Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 1999 Jul 15;518(Pt 2):385–399. doi: 10.1111/j.1469-7793.1999.0385p.x

Evidence for phosphorylation-dependent internalization of recombinant human ρ1 GABAC receptors

Natalia Filippova 1, Richard Dudley 1, David S Weiss 1
PMCID: PMC2269426  PMID: 10381587

Abstract

  1. Recombinant wild-type or mutant human ρ1 GABA receptors were expressed in human embryonic kidney (HEK) 293 or monkey COS-7 cells and studied using the patch clamp technique.

  2. Standard whole-cell recordings with 4 mM Mg-ATP in the patch pipette induced a time-dependent decrease in the GABA-activated current (IGABA) amplitude that was not the result of a decrease in GABA sensitivity. In contrast, IGABA remained stable when recordings were obtained using the perforated patch configuration or with standard whole-cell recording and no Mg-ATP in the patch pipette.

  3. The inhibitors of serine/threonine protein kinases KN-62 (20 μM) or staurosporine (20 nM) prevented the time-dependent decrease in the amplitude of IGABA seen in the presence of ATP. Alkaline phosphatase (220 U ml−1), when added to the patch pipette in the absence of ATP, induced a transient potentiation of IGABA. Although the protein kinase C (PKC) activator 4β-phorbol 12-myristate, 13-acetate (PMA) did not reduce the amplitude of IGABA, inclusion of the catalytic domain of PKC in the recording pipette accelerated the time-dependent decrease in current amplitude. These data suggest that phosphorylation is involved in the regulation of the amplitude of IGABA.

  4. Mutation of the three PKC consensus sequences of the ρ1 receptor had no significant effect on the decline in IGABA, indicating that direct phosphorylation of these putative sites on the ρ1 receptor does not underlie the time-dependent decrease in amplitude.

  5. In COS-7 cells transfected with wild-type ρ1 receptors, the amplitude of IGABA had completely recovered to the original value when the same cells were repatched after 30-40 min, indicating that the decline in IGABA was a reversible process.

  6. The inhibitor of actin filament formation cytochalasin B, when added to the patch pipette in the absence of ATP, induced a time-dependent inactivation suggesting that the actin cytoskeleton may play a role in the regulation of the amplitude.

  7. Coincident with the decrease in the amplitude of IGABA, the cell capacitance significantly decreased in the presence of ATP in the patch pipette. This decrease in capacitance was not observed in the absence of Mg-ATP. The decrease in the membrane surface area suggests that receptor internalization could be a potential mechanism for the observed inactivation.

  8. At 32 °C, compared with 22 °C, the rate and magnitude of the decline was increased dramatically. In contrast, at 16 °C, no significant change in IGABA was observed over the 20 min recording time. This marked temperature sensitivity is consistent with receptor internalization as a mechanism for the time-dependent decline in IGABA.

  9. The specificity of the decrease in IGABA was assessed by coexpressing the voltage-dependent potassium channel Kv1.4 along with the ρ1 receptor in HEK293 cells. The amplitude of the potassium current (IKv1.4) exhibited very little decrement in comparison to IGABA suggesting that the putative GABA receptor internalization was not the consequence of a non-specific membrane retrieval.

GABAC receptors, presumably consisting of rho subunits (Cutting et al. 1991; Wang et al. 1994), are a class of inhibitory GABA-activated channels in the central nervous system (Polenzani et al. 1991; Feigenspan et al. 1993; Qian & Dowling, 1994; Strata & Cherubini, 1994; Enz et al. 1996). Similar to GABAA receptors, they possess a high permeability to Cl−, but in contrast to GABAA channels, they are insensitive to bicuculline, barbiturates and benzodiazepines (Polenzani et al. 1991). The activity of GABAC receptors is regulated by extracellular agents, such as Zn2+, H+, Ca2+ (Wang et al. 1995; Kaneda et al. 1997), and also by intracellular factors, such as Ca2+, phosphatases and kinases (Feigenspan & Bormann, 1994b; Kusama et al. 1995). Protein phosphorylation is postulated to be an important physiological mechanism for regulating GABA-mediated synaptic inhibition (Moss et al. 1992; Moss & Smart, 1996).

Neuronal GABAC channel currents demonstrate a Ca2+- and ATP-dependent decrease in amplitude during prolonged whole-cell recording (Feigenspan & Bormann, 1994b). In addition, the protein kinase C (PKC) activator 4β-phorbol 12-myristate, 13-acetate (PMA) decreased the amplitude of GABA-activated currents from homomeric ρ1 receptors expressed in Xenopus oocytes (Kusama et al. 1995; Chapell et al. 1998). Consensus sites for phosphorylation are located in the large intracellular loop between the putative third and fourth transmembrane domains of the ρ1 homomeric GABA receptor (Cutting et al. 1991), suggesting that direct phosphorylation of the ρ1 receptor could be a potential mechanism for regulation of the current amplitude. Alternatively, the modulation could be due to a PKC-dependent alteration in other proteins that interact with, and regulate, the ρ1 receptor. The present study was designed to distinguish between these and other possible regulatory mechanisms of the ρ1 receptor.

ρ1 receptors were expressed in human embryonic kidney (HEK) 293 or monkey COS-7 cells and examined with whole-cell recording techniques. When ATP was included in the recording pipette, a time-dependent decrement in the amplitude of the GABA-activated current (IGABA) was observed. The decrement in amplitude was not observed in the absence of ATP or with perforated patch recordings. Although kinase inhibitors prevented the decline in IGABA, and the presence of the catalytic domain of PKC in the patch pipette hastened the decline, elimination of the three PKC consensus sites did not prevent the decrease indicating that direct phosphorylation of these particular residues is not required for the modulation. Experiments examining the actions of cytochalasin B, as well as those monitoring the cell capacitance throughout the recordings, suggest a phosphorylation-dependent receptor internalization via interaction with the cytoskeleton as a mechanism of the time-dependent inactivation of IGABA.

METHODS

Molecular biology

The human ρ1 subunit was obtained via the polymerase chain reaction as described previously (Amin & Weiss, 1994) and subcloned into the pALTER-1 vector (Promega, Madison, WI, USA) for site-directed mutagenesis using Altered Sites (Promega). Mutagenesis was verified by cDNA sequencing and both the wild-type and mutant ρ1 receptors were subcloned into pCDNA3 (Invitrogen, San Diego, CA, USA). The human Kv1.4 clone was kindly provided by C. Garner (UAB, Birmingham, AL, USA) in the pCMV vector.

