Characterization of selectivity and pharmacophores of type 1 sea anemone toxins by screening seven Na(v) sodium channel isoforms.

During their evolution, animals have developed a set of cysteine-rich peptides capable of binding various extracellular sites of voltage-gated sodium channels (VGSC). Sea anemone toxins that target VGSCs delay their inactivation process, but little is known about their selectivities. Here we report the investigation of three native type 1 toxins (CGTX-II, δ-AITX-Bcg1a and δ-AITX-Bcg1b) purified from the venom of Bunodosoma cangicum. Both δ-AITX-Bcg1a and δ-AITX-Bcg1b toxins were fully sequenced. The three peptides were evaluated by patch-clamp technique among Nav1.1-1.7 isoforms expressed in mammalian cell lines, and their preferential targets are Na(v)1.5>1.6>1.1. We also evaluated the role of some supposedly critical residues in the toxins which would interact with the channels, and observed that some substitutions are not critical as expected. In addition, CGTX-II and δ-AITX-Bcg1a evoke different shifts in activation/inactivation Boltzmann curves in Nav1.1 and 1.6. Moreover, our results suggest that the interaction region between toxins and VGSCs is not restricted to the supposed site 3 (S3-S4 linker of domain IV), and this may be a consequence of distinct surface of contact of each peptide vs. targeted channel. Our data suggest that the contact surfaces of each peptide may be related to their surface charges, as CGTX-II is more positive than δ-AITX-Bcg1a and δ-AITX-Bcg1b.

Revisiting cangitoxin, a sea anemone peptide: purification and characterization of cangitoxins II and III from the venom of Bunodosoma cangicum.

Sodium channel toxins from sea anemones are employed as tools for dissecting the biophysical properties of inactivation in voltage-gated sodium channels. Cangitoxin (CGTX) is a peptide containing 48 amino acid residues and was formerly purified from Bunodosoma cangicum. Nevertheless, previous works reporting the isolation procedures for such peptide from B. cangicum secretions are controversial and may lead to incorrect information. In this paper, we report a simple and rapid procedure, consisting of two chromatographic steps, in order to obtain a CGTX analog directly from sea anemone venom. We also report a substitution of N16D in this peptide sample and the co-elution of an inseparable minor isoform presenting the R14H substitution. Peptides are named as CGTX-II and CGTX-III, and their effects over Nav1.1 channels in patch clamp experiments are demonstrated.

Crotamine inhibits preferentially fast-twitching muscles but is inactive on sodium channels.

Crotamine is a peptide toxin from the venom of the rattlesnake Crotalus durissus terrificus that induces a typical hind-limb paralysis of unknown nature. Hind limbs have a predominance of fast-twitching muscles that bear a higher density of sodium channels believed until now to be the primary target of crotamine. Hypothetically, this makes these muscles more sensitive to crotamine and would explain such hind-limb paralysis. To challenge this hypothesis, we performed concentration vs. response curves on fast (extensor digitorum longus (EDL)) and slow (soleus) muscles of adult male rats. Crotamine was tested on various human Na+ channel isoforms (Na(v)1.1-Na(v)1.6 alpha-subunits) expressed in HEK293 cells in patch-clamp experiments, as well as in acutely dissociated dorsal root ganglion (DRG) neurons. Also, the behavioral effects of crotamine intoxication were compared with those of a muscle-selective sodium channel antagonist mu-CgTx-GIIIA, and other sodium-acting toxins such as tetrodotoxin alpha- and beta-pompilidotoxins, sea anemone toxin BcIII, spider toxin Tx2-6. Results pointed out that EDL was more susceptible to crotamine than soleus under direct electrical stimulation. Surprisingly, electrophysiological experiments in human Na(v)1.1 to Na(v)1.6 Na+ channels failed to show any significant change in channel characteristics, in a clear contrast with former studies. DRG neurons did not respond to crotamine. The behavioral effects of the toxins were described in detail and showed remarkable differences. We conclude that, although differences in the physiology of fast and slow muscles may cause the typical crotamine syndrome, sodium channels are not the primary target of crotamine and therefore, the real mechanism of action of this toxin is still unknown.

Four disulfide-bridged scorpion beta neurotoxin CssII: heterologous expression and proper folding in vitro.

The gene of the four disulfide-bridged Centruroides suffusus suffusus toxin II was cloned into the expression vector pQE30 containing a 6His-tag and a FXa proteolytic cleavage region. This recombinant vector was transfected into Escherichia coli BL21 cells and expressed under induction with isopropyl thiogalactoside (IPTG). The level of expression was 24.6 mg/l of culture medium, and the His tagged recombinant toxin (HisrCssII) was found exclusively in inclusion bodies. After solubilization the HisrCssII peptide was purified by affinity and hydrophobic interaction chromatography. The reverse-phase HPLC profile of the HisrCssII product obtained from the affinity chromatography step showed several peptide fractions having the same molecular mass of 9392.6 Da, indicating that HisrCssII was oxidized forming several distinct disulfide bridge arrangements. The multiple forms of HisrCssII after reduction eluted from the column as a single protein component of 9400.6 Da. Similarly, an in vitro folding of the reduced HisrCssII generated a single oxidized component of HisrCssII, which was cleaved by the proteolytic enzyme FXa to the recombinant CssII (rCssII). The molecular mass of rCssII was 7538.6 Da as expected. Since native CssII (nCssII) is amidated at the C-terminal residue whereas the rCssII is heterologously expressed in the format of free carboxyl end, there is a difference of 1 Da, when comparing both peptides (native versus heterologously expressed). Nevertheless, they show similar toxicity when injected intracranially into mice, and both nCssII and rCssII show the typical electrophysiological properties of beta-toxins in Na(v)1.6 channels, which is for the first time demonstrated here. Binding and displacement experiments conducted with radiolabelled CssII confirms the electrophysiological results. Several problems associated with the heterologously expressed toxins containing four disulfide bridges are discussed.