- Email: [email protected]
Elements of concern in fillets of bighead and silver carp from the Illinois River, Illinois
Chemosphere 104 (2014) 63–68
Contents lists available at ScienceDirect
Chemosphere journal homepage: www.elsevier.com/locate/chemosphere
Elements of concern in fillets of bighead and silver carp from the Illinois River, Illinois Jeffrey M. Levengood ⇑, David J. Soucek, Gregory G. Sass 1, Amy Dickinson, John M. Epifanio Illinois Natural History Survey, University of Illinois at Urbana-Champaign, 1816 S. Oak St., Champaign, IL 61820, United States
h i g h l i g h t s Concentrations of total As were below detection limits in most carp fillets. Mean Hg concentrations in fillets were below the USFDA action level. Mercury concentrations in some carp fillets would invoke consumption restrictions. Mean Se concentrations were below thresholds for consumption restrictions. The mean molar ratio of Se:Hg in fillets was lower in bighead than in silver carp.
a r t i c l e
i n f o
Article history: Received 5 July 2013 Received in revised form 23 October 2013 Accepted 26 October 2013 Available online 2 December 2013 Keywords: Toxic elements Fish flesh Invasive carp Consumption
a b s t r a c t Efforts to control invasive bighead (Hypophthalmichthys nobilis) and silver carp (H. molitrix) may include harvest for human consumption. We measured concentrations of arsenic (As), mercury (Hg), and selenium (Se) in fillets from silver and bighead carp collected from the lower Illinois River, Illinois, USA, to determine whether concentrations were of health concern and differed by species, size, and location. Concentrations of total As were below detection limits in most bighead (92%) and silver (77%) carp fillets, whereas inorganic As was below detection limits in all samples. Mean Hg concentrations were greater in bighead (0.068 mg kg1) than in silver carp (0.035 mg kg1), and were smallest in carp from the confluence of the Illinois and Mississippi rivers. Mercury concentrations in fillets were positively correlated with body mass in both species. Concentrations of Hg were below the US Food and Drug Administration’s (USFDA) action level (1 ppm as methyl-Hg); however, concentrations in some bighead (70%) and silver (12%) carp fell within the range that would invoke a recommendation to limit meals in sensitive cohorts. Mean Se concentrations were greater in silver (0.332 mg kg1) than in bighead (0.281 mg kg1) carp fillets, and were below the 1.5 mg kg1 limit for an unrestricted number of meals/month. The mean molar ratio of Se:Hg in fillets was lower in bighead (14.0) than in silver (29.1) carp and was negatively correlated with mass in both species Concentrations of Hg in bighead and silver carp fillets should be considered when assessing the risks associated with the use of these species as a protein source. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Bighead (Hypophthalmichthys nobilis) and silver carp (H. molitrix) are invasive fish species that have spread throughout the Upper Mississippi River Basin (UMRB) since their introduction (Chick and Pegg, 2001; Kelly et al. 2011). These Asian carp species have established naturally-reproducing populations in much of the UMRB and have reached high density populations (Sass et al. 2010). Given their abundance and high bioenergetic demand, they have the potential to negatively influence aquatic ecosystems in ⇑ Corresponding author. Tel.: +1 217 333 6767; fax: +1 217 244 0802. E-mail address: [email protected] (J.M. Levengood). Current address: Wisconsin Department of Natural Resources, 10810 County Hwy N, Boulder Junction, WI 54512, United States. 1
0045-6535/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.chemosphere.2013.10.058
their non-native range (Irons et al. 2007). As a result, considerable effort has focused on containing the spread of these species from reaching new waters outside of the Mississippi River Basin. Control efforts have included increasing commercial harvest for human consumption. To ensure food safety and to dismiss any human health concerns associated with their consumption, contaminant loads in these species must be evaluated. Studies of contaminant concentrations in bighead and silver carp have been limited. Rogowski et al. (2009) examined contaminant concentrations in bighead and silver carp in the Illinois and Mississippi rivers. That initial study was limited (5 fish per site at three locations), but was designed to investigate potential factors that may influence contaminant concentrations (e.g., diet, location, species, and total length) (Rogowski et al. 2009). According to Rogowski et al. (2009), Hg concentrations in some
64
J.M. Levengood et al. / Chemosphere 104 (2014) 63–68
individuals of both species would trigger consumption advisories. Selenium (Se) concentrations in some fish exceeded the USFDA tolerance levels for feed additives (0.3 ppm), and given human consumption of a large meal of these individuals, could result in exceedances of the recommended daily allowance of Se for pregnant woman (60 lg d1). Despite low sample sizes, Rogowski et al. (2009) observed that larger fish tended to have greater Hg burdens and Hg concentrations were positively correlated with trophic position. Concentrations of arsenic (As) and Se also differed by species and location. A more robust study is critically needed to better quantify the effects of species, size, and location on contaminant burdens in bighead and silver carp. We measured concentrations of As, Hg, and Se in skinless fillets of bighead and silver carp from the Illinois River from near its confluence with the Mississippi River upstream to Peoria, Illinois, which encompasses the three largest reaches of the river. Our objective was to test whether concentrations of selected elements in fillets were great enough to be of health concern to human consumers, and if so, whether this differed by species, size, and location. Gaining a better understanding of contaminant burdens in bighead and silver carp and the factors that affect those burdens will allow managers to better assess the risks associated with the potential use of these species as a commercial protein source.
