Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

A multiplicity of K⁺ channels are widely distributed in both peripheral nervous system and CNS tissue (Hille, 1992), where they regulate neuronal excitability and clearly modulate synaptic events governing neurotransmission. One important mechanism by which presynaptic K⁺ channels affect synaptic transmission is by controlling neurotransmitter release. Several hypotheses suggest that the blockade of K⁺ channels prolongs the action potential duration, which leads to an increased influx of extracellular Ca⁺⁺ through voltage-sensitive Ca⁺⁺ channels and results in an enhanced release of neurotransmitter (Thesleff, 1980; Rudy, 1988).

Aminopyridines and TEA have been used as standard reference compounds in a variety of studies involving the functions and properties of K⁺ channels (Glover, 1982; Rudy, 1988). These compounds have been classically employed as blockers of K⁺ efflux and conductances in a number of physiological preparations from both central and peripheral tissues (Rudy, 1988). Although there are differences in the selectivities of 4-AP and TEA for various K⁺ channels, these compounds share the ability to block presynaptic voltage-dependent K⁺ channels and modulate the release of a variety of neurotransmitters. The facilitating effects of 4-AP on neurotransmitter release have been reported for norepinephrine (Hu and Fredholm, 1991), ACh (Tapia and Sitges, 1982; Dolezal and Tucek, 1983; Drukarch et al., 1989), dopamine (Boireau et al., 1991; Scheer and Lavoie, 1991), -aminobutyric acid (Tapia et al., 1985) and glutamate (Tapia and Sitges, 1982; Tibbs et al., 1989). The stimulatory effects of 4-AP on ACh and dopamine release have been further corroborated in vivo using intrastriatal dialysis (Damsma et al., 1988; Dawson and Routledge, 1995). In addition, the K⁺ channel blocker TEA has been demonstrated to induce the release of ACh (Drukarch et al., 1989, norepinephrine (Hu et al., 1991) and dopamine (Boireau et al., 1991).

The indirect modulation of ligand-gated K⁺ channel function by 5-HT through second-messenger coupling has been well documented (see Belardetti and Siegelbaum, 1988). In contrast, the ability of K⁺ channels to modulate the neurochemical characteristics of 5-HT neurotransmission has received little attention. Anden and Leander (1979) reported that 4-AP administered peripherally did not change the turnover of 5-HT or dopamine but markedly accelerated that of norepinephrine in the brain and spinal cord. In contrast, recent data from Pei et al. (1995) using in vivo microdialysis demonstrated that 4-AP and TEA, tested at single concentrations and perfused directly into the hippocampus, enhanced 5-HT efflux in this brain region. Discrepancies between the two studies may be explained on the basis of routes of administration and differences in technical methods used to quantitate the levels of neurotransmitters.

The present study extends the previous findings involving K⁺ channel blockers and neurotransmitter release by examining the ability of 4-AP, pyridine analogs and TEA to enhance the spontaneous basal release of [³H]5-HT from rat hippocampal slices. In contrast to the previous study by Pei et al. (1995), experiments were designed to determine the full concentration-effect relationships for these agents and to examine the additivity of neurochemical responses between K⁺ channel blockers and chemical depolarizing agents such as veratridine and KCl. Finally, investigations examining the biochemical mechanism by which these agents facilitated the spontaneous release of [³H]5-HT were determined by studying the Ca⁺⁺ and Na⁺ dependence of these effects.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Animals. Male albino Sprague-Dawley rats (Charles River, Kingston, NY) were group-housed on a 12-hr light/12-hr dark lighting cycle under standard laboratory conditions. Animals were acclimated for a period of at least 7 days before experimentation and weighed between 300 and 400 g.

Materials. The following drugs and chemicals were used in this study: [³H]5-HT (specific activity = 97-99 Ci/mmol) (Amersham Corporation, Arlington Hts., Il), 4-AP, 3,4-DAP, 3-AP, 2-AP, pyridine and glybenclamide (Sigma Chemical Co., St. Louis, MO), TTX, veratridine, charybdotoxin, TEA and quinine (RBI, Natick, MA) and Apamin (Latoxan, Rosans, France). Fluoxetine was kindly provided by Lilly-Laboratories (Indianapolis, IN). All other reagents utilized were of the highest chemical grade and purity.

