Snake venoms are integrated systems, but abundant venom proteins evolve more rapidly.

BACKGROUND: While many studies have shown that extracellular proteins evolve rapidly, how selection acts on them remains poorly understood. We used snake venoms to understand the interaction between ecology, expression level, and evolutionary rate in secreted protein systems. Venomous snakes employ well-integrated systems of proteins and organic constituents to immobilize prey. Venoms are generally optimized to subdue preferred prey more effectively than non-prey, and many venom protein families manifest positive selection and rapid gene family diversification. Although previous studies have illuminated how individual venom protein families evolve, how selection acts on venoms as integrated systems, is unknown. RESULTS: Using next-generation transcriptome sequencing and mass spectrometry, we examined microevolution in two pitvipers, allopatrically separated for at least 1.6 million years, and their hybrids. Transcriptomes of parental species had generally similar compositions in regard to protein families, but for a given protein family, the homologs present and concentrations thereof sometimes differed dramatically. For instance, a phospholipase A2 transcript comprising 73.4 % of the Protobothrops elegans transcriptome, was barely present in the P. flavoviridis transcriptome (<0.05 %). Hybrids produced most proteins found in both parental venoms. Protein evolutionary rates were positively correlated with transcriptomic and proteomic abundances, and the most abundant proteins showed positive selection. This pattern holds with the addition of four other published crotaline transcriptomes, from two more genera, and also for the recently published king cobra genome, suggesting that rapid evolution of abundant proteins may be generally true for snake venoms. Looking more broadly at Protobothrops, we show that rapid evolution of the most abundant components is due to positive selection, suggesting an interplay between abundance and adaptation. CONCLUSIONS: Given log-scale differences in toxin abundance, which are likely correlated with biosynthetic costs, we hypothesize that as a result of natural selection, snakes optimize return on energetic investment by producing more of venom proteins that increase their fitness. Natural selection then acts on the additive genetic variance of these components, in proportion to their contributions to overall fitness. Adaptive evolution of venoms may occur most rapidly through changes in expression levels that alter fitness contributions, and thus the strength of selection acting on specific secretome components.

Quantitative high-throughput profiling of snake venom gland transcriptomes and proteomes (Ovophis okinavensis and Protobothrops flavoviridis).

BACKGROUND: Advances in DNA sequencing and proteomics have facilitated quantitative comparisons of snake venom composition. Most studies have employed one approach or the other. Here, both Illumina cDNA sequencing and LC/MS were used to compare the transcriptomes and proteomes of two pit vipers, Protobothrops flavoviridis and Ovophis okinavensis, which differ greatly in their biology. RESULTS: Sequencing of venom gland cDNA produced 104,830 transcripts. The Protobothrops transcriptome contained transcripts for 103 venom-related proteins, while the Ovophis transcriptome contained 95. In both, transcript abundances spanned six orders of magnitude. Mass spectrometry identified peptides from 100% of transcripts that occurred at higher than contaminant (e.g. human keratin) levels, including a number of proteins never before sequenced from snakes. These transcriptomes reveal fundamentally different envenomation strategies. Adult Protobothrops venom promotes hemorrhage, hypotension, incoagulable blood, and prey digestion, consistent with mammalian predation. Ovophis venom composition is less readily interpreted, owing to insufficient pharmacological data for venom serine and metalloproteases, which comprise more than 97.3% of Ovophis transcripts, but only 38.0% of Protobothrops transcripts. Ovophis venom apparently represents a hybrid strategy optimized for frogs and small mammals. CONCLUSIONS: This study illustrates the power of cDNA sequencing combined with MS profiling. The former quantifies transcript composition, allowing detection of novel proteins, but cannot indicate which proteins are actually secreted, as does MS. We show, for the first time, that transcript and peptide abundances are correlated. This means that MS can be used for quantitative, non-invasive venom profiling, which will be beneficial for studies of endangered species.

Domain loss enabled evolution of novel functions in the snake three-finger toxin gene superfamily.

