Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Mechanisms of GABAB receptor enhancement of extrasynaptic GABAA receptor currents in cerebellar granule cells

Abstract

Many neurons, including cerebellar granule cells, exhibit a tonic GABA current mediated by extrasynaptic GABAA receptors. This current is a critical regulator of firing and the target of many clinically relevant compounds. Using a combination of patch clamp electrophysiology and photolytic uncaging of RuBi-GABA we show that GABAB receptors are tonically active and enhance extrasynaptic GABAA receptor currents in cerebellar granule cells. This enhancement is not associated with meaningful changes in GABAA receptor potency, mean channel open-time, open probability, or single-channel current. However, there was a significant (~40%) decrease in the number of channels participating in the GABA uncaging current and an increase in receptor desensitization. Furthermore, we find that adenylate cyclase, PKA, CaMKII, and release of Ca2+ from intracellular stores are necessary for modulation of GABAA receptors. Overall, this work reveals crosstalk between postsynaptic GABAA and GABAB receptors and identifies the signaling pathways and mechanisms involved.

Introduction

The γ-amino butyric acid A receptors (GABAARs) are major contributors of cellular inhibition in the central nervous system. These receptors can be broadly divided into two classes based on composition and location, synaptic and extrasynaptic receptors. GABAARs are pentameric ligand gated ion channels, generally consisting of two α-subunits, two β-subunits, and either a γ- or δ-subunit. GABAARs containing a γ-subunit are targeted to the synaptic space where they are activated by synaptically released GABA and mediate phasic inhibition1,2. GABAARs containing a δ-subunit (δ-GABAARs), on the other hand, are found primarily in the extrasynaptic space where they experience limited exposure to synaptic GABA release and play little3 or no role in phasic inhibition4. However, δ-GABAARs show a higher affinity for GABA and can be tonically activated by ambient GABA in the extracellular space5. This produces a small but powerful tonic inhibitory current in cells expressing δ-subunit containing receptors, capable of dramatically influencing firing6,7. This has made extrasynaptic δ-GABAARs an important target for modulation. For example, the drug 4,5,6,7-Tetrahydroisothiazolo-[5,4-c]pyridine-3-ol (THIP or gaboxadol)8, the anesthetic propofol9, and ethanol10,11,12,13, preferentially activate δ-GABAARs. Likewise, modulation of δ-GABAARs by endogenous neurosteroids contributes to responses to stress and postpartum depression14,15,16.

Recent studies have shown that activation of GABAB receptors (GABABRs) can also modulate extrasynaptic δ-GABAARs. Application of GABABR agonists increases tonic GABAAR currents in cerebellar granule cells, dentate granule cells, and thalamic relay neurons, while GABABR antagonists decrease the tonic current17,18. Furthermore, our studies suggest that presynaptic GABAARs may also be enhanced by activation of GABABRs in parallel fiber presynaptic terminals19,20. These findings open new opportunities to indirectly target tonic GABA currents mediated by δ-GABAARs. However, the signaling mechanisms linking GABABR activity to δ-GABAARs and the biophysical mechanisms of GABAAR enhancement are not currently understood.

To address these questions, we made whole-cell patch-clamp recordings from granule cells in acute cerebellar slices, and GABAAR currents were evoked by photolytic release of GABA from RuBi-GABA. We found that blocking GABABRs by bath application of CGP55845 or 2hydroxy-saclofen reduced GABAAR currents to ~60% of control, suggesting GABAAR currents are constitutively enhanced by GABABRs. GABABR antagonists had no effect on GABAAR currents in Purkinje or stellate cells, or on GABAAR currents evoked by synaptic GABA release in granule cells. Non-stationary fluctuation analysis suggests the change in GABAAR current results from a reduction in number of GABAARs activated, at least partially due to increased desensitization. Enhancement of GABAAR currents by GABABRs requires G-protein signaling, adenylate cyclase, PKA, and Ca+2/calmodulin-dependent protein kinase II (CaMKII). The reduction in GABAAR current by GABABR antagonists was mimicked and occluded by high intracellular EGTA, suggesting changes in intracellular calcium are necessary. Inhibition of ryanodine receptors, but not L-type calcium channels, blocked the effects of GABABR antagonists, suggesting GABABR modulation works through release of calcium from intracellular stores.

