Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2001 May 1;532(Pt 3):673–684. doi: 10.1111/j.1469-7793.2001.0673e.x

Pregnenolone sulfate block of GABAA receptors: mechanism and involvement of a residue in the M2 region of the α subunit

Gustav Akk 1, John Bracamontes 1, Joe Henry Steinbach 1
PMCID: PMC2278584  PMID: 11313438

Abstract

  1. Neurosteroids are produced in the brain, and can have rapid actions on membrane channels of neurons. Pregnenolone sulfate (PS) is a sulfated neurosteroid which reduces the responses of the γ-aminobutyric acid A (GABAA) receptor. We analysed the actions of PS on single-channel currents from recombinant GABAA receptors formed from α1, β2 and γ2L subunits.

  2. Currents were elicited by a concentration of GABA eliciting a half-maximal response (50 μm) and a saturating concentration (1 mm). PS reduced the duration of clusters of single-channel activity at either concentration of GABA.

  3. PS had no discernable effect on rapid processes: no effects were apparent on channel opening and closing, nor on GABA affinity, and a rapidly recovering desensitised state was not affected. Instead, PS produced a slowly developing block which occurred at a similar rate for receptors with open or closed channels and with one or two bound GABA molecules.

  4. The rate of block was independent of membrane potential, implying that the charged sulfate moiety does not move through the membrane field.

  5. Change in a specific residue near the intracellular end of the channel lining portion of the α1 subunit had a major effect on the rate of block. Mutation of the residue α1 V256S reduced the rate of block by 30-fold. A mutation at the homologous position of the β2 subunit (β2 A252S) had no effect, nor did a complementary mutation in the γ2L subunit (γ2L S266A). It seems likely that this residue is involved in a conformational change underlying block by PS, instead of forming part of the binding site for PS.

Steroids can have rapid effects on mood and behaviour of animals, and on the function of neurons. A number of steroids have been identified that are produced in the brain; these steroids have been termed neurosteroids (Mensah-Nyagan et al. 1999). Pregnenolone sulfate (PS) is a sulfated neurosteroid that is present in brain at a relatively high concentration compared with many other neurosteroids (e.g. Wang et al. 1997). It has been found that the level of PS is reduced in the hippocampus of some aged rats and that there is a correlation between the reduction in PS levels and reduced performance in a behavioural test for memory of a novel place (Vallee et al. 1997), which could be restored by injection of PS (Vallee et al. 1997). Even in young adult rats, injection of PS into the brain can improve performance on multiple tests of memory (Mayo et al. 1993; Flood et al. 1995). The functional role of PS in the brain is not known, but it has been reported to have direct effects on two receptors for neurotransmitters: it blocks activation of γ-aminobutyric acid A (GABAA) receptors (Majewska et al. 1988) and it potentiates activation of N-methyl-d-aspartate (NMDA) receptors (Wu et al. 1991). We are particularly interested in the blocking of GABAA receptors, since many other neurosteroids are known to potentiate the activation of these receptors (Lambert et al. 1995).

Most previous studies of the action of PS and other sulfated steroids (in particular dehydro-epiandrosterone sulfate, DHEAS) on GABAA receptors have utilised studies of macroscopic currents elicited from whole cells (Majewska et al. 1988; Woodward et al. 1992; Zaman et al. 1992; Park-Chung et al. 1999; Shen et al. 1999). These studies have shown that PS can block receptors when applied before GABA (Zaman et al. 1992), indicating that receptors without bound transmitter and with closed channels can be blocked. However, PS or DHEAS block in a non-competitive fashion (Majewska et al. 1988; Majewska et al. 1990; Woodward et al. 1992), indicating that receptors with bound GABA can also be blocked. Experiments using steroids that potentiate GABAA receptor responses have shown that potentiating steroids are equally effective in the presence of PS (Zaman et al. 1992) and that PS or DHEAS is equally effective in the presence of potentiating steroids (Zaman et al. 1992; Park-Chung et al. 1999), indicating that the potentiating and blocking steroids do not have a common binding site or mechanism of action. A single report has been made of the effects of PS on single-channel currents elicited from GABAA receptors (Mienville & Vicini, 1989). In that study, a high concentration of PS (50 μm) had no effect on the duration or amplitude of openings elicited by a low concentration of GABA (1 μm), but did reduce the frequency of openings (by about 25 %). Although these studies have provided insights into the actions of PS, no detailed study has been made of the possible mechanism by which PS has its effects, and several possible mechanisms are consistent with these observations. For example, the channel opening rate could be reduced, a slowly developing block of both open and closed channels could be produced, or desensitisation could be altered (Shen et al. 2000).

We examined the mechanism of action of PS using recordings of single-channel activity from recombinant GABAA receptors. Our first goal was to determine whether PS affected rapid steps in receptor activation, including agonist binding or channel opening and closing. The second was to explore with more precision the possibility that block by PS occurred from some specific state of the receptor (e.g. unliganded, liganded closed, liganded open). The third was to examine, at the single-channel level, whether PS block was affected by potentiation produced by other steroids or by barbiturates. The final goal was to determine residues in the GABAA receptor that are required for the action of PS. The overall objective was to provide additional understanding of the mechanism of action for this inhibitory steroid on the GABAA receptor.

METHODS

Rat GABA receptor cDNA was generously provided by Drs A. Tobin (University of California Los Angeles; α1, β2) and D. Weiss (University of Alabama, Birmingham; γ2L) and subcloned into the expression vector pcDNAIII (Invitrogen Corp., San Diego, CA, USA). Point mutations were produced using QuikChange (Stratagene, San Diego, CA, USA). The mutated subunits were sequenced to confirm that only the desired mutation had been produced.

The methods used are described in Akk & Steinbach (2000). In brief, GABA receptors were expressed in HEK 293 cells using transient transfection based on calcium phosphate precipitation (Ausubel et al. 1992). Electrophysiological experiments were usually performed using the cell-attached patch-clamp configuration (Hamill et al. 1981) although in some cases inside-out patches were used. The interior of the pipette was held at +60 mV unless indicated otherwise. We assume that the cell membrane potential was about -40 mV, thus the total potential difference across the patch was usually about -100 mV. Experiments were performed at 22 °C. Single-channel currents were recorded with an Axopatch 200B amplifier (Axon Instruments, Foster City, CA, USA), low-pass filtered at 10 kHz, acquired with a Digidata 1200 Series Interface at 50 kHz using pCLAMP 7 software (Axon Instruments) and stored on a PC hard drive for further analysis.

