Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Ciguatera is a form of ichthyosarcotoxism caused by ingestion of particular reef fish species from tropical and subtropical waters. Ciguatera affects around 25,000 to 50,000 people worldwide each year (Higerd, 1983) causing significant morbidity, particularly in the atoll island countries of the Pacific basin. The principle toxin involved, ciguatoxin (CTX), is a heat-stable, lipophilic cyclic polyether derived from gambiertoxins, that are found in strains of the benthic dinoflagellate, Gambierdiscus toxicus. Gambiertoxins are thought to be precursors that are oxidized to CTX by fish liver cytochrome enzymes and are bioaccumulated in the fish through their natural predatory food chain (Lewis and Holmes, 1993; Murata et al., 1990). Presently fourteen strains of CTX have been identified from the Pacific region (Lewis and Jones, 1997). Of these, Pacific CTX-1 (P-CTX-1) is the most abundant and the most potent strain, with an LD₅₀ in mice of 0.25 µg/kg i.p. (Lewis et al., 1991; Murata et al., 1990).

The symptoms of ciguatera poisoning typically involve short-term gastrointestinal disturbances followed by longer-term sensory disturbances affecting the peripheral nervous system, such as paraesthesias and dysesthesias. The neurological symptoms are believed to be the consequence of the direct interaction of CTX with voltage-gated sodium channels present on the membranes of excitable cells. Previous radioligand binding experiments indicate that CTX competes with another cyclic polyether, brevetoxin (PbTX), from the marine dinoflagellate Ptychodiscus brevis, for neurotoxin receptor site 5 on the voltage-gated sodium channel (Lombet et al., 1987). This orphan receptor is believed to be located near the S5-S6 extracellular loop of domain IV of the sodium channel  subunit (Trainer et al., 1991, 1994). In amphibian nerve preparations, CTX acts to cause spontaneous repetitive firing of action potentials in addition to hyperpolarizing shifts in the voltage dependence of sodium channel activation (Benoit et al., 1986). However, little is known about how CTX interacts with mammalian peripheral sensory neurons to produce the neurological symptoms associated with ciguatera fish poisoning.

The present study was developed to differentiate the effects of P-CTX-1 on the gating and kinetics of mammalian sodium channel subtypes using whole-cell patch clamp recording techniques. Dorsal root ganglion (DRG) neurons were chosen to examine the actions of P-CTX-1, as their cell bodies and/or afferent fibers are presumably the origin of the characteristic sensory neurological disturbances. In addition these neurons express two sodium channel subtypes: tetrodotoxin-sensitive (TTX-S) sodium channels, which are readily blocked by tetrodotoxin (K_i = 0.3 nM), and TTX-resistant (TTX-R) sodium channels, which are largely resistant to the action of tetrodotoxin (K_i = 100 µM) (Roy and Narahashi, 1992). These TTX-R sodium channels, which have also been found in nodose (Ikeda and Schofield, 1987) and superior cervical ganglia (Schofield and Ikeda, 1988), appear to be confined to the peripheral nervous system and are believed to be important for sensory integration and thus a potential site of action for CTX. We report that P-CTX-1 produces differential modulation on the voltage dependence of activation, the rate of recovery from inactivation and on steady-state channel inactivation of TTX-S and TTX-R voltage-gated sodium channels.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

DRG Isolation and Preparation. All experiments were performed on acutely dissociated rat DRG neurons, which were prepared by a modification of the methods described by Nicholson et al. (1994). Briefly, a 2- to 12-day-old Wistar rat of either sex was anaesthetized with isoflurane, and DRGs were isolated from the vertebral column. DRG cells were then incubated in Ca⁺⁺- and Mg⁺⁺-free phosphate-buffered saline containing 2.5 mg/ml trypsin (Type XI; Sigma, St. Louis, MO), in a water bath at 37°C for 18 to 40 min, depending on the age of the animal. Following enzyme treatment, the ganglia were washed twice with sterile Dulbecco's modified Eagle's medium (DMEM) (Gibco, Grand Island, NY) containing 10% (v/v) newborn calf serum (Gibco) and 80 µg/ml gentamycin (Sigma). Neurons were then mechanically dissociated by trituration through a sterile Pasteur pipette. The cells were subsequently distributed into 8 wells of a 24-well tissue culture plate that contained 12-mm round glass coverslips (Assistent, Germany) previously coated with poly-L-lysine. Cells were incubated overnight in 1 ml of DMEM at 37°C (10% CO₂, 90% O₂, and 100% relative humidity) to allow the isolated neurons to settle and adhere to the coverslips.