Cell culture and transfection

HEK293 cells (ATCC CRL 1573) or COS-7 cells (ATCC CRL 1651) were plated onto poly-L-lysine-coated glass coverslips (12 mm) placed in a 35 mm culture dish and maintained in Dulbecco's modified Eagle's medium-Ham's F12 (DMEM-F12; Gibco BRL, Gaithersburg, MD, USA) (HEK293) or DMEM (COS-7) supplemented with 10 % fetal bovine serum (Atlanta Biologicals, Norcross, GA, USA). The following day, transient transfections were performed using Lipofectamine (Gibco BRL) in the same media without serum or antibiotic according to the manufacturer's protocol. The ρ1-containing plasmid was cotransfected with the green fluorescent protein (GFP)-containing plasmid pGreenLantern (Gibco BRL) in order to enable visual identification of transfected cells. A total of 2 μg of cDNA was used per 35 mm plate; typically 1:1 for the ρ1:pGreenLantern plasmids. Cells were maintained in normal serum-containing medium and recordings were made 2-4 days post transfection.

The HEK293 cell line that stably expressed ρ1 receptors was produced by carrying out a standard transfection as described above and performing cell passages for 1-2 months in the presence of 400 mg l−1 of G-418 (Gibco BRL). Cells were then plated at a low density and islands of cells were removed with a pipette after light trypsinization. These clonal cell lines were then tested for ρ1 receptor expression. Stable cells were maintained in 200 mg l−1 G-418.

Electrophysiology

Experiments were performed at room temperature (20-24°C) using the perforated patch and whole-cell recording techniques. Cells were visualized with an inverted microscope equipped with fluorescence. Currents were recorded using an Axopatch-1D amplifier (Axon Instruments, Foster City, CA, USA) and digitized (5-10 kHz) on a Macintosh computer using Pulse (Heka Electronik, Lambrecht, Germany). Analysis was carried out with Igor (Wavemetrics Inc., Lake Oswego, OR, USA). The external recording solution contained (mM): NaCl, 140; KCl, 3·5; glucose, 10; CaCl2, 2; Hepes, 10 (pH 7·2). The recording borosilicate glass pipettes had resistances of 2-4 MΩ when filled with the internal solution containing (mM): CsCl, 130; CaCl2, 0·25; EGTA, 1·1 (free Ca2+, ∼5 × 10−8 M); Hepes, 10 (pH 7·4). When indicated, 4 mM Mg-ATP, 220 U ml−1 alkaline phosphatase, 20 nM staurosporine, 20 μM KN-62, 0·4-4 nM of the catalytic domain of PKC, or 2 μg ml−1 cytochalasin B was added to the recording pipette. Staurosporine, KN-62 and cytochalasin B were dissolved in DMSO and diluted 1000-fold prior to use. The holding potential was -50 mV in all experiments. GABA (from 0·5 to 20 μM), dissolved in external solution, was applied to the cells through a double-barrel fast piezoelectric perfusion system. In some experiments, bicuculline (10 μM) was also added to the external solution. For the temperature experiments, the perfusion solution was warmed using a TC-344A heater controller (Warner Instruments, Hamden, CT, USA) or cooled by passing the inlet line through a cold water bath. In both cases, the temperature was monitored in the bath just beyond the target cell. For the experiments examining activation of PKC by PMA, HEK293 cells were switched to serum-free medium 24 h prior to the addition of PMA. PMA (0·5-0·8 μM) was introduced to the cells 1 h prior to recording.

Perforated patch recordings were performed using amphotericin B (240 mg ml−1) in the pipette solution containing (mM): NaCl, 140; KCl, 3·5; glucose, 10; Hepes, 10 (pH 7·2). Amphotericin B was dissolved in DMSO and diluted 1000-fold for use. The tip of the recording pipette was dipped for ∼1 s in control solution (without amphotericin) before the pipette was backfilled with the amphotericin B solution. Access resistances typically stabilized within 3-15 min after the patch formation and reached 15-40 MΩ. The value of the access resistance was estimated using a 10 mV depolarizing pulse. Changes in the cell capacitance were approximated by integrating the capacitive current transients in response to this 10 mV voltage pulse (Lindau & Neher, 1988).

The current amplitude and cell capacitance were normalized to the initial value (30-60 s from the begining of recording) in each of the experiments. All results are presented as means ±s.e.m. Data were compared statistically by Student's t test. Statistical significance was determined at the 5 % level.

The experiments examining the temperature dependence of the decrease in IGABA and the effects of PKC in the patch pipette were carried out on a HEK293 cell line stably expressing ρ1 receptors. The properties of the receptors in the stable cell line were indistinguishable from those of the transiently expressed ρ1 receptors, but the expression level was less variable and the magnitude of the time-dependent decline in IGABA over time was slightly less than that observed in the transient transfections.

Currents from Kv1.4 channels were examined by applying a 200 ms voltage step from -80 to +40 mV. The leak was determined by stepping the membrane potential from a holding potential of -40 mV to +40 mV. The plotted amplitude (IKv1.4) was the peak of the leak subtracted current trace.

Drugs

Cytochalasin B, staurosporine, KN-62, and alkaline phosphatase were obtained from RBI (Natick, MA, USA), bicuculline and amphotericin B were obtained from Sigma, and PMA and the catalytic domain of protein kinase C were obtained from Calbiochem (San Diego, CA, USA).

RESULTS

Basic properties of recombinant ρ1 receptors expressed in HEK293 cells

Figure 1A shows typical GABA-activated whole-cell currents (IGABA) recorded at a range of membrane potentials from HEK293 cells expressing recombinant homomeric ρ1 GABA receptors. As previously demonstrated, ρ1 GABA-activated currents exhibited little desensitization with prolonged GABA application. With an IGABA greater than 2 nA, however, we typically observed a slow decay during continuous GABA application. This decay was associated with a slight shift in the reversal potential (not shown), suggesting a change in the intracellular Cl− concentration. Voltage ramps indicated that the current-voltage (I-V) relationship was linear (Fig. 1A; I-V plot). The 10-90 % rise times were 1132 ± 80 and 495 ± 41 ms (n= 5 cells) for currents activated by 10 and 20 μM GABA, respectively (not shown). The time constant of decay upon GABA removal (deactivation) was 9·4 ± 0·9 s (n= 23; Fig. 1B).

Figure 1. Properties of wild-type recombinant ρ1 receptors transiently expressed in HEK293 cells.

Figure 1

Open in a new tab

A, responses to GABA (10 μM, 20 s duration) applied through a fast perfusion system at the indicated holding potentials. The whole-cell current-voltage relationship (I-V), as determined with voltage ramps, was linear. B, decay of the current upon GABA removal (deactivation) was well described by a single exponential component with a time constant (τ) of 11 s. In 23 cells, the time constant of decay was 9·4 ± 0·9 s. C, whole-cell currents evoked at a holding potential of -50 mV with GABA (0·5, 1, 5 and 10 μM) application through a fast perfusion system (top) and normalized concentration-response curve for GABA-evoked currents (bottom). Hill, Hill coefficient. D, bicuculline (10 μM) did not block the response to GABA (10 μM, 20 s duration).

Figure 1C shows currents activated by 0·5, 1, 5 and 10 μM GABA and the corresponding dose-response relationship. The continuous line is the best fit of the Hill equation to these data and revealed an EC50 for GABA (concentration required for half-maximal activation) of 0·90 ± 0·09 μM and a Hill coefficient of 3·0 ± 0·5 (n= 3).

