Joanna Burger, Michael Gochfeld, Tara Shukla, Christian Jeitner, Sean Burke, Mark Donio, Sheila Shukla, Ronald Snigaroff, Daniel Snigaroff, Timothy Stamm, Conrad Volz

Considerable attention has been devoted to the risks from mercury (Hg) and polychlorinated biphenyls (PCB) to high-level predators (including humans) who consume freshwater fish. Although the U.S. Food and Drug Administration (FDA) issued advisories because of Hg for four marine fish species, there are few data on lead (Pb), cadmium (Cd), or other metals in Bering Sea fish generally, or on the risk these levels pose to the fish themselves or to consumers of marine fish. Levels of arsenic (As), Cd, chromium (Cr), Pb, Hg, and selenium (Se) levels were examined in muscle and liver of 142 Pacific Cod (Gadus macrocephalus) collected in 2004 at Nikolski, Adak, Amchitka, and Kiska Islands in the Aleutian Chain (Alaska) in the Bering Sea/North Pacific Ocean, a major source of commercial fishing. One key objective was whether there were location, age, gender, and size effects on tissue concentration that might pose a risk to the fish or their predators (including humans). All fish were measured and weighed, and a subset was aged by examining otolith layers. As was higher in liver than in muscle (geometric mean 2420 versus 1590 ng/g or ppb wet weight), as were Cd (GM 224 versus 1.92) and Se (GM 1380 versus 165). Conversely, Cr was higher in muscle (76.8 versus 45 ppb), as were Pb (23.7 vs 12 ppb) and surprisingly Hg (128 versus 82 ppb). Adak, until recently a large military base, had the highest levels of As, Hg, and Se, while Amchitka had the highest Pb levels, but Nikolski, which generally had the lowest levels, had relatively high Pb in liver. In general, interisland differences were significant for most metals in muscle, but only for Cr in liver. Weight and length were positively related to age, but age tended to explain more of the variance in metal levels. The multiple regression relationships differed by tissue in an unanticipated manner. Location contributed significantly to the models for muscle Cd, Pb, Hg, and Se, but not for liver levels. Conversely the length by weight interaction entered all of the liver models but none of the muscle models. Se and Hg were positively but weakly correlated in both liver (tau = +0.16) and muscle tissue (tau = 0.12). Hg was positively correlated with length, weight, and age in muscle, but not in liver. As showed a significant negative correlation with size variable in both tissues, and Cr was negatively correlated in muscle. Cd was positively correlated with Hg, Se, and As. Between liver and muscle there were significant positive correlations for Hg (tau = .24), As (tau = .407), and Cr (tau = 0.17), but not for Pb, Cd, or Se. In this study, the only metals that might pose a risk to cod-eating predators is Hg, as well as some of the higher values of Pb at Amchitka and Nikolski . The U.S. Environmental Protection Agency (EPA) reference dose (RfD) (not available for lead) was used to evaluate the risk to people consuming an 8-ounce (228g) meal of cod once per day and once per week, and to calculate risk using the levels found in this study. If a subsistence fisher from one of the Aleut villages ate one meal of cod per week for As, or one meal per day for Hg, they would exceed the U.S. EPA reference dose for As and Hg (set at a level to be without adverse effect for any person with this average daily exposure).