Environmental Toxicology and Chemistry, Vol. 34, No. 1, pp. 152–160, 2015 # 2014 SETAC Printed in the USA
SELENIUM TISSUE BURDEN COMPARTMENTALIZATION IN RESIDENT WHITE STURGEON (ACIPENSER TRANSMONTANUS) OF THE SAN FRANCISCO BAY DELTA ESTUARY JAVIER LINARES-CASENAVE,*yz R. LINVILLE,yx J.P. VAN EENENNAAM,y J.B. MUGUET,y** and S.I. DOROSHOVy yDepartment of Animal Science, University of California, Davis, California, USA zUS Fish & Wildlife Service, Pacific Southwest Regional Office, Sacramento, California xOffice of Environmental Health Hazard Assessment, California Environmental Protection Agency, Sacramento, California, USA (Submitted 21 January 2014; Returned for Revision 19 February 2014; Accepted 5 October 2014) Abstract: High selenium (Se) loads in the San Francisco Bay Delta are bioaccumulated and biomagnified in food webs and can impair the reproduction of resident oviparous animals such as white sturgeon. The objective of the present study was to determine the Se tissue burden in San Francisco Bay Delta–resident white sturgeon to assess Se bioaccumulation in different organs, including ovaries and liver where egg yolk precursor proteins are synthesized. The authors obtained 54 San Francisco Bay Delta–resident white sturgeon including 26 female and 28 male subadults with immature gonads, 8 females with vitellogenic eggs, and 13 males with maturing gonads. Length, weight, age, reproductive stage of development, and kidney, liver, gonad, and muscle Se concentrations were determined in all fish. Concentrations of Se in muscle, gonads, and liver significantly increased with fish size, whereas kidney Se was not correlated to body size and was at the highest level compared with other organs. There was no difference between the sexes (p > 0.05) in Se concentrations in kidney (12.83 0.51 mg g1 dry wt), liver (11.85 1.04 mg g1 dry wt), and muscle (7.09 0.52 mg g1 dry wt; mean standard error, n ¼ 47); but Se concentration was higher in the ovary than in testis (p ¼ 0.04). Females with vitellogenic eggs had higher Se concentrations in the ovaries (20.77 4.11 mg g1 dry wt vs 5.22 2.50 mg g1 dry wt), liver (21.84 2.07 mg g1 dry wt vs 8.03 1.03 mg g1 dry wt), and muscle (10.18 1.93 mg g1 dry wt vs 5.48 0.64 mg g1 dry wt) compared with less advanced, previtellogenic females (p < 0.05). The elevated Se concentrations in the ovaries and liver of vitellogenic San Francisco Bay Delta white sturgeon were comparable with levels previously shown to cause reproductive toxicity in dietary Se experiments with captive white sturgeon. Environ Toxicol Chem 2015;34:152–160. # 2014 SETAC Keywords: White sturgeon
Selenium
Tissue burden
San Francisco Bay Delta
Bioaccumulation
Biomagnification
exceeds the levels shown to cause toxicity in animals consuming Se-contaminated diets. Prey items containing greater than 10 mg g1 Se have been shown to induce Se toxicity in birds and fish [13–18]. White sturgeon sampled from the San Francisco Bay Delta between 1986 and 1990 contained Se at concentrations ranging from 9 mg g1 to 30 mg g1 dry weight in liver (n ¼ 52) and 7 mg g1 to 15 mg g1 dry weight in muscle (n ¼ 99) [4,9]. Concentrations of Se in white sturgeon in 1990 reached levels of up to 72 mg g1 dry weight in the ovaries, which have been previously linked to adverse effects in other fish [8,9,19,20]. Sturgeon may be more prone to Se bioaccumulation compared with other fish because of their long life span and benthic feeding habits. The unique reproductive biology of white sturgeon also increases the risk of transferring bioaccumulated Se to offspring. Sturgeon have a prolonged period (at least 2 yr [21]) of yolk deposition in their eggs, which is a mechanism of Se maternal transfer to offspring [8]. Transfer and storage of Se in the egg yolk compromise the development and survival of embryos and yolk-sac larvae [19,20,22–24] lacking efficient mechanisms and organs (gills, liver, and kidney) for Se detoxification and excretion. The Se-enriched yolk causes toxicity, developmental defects, and mortality of sturgeon embryos and yolk-sac larvae and, therefore, may affect sturgeon recruitment [24]. To mitigate the Se pollution in the San Francisco Bay Delta, the California Regional Water Quality Control Board required oil refineries in the San Francisco Bay to reduce the amount of Se in effluents discharged into the bay; but refinery effluent continues to be a significant source of Se in this region [25]. The annual Se input loads from agricultural drainage into the San Joaquín River remain within historical levels after the US
INTRODUCTION
