Subcellular Distribution of Cadmium and Nickel in Chronically Exposed Wild Fish: Inferences Regarding Metal Detoxification Strategies and Implications for Setting Water Quality Guidelines for Dissolved Metals

ABSTRACT

The objective of this study was to investigate metal detoxification in chronically exposed juvenile yellow perch (YP: Perca flavescens) and to field test the commonly assumed threshold toxicity model. Fish were collected from lakes located along a cadmium (Cd) and nickel (Ni) concentration gradient. Ambient dissolved metal concentrations were measured to evaluate exposure and total hepatic metal concentrations were determined as a measure of metal bioaccumulation. Hepatic metal partitioning among potentially metal-sensitive fractions (heat-denatured proteins, organelles) and detoxified metal fractions (metallothionein) was determined after differential centrifugation of YP liver homogenates. Major proportions of hepatic Cd were found in the heat-stable cytosolic peptides and proteins fraction (HSP; including metallothioneins), whereas Ni was mainly found in the potentially metal-sensitive heat-denaturable proteins fraction (HDP). For these chronically exposed fish there was no threshold exposure concentration below which binding of Cd or Ni to the heat-denaturable protein fraction or the organelle fraction did not occur. Metal detoxification was clearly incomplete and P. flavescens was subject to some metal-related stress, as evidenced notably by endocrine perturbations. Similar subcellular partitioning results were obtained when juvenile yellow perch were transferred from a reference lake to a Cd-contaminated lake and Cd accumulation was followed over time; there was no accumulation threshold below which Cd binding to the putative metal-sensitive fractions (HDP and organelles) did not occur. The presence of Cd and Ni in these fractions, even for low exposure concentrations and low hepatic accumulation, contradicts the threshold toxicity model that underpins metal toxicology theory and that is implicitly used in setting water quality guidelines for metals. Chronically exposed YP appear to have settled for a tradeoff between the cost of turning on their detoxification apparatus at full capacity, to completely suppress metal binding to metal-sensitive sites, and the alternative cost of allowing some binding of inappropriate metals to metal-sensitive sites.

Key Words:

• yellow perch
• Cd
• Ni
• bioaccumulation
• detoxification
• threshold toxicity model
• PNEC

ACKNOWLEDGMENTS

This research was supported by the Metals in the Environment Research Network (MITE-RN). This network received financial contributions from the Natural Sciences and Engineering Research Council of Canada (NSERC), the Mining Association of Canada, Ontario Power Generation Inc., the International Copper Association, the International Lead Zinc Research Organization, and the Nickel Producers Environmental Research Association. The insightful comments of two anonymous referees are gratefully acknowledged. P. G. C. Campbell and A. Hontela are supported by the Canada Research Chair program.

Notes

¹Under the influence of the ongoing metals risk assessments in Europe (and with the increasing acceptance of the Biotic Ligand Model in regulatory agencies), there is widespread agreement that metal speciation in the water column (and water chemistry in general) must be taken into account (ICMM 2007).

^aRatio of “maximum concentration : minimum concentration”, as an expression of the water chemistry gradient. For pH, the ratio was calculated on the basis of the H⁺-ion concentrations.

²Analogous results have been reported for a freshwater mollusc living in lakes in the same Rouyn-Noranda study area (Campbell et al. 2005; Bonneris et al. 2005a,b).

^aValue taken from the CCME website (http://www.ccme.ca/sourcetotap/cadmium.html), for an unspecified water hardness (site visited 2007–03-06).

^bValue taken from USEPA (2006) “Criterion Continuous Concentration” or CCC, an estimate of the highest concentration of a material in surface water to which an aquatic community can be exposed indefinitely without resulting in an unacceptable effect. The value was corrected for a hardness of 5 or 65 mg·L⁻¹ CaCO₃, giving a range from 29 ng·L⁻¹ at the low hardness to 170 ng·L⁻¹ at the higher hardness (see USEPA 2006, Appendices A and B).

^cPredicted No Effect Concentration (PNEC) taken from the draft European Community risk assessment for Cd, for an unspecified hardness (EU 2003).

^dTrigger value taken from the Australia/New Zealand Water Quality Guidelines (A-NZ 2000), for protection of 99% of indigenous species in freshwater ecosystems. The value was corrected for a hardness of 5 or 65 mg·L⁻¹ CaCO₃, giving a range from 12 ng·L⁻¹ at the low hardness to 120 ng·L⁻¹ at the higher hardness (see A-NZ 2000, Tables 3.4.1 and 3.4.3).

^eValue taken from Demers et al. (2006), who used Ni as an example of the new Canadian protocol for developing Water Quality Guidelines, for an unspecified hardness.

^fCCC value taken from USEPA (2006). The value was corrected for a hardness of 5 or 65 mg·L⁻¹ CaCO₃, giving a range from 4 μg·L⁻¹ at the low hardness to 36 μg·L⁻¹ at the higher hardness (see USEPA 2006, Appendices A and B).

^gTentative PNEC values pending the finalization of the EU Existing Substances Risk Assessment of Nickel (Schlekat C, NiPERA, pers comm, 2007–03-07). PNECs were derived by normalizing a chronic nickel aquatic toxicity database (27 species) with chronic Biotic Ligand Models (BLMs). Water quality parameters (pH, hardness, and dissolved organic carbon [DOC]) from seven freshwater scenarios representing typical EU surface waters were used to parameterize the BLMs. The HC5 from a log-normal distribution of species-mean EC10/NOEC values was divided by an Assessment Factor of 2 to yield PNEC values. The lowest PNEC came from a water with pH = 7.7, hardness = 48 mg/L, and DOC = 2.8 mg /L. The highest PNEC came from a water with pH = 6.9, hardness = 260 mg/L, and DOC = 12 mg/L.

^hTrigger value taken from the A-NZ (2000), guidelines for protection of 99% of indigenous species in freshwater ecosystems. The value was corrected for a hardness of 5 or 65 mg·L⁻¹ CaCO₃, giving a range from 1.7 μg·L⁻¹ at the low hardness to 15.4 μg·L⁻¹ at the higher hardness (see A-NZ 2000, Tables 3.4.1 and 3.4.3).