NEW YORK (GenomeWeb) – A new method for genetic interaction screening has uncovered 124 new synthetic lethality combinations in several cancer cell lines.
Led by University of California, San Diego professors Trey Ideker and Prashant Mali, and joined by scientists from Stanford University and UCSF, the researchers used CRISPR/Cas9 to knock out all possible combinations of 73 genes, two at a time, in three different cell lines — HeLa cells, A549 KRAS mutant lung cancer cells, and 293T transformed embryonic kidney cells. The 73 genes included tumor-suppressor genes, cancer relevant drug targets, and verified oncogenes.
As they noted today in Nature Methods, the team found 152 total synthetic lethalities, 28 of which had been previously reported, out of a total of 141,912 interactions. Interactions were evaluated using a new computational analysis pipeline.
They partially validated their results with eight assays using two-drug combinations targeting the genes, instead of CRISPR knockout. Of those, six replicated the growth pattern, or lack thereof, seen in the CRISPR screen.
"Identifying underlying genetic interactions in this way can reveal important functional relationships between genes, such as contributions to the same protein complex or pathway," Ideker, a co-senior author on the study, said in a statement. "This in turn can impact both our fundamental understanding of biological systems, as well as therapeutics development."
The researchers said the results underscored the potential for leveraging synthetic lethality, the scenario where the combined loss of function in two enzymes spells doom for the cell when loss of just one does not.
"The [AstraZeneca] ovarian cancer drug [Lynparza (olaparib)] works by synthetic lethality — it inhibits a gene that, when a BRCA gene is also mutated, kills just those cancer cells," UCSD postdoc and co-first author John Paul Shen, said in a statement. The PARP1-BRCA synthetic lethality was among the gene-gene interactions uncovered by the screen. "Many other cancers could likely be treated this way as well, but we don't yet know which gene mutation combinations will be synthetic-lethal."
Gene-gene interaction networks and synthetic lethality has been an area of active research in pioneering labs across the country, including Michael Bassik's lab at Stanford and David Sabatini's lab at the Whitehead Institute. Both have used CRISPR/Cas9 screening to find gene-gene interactions in cancer that could be used to develop new therapeutics or new treatment regimens using existing US Food and Drug Administration-approved therapeutics.
But the new paper also suggested that synthetic lethality featuring a gene-gene interaction is not universal and may be limited to specific cell types. Only 16 of the synthetic lethalities were identified in more than one cell line and none were found in all three. The remaining 136 were only found in one of the three cell lines.
"Recognizing that there may be great diversity in genetic interactions among different tumors, it will be important to perform these studies across a large number of samples," the authors wrote, noting that their high-throughput method would enable that approach. "By allowing for genetic-interaction mapping directly in eukaryotic cells, our combinatorial CRISPR/Cas9 technology may pave the way for systematic determination of cancer pathways, with twofold applications: improving understanding of how networks of genes influence tumorigenesis and aiding in the development of precision therapeutics via new druggable synthetic-lethal interactions."
Drug-protein interaction was even brought into the study to validate some of the hits from the CRISPR screen. The researchers used drug combination assays targeting eight gene pairs and were able to recreate six of the eight interactions.
"Moving forward, we intend to further refine our technology platform and make it more robust," Mali said. "And we are scaling our cancer genetic networks maps so we can systematically identify new combination therapies."