Yu. N. Utkin

Interacting surfaces of neurotoxins and acetylcholine receptor

Binding of neurotoxin II Naja naja oxiana derivatives containing one spin label at various positions (Leu 1, Glu 2, Lys 15, Lys 25, Lys 26, His 31, Lys 44 and Lys 46) to purified solubilized acetylcholine receptor protein (AchR) from Torpedo marmorata was studied by EPR techniques. AchR interaction with several dansylated neurotoxin II derivatives was followed by difference fluorescence spectroscopy. A series of neurotoxin II p-azidobenzoyl derivatives were prepared and in three of them modified lysine residues were identified. In combination, spectroscopic data and photolabeling implicate a considerable area of the neurotoxin in association with AchR. Rigidity of the neurotoxin II conformation allowed to regard its binding surface as a mould of the AchR corresponding site and to estimate the minimal size of the latter. Conformation of the long-chain neurotoxins and their binding to AchR are briefly discussed basing on the 1H and 19F NMR studies of neurotoxin I Naja naja oxiana, toxin 3 Naja naja siamensis and its acetylated or trifluoroacetylated derivatives, as well as on Achr interaction with the derivatives spin labeled at Lys 27 and His 71.

Comparative Study of Structure and Activity of Cytotoxins from Venom of the Cobras Naja oxiana, Naja kaouthia, and Naja haje

Cytotoxins are positively charged polypeptides that constitute about 60% of all proteins in cobra venom; they have a wide spectrum of biological activities. By CD spectroscopy, cytotoxins CT1 and CT2 Naja oxiana, CT3 Naja kaouthia, and CT1 and CT2 Naja haje were shown to have similar secondary structure in an aqueous environment, with dominating β-sheet structure, and to vary in the twisting angle of the β-sheet and the conformation of disulfide groups. Using dodecylphosphocholine micelles and liposomes, CT1 and CT2 Naja oxiana were shown to incorporate into lipid structures without changes in the secondary structure of the peptides. The binding of CT1 and CT2 Naja oxiana with liposomes was associated with an increase in the β-sheet twisting and a sign change of the dihedral angle of one disulfide group. The cytotoxins were considerably different in cytotoxicity and cooperativity of the effect on human promyelocytic leukemia cells HL60, mouse myelomonocytic cells WEHI-3, and human erythroleukemic cells K562. The most toxic CT2 Naja oxiana and CT3 Naja kaouthia possessed low cooperativity of interaction (Hill coefficient h = 0.6-0.8), unlike 10-20-fold less toxic CT1 and CT2 Naja haje (h = 1.2-1.7). CT1 Naja oxiana has an intermediate position on the cytotoxicity scale and is characterized by h = 0.5-0.8. The cytotoxins under study induced necrosis of HL60 cells and failed to activate apoptosis. The differences in cytotoxicity are supposed to be related not with features of the secondary structure of the peptides, but with interactions of side chains of variable amino acid residues with lipids and/or membrane proteins.

Relationship between the binding sites for an α-conotoxin and snake venom neurotoxins in the nicotinic acetylcholine receptor from Torpedo californica

Photoinduced cross-links between the iodinated Lys26-p-azidobenzoyl derivative of neurotoxin II from Naja naja oxiana cobra venom and nicotinic acetylcholine receptor from Torpedo californica (AChR) have been studied in the presence of alpha-conotoxin GI from the marine snail C. geographus. Preincubation of the AChR-enriched membranes with increasing concentrations of alpha-conotoxin GI protects first the gamma subunit from photolabelling and then the delta subunit, the IC50 values being 0.76 and 5.01 microM, respectively. The results obtained, in view of the relevant data in literature, demonstrate that the (alpha + gamma) site, which is the high affinity site for d-tubocurarine, has also a higher affinity for an alpha-conotoxin than the (alpha + delta) containing site. The latter has a somewhat higher affinity than the (alpha + gamma) site towards some naturally occurring snake venom alpha-neurotoxins or their derivatives.

Diversity of nicotinic receptors mediating Cl- current in Lymnaea neurons distinguished with specific agonists and antagonist.

Diversity of nicotinic acetylcholine receptors (nAChRs) mediating Cl- current in voltage-clamped identifiable Lymnaea stagnalis neurons was studied using acetylcholine (ACh), three agonists and alpha-conotoxin ImI (ImI). Cytisine, nicotine, and choline, full agonists at alpha7 subunit-containing nAChRs of vertebrates, were found to evoke at saturating concentration 84-92% of the maximal current elicited by ACh. ImI, known to block selectively alpha7 and alpha9 nAChRs, markedly diminished the responses to ACh. The average maximal ImI-induced block was 80%, leaving a residual current which had very slow kinetics. The choline-, cytisine-, and nicotine-induced currents were blocked by ImI almost completely, suggesting that they activate only ImI-sensitive receptors. Two groups of cells which differ in desensitization kinetics and in sensitivity to ImI were revealed. IC50 values for ImI against ACh were 10.3 and 288 nM, respectively, with the rapidly desensitizing current being the more sensitive to ImI. The data obtained suggest the existence of at least three pharmacologically distinct subtypes of nicotinic receptors in Lymnaea neurons. Two of the subtypes are similar to alpha7 nAChRs of vertebrates, but differ from each other in their affinity for ImI and in their desensitization kinetics. The third subtype is quite distinct, in that it is resistant to ImI, is not activated by nicotine, cytisine or choline, and mediates a very slowly developing current.