Results

Extrasynaptic GABAARs are modulated by tonically active GABABRs

In order to investigate the mechanisms by which GABABRs enhance GABAAR-mediated currents we made whole-cell patch clamp recordings from granule cells in acute cerebellar slices. GABA currents were evoked by full field uncaging of RuBi-GABA (60 μM) using a brief (5 ms) light pulse from a 470 nm LED or 473 nm laser (Fig. 1A). The resultant current is referred to as the “uncaging current”. Uncaging currents were completely blocked by application of picrotoxin (100 μM), indicating these currents are mediated by GABAARs (Fig. 1B, top). G-protein coupled inward rectifying K+ (GIRK) currents potentially linked to GABABR activity are likely inhibited in our experiments by CsCl and QX-314 in the internal solution21,22. Despite the lack of GABABR-mediated currents, bath application of the GABABR antagonists CGP55845 (CGP, 3 μM) or saclofen (200 μM) reduced the amplitude of the uncaging current (CGP: 58.8 ± 2.7% of control, n = 9, p < 0.001; saclofen: 65.6 ± 5.7% of control, n = 6, p = 0.004; Fig. 1B), suggesting GABAARs are constitutively enhanced by GABABR activity. Application of CGP or saclofen also reduced the 10–90 rise-time (24.9 ± 5.1 vs. 17.6 ± 3.1 ms, n = 15, p = 0.02) and decay-time constant (719.3 ± 101.3 vs. 497.0 ± 68.0 ms, p = 0.003) of the uncaging current (Fig. 1B,C). In the absence of CGP application, uncaging current amplitudes in granule cells were stable over 15 minutes of repeated uncaging (last response: 95.8 ± 2.7% of first response, n = 7, p = 0.18; Fig. 1D), suggesting the decrease in uncaging current amplitude is not due to rundown of the response or accumulation of GABA in the slice or bath solution. In Purkinje or stellate cells of the cerebellum, application of CGP had no effect on GABA uncaging current amplitudes (Purkinje: 103.4 ± 9.9% of control, n = 6, p = 0.83; stellate: 101.2 ± 4.6% of control, n = 7, p = 0.67; Fig. 1E,F), indicating enhancement of GABAARs is specific to granule cells. Furthermore, synaptic IPSCs evoked by electrical stimulation in granule cells were not affected by application of CGP (119.9 ± 14.7% of control, n = 8, p = 0.21; Fig. 1G), suggesting that modulation by GABABRs is specific to extrasynaptic GABAARs. This profile of activity matches δ-subunit containing GABAARs expressed in cerebellar granule cells which mediate tonic inhibition23. This suggests GABABR modulation is specific to δ-GABAARs, as has been found previously17,18. However, our GABA uncaging protocol may activate both synaptic and extrasynaptic GABAARs, in order to demonstrate the same effect of CGP on isolated currents from extrasynaptic receptors, we recorded tonic GABAAR currents by measuring the change in the holding current following application of bicuculline (20 μM), a GABAAR antagonist. In the presence of CGP the tonic GABAAR current was decreased to 42.4% of control (n = 7, p < 0.05 unpaired t-test, Fig. 1H) similar to the reduction of uncaging currents in the presence of CGP. We therefor used GABA uncaging in further experiments to investigate the signaling mechanisms involved in extrasynaptic GABAAR modulation.

The observation that application of a GABABR antagonist alone is sufficient to reduce the uncaging current suggests that at least a fraction of GABABRs are tonically activated in our experiments. This is consistent with a previous study which found that ambient GABA tonically activates GABABRs on Golgi cell terminals in the granule cell layer24. Furthermore, application of 100 μM baclofen, a GABABR agonist, did not increase the uncaging GABA current amplitude (94.8 ± 6.2% of control, n = 6, p = 0.48; Fig. 2A), consistent with constitutive activation of GABABRs. In these experiments, GABABRs could be activated by endogenous ambient GABA in the granule cell layer6 or build-up of GABA in the slice due to repeated uncaging. In order to test for the presence of constitutive activation of GABABRs in the absence of repeated GABA uncaging, we activated GABAARs by local pressure application of 10 μM muscimol (a GABAAR specific agonist highly selective for δ-subunit GABAARs and tonic currents at low concentrations25,26) onto granule cells. Under these conditions we found that CGP still reduced the GABAAR current (67.0 ± 4.2% of control, n = 8, p = 0.002; Fig. 2B), similar to results from GABA uncaging. This suggests that GABABRs in granules cells are tonically activated by endogenous ambient GABA, not buildup of uncaged GABA in the bath solution.

In addition to granule cells, GABABRs are also expressed on the soma and axon terminals of GABAergic Golgi cells within the granule cell layer24,27, raising the possibility that the effects of CGP on GABAAR currents is mediated by network effects in the granule cell layer. A previous study found that GABABRs in Golgi cell terminals inhibit GABA release24, suggesting that blocking these receptors could increase GABA release in the granule cell layer. Moreover, in cell attached recordings we found that application of CGP increased the spontaneous firing rate of Golgi cells (226.4 ± 39.81% of control, p = 0.02, n = 3), which is also expected to increase GABA release. This raises the possibility that CGP acts on GABAARs through increasing the ambient extracellular GABA concentration in the granule cell layer. This could reduce the uncaging GABA current through either increasing desensitization of GABAARs5 or increasing the fraction of receptors tonically activated (and reducing the fraction of receptors available to be activated by GABA uncaging). We tested this possibility by applying 10 µM NNC-711, a GABA transporter blocker, to raise the ambient GABA level. In four out of four cells NNC-711 increased the tonic GABA current (data not shown), confirming an increase in ambient GABA. NNC-711 reduced the uncaging current amplitude (71.4 ± 4.4% of control, n = 7, p = 0.007), but subsequent application of CGP was still able to reduce the uncaging current amplitude further (72.2 ± 4.4% of NNC-711, n = 7, p = 0.002, Fig. 2C). These data indicate that increasing the ambient GABA level can alter GABA uncaging currents, but this does not occlude the reduction in uncaging current amplitude by CGP. In a second set of experiment, GDP-β-S was included in the internal solution to inhibit G-protein coupled signaling in the patched cell. In this case, CGP had little effect on the uncaging current amplitude (92.8 ± 3.3% of control, n = 9, p = 0.05; Fig. 2D), suggesting that GABABR signaling within the patched granule cell is necessary for enhancement of extrasynaptic GABAARs and ruling out network effects as a likely mechanism. This finding also confirms that CGP acts through a G-protein coupled receptors (GPCR) rather than binding to or altering GABAARs directly.

