New evidence challenges our current understanding about how cancers develop. Current thinking argues that cancer is caused either by oncogenes, normal genes that can cause cancer when inappropriately activated by mutations, or by inactivation, a loss of genes called tumor suppressor genes, again by mutation, which normally function to stop cancer development. Oncogenes drive cells to grow faster and survive longer, while tumor suppressor genes put the brakes on cells to stop their growth and to prevent cancer from forming. In this new study reported in the journal Nature Genetics, a team of researchers proposes that a mechanism that triggers cancer development does not involve oncogene activation, but is largely due to the repression of tumor suppressor genes, or taking the foot off of the brake.

“Oncogenes and tumor suppressor genes have long been involved in tumor development,” said senior author Dr. Wei Li, professor at Baylor College of Medicine and member of the Dan L Duncan Comprehensive Cancer Center. “Traditionally, much attention has been focused on studying mutations in these genes that can lead to cancer development, but this approach has not been sufficient to explain all cancers. We took a different approach.”

Li and his colleagues did not look to identify mechanisms that change the oncogenes or tumor suppressor genes themselves. They searched for mechanisms that could disrupt the normal regulation of their expression. They focused on a phenomenon widely spread in cancer cells called mRNA 3ʹUTR shortening. 

“The prevailing hypothesis, which is based on studies of cancer cell lines grown in the laboratory, is that mRNA 3ʹUTR shortening induces oncogene activation, but recent data has challenged that view,” said co-first author Dr. Ping Ji, assistant professor of biochemistry and molecular biology at The University of Texas Medical Branch in Galveston.  “Here, we studied samples from breast cancer patients and found evidence suggesting that the major role of mRNA 3ʹUTR shortening is in mediating repression of tumor suppressor genes, rather than inducing oncogene activation.”

Due to its complexity, the researchers studied this mechanism by combining two independent approaches.

“We combined a computational biology approach to analyze big data with classical molecular experiments in the laboratory,” said co-first author Dr. Hyun Jung Park, who was a postdoctoral associate in the Li lab during the development of this project and currently is an assistant professor of human genetics at the University of Pittsburgh. “Our statistical model accurately predicted gene expression changes mediated by mRNA 3ʹUTR shortening in human breast cancer.”

The laboratory studies provided a functional validation of these predictions confirming the major role of 3’UTR shortening plays in leading the repression of tumor suppressor genes, including PTEN a gene well-known in breast cancer.

“This study is highly significant because it sheds light on new ways that tumors disrupt gene expression networks. It was shocking to me to see the lengths that cancers such as breast tumors will go in order to reduce the expression of tumor suppressor genes,” said co-senior author Dr. Eric Wagner, associate professor of biochemistry and molecular biology at UTMB. “My lab and the Li lab have collaborated for many years, and I hope that we will continue to drill down on the mechanisms underlying cancer gene expression regulation.”

“We think that our findings can lead to future ways to manipulate the 3’UTR shortening mechanism and thus regulate PTEN and other genes involved in breast cancer to prevent or treat this condition,” Li said.

Other contributors to this work include Soyeon Kim, Zheng Xia, Benjamin Rodríguez, Lei Li, Jianzhong Su, Kaifu Chen, Chioniso P. Masamha, David Baillat, Camilla R. Fontes, Ann-Bin Shyu and Joel R. Neilson.

This work was supported by the US National Institutes of Health (NIH) grants R01HG007538, R01CA193466, RO1GM046454 and U54CA217297, the Cancer Prevention Research Institute of Texas (CPRIT) grant RP150292, CPRIT RP100107, CPRIT RP140800 and the Welch Foundation grant AU-1889.