Barry C. Kelly

Intestinal absorption and biomagnification of organic contaminants in fish, wildlife, and humans

Methods for the regulatory assessment of the bioaccumulation potential of organic chemicals are founded on empirical measurements and mechanistic models of dietary absorption and biomagnification. This study includes a review of the current state of knowledge regarding mechanisms and models of intestinal absorption and biomagnification of organic chemicals in organisms of aquatic and terrestrial food chains and also includes a discussion of the implications of these models for assessing the bioaccumulation potential of organic chemicals. Four mechanistic models, including biomass conversion, digestion or gastrointestinal magnification, micelle-mediated diffusion, and fat-flush diffusion, are evaluated. The models contain many similarities and represent an evolution in understanding of chemical bioaccumulation processes. An important difference between the biomagnification models is whether intestinal absorption of an ingested contaminant occurs solely via passive molecular diffusion through serial resistances or via facilitated diffusion that incorporates an additional advective transport mechanism in parallel (i.e., molecular ferrying within gastrointestinal micelles). This difference has an effect on the selection of physicochemical properties that best anticipate the bioaccumulative potential of commercial chemicals in aquatic and terrestrial food chains. Current regulatory initiatives utilizing Kow threshold criteria to assess chemical bioaccumulation potential are shown to be unable to identify certain bioaccumulative substances in air-breathing animals. We urge further research on dietary absorption and biomagnification of organic chemicals to develop better models for assessing the bioaccumulative nature of organic chemicals.

Bioaccumulation behaviour of polybrominated diphenyl ethers (PBDEs) in a Canadian Arctic marine food web

A comparative analysis of the bioaccumulation behaviour of polybrominated diphenyl ethers (PBDEs) and polychlorinated biphenyls (PCBs) was conducted involving simultaneous measurements of PBDE and PCB concentrations in organisms of a Canadian Arctic marine food web. Concentrations of individual PBDE congeners (BDE-28, -47, -99, -153, -154 and -183) in Arctic marine sediments (0.001-0.5 ng.g(-1) dry wt) and biota (0.1-30 ng.g(-1) wet wt) were low compared to those concentrations in biota from urbanized/industrial regions. While recalcitrant PCB congeners exhibited a high degree of biomagnification in this food web, PBDE congeners exhibited negligible biomagnification. Trophic magnification factors (TMFs) of PCBs ranged between 2.9 and 11, while TMFs of PBDEs ranged between 0.7 and 1.6. TMFs of several PBDE congeners (BDE-28, -66, -99, -100, -118, -153 and -154) were not statistically greater than 1, indicating a lack of food web magnification. BDE-47 was the only PBDE with a TMF (i.e. 1.6) statistically greater than 1, hence showing evidence of biomagnification in the food web. However, the TMF of BDE-47 (1.6) was substantially lower than TMFs of recalcitrant Cl(5)-Cl(7) PCBs (TMFs~9-11). Species-specific bioaccumulation factors (BAFs) of PBDEs in homeotherms were much smaller than those for PCBs. This further indicates the low degree or absence of biomagnification of PBDEs compared to PCBs in this food web. The field observations suggest PBDEs exhibit a relatively rapid rate of depuration though biotransformation in Arctic marine organisms, which is consistent with laboratory studies in fish and rats.

Mercury and other trace elements in farmed and wild salmon from british Columbia, Canada