GABAB receptors increase the number of available GABAA receptors

While previous studies and our data show that GABABRs enhance extrasynaptic GABAARs17,18, the biophysical mechanisms of this enhancement have not been investigated. We tested several potential mechanisms that could alter GABAAR currents, including changes in GABA potency, desensitization, single channel open time, single channel current, open probability, and number of receptors available.

To test whether GABABR activity alters the GABA potency of GABAARs in granule cells we used a range of blue laser intensities to uncage increasing concentrations of GABA and recorded the resulting currents (Fig. 3A). From this data, we created a dose-response curve of GABAARs in control solutions and in the presence of CGP. We found that the uncaging current was reduced to ~60% of control for each laser intensity tested, with no right or leftward shift in the curve (Fig. 3B). In fact, when the responses in CGP were scaled to the maximum response in control, the curves show significant overlap (Fig. 3B, open circles), suggesting GABABR activity does not change the GABA potency of GABAARs.

We also measured the mean-open times of GABAARs using power spectra analysis of GABA currents during long (12 sec) pressure applications of GABA in control ACSF and in the presence of CGP (see methods; Fig. 3C, inset). From these data we constructed plots of spectral density versus frequency for each cell, which were fit with Lorentzian functions. From these fits we were able to estimate the mean channel open times for each cell. We found a small, but significant, reduction in the fast-open time in the presence of CGP (0.45 ± 0.02 ms vs 0.33 ± 0.03 ms, n = 6, p < 0.001; Fig. 3D), and no change in the slower open times (p > 0.05). Baclofen had no effect on the fast open time (0.56 ± 0.12 ms vs 0.48 ± 0.15 ms, n = 5, p = 0.66). The faster open-time may contribute to the more rapid rise-time and decay time of the GABA current, but likely does not account for the relatively large (~40%) reduction in the macroscopic current.

We then used non-stationary fluctuation analysis28,29,30 of currents evoked by uncaging GABA to discern whether CGP induced a change in the single-channel conductance, open probability, or number of available GABAARs. For each cell we measured the current amplitude and variance during the decay phase of the current, from which a plot of variance versus the mean amplitude of the current was constructed and fit with a binomial model (see methods). From these fits we estimated the single channel conductance, open probability, and number of channels participating in the current for each cell before and after application of CGP (Fig. 3E). Application of CGP did not change the single channel conductance (118.8 ± 12.4%, n = 12, p = 0.24) or maximum open probability (PO max; 99.3 ± 2.9%, n = 12, p = 0.56) of GABAARs, however, the number of channels participating in the current was reduced to 61.8 ± 5.9% of control (n = 12, p = 0.005; Fig. 3F). This matches closely the reduction in the macroscopic uncaging current observed (~60% of control). The decrease in number of channels participating in the current could represent removal of GABAARs from the membrane or accumulation of receptors into long-lasting desensitized states31.

Previous investigations of desensitization of extrasynaptic/δ-subunit containing GABAAR have produced conflicting results, with some studies showing little or no desensitization32,33 and other showing moderate to profound desensitization2,5,34. To test for desensitization in our preparation and whether it is altered by GABABR activity, we delivered a pair of uncaging light pulses (1 second interval) and measured the peak current amplitudes. The second peak (measured from the baseline prior to the first peak) was consistently smaller than the first peak, indicating desensitization of GABAARs following the first GABA application (peak2/peak1: 0.80 ± 0.01, n = 5, p = 0.004; Fig. 3G). Bath application of CGP, in addition to reducing both peak current amplitudes (p = 0.007), also significantly reduced the ratio of the second peak to the first (peak2/peak1: 0.80 ± 0.01 versus 0.73 ± 0.01, p = 0.005; Fig. 3H), indicating greater desensitization. This suggests that increased desensitization contributes to the decreased GABAAR current and kinetics. However, these changes in desensitization are likely not sufficient to completely account for the large (~40%) decrease in the number of GABAARs participating in the GABA current following CGP application, suggesting changes in receptor trafficking and internalization are also involved.