Selenium (Se) is an essential element that can be toxic at concentrations above nutritional requirements. Selenium threatens higher–trophic level species because of its efficient food-web transfer [1,2]. The San Francisco Bay Delta receives Se in agricultural drainage from the Central Valley and in effluents from oil refineries. White sturgeon (Acipenser transmontanus) in San Francisco Bay are exposed to high levels of Se through their diet, evidenced by high Se levels in the common prey of white sturgeon [3–5], as well as in sturgeon muscle, liver, and eggs [4,6–9]. Risebrough et al. [10] reported concentrations of 8 mg g1 to 11 mg g1 Se dry weight in transplanted bivalves in northern San Francisco Bay. Johns and Luoma [3] reported an Se concentration of 6 3 mg g1 dry weight in resident bivalves Corbicula spp. sampled near Carquinez Strait. The Se Verification Study found Se concentrations of 5.13 mg g1 to 7.90 mg g1 dry weight in Corbicula spp. from the north bay between 1987 and 1990 [4,7,9]. The filter feeding exotic bivalve Corbula amurensis contains an average Se level of 15 mg g1 dry weight [5]. This nonnative species was introduced into San Francisco Bay in the mid-1980s and has become the dominant bivalve in the bay [11,12] and a major food source of white sturgeon [9]. The high Se level and wide distribution of this mollusk is of great concern because its Se burden significantly All Supplemental Data may be found in the online version of this article. * Address correspondence to [email protected] **Deceased Published online 15 October 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/etc.2775 152
Tissue Se in resident white sturgeon of San Francisco Bay
Bureau of Reclamation implemented the Grassland Bypass Project in 1996 and represent a significant source of Se entering the San Francisco Bay Delta [25]. Therefore, Se contamination is a high priority for the management of the San Francisco Bay Delta ecosystem. White sturgeon is 1 of the 2 endemic sturgeons inhabiting the North American Pacific coast and major river systems. Both white and green (A. medirostris) sturgeon have high zoological value and are important to biodiversity, but white sturgeon has a higher value as a food and game fish. Both species belong to a phylogenetically unique group of ancient ray-finned fish that has survived for over 200 million years but has experienced rapid decline over the past century [26]. White sturgeon are more common in the San Francisco Bay Delta, aggregating in northern San Francisco Bay (Suisun and San Pablo Bays) and migrating into Central Valley Sacramento and San Joaquín River systems to spawn [27,28]. We hypothesized that the long-lived white sturgeon residing in the San Francisco Bay Delta exhibit increasing Se tissue burden as they age. In the present study we determine the age and Se tissue burden of San Francisco Bay Delta–resident white sturgeon. MATERIALS AND METHODS
A total of 54 fish were used in the present study, which included the same data set (n ¼ 46 fish) used in our 2004 CALFED report [29] and 8 additional fish caught from 17 April to 11 May 2005, that is, after the 2004 CALFED had been finalized (Supplemental Data, Table S1). In both the present study and the 2004 CALFED report, the same 7 fish were excluded from Se analysis (Supplemental Data, Table S1); thus, the numbers of fish used in Se concentration analyses were 47 and 39 for the present study and the 2004 CALFED report, respectively. Wild sturgeon capture in the San Francisco Bay Delta
A total of 54 fish were used in the present study. Most wild sturgeon were obtained from the California Department of Fish and Wildlife’s study of set-line capture efficiency in Suisun Bay on 22 and 23 July 2003 (n ¼ 20) and a CALFED-funded green sturgeon radio tagging study where white sturgeon were caught as bycatch in San Pablo Bay from 21 April to 11 May 2005 (n ¼ 23). The gill nets used in the latter study were 19.1 cm stretch-bar mesh, 91.4 m long and 6.1 m deep, or 23.5 cm stretch-bar mesh, 182.9 m long and 3.7 m deep; and fish were caught primarily between the mouths of the Sonoma Creek and the Petaluma River, approximately 3.2 km to 4.0 km offshore in water 1.2 m to 3.0 m deep. Three other sources of fish were the Bureau of Reclamation Fish Collection Facility in Tracy from January 2003 to January 2004 (n ¼ 6), the fish treadmill project at the University of California–Davis from 17 April to 7 May 2004 (n ¼ 2), and the California Department of Fish and Wildlife tagging study in San Pablo Bay on 31 October 2002 (n ¼ 3). A total of 7 fish (6 from the Bureau of Reclamation Fish Collection Facility and 1 from the treadmill project at University of California–Davis) were excluded from the Se analysis because they were held 1 mo to 11 mo before tissue was collected, but these fish were used for length age comparisons. Capture locations for fish used in the tissue Se analysis are shown in Figure 1. Tissue sampling and Se analysis
Fish were anesthetized (100 mg L1 MS-222 in the water) and then euthanized by increasing the MS-222 concentration in the water to 500 mg L1. We recorded fork and total lengths and live weight. The pectoral fins’ leading ray was removed for aging.
Environ Toxicol Chem 34, 2015
153
Figure 1. Locations of white sturgeon captures for the Se study. 1 ¼ north San Pablo Bay 3.2 km to 4.0 km offshore (n ¼ 23); 2 ¼ northeast San Pablo Bay (n ¼ 3); 3 ¼ Suisun Bay cutoff (n ¼ 1); 4–6 ¼ Honker Bay, Channel Island, and Broad Slough North, respectively (n ¼ 20).
