SEQUENCE SLIDER: integration of structural and genetic data to characterize isoforms from natural sources

Abstract

Proteins isolated from natural sources can be composed of a mixture of isoforms with similar physicochemical properties that coexist in the final steps of purification. Yet, even where unverified, the assumed sequence is enforced throughout the structural studies. Herein, we propose a novel perspective to address the usually neglected sequence heterogeneity of natural products by integrating biophysical, genetic and structural data in our program SEQUENCE SLIDER. The aim is to assess the evidence supporting chemical composition in structure determination. Locally, we interrogate the experimental map to establish which side chains are supported by the structural data, and the genetic information relating sequence conservation is integrated into this statistic. Hence, we build a constrained peptide database, containing most probable sequences to interpret mass spectrometry data (MS). In parallel, we perform MS de novo sequencing with genomic-based algorithms to detect point mutations. We calibrated SLIDER with Gallus gallus lysozyme, whose sequence is unequivocally established and numerous natural isoforms are reported. We used SLIDER to characterize a metalloproteinase and a phospholipase A2-like protein from the venom of Bothrops moojeni and a crotoxin from Crotalus durissus collilineatus. This integrated approach offers a more realistic structural descriptor to characterize macromolecules isolated from natural sources.