Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Although strychnine (fig. 1) is a well-known competitive antagonist of glycine at glycine-gated Cl channels (Saitoh et al., 1994), its binding to glycine-resistant sites has been recognized for many years. For instance, numerous reports have provided evidence that strychnine can inhibit cholinergic transmission at the neuromuscular junction and in sympathetic ganglia (Lanari and Luco, 1939; Alving, 1961). A presynaptic nicotinic mechanism appears to account for the ability of the toxin (1-100 µM; Koelle and McKinstry, 1976) to block the release of ACh from sympathetic ganglia (Collier and Katz, 1970) and to block the release of catecholamines from adrenal medullary chromafin cells (10-30 µM; Kuijpers et al., 1994). However, it is unlikely that strychnine poisoning could be accounted for by these effects, because they are observed at concentrations that are considerably greater than those at which the alkaloid exerts its excitatory and depressant effects in the CNS (Franz, 1975). Nevertheless, findings that strychnine produces spiking in the cortex (Amato et al., 1969; García Ramos et al., 1977), where glycine receptors are rare (Frostholm and Rotter, 1985), strongly support the notion that some of the toxic effects of this alkaloid can be accounted for by its action on receptors other than the glycine receptors in the CNS.

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Structure of strychnine.

In contrast to the previously noted low potency of strychnine at nonneuronal nAChRs, the toxin has been shown to have potent actions on certain subtypes of neuronal nAChRs. On the outer hair cells of the vestibular system, strychnine is even more potent than curare or -BGT in antagonizing the nicotinic actions of ACh (Bartolami et al., 1993; Erostegui et al., 1994; Guth et al., 1994). The anticholinergic effects of strychnine in the auditory system (Guth et al., 1994) have been observed at concentrations as low as 0.01 µM (Doi and Ohmori, 1993), and strychnine binds to the neuronal nAChRs of the outer hair cells with a K_d of approximately 35 nM (Lawoko et al., 1995). Of interest, in addition to the outer hair cell nAChRs, which are likely to be made up of the 9 subunit (Elgoyhen et al., 1994), homomeric 7 nAChRs, homomeric 8 nAChRs and homomeric 9 nAChRs ectopically expressed in Xenopus oocytes are highly sensitive to inhibition by toxicologically relevant concentrations of strychnine (Anand et al., 1993; Elgoyhen et al., 1994; Gerzanich et al., 1994; Peng et al., 1994).

Studies directed at determining the ability of centrally acting compounds to interact with native neuronal nAChRs have been helpful in improving the understanding of the functions of these receptors in the CNS. For instance, studies of the interactions of epibatidine, -conotoxin-ImI and amantadine with neuronal nAChRs have suggested that different subtypes of these receptors are likely to be involved in analgesia, convulsions and Parkinson's disease. In fact, 1) epibatidine, a toxin isolated from the skin of the frog Epipedobates tricolor and known to cause analgesia when injected i.p. in mice (Badio and Daly, 1994), is a potent agonist at 42-containing nAChRs in hippocampal neurons (Alkondon and Albuquerque, 1995); 2) -conotoxin-ImI, a toxin isolated from the venom of Conus imperialis and shown to cause convulsions when injected intracerebroventricularly in mice and rats (Johnson et al., 1995), is a competitive antagonist selective for native 7-bearing nAChRs (Pereira et al., 1996) and 3) at therapeutically relevant concentrations, amantadine, a drug used to treat Parkinson's disease, acts as a noncompetitive antagonist of ACh at 7-containing nAChRs in hippocampal neurons (Matsubayashi et al., 1997).

Considering that, 1) homomeric 7 nAChRs heterologously expressed in Xenopus oocytes are sensitive to blockade by strychnine (Peng et al., 1994), 2) convulsions are among the well-documented toxic effects of strychnine, 3) hippocampal foci can be developed as a consequence of exposure to strychnine (Baker, 1965), and 4) 7-bearing nAChRs, which are expressed in the majority of the hippocampal neurons (Alkondon and Albuquerque, 1993), appear to be involved in convulsions, we decided to investigate the interactions of the toxin with the different subtypes of nAChRs (including the 7-bearing receptors) present on hippocampal neurons. As reported previously, hippocampal neurons can respond to nicotinic agonists with at least one of three pharmacologically and kinetically distinct types of nicotinic currents, namely types IA, II and III that arise from the activation of nAChRs composed of 7, 42, and 34 subunits, respectively (Alkondon and Albuquerque, 1993; Alkondon et al., 1994; Ishihara et al., 1995; Barbosa et al., 1996).