Neurotransmitter release. The rats were sacrificed by decapitation, and the brains were rapidly removed and placed on ice for dissection. The hippocampus was removed and washed in ice-cold Krebs buffer. Hippocampal tissue was subsequently chopped into slices (0.25 mm by 0.25 mm) using a McIlwain tissue chopper and suspended in a volume of oxygenated (95% O₂/CO₂) Krebs buffer (pH 7.4; 37°C). The composition of the buffer was (mM): NaKH₂PO₄, 1.2; NaCl, 118; KCl, 4.8; glucose, 10; CaCl₂, 1.3; MgSO₄, 1.2; NaHCO₃, 25; ascorbic acid, 0.1; pargyline, 0.128. The slices were subsequently incubated with 100 nM [³H]5-HT at 37°C for 60 min with occasional agitation. After incubation with [³H]5-HT, the slices were gently triturated and 200-µl aliquots of the preparation were loaded into the chambers of a Brandel (Gaithersburg, MD) superfusion apparatus that were enclosed by polyethylene filter discs. During the remainder of the experiment, the tissue preparation was superfused with Krebs buffer containing 10 µM fluoxetine to prevent the reuptake of 5-HT. After loading of the tissue into the apparatus, each chamber was continuously superfused with buffer for 45 min at a flow rate of 0.6 ml/min to remove excess radioactivity that was not incorporated into the tissue preparation. The flow rate was reduced to 0.3 ml/min during the remainder of the experiment, which was determined in preliminary studies to produce the optimal conditions for a stable base line, adequate oxygenation and fraction collection volumes. After the washout period, eighteen 5-min fractions were collected at a superfusion flow rate of 0.3 ml/min. Spontaneous release was recorded at base-line levels during the initial three collection fractions. Drug or appropriate vehicle was added during a 15-min period that began after the collection of the base-line fractions. Experiments designed to test the effects of K⁺-induced depolarization and release were performed by substituting an equimolar concentration of K⁺ for Na⁺. Studies investigating the Ca⁺⁺-dependent effects of [³H]5-HT release were performed by substituting 1 mM EGTA for CaCl₂. Compound or vehicle was subsequently washed out of the system by returning to standard buffer superfusion for the remainder of the experiment. Upon completion of the experiment, the total amount of radioactivity for each fraction, as well as that remaining in the slices and filters, was counted using standard scintillation techniques. Filters and slices were solubilized in 1 ml of NCS Solubilizer (0.6 N; Amersham) before the addition of Ready Organic scintillation cocktail (Beckman, Fullerton, CA).

Calculation of fractional release of [³H]5-HT. The release of [³H]5-HT was expressed as a percentage of the labeled transmitter present in the superfusion chamber at the time of collection such that Because fractional release is expressed as a percentage, the value obtained is independent of the total DPM present in any experiment, as previously reported (Snyder et al., 1992). Comparison of the maximal effects of compounds were made by subtracting the mean of the base-line fractions from the peak effect obtained during drug perfusion.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Pharmacological profile for the ability of K⁺ channel blockers to enhance spontaneous basal [³H]5-HT release. The pyridine derivatives increased the spontaneous basal release of [³H]5-HT from rat hippocampal slices in a concentration-dependent manner (fig. 1). The potencies of 4-AP and 3,4-DAP were similar, with calculated EC₅₀ values of approximately 1.2 mM and 0.88 mM, respectively. These two pyridine derivatives were also equally efficacious; the maximal responses at 30 mM were identical (~ 30%). In contrast, the structurally related compounds, pyridine, 2-AP and 3-AP, were less potent than 4-AP and 3,4-DAP. The apparent maximal effects observed at 300 mM indicated that 2-AP was more efficacious than 3-AP or pyridine in modulating the release of [³H]5-HT. The rank order of potencies from the estimated EC₅₀ values indicated that 3,4-DAP (0.88 mM)  4-AP (1.2 mM) > 2-AP (89 mM) > 3-AP(100 mM) > pyridine (256 mM).

