3p1v

Protein Summary

First structure of M64 metallopeptidase. This structure contains a N-terminal imminoglobin-like domain and a C-terminal domain which likely play a role in specificity. The C-terminal domain contains another zinc. Based on the structure, a region around within the C-terminal domain  is flexible and very likely play a role of trapping substrate (specificity loop). The N-terminal imminoglobin-like domain appears to have a unique topology, this needs to be examined more carefully.

The IgA protease from Clostridium ramosum is longer and share the 2nd and 3rd domain (not the first domain). Highly homologous proteins are also present in other Bacteriodes species.

This is a plan to experimentally verify that this protein can indeed use IgA as substrate (experiments had shown that this protein does not cleave IgA).

Information about Bacteriodes infection: http://emedicine.medscape.com/article/233339-overview.

Figure. 1. Structure of a putative Bacteriodes IgA peptidase

Fig 2. A highly restrictive active site defined by the C-terminal domain

From Neil Rawlings:

From: Neil Rawlings <ndr@sanger.ac.uk>
To: Alex Bateman <agb@sanger.ac.uk>
CC: Xu, Qingping <qxu@slac.stanford.edu>

The known peptidase in M64 is from Clostridium ramosum and as part of 
the infection process the peptidase inactives IgA.  There are several 
bacterial toxins that do this, cleaving within the polyproline 
hinge-region.  The Clostridium peptidase in particular cleaves Pro+Val 
bonds in human IgA1 and IgA2 (Kosowska et al., 2002).  So there is no 
surprise that a large domain from this structure is immunoglobulin-like, 
interactions with this domain is presumably how the specificity is 
restricted to IgA.  The structure of the peptidase domain is also not a 
surprise, family M64 has been in the same clan as adamalysins probably 
since it was created in MEROPS.  The catalytic zinc-binding motifs are 
similar in both families.  The third catalytic zinc ligand in family M64 
is Asp (within the motif HEXXHXXGXXD), which is more unusual (usually 
His), but peptidases with Asp in this position have been crystallized 
previously: snapalysin (MEROPS ID M07.001; PBD 1C7K, 1KUH), deuterolysin 
(MEROPS ID M35.002; PDB 1EB6), peptidyl-Lys metalloendopeptidase (MEROPS 
ID M35.004; PDB 1G12, 1GE5, 1GE6, 1GE7).

The source organism for this protein is Bacteroides ovatus, which is a 
normal human intestinal commensal.  Clostridium ramosum is also a 
component of the normal human gut flora (as well as other human 
cavities), and the IgA peptidase is only present in some strains 
(Kosowska et al., 2002).  Clostridium ramosum can become pathogenic in a 
number of diseases including appendicitis, liver abscesses, septicaemia, 
mastitis.  I can find no mention of Bacteroides ovatus being pathogenic, 
so why it would possess an IgA peptidase is an interesting question. 
There are several families of IgA proteinases, including some with 
different catalytic types (serine-type in MEROPS family S6), but none 
have ever been crystallized before to my knowledge.  Cleaving at prolyl 
bonds is unusual so understanding how the substrate binds would also be 
interesting.  Is it possible to predict substrate binding pockets?

Reference: Kosowska, K., Reinholdt‖, J., Rasmussen, L. K., Sabat, A., 
Potempa, J., Kilian, M. & Poulsen, K. (2002) The Clostridium ramosum IgA 
Proteinase Represents a Novel Type of Metalloendopeptidase.  J. Biol. 
Chem. 277, 11987-11994.

Ligand Summary