Our study demonstrates that strychnine inhibits both 7 nAChR-mediated type IA currents and 42 nAChR-mediated type II currents recorded from rat hippocampal neurons. It is likely that some of the symptoms of strychnine-induced intoxication are associated with its ability to block 7 nAChRs in the CNS, because we provide evidence that the concentrations at which strychnine inhibits the activation of native 7 nAChRs present on hippocampal neurons are similar to those at which it interacts with glycine receptors in the CNS.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Tissue culture. Cultured hippocampal neurons were prepared by a procedure similar to that described previously (Alkondon and Albuquerque, 1993). The hippocampi of 17- to 18-day-old rat fetuses (Sprague-Dawley strain) were dissected out, minced and incubated with 0.25% trypsin (Gibco BRL, Grand Island, NY) for 30 min at 36°C. Using a sterile Pasteur pipette, hippocampal neurons were dissociated and plated at a density of approximately 700,000 cells per 35-mm culture dish precoated with collagen (Vitrogen 100, Celtrix Laboratories, Palo Alto, CA). The cells were cultured in an incubator at 36°C in a water-saturated, 10%-CO₂/90%-air atmosphere. The medium surrounding the cells, which consisted of MEM (Gibco BRL) enriched with fetal bovine serum (10%; Gibco BRL), horse serum (10%, Gibco BRL), glutamine (2 mM, Sigma Chemical Co., St. Louis, MO) and DNase (40 µg/ml, Sigma), was replaced twice a week with fresh medium consisting of MEM supplemented with horse serum (10%) and glutamine (2 mM). On the 7th day after plating the cells, uridine (final concentration = 6.7 µg/ml) and 5-fluoro-2-deoxyuridine (final concentration = 13.3 µg/ml) were added for 24 hr to the culture medium to inhibit the proliferation of nonneuronal cells. Neurons cultured for 14 to 30 days were used throughout this study.

Whole-cell current recording. Nicotinic currents were recorded from hippocampal neurons according to the standard whole-cell patch-clamp technique using an LM-EPC-7 patch-clamp system (List Electronic, Darmstadt, FRG) (Hamill et al., 1981). Patch pipettes were pulled from borosilicate capillary glass and had resistances between 2 and 5 M when filled with internal solution. The series resistance of the patches was 10 to 25 M and was not compensated. Currents were filtered at 3 kHz and either sampled directly by a microcomputer using the program pCLAMP (Axon Instruments, Foster City, CA) or stored for off-line analysis on videocassette tapes after passage through a pulse code modulation device (Neuro-Corder DR-384; Neuro-Data Instruments Corp., New York, NY). All experiments were performed at room temperature (20-22°C).

Drug applications. The drugs were delivered to the neurons according to the method described previously (Albuquerque et al., 1991). A U-shaped tube ("U-tube") with a pore of 250 to 400 µm in diameter at the apex was positioned about 50 µm directly above the neuronal soma. The input to the U-tube was connected to a manual switch valve, which was used to select the desired drug solution. The outlet of the U-tube was connected via an electric valve to a polyethylene tube for removal of drug solution from the neuronal surroundings. To prevent any leakage of the drug solution through the pore of the U-tube, the outlet was under continuous vacuum so that the drug solution and a small amount of the bath solution from the dish were removed when the valve was not activated. Upon activation of the valve by a 0.5 to 2-sec electric pulse, the drug solution flowed rapidly out of the apical pore and displaced completely the solution surrounding the neuronal soma and dendrites. Using this U-tube system, pulses of ACh alone or in an admixture with strychnine (0.1-300 µM) were applied to the neurons. Alternatively, in some experiments, strychnine was applied through the external bath perfusion system, in which case the same concentration of strychnine was added to the agonist-containing solution in the U-tube.

Data analysis. The peak amplitude, the rise time (10-90%) and the exponential decay-time constant () of the whole-cell currents were determined using the pCLAMP program. EC₅₀, IC₅₀ and the n_H values were determined with the SigmaPlot program according to the following equation for activation: and for inhibition: where I, I_max, [ACh] and [strychnine] were the peak whole-cell current, the maximum obtainable peak current, the concentration of ACh, and the concentration of strychnine, respectively. Values were expressed as the mean ± S.E.M.

Drugs used. Acetylcholine chloride, tetrodotoxin, atropine sulfate and strychnine hydrochloride were obtained from Sigma. Dihydro--erythroidine hydrobromide was a gift from Merck, Sharp & Dohme Research Laboratories (Rahway, NJ). ATP (Tris salt), phosphocreatine (di-Tris salt) and creatine phosphokinase (type I) were obtained from Sigma Chemical Co.