Gonads were dissected out and weighed, and a section was fixed in 10% buffered formalin for histology. Tissue sections of gonad, liver, kidney, and muscle for Se measurements were collected, frozen on dry ice, and stored at –80 8C. A section of each tissue was removed for the determination of moisture content. Tissue samples were homogenized in Millipure water and lyophilized, and Se content was determined using inductively coupled plasma–atomic emission spectroscopy (protocol 8137) at the California Animal Health and Food Safety Laboratory System, School of Veterinary Medicine, University of California–Davis. Assay minimum detection limit was 5 mg kg1, and quality control was achieved by including standard reference material samples, spiked samples, and blanks using duplication in all analytical runs. Throughout the present study the Se concentration is expressed as micrograms per gram dry weight. Age determination
Age was determined from cross sections (0.2 mm–0.6 mm) of proximal portions of dried pectoral fin leading ray [30]. Sections were polished with increasing grades of wet–dry sandpaper and mounted on microscope slides with a coverslip adhesive. Slides were examined under a dissecting microscope with a camera lucida and a Nikon Microplan II digital image analyzer with a fine-point light cursor to count the annuli and trace their continuity throughout the section plane. In counting annuli we followed the guidelines of Cuerrier [31], Probst and Cooper [32], Semakula and Larkin [33], Sokolov and Akimova [34], and Brennan and Cailliet [35]. Histology
Histological preparations of gonadal tissue were obtained from fixed tissue as follows. Tissue was dehydrated in a series of alcohol, cleared in xylene, and infiltrated with paraffin in an automated tissue processor (Tissue-Tek VIP; Sakura Finetex). Gonad tissue was embedded in paraffin, sectioned at 5 mm, and stained by periodic acid Schiff stain (PAS) and hematoxylin and eosin [36]. Histological preparations were microscopically examined (Olympus BHS). Females were classified by examination of histological preparations of gonad tissue as previtellogenic or vitellogenic based on the most advanced clutch of ovarian follicles present (Figure 2). Follicles of previtellogenic females were in the endogenous growth phase and exhibited either small ( 0.10, t ¼1.542, t0.05 a(2), 41df ¼ 2.02) and elevation (p > 0.20, t ¼ 1.158, t0.05 a(2), 41df ¼ 2.02) of total length and age regression lines between the sexes. Figure 6 shows the results of 2-way ANOVA on the effect of total length and sex of fish on Se concentration of gonad, liver, and kidney and Kruskal-Wallis 1-way ANOVA on the effect of total length of fish on Se concentration in muscle. Comparisons between the sexes within each fish size class on the mean muscle Se concentration were done using Mann-Whitney rank sum test. The size and sex of fish had a significant effect on gonad Se concentration (p < 0.001). The size of fish also had a significant effect on muscle (p < 0.001) and liver (p < 0.001) Se concentrations (Figure 6). Larger fish (size class III) had significantly higher muscle and liver Se loads than smaller fish. No significant interaction between sex and size of fish was detected by the ANOVA for Se concentration of gonad (p ¼ 0.293), liver (p ¼ 0.319), and kidney (p ¼ 0.995). The Se concentration in gonads, muscle, and liver significantly increased with the size of fish for both sexes. The increase in gonadal tissue Se burden with size of fish was significantly more accentuated in females than in males (pairwise comparison of mean concentration of Se between sexes within each fish size class revealed a significant
100 60
50 40
W = 0.39 * (Age1.439) n = 19, R2 = 0.887
30
W
156
20 10 0 5
10
15
20
25
Age (yr)
Figure 5. Scatter plots and linear regressions of male length (fork and total) and weight on age. FL ¼ fork length; TL ¼ total length; W ¼ weight.
difference in gonad tissue in size class II, p < 0.001, and size class III, p ¼ 0.002; indicated by asterisk in Figure 6). Gonad tissue exhibited very high variation in Se concentration compared to other tissues. Small females exhibited very low Se concentration relative to larger females, which exhibited large variation and one of the highest Se concentrations measured (46.7 mg g1, gonad tissue of a previtellogenic female) in the present study. The Se burden in kidney tissue was similar in all 3 size classes of fish. Kidney, relative to other tissues, had elevated concentrations of Se, especially in smaller fish. The effect of sexual maturity on Se concentration in different tissues for each sex is illustrated on Figure 7. The mean gonad, liver, and muscle Se concentrations were significantly higher in vitellogenic females than in previtellogenic females (for gonads the mean standard error Se concentration was 5.22 2.50 mg g1 for previtellogenic and 20.77 4.11 mg g1 for vitellogenic females, whereas those for the liver and muscle were 8.03 1.03 mg g1 and 21.84 2.07 mg g1, and 5.48 0.64 mg g1 and 10.18 1.93 mg g1, respectively). Although vitellogenic females had slightly higher mean kidney Se concentration than previtellogenic females, this difference was not statistically significant (for the kidney Se load was 12.03 0.95 mg g1 and 14.71 1.45 mg g1). For males, the mean gonad Se concentration was significantly higher in the spermatogenesis group (3.73 1.04 mg g1) than in the early gonial group (1.33 0.33 mg g1), indicated by an asterisk in Figure 7. The mean liver, muscle, and kidney Se concentrations were not significantly different between males at different maturity stages (for liver the mean standard error Se concentration was 10.18 2.68 mg g1 for the early gonial group and 12.01 1.34 mg g1 for the spermatogenesis group, for muscle it was
Tissue Se in resident white sturgeon of San Francisco Bay
GONAD Gonad
B
*
*
40
MUSCLE Muscle
25
B
20
Selenium (µg·g-1)
Selenium (µg·g-1)
25
30
A
A 15
10
B
20
A 15
10
5
5
0
0
Selenium (µg·g-1)
30
Environ Toxicol Chem 34, 2015 Pre-Vitellogenic (18) Pre-vitellogenic
157
Vitellogenic Vitellogenic(8)(8)
30
*
* 20
* 10
0
II
III 30
LIVER Liver
B A
20
15
II
A
10
Earlygonial gonial(8) (8) Early
15
10
5
0
0
LIVER Liver
KIDNEY Kidney
40
KIDNEY Kidney
20
5
MUSCLE Muscle
III
25
Selenium (µg·g-1)
Selenium (µg·g-1)
25
GONAD Gonad I
Selenium (µg·g-1)
30
I
Spermatogenesis (13) Spermatogenesis (13)
30
20
10
*
0
I
II
III
GONAD I
II
(4) (13) (10) (4) (7) (9)
Figure 6. Selenium concentration (mean standard error) of male (solid bars) and female (open bars) white sturgeon in gonad (top left), kidney (top right), liver (bottom left), and muscle (bottom right). Different letters indicate significantly different means between size classes (2-way analysis of variance [ANOVA] for gonad, liver, and kidney and Kruskal-Wallis 1-way ANOVA on ranks for muscle), and asterisk indicates significant difference between sexes within each size class (p < 0.05). Numbers in parentheses under liver graph’s x-axis indicate sample size.
