Base-specific arrest of in vitro DNA replication by carcinogenic chromium: relationship to DNA interstrand crosslinking.

Abstract

We have previously shown that trivalent chromium can bind to purified DNA and form lesions capable of obstructing DNA replication in vitro. Trivalent chromium is not, however, carcinogenic to humans. Rather, it is the end product of the intracellular reduction of hexavalent chromium, which is carcinogenic. The process of chromium reduction yields several reactive intermediates which may also interact with DNA, perhaps producing different lesions than those generated when trivalent chromium binds DNA. The present study was undertaken to determine whether the treatment of DNA with hexavalent chromium in the presence of ascorbate (the intracellular reductant responsible for most in vivo chromium reduction), would also generate DNA lesions capable of obstructing replication. Using increasing chromium concentrations and a constant ascorbate:chromium ratio of 0.5:1 to generate biologically relevant adduct levels, a DNA polymerase arrest assay revealed that polymerase arresting lesions were formed and were indistinguishable from those generated by trivalent chromium, in that the most prominent arrests sites were one base upstream of guanine residues on the template strand. Measurement of the amount of chromium bound to template DNA in relation to the number of arrests demonstrated that only a subset (18.5%) of the chromium adducts were capable of causing polymerase arrest. Arrest assays performed with increasing ratios of ascorbate to chromium showed that high ratios (> or = 5:1) resulted in decreased polymerase arrests. DNA interstrand crosslinks in the arrest assay template were detected by renaturing agarose gel electrophoresis, and were shown to decrease markedly with increasing ascorbate to chromium ratios, whereas chromium binding levels remained unchanged. These results strongly implicate DNA interstrand crosslinks as the polymerase arresting lesion. The present study confirms and extends our previous study with trivalent chromium, and suggests that while the initial chemical nature of the DNA lesions formed by either trivalent chromium or reductive intermediates of hexavalent chromium may differ, their effect on DNA replication is the same.