View larger version (K): [in this window] [in a new window]   Fig. 1.   Concentration-effect curves for the ability of K+ channel blockers to enhance spontaneous basal release of [3H]5-HT. Hippocampal slices were perfused with increasing concentrations of 4-AP, 3,4-DAP, 3-AP, 2-AP, pyridine and TEA as described in "Materials and Methods." Data represent the means ± S.E.M. for peak percent fractional release (n  3 separate experiments). Individual data points were performed in triplicate.

Combined effects of K⁺ channel blockers and chemical depolarizing agents on the release of [³H]5-HT. Figure 2 displays the results of experiments designed to determine whether the effects of K⁺ channel blockers and chemical depolarizing agents are additive in their abilities to enhance [³H]5-HT release. Figure 2A demonstrates that at concentrations of 4-AP and TEA that produce approximately half-maximal effects (1 mM and 100 mM, respectively), the combined actions of the two compounds were greater than the sum of the individual responses elicited by each (4-AP: 11.39 ± 0.68%; TEA: 7.61 ± 0.14%; 4-AP + TEA: 31.77 ± 3.23%). In contrast, responses elicited by 4-AP (1 mM) and 3,4-DAP (1 mM) were not additive (fig. 2B). Interestingly, the effects of 4-AP (1 mM) were additive with the depolaring agent veratridine (5 µM), which acts by opening voltage-sensitive Na⁺ channels (fig. 2C). In fact, the combined response produced by 4-AP and veratridine was almost equal to the sum of their separate contributions (4-AP: 11.39 ± 0.68%; veratridine: 11.51 ± 2.63%; 4-AP + veratridine: 21.98 ± 1.69%). The simultaneous superfusion of 4-AP (1 mM) with 50 mM KCl did not result in any significant additivity in terms of maximal response, although the peak effect was prolonged in the presence of the two agents in comparison with KCl alone (fig. 2D). Nor was additivity obtained with 1 mM 4-AP and 25 mM KCl (data not shown).

View larger version (K): [in this window] [in a new window]   Fig. 2.   Combined effects of 4-AP + TEA (panel A), 4-AP + 3,4-DAP (panel B), 4-AP + Veratridine (panel C) and 4-AP + KCl (panel D) on the release of [3H]5-HT. Compounds were perfused simultaneously at concentrations approximately equal to their respective EC50 values indicated above and as described in "Materials and Methods." Data are the means ± S.E.M.; n  3 experiments per data point.

Ca⁺⁺ dependence of [³H]5-HT release induced by K⁺ channel blockers. The Ca⁺⁺ dependence of [³H]5-HT release induced by K⁺ channel blockers was investigated by testing the ability of these agents to act in a Ca⁺⁺-free buffer containing 1 mM EGTA. In control experiments, the spontaneous basal release of [³H]5-HT was not changed by a Ca⁺⁺-free buffer supplemented with EGTA. In agreement with previous studies, the increase in release produced by 50 mM KCl was significantly attentuated by decreasing extracellular Ca⁺⁺ (fig. 3A). Furthermore, the increase in spontaneous basal release of [³H]5-HT induced by 1 mM 4-AP and 100 mM TEA was significantly reduced under Ca⁺⁺-free conditions (fig. 3,B and C, respectively). In agreement with these results, pretreatment of the hippocampal preparation with 1 mM cadmium, a nonselective blocker of voltage-sensitive Ca⁺⁺ channels, reduced the response induced by 1 mM 4-AP by 63.3 ± 5.8%.

View larger version (K): [in this window] [in a new window]   Fig. 3.   Ca++ dependence of [3H]5-HT release induced by KCl (panel A), 4-AP (panel B) and TEA (panel C). Fractional release for each compound was measured in the absence and presence of 1.3 mM Ca++ as a component of the perfusion buffer. EGTA was added in the absence of added Ca++. Data are the means ± S.E.M. as calculated in four or more experiments.

