| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Journal of Physiology (2001), 537.3, pp. 715-733
© Copyright 2001 The Physiological Society
 |
ABSTRACT |
|---|
 TopAbstract IntroductionMethodsResultsDiscussionReferences |
|---|
1
2
2L GABAA receptors transiently expressed in human embryonic kidney 293 cells, using the cell-attached, single-channel patch-clamp technique. The receptors were activated by GABA, -alanine or piperidine-4-sulfonic acid (P4S), and the effects of pentobarbital (PB) on single-channel activity were examined.
) was determined from the value of the effective opening rate at a saturating agonist concentration. was about 1900 s-1 when the receptors were activated by GABA, 1500 s-1 when activated by -alanine, and too low to be determined when P4S was administered. In the presence of 40 µM PB, was about 1500 s-1 when the receptors were activated by GABA, 1400 s-1 when activated by -alanine, and 50 s-1 when activated by P4S. Hence, the potentiating effect of PB is not mediated by a change in . The concentration of agonist producing a half-maximal effective opening rate also remained unaffected in the presence of PB, indicating that receptor affinity for agonists is not influenced by PB.
-alanine could be described by the sum of two exponential components (1.1 and 3.5 ms). However, in the presence of 40 µM PB the open-time distribution contained three exponential components (0.2, 2 and 10 ms). Finally, openings elicited by P4S exhibited two components (0.3 and 0.9 ms). In the presence of 40 µM PB, three components could be distinguished (0.5, 2.5 and 13 ms).
-alanine are also broadly consistent with this idea. However, the changes in open-time distributions caused by PB appear to be more complex. Possible explanations of the effects of PB on gating by different agonists are considered.
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The GABA type A (GABAA) receptor is a ligand-gated ion channel whose activation leads to cellular inhibition in the central nervous system. This receptor is normally activated by GABA and can be modulated by a number of compounds including benzodiazepines, barbiturates and steroidal anaesthetics (reviewed in, for example, Olsen & Tobin, 1990).
The sedative and hypnotic effects of barbiturates are thought to result, at least in part, from actions on GABAA receptors. However, barbiturates have several distinct effects that have complicated efforts to reach a mechanistic understanding of their actions. Depending upon the dose, pentobarbital (PB) can have three effects on GABAA receptor function. At low micromolar concentrations, PB potentiates the activation of the receptor by GABA (Akaike et al. 1990). At high micromolar concentrations, PB directly gates the GABAA receptor (Akaike et al. 1985, 1990; Rho et al. 1996; Akk & Steinbach, 2000; Serafini et al. 2000), and at millimolar concentrations PB blocks the receptors (Akaike et al. 1985, 1990; Rho et al. 1996; Akk & Steinbach, 2000; Serafini et al. 2000). The effects can be separated functionally, and appear to be mediated via different interaction sites (for example, Dalziel et al. 1999; Serafini et al. 2000). In spite of the plethora of information available on the potentiating effect of PB from whole-cell experiments, relatively few studies have examined the actions of PB on single-channel currents from GABAA receptors. Macdonald et al. (1989) showed that in mouse spinal cord neurones, co-application of 50 µM PB with 2 µM GABA leads to an increase in the mean open duration of the channel. In both the presence and absence of PB, the open-time distribution was described by three exponential components; the increase in mean open time resulted from an increase in the relative contribution of the long-lived openings, while the time constants were unaffected by PB. In contrast, none of the five closed-time components were altered in the presence of PB. Porter et al. (1992) found that in Chinese hamster ovary cells stably expressing bovine 1 and 1 subunits, 50 µM PB mainly increased the proportion of long-duration openings elicited by 5 µM GABA. However, a small increase in the duration of openings was also observed. Of the five closed-time components, PB prolonged the durations of the four longer-duration components. In both of these studies, it was concluded that the major action of PB was to increase the relative opening rate for the long-duration open component, with minimal effects on channel closing rates.
In the present work, we have used the single-channel patch-clamp technique to investigate the gating mechanisms of recombinant GABAA receptors (122L) transiently expressed in human embryonic kidney (HEK) 293 cells. We used three agonists, GABA, -alanine and piperidine-4-sulfonic acid (P4S) to activate the GABAA receptor, to allow us to study more accurately the kinetic steps in activation that are affected by pentobarbital. P4S is a high-affinity, low-efficacy agonist for the GABAA receptor (Ebert et al. 1994). The maximal response to P4S is significantly lower than that obtained in the presence of GABA, suggesting that channel gating is impaired in the presence of P4S. A reduction in the maximal response can be achieved by a reduction in the channel-opening rate constant and/or an increase in the channel-closing rate constant. -Alanine, on the other hand, is a low-affinity, high-efficacy agonist for the GABAA receptor (Jones et al. 1998). The maximal response to -alanine is comparable to the maximal response to GABA, but the concentration dependence is shifted to higher concentrations of agonist.
In order to overcome some of the obstacles to interpreting single-channel data, we examined receptor function under conditions where the activity takes place in clusters. Clusters can be elicited by high, desensitizing concentrations of the agonist, and each cluster originates from the activity of a single receptor (Sakmann et al. 1980). Thus, the uncertainty in the number of active channels is eliminated. Single channel clusters are groups of channel openings and closings reflecting rapid kinetic events such as agonist association and dissociation, channel opening and closing, and dwells in various short-lived desensitized and blocked states. The cluster is terminated by entry of the receptor into a long-lived desensitized state.
|
METHODS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Expression systems and electrophysiology
Rat GABAA receptor 1, 2 and 2L subunit cDNAs were subcloned into a cytomegalovirus promoter-based vector pcDNAIII (Invitrogen, San Diego, CA, USA), and transiently expressed in HEK 293 cells using a calcium phosphate precipitation-based transfection technique (Ausubel et al. 1992).
P4S was purchased from Sigma-RBI. All other drugs and chemicals were purchased from Sigma Chemical (St Louis, MO, USA).
The single-channel currents were recorded using a patch-clamp technique in the cell-attached configuration (Hamill et al. 1981). The bath solution contained (mM): 140 NaCl, 5 KCl, 1 MgCl2, 2 CaCl2, 10 glucose and 10 Hepes; pH 7.4. The pipette solution contained (mM): 120 NaCl, 5 KCl, 10 MgCl2, 0.1 CaCl2, 20 TEA, 5 4-aminopyridine, 10 glucose, 10 Hepes; pH 7.4. In addition, the pipette solution contained the indicated concentrations of the agonist (GABA, -alanine or P4S) and PB, when needed. Agonist was added to the saline, and no correction for osmolarity was made. The pipette potential was held at +60 to +80 mV. We assume that the cell membrane potential was ~-40 mV, and thus the total potential across the patch membrane was between -100 and -120 mV. The channel activity was recorded with an Axopatch 200B amplifier, low-pass filtered at 10 kHz, acquired with a Digidata 1200 series interface at 50 kHz using pCLAMP 7 software (Axon Instruments, Foster City, CA, USA) and stored on a PC hard drive.
