Page 327 - The Toxicology of Fishes

Reactive Oxygen Species and Oxidative Stress 307

concern is that most fish populations are highly outbred with a high degree of genetic variability
(compared to typical laboratory mammalian models), which may contribute additional variability to the
responses observed. Finally, as mentioned earlier, the various antioxidant isoenzymes potentially repre-
sented in fish are not well characterized, and it may be that potentially good biomarkers are being
overlooked due to nonspecific substrates, antibodies, etc. An improved understanding of biological and
environmental modulation of oxidative stress and antioxidant defenses in fish, along with careful exper-
imental design and improved techniques, should enhance the utility of biomarkers for oxidative stress
in the future; however, as discussed above, no single broadly applicable biomarker for oxidative stress
is likely to emerge in the foreseeable future.

Oxidative Stress and the Health of Wild Fish Populations
As described above, oxidative stress is known to play an important role in many disease processes and
chemical-mediated toxicity in humans. Furthermore, it is clear that there are many similarities among
humans, mammalian models, and fish in terms of production and cellular effects of oxidative stress, as
well as in terms of antioxidant defenses; nonetheless, the importance of oxidative stress to fish populations
in the wild is not yet fully understood. Biomarkers of effect of exposure to ROS such as lipid peroxidation
(Di Giulio et al., 1993; Eufemia et al., 1997; Livingstone et al., 1993) and DNA damage (Di Giulio et
al., 1993; Eufemia et al., 1997; Malins and Gunselman, 1994; Payne et al., 1998; Rodríguez-Aziza et
al., 1999; Stein et al., 1992; Stephensen et al., 2000; Sugg et al., 1996; Theodorakis et al., 1997, 1999)
indicate that fish inhabiting polluted environments exhibit oxidative stress (although not all of these
DNA damage studies are necessarily strictly oxidative-stress related). Some common fish diseases, such
as jaundice and methemoglobinemia (“brown blood disease”) have been associated with oxidative stress
(Jensen, 1996; Sakai et al., 1998). Some studies suggest that the highly teratogenic effects of TCDD are
related to its ability to produce ROS mediated by the catalytic activity of cytochrome P4501A (Cantrell
et al., 1996; Dong et al., 2002; Schlezinger et al., 1999; Teraoka et al., 2003); however, this remains
controversial (Carney et al., 2004).
Fish populations adapted to inhabit ecosystems contaminated with chemical mixtures containing
prooxidants such as creosote (Meyer and Di Giulio, 2003) and metals (Weis et al., 1999; Xie and Klerks,
2003) have been shown to be less fit than their counterparts from clean sites in several contexts. There
are many published examples of fish populations inhabiting sites contaminated with prooxidant chemicals
that exhibit high rates of DNA damage and cancer (Black, 1983; Malins and Gunselman, 1994; Malins
et al., 1984; Mix, 1986; Vogelbein et al., 1990). Although the chemicals present at such sites are not
exclusively toxic via prooxidant mechanisms, a strong relationship exists between oxidative stress and
cancer (Klaunig et al., 1998). Furthermore, the ability of channel catfish to avoid cancer in the same
polluted environments that cause liver cancer in brown bullhead catfish has been associated with higher
antioxidant defenses (Hasspieler et al., 1994a,b; Ploch et al., 1999). Dietary exposure to hydrogen
peroxide enhanced the development of oxidative DNA damage (8-OHdG) and hepatic tumors in rainbow
trout exposed to the carcinogen N-methyl-N′-nitro-N-nitrosoguanidine in a laboratory study (Kelly et
al., 1992). As in the case of human epidemiological studies, however, it is difficult to unequivocally link
a given pollutant or form of toxicity with population-level measurements of health. Long-term laboratory
exposures will likely be needed to better define the links between exposure to oxidative stress and
ecologically relevant alterations in fish populations, and improved markers of oxidative stress will be
necessary to analyze the responses of populations of wild fish to oxidative stress.

Future Directions
A better understanding of which antioxidant genes are present in fish, how they are regulated, and what
their biochemical characteristics are will greatly improve our ability to understand the impact of oxidative
stress and the impact of prooxidant chemicals on fish. Similarly, application of the knowledge gained
in studies of oxidative stress in mammalian models will inform improved generation of hypotheses and
experimental design; for example, we know now that very large inductions in expression of classical