All of the analysis was performed on events in clusters of activity, elicited by 50 μm or 1 mm GABA. In the present experiments, most clusters were separated from each other by prolonged periods of inactivity lasting several seconds. Files that contained persistent overlapping activity were not analysed. We assume that during the intercluster silent periods all the receptors in the patch are desensitised (cf. Sakmann et al. 1980). We could not estimate the rate of recovery from desensitisation from these data, since the long closed times will depend on both the intrinsic rate of recovery and the number of receptors in the patch (which is not known). A cluster was defined as a series of openings separated by closed intervals shorter than a critical duration (tcrit). We chose tcrit to be 200-500 ms, which is ∼10 times longer than the mean duration of the longest closed-time constant within clusters, and ∼10 times shorter than the mean interval between clusters.

The cluster duration was defined as the time between the first opening and the last closing transition. We imposed a minimum duration for a cluster to be accepted (tmin). This was done to eliminate the contribution of isolated events (clusters with a single opening or artifacts). The value for tmin ranged from 25 ms (in the presence of 50 μm PS) to 200 ms (in the absence of PS). As a result of the application of a minimal duration, the apparent mean cluster duration will be longer that the true duration. Assuming that the cluster durations are well described by a single exponential distribution (data not shown), the mean cluster duration (T([PS])) is the apparent cluster duration minus tmin. For all data sets except those obtained in the presence of 50 μm PS, tmin < 0.1T([PS]), and so the correction was negligible. Accordingly, the correction was only applied to the data obtained in the presence of 50 μm PS.

Open and closed intervals were measured for events within clusters. To do this, the data in accepted clusters were digitally low-pass filtered at 2-4 kHz and idealised using the segmented-k-means algorithm (program SKM; www.qub.buffalo.edu). Data from each patch was fitted with kinetic schemes incorporating three open (for open duration histograms) or three closed states, respectively, to obtain the mean durations and relative occurrences presented (programme MIL; Qin et al. 1996, 1997). Cluster durations as a function of [PS] were fitted using Nfit (UTMB, Galveston, TX, USA). Parameters are presented as best fitting value ±s.d. of the fit.

All chemicals used were obtained from Sigma (St Louis, MO, USA).

RESULTS

Action of pregnenolone sulfate on clusters of openings

Our studies were made by analysing the properties of ‘clusters’ of openings elicited by relatively high concentrations of agonists. A cluster results from the activity of a single receptor. At the start of a cluster a receptor recovers from the long-lived desensitised state(s) and begins a rapid series of sojourns in open and closed channel states. At the end of a cluster, the receptor re-enters a long-lived desensitised state. During the cluster, it is possible to measure the probability that the channel is open (P open). The distribution of the durations of openings gives insight into the number of kinetically distinct open states and their intrinsic rates of closing. The durations of the closed periods within a cluster provides information about the closed states in rapid equilibrium with the open states. At the agonist concentrations which produce clusters, these closed states involve states with various numbers of ligands bound, and short-lived desensitised states. Finally, the overall duration of the cluster provides information on the rate at which the long-lived desensitised state(s) develop. Accordingly, we reasoned that examination of the effects of PS on the properties of clusters would provide the largest amount of information on the kinetic mechanism for its action.

We used a high concentration of GABA (1 mm) and a lower concentration (50 μm). Clusters of openings are clear at both concentrations (Fig. 1). Activation of the GABAA receptor by GABA is known to have a relatively complicated mechanism (Macdonald et al. 1989; Weiss & Magleby, 1989; Twyman et al. 1990), which we have confirmed (G. Akk & J.H. Steinbach, in preparation). A GABA concentration of 1 mm appears to saturate activation, producing the maximal P open of about 0.8, so that the channel is about 4 times more likely to be open than closed. A GABA concentration of 50 μm produces about a half-maximal P open (0.4) and, accordingly, the receptor has a closed channel for more than half the time during the cluster. The clusters at 1 mm and 50 μm are very similar in duration (Fig. 1 and 2). Since termination of the cluster under control conditions is the result of entry into a long-lived desensitised state, this observation indicates that long-lived desensitisation can arise from both partially and fully liganded states.

Figure 1. PS has similar effects on clusters at two GABA concentrations.

Figure 1

Open in a new tab

Single clusters elicited by 50 μm or 1 mm GABA are shown in the top row (the current through an open channel is downwards). The middle row shows typical clusters in the presence of 10 μm PS, while the bottom row shows two clusters in the presence of 50 μm PS. Note that the clusters are of similar duration at the two GABA concentrations, and are reduced similarly by increasing concentrations of PS. The probability of being open is about twice as high in the presence of 1 mm GABA compared with 50 μm, as can be seen in the top and middle traces. Data recorded cell attached with a membrane potential of about -100 mV (see Methods).

Figure 2. Dependence on [PS] of the reduction in cluster duration and the cluster P open.

Figure 2

Open in a new tab

A, mean cluster duration (T([PS])) for data from individual patches plotted against [PS] (note logarithmic abscissal co-ordinates; the points on the left are for GABA with no PS). The data for activity elicited by 50 μm and 1 mm GABA overlap at all [PS]. The lines show the fits of eqn (2). When eqn (1) is fitted to the data, the best-fitting parameter estimates are: 50 μm GABA, Tmax 2652 ± 1158 ms, nH -0.8 ± 0.2, IC50 0.9 ± 1.1 μm; 1 mm GABA, Tmax 2149 ± 362 ms, nH -1.0 ± 0.1, IC50 2.1 ± 1.0 μm. For eqn (2), the values are k+D 0.52 ± 0.11 s−1, k+PS 0.21 ± 0.04 μm−1 s−1 (50 μm) and k+D 0.48 ± 0.06 s−1, k+PS 0.20 ± 0.03 μm−1 s−1 (1 mm). B shows there is no change in P open over the same range of [PS]. Each point shows data from a single patch.

We then made recordings in the presence of different concentrations of PS. As shown in Fig. 1, PS reduced the durations of the clusters elicited by either 50 μm or 1 mm GABA. The combined data show no differences in the concentration effect curve for PS-induced shortening for clusters elicited at the two GABA concentrations (Fig. 2).

The data shown in Fig. 2 were fitted with two simple equations. Equation 1 was the Hill equation,

graphic file with name tjp0532-0673-m1.jpg (1)

where T([PS]) is the mean cluster duration in the presence of a given concentration of PS, Tmax is the maximal (control) mean cluster duration, IC50 is the concentration of PS producing half-maximal reduction in T([PS]), and nH is the Hill coefficient. Equation (1) does not imply any particular mechanism, but can be used to describe the shape of the curves. The parameter estimates for the data at the two GABA concentrations agree well (Fig. 2).