We then tested whether a simple change in the rate of desensitization could account for the changes observed following CGP application using a computer model of GABAAR gating based on the kinetic scheme first described by Jones and Westbrook35. The transition rates in the model were first optimized to fit the relatively slow rise-time and decay kinetics of GABAAR currents evoked by GABA uncaging in cerebellar granule cells (Fig. 3I, Table 1). In the model receptors were activated by a 5 ms square pulse of 60 μM GABA followed by an exponential decay to mimic the time course and relatively low concentrations of GABA likely to result from our uncaging protocol. In order to test the effects of increasing desensitization, we increased the transition rate to the doubly bound desensitized state 5-fold. We found that this change alone was sufficient to closely replicate the observed effects of CGP on GABAAR currents, including reduced peak current, reduced rise-time of the current, and reduced rate of decay (Fig. 3I, Table 1). Furthermore, both the experimental and model currents were well fit by single exponential curves. The faster current decay in the model is surprising given that previous modeling of GABAARs have found that increasing desensitization results in slower, bi-exponential current decay35,36 due to reopening of desensitized receptors before unbinding GABA. This difference in our results appears to be due primarily to the relatively slow back rate from the doubly bound desensitized state in our model, and may reflect unique properties of the extrasynaptic (δ-subunit containing) GABAARs modeled here as opposed to synaptic (γ-subunit containing) receptors. While this modelling data replicates the experimental data in many aspects, this does not necessarily mean that same process takes place in granule cells. However, this does provide theoretical evidence that it is possible for an increase in the rate of desensitization to produce faster macroscopic current kinetics.

Signaling mechanisms of GABAA receptor enhancement

How does activation of adenylate cyclase result in reduced GABAAR currents? Previous reports have shown that several conserved intracellular serine or tyrosine residues on GABAAR subunits are phosphorylation sites for kinases, including PKA, PKC, and CaMKII41,42,43,44. Phosphorylation of these residues maintains surface expression of GABAARs, whereas, dephosphorylation of these subunits triggers receptor internalization45,46. We first tested whether PKA is required for GABAAR enhancement by GABABRs, as it is a primary target of adenylate cyclase/cAMP activity. To block PKA activity we included the PKA inhibitory peptide PKI 14–22 (1 μM) in the intracellular pipette solution. In the presence of this inhibitor the effects of CGP were highly variable. Overall, the change in uncaging current amplitude was not significant (80.1 ± 10.9% of control, n = 8, p = 0.26; Fig. 4C), suggesting PKA activity may be required for the effect of CGP on the GABA current. This is consistent with previous findings in many cells that PKA is normally inhibited by Gi/o GPCRs (including GABABRs) and application of CGP results in activation/disinhibition of PKA37,38.

In order to identify other kinases involved in GABABR-mediated modulation of the uncaging current, we next tested CaMKII and PKC. The inclusion of the PKC inhibitor GF-109203 × (1 µM) in the bath solution did not inhibit the effects of CGP on the uncaging current (77.7 ± 4.7% of control, n = 6, p = 0.01; Fig. 4D). However, the inclusion of the CaMKII inhibitor KN-62 (2 μM) in the bath solution abolished the effects of CGP on the uncaging current (90.9 ± 7.9 of control, n = 6, p = 0.24, Fig. 4E). Furthermore, application of KN-62 alone was sufficient to reduce the uncaging current (68.5 ± 5.3% of control, n = 9; p < 0.001), indicating KN-62 mimics and occludes the effects of CGP. These data suggest that CGP reduces the uncaging current through inhibition of CaMKII.

CaMKII is regulated by intracellular calcium47,48, raising the possibility that the pathway from GABABRs to GABAARs involves changes in intracellular calcium. To test this we bath applied the cell permeable calcium chelator EGTA-AM, which becomes enriched inside the cell due to cleavage of the AM group. Application of 20 µM EGTA-AM alone was sufficient to reduce uncaging current amplitudes (77.21 ± 2.86% of control, n = 9, p < 0.001; Fig. 5A). In the presence of EGTA-AM, further application of CGP had only a small effect on the current amplitude (88.92 ± 3.33% of EGTA-AM, n = 9, p = 0.007), possibly due to continued accumulation of EGTA inside the cell. When using a high EGTA (10 mM) internal solution, CGP application had no effect on GABAAR current amplitudes (105.97% ± 5.52 of control, n = 9, p = 0.43; Fig. 5B). Furthermore, in slices bathed in calcium-free ACSF, application of CGP did not alter the uncaging current amplitude (98.1 ± 2.04% of control, n = 5, p = 0.26; Fig. 5C). These data suggest that changes in intracellular Ca+2 are required for enhancement of GABAARs by GABABRs.

