. For her rst AI attempt, she spontaneously ovulated 18 d after Provera withdrawal, before she could be articially inseminated. For her second AI attempt, ovulation was induced with GnRH at Day 12 following Provera withdrawal. For the remaining two attempts using Regumate, ovulation was induced with GnRH at 16 and 17 d post-Regu-mate, respectively. All four AIs with Female 1 were unsuccessful. Female 3 began her LH surge spontaneously 25 d post regumate and was inseminated 30 h post-LH surge initiation, but she did not conceive. In the nine (of 10) AI attempts when females were given GnRH, the test line on the cLH kit was negative (colorless) for serum LH immediately before the rst injection. The cLH kit test line became positive (color was similar or darker than the control line) just prior to the second GnRH injection 1.5 to 2 h later. Ovulation occurred in eight of the nine induction attempts. Although ovulation could be conrmed or denied via ultrasonography during all attempts within 24 h postAI, more precise timing of ovulation (to within a few h) was obtained in only four attempts within 12 h post-AI.

For those attempts, ovulation occurred 36.6 2.6 h (n 4) post-initial GnRH injection. The mean diameter and circumference of the largest follicle at the time of GnRH injection from the eight animals that ovulated post induction was 2.8 0.6 cm and 8.3 1.8 cm, respectively. In the one instance where ovulation failed to occur within 24 h post-GnRH administration (Female 4), the ovulation induction protocol was erroneously administered when her follicle was only 1.78 cm in diameter (5.49 cm in circumference). Trans-abdominal ultrasonographic images obtained before the GnRH injections were inaccurate and it was falsely believed that the POF was at least 2.5 cm in diameter; it was only during the ultrasnographic examination just prior to the AI that an accurate measurement obtained. Secondary follicles, 2.2 0.5 cm in diameter, were observed in 7 of 10 inseminations. Five of seven secondary follicles ovulated within 12 h of the primary follicles. The primary follicle was located on the left ovary 50% of the time, whereas the secondary follicles were located on the contra-lateral ovary to the primary follicle 57% of the time. In the three AI attempts with the two animals that could be trained for urine collection (Females 3 4), peak urinary EC values were 83.8 9.1 ng/mg cr (n 3; Fig. 1). Peak EC at time of GnRH injection (n 2), peak LH for natural (n 1) and induced cycles (n 1) were 76.3 21.3 ng/mg cr, 46.1 ng/mg cr and 183.3

T.R. Robeck et al. / Theriogenology 74 (2010) 989 1001

100 200 10

995

Estrone conjugates ng/mg cr

80 150 AI

60 100 40 GnRH 50 20

0 160 -24
10

0 -20 -16 -12 -8 -4 0 4 8 12 16 20 10

140
Progesterone ng/mL

8 P4 conceptive P4 non-conceptive 6

* *

AI 8

120

E2 pg/mL

100

6
0 -8 -4 0 4 8 12 16 20 24 28

80

Day post AI

GnRH 60

40

20 0 -24 -20 -16 -12 -8 -4 0 4 8 12 16 20

Day
Fig. 1. Hormone proles of two beluga during AI attempts after a successful (top graph; Female 3) and a non-successful (lower graph; Female 1) AI. Top graph shows urinary LH and EC (estrone conjugates) and serum progesterone proles. Note the occurrence of the LH surge in response to GnRH administration. Also note the continually elevated serum progesterone following the LH surge. The lower graph has serum estrogen (E2) and progesterone proles. Note that the serum progesterone post-AI did not remain elevated. Also note the composite serum progesterone proles (box in upper left of lower graph) comparing weekly progesterone post-AI in conceptive (n 3) and non-conceptive (n 8) AIs. Means from Days 14 and 21 were different (*; P 0.05) between groups (Day 28 was not tested due to lack of samples from conceptive groups).

