Structural features important for the biological activity of the potassium channel blocking dendrotoxins

Conkunitzin-S1 is the first member of a new Kunitz-type neurotoxin family. Structural and functional characterization

Conkunitzin-S1 (Conk-S1) is a 60-residue neurotoxin from the venom of the cone snail Conus striatus that interacts with voltage-gated potassium channels. Conk-S1 shares sequence homology with Kunitz-type proteins but contains only two out of the three highly conserved cysteine bridges, which are typically found in these small, basic protein modules. In this study the three-dimensional structure of Conk-S1 has been solved by multidimensional NMR spectroscopy. The solution structure of recombinant Conk-S1 shows that a Kunitz fold is present, even though one of the highly conserved disulfide cross-links is missing. Introduction of a third, homologous disulfide bond into Conk-S1 results in a functional toxin with similar affinity for Shaker potassium channels. The affinity of Conk-S1 can be enhanced by a pore mutation within the Shaker channel pore indicating an interaction of Conk-S1 with the vestibule of potassium channels.

Atypical Kunitz-type serine proteinase inhibitors produced by the ruminant placenta

Recently, an unusual family of genes was identified with expression confined to the trophoblast of ruminant ungulate species. The members of this family (the trophoblast Kunitz domain proteins, or TKDPs) are characterized by the presence of one or more similar, approximately 80-residue repeat sequences placed ahead of a Kunitz serine proteinase-inhibitor domain. To examine the specificity of the Kunitz moiety, the Kunitz domains of selected TKDPs and a control Kunitz protein, bovine pancreatic trypsin inhibitor (BPTI), were produced as glutathione S-transferase fusions, and their abilities to inhibit six serine proteinases were examined. Circular dichroism spectroscopy confirmed that the Kunitz fold was intact. Three of the TKDPs had unusual residues at their P1 "warhead" (ovine TKDP-1, Asn; bovine TKDP-3, Thr; and bovine TKDP-5, Ile) and exhibited no measurable inhibitory activity toward any of the proteinases. Three (ovine TKDP-3, bovine TKDP-3, and bovine TKDP-4) lacked the conserved cysteines at residues 14 and 38 that form one of the highly conserved disulfide bonds that are structurally important in all known mammalian Kunitz proteins. Ovine TKDP-3 and bovine TKDP-4 had P1 lysines and inhibited trypsin and plasmin with K(i) values only approximately 10-fold higher than that of BPTI. Bovine TKDP-2 had a P1 lysine and the three conserved disulfides, but it possessed an unusual residue (Asp) at P2. It exhibited no inhibitory activity. These data suggest that the function of the TKDP, like certain Kunitz proteins found in snake venoms, may not be in proteinase inhibition.

Insight into the factors influencing the backbone dynamics of three homologous proteins, dendrotoxins I and K, and BPTI: FTIR and time-resolved fluorescence investigations

Attenuated total reflection Fourier transform infrared (ATR-FTIR) spectroscopy, combined with hydrogen/deuterium exchange technique and time-resolved fluorescence spectroscopy, has been used to investigate the changes in structure and dynamics that underlie the thermodynamic stability differences observed for three closely homologous proteins: dendrotoxins I and K, and bovine pancreatic trypsin inhibitor (BPTI). The experiments were performed on proteins under their native state and a modified form, obtained by selective reduction of a disulfide bond at the surface of the molecule, increasing slightly the backbone flexibility without changing the average structure. The data confirmed the high local as well as global rigidity of BPTI. In protein K, the exchange process was slow during the first 2 h of exchange, presumably reflecting a compact three-dimensional conformation, and then increased rapidly, the internal amide protons of the beta-strands exchanging 10-fold faster than in BPTI or protein I. The most probable destabilizing element was identified as Pro32, in the core of the beta-sheet. Protein I was found to present a 10% more expanded volume than protein K or BPTI, and there is a possible correlation between the resulting increased flexibility of the molecule and the lower thermodynamic stability observed for this protein. Interestingly, the interior amide protons of the beta-sheet structure were found to be as protected against exchange in protein I as in BPTI, suggesting that, although globally more flexible than that of Toxin K or BPTI, the structure of Toxin I could be locally quite rigid. The structural factors suspected to be responsible for the differences in internal flexibility of the two toxins could play a significant role in determining their functional properties.

Simultaneous binding of basic peptides at intracellular sites on a large conductance Ca2+-activated K+ channel. Equilibrium and kinetic basis of negatively coupled ligand interactions

