A subfamily of acidic alpha-K(+) toxins

Three homologous acidic peptides have been isolated from the venom of three different Parabuthus scorpion species, P. transvaalicus, P. villosus, and P. granulatus. Analysis of the primary sequences reveals that they structurally belong to subfamily 11 of short chain alpha-K(+)-blocking peptides (Tytgat, J., Chandy, K. G., Garcia, M. L., Gutman, G. A., Martin-Eauclaire, M. F., van der Walt, J. J., and Possani, L. D. (1999) Trends Pharmacol. Sci. 20, 444-447). These toxins are 36-37 amino acids in length and have six aligned cysteine residues, but they differ substantially from the other alpha-K(+) toxins because of the absence of the critical Lys(27) and their total overall negative charge. Parabutoxin 1 (PBTx1), which has been expressed by recombinant methods, has been submitted to functional characterization. Despite the lack of the Lys(27), this toxin blocks several Kv1-type channels heterologously expressed in Xenopus oocytes but with low affinities (micromolar range). Because a relationship between the biological activity and the acidic residue substitutions may exist, we set out to elucidate the relative impact of the acidic character of the toxin and the lack of the critical Lys(27) on the weak activity of PBTx1 toward Kv1 channels. To achieve this, a specific mutant named rPBTx1 T24F/V26K was made recombinantly and fully characterized on Kv1-type channels heterologously expressed in Xenopus oocytes. Analysis of rPBTx1 T24F/V26K displaying an affinity toward Kv1.2 and Kv1.3 channels in the nanomolar range shows the importance of the functional dyad above the acidic character of this toxin.     

Identification of a peptide toxin from Grammostola spatulata spider venom that blocks cation-selective stretch-activated channels

We have identified a 35 amino acid peptide toxin of the inhibitor cysteine knot family that blocks cationic stretch-activated ion channels. The toxin, denoted GsMTx-4, was isolated from the venom of the spider Grammostola spatulata and has <50% homology to other neuroactive peptides. It was isolated by fractionating whole venom using reverse phase HPLC, and then assaying fractions on stretch-activated channels (SACs) in outside-out patches from adult rat astrocytes. Although the channel gating kinetics were different between cell-attached and outside-out patches, the properties associated with the channel pore, such as selectivity for alkali cations, conductance ( approximately 45 pS at -100 mV) and a mild rectification were unaffected by outside-out formation. GsMTx-4 produced a complete block of SACs in outside-out patches and appeared specific since it had no effect on whole-cell voltage-sensitive currents. The equilibrium dissociation constant of approximately 630 nM was calculated from the ratio of association and dissociation rate constants. In hypotonically swollen astrocytes, GsMTx-4 produces approximately 40% reduction in swelling-activated whole-cell current. Similarly, in isolated ventricular cells from a rabbit dilated cardiomyopathy model, GsMTx-4 produced a near complete block of the volume-sensitive cation-selective current, but did not affect the anion current. In the myopathic heart cells, where the swell-induced current is tonically active, GsMTx-4 also reduced the cell size. This is the first report of a peptide toxin that specifically blocks stretch-activated currents. The toxin affect on swelling-activated whole-cell currents implicates SACs in volume regulation.     

Phospholipase A2 engineering. X-ray structural and functional evidence for the interaction of lysine-56 with substrates.