6.35 1.13 mg g1 for the early gonial group and 7.87 0.58 mg g1 for the spermatogenesis group, and for kidney it was 11.87 1.08 mg g1 for the early gonial group and 13.36 0.63 mg g1 for the spermatogenesis group). Table 1 shows the average Se concentration in gonad, kidney, liver, and muscle in male (n ¼ 21) and female (n ¼ 26) white sturgeon. Mean gonad Se concentration was significantly higher in females than in males, but the average kidney, liver, and muscle Se concentrations were not significantly different between the sexes. In Tables 2 and 3, we show the tissue Se means of each sex for fish captured either in San Pablo Bay (capture sites 1 and 2 in Figure 1) or in Suisun Bay (capture sites 3–5 in Figure 1). Females and males captured in San Pablo Bay had significantly higher gonad, liver, and muscle Se concentrations than those captured in Suisun Bay. DISCUSSION
The main goal of the present study was to assess Se bioaccumulation in various tissues of different age white sturgeon caught in the San Francisco Bay Delta. White sturgeon is a highly valuable native species of the San Francisco Bay Delta, playing important ecosystem roles and supporting recreational fishery. As a semianadromous, long-lived, late maturing, and benthophagous species, white sturgeon has a propensity to accumulate more Se via its food chain compared to other fish of the San Francisco Bay Delta ecosystem. The present study revealed mean Se
MUSCLE
Gonad
III
Muscle
LIVER Liver
KIDNEY Kidney
Figure 7. Selenium concentration (mean standard error) in gonad, muscle, liver, and kidney of female (top graph) and male (bottom graph) white sturgeon with gonads in different stages of the reproductive cycle. Asterisk indicates significant difference (t test or Mann-Whitney rank sum test) between gonadal maturity stage within each tissue (p < 0.05). Table 1. Mean standard error of tissue selenium concentration (mg g1) for female (n ¼ 26) and male (n ¼ 21) white sturgeona
Gonad Kidney Liver Muscle
Males
Females
p
2.89 0.72 12.79 0.57 11.31 1.29 7.29 0.57
10.00 2.54 12.86 0.81 12.28 1.58 6.93 0.84
0.035 0.799 0.676 0.340
a The difference in the mean tissue selenium between the sexes was tested by Mann-Whitney rank sum test (gonad and liver) and t test (kidney and muscle).
Table 2. Comparison (t test) of selenium tissue burden (mean standard error) between male white sturgeon from San Pablo (Figure 1, sites 1 and 2) and Suisun (Figure 1, sites 3–5) Baysa Selenium (mg g1)
Gonad Kidney Liver Muscle
San Pablo Bay (n ¼ 15)
Suisin Bay (n ¼ 6)
p
3.69 0.96 12.86 0.71 13.18 1.54 8.12 0.63
1.02 0.17 12.63 1.02 6.65 0.78 5.23 0.74
0.002 0.881 0.005 0.014
a Number in parentheses indicates fish sample size. Significance level p < 0.05.
concentrations of 12.83 0.51 mg g1, 11.85 1.04 mg g1, 7.09 0.52 mg g1, and 6.91 1.51 mg g1 (mean standard error) in the kidney, liver, muscle, and immature gonads of San Francisco Bay Delta–resident white sturgeon. Levels of Se in 4-yr-old to 8-yr-old fish were lower than levels in older fish.
