Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

GABA is an important and abundant inhibitory neurotransmitter in the central nervous system (Krnjevi and Schwartz, 1966) that activates two families of receptors: GABA_ARs and GABA_BRs. Members of both receptor families gate specialized ion channels. For example, the GABA_AR is a hetero-oligomeric integral membrane protein complex that includes a GABA-gated chloride channel and various recognition sites enabling allosteric modulation. Heterogeneity of the GABA_AR has been studied with various techniques, including DNA recombination. Recombinant DNA coding for rat GABA_AR revealed that both its conductance and gating properties depend on the subunit composition (Verdoorn et al., 1990).

Interestingly, molecular studies reveal structural homology between the receptor/channel proteins selectively interacting with GABA, glycine and nicotine (Greeningloh et al., 1987; Schofield et al., 1987; Karlin and Akabas, 1995). These structural similarities may underlie the shared sensitivity of these receptor types to pharmacological agents. For example, in addition to its well known effect on peripheral nicotinic acetylcholine receptors d-tubocurare also inhibits the response of neurons to inhibitory transmitters (Hill et al., 1976; Scholfield, 1980; Lebeda et al., 1982). Waglerin-1 is a lethal peptide purified from the venom of Wagler's pit viper (Trimeresurus wagleri; Weinstein et al., 1991; Schmidt et al., 1992) which also interacts with the nicotinic acetylcholine receptor of mature muscle (Aiken et al., 1992; Sellin et al., 1996). For this reason, we explored the effect of Waglerin-1 on I_GABA of hypothalamic neurons. We found that Waglerin-1 had dual effects on I_GABA; i.e., Waglerin-1 potentiated I_GABA in some neurons and depressed I_GABA in others. We suggest that these opposite effects of Waglerin-1 relate to heterogeneity of the GABA_AR which native neurons express. Part of these data have appeared in abstract form (Ye and McArdle, 1995a).

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Isolation of neurons and electrophysiological recording. VMH neurons were prepared as described previously (Ye and McArdle, 1995b). Briefly, 7- to 17-day-old rats or mice were subjected to cervical dislocation. Their brains were quickly excised and placed into iced "standard external solution" containing (mM) NaCl 140, KCl 5, MgCl₂ 1, CaCl₂ 2, glucose 10, HEPES 10; (pH was adjusted to 7.4 with Tris base and osmolarity to 320 mmol/kg with sucrose). The brain was then glued to the chilled stage of a vibratome (Campden Instrument, LTD, Cambridge, England) and sliced to a thickness of 350 µm. Slices were transferred to standard solution containing 1 mg pronase/6 ml and incubated (31°C) for 20 min. After an additional 20 min incubation in solution containing 1 mg thermolysine/6 ml, micro-punches of the VMH were then isolated and transferred to a 35-mm culture dish. Mild trituration of these punches through heat polished pipettes of progressively smaller tip diameter served to dissociate single neurons. Within 20 min of trituration, isolated neurons had attached to the bottom of the culture dish and were ready for electrophysiological experiments. In the earlier part of this study, we did experiments on mice. Subsequently, we focused on rats. There were no observable differences of I_GABA and the effects of Waglerin-1 between these two murine species.

Toxin preparation. The venom of T. wagleri was obtained from Ventoxin Laboratories (Frederick, MD). Waglerin-1 was extracted and purified at USAMRIID as described by Weinstein et al. (1991). The preparation of Waglerin-1 is as described by Aiken et al. (1992). Briefly, the toxin is a polypeptide of 22 amino acids with a sequence of GGKPDLRPCHPPCHYIPRPKPR. Nominally pure samples (approximately 1 g/liter or 400 µM) were stored at -25°C in a buffer solution (pH 8.5) containing (mM): Tris (50), N-ethylmorpholine (6.0) and trifluoracetic acid (4.0). This stock solution was added directly to the perfusate. Experiments were also performed using synthetic Waglerin-1 produced with a model 431A Applied Biosystems peptide synthesizer (Foster City, CA) using N(9-fluorenylmethyl-oxy-carbonyl) synthetic chemistry. After synthesis, the single disulfide bond was allowed to form by incubation at pH 8.0 (21°C) overnight at a concentration of 0.5 to 1 mg/ml (Sellin et al., 1996). As noted previously (Aiken et al., 1992), the effect of the natural and synthetic form of Waglerin-1 was equivalent.