Electrophysiological Recordings. A single coverslip with attached DRG cells was transferred to a 1-ml perfusion chamber (RC-13; Warner Instruments, Hamden, CT) mounted on the stage of an inverted phase-contrast microscope (Lietz Labovert FS, Sydney, Australia). The neurons were bathed in an iso-osmotic external solution that contained (in mM); 30 NaCl, 5 CsCl, 1.8 CaCl₂, 1 MgCl₂, 25 D-glucose, 5 N-2-[hydroxyethyl]piperazine-N'-[2-ethanesulfonic acid] (HEPES-acid), 70 tetramethylammonium chloride (TMA-Cl), and 20 tetraethylammonium chloride (TEA-Cl). The pH of the external solution was adjusted to 7.4 with 1 M TEA-hydroxide, and the osmolarity was monitored by a vapor pressure osmometer (Gonotec, Berlin, Germany) and adjusted to 290 to 300 mmol/kg/l using sucrose. A gravity-fed perfusion system with a Gilmont flowmeter (Barrington, IL) maintained external solution flow at a rate between 0.3 to 0.6 ml/min, whereas a Peltier device maintained the bath at ambient room temperature (18-23°C). During experiments the temperature did not vary more than 1°C.

Data Analysis. Numerical data are presented as the mean ± S.E. (n, number of observations) and statistical differences were determined using a Student's t test, at P < 0.05. Mathematical curve fitting was accomplished by a Apple Macintosh Quadra-700 computer using SigmaPlot for the Macintosh. All curve-fitting routines were performed using a nonlinear least-squares method and splining routines.

(1)

(2)

h

(3)

(4)

Pacific CTX-1 Purification. P-CTX-1 was isolated from the viscera of moray eels (Lycodontis javanicus) that were collected from a region of Tarawa (1.3°N, 173°E) in the Republic of Kiribati (central Pacific Ocean) where ciguatera is endemic. The isolation and purification techniques required to extract CTX have been described fully by Lewis et al. (1991). Briefly, the purification technique involved heating the viscera to 70°C and extracting the lipid-soluble components using acetone. This product was then subjected to silica gel vacuum liquid chromatography followed by five chromatographic steps. Samples of isolated CTXs were reapplied to HPLC columns and eluted with different polarity solvents to confirm homogeneity, and an array detector was used to determine the UV profile and establish purity. P-CTX-1 stock was dissolved in 50% aqueous methanol and stored at -20°C in a glass vial, since CTX binds strongly to plastic. P-CTX-1 was superfused through the bath at various concentrations by dilution with external solution. Control experiments were performed with 50% aqueous methanol at a maximum concentration of 2.2 mg/ml to assess effects of the vehicle on sodium channel function.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Effects on Sodium Channel Activation and Inactivation Kinetics. To assess the actions of P-CTX-1 on the kinetics of sodium channel activation and inactivation, initial experiments measured the amplitude and timecourse of sodium currents after perfusion with P-CTX-1. A voltage-step protocol that applied depolarizing test pulses from a holding potential of 80 mV to 10 mV for 50 ms was employed to generate sodium currents (Fig. 1 A and B). The percentage change in peak amplitude was calculated for sodium currents from both TTX-S and TTX-R sodium currents, before and 10 min after perfusion with P-CTX-1 at concentrations of 0.2, 2, 5, 10 and 20 nM. P-CTX-1 produced a potent, concentration-dependent block of both TTX-S and TTX-R sodium currents (see Fig. 1). In the presence of 5 nM P-CTX-1 peak TTX-S sodium current amplitude was reduced by 38.7 ± 4.6% (n = 47, P < 0.001), but there was a 39.5 ± 5.5% (n = 34, P < 0.001) decrease in TTX-R sodium currents amplitude. A concentration of 5 nM P-CTX-1 was chosen for the remaining experiments in this study, first because it is close to the EC₅₀ value and second because our supplies of P-CTX-1 were limited.