Two primary sources of intracellular calcium are voltage-gated calcium channels and calcium release from intracellular stores. We first tested whether Ca+2 influx through L-type calcium channels is necessary for the CGP effect, as these channels make up approximately 90% of voltage-gated calcium channels (VGCC) in granule cell bodies49. In the presence of nifedipine (an L-type channel blocker), we still observed a significant reduction of the uncaging current amplitude following application of CGP (74.6 ± 4.6% of nifedipine, n = 9, p = 0.003), suggesting L-type calcium channels are not necessary for the effects of CGP. We then tested whether calcium release from intracellular stores is necessary for the effects of CGP. In slices preincubated in dantrolene (10 µM), uncaging currents were not altered by application of CGP (95.83 ± 4.8% of control, n = 7, p = 0.85; Fig. 5F), confirming that ryanodine receptor-mediated release of calcium from intracellular stores is required for the effects of CGP on GABAARs. These data suggest that increases in intracellular calcium from intracellular stores can activate CaMKII and enhance GABAAR-mediated currents.

Discussion

We find that GABABRs in cerebellar granule cells are tonically active and selectively enhance currents through extrasynaptic (presumably δ-subunit containing) GABAARs. Blocking GABABRs reduces the number of GABAA channels activated by GABA, due to changes in desensitization and trafficking of receptors. Further, we find that GABABR-mediated enhancement of GABAARs is dependent on signaling pathways including adenylate cyclase, PKA, CaMKII, and release of calcium from intracellular stores. Identification of the biophysical mechanisms and signaling pathway involved in regulation of tonic GABAAR currents will be important for understanding how ambient GABA and tonic inhibition in the granule cell layer regulates activity at the input layer of the cerebellum. Furthermore, the same mechanisms and pathways are likely employed in other cell types expressing extrasynaptic δ-GABAARs, including thalamic relay neurons and dentate granule cells of the hippocampus. Our recent studies show that presynaptic GABABRs can also modulate GABAARs expressed in the presynaptic terminals of parallel fibers19,20. Many other presynaptic terminals throughout the central nervous system also co-express GABAA and GABABRs, raising the possibility that GABABR-mediated enhancement of GABAARs is a widespread mechanism of synaptic modulation in presynaptic terminals.

Tonic GABAB receptor activity

The granule cell layer maintains basal levels of ambient GABA providing an additional layer of inhibition through extrasynaptic GABAARs6,50,51. While many studies have shown that ambient GABA tonically activates GABAARs in granule cells, tonic activation of GABABRs has received relatively little attention. Estimates of the GABA EC50 for GABABRs can vary widely, however, several reports put this value in the low nanomolar range (5–20 nM38,52,53), well within the range for activation by ambient GABA (150–250 nM5,54). We have recently shown that GABABRs in the axons of granule cells respond to lower concentrations of GABA than GABAARs19, suggesting GABABRs may also be tonically activated in granule cells. This possibility is further supported by findings that GABABRs expressed on synaptic terminals of Golgi cells are tonically activated24. In this study, we find that application of a GABABR antagonist alone is sufficient to reduce tonic GABA/muscimol evoked GABAAR currents (Figs 1H, 2B), suggesting GABABRs are constitutively activated. In addition, application of a GABABR agonist, baclofen, does not alter the uncaging GABA current, consistent with prior activation of GABABRs by ambient GABA. These results differ from an earlier study which showed that baclofen increases the tonic current in cerebellar granule cells while CGP slightly decreases the tonic current17. These conflicting results may be due to the fact that the previous study used tetrodotoxin to block neuronal activity, shown to lower the level of ambient GABA in the granule cell layer at this age6,55, and limiting tonic GABABR activity.

Biophysical mechanisms of GABAA receptor enhancement

There are several biophysical mechanisms that can account for changes in GABAAR-mediated currents, including changes in chloride reversal potential, GABA potency, channel open time, desensitization, single channel conductance, channel open probability, or the number of channels in the membrane. Recent work has shown that GABABRs can regulate the chloride transporter KCC2 altering the chloride reversal potential in neurons56. However, this is unlikely to account for the change in uncaging current in our experiments as the chloride reversal potential is determined by the relative chloride concentrations in the internal and bath solutions. We did not find changes in GABA potency (Fig. 3B), single channel conductance, or open probability (Fig. 3F). However, our data does suggest changes in channel gating contribute to the effects of blocking GABABRs. We observed a consistent decrease in the fast mean open time of the channel (Fig. 3D) and increased desensitization following GABA application (Fig. 3G), which may account for the increased rise-time and decay kinetics of the macroscopic GABA current. In a computer model of GABAAR gating we tested whether increasing the rate of desensitization could account for these changes and found that this change alone was sufficient to reduce the peak current amplitude, the rise-time, and the decay time. However, the fast open time account for only 3–5% of current fluctuations and the change in desensitized receptors was relatively small (5–10%), suggesting other mechanisms may also contribute to the relatively large change in the uncaging current. Using non-stationary fluctuation analysis we found that the number of channels participating in the GABA current was significantly reduced by CGP application (Fig. 3F). This decrease in available channels can be explained, at least in part, by increased accumulation of receptors into long-lasting desensitized states in response to low levels of ambient GABA, as has been reported at other GABAARs5,31. The literature on desensitization of δ-subunit containing GABAARs has been mixed. Initial reports suggest these receptors show little or no desensitization32. More recent studies using α6β2/3δ receptors (the subunit composition thought to mediate the tonic current in cerebellar granule cells) or extrasynaptic receptors in granule cells suggest moderate to profound desensitization of these receptors2,5,33,34. However, changes in receptor trafficking and surface expression may also be necessary to account for the 40–50% reduction of the macroscopic GABAAR current. We therefor argue that GABABR activity/blockade also alters trafficking and surface expression of GABAARs. Previous studies provide ample evidence that phosphorylation of GABAARs by common kinases (CaMKII, PKA, PKC) can alter GABAAR trafficking and surface expression41,42,43,44.