Chemical application. Solutions of GABA (Sigma Chemical Company, St Louis, MO) and Waglerin-1 were prepared on the day of the experiment. Flumazenil (a gift from Hoffmann-La Roche, Nutley, NJ) was dissolved in DMSO (Sigma Chemical Company, St Louis, MO) to give a stock solution of 10 mM which was added to the external solution to produce the desired concentration on the day of experiments. The final concentration of DMSO in test solutions was 0.1% (v/v) or less and had no effect on I_GABA. Picrotoxin was dissolved in methanol. The final concentration of methanol in test solutions was 0.1% (v/v) or less and had no effect on I_GABA. These solutions were applied to a dissociated neuron with a fast superfusion system having a multibarreled pipette as described previously (Ye and McArdle, 1995b). The tip of the superfusion pipette was normally placed 50 to 100 µm away from the cell, a position allowing rapid as well as uniform drug application and preserved the mechanical stability of the neuron. Throughout all experimental procedures the bath was continuously perfused with the standard external solution.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

GABA induced currents in VMH neurons recorded before and during coapplication of Waglerin-1. Approximately 89% of the VMH neurons examined in this study produced currents in response to GABA. The threshold concentration for I_GABA varied from 3 to 15 µM GABA. In response to threshold agonist concentrations, I_GABA slowly reached a peak from which it did not decay during a few seconds of continuing agonist application. At elevated GABA concentrations, I_GABA reached its peak amplitude more rapidly and clearly decayed. The latter can be attributed to desensitization of the GABA_AR (Akaike et al., 1985; Bormann and Clampham, 1985). I_GABA reversed direction at approximately -20 mV, a value equivalent to the Nernst equilibrium potential (-21 mV) calculated for the Cl concentration of the solutions used. As expected, I_GABA of VMH neurons was sensitive to picrotoxin (data not shown). Three groups of neurons could be distinguished on the basis of the effects of Waglerin-1 on I_GABA. That is, I_GABA was potentiated (fig. 1) for 55% and depressed (figs. 6, 7, 8, 9) for 31% of the neurons to which Waglerin-1 and GABA were coapplied. The remaining 14% of the neurons examined showed no significant change of I_GABA with the coapplication of Waglerin-1. To facilitate a description of the effects of Waglerin-1 on I_GABA, we describe the two populations of neurons responding to the peptide separately.

View larger version (K): [in this window] [in a new window]   Fig. 1.   Waglerin-1 enhances IGABA of hypothalamic neurons in response to subsaturating concentrations of agonist. IGABA was recorded in response to 10 µM GABA alone (open horizontal bars) or GABA plus 32 µM Waglerin-1 (filled bars). The holding potential was 0 and -60 mV for the upper and lower row of records, respectively. Comparison of the records in columns a and b demonstrates that Waglerin-1-enhanced IGABA in response to 10 µM GABA. Waglerin-1 also increased the rate of decay of IGABA in response to 10 µM GABA. In contrast, comparison of the records in columns c and d demonstrates that when GABA was increased 100-fold to 1 mM Waglerin-1 no longer potentiated IGABA. All recordings were obtained from the same mouse neuron subjected to the nystatin perforated patch technique. The amplitude calibration is 300 pA for columns a and b and 1000 pA for columns c and d.

View larger version (K): [in this window] [in a new window]   Fig. 6.   Waglerin-1-induced depression of IGABA is voltage dependent. A and B show IGABA in response to 10 µM GABA alone and coapplied with 20 µM Waglerin-1 at holding potentials of 0 and -40 mV, respectively. Note the greater depression of IGABA at -40 mV. The tracings in the right column of rows A and B are superimposed IGABA recorded before (solid line) and after (data points) Waglerin-1 that have been normalized to the control peak amplitude. Note that the decay of IGABA was slower when Waglerin-1 depressed peak amplitude. C and D are examples of IGABA when the neuron is subjected to a voltage ramp protocol. During this protocol, membrane potential was gradually stepped from -100 to 60 mV over an interval of 1.6 sec. The initial membrane response shown in the records of C and D is the membrane leakage current that was subtracted from the current recorded when the ramp was applied in the presence of 50 µM GABA alone or plus 32 µM Waglerin-1. E, Current-voltage relations obtained from the ramp protocol applied in the presence of GABA alone () or GABA plus Waglerin-1 (). F, Normalized current-voltage relations from the same experiment of E showing the stronger effect of Waglerin-1 on inward currents. Arbitrary unit 1 corresponds to 555 pA outward current for GABA () and 164 pA in the presence of Waglerin-1 ().

