Abstract:

Spider venoms are a rich source of highly specific peptide ligands that bind to voltage-gated calcium, potassium and sodium channels. These toxins have been used to help elucidate the structure and physiological roles of these channels and can potentially be useful models for drug design. ProTx-II, a peptide toxin originally purified from the venom of the tarantula Thrixopelma pruriens, reversibly inhibits Nav channel isoforms (Middleton et al., Biochemistry, 2002 ). Unlike other toxins that inhibit Nav channel function by occluding the pore, ProTx-II is thought to interact with the voltage sensor region of the channel to modify activation gating. Upon toxin binding, the voltage-dependence of activation is shifted toward more depolarized potentials and current is reduced. ProTx-II is also unique in that it demonstrates promiscuity across voltage-gated ion channels by inhibiting T-type Cav channels, suggesting a conserved structural motif among Nav and Cav channels. ProTx-II conforms to the inhibitory cystine knot (ICK) motif described previously for some toxins interacting with voltage-gated ion channels. This structural motif consists of a ring formed by two disulfide bonds that is threaded by a third disulfide bond. We have cloned a synthetic ProTx-II gene and developed a recombinant expression system that allows us to produce properly folded peptide toxin. Characterization by MALDI-TOF and circular dichroism indicates that the toxin has the correct disulfide pairings. We have confirmed the activity of recombinant ProTx-II by testing its function on HEK cells stably transfected with Na v 1.5 channels via a voltage-clamp assay.

Based upon studies using gating modifier toxins and artificial phospholipid membranes, it has been proposed that toxins exhibiting high affinity inhibition for voltage-gated ion channels bind their receptor sites by partitioning into the lipid membrane (Lee, SY & MacKinnon, R, Nature, 2004; Wang et al., JGP, 2004). This theory has important implications for the structure of the voltage sensors and the mechanism of channel activation and while it has been proposed for Kv channels and their ligands, the model should be consistent throughout the superfamily of ion channels. Therefore, to test this hypothesis for a Nav channel ligand, I have analyzed the phospholipid binding capabilities of ProTx-II and compared these results to the phospholipid binding properties of other functionally distinct gating modifier toxins. In addition, I have analyzed the interaction between Nav 1.5 and single alanine mutants created at every non-cysteine position of ProTx-II to identify residues important in binding. Furthermore, in order to determine the channel binding site I created 72 Nav 1.5 mutants and analyzed the interactions between channel mutants and wild-type ProTx-II.

Toxin-lipid interaction studies identified ProTx-II as a gating modifier toxin with an affinity for phospholipids and the alanine scan of ProTx-II identified several hydrophobic and cationic residues important in modification of Nav 1.5. These findings are consistent with a membrane-access mechanism of inhibition but also implicate anionic channel residues in formation of the toxin-channel complex. Our channel mutagenesis established ProTx-II as a novel neurotoxin acting at an unknown receptor site but additional studies are necessary in order to fully define this site.