Recombinant homomeric ρ1 receptors are insensitive to the GABAA receptor antagonist bicuculline. Figure 1D demonstrates that 10 μM bicuculline had no effect on GABA-activated currents from HEK293 cells transfected with ρ1 receptors (n= 3), indicating the lack of endogenous GABAA currents that are sometimes observed in HEK293 cells (Ueno et al. 1996). Thus, the data in Fig. 1 demonstrate that the main properties of recombinant ρ1 receptors expressed in HEK293 cells are similar to those of recombinant ρ1 receptors expressed in oocytes (Cutting et al. 1991; Amin & Weiss, 1994) as well as those of GABAC receptors in neurons (Feigenspan & Bormann, 1994a; Qian & Dowling, 1994).

Inactivation of ρ1 receptors during whole-cell recording

Figure 2A shows GABA-activated currents measured with standard whole-cell recording techniques during a 25 min recording period with 4 mM Mg-ATP in the patch pipette. Figure 2C shows a plot of the mean amplitude of IGABA normalized to that at the initial application of GABA. Note that the amplitude declined to 0·56 ± 0·09 (n= 6) of the initial value during the first 15 min of recording and then stabilized. The ratios of the current amplitudes with 1 and 5 μM GABA after 1 and 15 min of whole-cell recording were 0·74 ± 0·14 and 0·71 ± 0·10, respectively (Fig. 2B, n= 3). Thus, the decline in the amplitude of IGABA was not the result of a decrease in agonist sensitivity.

Figure 2. ATP in the recording pipette induced inactivation of GABA-activated currents in HEK293 cells expressing ρ1 receptors during whole-cell recording.

Figure 2

Open in a new tab

A, currents evoked by GABA (20 μM) application through a fast perfusion system decreased by ≈50 % during 15 min of whole-cell recording in the presence of ATP (4 mM) in the patch pipette. In B, the traces represent the currents activated by 1 and 5 μM GABA at different times during the recording in the presence of ATP in the patch pipette. The ratios of current amplitudes activated by 1 and 5 μM GABA were 0·62 and 0·68 before and after inactivation of the current amplitude, respectively. C, mean normalized GABA-activated current amplitude in the presence of ATP during prolonged recording. In six cells, the amplitude fell to 0·56 ± 0·09 during 15 min of recording.

Wild-type ρ1 GABA-activated currents are stable with perforated patch recording or in the absence of Mg-ATP in the patch pipette

HEK293 cells expressing homomeric ρ1 receptors were voltage clamped using the perforated patch technique and 10 μM GABA was applied at 5 min intervals. As illustrated in Fig. 3A, the amplitude of IGABA was stable during prolonged recording (the normalized amplitude was 1·05 ± 0·04 after 25 min of recording; n= 5). Figure 3B shows a plot of the amplitude of IGABA (filled circles) and the access resistance (Raccess; open circles), both of which remained stable over this time period. Thus, the amplitude of IGABA in HEK293 cells transfected with ρ1 receptors and recorded using the perforated patch technique remained relatively constant during long term recording.

Figure 3. The amplitude of the GABA-activated current remained stable during perforated patch recording or without ATP in the recording pipette.

Figure 3

Open in a new tab

A, currents evoked by GABA application through a fast perfusion system to cells expressing wild-type ρ1 receptors at different times during perforated patch recording. B, mean GABA-activated current amplitude (•) and access resistance (○) during prolonged recording. Note that the amplitude and the access resistance remained stable. The normalized values were 1·05 ± 0·04 and 0·98 ± 0·12 (n= 5) after 25 min of recording, respectively. C, currents evoked by GABA application through a fast perfusion system to cells expressing ρ1 receptors at different times during standard whole-cell recording and without ATP in the recording pipette. D, mean GABA-activated current amplitude in the absence of ATP in the recording pipette (•). ○indicates the mean amplitude in the presence of ATP (Fig. 2C) replotted for comparison. Note that, in the absence of ATP, the current amplitude remained stable for the first 20 min of recording.

Similarly, without the addition of Mg-ATP to the patch pipette (Fig. 3C), the current amplitude did not change significantly during 15 min of recording (0·97 ± 0·09 of the initial value, n= 7; Fig. 3D, filled circles). In most cells, a decrease in the amplitude began to develop after 20 min of recording. The open circles in Fig. 3D are the data in the presence of ATP replotted for comparison. The requirement for ATP in this time-dependent inactivation of IGABA suggests phosphorylation as a possible mechanism.

Serine/threonine-dependent phosphorylation is involved in the regulation of ρ1 receptors

To gain further insight into the mechanism of this ATP-dependent inactivation of IGABA, we examined the effects of KN-62 (20 μM), an inhibitor of Ca2+-calmodulin (CaM)-dependent protein kinase, and staurosporine (20 nM), an inhibitor of PKC. As shown in Fig. 4A and B and the plot of IGABA over time (Fig. 4D), these protein kinase inhibitors prevented the ATP-dependent decrease in the amplitude of IGABA. The normalized amplitudes were 0·97 ± 0·1 (n= 5) and 0·96 ± 0·16 (n= 5) after 15 min of recording for KN-62 and staurosporine, respectively. Alkaline phosphatase (220 U ml−1), added to the recording pipette without ATP, induced an initial (although transient and highly variable) potentiation of IGABA during the first 10 min of recording (Fig. 4C and D). The data in Fig. 4 implicate PKC- and Ca2+-CaM-dependent phosphorylation as candidate pathways for the ATP-dependent inactivation of homomeric ρ1 GABA receptors.

Figure 4. PKC and Ca2+-CaM-dependent phosphorylation may be involved in the modulation of the amplitude of IGABA.

Figure 4

Open in a new tab

In A-C, traces represent currents evoked by GABA (20 μM) application through a fast perfusion system to cells expressing wild-type ρ1 receptors in the presence of KN-62 (20 μM; A), staurosporine (20 nM; B) and alkaline phosphatase (220 U ml−1; C). Note that, in all cases, the GABA-evoked current did not decrease during 25 min of recording. D, mean normalized GABA-activated current amplitudes in the presence of staurosporine (▪), KN-62 (○) or alkaline phosphatase (•) in the recording pipette.

In order to examine whether the PKC pathway was involved in the ATP-dependent inactivation, we compared the amplitude of IGABA with and without a 1 h incubation in the presence of the PKC activator PMA (0·5-0·8 μM). The IGABA amplitudes were 796 ± 180 pA (n= 6) and 747 ± 70 pA (n= 6), with and without PMA, respectively. We also tested the effects of including the catalytic domain of PKC (0·4-4 nM) in the patch pipette on the ATP-dependent decline in IGABA. Figure 5A and B shows examples of IGABA throughout a 20 min recording period in the absence and presence, respectively, of 1 nM PKC. The mean peak values of IGABA from five cells with and without PKC in the recording pipette are plotted in Fig. 5C. The presence of PKC increased the ATP-dependent decrease in IGABA. The amplitude of IGABA decreased to 0·77 ± 0·05 (n= 5) of the original value in comparison with a decrease to 0·46 ± 0·08 (n= 5) in the presence of PKC. The experiments in which the HEK293 cells were incubated in PMA suggest that either PKC was already maximally stimulated or the PKC pathway may not be directly involved in the ATP-dependent decline. Nevertheless, the enhanced decline in the presence of PKC supports the conclusion that a phosphorylation-dependent mechanism could underlie the time-dependent decrease in IGABA.