View larger version (K): [in this window] [in a new window]   Fig. 7.   A, The depressant effect of Waglerin-1 on IGABA is fast in onset and offset. Brief pulses of 20 µM Waglerin-1 depress IGABA in response to a longer pulse of 10 µM GABA alone. Holding potential -60 mV. B, Waglerin-1 depresses IGABA of hypothalamic neurons IGABA evoked in response to 10 µM GABA. The holding potential was -50 mV. C, In contrast, when GABA was increased 50-fold to 500 µM Waglerin-1 no longer depressed IGABA. The amplitude calibration is: 400 pA for B; 2000 pA for C.

View larger version (K): [in this window] [in a new window]   Fig. 8.   A, Concentration-response relationships for GABA (1-1000 µM) in the control () and in the presence of 32 µM Waglerin-1 (). Peak IGABA was plotted as a function of GABA concentration. Solid lines are least square fit of the Michaelis-Menten equation (see legend of fig. 4) to the experimental data. The EC50 was 15 µM and 48 µM, although n was 1.2 and 1.2, for GABA alone and GABA plus Waglerin-1, respectively. B, The concentration-response relation for Waglerin-1 induced depression of IGABA. The normalized peak IGABA in response to 10 µM GABA plus varying concentrations of Waglerin-1 is plotted as a function of the Waglerin-1 concentration; peak IGABA was first normalized to the value obtained in response to 10 µM GABA alone. Each data point is the mean ± S.E.M. of four to six neurons. IGABA was recorded at a holding potential of -60 mV. Solid lines are fit of the following form of the Logistic equation to the experimental data: I/Imax = 1/{1 + (C/IC50)n}, where I, IMax, C, IC50 and n are IGABA, maximal IGABA, Waglerin-1 concentration, the concentration at which IGABA is 50% of maximum, and the Hill coefficient, respectively. The IC50 was 24 µM and n was 0.9 for Waglerin-1.

View larger version (K): [in this window] [in a new window]   Fig. 9.   A, Zn++ -induced suppression of IGABA is voltage dependent. Left panel, current-voltage relations obtained from the ramp protocol (see legend to fig. 6) applied in the presence of GABA alone () or GABA plus 1 µM Zn++ (). Right panel, normalized current-voltage relations from the same experiment showing the stronger effect of Zn++ on inward currents. B, Depression of IGABA by 24 µM Waglerin-1 and 1 µM Zn++ applied to the same neurons; IGABA was induced by 20 µM GABA. Each data point was obtained from a neuron exposed to Waglerin-1 and Zn++ separately. The regression line (continuous line) has a slope of 1.02.

Potentiation of I_GABA by Waglerin-1. The records of I_GABA presented in figure 1 demonstrate that the potentiating effect of Waglerin-1 was not dependent on voltage. That is, 32 µM Waglerin-1 increased peak I_GABA in response to 10 µM GABA when neurons were held at 0 (upper row) and -60 mV (lower row). Figure 2A summarizes the current-voltage relations of peak I_GABA for five neurons before () and during () coapplication of 32 µM Waglerin-1. The peptide potentiated I_GABA at both positive and negative holding potentials. For example, at -60 and 20 mV 32 µM Waglerin-1 potentiated I_GABA by 66.6 ± 20.0% and 60.1 ± 10% of control (mean ± S.E.M., n = 5, P > .5). In contrast to peak I_GABA, the current amplitude measured at the end of a 6-sec GABA application decreased; i.e., 32 µM Waglerin-1 reduced the end I_GABA to 48.5 ± 15.7% (n = 7) of control end current (see fig. 3). This was associated with an increase in the rate of I_GABA decay. In contrast to its effect on current amplitude and decay, Waglerin-1 did not alter the apparent reversal potential (-21 mV) of peak I_GABA. However, the potentiating effect of Waglerin-1 on I_GABA disappeared or changed as the GABA concentration was increased. This phenomenon is illustrated in figure 1c and d; coapplication of 1 mM GABA and 32 µM Waglerin-1 resulted in slight depression of I_GABA. These records were obtained from the same neuron presented in figure 1a and b whose I_GABA in response to 10 µM GABA increased when 32 µM Waglerin-1 was coapplied. This concentration dependence is further analyzed in the study of the effect of Waglerin-1 on the concentration-response curves of GABA (fig. 4).