The second equation was derived from a very simple picture of block by PS. In the control condition, the mean duration of a cluster is given by the inverse of the overall rate of entering long-lived desensitised states, T([0]) = 1/(k+D). In the presence of PS, it is assumed that there is an additional mechanism for terminating a cluster, that in the simplest case has a rate which is directly proportional to the concentration of PS. Hence, in the presence of PS the total rate for leaving a cluster will be the sum (k+D+k+PS[PS]), where k+PS is the association rate constant for the hypothesised blocking step. Following this idea, the second equation used was

graphic file with name tjp0532-0673-m2.jpg (2)

As the lines in Fig. 2 show, eqn (2) can describe the data well, and k+PS is very similar for clusters elicited by the two concentrations of GABA (Fig. 2). We will use the parameter k+PS to characterise the action of PS in the rest of this paper, keeping in mind that we do not know that block is actually described by a first-order reaction. Equation (2) is appropriate to describe the concentration dependence for the rate of development of block produced by an occupancy mechanism (e.g. direct occlusion of the channel). In addition, it describes the rate of development for other mechanisms of block so long as the concentration of PS is low compared with its affinity at the binding site mediating block. For example, if binding and dissociation were rapid and block were produced by a slower conformational change, the rate of development of block would increase linearly with [PS] at low concentrations. Thus, k+PS is a suitable parameter to describe block by PS by a variety of mechanisms, although it should not be interpreted as a true first-order association rate constant. It is straightforward to estimate k+PS, since it can be obtained simply by measuring the mean cluster duration at a concentration of PS that greatly reduces the cluster duration. In this case, k+PS[PS] ≫ k+D, and k+PS can be calculated directly from the mean cluster duration and the concentration of PS (k+PS= 1/{(T([PS]))[PS]}).

The results using 50 μm and 1 mm GABA show that PS blocks equally rapidly at half-maximal and maximal concentrations of GABA, confirming macroscopic measurements showing non-competitive block by PS. In addition, the single-channel data suggest that PS can block essentially equally well when the receptor has one or two GABA molecules bound to it. At a high concentration of PS, the cluster duration reflects the overall rate of block for receptors in all states within a cluster. The overall rate can be approximately separated into block of receptors with two bound GABA molecules compared with other states of ligation, in the following way. If F2 is the fraction of time spent with two bound GABA molecules, ρ is the rate of blocking receptors with two bound GABA molecules and Q is the relative rate for blocking receptors with less than two bound GABA, then the aggregate rate for blocking a cluster would be (F2ρ+ (1 - F2)Qρ). In applying this approach, multiple states of the receptor (open channel, closed channel, desensitised) with the same degree of ligation will be lumped together. When 1 mm GABA is used to activate, the receptor is doubly-liganded essentially throughout the cluster (see below), so F2 = 1. On the other hand, when 50 μm GABA is used the channel is open about 40 % of the time and closed 60 %. If we assume that half of the time when the channel is closed at 50 μm GABA it is doubly liganded (and closed or desensitised), then F2 = 0.7. The mean cluster duration at 50 μm PS did not differ for 50 μm and 1 mm GABA (Fig. 2), so there is less than a 2-fold difference in cluster durations. With these assumptions, the rate for PS block must differ by less than 4-fold for receptors with two GABA molecules bound compared with those with fewer.

An analogous calculation can be made to examine the rate of block of receptors with open or closed channels. If Fc is the fraction of time spent closed, ρ is the rate of blocking receptors with closed channels and Q is the relative rate for blocking open channels, then the aggregate rate for blocking a cluster would be (Fcρ+ (1 - Fc)Qρ). As an example, if the rate of block were 10-fold higher for open channels (Q = 10), then the mean cluster duration should have been only half as long with 1 mm GABA (where Fc is 0.2) than at 50 μm GABA (where Fc is 0.6). This analysis indicates, therefore, that the difference in rates is less than 10-fold between receptors with open and closed channels.

We used an additional approach to examine the question of selectivity between open and closed channel states. Piperidine-4-sulfonic acid (P4S) has been described as an agonist at the GABAA receptor that binds with high affinity but has a low maximal response (Woodward et al. 1993; Ebert et al. 1994). This suggested to us that P4S has low efficacy for opening the channel, so that we could obtain clusters of activity reflecting fully liganded receptors which spend most of the time with closed channels (a low maximal P open). This appears to be the case, as clusters produced by 1 mm P4S are very sparse (Fig. 3), with an estimated P open of about 0.1. In the presence of 50 μm PS the cluster duration was clearly reduced, with a mean duration indistinguishable from those produced by GABA in the presence of 50 μm PS (Table 1). Given the P open values for 1 mm GABA and 1 mm P4S, the qualitative analysis described above indicates that the difference in blocking rates for receptors with open or closed channels is less than 3-fold. In sum, the use of a low efficacy agonist also gives no indication that the di-liganded closed state is blocked any differently than the di-liganded open state.

Figure 3. PS reduces the durations of clusters elicited by a low efficacy agonist.

Figure 3

Open in a new tab

A single cluster elicited by 1 mm piperidine-4-sulfonic acid (P4S) is shown in the upper row, while the lower row shows typical clusters elicited by 1 mm P4S in the presence of 50 μm PS. The probability of being open in clusters elicited by P4S is very low (compare with Fig. 1), but PS reduces the cluster duration. Data recorded cell attached with a membrane potential of about -100 mV (see Methods).

Table 1.

Mean cluster duration and forward rate constant with 50 μM PS

Condition T([0]) (ms) N clusters (N patches) T([50 μM]) (ms) k+PS (μM−1 S−1) N clusters (N patches)
1 mM GABA 2881 ± 1978 33 (2) 101 ± 80 0.20 135 (2)
50 μM GABA 3586 ± 3150 25 (2) 81 ± 78 0.25  46 (2)
1 mM P4S n.d. — 128 ± 96 0.16 115 (3)
50 μM GABA + 40 μM PB 6701 ± 5716  9 (1) 100 ± 81 0.20 131 (4)
50 μM GABA + 1 μM CAN 10141 ± 10639*  5 (1) 84 ± 60 0.24  89 (2)
Open in a new tab

The mean cluster durations (± S.D.) in the absence of PS (T([0])) and in the presence of 50 μM PS T([50 μM]) are shown, with the calculated value for k+PS. P4S, piperidine−4-sulfonic acid; PB, pentobarbitone. The number of clusters analysed in each condition is shown (from N patches). Note that for a single exponential distribution the mean and standard deviation of the durations should be equal.

*

Data for T([0]) in the presence of ACN were obtained with 0.5 μM ACN. n.d., not determined.