The present study reports measured levels of Hg and other trace elements in commercial salmon feed; farmed Atlantic, coho, and chinook salmon (n 110); and wild coho, chinook, chum, sockeye, and pink salmon (n 91). Metal concentrations in farmed and wild salmon from British Columbia, Canada, were relatively low and below human health consumption guidelines. Methylmercury in all salmon samples (range, 0.03-0.1 g/g wet wt) were below the 0.5 g/g guideline set by Health Canada. Negligible differences in metal concentrations were observed between the various species of farmed and wild salmon. Metal concentrations generally were higher in commercial salmon feed compared to farmed salmon. Mercury showed slight bioaccumulation potential in farmed salmon, with biomagnification factors (BMFs) ranging between 0.8 and 1.9. Other metals, such as Cd, Pb, and Ni, exhibited biodilution, with BMFs of much less than one. The relatively low degree of biomagnification of metals observed in farmed salmon likely resulted from the combination of low gastrointestinal absorption efficiency, negligible transfer to muscle tissue relative to other compartments, and a high degree of growth dilution in these fish. Human dietary exposure calculations indicate intakes of Hg, Cd, Pb, Cu, As, and Ni via farmed and wild British Columbia salmon are a relatively small percentage of total intakes (0.05-32%) compared to other Canadian foodstuffs, such as fruits, vegetables, chicken, and beef (68-99%). Although total dietary exposure of Cd, Pb, and Cu approached provisional tolerable daily intake levels, the contribution from British Columbia salmon was less than 2%. Our findings indicate farmed and wild British Columbia salmon remain a safe source of omega-3 highly unsaturated fatty acid intake for cardioprotective and, possibly, other health benefits.

Tissue residue concentrations of organohalogens and trace elements in adult Pacific salmon returning to the Fraser River, British Columbia, Canada

We report measured concentrations of organohalogens and trace elements in muscle and eggs of returning wild Pacific sockeye and chinook salmon during their 2007 migration through the Fraser River watershed in Canada. Chemical analyses revealed the presence of ppb to ppm levels of a wide variety of contaminants in these fish, including polychlorinated biphenyls (PCBs); polychlorinated dibenzo-p-dioxins (PCDDs); polychlorinated dibenzofurans (PCDFs); polybrominated diphenyl ethers (PBDEs); organochlorine pesticides (OCPs) such as DDTs, hexachlorocyclohexanes (HCHs), octachlorostyrene, and cyclodienes; and Hg, As, Cd, Pb, and several other trace elements. Body weights and flesh lipid contents declined during upstream migration, resulting in significantly higher (p < 0.05) lipid-normalized concentrations of lipophilic organohalogens (PCBs, PCDD/Fs, pesticides) in those spawning salmon. Postmigration magnification factors (MFs) of organohalogens (0.1-10) were comparable to previous observations and model predictions. MFs generally increased with increasing hydrophobicity (K(OW)). For example, MFs of tetra- and pentachlorobenzenes and HCH isomers (log K(OW) range: 3.8-5) were relatively low (between 0.1 and 1.7) compared with those of more lipophilic compounds (log K(OW) > 6) such as PCBs, DDTs, and mirex (MFs between 5 and 10). Lipid-normalized muscle:egg ratios in female salmon, which varied between 0.1 and 8, also exhibited a positive relationship with chemical K(OW). The results indicate that lipophilic compounds (K(OW) > 10(6)) can be magnified in flesh lipids of Pacific salmon during spawning migration, but maternal transfer kinetics (deposition to eggs) of those chemicals are relatively slow compared with less hydrophobic compounds. 2,3,7,8-TCDD toxic equivalents (ΣTEQs) in eggs of these spawning salmon, calculated using WHO toxic equivalency factors (WHO-TEFs) for fish health, in some cases exceeded the 0.3 pg·g(-1) threshold level associated with 30% salmonid egg mortality, indicating the potential for reproductive impacts in Fraser River salmon populations.

Gene expression profiling and environmental contaminant assessment of migrating Pacific salmon in the Fraser River watershed of British Columbia.

