A recent report by Czene et al. (1) in this journal described both 1–2 hydroxypropyl-adenine DNA and HP-valine hemoglobin adducts in workers exposed to PO.1 The methods used were elegant and the findings convincing. Concentration levels of these two adducts were highly correlated. Biomarkers of exposure such as these are required to interpret biomarkers of genotoxic effect, which serve to reflect the biological consequences of adverse environmental/occupational exposures.
In this same report, SCEs were measured as indicators of effect. Unlike the elevations of adducts, however, the claim that SCEs were increased in frequency in response to PO exposures must be viewed with caution. It is noteworthy that the controls did not come from the same industrial facility that the workers did. Although matched for sex and age, it is not clear how well the individuals from an institute of industrial health otherwise compared with the workers being tested. In addition, the workers themselves might have experienced other chemical exposures, because it was only PO that was measured. As clearly described in this report, SCE frequencies were remarkably low in both exposed workers and controls. These low SCE frequencies were attributable to very low levels of BrdUrd incorporation into DNA in all lymphocyte cultures necessitating the use of an alternative antibody technique for sister chromatid identification. This alternative method should not in itself have skewed frequency distributions between the two groups, but differences in BrdUrd incorporation could have, given that BrdUrd in part determines in vitro exchange frequencies. Had this occurred, it could have accounted for group mean differences but not for the associations found between adduct concentrations and SCE frequencies (the strength of having these individual biomarkers of exposure). However, on closer inspection, even these reported associations are not as convincing as they might appear. For both kinds of adducts, the correlations between adduct concentrations and SCE frequencies were made over the entire data set, i.e., over both controls and exposed workers. The controls by definition are unexposed. In addition, for whatever reason, their SCE frequencies are low. The inclusion of the controls in the regressions may have produced an association between adduct concentrations (reflecting PO exposure levels) and SCEs that simply reflects the high density of data points in the bottom left quadrant (Table 2B in this study). These data points are mostly from the controls. A true association between adducts and SCEs should be seen also for the exposed workers alone who apparently experienced a wide range of PO exposures as reflected by the adduct concentrations. This association is not apparent by visual inspection of Table 2B; however, there are too few data points to be certain, a point acknowledged by the authors.
Focusing on SCEs in this regard might itself be a bit misleading. These end points do not reflect an adverse biological consequence of an environmental exposure in that they do not represent irreversible in vivo genotoxic effects. It would be important to know the results of studies of chromosome aberrations, especially translocations, in these PO-exposed workers if any were performed. Structural chromosome changes (or somatic mutations) are convincing indicators of genotoxic effects in exposed individuals.
The issue of whether or not a genotoxic effect was produced by a known exposure to an occupational chemical is important and becoming more so as such data are finding their way into the human cancer risk assessment process. Positive reports of effects must be examined rigorously. The authors do not claim in this study that their finding of PO induction of SCEs in exposed workers is definitive and recommend that additional studies be done to investigate this important point. They are certainly correct and have provided valuable tools for conducting such studies.
The abbreviations used are: PO, propylene oxide; SCE, sister chromatid exchange.
RSS Feeds