ng/mg cr, respectively. Peak estrogens (urinary EC or serum E2) occurred 18 15.9 h (n 4) prior to GnRH injection. For the one animal with a natural cycle, peak EC (93.7 ng/mg cr) occurred 12 h prior to peak LH. For all animals injected with GnRH (n 9), the cLH kit test lines were negative at the time of the rst injection,

and strongly positive just prior to the second and third injections. Progesterone concentrations at the time of the rst GnRH injection and at the time of the AI were 0.34 0.18 ng/mL and 1.41 1.03 ng/mL, respectively (Fig. 1). For the attempt when the female (4) was erroneously injected with GnRH prior to devel-

Progesterone ng/mL

Progesterone ng/mL

LH ng/mg cr

996

T.R. Robeck et al. / Theriogenology 74 (2010) 989 1001

Fig. 2. Beluga hysteroscopic anatomical characteristics. A: Smooth longitudinal folds of the caudal vagina; B: Transition from longitudinal folds to circumferential folds (CF) of the mid-vagina; C: CF in the mid-vagina, continuing cranially; D: Distal vaginal CF with the external cervical os located centrally.

oping a POF of at least 2.5 cm, peak EC concentration was 24.3 ng/mg cr. Retrospectively, this lower value was compatible with the smaller POF size (1.78 cm in diameter). In Female 1, serum E2 concentrations peaked at 72.0 pg/mL prior to spontaneously ovulating a 2.9 cm diameter follicle in March 2008. These values were subsequently used to help determine the appropriate timing for three AI attempts. During these attempts, her mean POF diameter and peak E2 concentrations were 3.4 0.7 cm and 111.9 40.4 pg/mL, respectively (n 3; Fig. 1). 3.3. Articial insemination and pregnancy monitoring The number of progressively motile spermatozoa (PMS) that were used for the inseminations ranged from 285 x 106 to 3291 x 106. The two conceptions occurred after insemination of 525 x 106 spermatozoa (MED, minimum effective dose) and 3291 x 106 PMS. The rst two inseminations were below the MED (301 x 106 and 285 x 106 PMS, respectively), with the remaining AIs utilizing numbers which were 1.5- to 6.2-fold higher than the MED.

The rst insemination attempt (Female 7) occurred at an estimated 3 h after ovulation. When time of ovulation could be determined, the remaining inseminations occurred 8.5 7.6 h prior to ovulation. The distance from the genital opening to the cEOS, the internal uterine bifurcation, and nal semen deposition was 44.8 9.3, 69.1 13.2, and 92.6 16.2 cm, respectively. The prominent vaginal rings, estimated from ve to seven in total, presented an initial impediment, particularly when the vagina was not sufciently insufated. After becoming familiar with beluga vaginal and cervical anatomy through multiple endoscopic exams, subsequent attempts to reach the cervix became considerably easier (Fig. 2). After identifying the cervix, the scope was aligned with the transverse cervical folds to help identify the cEOS, and then passed into the uterus (Fig. 3). During AI procedures, the internal bifurcation of the uterus had to be identied by slowly retracting the endoscope until the opening of the second horn was seen. Using this method, the bifurcation was identied in eight of the last nine attempts. During the eighth AI attempt with Female 4, the endoscope could

T.R. Robeck et al. / Theriogenology 74 (2010) 989 1001

997

Fig. 3. Beluga hysteroscopic anatomical characteristics during deep intra-uterine AI. (A) Entry of the endoscope in the cervical lumen (white arrow). Note the longitudinal folds of the cervix. (B) Entry of the endoscope in the cervical lumen with the opening into the uterus in the upper right quadrant (white arrow head). (C) uterine body and the beginning of the uterine horn (white arrow). (D) Apex of the uterine horn, with the tip of the insemination catheter (white tube) at the presumptive utero-tubal junction.