The health and physiological condition of anadromous salmon is of concern as their upriver migration requires navigation of human-impacted waterways and metabolism of stored energy reserves containing anthropogenic contaminants. Such factors may affect reproductive success of fish stocks. This study investigates chemical contaminant burdens and select gene expression profiles in Pacific Sockeye (Oncorhynchus nerka) and Chinook (Oncorhynchus tshawytscha) salmon which traverse the Fraser River watershed during their spawning migration. Chemical analyses of muscle tissue and eggs of salmon collected from the lower Fraser River (pre-migration) and from upstream spawning grounds (post-migration) during the 2007 migration revealed the presence of numerous chemical contaminants, including PCBs, dioxins/furans, pesticides, and heavy metals. However, muscle tissue residue concentrations were well below human health consumption guidelines and 2,3,7,8 TCDD toxic equivalents (SigmaTEQs) in salmon eggs, calculated using WHO toxic equivalency factors (WHO-TEFs) for fish health, did not exceed the 0.3pgg(-1) wet weight toxicological threshold level previously associated with 30% egg mortality in salmon populations. Quantitative real-time PCR probes were generated and used to assess differences in abundance of key mRNA transcripts encoding nine gene products associated with reproduction, stress, metal toxicity, and exposure to environmental contaminants. Gene expression profiles were characterized in liver and muscle tissue of pre- and post-migration Sockeye and Chinook salmon. The results of stock-matched animals indicate that dynamic changes in mRNA levels occur for a number of genes in both species during migration and suggest that Sockeye salmon exhibit a greater level of biological stress compared to the Chinook salmon population. Using a male-specific genotypic marker, we found that out of the 154 animals examined, one Sockeye was genotypically male but phenotypically female. This individuals gene expression profile in liver and muscle was reminiscent of, but not identical to, the female expression profile. These studies provide the first glimpse of the dynamic yet common nature of changes in the transcriptome that are shared between species during in-migration and highlight differences that may relate to population success. Continued longitudinal assessment will further define the association between contaminant burden, physiological stress, and modulation of gene expression in migrating Pacific salmon.

Flesh residue concentrations of organochlorine pesticides in farmed and wild salmon from British Columbia, Canada

The present study reports measured levels of organochlorine pesticides (OCPs) in commercial salmon feed (n = 8) and farmed Atlantic, coho, and chinook salmon (n = 110), as well as wild coho, chinook, chum, sockeye, and pink salmon (n = 91). Flesh residue concentrations (ng/g wet weight) of dichlorodiphenyltrichloroethanes (DDTs), hexachlorocyclohexanes (HCHs), chlordanes, chlorobenzenes (CBz) and cyclodiene pesticides (e.g., dieldrin, mirex) were 2 to 11 times higher (p < 0.05) in farmed salmon compared with wild salmon. Concentrations were positively correlated with flesh lipid levels. Farmed Atlantic salmon (12-15% lipid) typically exhibited the greatest OCP burdens compared with other salmon species. However, when expressed on a lipid weight basis, concentrations of OCPs (ng/g lipid weight) in wild salmon, in many cases, exceeded those levels in farmed salmon. Observed interspecies and site-specific variations of OCP concentrations in farmed and wild salmon may be attributed to divergent life history, prey/feed characteristics and composition, bioenergetics, or ambient environmental concentrations. Calculated biomagnification factors (BMF = C(F)/C(D), lipid wt) of OCPs in farmed salmon typically ranged between two and five. Biomagnification of chemicals such as DDTs, chlordanes, and mirex was anticipated, because those compounds tend to exhibit high dietary uptake and slow depuration rates in fish because of relatively high octanol-water partition coefficients (K(OW)s > 10⁵). Surprisingly, less hydrophobic pesticides such as hexachlorocyclohexanes and endosulfans (K(OW) s < 10⁵) consistently exhibited a high degree of biomagnification in farmed salmon species (BMFs > 5). This is contrary to previous laboratory and field observations demonstrating fish BMFs less than 1 for low K(OW) chemicals, because of efficient respiratory elimination of those compounds via gills. The results suggest that ambient seawater concentrations and bioconcentration-driven accumulation may play a key role in the bioaccumulation of these relatively more water-soluble contaminants in farmed salmon. Finally, OCP exposure through consumption of British Columbian salmon is found to be low relative to United States national average per capita total exposure levels and provisional tolerable daily intakes.