158
Environ Toxicol Chem 34, 2015
J. Linares-Casenave et al.
Table 3. Comparison of selenium tissue burden (mean standard error) between female white sturgeon from San Pablo (Figure 1, sites 1 and 2) and Suisun (Figure 1, sites 3–5) Baysa Selenium (mg g1)
Gonad Kidney Liver Muscle
San Pablo Bay (n ¼ 11)
Suisin Bay (n ¼ 15)
p
16.90 3.59 13.75 1.29 18.04 2.49 9.26 1.61
4.95 2.99 12.20 1.05 8.05 1.21 5.21 0.56
SELENIUM TISSUE BURDEN COMPARTMENTALIZATION IN RESIDENT WHITE STURGEON (ACIPENSER TRANSMONTANUS) OF THE SAN FRANCISCO BAY DELTA ESTUARY JAVIER LINARES-CASENAVE,*yz R. LINVILLE,yx J.P. VAN EENENNAAM,y J.B. MUGUET,y** and S.I. DOROSHOVy yDepartment of Animal Science, University of California, Davis, California, USA zUS Fish & Wildlife Service, Pacific Southwest Regional Office, Sacramento, California xOffice of Environmental Health Hazard Assessment, California Environmental Protection Agency, Sacramento, California, USA (Submitted 21 January 2014; Returned for Revision 19 February 2014; Accepted 5 October 2014) Abstract: High selenium (Se) loads in the San Francisco Bay Delta are bioaccumulated and biomagnified in food webs and can impair the reproduction of resident oviparous animals such as white sturgeon. The objective of the present study was to determine the Se tissue burden in San Francisco Bay Delta–resident white sturgeon to assess Se bioaccumulation in different organs, including ovaries and liver where egg yolk precursor proteins are synthesized. The authors obtained 54 San Francisco Bay Delta–resident white sturgeon including 26 female and 28 male subadults with immature gonads, 8 females with vitellogenic eggs, and 13 males with maturing gonads. Length, weight, age, reproductive stage of development, and kidney, liver, gonad, and muscle Se concentrations were determined in all fish. Concentrations of Se in muscle, gonads, and liver significantly increased with fish size, whereas kidney Se was not correlated to body size and was at the highest level compared with other organs. There was no difference between the sexes (p > 0.05) in Se concentrations in kidney (12.83 0.51 mg g1 dry wt), liver (11.85 1.04 mg g1 dry wt), and muscle (7.09 0.52 mg g1 dry wt; mean standard error, n ¼ 47); but Se concentration was higher in the ovary than in testis (p ¼ 0.04). Females with vitellogenic eggs had higher Se concentrations in the ovaries (20.77 4.11 mg g1 dry wt vs 5.22 2.50 mg g1 dry wt), liver (21.84 2.07 mg g1 dry wt vs 8.03 1.03 mg g1 dry wt), and muscle (10.18 1.93 mg g1 dry wt vs 5.48 0.64 mg g1 dry wt) compared with less advanced, previtellogenic females (p < 0.05). The elevated Se concentrations in the ovaries and liver of vitellogenic San Francisco Bay Delta white sturgeon were comparable with levels previously shown to cause reproductive toxicity in dietary Se experiments with captive white sturgeon. Environ Toxicol Chem 2015;34:152–160. # 2014 SETAC Keywords: White sturgeon
Selenium
Tissue burden
San Francisco Bay Delta
Bioaccumulation
Biomagnification
exceeds the levels shown to cause toxicity in animals consuming Se-contaminated diets. Prey items containing greater than 10 mg g1 Se have been shown to induce Se toxicity in birds and fish [13–18]. White sturgeon sampled from the San Francisco Bay Delta between 1986 and 1990 contained Se at concentrations ranging from 9 mg g1 to 30 mg g1 dry weight in liver (n ¼ 52) and 7 mg g1 to 15 mg g1 dry weight in muscle (n ¼ 99) [4,9]. Concentrations of Se in white sturgeon in 1990 reached levels of up to 72 mg g1 dry weight in the ovaries, which have been previously linked to adverse effects in other fish [8,9,19,20]. Sturgeon may be more prone to Se bioaccumulation compared with other fish because of their long life span and benthic feeding habits. The unique reproductive biology of white sturgeon also increases the risk of transferring bioaccumulated Se to offspring. Sturgeon have a prolonged period (at least 2 yr [21]) of yolk deposition in their eggs, which is a mechanism of Se maternal transfer to offspring [8]. Transfer and storage of Se in the egg yolk compromise the development and survival of embryos and yolk-sac larvae [19,20,22–24] lacking efficient mechanisms and organs (gills, liver, and kidney) for Se detoxification and excretion. The Se-enriched yolk causes toxicity, developmental defects, and mortality of sturgeon embryos and yolk-sac larvae and, therefore, may affect sturgeon recruitment [24]. To mitigate the Se pollution in the San Francisco Bay Delta, the California Regional Water Quality Control Board required oil refineries in the San Francisco Bay to reduce the amount of Se in effluents discharged into the bay; but refinery effluent continues to be a significant source of Se in this region [25]. The annual Se input loads from agricultural drainage into the San Joaquín River remain within historical levels after the US
INTRODUCTION
