![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 286, Issue 1, 243-255, July 1998
-Aminobutyric Acid Receptor
Function: Role of Protein Tyrosine Phosphorylation and
Calcineurin1
Department of Pharmacology, University of North Texas Health Science Center at Fort Worth, Fort Worth, Texas
 |
Abstract |
|---|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
In the present study, rundown of 
-aminobutyric acid (GABA)-activated
Cl
channels was studied in recombinant GABAA
receptors stably expressed in human embryonic kidney cells (HEK 293),
with conventional whole-cell and amphotericin B-perforated patch
recording. When [ATP]i was lowered to 1 mM and resting
[Ca++]i was buffered to a relatively high
level, the response of 
3 
2 2 GABAA receptors to
relatively low [GABA] (up to 50 µM) did not show rundown in the
whole-cell configuration. However, high [GABA] (greater than 200 µM) induced significant rundown, which was observed by decreases in
both the maximum GABA-induced current and GABA EC50.
Rundown was prevented completely with a solution containing 4 mM
Mg++-ATP and low resting [Ca++]i,
or during perforated patch recording. The magnitude of rundown was
comparable in 1 2 2 and 2 2 receptors. Neither
stimulation nor inhibition of protein kinase A or protein kinase C had
a significant effect on rundown. However, sodium metavanadate, an
inhibitor of protein tyrosine phosphatase, significantly reduced
rundown. In addition, inhibition of protein tyrosine kinase activity by either genistein or lavendustin A induced rundown of the GABA response.
Inhibition of the Ca++/calmodulin-dependent phosphatase
calcineurin with fenvalerate also prevented rundown of the response to
GABA. Our results demonstrate that rundown of GABAA
receptor function is concentration-dependent, due to depletion of ATP
and/or unbuffered [Ca++]i, and does not
depend on the presence or subtype of the alpha subunit.
We propose that protein phosphorylation at a tyrosine kinase-dependent
site, and a distinct unidentified site, which is dephosphorylated by
calcineurin, maintains the function of GABAA receptors.
| |
Introduction |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Rundown,
a phenomenon characterized by a time-dependent decline in ion channel
activity (distinct from desensitization), has been observed during
patch-clamp recording in several ion channels (Armstrong and Eckert,
1987
; Chen et al., 1990; Rosenmund and Westbrook, 1993). The
working hypothesis adopted in most studies is that rundown mainly
reflects a change in the tonic activity of some intracellular channel
modulator(s). Analysis of factors that modulate rundown provides
insight into the mechanisms responsible for the control of receptor or
channel function. Low intracellular [ATP] and high intracellular
[Ca++] have been shown to be involved in the
rundown of several ion channels (Armstrong and Eckert, 1987; Chen
et al., 1990; Rosenmund and Westbrook, 1993).
The Cl channel associated with neuronal
GABAA receptors is one such channel susceptible
to rundown. Reports from several laboratories have indicated that ATP
(Stelzer and Wong, 1988; Gyenes et al., 1988, 1994; Chen
et al., 1990; Shirasaki and Akaike, 1992) and Ca++ (Mouginot and Schlichter, 1991) may
modulate rundown of these channels. The mechanism(s) through which ATP
and Ca++ influence rundown, however, is not
clear. Phosphorylation is a common mechanism for regulation of
receptor/channel function (Levitan, 1994). Previous studies on native
receptors have suggested that ATP limits rundown of the
GABAA receptor by acting as a substrate for
protein phosphorylation (Stelzer and Wong, 1988; Chen et
al., 1990; Gyenes et al., 1994). Several studies have
demonstrated that PKA- and/or PKC-mediated phosphorylation of serine or
threonine residues influences activity of the
GABAA receptor (Sigel et al., 1991;
Moss et al., 1992; Krishek et al., 1994; Lin
et al., 1996). It is reasonable to hypothesize that
disturbance of pertinent phosphorylation factors by the patch pipette
may be responsible for rundown. To date, however, reports on the
functional role of these kinases in regulation of
GABAA receptors have sometimes been in conflict
(Krishek et al., 1994; Lin et al., 1996).
Moreover, there is no direct evidence that phosphorylation of these
residues modulates rundown of the GABAA receptor.
It has been demonstrated that tyrosine phosphorylation, mediated
via PTK, enhances activity of neuronal and recombinant
GABAA receptors (Moss et al., 1995;
Wan et al., 1997). Evidence also has been presented which
suggests that at least some of the PTK effect on the
GABAA receptor (Bureau and Laschet, 1995; Wan
et al., 1997), and other neurotransmitter receptors (Chen
and Leonard, 1996), is caused by endogenously active PTK. However, it
is not known if activity of PTK is necessary to maintain function of the GABAA receptor.
The mechanism of the Ca++-mediated enhancement of
GABAA receptor rundown is also unclear. It has
been proposed that Ca++ could be inducing rundown
by directly inhibiting the receptor/channel complex (Behrends et
al., 1988), or by activating a
Ca++/calmodulin-dependent phosphatase (Chen
et al., 1990). In addition, activation of a
Ca++-dependent kinase that inhibits receptor
function could induce rundown (Krishek et al., 1994).