Signaling mechanisms of GABAAR enhancement

The primary signaling pathways of GABABRs, like all Gi/o coupled GPCRs, is through inhibition of adenylate cyclase and reduction of cAMP production37,38. In our experiments, GABABRs are tonically activated, thus application of the GABABR antagonist, CGP-55845, would disinhibit (or activate) adenylate cyclase and cAMP production. Consistent with this, we find that forskolin, an activator of adenylate cyclase, mimics and occludes the effects of CGP. Further, in the presence of SQ 22536, an adenylate cyclase inhibitor, CGP has no effect on the GABAAR current because there is no disinhibition of adenylate cyclase in this case. One of the main targets of cAMP in intracellular signaling cascade is PKA, thus disinhibition (or activation) of adenylate cyclase/cAMP production would activate PKA37,38. Consistent with this, we find that if activation of PKA is prevented by including a PKA inhibitory peptide in the intracellular solution, application of CGP does not significantly reduce the GABAAR current. This suggests activation of PKA is required for the CGP effect, and is consistent with earlier findings that PKA activity decreases tonic GABAAR currents in thalamo-cortical neurons17. However, phosphorylation of GABAARs by PKA can either increase or decrease surface expression of receptors depending on β subunit expression (β1 subunit containing receptor expression is decreased while β3 subunit containing receptor expression is increased)43,44,57. Cerebellar granule cells primarily express β2 and β3, though β1 expression is also present at lower levels58,59. This suggests phosphorylation of GABAARs by PKA would primarily increase GABAAR expression, while we find that disinhibition/activation of PKA results in reduced GABAAR-mediated currents. It is therefore unlikely that PKA activity is directly involved in phosphorylating GABAARs. Rather, other signaling molecules/kinases must be involved.

Several studies have shown that CaMKII and PKC can also phosphorylate GABAARs and increase cell surface expression41,60,61,62. In cell expression system or in cerebellar granule cells, activation of CaMKII enhanced surface expression of GABAARs containing a β3 subunit but not receptors containing a β2 subunit47. Furthermore, the expression of β3 subunits may contribute to receptor trafficking and the cellular location of receptor expression63. In our experiments blocking CaMKII, but not PKC, abolished the effect of CGP application on GABAAR currents. In fact, application of KN-62 (a CaMKII inhibitor) alone was sufficient to reduce the GABAAR current, suggesting that CGP acts through inhibition of CaMKII. This raises the possibility that GABABR modulation acts through CaMKII dependent phosphorylation of extrasynaptic, but not synaptic receptors, based on β subunit expression.

CaMKII activity is highly dependent on calcium, raising the possibility that changes in intracellular calcium levels are involved in the CGP effect. Indeed, we find that the effects of CGP are absent in the presence of high EGTA or following removal of extracellular calcium (Fig. 5B,C). There are two primary sources of intracellular calcium in neurons, influx of calcium through voltage-gated calcium channels and release of calcium from intracellular stores. Inhibition of calcium release from intracellular stores, but not inhibition of L-type calcium channels (which constitute ~90% of voltage-gated calcium channels in granule cells49), abolished the effects of CGP (Fig. 5F), suggesting calcium release from intracellular stores is more important than calcium influx for this effect. However, the observation that the CGP effect is abolished by removal of calcium from the bath solution suggests influx of calcium may also be necessary. It is tempting to speculate that calcium influx (through channels other than L-type channels) induces calcium release from intracellular stores. However, mature granule cells show little evidence of calcium induced calcium release64,65,66, suggesting this is not the case. Another possibility is that bathing the slice in 0 Ca2+ aCSF results in a slow depletion of intracellular stores, and subsequent loss of the CGP effect.

The requirement of release of calcium from intracellular stores would seem to implicate Gq coupled receptors, not Gi/o coupled receptors such as the GABABR, which signal primarily through adenylate cyclase and cAMP. However, a number of previous studies have also observed a rise in calcium from intracellular stores as a result of GABABR activation in a variety of cells67,68,69 including cerebellar granule cells70. While the entire signaling pathway remains unclear, it has been shown to depend on signaling through IP3 and IP3 receptors71. One possible mechanism is through activity of the Gβγ subunit, which has been shown to stimulate PLCβ1–372 and/or formation of IP373,74,75. This raises the possibility that GABABRs signal through PKA with the Gα subunit and stimulate release of calcium from intracellular stores with the Gβγ subunit. While this possibility is consistent with our data, we do not have any direct evidence of GABABR-mediated activation of PLC or formation of IP3.