Site-directed mutagenesis studies of bovine pancreatic phospholipase A2 (PLA2, overproduced in Escherichia coli) showed that replacement of surface residue Lys-56 by a neutral or hydrophobic amino acid residue resulted in an unexpected and significant change in the function of the enzyme. The kcat for phosphatidylcholine micelles increases 3-4-fold for K56M, K56I, and K56F and ca. 2-fold for K56N and K56T but does not change for K56R. These results suggest that the side chain of residue 56 has significant influence on the activity of PLA2. In order to probe the structural basis for the enhanced activity, the crystal structures of wild-type and K56M PLA2 were determined by X-ray crystallography to a resolution of 1.8 A. The results suggest that the mutation has not only perturbed the conformation of the side chain of Met-56 locally but also caused conformational changes in the neighboring loop (residues 60-70), resulting in the formation of a hydrophobic pocket by residues Met-56, Tyr-52, and Tyr-69. Docking of a phosphatidylcholine inhibitor analogue into the active site of K56M, according to the structure of the complex of cobra venom PLA2-phosphatidylethanolamine inhibitor analogue [White, S.P., Scott, D. L., Otwinowski, Z., Gleb, M. H., & Sigler, P. (1990) Science 250, 1560-1563], showed that the choline moiety [N(CH3)3]+ is readily accommodated into the newly formed hydrophobic pocket with a high degree of surface complementarity. This suggests a possible interaction between residue 56 and the head group of the phospholipid, explaining the enhanced activities observed when the positively charged Lys-56 is substituted by apolar residues, viz., K56M, K56I, and K56F. Further support for this interpretation comes from the 5-fold enhancement in kcat for the mutant K56E with a negatively charged side chain, where there would be an attractive electrostatic interaction between the side chain of Glu-56 and the positively charged choline moiety. Our results also refute a recent report [Tomasselli, A. G., Hui, J., Fisher, J., Zürcher-Neely, H., Reardon, I.M., Oriaku, E., Kézdy, F.J., & Heinrikson, R.L. (1989) J. Biol. Chem. 264, 10041-10047] that substrate-level acylation of Lys-56 is an obligatory step in the catalysis by PLA2.     

Binding of correolide to K(v)1 family potassium channels. Mapping the domains of high affinity interaction.

Correolide, a novel nortriterpene natural product, potently inhibits the voltage-gated potassium channel, K(v)1.3, and [(3)H]dihydrocorreolide (diTC) binds with high affinity (K(d) approximately 10 nM) to membranes from Chinese hamster ovary cells that express K(v)1.3 (Felix, J. P., Bugianesi, R. M., Schmalhofer, W. A., Borris, R., Goetz, M. A., Hensens, O. D., Bao, J.-M., Kayser, F. , Parsons, W. H., Rupprecht, K., Garcia, M. L., Kaczorowski, G. J., and Slaughter, R. S. (1999) Biochemistry 38, 4922-4930). Mutagenesis studies were used to localize the diTC binding site and to design a high affinity receptor in the diTC-insensitive channel, K(v)3.2. Transferring the pore from K(v)1.3 to K(v)3.2 produces a chimera that binds peptidyl inhibitors of K(v)1.3 with high affinity, but not diTC. Transfer of the S(5) region of K(v)1.3 to K(v)3.2 reconstitutes diTC binding at 4-fold lower affinity as compared with K(v)1.3, whereas transfer of the entire S(5)-S(6) domain results in a normal K(v)1.3 phenotype. Substitutions in S(5)-S(6) of K(v)1.3 with nonconserved residues from K(v)3.2 has identified two positions in S(5) and one in S(6) that cause significant alterations in diTC binding. High affinity diTC binding can be conferred to K(v)3.2 after substitution of these three residues with the corresponding amino acids found in K(v)1.3. These results suggest that lack of sensitivity of K(v)3.2 to diTC is a consequence of the presence of Phe(382) and Ile(387) in S(5), and Met(458) in S(6). Inspection of K(v)1.1-1.6 channels indicates that they all possess identical S(5) and S(6) domains. As expected, diTC binds with high affinity (K(d) values 7-21 nM) to each of these homotetrameric channels. However, the kinetics of binding are fastest with K(v)1.3 and K(v)1.4, suggesting that conformations associated with C-type inactivation will facilitate entry and exit of diTC at its binding site. Taken together, these findings identify K(v)1 channel regions necessary for high affinity diTC binding, as well as, reveal a channel conformation that markedly influences the rate of binding of this ligand.     

