Back

Structural studies of MMP-3 interaction with triple-helical collagen introduce the enzyme’s new roles in tissue remodelling

Manka, S. W.; Bihan, D.; Farndale, R. W.

2019-09-03 biochemistry
10.1101/756460 bioRxiv
Show abstract

Matrix metalloproteinase-3 (MMP-3 or stromelysin 1) participates in normal extracellular matrix (ECM) turnover during embryonic development, organ morphogenesis and wound healing, and in tissue-destructive diseases, such as aneurysm, cancer, arthritis and heart failure. Despite its ability to hydrolyse numerous proteins in the ECM, MMP-3 fails to cleave the triple helix of interstitial fibrillar collagens. Nonetheless, it can still bind to these collagens although the mechanism, location and role of binding are not known. We used the Collagen Toolkits, libraries of triple-helical peptides that embrace the entire helical domains of collagens II and III, to map MMP-3 interaction sites. The enzyme recognises five sites on collagen II and three sites on collagen III. They share a glycine-phenylalanine-hydroxyproline/alanine (GFO/A) motif that is recognised by the enzyme in a context-dependent manner. Neither MMP-3 zymogen (proMMP-3) nor the individual catalytic (Cat) and hemopexin (Hpx) domains of MMP-3 interact with the peptides, revealing cooperative binding of both domains to the triple helix. The Toolkit binding data combined with molecular modelling enabled us to deduce the putative collagen-binding mode of MMP-3, where all three collagen chains make contacts with the enzyme in the valley running across both Cat and Hpx domains. The observed binding pattern casts light on how MMP-3 could regulate collagen turnover and compete with various collagen-binding proteins regulating cell adhesion and proliferation.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.