RASSF2 Under the Computational Microscope: Prediction of Destabilizing Missense Variants

Abstract

Background: RASSF2 is a tumor suppressor gene involved in multiple cancers such as colorectal and lung cancer. Different studies have focused on promoter hypermethylation as the main inactivation pathway of this gene, which leads to cancer proliferation and metastasis. Here, using a comprehensive in silico pipeline, we investigate, for the first time, the impact of missense variants on protein structure, stability, and interactions. Methods: Multitiered computational tools were used to predict the most likely pathogenic and destabilizing nsSNPs; the impact of coding mutations on different gene expression was analyzed, and the impact of nsSNPs on PPIs using mCSM-PPI2 and MutPred-PPI was analyzed using predicted AlphaFold protein complexes. Results: Of 454 variants, 19 variants were consistently predicted to be likely pathogenic and destabilizing, 14 variants were in the RA domain, and two variants were in the SARAH domain. V193D, L235P, L247P, and Y275C were conserved and exhibited severe structural and molecular alterations. Also, Y275C was at the interface, and it was predicted to disrupt the RASSF2-MST2 complex. KRT23 and BCL2L14 were significantly downregulated in samples containing mutated RASSF2 in COAD and LUAD, respectively. STRING illustrates the Hippo signaling pathway as the main pathway involved in the network. Conclusions: These findings provide promising candidates that can be important for future functional investigations. However, in vivo and in vitro experiments to validate our findings are necessary.