Kinetic analysis
The kinetic analysis was performed on single-channel clusters. A cluster is defined as a series of openings and closings of a single ion channel, which starts as the channel returns from the long-lived desensitized state(s) and ends when the channel enters the long-lived desensitized state(s). The duration of a typical cluster is 2-3 s, while the average lifetime of the receptor in the desensitized state(s) can be > 10 s (cf. Jones & Westbrook, 1995). Due to the low number of receptors in the patch and the relatively short lifetime of clusters compared to the dwells in long-lived desensitized state(s), the activity in most patches consisted of episodes of intense activity of a single channel (a cluster), separated from other clusters by intervals lasting up to tens of seconds. In the present experiments, we observed the clustering activity of the receptor at GABA concentrations ([GABA]) of 
20 µM, and -alanine concentrations ([-alanine]) above 2 mM. Clusters of activity produced by P4S were difficult to resolve in the absence of PB (see Results), because the combination of brief open times and long closed times within clusters made the clusters hard to recognize.
To isolate single channel clusters we used a cutoff duration, 
crit. A cluster termination was called when the duration of the closed-time interval between two neighbouring openings exceeded that of crit. In theory, the optimal duration of crit depends upon the intracluster closed times and should be at least five times longer than the mean duration of the longest intracluster closed time. This assures that the misclassification of closed-time events is insignificant. In the present case, clusters at all agonist concentrations contained relatively long-lived gaps with an overall mean duration of 27 ms. To include such gaps in the analysis, we used a crit of at least 500 ms.
The great majority of openings in clusters were to a single level. However, we noted that 5 % or less of the openings in clusters elicited by GABA (or GABA plus PB) were to a level with a conductance of about 75 % of the most common one. These openings were clearly from the same channel as the most common ones, as they occurred during the high activity of a cluster. In clusters elicited by -alanine (or -alanine plus PB) the lower-conductance openings also constituted 5 % or less of the openings, while we did not observe the lower conductance level in activity elicited by P4S. The lower-conductance openings were treated as full openings in all analyses.
The isolated clusters were low-pass filtered at 3 kHz and idealized using the segmented-k-means algorithm (program SKM, Qin et al. 1996). The intracluster open and closed times were estimated using maximum likelihood methods, which incorporate a correction for missed events (program MIL, Qin et al. 1997). Since we do not have a complete scheme for activation of the GABAA receptor, we analysed the closed- and open-time distributions separately. For example, we analysed the open-time distributions by assuming that a single closed state existed, which was connected to one or more open states. The open states were assumed to be unconnected to each other:
In this simplistic approach, the rates for entering and leaving the open states are independent, and we do not examine how particular closed states might be connected to particular open states. The fit commenced using a simple two-state model corresponding to a single open and a single closed state. Additional states were then added to the scheme until the increase in the log-likelihood was below the level of significance (P < 0.05, see Colquhoun & Sigworth, 1995). The optimal fit for the intracluster open time histogram was obtained using two or three open states (see Results).
A similar approach was used to analyse the distribution of closed durations. The optimal fits for the intracluster closed-time distributions were obtained with three or four closed states, depending on the agonist concentration (see Results). The values for the best fitting parameter estimates are given in the text, with the estimated S.D. of the parameter derived from the fitting.
We note that this approach ignores the information that is contained in the association between kinetically identifiable states; for example, whether brief openings are preferentially associated with long closings. However, we do not have a complete kinetic model that incorporates all of the open and closed states that can be distinguished kinetically. Much of our data is qualitatively consistent with models proposed by others (Weiss & Magleby, 1989; Twyman et al. 1990; Haas & Macdonald, 1999).
Single-channel activity elicited by lower [GABA] and [-alanine] had closed-time distributions with three exponential components, with time constants of about 0.2, 1.5 and 50-100 ms. Bursts were identified using a crit of 10-15 ms.
|
RESULTS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Receptor activation by GABA
Sample GABAA receptor activity elicited by 20 and 1000 µM GABA is shown in Fig. 1. In the presence of [GABA] below 20 µM, the receptor openings were not condensed into clusters (data not shown). At higher values of [GABA], channel activity occurred in clusters that represent a sequence of channel events arising from a single ion channel. As the agonist concentration was increased, the intracluster open probability (Po) increased. Figure 2A shows the relationship between Po and [GABA]. The fit of the Hill equation demonstrates that the curve saturates at a maximum Po (Po,max) of 0.82, with a concentration producing a half-maximal Po (EC50) of 70 µM and a Hill coefficient of 1.2.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 1. Single-channel currents elicited by GABA Sample traces showing clusters of single-channel activity elicited by GABA in the absence or presence of 40 µM pentobarbital (PB), with the open-time and closed-time histograms presented under the traces. The scale bars apply to all traces in the figure. Note that the openings (openings shown downwards) are longer in the presence of PB. Each histogram is shown as the square root of the fraction of the total number of events in each bin, while the durations are binned logarithmically (Sigworth & Sine, 1987). Note that PB results in an increase in the prevalence and mean duration of long-duration openings (compare right column to left column). The lines show the fits of exponential components to the data. Each open-time histogram is fitted with the sum of three exponential components (for values see Table 1A). The closed-time histograms with 20 µM GABA are fitted with the sum of three exponential components, while those with 1 mM GABA are fitted with the sum of four (Table 1B). The total number of dwells (open and closed) are: 20 µM GABA, 3091; 20 µM GABA + 40 µM PB, 4091; 1 mM GABA, 2937; 1 mM GABA + 40 µM PB, 8167. | ||
 |
View larger version [in this window] [in a new window] |
|
|
Figure 2. Activation by GABA A, relationship between the intracluster open probability (Po) and the concentration of GABA ([GABA]), in the absence ( | ||
We do not propose here a kinetic scheme that will account fully for the distributions of closed and open durations. Instead, we have analysed the open- and closed-time distributions separately (see Methods).
The open-time histograms were better fitted with the sum of three exponentials than with either two or one. The three time constants were well separated from each other, with the shortest at ~0.4 ms (called brief openings, or OT1), the openings of intermediate duration at ~2 ms (called intermediate openings, or OT2) and the long-lived openings at ~6 ms (called long openings, or OT3). The values for the three open-time components obtained at different values of [GABA] are summarized in Table 1A. The durations and fractions of the three classes of openings are independent of [GABA] at 20-1000 µM. Thus, we believe that they all originate from fully liganded states (i.e. they do not contain monoliganded open states).
The closed-time histograms were best fitted with the sum of three or four components; the best fit was obtained with three closed states for data recorded with 20-200 µM GABA, while in the presence of 500-1000 µM GABA, four closed-time components were observed. Table 1B shows the results of fits of closed-time histograms by a model containing a single open state connected to three or four closed states (see Methods). In this approach, the closed states are all viewed as disconnected from each other. The rate for entering state i is given as k-i while the rate of leaving the state (the inverse of the mean duration) is given as k+i. The results demonstrate the following:
(1) The data from all patches contained a short-lived closed-time component (designated as CT1). Its mean duration was 0.2 ms, and this value did not depend upon [GABA]. This component made up 60 % of all intracluster closed times, while the entry rate into this closed state was 286 s-1, i.e. the channel entered into this state after it had been open for an average of 4 ms.