The homologous Kunitz inhibitor proteins, bovine pancreatic trypsin inhibitor (BPTI) and dendrotoxin I (DTX-I), interact with large conductance Ca2+-activated K+ channels (maxi-KCa) by binding to an intracellular site outside of the pore to produce discrete substate events. In contrast, certain homologues of the Shaker ball peptide produce discrete blocking events by binding within the ion conduction pathway. In this study, we investigated ligand interactions of these positively charged peptide molecules by analysis of single maxi-KCa channels in planar bilayers recorded in the presence of DTX-I and BPTI, or DTX-I and a high-affinity homologue of ball peptide. Both DTX-I (Kd, 16.5 nM) and BPTI (Kd, 1, 490 nM) exhibit one-site binding kinetics when studied alone; however, records in the presence of DTX-I plus BPTI demonstrate simultaneous binding of these two molecules. The affinity of BPTI (net charge, +6) decreases by 11.7-fold (Kd, 17,500 nM) when DTX-I (net charge, +10) is bound and, conversely, the affinity of DTX-I decreases by 10.8-fold (Kd, 178 nM) when BPTI is bound. The ball peptide homologue (BP; net charge, +6) exhibits high blocking affinity (Kd, 7.2 nM) at a single site when studied alone, but has 8.0-fold lower affinity (Kd, 57 nM) for blocking the DTX-occupied channel. The affinity of DTX-I likewise decreases by 8.4-fold (Kd, 139 nM) when BP is bound. These results identify two types of negatively coupled ligand-ligand interactions at distinct sites on the intracellular surface of maxi-KCa channels. Such antagonistic ligand interactions explain how the binding of BPTI or DTX-I to four potentially available sites on a tetrameric channel protein can exhibit apparent one-site kinetics. We hypothesize that negatively coupled binding equilibria and asymmetric changes in transition state energies for the interaction between DTX-I and BP originate from repulsive electrostatic interactions between positively charged peptide ligands on the channel surface. In contrast, there is no detectable binding interaction between DTX-I on the inside and tetraethylammonium or charybdotoxin on the outside of the maxi-KCa channel.

Cloning and functional expression of B chains of beta-bungarotoxins from Bungarus multicinctus (Taiwan banded krait)

The cDNA species encoding the B chains (B1 and B2) of beta-bungarotoxins (beta-Bgt) were constructed from the cellular RNA isolated from the venom glands of Bungarus multicinctus (Taiwan banded krait). The deduced amino acid sequences of the B chains were different from those determined previously by a protein sequencing technique. One additional Arg residue is inserted between Val-19 and Arg-20 of the B1 chain. Similarly the insertion of one additional Val residue between Val-19 and Arg-20 of the B2 chain is noted. Thus the B chains should comprise 61 amino acid residues. Moreover, the residues at positions 44-46 are Gly-Asn-His, in contrast with a previous result showing the sequence His-Gly-Asn. Instead of Asp, the residues at positions 41 and 43 are Asn. The B chain was subcloned into the expression vector pET-32a(+) and transformed into Escherichia coli strain BL21(DE3). The recombinant B chain was expressed as a fusion protein and purified on a His-Bind resin column. The yield of affinity-purified fusion protein was increased markedly by replacing Cys-55 of the B chain with Ser. However, the isolated B(C55S) chain became insoluble in aqueous solution after removal of the fused protein from the affinity-purified product, suggesting that protein-protein interactions might be crucial for stabilizing the structure of the B chain. The B(C55S) chain fusion protein showed activity in blocking the voltage-dependent K+ channel, but did not inhibit the binding of beta-Bgt to synaptosomal membranes. These results, together with the finding that modification of His-48 of the A chain of beta-Bgt caused a marked decrease in the ability to bind toxin to its acceptor proteins, suggest that the B chain is involved in the K+ channel blocking action observed with beta-Bgt, and that the binding of beta-Bgt to neuronal receptors is not heavily dependent on the B chain.

Recent studies on dendrotoxins and potassium ion channels

1. Dendrotoxins are small proteins isolated from mamba (Dendroaspis) snake venoms. They block some subtypes of voltage-dependent potassium channels in neurons. 2. Dendrotoxins contain 57-60 amino acid residues crosslinked by three disulfide bridges. They are homologous to Kunitz-type serine protease inhibitors, such as aprotinin, although they have little or no antiprotease activity. 3. Dendrotoxins act mainly on neuronal K+ channels. Studies with cloned K+ channels indicate that alpha-dendrotoxin from green mamba Dendroaspis angusticeps blocks Kv1.1 and Kv1.2 channels in the nanomolar range. In native cells, dendrotoxin appears preferentially to block inactivating forms of K+ current. 4. Dendrotoxins can induce repetitive firing in neurons and facilitate transmitter release. On direct injection to the CNS, dendrotoxins can induce epileptiform activity. 5. Radiolabeled dendrotoxins are useful markers of subtypes of K+ channels in vivo, and structural analogs help to define the molecular recognition properties of different types of K+ channels.

Molecular structure, conformational analysis, and structure-activity studies of Dendrotoxin and its homologues using molecular mechanics and molecular dynamics techniques