View larger version (K): [in this window] [in a new window]   Fig. 2.   A, Waglerin-1-induced potentiation of IGABA is not voltage dependent. Current-voltage relations for IGABA of five separate neurons exposed to 10 µM GABA alone () or simultaneously with 32 µM Waglerin-1 (). Peak IGABA was normalized to that recorded at -60 mV in the absence of Waglerin-1. Each data point is the mean ± S.E.M. Waglerin-1 enhanced IGABA at all holding potentials without changing the apparent reversal potential of IGABA. B, Concentration-response relation for Waglerin-1-induced potentiation of IGABA. The normalized peak IGABA in response to 10 µM GABA plus varying concentrations of Waglerin-1 is plotted as a function of the Waglerin-1 concentration. Each data point is the mean ± S.E.M. of four to six neurons. IGABA was recorded at a holding potential of -60 mV.

View larger version (K): [in this window] [in a new window]   Fig. 3.   Waglerin-1 enhancement of IGABA is rapid in onset and offset. Record a presents two currents separated by an interval of 50 sec; the first current is in response to 10 µM GABA alone and the second is in response to GABA plus 32 µM Waglerin-1. Record b was similarly obtained from the same mouse neuron, with the exception that Waglerin-1 was continuously present before and during the second application of GABA. Record c demonstrates that a brief pulse of Waglerin-1 plus GABA enhances IGABA in response to a longer pulse of GABA alone. Holding potential -60 mV.

View larger version (K): [in this window] [in a new window]   Fig. 4.   Waglerin-1 increases the apparent affinity of the GABAA receptor for GABA. A, Peak IGABA was plotted as a function of GABA concentration. The neuron was exposed to either GABA alone () or GABA plus 32 µM Waglerin-1 (). Solid lines are least square fit of the following form of the Michaelis-Menten equation to the experimental data: I = (IMax * Cn)/(Cn + EC50n) where I, IMax, C, EC50 and n are IGABA, maximal IGABA, GABA concentration, the concentration at which IGABA is 50% of maximum and the Hill coefficient, respectively. The EC50 was 45 and 20 µM for GABA alone and GABA plus Waglerin-1, respectively; the corresponding Hill coefficients were 1.7 and 1.6.

Correlation of the effects of Waglerin-1 and benzodiazepine. The above results suggest that the potentiating effect of Waglerin-1 on I_GABA is similar to that of benzodiazepines. To test this hypothesis, we carried out the following two series experiments. First, we correlated the magnitude of the potentiation of I_GABA by Waglerin-1 with that of diazepam for the same population of neurons (fig. 5A). Large potentiation of I_GABA by Waglerin-1 correlated very well with large potentiation by diazepam. Second, we performed competition experiments using flumazenil, an antagonist of the benzodiazepine site of the GABA_AR. Flumazenil antagonized the potentiating effect of Waglerin-1 (fig. 5B). For this series of neurons, 32 µM Waglerin-1 enhanced I_GABA in response to 5 µM GABA to 235.3 ± 27% (n = 4) of control. Flumazenil (25 µM) alone suppressed I_GABA to 75.1 ± 4% (n = 4) of control. When 25 µM flumazenil and 32 µM Waglerin-1 were coapplied with 5 µM GABA, I_GABA was 109.0 ± 6% of control current (n = 4). These two series of experiments suggest that Waglerin-1 and benzodiazepins share a common binding site or act on sites that are functionally related.

View larger version (K): [in this window] [in a new window]   Fig. 5.   Interaction of Waglerin-1 and benzodiazepine. A, IGABA was recorded with 24 µM Waglerin-1 and 1 µM diazepam. The currents were induced by 15 µM GABA. Each point represents data obtained from a single neuron. The regression line (continuous line) has a slope of 0.87 indicating a strong correlation. B, The potentiating effect of 32 µM Waglerin-1 on IGABA (5 µM GABA) is antagonized by 25 µM flumazenil. Bars are the mean + S.E.M. for four separate neurons.

