Elsevier

Toxicon

Volume 42, Issue 7, December 2003, Pages 801-808
Toxicon

Snake venom metalloproteinases: structure/function relationships studies using monoclonal antibodies

https://doi.org/10.1016/j.toxicon.2003.10.010Get rights and content

Abstract

Snake Venom Metalloproteinases (SVMPs) are synthesized as zymogens and undergo proteolytic processing resulting in a variety of multifunctional proteins. Jararhagin is a P-III SVMP, isolated from the venom of Bothrops jararaca, comprising metalloproteinase, disintegrin-like and cysteine-rich domains. The catalytic domain is responsible for the hemorrhagic activity. The disintegrin-like/cysteine-rich domains block α2β1 integrin binding to collagen and apparently enhance the hemorrhagic activity of SVMPs. The relevance of disintegrin-like domain is described in this paper using a series of mouse anti-jararhagin monoclonal antibodies (MAJar 1–7). MAJar 3 was the only antibody able to completely neutralize jararhagin hemorrhagic activity. Neutralization of catalytic activity was partial by incubation with MAJar 1. MAJars 1 and 3 efficiently neutralized jararhagin binding to collagen with IC50 of 330 and 8.4 nM, respectively. MAJars 1 and 3 recognized the C-terminal portion of the disintegrin domain, which is apparently in conformational proximity with the catalytic domain according to additivity tests. These data suggest that disintegrin-like domain epitopes are in close contact with catalytic site or functionally modulate the expression of hemorrhagic activity in SVMPs.

Introduction

Snake venoms have long been recognized for their biochemical complexity. A typical viperid venom, for instance, may contain several thousand proteins (Fox et al., 2002). This has implications for understanding the complex mechanisms involved in the pathology of snakebite envenomations and in the amplitude of antibody specificity necessary for the efficacy of antivenoms. Moreover, the complexity and variety of targets to which venom toxins specifically interact make them important candidates for the design of drugs or as useful biological tools. In these aspects, zinc-dependent metalloproteinases (SVMPs) are relevant components in the venoms of snakes of the family Viperidae (Bjarnason and Fox, 1994). They are responsible for the conspicuous hemorrhage often associated with their bites, representing therefore an important antigen to target in antivenom neutralization. Structurally, they are closely related to the ADAMs (a disintegrin and metalloproteinase), mammalian proteins involved in cell activation, and communication and to MMPs (matrix metalloproteinases) (Fox and Long, 1998).

SVMPs are present in high concentration in venoms of viperid snakes and more recently they have been reported, although in minor proportions, also in venoms of snakes from the family Elapidae (Matsui et al., 2000). They are found in different processing states of similar zymogen multidomain precursors: P-I SVMPs contain only the catalytic domain, while in P-III SVMPs, disintegrin-like/cysteine-rich domains are added to the catalytic domain (Fox and Long, 1998). The catalytic domain shares great functional and structural similarity with the metalloproteinase domain of MMPs (Blundel, 1994), conserving all the zinc-binding residues and structural constraints involved in catalysis. In opposition, the disintegin-like/cysteine-rich domains are found in ADAMs replacing the hemopexin C domain frequently found in secreted MMPs. The specificity for substrates of SVMPs is also similar to MMPs, involving extracellular matrix components located in the basement membrane of the microvasculature. Cleavage of proteins at the basement membrane, with the consequent weakening of the capillary structure, constitutes the mechanism by which SVMPs induce hemorrhage (Gutierrez and Rucavado, 2000). Moreover, catalytic activity of both SVMPs and ADAMs is also related to shedding of cell surface molecules, thus regulating cell communication and activation (Moura-da-Silva et al., 1996, Black et al., 1997).

