Venomics: unravelling the complexity of animal venoms with mass spectrometry.

Animal venoms and toxins are now recognized as major sources of bioactive molecules that may be tomorrow's new drug leads. Their complexity and their potential as drug sources have been demonstrated by application of modern analytical technologies, which have revealed venoms to be vast peptide combinatorial libraries. Structural as well as pharmacological diversity is immense, and mass spectrometry is now one of the major investigative tools for the structural investigation of venom components. Recent advances in its use in the study of venom and toxins are reviewed. The application of mass spectrometry techniques to peptide toxin sequence determination by de novo sequencing is discussed in detail, in the light of the search for novel analgesic drugs. We also present the combined application of LC-MALDI separation with mass fingerprinting and ISD fragmentation for the determination of structural and pharmacological classes of peptides in complex spider venoms. This approach now serves as the basis for the full investigation of complex spider venom proteomes, in combination with cDNA analysis.

Delta-atracotoxins from Australian funnel-web spiders compete with scorpion alpha-toxin binding on both rat brain and insect sodium channels.

Atracotoxins are novel peptide toxins from the venom of Australian funnel-web spiders that slow sodium current inactivation in a similar manner to scorpion alpha-toxins. To analyse their interaction with known sodium channel neurotoxin receptor sites we determined their effect on scorpion toxin, batrachotoxin and saxitoxin binding. Nanomolar concentrations of delta-atracotoxin-Hv1 and delta-atracotoxin-Ar1 completely inhibited the binding of the scorpion alpha-toxin AaH II to rat brain synaptosomes as well as the binding of LqhalphaIT, a scorpion alpha-toxin highly active on insects, to cockroach neuronal membranes. Moreover, delta-atracotoxin-Hv1 cooperatively enhanced batrachotoxin binding to rat brain synaptosomes in an analogous fashion to scorpion alpha-toxins. Thus the delta-atracotoxins represent a new class of toxins which bind to both mammalian and insect sodium channels at sites similar to, or partially overlapping with, the receptor binding sites of scorpion alpha-toxins.

Isolation and pharmacological characterisation of delta-atracotoxin-Hv1b, a vertebrate-selective sodium channel toxin.

delta-Atracotoxins (delta-ACTXs) are peptide toxins isolated from the venom of Australian funnel-web spiders that slow sodium current inactivation in a similar manner to scorpion alpha-toxins. We have isolated and determined the amino acid sequence of a novel delta-ACTX, designated delta-ACTX-Hv1b, from the venom of the funnel-web spider Hadronyche versuta. This 42 residue toxin shows 67% sequence identity with delta-ACTX-Hv1a previously isolated from the same spider. Under whole-cell voltage-clamp conditions, the toxin had no effect on tetrodotoxin (TTX)-resistant sodium currents in rat dorsal root ganglion neurones but exerted a concentration-dependent reduction in peak TTX-sensitive sodium current amplitude accompanied by a slowing of sodium current inactivation similar to other delta-ACTXs. However, delta-ACTX-Hv1b is approximately 15-30-fold less potent than other delta-ACTXs and is remarkable for its complete lack of insecticidal activity. Thus, the sequence differences between delta-ACTX-Hv1a and -Hv1b provide key insights into the residues that are critical for targeting of these toxins to vertebrate and invertebrate sodium channels.

Differing actions of convulsant and nonconvulsant barbiturates: an electrophysiological study in the isolated spinal cord of the rat.