Lipid membrane interaction and antimicrobial activity of GsMTx-4, an inhibitor of mechanosensitive channel

GsMTx-4, a polypeptide from the spider Grammostola spatulata, is an inhibitor of mechanosensitive channels. It is known to interact with lipid membranes, suggesting it partitions into the membrane to alter the channel gating, but the effect of the membrane charge on GsMTx-4 activity remains unknown. In this study, we found that GsMTx-4 more effectively interacts with anionic lipids than zwitterionic ones. The effect of GsMTx-4 on negatively charged membranes was similar to that of the antimicrobial peptide melittin, which led us to assess GsMTx-4's antimicrobial activity. Interestingly, we found that, in contrast to other neurotoxins, GsMTx-4 exhibited antimicrobial properties and was more active against Gram-positive than Gram-negative bacteria. These results suggest that GsMTx-4 exerts its antimicrobial effect by altering the packing of the membrane and/or inhibiting mechanosensitive channels. These findings could point the way towards a new class of antimicrobial peptides.     

A single arginyl residue in plastocyanin and in cytochrome c(6) from the cyanobacterium Anabaena sp. PCC 7119 is required for efficient reduction of photosystem I

Positively charged plastocyanin from Anabaena sp. PCC 7119 was investigated by site-directed mutagenesis. The reactivity of its mutants toward photosystem I was analyzed by laser flash spectroscopy. Replacement of arginine at position 88, which is adjacent to the copper ligand His-87, by glutamine and, in particular, by glutamate makes plastocyanin reduce its availability for transferring electrons to photosystem I. Such a residue in the copper protein thus appears to be isofunctional with Arg-64 (which is close to the heme group) in cytochrome c(6) from Anabaena (Molina-Heredia, F. P., Diaz-Quintana, A., Hervás, M., Navarro, J. A., and De la Rosa, M. A. (1999) J. Biol. Chem. 274, 33565-33570) and Synechocystis (De la Cerda, B., Diaz-Quintana, A., Navarro, J. A. , Hervás, M., and De la Rosa, M. A. (1999) J. Biol. Chem. 274, 13292-13297). Other mutations concern specific residues of plastocyanin either at its positively charged east face (D49K, H57A, H57E, K58A, K58E, Y83A, and Y83F) or at its north hydrophobic pole (L12A, K33A, and K33E). Mutations altering the surface electrostatic potential distribution allow the copper protein to modulate its kinetic efficiency: the more positively charged the interaction site, the higher the rate constant. Whereas replacement of Tyr-83 by either alanine or phenylalanine has no effect on the kinetics of photosystem I reduction, Leu-12 and Lys-33 are essential for the reactivity of plastocyanin.     

Sequence comparison of the 63-, 61-, and 59-kDa calmodulin-dependent cyclic nucleotide phosphodiesterases.

Partial protein sequences from the 59-kDa bovine heart and the 63-kDa bovine brain calmodulin-dependent phosphodiesterases (CaM-PDEs) were determined and compared to the sequence of the 61-kDa isozyme reported by Charbonneau et al. [Charbonneau, H., Kumar, S., Novack, J. P., Blumenthal, D. K., Griffin, P. R., Shabanowitz, J., Hunt, D. F., Beavo, J. A. & Walsh, K. A. (1991) Biochemistry (preceding paper in this issue)]. Only a single segment (34 residues) at the N-terminus of the 59-kDa isozyme lacks identity with the 61-kDa isozyme; all other assigned sequence is identical in the two isozymes. Peptides from the 59-kDa isozyme that correspond to residues 23-41 of the 61-kDa protein bind calmodulin with high affinity. The C-terminal halves of these calmodulin-binding peptides are identical to the corresponding 59-kDa sequence; the N-terminal halves differ. The localization of sequence differences within this single segment suggests that the 61- and 59-kDa isozymes are generated from a single gene by tissue-specific alternative RNA splicing. In contrast, partial sequence from the 63-kDa bovine brain CaM-PDE isozyme displays only 67% identity with the 61-kDa isozyme. The differences are dispersed throughout the sequence, suggesting that the 63- and 61-kDa isozymes are encoded by separate but homologous genes.     