Lack of voltage dependence of block by PS

PS is an anion at physiological pH. We determined whether the charge on PS interacts with the membrane field during the blocking action by measuring the mean cluster duration at different membrane potentials in the presence of 1 mm GABA and 50 μm PS. There was no discernable consequence of changing the membrane potential by 65 mV (data not shown; the regression slope gave a voltage dependence of e-fold per 4125 mV). The implication of this low slope is that the single negative charge on PS must traverse a very small fraction of the membrane field (Woodhull, 1973), and the sign of the dependence is the opposite expected for movement of an anion from the external medium into the membrane field. Accordingly, it is unlikely that the action of PS involves interaction of the sulfate group with a residue that is very far into the membrane field.

Actions of PS on rapid processes within the cluster

PS did not affect the P open within a cluster (Fig. 2). This observation suggests that PS has relatively little effect on the rapid processes involved in receptor activation, although it could be that equal changes were made in both the open and closed durations (for example, both could be prolonged).

However, PS did not alter the open times within a cluster (Fig. 4). The histograms of open-time durations contain three components in control conditions (Fig. 4), which reflect three distinct open states. Such complexity has been reported before (Macdonald et al. 1989; Weiss & Magleby, 1989; Twyman et al. 1990), and we will not propose a specific kinetic scheme to accommodate the observations. However, PS did not affect the number of open-time components, the fitted time constant of any of the three components, or the relative prevalence of the components (Fig. 5). The values of the three open-time constants and the three closed-time constants in the absence and presence of 50 μm PS are given in Fig. 4. Accordingly, there is no indication that PS acts preferentially on a particular open state. Furthermore, PS did not appear to increase the rate of channel closing, nor to produce a rapid open-channel block. The absence of any systematic changes in the mean open duration in the absence and presence of various concentrations of PS is shown in Fig. 5. For a single patch recorded with 50 μm GABA and no PS, the mean open duration was 3.7 ms and the mean closed duration was 6.9 ms. For two patches recorded with 50 μm GABA + 50 μm PS, the mean open durations were 1.5 and 1.9 ms, while the mean closed times were 2.4 and 2.9 ms. For a single patch recorded in the presence of 1 mm GABA, the mean open duration was 2.3 ms and the mean closed duration was 0.7 ms. For one patch obtained in the presence of 1 mm GABA + 50 μm PS, the mean open time was 3.5 ms and the mean closed time was 1.1 ms.

Figure 4. PS does not change the histograms of open or closed times within clusters.

Figure 4

Open in a new tab

Histograms of the open durations (first and second columns) and closed durations (third and fourth columns) are shown for clusters elicited by 50 μm (a) or 1 mm GABA (b) in the absence (first and third columns) or presence (second and fourth columns) of 50 μm PS. Each histogram shows data from a single patch. In each case the data have been fitted with the sum of 3 exponential components (time constants given in the panels), shown by the continuous lines. The collected data on the mean durations and fractional areas of the 3 components are shown in Fig. 5. Note that the histograms are displayed with the square root of the number of counts in each bin, and bins are logarithmically scaled.

Figure 5. PS does not change the distributions of open or closed times within clusters.

Figure 5

Open in a new tab

The mean durations and relative areas of the components fitted to the duration histograms are shown for clusters elicited by 50 μm GABA (A) and 1 mm GABA (b) at a range of PS concentrations. The data from a patch (1000-20 000 sojourns) were well fitted with the sum of 3 exponential components (identified as O1, O2, O3 and C1, C2, C3, respectively) with time constants separated 3- to 10-fold. The plots show results from individual patches plotted against [PS]; note that the points on the left in each panel are for patches with no PS. The lines in panels showing mean durations denote the weighted average durations of all 3 components of the respective histograms. The plots in the first column show the mean open times (in ms) for each component while the plots in the second column show the fraction of the total open times in each component. Note that PS has no effect on the durations or prevalence of open-time components nor the weighted average duration. There is also no apparent difference between open-time distributions in the presence of 50 μm and 1 mm GABA. The third and fourth columns show the data for closed times in the same format: the mean closed time for each component (third column), then the fraction of the total closed times in each component (fourth column). In contrast to the data on open times, there is a change in the distributions of closed times between the two GABA concentrations, most clearly seen in the fractional representation in each component (fourth column; compare A with B). However, again PS has no effect on the duration or prevalence of the closed-time components nor on the weighted average duration.

The closed periods are informative because they can provide some information on the agonist binding rate, the channel opening rate and the rates for rapidly recovering desensitisation. Again, three components were present for closed periods within a cluster at either GABA concentration, and PS did not affect the mean durations or relative amplitudes for the closed periods at either GABA concentration. The lack of effect on the histograms demonstrates that PS has little effect on the channel opening rate, GABA association or dissociation rates, or a rapidly equilibrating desensitised state. A quantitative demonstration of this lack of effect will require a full model for GABAA receptor activation, which is not presently available. However, these conclusions are supported by inspection of the histograms. In particular, we used a high concentration of GABA (1 mm) to examine the channel opening rate and a rapidly equilibrating desensitised state. In contrast, the closed-time histograms for data obtained using a concentration of GABA producing a half-maximal value for P open will contain information about GABA association and dissociation rates.

In qualitative terms, one of the two closed-time components (C1 and C2 in Fig. 5) in the presence of 1 mm GABA reflects the channel opening process (durations of 0.25 or 1 ms imply a channel opening rate of about 1000-4000 s−1). There is no effect of PS on the duration or prevalence of either component. Accordingly, we find no evidence that PS acts by reducing the channel opening rate. There is also a much longer component in the closed-time histograms at 1 mm GABA (identified as C3 in Fig. 5), which is related to a short-lived desensitised state (see Jones & Westbrook, 1995). This component has a mean duration of 20-50 ms and constitutes about 2 % of the closed periods within a cluster. If PS produced its block by increasing the rate of occurrence of this type of desensitisation, we would have expected to see an increase in its prevalence. Alternatively, if PS increased the duration of this desensitised state we would expect to see an increase in the duration of this component with increased [PS]. Since there was no change in the presence of PS, we find no evidence that PS acts by affecting this component of desensitisation.

The closed-time histograms obtained with 50 μm GABA are harder to interpret. However, PS has no effect on the distributions (Fig. 4 and 5). This concentration of GABA produces a Popen which is about half-maximal. At this concentration the closed times reflect both channel opening and GABA association and dissociation, as indicated most clearly by the observation that the apparent opening rate for channels is far from maximal and continues to increase with GABA concentration (e.g. Maconochie et al. 1994). The fact that no changes are observed in the closed-time histograms at a half-maximally effective concentration of GABA suggests that PS does not exert its major effect by reducing the association rate for GABA to the second site on the receptor, or by increasing the dissociation rate for GABA from this site.