The effects of various pairs of convulsant and nonconvulsant barbiturates on mono- and polysynaptic activity were studied in the isolated spinal cord of the immature rat, using extracellular recording. The convulsant barbiturates, 5-ethyl-5-(3-methylbut-2'-enyl) barbituric acid (3M2B), 5-ethyl-5-(1,3-dimethylbut-1'-enyl) barbituric acid (1,3M1B) and (+)-5-(1,3-dimethylbutyl)-5-ethyl barbituric acid [(+) DMBB] all increased the monosynaptic reflex at concentrations between 5 and 50 microM with no change in polysynaptic activity. When the concentration was raised to between 100 and 300 microM, however, the convulsants all reduced the monosynaptic reflex, thus producing a biphasic dose-response relationship. The nonconvulsant barbiturates phenobarbital, 5-ethyl-5-(3-methylbut-1'-enyl) barbituric acid (3M1B), amylobarbital (3MB) and (-)-5-(1,3-dimethylbutyl)-5-ethyl barbituric acid [(-)DMBB] produced only a decrease in mono- and polysynaptic reflexes. At concentrations which enhanced the monosynaptic reflex, the responses of motoneurones to glycine and eledoisin-related peptide (an analogue of substance P) were reduced by (+)DMBB, while 1,3M1B and 3M2B had no significant effects upon any of the neurotransmitters tested. At concentrations which depressed the monosynaptic reflex, the convulsants all reduced the response to glycine whereas the nonconvulsant barbiturates all increased the response to GABA. With the exception of phenobarbital, both convulsant and nonconvulsant barbiturates produced a direct depolarisation of the presynaptic terminal membrane, with only the convulsants producing a depolarisation of the membrane of the motoneurone. Using another convulsant barbiturate, 5-(2-cyclohexylideneethyl)-5-ethyl barbituric acid (CHEB), this direct depolarising action was found to be calcium-dependent.

Strychnine-like action of the convulsant barbiturate, CHEB.

The effect of 5-(2-cyclohexylideneethyl)-5-ethyl barbituric acid (CHEB) on the isolated spinal cord of the immature rat was examined using extracellular recording. At concentrations less than 20 microM CHEB increased the monosynaptic reflex (MSR) but depressed the reflex at greater concentrations (30-100 microM). At concentrations which enhanced the monosynaptic reflex, CHEB reduced the responses of motoneurones to glycine and to a lesser extent to those of L-glutamate. In the presence of strychnine (5 microM), which enhanced both mono- and polysynaptic reflexes, CHEB produced only slight enhancement of the monosynaptic reflex. At concentrations of 30-100 microM the responses to gamma-aminobutyric acid (GABA), glycine, L-glutamate and eledoisin-related peptide (ERP a substance P and analogue) were all reduced. At these concentrations CHEB directly depolarised the motoneurone membrane. Increases in [Mg2+]0, which reduced spontaneous activity, blocked the enhancement, by CHEB, of the monosynaptic reflex. The actions of CHEB in small doses may be due therefore to its ability to block the action of glycine and thus block tonic inhibition.

Effects of a depressant/convulsant pair of glutarimides on neuronal activity in the isolated spinal cord of the immature rat.

The effect of depressant/convulsant pair of glutarimides on the isolated spinal cord of the immature rat was examined using extracellular recording. At concentrations of 300 microM the depressant beta-butyl-beta-methyl glutarimide enhanced the response of motoneurones and dorsal root fibres to gamma-aminobutyric acid (GABA) while the convulsant bemegride (beta-ethyl-beta-methyl glutarimide) decreased both responses to GABA. At this concentration both the convulsant and depressant reduced mono- and polysynaptic reflex activity. Neither the convulsant or depressant had prominent direct actions, with only a small hyperpolarization being produced by both glutarimides on dorsal root fibres. The overall depressant or convulsant properties of these glutarimides may be due in part therefore to a differential effect on the postsynaptic action of the inhibitory transmitter GABA. Furthermore, the depressant glutarimide reduced the excitatory effects of L-glutamate and the convulsant reduced the inhibitory effects of glycine on spinal neurones; thus, actions on these transmitters may also contribute to the overall effects of these glutarimides.

ATP-sensitive potassium channel blockade enhances spontaneous alternation performance in the rat: a potential mechanism for glucose-mediated memory enhancement.