(2) Clusters obtained in the presence of 100-1000 µM GABA contained an infrequent, relatively long-lived closed state (designated as CT3). Its mean duration was 20 ms, it occurred at a rate of 8 s-1 and constituted ~1 % of all closed dwells. Even though the component was observed only at relatively high agonist concentrations (100-1000 µM), we believe that it is also present in patches obtained in the presence of 20 and 50 µM GABA. It is likely that we could not resolve it under these conditions due to the presence of another component that has a similar duration and is observed at a higher frequency at low values of [GABA] (see below).
(3) In the presence of low (20-50 µM) and high (500-1000 µM) [GABA], we detected a closed-time component with a mean duration of 2.2 ms (designated as CT2). It occurred at a mean rate of 60 s-1, and its relative weight was 12 %. It was not resolved at 100-200 µM, probably because of the presence of another closed component with a similar duration (Cact). Its relative weight was lower at high than at low [GABA] (4 %). This may be caused by the similarity in values for the mean durations of this component and Cact at high [GABA].
(4) Finally, a closed state whose duration scaled with [GABA] was observed (designated as Cact). Its duration was similar to that of CT3 at low [GABA], and to that of CT2 at intermediate [GABA]. This component had a relative weight of 26 %, while the rate of entry into it was 135 s-1.
The Cact component in the intracluster closed-time histograms is likely to reflect steps in the channel activation pathway. There are two major reasons for this assignment. First, the duration of this component differs for GABA and P4S, and the dependence on agonist concentration differs between GABA and -alanine (see below), while the properties of the other components are less critically dependent upon the nature of the agonist. Second, the duration of the Cact component depends upon [GABA] (Fig. 2B), as well as [-alanine] and the concentration of P4S ([P4S]; see below). In many models for channel activation, the effective opening rate is predicted to increase with increasing agonist concentrations until it saturates at high agonist concentrations. The high concentration asymptote is the channel-opening rate constant (). In Fig. 2B, it may be seen that the rate for leaving Cact (that is, the inverse of the mean duration) saturates at a [GABA] of about 1 mM. A fit by the Hill equation yields = 1883 s-1. The midpoint of the curve (EC50) is at 359 µM and the Hill coefficient is 1.7. The estimate for is comparable to estimates from rapid applications of high [GABA] (1800-6000 s-1; Maconochie et al. 1994; Lavoie et al. 1997; Li & Pearce, 2000). If Cact has been correctly identified with channel reopening from the liganded, closed state, then the rate of occurrence of this component will give the channel closing rate. Our data show this to be about 220 s-1 for GABA.
We are not sure how closed and open states would be connected in a complete kinetic scheme for gating of the GABAA receptor. We note, however, that our observations are in agreement with the scheme proposed by Haas & Macdonald (1999), allowing for the fact that our approach analysed open and closed states separately. These authors also observed three components in the open-time distribution, and associated two of them with diliganded receptors. We would modify this by associating even the third, brief-duration open component with diliganded receptors. They identify six brief-duration, non-conducting states as representing transitions from the open states to closed states that are not on the return pathway to the resting, unliganded state. These would probably underlie the CT1 and CT2 components we have seen, consolidated into a single brief closed state and a single longer-duration closed state. Finally, they identify a rapidly recovering desensitized state, which might correspond to the CT3 component. They also postulate two longer-duration desensitized states, which in our records would correspond to cluster terminations. Haas & Macdonald (1999) estimated that for the most common open state in records from expressed recombinant 132 receptors, the channel closing rate is about 280 s-1 and is about 1800 s-1. These estimates are similar to those presented here.
Receptor activation by GABA in the presence of 40 µM PB
To establish the concentration of PB at which its effects would be studied, we first measured single-channel activity of the receptors in the presence of 0.05-100 µM PB. The [GABA] in these experiments was 50 µM to allow easy recognition of clusters yet have a low enough Po to observe the potentiating effect of PB. Figure 3 shows the relationship between the Po and PB concentration. The curve was fitted using a Hill equation with an offset. The results show that the saturating Po is 0.57, the EC50 is 2.5 µM PB and the Hill coefficient is 1.9. The fit offset is at 0.36 (the fit Po of clusters elicited by 50 µM GABA in the absence of PB). We chose to use 40 µM PB in further experiments, since this concentration produced an essentially maximal increase in Po, while being low enough that direct gating and block by PB are minimal (Akk & Steinbach, 2000).
 |
View larger version [in this window] [in a new window] |
|
|
Figure 3. Potentiation by PB The presence of PB increased Po for clusters elicited by 50 µM GABA, reaching a maximum at about 10 µM PB. The smooth curve shows the predictions of the Hill equation, with the best fitting parameter estimates (± S.D. for the parameter estimated from the fit): Po,max = 0.57 ± 0.03, Po,min = 0.36 ± 0.02, EC50 = 2.5 ± 0.8 µM and the Hill coefficient is 1.9 ± 1.0. Each symbol shows data from a single patch. | ||
Using a constant concentration of 40 µM PB, we then examined how PB affects the channel activity elicited by GABA. Figure 1 shows sample clusters from patches recorded in the presence of 40 µM PB and 20 or 1000 µM GABA. The intracluster Po was increased at the lower values of [GABA] (Fig. 2A). When fitted with the Hill equation, the EC50 was decreased to 28 µM GABA, while Po,max and the Hill coefficient were essentially unchanged at 0.83 and 1.3, respectively.
Comparison of open-time histogram components (see Table 1A) shows that both the fraction of long openings (OT3) and the mean duration of this component were increased in the presence of PB. The parameters for the two briefer components were less affected, with the exception of a reduction in the fraction of intermediate openings (OT2).
The intracluster closed times were little affected by PB (Table 1B). Four components were identified in the distributions of closed times (Table 1B). The mean durations of the components were quite similar in the presence and absence of PB, and the fractional representation. However, one clear difference was seen: the rates of entry into the closed states were all reduced in the presence of PB, by about twofold (Table 1B).
Figure 2B shows the dependence of the inverse of the mean duration of Cact on [GABA]. When the data were fitted with the Hill equation, = 1505 s-1, EC50 = 329 µM and the Hill coefficient is 1.6. Neither the concentration dependence nor the estimate of was affected by PB.
The observation that the open- and closed-time distributions contained multiple components in both the absence and presence of PB raised the possibility that the GABAA receptors in our patches may be structurally heterogeneous. In this case, they might differ both in activation properties and in the response to PB. As an initial approach to this question, we made cumulative distributions of data for individual clusters from selected records and compared them to the total cumulative distribution for all data from a patch (data not shown). This approach indicated that the dwell distributions for individual clusters resemble those for the entire data from a patch. There was no indication that some clusters reflect the activity of 'long opening' receptors while others reflect 'brief opening' receptors, in either the absence or presence of PB.
Gating by GABA and the actions of PB
The main observations made in studies of GABA and PB are: (1) PB increases the Po for currents elicited by low values of [GABA] and shifts the EC50 to a lower concentration, but there is no effect on Po,max; (2) PB has no major effect on the distribution of closed times at any concentration of GABA; and (3) PB increases the relative rate of occurrence of the longest duration class of openings, and also increases the duration of that component. PB apparently reduces all the rates for leaving that open state.