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Dose-response effects of P-CTX-1 on TTX-S and TTX-R sodium currents. TTX-S (A) and TTX-R (B) sodium currents were elicited using a voltage-step shown above (A), where 50 ms depolarizing test pulses to -10 mV were applied from a holding potential of -80 mV every 10 s. Control traces (a) were recorded before a 10-min perfusion with 5 nM P-CTX-1 (b). TTX-R sodium currents recorded in the presence of 200 nM TTX to eliminate residual TTX-S sodium currents. A dose-response histogram (C) for the reduction in peak sodium current was constructed for TTX-S () and TTX-R () sodium currents. Currents were measured before and after a 10-min perfusion in 0.2, 2, 5, 10, and 20 nM P-CTX-1, and the percentage change in peak sodium current was calculated. Note n  3 for all concentrations tested.

Effects on the Leakage Current. An interesting finding during our study was that P-CTX-1 induced a large rise in leakage current. Only DRG cells eliciting predominantly TTX-S sodium currents exhibited this significant increase in leakage current. This can be observed in Fig. 2A, where a TTX-S sodium current was recorded without leakage current subtraction in the absence (left trace) and presence (right trace) of 5 nM P-CTX-1. There was a development of a marked leakage current after perfusion with 5 nM P-CTX-1 of approximately -1 nA. After a 10-min perfusion in 5 nM P-CTX-1, the mean rise in leakage current in these cells was -0.63 ± 0.12 nA (n = 69), as seen in Fig. 2B. The rise in the leakage current commenced immediately on perfusion with P-CTX-1 and was concentration dependent. This is shown in Fig. 2B, where the leakage currents recorded after the addition of 5 nM P-CTX-1 were plotted for both TTX-S (n = 69) and TTX-R (n = 19) sodium channels. In order to determine if this leakage current was mediated through sodium channels, 200 nM TTX was added to the external solution. Figure 2C shows a typical experiment from a cell expressing a TTX-S sodium current, in which the leakage current increased from -0.074 nA, under control conditions, to -0.204 nA following a 10-min perfusion with 5 nM P-CTX-1. A subsequent 8-min application of 200 nM TTX resulted in a rapid decline in the leakage current, which later increased during washout with TTX-free external solution. The reversal of this leakage current by 200 nM TTX was typical in eight additional experiments.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Increases in the leakage current produced by 5 nM P-CTX-1. A, typical current traces recorded in the absence (left) and presence (right) of 5 nM P-CTX-1. The dotted line represents zero current. Currents were recorded without leakage current subtraction. Note the development of a marked leakage current in the right hand trace. B, the timecourse of changes in the leakage current following the addition of 5 nM P-CTX-1 where data represents the mean ± S.E. of 69 cells expressing TTX-S sodium currents () and 19 cells expressing TTX-R sodium currents (). C, a typical experiment illustrating reversal by 200 nM TTX of this increase in leakage current in a cell expressing TTX-S sodium currents. Washout was performed in toxin-free external solution.

