Atragin

Introduction

'Atragin (MW 49.7 kDa and theoretical pI 8.6) is a Snake Venom MetalloProtease (SVMP) belonging to the ADAM/adamlysin/ reprolysin family and is found in the venom of the Naja Atra, or Chinese cobra(1). It contains a metalloprotease domain (M),disintegrin-like domain(D), and cysteine-rich domain (C), making it a P-III SVMP (2). When injected into the bloodstream, P-III SVMPs cause hemorrhagic effects, inhibit platelet formation and inhibit cell migration activity(3,1). These P-III SVMPs, and especially their cysteine residues, bear very close resemblance to mammalian ADAM (A Disintegrin And Metalloprotease) proteins, which also contain the MDC domain architecture (4). The ADAMs are a group of around 40 transmembrane proteins that have been found so far in mammals, 19 of which genes are found in humans (5,6). These proteins play a role in the production of cytokines and growth factors, such as ADAM 17 which releases tumor necrosis factor alpha (TNFa) as an immunomodulatory and pro-inflammatory cytokine (Moss et al., 1997); and ADAM 9 which has been shown to be involved in the release of heparin-binding EGF, which inhibits the proliferation of neighboring cells (Raab and Klagsbrun, 1997). ADAM proteins have also been shown to be involved with multiple human diseases today, including cancer, asthma, cardiac hypertrophy and SARS (Asakura et al., 2002; Haga et al., 2008; Van Eerdewegh et al., 2002; Wu et al., 1997). Unfortunately, the full structure of the ADAM proteins is not yet available and our current understanding of the ADAM structure is based primarily on viper P-III SVMPs. (Igarashi et al., 2007; Muniz et al., 2008; Takeda et al., 2006, 2007; Zhu et al., 2009). Atragin, however, is an elapid P-III SVMP and provides more clues as to how the MDC domains work together.

Structure

Atragin is a 63x52x67 Å protein that contains a Metalloprotease domain, Disintegrin-like domain, and a cysteine-rich domain(1). The overall architecture of the protein gives a C-shaped structure that is similar to other P-III SVMPs such as VAP1 (Takeda et al., 2006) in that the recognition site of the C domain faces the M domain’s catalytic cleft in a linear orientation, which means the target recognized by the C domain could be the same as the substrate catalyzed by M domain. This is a contrast to other P-III SVMPs which can have an I-shaped structure that contains a non-linear orientation between the C and M domains, in which the target recognized by the C domain can be different than the catalyzed substrate molecule (1).

'Metalloprotease Domain

The Metalloprotease domain (M) consists of six alpha-helixes and 5 beta-sheets and is a metalloendopeptidase. Every endometalloprotease contains a HEXXHXXGXXH strain of residues and the presence of a methionine turn needed for zinc and substrate binding for proteolysis. In Atragin, the binding sequence can be found from aa 341-351. The zinc atom is ligated by the side chains of His341 and 345 on the a-helix, and this allows His 351 at the turn the be repsonislbe for the catalytic reaction (Gomis-Ruth, 2003).

Most P-III SVMPs contain seven cysteine residues, but in Atragin, as in other elapid SVMPs, only six residues are found. This is also thought to help in the regulation of autolytic activity (Fox and Solange, 2005)

'Disintegrin-like Domain

Following the M domain and a linker S-region, is the Disintegrin-like (D) domain. This domain is thought to play an important role in the relative orientation of the M and C domains in P-III SVMPs (Think back to C-shaped versus I-shaped). Atragin maintains the C-shaped arhcitexture as its D-shoulder domain has three disulfide bonds, and its D-arm domain conatins another 3 disulfide bonds, which is similar to other C-shaped proteins (Igarashi et al., 2007; Takeda et al., 2006). One disulfide bond also connects these two subdomains and another connects the D-arm domain to the C- wrist domain (discussed below).

The disulfide bond pattern in the D domain alters the oreintations of the other domains in the ADAM/adamalysin/ reprolysins family (1). The different orientations could explain some of the ADAM enzymatic processes, seeing as the different lengths or altering the disulfide pairs of the D-domain can increase or decrease the size of the cleft and the orientation. This could cause for substrates of different sizes and shapes to able to be cleaved by the M-domain in ADAMs (1).

'Cysteine-rich Domain

The Cysteine-rich (C) domain contains the C-wrist and C-hand domains. The C-hand subdomain is the most intriguing to most researches because it conatins the Hypr Variable Region (HVR). This region is through to play an important role in target selection *since it sits on the inside of the C-shaped strucutres cleft* (Takeda et al., 2006). The C domain of Atragin consists of seven disulfide bonds, one of which is connected to the HVR, and this suggests it may play a functional role. The HVR of Atragin is also though to be responsible for the inhibitory affect on cell migration activity.*