Page 554 - The Toxicology of Fishes

534 The Toxicology of Fishes

found cholangiocellular neoplasms in 3 of 12 white suckers (Catostomus commersoni) and one hepato-
cellular neoplasm among 100 brown bullheads (Ictalurus nebulosus). This small set of findings accom-
plished three things: (1) it showed that the salmonid family is not the only family of teleosts in which
liver tumors can arise, (2) it triggered the suspicion that anthropogenic carcinogens in aqueous habitats
might be responsible for liver neoplasms in feral fishes, and (3) it raised the question as to whether
hepatic and other neoplasms in fishes might serve as indicators of the presence of chemical carcinogens
in geographically widespread aqueous habitats. Within 2 years of the Deep Creek Lake survey, Stanton
(1965, 1966), using a small fish of the cyprinid (minnow) family, Brachydanio rerio (or “zebra danio”),
demonstrated that diethylnitrosamine (DEN) and cycasin are hepatocarcinogenic in that species.
These three discoveries, all made in rapid succession within a 5-year period (1961 to 1966) were
largely to determine the pathways taken by investigations of neoplasms in fish up to the present time.
The main directions these investigations took can be categorized as follows:

1. Rainbow trout have been evaluated experimentally, as they are a highly sensitive animal for
the detection and bioassay of the hepatocarcinogenic action of chemicals and are an apparently
ideal species for analysis of the histo- and cytogenesis of liver tumors. More recent investiga-
tions have proven this to be a multi-organ carcinogenesis model.
2. Surveys of polluted marine and freshwater habitats have been conducted with the purpose of
finding epizootics of neoplasms in fishes as evidence of environmental exposure to carcinogens
(for reviews, see Couch and Harshbarger, 1985; Harshbarger and Clark, 1990).
3. Many different species of small “aquarium fishes” have been tested to determine their sensitivity
to a wide variety of known chemical carcinogens and their future usefulness in bioassays of
chemicals for carcinogenic potency (reviewed in Boorman et al., 1997; Bunton, 1996; Hawkins
et al., 1995; Hoover, 1984). Work with these and other aquarium species underwent rapid
expansion. As a consequence of such successful employment of fishes in carcinogen testing,
an investigation of the underpinning molecular etiology of cancer has been facilitated by three
aquaria-held species. Xiphophorus, medaka, and zebrafish have each emerged as models of
particular human cancers (for a general review, see Berghmans et al., 2005; Ostrander and
Rotchell, 2005; Stern and Zon, 2003).

Association of Fish Cancers with Environmental Contaminants
The Dawe et al. (1964) study is one of the earliest if not the first report that suggested an association
between anthropogenic carcinogens in the aquatic medium and neoplasia in fishes. Epizootics of neo-
plasia in wild fishes have been reported following carefully conducted, often multi-year, investigations
in the United States (Harshbarger and Clark, 1990; Malins et al., 1985), Canada (Goyette et al., 1988),
various countries bordering the North Sea (Vethaak and Rheinallt, 1992), and Japan (Kimura et al.,
1984). The approach used in searching for epizootics in wild fishes has been to establish the prevalence
of neoplasia in organisms from specific sites and to compare this with that of organisms from reference
sites. Because cancer is a form of chronic toxicity, there has been an association between advanced age
of fish and the prevalence of tumors (Baumann et al., 1990). In addition, those fish residing on or near
sediment (benthic species) have often been those in which tumors have been found (Murchelano and
Wolke, 1985; Myers et al., 1994; Peters et al., 1987; Vethaak and Rheinallt, 1992). In general, the
characteristics of the tumor-prone fish include multiple years of age, residence in polluted harbors or
estuaries, and a bottom- or benthic-dwelling lifestyle (Harshbarger and Clark, 1990).
A more thorough coverage of field studies and fish neoplasia is provided later in this chapter; however,
an overview of approaches and findings to establish the association between environmental contamination
and neoplasia is presented here and organized around principles presented by Peters et al. (1987) and
reviewed and expanded by Vethaak and Rheinallt (1992). According to these principles: “A correlation
must exist between disease (neoplasia) prevalence and distribution of a pollutant within a relatively
small area.” Studies by Malins and associates (1985, 1987) meet these conditions. English sole
(Parophrys vetulus) were collected from Eagle Harbor and from a reference site in Puget Sound,