Matrix metalloproteinases - From the cleavage data to the prediction tools and beyond

Abstract

Understanding the physiological role of any protease requires identification of both its cleavage substrates and their relative cleavage efficacy as compared with other substrates and other proteinases. Our review manuscript is focused on the cleavage preferences of the individual matrix metalloproteinases (MMPs) and the cleavage similarity and distinction that exist in the human MMP family. The recent in-depth analysis of MMPs by us and many others greatly increased knowledge of the MMP biology and structural-functional relationships among this protease family members. A better knowledge of cleavage preferences of MMPs has led us to the development of the prediction tools that are now capable of the high throughput reliable prediction and ranking the MMP cleavage sites in the peptide sequences in silico. Our software unifies and consolidates volumes of the pre-existing data. Now this prediction-ranking in silico tool is ready to be used by others. The software we developed may facilitate both the identification of the novel proteolytic regulatory pathways and the discovery of the previously uncharacterized substrates of the individual MMPs. Because now the MMP research may be based on the mathematical probability parameters rather than on either random luck or common sense alone, the researchers armed with this novel in silico tool will be better equipped to fine-tune or, at least, to sharply focus their wet chemistry experiments. This article is part of a Special Issue entitled: Matrix Metalloproteinases edited by Rafael Fridman.

Keywords: CleavPredict; MMPs; cleavage prediction tools; matrix metalloproteinases; software; substrate cleavage; substrate phage display.

Figure 1

An example of the main page of the CleavPredict analysis. The page shows the protein sequence of the substrate protein, the cleavage site sequence with the corresponding PWM score, the mass of the generated N-terminal and C-terminal fragments, the presence (or the absence) of the signal peptide, the most probable subcellular compartment of the protein substrate, the co-expression probability of the protease and its protein substrate, the type and the position of SNPs and post-translational modifications (PTM) and other relevant information. The arrangement of the main page allows for the analysis to be continued via the follow-on structural page (Figure 2) and virtual mass spectrometry (VMS) page (Figure 3).

Figure 2

An example of the structure page of the CleavPredict analysis. The page shows the sequence and the structure of the substrate protein with the color-coded positions of SNPs and post-translational modifications (PTM) relative to the P1 position of the cleavage sequence.

Figure 3

An example of the CleavPredict virtual mass spectrometry analysis (VMS button at the main page). The page shows the sequence of the protein with the predicted scissile bonds and the sequence, the mass and the “intensity” parameter of the cleavage fragments. The intensity parameter is the probability of the fragment release and it is based on the primary PWM scores for the N- and C-terminal cleavage sites for an individual peptide fragment.

Figure 4

The number of the unique and common cleavage sites, and substrate proteins for MMP-2, MMP-9 and MMP-14/MT1-MMP in the secreted soluble and membrane proteins encoded by the human genome. Soluble proteins exhibit a signal peptide but lack a transmembrane domain. Membrane proteins exhibit both a signal peptide and a transmembrane domain.