Three-dimensional structures of Dendrotoxin (DtX), Toxin-I (DpI), and Toxin-K (DpK) were determined using molecular mechanics and molecular dynamics techniques. The overall molecular conformation and protein folding of the three dendrotoxins are very similar to the published crystal structures of bovine pancreatic trypsin inhibitor (BPTI) and alpha-DtX. Major secondary structural regions of the dendrotoxins are stable without much fluctuation during the dynamics simulation; the regions corresponding to the turns and bends (rich in lysines and arginines) exhibit more fluctuations. The conformational angles and the C alpha...C alpha' distances of the three disulfides (in each of the dendrotoxins) are different from each other. Comparative model building studies, involving the dendrotoxins and the proteinases, reveal that the key interactions (observed in BPTI-trypsin complex) needed for anti-protease activity are absent due to structural differences between the dendrotoxins and BPTI at the anti-protease loop; this explains the inability of the dendrotoxins to inhibit proteinases. The model also suggests that the solvent-exposed beta-turn region, rich in lysines (residues 26-28), might bind directly to the extracellular anionic sites of the receptors (K+ channels) by ionic interactions. The strikingly homologous cysteine distribution (Cys-x-x-x-Cys) in DtX, DpI, and DpK, at the C-terminus, induces the occurrence of a characteristic conformational motif, consisting of an alpha-helix (in an amphiphilic environment) stabilized by two disulfides, one involving a cysteine at the beta-strand, and the other at the N-terminus. This amphiphilic secondary structural element seems to provide the rigid frame work needed for exposing the proposed active site region of the dendrotoxins to the anionic sites of the K+ channel receptors.

Cloning and expression of mamba toxins

Mamba venoms contain pharmacologically active proteins that interfere with neuromuscular transmission by binding to and altering the normal functioning of neuronal proteins involved, directly or indirectly, with regulating nerve transmission. Of the mamba toxins studied to date, many act on voltage-sensitive K+ channels, nicotinic or muscarinic acetylcholine receptors, or acetylcholinesterase. In an attempt to clone, characterize, and express the genes encoding these toxins, as well as other genes specifying activities not completely elucidated as yet, a cDNA library was constructed from mRNA isolated from the glands of the black mamba. Clones from the library harboring sequences encoding 14 different mamba toxins were isolated and characterized by nucleotide sequence analysis. Genes coding for three proteins, dendrotoxins (DTX) K, I, and E, were expressed as maltose-binding (MBP) fusion proteins in the periplasmic space of Escherichia coli. The DTXK-MBP fusion protein was affinity purified, cleaved from its chaperon, and the recombinant DTXK purified from MBP. Recombinant DTXK was shown to be identical to native DTXK in its N-terminal sequence, chromatographic behavior, convulsion-inducing activity, and binding to voltage-activated K+ channels in bovine synaptic membranes. Computer modeling was employed to create three-dimensional structures of DTXK and DTX1 from the X-ray crystal structure of alpha-DTX utilizing both structural and sequence homologies. Comparisons were made between the three toxins, providing a framework for site-directed mutagenesis.

Blockade by dendrotoxin homologues of voltage-dependent K+ currents in cultured sensory neurones from neonatal rats

1. Homologues of dendrotoxin (Dtx) were isolated from the crude venom of Green and Black Mamba snakes and examined for K+ channel blocking activity in neonatal rat dorsal root ganglion cells (DRGs) by whole-cell patch clamp recording. 2. Outward potassium current activated by depolarization was composed of two major components: a slowly inactivating current (SIC, tau decay approximately 50 ms, 200 ms and 2s), and a non-inactivating current (NIC, tau decay > 2 min). Tail current analysis revealed two time constants of deactivation of total outward current, 3-12 ms and 50-150 ms (at -80 mV) which corresponded to SIC and NIC, respectively. 3. All the homologues (alpha-, beta-, gamma- and delta-Dtx and toxins I and K) blocked outward current activated by depolarization in a dose-dependent manner. The most potent in blocking total outward current was delta-Dtx (EC50 of 0.5 +/- 0.2 nM), although there were no statistically significant differences in potency between any of the homologues. 4. Qualitative differences in the nature of the block were noted between homologues. In particular, the block by delta-Dtx was time-dependent, whereas that by alpha-Dtx was not. 5. alpha-Dtx was a much better blocker of SIC (EC50 = 1.0 +/- 0.4 nM) than was delta-Dtx (EC50 = 17.6 +/- 5.8 nM). Furthermore, delta-Dtx was selective for NIC (EC50 +/- 0.24 +/- 0.03 nM) over SIC and reduced the slow component of tail currents (NIC), preferentially. On the other hand, a-Dtx did not significantly distinguish between SIC and NIC although tail current analysis showed that a-Dtxpreferentially reduced the fast component of tail currents (SIC).6. The results confirm, using direct electrophysiological methods, that homologues of dendrotoxins from Mamba snake venom block K+ channels in rat sensory neurones. Furthermore, a-Dtx and 6-Dtx distinguish between sub-types of K+ channels in these cells and may thus be useful pharmacological tools in other neuronal K+ channel studies.

Proteinase inhibitor homologues as potassium channel blockers

We report here the NMR structure of dendrotoxin I, a powerful potassium channel blocker from the venom of the African Elapidae snake Dendroaspis polylepis polylepis (black mamba), calculated from an experimentally-derived set of 719 geometric restraints. The backbone of the toxin superimposes on bovine pancreatic trypsin inhibitor (BPTI) with a root-mean-square deviation of < 1.7 A. The surface electrostatic potential calculated for dendrotoxin I and BPTI, reveal an important difference which might account for the differences in function of the two proteins. These proteins may provide examples of adaptation for specific and diverse biological functions while at the same time maintaining the overall three-dimensional structure of a common ancestor.