Three-finger toxins (3FTXs) are a functionally diverse family of toxins, apparently unique to venoms of caenophidian snakes. Although the ancestral function of 3FTXs is antagonism of nicotinic acetylcholine receptors, redundancy conferred by the accumulation of duplicate genes has facilitated extensive neofunctionalization, such that derived members of the family interact with a range of targets. 3FTXs are members of the LY6/UPAR family, but their non-toxin ancestor remains unknown. Combining traditional phylogenetic approaches, manual synteny analysis, and machine learning techniques (including AlphaFold2 and ProtT5), we have reconstructed a detailed evolutionary history of 3FTXs. We identify their immediate ancestor as a non-secretory LY6, unique to squamate reptiles, and propose that changes in molecular ecology resulting from loss of a membrane-anchoring domain and changes in gene expression, paved the way for the evolution of one of the most important families of snake toxins.

Introduction to the Toxins Special Issue on Identification and Functional Characterization of Novel Venom Components.

Throughout most of the 20th century, the toxinological literature consisted largely of pharmacological and functional characterizations of crude venoms and venom constituents, often constituents that could not be identified unambiguously [...].

Snake venom NAD glycohydrolases: primary structures, genomic location, and gene structure.

NAD glycohydrolase (EC 3.2.2.5) (NADase) sequences have been identified in 10 elapid and crotalid venom gland transcriptomes, eight of which are complete. These sequences show very high homology, but elapid and crotalid sequences also display consistent differences. As in Aplysia kurodai ADP-ribosyl cyclase and vertebrate CD38 genes, snake venom NADase genes comprise eight exons; however, in the Protobothrops mucrosquamatus genome, the sixth exon is sometimes not transcribed, yielding a shortened NADase mRNA that encodes all six disulfide bonds, but an active site that lacks the catalytic glutamate residue. The function of this shortened protein, if expressed, is unknown. While many vertebrate CD38s are multifunctional, liberating both ADP-ribose and small quantities of cyclic ADP-ribose (cADPR), snake venom CD38 homologs are dedicated NADases. They possess the invariant TLEDTL sequence (residues 144-149) that bounds the active site and the catalytic residue, Glu228. In addition, they possess a disulfide bond (Cys121-Cys202) that specifically prevents ADP-ribosyl cyclase activity in combination with Ile224, in lieu of phenylalanine, which is requisite for ADPR cyclases. In concert with venom phosphodiesterase and 5'-nucleotidase and their ecto-enzyme homologs in prey tissues, snake venom NADases comprise part of an envenomation strategy to liberate purine nucleosides, and particularly adenosine, in the prey, promoting prey immobilization via hypotension and paralysis.

The complete mitochondrial genome of the aquatic coralsnake Micrurus surinamensis (Reptilia, Serpentes, Elapidae).

In this study, we report the first complete mitochondrial genome sequence of the Aquatic Coralsnake Micrurus surinamensis. The mitochondrial genome lengthis 17,375 bp, comprising 13 protein-coding genes, 2 rRNA (12S and 16S) and 22 tRNA, as well as two typical control regions. Phylogenetic analysis based upon 13 protein-coding genes showed clusters based on terrestrial and marine species.

Snake venom dipeptidyl peptidase IV: taxonomic distribution and quantitative variation.

The present study examined the taxonomic distribution of dipeptidyl peptidase IV (DPP IV) activity in venoms of 59 ophidian taxa, representing seven subfamilies of the Families Elapidae and Viperidae. DPP IV activity is extremely variable at all taxonomic levels. It ranged from essentially none in laticaudine, hydrophiine, and some bungarine and elapine venoms, to 10.72 mumol 4-methoxy-beta-naphthylamine liberated per min per 200 mug venom, for Ophiophagus hannah. Intra- and interpopulational variation were examined among eight populations of prairie rattlesnakes (Crotalus viridis viridis), Great Basin rattlesnakes (Crotalus viridis lutosus) and southern Pacific rattlesnakes (Crotalus viridis helleri). Among these populations, the mean weighted range of variation was 4.9-fold, and even among litter mates of C. v. lutosus, DPP IV activity varied as much as 5.6-fold. The two most salient findings, the near ubiquity of DPP IV in snake venoms and its great quantitative variability, even among full siblings, are paradoxical. The widespread distribution of the enzyme suggests an important role in envenomation, while the variable activity levels suggest that DPP IV and by extension, other individual enzymatic constituents, may not be under much individual selective pressure.

Nucleoside composition of Heloderma venoms.

