Page 257 - The Toxicology of Fishes

Receptor-Mediated Mechanisms of Toxicity 237

Several features of fish make them valuable as models in toxicology (Ballatori and Villalobos, 2002;
Hinton et al., 2005; Kelly et al., 1998). As vertebrates, fish have a close evolutionary relationship to
humans, with shared genes and biochemical pathways that have become even more apparent as a result
of recent whole-genome analyses (Aparicio et al., 2002; Jaillon et al., 2004). Most of the fish species
used in toxicological research are small, develop rapidly with a short generation time, and have trans-
parent, externally developing embryos that facilitate experiments in developmental toxicology. For some
species, such as zebrafish, methods for transgenesis (Linney and Udvadia, 2004; Udvadia and Linney,
2003) and gene knock-down (Nasevicius et al., 2000) are well developed and serve as powerful tools
for mechanistic research. These and other advantages of small fish models have been described in detail
elsewhere for zebrafish (Danio rerio) (Carvan et al., 2005; Hill et al., 2005; Linney et al., 2004b), medaka
(Oryzias latipes) (Oxendine et al., 2006), Atlantic killifish (Fundulus heteroclitus) (Burnett et al., 2007),
and rainbow trout (Oncorhynchus mykiss) (Bailey et al., 1996).

Comparative Toxicology: Complications and Opportunities

The use of fish as models for human health requires careful consideration of the similarities and
differences between fish and mammals. First, it is important to keep in mind that “fish” as a group
encompasses an estimated 29,000 species (compared to approximately 5500 for mammals), including
bony, cartilaginous, and jawless fishes and representing an enormous range of genetic, biochemical, and
physiological diversity. Second, although as vertebrates fish share with humans and other mammals most
features of key biochemical pathways involved in mechanisms of toxicity, important differences between
fish and mammals in the details of these pathways have emerged in recent years (Table 5.1). Most notable
is the finding that a whole-genome duplication occurred in the teleost fish lineage approximately 350
million years ago (MYA), after its divergence from the lineage that would become tetrapods (including
humans) (Amores et al., 1998; Christoffels et al., 2004; Crow et al., 2006; Hoegg et al., 2004; Postlethwait
et al., 2004; Taylor et al., 2001). The result of this is that many fish species have retained extra copies
(paralogs*) of some genes as compared to humans. Although only approximately 15 to 30% of the
duplicated genes have been retained (Brunet et al., 2006), many of these encode transcription factors
(Brunet et al., 2006; Steinke et al., 2006). Examples of toxicologically relevant genes for which fish
have additional paralogs include CYP19 (Kishida and Callard, 2001; Tchoudakova and Callard, 1998),
estrogen receptors (ERs) (Bardet et al., 2002; Hawkins et al., 2000), and aryl hydrocarbon receptors (see
below). Other differences in genetic diversity have arisen from independent expansion of some gene
families in mammals and fishes (e.g., the CYP2 family) as well as from lineage-specific gene losses in
these groups. The resulting differences in the toxicological toolkit raise important questions about the
extrapolation of findings between fish and mammals. The existence of differences in receptor diversity
among species highlights the need to understand the nature of homologous relationships among these
genes; interpretations and extrapolation will depend on whether the genes under consideration are related
as orthologs or paralogs (Sanetra et al., 2005).
The presence of extra copies of fish genes as a result of the fish-specific whole-genome duplication
is more than just an annoying complication. It also provides an opportunity for mechanistic insights,
because duplicated fish genes might be exploited to obtain new information about the function of their
single mammalian counterpart (Amores et al., 1998; Force et al., 1999; Lynch and Force, 2000; Post-
lethwait et al., 2004; Taylor et al., 2001); for example, the duplication, degeneration, complementation
(DDC) model of gene evolution (Force et al., 1999; Lynch and Force, 2000) predicts that the multiple
functions of a mammalian gene may be partitioned between its fish “co-orthologs” (referred to as
subfunction partitioning) (Postlethwait et al., 2004). Most of the evidence in support of this model comes

* Distinct types of homologous genes occur within and among species, and specialized terms are used to distinguish among
them. Genes in two different species are orthologous (Fitch, 1970) if they are descended from the same gene in the most recent
common ancestor of the two species (i.e., the two genes are separated only by a speciation event). Paralogous genes (Fitch,
1970) are homologs that resulted from a gene duplication event such as a tandem duplication or a whole-genome duplication.
If a gene duplication has occurred in one lineage but not another, the term co-ortholog (Gates et al., 1999; Taylor et al., 2001)
is used to describe the relationship between each of the two duplicated genes in one species and their single ortholog in the
other species. Other types of homology have also been described (Meyer and Mindell, 2001).