Shifts in the Voltage Dependence of Activation. Previously it had been reported that partially purified CTX caused a significant hyperpolarizing shift in the voltage dependence of sodium channel activation in amphibian sciatic nerve (Benoit et al., 1986). Accordingly, we examined the actions of P-CTX-1 on TTX-S and TTX-R sodium channel activation and found that P-CTX-1 also caused hyperpolarizing shifts in DRG neurons. A typical experiment illustrating this shift is shown in Fig. 3, A-C. Under control condition, TTX-S sodium currents activated at -50 mV reached a maximum inward current at -20 mV and reversed at approximately +35 mV. However, following a 10-min perfusion with 5 nM P-CTX-1, the threshold of activation was reduced to -65 mV, reached a maximum at -30 mV, and reversed at +30 mV, indicating a negative shift in the threshold of channel activation of approximately -15 mV. This experiment was typical of the results obtained in seven additional experiments. There was, however, a much smaller shift in the voltage dependence of activation of TTX-R sodium currents exposed to 5 nM P-CTX-1 (see Fig. 3, D-F). In a typical experiment representative of nine other experiments, the threshold of channel activation was shifted only around 5 mV in the hyperpolarizing direction by P-CTX-1 from -40 to -45 mV. Despite these shifts in the threshold of channel activation, there were no significant changes in the reversal potentials for either TTX-S and TTX-R sodium current subtypes in the presence of 5 nM P-CTX-1.

View larger version (28K): [in this window] [in a new window]   Fig. 3.   Typical effects of 5 nM P-CTX-1 on the voltage dependence of activation of TTX-S and TTX-R sodium currents. Test pulse protocols (A) used 50-ms depolarizing test pulses from a holding potential of -80 mV to +70 mV in 5-mV increments before (A and D) and 10 min after (B and E) perfusion with 5 nM P-CTX-1. For clarity, only sodium currents recorded in 20-mV steps are presented. The peak sodium current at each voltage step was measured and plotted as a function of membrane potential (C and F) in the absence () and presence of 5 nM P-CTX-1 ().

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Typical effects of P-CTX-1 on the voltage dependence of sodium conductance (gNa). The peak current amplitudes obtained from I/V experiments in Fig. 2 were converted to gNa/V relationships using eq. 1 detailed in Materials and Methods. The gNa values were subsequently normalized against maximal sodium conductance (gmax) and plotted against membrane potential for TTX-S (A) and TTX-R (B) sodium currents before () and 10 min after perfusion with 5 nM P-CTX-1 (). Curves were fitted to a Boltzmann function as described in Materials and Methods (eq. 2).

Effects on Steady-State Sodium Channel Inactivation (h). Certain toxins and drugs, such as local anesthetics, preferentially bind to, and stabilize, the inactivated state of the sodium channel (Hille, 1977). To determine whether P-CTX-1 stabilizes the inactivated state a standard two-pulse protocol with 0.5-ms interpulse interval was applied to measure steady-state inactivation (h) of TTX-S and TTX-R sodium currents. This consisted of a 1-s conditioning prepulse, in which the holding potential of -80 mV was stepped to potentials ranging from -130 to 0 mV in 5-mV increments, followed by a 40-ms test pulse to -10 mV (Fig. 5A, inset). The peak currents recorded during the test pulse were normalized to the maximum amplitude and plotted against prepulse potential. Figure 5A shows the TTX-S h/V relationship following exposure to 5 nM P-CTX-1. It was found that P-CTX-1 caused a hyperpolarizing shift of 22 ± 4 mV (n = 8, P < 0.002) in the TTX-S sodium channel h curve, which occurred from -71 ± 5 mV to -93 ± 5 mV with no significant change in the slope factor (k). In contrast, 5 nM P-CTX-1 failed to produce a significant shift in the V_1/2 of TTX-R sodium channel steady-state inactivation (Fig. 5C). Moreover, P-CTX-1 did not produce inward sodium currents at prepulse potentials, which fully inactivate the channel (>-40 mV for TTX-S and >-10 mV for TTX-R sodium channels), as has been observed with other toxins interacting with sodium channels such as sea anemone, -scorpion, and funnel-web spider toxins (Nicholson et al., 1994, 1998; Strichartz and Wang, 1986; Wasserstrom et al., 1993). In Fig. 5, B and D, the currents elicited during perfusion with 5 nM P-CTX-1 were expressed as a fraction of the maximum control current. This shows that the hyperpolarizing prepulses as negative as -130 mV did not reverse the P-CTX-1-induced reductions in peak sodium current. It has been reported that steady-state inactivation V_1/2 values are shifted over time in whole-cell patch clamp measurements (Fernandez et al., 1984), however, such time-dependent shifts are relatively small (in our recordings ranging between 0.1 and 3 mV) over this period and would therefore not influence the results on TTX-S sodium currents observed with P-CTX-1.