Venoms of Heloderma horridum and Heloderma suspectum were analyzed for the possible presence of purine and pyrimidine nucleosides. Adenosine, cytidine, guanosine, hypoxanthine, inosine, and uridine were found in mug quantities. These amounts are much smaller than those seen in many elapid or viperine venoms, but greater and more varied than those found in crotaline venoms. While their contribution to the hypotension induced by Heloderma venoms may be minor, venom nucleosides nonetheless act in concert with kallikreins/hemorrhagins, alkaline phosphomonoesterase, 5'-nucleotidase, helodermin, helospectins, helothermine, and serotonin. The use of nucleosides as toxins is therefore a generalized squamate strategy, rather than the exclusive province of snakes. Both Heloderma venoms were found to be devoid of NADase and phosphodiesterase activities. Enzymes to release endogenous purines in the prey, are not significant components of Heloderma venoms.

Organic and Peptidyl Constituents of Snake Venoms: The Picture Is Vastly More Complex Than We Imagined.

Small metabolites and peptides in 17 snake venoms (Elapidae, Viperinae, and Crotalinae), were quantified using liquid chromatography-mass spectrometry. Each venom contains >900 metabolites and peptides. Many small organic compounds are present at levels that are probably significant in prey envenomation, given that their known pharmacologies are consistent with snake envenomation strategies. Metabolites included purine nucleosides and their bases, neurotransmitters, neuromodulators, guanidino compounds, carboxylic acids, amines, mono- and disaccharides, and amino acids. Peptides of 2⁻15 amino acids are also present in significant quantities, particularly in crotaline and viperine venoms. Some constituents are specific to individual taxa, while others are broadly distributed. Some of the latter appear to support high anabolic activity in the gland, rather than having toxic functions. Overall, the most abundant organic metabolite was citric acid, owing to its predominance in viperine and crotaline venoms, where it chelates divalent cations to prevent venom degradation by venom metalloproteases and damage to glandular tissue by phospholipases. However, in terms of their concentrations in individual venoms, adenosine, adenine, were most abundant, owing to their high titers in Dendroaspis polylepis venom, although hypoxanthine, guanosine, inosine, and guanine all numbered among the 50 most abundant organic constituents. A purine not previously reported in venoms, ethyl adenosine carboxylate, was discovered in D. polylepis venom, where it probably contributes to the profound hypotension caused by this venom. Acetylcholine was present in significant quantities only in this highly excitotoxic venom, while 4-guanidinobutyric acid and 5-guanidino-2-oxopentanoic acid were present in all venoms.

Elution of tightly bound solutes from concanavalin A Sepharose. Factors affecting the desorption of cottonmouth venom glycoproteins.

Some glycoproteins bind so tightly to concanavalin A Sepharose that common desorption techniques are ineffective, so a systematic exploration of factors affecting desorption of cottonmouth venom glycoproteins was undertaken. Glycoprotein desorption is greatly improved by introducing up to four pauses of 5-10 min duration into the elution step. Eluent concentrations above 250 mM methylglucoside or 500 mM methyl-mannoside reduced glycoprotein desorption. Eluent NaCl diminished glycoprotein desorption. Most venom glycoproteins desorb more readily as pH diminishes from 6.0 to 4.0, but phosphodiesterase shows the opposite pattern. Eluents recommended by the supplier for desorbing solutes or for column cleaning were ineffectual.

Population Genomic Analysis of a Pitviper Reveals Microevolutionary Forces Underlying Venom Chemistry.

Venoms are among the most biologically active secretions known, and are commonly believed to evolve under extreme positive selection. Many venom gene families, however, have undergone duplication, and are often deployed in doses vastly exceeding the LD50 for most prey species, which should reduce the strength of positive selection. Here, we contrast these selective regimes using snake venoms, which consist of rapidly evolving protein formulations. Though decades of extensive studies have found that snake venom proteins are subject to strong positive selection, the greater action of drift has been hypothesized, but never tested. Using a combination of de novo genome sequencing, population genomics, transcriptomics, and proteomics, we compare the two modes of evolution in the pitviper, Protobothrops mucrosquamatus. By partitioning selective constraints and adaptive evolution in a McDonald-Kreitman-type framework, we find support for both hypotheses: venom proteins indeed experience both stronger positive selection, and lower selective constraint than other genes in the genome. Furthermore, the strength of selection may be modulated by expression level, with more abundant proteins experiencing weaker selective constraint, leading to the accumulation of more deleterious mutations. These findings show that snake venoms evolve by a combination of adaptive and neutral mechanisms, both of which explain their extraordinarily high rates of molecular evolution. In addition to positive selection, which optimizes efficacy of the venom in the short term, relaxed selective constraints for deleterious mutations can lead to more rapid turnover of individual proteins, and potentially to exploration of a larger venom phenotypic space.