Based on our data we propose a signaling pathway following GABABR activation involving inhibition of adenylate cyclase and PKA (Fig. 6). This directly or indirectly produces a downstream increase in calcium release from intracellular stores and activation of CaMKII. Finally, CaMKII phosphorylates extrasynaptic GABAARs resulting in decreased desensitization and/or insertion of receptors into the surface membrane.

Physiological relevance

The granule cell layer is the primary input layer of the cerebellum, where granule cells integrate and process information received from across many brain regions76. Spike processing in granule cells is influenced by phasic inhibition from Golgi cell terminals, and tonic inhibition mediated by extrasynaptic GABAARs. It has been estimated that tonic inhibition accounts for 25–75% of chloride influx in granule cells77 and regulates the gain of the input/output function of these cells78. Previous work has shown that the level of ambient GABA in the cerebellum can be modulated, affecting granule cell firing and cerebellar function. For example, neuroinflammation in the cerebellum has been shown to increase ambient GABA, resulting in impaired motor coordination79. Our data suggest that GABABR enhancement of GABAARs may amplify the effects of ambient GABA on granule cell firing. Low levels of ambient GABA likely produce little activation of GABAA or GABABRs, resulting in little or no tonic GABAAR current. However, as the ambient GABA concentration increases the tonic GABAAR current is expected to increase due to both direct activation of GABAARs and due to increased activation of GABABRs and the resulting enhancement of extrasynaptic GABAARs. This creates a much steeper response of GABAARs to changes in ambient GABA than would be expected from the biophysical properties of the GABAARs alone and increases the sensitivity of granule cells to changes in ambient GABA.

Future directions

Cerebellar granule cells express a range of G-protein coupled receptors including, GABABRs80, adenosine A1 receptors81, orexin receptors82, and glutamate receptors (mGluRs)83. This raises the question of whether modulation of extrasynaptic GABAARs is specific to GABABRs in these cells or if other types of G-protein coupled receptors can also activate/inhibit this pathway. It was recently shown that orexin receptors (OX1 and OX2) are expressed in granule cells and their activation decreases GABAAR-mediated currents in a calcium dependent manner82. This suggests that orexin receptors may employ the same mechanisms as GABABRs to modulate GABAAR function, though in the opposite direction. The opposite effect of orexin receptors on GABAAR-mediated currents may be accounted for by the fact that orexin receptors couple to Gs subunits (in addition to Gi/o and Gq)84 which generally activate adenylate cyclase, unlike Gi/o subunits which inhibit adenylate cyclase. These data suggests that orexin receptors, and possibly other Gs coupled receptors, may regulate tonic GABAAR-mediated currents in opposition to GABABRs using the same signaling pathways. Future experiments will be required to determine if other G-protein coupled receptors, besides GABABRs, also employ the signaling pathway/biophysical mechanisms to regulate GABAARs.

Our results indicate that only extrasynaptic receptors in granule cells are modulated by GABABR activity. This result is consistent with δ-subunit expression patterns, but does not directly implicate these receptors. It remains possible that the δ-subunit itself is not necessary for this form of modulation and other extrasynaptic receptors lacking the δ-subunit can also be modulated by this mechanism. In the future these questions can be addressed using cell-type specific δ-subunit knock-out mice.

Experimental Procedures

Animals

All experimental procedures involving animals were approved by the Institutional Animal Care and Use Committee at UT Health San Antonio and followed the guidelines of the National Institutes of Health’s Guide for the Care and Use of Laboratory Animals. Male and female C57BL/6 mice (Charles River, MA) 14–30 days old were used for all experiments. Animals were kept on a 12/12 hour light dark cycle with ad libitum access to food and water.

Slice preparation

Mice were deeply anesthetized with isoflurane and the cerebellum was rapidly dissected following decapitation and placed in ice-cold oxygenated ACSF containing the following (in mM): 119 NaCl, 26.2 NaHCO3, 2.5 KCl, 1.0 NaH2PO4, 11 glucose, 2 CaCl2, 1.3 MgCl2 as described previously19. Parasagittal slices (250–300 µm) were cut from the vermis of the cerebellum using a vibratome (Leica Biosystems, IL) and then incubated at 34 °C for 30 min, after which they were used for electrophysiological recordings at room temperature.