Figure 5. Inclusion of the catalytic domain of PKC in the patch pipette enhanced the decline in IGABA.

Figure 5

Open in a new tab

A, GABA-activated currents in a HEK293 cell stably expressing ρ1 receptors over a 20 min recording period. B, same as in A, but 1 nM of the catalytic domain of PKC was included in the recording pipette. Note the enhanced decline in IGABA in the presence of PKC. C, mean normalized GABA-activated current amplitudes in the absence and presence of PKC in the patch pipette. After 20 min IGABA decreased to 0·77 ± 0·05 (n= 5) of its original value while, in the presence of PKC, IGABA decreased to 0·46 ± 0·08 (n= 5) of its original value.

Phosphorylation of the PKC consensus sites S410, S419 and S426 is not required for the IGABA inactivation

The ρ1 receptor has three serine residues in the intracellular loop between the putative third and fourth transmembrane domains that are potential PKC phosphorylation sites (Cutting et al. 1991). One possible mechanism for the ATP-dependent decline in IGABA might involve direct phosphorylation of the ρ1 subunit. To evaluate whether phosphorylation of these sites may be directly involved in the ATP-dependent inactivation of GABA receptor function, we examined recombinant ρ1 receptors in which all three serines were mutated to alanine (S/410-419-426/A). Figure 6A shows GABA-activated currents from HEK293 cells transfected with the triple mutant. Note that, similar to the wild-type receptor (Fig. 2A), the current continuously decreased during 20 min of recording. The filled circles in Fig. 6B show IGABA normalized to the value at the initial GABA application. The dashed line and open symbols are the wild-type data (Fig. 2C) replotted for comparison. The current amplitude of the mutant decreased to 0·59 ± 0·09 (n= 7) of the initial value during 25 min of recording. These results indicate that phosphorylation of the serine residues S410, S419 or S426 is not necessary for the ATP-dependent decrease in the amplitude of IGABA.

Figure 6. Mutation of the PKC consensus sites on the ρ1 receptor did not prevent the inactivation of IGABA.

Figure 6

Open in a new tab

A, currents evoked by GABA (10 μM) applications to HEK293 cells expressing the triple mutant ρ1 receptors (S/410-419-426/A) in the presence of ATP (4 mM) in the intracellular solution during whole-cell recording. Note that the current amplitude decreased ≈33 % during 20 min of whole-cell recording. B, mean normalized GABA-activated current amplitude of the triple mutant ρ1 receptors in the presence of ATP in the intracellular solution (•). Note the significant decrease (0·59 ± 0·09, n= 7, P < 0·05) in the current amplitude during 25 min of recording. ○, current amplitude of the wild-type ρ1 receptor replotted for comparison.

The time-dependent inactivation of ρ1 receptors is a reversible process

Next, we set out to determine whether, after declining, the amplitude of IGABA could recover back to its original value. Reversibility would place some constraints on potential mechanisms. These particular experiments were carried out in COS-7 cells, rather than in HEK293 cells, since (at least in our hands) the COS-7 cells were much more easily repatched. In the presence of ATP, the amplitude of the GABA-induced current continuously dropped to 0·60 ± 0·06 (n= 7) of the initial value during 30 min of recording (first four current traces in Fig. 7A and the filled circles in Fig. 7B). Thus, the features of the time-dependent inactivation of IGABA in COS-7 cells were qualitatively similar to those observed in HEK293 cells.

Figure 7. The ATP-dependent inactivation of IGABA is reversible in COS-7 cells transfected with wild-type ρ1 receptors.

Figure 7

Open in a new tab

A, traces are GABA-activated currents during whole-cell recording. The trace on the right is the GABA-activated current when the same cell was repatched 30 min after the initial recording was terminated. Note that during 28 min of initial recording the amplitude decreased ≈38 % and then fully recovered to its original value. B, the normalized GABA-activated current amplitude continuously decreased during prolonged recording in the presence of ATP in the intracellular solution. Upon repatching (after 30 min) the amplitude had recovered to its original value and then decayed again. C, mean GABA-evoked current amplitudes normalized to the amplitude at the first minute of initial recording. Note that upon repatching after 25-35 min from the end of the initial recording, the current amplitude had returned to 0·94 ± 0·07 of the original value.

We then repatched the same cell after 30 min from the end of the initial whole-cell recording (right-hand trace in Fig. 7A). In this experiment, the current amplitude dropped to almost half during 38 min of initial recording and recovered to nearly its original value after repatching, before declining again (Fig. 7B, open circles). Figure 7C shows the mean results from four such experiments. IGABA decreased to 0·55 ± 0·06 (n= 4) during 30-40 min of initial recording and recovered to 0·94 ± 0·07 (n= 4) of the initial value after repatching. A recovery in the amplitude was not observed if we maintained the initial recording for 15-20 min after inactivation, but without GABA application (n= 2; data not shown). These data eliminate irreversible mechanisms such as receptor degradation for the decline in the amplitude of IGABA.

The actin cytoskeleton may be involved in the regulation of ρ1 GABA receptors

The cytoskeleton has been shown to interact with, and regulate, many different types of ion channels. One possible explanation of our results thus far is that the ρ1 receptor interacts with cytoskeletal elements in a phosphorylation-dependent manner and this interaction regulates the ρ1 receptor secondarily. To test for a potential involvement of the cyoskeleton, the actin depolymerizing agent cytochalasin B (2 μg ml−1) was included in the patch pipette. Cytochalasin B induced a significant decrease in the amplitude of IGABA (0·65 ± 0·03; n= 5) during 15 min of whole-cell recording without ATP in the intracellular solution (Fig. 8A and B, filled circles). The open circles and dashed line in Fig. 8B are the wild-type data in the absence of ATP replotted for comparison (Fig. 3D). These data indicate that the actin cytoskeleton may play a role in the ATP-dependent inactivation of IGABA.

Figure 8. Cytochalasin B induced a time-dependent inactivation of IGABA amplitude in HEK293 cells expressing wild-type ρ1 receptors.

Figure 8

Open in a new tab

A, currents evoked by GABA (10 μM) applications through a fast perfusion system at different times during whole-cell recording in the presence of cytochalasin B (ATP free) in the recording pipette. B, mean amplitude of IGABA during prolonged recording with cytochalasin B in the patch pipette (•). Note that, in contrast to recordings in the absence of cytochalasin B (○, replotted from Fig. 3D), the amplitude decreased (0·65 ± 0·03, n= 5, P < 0·05) during 15 min of recording.