We calculated values for the mean closed times and mean open times from the values fitted to the closed-time distributions (Table 1B). As shown in Fig. 4A, the addition of PB prolonged the mean open time at all values of [GABA], with no effect on the mean closed time. A predicted Po was then calculated from the mean open and closed times estimated from the kinetic analysis. As shown in Fig. 4B, the calculated Po was close to the experimentally determined values, and showed the same shift in the dependence upon [GABA].
 |
View larger version [in this window] [in a new window] |
|
|
Figure 4. The increased mean open time in the presence of PB can account for potentiation of responses to GABA A, calculated mean open and closed times for GABA in the absence ( | ||
We then examined two additional agonists: -alanine and P4S. -Alanine has been reported to have a lower affinity for the receptors than GABA, but an essentially equal efficacy (Jones et al. 1998). In contrast, P4S has been reported to have a similar affinity to GABA, but significantly lower efficacy (Ebert et al. 1994). These agonists were chosen to help associate some of the components in the closed-time histograms with receptor activation, and to determine whether the potentiating action of PB was qualitatively consistent for multiple agonists.
Receptor activation by -alanine
Activation by -alanine resulted in the appearance of clusters of channel activity at concentrations above 2 mM (see Fig. 5 for clusters elicited by 10 mM -alanine). The Po increased with concentration (Fig. 6A), and a fit of the Hill equation gave parameter estimates for EC50 of 12.6 mM, Po,max of 0.80, and a Hill coefficient of 1.8. The EC50 was much greater than that estimated for GABA, while Po,max was similar.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 5. Single-channel currents elicited by Sample traces show clusters elicited by 10 mM | ||
 |
View larger version [in this window] [in a new window] |
|
|
Figure 6. Activation by A, relationship between Po and [ | ||
The distributions of open times within clusters differed from those observed with GABA, in that only two exponential components were consistently resolved (Fig. 5). The mean durations for these components were similar to the intermediate- and long-duration openings seen with GABA (Table 2A). As seen in the observations with GABA, the mean durations and relative prevalence of the open-time components were relatively constant across different values of [-alanine], indicating that over that range of concentrations both open states reflect fully liganded receptors.
The distribution of closed times required three or four components, as was observed for currents elicited by GABA (Fig. 5). Three components (CT1, CT2 and CT3) appeared to be independent of [-alanine] (Table 2B). It should be noted, however, that the same possible confusion existed for this data as for data obtained with GABA, in that the fourth component (Cact) had a duration that overlapped either CT2 or CT3 at some values of [-alanine]. The mean durations and rates of occurrence of CT1, CT2 and CT3 were similar (within a factor of 2) for -alanine and GABA. Cact became briefer as [-alanine] increased, although the concentration range was much higher than for GABA (Fig. 6B). When the Hill equation was fitted to the data for the concentration dependence of the effective opening rate, the parameter values obtained were = 1470 s-1, EC50 = 19 mM and the Hill coefficient is 2.1. The estimates for and the Hill coefficient were similar to those for GABA activation, while the EC50 was much greater.
Effect of 40 µM PB on the clusters of activity elicited by -alanine
In the presence of 40 µM PB, the relationship between Po and [-alanine] was shifted to lower concentrations (Fig. 6A). A fit of the Hill equation gave the following parameter estimates: EC50 = 5.6 mM, Po,max = 0.77, and the Hill coefficient is 2.4. The EC50 was reduced compared to the response to -alanine in the absence of PB, while Po,max and the Hill coefficient were not altered.
In the presence of PB, the open-time distributions now required three components (Fig. 5), with time constants and proportions similar to the components seen for GABA in the presence of PB (Table 2A). As was the case with GABA, the proportions and time constants for the components showed no consistent trend with agonist concentration. The major consequences of the presence of PB were the appearance of a brief component in the open-time distribution and a prolongation of the slowest component.
In contrast to the open-time distributions, the closed-time distributions were not qualitatively altered by PB (Table 2A). The major effect was a reduction in the rates of occurrence of the components, consistent with the prolongation of open times. When the Hill equation was fitted to the data for the concentration dependence of effective opening rate in the presence of PB, the values for (1400 s-1), the EC50 (24 mM) and the Hill coefficient (1.8) were not different from those obtained in the absence of PB (Fig. 6B).
The action of PB on the currents elicited by -alanine, therefore, has features in common with its action on currents elicited by GABA: the major effect was to prolong open times, with less effect on closed times and no apparent effect on . However, the number of components in the open-time distributions was altered.
Again, the changes in the calculated mean open times could quantitatively predict the ability of PB to enhance Po (Fig. 7).
 |
View larger version [in this window] [in a new window] |
|
|
Figure 7. The increased mean open time in the presence of PB can account for the potentiation of responses to A, calculated mean open and closed times for | ||
Receptor activation by P4S
We could not measure the gating kinetics for activation by P4S, since the channel opening rate was very low. Figure 8 shows single-channel currents activated by 1 mM P4S. The activity consisted of infrequent openings, with no clear clustering. Accordingly, we looked at the overall distributions of open and closed times. The open-time distribution for the activity elicited by P4S showed only two components, roughly corresponding to the brief and intermediate components elicited by GABA (Table 3A).
 |
View larger version [in this window] [in a new window] |
|
|
Figure 8. Single-channel currents elicited by P4S A sample of the activity recorded in the presence of 1 mM P4S, in the absence of PB, is shown at the top. No clusters could be resolved under these conditions. The second trace shows an expanded segment of the activity, to demonstrate the brief duration of the openings. A cluster elicited by 1 mM P4S in the presence of 40 µM PB is shown in the third trace, with an expanded segment below it. Note that the openings are clearly longer in the presence of PB. The duration histograms are displayed in the same fashion as in Fig. 1. For clarity, the individual components fitted to the open-time distributions (two for 1 mM P4S in the absence of PB, three in the presence of PB) are shown. The total number of dwells are: 1 mM P4S, 1741; 1 mM P4S + 40 µM PB, 7532. | ||
The closed-time distributions showed components resembling CT1 and CT2 in the presence of GABA. The inverses of the mean durations were about 1600 s-1 and 350 s-1. However, we could not interpret the longer-duration closed times for activity elicited by this agonist since we do not know the number of receptors active at a given time, and could not decide which closures occurred within clusters.
Effect of 40 µM PB on the clusters of activity elicited by P4S
In contrast to the observations made using P4S alone, in the presence of PB, clear clusters of activity could be resolved (Fig. 8). The Po curve for P4S in the presence of 40 µM PB is shown in Fig. 6A, and a fit of the Hill equation yielded Po,max = 0.31, EC50 = 115 µM and the Hill coefficient is 0.7.