Selenium (Se) is an essential element that can be toxic at concentrations above nutritional requirements. Selenium threatens higher–trophic level species because of its efficient food-web transfer [1,2]. The San Francisco Bay Delta receives Se in agricultural drainage from the Central Valley and in effluents from oil refineries. White sturgeon (Acipenser transmontanus) in San Francisco Bay are exposed to high levels of Se through their diet, evidenced by high Se levels in the common prey of white sturgeon [3–5], as well as in sturgeon muscle, liver, and eggs [4,6–9]. Risebrough et al. [10] reported concentrations of 8 mg g1 to 11 mg g1 Se dry weight in transplanted bivalves in northern San Francisco Bay. Johns and Luoma [3] reported an Se concentration of 6 3 mg g1 dry weight in resident bivalves Corbicula spp. sampled near Carquinez Strait. The Se Verification Study found Se concentrations of 5.13 mg g1 to 7.90 mg g1 dry weight in Corbicula spp. from the north bay between 1987 and 1990 [4,7,9]. The filter feeding exotic bivalve Corbula amurensis contains an average Se level of 15 mg g1 dry weight [5]. This nonnative species was introduced into San Francisco Bay in the mid-1980s and has become the dominant bivalve in the bay [11,12] and a major food source of white sturgeon [9]. The high Se level and wide distribution of this mollusk is of great concern because its Se burden significantly All Supplemental Data may be found in the online version of this article. * Address correspondence to [email protected] **Deceased Published online 15 October 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/etc.2775 152
Tissue Se in resident white sturgeon of San Francisco Bay
Bureau of Reclamation implemented the Grassland Bypass Project in 1996 and represent a significant source of Se entering the San Francisco Bay Delta [25]. Therefore, Se contamination is a high priority for the management of the San Francisco Bay Delta ecosystem. White sturgeon is 1 of the 2 endemic sturgeons inhabiting the North American Pacific coast and major river systems. Both white and green (A. medirostris) sturgeon have high zoological value and are important to biodiversity, but white sturgeon has a higher value as a food and game fish. Both species belong to a phylogenetically unique group of ancient ray-finned fish that has survived for over 200 million years but has experienced rapid decline over the past century [26]. White sturgeon are more common in the San Francisco Bay Delta, aggregating in northern San Francisco Bay (Suisun and San Pablo Bays) and migrating into Central Valley Sacramento and San Joaquín River systems to spawn [27,28]. We hypothesized that the long-lived white sturgeon residing in the San Francisco Bay Delta exhibit increasing Se tissue burden as they age. In the present study we determine the age and Se tissue burden of San Francisco Bay Delta–resident white sturgeon. MATERIALS AND METHODS
A total of 54 fish were used in the present study, which included the same data set (n ¼ 46 fish) used in our 2004 CALFED report [29] and 8 additional fish caught from 17 April to 11 May 2005, that is, after the 2004 CALFED had been finalized (Supplemental Data, Table S1). In both the present study and the 2004 CALFED report, the same 7 fish were excluded from Se analysis (Supplemental Data, Table S1); thus, the numbers of fish used in Se concentration analyses were 47 and 39 for the present study and the 2004 CALFED report, respectively. Wild sturgeon capture in the San Francisco Bay Delta
A total of 54 fish were used in the present study. Most wild sturgeon were obtained from the California Department of Fish and Wildlife’s study of set-line capture efficiency in Suisun Bay on 22 and 23 July 2003 (n ¼ 20) and a CALFED-funded green sturgeon radio tagging study where white sturgeon were caught as bycatch in San Pablo Bay from 21 April to 11 May 2005 (n ¼ 23). The gill nets used in the latter study were 19.1 cm stretch-bar mesh, 91.4 m long and 6.1 m deep, or 23.5 cm stretch-bar mesh, 182.9 m long and 3.7 m deep; and fish were caught primarily between the mouths of the Sonoma Creek and the Petaluma River, approximately 3.2 km to 4.0 km offshore in water 1.2 m to 3.0 m deep. Three other sources of fish were the Bureau of Reclamation Fish Collection Facility in Tracy from January 2003 to January 2004 (n ¼ 6), the fish treadmill project at the University of California–Davis from 17 April to 7 May 2004 (n ¼ 2), and the California Department of Fish and Wildlife tagging study in San Pablo Bay on 31 October 2002 (n ¼ 3). A total of 7 fish (6 from the Bureau of Reclamation Fish Collection Facility and 1 from the treadmill project at University of California–Davis) were excluded from the Se analysis because they were held 1 mo to 11 mo before tissue was collected, but these fish were used for length age comparisons. Capture locations for fish used in the tissue Se analysis are shown in Figure 1. Tissue sampling and Se analysis
Fish were anesthetized (100 mg L1 MS-222 in the water) and then euthanized by increasing the MS-222 concentration in the water to 500 mg L1. We recorded fork and total lengths and live weight. The pectoral fins’ leading ray was removed for aging.
Environ Toxicol Chem 34, 2015
153
Figure 1. Locations of white sturgeon captures for the Se study. 1 ¼ north San Pablo Bay 3.2 km to 4.0 km offshore (n ¼ 23); 2 ¼ northeast San Pablo Bay (n ¼ 3); 3 ¼ Suisun Bay cutoff (n ¼ 1); 4–6 ¼ Honker Bay, Channel Island, and Broad Slough North, respectively (n ¼ 20).
