Gene-specific differences in the aflatoxin B1 adduction of chicken erythrocyte chromatin.

Mature and immature chicken erythrocyte nuclei were treated with activated aflatoxin B1 (2,3-dichloroaflatoxin B1), producing covalently bound DNA adducts. This reaction produces alkali-labile sites in the DNA which can be identified by using a variation of the Maxam-Gilbert sequencing procedure. We determined the aflatoxin B1 accessibility of defined regions of the erythroid genome by using different specific probes and monitoring the disappearance of similar-sized fragments generated by restriction enzyme digestion. The genes studied were the erythroid-specific beta-globin and histone H5 genes, which are potentially active in mature erythroid nuclei and transcriptionally active in immature erythrocytes, and the vitellogenin and ovalbumin genes, which are both transcriptionally inactive in these cells. The beta-globin and histone H5 genes were more accessible than the repressed vitellogenin and ovalbumin genes to aflatoxin B1 modification in mature and immature erythroid chromatin. Micrococcal nuclease was used to probe the nucleosomal organization of active (beta-globin and histone H5) and repressed (vitellogenin and ovalbumin) genes in chicken erythrocytes. The vitellogenin and ovalbumin genes show a canonical nucleosome repeat pattern in mature and immature chicken erythrocyte nuclei. In contrast, the beta-globin and histone H5 genes lack a distinct nucleosomal repeat pattern in these cells. These results support the hypothesis that transcriptionally active genes are preferentially accessible to carcinogen modification because of their disrupted chromatin structure.