Revisiting the identification of breast cancer tumour suppressor genes defined by copy number loss of the long arm of chromosome 16.

Abstract

In breast cancer loss of the long-arm of chromosome 16 is frequently observed, suggesting this is the location of tumour suppressor gene or genes. Previous studies localised two or three minimal regions for the LOH genes in the vicinity of 16q22.1 and 16q24.3, however the identification of the relevant tumour suppressor genes has proved elusive. The current availability of large datasets from breast cancers, that include both gene expression and gene dosage of the majority of genes on the long-arm of chromosome 16 (16q), provides the opportunity to revisit the identification of the critical tumour suppressor genes in this region. Utilising such data it was found 37% of breast cancers are single copy for all genes on 16q and this was more frequent in the luminal A and B subtypes. Since luminal breast cancers are associated with a superior prognosis this is consistent with previous data associating loss of 16q with breast cancers of better survival. Previous chromosomal studies found a karyotype with a der t(1;16) to be the basis for a proportion of breast cancers with loss of 16q. Use of data indicating the dosage of genes 21.9% of breast cancers were consistent with a der t(1;16) as the basis for loss of 16q. In such cases there is both loss of one dose of 16q and three doses of 1q suggesting a tumour suppressor function associated with long-arm of chromosome 16 and an oncogene function for 1q. Previous studies have approached the identification of tumour suppressor genes on 16q by utilising breast cancers with partial loss of 16q with the assumption regions demonstrating the highest frequency of loss of heterozygosity pinpoint the location of tumour suppressor genes. Sixty one of 816 breast cancers in this study showed partial loss of 16q defined by dosage of 357 genes. There was no compelling evidence for "hot-spots" of localised LOH which would pinpoint major tumour suppressor genes. Comparison of gene expression data between various groups of breast cancers based on 16q dosage was used to identify possible tumour suppressor genes. Combining these comparisons, together with known gene functional data, allowed the identification of eleven potential tumour suppressor genes spread along 16q. It is proposed that breast cancers with a single copy of 16q results in the simultaneous reduction of expression of several tumour suppressor genes. The existence of multiple tumour suppressor genes on 16q would severely limit any attempt to pinpoint tumour suppressor genes locations based on localised hot-spots of loss of heterozygosity. Interestingly, the majority of the identified tumour suppressor genes are involved in the modulation of wild-type p53 function. This role is supported by the finding that 80.5% of breast cancers with 16q loss have wild-type p53. TP53 is the most common mutated gene in cancer. In cancers with wild-type p53 would require other strategies to circumvent the key tumour suppressor role of p53. In breast cancers with complete loss of one dose of 16q it is suggested this provides a mechanism that contributes to the amelioration of p53 function.