These observations, taken together, indicate that PS does not have major effects on the channel opening rate, closing rate, nor on GABA binding or dissociation. Accordingly, the reduction of response by PS is the result of a slow process which shortens clusters of activity. We are unable to determine, from the present results, whether PS acts by increasing the rate of a slow desensitisation process or by an independent mechanism.

Interaction of PS with potentiators of GABAA receptors

Barbiturates and many steroids act at low concentrations to potentiate activation of GABAA receptors. The mechanism of potentiation is not fully understood, although barbiturates and steroids appear to act after binding to sites which differ between the two classes of drugs, as well as from the sites which bind GABA, benzodiazepines or convulsants (e.g. Amin & Weiss, 1993; Lambert et al. 1995; Ueno et al. 1997). We used pentobarbital and 3α,5α,17β-3-hydroxyandrostane-17-carbonitrile (ACN) as representatives of these two classes of potentiating drugs.

There was no change in k+PS for PS when 50 μm GABA and 40 μm pentobarbital, or 50 μm GABA and 1 μm ACN were co-applied (Table 1). These results, therefore, suggest that the mechanism by which PS reduces cluster duration is independent of potentiation by barbiturates or potentiating steroids.

The reduction of cluster duration by PS is dependent on a particular residue in the α1 subunit

The ability of some non-competitive blocking drugs to produce their effects is critically dependent on particular residues in the second membrane-spanning region of receptor subunits. This was first shown with non-competitive blocking agents for the muscle nicotinic receptor (Charnet et al. 1990). The actions of some positively charged local anaesthetics depended on the nature of the residues at the 6th (6′) and 10th (10′) positions in the aligned M2 regions of all four subunits. For convenience, the position in the M2 region will be indicated by a primed number, in which the 1st (1′) position is the closest to the N-terminus of the protein and also closest to the cytoplasmic end of the M2 helix. Studies of GABAA receptors have identified the 6′ residue as a critical residue for the blocking action of picrotoxin (Gurley et al. 1995), but additional studies have found that a mutation of the 6′ residue of the GABAA γ2 subunit which removes block by picrotoxin had no effect on the ability of PS to block responses (Shen et al. 1999).

Work with the GABA-activated rdl (Ffrench-Constant et al. 1993) and ρ (Wang et al. 1995) receptors had indicated that the 2′ position can also have significant effects on picrotoxin block. A mutation of the rdl subunit which converted an alanine at the 2′ position to serine (A2′S) reduced block by picrotoxin (Ffrench-Constant et al. 1993). Accordingly, we set out to determine whether the nature of the residue at the 2′ position was critical for PS action. The GABAA α1 subunit has valine (V256) at the 2′ position, the β2 subunit has alanine (A252), while the γ2 subunit has serine (S266). We produced the point mutants α1(V2′S), β2(A2′S) and γ2(S2′A).

In all cases, the receptors containing mutated subunits had essentially normal single-channel amplitudes, mean open times and Popen values (data not shown), indicating that the mutations did not have major effects on these aspects of receptor function.

Cotransfection of α1(V2′S)β2γ2 subunits resulted in expression of receptors with altered block by PS. The cluster duration was prolonged. In many cases the extent of overlap of activity was such that it was not possible to estimate the cluster duration accurately. In one patch with very low activity the mean cluster duration was about 11 s. When recordings were made in the presence of 50 μm PS the cluster duration was reduced enough to allow clear resolution of clusters, which had a mean duration significantly longer than control (Fig. 6, Table 2). The apparent association rate (k+PS) for PS was reduced by about 30-fold by the mutation. This indicates that the presence of serine at the 2′ position of the α1 and γ2 subunits affects the ability of PS to block. It is important to note that clusters which showed overlapping activity (the occurrence of periods when two or more channels were open at the same time) were eliminated in calculating the mean duration. Since longer clusters have a greater likelihood of participating in an overlap, the measured mean durations for clusters from these receptors are actually a lower estimate of the mean duration (and so the estimate for k+PS also is a lower limit).

Figure 6. Mutation in the α subunit reduces the ability of PS to shorten cluster duration.

Figure 6

Open in a new tab

Representative clusters elicited by 1 mm GABA in the presence of 50 μm PS are shown. Note that the clusters in the presence of PS are much longer when the receptor contains the mutated α1(V2′S) subunit, while the mutated β2(A2′S) or γ2(S2′A) subunits have no effect by themselves, nor do they alter the consequences of incorporating the α1(V2′S) subunit. The bottom trace has a different time scale from the others

Table 2.

Mean cluster duration and forward rate constant for wild-type and mutated subunits

Receptor T([0]) (ms) N clusters (N patches) T([50 μM]) (ms) k+PS (μM−1 s−1) N clusters (N patches)
α1β2γ2 2881 ± 1978 33 (2) 81 ± 78 0.25  46 (2)
α1(V2'S)β2γ2 n.d. — 2913 ± 2502 0.007  38 (7)
α1β2(A2'S)γ2 5396 ± 2997 12 (2) 68 ± 50 0.29 133 (5)
α1β2γ2(S2'A) 3417 ± 242 17 (2) 97 ± 76 0.21  80 (3)
α1(V2'S)β2γ2(S2'A) 3916 ± 2977 23 (3) 7240 ± 1678 0.003  5 (2)
α1(V2'S)β2(A2'S)γ2 n.d. — 5140 ± 7600 0.004  10 (3)
Open in a new tab

Data displayed as in Table 1.

This action on PS block was quite specific to the mutation in the α1 subunit. Cotransfection of α1β2(A2′S)γ2 subunits resulted in expression of receptors that had somewhat prolonged mean cluster durations (about 6 s, compared with control values of about 3 s). When clusters were examined in the presence of 50 μm PS, however, the cluster duration was reduced to the same extent as for wild-type receptors and k+PS was unchanged (Fig. 6, Table 2). Clearly, the presence of serine at the 2′ position of the β2 and γ2 subunits does not affect the ability of PS to block.

Cotransfection of α1β2γ2(S2′A) subunits resulted in expression of receptors with normal cluster durations (about 3 s), and the action of PS was not affected (Fig. 6, Table 2). Hence, removal of all serines at the 2′ position did not affect PS block.

Cotransfection of α1(V2′S)β2(A2′S)γ2 subunits or α1(V2′S)β2γ2(S2′A) subunits produced receptors with properties very similar to those seen following transfection with α1(V2′S)β2γ2 subunits (Fig. 6, Table 2). These observations demonstrate that the effect of the serine residue in the α1 subunit is independent of the presence (or absence) of homologous serines in the β2 or γ2 subunits.