The human embryonic kidney cell line HEK 293 is used widely for the
expression of cloned channels, in part because this cell line expresses
few endogenous channels (Marshall et al., 1995). The
characteristics of recombinant GABAA receptors
expressed in this system are similar to those seen in native receptors
(Hamilton et al., 1993). Recombinant receptor preparations
offer several advantages to study long-term receptor function. Their
use eliminates possible interaction among receptors/channels that may
exist in neuronal preparations, and allows one to determine whether
rundown is influenced by receptor subunit composition. This in turn
provides information about modulatory sites present on the receptor
subunits. However, there is little information about maintenance of
function in recombinant GABAA receptors.
Therefore, in the present study, we characterized rundown of
recombinant GABAA receptors, and the role of ATP
and Ca++ in this process. In addition, we
examined the mechanisms by which ATP and Ca++
modulate rundown. We report here that ATP maintains long-term receptor
function by promoting activation of an endogenous protein tyrosine
kinase, stimulation of which enhances channel function. Conversely,
Ca++ enhances channel rundown by stimulating
calcineurin. We also demonstrate that rundown is unaffected by change
or deletion of the subunit of the GABAA
receptor.
| |
Materials and Methods |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Cloned GABAA receptors.
Rat
GABAA receptors composed of 3 2 2S, 1
2 2S or 2 2S subunits have been stably expressed in human
embryonic kidney cell lines (HEK293) as described previously (Hamilton
et al., 1993). The cells were transfected with plasmids
containing cDNA and a plasmid encoding G418 resistance. After 2 weeks
of selection in 1 mg/ml G418, resistant cells were assayed by Northern
blotting for the ability to synthesize GABAA
receptor mRNAs. Cells that have been shown to express all subunits were
used for electrophysiology. All studies were conducted on cells
expressing the above receptor configurations.
Electrophysiological recordings.
The conventional whole-cell
configuration of the patch-clamp technique was used to study
GABA-induced Cl currents. Patch pipettes were
pulled from thin-walled borosilicate glass by a horizontal micropipette
puller (Sutter Instrument, P-87) and had a resistance of 1 to 3 megohm
when filled with the following minimal pipette solution (in mM): CsCl,
140; EGTA, 4; Mg++-ATP,1; HEPES, 10; pH 7.2. In
some experiments, Mg++-ATP and/or EGTA were
altered. Free [Ca++]i was
approximated with MaxChelator software (WinMaxc, version 1.70, developed by C. Patton, Stanford University).
5 megohm) was
obtained. The initial control GABA responses were measured 3 min after
stabilization of the response to 20 µM GABA. Fresh pipette solution,
which was kept on ice, maintained effective perforation of the membrane
for about 3 hr. In some perforated-patch recordings, a solution
containing the following was used (in mM): potassium methanesulfonic
acid, 130; KCl, 20; HEPES, 5; and EGTA, 1. Recordings obtained with
either solution produced similar results. pH of all internal solutions
was adjusted to 7.2 to 7.25.
Coverslips containing the cultured cells were transferred to a small
chamber (1 ml) on the stage of an inverted light microscope (Olympus,
IMT-2) and superfused continuously (5-8 ml/min) with the following
external solution (in mM): NaCl, 125; KCl, 5.5; CaCl2, 3.0; MgCl2, 0.8;
HEPES-Na, 20; glucose, 25; pH, 7.3. In the
Ca++-free external solution,
CaCl2 was replaced with an additional 6 mM NaCl.
Whole-cell currents were recorded with an Axopatch 200A amplifier (Axon
Instruments, Foster City, CA) equipped with a CV-4 headstage.
GABA-induced Cl currents were monitored
simultaneously on a storage oscilloscope and a thermal-head pen
recorder, and stored on a computer with an on-line data acquisition
system (pClamp; Axon Instruments). To minimize the possibility that
changes in access resistance could affect current recordings over time,
we measured and stored on our digital oscilloscope, at the initiation
of each recording, the current response to a 5-mV voltage pulse. This
stored trace was referenced continually throughout the recording. If a
change in access resistance was observed throughout the recording
period, the patch was aborted and the data were not included in the
analysis. All recordings were made at room temperature, and with the
exception of collection of current-voltage recordings, all cells were
voltage-clamped at 
60 mV.
Pharmacological agents. Drugs used in the experiments were: GABA, amphotericin B, H-7, NaVO3 (Sigma); D-cAMP, PMA (Research Biochem Int., Natick, MA); fenvalerate, resmethrin, daidzein, lavendustin A and genistein (all from Calbiochem-Novabiochem Co., La Jolla, CA).
Experimental protocol.
GABA was dissolved in the external
solution (above) and applied (10 sec) to the target cell through a
Y-tube positioned within 100 µm of the cell. In experiments where
preincubation with drugs was required, the cells were bathed in the
external solution containing the drug under study at indicated
concentrations for 5 min. Based on our initial characterization of the
recombinant 3 2 2S receptors, 20 µM GABA is approximately
half the EC50 concentration, and 2 mM GABA
induces a maximal GABA response. We thus studied
Cl currents generated in response to these two
concentrations of GABA to examine the potential rundown at both
relatively low and high GABA concentrations. GABA-activated currents
generally were studied during a 60-min recording period, which began
once a consistent response to low GABA (20 µM) was obtained. In
experiments on cells expressing 1 2 2S and 2 2S
receptors, we used 5 and 10 µM GABA as a low concentration, and 500 µM and 2 mM as the maximum concentration, respectively, to activate
the channel.