Peripheral and central injections of D-glucose enhance learning and memory in rats, and block memory impairments produced by morphine. The mechanism(s) for these effects is (are) as yet unknown. One mechanism by which glucose might act on memory and other brain functions is by regulating the ATP-sensitive potassium channel. This channel may couple glucose metabolism and neuronal excitability, with channel blockade increasing the likelihood of stimulus-evoked neurotransmitter release. The present experiments explored the effects of intra-septal injections of glucose and the ATP-sensitive potassium channel blocker glibenclamide on spontaneous alternation behavior in the rat. Intra-septal injections of glucose (20 nmol) or glibenclamide (10 nmol), 30 min prior to plus-maze spontaneous alternation performance, significantly enhanced alternation scores compared to rats receiving vehicle injections. Glibenclamide enhanced spontaneous alternation performance in an inverted-U dose-response manner. Individually sub-effective doses of glucose (5 nmol) and glibenclamide (5 nmol) significantly enhanced plus-maze alternation scores when co-injected into the septal area. Glibenclamide (10 nmol), when co-administered with morphine (4 nmol) 30 min prior to Y-maze spontaneous alternation performance, attenuated the performance-impairing effects of morphine alone. The present findings show that intra-septal injections of the direct ATP-sensitive potassium channel blocker glibenclamide, both alone and in conjunction with a sub-effective dose of glucose, enhance spontaneous alternation performance and attenuate the performance-impairing effects of morphine. The similarity of the results obtained with glibenclamide and glucose, together with their similar actions on ATP-sensitive potassium channel function, suggests that glucose may modulate memory-dependent behavior in the rat by regulating the ATP-sensitive potassium channel.

Neurotoxic activity of venom from the Australian eastern mouse spider (Missulena bradleyi) involves modulation of sodium channel gating.

Mouse spiders represent a potential cause of serious envenomation in humans. This study examined the activity of Missulena bradleyi venom in several in vitro preparations. Whilst female M. bradleyi venom at doses up to 0.05 microl ml(-1) failed to alter twitch or resting tension in all preparations used, male venom (0.02 and 0.05 microl ml(-1)) produced potent effects on transmitter release in both smooth and skeletal neuromuscular preparations. In the mouse phrenic nerve diaphragm preparation, male M. bradleyi venom (0.02 microl ml(-1)) caused rapid fasciculations and an increase in indirectly evoked twitches. Male venom (0.02 and 0.05 microl ml(-1)) also caused a large contracture and rapid decrease in indirectly evoked twitches in the chick biventer cervicis muscle, however had no effect on responses to exogenous ACh (1 mM) or potassium chloride (40 mM). In the chick preparation, contractile responses to male M. bradleyi venom (0.05 microl ml(-1)) were attenuated by (+)-tubocurarine (100 microM) and by tetrodotoxin (TTX, 1 microM). Both actions of male M. bradleyi venom were blocked by Atrax robustus antivenom (2 units ml(-1)). In the unstimulated rat vas deferens, male venom (0.05 microl ml(-1)) caused contractions which were inhibited by a combination of prazosin (0.3 microM) and P(2X)-receptor desensitization (with alpha,beta-methylene ATP 10 microM). In the rat stimulated vas deferens, male venom (0.05 microl ml(-1)) augmented indirectly evoked twitches. Male venom (0.1 microl ml(-1)) causes a slowing of inactivation of TTX-sensitive sodium currents in acutely dissociated rat dorsal root ganglion neurons. These results suggest that venom from male M. bradleyi contains a potent neurotoxin which facilitates neurotransmitter release by modifying TTX-sensitive sodium channel gating. This action is similar to that of the delta-atracotoxins from Australian funnel-web spiders.

Depolarizing actions of convulsant barbiturates on isolated rat dorsal root ganglion cells.

The actions of convulsant barbiturates were studied on dorsal root ganglion (DRG) cells in vitro using intracellular recording techniques. Only the convulsant barbiturates (+)-DMBB and CHEB produced a concentration-dependent depression in the responses to gamma-aminobutyric acid (GABA). All convulsant barbiturates were found to produce a direct depolarization of the DRG cell membrane which was accompanied by a decrease in the input resistance of the cell and a reduction in the orthodromic action potential. A sub-population of DRG cells were found to be refractory to these actions but there was no relationship between the cell type (A beta, A delta and C) and ability to respond to the convulsant barbiturates.

Presynaptic snake beta-neurotoxins produce tetanic fade and endplate potential run-down during neuromuscular blockade in mouse diaphragm.