View larger version (29K): [in this window] [in a new window]   Fig. 5.   Effects of P-CTX-1 on steady-state sodium channel inactivation (h). Steady-state inactivation was determined using the two-pulse protocol shown (A, inset). A, 1-s conditioning prepulse from -130 to 0 mV was stepped in 5-mV increments and followed by a 40-ms test pulse to -10 mV every 10 s. The amount of sodium current that is available for activation under control conditions () and during perfusion with 5 nM P-CTX-1 () are shown for TTX-S (A and B) and TTX-R (C and D) sodium currents. (A and C) All currents were normalized to the maximum control current, whereas (B and D) currents recorded in the presence of P-CTX-1 were expressed as a fraction of the control current. Curves were fitted to eq. 3 in Materials and Methods. Values represent the mean ± S.E. of eight experiments on TTX-S sodium currents and six experiments for TTX-R sodium currents.

Use-Dependent Effects on Sodium Channel Gating. High-frequency stimulation has been found to modify the action of certain sodium channel modulators such as carbamazepine and phenytoin (Willow et al., 1985), local anesthetics (Hille, 1977) and sea anemone toxins (Wasserstrom et al., 1993). These effects are termed use-dependent because they directly depend on the frequency of channel gating. To determine any possible use-dependent actions of 5 nM P-CTX-1, a protocol that applied a train of 20 depolarizing pulses from a holding potential of -80 to -10 mV for 25 ms at a frequency of 1, 10, or 30 Hz was used. P-CTX-1 at 5 nM concentrations failed to produce any significant use-dependent changes in TTX-S sodium current amplitude at all stimulation frequencies tested (data not shown). Although there appeared to be a small use-dependent block by P-CTX-1 on TTX-R sodium channels, this was statistically nonsignificant.

Effects on the Rate of Recovery from Inactivation. Since a partially purified CTX has been shown to promote repetitive firing in frog myelinated nerve fibers (Benoit et al., 1986), the possibility exists that an increase in the rate at which sodium channels recover from inactivation may contribute to the increase in neuronal excitability. To investigate this, a standard two-pulse protocol with a variable interpulse interval (T) was used (Fig. 6). A conditioning prepulse to -10 mV was used to inactivate sodium channels, after which a 40-ms depolarizing test pulse to -10 mV was applied. The prepulse duration was 100 ms for TTX-S sodium currents and 300 ms for TTX-R sodium currents to allow for complete sodium channel inactivation. The interpulse interval (T) between the conditioning and test pulses was varied between 0.5 ms and 1 s. Peak current recorded during the test pulse was normalized against the current amplitude during the conditioning prepulse and plotted as a function of the interpulse interval. Figure 6A demonstrates that P-CTX-1 failed to alter the rate of recovery of inactivation of TTX-S sodium currents, as complete recovery from inactivation for the currents recorded in the absence and presence of 5 nM P-CTX-1 occurred at approximately 200 ms. These results are similar to the combined results from six additional cells, in which no significant differences existed between the time constants of recovery or their respective coefficients (Table 1).

View larger version (23K): [in this window] [in a new window]   Fig. 6.   Typical experiments showing the effects of 5 nM P-CTX-1 on the rate of recovery from sodium channel inactivation in TTX-S and TTX-R sodium channels. The rate of inactivation was assessed using a standard two-pulse protocol as shown where T represents a variable interpulse interval. Peak currents elicited during the 40-ms test pulse were normalized to the peak current during a conditioning prepulse was 300 ms for TTX-R sodium currents and 100 ms for TTX-S sodium currents. Recovery from inactivation was plotted as a function of interpulse interval (A and C) under control conditions () and 10 min after perfusion with 5 nM P-CTX-1 (), and the magnitude of the unrecovered fraction of current (A and C) was plotted on a semilogarithmic scale (B and D). Data was fitted to the sum of two exponential functions, according to eq. 3 described in Materials and Methods.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The present study shows that low nanomolar concentrations of P-CTX-1 significantly alter the gating of TTX-S and TTX-R sodium channels in mammalian sensory neurons. P-CTX-1 reduced peak sodium current amplitude in a concentration-dependent fashion and produced differential effects on other aspects of sodium channel gating in the two channel subtypes. In TTX-S sodium channels, the major action was a hyperpolarizing shift in the voltage dependence of activation and inactivation, whereas in TTX-R sodium channels, P-CTX-1 produced an increase in the rate of recovery of channel inactivation. The most intriguing finding of the present study, however, was the P-CTX-1-induced increase in the leakage current that was mediated through TTX-S sodium channels.