Data analysis. All data were reported as mean ± S.E.M. Significance was assessed by Student's t test (paired or unpaired). Peak current amplitudes were measured directly from the computer screen by pClamp software.
Current amplitude was expressed as It/Io, where It was the current amplitude recorded at t min after initial recording, and Io was the initial recording. In experiments examining the potential rundown of currents with time, the initial current amplitude in response to 2 mM GABA was assigned a value of 100%. All subsequent currents were expressed as a percentage of this current. The time constant for current decay was obtained by fitting an exponential function to time course-current profiles with the aid of a computer software program (pClamp, Axon Instruments). To construct concentration-response curves, the data were normalized relative to the value obtained at 2 mM GABA (Imax) and fitted with the equation I/Imax = cn/(cn + EC50n), where I/Imax is normalized current, c is GABA concentration, EC50 is the half-maximal effective GABA concentration and n is the Hill coefficient.| |
Results |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Rundown of GABAA receptors is caused by
intracellular dialysis during conventional whole-cell recording.
Initial whole-cell voltage-clamp recordings were obtained from HEK293
cells stably expressing rat 3 2 2S GABAA
receptors. Our initial experiments examined currents elicited by a
relatively low (20 µM) and a high (2 mM) concentration of GABA,
applied in sequence to the same cell at 10-min intervals. Responses to
20 µM GABA were usually nondesensitizing, whereas 2 mM induced rapid desensitization of the GABAA receptor.
current of the same amplitude (101 ± 3.6% of the first response, n = 6). Characteristics
of rundown were similar in cells expressing rat 1 2 2S and
2 2S GABAA receptors. For example, in 1
2 2S receptors, the peak maximum current induced by 500 µM GABA declined to 74 ± 10.2% of the control response 30 min after
initiation of whole-cell patch recording, whereas the response to 5 µM GABA remained unchanged or increased (n = 4).
Maximum currents (2 mM GABA) in receptors composed of only 2 2
subunits ran down to 74.2 ± 8.8% of control maximal current
(n = 4). Thus, rundown apparently does not depend on
changes and/or deletion of the subunit.
|
driving force.
Rundown of other ion channels and neuronal GABAA
receptors can be prevented by use of the perforated patch recording
configuration (Gyenes et al., 1994), which prevents
disruption of the intracellular environment. We tested the possibility
that rundown of recombinant GABAA receptors can
be prevented by use of the amphotericin B perforated patch technique.
As shown in figure 1, C and D, there was no rundown in the response to
either 20 µM or 2 mM GABA during recording for up to 60 min when the
perforated patch method was used. Absence of time-dependent current
decline with the perforated patch recording configuration indicates
that dialysis by the patch pipette of soluble cytoplasmic factors
results in receptor rundown.
The observation that rundown selectively decreases the response to
high, but not low concentrations of GABA indicated that there are
likely changes in both the efficacy and potency of GABA. To examine
this directly, GABA (2-2000 µM) concentration-response curves were
determined for individual cells during the early stage of recording
(within 0-15 min after whole-cell configuration) and after rundown was
fully established (45-60 min). To minimize the possibility of rundown
on the response to high GABA concentration during the first 15 min, we
determined the GABA concentration-response profiles in reverse order
(i.e., from 2000 to 2 µM). GABA applications were
separated by at least 3 min to ensure that receptors had adequate time
to recover from desensitization. Figure
2A shows that rundown is significant at
only the higher concentrations of GABA, whereas no rundown exists up to
200 µM GABA. As a result, the EC50 for GABA was
reduced approximately 3-fold after rundown.
|
Effect of intracellular ATP and Ca++ on
rundown of recombinant GABAA receptors.
Several lines of evidence suggest that rundown of native
GABAA receptors is caused by depletion of
intracellular ATP (Stelzer et al., 1988; Shirasaki et
al., 1992; Gyenes et al., 1988, 1994). To examine this
hypothesis in the recombinant receptors expressed in HEK 293 cells, we
examined the effects of ATP on the response to GABA by altering [ATP]
in the internal solution. Figure 3
exhibits the time course of peak GABA responses recorded in different
Mg++-ATP solutions; all other intracellular
contents were unchanged. Intracellular ATP was varied from 0 to 4 mM,
and resting [Ca++]i was
adjusted to approximately 108 M with a
relatively large concentration of the calcium chelator EGTA (10 mM).
The response to 20 µM GABA either increased, or was unchanged over
time, depending on intracellular [ATP] (fig. 3A). Conversely, it is
evident that exclusion of ATP from the internal solution induced
rundown of maximum GABA-evoked currents (fig. 3B). As intracellular
[ATP] was increased from 0 to 1 mM, the rate and extent of rundown
was retarded. No rundown was evident when [ATP] was increased to 4 mM. These results indicate that some minimal level of intracellular
ATP, approximately 4 mM, is required to maintain the normal
function of recombinant GABAA receptors.