Selectivity filter residues contribute unequally to pore stabilization in voltage-gated sodium channels

Mutations in the putative selectivity filter region of the voltage-gated Na+ channel, the so-called DEKA-motif, not only affect selectivity but also alter the channel's gating properties, suggesting functional coupling between permeation and gating. We have previously reported that charge-altering mutations at position 1237 in the P-loop of domain III (position K of the DEKA-motif in the adult rat skeletal muscle Na+ channel, rNa(v)1.4) dramatically enhanced entry to an inactivated state from which the channels recovered with a very slow time constant on the order of approximately 100 s (Todt, H., Dudley, S. C. J., Kyle, J. W., French, R. J., and Fozzard, H. A. (1999) Biophys. J. 76, 1335-1345). This state, termed "ultra-slow inactivation", may reflect a complex molecular rearrangement of the channel's pore region that involves both the extracellular and the cytoplasmic pore. Here, we tested whether charged DEKA-motif residues other than K1237 were also important determinants of a channel's gating properties. Therefore, we constructed the charge-neutralizing mutations D400A, E755A, and K1237A and studied the effects of these mutations on I(US). We found that, compared to wild-type rNa(v)1.4 channels, mutant D400A and K1237A but not E755A channels exhibited enhanced entry into ultra-slow inactivation. Selectivity for Na+ over K+, as judged from shifts in reversal potentials, was preserved in D400A, reduced in E755A, and completely lost in K1237A. These data suggest that an electrostatic interaction between the positively charged residue K1237 and the negatively charged residue D400 stabilizes the structure of the pore and thereby prevents I(US). Moreover, the interaction between K1237 and E755 appears to provide the basis for selective permeation of Na+ over K+.     

Why is quinidine an inhibitor of cytochrome P450 2D6? The role of key active-site residues in quinidine binding

We have previously shown that Phe(120), Glu(216), and Asp(301) in the active site of cytochrome P450 2D6 (CYP2D6) play a key role in substrate recognition by this important drug-metabolizing enzyme (Paine, M. J., McLaughlin, L. A., Flanagan, J. U., Kemp, C. A., Sutcliffe, M. J., Roberts, G. C., and Wolf, C. R. (2003) J. Biol. Chem. 278, 4021-4027 and Flanagan, J. U., Maréchal, J.-D., Ward, R., Kemp, C. A., McLaughlin, L. A., Sutcliffe, M. J., Roberts, G. C., Paine, M. J., and Wolf, C. R. (2004) Biochem. J. 380, 353-360). We have now examined the effect of mutations of these residues on interactions of the enzyme with the prototypical CYP2D6 inhibitor, quinidine. Abolition of the negative charge at either or both residues 216 and 301 decreased quinidine inhibition of bufuralol 1'-hydroxylation and dextromethorphan O-demethylation by at least 100-fold. The apparent dissociation constants (K(d)) for quinidine binding to the wild-type enzyme and the E216D and D301E mutants were 0.25-0.50 microm. The amide substitution of Glu(216) or Asp(301) resulted in 30-64-fold increases in the K(d) for quinidine. The double mutant E216Q/D301Q showed the largest decrease in quinidine affinity, with a K(d) of 65 microm. Alanine substitution of Phe(120), Phe(481),or Phe(483) had only a minor effect on the inhibition of bufuralol 1'-hydroxylation and dextromethorphan O-demethylation and on binding. In contrast to the wild-type enzyme, a number of the mutants studied were found to be able to metabolize quinidine. E216F produced O-demethylated quinidine, and F120A and E216Q/D301Q produced both O-demethylated quinidine and 3-hydroxyquinidine metabolites. Homology modeling and molecular docking were used to predict the modes of quinidine binding to the wild-type and mutant enzymes; these were able to rationalize the experimental observations.     

