Allele-specific hypermethylation of the retinoblastoma tumor-suppressor gene

Abstract

Inactivation of the retinoblastoma gene appears to have a fundamental role in the genesis of retinoblastoma, osteosarcoma, and other malignant tumors. The gene is generally inactivated because of loss-of-function mutations, although epigenetic phenomena, such as hypermethylation of the promoter region, could possibly have the same effect. We investigated the methylation pattern at the 5' end of the retinoblastoma gene, including its promoter region and exon 1, in DNA purified from 56 primary retinoblastomas. We found five tumors with evidence for hypermethylation, all from unilateral, simplex patients. No methylation abnormalities were detected in DNA purified from the leukocytes from these patients. It is interesting that in one of these tumors the hypermethylation was confined to one allele. There were no mutations in a 1,306-bp sequence including the hypermethylated region that might account for the allele-specific hypermethylation. We believe that the hypermethylation of the retinoblastoma gene that we found in these tumors corresponds to the allelic inactivation of the gene, and we speculate that erroneous hypermethylation without alteration of nucleotide sequence occasionally plays a role in the genesis of this cancer. If this is true, then retinoblastomas with hypermethylation might be treatable with chemotherapeutic agents that interfere with methylation of DNA.