|
; Mouginot et al., 1991; Krishek et al., 1994). We
thus assessed the possible contribution of Ca++
to rundown of recombinant GABAA receptors. When
[Ca++]i was maintained at
a relatively low level (with 10 mM EGTA), no significant rundown of the
response to 2 mM GABA occurred, and the response to 20 µM GABA
increased (fig. 3, C and D). Rundown also was prevented in the 1
2 2 subunit configuration with this internal solution
(n = 6, data not shown). As
[Ca++]i was increased,
rundown of the response to high GABA became noticeably faster, even in
the presence of high ATP (4 mM). Moreover, the response to low GABA
also showed rundown with elevated
[Ca++]i (fig. 3C). Under
these conditions, peak current amplitudes were reduced to 77.5 ± 6.2% and 44.7 ± 4.9% of the original response to 20 µM and 2 mM GABA, respectively (n = 10). There was no
significant difference in time or extent of rundown of the GABA
responses when recordings were obtained in a
Ca++-free medium, excluding the contribution of
extracellular calcium to recombinant GABAA
receptor rundown (n = 10). Thus, in addition to a
requirement for some minimal level of intracellular ATP, complete
GABAA receptor function is maintained over time
only when [Ca++]i is
tightly regulated.
We subsequently tested whether changes in intracellular
Ca++ and/or ATP affected the potency of GABA at
its receptor. Figure 4A demonstrates that
increasing intracellular [ATP] from 1 to 4 mM, while holding
[Ca++]i constant, induced
a roughly 2-fold decrease in EC50 of GABA for the
receptor. Changes in
[Ca++]i had the opposite
effect; increasing
[Ca++]i from roughly
108 M (10 mM EGTA) to roughly
107 M (4 mM EGTA) resulted in a 2-fold
increase in EC50 of GABA for its receptor (fig.
4B).
|
Activity of PKA or PKC does not modulate rundown of the GABA
response.
Based on the results above, it is reasonable to
hypothesize that the role of ATP is to act as a substrate for a
kinase-mediated phosphorylation of the receptor. One possibility is
that ATP is converted to cAMP, which then stimulates cAMP-dependent
kinase (PKA), a known modulator of recombinant and native
GABAA receptors (Moss et al., 1992).
In this scenario, direct stimulation of PKA should block receptor
rundown. Figure 5 shows the effect of
D-cAMP, a cell-permeable cAMP derivative and activator of PKA, on
receptor rundown during conventional whole-cell recording. GABA-induced currents after application of D-cAMP (3 µM) desensitized more rapidly
(see below) but exhibited virtually the same degree of rundown in
response to 2 mM GABA (fig. 5B). D-cAMP also had no effect on the
amplitude of the responses to 20 µM or 2 mM GABA during perforated
patch recording (fig. 5, C and D); the desensitization rate of the GABA
response, however, was enhanced.
|
).
If a similar effect of PKC exists in our preparation, it is reasonable to anticipate that washout of cytoplasmic PKC by the patch pipette during whole-cell recording may result in a decline of GABA-mediated currents. We tested this hypothesis by stimulating PKC during our
normal rundown conditions (1 mM ATP, 4 mM EGTA in the pipette solution). Figure 6A shows the effect of
PMA, a cell-permeable phorbol ester that stimulates PKC, on rundown of
the GABA response during conventional whole-cell recording. PMA (20 nM)
was applied in the bath for 5 min after the onset of rundown. PMA did
not influence the progression or amplitude of rundown. In addition, PMA
could not prevent rundown when it was applied before its onset. To
determine whether the lack of response to PMA may have been caused by
dialysis of PKC during conventional whole-cell recording, the effect of
PKC activation during perforated patch recording also was studied.
Figure 6B illustrates that application of PMA (20 nM) also had no
effect on GABA-induced currents under these conditions. Similar results
were obtained in experiments with a higher concentration of PMA (40 nM). Furthermore, intracellular introduction of 100 µM H-7, an
inhibitor of both PKA and PKC, had no apparent effect on rundown of the
response (fig. 6, C and D). Thus, phosphorylation mediated
via activation of PKC or PKA does not appear to influence
rundown of recombinant 3 2 2S GABAA
receptors.
|
Effect of PTK activity on recombinant GABAA
receptor function.
PTK-dependent phosphorylation recently has been
shown to regulate function of GABAA receptors.
Genistein, a broad spectrum inhibitor of tyrosine kinase (Akiyama
et al., 1987), reduces the GABA response in recombinant
receptors expressed in both HEK293 cells (Moss et al., 1995;
Wan et al., 1997) and Xenopus oocytes (Valenzuela
et al., 1995). In addition, evidence suggests tyrosine residues of bovine cortical GABAA receptors are
phosphorylated endogenously (Bureau and Laschet, 1995). In light of
these studies, we examined the possibility that ATP may inhibit rundown
by acting as a substrate for PTK-mediated phosphorylation.
). As seen in figure
7, NaVO3 (100 µM)
greatly retarded the rate of rundown; 2 mM GABA-induced currents were
not significantly reduced compared with control until the 60-min
timepoint. Moreover, the PTK inhibitor genistein (100 µM in the
pipette) induced a significant decrease in response to 2 mM GABA during
conventional whole-cell recording conditions (4 mM ATP, 10 mM EGTA)
that usually prevent rundown (fig 8).
Under these conditions, addition of intracellular genistein (100 µM)
induced significant rundown (to 58% of control after 60 min) of the
response to 2 mM GABA, whereas the response to 20 µM GABA was
unaffected. During perforated patch recording, GABA current amplitudes
were inhibited similarly by bath perfusion with genistein (fig. 9).