Inhibition of a cathepsin L-like cysteine protease by a chimeric propeptide-derived inhibitor

Like other papain-related cathepsins, congopain from Trypanosoma congolense is synthesized as a zymogen. We have previously identified a proregion-derived peptide (Pcp27), acting as a weak and reversible inhibitor of congopain. Pcp27 contains a 5-mer YHNGA motif, which is essential for selectivity in the inhibition of its mature form [Lalmanach, G., Lecaille, F., Chagas, J. R., Authié, E., Scharfstein, J., Juliano, M. A., and Gauthier, F. (1998) J. Biol. Chem. 273, 25112-25116]. In the work presented here, a homology model of procongopain was generated and subsequently used to model a chimeric 50-mer peptide (called H3-Pcp27) corresponding to the covalent linkage of an unrelated peptide (H3 helix from Antennapedia) to Pcp27. Molecular simulations suggested that H3-Pcp27 (pI = 9.99) maintains an N-terminal helical conformation, and establishes more complementary electrostatic interactions (E(coul) = -25.77 kcal/mol) than 16N-Pcp27, the 34-mer Pcp27 sequence plus the 16 native residues upstream from the proregion (E(coul) = 0.20 kcal/mol), with the acid catalytic domain (pI = 5.2) of the mature enzyme. In silico results correlated with the significant improvement of congopain inhibition by H3-Pcp27 (K(i) = 24 nM), compared to 16N-Pcp27 (K(i) = 1 microM). In addition, virtual alanine scanning of H3 and 16N identified the residues contributing most to binding affinity. Both peptides did not inhibit human cathepsins B and L. In conclusion, these data support the notion that the positively charged H3 helix favors binding, without modifying the selectivity of Pcp27 for congopain.     

Role of the ortholog and paralog amino acid invariants in the active site of the UDP-MurNAc-L-alanine:D-glutamate ligase (MurD).

To evaluate their role in the active site of the UDP-N-acetylmuramoyl-L-alanine:D-glutamate ligase (MurD) from Escherichia coli, 12 residues conserved either in the Mur superfamily [Eveland, S. S., Pompliano, D. L., and Anderson, M. S. (1997) Biochemistry 36, 6223-6229; Bouhss, A., Mengin-Lecreulx, D., Blanot, D., van Heijenoort, J., and Parquet, C. (1997) Biochemistry 36, 11556-11563] or in the sequences of 26 MurD orthologs were submitted to site-directed mutagenesis. All these residues lay within the cleft of the active site of MurD as defined by its 3D structure [Bertrand, J. A., Auger, D., Fanchon, E., Martin, L., Blanot, D., van Heijenoort, J., and Dideberg, O. (1997) EMBO J. 16, 3416-3425]. Fourteen mutant proteins (D35A, K115A, E157A/K, H183A, Y194F, K198A/F, N268A, N271A, H301A, R302A, D317A, and R425A) containing a C-terminal (His)(6) extension were prepared and their steady-state kinetic parameters determined. All had a reduced enzymatic activity, which in many cases was very low, but no mutation led to a total loss of activity. Examination of the specificity constants k(cat)/K(m) for the three MurD substrates indicated that most mutations affected both the binding of one substrate and the catalytic process. These kinetic results correlated with the assigned function of the residues based on the X-ray structures.     

Residues in Na(+) channel D3-S6 segment modulate both batrachotoxin and local anesthetic affinities

Batrachotoxin (BTX) alters the gating of voltage-gated Na(+) channels and causes these channels to open persistently, whereas local anesthetics (LAs) block Na(+) conductance. The BTX and LA receptors have been mapped to several common residues in D1-S6 and D4-S6 segments of the Na(+) channel alpha-subunit. We substituted individual residues with lysine in homologous segment D3-S6 of the rat muscle mu1 Na(+) channel from F1274 to N1281 to determine whether additional residues are involved in BTX and LA binding. Two mutant channels, mu1-S1276K and mu1-L1280K, when expressed in mammalian cells, become completely resistant to 5 microM BTX during repetitive pulses. The activation and/or fast inactivation gating of these mutants is substantially different from that of wild type. These mutants also display approximately 10-20-fold reduction in bupivacaine affinity toward their inactivated state but show only approximately twofold affinity changes toward their resting state. These results demonstrate that residues mu1-S1276 and mu1-L1280 in D3-S6 are critical for both BTX and LA binding interactions. We propose that LAs interact readily with these residues from D3-S6 along with those from D1-S6 and D4-S6 in close proximity when the Na(+) channel is in its inactivated state. Implications of this state-dependent binding model for the S6 alignment are discussed.     