View larger version (K): [in this window] [in a new window]   Fig. 4.   Concentration-dependent effect of 4-AP (panel A) and TEA (panel B) to induce [3H]5-HT release in the absence and presence of Ca++. Data were normalized according to control data obtained in normal Krebs containing 1.3 mM Ca++ and represent the means ± S.E.M. in four separate experiments. Asterisks indicate significant differences from the respective controls at each drug concentration. * P < .05 according to Student's t test.

Na⁺ dependence of [³H]5-HT release induced by K⁺ channel blockers. In order to determine whether the enhancement in the spontaneous basal release of [³H]5-HT release induced by K⁺ channel blockers was mediated by the opening of voltage-sensitive Na⁺ channels, the hippocampal preparations were pretreated with the Na⁺ channel blocker TTX at a concentration of 1 µM. TTX produced a slightly lower rate of spontaneous basal release of neurotransmitter and, furthermore, did not block the increase in chemically depolarized release produced by 50 mM KCl in hippocampal slices (fig. 5A). The releasing effects of 4-AP and TEA were also not inhibited by the presence of TTX, which is consistent with the lack of a direct membrane-depolarizing action of these compounds (fig. 5, B and C). In parallel experiments, 1 µM TTX blocked the release of [³H]5-HT induced by 5 µM veratridine, a Na⁺ channel opener (fig. 5D). This demonstrated that under the present experimental conditions, TTX did not block the effects of 4-AP and TEA at a concentration that was able to block voltage-dependent Na⁺ channels.

View larger version (K): [in this window] [in a new window]   Fig. 5.   Na+ dependence of [3H]5-HT release induced by KCl (panel A), 4-AP (panel B), TEA (panel C) and veratridine (panel D). The role of Na+ in the measured responses was assessed by pretreating the tissue preparation for 15 min with the Na+ channel blocking agent TTX at a concentration of 1 µM. The effects of KCl, 4-AP, TEA and veratridine were tested in the absence and presence of TTX during a 15-min period after the initial pretreatment. Data are the means ± S.E.M.; n  3 experiments per data point.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The results of the present report demonstrate that the aminopyridines and TEA can enhance the spontaneous basal release of [³H]5-HT in a rat hippocampal slice preparation. The rank order of potencies for the aminopyridine congeners was 3,4-DAP  4-AP > 2-AP > 3-AP > pyridine. This is consistent with the pharmacological profiles of these agents to block K⁺ channels in electrophysiological preparations (Glover, 1982; Rudy, 1988). Interestingly, the enhancement of release produced by the simultaneous addition of 1 mM 4-AP and 100 mM TEA was greater than that produced by either agent tested alone. Furthermore, the enhancement of release induced by 1 mM 4-AP was additive with 5 µM veratridine but not with 25 or 50 mM KCl. The effects of 4-AP and TEA are dependent on extracellular Ca⁺⁺ and do not appear to be mediated by modulating Na⁺ channel function. Taken together, the results demonstrate that K⁺ channel blockade can enhance the spontaneous basal release of [³H]5-HT in rat hippocampal slices.

This study confirms and extends previous findings that K⁺ channel blocking agents can modulate the release of a variety of neurotransmitters (see the Introduction). Indeed, a plethora of data demonstrate that aminopyridine derivatives and TEA block K⁺ efflux and conductances in a number of different physiological preparations (Rudy, 1988). For example, 4-AP and TEA have been shown to block K⁺ channels in rat brain synaptosomes (Bartschat and Blaustein, 1985) and in axonal preparations from the squid (Llinas et al., 1976; Augustine, 1990) and frog (Hille, 1967). However, one may argue that the effects of these agents may be mediated through G protein-coupled presynaptic receptors. In this regard, Drukarch and colleagues (1989) reported that 4-AP and TEA may displace alpha-2 adrenergic radioligands. This mechanism does not appear to be involved in the effects of these compounds on the basal release of [³H]5-HT, because it has previously been demonstrated that alpha-2 adrenergic compounds do not modulate the spontaneous basal release of [³H]5-HT (Raiteri et al., 1990; L.E. Schechter, unpublished observation). Furthermore, the effects of 4-AP and TEA do not appear to be mediated through presynaptic serotonergic autoreceptors, because methiothepin, a 5-HT autoreceptor antagonist that enhances the evoked release of [³H]5-HT (Cerrito and Raiteri, 1979), has no effect on the spontaneous basal release of this indoleamine transmitter in the presence of 5-HT uptake blockers (Gothert, 1980; L.E. Schechter, unpublished observation). Thus the results strongly suggest that the mechanism of action responsible for the enhancement of the basal release of [³H]5-HT by aminopyridines and TEA is related to their abilities to block K⁺ channels.