Internalization of ρ1 receptors is a potential mechanism of ATP-dependent inactivation

One possibility is that the ATP-dependent inactivation of IGABA could be due to an internalization of ρ1 receptors. This receptor internalization would probably involve vesicle endocytosis and might therefore be associated with an observable decrease in the membrane surface area. To test this hypothesis, we estimated the cell capacitance during whole-cell recording. Note that, in the presence of ATP, the cell capacitance (Fig. 9, filled circles) continuously decreased concomitantly with the decrease in the amplitude of IGABA (Fig. 9, filled bars). However, in the absence of ATP, the cell capacitance remained stable (Fig. 9, open circles; 1·03 ± 0·02, n= 7). These data indicate a concomitant decrease in the amplitude of IGABA and the surface area of the membrane and support internalization of ρ1 receptors as a possible mechanism of the time-dependent inactivation of IGABA.

Figure 9. The membrane capacitance decreased during the inactivation of IGABA.

Figure 9

Open in a new tab

Mean membrane capacitance during prolonged whole-cell recording in the presence (•) and absence (○) of ATP in the recording pipette. A representative capacitive transient from the -50 to -40 mV voltage step is shown at the top. The normalized amplitude of IGABA with ATP in the recording pipette is plotted as filled bars. Note that, in the presence of ATP, the normalized amplitude and the normalized cell capacitance concomitantly decreased during 20 min of recording (0·65 ± 0·06 and 0·80 ± 0·06, respectively; n= 5, P < 0·05).

The ATP-dependent inactivation of IGABA is highly temperature dependent

One hallmark of receptor internalization by an endocytotic mechanism is a marked temperature sensitivity (Silverstein et al. 1977; Weigel & Oka, 1981). We therefore examined inactivation at temperatures both above and below room temperature. Figure 10A shows representative GABA-activated currents in a HEK293 cell expressing ρ1 receptors at 22 and 34°C, then upon return to 22°C. Note the slight increase in amplitude as well as the faster deactivation rate at the more elevated temperature. Figure 10B and C shows representative GABA-activated currents at 32 and 22°C, respectively, recorded over a 20 min time period. A plot of the time dependence of the amplitude of IGABA for recordings obtained at 16, 22 and 32°C is shown in Fig. 10D. Both the rate of the ATP-dependent decline in current amplitude (time constants of 35 and 5·5 min for 22 and 32°C, respectively) as well as the final fractional inactivation were highly correlated to the recording temperature, as would be expected for an endocytotic mechanism.

Figure 10. Temperature dependence of the decline in IGABA.

Figure 10

Open in a new tab

A, GABA-activated currents in a HEK293 cell stably expressing ρ1 receptors at 22 and 34 °C, then upon return to 22 °C. Note the increase in amplitude and deactivation rate (τ) at 34 °C. B and C, GABA-activated currents obtained at 32 and 22 °C, respectively, during 20 min of recording. Note the increased decline observed at the higher recording temperature. D, mean amplitude of IGABA over time at 16, 22 and 32 °C. At the lower temperature, there was no decline in IGABA after a 20 min recording period with a normalized amplitude of 1·03 ± 0·06 (n= 4). In contrast, IGABA decreased to 0·75 ± 0·05 (n= 4) and 0·51 ± 0·05 (n= 5) at 22 and 32 °C, respectively. Fitting an exponential function to the mean decay revealed time constants of 35 and 5·5 min for 22 and 32 °C, respectively.

The putative internalization is specific for GABA receptors

It is possible that ATP induces a general membrane internalization and that the GABA receptors are thereby non-specifically internalized during the process. To test this hypothesis we coexpressed the voltage-dependent potassium channel Kv1.4 with ρ1 GABA receptors and the results are presented in Fig. 11. Representative Kv1.4 channel currents, and the method of leak subtraction, are shown in Fig. 11A. Figure 11B shows Kv1.4 and ρ1 channel currents in the same HEK293 cell at 1 and 20 min after the start of recording. The mean time course of the current amplitudes is plotted in Fig. 11C. While IGABA decreased to 0·66 ± 0·04 (n= 4) over the 20 min recording time, IKv1.4 decreased to only 0·89 ± 0·06 of its original value (n= 4). Note that by 10 min IGABA was reduced to near the plateau level whereas there was no significant decrement in IKv1.4. These experiments rule out a non-specific membrane internalization since, assuming a homogeneous distribution of both membrane proteins, a comparable rate and magnitude of decrement would be expected for IGABA and IKv1.4. It appears that ρ1 recepors are specifically removed, in a phosphorylation-dependent manner, from the cell surface.

Figure 11. Comparison of the amplitude of IKv1.4 and IGABA over time.

Figure 11

Open in a new tab

A, Kv1.4 receptors were expressed transiently in a HEK293 cell line stably expressing ρ1 receptors. A family of voltage steps was applied to activate the K+ channels (left traces). The middle trace shows a voltage step from -80 to -40 mV (a) superimposed on a voltage step from -40 to +40 mV (b) to determine the leak. The amplitude was the difference of these two currents (right-hand trace). B, representative Kv1.4 and ρ1 channel currents in a single HEK293 cell at 1 and 20 min after the start of recording. Note that, in this particular cell, IGABA decreased by nearly 50 % after 20 min of recording compared with an ≈10 % decrease in the amplitude of the Kv1.4 channel current. C, mean amplitude of IGABA and IKv1.4 over time. After 20 min, IKv1.4 decreased to 0·89 ± 0·06 compared with 0·66 ± 0·04 for IGABA (n= 4). Note that by 10 min IGABA was near its plateau value whereas IKv1.4 was essentially unchanged.

DISCUSSION

Whole-cell recordings from HEK293 cells transiently expressing ρ1 GABA receptors demonstrated a gradual decline in the amplitude of IGABA. We have provided evidence that this decline depends on phosphorylation, although it does not involve phosphorylation of the ρ1 receptor at the three originally identified PKC consensus sequences (Cutting et al. 1991). Studies investigating the actions of a cytoskeletal disrupting agent and measurements of the cell capacitance during whole-cell recording suggest that phosphorylation-dependent alterations in the interactions of the ρ1 receptor with the cytoskeletal network, leading to receptor internalization, may underlie the decline in the amplitude of IGABA.