The distributions of intracluster open times elicited by P4S in the presence of PB required three components (Fig. 8). The components resembled qualitatively the components seen for GABA or -alanine in the presence of PB. Again, there was no consistent change in either the proportions or the mean durations of the components with [P4S] (Table 3A). The major effect of PB was the appearance of a long-duration open-time component (OT3). As will be shown (Fig. 9), increasing concentrations of PB also prolonged the duration of this component.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 9. PB first prolongs, then shortens the duration of long openings elicited by P4S This figure shows the inverse of the mean duration of the long-duration open-time component (OT3), for opening elicited by P4S in the presence of various concentrations of PB. Between 1 and 100 µM PB the mean duration increases. At higher concentrations of PB, the mean duration decreases rapidly. The continuous line shows predictions of a simple model for prolongation and block of the long duration open state by PB (see Results), while the dotted line shows the predicted prolongation that would occur with this model if no block were present. Each symbol shows data from a single patch. | ||
The intracluster closed-time distributions required three or four exponential components for description (Fig. 8, Table 3B). The mean duration of the briefest closed-time component was 0.35 ms (analogous to CT1 in the presence of GABA). The duration of the intermediate closed-time was 2.7 ms (analogous to CT2 in the presence of GABA). The durations of these closed times did not depend upon the [P4S]. There was an infrequent, long-duration component, perhaps equivalent to CT3. Finally, there was a component whose duration appeared to change with [P4S]. This component had a long mean duration that overlapped with the mean duration of CT3 at many values of [P4S]. However, the rate of occurrence of CT3 appeared to be as low as it was for the closed-time distributions for activity elicited by GABA or -alanine (see Tables 1, 2 and 3). Accordingly, it appeared likely that the mean durations of this component reflected the duration of a component analogous to Cact for activity elicited by GABA or -alanine. The relationship between the inverse of the mean duration of this component and [P4S] is shown in Fig. 6B. A fit of the Hill equation yielded the following estimates: = 54 s-1, EC50 = 196 µM and the Hill coefficient is 1.3.
Again, the major effect of PB was to prolong open times. In the case of P4S, we did not obtain clear clustering in the absence of PB, which makes it impossible to distinguish the closed times within clusters. As was the case with -alanine, the number of components in the open-time distributions was increased in the presence of PB.
The effect of PB on the bursts elicited by lower values of [GABA] or [-alanine]
The data discussed up to this point were obtained during periods of activity at relatively high agonist concentrations. At these concentrations, rebinding of agonist is rapid so that some sojourns in closed-channel states probably reflect the sequence: channel closure, agonist dissociation, agonist binding, channel opening. At lower concentrations of agonist, however, the agonist association rate is sufficiently low that agonist dissociation results in a long closed sojourn. Accordingly, the 'bursts' of activity recorded at lower agonist concentrations can contain information about the relative rates of agonist dissociation and channel opening (Colquhoun & Hawkes, 1981).
The presence of 40 µM PB increased the mean open time for events in bursts elicited by GABA (20 µM GABA: 2.6 ms; 10 µM GABA + 40 µM PB: 4.9 ms; 1.9-fold) and -alanine (1 mM -alanine: 2.7 and 2.4 ms; 1 mM -alanine + 40 µM PB: 4.4 and 4.8 ms; 1.8-fold). The increase in mean open times agrees well with the values seen at higher concentrations (Tables 1 and 2). The mean burst duration was also increased, somewhat more than the mean open time (20 µM GABA: 16 ms; 10 µM GABA + 40 µM PB: 43 ms; 2.7-fold; 1 mM -alanine: 5.1 and 6.4 ms; 1 mM -alanine + 40 µM PB: 19 and 24 ms; 3.8-fold).
The definition of a burst is a series of openings that are separated by closed sojourns that last less than a critical time (crit; in the present cases 10-15 ms). For a kinetically simple receptor such as the muscle nicotinic receptor, this is largely determined by the ratio of the channel opening rate to the agonist dissociation rate (cf. Colquhoun & Sakmann, 1981). This situation arises because almost all dwells in closed states originate from a channel closing step, and the probability of reopening rather than losing a bound agonist molecule is determined by the ratio of the opening rate to the sum of the opening rate plus the dissociation rate. For the GABAA receptor, it can be problematic to define what event terminates an opening, since closed periods can arise from sojourns in multiple states (see Weiss & Magleby, 1989; Twyman et al. 1990). However, the data obtained with higher concentrations of agonists allow us to assess the rates for entering several of the closed states. The mean durations of CT1 and CT2 are much less than crit for determining a burst, and so a sojourn in a non-conducting state in these components would produce an intraburst closing. The duration of CT3 was longer than crit, and so entry into a state in this component would produce a burst termination. The rate of occurrence of CT3 was quite low, however. Finally, if our interpretation of Cact is correct, about 30 % of all terminations of sojourns in an open state reflect entry into a non-conducting state by a classical channel closure.
We estimated the probability of reopening within a burst (Preopen) from the data by determining the numbers of bursts with j = 1, 2, 3, etc. events. To smooth the data we calculated the cumulative distribution, and then calculated Preopen by taking the ratio of the fraction of bursts with more than j + 1 events to that with more than j events. Preopen was estimated as the mean value for the ratio until 90 % of bursts had j + 1 or fewer events. As seen in Table 4, there was an increase in Preopen in the presence of PB, which would account for the additional increase in the mean burst duration compared to the mean open time.
We then calculated a value for Preopen from the data obtained at high agonist concentrations in the absence and presence of PB (Table 4). The agreement was reasonably good for -alanine. In the case of GABA, the predicted Preopen was lower than the observed value. Our initial assumption was that reopening within a burst occurred only from CT1 and CT2, but not from Cact. That is, in terms of the simple situation discussed above, that agonist dissociation proceeded at a much higher rate than channel opening. However, the predicted Preopen matched the experimental values very well if it was assumed that a closed channel with GABA bound to the receptor reopened about 40 % of the time (i.e. the opening rate was about two-thirds the dissociation rate). The presence of PB had no effect on the assumed reopening ratio (Table 4). For both GABA and -alanine, the increase in Preopen arose because Cact was relatively less common than CT1 or CT2 in the presence of PB (in other words, the channel closing rate was reduced somewhat more than the rate for entering CT1 and CT2).
Some previous studies have estimated that the channel opening rate is severalfold higher than the dissociation rate when GABA is used as an agonist (Twyman et al. 1990; Jones & Westbrook, 1995). Such a high ratio would have been manifest by a much higher probability of reopening, and the presence of bursts with many openings. The interpretation of Preopen is complicated because we do not know the connections among all the states that the GABAA receptor commonly adopts. However, the conclusion that the dissociation rate is comparable to the opening rate is likely to be correct, for the following reason. Assume that there are three competing reactions for a single closed state: agonist dissociation, channel opening and transition to a non-conducting state X. Assume then that X is a long-lived non-conducting state. However, the only long-lived closed state we see (CT3) occurs rarely, less than 10 % as often as Cact. Accordingly, a transition to this closed state would not affect the conclusion. Alternatively, if we assume that X is a shorter-lived non-conducting state, and we also assume that X resembles a desensitized state so that agonist dissociation from X is slow, in this case, the ratio of opening to dissociation rate constants would not be altered, although the estimated values for the rates would be affected.
These observations of the effects of PB on the activity elicited by lower values of [GABA] and [-alanine] are predictable from the results obtained at higher agonist concentrations. There is no indication that PB affected agonist dissociation rates. In conjunction with the finding that the concentration dependence of the apparent opening rate is unchanged, this indicates that the association rate is unlikely to be affected, and hence that agonist binding is not affected by PB.