2. Methods We collected 15 fillets each from silver and bighead carp from the Alton (upper and lower), LaGrange, and Peoria reaches of the Illinois River using trammel netting and electrofishing, for a total of 120 samples (Fig. 1). Tissue samples were analyzed for As, Hg, and Se via a Thermo X-Series II with Collision Cell Technology Inductively Coupled Plasma Mass Spectrometer according to US Environmental Protection Agency’s (USEPA) Method 6020. Samples were prepared in accordance with USEPA Method 3050B. Inorganic As (iAs) and trivalent As (pentavalent As was determined by difference) were determined by Hydride Generation-Cryogenic Trapping-Atomic Absorption Spectrometry (HG-CT-AAS) using USEPA Method 1632 (modified). Results of the analysis of Quality Control samples were evaluated and subsequent data were entered into a comprehensive database and checked for data entry errors. Quality control measures included percent recovery of laboratory fortified blanks, certified reference material, and matrix spikes. Other quality control data included relative percent difference (RPD) between spiked duplicates and analysis of laboratory blanks. Elemental concentrations were generally below detection limits in laboratory blanks. Most recoveries of laboratory fortified blanks and other spiked quality control samples were within acceptable limits. For iAs, percent recoveries ranged from 66% to 122% with only two samples falling outside of acceptable limits (45%, 5%). Percent recoveries for total As, Hg, and Se were excellent, ranging from 97% to 111%. Most of the RPDs between matrix spikes and matrix spike duplicates were below acceptable limits, which ranged from 20% to 35% depending on the analyte. Values for iAs ranged from 0% to 29%, and for total As, Hg, and Se ranged from 0.2% to 8%. To compare element concentrations in fillets between species and among sites, we conducted a two-way analysis of variance (ANOVA) using JMPÒ version 9.0.0 software. We used a student’s t-test to conduct post hoc pairwise comparisons. We also used JMP to conduct regression analyses between element concentrations and fish carcass mass and between molar ratios of Se:Hg in fillets and carcass mass. For these analyses, mass was calculated from fish length using regression equations derived from whole fish data collected for another portion of this study. The bighead and silver carp length-mass
relationships were given by mass (g) = 0.00001 (length2.9405) and mass (g) = 0.000039 (length2.7948), respectively. The null hypothesis in all statistical tests was rejected at the a = 0.05 level. 3. Results and discussion Concentrations of total As (tAS) were below detection limits in 92% of bighead and 77% of silver carp fillets (Fig. 2). Arsenic was detected at a much greater frequency in silver carp, although maximum values were similar in both species (Table 1). These maximum observed values were greater than reported for composited fillet samples of bighead and silver carp from the Missouri River (0.034 and 0.039 mg kg1 ww, respectively, Orazio et al. 2011). Inorganic As concentrations were below detection limits (0.0015 mg kg1) in all samples examined. According to the USEPA’s risk-based consumption limit for non-cancer endpoints, iAs concentrations in fish >0.08 mg kg1 would result in some recommended consumption limits (based on RfD). Concentrations of iAs >0.002 mg kg1 would invoke consumption limits with regard to cancer endpoints. Thus, iAs concentrations in bighead and silver carp fillets from the lower Illinois River were not of human health concern. In the absence of iAs, the tAs content of samples was likely comprised primarily of the organoarsenic compound arsenobetaine, which predominates in many fish species (Rattanachongkiat et al., 2004, Miyashita et al. 2009) and has shown little to no toxicity in humans (e.g., Kaise et al. 1985; Sabbioni et al. 1991). Mean Hg concentrations were greater in bighead carp (F1 = 48.4, p < 0.001; Table 1) and were smaller (F3 = 11.6, p < 0.001) in carp from the lower Alton reach, i.e., near the confluence of the Illinois and Mississippi rivers (Fig. 2). Orazio et al. (2011) also found greater Hg concentrations in skinless, bone-in fillets of bighead carp 1 ð as compared to silver carp x ¼ 0:202 mg kg wwÞ 1 ðx ¼ 0:076 mg kg wwÞ from the Missouri River; these mean Hg concentrations, especially in bighead carp, were much greater than 1 in bighead ð x ¼ 0:019Þ and silver carp ð x ¼ 0:035 mg kg Þ, respectively, from the Illinois River. Our mean value in silver carp was the same as that of a single skin-on fillet from a silver carp collected from the Patoka River, though smaller than that of a composite of two silver carp from the Wabash River (0.065 mg kg1), both in Indiana (Indiana Department of Environmental Management data). After converting the mean concentration in silver carp to dry weight (ww 3.33 = dw, 0.117 mg kg1 dw), the concentration we observed was less than in silver carp from a reservoir in Iran having elevated levels of Hg in its water (367 ng g1 or 0.367 mg kg1 dw in white muscle, Khoshnamvand et al. 2013). In contrast to our findings, the Hg concentration was relatively large in a single skin-on muscle sample from one bighead carp collected from the Mississippi River in Minnesota in 2003 (0.382 mg kg1, Minnesota Department of Health data). However, Hg was not detected (detection limits 0.015 and 0.017 mg kg1) in two skin-on composite samples of bighead carp fillets collected from the Peoria reach of the Illinois River in 2004 (Illinois Environmental Protection Agency data). Concentrations of Hg were well below the USFDA action level of 1 ppm methyl-Hg in edible portions (USFDA 2009; Fig. 3). We assumed that all Hg present was methyl-Hg (agencies typically assume that methyl-Hg in fish muscle tissue constitutes 100% of total Hg). Concentrations were below the USEPA Screening Value for Recreational Fishers (USEPA 2000; 0.4 mg kg1). However, fillets from 48 bighead and 14 silver carp contained Hg concentrations above the USEPA Screening Value for Subsistence Fishers. Concentrations in 42 bighead and seven silver carp fell within the range that would invoke a recommendation to limit meals in sensitive cohorts to 1 meal week1 (>0.05–0.22 mg kg1) according to the Great Lakes Fish Advisory Workgroup (2007) (Fig. 3). However, most fish consumption guidelines are based on mean
65
J.M. Levengood et al. / Chemosphere 104 (2014) 63–68
Joliet
Des Plaines River
Starved Rock Dam
Kankakee River
Peoria Reach Peoria Peoria Dam
Illinois River Havana
La Grange Reach La Grange Dam
Alton Reach
Confluence of the Mississippi and Illinois Rivers
Alton
Joliet Havana Alton
Fig. 1. Map showing the Illinois River and locations of dams and river reaches sampled for Asian carp in this study.
concentrations in a sample of fish and consumption of a number of meals over time. This lessens the impact of a single fish with high contaminant concentrations; however, using the higher concentrations observed in a single fish does provide a more conservative approach to risk assessment. Using the mean value approach, the Hg concentrations observed in bighead carp fillets were above, and in silver carp were below, lowest threshold to restrict consumption (0.05 mg kg1) by sensitive cohorts. According to the USEPA’s risk-based consumption limit for non-cancer endpoints, methyl-
Hg concentrations in fish >0.030 mg kg1 would result in some recommended consumption limits. Our greatest observed values would result in a recommendation to limit consumption to four, 8-oz meals per month. Furthermore, fish size, particularly in silver carp, should be considered in consumption recommendations, as Hg concentrations were positively correlated with carcass mass (Fig. 4). Concentrations of Se in carp fillets were well below the 1.5 mg kg1 limit for an unrestricted number of monthly meals
66
J.M. Levengood et al. / Chemosphere 104 (2014) 63–68
Fig. 3. Concentrations of total mercury (Hg) in fillets of bighead (Hypophthalmichthys nobilis) and silver carp (H. molitrix) from the lower Illinois River.
Fig. 2. Concentrations of elements of concern in fillets of a) bighead (Hypophthalmichthys nobilis) and b) silver carp (H. molitrix) from the lower Illinois River. The collection site at the confluence of the Illinois and Mississippi rivers was designed as Alton Lower. Data are arithmetic mean + sd. Concentrations of As below the detection limit are presented as 1/2 the detection limit.
per USEPA. The Se content of fillets, when corrected for meal size, were below the tolerable upper intake for children 4–8 years of age of 150 lg Se1 d1 (NIH, 2012, Fig. 5). Selenium concentrations were greater in silver carp than bighead carp fillets (F1 = 31.5, p < 0.001; Table 1); these concentrations were less than in bighead ð x ¼ 0:56Þ and silver carp ð x ¼ 0:68Þ from the Missouri River (Orazio et al. 2011). Mean Se concentrations were smaller in carp from the confluence of the Illinois and Mississippi rivers and larger in carp from the Alton reach proper (F3 = 21.0, p < 0.001). Concentrations were greatest in silver carp from the upper portion of the Alton reach and lowest in bighead carp from the lower Alton and Peoria reaches (F3 = 4.9, p = 0.003; Fig. 2). Selenium is known to ameliorate Hg toxicity. It appears that Hg binds Se, making it unavailable for critical physiological processes (Ralston et al.2007) and Se deficiency upon methyl-Hg exposure may result in the neurobehavioral toxicity observed with mercury
Fig. 4. Relationship between mercury concentrations (Hg) in fillets and whole carcass mass of bighead (Hypophthalmichthys nobilis) and silver carp (H. molitrix) from the lower Illinois River, Illinois.
exposures (e.g. Watanabe et al. 1999). Thus, Se molar concentrations above that of Hg are needed for proper physiological functioning. For example, maternal consumption of fish having a molar excess of Hg over Se can result in developmental harm (Hall et al. 1978; Julshamn et al. 1987). Conversely, an excess of Se over Hg protects against Hg toxicity (Ganther et al. 1972; Watanabe 2001; Ralston et al. 2007). Although a molar ratio of 1:1 for Se:Hg has been considered the threshold for protection, more research needs to be conducted to better understand variability within and between fish species and uncertainties about what ratios are protective (Burger and Gochfeld, 2012). Currently, Se:Hg should likely be used as a measure of relative risk. In the present study, the molar ratio of Se:Hg concentrations in fillets was significantly (p < 0.001) lower in
Table 1 Concentrations (all in mg kg1 wet weight) of inorganic arsenic (iAS), total arsenic (tAs), total mercury (Hg) and total Se (Se) in bighead (Hypophthalmichthys nobilis) (BH) and silver carp (H. molitrix) (SI) fillets from the the Illinois River, IL.
a
iAs
tAs
Hg
Se
BH fillets
Mean ± s.d. Min–max % Detect
0.016 ± 0.009
0.068 ± 0.038 0.015a–0.220 100
0.281 ± 0.058 0.170a–0.430 100
SI fillets
Mean ± s.d. Min–max % Detect
0.021 ± 0.014
0.035 ± 0.018 0.014a–0.120 100
0.332 ± 0.064 0.210–0.600 100
Estimated concentration above adjusted method detection limit and below reporting limit.
J.M. Levengood et al. / Chemosphere 104 (2014) 63–68
67
human consumers. However, we recommend that fewer and smaller bighead carp be consumed compared to silver carp due to greater Se:Hg ratios and the positive correlation observed between bighead carp mass and Hg concentration. Acknowledgements This study was conducted with USEPA Great Lakes Restoration Initiative funds provided by the US Fish and Wildlife Service through the Illinois Department of Natural Resources (IDNR). Additional support was provided by Illinois-Indiana Sea Grant. Special thanks to Kevin Irons, Tom Heavisides and Steve Schultz, IDNR, David Glover and John Bowzer, Southern Illinois University at Carbondale, and Diane Greer, Illinois Natural History Survey. Fig. 5. Calculated selenium (Se) content of fillets from bighead (Hypophthalmichthys nobilis) and silver carp (H. molitrix) from the lower Illinois River, Illinois, in comparison to the maximum recommended tolerable daily intake for children 4– 8 years of age.
Fig. 6. Molar selenium (Se):mercury (Hg) ratios in fillets of bighead (Hypophthalmichthys nobilis) and silver carp (H. molitrix) from the Illinois River, Illinois.
bighead carp ð x ¼ 14:0 8:4Þ than in silver carp ð x ¼ 29:1 12Þ. Molar ratios in both species were variable and declined with increasing fish carcass mass (Fig. 6). Although statistically significant, bighead carp mass only explained a low amount of variability in the molar ratio of Se:Hg.
4. Conclusions Arsenic and Se concentrations in the bighead and silver carp fillets we examined did not pose a risk to human consumers. Inorganic arsenic concentrations were not detectable. Selenium and Hg concentrations varied by species and river reach. Mercury concentrations were greater in bighead carp and were elevated in both species from the confluence of the Illinois and Mississippi rivers. Mercury in fillets was below the USFDA Action Level and USEPA Screening Value for recreational fishers, though some individual fish fell within more conservative guidelines for consumption restrictions. Mercury concentrations in fillets were positively correlated with carcass mass, especially in silver carp, which were smaller on average and less variable in size. Selenium concentrations were greatest in silver carp from the upper portion of the Alton reach and lowest in bighead carp from the lower Alton and Peoria reaches, though were not of human consumer health concern. Selenium:Hg molar ratios were considerably lower in bighead carp, reflecting the greater Hg loads in these zooplanktivorous filter-feeders. Lower Se:Hg ratios are considered less protective of Hg intoxication in consumers of fish flesh. Overall, concentrations of As, Se, and Hg in bighead and silver carp from the lower Illinois River did not appear to be a health concern for
References Burger, J., Gochfeld, M., 2012. Selenium and mercury ratios in saltwater fish from New Jersey: individual and species variations complicate possible use in human health consumption advisories. Environ. Res. 114, 12–23. Chick, J.H., Pegg, M.A., 2001. Invasive Asian carp in the Mississippi River basin. Science 292, 2250–2251. Ganther, H.E., Goudie, C., Sunde, M.L., Kopecky, M.J., Wagner, P., Oh, S.H., Hoekstra, W.G., 1972. Selenium: relation to decreased toxicity of methylmercury added to diets containing tuna. Science 175, 1122. Great Lakes Fish Advisory Workgroup. 2007. A protocol for mercury-based fish consumption advice: an addendum to the 1993 protocol for a uniform Great Lakes sport fish consumption advisory, 30 pp.
68
J.M. Levengood et al. / Chemosphere 104 (2014) 63–68
USEPA, 2000. Guidance for Assessing Chemical Contaminant Data for Use in Fish Advisories, vol. 1. Third Ed. Fish Sampling and Analysis. EPA 823-B-00-007, 2000, 485 pp. USFDA, 2009. Inspections, Compliance, Enforcement, and Criminal Investigations. CPG Sec. 540.600 Fish, Shellfish, Crustaceans and other Aquatic Animals - Fresh, Frozen or Processed – Methyl Mercury.
Watanabe, C., Yin, K., Kasanuma, Y., Satoh, H., 1999. In utero exposure to methylmercury and Se deficiency converge on the neurobehavioral outcome in mice. Neurotox. Teratol. 21, 83–88. Watanabe, C., 2001. Selenium deficiency and brain functions: the significance for methylmercury toxicity. Nippon Eiseigaku Zasshi 55, 581–589.
Contents lists available at ScienceDirect
Chemosphere journal homepage: www.elsevier.com/locate/chemosphere
Elements of concern in fillets of bighead and silver carp from the Illinois River, Illinois Jeffrey M. Levengood ⇑, David J. Soucek, Gregory G. Sass 1, Amy Dickinson, John M. Epifanio Illinois Natural History Survey, University of Illinois at Urbana-Champaign, 1816 S. Oak St., Champaign, IL 61820, United States
h i g h l i g h t s Concentrations of total As were below detection limits in most carp fillets. Mean Hg concentrations in fillets were below the USFDA action level. Mercury concentrations in some carp fillets would invoke consumption restrictions. Mean Se concentrations were below thresholds for consumption restrictions. The mean molar ratio of Se:Hg in fillets was lower in bighead than in silver carp.
a r t i c l e
i n f o
Article history: Received 5 July 2013 Received in revised form 23 October 2013 Accepted 26 October 2013 Available online 2 December 2013 Keywords: Toxic elements Fish flesh Invasive carp Consumption
a b s t r a c t Efforts to control invasive bighead (Hypophthalmichthys nobilis) and silver carp (H. molitrix) may include harvest for human consumption. We measured concentrations of arsenic (As), mercury (Hg), and selenium (Se) in fillets from silver and bighead carp collected from the lower Illinois River, Illinois, USA, to determine whether concentrations were of health concern and differed by species, size, and location. Concentrations of total As were below detection limits in most bighead (92%) and silver (77%) carp fillets, whereas inorganic As was below detection limits in all samples. Mean Hg concentrations were greater in bighead (0.068 mg kg1) than in silver carp (0.035 mg kg1), and were smallest in carp from the confluence of the Illinois and Mississippi rivers. Mercury concentrations in fillets were positively correlated with body mass in both species. Concentrations of Hg were below the US Food and Drug Administration’s (USFDA) action level (1 ppm as methyl-Hg); however, concentrations in some bighead (70%) and silver (12%) carp fell within the range that would invoke a recommendation to limit meals in sensitive cohorts. Mean Se concentrations were greater in silver (0.332 mg kg1) than in bighead (0.281 mg kg1) carp fillets, and were below the 1.5 mg kg1 limit for an unrestricted number of meals/month. The mean molar ratio of Se:Hg in fillets was lower in bighead (14.0) than in silver (29.1) carp and was negatively correlated with mass in both species Concentrations of Hg in bighead and silver carp fillets should be considered when assessing the risks associated with the use of these species as a protein source. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Bighead (Hypophthalmichthys nobilis) and silver carp (H. molitrix) are invasive fish species that have spread throughout the Upper Mississippi River Basin (UMRB) since their introduction (Chick and Pegg, 2001; Kelly et al. 2011). These Asian carp species have established naturally-reproducing populations in much of the UMRB and have reached high density populations (Sass et al. 2010). Given their abundance and high bioenergetic demand, they have the potential to negatively influence aquatic ecosystems in ⇑ Corresponding author. Tel.: +1 217 333 6767; fax: +1 217 244 0802. E-mail address: [email protected] (J.M. Levengood). Current address: Wisconsin Department of Natural Resources, 10810 County Hwy N, Boulder Junction, WI 54512, United States. 1
0045-6535/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.chemosphere.2013.10.058
their non-native range (Irons et al. 2007). As a result, considerable effort has focused on containing the spread of these species from reaching new waters outside of the Mississippi River Basin. Control efforts have included increasing commercial harvest for human consumption. To ensure food safety and to dismiss any human health concerns associated with their consumption, contaminant loads in these species must be evaluated. Studies of contaminant concentrations in bighead and silver carp have been limited. Rogowski et al. (2009) examined contaminant concentrations in bighead and silver carp in the Illinois and Mississippi rivers. That initial study was limited (5 fish per site at three locations), but was designed to investigate potential factors that may influence contaminant concentrations (e.g., diet, location, species, and total length) (Rogowski et al. 2009). According to Rogowski et al. (2009), Hg concentrations in some
64
J.M. Levengood et al. / Chemosphere 104 (2014) 63–68
individuals of both species would trigger consumption advisories. Selenium (Se) concentrations in some fish exceeded the USFDA tolerance levels for feed additives (0.3 ppm), and given human consumption of a large meal of these individuals, could result in exceedances of the recommended daily allowance of Se for pregnant woman (60 lg d1). Despite low sample sizes, Rogowski et al. (2009) observed that larger fish tended to have greater Hg burdens and Hg concentrations were positively correlated with trophic position. Concentrations of arsenic (As) and Se also differed by species and location. A more robust study is critically needed to better quantify the effects of species, size, and location on contaminant burdens in bighead and silver carp. We measured concentrations of As, Hg, and Se in skinless fillets of bighead and silver carp from the Illinois River from near its confluence with the Mississippi River upstream to Peoria, Illinois, which encompasses the three largest reaches of the river. Our objective was to test whether concentrations of selected elements in fillets were great enough to be of health concern to human consumers, and if so, whether this differed by species, size, and location. Gaining a better understanding of contaminant burdens in bighead and silver carp and the factors that affect those burdens will allow managers to better assess the risks associated with the potential use of these species as a commercial protein source.
2. Methods We collected 15 fillets each from silver and bighead carp from the Alton (upper and lower), LaGrange, and Peoria reaches of the Illinois River using trammel netting and electrofishing, for a total of 120 samples (Fig. 1). Tissue samples were analyzed for As, Hg, and Se via a Thermo X-Series II with Collision Cell Technology Inductively Coupled Plasma Mass Spectrometer according to US Environmental Protection Agency’s (USEPA) Method 6020. Samples were prepared in accordance with USEPA Method 3050B. Inorganic As (iAs) and trivalent As (pentavalent As was determined by difference) were determined by Hydride Generation-Cryogenic Trapping-Atomic Absorption Spectrometry (HG-CT-AAS) using USEPA Method 1632 (modified). Results of the analysis of Quality Control samples were evaluated and subsequent data were entered into a comprehensive database and checked for data entry errors. Quality control measures included percent recovery of laboratory fortified blanks, certified reference material, and matrix spikes. Other quality control data included relative percent difference (RPD) between spiked duplicates and analysis of laboratory blanks. Elemental concentrations were generally below detection limits in laboratory blanks. Most recoveries of laboratory fortified blanks and other spiked quality control samples were within acceptable limits. For iAs, percent recoveries ranged from 66% to 122% with only two samples falling outside of acceptable limits (45%, 5%). Percent recoveries for total As, Hg, and Se were excellent, ranging from 97% to 111%. Most of the RPDs between matrix spikes and matrix spike duplicates were below acceptable limits, which ranged from 20% to 35% depending on the analyte. Values for iAs ranged from 0% to 29%, and for total As, Hg, and Se ranged from 0.2% to 8%. To compare element concentrations in fillets between species and among sites, we conducted a two-way analysis of variance (ANOVA) using JMPÒ version 9.0.0 software. We used a student’s t-test to conduct post hoc pairwise comparisons. We also used JMP to conduct regression analyses between element concentrations and fish carcass mass and between molar ratios of Se:Hg in fillets and carcass mass. For these analyses, mass was calculated from fish length using regression equations derived from whole fish data collected for another portion of this study. The bighead and silver carp length-mass
relationships were given by mass (g) = 0.00001 (length2.9405) and mass (g) = 0.000039 (length2.7948), respectively. The null hypothesis in all statistical tests was rejected at the a = 0.05 level. 3. Results and discussion Concentrations of total As (tAS) were below detection limits in 92% of bighead and 77% of silver carp fillets (Fig. 2). Arsenic was detected at a much greater frequency in silver carp, although maximum values were similar in both species (Table 1). These maximum observed values were greater than reported for composited fillet samples of bighead and silver carp from the Missouri River (0.034 and 0.039 mg kg1 ww, respectively, Orazio et al. 2011). Inorganic As concentrations were below detection limits (0.0015 mg kg1) in all samples examined. According to the USEPA’s risk-based consumption limit for non-cancer endpoints, iAs concentrations in fish >0.08 mg kg1 would result in some recommended consumption limits (based on RfD). Concentrations of iAs >0.002 mg kg1 would invoke consumption limits with regard to cancer endpoints. Thus, iAs concentrations in bighead and silver carp fillets from the lower Illinois River were not of human health concern. In the absence of iAs, the tAs content of samples was likely comprised primarily of the organoarsenic compound arsenobetaine, which predominates in many fish species (Rattanachongkiat et al., 2004, Miyashita et al. 2009) and has shown little to no toxicity in humans (e.g., Kaise et al. 1985; Sabbioni et al. 1991). Mean Hg concentrations were greater in bighead carp (F1 = 48.4, p < 0.001; Table 1) and were smaller (F3 = 11.6, p < 0.001) in carp from the lower Alton reach, i.e., near the confluence of the Illinois and Mississippi rivers (Fig. 2). Orazio et al. (2011) also found greater Hg concentrations in skinless, bone-in fillets of bighead carp 1 ð as compared to silver carp x ¼ 0:202 mg kg wwÞ 1 ðx ¼ 0:076 mg kg wwÞ from the Missouri River; these mean Hg concentrations, especially in bighead carp, were much greater than 1 in bighead ð x ¼ 0:019Þ and silver carp ð x ¼ 0:035 mg kg Þ, respectively, from the Illinois River. Our mean value in silver carp was the same as that of a single skin-on fillet from a silver carp collected from the Patoka River, though smaller than that of a composite of two silver carp from the Wabash River (0.065 mg kg1), both in Indiana (Indiana Department of Environmental Management data). After converting the mean concentration in silver carp to dry weight (ww 3.33 = dw, 0.117 mg kg1 dw), the concentration we observed was less than in silver carp from a reservoir in Iran having elevated levels of Hg in its water (367 ng g1 or 0.367 mg kg1 dw in white muscle, Khoshnamvand et al. 2013). In contrast to our findings, the Hg concentration was relatively large in a single skin-on muscle sample from one bighead carp collected from the Mississippi River in Minnesota in 2003 (0.382 mg kg1, Minnesota Department of Health data). However, Hg was not detected (detection limits 0.015 and 0.017 mg kg1) in two skin-on composite samples of bighead carp fillets collected from the Peoria reach of the Illinois River in 2004 (Illinois Environmental Protection Agency data). Concentrations of Hg were well below the USFDA action level of 1 ppm methyl-Hg in edible portions (USFDA 2009; Fig. 3). We assumed that all Hg present was methyl-Hg (agencies typically assume that methyl-Hg in fish muscle tissue constitutes 100% of total Hg). Concentrations were below the USEPA Screening Value for Recreational Fishers (USEPA 2000; 0.4 mg kg1). However, fillets from 48 bighead and 14 silver carp contained Hg concentrations above the USEPA Screening Value for Subsistence Fishers. Concentrations in 42 bighead and seven silver carp fell within the range that would invoke a recommendation to limit meals in sensitive cohorts to 1 meal week1 (>0.05–0.22 mg kg1) according to the Great Lakes Fish Advisory Workgroup (2007) (Fig. 3). However, most fish consumption guidelines are based on mean
65
J.M. Levengood et al. / Chemosphere 104 (2014) 63–68
Joliet
Des Plaines River
Starved Rock Dam
Kankakee River
Peoria Reach Peoria Peoria Dam
Illinois River Havana
La Grange Reach La Grange Dam
Alton Reach
Confluence of the Mississippi and Illinois Rivers
Alton
Joliet Havana Alton
Fig. 1. Map showing the Illinois River and locations of dams and river reaches sampled for Asian carp in this study.
concentrations in a sample of fish and consumption of a number of meals over time. This lessens the impact of a single fish with high contaminant concentrations; however, using the higher concentrations observed in a single fish does provide a more conservative approach to risk assessment. Using the mean value approach, the Hg concentrations observed in bighead carp fillets were above, and in silver carp were below, lowest threshold to restrict consumption (0.05 mg kg1) by sensitive cohorts. According to the USEPA’s risk-based consumption limit for non-cancer endpoints, methyl-
Hg concentrations in fish >0.030 mg kg1 would result in some recommended consumption limits. Our greatest observed values would result in a recommendation to limit consumption to four, 8-oz meals per month. Furthermore, fish size, particularly in silver carp, should be considered in consumption recommendations, as Hg concentrations were positively correlated with carcass mass (Fig. 4). Concentrations of Se in carp fillets were well below the 1.5 mg kg1 limit for an unrestricted number of monthly meals
66
J.M. Levengood et al. / Chemosphere 104 (2014) 63–68
Fig. 3. Concentrations of total mercury (Hg) in fillets of bighead (Hypophthalmichthys nobilis) and silver carp (H. molitrix) from the lower Illinois River.
Fig. 2. Concentrations of elements of concern in fillets of a) bighead (Hypophthalmichthys nobilis) and b) silver carp (H. molitrix) from the lower Illinois River. The collection site at the confluence of the Illinois and Mississippi rivers was designed as Alton Lower. Data are arithmetic mean + sd. Concentrations of As below the detection limit are presented as 1/2 the detection limit.
per USEPA. The Se content of fillets, when corrected for meal size, were below the tolerable upper intake for children 4–8 years of age of 150 lg Se1 d1 (NIH, 2012, Fig. 5). Selenium concentrations were greater in silver carp than bighead carp fillets (F1 = 31.5, p < 0.001; Table 1); these concentrations were less than in bighead ð x ¼ 0:56Þ and silver carp ð x ¼ 0:68Þ from the Missouri River (Orazio et al. 2011). Mean Se concentrations were smaller in carp from the confluence of the Illinois and Mississippi rivers and larger in carp from the Alton reach proper (F3 = 21.0, p < 0.001). Concentrations were greatest in silver carp from the upper portion of the Alton reach and lowest in bighead carp from the lower Alton and Peoria reaches (F3 = 4.9, p = 0.003; Fig. 2). Selenium is known to ameliorate Hg toxicity. It appears that Hg binds Se, making it unavailable for critical physiological processes (Ralston et al.2007) and Se deficiency upon methyl-Hg exposure may result in the neurobehavioral toxicity observed with mercury
Fig. 4. Relationship between mercury concentrations (Hg) in fillets and whole carcass mass of bighead (Hypophthalmichthys nobilis) and silver carp (H. molitrix) from the lower Illinois River, Illinois.
exposures (e.g. Watanabe et al. 1999). Thus, Se molar concentrations above that of Hg are needed for proper physiological functioning. For example, maternal consumption of fish having a molar excess of Hg over Se can result in developmental harm (Hall et al. 1978; Julshamn et al. 1987). Conversely, an excess of Se over Hg protects against Hg toxicity (Ganther et al. 1972; Watanabe 2001; Ralston et al. 2007). Although a molar ratio of 1:1 for Se:Hg has been considered the threshold for protection, more research needs to be conducted to better understand variability within and between fish species and uncertainties about what ratios are protective (Burger and Gochfeld, 2012). Currently, Se:Hg should likely be used as a measure of relative risk. In the present study, the molar ratio of Se:Hg concentrations in fillets was significantly (p < 0.001) lower in
Table 1 Concentrations (all in mg kg1 wet weight) of inorganic arsenic (iAS), total arsenic (tAs), total mercury (Hg) and total Se (Se) in bighead (Hypophthalmichthys nobilis) (BH) and silver carp (H. molitrix) (SI) fillets from the the Illinois River, IL.
a
iAs
tAs
Hg
Se
BH fillets
Mean ± s.d. Min–max % Detect
0.016 ± 0.009
0.068 ± 0.038 0.015a–0.220 100
0.281 ± 0.058 0.170a–0.430 100
SI fillets
Mean ± s.d. Min–max % Detect
0.021 ± 0.014
0.035 ± 0.018 0.014a–0.120 100
0.332 ± 0.064 0.210–0.600 100
Estimated concentration above adjusted method detection limit and below reporting limit.
J.M. Levengood et al. / Chemosphere 104 (2014) 63–68
67
human consumers. However, we recommend that fewer and smaller bighead carp be consumed compared to silver carp due to greater Se:Hg ratios and the positive correlation observed between bighead carp mass and Hg concentration. Acknowledgements This study was conducted with USEPA Great Lakes Restoration Initiative funds provided by the US Fish and Wildlife Service through the Illinois Department of Natural Resources (IDNR). Additional support was provided by Illinois-Indiana Sea Grant. Special thanks to Kevin Irons, Tom Heavisides and Steve Schultz, IDNR, David Glover and John Bowzer, Southern Illinois University at Carbondale, and Diane Greer, Illinois Natural History Survey. Fig. 5. Calculated selenium (Se) content of fillets from bighead (Hypophthalmichthys nobilis) and silver carp (H. molitrix) from the lower Illinois River, Illinois, in comparison to the maximum recommended tolerable daily intake for children 4– 8 years of age.
Fig. 6. Molar selenium (Se):mercury (Hg) ratios in fillets of bighead (Hypophthalmichthys nobilis) and silver carp (H. molitrix) from the Illinois River, Illinois.
bighead carp ð x ¼ 14:0 8:4Þ than in silver carp ð x ¼ 29:1 12Þ. Molar ratios in both species were variable and declined with increasing fish carcass mass (Fig. 6). Although statistically significant, bighead carp mass only explained a low amount of variability in the molar ratio of Se:Hg.
4. Conclusions Arsenic and Se concentrations in the bighead and silver carp fillets we examined did not pose a risk to human consumers. Inorganic arsenic concentrations were not detectable. Selenium and Hg concentrations varied by species and river reach. Mercury concentrations were greater in bighead carp and were elevated in both species from the confluence of the Illinois and Mississippi rivers. Mercury in fillets was below the USFDA Action Level and USEPA Screening Value for recreational fishers, though some individual fish fell within more conservative guidelines for consumption restrictions. Mercury concentrations in fillets were positively correlated with carcass mass, especially in silver carp, which were smaller on average and less variable in size. Selenium concentrations were greatest in silver carp from the upper portion of the Alton reach and lowest in bighead carp from the lower Alton and Peoria reaches, though were not of human consumer health concern. Selenium:Hg molar ratios were considerably lower in bighead carp, reflecting the greater Hg loads in these zooplanktivorous filter-feeders. Lower Se:Hg ratios are considered less protective of Hg intoxication in consumers of fish flesh. Overall, concentrations of As, Se, and Hg in bighead and silver carp from the lower Illinois River did not appear to be a health concern for
References Burger, J., Gochfeld, M., 2012. Selenium and mercury ratios in saltwater fish from New Jersey: individual and species variations complicate possible use in human health consumption advisories. Environ. Res. 114, 12–23. Chick, J.H., Pegg, M.A., 2001. Invasive Asian carp in the Mississippi River basin. Science 292, 2250–2251. Ganther, H.E., Goudie, C., Sunde, M.L., Kopecky, M.J., Wagner, P., Oh, S.H., Hoekstra, W.G., 1972. Selenium: relation to decreased toxicity of methylmercury added to diets containing tuna. Science 175, 1122. Great Lakes Fish Advisory Workgroup. 2007. A protocol for mercury-based fish consumption advice: an addendum to the 1993 protocol for a uniform Great Lakes sport fish consumption advisory, 30 pp.
68
J.M. Levengood et al. / Chemosphere 104 (2014) 63–68
USEPA, 2000. Guidance for Assessing Chemical Contaminant Data for Use in Fish Advisories, vol. 1. Third Ed. Fish Sampling and Analysis. EPA 823-B-00-007, 2000, 485 pp. USFDA, 2009. Inspections, Compliance, Enforcement, and Criminal Investigations. CPG Sec. 540.600 Fish, Shellfish, Crustaceans and other Aquatic Animals - Fresh, Frozen or Processed – Methyl Mercury.
Watanabe, C., Yin, K., Kasanuma, Y., Satoh, H., 1999. In utero exposure to methylmercury and Se deficiency converge on the neurobehavioral outcome in mice. Neurotox. Teratol. 21, 83–88. Watanabe, C., 2001. Selenium deficiency and brain functions: the significance for methylmercury toxicity. Nippon Eiseigaku Zasshi 55, 581–589.