only be passed just beyond the internal cervical Os and thus the uterine bifurcation was not identied. During the course of developing this insemination procedure, it was hypothesized that when the animal is held out of the water in lateral recumbency, gravitational effects on the body mass likely compress the downside uterine horn. Therefore, the horn with the least compression was the rst to be insufated and, most likely, was the rst one entered after passing beyond the cervix. With the exception of the rst AI procedure, the animal was placed in lateral recumbency, with the side ipsilateral to the ovary containing the POF positioned dorsally. Once the endoscope was passed as deep in the uterine horn as visually possible, the catheter was advanced until slight resistance was met (Fig. 3). At that point, semen was slowly introduced into the uterine horn, followed by 5-20 mL of BF5F to rinse residual spermatozoa from the catheter. The air source was discontinued once the insemination had begun, to minimize air accumulation in the uterus,

and to allow for air to be expelled during normal uterine contractions. After AI, semen was visualized to have pooled in the uterine horn, and the endoscope was slowly removed. Of the ten AI attempts, two resulted in pregnancy (Females 3 and 5). Frozen-thawed semen was deposited in one uterine horn for seven inseminations, and in both horns for the remaining three. Of the two animals that became pregnant, one had semen deposited in both horns (Female 5) and the other in one horn (Female 3). The total volume of uid (frozen-thawed semen and BF5F from ushing of the catheter) introduced into the uterus across all AIs was 45 18 mL. The time required to complete the inseminations (time from endoscope introduction to withdrawal) was 14 3 min. The weekly post-AI progesterone samples revealed a normal luteal rise in progesterone of 4.4 3.0 ng/mL by Day 14. The progesterone proles to Day 30 post-AI from three conceptive AIs (two from the present study using frozen-thawed semen and one from a previous

998

T.R. Robeck et al. / Theriogenology 74 (2010) 989 1001

Fig. 4. The rst successful beluga born from articial insemination with frozen-thawed semen (SeaWorld San Antonio, 2009). Photo Credit to White Whale and Dolphin Staff (SWSA).

study [14] using liquid-stored semen) were compared to eight non-conceptive AIs. Progesterone concentrations from non-pregnant animals diverged (P 0.05) from pregnant animals after Day 14 post-AI (Fig. 1). The rst pregnancy (Female 5) resulted in twin calves being born at 442 d post-AI. Only the larger of the twin calves was alive at birth and is currently growing normally (Fig. 4). For the second pregnancy (Female 3), fetal cross-sectional thorax and biparietal measurements were 5.3 cm and 4.0 cm by Day 156 of gestation. This pregnancy is ongoing and parturition is anticipated in July 2010. 4. Discussion The production of beluga calves reported herein using assisted reproductive technology is a milestone from 16 y of research on reproductive biology of this species. An important discovery during this research was that unlike other odontocetes examined thus far, beluga are facultatively induced ovulators [16]. Consequently, intensive endocrine monitoring and ovarian ultrasonographic assessments were necessary to devise a protocol for ovulation induction, thereby enabling determination of ovulation timing and development of AI. Research in beluga sperm biology revealed another aspect of the species divergent reproductive physiol-

ogy compared to other odontocetes. Since beluga spermatozoa were shown to be highly susceptible to damage during conventional freeze-thaw processes [15], a novel trehalose-based cryodiluent, combined with directional freezing technology, was used to achieve post-thaw sperm characteristics satisfactory for AI. Not surprisingly, using this cryopreservation method, our semen freezing efforts in the present study produced similar post-thaw results (38% progressive motility) to those previously reported [15]. Demonstration of the fertility of cryopreserved spermatozoa provided justication for the development of a beluga gamete resource bank which will allow for global exchange of beluga spermatozoa in conjunction with AI. In addition, ex situ AI programs may now be expanded to include spermatozoa collected post-mortem from incidental deaths of captive animals, or from wild animals killed during subsistence hunting in the arctic. In contrast to bottlenose dolphins [9,21], Pacic white-sided dolphins [10], killer whales [8], and indopacic humpback dolphins [25] where ovulation occurs on the left ovary greater than 70% of the time, based on the present study, beluga ovulate equally (50%) from each ovary. The observed lack of ovulation bias in beluga ovaries was consistent with post-mortem ndings in wild beluga, where there is equal distribution of corpora lutea on both ovaries [17]. Furthermore, animals in the present study exhibited double ovulations

