Germline mutations in mismatch repair genes predispose individuals to Lynch Syndrome, the most common colorectal cancer predisposition syndrome. MLH1 is a key mismatch repair gene that is mutated in Lynch syndrome and about a third of the genetic alterations identified in MLH1 are missense variants of unclassified clinical significance. We hypothesize that missense alterations in distinct domains of MLH1 likely affect its expression and function(s) to varying degrees. To address this we utilized several approaches to investigate the molecular basis of the pathogenicity of a panel of unclassified variants. Our results demonstrate that the MLH1 variants p.R265C and p.K618A significantly decrease the stability of the MLH1 protein, while the variant p.L749Q compromises heterodimerization of the MLH1-PMS2 complex. Given the limitations and complexity of in vitro assessment strategies, we conducted a proof-of-principle study to investigate whether missense variants in MLH1 lead to allelic imbalances at the transcriptional level. Our analysis using the PeakPicker software indicated that the missense variants c.350C>T, c.793C>T, c.1852_1853AA>GC, as well as the truncating variant c.1528C>T were associated with significantly unbalanced allelic expression. This illustrates a novel method of investigating the pathogenicity of unclassified genetic variants, which has the potential to be applied in the diagnostic setting. Previous genetic epidemiology studies from our laboratory have demonstrated that the MLH1 c.-93G>A promoter variant is strongly associated with the microsatellite instability phenotype in colorectal tumours. Additionally, this promoter variant was associated with an elevated risk of endometrial cancer in case-control studies. Results from our functional studies indicate that the c.-93G>A variant significantly alters the promoter activity of MLH1. The MLH1 promoter is bi-directional with the EPM2AIP1 gene located on the antisense strand. Interestingly, we observed that this variant significantly affected EPM2AIP1 transcription as well. Furthermore, our experiments suggest that c.-93G>A variant affects transcription by altering the affinity of nuclear factors that bind this region. Combined, these findings shed light on the possible mechanisms by which missense variants affect MLH1 expression and function, which in conjunction with results from other functional assays will help increase the accuracy and efficiency of genetic testing of inherited cancers.

Germline mutations in mismatch repair genes predispose individuals to Lynch Syndrome, the most common colorectal cancer predisposition syndrome. MLH1 is a key mismatch repair gene that is mutated in Lynch syndrome and about a third of the genetic alterations identified in MLH1 are missense variants of unclassified clinical significance. We hypothesize that missense alterations in distinct domains of MLH1 likely affect its expression and function(s) to varying degrees. To address this we utilized several approaches to investigate the molecular basis of the pathogenicity of a panel of unclassified variants. Our results demonstrate that the MLH1 variants p.R265C and p.K618A significantly decrease the stability of the MLH1 protein, while the variant p.L749Q compromises heterodimerization of the MLH1-PMS2 complex. Given the limitations and complexity of in vitro assessment strategies, we conducted a proof-of-principle study to investigate whether missense variants in MLH1 lead to allelic imbalances at the transcriptional level. Our analysis using the PeakPicker software indicated that the missense variants c.350C>T, c.793C>T, c.1852_1853AA>GC, as well as the truncating variant c.1528C>T were associated with significantly unbalanced allelic expression. This illustrates a novel method of investigating the pathogenicity of unclassified genetic variants, which has the potential to be applied in the diagnostic setting. Previous genetic epidemiology studies from our laboratory have demonstrated that the MLH1 c.-93G>A promoter variant is strongly associated with the microsatellite instability phenotype in colorectal tumours. Additionally, this promoter variant was associated with an elevated risk of endometrial cancer in case-control studies. Results from our functional studies indicate that the c.-93G>A variant significantly alters the promoter activity of MLH1. The MLH1 promoter is bi-directional with the EPM2AIP1 gene located on the antisense strand. Interestingly, we observed that this variant significantly affected EPM2AIP1 transcription as well. Furthermore, our experiments suggest that c.-93G>A variant affects transcription by altering the affinity of nuclear factors that bind this region. Combined, these findings shed light on the possible mechanisms by which missense variants affect MLH1 expression and function, which in conjunction with results from other functional assays will help increase the accuracy and efficiency of genetic testing of inherited cancers.