A research team has created a method for using widely available data to identify proteins that could play key roles in tumor formation. Called master transcription factors (MTFs), these proteins vary with each cancer type and could provide targets for therapies, according to a news release from Cedars-Sinai.
The study was published in Science Advances.
MTFs help control the expression of genes and other transcription factors, which are proteins that signal specific genes to turn "on" or "off," directing the division, growth and death of cells. Like network radio stations sending out signals to their affiliates, MTFs direct the activity of groups of transcription factors.
"Cancers hijack these MTFs and use them to make cells grow more quickly than they should or survive better than they should, allowing tumors to grow," said Kate Lawrenson, PhD, Associate Professor of Obstetrics and Gynecology at Cedars-Sinai. "Each cancer type has its own handful of MTFs. In some cancers, we know what they are, for others, we don't."
Lawrenson explained that the current technology for identifying potential MTFs, known as chromatin immunoprecipitation sequencing (ChIP-seq), isn't widely available and relies on tissue samples, which aren't available for every tumor type. "We have developed a novel method based on data that is more widely available so that it can be used by a much broader group of researchers," she said.
Lawrenson and colleagues from Cedars-Sinai, Dana-Farber Cancer Institute, University of Pennsylvania, MIT, UCLA and St. Jude Children's Research Hospital created an algorithm based on data from RNA sequencing, the most common method currently used to look at gene expression in cells. Large libraries of this data are widely available.
The new algorithm, called Cancer Core Transcription factor Specificity (CaCTS), is based on information from The Cancer Genome Atlas (TCGA), a joint effort between the National Cancer Institute and the National Human Genome Research Institute that includes 9,691 patient samples representing 34 tumor types.
For each of these tumor types, the researchers calculated the expression levels of each of the transcription factors. They then assigned a CaCTS score to each transcription factor for each type of tumor, based on its expression level in that type compared with its levels of expression in the others. For each tumor type, the most uniquely expressed transcription factors received the highest CaCTS scores; transcription factors with the highest CaCTS scores and highest expression levels were identified as candidate MTFs. The CaCTS algorithm yielded candidate MTFs for 34 major tumor types and 140 tumor subtypes.
Using other, larger public datasets, such as the Cancer Dependency Map (DepMap), the team verified that the MTFs they had identified possessed a constellation of necessary features. The MTFs worked together cooperatively, forming a circuit that positively regulated their own expression and that of other MTF proteins that were essential to the survival and growth of the cancer cells.
Using ovarian cancer tumor tissue from patients at Cedars-Sinai, Dana Farber and other participating institutions, the investigators also used ChIP-seq to confirm the effect of their MTF candidates on cancer cells. In this way, they were able to confirm three previously unidentified MTFs for aggressive ovarian cancers.