DISCUSSION

Mechanism of block by PS

Our results indicate that PS produces block of GABAA receptors by a slow process, with no discernable effects on rapid processes of GABA binding and unbinding or channel opening and closing. The block also does not appear to involve change in a rapid form of desensitisation.

The apparent forward rate constant, k+PS, is about 2 × 105m−1 s−1, a slow value for an open channel blocker (cf. Charnet et al. 1990). We could not estimate the dissociation rate, but recovery times in whole-cell experiments suggest that the dissociation rate is ∼3 × 10−2 s−1 (Woodward et al. 1992). These values generate a value for the apparent IC50 for steady-state block of ∼100 nm. This value is lower than estimates in the literature obtained with recombinant α1β2γ2 receptors (e.g. 1.2 μm; Shen et al. 1999), but it is not certain that the steady-state level of block had been reached at the lower concentrations of PS tested in the whole-cell recordings. The total content of PS in rat brain is estimated to be 10-100 nmol (kg wet weight of tissue)−1 (Robel et al. 1987; Corpechot et al. 1997; Wang et al. 1997), which would suggest that some steady-state block might be present. However, it is not known what the effective concentration of PS at the GABAA receptor in the brain would be, so the significance of the comparison is not clear.

Block appears to develop at the same rate independent of whether the channel is open or closed, and of whether the receptor has one or two bound GABA molecules. These observations confirm previous studies of macroscopic currents, which had found that PS can block GABAA receptors in the absence of agonist (Zaman et al. 1992) and that block is not competitive with GABA (Majewska et al. 1988, 1990; Woodward et al. 1992). However, these previous results were all obtained using steady-state measurements of block. We have extended the previous observations using studies of clusters of single-channel events which allowed us to identify the distinct states of the receptor with greater precision, and using conditions (GABA concentration or P4S as agonist) that altered the probability a channel is open in a cluster.

At a microscopic level clusters are shortened to the same mean duration irrespective of whether pentobarbital or ACN is present as well as PS. These observations at the single-channel level agree with previous studies of evoked whole-cell currents, which had found that blocking steroids act independently of potentiating steroids (Zaman et al. 1992; Park-Chung et al. 1999). All of these results imply that there is no significant overlap between the binding sites occupied by potentiating and blocking steroids, in agreement with results from studies of binding interactions (e.g. Gee et al. 1989).

Structural basis for block by PS

In general, it is felt that the sulfate moiety (or another negatively charged group) is critical in producing a steroid that blocks rather than potentiates GABAA receptors (e.g. Park-Chung et al. 1999). This conclusion, however, must be tempered by the observations that dehydroepiandrosterone has been reported to be essentially as potent at blocking responses as its sulfated derivative (Demirgoren et al. 1991; LeFoll et al. 1997), and that some sulfated steroids are weak blocking agents (El-Etr et al. 1998). In any case, the idea that an anionic group was critical had suggested to us that the sulfate might actually interact with residues forming the binding site which mediated block. We found that the rate of development of block is independent of the membrane potential, which suggests that the charged sulfate moiety does not interact significantly with the membrane field as PS approaches the transition state between unbound and unblocked to bound and blocked (Woodhull, 1973). It has already been reported that equilibrium block by PS is independent of membrane potential (Majewska et al. 1988), so the reverse step is also voltage independent. Accordingly, the lack of voltage dependence renders it highly unlikely that the sulfate moiety interacts with a site deep within the channel.

It was a surprise, therefore, to find that a residue deep in the channel at the 2′ position in the M2 region of the α1 subunit is critical for block by PS. In thinking about the role of this residue, we considered whether it was likely that it formed part of a binding site for PS. It seems likely that a large, rigid molecule such as a steroid, if it bound in the channel, would bind end-on rather than with its long axis across the channel. In this case, either the A ring (where the sulfate moiety is attached) or the D ring (the other end of the molecule) would penetrate most deeply. The lack of voltage dependence indicates that the A ring is unlikely to penetrate to the 2′ position. However, the structure of the D ring of PS is identical to that for many potentiating pregnane steroids. Accordingly, we think it is less likely that other (uncharged) portions of the PS molecule interact with the 2′ residue. A more circumstantial argument is the finding that the presence of a serine residue is effective at altering PS block only when in the α1 subunit, not when in the β2 or γ2 subunit. Studies of the 2′ residue in nicotinic receptors have led to the suggestion that the side-chains from all five subunits form a ring in the channel (Villarroel et al. 1992), and it might be expected from symmetry arguments that there would be closer equivalence in the subunits for contributions to a binding site. Studies of the effects of mutations of the 6′ and 10′ residues in muscle nicotinic receptor subunits have also indicated that the residues from the different subunits participate relatively equally in forming a binding site for local anaesthetics (Charnet et al. 1990).

Accordingly, we think that it is likely that the specific mutation we have found to affect block by PS, GABAA α1(V256S), exerts its actions by altering an allosteric mechanism rather than by directly altering a binding site. The studies of local anaesthetic block of nicotinic receptors (Charnet et al. 1990) have clearly shown that residues in M2 can participate in binding sites. However, it is important to note that several reports have found that mutations in M2 also affect block by allosteric mechanisms ranging from converting block by dihydro-β-erythroidine in nicotinic α7 receptors to activation (Devillers-Thiery et al. 1992), to altering block by pseudocompetitive blocking agents in GABA- and glycine-activated receptors (Wang et al. 1995; Steinbach et al. 2000).

Conclusions

The question remains, however, as to what the actual mechanism is for block by PS. We have demonstrated that it is not the result of an effect on GABA binding or channel opening or closing rates. Two alternatives are that PS acts to produce a novel change in channel structure, or acts to enhance a normal mode of behaviour, particularly to increase the rate of development of slow desensitisation (as has been proposed previously; Shen et al. 2000). An observation that might support this conclusion is that receptors containing the mutated α1 subunit have prolonged cluster durations as well as reduced k+PS. However, it seems unlikely that a simple multiplicative increase in the rate of desensitisation by PS can explain all the observations. For example, the observation that PS blocks receptors in the absence of GABA as well as in its presence is difficult to accommodate in a simple version of this idea. In addition, some of our observations are problematic. The α1(V2′S) mutation's effect of reducing k+PS (30-fold) is greater than its effect of reducing k+D (∼5-fold), and other conditions that prolonged clusters (presence of pentobarbitone or ACN) did not affect k+PS. Thus, although our studies indicate that PS blocks by an allosteric mechanism, further work will be required to identify the molecular mechanism by which PS acts.

Acknowledgments

This research was supported by grant P01 GM47969 to J.H.S. J.H.S is the Russell and Mary Shelden Professor of Anesthesiology, and G.A. is a fellow of the McDonnell Center for Cellular and Molecular Neurobiology.

