Discovery of Enhanced Dependency on the Enzyme PRMT5 in Cancer Cells Bearing Deletion of MTAP.MTAP, located on 9p21 adjacent to the CDKN2A/CDKN2B locus, which is commonly deleted in cancer, encodes a critical metabolic enzyme that cleaves MTA to generate precursor substrates for methionine and adenine salvage pathways. Interestingly, PRMT5 is inhibited by MTA (due to the fact that MTA acts as a competitive inhibitor of SAM which is required for PRMT5 function) and is therefore partially inhibited in tumors lacking MTAP. Because PRMT5 is essential for cell survival, cells lacking MTAP are especially sensitive to pharmacologic inhibition of PRMT5. MTA, methylthioadenosine; MTAP, methylthioadenosine phosphorylase; MTR-1-P, methylthioribose-1-phosphate; PRMT5, protein arginine methyltransferase 5; SAM, s-adenosyl methionine.
Discovery of Enhanced Dependency on the Enzyme PRMT5 in Cancer Cells Bearing Deletion of MTAP.MTAP, located on 9p21 adjacent to the CDKN2A/CDKN2B locus, which is commonly deleted in cancer, encodes a critical metabolic enzyme that cleaves MTA to generate precursor substrates for methionine and adenine salvage pathways. Interestingly, PRMT5 is inhibited by MTA (due to the fact that MTA acts as a competitive inhibitor of SAM which is required for PRMT5 function) and is therefore partially inhibited in tumors lacking MTAP. Because PRMT5 is essential for cell survival, cells lacking MTAP are especially sensitive to pharmacologic inhibition of PRMT5. MTA, methylthioadenosine; MTAP, methylthioadenosine phosphorylase; MTR-1-P, methylthioribose-1-phosphate; PRMT5, protein arginine methyltransferase 5; SAM, s-adenosyl methionine.
Deletion of the 9p21 locus, encompassing the tumor suppressors CDKN2A and CDNK2B, is one of the most common genetic events in cancer, yet there are no therapies targeting cancer cells with this alteration. Interestingly, two recent studies have identified synthetic lethal interactions created by deletion of genes in the neighboring genome around this 9p21 locus. Homozygous deletions of CDKN2A are commonly found in multiple solid tumors as well as in roughly 25 percent of diffuse large B-cell lymphoma cases, 21 percent of patients with B-cell acute lymphocytic leukemia, and 50 percent of cases of therapy-related acute lymphocytic leukemia. Seemingly incidentally, a nearby gene encoding methylthioadenosine phosphorylase (MTAP) is frequently co-deleted due to its close proximity to CDKN2A (Figure). The findings reported by Dr. Gregory Kryukov and colleagues1 show that MTAP deletion results in selective dependence on protein arginine methyltransferase 5 (PRMT5) and its binding partner WDR77, generating a therapeutic vulnerability that can be genetically or pharmacologically targeted. These results were independently verified by Dr. Konstantinos Mavrakis and colleagues2 in an article simultaneously released online.
In their study, Dr. Kryukov and colleagues used data from a large genome-wide short-hairpin RNA (shRNA) screen known as Project Achilles.3 This project, completed in 2014, used 50,529 shRNAs across 216 cancer cell lines. The authors inferred MTAP deletion status from copy number and mRNA expression data in the Cancer Cell Line Encyclopedia and categorized the cell lines into two groups: MTAP-deleted (n=50) or normal MTAP (n=166). This allowed them to correlate shRNA sensitivity profiles with MTAP status and identified two hits with selectivity for MTAP-deleted cell lines. One shRNA targeted PRMT5 and the other targeted WDR77. Interestingly, the MTAP-deleted cells showed cross-sensitivity to either shRNA, and indeed, PRMT5 and WDR77 are known to interact in a methyltransferase complex called the methylosome.
To confirm that the sensitivity was due to MTAP deletion and not to some other confounder, the authors reintroduced MTAP into four MTAP-deleted cell lines and then knocked down PRMT5 and WDR77. As hypothesized, reconstitution of MTAP in MTAP-deleted cell lines rescued the sensitivity to PRMT5 or WDR77 suppression. Next, the authors sought to determine the link between MTAP deletion and sensitivity to PRMT5/WDR77 inhibition. To do this, they followed prior studies, which revealed that methylthioadenosine (MTA), the substrate for MTAP, might inhibit PRMT5 activity. Measurement of intracellular MTA levels from 40 MTAP-deleted and normal MTAP cell lines demonstrated a strong correlation between MTA levels and MTAP loss (Figure). Furthermore, they showed that MTA does inhibit PRMT5 by competing for its substrate, s-adenosyl methionine, and observed a greater than 100-fold selectivity of MTA for PRMT5 compared with 30 other methyltransferases.
In Brief
The above findings about genetic dependence on PRMT5 in MTAP-deficient cancer cells serendipitously converge with the recent development of a first-in-class small molecule inhibitor of PRMT5. Given this, both groups substituted a pharmacologic small molecule inhibitor of PRMT5 for the shRNA and again tested for selectivity in isogenic cell lines with and without MTAP. They found a modest difference favoring sensitivity of the MTAP-deleted cell lines towards PRMT5 inhibition; however, this was not seen in all pairs, and the mean IC50 difference did not reach statistical significance. Whether this was due to the efficacy of the drug at inhibiting PRMT5 or an alternate function of PRMT5 was not clear and was not investigated in this study. Nonetheless, this report establishes a novel cancer susceptibility due to co-deletion of MTAP with CDKN2A and exemplifies the need for detailed functional characterization of molecular alterations to guide development of new treatment strategies. This also opens up the field to studying other common “passenger” alterations for prospective cancer susceptibilities.