Several neurotoxins that interact with a variety of receptor sites on the voltage-gated sodium channel can produce repetitive firing of nerves by altering the kinetics of activation and/or inactivation such as veratridine (Ulbricht, 1969), -scorpion toxins (Wang and Strichartz, 1983), funnel-web spider toxins (Nicholson et al., 1994, 1998), and brevetoxins (Baden et al., 1994). These neurotoxins slow or remove inactivation to maintain sodium channels in the open state and cause repetitive activity resulting from a prolonged depolarizing after-potential. Unlike these neurotoxins, the present study has shown that P-CTX-1 had no effect on the activation or inactivation kinetics of both TTX-S and TTX-R sodium channel currents. As a result, a slower activation or inactivation gating mechanism cannot explain the CTX-induced increase in neuronal excitability seen in other studies (Benoit et al., 1986, 1996; Bidard et al., 1984; Brock et al., 1995; Hamblin et al., 1995). This finding is in accordance with previous voltage-clamp experiments on guinea pig cardiac muscle (Seino et al., 1988) but are in contrast to observations in voltage-clamped frog myelinated nerve (Benoit et al., 1986) where late inward currents during long depolarization's were observed. This may reflect the use of partially purified CTX in the frog myelinated nerve study or possibly differences in channel gating kinetics between frog and rat sodium channels. Interestingly a suppression of fast inactivation has also been observed in the other site-5 cyclic polyether neurotoxin, brevetoxin (PbTX-3), in single-channel experiments on rat nodose ganglia (Baden et al., 1994; Jeglitsch et al., 1998); but others report no change in single-channel mean open lifetime of NG108 to 15 neuroblastoma cells (Sheridan and Adler, 1989). These differences in the actions of toxins that interact with neurotoxin receptor site-5 may be caused by structural differences between P-CTX-1 and PbTX that result in differing interactions with the sodium channel, especially in the region of the inactivation gate. In addition, they may also represent differences in sodium channel inactivation mechanisms between cell types.

Repetitive firing of frog myelinated nerves induced by CTX (Benoit et al., 1986) could also be caused by after-depolarization arising from a slowing of sodium channel deactivation. Under voltage-clamp conditions, this results in the development of tail currents that arise from the slowing of the activation (m) gate closure between the inactivated and resting states of the channel. Certain neurotoxins such as the pyrethroid tetramethrin (Tatebayashi and Narahashi, 1994), -scorpion toxins (Centruroides sculpturatus toxins II, III, and III; Wang and Strichartz, 1983) and DDT (Lund and Narahashi, 1981) induce prolonged tail currents, which can last up to several seconds. The absence of prolonged tail currents in the presence of P-CTX-1, however, indicates that repetitive firing of nerves is not caused by alterations in the deactivation of voltage-gated sodium channels.

One of the major effects likely to increase neuronal excitability with P-CTX-1 is the concentration-dependent hyperpolarizing shift in the voltage dependence of activation of TTX-S sodium channels. This shift occurred in the absence of alterations in the reversal potential, which would indicate that the ion selectivity of the channels is unaltered. Both sodium channel subtypes shifted the activation voltage to membrane potentials more negative than normal, but this was very small in TTX-R sodium channels. Indeed this was only just significant at P-CTX-1 concentrations higher than 5 nM. Interestingly, however, these hyperpolarizing shifts in activation were markedly less than those found in other voltage-clamp studies on CTX (Benoit et al., 1986) and PbTx (Huang et al., 1984; Baden et al., 1994; but see Sheridan and Adler, 1989).