References

  1. Akk G, Steinbach JH. Activation and block of recombinant GABAA receptors by pentobarbitone: a single-channel study. British Journal of Pharmacology. 2000;130:249–258. doi: 10.1038/sj.bjp.0703335. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Amin J, Weiss DS. GABAA receptor needs two homologous domains of the β-subunit for activation by GABA but not by pentobarbital. Nature. 1993;366:565–569. doi: 10.1038/366565a0. [DOI] [PubMed] [Google Scholar]
  3. Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K. Short Protocols in Molecular Biology. 2. New York: John Wiley; 1992. [Google Scholar]
  4. Charnet P, Labarca C, Leonard RJ, Vogelaar NJ, Czyzyk L, Gouin A, Davidson N, Lester HA. An open-channel blocker interacts with adjacent turns of α-helices in the nicotinic acetylcholine receptor. Neuron. 1990;4:87–95. doi: 10.1016/0896-6273(90)90445-l. [DOI] [PubMed] [Google Scholar]
  5. Corpechot C, Collins BE, Carey MP, Tsouros A, Robel P, Fry JP. Brain neurosteroids during the mouse oestrous cycle. Brain Research. 1997;766:276–280. doi: 10.1016/s0006-8993(97)00749-x. [DOI] [PubMed] [Google Scholar]
  6. Demirgoren S, Majewska MD, Spivak CE, London ED. Receptor binding and electrophysiological effects for dehydroepiandrosterone sulfate, an antagonist of the GABAA receptor. Neuroscience. 1991;45:127–135. doi: 10.1016/0306-4522(91)90109-2. [DOI] [PubMed] [Google Scholar]
  7. Devillers-Thiery A, Galzi JL, Bertrand S, Changeux JP, Bertrand D. Stratified organization of the nicotinic acetylcholine receptor channel. NeuroReport. 1992;3:1001–1004. doi: 10.1097/00001756-199211000-00014. [DOI] [PubMed] [Google Scholar]
  8. Ebert B, Wafford KA, Whiting PJ, Krogsgaard-Larsen P, Kemp JA. Molecular pharmacology of γ-aminobutyric acid type A receptor agonists and partial agonists in oocytes injected with different α, β, and γ receptor subunit combinations. Molecular Pharmacology. 1994;46:957–963. [PubMed] [Google Scholar]
  9. El-Etr M, Akwa Y, Robel P, Baulieu EE. Opposing effects of different steroid sulfates on GABA receptor-mediated chloride uptake. Brain Research. 1998;790:334–338. doi: 10.1016/s0006-8993(98)00145-0. [DOI] [PubMed] [Google Scholar]
  10. Ffrench-Constant RH, Rocheleau TA, Steichen JC, Chalmers AE. A point mutation in a Drosophila GABA receptor confers insecticide resistance. Nature. 1993;363:449–451. doi: 10.1038/363449a0. [DOI] [PubMed] [Google Scholar]
  11. Flood JF, Morley JE, Roberts E. Pregnenolone sulfate enhances post-training memory processes when injected in very low doses into limbic system structures: the amygdala is by far the most sensitive. Proceedings of the National Academy of Sciences of the USA. 1995;92:10806–10810. doi: 10.1073/pnas.92.23.10806. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Gee KW, Joy DS, Belelli D. Complex interactions between pregnenolone sulfate and the t-butylbicyclophosphorothionate-labeled chloride ionophore in rat brain. Brain Research. 1989;482:169–173. doi: 10.1016/0006-8993(89)90556-8. [DOI] [PubMed] [Google Scholar]
  13. Gurley D, Amin J, Ross PC, Weiss DS, White G. Point mutations in the M2 region of the α, β, or γ subunit of the GABAA channel that abolish block by picrotoxin. Receptors and Channels. 1995;3:13–20. [PubMed] [Google Scholar]
  14. Hamill OP, Marty A, Neher E, Sakmann B, Sigworth FJ. Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches. Pflügers Archiv. 1981;391:85–100. doi: 10.1007/BF00656997. [DOI] [PubMed] [Google Scholar]
  15. Jones MV, Westbrook GL. Desensitized states prolong GABAA channel responses to brief agonist pulses. Neuron. 1995;15:181–191. doi: 10.1016/0896-6273(95)90075-6. [DOI] [PubMed] [Google Scholar]
  16. Lambert JJ, Belelli D, Hill-Venning C, Peters JA. Neurosteroids and GABAA receptor function (Review) Trends in Pharmacological Sciences. 1995;16:295–303. doi: 10.1016/s0165-6147(00)89058-6. [DOI] [PubMed] [Google Scholar]
  17. Lefoll F, Louiset E, Castel H, Vaudry H, Cazin L. Electrophysiological effects of various neuroactive steroids on the GABA receptor in pituitary melanotrope cells. European Journal of Pharmacology. 1997;331:303–311. doi: 10.1016/s0014-2999(97)01042-x. [DOI] [PubMed] [Google Scholar]
  18. Macdonald RL, Rogers CJ, Twyman RE. Kinetic properties of the GABAA receptor main conductance state of mouse spinal cord neurones in culture. Journal of Physiology. 1989;410:479–499. doi: 10.1113/jphysiol.1989.sp017545. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Maconochie DJ, Zempel JM, Steinbach JH. How quickly can GABAA receptors open? Neuron. 1994;12:61–71. doi: 10.1016/0896-6273(94)90152-x. [DOI] [PubMed] [Google Scholar]
  20. Majewska MD, Demirgoren S, Spivak CE, London ED. The neurosteroid dehydroepiandrosterone sulfate is an allosteric antagonist of the GABAA receptor. Brain Research. 1990;526:143–146. doi: 10.1016/0006-8993(90)90261-9. [DOI] [PubMed] [Google Scholar]
  21. Majewska MD, Mienville JM, Vicini S. Neurosteroid pregnenolone sulfate antagonizes electrophysiological responses to GABA in neurons. Neuroscience Letters. 1988;90:279–284. doi: 10.1016/0304-3940(88)90202-9. [DOI] [PubMed] [Google Scholar]
  22. Mayo W, Dellu F, Robel P, Cherkaoui J, Le Moal M, Baulieu EE, Simon H. Infusion of neurosteroids into the nucleus basalis magnocellularis affects cognitive processes in the rat. Brain Research. 1993;607:324–328. doi: 10.1016/0006-8993(93)91524-v. [DOI] [PubMed] [Google Scholar]
  23. Mensah-Nyagan AG, Do-Rego JL, Beaujean D, Luu-The V, Pelletier G, Vaudry H. Neurosteroids: expression of steroidogenic enzymes and regulation of steroid biosynthesis in the central nervous system (Review) Pharmacological Reviews. 1999;51:63–81. [PubMed] [Google Scholar]