Coralsnake Venomics: Analyses of Venom Gland Transcriptomes and Proteomes of Six Brazilian Taxa.

Venom gland transcriptomes and proteomes of six Micrurus taxa (M. corallinus, M. lemniscatus carvalhoi, M. lemniscatus lemniscatus, M. paraensis, M. spixii spixii, and M. surinamensis) were investigated, providing the most comprehensive, quantitative data on Micrurus venom composition to date, and more than tripling the number of Micrurus venom protein sequences previously available. The six venomes differ dramatically. All are dominated by 2-6 toxin classes that account for 91-99% of the toxin transcripts. The M. s. spixii venome is compositionally the simplest. In it, three-finger toxins (3FTxs) and phospholipases A2 (PLA2s) comprise >99% of the toxin transcripts, which include only four additional toxin families at levels ≥0.1%. Micrurus l. lemniscatus venom is the most complex, with at least 17 toxin families. However, in each venome, multiple structural subclasses of 3FTXs and PLA2s are present. These almost certainly differ in pharmacology as well. All venoms also contain phospholipase B and vascular endothelial growth factors. Minor components (0.1-2.0%) are found in all venoms except that of M. s. spixii. Other toxin families are present in all six venoms at trace levels (<0.005%). Minor and trace venom components differ in each venom. Numerous novel toxin chemistries include 3FTxs with previously unknown 8- and 10-cysteine arrangements, resulting in new 3D structures and target specificities. 9-cysteine toxins raise the possibility of covalent, homodimeric 3FTxs or heterodimeric toxins with unknown pharmacologies. Probable muscarinic sequences may be reptile-specific homologs that promote hypotension via vascular mAChRs. The first complete sequences are presented for 3FTxs putatively responsible for liberating glutamate from rat brain synaptosomes. Micrurus C-type lectin-like proteins may have 6-9 cysteine residues and may be monomers, or homo- or heterodimers of unknown pharmacology. Novel KSPIs, 3× longer than any seen previously, appear to have arisen in three species by gene duplication and fusion. Four species have transcripts homologous to the nociceptive toxin, (MitTx) α-subunit, but all six species had homologs to the ß-subunit. The first non-neurotoxic, non-catalytic elapid phospholipase A2s are reported. All are probably myonecrotic. Phylogenetic analysis indicates that the six taxa diverged 15-35 million years ago and that they split from their last common ancestor with Old World elapines nearly 55 million years ago. Given their early diversification, many cryptic micrurine taxa are anticipated.

Polyamines as Snake Toxins and Their Probable Pharmacological Functions in Envenomation.

While decades of research have focused on snake venom proteins, far less attention has been paid to small organic venom constituents. Using mostly pooled samples, we surveyed 31 venoms (six elapid, six viperid, and 19 crotalid) for spermine, spermidine, putrescine, and cadaverine. Most venoms contained all four polyamines, although some in essentially trace quantities. Spermine is a potentially significant component of many viperid and crotalid venoms (≤0.16% by mass, or 7.9 µmol/g); however, it is almost completely absent from elapid venoms assayed. All elapid venoms contained larger molar quantities of putrescine and cadaverine than spermine, but still at levels that are likely to be biologically insignificant. As with venom purines, polyamines impact numerous physiological targets in ways that are consistent with the objectives of prey envenomation, prey immobilization via hypotension and paralysis. Most venoms probably do not contain sufficient quantities of polyamines to induce systemic effects in prey; however, local effects seem probable. A review of the pharmacological literature suggests that spermine could contribute to prey hypotension and paralysis by interacting with N-methyl-d-aspartate (NMDA) and α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors, nicotinic and muscarinic acetylcholine receptors, γ-Aminobutyric acid (GABA) receptors, blood platelets, ryanodine receptors, and Ca2+-ATPase. It also blocks many types of cation-permeable channels by interacting with negatively charged amino acid residues in the channel mouths. The site of envenomation probably determines which physiological targets assume the greatest importance; however, venom-induced liberation of endogenous, intracellular stores of polyamines could potentially have systemic implications and may contribute significantly to envenomation sequelae.

Taxonomic distribution and quantitative analysis of free purine and pyrimidine nucleosides in snake venoms.

