Possums for snake venoms

Snakebites are a major problem in tropical developing countries, despite the advent of antidotes. It has been estimated that there are several million cases of snakebite annually worldwide, with up to 100,000 deaths. In Brazil, one of the affected countries, the National Ministry of Health cited more than 28,000 reports of envenomation in 2005, although the real figure is likely to be far higher due to under-reporting.

The development of antidotes requires knowledge of the toxin composition and modern proteomics techniques have been instrumental in this area, helping to unravel the complex venoms. In particular, mass spectrometry is well-placed to identify the proteins and peptides present.

Most snake venoms consist of relatively few protein classes, such as serine proteases or metalloproteases, but there are a myriad of post-translational modifications and large ranges of protein concentrations to complicate analyses. One common technique for simplifying the analyses is some sort of pre-fractionation but a team of scientists in Brazil has noted the paucity of this practice in toxin studies. So, they decided to see what effect this would have on the characterisation of venoms from a series of local pit vipers.

Jonas Perales from the Instituto Oswaldo Cruz in Fiocruz, with colleagues from the Federal University of Rio de Janeiro, all of whom belonged to the Proteome Network of Rio de Janeiro, evaluated an unusual pre-fractionation procedure. They recognised that the big-eared opossum (Didelphis aurita), a native of Brazil, is resistant to snake venom due to particular neutralising factors present in their blood. The factors form complexes with the metalloproteinase toxins from the venoms and prevent them from inflicting internal damage in the victims.

So, the team used one of these factors, an anti-haemorrhagic protein known as DM43, as a purification aid in affinity chromatography to separate the abundant metalloproteinases from the other proteins. Venoms from four pit vipers were analysed. Bothrops atrox (common lancehead) is responsible for most snake bites in the Amazon region. Bothrops jararaca (jararaca) is a dominant threat in the southeast of Brazil. Bothrops insularis (golden lancehead) is closely related to B. jararaca and is found on a small coastal island but its toxins are poorly studied. The fourth snake was Crotalus atrox (western diamondback rattlesnake) from North America.

The venoms were passed though a DM43 affinity column and eluting proteins were detected at 280 nm. The flow-through fractions were separated from the bound fractions, which were eluted and collected separately. This simple process dramatically reduced the complexity of the venom, since 19-41% of the proteins present were bound to DM43, depending on the snake. The bound fraction consists primarily of the metalloproteinases and the proportions are consistent with some literature estimates.

By removing them from the venom, less abundant proteins become enriched, making them easier to analyse and presenting a more complete picture of the proteome. The bound and flow-through protein mixtures were each separated further by 2D gel electrophoresis and the isolated protein spots were cut from the gel and in-gel digested with trypsin for MALDI mass spectrometric analysis of the resulting peptides in MS and tandem MS modes.

The mass spectra confirmed that metalloproteinases and their fragments made up the majority of the DM43-bound proteins, although some phospholipases A2 were also trapped. The metalloproteinase sub-proteomes of B. atrox, B. jararaca and B. insularis were comparable. However, those of B. atrox and C. atrox were markedly different, which was unexpected since the venoms induce similar pathologies.

The flow-through fraction consisted primarily of serine proteinases, C-type lectins, C-type lectin-like proteins and L-amino acid oxidases, Many of these proteins also existed in abundant isoforms, contributing to a heterogeneous sub-proteome. Some metalloproteinases were also present, their passage through the column being attributed to proteolysis or column saturation.

The low number of protein families in the venoms was countered by the number of different isoforms present, with several hundred proteins identified in total from the four venoms. This degree of characterisation would not have been possible without the help of the opossum protein.

The views represented in this article are solely those of the author and do not necessarily represent those of John Wiley and Sons, Ltd.

Snake venom research associate