This inhibitory effect of genistein in both recording conditions
was concentration-dependent and reversible. Daidzein (100 µM), an
analog of genistein that lacks PTK inhibitory activity (Akiyama
et al., 1987; Wan et al., 1997), had
no effect on GABA currents (fig. 8C). Genistein (100 µM, bath
application) also inhibited 50 µM GABA-induced current (to 47.8 ± 4.9% of the control, P < .01) in recombinant rat 1 2
2S GABAA receptors (n = 5).
Because genistein also inhibited the activity of other protein kinases
(Steadman et at., 1996), we examined the effects of lavendustin A, a
highly selective inhibitor of PTK that has little effect on PKA and PKC
(Hsu et al., 1991). Run-up of the 20 µM GABA response
persisted with inclusion of 5 µM Lavendustin A in the whole-cell
recording solution (fig. 8A). However,
Lavendustin A induced a significant rundown of the response to 2 mM
GABA (fig. 8B). These data suggest that recombinant
GABAA receptors may require PTK to be active for
maintenance of complete receptor function.
|
|
|
Role of Ca++/calmodulin-dependent
phosphatase (calcineurin) in the GABA response.
We next sought to
determine the mechanism through which Ca++
facilitates rundown of the receptor. A potential mechanism for the
effect of Ca++ in the regulation of
GABAA receptor, with which both our
Ca++ and ATP data are consistent, is activation
of the Ca++/calmodulin-dependent phosphatase
calcineurin. We tested this possibility with fenvalerate (120 nM), a
potent cell-permeable calcineurin inhibitor (Enan and Matsumura, 1992).
Figure 10 shows that in the presence of
fenvalerate rundown of the response to 2 mM GABA was prevented
completely during conditions that would otherwise induce rundown.
Resmethrin (120 nM), a negative control for studies of calcineurin
inhibition with fenvalerate (Enan and Matsumura, 1992), did not alter 2 mM GABA-induced rundown significantly. These data support the
hypothesis that Ca++ facilitates
GABAA receptor rundown by activation of
calcineurin.
|
Effect of rundown and phosphorylation state on desensitization
kinetics.
There is some enhancement of desensitization during
rundown of neuronal GABAA receptors (Gyenes
et al., 1988, 1994). Gyenes et al. reported that
current decay during perforated patch recording was considerably slower
than that obtained during conventional whole-cell patch recording. In
addition, stimulation of PKA (Tehrani et al., 1989) and PKC
(Leidenheimer et al., 1992) may influence GABAA receptor desensitization kinetics. We thus
analyzed desensitization kinetics of recombinant receptors during
different recording conditions, and the influence of PKA, PKC and PTK
on desensitization. Time constants for initial GABA (2 mM)-induced
current decay were similar (5.6 ± 0.6 vs. 6.1 ± 1.6 sec) during conventional whole-cell and perforated patch recording
conditions. During rundown conditions (whole-cell configuration, 1 mM
ATP, 4 mM EGTA), we noted a tendency for a slowing of the decay time
constant, although this effect was significant only at 40 min (to
7.3 ± 1.1 sec). When recordings were obtained with a non-rundown
pipette solution (4 mM ATP and 10 mM EGTA), time constants for current
decay tended to decrease with time, from 5.8 ± 0.7 at control to
4.7 ± 0.8 and 4.2 ± 0.6 sec at 50 and 60 min, respectively
(P < .05). Stimulation of PKC with the phorbol ester PMA (20 nM)
did not affect desensitization. However, exogenous addition of D-cAMP,
during either conventional whole-cell or perforated patch recording
(fig. 5D), reversibly enhanced desensitization. Decaying currents in
the presence of D-cAMP were best fitted with a biexponential function,
with the fast component typically about 1 sec. PTK activity also may
influence desensitization kinetics, because the inhibitory effect of
100 µM genistein on the current amplitudes induced by 2 mM GABA (fig. 8B) was accompanied by a reversible increase in time constant (from
5.3 ± 1.2 to 7.7 ± 1.2 P < .05; then recovered to
4.4 ± 1.6 sec). These data indicate that desensitization of
GABA-activated currents may be enhanced by intracellular activation of
protein phosphorylation via cAMP-dependent kinase and/or
protein tyrosine kinase. Washout of these phosphorylation factors
during whole-cell recording may contribute to the decline in
desensitization rate of GABA-activated currents.
| |
Discussion |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Characteristics of rundown in recombinant
GABAA receptors.
To better understand
long-term maintenance of GABAA receptor function,
we studied the mechanism of rundown in recombinant GABAA receptors stably expressed in HEK 293 cells. Our data show that the GABA response runs down as early as 10 min during conventional whole-cell recording with 1 mM ATP and 4 mM
EGTA in the internal solution. Significant rundown occurred only at
relatively high concentrations of GABA, with an approximately 30%
reduction in maximum current amplitude 60 min after the initial
response. We also determined that rundown of 1 2 2 and 2
2 receptors was similar to that seen in 3 2 2 receptors,
which indicates that rundown does not depend on the subunit. At a
relatively low concentration of GABA (20 µM), current amplitudes
tended to "run-up." Run-up at low concentrations also was observed
with native GABAA receptors (Gyenes et
al., 1994). Although we did not attempt to fully characterize
run-up conditions, our data did permit a few conclusions. Although some
minimal level of ATP and a low level of Ca++ was
necessary for run-up to occur, the magnitude of run-up was not related
linearly to either [ATP] or [Ca++] (fig. 3, A
and C). Thus, additional factors, not examined it this study, also
apparently influence run-up.