The nucleoside content of 32 elapid and viperid venoms was examined. Free purines, principally adenosine (ADO), inosine (INO), and guanosine (GUA), comprised as much as 8.7% of the solid components of some venoms. Thus, purines are far more abundant in some venoms than many proteinaceous toxins. Hypoxanthine (HYP) was found in about half of elapid and viperine venoms, in which it is a relatively minor constituent (<60 microg/g). Adenosine monophosphate (AMP) was tentatively identified in only three elapid and two viperid venoms. The pyrimidines, uridine (URI) and cytidine (CYT), were also found in most elapid and viperine venoms. In most of these, the amount of uridine was substantially greater than that of cytidine. Thymidine (THY) was not found in any venom, indicating that DNA from disintegration of glandular cells is not the source of venom nucleosides. In contrast to elapid and viperine venoms, most crotaline venoms are devoid of free nucleosides. Elapid and viperine venoms also contained other minor, low molecular weight constituents that could not be positively identified. Some had spectra identical to those of adenosine, nicotinamide adenine dinucleotide (NAD), inosine, xanthosine (XAN), and guanosine, while others had unique spectra. There is no apparent correlation between quantities of venom nucleosides and literature values for the three dominant venom enzymes that release endogenous nucleosides, 5'-nucleotidase (5NUC), phosphodiesterase (PDE), and alkaline phosphomonoesterase (PME).

Ophidian envenomation strategies and the role of purines.

Snake envenomation employs three well integrated strategies: prey immobilization via hypotension, prey immobilization via paralysis, and prey digestion. Purines (adenosine, guanosine and inosine) evidently play a central role in the envenomation strategies of most advanced snakes. Purines constitute the perfect multifunctional toxins, participating simultaneously in all three envenomation strategies. Because they are endogenous regulatory compounds in all vertebrates, it is impossible for any prey organism to develop resistance to them. Purine generation from endogenous precursors in the prey explains the presence of many hitherto unexplained enzyme activities in snake venoms: 5'-nucleotidase, endonucleases (including ribonuclease), phosphodiesterase, ATPase, ADPase, phosphomonoesterase, and NADase. Phospholipases A(2), cytotoxins, myotoxins, and heparinase also participate in purine liberation, in addition to their better known functions. Adenosine contributes to prey immobilization by activation of neuronal adenosine A(1) receptors, suppressing acetylcholine release from motor neurons and excitatory neurotransmitters from central sites. It also exacerbates venom-induced hypotension by activating A(2) receptors in the vasculature. Adenosine and inosine both activate mast cell A(3) receptors, liberating vasoactive substances and increasing vascular permeability. Guanosine probably contributes to hypotension, by augmenting vascular endothelial cGMP levels via an unknown mechanism. Novel functions are suggested for toxins that act upon blood coagulation factors, including nitric oxide production, using the prey's carboxypeptidases. Leucine aminopeptidase may link venom hemorrhagic metalloproteases and endogenous chymotrypsin-like proteases with venom L-amino acid oxidase (LAO), accelerating the latter. The primary function of LAO is probably to promote prey hypotension by activating soluble guanylate cyclase in the presence of superoxide dismutase. LAO's apoptotic activity, too slow to be relevant to prey capture, is undoubtedly secondary and probably serves principally a digestive function. It is concluded that the principal function of L-type Ca(2+) channel antagonists and muscarinic toxins, in Dendroaspis venoms, and acetylcholinesterase in other elapid venoms, is to promote hypotension. Venom dipeptidyl peptidase IV-like enzymes probably also contribute to hypotension by destroying vasoconstrictive peptides such as Peptide YY, neuropeptide Y and substance P. Purines apparently bind to other toxins which then serve as molecular chaperones to deposit the bound purines at specific subsets of purine receptors. The assignment of pharmacological activities such as transient neurotransmitter suppression, histamine release and antinociception, to a variety of proteinaceous toxins, is probably erroneous. Such effects are probably due instead to purines bound to these toxins, and/or to free venom purines.

Primary structures of four trypsin inhibitor E homologs from venom of Dendroaspis angusticeps: structure-function comparisons with other dendrotoxin homologs.