Phosphorylation-dependent regulation of GABA receptors

Regulation by phosphorylation has been extensively studied in native and recombinant (αβγ) GABAA receptors; however, the results of these studies have been variable and, in some cases, contradictory (Moss & Smart, 1996). The amplitudes of GABA-activated current in neurones from the spinal cord, hippocampus, cerebellum and retina demonstrated a time-dependent decline that was prevented by the inclusion of Mg-ATP in the patch pipette (Gyenes et al. 1988; Stelzer et al. 1988; Robello et al. 1993; Gillette & Dacheux, 1996). Recombinant α1β1γ2L receptors expressed in HEK293 cells and GABAA (mRNA injection) or recombinant α1β1γ2S receptors expressed in Xenopus oocytes demonstrated a reduction in the amplitude of IGABA in response to PKC activators (Sigel & Baur, 1988; Kellenberger et al. 1992; Krishek et al. 1994). In contrast, introduction of catalytically active PKC into L929 fibroblasts expressing recombinant α1β1γ2L receptors caused an enhancement of IGABA (Lin et al. 1994, 1996). The first report on modulation of recombinant GABAA receptors by cAMP-dependent phosphorylation (protein kinase A (PKA)) demonstrated a direct phosphorylation of the β1 subunit associated with a decrease in the amplitude of IGABA (Moss et al. 1992). Nevertheless, opposing effects have also been observed for PKA-dependent regulation of both native and recombinant GABAA receptors (Harrison & Lambert, 1989; Tehrani et al. 1989; Porter et al. 1990; Ticku & Mehta, 1990; McDonald et al. 1998). It is not clear whether these differences in the modulation of GABAA receptors relate to the choice of expression system or the particular methods used to manipulate the kinase pathway (Moss & Smart, 1996).

Whatever the direction of the phosphorylation-dependent modulation of GABAA receptors (potentiation versus inhibition), our studies on the ρ1 receptor show two key differences. First, while GABAA receptors demonstrated a gradual decline in the amplitude of IGABA that was slowed by the inclusion of ATP in the patch pipette (Gyenes et al. 1988; Stelzer et al. 1988; Robello et al. 1993), the decline of ρ1 GABA-activated currents was facilitated by the inclusion, and prevented by the exclusion, of ATP in the patch pipette. Thus, intracellular application of ATP seems to have opposing effects on the two different classes of GABA receptors. Native GABAC receptors in rat retinal bipolar cells demonstrate a decline in the amplitude of IGABA that was prevented by PKC inhibitors and enhanced by PKC activators (Feigenspan & Bormann, 1994b), indicating that our results are not an anomaly of recombinant receptor expression in HEK293 cells.

A second difference between studies of αβγ and ρ1 GABA receptors is that PKA- or PKC-dependent modulation of αβγ GABA receptor function involved direct phosphorylation of the GABA receptor. In contrast, the phosphorylation-dependent decline in the amplitude of IGABA in the present study was not prevented by mutation of the three PKC consensus sequences. (Thus far, we have not observed any functional effects with elimination of the three PKC consensus sequences, and with expression of the mutant in HEK293 cells, COS-7 cells or Xenopus oocytes.) We cannot rule out the possibility that there are other potential sites on the ρ1 receptor (Kennelly & Krebs, 1991) that may be phosphorylated and play a role in the ATP-dependent decline of IGABA. Clearly, a thorough biochemical analysis of phosphorylation sites, similar to that carried out on the β2 and β3 subunits (McDonald & Moss, 1997), is needed for the ρ1 recptor. Nevertheless, many retinal cells express both GABAA and GABAC (ρ1) receptors and these differences in modulation could reflect an essential feature for differentially regulating the two inhibitory pathways (Yeh et al. 1990; Feigenspan et al. 1993; Feigenspan & Bormann, 1994b).

The cytoskeletal environment is important for ρ1 receptor function

The actin-based cytoskeleton plays an essential role in the regulation of cell shape, cell motility, and the function/distribution of integral membrane proteins (Lynch & Baudry, 1987; Mammen et al. 1997). Cytoskeletal proteins have been localized with GABAA receptors, NMDA receptors, glycine receptors, nicotinic acetylcholine (nACh) receptors, cyclic nucleotide-gated channels and potassium channels (Qu et al. 1996; Kannenberg et al. 1997). In addition, disruption of actin filaments has been shown to alter the properties of amiloride-sensitive sodium channels, GABAA receptors, nACh receptors and cyclic nucleotide-gated channels (Connolly, 1984; Rosenmund & Westbrook, 1993; Gordon et al. 1995; Berdiev et al. 1996). It is noteworthy that GABAC receptors colocalize with syntaxin 3 (Koulen et al. 1998), a cytosolic protein that also binds actin filaments.

In our experiments, both ATP and cytochalasin B (in the absence of ATP) caused a gradual reduction in the amplitude of IGABA. Mg-ATP has been shown to facilitate the polymerization of actin filaments (Bennet & Weeds, 1986; Pollard & Cooper, 1986) and might therefore be expected to have an effect opposite to that of cytochalasin B, that is to stabilize actin filaments and oppose the time-dependent decrease in amplitude. However, PKC and Ca2+-CaM-dependent protein kinase have been implicated in the control of the structure of the cytoskeletal network (Prekeris et al. 1996; Audesirk et al. 1997). Therefore, the regulation of kinase pathways, either via ATP application or the use of kinase inhibitors, could alter the properties of the network of actin filaments which in turn could alter the interaction of the cytoskeletal environment with the ρ1 receptors. Thus, the stabilization of actin filaments by ATP and the phosphorylation-dependent alteration in the cytoskeletal network could exert opposing effects on the regulation of IGABA.

In summary, we propose the following mechanism for ATP-dependent modulation of the ρ1 GABA receptor. Recombinant ρ1 GABA receptors normally interact with the actin cytoskeletal network via a protein matrix. The phosphorylation of one or more proteins alters the interaction of the ρ1 receptor with the cytoskeletal network, which ultimately leads to the internalization of ρ1 receptors. This pool of internalized receptors can be readily retargeted to the cell surface. In a similar fashion, activation of the insulin receptor in hippocampal neurones can translocate GABAA receptors between an intracellular compartment and the cell surface (Wan et al. 1997). In this manner, insertion and retrieval of GABA receptors to and from the plasma membrane may be a common and powerful mechanism for regulating inhibitory synaptic transmission.

Acknowledgments

We thank R. A. J. Lester for helpful comments. The work was supported by NIH grants NS35291 and NS36195 and a grant from the W. M. Keck Foundation (931360).