, 1994) on rundown of the GABA response in spinal cord neurons.
Our results support their argument that rundown modifies the
GABAA receptor so that GABA potency is increased,
whereas efficacy is decreased. The mechanism for this effect, however,
is not clear. It does not relate to the presumed depletion of ATP
during rundown, as a reduction in intracellular [ATP] decreases GABA
potency, as shown in this (fig. 4) and other studies (Shiraski et
al., 1992; Akaike, 1995). It might be argued that the increased
potency after rundown is caused by chelation of
Ca++ that may have been liberated during initial
formation of the whole-cell patch configuration. This scenario is
consistent with the effect of Ca++ on GABA
EC50 (fig. 4). However, when one considers the
overall phenomenon of rundown, and the prominent role of
Ca++ in rundown, it is highly unlikely that
[Ca++]i is decreasing
throughout the recording period. Thus, the mechanism for the change in
GABA potency after rundown remains to be determined.
Whole-cell recordings from neuronal GABAA
receptors (Stelzer and Wong, 1988; Gyenes et al., 1988,
1994; Chen et al., 1990; Shirasaki and Akaike, 1992) display
a greater degree of rundown than we report here for recombinant
receptors. Although the exact reason for this has not been defined, one
possibility is that there are differences in endogenous phosphatase
and/or kinase activity in the two preparations. Alternatively, it may
relate to factors other than regulation of the receptor itself. For
instance, compared with HEK cells, the neuron is relatively complex;
maintenance of its function requires regulation of many other proteins
that in turn may influence the neuron's ability to respond to GABA. This further illustrates the utility of recombinant expression systems
such as the HEK cells for studying isolated receptor regulation, without potential confounding factors that may be present in neurons.
Maintenance of GABAA receptor function by
PTK.
There is evidence that PTK and PTP are key enzymes in
signal-transduction pathways for a wide range of cellular processes. The function of NMDA receptors is enhanced by tyrosine phosphorylation (Wang et al., 1994; Wang and Salter, 1994; Chen and Leonard,
1996). Recent work showed that protein tyrosine phosphorylation also may enhance GABAA receptors, because the
whole-cell GABA response was increased by exogenous tyrosine
phosphorylation (Moss et al., 1995; Wan et al.,
1997) and depressed by genistein in HEK 293 cells (Wan et
al., 1997), cortical microsacs (Valenzuala et al., 1995) and Xenopus oocytes (Valenzuala et al.,
1995). In addition, biochemical studies have shown that purified brain
GABAA receptors can be phosphorylated by an
endogenously active tyrosine kinase (Bureau and Laschet, 1995).
Preliminary experiments in our laboratory indicate that genistein also
inhibits posterior hypothalamic GABAA receptors
recorded in thin brain slices.
), interpretation of results with use of this compound alone as a PTK inhibitor is difficult. Thus, we also performed
experiments with lavendustin A, a more specific inhibitor of PTK (Enan
and Matsumura, 1992). Lavendustin A had a rundown-inducing effect
similar to did genistein. Neither of these compounds induced rundown of
the response to low [GABA]. In fact, some degree of run-up, similar
to that seen with control pipette solution, was observed with
lavendustin A. This suggests that modulation by PTK may depend on
receptor state. Nonetheless, these results together indicate that one
of the previously unidentified kinases that maintains long-term
function of GABAA receptors (Stelzer et
al., 1988; Gyenes et al., 1988, 1994; Chen et
al., 1990), and thus prohibits rundown, is PTK. This tyrosine
kinase is endogenously active, as blockade of the corresponding
tyrosine phosphatase by the PTP inhibitor NaVO3
prevented rundown.
The exact site of action of this tyrosine kinase that maintains
GABAA receptor function has not been determined.
Gyenes et al. (1994) proposed that factors required to
maintain the GABA response must be associated closely with the plasma
membrane, because the rundown could be reversed by adding ATP--S in
excised, outside-out patches. The subunit that is phosphorylated by PTK has been investigated recently by a few laboratories. Our present results are consistent with the hypothesis that tyrosine kinase modulation of rundown is not the result of an effect on the subunit. Recently, Wan et al. (1997) proposed that in
cultured central nervous system neurons and recombinant 1 2 2
or 2 2 GABAA receptors transiently
expressed in HEK 293 cells, the possible endogenous tyrosine kinase
that modulates receptor function does so via phosphorylation of the
2 subunit. Conversely, Moss et al. (1995) suggested
exogenous vSRC-mediated modulation of GABAA
receptor function is caused by phosphorylation of the 2 subunit.
Phosphorylation by exogenous vSRC of tyrosine residues on both 1 and
2L subunits of bovine GABAA receptors also has been observed (Valenzuala et al., 1995), whereas endogenous
tyrosine phosphorylation of 2 subunits of cortical
GABAA receptors has been reported by others
(Bureau and Laschet, 1995). The relative importance of these tyrosine
sites, with regard to maintenance of GABAA
receptor function, requires additional investigation.
Activity of PKA or PKC does not influence
GABAA receptor rundown.