Another possible mechanism for the observed effects on [³H]5-HT overflow may be related to the ability of 4-AP and TEA to enhance the release of other neurotransmitters that can indirectly modulate serotonergic neuronal function. It has been reported that 4-AP can facilitate the release of various neurotransmitters, such as glutamate, norepinephrine, dopamine and -aminobutyric acid (see the Introduction). In this regard, it has been demonstrated that stimulating NMDA and non-NMDA receptors can actually increase the overflow of [³H]5-HT in cortical slices (Fink et al., 1995). This suggests that glutamate would be able to contribute to the effects observed in this paper on [³H]5-HT release. However, Fink and colleagues reported that glutamate was effective only at concentrations of  1 mM, and because of the efficiency of the glutamate transporter, it is questionable whether these high concentrations are reached in brain (Robinson and Dowd, 1997). Furthermore, norepinephrine can be released by stimulating NMDA and non-NMDA receptors (Fink et al., 1992) in addition to the release of this neurotransmitter induced by 4-AP. The effects of ACh on the release of [³H]5-HT have not been well studied. Taken together, these results suggest that the actions of 4-AP and TEA may be very complex and that at least part of their final effects on [³H]5-HT release may be due to other indirect mechanisms. Further studies are planned to examine systematically the contributions of these separate indirect effects on the release of [³H]5-HT induced by K⁺ channel blockade.

It is quite clear that the aminopyridines and TEA are not selective for any particular K⁺ channel. On the basis of a channel classification scheme designed using conductance parameters, 4-AP has been shown to block voltage-dependent K⁺ channels that have been termed non-inactivating delayed rectifiers (Meves and Pichon, 1977) and transient 'A'-type channels (Thompson, 1982). TEA also appears to block the non-inactivating delayed rectifier and transient 'A'-type channels (Foehring and Surmeier, 1993; Ruppersburg et al., 1993), but it is also capable of blocking the inward rectifier (Standen and Stanfield, 1980). TEA may also block a Ca⁺⁺-activated K⁺ channel current (Farley and Rudy, 1988). Pharmacological studies performed on clonal cell lines have revealed that the affinities of 4-AP and TEA differ significantly among the voltage-gated K⁺ channels (Rehm, 1991; Grissmer et al., 1994). It is important to note that compounds capable of blocking K_ATP or Ca⁺⁺-activated channels, such as glybenclamide, apamin and charybdotoxin, were ineffective in modulating release. On the basis of the present data, this suggests that voltage-dependent K⁺ channels are involved in the actions of 4-AP and TEA. In addition, the K_ATP opener lemakalim had no effect on the spontaneous basal release of [³H]5-HT. Concentrations of these compounds tested were based on their affinities in binding studies and were supramaximally effective in various physiological assays (Longman and Hamilton, 1992; Habermann and Horvath, 1980; Lucchesi et al., 1989). Furthermore, compounds capable of modulating K_ATP or Ca⁺⁺-activated channels have not been found to enhance the spontaneous basal release of neurotransmitters under normal physiological conditions.