References

  1. Amin J, Weiss DS. Homomeric ρ1 GABA channels: activation properties and domains. Receptors and Channels. 1994;2:227–236. [PubMed] [Google Scholar]
  2. Audesirk G, Cabell L, Kern M. Modulation of neurite branching by protein phosphorylation in cultured rat hippocampal neurons. Developmental Brain Research. 1997;102:247–260. doi: 10.1016/s0165-3806(97)00100-4. [DOI] [PubMed] [Google Scholar]
  3. Bennet J, Weeds A. Calcium and the cytoskeleton. British Medical Bulletin. 1986;42:385–390. doi: 10.1093/oxfordjournals.bmb.a072156. [DOI] [PubMed] [Google Scholar]
  4. Berdiev B, Prat A, Cantiello H, Ausiello D, Fuller C, Jovov B, Benos D, Ismailov I. Regulation of epithelial sodium channels by short actin filaments. Journal of Biological Chemistry. 1996;271:17704–17710. doi: 10.1074/jbc.271.30.17704. [DOI] [PubMed] [Google Scholar]
  5. Chapell R, Bueno OF, Alvarez-Hernandez X, Robinson LC, Leidenheimer NJ. Activation of protein kinase C induces γ-aminobutyric acid type A receptor internalization in Xenopus oocytes. Journal of Biological Chemistry. 1998;273:32595–32601. doi: 10.1074/jbc.273.49.32595. [DOI] [PubMed] [Google Scholar]
  6. Connolly JA. Role of the cytoskeleton in the formation, stabilization, and removal of acetylcholine receptor clusters in cultured muscle cells. Journal of Cell Biology. 1984;99:148–154. doi: 10.1083/jcb.99.1.148. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Cutting GR, Lu L, O'Hara BF, Kasch LM, Montrose-Rafizadeh C, Donovan DM, Shimada S, Antonarakis SE, Guggino WB, Uhl GR, Kazazian HH. Cloning of the γ-aminobutyric acid (GABA) ρ1 cDNA: A GABA receptor subunit highly expressed in the retina. Proceedings of the National Academy of Sciences of the USA. 1991;88:2673–2677. doi: 10.1073/pnas.88.7.2673. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Enz R, Brandstatter JH, Wassle H, Bormann J. Immunocytochemical localization of the GABAc receptor rho subunits in the mammalian retina. Journal of Neuroscience. 1996;16:4479–4490. doi: 10.1523/JNEUROSCI.16-14-04479.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Feigenspan A, Bormann J. Differential pharmacology of GABAA and GABAC receptors on rat retinal bipolar cells. European Journal of Pharmacology. 1994a;288:97–104. doi: 10.1016/0922-4106(94)90014-0. [DOI] [PubMed] [Google Scholar]
  10. Feigenspan A, Bormann J. Modulation of GABAC receptors in rat retinal bipolar cells by protein kinase C. The Journal of Physiology. 1994b;481:325–330. doi: 10.1113/jphysiol.1994.sp020442. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Feigenspan A, Wassle H, Bormann J. Pharmacology of GABA receptor Cl− channels in rat retinal bipolar cells. Nature. 1993;361:159–162. doi: 10.1038/361159a0. [DOI] [PubMed] [Google Scholar]
  12. Gillette MA, Dacheux RF. Protein kinase modulation of GABAA currents in rabbit retinal rod bipolar cells. Journal of Neurophysiology. 1996;76:3070–3086. doi: 10.1152/jn.1996.76.5.3070. [DOI] [PubMed] [Google Scholar]
  13. Gordon SE, Downing-Park J, Zimmerman AL. Modulation of the cGMP-gated ion channel in frog rods by calmodulin and an endogenous inhibitory factor. The Journal of Physiology. 1995;486:533–546. doi: 10.1113/jphysiol.1995.sp020832. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Gyenes M, Farrant M, Farb DH. ‘Run down’ of γ-aminobutyric acidA receptor function during whole-cell recording: a possible role for phosphorylation. Molecular Pharmacology. 1988;34:719–723. [PubMed] [Google Scholar]
  15. Harrison NL, Lambert NA. Modification of GABAA receptor function by an analog of cyclic AMP. Neuroscience Letters. 1989;105:137–142. doi: 10.1016/0304-3940(89)90025-6. [DOI] [PubMed] [Google Scholar]
  16. Kaneda M, Mochizuki M, Aoki K, Kaneko A. Modulation of GABA (C) response by Ca2+ and other divalent cations in horizontal cells of the catfish retina. Journal of General Physiology. 1997;110:741–747. doi: 10.1085/jgp.110.6.741. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Kannenberg K, Baur R, Sigel E. Proteins associated with alpha1-subunit-containing GABAA receptors from bovine brain. Journal of Neurochemistry. 1997;68:1352–1360. doi: 10.1046/j.1471-4159.1997.68041352.x. [DOI] [PubMed] [Google Scholar]
  18. Kellenberger S, Malherbe P, Sigel E. Function of the α1β2γ2S γ-aminobutyric acid type A receptor is modulated by protein kinase C via multiple phosphorylation sites. Journal of Biological Chemistry. 1992;267:25660–25663. [PubMed] [Google Scholar]
  19. Kennelly PJ, Krebs EG. Consensus sequences as substrate specificity determinants for protein kinases and protein phosphatases. Journal of Biological Chemistry. 1991;266:15555–15558. [PubMed] [Google Scholar]
  20. Koulen P, Brandstatter JH, Enz R, Bormann J, Wassle H. Synaptic clustering of GABAC receptor ρ-subunits in the rat retina. European Journal of Neuroscience. 1998;10:115–127. doi: 10.1046/j.1460-9568.1998.00005.x. [DOI] [PubMed] [Google Scholar]
  21. Krishek B, Xie X, Blackstone C, Huganir R, Moss S, Smart T. Regulation of GABAA receptor function by protein kinase C phosphorylation. Neuron. 1994;12:1081–1095. doi: 10.1016/0896-6273(94)90316-6. [DOI] [PubMed] [Google Scholar]
  22. Kusama T, Sakurai M, Kizawa Y, Uhl GR, Murakami H. GABA rho 1 receptor: inhibition by protein kinase C activators. European Journal of Pharmacology. 1995;291:431–434. doi: 10.1016/0922-4106(95)90086-1. [DOI] [PubMed] [Google Scholar]
  23. Lin Y-F, Angelotti TP, Dudek EM, Browning MD, Macdonald RL. Enhancement of recombinant α1β1γ2L γ-aminobutyric acidA receptor whole-cell currents by protein kinase C is mediated through phosphorylation of both β1 and γ2L subunits. Molecular Pharmacology. 1996;50:185–195. [PubMed] [Google Scholar]
  24. Lin Y-F, Browning MD, Dudek EM, Macdonald RL. Protein kinase C enhances recombinant bovine α1β1γ2L GABAA receptor whole-cell currents expressed in L929 fibroblasts. Neuron. 1994;13:1421–1431. doi: 10.1016/0896-6273(94)90427-8. [DOI] [PubMed] [Google Scholar]
  25. Lindau M, Neher E. Patch-clamp techniques for time-resolved capacitance measurements in single cells. Pflügers Archiv. 1988;411:137–146. doi: 10.1007/BF00582306. [DOI] [PubMed] [Google Scholar]
  26. Lynch G, Baudry M. Brain spectrin, calpain and long-term changes in synaptic efficacy. Brain Research Bulletin. 1987;18:809–815. doi: 10.1016/0361-9230(87)90220-6. [DOI] [PubMed] [Google Scholar]