Unlike a number of other agents that modulate sodium channel function such as antiarrhythmic drugs (Ragsdale et al., 1991), local anesthetics (Hille, 1977), batrachotoxin (Tanguy and Yeh, 1991), sea anemone toxins (Wasserstrom et al., 1993), and versutoxin (Nicholson et al., 1994), P-CTX-1 did not produce a use-dependent action. Pacific CTX-1 did, however, precipitate large increases in the leakage current in cells expressing TTX-S sodium currents, which were reversed when TTX was added, indicating an action mediated through TTX-S sodium channels. This current is no doubt responsible for the tetrodotoxin-sensitive membrane depolarization observed in a variety of other studies (Bidard et al., 1984; Benoit et al., 1986; Lewis and Endean, 1986; Seino et al., 1988). In particular, Benoit et al. (1986, 1996) found that in isolated frog myelinated nerves, a fraction of sodium channels failed to inactivate after perfusion with CTX (CTX-1B). In support of the present study it was concluded that CTX acts to increase neuronal excitability by modifying a fraction of sodium channels so that they remain in the open state permanently.

Increases in the rate of transition between the inactivated and resting states of the channel may provide a mechanism by which an increase in neuronal excitability can also occur. A significant increase in the rate of recovery from sodium channel inactivation was observed in TTX-R sodium currents, but not TTX-S sodium currents, perfused with P-CTX-1. This indicates that P-CTX-1 acts on TTX-R sodium channels primarily by increasing the rate at which channels undergo transition from the inactivated to the resting state during repolarization. This indicates that P-CTX-1 induced increases in the repriming kinetics, which contribute to an increase in neuronal excitability only in the case of TTX-R sodium channels.

The CTX-induced effects appeared resistant to sustained (20-30 min) washout with external solution, although some reversal of the changes in peak current amplitude usually occurred after 10 to 15 min. This lack of reversibility has also been observed in guinea pig heart and rat phrenic-hemidiaphragm nerve-muscle preparations (Lewis and Wong Hoy, 1993; Wong Hoy and Lewis, 1992) and is consistent with the high lipid solubility of the toxin and its retention in the neuronal membrane (Scheuer et al., 1967). Undoubtedly this underlies the reemergence of the leakage current, following blockage by TTX, when the cells were washed with P-CTX-1 free external solution and presumably contributes to the long-term clinical symptoms following ciguatera poisoning.

These differential actions on TTX-S (PN1 subtype) versus TTX-R (PN3 subtype) sodium channels, described above, are not without precedent. Scorpion -toxins, sea anemone toxins, and funnel-web spider toxins have all been shown to target the TTX-S rather than the TTX-R sodium channel subtype in dorsal root ganglion neurons (Roy and Narahashi, 1991; Nicholson et al., 1994, 1998; H. Wilson and G. Nicholson, unpublished observations). It would therefore appear that the variation in channel subtype may prevent binding of these sodium channel modulators or drastically alter their actions on channel gating and kinetics, in particular, channel inactivation.

In summary, P-CTX-1, the most potent strain of CTX isolated thus far, acts in a differential manner on the two types of sodium channel subtypes found in rat dorsal root ganglion neurons. P-CTX-1 decreases peak sodium current in both TTX-S and TTX-R sodium channels but differentially alters the voltage dependence of gating and the rate of recovery from sodium channel inactivation. These effects suggest P-CTX-1 causes TTX-S sodium channels to open closer to the normal resting membrane potential, whereas TTX-R sodium channels recover from inactivation more quickly, enabling an earlier transition to the open state. Moreover, P-CTX-1 appears to induce a permanently open state of TTX-S sodium channels as evidenced by a significant increase in leakage current. These effects to modulate sodium channel gating may provide an explanation for the increased neuronal excitability and generalized disturbance in nerve conduction observed in ciguatera patients.