The present study investigated the ability of a number of presynaptic snake neurotoxins (snake beta-neurotoxins) to produce nerve-evoked train-of-four fade, tetanic fade and endplate potential run-down during the development of neuromuscular blockade in the isolated mouse phrenic-hemidiaphragm nerve-muscle preparation. All the snake beta-neurotoxins tested, with the exception of notexin, produced train-of-four and tetanic fade of nerve-evoked isometric muscle contractions. Train-of-four fade was not present during the initial depressant or facilitatory phases of muscle tension produced by the snake beta-neurotoxins but developed progressively during the final depressant phase that precedes complete neuromuscular blockade. The 'non-neurotoxic' bovine pancreatic phospholipase A2 and the 'low-toxicity' phospholipase A2 from Naja naja atra venom failed to elicit train-of-four fade, indicating that the phospholipase activity of the snake beta-neurotoxins is not responsible for the development of fade. Intracellular recording of endplate potentials (EPPs) elicited by nerve-evoked trains of stimuli showed a progressive run-down in EPP amplitude during the train following incubation with all snake beta-neurotoxins except notexin. Again this run-down in EPP amplitude was confined to the final depressant phase of snake beta-neurotoxin action. However when EPP amplitude fell to near uniquantal levels (< 3 mV) the extent of toxin induced-fade was reduced. Unlike postjunctional snake alpha-neurotoxins, prejunctional snake beta-neurotoxins interfere with acetylcholine release at the neuromuscular junction during the development of neuromuscular blockade. This study provides further support for the hypothesis that fade in twitch and tetanic muscle tension is due to an underlying rundown in EPP amplitude resulting from a prejunctional alteration of transmitter release rather than a use-dependent block of postjunctional nicotinic receptors.

Induction of giant miniature end-plate potentials during blockade of neuromuscular transmission by textilotoxin.

The present study investigated the action of textilotoxin, isolated from the venom of the Australian common brown snake Pseudonaja textilis, on neuromuscular transmission in isolated toad nerve-muscle preparations. Initial muscle twitch tension experiments revealed a triphasic pattern of changes in muscle tension and a irreversible binding action of textilotoxin (10 micrograms/ml) similar to other snake beta-neurotoxins. This was characterised by an initial depression of twitch tension, followed by a period of enhanced tension, eventually leading to a reduction in tension to complete neuromuscular blockade. These actions on muscle tension were investigated further by assessing the action of textilotoxin on end-plate potential amplitude (EPP). This revealed a similar triphasic alteration of the nerve-evoked release of acetylcholine from the motor nerve terminal. These actions on acetylcholine release were confirmed to be of a presynaptic origin since the modal amplitude of miniature end-plate potentials (MEPPs) was not reduced and in twitch tension experiments the muscle still contracted in response to direct muscle stimulation when nerve-evoked release was completely blocked. Interestingly dramatic effects were observed on the spontaneous release of acetylcholine, including an marked increase in MEPP frequency, a skewing of the MEPP amplitude frequency histogram to the right, and a resultant increase in the number of 'giant' MEPPs. These results indicate that textilotoxin causes a presynaptic blockade of neuromuscular transmission involving a disruption of the regulatory mechanism that controls acetylcholine release.

Selective alteration of sodium channel gating by Australian funnel-web spider toxins.

The actions of potent mammalian neurotoxins isolated from the venom of two Australian funnel-web spiders were investigated using both electrophysiological and neurochemical techniques. Whole-cell patch clamp recording of sodium currents in rat dorsal root ganglion neurons revealed that versutoxin (VTX), isolated from the venom of Hadronyche versuta, produced a concentration-dependent slowing or removal of tetrodotoxin-sensitive (TTX-S) sodium current inactivation and a reduction in peak TTX-S sodium current. In contrast, VTX had no effect on tetrodotoxin-resistant (TTX-R) sodium currents or potassium currents. VTX also shifted the voltage dependence of sodium channel activation in the hyperpolarizing direction and increased the rate of recovery from inactivation. Ion flux studies performed in rat brain synaptosomes also revealed that robustoxin (RTX), from the venom of Atrax robustus, and VTX both produced a partial activation of 22Na+ flux and an inhibition of batrachotoxin-activated 22Na+ flux. This inhibition of flux through batrachotoxin-activated channels was not due to an interaction with neurotoxin receptor site 1 since [3H]saxitoxin binding was unaffected. In addition, the partial activation of 22Na+ flux was not enhanced in the presence of alpha-scorpion toxin and further experiments suggest that VTX also enhances [3H]batrachotoxin binding. These selective actions of funnel-web spider toxins on sodium channel function are comparable to those of alpha-scorpion and sea anemone toxins which bind to neurotoxin receptor site 3 on the channel to slow channel inactivation profoundly. Also, these modifications of sodium channel gating and kinetics are consistent with actions of the spider toxins to produce repetitive firing of action potentials.