Depression of I_GABA by Waglerin-1. In addition to the potentiating effect observed in the majority of the neurons tested, Waglerin-1 depressed I_GABA for 31% of the neurons examined. Figure 6 illustrates I_GABA in response to 10 µM GABA alone or coapplied with 20 µM Waglerin-1 for one of these neurons. Waglerin-1 reduced both the peak amplitude and the rate of the decay of I_GABA. The records presented in figure 6 also demonstrate that the depressant effect of Waglerin-1 on I_GABA was dependent on voltage. That is, although 20 µM Waglerin-1 decreased peak I_GABA in response to 10 µM GABA when neurons were held at 0 (A) and -40 mV (B), the magnitude of depression was greater at -40 mV. This observation was confirmed in four neurons; i.e., at -40 and 0 mV Waglerin-1 reduced I_GABA to 0.25 ± 0.02 and 0.49 ± 0.04 of the control value, respectively (P < .001, Student's t test). For the records at the right of rows A and B, the GABA alone and the GABA plus Waglerin-1 responses were normalized to the same amplitude and then superimposed. Note that, in contrast to the increased decay that accompanied the potentiation of I_GABA, the depression of I_GABA by Waglerin-1 was associated with decreased decay of I_GABA. Again, this effect of Waglerin-1 was voltage dependent because it was more pronounced at -40 mV than at 0 mV. Such voltage dependence was confirmed in four other neurons. The voltage dependence of the Waglerin-1 depressant effect was also observed when a "ramp" protocol was used to obtain current-voltage relations (fig. 6C-F). The control data reveal outward rectification as indicated by the ratio of I_GABA at +40 mV and -80 mV. Such outward rectification is characteristic of I_GABA recorded from GABARs consisting of 1 and 2 subunits (Verdoorn et al., 1990). These ratios of I_GABA were 1.7 ± .2 and 4.8 ± 1.2 (n = 5) in the absence and presence of Waglerin-1, respectively. Thus, Waglerin-1 significantly increased outward rectification (P < .01; t test). Furthermore, analysis of voltage ramps revealed that although Waglerin-1 depressed I_GABA at all holding potentials, the depressant effect was more pronounced at negative potentials. Waglerin-1 achieved this depressant effect without altering the apparent reversal potential of I_GABA.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

We describe the effects of Waglerin-1, a small peptide purified from the venom of Wagler's pit viper (T. wagleri), on I_GABA of murine hypothalamic neurons. We observed that although Waglerin-1 increased peak I_GABA for the majority of the neurons tested, the peptide suppressed or had no effect on I_GABA of other neurons. This variation was observed for neurons derived from the same animal and examined on the same day under the same experimental conditions. However, in a given neuron, the results were very consistent. Recent molecular biological studies provide direct evidence for the presence of a variety of GABA_ARs composed of various combinations of subunits (Verdoorn et al., 1990). Varying combinations of  and  subunits may occur in different cell populations or even in different cell compartments to yield a mosaic of different oligomers (Khrestchatisky et al., 1989). It has been reported that ,  and  subunits not only allow positive or negative functional cooperativity, but also determine the quality, potency and efficacy of various benzodiazepine interactions with the GABA_AR (Puia et al., 1989). However, Zn⁺⁺ selectively inhibits benzodiazepine-insensitive GABA_AR (Smart et al., 1991; Davies et al., 1993). Thus, heterogeneity of the GABA_AR may be one reason for variation in response to Waglerin-1. For this reason, we suggest that Waglerin-1 is a useful new tool with which to explore structure activity relations of the GABA_AR.

The most frequent effect of Waglerin-1 was to increase peak I_GABA in response to subsaturating GABA concentrations. This potentiating effect was very rapid in onset as well as offset and had an apparent EC₅₀ of 24 µM Waglerin-1. Our concentration-response data suggest that Waglerin-1 increased the affinity of the GABA_AR for agonist without changing the peak response; the apparent EC₅₀ for GABA was 20 and 45 µM in the presence and absence of 32 µM Waglerin-1, respectively. This effect of Waglerin-1 is consistent with preliminary experiments (Ye and McArdle, 1995a) showing that Waglerin-1 increases the frequency of single channel opening without altering conductance.

The potentiation of I_GABA was associated with an increase of the decay of I_GABA. There are several ways in which Waglerin-1 could increase the rate of I_GABA decay. For example, Waglerin-1 may affect the desensitization of the GABA_A receptor, block the open channel or allosterically modulate the channel's opening or closing frequency. Although the actual mechanism(s) whereby Waglerin-1 increases the decay of I_GABA requires further investigation, it is interesting that a decrease in the decay of I_GABA accompanied the depressant effect of Waglerin-1. These observations suggest a working hypothesis to explain the changes of I_GABA decay. That is, the decay rate of I_GABA is proportional to the number of receptors activated. Thus, for those neurons in which Waglerin-1 potentiated I_GABA, the peptide increased the affinity of receptor for the agonist. Consequently, more receptors were activated in response to a given concentration of GABA and, in turn, more receptors entered the desensitized state. However, desensitization was less complete for those neurons in which Waglerin-1 depressed I_GABA. This hypothesis assumes that there is a relationship between receptor affinity and desensitization. In fact, Jones and Westbrook (1995) demonstrated for hippocampal neurons that desensitization during long (5-10 sec) applications of -alanine was less complete than with GABA. Because GABA has a higher affinity than -alanine for GABA receptors, Jones and Westbrook (1995) suggested that the accumulation of receptors in the desensitized state relates to the degree of receptor occupancy. Thus, it is possible that Waglerin-1 changes the affinity of the GABA_AR for GABA and consequently the degree of receptor occupancy and desensitization.