The dependence of channel open times on the concentration of PB
Since currents elicited by P4S in the absence of PB showed only brief and intermediate-duration openings, it was easier to see the changes in the open-time distribution produced by PB. Accordingly, the changes produced by increasing concentrations of PB were re-examined, using 1 mM P4S to activate the receptors. As shown in Fig. 9, the mean duration of OT3 increased with increasing PB concentrations over the range of 1-100 µM PB, then decreased. The decrease was expected, since it has already been shown that PB reduces the duration of channel openings by a blocking mechanism (Akk & Steinbach, 2000). The fraction of the total openings that fall into the OT3 component increased steadily with increasing PB concentration (data not shown).
Since PB both prolonged openings and shortened them at higher concentrations, the concentration dependence of the prolongation cannot be immediately determined from the data on open times. We fitted the inverse of the mean duration of OT3 with a particular equation:
where k is the inverse of the mean duration of OT3 (i.e. the rate of termination of a long-duration opening), [PB] is the concentration of pentobarbital, a is the rate when PB is not bound to the receptor, b is the rate when PB is bound to the receptor, k+b is the forward rate for channel block, and X = [PB]/([PB] + Kd), where Kd is the dissociation constant for PB binding to the postulated prolongation site. This equation is based on a particular model in which PB binding is relatively rapid compared to gating, and block occurs equally for receptors with or without PB bound to the potentiation site. As shown in Fig. 9, this equation can describe the data and provides the following estimates: a = 242 ± 33 s-1 (± estimated S.E.M. of fitting), b = 37 ± 23 s-1, Kd = 7.4 ± 5 µM and k+b = (6.5 ± 0.05) 
105 M-1 s-1. According to this model, the potency of PB to produce potentiation is much higher than its potency to effect direct activation (with estimated dissociation constants greater than 1 mM; Akk & Steinbach, 2000; Serafini et al. 2000). In the absence of block, the maximal prolongation of open times would be from 4 ms to 27 ms. The rate for block is comparable to the rates estimated earlier for block of channels activated by GABA (3.2 105 M-1 s-1) or PB (5.6 105 M-1 s-1; Akk & Steinbach, 2000).
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Our results show that PB does not affect the distribution of closed times, the concentration dependence of the effective channel opening rate, or the channel-opening rate constant (). These observations indicate that the channel opening and agonist dissociation rates are unlikely to be significantly affected. In contrast, PB has major effects on the distribution of channel open times and the mean open time, which can fully explain the potentiation observed in terms of the probability of a channel being open in a cluster. We will consider these observations in some more detail, starting with a discussion of gating by agonists in the absence of PB.
Activation by GABA
Our data are consistent with previous conclusions that the kinetic scheme for the activation of the GABAA receptor is complicated. Indeed, we did not attempt to fit all of our data with a single, comprehensive model. There were two major reasons for this. The first was that we have not resolved which scheme would be the most appropriate for activation by GABA. The second is that activation by -alanine and P4S, while similar to activation by GABA, showed enough differences that we did not know how to incorporate all of the observations into a single model. Instead, we have adopted an approach that examines a number of agonists across a relatively wide concentration range to extract conclusions that are less dependent upon any particular kinetic model.
The maximal Po in a cluster elicited by GABA was 0.8, with an EC50 of 70 µM. These data compare relatively well with previous data on the activation of recombinant 122 receptors (cluster Po = 0.5 for single-channel currents elicited by 50 µM GABA; Brickley et al. 1999; EC50 = 18 µM for peak current elicited by rapid applications of GABA to excised, outside-out patches; Li & Pearce, 2000). Other values of EC50 estimated from macroscopic responses to rapid applications of GABA fall in the range 10-40 µM (Maconochie et al. 1994; Lavoie et al. 1997; Jones et al. 1998).
Several kinetic schemes have been proposed for activation (see Fig. 10). As already indicated, our observations are basically consistent with the gating scheme proposed by Twyman et al. (1990; Fig. 10B, see also Gingrich et al. 1995; Haas & Macdonald, 1999). Our data agree with the idea that there are two (CT1, CT2) brief-duration (0.1-0.4 and 1-2 ms) closed states that arise from the open states, and that are not on the return pathway to the resting, unliganded receptor (states identified as G in Fig. 10B). These states could provide significant difficulties in kinetic analyses that assumed that they actually corresponded to the closed liganded state. Our data also show a longer-duration closed state (CT3, 20-40 ms), which probably corresponds to the short-duration desensitized state seen by others (Maconochie et al. 1994; Jones & Westbrook, 1995; Haas & Macdonald, 1999). Our data do not address directly the question of whether this desensitized state arises exclusively from a closed-channel state(s) of the receptor (see Haas & Macdonald, 1999). However, the fact that this state occurred at a similar rate across a wide range of concentrations suggests that receptors with either one or two bound agonist molecules can enter the state. Our data do not examine critically the rate of cluster termination, which would correspond to the rate of entry into the long-duration desensitized state(s). However, the fact that the cluster duration does not depend upon [GABA] over the range of 50 µM to 1 mM (see Akk et al. 2001) is in agreement with the conclusion made by others that fully liganded receptors can enter this state (Gingrich et al. 1995; Haas & Macdonald, 1999). That is, both the short- and long-duration desensitized states (DF and DS in Fig. 10) can be entered by doubly liganded receptors. In sum, our data are not consistent with one specific model (Fig. 10C; Jones & Westbrook, 1995; Jones et al. 1998), in terms of the numbers of closed states and the connections between desensitized and non-desensitized states.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 10. Kinetic schemes for activation of GABAA receptors by GABA Three schemes which have been proposed for GABAA receptor activation are shown. A, Weiss & Magleby (1989). B, Haas & Macdonald (1999). C, Jones & Westbrook (1995). The state of the receptor is indicated by the letters C (closed channel), O (open channel), D (closed channel/desensitized) and G (brief-duration closed channel state). The letter A denotes an agonist molecule. The subscripts indicate kinetically distinguishable states within a given overall state. The desensitized states are identified as rapidly equilibrating (F), intermediate equilibration rate (I) and slowly equilibrating (S). Note that in the scheme shown in C, the long-lived desensitized state arises from monoliganded receptors, while the short-lived desensitized state arises from diliganded receptors. Overall, our data agree qualitatively best with the scheme shown in B, including the numbers of open and closed states, and connections among them. The hypothesized connection between open states (see Discussion) is indicated by the dotted arrows between A2O2 and A2O3 in B (see also A). | ||
We estimated a value for the channel opening rate constant () of about 1900 s-1 for GABA. The EC50 for the effective opening rate was 359 µM GABA. There has been one study of recombinant 122 receptors examining macroscopic responses to rapid GABA applications, in which quite similar estimates were obtained ( = 2000 s-1, EC50 = 307 µM; Li & Pearce, 2000). The estimate we made for is also in good agreement with some estimates from analyses of single-channel currents using receptors of either differing or unknown subunit composition (1100-2000 s-1; Weiss & Magleby, 1989; Haas & Macdonald, 1999) and responses to fast agonist applications (2000-6000 s-1; Maconochie et al. 1994; Lavoie et al. 1997). However, we note that some estimates for are lower (e.g. 20 s-1; Twyman et al. 1990). The EC50 is also comparable to previous estimates (300 µM to 1 mM; Maconochie et al. 1994; Lavoie et al. 1997).