Frequency-dependent neuromuscular blockade by textilotoxin in vivo.

The effect of stimulation frequency on the timecourse of neuromuscular blockade, following the administration of textilotoxin (20 micrograms/kg) or beta-bungarotoxin (50 micrograms/kg), was examined in the interdigital muscles of the hindlimb in anaesthetized mice. While the time of death was variable, neuromuscular blockade of the interdigital muscles occurred at the same time as respiratory failure with both textilotoxin and beta-bungarotoxin only at stimulation rates of 0.5 Hz and above. Textilotoxin (50 micrograms/kg) produced an increase in the heart rate prior to death but no change in the shape of the electrocardiogram.

Cross-reactivity of Sydney funnel-web spider antivenom: neutralization of the in vitro toxicity of other Australian funnel-web (Atrax and Hadronyche) spider venoms.

Australian funnel-web spiders are recognized as one of the most venomous spiders to humans world-wide. Funnel-web spider antivenom (FWS AV) reverses clinical effects of envenomation from the bite of Atrax robustus and a small number of related Hadronyche species. This study assessed the in vitro efficacy of FWS AV in neutralization of the effects of funnel-web spider venoms, collected from various locations along the eastern seaboard of Australia, in an isolated chick biventer cervicis nerve-muscle preparation. Venoms were separated by SDS-PAGE electrophoresis to compare protein composition and transblotted for Western blotting and incubation with FWS AV.SDS-PAGE of venoms revealed similar low and high molecular weight protein bands. Western blotting with FWS AV showed similar antivenom binding with protein bands in all the venoms tested. Male funnel-web spider venoms (7/7) and female venoms (5/10) produced muscle contracture and fasciculation when applied to the nerve-muscle preparation. Venom effects were reversed by subsequent application of FWS AV or prevented by pretreatment of the preparation with antivenom.FWS AV appears to reverse the in vitro toxicity of a number of funnel-web spider venoms from the eastern seaboard of Australia. FWS AV should be effective in the treatment of envenomation from most, if not all, species of Australian funnel-web spiders.

Isolation of a funnel-web spider polypeptide with homology to mamba intestinal toxin 1 and the embryonic head inducer Dickkopf-1.

We have isolated and determined the amino acid sequence of a novel peptide component from the venom of the Australian funnel-web spider Hadronyche versuta. This 68-residue toxin, ACTX-Hvf17, does not function like classical neurotoxins in modulating ion channel function as evidenced by its lack of insecticidal activity and its inability to affect vertebrate smooth or skeletal muscle contractility. The peptide shows significant sequence homology with mamba intestinal toxin 1 (MIT1) and to a lesser extent with a variety of colipases. The strong structural homology between MIT1 and porcine colipase leads us to propose that ACTX-Hvf17 also adopts the MIT1/colipase three-dimensional fold. However, we show that ACTX-Hvf17 has no colipase activity and does not stimulate muscle contractility like MITI. We also show that MIT1 and ACTX-Hvf17 display significant sequence homology with the C-terminal cysteine-rich domain of the Dickkopf-1 family of proteins that induce head formation in developing embryos, which leads us to propose that this domain of Dickkopf-1 also adopts the MIT1 colipase fold.

Red-back spider (Latrodectus hasselti) antivenom prevents the toxicity of widow spider venoms.