It is useful to compare the potentiating effect of Waglerin-1 on I_GABA with that of benzodiazepines. As with the benzodiazepines, Waglerin-1 potentiates I_GABA without a direct GABAmimetic effect. This observation and the close correlation between the potentiation of I_GABA by diazepam and Waglerin-1 suggests a similarity of their effects. Furthermore, the observation that the benzodiazepine antagonist flumazenil also antagonized the potentiating effect of Waglerin-1 suggests an interaction with the benzodiazepine site/mechanism of responsive GABA_ARs. It is conceivable that flumazenil and Waglerin-1 act at different sites of the GABA_AR/channel complex to simply offset each others effect on I_GABA. However, this seems unlikely because flumazenil had a much greater effect on I_GABA when Waglerin-1 was present. Therefore, we suggest that both flumazenil and Waglerin-1 compete for the benzodiazepine binding site of the GABA_AR. When flumazenil occupies this site, Waglerin-1 can no longer bring about potentiation of I_GABA.

Cloning studies reveal that a large variety of GABA_AR subunits assemble into GABA_AR subtypes having different functional properties. As noted above, it is known that the subunit composition of GABA_ARs is related to a functional distinction between Zn⁺⁺-sensitive and Zn⁺⁺-insensitive subtypes (Draguhn et al., 1990). Specifically, although the  subunit is necessary for the production of a high-affinity benzodiazepine binding site (Pritchett et al., 1989), this subunit also renders the GABA_AR receptor insensitive to Zn⁺⁺ (Draguhn et al., 1990). In this study, we observed that neurons in which Waglerin-1 potentiated I_GABA were sensitive to benzodiazepines. However, neurons in which Waglerin-1 suppressed I_GABA were also sensitive to Zn⁺⁺. These observations suggest that the  subunit may contribute to variations in response to Waglerin-1. Along this line of thought, it is interesting that the concentration of Waglerin-1 required for 50% potentiation or suppression of I_GABA is 24 µM. One interpretation of this observation is that although various forms of the GABA_AR can have the same affinity for Waglerin-1, the subunit composition causes them to react to occupation of the Waglerin-1 site in a different fashion. To test this hypothesis and the role of the  subunit, experiments with recombinant GABA_ARs are necessary.

Subunit composition of the GABA_AR affects the sensitivity to GABA and the Hill number (Sigel et al., 1990). Matthews et al. (1996) have recently reported that GABA_ARs containing the  subunit have a larger EC₅₀ and Hill number for GABA than those without . In accord with this observation, we observed that the EC₅₀ and Hill number of GABA are larger for the group of neurons in which Waglerin-1 potentiated I_GABA (EC₅₀ = 45 µM, n = 1.6) as compared to these values (15 µM, 1.2) for the group in which Waglerin-1 suppressed I_GABA. This correlation further supports our working hypothesis that the  subunit may play a role in determining the response of the GABA_AR to Waglerin-1.

In summary, our data suggest that Waglerin-1 distinguishes between three types of GABA_AR in freshly isolated hypothalamic neurons; i.e., GABA_ARs that are nonresponsive or are either potentiated or inhibited by Waglerin-1. Although the potentiating effect appears to be allosteric and voltage independent, suppression of I_GABA appears to be competitive and voltage dependent. The similarity between the EC₅₀ and IC₅₀ for Waglerin-1 potentiation and inhibition of I_GABA suggests that the peptide has equal potency for the receptor sites involved in either effect. It is reasonable to propose that we may divide the GABARs into groups based on their response to Waglerin-1. Thus, we suggest that the Waglerin-1 potentiating group of GABARs belong to the benzodiazepine sensitive group and the Waglerin-1 suppression group of GABARs belong to the Zn⁺⁺-sensitive group. Further work is required to reveal the structural features of the GABA_A receptor responsible for the differential effects of Waglerin-1.