Homology Modeling and Virtual Screening of Proteins Related to PXE and PXE-like Diseases: Insights for Overlapping Metabolites

Abstract

Background: The molecular etiology of Pseudoxanthoma Elasticum (PXE), an autosomal recessive connective tissue disorder, has become increasingly complex as not only mutations in the ABCC6, but also in ENPP1 and GGCX, can cause resembling phenotypes. Methods: To get insights on the common pathway, the overlapping metabolites for these three proteins were predicted through 3D homology modeling and virtual screening. 3D homology models of ABCC6, ENPP1, and GGCX were generated by the MODELLER program, which were further validated using RAMPAGE and ERRAT servers. Substrate binding sites of ABCC6 were predicted using blind docking of reported in vitro substrates. Results: Virtual screening against the substrate binding site of ABCC6 using metabolites listed in Human Metabolome Databases (HMDB) revealed the best possible substrate of ABCC6. Those listed metabolites were further docked against predicted substrate binding sites of GGCX and ENPP1. Molecular docking and virtual screening revealed a list of 133 overlapping metabolites of these three proteins. Most of them are Phosphatidylinositol (PI), Phosphatidylserine (PS), Diacylglycerol (DAG), phosphatidic acid, oleanolic acid metabolites and were found to have links with calcification. Conclusion: These predicted overlapping metabolites may give novel insights for searching common pathomechanism for PXE and PXE-like diseases.