Top killers share genes

Cancer, obesity, and even atherosclerosis share a common set of differentially expressed genes, suggesting a diverse number of human diseases share the same disrupted biological pathways, according to new research published this week in Cancer Cell. The genetic link also suggests that drugs currently used for the treatment of metabolic and cardiovascular diseases might also be used against cancer, researchers say.

DNA microarray
Image: Wikimedia commons,
Guillaume Paumier

"In any year, there are probably ten big papers in a field that help push a concept forward," said linkurl:Reuben Shaw,;http://www.salk.edu/faculty/shaw.html an HHMI investigator at the Salk Institute in California who was not involved in the research. "I think this is one of them." linkurl:Kevin Struhl;https://struhl.med.harvard.edu/ and colleagues at Harvard Medical School compared the transcription profiles of two genetically identical human mammary epithelial cell lines from the same individual -- one normal and one cancerous, or transformed, through the addition of an oncoprotein -- to determine which genes were differentially expressed in the transformed cells. They then performed the same experiment with human fibroblasts, and juxtaposed the two models to identify 343 genes that were differentially expressed in both models -- a common "cancer gene signature." "The use of the two systems made a huge difference," said Struhl. With a single model, researchers isolate thousands of genes that appear to be expressed differently in cancerous cells versus normal cells. But by comparing data sets from two totally different cell models, mammary cells and fibroblasts, "it cut down the list of genes from thousands to hundreds. And hundreds is something you can really work with," added Struhl. Once the team had a list of 343 genes, they set out to identify them. Using basic bioinformatics techniques, they identified several families of transcription factors involved in inflammation and lipid metabolism. "I was completely stunned," said Struhl. A significant amount of research links inflammation and cancer, but the association between metabolic disease and cancer is much less understood. Scanning the literature, the researchers found that many of the 343 genes they implicated in cancer have already been implicated in diabetes, obesity, heart disease, and other top killers. To validate those metabolic genes that had not previously been linked to cancer, Struhl and his team knocked down the genes prior to transformation and found seven of eleven were important to the development of cancer in both experimental models. In additional work, they demonstrated that low-density lipoprotein (LDL) and its receptor (OLR1), both highly involved in atherosclerosis, also play a role in cancer. "That seems very novel, very unexpected," said Shaw, who studies genes at the interface between cancer and diabetes. "This further illustrates how broadly some of these proteins are involved in different disease states." Looking at the results, Struhl believes there may be common molecular pathways underlying many disparate diseases. "The idea is that there's essentially a diseased cell state, but the specific phenotype and disease you get depends on what cell type it is," said Struhl. It's a broad idea, he admitted, which may be part of the reason it took him two years to get the paper published. As a crude test of that idea, the team tested drugs for other diseases -- including metformin (diabetes), celecoxib (arthritis), and simvastatin (atherosclerosis) -- against cancerous cells. "If all these diseases have some common pathways, then drugs against these diseases should also work against cancer," said Struhl. Eleven out of thirteen drugs inhibited cancer development in the cell lines. The drugs had a dramatic effect on the cancer cells, said linkurl:Sam Hanash,;http://www.fhcrc.org/science/phs/molec/faculty/hanash.html a molecular diagnostician at the Fred Hutchinson Cancer Research Center who was not part of the study. "The findings are very interesting, and we should give more consideration to those agents in developing strategies for chemoprevention," he said. Last October, Struhl published linkurl:additional results;http://www.ncbi.nlm.nih.gov/pubmed/19752085 in Cancer Research demonstrating that metformin selectively killed cancer stem cells in four different types of breast cancer. "A lot of work is going on concerning the role of metabolism in cancer development," said Hanash. "I think we will see more and more similar studies."
**__Related stories:__***linkurl:Cancer genetics gets personal;http://www.the-scientist.com/blog/display/57150/
[18th February 2010]*linkurl:Antioxidants help cancer cells?;http://www.the-scientist.com/blog/display/55896/
[19th August 2009]*linkurl:Computing Gene Regulation;http://www.the-scientist.com/article/display/14772/
[21th June 2004]