Four trypsin inhibitor homologs, the first known from Dendroaspis angusticeps venom, were characterized using a combination of gel filtration, cation exchange, reverse-phase liquid chromatography, Edman degradation and mass spectrometry. The four toxins comprise two 57 residue and two 59 residue isoforms. The long toxins possess a Lys-Gln N-terminal extension lacked by the short toxins. The only other structural difference is an Arg/His replacement at position 55. The long Arg55 variant is identical to trypsin inhibitor E from the venom of Dendroaspis polylepis. The name epsilon-dendrotoxin is suggested so as to follow the nomenclature of Benishin, C.G., Sorensen, R.G., Brown, W.E., Krueger, B.K., Blaustein, M.P., 1988. Four polypeptide components of green mamba venom selectively block certain potassium channels in rat brain synaptosomes. Mol. Pharmacol. 34, 152-159. Among snake venom protease inhibitors, the epsilon-dendrotoxins are structurally most like the delta-dendrotoxins, with which they share only 64% of their residues. In addition, the epsilon-dendrotoxins display hydropathy profiles more like those of the alpha- and delta-dendrotoxins, than those of the trypsin inhibitors from snake venoms. Given the strong protease inhibitory activity of trypsin inhibitor E and the recently demonstrated weak K(+) channel inhibitory activity of two of these variants (Tytgat, J., Vandenberghe, I., Ulens, C., Van Beeumen, J., 2001. New polypeptide components purified from mamba venom. FEBS Lett. 491, 217-221), the epsilon-dendrotoxins represent structural and functional intermediates between the facilitatory toxins and the protease inhibitors.

A novel peptide from the ACEI/BPP-CNP precursor in the venom of Crotalus durissus collilineatus.

In crotaline venoms, angiotensin-converting enzyme inhibitors [ACEIs, also known as bradykinin potentiating peptides (BPPs)], are products of a gene coding for an ACEI/BPP-C-type natriuretic peptide (CNP) precursor. In the genes from Bothrops jararaca and Gloydius blomhoffii, ACEI/BPP sequences are repeated. Sequencing of a cDNA clone from venom glands of Crotalus durissus collilineatus showed that two ACEIs/BPPs are located together at the N-terminus, but without repeats. An additional sequence for CNP was unexpectedly found at the C-terminus. Homologous genes for the ACEI/BPP-CNP precursor suggest that most crotaline venoms contain both ACEIs/BPPs and CNP. The sequence of ACEIs/BPPs is separated from the CNP sequence by a long spacer sequence. Previously, there was no evidence that this spacer actually coded any expressed peptides. Aird and Kaiser (1986, unpublished) previously isolated and sequenced a peptide of 11 residues (TPPAGPDVGPR) from Crotalus viridis viridis venom. In the present study, analysis of the cDNA clone from C. d. collilineatus revealed a nearly identical sequence in the ACEI/BPP-CNP spacer. Fractionation of the crude venom by reverse phase HPLC (C(18)), and analysis of the fractions by mass spectrometry (MS) indicated a component of 1020.5 Da. Amino acid sequencing by MS/MS confirmed that C. d. collilineatus venom contains the peptide TPPAGPDGGPR. Its high proline content and paired proline residues are typical of venom hypotensive peptides, although it lacks the usual N-terminal pyroglutamate. It has no demonstrable hypotensive activity when injected intravenously in rats; however, its occurrence in the venoms of dissimilar species suggests that its presence is not accidental. Evidence suggests that these novel toxins probably activate anaphylatoxin C3a receptors.

Chromatographic behavior of Bothrops erythromelas phospholipase and other venom constituents on Superdex 75.

In a chromatographic method modification intended to preserve protease activity in Bothrops erythromelas venom, 2 mM CaCl2 was added to the gel filtration buffer [50mM Tris/HCl/150mM NaCl (pH 8.0)], in lieu of an equimolar portion of NaCl. This minor compositional change induced significant differences in the venom elution profile on Superdex 200. For this reason, the influence of buffer composition on chromatographic behavior was investigated using an analytical Superdex 75 HR 10/30 column. Phospholipase (PLA) was used as a marker because Naja atra PLA had previously been observed to interact hydrophobically with this resin. PLA elution volumes generally increased as buffer pH decreased. Addition of 20% acetonitrile to the Tris buffer with CaCl2, reduced hydrophobic interaction of the PLA so significantly that its elution was non-overlapping in the two buffers. Other venom constituents, including bradykinin-potentiating peptides and probable hemorrhagic metalloproteases, were similarly affected. Buffer calcium, bound by vicinal dextran hydroxyl groups, appears to retard elution of this acidic PLA.