Gonads were dissected out and weighed, and a section was fixed in 10% buffered formalin for histology. Tissue sections of gonad, liver, kidney, and muscle for Se measurements were collected, frozen on dry ice, and stored at –80 8C. A section of each tissue was removed for the determination of moisture content. Tissue samples were homogenized in Millipure water and lyophilized, and Se content was determined using inductively coupled plasma–atomic emission spectroscopy (protocol 8137) at the California Animal Health and Food Safety Laboratory System, School of Veterinary Medicine, University of California–Davis. Assay minimum detection limit was 5 mg kg1, and quality control was achieved by including standard reference material samples, spiked samples, and blanks using duplication in all analytical runs. Throughout the present study the Se concentration is expressed as micrograms per gram dry weight. Age determination
Age was determined from cross sections (0.2 mm–0.6 mm) of proximal portions of dried pectoral fin leading ray [30]. Sections were polished with increasing grades of wet–dry sandpaper and mounted on microscope slides with a coverslip adhesive. Slides were examined under a dissecting microscope with a camera lucida and a Nikon Microplan II digital image analyzer with a fine-point light cursor to count the annuli and trace their continuity throughout the section plane. In counting annuli we followed the guidelines of Cuerrier [31], Probst and Cooper [32], Semakula and Larkin [33], Sokolov and Akimova [34], and Brennan and Cailliet [35]. Histology
Histological preparations of gonadal tissue were obtained from fixed tissue as follows. Tissue was dehydrated in a series of alcohol, cleared in xylene, and infiltrated with paraffin in an automated tissue processor (Tissue-Tek VIP; Sakura Finetex). Gonad tissue was embedded in paraffin, sectioned at 5 mm, and stained by periodic acid Schiff stain (PAS) and hematoxylin and eosin [36]. Histological preparations were microscopically examined (Olympus BHS). Females were classified by examination of histological preparations of gonad tissue as previtellogenic or vitellogenic based on the most advanced clutch of ovarian follicles present (Figure 2). Follicles of previtellogenic females were in the endogenous growth phase and exhibited either small ( 0.10, t ¼1.542, t0.05 a(2), 41df ¼ 2.02) and elevation (p > 0.20, t ¼ 1.158, t0.05 a(2), 41df ¼ 2.02) of total length and age regression lines between the sexes. Figure 6 shows the results of 2-way ANOVA on the effect of total length and sex of fish on Se concentration of gonad, liver, and kidney and Kruskal-Wallis 1-way ANOVA on the effect of total length of fish on Se concentration in muscle. Comparisons between the sexes within each fish size class on the mean muscle Se concentration were done using Mann-Whitney rank sum test. The size and sex of fish had a significant effect on gonad Se concentration (p < 0.001). The size of fish also had a significant effect on muscle (p < 0.001) and liver (p < 0.001) Se concentrations (Figure 6). Larger fish (size class III) had significantly higher muscle and liver Se loads than smaller fish. No significant interaction between sex and size of fish was detected by the ANOVA for Se concentration of gonad (p ¼ 0.293), liver (p ¼ 0.319), and kidney (p ¼ 0.995). The Se concentration in gonads, muscle, and liver significantly increased with the size of fish for both sexes. The increase in gonadal tissue Se burden with size of fish was significantly more accentuated in females than in males (pairwise comparison of mean concentration of Se between sexes within each fish size class revealed a significant
100 60
50 40
W = 0.39 * (Age1.439) n = 19, R2 = 0.887
30
W
156
20 10 0 5
10
15
20
25
Age (yr)
Figure 5. Scatter plots and linear regressions of male length (fork and total) and weight on age. FL ¼ fork length; TL ¼ total length; W ¼ weight.
difference in gonad tissue in size class II, p < 0.001, and size class III, p ¼ 0.002; indicated by asterisk in Figure 6). Gonad tissue exhibited very high variation in Se concentration compared to other tissues. Small females exhibited very low Se concentration relative to larger females, which exhibited large variation and one of the highest Se concentrations measured (46.7 mg g1, gonad tissue of a previtellogenic female) in the present study. The Se burden in kidney tissue was similar in all 3 size classes of fish. Kidney, relative to other tissues, had elevated concentrations of Se, especially in smaller fish. The effect of sexual maturity on Se concentration in different tissues for each sex is illustrated on Figure 7. The mean gonad, liver, and muscle Se concentrations were significantly higher in vitellogenic females than in previtellogenic females (for gonads the mean standard error Se concentration was 5.22 2.50 mg g1 for previtellogenic and 20.77 4.11 mg g1 for vitellogenic females, whereas those for the liver and muscle were 8.03 1.03 mg g1 and 21.84 2.07 mg g1, and 5.48 0.64 mg g1 and 10.18 1.93 mg g1, respectively). Although vitellogenic females had slightly higher mean kidney Se concentration than previtellogenic females, this difference was not statistically significant (for the kidney Se load was 12.03 0.95 mg g1 and 14.71 1.45 mg g1). For males, the mean gonad Se concentration was significantly higher in the spermatogenesis group (3.73 1.04 mg g1) than in the early gonial group (1.33 0.33 mg g1), indicated by an asterisk in Figure 7. The mean liver, muscle, and kidney Se concentrations were not significantly different between males at different maturity stages (for liver the mean standard error Se concentration was 10.18 2.68 mg g1 for the early gonial group and 12.01 1.34 mg g1 for the spermatogenesis group, for muscle it was
Tissue Se in resident white sturgeon of San Francisco Bay
GONAD Gonad
B
*
*
40
MUSCLE Muscle
25
B
20
Selenium (µg·g-1)
Selenium (µg·g-1)
25
30
A
A 15
10
B
20
A 15
10
5
5
0
0
Selenium (µg·g-1)
30
Environ Toxicol Chem 34, 2015 Pre-Vitellogenic (18) Pre-vitellogenic
157
Vitellogenic Vitellogenic(8)(8)
30
*
* 20
* 10
0
II
III 30
LIVER Liver
B A
20
15
II
A
10
Earlygonial gonial(8) (8) Early
15
10
5
0
0
LIVER Liver
KIDNEY Kidney
40
KIDNEY Kidney
20
5
MUSCLE Muscle
III
25
Selenium (µg·g-1)
Selenium (µg·g-1)
25
GONAD Gonad I
Selenium (µg·g-1)
30
I
Spermatogenesis (13) Spermatogenesis (13)
30
20
10
*
0
I
II
III
GONAD I
II
(4) (13) (10) (4) (7) (9)
Figure 6. Selenium concentration (mean standard error) of male (solid bars) and female (open bars) white sturgeon in gonad (top left), kidney (top right), liver (bottom left), and muscle (bottom right). Different letters indicate significantly different means between size classes (2-way analysis of variance [ANOVA] for gonad, liver, and kidney and Kruskal-Wallis 1-way ANOVA on ranks for muscle), and asterisk indicates significant difference between sexes within each size class (p < 0.05). Numbers in parentheses under liver graph’s x-axis indicate sample size.
