Page 325 - The Toxicology of Fishes

Reactive Oxygen Species and Oxidative Stress 305

Biomarkers of Oxidative Stress
How is it possible to know if oxidative stress is a problem in a given field situation? Many studies have
attempted to identify and evaluate reliable biomarkers of exposure to pollution (see Chapter 16). Despite
a large number of laboratory exposures to pure compounds (Almeida et al., 2004; Bergman et al., 1994;
de Pandey et al., 2001; Dorval et al., 2003; Rau et al., 2004) and environmentally relevant contaminant
mixtures (Bergman et al., 1994; Celander et al., 1994; Di Giulio et al., 1993; Livingstone et al., 1992;
Meyer et al., 2003; Nishimoto et al., 1995; Steadman et al., 1991) and field studies (Bacanskas et al.,
2004; Bainy et al., 1996; Eufemia et al., 1997; Livingstone et al., 1992, 1995; McClain et al., 2003;
McFarland et al., 1999; Otto and Moon, 1996; Porte et al., 2002; Rodríguez-Aziza et al., 1992; Stein et
al., 1992; Stephensen et al., 2000; van der Oost et al., 1996; Ventura et al., 2002), no single biomarker
of oxidative stress has emerged that is as sensitive and specific as other established biochemical biom-
arkers such as acetylcholinesterase activity for organophosphate insecticides, δ-ALAD activity for
exposures to lead, or CYP1A expression/activity for aryl hydrocarbon receptor agonists (see Chapter
16). Various studies have identified GSSG:GSH ratios, levels of MT or lipid peroxidation, and activities
of GR, microsomal GST, or microsomal GPX as the most sensitive indicators in the system being studied,
but these markers have been completely unresponsive in other contexts. An extensive review of biomarker
studies that included a specific review of biomarkers of oxidative stress (van der Oost et al., 2003) failed
to identify any marker that was responsive in a high percentage of the studies reviewed, with lipid
peroxidation being perhaps the most consistent marker. Representative examples of field studies employ-
ing biomarkers of oxidative stress are provided in Table 6.5.
Why have better markers of oxidative stress in wild fish not been identified? A variety of factors may
be involved. First of all, as in mammals, large inductions (e.g., comparable to the inductions in CYP1A
observed after exposure to certain xenobiotics, as discussed in Chapters 4 and 16) in antioxidant enzymes
have not been observed in fish exposed to prooxidants. As a result, any confounding factors present have
a strong chance of hiding an otherwise observable effect. Unfortunately, there are many such confounding
factors. An environmental pollution mixture will rarely be expected to exert toxicity only by oxidative
stress, so other forms of toxicity may be occurring. Additionally, sex and reproductive condition can
affect many antioxidant parameters (Livingstone et al., 1995; McFarland et al., 1999; Meyer et al., 2003;
Winzer et al., 2001, 2002a,b) but are not always taken into account. In that case, the variance associated
with sex becomes “noise,” potentially obscuring real differences. Similarly, temperature can greatly alter
the metabolic capacity of poikilotherms and has been shown to affect antioxidant defenses in fish (Heise
et al., 2003; Olsen et al., 1999; Parihar and Dubey, 1995; Parihar et al., 1996). Diet alters the activity
of many antioxidant enzymes (George et al., 2000; Hidalgo et al., 2002; Mourente et al., 2000, 2002;
Pascual et al., 2003), as well as altering the tissue concentrations of nonenzymatic antioxidants, as
mentioned above. Dissolved oxygen (Cooper et al., 2002; Hermes-Lima and Zenteno-Savín, 2002;
Lushchak et al., 2001; Ritola et al., 2002; Ross et al., 2001) and salinity (Kolayli and Keha, 1999;
Martínez-Álvarez et al., 2002) have also been observed to affect antioxidant parameters. Seasonal effects
have been observed (Bacanskas et al., 2004; Ronisz et al., 1999) and are likely to incorporate many
other biological and environmental variables, such as temperature, reproductive status, and food sources.
Additionally, time courses can be complicated; for example, although total glutathione levels can increase
dramatically in response to prooxidant exposure, the initial response is often depletion, and degree of
induction is very likely to be additionally affected by diet, as the availability of cysteine can be limiting
for the production of GSH. Antioxidant enzymes have also been observed sometimes to be depressed
at the level of activity or expression after exposure to prooxidants (Fujii and Taniguchi, 1999; Kim and
Lee, 1997; Pedrajas et al., 1995; Radi and Matkovics, 1988; Stephensen et al., 2002; Zikic et al., 1997).
Developmental stage is another variable to be considered (Peters and Livingstone, 1996).
Physiological or genetic adaptation to pollution has been shown to lead to an altered response even
in biomarkers that are usually fairly robust, such as CYP1A (Elskus et al., 1999; Hahn 1998; Meyer et
al., 2002; Roy et al., 2001; see also Chapter 15). Although the possibility of adaptation to oxidative
stress in fish has been less studied, it may also be a significant factor in some cases (Bacanskas et al.,
2004; McFarland et al., 1999; Meyer et al., 2003). Markers of damage, such as lipid peroxidation and
DNA damage, may only reflect recent or constant damage, as both can be repaired. Another important