Synthetic peptide analogues differentially alter the binding affinities of cyclic nucleotide dependent protein kinases for nucleotide substrates

Analogues of a synthetic heptapeptide substrate corresponding to the sequence around a phosphorylation site in histone H2B [Glass, D. B. & Krebs, E. G. (1982) J. Biol. Chem. 257, 1196-1200] were used to assess interactions between the peptide substrate and the ATP binding sites of cGMP-dependent protein kinase and the catalytic subunit of cAMP-dependent protein kinase. The affinity of each protein kinase for lin-benzo-ADP was determined in the absence and presence of substrate peptide by fluorescence anisotropy titrations [Bhatnagar, D., Roskoski, R., Jr., Rosendahl, M. S., & Leonard, N. J. (1983) Biochemistry 22, 6310-6317]. The Kd values of cGMP-dependent protein kinase for lin-benzo-ADP in the absence and presence of cGMP were 7.6 and 9.7 microM, respectively. Histone H2B(29-35) (Arg-Lys-Arg-Ser-Arg-Lys-Glu) had no effect on nucleotide affinity in either the absence or presence of cGMP. However, when lysine-34 located two residues after the phosphorylatable serine is replaced with an alanyl residue, the resulting [Ala34]histone H2B(29-35) and its analogue peptides interact with cGMP-dependent protein kinase and/or the nucleotide in a fashion that decreases nucleotide binding affinity approximately 3-fold. This amino acid replacement had previously been shown to cause an increase in Vmax and a decrease in the pH optimum for the phosphotransferase reaction. Replacement of positively charged residues at positions 30 and 31 of the peptide also decreased nucleotide affinity. Other analogues of histone H2B(29-35) failed to affect binding of lin-benzo-ADP to the active site of the cGMP-dependent enzyme.(ABSTRACT TRUNCATED AT 250 WORDS)     

The primary structure of rat platelet phospholipase A2

In our previous report (Hayakawa, M., Kudo, I., Tomita, M., & Inoue, K. (1988) J. Biochem. 103, 263-266), we have shown that phospholipases A2 purified from rat platelet membrane fractions and an extracellular medium of thrombin-stimulated rat platelets were essentially identical to each other. Both purified enzymes were digested with proteases, and the resulting peptides were subjected to primary sequence determination. The sequence analysis of the HPLC-separated peptides and the alignment of the sequences showed a tentative primary structure of rat platelet phospholipase A2, which was composed of 125 amino acid residues. It showed 47% homology with snake venom Agkistrodon halys blomhoffii phospholipase A2.     

Purification and cloning of toxins from elapid venoms that target cyclic nucleotide-gated ion channels

In 1999, we purified pseudechetoxin (PsTx), the first peptide toxin known to block cyclic nucleotide-gated (CNG) ion channels, from the venom of Pseudechis australis [Brown, R. L., Haley, T. L., West, K. A., and Crabb, J. W. (1999) Proc. Natl. Acad. Sci. U.S.A. 96, 754-759]. Here we report the cloning of the cDNA encoding PsTx, as well as the discovery and cloning of pseudecin, a homologous toxin from the venom of Pseudechis porphyriacus. The mature proteins are 211 and 210 amino acids in length, and the amino acid sequences are 96.7% identical, differing in only seven residues. The purified toxins were applied to outside-out patches excised from Xenopus oocytes expressing CNG channels composed of the rod CNGA1 or olfactory CNGA2 channel subunits. Surprisingly, these patch-clamp studies revealed a 30-fold difference in affinity between PsTx and pseudecin for channels composed of CNGA2 subunits. The apparent K(i) of PsTx was 15 nM, while the affinity of pseudecin was 460 nM. The difference in affinities for the CNGA1 subunit from rod photoreceptors was less pronounced, but the affinity of PsTx was 70 nM, compared with 1000 nM for pseudecin. This difference in affinity may be instructive as we attempt to identify the regions of the toxins that contact CNG channels. As the only known protein blockers of CNG channels, these toxins promise to be valuable tools to study the structure of the external face of these channels.     

