Page 386 - The Toxicology of Fishes

366 The Toxicology of Fishes

as compared with fish depurated for more than 4 weeks in clean water in the laboratory; however, in
the field-sampled fish, a gradient of hepatic Pgp expression correlating with proximity to the pulp mill
effluents was not evident. In laboratory experiments, blennies exposed to oiled sediments exhibited
significantly elevated hepatic Pgp levels, while blennies exposed to contaminated sediments from the
pulp mill sites or injected with β-naphthoflavone showed no elevation of hepatic Pgp levels.
When a creosote-resistant killifish population from a polluted site in the Elizabeth River in Virginia
was studied (Cooper et al., 1999), adult fish showed a high prevalence of hepatic tumors. In immunoblots
with the C219 antibody, both normal and neoplastic livers of killifish from the polluted site contained
significantly higher levels of Pgp-like proteins than did livers of reference killifish. This finding may be
taken as an indication that Pgp contributes to creosote resistance, since biliary excretion is a major
elimination route for PAH metabolites.
In contrast, studies on killifish in New Bedford Harbor in Massachusetts (Bard et al., 2002a) revealed
an acquired resistance to planar halogenated aromatic hydrocarbons (PHAHs), and the contaminant-
resistant fish showed lower levels of hepatic C219-immunoreactive protein than did killifish from a
relatively unpolluted site. Interestingly, fish from New Bedford Harbor displayed Pgp-like protein in the
intestine, whereas fish from the relatively unpolluted site exhibited no detectable Pgp in the intestine.
The authors speculated that fish at New Bedford Harbor are exposed to xenobiotics that are inducers of
a gut-specific Pgp isoform and that intestinal uptake is the dominating route of xenobiotics uptake.
The question remains whether elevated levels of hepatic or intestinal Pgp observed in contaminant-
resistant killifish from polluted sites reflect an exposure to Pgp inducers. Because fish from the polluted
field sites showed strongly declining levels of Pgp expression after culture in clean water (Bard et al.,
2002b), the argument can be made for a role of xenobiotics in the induction of Pgp expression; however,
results from laboratory exposure experiments that directly demonstrate a role of environmental chemicals
in elevating the Pgp expression of killifish are elusive. Exposure of killifish to the PHAH 2,3,7,8-
tetrachlorodibenzofuran, known to be present in New Bedford Harbor, did not stimulate intestinal or
hepatic Pgp expression (Bard et al., 2002a). Likewise, when Cooper et al. (1999) injected both reference
and creosote-resistant killifish with 3-methylcholanthrene, a PAH contained in creosote, they found no
increase of hepatic levels of Pgp-like proteins. In summary, based on the current information, contam-
inant-resistant fish may show altered Pgp expression but the cause-and-effect relationship of xenobiotic
exposure lacks verification. Overall, the database on the relation between xenobiotics and Pgp in fish is
small and dispersed. Published reports do not yield a consistent pattern, and this may be explained in
part by interlaboratory differences in species, xenobiotics, and experimental design; however, it may
also be due to technical problems, such as the lack of fish-specific probes for Pgp, and to insufficient
knowledge on the basic parameters of Pgp regulation in fish. Pgp expression can be induced as part of
a general response to cellular stress, and this may be regulated by ligands of the PXR, thus sharing
elements of regulation with an important group of mammalian biotransformation enzymes, CYP3A.
Further, mammalian Pgp and CYP1A share some inducers, albeit they are not regulated by the same
mechanisms. For fish, no published evidence exists on PXR, and the understanding of piscine CYP3A,
despite some ortholog characterization in teleosts (Kullman and Hinton 2001; Kullman et al., 2000),
remains fragmentary. Several studies with fish were attempted to provoke a co-induction of CYP1A and
Pgp. When Fundulus heteroclitus were exposed to the AhR ligand 3-methylcholanthrene and observed
(Cooper et al., 1999), an induction of CYP1A but not of Pgp expression occurred. Similarly, Bard et al.
(2002a) exposed killifish to the AhR ligand 2,3,7,8-tetrachlorodibenzofuran with resultant induction of
CYP1A and not Pgp. In rainbow trout, prochloraz induced hepatic CYP1A but not hepatic Pgp (Sturm
et al., 2001a). A correlation between CYP1A and Pgp levels was found in blennies collected at polluted
field sites; however, when blennies were treated in the laboratory with the AhR ligand β-naphthoflavone,
only CYP1A (not Pgp) was elevated (Bard et al., 2002b). In flounders collected at three sites with
differing pollution levels in Seine Bay (English Channel), CYP1A and Pgp levels showed exactly opposite
trends (Minier et al., 2000). These findings suggest that Pgp and CYP1A are not coordinately regulated
in fish; however, these two proteins may act in complementary ways in cellular detoxification.
Virtually nothing is known about cholestasis in fish, and there is an urgent need for understanding
piscine cholestasis given that the pathogenesis of other toxic alterations in piscine liver is at least
qualitatively similar to that of mammalian liver. Two studies in medaka showed that chronic exposures