Statistical analysis. Two-way ANOVA was used to determine the significance of the results.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Nicotinic whole-cell currents were elicited by pulse application (0.5-2 sec) of ACh via the U-tube to hippocampal neurons in culture. In this study, 80 of the 121 neurons tested responded to ACh with type IA current, characterized by rapid onset after the start of the ACh pulse and rapid decay in the continued presence of ACh (fig. 2, top row; Alkondon and Albuquerque, 1993). Type II current, characterized by its slow decay phase (fig. 2, bottom row; Alkondon and Albuquerque, 1993), was recorded from 40 neurons. Type IB current, which is comprised of the rapidly decaying type IA current and the slowly decaying type II current (Alkondon and Albuquerque, 1993; Alkondon et al., 1994), was recorded from only one neuron. In our study, the percent probability of finding hippocampal neurons that responded to ACh with type II currents was somewhat higher than that reported previously (Alkondon and Albuquerque, 1993, 1995; Alkondon et al., 1994).

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Inhibitory effects of strychnine (10 µM) on the two major types of ACh (1 mM)-induced nicotinic currents in hippocampal neurons. Each row shows recording samples of one type of nicotinic current, each of which was recorded from a single neuron. ACh was applied to the neurons via the U-tube 1 min before (left column) and after (right column) the application of the admixture of ACh plus strychnine (10 µM) (middle column). ACh-evoked currents that decayed rapidly were characterized as type IA currents (top row) and were sensitive to reversible inhibition by strychnine. ACh-evoked currents that decayed slowly were characterized as type II currents (bottom row) and were insensitive to blockade by strychnine at the concentration shown. Holding potential, 60 mV.

Effects of strychnine on ACh-induced currents recorded from hippocampal neurons in culture. Application of an admixture of ACh (1 mM) plus strychnine (10 µM) to neurons that responded to ACh (1 mM) with type IA currents resulted in activation of currents whose amplitudes were approximately 60% smaller than those of currents evoked by ACh alone (fig. 2). The effect of strychnine on type IA currents was completely reversible within 1 min after washing of the neurons with external solution. In contrast, the peak amplitude and the kinetics of ACh (1 mM)-evoked type II currents were not affected when strychnine (10 µM) was applied to the neurons exclusively in admixture with ACh (1 mM) (fig. 2). Thus, strychnine was apparently more selective in inhibiting the 7 nAChR activity subserving type IA currents than in inhibiting the 42 nAChR activity subserving type II currents.

The potency and the mechanism of action of strychnine on type IA nicotinic currents recorded from hippocampal neurons. Because a single hippocampal neuron can express both 7 and 42 nAChRs (Alkondon and Albuquerque, 1993), contamination of type IA currents by slowly decaying currents corresponding to the activation of 42 nAChRs was avoided by the continuous perfusion of the neurons with DHE (0.1 µM)-containing external solution. At 0.1 µM, DHE selectively inhibits the activation of whole-cell currents subserved by 42 nAChRs (Alkondon and Albuquerque, 1993). Except in experiments where the voltage dependence of the effect of strychnine on nicotinic currents was addressed, the neurons were held at 60 mV.

View larger version (20K): [in this window] [in a new window]   Fig. 3.   Concentration-dependent inhibition of type IA current by strychnine and the mechanism underlying the action of the toxin. A, Concentration-response relationship for the inhibitory effect of strychnine on type IA currents. The peak amplitude of ACh (1 mM)-evoked currents was taken as 100% and was used to normalize the peak amplitude of the currents evoked in the presence of strychnine. The curves were fitted to the Hill equation. Symbols and bars represent the mean ± S.E.M. of results obtained from four to seven neurons. Holding potential, 60 mV. The IC50 and the nH values for strychnine obtained are shown in table 1. B, Effect of strychnine on the concentration-response relationship for ACh in evoking type IA currents. In each experiment, the peak amplitude of currents evoked by ACh (10 mM) was taken as 100% and was used to normalize the peak amplitude of currents evoked by other concentrations of ACh in the presence or in the absence of strychnine. Each symbol and bar represent the mean ± S.E.M. (n = 5-25). The EC50 and nH for ACh in evoking type IA currents are shown in table 2.

View larger version (17K): [in this window] [in a new window]   Fig. 4.   Inhibitory effect of strychnine (1 µM) on the peak amplitude of ACh (0.3 mM)-evoked type IA current at various holding potentials. A, Sample recordings of type IA currents evoked by application of ACh or ACh plus strychnine to a neuron held at various holding potentials. The admixture of ACh plus strychnine was applied via the U-tube to the neuron after its 10-min perfusion with strychnine-containing external solution. B, In each experiment, the peak amplitudes of the ACh-elicited currents at 120 mV were taken as 100% and were used to normalize the amplitude of the currents evoked by ACh in the presence or in the absence of strychnine at all other holding potentials. C, The relationship between the holding potential and the ratio of the amplitude of the currents evoked in the presence of strychnine (1 µM) to the corresponding amplitude of the currents evoked in the absence of strychnine at all holding potentials is shown. Symbols and bars represent the mean ± S.E.M. from experiments using seven neurons.

View larger version (21K): [in this window] [in a new window]   Fig. 5.   Concentration-dependent prolongation of decay phase of type IA current by strychnine. Top traces, Sample recordings of type IA currents evoked by application of ACh (0.3 mM) in the absence and in the presence of strychnine (0.3-3 µM) to neurons held at 60 mV. Each pair of traces was obtained from one neuron. To illustrate the effects of strychnine on the kinetics of type IA currents, the peak amplitudes of the currents evoked by ACh in the presence of strychnine were scaled to match those of the currents evoked by ACh alone. Bottom graph, The relationship between the membrane potentials and the decay-time constants of the currents evoked by ACh (0.3 mM) alone () or in the presence of 1 µM strychnine () shows the voltage independence of the strychnine-induced prolongation of the decay phase of type IA currents. Only the values of the fast component of the decay phase of type IA currents are shown. Quantification of the effects of strychnine on the decay phase and on the rising phase of type IA currents is shown in table 3.

The potency and the mechanism of action of strychnine as an inhibitor of the 42 nAChR activity subserving type II currents in hippocampal neurons. Because type II currents elicited by ACh (1 mM) were insensitive to strychnine (10 µM) (see fig. 2), the effects of the toxin on these currents were investigated at concentrations 10-fold higher than those used to test the inhibition of the type IA current.

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Concentration-dependent inhibition of type II current by strychnine. Traces, top row, Sample recordings of type II currents recorded from a hippocampal neuron exposed at 1-min intervals to ACh (0.3 mM) (left trace) and ACh (0.3 mM)-plus strychnine (100 µM) (right trace). Traces, bottom row, Sample recordings of ACh-evoked type II current recorded from a neuron before (left trace) and after (right trace) its 10-min perfusion with strychnine (100 µM)-containing external solution. Right graph, Concentration-response relationship for the inhibitory effect of strychnine on type II currents. In each experiment, the peak amplitude of ACh (0.3 mM)-evoked currents was taken as 100% and was used to normalize the peak amplitude of the currents evoked in the presence of strychnine. The curves were fitted to the Hill equation; the EC50 and the nH for strychnine in blocking type II currents under the two experimental conditions are shown in table 1. Symbols and bars represent the mean ± S.E.M. of results obtained from six to nine neurons. Holding potential, 60 mV.

View larger version (19K): [in this window] [in a new window]   Fig. 7.   Voltage-dependent inhibition by strychnine of type II currents evoked by ACh (0.3 mM). A, Sample recordings of type II currents evoked by application of ACh or ACh plus strychnine to a neuron held at various membrane potentials. The admixture of ACh plus strychnine was applied to the neuron after its 10-min perfusion with strychnine-containing external solution. B, In each experiment the peak amplitude of the ACh-elicited currents at 120 mV was taken as 100% and was used to normalize the amplitude of the currents evoked at all other holding potentials by ACh in the presence or in the absence of strychnine. C, The relationship between the holding potentials and the ratio of the amplitude of the currents evoked in the presence of strychnine (30 µM) to the corresponding amplitude of the currents evoked by ACh alone at all holding potentials is illustrated. Symbols and bars represent the mean ± S.E.M. from experiments using three neurons.

View larger version (28K): [in this window] [in a new window]   Fig. 8.   Concentration-dependent effect of strychnine on the decay phase of type II currents. The peak amplitudes of type II currents evoked by ACh (0.3 mM) in the presence of increasing concentrations of strychnine were scaled to match the peak amplitudes of the currents evoked by ACh alone. Each pair of traces was obtained from one neuron. A, Strychnine was applied to the neurons exclusively as an admixture with ACh. B, The admixture of strychnine plus ACh was applied to the neurons after their 10-min perfusion with strychnine-containing external solution. Holding potential, 60 mV. The quantification of the effect of strychnine applied to the neurons via the bath and the U-tube on the decay phase of type II currents is shown in table 4.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Our study revealed that strychnine, a toxin well known for its convulsant effects and for its ability to block glycine-activated Cl channels, acts as a competitive nicotinic antagonist at the 7 nAChRs, which give rise to type IA currents in hippocampal neurons, and inhibits via noncompetitive mechanisms the 42 nAChR activity. On the basis of the IC₅₀ values obtained in our study, strychnine appears to be 30 times more potent in blocking type IA currents (IC₅₀ = 1.2 µM) than in blocking type II currents (IC₅₀ = 38 µM).