Enhanced binding of a rationally designed peptide ligand of concanavalin a arises from improved geometrical complementarity

The structural basis of affinity enhancement was addressed by analyzing the interactions between concanavalin A and the carbohydrate-mimicking peptide ligands. Based on the crystal structures of concanavalin A in complex with these peptides [Jain, D., Kaur, K. J., Sundaravadivel, B., and Salunke, D. M. (2000) J. Biol. Chem. 275, 16098-16102; Jain, D., Kaur, K. J., and Salunke, D. M. (2001) Biophys. J. 80, 2912-2921], a high-affinity analogue was designed. This analogue (acetyl-MYWYPY-amide) binds to the lectin with 32-fold enhanced affinity compared to the corresponding precursor peptides. The crystal structure of concanavalin A bound to the designed peptide has been determined. A peptide molecule binds to each of the crystallographically independent monomers of the tetrameric lectin. The four bound peptide molecules exhibit two major conformations both of which are extended. Unlike in the case of other concanavalin A binding peptides, the structural variations within different conformers of this analogue are marginal. It is apparent that the deletion of the structurally variable region of the larger peptides has led to an improved complementarity and increased buried surface area in the case of the designed peptide. The crystal structure also showed the formation of two backbone hydrogen bonds between the ligand and the ligate which were not present in the complexes of the precursor peptides. The observed structural features adequately explain the enhanced binding of the designed analogue.     

Synthesis of a stable form of tertiapin: a high-affinity inhibitor for inward-rectifier K+ channels

Tertiapin (TPN), a small protein derived from honey bee venom, inhibits the GIRK1/4 and ROMK1 channels with nanomolar affinities. Methionine residue 13 in TPN interacts with residue F148 in the channel, located just outside of the narrow region of the ROMK1 pore. The methionine residue in TPN can be oxidized by air, which significantly hinders TPN binding to the channels. To overcome the reduction in TPN affinity due to oxidation of M13, we replaced M13 in TPN with fourteen different residues. Out of the fourteen derivatives, only the one in which M13 was replaced by glutamine, TPNQ, binds to the channel with a Ki value very similar to that of native TPN. Since TPNQ is stable and functionally resembles native TPN, it will be a very useful molecular probe for studying the inward-rectifier K+ channels.     

Synthesis,characterization, and application of cy-dye- and alexa-dye-labeled hongotoxin(1) analogues. The first high affinity fluorescence probes for voltage-gated K+ channels

Hongotoxin(1) (HgTX(1)), a 39-residue peptide recently isolated from the venom of Centruroides limbatus, blocks the voltage-gated K+ channels K(v)1.1, K(v)1.2, and K(v)1.3 at picomolar toxin concentrations (Koschak, A., Bugianesi, R. M., Mitterdorfer, J., Kaczorowski, G. J., Garcia, M. L., and Knaus, H. G. (1998) J. Biol. Chem. 273, 2639-2644). In this report, we determine the three-dimensional structure of HgTX(1) using NMR spectroscopy (PDB-code: 1HLY). HgTX(1) was found to possess a structure similar to previously characterized K+ channel toxins (e.g. margatoxin) consisting of a three-stranded antiparallel beta-sheet (residues 2-4, 26-30, and 33-37) and a helical conformation (part 3(10) helix and part alpha helix; residues 10-20). Due to the importance of residue Lys-28 for high-affinity interaction with the respective channels, lysine-reactive fluorescence dyes cannot be used to label wild-type HgTX(1). On the basis of previous studies (see above) and our NMR data, a HgTX(1) mutant (HgTX(1)-A19C) was engineered, expressed, and purified. HgTX(1)-A19C-SH was labeled using sulfhydryl-reactive Cy3-, Cy5-, and Alexa-dyes. Pharmacological characterization of fluorescently labeled HgTX(1)-A19C in radioligand binding studies indicated that these hongotoxin(1) analogues retain high-affinity for voltage-gated K+ channels and a respective pharmacological profile. Cy3- and Alexa-dye-labeled hongotoxin(1) analogues were used to investigate the localization of K+ channels in brain sections. The distribution of toxin binding closely follows the distribution of K(v)1.2 immunoreactivity with the highest expression levels in the cerebellar Purkinje cell layer. Taken together, these results demonstrate that fluorescently labeled HgTX(1) analogues comprise novel probes to characterize a subset of voltage-gated K+ channels.     

Solution structure of omega-grammotoxin SIA,a gating modifier of P/Q and N-type Ca(2+) channel

omega-Grammotoxin SIA (GrTx) is a 36 amino acid residue protein toxin from spider venom that inhibits P/Q and N-type voltage-gated Ca(2+) channels by modifying voltage-dependent gating. We determined the three-dimensional structure of GrTx using NMR spectroscopy. The toxin adopts an "inhibitor cystine knot" motif composed of two beta-strands (Leu19-Cys21 and Cys30-Trp32) and a beta-bulge (Trp6, Gly7-Cys30) with a +2x, -1 topology, which are connected by four chain reversals. Although GrTx was originally identified as an inhibitor of voltage-gated Ca(2+) channel, it also binds to K(+) channels with lower affinity. A similar cross-reaction was observed for Hanatoxin1 (HaTx), which binds to the voltage-sensing domains of K(+) and Ca(2+) channels with different affinities. A detailed comparison of the GrTx and HaTx structures identifies a conserved face containing a large hydrophobic patch surrounded by positively charged residues. The slight differences in the surface shape, which result from the orientation of the surface aromatic residues and/or the distribution of the charged residues, may explain the differences in the binding affinity of these gating modifiers with different voltage-gated ion channels.     

An ERG channel inhibitor from the scorpion Buthus eupeus

The isolation of the peptide inhibitor of M-type K(+) current, BeKm-1, from the venom of the Central Asian scorpion Buthus eupeus has been described previously (Fillipov A. K., Kozlov, S. A., Pluzhnikov, K. A., Grishin, E. V., and Brown, D. A. (1996) FEBS Lett. 384, 277-280). Here we report the cloning, expression, and selectivity of BeKm-1. A full-length cDNA of 365 nucleotides encoding the precursor of BeKm-1 was isolated using the rapid amplification of cDNA ends polymerase chain reaction technique from mRNA obtained from scorpion telsons. Sequence analysis of the cDNA revealed that the precursor contains a signal peptide of 21 amino acid residues. The mature toxin consists of 36 amino acid residues. BeKm-1 belongs to the family of scorpion venom potassium channel blockers and represents a new subgroup of these toxins. The recombinant BeKm-1 was produced as a Protein A fusion product in the periplasm of Escherichia coli. After cleavage and high performance liquid chromatography purification, recombinant BeKm-1 displayed the same properties as the native toxin. Three BeKm-1 mutants (R27K, F32K, and R27K/F32K) were generated, purified, and characterized. Recombinant wild-type BeKm-1 and the three mutants partly inhibited the native M-like current in NG108-15 at 100 nm. The effect of the recombinant BeKm-1 on different K(+) channels was also studied. BeKm-1 inhibited hERG1 channels with an IC(50) of 3.3 nm, but had no effect at 100 nm on hEAG, hSK1, rSK2, hIK, hBK, KCNQ1/KCNE1, KCNQ2/KCNQ3, KCNQ4 channels, and minimal effect on rELK1. Thus, BeKm-1 was shown to be a novel specific blocker of hERG1 potassium channels.