Previous studies indicate that 4-AP, aminopyridine analogs and TEA are relatively selective for the voltage-gated K⁺ channels as a family, although, as noted above, their affinities differ for the individual subtypes (Rudy, 1988). Notably, the potencies of these compounds in enhancing the release of neurotransmitters depend on the system being studied (Tapia and Sitges, 1982; Huang et al., 1989). For example, 3,4-DAP is 2 to 4 times more active than 4-AP in releasing [³H]norepinephrine from hippocampal slices (Huang et al., 1989; L.E. Schechter, unpublished data). In contrast, 3,4-DAP and 4-AP are approximately equipotent in their effects on [³H]dopamine (Scheer and Lavoie, 1991) and [³H]5-HT as determined in this study. In fact, the rank order of potencies for the aminopyridines and TEA on [³H]5-HT release are in agreement with previous results obtained for the release of [³H]dopamine (Scheer and Lavoie, 1991). Interestingly, the results from a recent in vivo microdialysis study investigating extracellular neurotransmitter concentrations in the striatum demonstrated a lack of effect of 4-AP on 5-HT levels, although 4-AP induced a dose-dependent increase in extracellular dopamine (Dawson and Routledge, 1995). In contrast, TEA produced increases in both 5-HT and dopamine levels in the striatum. This suggests that the K⁺ channels that control 5-HT release in the striatum and the hippocampus differ. With the future development of selective agents for specific K⁺ channels, it will be of interest to determine which voltage-dependent K⁺ channels control the release of 5-HT as opposed to norepinephrine, ACh and other neurotransmitter systems found in brain.

Indeed, the differences in potencies and efficacies for the various neurotransmitter systems suggest that different K⁺ channels are responsible for the various pharmacological actions of 4-AP and TEA. Notably, the simultaneous administration of 4-AP and TEA produced an enhancement of release greater than the sums of their separate contributions. The synergistic effect obtained from the combination of 4-AP and TEA may indicate that the compounds act at unique sites on the same K⁺ channel and more efficaciously block the ionic current. In this regard, 4-AP and TEA bind to different regions of the K⁺ channel pore and demonstrate different intracellular and extracellular pharmacological properties (MacKinnon and Yellen, 1990; Kavanaugh et al., 1992; Kirsch et al., 1993). Alternatively, because each compound shows different affinities for various channels, the synergistic effect may be related to the blockade of distinct K⁺ channels by each agent. Notably, 4-AP and 3,4-DAP were not additive in their enhancement of [³H]5-HT release. These agents compete for a common binding site and appear to possess similar properties for K⁺ channels (Thesleff, 1980). Furthermore, 4-AP and veratridine, which clearly differ in their mechanisms and sites of action, were additive in that the observed effect was equal to the sum of their individual responses. This suggests that K⁺ and Na⁺ channels may be differentially targeted but act synergistically to increase transmitter release. The lack of a significant interaction observed between 4-AP and KCl may be due to the inability of 4-AP to act in the presence of a depolarized tissue preparation, which would be in agreement with a previously reported study (Tapia and Sitges, 1982). It has been demonstrated that the ability of K⁺ blocking agents to bind to specific channels is dependent on the transitional state of channel opening or closing (Rudy, 1988). In regard to this, 4-AP appears to block closed channels preferentially, because the blockade induced by 4-AP is voltage-dependent and decreases with increasing depolarization (Yeh et al., 1976; Meves and Pichon, 1977). Interestingly, the effect on [³H]5-HT release appeared to be prolonged in the presence of 4-AP and KCl. This may reflect a blockade of K⁺ channels by 4-AP that would alter the ability of the neurons to repolarize and return to equilibrium potential.