An interlaboratory comparison study for the determination of dialkyl phthalate esters in environmental and biological samples

A series of interlaboratory comparison exercises were conducted to assess the accuracy of dialkyl phthalate ester (DPE) concentration measurements in environmental and biological samples. Five laboratories participated in analyses to determine DPE concentrations in standard test solutions; marine sediments; three certified reference materials, including CARP-2 (fish muscle) and BCR-07 (fortified milk powder); and several livestock samples (sheeps milk, liver, and muscle). In addition, one laboratory determined DPE residue concentrations in 20 municipal sewage sludge samples, previously analyzed as part of the 2006/2007 U.S. Environmental Protection Agencys Targeted National Sewage Sludge Survey (TNSSS). The results showed relatively good interlaboratory agreement for analyses of di-ethylhexyl phthalate (DEHP). Three independent laboratories (Labs A, B, and C) reported concentrations of DEHP (ng/g wet wt) in fish muscle (CARP-2) of 1,550 ± 148, 1,410 ± 193, and 1,380 ± 187, respectively. Similarly, DEHP concentration measurements in sewage sludge samples showed good agreement with those reported in the 2006/2007 TNSSS report. Measured concentrations of individual DPEs and C6-C10 isomeric mixtures in these samples of municipal sewage sludge, which have not been previously reported, ranged between 1 and 200,000 ng/g dry weight. The results demonstrate that environmental monitoring of DPEs is often hampered by high method detection limits (MDLs), due to contamination of procedural blanks. It is important to note, however, that when background contamination is minimized (<10 ng/sample), relatively low MDLs (<0.1 ng/g) can be achieved, allowing for low-level quantification of DPEs in environmental and biological samples. Future efforts to develop better protocols to lower MDLs, as well to develop reference materials, would greatly benefit future DPE monitoring initiatives.

Isomer-Specific Transplacental Transfer of Perfluoroalkyl Acids: Results from a Survey of Paired Maternal, Cord Sera, and Placentas

Currently, information regarding isomer-specific concentrations of PFHxS, PFOS, and PFOA in human placenta, and corresponding placental-maternal ratios (RPM) of these compounds does not exist. The objective of the present study was to assess the occurrence, and distribution of different PFHxS, PFOS, and PFOA isomers in maternal serum, umbilical cord serum, and placenta to gain a better understanding of transplacental transport efficiency and prenatal exposure risks. The study involved quantitative determination of isomer-specific concentrations of PFHxS, PFOS, and PFOA in samples of maternal serum (n = 32), cord serum (n = 32), and placenta (n = 32) from pregnant women in Wuhan, China. The results indicate that both linear and branched PFHxS, PFOS and PFOA can be efficiently transported across the placenta, with exposure levels ordered maternal serum > cord serum > placenta. For PFOS isomers, the concentration ratios between cord serum and maternal serum (RCM) were ordered n < iso < 4m < (3 + 5)m < 1m < ∑m2. The RPM values exhibited a similar trend for branched PFOS isomers: iso < 4m ≈ (3 + 5)m < 1m ≈ ∑m2. Conversely, PFOA isomers did not exhibit an obvious structure-activity relationship for RCM and RPM. n-PFHxS transported across the placenta to a greater extent than br-PFHxS. To the best of our knowledge, this is the first study to report the occurrence of PFHxS, PFOS, and PFOA isomers in human placenta. Further, RPM values of these compounds are reported here for the first time. The findings help to better understand the mechanisms of the placental transfer and neonatal exposure to these important contaminants of concern.