There are three components in the open-time distributions for activity elicited by GABA, again as has been described previously (Twyman et al. 1990). In our hands, the three components occurred with similar proportions for activity elicited by 20-1000 µM GABA, so it is likely that none of the components reflect the activity of receptors with only one GABA molecule bound (see also Haas & Macdonald, 1999, page 42). At lower values of [GABA], it is very likely that at least some fraction of the brief-duration openings reflect the activation of receptors with only one GABA molecule bound (Mathers & Wang, 1988; Twyman et al. 1990; G. Akk, unpublished observations).
Activation by -alanine
The relationship between Po and the concentration of agonist was shifted to much higher concentrations for -alanine than for GABA (EC50 13 mM). This shift agrees well with EC50 estimates from macroscopic responses of hippocampal neurones in culture (EC50 for -alanine = 6 mM; Jones et al. 1998). The maximal Po was similar for -alanine and GABA, about 0.8. This also agrees with the finding that the maximal macroscopic responses to GABA and -alanine are quite similar (Jones et al. 1998).
There have been no previous analyses of single-channel activity elicited by -alanine. The distributions of closed times within clusters elicited by -alanine are very similar to those seen with GABA, in terms of the numbers of components, their mean durations and their rate of occurrence. The major difference lies in the effective opening rate. With -alanine, we estimate that = 1500 s-1, similar to the figure obtained for GABA, but the EC50 is shifted to much higher concentrations (19 mM compared to 360 µM). This observation indicates that the major difference between GABA and -alanine lies in the affinity of the two agents (see Jones et al. 1998).
It was surprising that the distributions of open times elicited by -alanine exhibited only two components. The components were comparable in mean duration to the intermediate (OT2) and long (OT3) open times seen when GABA was used. The brief open time (OT1) makes a negligible contribution to the charge transfer during activation. However, the fact that it was not observed with -alanine raises the possibility that the states of the receptor that are commonly reached when -alanine is used as an agonist differ from the major states reached when GABA is used.
Activation by P4S
There have been no previous analyses of single-channel activity elicited by P4S. P4S was an extremely ineffective agonist. In the absence of PB, activity consisted of isolated, brief openings. This observation agrees with the finding that the maximal macroscopic response to P4S is much less than that elicited by GABA (Ebert et al. 1994). We were not able to estimate EC50 values for either Po or the effective opening rate, since we could not resolve clusters. However, our data do not disagree with the observation that P4S activates macroscopic currents with an EC50 comparable to that of GABA (Ebert et al. 1994).
The distributions of open times elicited by P4S exhibited only two components. However, in this case brief and intermediate open times were observed. Again, the fact that only two components were seen with P4S suggests that the states of the receptor that are commonly reached differ from those elicited when GABA is used.
Potentiation by PB
The presence of 40 µM PB had remarkably little effect on the closed times for activity elicited by either GABA or -alanine. In particular, there was at most a small decrease in the estimated channel opening rate, and no apparent change in agonist association or dissociation rates. However, the relationship between Po and agonist concentration was shifted to lower concentrations for both GABA and -alanine. This shift could be entirely accounted for by an increase in the mean open time.
Our observations extend and clarify some previous studies of potentiation by PB. These studies used low values of [GABA] to activate receptors. PB increased the mean open time for channels activated by GABA (Macdonald et al. 1989; Porter et al. 1992; Rho et al. 1996). In terms of the individual open-time components seen in the distributions, PB either acted solely to increase the fraction of openings that fell in the longest of three components, with no change in the mean duration (Macdonald et al. 1989), or to increase both the proportion and duration of the longer of the two components (Porter et al. 1992). There was either no effect on the closed-time distribution (Macdonald et al. 1989), or a prolongation of the four longer of the five components seen in the closed-time distribution (Porter et al. 1992). The data were interpreted to indicate that the action of PB resulted from an increase in the opening rate for the longest-duration open state of the receptor, with minimal effects on the rate for leaving the open state (Macdonald et al. 1989; Porter et al. 1992). The present data were obtained over a wider range of agonist concentrations and utilized several agonists, to more critically evaluate possible changes in channel kinetics.
For both GABA and -alanine, the increase in open time resulted from a change in the duration of the long-duration component (OT3), with an increase in the fraction of openings that fell in the long-duration component. The increased duration was the result of a decrease in the rates of occurrence of all of the closed-time components. That is, it seemed that the long-duration open state was stabilized in the presence of PB. The increase in the fraction of openings in the long duration component is difficult to explain by this single mechanism. However, if the activation scheme for the GABAA receptor includes connections between open states (dotted arrow in Fig. 10B; see Fig. 10A and Weiss & Magleby, 1989), a change in rates between open states could result in an increase in the fraction of openings in the long-duration component. Alternatively, transitions between closed states could increase the occupancy of the closed state from which the long-duration open state occurs (e.g. A2C3 in Fig. 10B) so that the rate of occurrence of the long open state was increased. We think that it is less likely that the opening rate constant for the long-duration open state was increased, since there was no indication in the data that such an increase occurred.
Two other observations require further discussion. These are that PB results in the appearance of an additional brief duration component in the open-time distribution for activity elicited by -alanine and a long duration component for activity elicited by P4S. In the case of P4S, the new component is similar in duration to the long-duration component seen in activity elicited by GABA, and this component is prolonged in a similar fashion by PB. A possible mechanism, which would be consistent with the simple stabilizing action discussed above is that activation by P4S (in the absence of PB) actually results in three-component open-time distributions, but two have indistinguishable mean durations (about 1 ms). PB then acts to reveal the longer-duration component by stabilizing it. This simple idea also has to account for the observation that at low concentrations of PB the openings in the long-duration component are rare, but the mean duration is similar to that seen in the presence of GABA alone. This could mean that the mean duration of the 'stabilized state' (reflecting either occupation by PB or a conformational change) corresponds roughly to the mean duration of the long-duration open component. This argument does not explain the observations made with -alanine as agonist, however. In the case of -alanine, the presence of PB results in the appearance of a brief-duration open state. This is particularly hard to explain, since for activity elicited by GABA PB had essentially no effect on the mean durations of the OT1 and OT2 components. We note, however, that it is difficult to distinguish between multiple exponential components when the time constants differ by less than severalfold. In the case of activity elicited by -alanine, it is possible that there are actually two components (with time constants of 0.4 and about 1.5 ms) that are not resolved. If the duration of the 1.5 ms component were increased by PB, then the brief-duration component might be revealed.
In sum, the mechanism by which PB potentiates the activation of GABAA receptors seems to be very simple in that it appears to involve only changes in the rates for leaving the long-duration open state. This mechanism is adequate to account for all of the effects of PB on activation by the natural transmitter, GABA. However, the actions of PB on currents elicited by -alanine and P4S suggest that the mechanism may be more complex. Two alternative mechanisms immediately come to mind. The first is to say that activation by all three agonists (GABA, -alanine and P4S) actually results in the appearance of three open states that can be distinguished by mean duration in the case of GABA, but are degenerate in the cases of -alanine and P4S. PB then acts to stabilize one open state (the longest-duration state for GABA).
The second explanation is to say that different kinetic schemes are necessary to account for the activity elicited by the three different agonists in the absence of PB. By this we mean that, for example, in the presence of -alanine the rate constants connecting states have values so that the brief-duration open state occurs only very rarely. PB then acts to convert the receptor into a state in which all agonists produce activity resembling that elicited by GABA in the absence of PB (in other words, PB determines the kinetic mode). The action of PB to prolong the duration of the long-duration open state could possibly reflect the same process.