Nicotinic receptors in Lymnaea stagnalis neurons are blocked by α-neurotoxins from cobra venoms

Abstract The influence of cobra neurotoxins on the Cl-dependent responses to acetylcholine (ACh) of Lymnaea neurons was studied by the voltage-clamp technique. It was found that a short chain neurotoxin II (NT II), a long chain cobratoxin (CTX) and weak neurotoxin (WTX) diminished the ACh-induced currents, the block being concentration-dependent and competitive. The IC 50 values of 130 nM for CTX, 11 μM for NT II, and 67 μM for WTX were determined. The block induced by NT II was quickly reversible upon toxin washout, whereas the action of CTX and WTX was only partially reversible even after an hour of intensive washing. The data obtained suggest that acetylcholine receptors (AChRs) in Lymnaea neurons have common features with cation-selective α7 AChRs of vertebrates and one type of Aplysia Cl-conducting AChRs.

Bacterial production and refolding from inclusion bodies of a “Weak” toxin, a disulfide rich protein

The gene for the “weak” toxin of Naja kaouthia venom was expressed in Escherichia coli. “Weak” toxin is a specific inhibitor of nicotine acetylcholine receptor, but mechanisms of interaction of similar neurotoxins with receptors are still unknown. Systems previously elaborated for neurotoxin II from venom of the cobra Naja oxiana were tested for bacterial production of “weak” toxin from N. kaouthia venom. Constructs were designed for cytoplasmic production of N. kaouthia “weak” toxin in the form of a fused polypeptide chain with thioredoxin and for secretion with the leader peptide STII. However, it became possible to obtain “weak” toxin in milligram amounts only within cytoplasmic inclusion bodies. Different approaches for refolding of the toxin were tested, and conditions for optimization of the yield of the target protein during refolding were investigated. The resulting protein was characterized by mass spectrometry and CD and NMR spectroscopy. Experiments on competitive inhibition of 125I-labeled α-bungarotoxin binding to the Torpedo californica electric organ membranes containing the muscle-type nicotine acetylcholine receptor (α12β1γδ) showed the presence of biological activity of the recombinant “weak” toxin close to the activity of the natural toxin (IC50 = 4.3 ± 0.3 and 3.0 ± 0.5 µM, respectively). The interaction of the recombinant toxin with α7 type human neuronal acetylcholine receptor transfected in the GH4C1 cell line also showed the presence of activity close to that of the natural toxin (IC50 31 ± 5.0 and 14.8 ± 1.3 µM, respectively). The developed bacterial system for production of N. kaouthia venom “weak” toxin was used to obtain 15N-labeled analog of the neurotoxin.

Benzophenone-type photoactivatable derivatives of alpha-neurotoxins and alpha-conotoxins in studies on Torpedo nicotinic acetylcholine receptor.

By chemical modification of different lysine residues, benzoylbenzoyl (BzBz) groups were introduced into neurotoxin II Naja naja oxiana (NT-II), a short-chain snake venom alpha-neurotoxin, while p-benzoylphenylalanyl (Bpa) residue was incorporated in the course of peptide synthesis at position 11 of alpha-conotoxin G1, a neurotoxic peptide from marine snails. Although the crosslinking yields for iodinated BzBz derivatives of NT-II and for Bpa analogue of G1 to the membrane-bound Torpedo californica nicotinic acetylcholine receptor (AChR) are relatively low, the subunit labeling patterns confirm the earlier conclusions, derived from arylazide or diazirine photolabels, that alpha-neurotoxins and alpha-conotoxins bind at the subunit interfaces. Detecting the labeled alpha-subunit with iodinated Bpa analogue of G1 provided a direct proof for the contact between this subunit and alpha-conotoxin molecule.

Influence of phospholipases A2 from snake venoms on survival and neurite outgrowth in pheochromocytoma cell line PC12.

To determine whether the ability to induce neurite outgrowth in rat pheochromocytoma cell line PC12 is characteristic of phospholipases of different types, we have studied the influence of phospholipase A2 (PLA2) from cobra Naja kaouthia venom and two PLA2s from viper Vipera nikolskii venom on PC12 cells. Phospholipases from the viper venom are heterodimers in which only one of the subunits is enzymatically active, while PLA2 from the cobra venom is a monomer. It was found that all three PLA2s induce neurite outgrowth in PC12. The PLA2 from cobra venom exhibits this effect at higher concentrations as compared to the viper enzymes. We have not observed such an activity for isolated subunits of viper PLA2s, since the enzymatically active subunits have very high cytotoxicity, while the other subunits are not active at all. However, co-incubation of active and inactive subunits before addition to the cells leads to a marked decrease in cytotoxicity and to restoration of the neurite-inducing activity. It has also been shown that all enzymatically active PLA2s are cytotoxic, the PLA2 from cobra venom being the least active. Thus, for the first time we have shown that PLA2s from snake venoms can induce neurite outgrowth in PC12 cells.

The MALDI mass spectrometry in the identification of new proteins in snake venoms

By MALDI MS, we searched cobra venoms for new low-content polypeptides. A number of new proteins with molecular masses 7–25 kDa, characteristic of the known snake protein toxins, were identified, with the content of one of them less than 0.02%.

New weak toxins from the cobra venom

A protein with M 7485 Da containing five disulfide bonds was isolated from the venom of cobra Naja oxiana using various types of liquid chromatography. The complete amino acid sequence of the protein was determined by protein chemistry methods, which permitted us to assign it to the group of weak toxins. This is the first weak toxin isolated from the venom of N. oxiana. In a similar way, two new toxins with M 7628 and 7559 Da, which fall into the range of weak toxin masses, were isolated from the venom of the cobra N. kaouthia. The characterization of these proteins using Edman degradation and MALDI mass spectrometry has shown that one of these proteins is a novel weak toxin, and the other is the known weak toxin WTX with an oxidized methionine residue in position 9. Such a modification was detected in weak toxins for the first time. A study of the biological activity of the toxin from N. oxiana showed that, like other weak toxins, it can be bound by α7 and muscle-type nicotinic acetylcholine receptors.