The problem is that both copies of a gene can also be frequently deleted from cancers in regions called fragile sites. The underlying DNA structure of fragile sites appears to make them particularly prone to breakage and hence to being jettisoned from the cancer genome. The deletions at fragile sites are, most often, likely to be passengers. Therefore the challenge is how to distinguish between the passenger deletions over fragile sites and the driver deletions over tumour suppressor genes.
"We analysed almost 750 cancer cell samples for deletions at known tumour suppressor genes and compared them to deletions at known fragile sites," explains Dr Graham Bignell, a lead author on the paper. "We found clear differences between the two which allowed us to construct 'signatures' of deletions that are associated with tumour suppressor genes and signatures associated with fragile sites."
For example, one crucial difference between the tumour suppressors and fragile sites was that both copies of tumour suppressor genes were usually deleted and it was rare to find only one copy deleted. In contrast, it was often the case that only one copy was deleted at fragile sites. This provides a first-pass test to sift through for tumour suppressor genes: if both copies of a gene are usually deleted and one copy deleted only rarely then these deletions may be drivers and the gene may be a tumour suppressor. However, if it is common to find only one copy deleted then the deletions may be passengers and it may well be a fragile site.
The authors then used these signatures to look at the deletions that they found in cancers which were currently unexplained. Three regions had the signature of a tumour suppressor gene, but many looked like fragile sites.
"When it became clear in the 1990s that novel tumour suppressor genes could be discovered by looking at genome deletions, many hands were put to the pump," explains Dr Andy
|Contact: Don Powell|
Wellcome Trust Sanger Institute