Implications for the formation of abasic sites following modification of polydeoxycytidylic acid by acrolein in vitro.

Polydeoxycytidylic acid (poly dC) was incubated with excess acrolein. A Nensorb 20 nucleic acid purification cartridge was used to bind the polymeric material in the poly dC/acrolein reaction mixture. The non-polymeric material eluted from this column had a UV absorbance four times higher than that of the control. The fluorescence spectrum of the eluted material did not correspond to that of unmodified cytosine. Separate aliquots of the reaction mixture were digested to deoxynucleotide 3'-monophosphates by incubation with micrococcal nuclease and spleen phosphodiesterase. The products were converted to 32P-labeled deoxynucleotide 3',5'-bisphosphates by incubation with T4 polynucleotide kinase and excess [gamma-32P]ATP. The 3'-monophosphate was selectively removed by incubation with nuclease P1. Two-dimensional thin-layer chromatography (TLC) on polyethyleneimine cellulose (PEI)-cellulose and detection of 32P-labeled deoxynucleotide 5'-monophosphates by autoradiography failed to provide evidence for the formation of an acrolein adduct of deoxycytidine 5'-monophosphate. When acrolein-modified deoxycytidine 3'-monophosphate was 32P post-labeled, a new product, which co-chromatographed with UV markers synthesized by reaction of acrolein with deoxycytidine 5'-monophosphate, was detected. These data show that acrolein-modified deoxycytidine 3'-monophosphates are substrates for 32P labeling by T4 polynucleotide kinase and are stable under the assay conditions employed. The inability to detect the acrolein-modified nucleotides after reaction with poly dC in vitro suggests that the modified bases are lost from poly dC by cleavage of the N-glycosyl bond resulting in the formation of an abasic site.