  27. McDonald BJ, Amato A, Connolly CN, Benke D, Moss SJ, Smart TG. Adjacent phosphorylation sites on GABAA receptor β subunits determine regulation by cAMP-dependent protein kinase. Nature Neuroscience. 1998;1:23–28. doi: 10.1038/223. [DOI] [PubMed] [Google Scholar]
  28. McDonald BJ, Moss SJ. Conserved phosphorylation of the intracellular domains of GABAA receptor β2 and β3 subunits by cAMP-dependent protein kinase, cGMP-dependent protein kinase, protein kinase C and Ca2+/calmodulin type II-dependent protein kinase. Neuropharmacology. 1997;36:1377–1385. doi: 10.1016/s0028-3908(97)00111-1. [DOI] [PubMed] [Google Scholar]
  29. Mammen A, Huganir R, O'Brien R. Redistribution and stabilization of cell surface glutamate receptors during synapse formation. Journal of Neuroscience. 1997;17:7351–7358. doi: 10.1523/JNEUROSCI.17-19-07351.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Moss SJ, Smart TG. Modulation of amino acid-gated ion channels by protein phosphorylation. International Review of Neurobiology. 1996;39:1–52. doi: 10.1016/s0074-7742(08)60662-5. [DOI] [PubMed] [Google Scholar]
  31. Moss SJ, Smart TG, Blackstone CD, Huganir RL. Functional modulation of GABAA rceptors by cAMP-dependent protein phosphorylation. Science. 1992;257:661–665. doi: 10.1126/science.1323140. [DOI] [PubMed] [Google Scholar]
  32. Polenzani L, Woodward RM, Miledi R. Expression of mammalian γ-aminobutyric acid receptors with distinct pharmacology in Xenopus oocytes. Neurobiology. 1991;88:4318–4322. doi: 10.1073/pnas.88.10.4318. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Pollard T, Cooper J. Actin and actin-binding proteins: a critical evaluation of mechanisms and functions. Annual Review of Biochemistry. 1986;55:987–1035. doi: 10.1146/annurev.bi.55.070186.005011. [DOI] [PubMed] [Google Scholar]
  34. Porter NM, Twyman RE, Uhler MD, Macdonald RL. Cyclic AMP-dependent protein kinase decreases GABAa receptor current in mouse spinal neurons. Neuron. 1990;5:789–796. doi: 10.1016/0896-6273(90)90338-g. [DOI] [PubMed] [Google Scholar]
  35. Prekeris R, Mayhew M, Cooper B, Terrian D. Identification and localization of an actin-binding motif that is unique to the epsilon isoform of protein kinase C and participates in the regulation of synaptic function. Journal of Cell Biology. 1996;132:77–90. doi: 10.1083/jcb.132.1.77. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Qian H, Dowling J. Pharmacology of novel GABA receptors found on rod horizontal cells of the white perch retina. Journal of Neuroscience. 1994;14:4299–4307. doi: 10.1523/JNEUROSCI.14-07-04299.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Qu Z, Apel E, Doherty C, Hoffman P, Merlie J, Huganir R. The synapse-associated protein rapsyn regulates tyrosine phosphorylation of proteins localized at nicotinic acetylcholine receptor clusters. Molecular and Cellular Neuroscience. 1996;8:171–184. [PubMed] [Google Scholar]
  38. Robello M, Amico C, Cupello A. Regulation of GABAA receptor in cerebellar granule cells in culture: differential involvement of kinase activities. Neuroscience. 1993;53:131–138. doi: 10.1016/0306-4522(93)90291-m. [DOI] [PubMed] [Google Scholar]
  39. Rosenmund C, Westbrook G. Calcium-induced actin depolymerization reduces NMDA channel activity. Neuron. 1993;10:805–814. doi: 10.1016/0896-6273(93)90197-y. [DOI] [PubMed] [Google Scholar]
  40. Sigel E, Baur R. Activation of protein kinase C differentially modulates neuronal Na+, Ca2+, and γ-aminobutyrate type A channels. Proceedings of the National Academy of Sciences of the USA. 1988;85:6192–6196. doi: 10.1073/pnas.85.16.6192. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Silverstein SC, Steinman RM, Cohn ZA. Endocytosis. Annual Review of Biochemistry. 1977;46:669–722. doi: 10.1146/annurev.bi.46.070177.003321. [DOI] [PubMed] [Google Scholar]
  42. Stelzer A, Kay AR, Wong RKS. GABAA-receptor function in hippocampal cells is maintained by phosphorylation factors. Science. 1988;241:339–341. doi: 10.1126/science.2455347. [DOI] [PubMed] [Google Scholar]
  43. Strata F, Cherubini E. Transient expression of a novel type of GABA response in rat CA3 hippocampal neurones during development. The Journal of Physiology. 1994;480:493–503. doi: 10.1113/jphysiol.1994.sp020378. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Tehrani MHJ, Hablitz JJ, Barnes EM. cAMP increases the rate of GABAA receptor desensitization in chick cortical neurons. Synapse. 1989;4:126–131. doi: 10.1002/syn.890040206. [DOI] [PubMed] [Google Scholar]
  45. Ticku MK, Mehta AK. γ-Aminobutyric acidA receptor desensitization in mice spinal cord cultured neurons: lack of involvement of protein kinase A and C. Molecular Pharmacology. 1990;318:719–724. [PubMed] [Google Scholar]
  46. Ueno S, Zorumski C, Bracamontes J, Steinbach JH. Endogenous subunits can cause ambiguities in the pharmacology of exogenous γ-aminobutyric acidA receptors expressed in human embryonic kidney 293 cells. Molecular Pharmacology. 1996;50:931–938. [PubMed] [Google Scholar]
  47. Wan Q, Xiong ZG, Man HY, Ackerley CA, Braunton J, Lu WY, Becker LE, MacDonald JF, Wang YT. Recruitment of functional GABAA receptors to postsynaptic domains by insulin. Nature. 1997;388:686–690. doi: 10.1038/41792. [DOI] [PubMed] [Google Scholar]
  48. Wang TL, Guggino WB, Cutting GR. A novel γ-aminobutyric acid receptor rubunit (ρ2) cloned from human retina forms bicuculline-insensitive homooligomeric receptors in Xenopus oocytes. Journal of Neuroscience. 1994;14:6524–6531. doi: 10.1523/JNEUROSCI.14-11-06524.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Wang TL, Hackam A, Guggino WB, Cutting GR. A single histidine residue is essential for zinc inhibition of GABA ρ1 receptors. Journal of Neuroscience. 1995;15:7684–7691. doi: 10.1523/JNEUROSCI.15-11-07684.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Weigel PH, Oka JA. Temperature dependence of endocytosis mediated by the asialoglycoprotein receptor in isolated rat hepatocytes. Evidence for two potentially rate-limiting steps. Journal of Biological Chemistry. 1981;256:2615–2617. [PubMed] [Google Scholar]
  51. Yeh HH, Lee MB, Cheun J. Properties of GABA-activated whole-cell currents in bipolar cells of the rat retina. Visual Neuroscience. 1990;4:349–357. doi: 10.1017/s0952523800004557. [DOI] [PubMed] [Google Scholar]

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

RESOURCES