6.35 1.13 mg g1 for the early gonial group and 7.87 0.58 mg g1 for the spermatogenesis group, and for kidney it was 11.87 1.08 mg g1 for the early gonial group and 13.36 0.63 mg g1 for the spermatogenesis group). Table 1 shows the average Se concentration in gonad, kidney, liver, and muscle in male (n ¼ 21) and female (n ¼ 26) white sturgeon. Mean gonad Se concentration was significantly higher in females than in males, but the average kidney, liver, and muscle Se concentrations were not significantly different between the sexes. In Tables 2 and 3, we show the tissue Se means of each sex for fish captured either in San Pablo Bay (capture sites 1 and 2 in Figure 1) or in Suisun Bay (capture sites 3–5 in Figure 1). Females and males captured in San Pablo Bay had significantly higher gonad, liver, and muscle Se concentrations than those captured in Suisun Bay. DISCUSSION
The main goal of the present study was to assess Se bioaccumulation in various tissues of different age white sturgeon caught in the San Francisco Bay Delta. White sturgeon is a highly valuable native species of the San Francisco Bay Delta, playing important ecosystem roles and supporting recreational fishery. As a semianadromous, long-lived, late maturing, and benthophagous species, white sturgeon has a propensity to accumulate more Se via its food chain compared to other fish of the San Francisco Bay Delta ecosystem. The present study revealed mean Se
MUSCLE
Gonad
III
Muscle
LIVER Liver
KIDNEY Kidney
Figure 7. Selenium concentration (mean standard error) in gonad, muscle, liver, and kidney of female (top graph) and male (bottom graph) white sturgeon with gonads in different stages of the reproductive cycle. Asterisk indicates significant difference (t test or Mann-Whitney rank sum test) between gonadal maturity stage within each tissue (p < 0.05). Table 1. Mean standard error of tissue selenium concentration (mg g1) for female (n ¼ 26) and male (n ¼ 21) white sturgeona
Gonad Kidney Liver Muscle
Males
Females
p
2.89 0.72 12.79 0.57 11.31 1.29 7.29 0.57
10.00 2.54 12.86 0.81 12.28 1.58 6.93 0.84
0.035 0.799 0.676 0.340
a The difference in the mean tissue selenium between the sexes was tested by Mann-Whitney rank sum test (gonad and liver) and t test (kidney and muscle).
Table 2. Comparison (t test) of selenium tissue burden (mean standard error) between male white sturgeon from San Pablo (Figure 1, sites 1 and 2) and Suisun (Figure 1, sites 3–5) Baysa Selenium (mg g1)
Gonad Kidney Liver Muscle
San Pablo Bay (n ¼ 15)
Suisin Bay (n ¼ 6)
p
3.69 0.96 12.86 0.71 13.18 1.54 8.12 0.63
1.02 0.17 12.63 1.02 6.65 0.78 5.23 0.74
0.002 0.881 0.005 0.014
a Number in parentheses indicates fish sample size. Significance level p < 0.05.
concentrations of 12.83 0.51 mg g1, 11.85 1.04 mg g1, 7.09 0.52 mg g1, and 6.91 1.51 mg g1 (mean standard error) in the kidney, liver, muscle, and immature gonads of San Francisco Bay Delta–resident white sturgeon. Levels of Se in 4-yr-old to 8-yr-old fish were lower than levels in older fish.
158
Environ Toxicol Chem 34, 2015
J. Linares-Casenave et al.
Table 3. Comparison of selenium tissue burden (mean standard error) between female white sturgeon from San Pablo (Figure 1, sites 1 and 2) and Suisun (Figure 1, sites 3–5) Baysa Selenium (mg g1)
Gonad Kidney Liver Muscle
San Pablo Bay (n ¼ 11)
Suisin Bay (n ¼ 15)
p
16.90 3.59 13.75 1.29 18.04 2.49 9.26 1.61
4.95 2.99 12.20 1.05 8.05 1.21 5.21 0.56