T.R. Robeck et al. / Theriogenology 74 (2010) 989 1001

999

50% of the time, a unique nding which has not been observed in other odontocete species. Post-mortem reports of wild beluga demonstrate that animals frequently possess two CLs during early pregnancy, which when combined with our results, provided conclusive evidence for the occurrence of double ovulations in this species [17]. Estrus synchronization in the beluga using Regumate has not been previously reported. The period of time from withdrawing Regu-mate to follicular recruitment and POF development (21 d) was similar to other odontocetes [8 10]. Additionally this was the rst report on the use of Provera as a substitute for Regu-mate for estrus synchronization in any odontocete, which produced similar results with both compounds in the same animal on different occasions. All cetaceans appear to have unique vaginal and cervical folds that probably function to inhibit salt water contamination during copulation [26 27]. Beluga have been described as having a series of 10 15 vaginal folds, beginning midway in the vagina and moving cranially toward the cervix [17]. In the current study, adequate vaginal insufation was essential to allow passage of the endoscope beyond the vaginal folds to the cervical opening. Due to the small and constricted vaginal and rectal openings in cetaceans [28], IU placement of the endoscope and catheter could not be veried by palpation or ultrasonography. However, the lengths documented herein for the vagina, cervix and internal uterine bifurcation seemed consistent with post-mortem reproductive tract descriptions by Kleineberg et al [17]. These post-mortem data, which described a mean total reproductive tract length of 80 cm, and the mean insemination depth of 93 cm during IU insemination attempts conducted in the present study, substantiated our belief that semen was deposited in the proximal uterine horn. Despite the use of a deep IU insemination method, the conception rate for the beluga was poor (20%) as compared with killer whales (40%; [8]), bottlenose dolphins (60%; [9]) and Pacic white-sided dolphins (50%; [10]). Deep IU insemination has been used successfully in other species to reduce the insemination dose [29 33], increase fertility [34,35], and in some species increase the success of AI using semen from subfertile males [36]. In an attempt to optimize the conception rate in beluga, deep IU insemination using high numbers of spermatozoa was employed. Despite numerous inseminations (n 7), where the total num-

ber of progressively motile spermatozoa was 1.3 to 6.3 times the minimum effective insemination dose of 524 million spermatozoa, and that inseminations occurred within 12 h of ovulation, only one additional conception was achieved. A possible explanation for the low conception rate relates to the fact that the majority (7/10) of inseminations were unicornual, presumably in the horn ipsilateral to the ovary destined to ovulate. Initially, we hypothesized that even if spermatozoa were deposited into the contralateral horn of the ovulation, sperm migration to the opposite horn would occur, as it does in other species [37 40]. Since the uterine side of sperm deposition can not be conrmed during AI in beluga, it is possible that some inseminations were performed in the uterine horn contralateral to the side with the POF. The observed low pregnancy rate under high sperm doses might therefore be reective of a reduced ability of frozen-thawed beluga spermatozoa to undergo uterine migration and that it is likely that some inseminations were performed in the uterine horn contralateral to the side with the POF. This impediment could be overcome in future AI procedures by performing bilateral, low dose, deep IU inseminations. The progesterone proles during the rst 30 d postinsemination from three conceptive cycles are the rst reported from beluga with known conception dates. In these proles, there was a clear divergence of serum progesterone after Day 14 post-AI; this was due to the demise of the CL in non-pregnant animals. In other species, the demise of the CL, as reected by a decline in circulating progesterone concentrations, typically occurred during a 2 to 3 d window if a conceptus is not present [41]. If this relationship holds true for the beluga, maternal recognition of pregnancy probably occurs, by a yet to be determined mechanism, between Days 14 and 21 post-ovulation. Through the application of previous research in sperm biology and advances in the endoscopic insemination technique, this study demonstrated the rst use of cryopreserved beluga semen for AI. Such ndings encourage the development of gamete resource banks and validate the use of the ovulation induction protocol previously developed for this species. With climate change in the arctic moving at increasing rates, future effects on beluga populations cannot be predicted [42]; however, application of the information gained by the present research may eventually prove to be an important part of the overall effort to maintain healthy, sustainable populations of this species.