Most
GABAA receptor subunits contain consensus
sequences for kinases such as cAMP-dependent protein kinase (PKA), and
or Ca++-phospholipid -dependent PKC (Sigel
et al., 1991; Moss et al., 1992; Krishek et
al., 1994; Lin et al., 1996). Purified preparations of
the GABAA receptors can be phosphorylated
in vitro by PKA (Browning et al., 1990; Tehrani
and Barnes, 1994) and PKC (Browning et al., 1990).
Furthermore, Bureau and Laschet (1995) found multiple endogenous kinase
activities on an and possibly subunit. However, the results
concerning the role of these kinases in regulation of the
GABAA receptors have been complex and sometimes
contradictory (Ticku and Mehta, 1990; Moss et al., 1992;
Feigenspan and Bormann, 1994; Krishek et al., 1994; Lin
et al., 1996).
; Shiraski et al., 1992). We did find
that desensitization kinetics are enhanced by D-cAMP. Consistent with
this, inhibition of PKA by H-7 tended to slow decay kinetics. In light
of these results, it seems likely that the dialysis of PKA by the
whole-cell pipette also may contribute to the decreased time constant
for current decay noted during the whole-cell recording period.
We found no effect of stimulation or inhibition of PKC during either
conventional whole-cell or amphotericin B-perforated patch recording.
Our results are in contrast to those of Lin et al. (1996),
who found that PKC enhances bovine 1 1 2L receptors transiently expressed in L929 fibroblasts. Different experimental preparations may partly explain this difference. Our results actually are supported by those of Krishek et al. (1994), who studied
effects of PKC on 1 2 2 receptors expressed in HEK293 cells.
Although they reported that PKC inhibits GABA-activated current, they
found no effect of PKC stimulation unless they also cotransfected PKC along with the GABAA receptor subunits. Our
results, where PKC was not cotransfected, are consistent with those of
Krishek et al. (1994). Thus, PKC does not appear to
contribute to maintaining the normal function of recombinant 3 2
2 GABAA receptors stably expressed in HEK293
cells.
Ca++ enhances rundown through stimulation
of calcineurin.
Our results presented here show that fenvalerate,
a specific inhibitor of the
Ca++/calmodulin-dependent phosphatase
calcineurin, prevented rundown of the GABA response during conditions
when [Ca++]i was
elevated. Other studies have shown that stimulation of phosphatase in
general (Chen et al., 1990; Gyenes et al., 1994; Wang et al., 1994), or a
Ca++-dependent phosphatase in particular (Chen
et al., 1990; Lieberman and Mody, 1994), regulates channel
function. Introduction of intracellular alkaline phosphatase caused a
complete rundown of the GABA response in hippocampal (Chen et
al., 1990) and spinal cord neurons (Gyenes et al.,
1994). The influence of Ca++-dependent
phosphatase on NMDA channel function was demonstrated by Lieberman and
Mody (1994), who reported that inhibition of calcineurin enhanced
activity of NMDA current. Chen et al. (1990) reported that
inhibition of calmodulin slowed GABAA receptor
rundown. The interpretation of these experiments is difficult, however, because calmodulin has many cellular effects in addition to stimulation of calcineurin (Braun and Schulman, 1995). In the present experiments, we used a specific inhibitor of calcineurin to demonstrate conclusively that Ca++ facilitates rundown through activation
of an endogenous Ca++/calmodulin-dependent
phosphatase (calcineurin).
, we propose
the following model for GABAA receptor function (fig. 11). With adequate ATP present,
GABAA receptors are maintained in a maximally
functional state by a phosphorylation process involving protein
tyrosine kinase and an as yet unidentified kinase. Dephosphorylation of
tyrosine residues is mediated via protein tyrosine
phosphatase. This dephosphorylation, in conjunction with
dephosphorylation of serine/threonine residues catalyzed by the
Ca++/calmodulin-dependent phosphatase
calcineurin, decreases function of the GABAA
receptor. The corresponding protein kinase that phosphorylates Ser/Thr
residues could not be identified in this study, but does not appear to
be PKA or PKC.
|
| |
Acknowledgments |
|---|
We thank Dr. Donald Carter for supplying the cell lines used in this study and Ms. Cathy Bell-Horner for her technical assistance.
| |
Footnotes |
|---|
Accepted for publication March 6, 1998.
Received for publication January 15, 1998.
1 Supported by National Institutes of Health grant ES07904, American Heart Association, Texas Affiliate grant 95G-158 and Texas Advanced Research Program grant 009768-027.
Send reprint requests to: Glenn H. Dillon, Ph.D., Dept. of Pharmacology, University of North Texas Health Science Center at Fort Worth, 3500 Camp Bowie Blvd., Fort Worth, TX 76107.
| |
Abbreviations |
|---|
GABA, -aminobutyric acid;
GABAA, type A GABA receptor;
D-cAMP, adenosine 3',5'-cyclic
monophosphate N6O6 dioctanoyl sodium salt;
EGTA, ethylene glycol-bis(-aminoethyl ether)-N,N,N',N'-tetraacetic
acid;
H-7, (5-iosquinolinylsulfonyl)-2 methyl-piperazine
dihydrochloride;
HEK, human embryonic kidney;
HEPES, N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid;
NaVO3, sodium metavanadate;
NMDA, N-methyl-D-aspartate;
PKC, protein kinase C;
PKA, protein kinase A;
PMA, phorbol 12-myristate
13-acetate;
PTK, protein tyrosine kinase;
PTP, protein tyrosine
phosphatase.
| |
References |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
subunit of the purified -aminobutyric acid A receptor.