Chlorinated Polyfluoroalkyl Ether Sulfonic Acids in Matched Maternal, Cord, and Placenta Samples: A Study of Transplacental Transfer

Currently, information regarding concentrations of chlorinated polyfluoroalkyl ether sulfonic acids (Cl-PFESAs) in human placenta does not exist. The main objective of this study was to assess the occurrence and distribution of two Cl-PFESAs, 6:2 Cl-PFESA and 8:2 Cl-PFESA, in maternal serum, umbilical cord serum, and placenta to better assess the transport pathways related to human prenatal exposure. The widely studied perfluorooctanesulfonate (PFOS) was studied for comparison. This study was a hospital-based survey involving quantitative determination of Cl-PFESA and PFOS concentrations in maternal serum (n = 32), cord serum (n = 32), and placenta (n = 32) samples from women in Wuhan, China. The results indicate that Cl-PFESAs can efficiently be transported across placenta, with median exposure levels of 0.60 and 0.01 ng/mL for 6:2 Cl-PFESA and 8:2 Cl-PFESA in the cord sera, respectively. Concentrations of the target compounds in maternal sera, cord sera, and placentas decreased in the following order: PFOS > 6:2 Cl-PFESA > 8:2 Cl-PFESA. Similar patterns were observed in maternal sera, cord sera, and placentas for Cl-PFESAs, with concentrations decreasing in the following order: maternal sera > cord sera > placentas. Significant correlations were observed among 6:2 Cl-PFESA, 8:2 Cl-PFESA, and PFOS concentrations in the maternal serum, cord serum, and placenta samples (r > 0.7; p < 0.001). The median value of RCM (ratio of cord serum to maternal serum concentration) of 6:2 Cl-PFESA was 0.403, indicating a relatively high (∼40%) placental transfer efficiency. 8:2 Cl-PFESA was transported across placenta to a greater extent than 6:2 Cl-PFESA was, likely because of its higher hydrophobicity and lower plasma protein binding affinity. To the best of our knowledge, this is the first study to report the occurrence and distribution of 6:2 Cl-PFESA and 8:2 Cl-PFESA in human placenta. The findings improve our understanding of the mechanisms of transplacental transfer and neonatal exposure to these important PFOS alternatives.

Bioaccumulation Behavior of Pharmaceuticals and Personal Care Products in Adult Zebrafish (Danio rerio): Influence of Physical-Chemical Properties and Biotransformation

The factors influencing bioaccumulation of pharmaceuticals and personal care products (PPCPs) in aquatic organisms are not well understood. The present study involved a comprehensive laboratory investigation to assess the bioaccumulation behavior of several PPCPs in adult zebrafish (Danio rerio). The studied PPCPs included several ionogenic organic compounds (IOCs) such as weak acids and weak bases. Experiments involved two exposure groups (high and low) and a control group, with a 6 day aqueous exposure, followed by a 7 day depuration phase under flow-through conditions. Uptake rate constants (ku) ranged between 0.19 and 8610 L·kg-1·d-1, while depuration rate constants (kd) ranged between 0.14 and 5.14 d-1 in different fish tissues. Steady-state bioconcentration factor (BCFss) values varied widely among the studied PPCPs, ranging from 0.09 to 6,460. In many cases, BCFss values of individual PPCPs differed substantially among different fish tissues. Positive linear relationships were observed between log BCFss values and physical-chemical properties such as octanol-water distribution coefficients (log Dow), membrane-water distribution coefficients (log Dmw), albumin-water distribution coefficients (log DBSAw), and muscle protein-water distribution coefficients (log Dmpw), indicating the importance of lipid-, phospholipid-, and protein-water partitioning. The results also showed that for many PPCPs, the estimated whole-body metabolism rate constant (km) values were comparable to the observed depuration rate (kd), indicating that metabolism plays a major role in the overall elimination of these compounds in zebrafish. An exception was sertraline, which exhibited a kd value (0.4-0.5 d-1) that was much higher than the estimated whole-body km (0.03 d-1). Overall, the results help to better understand the influence of physical-chemical properties and biotransformation on bioaccumulation behavior of these contaminants of concern in aquatic organisms.