Structural studies of novel potassium(+)-channel toxins derived from Pandinus imperator

Abstract

The Charybdotoxin, or alpha-K, family of scorpion toxins is a group of small (approx. 4 kD) peptides that have been successfully employed as both selective blockers and structural probes of various types of voltage gated K+ channels. This thesis describes two new members of this family, PiTX-Kalpha and PiTX-Kbeta, which are derived from Pandinus imperator. Structural analysis of these toxins by homonuclear NMR spectroscopy reveals that they have the same secondary structure and global fold as Charybdotoxin, but they also have unique characteristics of sequence, structure, and activity which place them in a separate subclass (alpha-K5.x) of the alpha-K family. Compared to other alpha-K toxins, both pandinotoxins contain a novel 3 residue deletion at the N-terminus, and a glycine, rather than an alanine residue at position 26, despite the previously held theory that any residue larger than alanine in this position would create a structural disruption. While both pandinotoxins selectively block rapidly inactivating Ca2+-dependant (A-type) K+ channels, the single mutation which distinguishes PiTX-Kalpha from PiTX-Kbeta (P10E), dramatically changes the toxin affinity. PiTX-Kalpha displays a high affinity for A-type channels in both rat brain synaptosomes (IC50=6 nM) and dorsal root ganglia (IC50=7.8 nM), but PiTX-Kbeta shows a much larger difference in these preparations (IC50= 42 nM in synaptosomes, (IC50=3600 nM in ganglia). A detailed analysis of sidechain positioning in these toxins has revealed that a group of residues at the more variable end regions of the ellipsoidal toxin molecule are probably responsible for the selective affinity that the pandinotoxins show for A-type channels, while a distinct group of residues which surround the critical lysine 27 residue are responsible for both the affinity of other alpha-K toxins towards Ca2+-dependant (Maxi-K) K+ channels and the affinity differences observed between PiTX-alpha and PiTX-Kbeta. The implications of these results on current structural models of the K+ channel vestibule, and their application towards studying native K+ channels are also explored. In addition to these results, the use of molecular modeling techniques in simulating toxin structures is addressed, and preliminary structural studies of TsTX-Kalpha, a selective blocker of delayed rectifier K+ channels in rat brain synaptosomes are presented.