Mechanism of action of strychnine on the 7 nAChRs subserving type IA currents. Studies of the concentration-response relationship for ACh in evoking type IA currents in the absence and in the presence of 2 µM strychnine (a concentration of strychnine that is close to its IC₅₀ in blocking type IA current) demonstrated that the toxin decreases the apparent potency of ACh; strychnine increased the EC₅₀ for ACh from 0.4 to 0.8 mM. In addition, neither the maximal responsiveness of the 7 nAChRs subserving type IA currents nor the Hill coefficient for ACh was affected by the toxin. These results suggested that strychnine competes with ACh for the agonist binding site(s) on the 7 nAChRs. The findings that 1) the reduction of the peak amplitude of type IA currents by strychnine was voltage independent and 2) the decay phase and the rise time of type IA currents were prolonged by the toxin favor the notion that the toxin is a competitive antagonist of ACh at the 7 nAChRs.

Mechanism of action of strychnine on the 42 nAChRs subserving type II currents. The acceleration of the decay phase of type II currents and the voltage-dependent reduction of the peak amplitude of these currents by strychnine strongly suggest that the toxin acts as an open-channel blocker of the 42 nAChR. This change in the decay phase of type II currents had the same magnitude regardless of whether the neurons were allowed to equilibrate with strychnine or were rapidly exposed to the admixture of this toxin with ACh. However, the apparent potency of strychnine in reducing the peak amplitude of type II currents was 3-fold higher when the neurons were perfused with strychnine-containing external solution before their exposure to the admixture of ACh plus strychnine, thus indicating that, in addition to blocking the 42 nAChR in its open state, strychnine can also interact with 42 nAChRs that are not open.

Structure-activity relationships of strychnine. The pharmacophore for molecules that interact with the glycine receptor appears to bear two negative regions contributed by a coplanar amide nitrogen and phenyl group and by a positive region at the opposite end of the molecule borne by a nitrogen with its adjacent carbon. The antagonist property of strychnine is carried in the carbonyl oxygen of the molecule (Aprison et al., 1995; Galvez-Ruano et al., 1995). In contrast, it has been proposed that strychnine's role as a nicotinic antagonist is met by the cationic nitrogen and the ether oxygen (fig. 1), both of which form a hydrogen bond with the nicotinic receptor (Beers and Reich, 1970; Sheridan et al., 1986). It is unclear whether the same pharmacophore would operate for the rat 7-containing nAChR. Nevertheless, this class of semi-rigid compounds may provide further insight to the structure of the neuronal nAChRs bearing the 7 subunit.

The relevance of the neuronal nAChR sensitivity to inhibition by strychnine. The toxic effects of strychnine have been associated with its ability to block glycine receptors. However, based on our study, the K_i for strychnine as a competitive antagonist of ACh at the 7 nAChRs is in the submicromolar range, as is the affinity of the toxin for the glycine receptors. Direct radioligand analysis has shown that strychnine binds to the glycine receptor with K_i of 32 nM (Saitoh et al., 1994), and according to a pA₂ analysis, strychnine interacts with glycine receptors with a K_d of 270 nM (Ito and Cherubini, 1991). In many studies, strychnine concentrations of 1 to 10 µM (Mercuri et al., 1990; Ito and Cherubini, 1991) or 10 to 50 µM (Wu et al., 1992) have been used to inhibit glycine-receptor-mediated responses. Thus, at the concentrations generally used, the toxin may block not only the function of the glycine receptors but also that of neuronal nAChRs in the CNS. Considering that the degree of inhibition of receptors in vivo will be determined by the level of each endogenous agonist versus its affinity for its receptor, a challenge is raised to the common belief that the toxic effects of strychnine are mediated by its inhibitory action on glycine receptors. The finding that the activity of the 7-bearing nAChRs in the rat hippocampus was particularly sensitive to inhibition by strychnine supports the concept that these receptors are likely to be involved in controlling the overall neuronal excitability (Sargent, 1993; Lindstrom, 1995; Role and Berg, 1996; Albuquerque et al., 1997).