1000

T.R. Robeck et al. / Theriogenology 74 (2010) 989 1001 articial insemination technology in killer whales (Orcinus orca). Biol Reprod 2004;71:650 60. Robeck TR, Steinman KJ, Yoshioka M, Jensen E, OBrien JK, Katsumata E, Gili C, McBain JF, Sweeney J, Monfort SL. Estrous cycle characterization and articial insemination using frozen-thawed spermatozoa in the bottlenose dolphin (Tursiops truncatus). Reproduction 2005;129:659 74. Robeck TR, Steinman KJ, Ramirez K, Greenwell M, Van Bonn W, Yoshioka M, Katsumata E, Dalton L, Osborn S, OBrien JK. Seasonality, estrous cycle characterization, estrus synchronization, semen cryopreservation and articial insemination in the Pacic white-sided dolphin (Lagenorhynchus obliquidens). Reproduction 2009:138:391 405. Robeck TR, OBrien JK. Effect of cryopreservation methods and pre-cryopreservation storage on bottlenose dolphin (Tursiops truncatus) spermatozoa. Biol Reprod 2004;70:1340 8. OBrien JK, Robeck TR. Development of sperm sexing and associated assisted reproductive technology for sex pre-selection of captive bottlenose dolphins (Tursiops truncatus). Reprod Fert Dev 2006;18:319 29. OBrien JK, Steinman KJ, Robeck TR. Application of sperm sorting and associated reproductive technology for wildlife management and conservation. Theriogenology 2009;71:98 107. OBrien JK, Steinman KJ, Schmitt T, Robeck TR. Semen collection, characterisation and articial insemination in the beluga (Delphinapterus leucas) using liquid-stored spermatozoa. Reprod Fertil Dev 2008;20:770 83. OBrien JK, Robeck TR. Preservation of beluga (Delphinapterus leucas) spermatozoa using a trehalose-based cryodiluent and directional freezing technology. Reprod Fertil Dev 2010:22:653 63. Steinman KJ, OBrien JK, Robeck TR. Characterization of reproductive cycles and the development of an ovulation induction method in the beluga (Delphinapterus leucas). In: Proceedings International Association of Aquatic Animal Medicine 2007;38:162 4. Kleinenberg SE, Yablokov AV, Belkovich BM, Tarasevich MN. Beluga (Delphinapterus leucas) Investigation of The Species. Izdatelstvo Nauka, Moskva 1964. Israel Program for Scientic Translation, Jerusalem, 1969. Robeck TR, Steinman KJ, OBrien JK. Understanding reproductive physiology and development of assisted reproductive technology (ART) in beluga, Delphinapterus leucas: Contributions of captive and wild research. In: 1st International Workshop on Beluga Whale Research, Husbandry and Management in Wild and Captive Environments, Valencia, Spain, 2007. p. 7 8. Taussky HH. A microcolorimetric determination of creatine in urine by the Jaffe reaction. J Biol Chem 1954;208:853 61. Graham LG, Boling J, Miller G, Pratt-Hawkes N, Joseph S. Enzyme-immunoassay for the measurement of luteinizing hormone in the serum of African elephants (Loxodonta africana). Zoo Biol 2002;21:403 8. Brook FM. Sonographic imaging of the reproductive tract of the female bottlenose dolphin, Tursiops truncatus aduncas. Reproduction 2001;121:419 28. Graack M, Grovhoug N. Semen collection training with a beluga whale (Delphinapterus leucas). In: Proceedings of the International Marine Animal Trainer Association 2003;31 42. Lacave G, Eggermont M, Verslycke T, Brook F, Salbany A, Roque L, Kinoshita R. Prediction from ultrasonographic measurements of the expected delivery date in two species of bot-

Acknowledgements We thank the animal care laboratory, animal training and animal care staff at SeaWorld San Antonio (SWSA), SeaWorld San Diego (SWSD), Kamogawa Sea World (KSW), and Mystic Aquarium and Institute for Exploration (MA-IFE) for their support. For technical assistance, we thank Michelle Morrisseau (SWSD). For assistance with ultrasonographic exams, we thank former MA-IFE veterinary intern Dr. Michelle Davis. For semen collection and processing, we thank Sherry Dickerson and Chris White (SWSA), and Wild Arctic staff, in particular Mitzi Synnott and Nicole Grovhoug (SWSD). We thank Dr. Hideki Koi (Kameda Medical Center) for serum E2 analysis. We thank Brad Andrews (SeaWorld Parks and Entertainment Corporation, SEA) and Kazutoshi Arai (KSW) for their support of this project. This research was conducted under the NMFS permit number 116-1691, was funded by SEA, KSW and MA-IFE, and is a SeaWorld Technical contribution number 2009-05-T.
[9]

[10]

[11]

[12]

[13]

[14]

References
[1] Jefferson TA, Karczmarski L, Laidre K, OCorry-Crowe G, Reeves RR, Rojas-Bracho L, Secchi ER, Slooten E, Smith BD, Wang JY, Zhou K. Delphinapterus leucas. In: IUCN 2009. IUCN Red List of Threatened Species. Version 2009.1. Available at: www.iucnredlist.org. Accessed on: 16 October 2009. [2] Andrews B. Beluga North American Studbook. Marine Mammal Taxon Advisory Group, American Zoo and Aquarium Association (AZA): Silver Spring, MD. 2009. [3] Calle PP, Cook RA, McClave C, Palma S. Circulating gestational progesterone and estradiol concentrations, parturition, and placental descriptions of two beluga whales (Delphinapterus leucas), In: Proceedings International Association of Aquatic Animal Medicine, 1993. p. 24 61. [4] Calle PP, Cook RA, Robeck TR, Young SJF, Jones MH. Circulating gestational progesterone and oestradiol concentrations in Beluga whales (Delphinapterus leucas). In Proceedings of the American Association of Zoo Veterinarians, 1996. p. 340 42. [5] Calle PP, Monfort SL, Dunn JL, Jensen E, Boehm J, Young SJF, Robeck TR. Seasonal testosterone secretion in male white whales (Delphinapterus leucas). In: Joint Proceedings of American Association of Zoo Veterinarians and International Association of Aquatic Animal Medicine, 2000. p. 49 50. [6] Russell JM, Simonoff JS, Nightingale J. Nursing behavior of beluga calves (Delphinapterus leucas) born in captivity. Zoo Biol 1997;16:247 62. [7] Robeck TR, Monfort SL, Calle PP, Dunn JL, Jensen E, Boehm JR, Young S, Clark ST. Reproduction, growth and development in captive beluga (Delphinapterus leucas). Zoo Biol 2005;24: 29 49. [8] Robeck TR, Steinman KJ, Gearhart S, Reidarson TR, McBain JF, Monfort SL. Reproductive physiology and development of

[15]

[16]

[17]

[18]

[19] [20]

[21]

[22]

[23]