Proc Natl Acad Sci USA
87:
1315-1318-aminobutyric acid type A receptor polypeptides by ser/thr and tyr kinase activities associated with the purified receptor.
J Biol Chem
279:
26482-26487.-aminobutyric acidA receptor function during whole-cell recording: a possible role for phosphorylation.
Mol Pharmacol
34:
719-723[Abstract].-aminobutyric acid type A receptor.
Mol Pharmacol
46:
542-549[Abstract].-aminobutyric acidA receptor: role of desensitization.
Mol Pharmacol
41:
1116-1123[Abstract].112L -aminobutyric acidA receptor whole-cell currents by protein kinase C is mediated through phosphorylation of both 1 and 2L subunits.
Mol Pharmacol
50:
185-195[Abstract].-Aminobutyric acidA receptor desensitization in mice spinal cord cultured neurons: lack of involvement of protein kinases A and C.
Mol Pharmacol
38:
719-724[Abstract]. subunits.
J Neurosci
17:
5062-5069This article has been cited by other articles:
|
|
 |
|
  Z.-H. Qi, M. Song, M. J. Wallace, D. Wang, P. M. Newton, T. McMahon, W.-H. Chou, C. Zhang, K. M. Shokat, and R. O. Messing Protein Kinase C{epsilon} Regulates {gamma}-Aminobutyrate Type A Receptor Sensitivity to Ethanol and Benzodiazepines through Phosphorylation of {gamma}2 Subunits J. Biol. Chem., November 9, 2007; 282(45): 33052 - 33063. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. M. Sanchez, W. Dai, R. E. Levada, J. J. Lippman, and F. E. Jensen AMPA/Kainate Receptor-Mediated Downregulation of GABAergic Synaptic Transmission by Calcineurin after Seizures in the Developing Rat Brain J. Neurosci., March 30, 2005; 25(13): 3442 - 3451. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Palma, G. Torchia, C. Limatola, F. Trettel, A. Arcella, G. Cantore, G. Di Gennaro, M. Manfredi, V. Esposito, P. P. Quarato, et al. BDNF modulates GABAA receptors microtransplanted from the human epileptic brain to Xenopus oocytes PNAS, February 1, 2005; 102(5): 1667 - 1672. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. J. Laschet, F. Minier, I. Kurcewicz, M. H. Bureau, S. Trottier, F. Jeanneteau, N. Griffon, B. Samyn, J. Van Beeumen, J. Louvel, et al. Glyceraldehyde-3-Phosphate Dehydrogenase Is a GABAA Receptor Kinase Linking Glycolysis to Neuronal Inhibition J. Neurosci., September 1, 2004; 24(35): 7614 - 7622. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D D Wang, D D Krueger, and A Bordey GABA Depolarizes Neuronal Progenitors of the Postnatal Subventricular Zone Via GABAA Receptor Activation J. Physiol., August 1, 2003; 550(3): 785 - 800. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Herring, R. Huang, M. Singh, L. C. Robinson, G. H. Dillon, and N. J. Leidenheimer Constitutive GABAA Receptor Endocytosis Is Dynamin-mediated and Dependent on a Dileucine AP2 Adaptin-binding Motif within the {beta}2 Subunit of the Receptor J. Biol. Chem., June 20, 2003; 278(26): 24046 - 24052. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Alix, F. Grolleau, and B. Hue Ca2+/Calmodulin-Dependent Protein Kinase Regulates GABA-Activated Cl- Current in Cockroach Dorsal Unpaired Median Neurons J Neurophysiol, June 1, 2002; 87(6): 2972 - 2982. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Cuppoletti, K. P. Tewari, A. M. Sherry, E. Y. Kupert, and D. H. Malinowska ClC-2 Cl{-} channels in human lung epithelia: activation by arachidonic acid, amidation, and acid-activated omeprazole Am J Physiol Cell Physiol, July 1, 2001; 281(1): C46 - C54. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Rusnak and P. Mertz Calcineurin: Form and Function Physiol Rev, October 1, 2000; 80(4): 1483 - 1521. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Herzig and J. Neumann Effects of Serine/Threonine Protein Phosphatases on Ion Channels in Excitable Membranes Physiol Rev, January 1, 2000; 80(1): 173 - 210. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R.-Q. Huang and G. H. Dillon Effect of Extracellular pH on GABA-Activated Current in Rat Recombinant Receptors and Thin Hypothalamic Slices J Neurophysiol, September 1, 1999; 82(3): 1233 - 1243. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Verrecchia, F. Duthe, S. Duval, I. Duchatelle, D. Sarrouilhe, and J. C. Herve ATP counteracts the rundown of gap junctional channels of rat ventricular myocytes by promoting protein phosphorylation J. Physiol., April 15, 1999; 516(2): 447 - 459. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Poisbeau, M. C. Cheney, M. D. Browning, and I. Mody Modulation of Synaptic GABAA Receptor Function by PKA and PKC in Adult Hippocampal Neurons J. Neurosci., January 15, 1999; 19(2): 674 - 683. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
