Mixture effects of environmental contaminants in Atlantic cod (Gadus morhua) : Combining experimental approaches to study contaminant-induced biological responses

Abstract

Release of chemicals from anthropogenic activities is considered a substantial threat to the global environment. Marine ecosystems are sinks for accumulation of contaminants, therefore marine organisms are highly susceptible to contaminant exposure and possible adverse effects. Importantly, organisms are continuously exposed to complex mixtures of chemicals and interactions among these chemicals may occur. Thus, acquiring knowledge of which components that provoke certain effects can be challenging and needs to be addressed in detail.

The main objective of this dissertation was to assess the biological responses of Atlantic cod (Gadus morhua) exposed to contaminant mixtures, focusing primarily on the liver as the target organ for effects. An additional aspect was to investigate mixture effects of contaminants using different experimental approaches, including a field/caging study (Paper I), an in vivo exposure study (Paper II), and an ex vivo exposure study with precision-cut liver slices (PCLS) (Paper III). The caging and in vivo studies (Paper I and II) assessed effects of environmentally relevant concentrations of contaminants, whereas a mechanistic approach with a wider range of concentrations was applied in the PCLS study (Paper III). While Paper I included several groups of environmental contaminants, the contaminants investigated in the studies comprising this dissertation were the polycyclic aromatic hydrocarbons (PAHs) (Paper II) and the per- and polyfluoroalkyl substances (PFAS) (Paper II and III). The distribution and possible adverse effects of these contaminants have previously been studied in isolation in marine species, however, limited research exists on their combined effects.

In Paper I, an in situ approach was applied to assess whether Atlantic cod would be affected by contaminants leaking from a capped waste disposal site in the bay of Kollevåg in Western Norway. To this end, farmed juveniles of Atlantic cod were caged for six weeks at different locations within Kollevåg bay, with three cages at, or close to the capped disposal site, and a reference cage at a nearby, unpolluted location. Chemical measurements and analyses of biological responses were combined to investigate the possible impact of the waste disposal site on caged cod. We observed higher concentrations of certain contaminant groups at the Kollevåg stations both in sediment, cod liver, and cod bile compared to the reference station. Contaminant concentrations in cod liver were greatest in cod caged at station 1, which was situated closest to the waste disposal site. In contrast, station 2 had the highest concentrations of PAHs and polychlorinated biphenyls (PCBs) in sediments, in addition to the highest levels of PAH metabolites in cod bile. Several biological responses were measured in cod tissues. Specifically, condition factor, hepatosomatic index, and hepatic antioxidant enzyme activities were reduced in cod caged at station 1. Station-dependent increases in expression of genes involved in lipid metabolism and steroidogenesis were also observed. Effects on reproductive parameters, including induction of steroid hormone synthesis and genes encoding proteins involved in steroidogenesis were most prominent among the biological responses assessed, and had the highest correlation with contaminant concentrations in the liver. These findings suggest that reproduction of cod caged in Kollevåg may be disrupted by contaminants leaking from the waste disposal site, which may also apply for local fish inhabiting the area.

In Paper II, an in vivo approach was applied by exposing Atlantic cod for two weeks to mixtures of PAHs and PFAS through two intraperitoneal injections (day 0 and day 7), using environmentally relevant (1x, L) and higher (20x, H) concentrations. PAH metabolite and PFAS concentrations were determined in cod bile and liver, respectively, and confirmed a dose-dependent accumulation of contaminants. Chemical analyses revealed trends of potential chemical interactions between PAHs and PFAS, with regards to accumulation and/or metabolization. Biological responses were assessed by applying both single biomarker assays and toxicogenomic approaches (proteomics and lipidomics). Observed changes in biological responses were mostly related to PFAS exposure. Proteomic and lipidomic analyses of liver samples revealed an increase in the levels of proteins involved in lipid degradation pathways and a decrease in levels of triacylglycerols in cod exposed to H-PFAS, respectively. In addition, hepatic antioxidant enzyme activities were significantly induced following PFAS exposure. For several other biomarker assays, including plasma vitellogenin concentrations, cytochrome P450 1a (Cyp1a) response (both at mRNA and protein levels/activity) and DNA fragmentation, no clear effects were observed following contaminant exposure. Nonetheless, our findings demonstrated that two weeks of exposure to environmentally relevant concentrations of PFAS were sufficient to affect the lipid composition and homeostasis in cod liver.

In Paper III, PCLS from Atlantic cod were applied as an ex vivo approach to expand on PFAS mixture toxicity, in addition to comparing mixture effects to effects of single PFAS. Cod PCLS were exposed for 48 h to perfluorooctanesulfonic acid (PFOS), perfluorooctanoic acid (PFOA), and perfluorononanoic acid (PFNA) at three different concentrations (10, 50 and 100 μM), and ternary mixtures of these PFAS (10, 50 and 100 μM of each compound). Changes in the transcriptome of exposed PCLS were assessed by RNA sequencing. When comparing single compounds, PFOS exposure generated a larger number of differentially expressed genes (DEGs) compared to PFOA and PFNA exposure. Affected pathways following PFOS exposure included sterol metabolism, nuclear receptor pathways and oxidative stress-related responses. Compared to single compound exposure, mixture exposure generated a 10-fold higher number of DEGs compared to PFOS exposure alone. PFOS clustered closer to the PFAS mixture in hierarchical clustering and principal component analyses compared PFOA and PFNA, and the majority of DEGs following PFOS exposure was shared with the DEGs following mixture exposure. The PFAS mixture caused significant enrichment of several stress-related pathways, including antioxidant responses, ferroptosis, cancer-related responses, and nuclear receptor pathways. Mixture effect analysis showed that approximately 10% of the DEGs from the mixture exposure displayed non-additive effects compared to the prediction from the single compound exposures. The majority of these genes showed synergistic effects. To summarize, PFAS exposure promoted effects on nuclear receptors, sterol metabolism and oxidative stress-related pathways in cod PCLS. Although some genes had synergistic or antagonistic expression patterns in the PFAS mixture, the majority of DEGs showed additive expression patterns, suggesting additivity to be the main mixture effect.