The effects of 4-AP and TEA on the spontaneous release of [³H]5-HT were partially dependent on extracellular Ca⁺⁺. The proposed mechanism for the enhancement of release induced by K⁺ channel blockers involves an increase in the influx of extracellular Ca⁺⁺ that is due to the maintained opening of voltage-sensitive Ca⁺⁺ channels (Thesleff, 1980). Subsequently, the influx of Ca⁺⁺ into the neuron would mediate stimulus-secretion coupling (Augustine et al., 1987). This mechanism has been hypothesized to be secondary to K⁺ channel blockade and to prolongation in the duration of the action potential. In the present study, the effects of these agents on [³H]5-HT release were not blocked completely by the removal of extracellular Ca⁺⁺ coupled to the addition of EGTA to the buffer. This appeared to be the case at both low and high concentrations of drug, although higher concentrations of drug were much less affected by the lack of Ca⁺⁺. In agreement with these experiments, the blockade of Ca⁺⁺ channels by cadmium, a nonselective blocker of voltage-sensitive Ca⁺⁺ channels (Lansman et al., 1986; Blaxter et al., 1989), did not completely reverse the effects of 1 mM 4-AP to increase the basal release of [³H]5-HT. Similar results were obtained in a previous study by Tapia and Sitges (1982), where it was reported that the enhancement in the release of glutamate, GABA and ACh induced by 4-AP was not totally eliminated by lowering extracellular Ca⁺⁺. Furthermore, it was demonstrated in that study that increasing concentrations of 4-AP appeared to overcome the lack of Ca⁺⁺ in the buffer. The Ca⁺⁺-independent effects could be explained by residual Ca⁺⁺ in the buffer or by the ability of these agents to stimulate directly the release of intracellular stores of Ca⁺⁺, which would not be affected by the experimental conditions employed in this study. Taken together, the present results further suggest that at high concentrations, 4-AP and TEA may have other mechanisms of action in addition to those related to K⁺ channel blockade.

The enhancement of [³H]5-HT release induced by 4-AP and TEA was not significantly attentuated by pretreatment with the Na⁺ channel blocking agent TTX at a concentration that clearly antagonized the effects of the Na⁺ channel opener veratridine. These results suggest that the actions of 4-AP and TEA were not dependent on the opening of Na⁺ channels and, accordingly, depolarization by ions flowing through Na⁺ channels. This appears to be consistent with the lack of a direct membrane-depolarizing effect through Na⁺ channels for 4-AP and TEA. Indeed, Agoston et al. (1983) have previously reported that 4-AP produces a negligible change in the resting membrane potential in synaptosomes.

The physiological significance of K⁺ channel modulation of 5-HT release has not been studied in detail. However, various lines of evidence suggest that important physiological interactions exist between K⁺ channels and the modulation of 5-HT release from serotonergic nerve terminals. In this regard, 4-AP can stimulate CNS activity when injected directly into the hippocampus or peritoneum (Fragoso-Veloz et al., 1990). Interestingly, rats injected with 4-AP display wet-dog shakes, a behavioral phenomenon associated with stimulating 5-HT_2A receptors in the brain (Lucki et al., 1984). Although wet-dog shakes produced by 4-AP are blocked by nonserotonergic agents such as the NMDA receptor antagonist MK-801 (Fragoso-Veloz and Tapia, 1992), more recent data have revealed that 4-AP-induced wet-dog shakes can also be blocked by the prior administration of ketanserin, a 5-HT_2A receptor antagonist (Gorman et al., 1995). Furthermore, the chronic administration of 4-AP (1 mg/kg s.c.) in rats over a 3-week period has been demonstrated to down-regulate the number of 5-HT_2A receptors in cortical brain tissue (Gorman et al., 1995). These effects may be mediated by the stimulation of 5-HT release, because 4-AP lacks any appreciable affinity at 5-HT_2A receptors (L.E. Schechter, unpublished observation). In addition, 4-AP has been shown to have antidepressant activity in animal models (Trella et al., 1995), and, notably, 5-HT_2A receptor desensitization is a biochemical change associated with chronic antidepressant administration. Thus the ability of 4-AP to induce the release of 5-HT appears to have biochemical and behavioral consequences that suggest that channel blockade can modulate serotonergic activity in vivo.

In conclusion, the present set of studies demonstrates that 4-AP and related pyridine derivatives, as well as TEA, stimulate the release of [³H]5-HT from serotonergic nerve terminals in hippocampal tissue. This effect appears to be mediated through the blockade of K⁺ channels, which is clearly a known property of the compounds utilized in this study. Future experiments will use site-selective agents to determine whether the effects on 5-HT release are occurring through specific K⁺ channels.