  24. Mienville JM, Vicini S. Pregnenolone sulfate antagonizes GABAA receptor-mediated currents via a reduction of channel opening frequency. Brain Research. 1989;489:190–194. doi: 10.1016/0006-8993(89)90024-3. [DOI] [PubMed] [Google Scholar]
  25. Park-Chung M, Malayev A, Purdy RH, Gibbs TT, Farb DH. Sulfated and unsulfated steroids modulate γ-aminobutyric acid receptor function through distinct sites. Brain Research. 1999;830:72–87. doi: 10.1016/s0006-8993(99)01381-5. [DOI] [PubMed] [Google Scholar]
  26. Qin F, Auerbach A, Sachs F. Estimating single-channel kinetic parameters from idealized patch-clamp data containing missed events. Biophysical Journal. 1996;70:264–280. doi: 10.1016/S0006-3495(96)79568-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Qin F, Auerbach A, Sachs F. Maximum likelihood estimation of aggregated markov processes. Proceedings of the Royal Society. 1997;264B:375–383. doi: 10.1098/rspb.1997.0054. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Robel P, Bourreau E, Corpechot C, Dang DC, Halberg F, Clarke C, Haug M, Schlegel ML, Synguelakis M, Vourch C, Baulieu EE. Neuro-steroids: 3 β-hydroxy-δ-5 derivatives in rat and monkey brain. Journal of Steroid Biochemistry. 1987;27:649–655. doi: 10.1016/0022-4731(87)90133-6. [DOI] [PubMed] [Google Scholar]
  29. Sakmann B, Patlak J, Neher E. Single acetylcholine-activated channels show burst-kinetics in presence of desensitizing concentrations of agonist. Nature. 1980;286:71–73. doi: 10.1038/286071a0. [DOI] [PubMed] [Google Scholar]
  30. Shen W, Mennerick S, Covey DF, Zorumski CF. Pregnenolone sulfate modulates inhibitory synaptic transmission by enhancing GABAA receptor desensitization. Journal of Neuroscience. 2000;20:3571–3579. doi: 10.1523/JNEUROSCI.20-10-03571.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Shen W, Mennerick S, Zorumski EC, Covey DF, Zorumski CF. Pregnenolone sulfate and dehydroepiandrosterone sulfate inhibit GABA-gated chloride currents in Xenopus oocytes expressing picrotoxin-insensitive GABAA receptors. Neuropharmacology. 1999;38:267–271. doi: 10.1016/s0028-3908(98)00172-5. [DOI] [PubMed] [Google Scholar]
  32. Steinbach JH, Bracamontes J, Yu L, Zhang PN, Covey DF. Subunit-specific action of an anticonvulsant thiobutyrolactone on recombinant glycine receptors involves a residue in the M2 membrane-spanning region. Molecular Pharmacology. 2000;58:11–17. doi: 10.1124/mol.58.1.11. [DOI] [PubMed] [Google Scholar]
  33. Twyman RE, Rogers CJ, Macdonald RL. Intraburst kinetic properties of the GABAA receptor main conductance state of mouse spinal cord neurones in culture. Journal of Physiology. 1990;423:193–220. doi: 10.1113/jphysiol.1990.sp018018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Ueno S, Bracamontes J, Zorumski C, Weiss DS, Steinbach JH. Bicuculline and gabazine are allosteric inhibitors of channel opening of the GABAA receptor. Journal of Neuroscience. 1997;17:625–634. doi: 10.1523/JNEUROSCI.17-02-00625.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Vallee M, Mayo W, Darnaudery M, Corpechot C, Young J, Koehl M, Lemoal M, Baulieu EE, Robel P, Simon H. Neurosteroids-deficient cognitive performance in aged rats depends on low pregnenolone sulfate levels in the hippocampus. Proceedings of the National Academy of Sciences of the USA. 1997;94:14865–14870. doi: 10.1073/pnas.94.26.14865. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Villarroel A, Herlitze S, Witzemann V, Koenen M, Sakmann B. Asymmetry of the rat acetylcholine receptor subunits in the narrow region of the pore. Proceedings of the Royal Society. 1992;249B:317–324. doi: 10.1098/rspb.1992.0121. [DOI] [PubMed] [Google Scholar]
  37. Wang MD, Wahlstrom G, Backstrom T. The regional brain distribution of the neurosteroids pregnenolone and pregnenolone sulfate following intravenous infusion. Journal of Steroid Biochemistry and Molecular Biology. 1997;62:299–306. doi: 10.1016/s0960-0760(97)00041-1. [DOI] [PubMed] [Google Scholar]
  38. Wang TL, Hackam AS, Guggino WB, Cutting GR. A single amino acid in γ-aminobutyric acid ρ1 receptors affects competitive and noncompetitive components of picrotoxin inhibition. Proceedings of the National Academy of Sciences of the USA. 1995;92:11751–11755. doi: 10.1073/pnas.92.25.11751. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Weiss DS, Magleby KL. Gating scheme for single GABA-activated Cl− channels determined from stability plots, dwell-time distributions, and adjacent-interval durations. Journal of Neuroscience. 1989;9:1314–1324. doi: 10.1523/JNEUROSCI.09-04-01314.1989. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Woodhull AM. Ionic blockage of sodium channels in nerve. Journal of General Physiology. 1973;61:687–708. doi: 10.1085/jgp.61.6.687. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Woodward RM, Polenzani L, Miledi R. Effects of steroids on γ-aminobutyric acid receptors expressed in Xenopus oocytes by poly(A)+ RNA from mammalian brain and retina. Molecular Pharmacology. 1992;41:89–103. [PubMed] [Google Scholar]
  42. Woodward RM, Polenzani L, Miledi R. Characterization of bicuculline/baclofen-insensitive (ρ-like) γ-aminobutyric acid receptors expressed in Xenopus oocytes. II. Pharmacology of γ-aminobutyric acidA and γ-aminobutyric acidB receptor agonists and antagonists. Molecular Pharmacology. 1993;43:609–625. [PubMed] [Google Scholar]
  43. Wu FS, Gibbs TT, Farb DH. Pregnenolone sulfate: a positive allosteric modulator at the N-methyl-D-aspartate receptor. Molecular Pharmacology. 1991;40:333–336. [PubMed] [Google Scholar]
  44. Zaman SH, Shingai R, Harvey RJ, Darlison MG, Barnard EA. Effects of subunit types of the recombinant GABAA receptor on the response to a neurosteroid. European Journal of Pharmacology. 1992;225:321–330. doi: 10.1016/0922-4106(92)90106-6. [DOI] [PubMed] [Google Scholar]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES