Search

Peptide toxins that adopt the ShK fold can inhibit the voltage-gated potassium channel K V 1.3 with IC 50 values in the pM range and are therefore potential leads for drugs targeting autoimmune and neuroinflammatory diseases. Nuclear magnetic resonance (NMR) relaxation measurements and pressure-dependent NMR have shown that, despite being cross-linked by disulfide bonds, ShK itself is flexible in solution. This flexibility affects the local structure around the pharmacophore for the K V 1.3 channel blockade and, in particular, the relative orientation of the key Lys and Tyr side chains (Lys22 and Tyr23 in ShK) and has implications for the design of K V 1.3 inhibitors. In this study, we have performed molecular dynamics (MD) simulations on ShK and a close homologue, HmK, to probe the conformational space occupied by the Lys and Tyr residues, and docked the different conformations with a recently determined cryo-EM structure of the K V 1.3 channel. Although ShK and HmK have 60% sequence identity, their dynamic behaviors are quite different, with ShK sampling a broad range of conformations over the course of a 5 μs MD simulation, while HmK is relatively rigid. We also investigated the importance of conformational dynamics, in particular the distance between the side chains of the key dyad Lys22 and Tyr23, for binding to K V 1.3. Although these peptides have quite different dynamics, the dyad in both adopts a similar configuration upon binding, revealing a conformational selection upon binding to K V 1.3 in the case of ShK. Both peptides bind to K V 1.3 with Lys22 occupying the pore of the channel. Intriguingly, the more flexible peptide, ShK, binds with significantly higher affinity than HmK.

Immunotoxins are proteins containing a cell-targeting element linked to a toxin that are under investigation for next-generation cancer treatment. However, these agents are difficult to synthesize, chemically heterogeneous, expensive, and show toxicity toward healthy cells. In this work, we describe the synthesis and characterization of a new type of immunotoxin that showed exquisite selectivity toward targeted cells. In our construct, targeting molecules were covalently attached or genetically fused to oligomeric pore-forming toxins. The activity of the immunotoxin was then caged by fusing a soluble protein to the transmembrane domain and activated via cleavage with furin, which is a protease that is overexpressed in many cancer cells. During the several coupling steps, directed evolution allowed the efficient synthesis of the molecules in E. coli cells, as well as selection for further specificity toward targeted cells. The final construct showed no off-target activity, while acquiring an additional degree of specificity toward the targeted cells upon activation. The pore-forming toxins described here do not require internalization to operate, while the many protomeric subunits can be individually modified to refine target specificity.

To realize the medicinal potential of peptide toxins, naturally occurring disulfide-rich peptides, as ion channel antagonists, more efficient pharmaceutical optimization technologies must be developed. Here, we show that the therapeutic properties of multiple cysteine toxin peptides can be rapidly and substantially improved by combining direct chemical strategies with high-throughput electrophysiology. We applied whole-molecule, brute-force, structure-activity analoging to ShK, a peptide toxin from the sea anemone Stichodactyla helianthus that inhibits the voltage-gated potassium ion channel Kv1.3, to effectively discover critical structural changes for 15× selectivity against the closely related neuronal ion channel Kv1.1. Subsequent site-specific polymer conjugation resulted in an exquisitely selective Kv1.3 antagonist (>1000× over Kv1.1) with picomolar functional activity in whole blood and a pharmacokinetic profile suitable for weekly administration in primates. The pharmacological potential of the optimized toxin peptide was demonstrated by potent and sustained inhibition of cytokine secretion from T cells, a therapeutic target for autoimmune diseases, in cynomolgus monkeys.

The sea anemone peptide APETx2 is a potent and selective blocker of acid-sensing ion channel 3 (ASIC3). APETx2 is analgesic in a variety of rodent pain models, but the lack of knowledge of its pharmacophore and binding site on ASIC3 has impeded development of improved analogues. Here we present a detailed structure-activity relationship study of APETx2. Determination of a high-resolution structure of APETx2 combined with scanning mutagenesis revealed a cluster of aromatic and basic residues that mediate its interaction with ASIC3. We show that APETx2 also inhibits the off-target hERG channel by reducing the maximal current amplitude and shifting the voltage dependence of activation to more positive potentials. Electrophysiological screening of selected APETx2 mutants revealed partial overlap between the surfaces on APETx2 that mediate its interaction with ASIC3 and hERG. Characterization of the molecular basis of these interactions is an important first step toward the rational design of more selective APETx2 analogues.

We have re-examined the utility of native chemical ligation at -Gln/Glu-Cys- [Glx-Cys] and -Asn/Asp-Cys- [Asx-Cys] sites. Using the improved thioaryl catalyst 4-mercaptophenylacetic acid (MPAA), native chemical ligation could be performed at -Gln-Cys- and Asn-Cys- sites without side reactions. After optimization, ligation at a -Glu-Cys- site could also be used as a ligation site, with minimal levels of byproduct formation. However, -Asp-Cys- is not appropriate for use as a site for native chemical ligation because of formation of significant amounts of β-linked byproduct. The feasibility of native chemical ligation at -Gln-Cys- enabled a convergent total chemical synthesis of the enantiomeric forms of the ShK toxin protein molecule. The D-ShK protein molecule was ~50,000-fold less active in blocking the Kv1.3 channel than the L-ShK protein molecule. Racemic protein crystallography was used to obtain high-resolution X-ray diffraction data for ShK toxin. The structure was solved by direct methods and showed significant differences from the previously reported NMR structures in some regions of the ShK protein molecule.

The voltage-gated potassium channel Kv1.3 is an attractive target for treatment of autoimmune diseases. ShK toxin from sea anemone is one of the most potent blockers of Kv1.3, and therefore ShK and its analogues have been proposed as therapeutic leads for such diseases. Increasing the selectivity of the proposed leads for Kv1.3 over other Kv1 channels is a major issue in this endeavor. Here we study binding of ShK toxin to Kv1 channels using free energy simulation methods. Homology models for Kv1.1 and Kv1.3 channels are constructed using the crystal structure of Kv1.2. The initial poses for the Kv1.x-ShK complexes are obtained using HADDOCK, which are then refined via molecular dynamics simulations. The binding mode in each complex is characterized by identifying the strongly interacting residues, which compare well with available mutagenesis studies. For each complex, the potential of mean force is calculated from umbrella sampling simulations, and the corresponding absolute binding free energy is determined. The computed binding free energies are in good agreement with the experimental data, which increases the confidence on the model complexes. The insights gained on Kv1.x-ShK binding modes will be valuable in the development of new ShK analogues with better selectivity properties.

Computational methods are employed to simulate the interaction of the sea anemone toxin ShK in complex with the voltage-gated potassium channel Kv1.3 from mice. All of the available 20 structures of ShK in the Protein Data Bank were considered for improving the performance of the rigid protein docking of ZDOCK. The traditional and novel binding modes were obtained among a large number of predicted complexes by using clustering analysis, screening with expert knowledge, energy minimization, and molecular dynamic simulations. The quality and validity of the resulting complexes were further evaluated to identify a favorable complex structure by 500 ps molecular dynamic simulations and the change of binding free energies with a computational alanine scanning technique. The novel and reasonable ShK-Kv1.3 complex structure was found to be different from the traditional model by using the Lys22 residue to block the channel pore. From the resulting structure of the ShK-Kv1.3 complex, ShK mainly associates the channel outer vestibule with its second helical segment. Structural analysis first revealed that the Lys22 residue side chain of the ShK peptide just hangs between C and D chains of the Kv1.3 channel instead of physically blocking the channel pore. The obvious loss of the ShK Ser20Ala and Tyr23Ala mutant binding ability to the Kv1.3 channel is caused by the conformational change. The five hydrogen bonds between Arg24 in ShK and H404(A) and D402(D) in Kv1.3 make Arg24 the most crucial for its binding to the Kv1.3 channel. Besides the detailed interaction between ShK and Kv1.3 at the atom level, the significant conformational change induced by the interaction between the ShK peptide and the Kv1.3 channel, accompanied by the gradual decrease of binding free energies, strongly implies that the binding of the ShK peptide toward the Kv1.3 channel is a dynamic process of conformational rearrangement and energy stabilization. All of these can accelerate the development of ShK structure-based immunosuppressants.

Type I sea anemone toxins are highly potent modulators of voltage-gated Na-channels (Na(v)s) and compete with the structurally dissimilar scorpion alpha-toxins on binding to receptor site-3. Although these features provide two structurally different probes for studying receptor site-3 and channel fast inactivation, the bioactive surface of sea anemone toxins has not been fully resolved. We established an efficient expression system for Av2 (known as ATX II), a highly insecticidal sea anemone toxin from Anemonia viridis (previously named A. sulcata), and mutagenized it throughout. Each toxin mutant was analyzed in toxicity and binding assays as well as by circular dichroism spectroscopy to discern the effects derived from structural perturbation from those related to bioactivity. Six residues were found to constitute the anti-insect bioactive surface of Av2 (Val-2, Leu-5, Asn-16, Leu-18, and Ile-41). Further analysis of nine Av2 mutants on the human heart channel Na(v)1.5 expressed in Xenopus oocytes indicated that the bioactive surfaces toward insects and mammals practically coincide but differ from the bioactive surface of a structurally similar sea anemone toxin, Anthopleurin B, from Anthopleura xanthogrammica. Hence, our results not only demonstrate clear differences in the bioactive surfaces of Av2 and scorpion alpha-toxins but also indicate that despite the general conservation in structure and importance of the Arg-14 loop and its flanking residues Gly-10 and Gly-20 for function, the surface of interaction between different sea anemone toxins and Na(v)s varies.

ShK, a peptide isolated from Stichodactyla helianthus venom, blocks the voltage-gated potassium channels, K(v)1.1 and K(v)1.3, with similar high affinity. ShK-Dap(22), a synthetic derivative in which a diaminopropionic acid residue has been substituted at position Lys(22), has been reported to be a selective K(v)1.3 inhibitor and to block this channel with equivalent potency as ShK [Kalman et al. (1998) J. Biol. Chem. 273, 32697-32707]. In this study, a large body of evidence is presented which indicates that the potencies of wild-type ShK peptide for both K(v)1.3 and K(v)1.1 channels have been previously underestimated. Therefore, the affinity of ShK-Dap(22) for both channels appears to be ca. 10(2)-10(4)-fold weaker than ShK. ShK-Dap(22) does display ca. 20-fold selectivity for human K(v)1.3 vs K(v)1.1 when measured by the whole-cell voltage clamp method but not in equilibrium binding assays. ShK-Dap(22) has low affinity for K(v)1.2 channels, but heteromultimeric K(v)1.1-K(v)1.2 channels form a receptor with ca. 200-fold higher affinity for ShK-Dap(22) than K(v)1.1 homomultimers. In fact, K(v)1.1-K(v)1.2 channels bind ShK-Dap(22) with only ca. 10-fold less potency than ShK and reveal a novel pharmacology not predicted from the homomultimers of K(v)1.1 or K(v)1.2. The concentrations of ShK-Dap(22) needed to inhibit human T cell activation were ca. 10(3)-fold higher than those of ShK, in good correlation with the relative affinities of these peptides for inhibiting K(v)1.3 channels. All of these data, taken together, suggest that ShK-Dap(22) will not have the same in vivo immunosuppressant efficacy of other K(v)1.3 blockers, such as margatoxin or ShK. Moreover, ShK-Dap(22) may have undesired side effects due to its interaction with heteromultimeric K(v)1.1-K(v)1.2 channels, such as those present in brain and/or peripheral tissues.

P-type ATPases such as the Na+,K+-ATPase (sodium pump) hydrolyze ATP to pump ions through biological membranes against their electrochemical gradients. The mechanisms that couple ATP hydrolysis to the vectorial ion transport are not yet understood, but unveiling structures that participate in ATP binding and in the formation of the ionophore might help to gain insight into this process. Looking at the alpha- and beta-phosphates of ATP as a pyrophosphate molecule, we found that peptides highly conserved among all soluble inorganic pyrophosphatases are also present in ion-transporting ATPases. Included therein are Glu48 and Lys56 of the Saccharomyces cerevisiae pyrophosphatase (SCE1-PPase) that are essential for the activity of this enzyme and have been shown in crystallographic analysis to interact with phosphate molecules. To test the hypothesis that equivalent amino acids are also essential for the activity of ion-transporting ATPases, Glu472 and Lys480 of the sodium pump alpha 1 subunit corresponding to Glu48 and Lys56 of SCE1-PPase were mutated to various amino acids. Mutants of the sodium pump alpha1 subunit were expressed in yeast and analyzed for their ATPase activity and their ability to bind ouabain in the presence of either ATP, Mg2+, and Na+ or phosphate and Mg2+. All four mutants investigated, Glu472Ala, Glu472Asp, Lys480Ala, and Lys480Arg, display only a fraction of the ATPase activity obtained with the wild-type enzyme. The same applies with respect to their ability to bind ouabain, where maximum ouabain binding to the mutants accounts for only about 10% of the binding obtained with the wild-type enzyme. On the basis of our results, we conclude that Glu472 and Lys480 are essential for the activity of the sodium pump. Their function is probably to arrest the alpha- and beta-phosphate groups of ATP in a proper position prior to hydrolysis of the gamma-phosphate group. The identification of these amino acids as essential components of the ATP-recognizing mechanism of the pump has resulted in a testable hypothesis for the initial interactions of the sodium pump, and possibly of other P-type ATPases, with ATP.

A new potassium channel toxin, HmK, has been isolated from the sea anemone Heteractis magnifica. It inhibits the binding of [125I]-alpha-dendrotoxin (a ligand for voltage-gated K channels) to rat brain synaptosomal membranes with a Ki of about 1 nM, blocks K+ currents through Kv 1.2 channels expressed in a mammalian cell line, and facilitates acetylcholine release at the avian neuromuscular junction. HmK comprises of 35 amino acids (Mr 4055) with the sequence R1TCKDLIPVS10ECTDIRCRTS20MKYRLNLCRK30TCGSC35. A full assignment of the disulfide linkages was made by using partial reduction with tri(2-carboxyethyl)phosphine (TCEP) at acid pH and rapid alkylation with iodoacetamide. The disulfide bridges were identified as Cys3-Cys35, Cys12-Cys28, and Cys17-Cys32. A cDNA clone encoding HmK was isolated using RT-PCR from the total RNA obtained from sea anemone tentacles, while the 5'- and 3'-flanking regions of the cDNA were amplified by RACE. The full-length cDNA was 563 bp long and contained a sequence encoding a signal peptide of 39 amino acids. The coding region for matured HmK toxin was cloned and expressed as a glutathione S-transferase (GST) fusion product in the cytoplasm of Escherichia coli. After affinity purification and cleavage, the recombinant toxin was shown to be identical to native HmK in its N-terminal sequence, chromatographic behavior, and binding to dendrotoxin binding sites on rat brain membranes.

Actinaria cytolysins are very potent basic toxins isolated from the venom of sea anemones, which are supposed to exert their toxic activity through formation of oligomeric pores in the host plasma membrane. To gain insight into their mechanism of action, the interaction of Stichodactyla helianthus sticholysin I (St-I) with lipid bilayers was studied. St-I increased the permeability of calcein-loaded lipid vesicles composed of different phospholipids. The rate of permeabilization improved when sphingomyelin (SM) was introduced into phosphatidylcholine (PC) vesicles, reaching an optimum value at equimolar concentrations of these two phospholipids. It was also a function of the pH, showing a local maximum of activity between pH 8 and 9 and a marked decrease at pH 10 and 11. Under optimal conditions (e.g., PC:SM 1:1, pH 8, toxin to vesicle ratio < 200), most of the toxin is bound to the lipid phase. The reduced toxin effect at low and high SM content, or at high pH, is principally due to a decreased toxin binding. From the dose dependence of the permeabilization, at constant lipid concentration, it was inferred that St-I increases membrane permeability by forming oligomeric pores comprising at least three cytolysin monomers. The involvement of oligomers was also suggested by the dependence of calcein release on the vesicle concentration at constant toxin dose. In fact, the time course of dye release was well described under all circumstances by a kinetic model which assumes that trimerization leads to a conductive pore. All the relevant equilibrium and rate constants were derived. Addition of St-I to one side of a planar lipid membrane increased the conductivity of the film in discrete steps of defined amplitude, indicating the formation of ion channels. The dose dependence of this effect was the same as with LUV. The channel was cation-selective and its conductance suggested a functional radius of about 1.0 nm, consistent with the size of the lesion previously observed in red blood cells. Pores exhibited rectification and voltage-dependent gating.

Transferrin, a serum glycoprotein, is a major regulator of cellular growth via its cellular receptor. Because transferrin receptors are absent from the plasma membranes of most normal adult resting cells, but are present on transformed, activated, and malignant cells, it can be used to address a toxin toward these cells. The cytolysin equinatoxin II, isolated from the sea anemone Actinia equina L., was coupled to human apo or diferric transferrin by using a heterobifunctional cross-linking reagent, N-succinimidyl 3-(2-pyridyldithio)propionate (SPDP). The conjugates were separated by column chromatography, and their composition was demonstrated by electrophoresis, antibody staining, and determination of the hemolytic activity in the absence or presence of a reducing agent. The average molar ratio of equinatoxin II to transferrin for the studied conjugates was found to be approximately 3.4. The activity of the conjugates against human erythrocytes and human tumor cells (Raji and Jurkat) was assessed. The conjugate is very active on tumor cells in vitro; however, the hybrid molecule maintains an unspecific hemolytic activity. This unspecific toxicity is due to the fact that transferrin-bound toxin partially retains its original ability to bind to the cell membrane directly. It could be strongly reduced (and even eliminated) by pretreating the conjugates with sphingomyelin, the natural ligand of sea anemone cytolysins. These conjugates were stable versus temperature (up to at least 40 degrees C), versus time (up to several weeks at 4 degrees C and at least 1 year at -80 degrees C), and versus repeated freeze-thaw cycles with liquid nitrogen (but not with -80 degrees C).

Lophotoxin is a small cyclic diterpene that irreversibly inhibits agonist binding to nicotinic acetylcholine receptors by reacting covalently with Tyr190 in the alpha-subunits of the receptor. Structure/activity and molecular modeling studies were undertaken to investigate the structural and conformational features responsible for this unique biological activity. A total of 18 naturally occurring and 7 chemically modified analogues were evaluated for their ability to inhibit the binding of [125I]-alpha-bungarotoxin to nicotinic acetylcholine receptors on membranes prepared from Torpedo electric organ. When the toxins were incubated with the receptor for short durations they did not slow the initial rate of binding of [125I]-alpha-bungarotoxin, suggesting that they have relatively low reversible affinity. However, their ability to inhibit the equilibrium binding of [125I]-alpha-bungarotoxin increased progressively with longer incubation times, consistent with an irreversible mechanism of action. Comparison of active and inactive analogues allowed identification of a conserved pharmacophore that appeared to be required for irreversible inhibition of the receptor. This pharmacophore contains lactone oxygens and an electron-deficient epoxide that may mimic the acetate oxygens and quaternary ammonium group of acetylcholine, respectively. Computer modeling of the toxins using molecular mechanics and dynamics revealed that the toxins have restricted conformational mobility, thus allowing identification of a minimum-energy conformation. The results allow speculation concerning the site of covalent reaction between Tyr190 and the toxins, the normal function of Tyr190 in binding acetylcholine, and the bound conformation of acetylcholine.

Eight different polypeptide toxins from sea anemones of four different origins (Anemonia sulcata, Anthopleura xanthogrammica, Stoichactis giganteus, and Actinodendron plumosum) have been studied. Three of these toxins are new; the purification procedure for the five other ones has been improved. Sea anemone toxins were assayed (i) for their toxicity to crabs and mice, (ii) for their affinity for the specific sea anemone toxin receptor situated on the Na+ channels of rat brain synaptosomes, and (iii) for their capacity to increase, in synergy with veratridine, the rate of 22Na+ entry into neuroblastoma cells via the Na+ channel. Some of the toxins are more active on crustaceans, whereas others are more toxic to mammals. A very good correlation exists between the toxic activity to mice, the affinity of the toxin for the Na+ channel in rat brain synaptosomes, and the stimulating effect on 22 Na+ uptake by neuroblastoma cells. The observation has also been made that the most cationic toxins are also the most active on mammals and the least active on crustaceans. Toxicities (LD50) to mice of the most active sea anemone toxins and of the most active scorpion toxins are similar, and sea anemone toxins at high enough concentrations prevent binding of scorpion toxins to their receptor. However, scorpion toxins have affinities for the Na+ channel which are approximately 60 times higher than those found for the most active sea anemone toxins. Three sea anemone toxins appear to be more interesting than toxin II from A. sulcata (the "classical" sea anemone toxin) for studies of the Na+ channel structure and mechanism when the source of the channel is of a mammalian origin. Two of these three toxins can be radiolabeled with iodine while retaining their toxic activity; they appear to be useful tools for future biochemical studies of the Na+ channel.

We have isolated a new toxin, calitoxin (CLX), from the sea anemone Calliactis parasitica whose amino acid sequence differs greatly from that of other sea anemone toxins. The polypeptide chain contains 46 amino acid residues, with a molecular mass of 4886 Da and an isoelectric point at pH 5.4. The amino acid sequence determined by Edman degradation of the reduced, S-carboxymethylated polypeptide chain and tryptic and chymotryptic peptides is Ile-Glu-Cys-Lys-Cys-Glu-Gly-Asp-Ala-Pro-Asp-Leu-Ser-His-Met-Thr-Gly-Thr- Val-Tyr - Phe-Ser-Cys-Lys-Gly-Gly-Asp-Gly-Ser-Trp-Ser-Lys-Cys-Asn-Thr-Tyr-Thr-Ala- Val-Ala - Asp-Cys-Cys-His-Glu-Ala. No cysteine residues were present in the peptide. Similarly to other sea anemone toxins, calitoxin interacts, in crustacean nerve muscle preparations, with axonal and not with muscle membranes, inducing a massive release of neurotransmitter that causes a strong muscle contraction. The low homology of CLX with RP II and ATX II toxins has implications regarding the role played by particular amino acid residues.

Four new toxins have been isolated from the sea anemone Radianthus paumotensis: RpI, RpII, RpIII, and RpIV. They are polypeptides comprised of 48 or 49 amino acids; the sequence of RpII has been determined. Toxicities of these toxins in mice and crabs are similar to those of the other known sea anemone toxins, but they fall into a different immunochemically defined class. The sequence of RpII shows close similarities with the N-terminal end (up to residue 20) of the previously sequenced long sea anemone toxins, but most of the remaining part of the molecule is completely different. Like the other sea anemone toxins, Radianthus toxins are active on sodium channels; they slow down the inactivation process. Through their Na+ channel action, Radianthus toxins stimulate Na+ influx into tetrodotoxin-sensitive neuroblastoma cells and tetrodotoxin-resistant rat skeletal myoblasts. The efficiency of the toxins is similar in the two cellular systems. In that respect, Radianthus toxins behave much more like scorpion neurotoxins than sea anemone toxins from Anemonia sulcata or Anthopleura xanthogrammica. In binding experiments to synaptosomal Na+ channels, Radianthus toxins compete with toxin II from the scorpion Androctonus australis but not with toxins II and V from Anemonia sulcata.