LONDON - A mission to sequence certain genes from more than 100 kidney cancer samples has identified several groups of mutations involved in packaging chromatin, the three-dimensional mix of proteins that support DNA.

As well as casting light on the genetic causes of that type of cancer, the findings provide fresh targets for the development of new cancer treatments.

The project forms part of a program of work aimed at exploring cancer genomes being undertaken by the Cancer Genome Project (CGP) at the Wellcome Trust Sanger Institute in Hinxton, UK. The CGP is one part of the International Cancer Genome Consortium, an international collaboration that aims to catalog genomic abnormalities - including mutations and abnormal gene expression - in 50 different cancer types.

For the study, researchers examined tumors from people with clear cell renal cell carcinoma (ccRCC). That is the most common type of kidney cancer, with 209,000 new cases and 102,000 deaths worldwide each year. They chose that type of kidney cancer because its appearance when stained and examined under the microscope is highly uniform. In addition, they already knew that a certain gene, called VHL, was mutated in the majority of cases.

A report of the study appeared in the Jan. 6, 2010, issue of Nature in a paper titled: "Systematic sequencing of renal carcinoma reveals inactivation of histone-modifying genes."

Mike Stratton, co-leader of the Cancer Genome Project, said: "This extensive mutational screen of clear cell renal cancer has uncovered abnormalities in genes that encode the machinery, which determines the structure of chromatin - the three-dimensional environment of proteins that hold DNA. This machinery is critical to the fine regulation of transcription of genes into RNA and therefore to the functioning of all cells. We have now shown that several components of the machinery are mutated in cancer and the consequent abnormalities in their function contribute to the development of the disease."

Andy Futreal, also co-leader of the Cancer Genome Project, said: "We knew there were very few examples of known 'cancer genes' being mutated in ccRCC, which made it a very interesting tumor in which to explore the contribution of novel genes in human cancer, and thus provide opportunities to more fully understand the processes that can be subverted to drive cancer."

In all, the researchers sequenced 3,544 genes in the 101 samples of ccRCC that they studied. As expected, they found that there were mutations in the VHL gene in more than half of the cases. But they also found many other mutations in other genes.

For example, there were inactivating mutations in ccRCC in two genes thought to modulate control of gene expression throughout the genome. Those genes, called JSRID1C and SETD2, encode proteins that methylate and demethylate histones, which form part of the chromatin packaging of DNA. Methylation and demethylation alter the functions of the histones, although more research will be needed to find out exactly how.

The discovery of the mutations in the two genes fits with the results of an earlier study by the same team, which found that mutations in a gene called UTX also were associated with ccRCC and two other cancers. UTX also controls histone activity.

"This paper shows clearly that the methylation and demethylation machinery that is controlling the structure of chromatin is critically implicated in the development of human cancer," Stratton told BioWorld International. "The identification of this machinery offers a completely new set of drug targets for the development of novel anticancer therapies."

The study also identified mutations in the gene NF2 in some cases of ccRCC. Mutations in that gene are known to cause the disease neurofibromatosis 2, an inherited condition that is associated with tumors of the nervous system. All of the mutations in NF2 in the ccRCC samples were in tumors that did not have mutations in VHL.

The researchers were surprised to find, however, that despite the presence of the same mutation in VHL in most cases of ccRCC and the homogeneous microscopic appearance of the tumor, none of the additional mutations accounted for more than 5 percent of cases.

"Even in a tumor type dominated by a single histology, with a prevalent signature, even in this clearest of cases, we see evidence for substantial genetic heterogeneity," Futreal said.

Because some mutations had been found only once or twice in such a large set of more than 100 ccRCC samples, he added, it was possible that the study had missed other mutations.

"To search comprehensively - to build the full picture we need of cancer genetics - we will need many hundred samples sequenced through the entire genome and all genes. The target for those cancers being analyzed by the International Cancer Genome Consortium is 500. This is the scale that will be required to really take this group of diseases apart," Futreal added.