STUDY OBJECTIVES: Widow spiders of the genus Latrodectus are found worldwide and produce similar clinical envenomation syndromes. In Australia, red-back spider antivenom (RBS-AV) is effective therapy for Latrodectus hasselti envenomation and it has been reported to reverse envenomation by other widow spiders. This study assessed the efficacy of RBS-AV in preventing in vitro and in vivo toxicity of widow spider venoms of North America and Europe. METHODS: The binding of RBS-AV to alpha-latrotoxin and Latrodectus venoms (Latrodectus spp mactans, hesperus, lugubris, tredecimguttatus, hasselti) was assayed using Western blotting. Prevention of in vitro toxicity to alpha-latrotoxin and the same venoms was tested by pretreating an isolated chick biventer cervicis nerve-muscle preparation with RBS-AV. Prevention of in vivo toxicity was determined by a lethality study in male Balb/c mice (2.5 to 5x median lethal dose [LD50]) or alpha-latrotoxin (10x LD50) preincubated with antivenom or without RBS-AV (control). RESULTS: In Western blots, RBS-AV bound to alpha-latrotoxin and similar widow spider proteins in all venoms tested, indicating antigenic similarity with proteins found in RBS venom. Antivenom prevented the typical in vitro muscle contracture and loss of twitch tension seen with alpha-latrotoxin and the venoms tested. Control mice rapidly developed signs of envenomation, but mice treated with RBS-AV remained free of signs of envenomation. CONCLUSION: RBS-AV prevented both in vitro and in vivo toxicity from Latrodectus venoms and alpha-latrotoxin in mice. These data suggest that RBS-AV may be clinically effective in the treatment of envenomation resulting from the bite of other widow spiders.

Calcium-dependent actions of the convulsant barbiturate, CHEB, on transmitter release at the rat neuromuscular junction.

1. The effect of convulsant barbiturates on spontaneous and evoked acetylcholine release was studied at the rat neuromuscular junction in vitro. 2. The convulsant barbiturates (+)-5-(1,3-dimethylbutyl)-5-ethyl barbituric acid [(+)-DMBB], 5-(2-cyclohexylideneethyl)-5-ethyl barbituric acid (CHEB), 5-ethyl-5-(3-methylbut-2'-enyl) barbituric acid (3M2B) and 5-ethyl-5-(1,3-dimethylbut-1'-enyl) barbituric acid (1,3M1B) all produced a concentration-dependent increase in miniature end-plate potential (MEPP) frequency. 3. With CHEB (100 microM) this increase in MEPP frequency was found to be dependent on the [Ca2+]o. CHEB in 0.5 mM [Ca2+]o did not alter MEPP amplitude, but in 1.3 and 2.5 mM [Ca2+]o CHEB significantly reduced the amplitude. 4. At a [Ca2+]o of 0.5 mM, CHEB produced an increase in both EPP amplitude and quantal content, while at 1.3 mM [Ca2+]o CHEB did not alter EPP amplitude or quantal content. 5. The plot of log quantal content vs log [Ca2+]o showed a parallel shift to the left in the presence of 100 microM CHEB. This change occurred without any alteration in the maximum quantal content. This suggests that the enhancement of transmitter release may be mediated by an effect on calcium fluxes in the pre-junctional nerve terminal.

Differential actions of pacific ciguatoxin-1 on sodium channel subtypes in mammalian sensory neurons.

Pacific ciguatoxin-1 (P-CTX-1), is a highly lipophilic cyclic polyether molecule originating from the marine dinoflagellate Gambierdiscus toxicus. Its effects were investigated on sodium channel subtypes present in acutely dissociated rat dorsal root ganglion neurons, using whole-cell patch clamp techniques. Concentrations of P-CTX-1 ranging from 0.2 to 20 nM had no effect on the kinetics of tetrodotoxin-sensitive (TTX-S) or tetrodotoxin-resistant (TTX-R) sodium channel activation and inactivation, however, a concentration-dependent reduction in peak current amplitude occurred in both channel types. The main actions of 5 nM P-CTX-1 on TTX-S sodium channels were a 13-mV hyperpolarizing shift in the voltage dependence of sodium channel activation and a 22-mV hyperpolarizing shift in steady-state inactivation (hinfinity). In addition, P-CTX-1 caused a rapid rise in the membrane leakage current in cells expressing TTX-S sodium channels. This effect was blocked by 200 nM TTX, indicating an action mediated through TTX-S sodium channels. In contrast, the main action of P-CTX-1 (5 nM) on TTX-R sodium channels was a significant increase in the rate of recovery from sodium channel inactivation. These results indicate that P-CTX-1 acts to modify voltage-gated sodium channels present in peripheral sensory neurons consistent with its action to increase nerve excitability. This provides an explanation for the sensory neurological disturbances associated with ciguatera fish poisoning.