Our data cannot be used to distinguish between the two explanations. If the blocking process could be removed (for example by mutation of the receptor) it would be possible to examine potentiation over a wider range of PB concentrations. Also, the use of rapid concentration jumps would provide insights into the rates for the onset and offset of PB action.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
| AKAIKE, N., HATTORI, K., INOMATA, N. & OOMURA, Y. (1985). -Aminobutyric-acid- and pentobarbitone-gated chloride currents in internally perfused frog sensory neurones. Journal of Physiology 360, 367-386 |
[Abstract] |
| AKAIKE, N., TOKUTOMI, N. & IKEMOTO, Y. (1990). Augmentation of GABA-induced current in frog sensory neurons by pentobarbital. American Journal of Physiology 258, C452-460 | [Medline] |
| AKK, G., BRACAMONTES, J. & STEINBACH, J. H. (2001). Pregnenolone sulfate block of GABAA receptors: mechanism and involvement of a residue in the M2 region of the subunit. Journal of Physiology 532, 673-684 |
[Abstract/Full Text] |
| AKK, G. & STEINBACH, J. H. (2000). Activation and block of recombinant GABAA receptors by pentobarbitone: a single-channel study. British Journal of Pharmacology 130, 249-258 | [Abstract/Full Text] |
| AUSUBEL, F. M., BRENT, R., KINGSTON, R. E., MOORE, D. D., SEIDMAN, J. G., SMITH, J. A. & STRUHL, K. (eds) (1992). Short Protocols in Molecular Biology, 2nd edn. John Wiley, New York | |
| BRICKLEY, S. G., CULL-CANDY, S. G. & FARRANT, M. (1999). Single-channel properties of synaptic and extrasynaptic GABAA receptors suggest differential targeting of receptor subtypes. Journal of Neuroscience 19, 2960-2973 | [Abstract/Full Text] |
| COLQUHOUN, D. & HAWKES, A. G. (1981). On the stochastic properties of single ion channels. Proceedings of the Royal Society B 211, 205-235 | |
| COLQUHOUN, D. & SAKMANN, B. (1981). Fluctuations in the microsecond time range of the current through single acetylcholine receptor ion channels. Nature 294, 464-466 | [Medline] |
| COLQUHOUN, D. & SIGWORTH, F. J. (1995). Fitting and statistical analysis of single-channel records. In Single-Channel Recording, ed. SAKMANN, B. & NEHER, E., pp. 483-589. Plenum, New York | |
| DALZIEL, J. E., COX, G. B., GAGE, P. W. & BIRNIR, B. (1999). Mutant human 11(T262Q) GABAA receptors are directly activated but not modulated by pentobarbital. European Journal of Pharmacology 385, 283-286 |
[Medline] |
| EBERT, B., WAFFORD, K. A., WHITING, P. J., KROGSGAARD-LARSEN, P. & KEMP, J. A. (1994). Molecular pharmacology of -aminobutyric acid type A receptor agonists and partial agonists in oocytes injected with different , , and receptor subunit combinations. Molecular Pharmacology 46, 957-963 |
[Abstract] |
| GINGRICH, K. J., ROBERTS, W. A. & KASS, R. S. (1995). Dependence of the GABAA receptor gating kinetics on the -subunit isoform: implications for structure-function relations and synaptic transmission. Journal of Physiology 489, 529-543 |
[Abstract] |
HAAS, K. F. & MACDONALD, R. L. (1999). GABAA receptor subunit 2 and  subtypes confer unique kinetic properties on recombinant GABAA receptor currents in mouse fibroblasts. Journal of Physiology 514, 27-45 |
[Abstract/Full Text] |
| HAMILL, O. P., MARTY, A., NEHER, E., SAKMANN, B. & SIGWORTH, F. J. (1981). Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches. Pflügers Archiv 391, 85-100 | [Medline] |
| JONES, M. V., SAHARA, Y., DZUBAY, J. A. & WESTBROOK, G. L. (1998). Defining affinity with the GABAA receptor. Journal of Neuroscience 18, 8590-8604 | [Abstract/Full Text] |
| JONES, M. V. & WESTBROOK, G. L. (1995). Desensitized states prolong GABAA channel responses to brief agonist pulses. Neuron 15, 181-191 | [Medline] |
| LAVOIE, A. M., TINGEY, J. J., HARRISON, N. L., PRITCHETT, D. B. & TWYMAN, R. E. (1997). Activation and deactivation rates of recombinant GABAA receptor channels are dependent on -subunit isoform. Biophysical Journal 73, 2518-2526 |
[Abstract] |
| LI, X. & PEARCE, R. A. (2000). Effects of halothane on GABAA receptor kinetics: evidence for slowed agonist unbinding. Journal of Neuroscience 20, 899-907 | [Abstract/Full Text] |
| MACDONALD, R. L., ROGERS, C. J. & TWYMAN, R. E. (1989). Barbiturate regulation of kinetic properties of the GABAA receptor channel of mouse spinal neurones in culture. Journal of Physiology 417, 483-500 | [Abstract] |
| MACONOCHIE, D. J., ZEMPEL, J. M. & STEINBACH, J. H. (1994). How quickly can GABAA receptors open? Neuron 12, 61-71 | [Medline] |
| OLSEN, R. W. & TOBIN, A. J. (1990). Molecular biology of GABAA receptors. FASEB Journal 4, 1469-1480 | [Abstract] |
| PORTER, N. M., ANGELOTTI, T. P., TWYMAN, R. E. & MACDONALD, R. L. (1992). Kinetic properties of 11-aminobutyric acidA receptor channels expressed in Chinese hamster ovary cells: regulation by pentobarbital and picrotoxin. Molecular Pharmacology 42, 872-881 |
[Abstract] |
| QIN, F., AUERBACH, A. & SACHS, F. (1996). Estimating single-channel kinetic parameters from idealized patch-clamp data containing missed events. Biophysical Journal 70, 264-280 | [Abstract] |
| QIN, F., AUERBACH, A. & SACHS, F. (1997). Maximum likelihood estimation of aggregated Markov processes. Proceedings of the Royal Society B 264, 375-383 | |
| RHO, J. M., DONEVAN, S. D. & ROGAWSKI, M. A. (1996). Direct activation of GABAA receptors by barbiturates in cultured rat hippocampal neurons. Journal of Physiology 497, 509-522 | [Abstract] |
| SAKMANN, B., P |
) and presence (
) of 40 µM PB. The smooth curves show the predictions of the Hill equation, with the following best fitting parameter estimates: Po,max = 0.82 ± 0.04 (no PB; best fitting value ± S.D. for the parameter estimated from the fit) and Po,max = 0.83 ± 0.02 (+40 µM PB); EC50 = 70 ± 9 µM and 25 ± 2 µM, respectively, and the Hill coefficient is 1.2 ± 0.2 and 1.3 ± 0.1, respectively. B, data for the effective opening rate (
, open times; 
, open times; 
) and presence (
) of 40 µM PB. Data represented by squares were obtained using