Ecotoxicology and Environmental Safety 136 (2017) 2430

Contents lists available at ScienceDirect

Ecotoxicology and Environmental Safety

journal homepage: www.elsevier.com/locate/ecoenv

Muscle mercury and selenium in shes and semiaquatic mammals from a

selenium-decient area
crossmark

Elzbieta Kalisinskaa, Natalia Lanocha-Arendarczyka, Danuta Kosik-Bogackaa, , Halina Budisb,
Bogumila Pilarczykc, Agnieszka Tomza-Marciniakc, Joanna Podlasinskad, Lukasz Cieslike,
Marcin Popiolekf, Agnieszka Pirogg, Ewa Jedrzejewskah
a
Department of Biology and Medical Parasitology, Pomeranian Medical University, in Szczecin, Poland
b
Department of Health Education, University of Szczecin, Poland
c
Chair of Animal Reproduction Biotechnology and Environmental Hygiene, West Pomeranian University of Technology in Szczecin, Poland
d
Department of Environmental Management and Protection, Western Pomeranian University of Technology,in Szczecin, Poland
e
Warta Mouth National Park, Poland
f
Department of Parasitology, Institute of Genetics and Microbiology, University of Wroclaw, Poland
g
Department of Invertebrate Systematics and Ecology, Institute of Biology, Wroclaw University of Environmental and Life Sciences, Poland
h
Ecoexpert Sp. z o.o., Poland

A R T I C L E I N F O A BS T RAC T

Keywords: The aim of this study was to investigate and compare total mercury (Hg), selenium (Se), and Se:Hg molar ratios
Piscivorous mammals in sh muscles (phytophages n=3; benthophages n=32; predators n=5) and semiaquatic carnivores, including
Fish piscivores (the European otter n=8, the feral American mink n=7) and the omnivorous raccoon (n=37) from a
Selenium/mercury molar ratio riverine European ecosystem in a Se-decient area. The Hg concentration in sh reached 0.337 g/g dry weight,
Riverine ecosystem
dw (0.084 g/g wet weight, ww). We found signicant dierences among Hg levels in tested vertebrate groups
(predators vs benthophages: 0.893 vs 0.281 g/g; piscivores vs omnivores: 6.085 vs 0.566 g/g dw). Fish
groups did not dier in Se concentrations, with a mean value of 0.653 g/g dw. Signicant dierences were
revealed between Se levels in piscivorous and omnivorous carnivores (0.360 vs 0.786 g/g dw, respectively).
Fish Se:Hg molar ratio values were > 2.2. Benthophages had higher the ratio than predators but similar to
phytophages. Among carnivores, piscivores had much lower the ratio than raccoon (0.14 vs 3.75) but raccoon
and sh medians did not signicantly dier. We found almost two times higher Se levels in sh and raccoons
compared to piscivores, possibly resulting from lower sh Se digestibility by piscivores in contrast to higher
absorption of plant Se by many sh and omnivorous raccoons. Considering that a tissue Se:Hg molar ratio < 1
may be connected with a Hg toxicity potential increase, we assume that piscivores in Se-decient area are in
worse situation and more exposed to Hg than sh and omnivores.

1. Introduction atmospheric deposition, and euent discharged into surface waters.

Mercury (Hg) in the environment can be of natural or anthropo-
genic origin. It is highly toxic heavy metal with high volatility. It is
estimated that only 19% of deposited Hg in the European Union (EU),
with an annual average of 138.5 t between 2005 and 2011, originate
from local sources (Chen et al., 2014). Since the 1990s, Poland and
Germany with coal-based energetics have emitted the largest amounts
of Hg within the EU. Between 1990 and 2012, emissions from these
two countries have declined and as a consequence, annual atmospheric
Hg input has decreased by 70%, from around 3210 t (EEA, 2014). In
inland aquatic environments, major sources of Hg are the bedrock,
Locally, the latter two sources can lead to Hg intoxication in sh and
other organisms. In water sediments, biotransformation of inorganic
mercury (InHg) through microorganisms results in the formation of
methylmercury, MeHg (AMAP/UNEP, 2013). Although all Hg forms
are toxic for warm-blooded vertebrates, MeHg is particularly danger-
ous because of its high bioavailability. It is almost completely absorbed
( > 90%) upon ingestion and is highly neurotoxic. All Hg species are
subjected to the bioaccumulation but MeHg additionally biomagnica-
tion in aquatic food chains to varying degrees, depending on Hg
amounts and forms. Methylmercury predominates in sh muscles ( >
80% of total Hg, THg), and its concentration is signicantly correlated

Correspondence to: Department of Biology and Medical Parasitology, Pomeranian Medical University, Powstancow Wielkopolskich 72, 70111 Szczecin, in Szczecin, Poland.
E-mail address: kodan@pum.edu.pl (D. Kosik-Bogacka).

http://dx.doi.org/10.1016/j.ecoenv.2016.10.028
Received 27 June 2016; Received in revised form 13 October 2016; Accepted 23 October 2016
Available online 31 October 2016
0147-6513/ 2016 Elsevier Inc. All rights reserved.
E. Kalisinska et al.
Fig. 1. Study area and the main source of its contamination.
with age, weight, and body length. Moreover, the MeHg level depends
on the trophic position of sh species. Monitoring toxic Hg concentra-
tions in biota are widely and increasingly used for assessing the
chemical status of aquatic ecosystems, mainly selecting common
bentophagous or predatory sh (Scheuhammer et al., 2007; Peterson
et al., 2009; Schfer et al., 2015).
Numerous ecotoxicological studies revealed the highest THg levels
in liver and kidneys of terrestrial wildlife feeding on aquatic animals
but both tissues account 46% of their body weight only (Lanszki et al.,
2008). Methylmercury demethylation mostly occurs in the liver and
kidneys, leading to the production of the less toxic InHg, which is
excreted with feces and urine. When nephritic and hepatic THg
concentrations are greater than 810 mg THg/kg dry weight (dw),
MeHg may be well below 4050% of THg in semiaquatic carnivores
(OConnor and Nielsen, 1981; Fortin et al., 2001). In avian and
mammalian muscles demethylation of MeHg is insucient or does
not occur at all, leading to MeHg percentages of around 90% of THg
(Scheuhammer et al., 2007). Skeletal muscles constitute 5055% of
carnivore bodies (Muchlinski et al., 2012) and play a greater role in
MeHg transmission from the aquatic to the terrestrial environment
than liver and kidneys. Determination of Hg muscle levels in sh and
semiaquatic carnivore allows conclusions about contamination of e.g.
riverine and riparian ecosystems. However, ecotoxicological studies of
muscle Hg concentrations in sh and semiaquatic predators from the
same area are rare (Chumchal et al., 2011).
In mammals and sh, an important detoxication mechanism is the
antagonistic impact of selenium (Se) in relation to Hg, which is well
documented for marine species, and there are various hypotheses for
Se protective mechanisms. Selenium is essential element for verte-
brates and in some of their proteins (selenoproteins). In comparison to
25
Ecotoxicology and Environmental Safety 136 (2017) 2430
bony sh birds and mammals produce less number of selenoproteins.
In mammals, the dierence between necessary and harmful Se amount
is very small. Selenium deciency results in muscle pathology, fetal
death and reproduction deterioration. In some regions, i.e. west North
America, Se is present in signicant quantities in surface waters and
soils, resulting in excessive Se bioaccumulation in sh, waterfowl, and
grazing animals (Fordyce, 2005). In other regions, e.g. some areas of
Germany, Poland, and Scandinavia Se deciency in soils was observed,
resulting in low bioavailability for mammals (Salminen, 2005;
Pilarczyk et al., 2011).
Ecotoxicologists are especially interested in various relations be-
tween Se and Hg in animals. In some marine vertebrate species, the
insoluble, biologically inactive forms are generated as nontoxic combi-
nations of SeHg, including equimolar HgSe compound. The presence
of HgSe was rst demonstrated in livers, kidneys and brains, and more
recently, in marine mammal muscles (Nakazawa et al., 2011). It is
likely that with low Hg molar concentrations in relation to Se in organs
and tissues, selenoenzyme remains active. However, when tissue Hg
exceeds Se, Se tissue portions vital for antioxidant selenoenzyme
production become sequestered and symptoms of Hg toxicity develop.
Therefore, Hg concentration in an organism might not be the critical
factor, but rather the ratio of tissue Hg and Se. Hence, Se:Hg molar
ratios below one may increase Hg toxicity potentials, while molar ratios
that approach or exceed one eectively may protect against Hg toxicity
(Peterson et al., 2009; Zhang, 2014). Information about Se and Hg
levels in freshwater sh and semiaquatic mammals is scarce (Wren,
1984; Burger and Gochfeld, 2013). In the literature no information on
correlation, proportion between Se and Hg or Se:Hg molar rations in
carnivore muscles has been found. Moreover, probably studies on of
these elements in muscles of freshwater sh and semiaquatic mammals
E. Kalisinska et al.
living on Se-decient area have not been performed.
The primary objective of the study was to assess bioaccumulation
and biomagnication of Hg in freshwater sh and carnivores in Se-
decient area and to analyze Se concentrations and Se:Hg molar ratios
in vertebrate muscles.
2. Material and methods
2.1. Study area and sampling
The Warta Mouth National Park (WMNP) in west Poland (near the
Polish-German border) covers around 80 km2 of the Warta River
Valley. The river is the largest tributary of the Odra River (Poland's
second largest river), and it's mouth is near Kostrzyn (52.5883N;
14.6567E). The park is periodically inundated by both rivers, whereby
the Odra transports a considerable amount of heavy metals from Polish
and German industrial areas south of Kostrzyn (metallurgical industry
in Germany, Eisenhuttenstadt, copper mining industry, Legnica-
Gogw Copper District) (Fig. 1). Various plants operate in Kostrzyn,
including a paper mill with over 80 years of production history
(Lanocha et al., 2014). In sediments from the Warta near Kostrzyn,
high mercury concentrations (sometimes > 1.5 g/g dw) have been
found (CSO, 2014).
Between 2009 and 2014 vertebrates were legally gathered.
Taxonomic positions of sh (eight species), semiaquatic carnivores
(three species) and trophic category of sh were established by the park
sta. Average body weight and ranges are given in the parentheses.
Fish predators (which feed mainly on sh) were represented by the
northern pike Esox lucius (n=1, 985 g), asp Leuciscus aspius (n=1,
1090 g), ide Leuciscus idus (n=3, 940 g, 6501470 g); benthophages
(feeding on benthic aquatic invertebrates) by roach Rutilus rutilus
(n=25, mean 58 g, 10332 g), bream Abramis brama (n=1, 1225 g),
white bream Blicca bjoerca (n=5, 160g, 155211 g), crucian carp
Carassius carassius (n=1, 490 g); and rudd Scardinius erythrophthal-
mus (n=3, 145 g, 100200 g) belonging to the group phythophages.
Carnivores, all belonging to mammals, were divided into two groups:
piscivores, including the European otter Lutra lutra (n=8, 4000
7800 g) and the feral American mink Neovision vision (n=7, 1150 g,
8001400 g) and omnivores, the raccoon Procyon lotor (n=37, 4000 g,
5508100 g). Taking into account literature data, we have assumed
that in the WMNP aquatic food (mainly sh) in diet of European otters,
American minks and raccoons constitute of 90%, 60% and 30% sh,
respectively.
Specimens were stored in individual plastic bags at 20 C until
weighing and dissection in the Laboratory of Pomeranian Medical
University in Szczecin. Samples from 450 g of sh muscle were taken
and at least 50 g from the mammal hindlimbs.
2.2. Chemical analysis
All samples were dried to constant weight at 55 C and then ground
in an agate mortar. Mean moisture contents in sh and carnivore
muscles were 74.5% and 73.4%, respectively; to facilitate comparison
of dry and wet weight, we used a dry-to-wet weight conversion rate for
sh and mammal muscles.
Total Hg concentrations were determined using atomic absorption
spectroscopy (AAS) at the Department of Environmental Management
and Protection, Western Pomeranian University of Technology in
Szczecin. The assays were run in an AMA 254 mercury analyser
(Altach Ltd, Czech Republic) in accordance with previously described
methods (Kalisinska et al., 2012). The apparatus has a detection limit
of 0.01 ng and a detection range of 0.05600 ng. The analytical
procedure used was checked determining Hg concentrations in two
reference materials: DOLT-4 (Dogsh Liver, n=3) and 8414 NIST
(Bovine Muscle, n=3). The mean Hg concentrations were 99.2% for
DOLT-4% and 108% for 8414 NIST of the reference values.
26
Ecotoxicology and Environmental Safety 136 (2017) 2430
Concentrations of total selenium in verteberate muscles were
determined using spectrouorimetric method (details see Pilarczyk
et al., 2011). All chemicals used were of analytical reagent grade. Most
of chemicals were obtained from Chempur, except 2,3-diaminonaph-
talene which was obtained from Sigma-Aldrich. The muscles were
digested in HNO3 at 230 C for 180 min and in HClO4 at 310 C for
20 min. Then 9% HCl was added to reduction Se(VI) to Se(IV).
Selenium was derivatised with 2,3-diaminonaphtalene (Sigma-
Aldrich) under the conditions of controlled pH (pH 12) with the
formation of selenodiazole complex. This complex was extracted into
cyclohexane. EDTA and hydroxylamine hydrochlorine were used as
masking agents. Selenium concentration was determined uorimetri-
cally using RF-5001 PC Shimadzu spectrophotouorimeter. The ex-
citation wavelength was 376 nm and the uorescence emission wave-
length was 518 nm. The detection limit of Se was 0.003 g/g. It was
validated with certied reference materials: NCS ZC 71001 (beef liver,
n=5), IAEA 407 Fish (sh tissue, n=4) and 8414 Bovine Muscle (bovine
muscle, n=4). The mean Se concentrations were 92.5%, 94% and 92%
of the reference value, respectively.
2.3. Statistical analysis
Statistical analyses were performed using Statistica PL v. 9.1. All
data are expressed throughout as medians, lower and upper quartiles.
All tissue concentrations of Se and Hg are presented as parts per
million (ppm, g/g) on dry weight; additionally median values on wet
weight. The distribution of Hg concentrations in the animal samples
deviated from the expected normal distribution, and so in the statistical
analysis the comparison of mean concentrations of Hg and Se, non-
parametric KruskalWallis (KW) or MannWhitney (MW) tests
were used when the number of means was respectively 3 or equal 2.
Relationships between Hg and Se concentrations in animal muscle and
an element body weight (in roach and raccoon) were evaluated by
calculation of Spearman correlation coecient (rS). In all tested
comparisons and relationships the dierences or correlation coe-
cients were considered statistically signicant with an alpha level of
0.05.
Dry and wet weight data coming from this study and various reports
were recalculated into nmol/g by the following formula: molar
concentration (nmol/g)=concentration (nmol/g)1000/atomic weight
(g/mol). Atomic weights of Hg and Se are 200.59 and 78.96 g/mol,
respectively. The recalculated Se and Hg concentrations were used to
calculation of Se:Hg molar ratios. Additionally, based on muscle
median concentrations of Hg in shes as potential preys of carnivores
biomagnication factors (BMF) were calculated as follows:
BMF=carnivore Hg concentration (g/g)/sh Hg concentration (g/g).
3. Results
3.1. Mercury
Mercury concentrations ranged from 0.1821.400 g/g in dry
muscle of sh and from 0.01315.877 g/g in mammals (Table 1).
Median of Hg concentration reached 0.337 g/g dw (0.084 g/g ww)
for all sh specimens. A signicant relationship between body weight
and Hg levels (Fig. 2) was revealed for the roach. We found no
dierence in Hg concentrations between benthophages and phyto-
phages; however, Hg concentrations were signicantly dierent be-
tween predators and benthivores (Table 2).
The correlation between body weight and Hg concentration in
raccoons (rS=0.377; p < 0.05) was found. The metal concentrations in
otters and minks were similar, however piscivores showed Hg levels
more than ten times higher than those in omnivorous raccoons
(Table 2). Mercury muscle concentrations varied among vertebrate
groups. The mercury levels in piscivores were signicantly higher than
in sh (M-W test: Z=5.66; p < 0.0001), whereas sh and raccoon Hg
E. Kalisinska et al. Ecotoxicology and Environmental Safety 136 (2017) 2430

Table 1 Table 2
Total mercury (Hg) and selenium (Se) concentrations in g/g dw (or ww) also Se:Hg Total mercury (Hg) and selenium (Se) concentrations, also Se:Hg molar ratios in muscles
molar ratio values in fish and mammalian muscles from the Warta Mouth National Park of fish and carnivore ecological groups from Se-deficient area.
(Med, median; Q1 and Q3 lower and upper quartiles; dw, dry weight; ww, wet weight; ad,
adult; imm, immature). Ecological groups n Median Quartils (Q1 & Range min
Q3) max
Species n THg Med dw Se Med dw (ww) Se:THg Med
(ww) Q1 & Q3 Q1 & Q3 Range Q1 & Q3 Mercury, g/g dw (median g/g ww)
Range Range
Fish
Fish Phytophage 3 0.399 (0.092) 0.265 & 0.436 0.2650.436
Northern pike 1 0.990 (0.222) 1.001 (0.205) 2.57 Benthophage 32 0.281 (0.075) 0.232 & 0.479 0.1820.613
Asp 1 1.400 (0.308) 0.781 (0.172) 1.42 Predator 5 0.893 (0.203) 0.740 & 0.990 0.7231.400
KW test H=13.08, p < 0.01
Ide 3 0.740 (0.189) 0.643 (0.164) 2.21
0.723 & 0.893 0.516 & 1.035 1.47 & 3.64 Mammals
0.7230.893 0.5161.035 1.473.64 Piscivore 15 6.085 (1.746) 3.648 & 9.184 1.93115.880
Omnivore 37 0.566 (0.152) 0.143 & 1.091 0.0132.773
Roach 26 0.258 (0.067) 0.667 (0.176) 6.56 MW test Z=5.1, p < 0.0001
0.227 & 0.321 0.535 & 0.801 4.72 & 7.72
0.1820.610 0.4111.190 2.5110.95 Selenium, g/g dw (median g/g ww)
Bream 1 0.408 (0.097) 1.131 (0.269) 7.04
Fish
White bream 5 0.516 (0.132) 0.584 (0.149) 2.89 Phytophage 3 0.537 (0.114) 0.452 & 0.541 0.4520.541
0.474 & 0.553 0.573 & 0.587 2.57 & 3.44 Benthophage 32 0.668 (0.173) 0.550 & 0.819 0.4111.186
0.3740.578 0.4280.750 2.293.89 Predator 5 0.781 (0.172) 0.643 & 1.001 0.5161.035
Crucian carp 1 0.460 (0.099) 1.046 (0.226) 5.78 KW test NS

Rudd 3 0.367 (0.092) 0.541 (0.114) 3.74 Mammals

0.265 & 0. 436 0.537 & 0.542 3.13 & 5.19 Piscivore 15 0.360 (0.100) 0.207 & 0.671 0.1780.776
0.2650.436 0.5370.542 3.135.19 Omnivore 37 0.786 (0.209) 0.457 & 0.920 0.2444.127
MW test Z=3.53, p <
Mammals 0.001

European otter 8 3.683 (0.972) 0.343 (0.088) 0.27 Molar ratio Se:Hg
2.525 & 5.133 0.194 & 0.589 0.17 & 0.37 Phytophage 3 3.44 3.13 & 4.33 3.134.33
1.936.085 0.1780.728 0.100.46 Benthophage 32 5.93 3.43 & 7.48 2.2910.95
Predator 5 2.21 1.47 & 2.57 1.423.64
American mink 7 9.184 (2.801) 0.360 (0.100) 0.11 KW test H=10.8, p < 0.01
8.145 & 12.593 0.214 & 0.671 0.06 & 0.14
6.76615.877 0.1780.776 0.040.23 Mammals
Piscivore 15 0.14 0.10 & 0.30 0.040.46
Raccoon 37 0.566 (0.152) 0.786 (0.209) 3.75 Omnivore 37 3.75 2.23 & 7.61 0.7365.99
0.143 & 1.091 0.457 & 0.920 2.23 & 7.61 MW test Z=5.59, p <
0.0132.773 0.2444.127 0.7365.99 0.0001

value was noted in an adult raccoon (Table 1). In immature and adult
raccoons, Se concentrations were similar and positively correlated with
body weights (rS=0.331; p < 0.05). Among carnivore species, we found
signicant dierences in Se levels (K-W test H=16.1; p < 0.001). The
median of Se level was higher in racoon than in otter, mink and all
piscivores, with no signicant dierence between piscivore species.
Selenium concentrations in sh and raccoon muscles were similar, but
we found signicantly lower element level in piscivores than sh (0.360
vs 0.653 ppm dw; MW test: Z=3.52, p < 0.001).

3.3. Selenium-mercury relationship and Se:THg molar ratio

Fig. 2. Relationship between body weights and muscle Hg levels in the roach from the
Warta Mouth National Park.
concentrations were similar (0.337 vs 0.574 ppm dw). On the basis of
Hg concentrations, the calculated BMF values for otter, mink, and
raccoon were 10.9, 27.3, and 1.7, respectively.
3.2. Selenium
In sh muscles, Se concentrations ranged from 0.4111.190 g/
g dw, with no dierence between sh groups (Table 1). Values for all
sh reached 0.653 g/g dw (0.168 g/g ww). Carnivore muscle sele-
nium levels were between 0.178 and 4.127 ppm dw, and the largest
In all sh samples Se:Hg molar ratios were > 1 with the highest and
lowest values in the roach (10.95) and the asp (1.42) (Table 1).
Benthophages had higher ratio than predators but similar to phyto-
phages (Table 2). In carnivores Se:Hg molar ratio values ranged from
0.04 in the mink to 66.0 in the raccoon. In otters and minks ratios were
similar and constantly < 0.50. In piscivores the Se:Hg ratio reached
0.14, which was 27 times lower than raccoons (Tables 1, 2). There was
no signicant dierence between sh and raccoon, but between sh
and piscivores was (MW test: Z=5.66, p < 0.0001). In raccoons, we
found a positive correlations between muscle Hg and Se concentrations
and between body mass and Se:Hg molar ratio (rS=0.640, p < 0.0001
and rS=0.331, p < 0.05, respectively).

27
E. Kalisinska et al.
4. Discussion
Mercury pollution is a global issue and linked with toxic metal
accumulation in aquatic vertebrates. Recently, there has been an
increased interest in the Se and Hg relationship in muscles of marine
mammals and marine and freshwater sh (Nakazawa et al., 2011;
Burger and Gochfeld, 2013; Zhang, 2014). However, data on Se, Hg
and Se:Hg ratios in terrestrial piscivores are limited (Wren, 1984;
Kalisinska et al., 2014). Most studies are dealing with Hg concentra-
tions in sh since high MeHg levels pose a serious threat to humans
and wildlife. In sh, Hg concentrations range from 0.00111.4 ppm
ww, with highest values in predatory marine species (WHO, 2011).
Variability of sh muscle Hg is considerably high and aected by
several abiotic and biotic factors. Age (strongly correlating with weight
and length) is the main dierentiating factor, which has been docu-
mented in many freshwater sh species, including the roach (Akerblom
et al., 2014; Jurgens, 2015; Nguetseng et al., 2015; Fig. 2).
4.1. Mercury
According to the EU Water Framework Directive (WFD),
Environmental Quality Standards (EQS) for some chemicals in biota
have been set, with mercury being dened as a priority hazardous
substance (Directive, 2008/105/EC). EQS is intended to protect top
predators against secondary poisoning and refers to THg; for fresh-
water sh, the EQS for Hg (EQS/Hg) is at 0.02 g/g ww. A maximum
Hg residue level of 0.50 ppm ww applies for sh intended for human
consumption; however, in some predatory species, including the pike,
the maximum residue level is twice higher (Commission Regulation,
EC, 2006). In the last three decades, Hg muscle concentrations above
2.5 ppm ww were rarely recorded in European freshwater sh, and
mean values did not exceed the EU maximum limit for THg in human
food (Akerblom et al., 2014; Nguetseng et al., 2015). Highest Hg
concentrations in predatory and non-predatory shes from the Warta/
Odra ranged from 0.230.31 and 0.160.31 ppm ww (Szkoda et al.,
2014; Table 2). In our study, EQS/Hg value exceeded these concentra-
tions 4.6, 3.7, and 10.1 times in phytophages, benthophages, and
predator sh. In roach from Swedish and Norwegian lakes, the Hg
muscle levels were two to seven times higher than in our roach and
exceeded the EQS/Hg (Julshamn et al., 1986; Myreng, 2013; Akerblom
et al., 2014).
Since the end of the 1990s, Hg emission has decreased considerably
in many EU countries, including Poland, as a result of economic
transformation and restrictive environmental protection law, especially
aquatic ecosystems. However, in studies on freshwater sh such as
breams from France, Sweden, the Netherlands, the UK and roach from
Germany and Sweden, the recommended value of EQS/Hg was
exceeded (Akerblom et al., 2014; Nguetseng et al., 2015). Roach from
four rivers in England, sampled between 2007 and 2011, showed mean
muscle Hg levels ranging from 0.0060.068 ppm ww, with 21%
showing levels below 0.02 g/g ww (Jurgens et al., 2013). Previous
studies in this eld do not indicate increased Hg concentrations in
waters and sediments but locally, they still show persistent elevated
levels of the metal, possibly resulting from former metal mining and
industry using Hg in technological processes (like chlor-alkali plants,
paper mills) and current Hg deposition and direct releases of Hg forms
from industry and urban wastewater treatment plants (Nguetseng
et al., 2015). The Europe-wide geochemistry survey (Salminen, 2005)
reported a median river bed-sediment concentration of 0.038 ppm Hg
dw, however, in sediment from the Warta/Odra site, higher values were
recorded, e.g. 1.57 ppm dw in 2013 (CSO, 2014). This indicates
sustained Hg contamination of riverine ecosystems, which is conrmed
by our study.
Unlike sh, Hg threshold values have not been established for other
wildlife. In the EU, semiaquatic carnivores represent the native and
protected European otter and two hunting invasive species: the
28
Ecotoxicology and Environmental Safety 136 (2017) 2430
American mink and the raccoon, originating from North America.
This allows various ecotoxicological comparisons. In experimental and
eld studies on Hg-poisoned terrestrial piscivores, extremely high
muscle Hg concentrations of up to 67120 ppm dw (1628 ppm ww)
were reported, which has been documented in the North American
river otter Lontra canadensis (OConnor and Nielsen, 1981; Sleeman
et al., 2010). In European otters (from Great Britain, Spain and
Finland), river otters from North America inhabiting uncontaminated
or slightly contaminated areas, and one control specimen used in the
experiment mean muscle THg concentrations reached up to 1 ppm ww
or 3.8 ppm dw (OConnor and Nielsen, 1981; Wren, 1984; Mason et al.,
1986; Ruiz-Olmo et al., 2000; Hyvrinen et al., 2003). In James Bay,
Canada, a region with high levels of naturally occurring Hg, muscle
THg concentrations were larger and reached 1.29 and 2.14 ppm ww in
river otters and minks, respectively (Fortin et al., 2001). Also in minks
Hg concentrations were higher than in otters from the WMNP (2.80 vs
0.97 ppm ww); however, the dierence was not statistically signicant,
probably due to the small sample number. Similar studies (e.g. Fortin
et al., 2001) indicate increased susceptibility on Hg intoxication for
minks compared to otters.
Raccoon meat, sometimes consumed by humans, contained sig-
nicantly lower Hg concentration than meat of piscivores, namely
0.15 ppm ww; however, this value is seven times higher than the
maximum permissible Hg level at 0.02 g/g ww in food of animal
origin, established for the National Residue Control Plan in Poland
(Szkoda et al., 2011). North American studies have found mean muscle
Hg concentrations in raccoons from unpolluted or moderately polluted
areas of 0.020.08 and 0.280.66 ppm ww, respectively (Wren, 1984;
Burger et al., 2002; Chumchal et al., 2011). In our study, raccoon had a
median Hg level of 0.15 ppm ww, but 30% of specimens exceeded
0.28 ppm ww. All our results indicate moderate Hg pollution of the
WMNP.
4.2. Selenium
Selenium concentration in rivers across Europe is generally low
does not exceed 0.15 mg/L; however, high levels of up to 1.50 g/L
have been found in some areas in south-east England (Salminen,
2005). In the EU, there is currently no EQS for Se either in water or
biota. The United States Environmental Protection Agency has pub-
lished a comprehensive draft on Se standards for aquatic life and
ambient water (US EPA, 2015), with Se limits for sh at a level of
11.3 g/g dw (or 2.94 g/g ww assuming that sh muscle contains 74%
water); this level is lower than the muscle value for the most sensitive
sh taxons, which were tested. Generally in USA and Canada, in
freshwater sh Se is present at levels of 0.420.64 ppm ww. However,
in high-Se inland waters, e.g. western USA, Se can be problematic as it
accumulates in the food chain (Fordyce, 2005). In the WMNP, muscle
Se concentrations in sh were below the US criterion and the typical
range of Se levels in North American freshwater sh, and Polish
extreme values were much lower than in roach from English rivers
(0.548.46 ppm dw) (Jurgens, 2015). In roach from east Norway and
the WMNP with similar body mass, around 50 g, Se concentrations
were comparable: 0.12 and 0.18 ppm ww (Julshamn et al., 1986;
Table 2). Higher Se values of 0.40 ppm ww were found in larger
specimens, weighing 50200 g, from South-eastern Norway (Myreng,
2013).
Compared our and the ndings of Julshamn et al. (1986) to Se
concentrations in 35 roach groups from dierent rivers and sites in
south-east England (with mean values ranging from 0.1351.006 ppm
ww), which were investigated by Jurgens (2015) a spectacular dier-
ence can be observed. On the basis of Jurgens (2015) data, we
calculated a median value of 0.46 ppm, which is 34 times higher
than in roach from the WMNP and east Norway.
River otter and raccoon muscle from North America contained
about 0.2 and 0.1 ppm ww of Se; with no dierence between raccoons
E. Kalisinska et al. Ecotoxicology and Environmental Safety 136 (2017) 2430

from contaminated and uncontaminated areas (Wren, 1984; Burger protein. Specialized piscivores have less bioavailable Se in their diet,
et al., 2002). In our study, in muscles of European otters and minks Se while sh, which are more exposed to MeHg, are the main causes of the
levels were similar (~0.10 ppm ww) and twice lower than level in observed dierences in Se, Hg concentrations, and Se:Hg molar ratios
raccoon. This suggests that, although the environmental abundance of in semiaquatic carnivores.
Se varies in Europe and North America, its bioaccumulation in
carnivore muscles in both continents is similar. It should also be noted
4.3. Selenium-mercury relationship and Se:THg molar ratio
that during evolution, there was a reduction in selenocysteine (SeCys)
participation in selenoproteins, which occurred in only in terrestrial
Mercury or Se, when present individually, may lead to poisoning of
animals. The evolution from sh to mammals was probably accom-
vertebrates in inland ecosystems. If, however, they co-occurrence in the
panied by decreased use of SeCys and increase of amino acid cysteine.
environment and bioaccumulate in vertebrates in moderately large
Recent analysis of selenoproteoms of 44 sequenced vertebrates showed
amounts, an antagonistic relationship between Hg and Se exists, with
that among 45 selenoprotein subfamilies, 28 were found in mammals
moderating eects of Se on InHg and MeHg toxicity (Peterson et al.,
and 41 in bony sh (Mariotti et al., 2012). Further research concerning
2009; Zhang, 2014). Although dietary Se reduces MeHg retention in
the role and demand for SeCys and use of selenomethionine (SeMet) as
freshwater sh, Hg and Se levels in sh and piscivorous bird muscle
Se source is required (Penglase et al., 2015). The SeMet is dominated
showed a weak or no correlation (Belzile et al., 2006; Bjerregaard et al.,
by Se-amino acid in higher plants and more than 90% of ingested
2011; Kalisinska et al., 2014). In our study, there was no correlation
SeMet is absorbed in the gastrointestinal tract in vertebrates. However,
between these elements in the roach however, we observed a signicant
in mammals, the physiological and biochemical importance of Se is
correlation in raccoon muscle (rS=0.64).
primarily achieved via the SeCys activity. It is important to note that
Values of Se:Hg molar ratios in freshwater sh vary, with a wide
the biological activity of Se is determined by its chemical form and the
range of 0.1268.3, depending on trophic and taxonomic position,
dose ingested (Fordyce, 2005; Zhang, 2014) (Table 3).
intraspecic variation connected with age, and abiotic factors such as
Our studies revealed that sh muscles contain almost two times
geographic location, local geochemical conditions, and the degree of
more Se than piscivores mammals, but concentrations did not dier
anthropogenic Hg and Se contamination (Peterson et al., 2009; Burger
between sh and the omnivorous raccoon. This is probably a result of
and Gochfeld, 2013). Using our own data and data from other studies,
lower digestibility of sh Se by piscivores in contrast to higher
we compared ratio values of roach in Europe. Muscle Se:Hg ratios were
absorption of plant Se by many sh and raccoons, having a diet with
always > 1, ranging from 1.848.6. In specimens from English rivers
signicant amounts of plants. In homoeothermic animals, 85100% of
with higher Se levels, ratios were 727 times higher than in other Hg
dietary plant Se is absorbed compared to 2050% of Se present in meat
areas in Se-decient areas. This can indicate that roach in England are
and sh (Fordyce, 2005). It cannot be excluded that evolutionary
less exposed to toxic inuence of Hg than these in Se-decient
changes have led to a lower demand for SeCys in terrestrial mammal
countries.

Table 3
Comparison of Se and THg concentrations (after calculation on mmol*) in muscle of fish and piscivorous mammals from different regions (ad, adults; F, female, M, male).

Species Site n ww or dw THg mmol Se mmol Se:THg References

Fish

Pike Esox lucius Poland 1 dw 4.94 12.68 2.57 This study

Norway 3 ww 1.30* 2.41* 1.86* (Julshamn et al., 1986)
Norway 47 ww 2.24* 2.03* 0.90* (Hartmann, 2014)
Asp Aspius aspius Poland 1 dw 6.98 9.89 1.42 This study

Ide Leuciscus idus Poland 1 dw 3.69 8.14 2.21 This study

Norway 5 ww 1.65* 1.14* 0.69* (Julshamn et al., 1986)

Roach Rutilus rutilus Poland 1 dw 1.29 8.45 6.56 This study

Finland 8 ww 0.60* 3.67* 6.14* Kari and Kauranen, 1978
Norway 14 ww 0.85* 1.52* 1.79* (Julshamn et al., 1986)
Norway 20 ww 2.79* 5.32* 1.91* (Myreng, 2013)
England 122 ww 0.03* 0.46* 48.58* (Jurgens, 2015)

Bream Abramis brama Poland 1 dw 2.57 7.40 2.89 This study

Norway 2 ww 0.95* 3.55* 3.74* (Julshamn et al., 1986)
White bream Blicca bjoerkna Poland 5 dw 2.57 7.40 2.89 This study
Crucian carp Carassius carassius Poland 1 dw 2.29 13.25 5.78 This study
Rudd Scardinius erythrophthalmus Poland 3 dw 1.83 6.85 3.74 This study
European eel Anguilla anguilla Poland 40 ww 1.04* 3.29* 3.16* (Polak-Juszczak and Robak, 2015)

Mammals
European otter Lutra lutra Poland 8 dw 18.36 4.34 0.27 This study
River otter Lontra canadensis Canada 4 ww 4.44* 2.53* 0.57* (Wren, 1984)
American mink Neovison vison Poland 7 ww 45.78 4.56 0.11 This study

Raccoon Procyon lotor Poland 37 dw 2.82 9.95 3.75 This study

Canada 4 ww 1.40* 2.53* 1.81* (Wren, 1984)
USA 11 ww 1.65* 1.39* 0.84* (Lord et al., 2002)
25 ww 0.10* 1.27* 12.66* (Burger et al., 2002)
American mink Neovison vison Poland 7 ww 45.78 4.56 0.11 This study

Saimaa ringed seal Phoca hispida siamensis Finland ad F 8 ww 5.98* 5.06* 0.85* (Lyytikinen et al., 2015)
ad M 6 10.97* 3.80* 0.35*

29
E. Kalisinska et al.
Considering that a tissue Se:Hg molar ratio < 1 is connected with
an increased Hg toxicity potential, it may be assumed that piscivores in
the WMNP are more polluted than raccoons because muscle Se:Hg
molar ratios were < 0.3 and > 3.7 in sh-eating and omnivorous
carnivore, respectively. Based on data collected by Wren (1984), we
calculated that in river otters and raccoons from Canada, muscle Se:Hg
ratios reached 0.57 and 1.81, respectively. These values, although
slightly dierent from our corresponding ones, may indicate that Se:Hg
molar ratios are improper in otters from the WMNP and Canada. Based
on the information on Se and Hg concentrations in muscles in one
raccoon group from South Carolina (USA) analyzed by Burger et al.
(2002) and Lord et al. (2002), we carried out a appropriate calculation.
In this group, muscle Se:Hg molar ratio reached 13 and was over thrice
higher than in raccoons from the WMNP. These results suggest greater
exposure of piscivores to the adverse impact of Hg than of omnivores,
even with a low degree of environmental Hg contamination. This can
be explained by higher Hg absorption rates and lower Se from the diet
by piscivores compared to omnivores, consuming a food containing
about 30% of sh and large amounts of easily digestible Se plant.
References
Akerblom, S., Bignert, A., Meili, M., Sonesten, L., Sundbom, M., 2014. Half a century of
changing mercury levels in Swedish freshwater sh. AMBIO 43, 91103.
AMAP/UNEP, 2013. Technical Background Report for the Global Mercury Assessment.
Arctic Monitoring and Assessment Programme, Oslo, Norway/UNEP Chemicals
Branch, Geneva, Switzerland, pp. 263.
Belzile, N., Che, Y.W., Gun, J.M., Ton, J., Alari, Y., Delonchamp, T., et al., 2006. The
eect of selenium on mercury assimilation by freshwater organisms. Can. J. Fish.
Aquat. Sci. 63, 110.
Bjerregaard, P., Fjordside, S., Hansen, M.G., Petrova, M.B., 2011. Dietary selenium
reduces retention of methylmercury in freshwater sh. Environ. Sci. Technol. 45,
97939798.
Burger, J., Gaines, K.F., Lord, C.G., Brisbin, I.L., Shukla, S., Gochfeld, M., 2002. Metal
levels in raccoon tissues: dierences on and o the department of Energy's Savannah
River site in South Carolina. Environ. Monit. Assess. 74, 6784.
Burger, J., Gochfeld, M., 2013. Selenium/mercury molar ratios in freshwater, marine,
and commercial sh from the USA: variation, risk, and health management. Rev.
Environ. Health 28, 129143.
Chen, L., Wang, H.H., Liu, J.F., Tong, Y.D., Ou, L.B., Zhang, W., et al., 2014.
Intercontinental transport and deposition patterns of atmospheric mercury from
anthropogenic emissions. Atm. Chem. Phys. 14, 1016310176.
Chumchal, M.M., Rainwater, T.R., Osborn, S.C., Roberts, A.P., Abel, M.T., Cobb, G.P.,
et al., 2011. Mercury speciation and biomagnication in the food web of Caddo Lake,
Texas and Louisiana, USA, a subtropical freshwater ecosystem. Environ. Toxicol.
Chem. 30, 11531162.
Commission Regulation, EC, 2006. Commission Regulation No 1881/2006/ setting
maximum levels for certain contaminants in foodstus. O.J. EU L364/5.
CSO, 2014. Central Statistical Oce. Environment, GUS, Warsaw, pp. 593.
Directive, 2008/105/EC. Directive on environmental quality standards in the eld of
water policy, O.J. EU L384/84.
EEA, 2014. European Union Emission Inventory Report 19902012 under the UNECE
Convention on Long-range Transboundary Air Pollution (LRTAP). European
Environment Agency, Luxembourg, 130.
Fordyce, F., 2005. Selenium deciency and toxicity in the environment. In: Selinus, O.,
Alloway, B., Centeno, J.A., Finkelman, R.B., Fuge, R., Lindh, U., Smedley, P. (Eds.),
Essentials of Medical Geology, Impacts of the Natural Environment on Public
Health. Elsevier Academic Press, Boston, San Diego, London, 373415.
Fortin, C., Beauchamp, G., Dansereau, M., Larivire, N., Blanger, D., 2001. Spatial
variation in mercury concentrations in wild mink and river otter carcasses from the
James bay territory, Quebec, Canada. Arch. Environ. Contam. Toxicol. 40, 121127.
Hartmann, S.S., 2014. Comparing Mercury (Hg) Concentration in Pike (Esox lucius)
from the Lakes Djupetjern, Holmetjern and Vistero in Southeast Norway: Eects of
Selenium (Se), Individual Growth and Water Chemistry (Master thesis). Norwegian
University of Life Sciences.
Hyvrinen, H., Tyni, P., Nieminen, P., 2003. Eects of moult, age, and sex on the
accumulation of heavy metals in the otter (Lutra lutra) in Finland. Bull. Environ.
Contam. Toxicol. 70, 278284.
Julshamn, K., Ringdal, O., Haugsnes, J., 1986. Minerals and trace elements in llets of
nine freshwater shes from Norway. Fisk. Dir. Skr. Ser. Ermring 2, 185191.
Jurgens, M.D., 2015. Biomonitoring of Wild Fish to Assess Chemical Pollution in English
Rivers - An Application of a Fish Tissue Archive (Ph.D. thesis). Lancaster
Environment Center, Lancaster University, Lancaster, UK, 338.
Jurgens, M.D., Johnson, A.C., Jones, K.C., Hughes, D., Lawlor, A.J., 2013. The presence
of EU priority substances mercury, hexachlorobenzene, hexachlorobutadiene and
PBDEs in wild sh from four English rivers. Sci. Total Environ. 461462, 441452.
30
Ecotoxicology and Environmental Safety 136 (2017) 2430
Kalisinska, E., Gorecki, J., Okonska, A., Pilarczyk, B., Tomza-Marciniak, A., Budis, H.,
et al., 2014. Mercury and selenium in the muscle of piscivorous common mergansers
(Mergus merganser) from a selenium-decient European country. Ecotoxicol.
Environ. Saf. 101, 107115.
Kalisinska, E., Budis, H., anocha, N., Podlasiska, J., Baraniewicz, E., 2012.
Comparison of hepatic and nephric concentrations of mercury between feral and
ranch American mink (Neovison vison) from NW Poland. Bull. Environ. Contam.
Toxicol. 88, 802806.
Kari, T., Kauranen, P., 1978. Mercury and selenium contents of seals from fresh and
brackish waters in nland. Bull. Environ. Contam. Toxicol. 19, 273280.
Lanocha, N., Kalisinska, E., Kosik-Bogacka, D.I., Budis, H., Podlasinska, J.,
Jedrzejewska, E., 2014. Mercury levels in raccoons (Procyon lotor) from the Warta
Mouth National Park, northwestern Poland. Biol. Trace Elem. Res. 159, 152160.
Lanszki, J., Sugr, L., Orosz, E., Nagy, D., 2008. Biological data from post mortem
analysis of otters in Hungary. Acta Zool. Acad. Sci. Hung. 54, 201212.
Lord, C.G., Gaines, K.F., Boring, C.S., Brisbin, I.L., Gochfeld, M., Burger, J., 2002.
Raccoon (Procyon lotor) as a bioindicator of mercury contamination at the U.S.
Department of Energy's Savannah River site. Arch. Environ. Contam. Toxicol. 43,
356363.
Lyytikinen, M., Ptynen, J., Hyvrinen, H., Sipil, T., Kunnasranta, M., 2015. Mercury
and selenium balance in endangered Saimaa ringed seal depend on age and sex.
Environ. Sci. Technol. 49, (11808-1116).
Mariotti, M., Ridge, P.G., Zhang, Y., Lobanov, A.V., Pringle, T.H., Guigo, R., et al., 2012.
Composition and evolution of the vertebrate and mammalian selenoproteomes. PLoS
One 7, e33066.
Mason, C.F., Last, N.I., Macdonald, S.M., 1986. Mercury, cadmium, and lead in British
otters. Bull. Environ. Contam. Toxicol. 37, 844849.
Muchlinski, M.N., Snodgrass, J.J., Terranova, C.J., 2012. Muscle mass scaling in
primates: an energetic and ecological perspective. Am. J. Primatol. 74, 395407.
Myreng, H., 2013. Bioaccumulation and Biomagnication of Mercury (Hg) to "At Risk
Levels" in the Fish Community in the Humic Lake vre Sandvannet, SE Norway.
Norwegian University of Life Sciences, Oslo, Norway, 68.
Nakazawa, E., Ikemoto, T., Hokura, A., Terada, Y., Kunito, T., Tanabe, S., et al., 2011.
The presence of mercury selenide in various tissues of the striped dolphin: evidence
from -XRF-XRD and XAFS analyses. Metallomics 3, 719725.
Nguetseng, R., Fliedner, A., Knopf, B., Lebreton, B., Quack, M., Rdel, H., 2015.
Retrospective monitoring of mercury in sh from selected European freshwater and
estuary sites. Chemosphere 134, 427434.
OConnor, D.J., Nielsen, S.W., 1981. Environmental survey of methylmercury levels in
wild mink (Mustela vison) and otter (Lutra canadensis) from Northeastern United
States and experimental pathology of methylmercurialism in the otter. In: Chapman
J.D., Pursley D. (eds), In: Proceedings of the World Furbearer Conference,
Frostburg, Maryland, pp. 17281745.
Penglase, S., Hamre, K., Ellingsen, S., 2015. The selenium content of SEPP1 versus
selenium requirements in vertebrates. Peer J 3, e1244.
Peterson, S.A., Ralston, N.V., Peck, D.V., Van Sickle, J., Robertson, J.D., Spate, V.L.,
et al., 2009. How might selenium moderate the toxic eects of mercury in stream sh
of the Western U.S.? Environ. Sci. Technol. 43, 39193925.
Pilarczyk, B., Tomza-Marciniak, A., Pilarczyk, R., Hendzel, D., Baszczyk, P., Bakowska,
M., 2011. Tissue distribution of selenium and eect of season and age on selenium
content in roe deer from northwestern Poland. Biol. Trace Elem. Res. 140, 299307.
Polak-Juszczak, L., Robak, S., 2015. Mercury toxicity and the protective role of selenium
in eel, Anguilla anguilla. Environ. Sci. Pollut. Res. Int. 22, 679688.
Ruiz-Olmo, J., Lafontaine, L., Prignioni, C., Lpez-Martin, J.M., Santos-Reis, M., 2000.
Pollution and its eects on otter populations in south-western Europe. In: Conroy J.
W.H., Yoxon P., Gutleb A.C.(eds) In: Proceedings of the First Otter Toxicology
Conference. Skye, J. Inter. Otter Surv. Fund 1, pp. 6382.
Salminen, R., 2005. FOREGS Geochemical Atlas of Europe. Part 1, Background
Information, Methodology and Maps. Geological Survey of Finland, Espoo.
Schfer, S., Buchmeier, G., Claus, E., Duester, L., Heininger, P., Krner, A., et al., 2015.
Bioaccumulation in aquatic systems: methodological approaches, monitoring and
assessment. Environ. Sci. Eur. 27, 5.
Scheuhammer, A.M., Meyer, M.W., Sandheinrich, M.B., Murray, M.W., 2007. Eects of
environmental methylmercury on health of wild birds, mammals, and sh. AMBIO
36, 1218.
Sleeman, J.M., Cristol, D.A., White, A.E., Evers, D.C., Gerhold, R.W., Keel, M.K., 2010.
Mercury poisoning in free-living northern river otter (Lontra canadensis). J. Wildl.
Dis. 46, 10351039.
Szkoda, J., Zmudzki, J., Nawrocka, A., Kmiecik, M., 2014. Toxic elements in free-living
freshwater sh, water and sediments. Bull. Vet. Inst. Pulawy 58, 589595.
Szkoda, J., Nawrocka, A., Kmiecik, M., mudzki, J., 2011. Monitoring study of toxic
elements in food of animal origin. Ochr. rod. Zas. Nat. 48, 475484, (In Polish).
US EPA, 2015. Draft Aquatic Life Ambient Water Quality Criterion for Selenium
Freshwater, US Environmental Protection Agency. Oce of Water. Oce of Science
and Technology, Washington, D.C., 740.
WHO, 2011. Evaluation of certain contaminants in food. WHO Technical Report Series,
No. 959.
Wren, C., 1984. Distribution of metals in tissues of beaver, raccoon and otter from
Ontario, Canada. Sci. Total Environ. 34, 177184.
Zhang, H., 2014. Impacts of selenium on the biogeochemical cycles of mercury in
terrestrial ecosystems in mercury mining areas. Springer-Verlag, Berlin Heidelberg,
193.