BBB Seminar: Nathalie Reuter

Nathalie Reuter, Computational Biology Unit, BCCS, University of Bergen

Serine proteinases of the polymorphonuclear neutrophils (PMN), including proteinase 3 (PR3) and human neutrophil elastase (HNE), are involved in proteolytic events associated with inflammation and play a role in several diseases such as emphysema, cystic fibrosis, rheumatoid arthritis and vasculitis. Because of their sequence similarity they have been thought for a long time to have similar localization, ligand specificity and function. However, recent studies indicate that they might have different and yet complementary, physiologic roles. In particular, PR3 has specific intracellular protein substrates resulting in the involvement of the enzyme in regulation of intracellular functions such as proliferation or apoptosis. It behaves as a peripheral membrane protein and its membrane expression is a risk factor in chronic inflammatory diseases. Moreover, PR3 is the preferred target antigen in Wegener's granulomatosis, a particular type of vasculitis. These properties were not observed for HNE.

Using a computational biology approach, we have been able to map the substrate binding sites of both enzymes and propose a sequence motif specific for human PR3. This led us to design a peptide, which was indeed shown by enzymatic assays, to be specific for human PR3 but not HNE. The same computational approach was used to show that peptidic substrates, efficiently and specifically cleaved by human PR3, have a poor interaction pattern with mouse and rat PR3s. We were later able to unravel the structural binding determinants of human PR3 to membranes. All our results have been subsequently validated by experimental observations.

This project illustrates the predictive power of computational biology, but also demonstrates the advantage of tightly integrating theoretical and experimental approaches when addressing complex biological problems.

Host: Aurora Martinez, Department of Biomedicine

BBB Seminars