T.R. Robeck et al. / Theriogenology 74 (2010) 989 1001 tlenose dolphin (Tursiops truncatus and Tursiops aduncus). Vet Rec 2004;154:228 33. Pursel VG, Johnson LA. Freezing of boar spermatozoa: fertilizing capacity with concentrated semen and a new thawing procedure. J Anim Sci 1975;40:99 102. Brook FM, Lim EHT, Chua FHC, Mackay B. Assessment of the reproductive cycle of the Indo-Pacic humpback dolphin, Sousa chinensis, using ultrasonography. Aquatic Mammal 2004;30: 137 48. Harrison RJ. Reproductive and reproductive organs. In: The biology of Marine Mammals Anderson HT (Ed.), Academic Press, New York, New York; 1969. p. 253348. Green RF. Anatomy of the reproductive organs in dolphins. In: Breeding Dolphins: Present Status, Suggestions for the Future, Ridgway SH, Benirschke K. (Eds.), NTIS PB-273 673, US Dept of Commerce, Washington, DC; 1977. p. 18591. Robeck TR, Curry BF, McBain JF, Kraemer, DC. Reproductive biology of the bottlenose dolphin (Tursiops truncates) and the potential application of advanced reproductive technologies. J Zoo Wild Med 1994;25:32136. Brinsko SP, Rigby SL, Lindsey AC, Blanchard TL, Love CC, Varner DD. Pregnancy rates in mares following hysteroscopic or transrectally-guided insemination with low sperm numbers at the utero-tubal papilla. Theriogenology 2003;59:10019. Martinez EA, Vazquez JM, Roca J, Lucas X, Gil MA, Parrilla I, Vazquez JL, Day BN. Minimum number of spermatozoa required for normal fertility after deep intrauterine insemination in non-sedated sows. Reproduction 2002;123:16370. Martinez EA, Vazquez JM, Parrilla I, Cuello C, Gil MA, Rodriguez-Martinez H, Roca J, Vazquez JL. Incidence of unilateral fertilization after low dose deep intrauterine insemination in spontaneously ovulating sows under eld conditions. Reprod Dom Anim 2006;41:41v7. Skidmore JA, Billah M. Comparison of pregnancy rates in dromedary camels (Camelus dromedarius) after deep intra-uter-

1001

[24]

[33]

[25]

[34]

[35]

[26]

[27]

[36]

[37]

[28]

[29]

[38] [39] [40]

[30]

[31]

[41]

[42]

[32]

ine versus cervical insemination. Theriogenology 2006; 66:292 6. Vazquez JM, Roca J, Gil MA, Cuello C, Parrilla I, Vazquez JL, Martnez EA. New developments in low-dose insemination technology. Theriogenology 2008;70:1216 24. Senger PL, Becker WC, Davidge ST, Hillers JK, Reeves JJ. Inuence of cornual insemination on conception rates in dairy cattle. J Anim Sci 1988;66:3010 6. Dalton JC, Nadir S, Barne JH, Saacke RG. Effect of a deep uterine insemination of spermatozoal accessibility to the ovum in cattle: a competitive insemination study. Theriogenology 1999;51:88390. Hunter RHF. Advances in deep uterine insemination: a fruitful way forward to exploit new sperm technologies in cattle. Anim Reprod Sci 2003;79:15770. Fukushima FB, Malm C, Henry M, Gheller VA, Serakides R, Neves MM, Macedo SP, Figueiredo MS, Andrade MEJ, Chaves MS, Silva MX, Rezende CMF, Melo EG. Site of intrauterine articial insemination in the bitch does not affect sperm distribution within the uterus. Reprod Dom Anim 2009 Jul 23 [Epub ahead of print]. Katila T. Sperm-uterine interactions: a review. Anim Reprod Sci 2001:26772. Troedsson MHT, Liu LKM, Crabo BG. Sperm transport and survival in the mare: a review. Theriogenology 1998;49:90515. Bourke DA, Lindsay FEF. Uterine contraction in the cow and associated movement of inseminate and other substances. In: Proceedings of the 11th International Congress on Animal Reproduction and AI; 2003. p. 233. Bazer FW, Ott TL, Spencer TE. Maternal Recognition of pregnancy: comparative aspects. A Review. Trophoblast Research 1998;12:375 86. Moore, SE. Climate Change. In: Encyclopedia of Marine Mammals, Second Edition. Academic Press. Elsevier, B.V.; 2009. p. 238 41.