Modification of sodium channel gating and kinetics by versutoxin from the Australian funnel-web spider Hadronyche versuta.

The effects of a neurotoxin (versutoxin VTX), purified from the venom of the Australian Blue Mountains funnel-web spider Hadronyche versuta, on the ionic currents in rat dorsal root ganglion cells were investigated under voltage-clamp conditions using the whole-cell patch-clamp technique. VTX had no effect on tetrodotoxin-resistant (TTX-R) sodium currents or potassium currents. In contrast VTX produced a dose-dependent slowing or removal of tetrodotoxin-sensitive (TTX-S) sodium current inactivation, a reduction in peak TTX-S sodium current but did not markedly slow tail current kinetics of TTX-S sodium currents. This steady-state sodium current was maintained during prolonged depolarizations at all test potentials and the reduction in sodium current amplitude produced by VTX had an apparent Ki of 37 nM. In the presence of 32 nM VTX the voltage dependence of steady-state sodium channel inactivation (h infinity) also showed a significant 7 mV shift in the voltage midpoint in the hyperpolarizing direction, with no change in the slope factor. In addition there was a steady-state or non-inactivating component present (14 +/- 2% of maximal sodium current) at prepulse potentials more depolarized than -40 mV, potentials which normally inactivate all TTX-S sodium channels. Finally, there was an observed increase in the rate of recovery from inactivation in the presence of VTX. These selective actions of VTX on sodium channel gating and kinetics are similar to those of alpha-scorpion and sea anemone toxins.

Characterisation of the effects of robustoxin, the lethal neurotoxin from the Sydney funnel-web spider Atrax robustus, on sodium channel activation and inactivation.

The present study investigates the actions of robustoxin (atracotoxin-Ar1) purified from the venom of the male Sydney funnel-web spider Atrax robustus on sodium channel gating. Using whole-cell patch-clamp techniques the study assessed the actions of robustoxin on tetrodotoxin-resistant (TTX-R) and tetrodotoxin-sensitive (TTX-S) sodium currents in rat dorsal root ganglion cells. Similar to the closely related funnel-web spider toxin versutoxin (delta-atracotoxin-Hv1) from Hadronyche versuta, robustoxin had no effect on TTX-R sodium currents but exerted potent effects on TTX-S sodium currents. The main action of robustoxin was a concentration-dependent slowing or removal of TTX-S sodium current inactivation. This steady-state current was maintained during long-lasting depolarisations at all test potentials. Robustoxin (30 nM) also caused a 13-mV hyperpolarising shift in the voltage midpoint of steady-state sodium channel inactivation (h infinity) leading to a reduced peak current at a holding potential of -80 mV. Moreover there was a steady-state or non-inactivating component present (18% of maximal sodium current) at prepulse potentials that normally inactivate all TTX-S sodium channels (more depolarised than -40 mV). In addition robustoxin produced a significant increase in the repriming kinetics of the sodium channel when channels returned to the resting state following activation. This increase in the rate of recovery of sodium current appears to explain the use-dependent effects on peak sodium current amplitude at high stimulation frequencies. Finally 30 nM robustoxin caused an 11-mV hyperpolarising shift in the voltage dependence of the channel but did not markedly modify tail current kinetics. These actions suggest that robustoxin inhibits conversion of the open state to the inactivated state of TTX-S sodium channels, thus allowing a fraction of the sodium current to remain at membrane potentials at which inactivation is normally complete. Given the recent reclassification of funnel-web spider toxins as atracotoxins, robustoxin should henceforth be known as delta-atracotoxin-Ar1 to reflect this main action on sodium channel inactivation. These present results further support the hypothesis that funnel-web spider toxins interact with neurotoxin receptor site 3 to slow channel inactivation in a manner similar to that of alpha-scorpion and sea anemone toxins.