Jararhagin, an archetypical P-III SVMP isolated from Bothrops jararaca, is one of the main venom components responsible for the local and systemic hemorrhage induced by this venom (Paine at al., 1992). Jararhagin has been shown to degrade extracellular matrix components and also some plasma proteins important for hemostasis (Kamiguti et al., 1996). However, the role of disintegrin-like/cysteine-rich domain in hemorrhagic activity is far from understood. In the venom of B. jararaca, the metalloproteinase domain of jararhagin may be proteolytically processed generating Jararhagin-C, a fragment representing the disintegrin-like and cysteine-rich domains of jararhagin (Moura-da-Silva et al., 2003), which has already been isolated and is an efficient antagonist for collagen-induced platelet-aggregation (Usami et al., 1994). Apparently, in the complete P-III SVMP, these domains act synergistically with the catalytic domain, enhancing hemorrhage by inhibiting platelet aggregation (Kamiguti et al., 1996) and probably by targeting the enzyme to relevant physiological components at basement membranes. Consequently, hemorrhagic activity of P-III SVMPs is generally 10 times higher than that of P-I SVMPs, lacking disintegrin-like/cysteine-rich domains (Bjarnason and Fox, 1994). Recently, Harrison and coworkers (2000) immunized mice with the cDNA coding for the disintegrin-like/cysteine-rich domains of jararhagin. Antibodies obtained as a result of the immunization with jararhagin.-C efficiently neutralized jararhagin-induced hemorrhage, further evidencing the relevance of these additional domains in hemorrhagic activity.

However, structural interactions between catalytic and disintegrin-like/cysteine-rich domains of jararhagin or other SVMPs have never been demonstrated since structural-three-dimensional data are lacking for both P-III SVMPs and ADAMs. In this contribution, we approached this aspect using monoclonal antibodies (MoAbs) directed to jararhagin disintegrin-like/cysteine-rich domains. A neutralizing MoAb described here supported the participation of disintegrin-like domain for expression of hemorrhagic activity.

Section snippets

Jararhagin, jararhagin C, recombinant fragments and BaP1

Jararhagin and jararhagin C were isolated from B. jararaca venom as previously described (Paine et al., 1992, Moura da Silva et al., 2003). BaP1 was isolated from Bothrops asper venom according to Gutierrez et al. (1995). Recombinant fragments corresponding to the jararhagin disintegrin (JD49 and JD89) and cysteine-rich (JC76 and JC116) domains were produced in fusion with Glutathione S-Transferase (GST). Briefly, the cDNA sequences were amplified by the polymerase chain reaction (PCR) from

Characterization of the monoclonal antibodies against Jararhagin (MAJar)

Fusion of myeloma cells SP2O with popliteal lymphocytes of mice immunized with jararhagin resulted in seven immortalized clones secreting anti-jararhagin antibodies: mouse anti-jararhagin antibodies (MAJar) 1–7. The seven MAJars are of IgG1 isotype, presenting dissociation constants (Kd) of antigen–antibody interactions between 10−7 and 10−10, indicating high affinities to jararhagin. MAJars 6 and 7 presented the highest affinity while MAJar 4 presented the lowest. The structure of the epitopes

Discussion

MoAbs have been extensively used for characterizing functional epitopes involved in biological activities of protein antigens. In this paper, we report the production and characterization of seven MoAbs against jararhagin. All antibodies reacted preferentially with jararhagin-C, which comprises the disintegrin-like/cysteine-rich domains of the toxin. Only a minor reaction was observed between MAJar 4 and BaP1, used as a representative of the catalytic domain. It is important to point out that

Acknowledgements

The authors thank Dr Maria Juliano, UNIFESP, Brazil, for supplying the synthetic peptide used in the enzymatic assays and Dr Solange Serrano, Instituto Butantan, Brazil, for her comments and review of the manuscript. The study was supported by FAPESP (grants No. 99/12432-3 and 00/13651-0), CNPq (grant No. 52.0636/1996.1) and the Wellcome Trust (grant No. 062043).

References (26)

Cited by (39)

View all citing articles on Scopus
View full text