Electrophysiology experiments

For recording, slices were gently place in a chamber that was perfused with room temperature ACSF containing the following (in mM): 119 NaCl, 26.2 NaHCO3, 2.5 KCl, 1.0 NaH2PO4, 11 glucose, 2 CaCl2, 1.3 MgCl2 (flow rate of ~2 ml/min) housed in a SliceScope Pro upright microscope (Scientifica Instruments, UK.). For experiments on somatodendritic GABA currents, whole cell patch clamp recordings were made from granule cells, Purkinje cells, and stellate cells with the following internal (in mM): 135 CsCl, 10 HEPES, 4 MgCl2, 5 EGTA, 4 Na-ATP, 0.5 Na-GTP, 2 QX-314. Additional drugs were included in the internal solution as indicated in the text. The pH of the internal solution was adjusted to 7.2–7.4 using CsOH and the osmolarity was 280–300 mosmol. Cells were patched using 4–6 MΩs borosilicate glass pipettes (Sutter instruments) that were pulled on Sutter pipette puller (Model P-100, Sutter instruments, CA, USA). Electrophysiological currents were recorded with a Multiclamp 700B amplifier (Molecular Devices, CA), filtered at 5 kHz and digitized at 50 kHz. Data were collected using pCLAMP software (Molecular Devices). For GABA uncaging experiments, 10 ml of ACSF containing 10 µM NBQX (Tocris, MN) and 60 μM RuBi-GABA (Tocris) was recirculated using a fluid pump (Cole-Parmer, IL). RuBi-GABA at this concentration does not alter membrane resistance of the cell20. GABA was uncaged by a brief (5 ms) illumination from a 470 nm light-emitting diode (LED) (CoolLED, Andover, UK) at 20% intensity (resulting in ~4 μM free GABA19) to preferentially activate high-affinity extrasynaptic GABAARs. We used a 30 second inter-sweep interval to allow for clearance of GABA between sweeps. For GABA dose-response experiments a 473 nm laser light source (PSU-III-LED, Opto Engine LLC, Midvale, UT) was used for GABA uncaging. The laser power was modulated over a range of intensities (0.28 µW–315 µW) to modulate the amount of GABA released. The maximum GABA concentration in these experiments was not more than 60 μM (the concentration of RuBi-GABA in the bath) and likely much less (20–40 μM). For this reason the saturation of responses at high laser powers likely represents saturation of GABA uncaging rather than saturation of GABAA receptors. The uncaging laser intensity was measured using a photometer placed under the objective of the microscope. For each experiment uncaging current amplitudes were first measured over 3–5 minutes to obtain a stable baseline amplitude before adding CGP55845 or other drugs. Cells in which the uncaging current amplitude was not stable over this period were not included for further analysis. The tonic GABAA current in granule cells was measured by the net change in holding current following application of 20 µM bicuculline (Tocris, MN). In tonic current recordings the ACSF contained 10 µM NBQX and 10 µM CPP. Access resistances of all cells were monitored and cells deviating from 20 ± 10 MΩs were discarded.

Where indicated, ACSF contained one or more of the following: 3 µM CGP 55845 (CGP, Abcam, Cambridge, MA, USA); 200 µM picrotoxin (PTX; Abcam); 200 µM 2-hydroxysaclofen (Tocris); 20 µM Bicuculline (Tocris), 100 µM Baclofen (Abcam), 10 µM NN711, 1 mM GDP-β-S (Sigma), 10 µM Forskolin (sigma), 100 µM SQ-22536 (Tocris), 1 µM PKI 14–22, 1 µM GF-109203 × , 3–6 µM Nifedipine (Tocris), 2 µM KN-62 (Tocris), 20 µM EGTA-AM (Tocris), or 10 µM Dantrolene (Tocris).

Data analysis

Data were analyzed using IgorPro (Wavemetrics, Lake Oswego, OR, USA) using the Neuromatic plug-in85 and Excel (Microsoft, Redmond, WA, USA). Statistical significance was determined using 2-tailed paired Student’s t-tests in Excel (Microsoft, Redmond, WA, USA) or Prism 6 (GraphPad Software, La Jolla, CA, USA). Statistical values of p ≤ 0.05 were considered significant.

Power-spectra analysis: In order to measure the power-spectra of GABAAR mediated currents, GABA was applied to cerebellar granule cells by prolonged uncaging of 60 μM RuBi-GABA (12 second pulse from 473 nM LED). GABA uncaging sweeps were interleaved with background sweeps in which the LED was not activated. GABAAR mediated currents were first recorded in standard ACSF and then following bath application of 3 µM CGP55845. For each sweep, the power spectra was computed by fast Fourier transform over 16 segments of 8192 points each and averaged. In order to avoid time dependent changes in current amplitude, the time segments analyzed began at 5 seconds after the onset of GABA uncaging where the macroscopic GABA current was relatively stable. The power spectra of interleaved background sweeps (lacking GABA uncaging) were computed by the same method and subtracted from GABA power spectra. The power spectra was fitted with the sum of three Lorentzian functions of the form:

$${\rm{L}}({\rm{f}})={\rm{S}}/({\rm{1}}+{(2{\rm{\pi }}\ast {\rm{f}}\ast {\rm{\tau }})}^{{\rm{2}}})$$

where L(f) is the spectral density at frequency f, S is the power of the spectrum at f = 0, and τ is the time constant.

Non-stationary fluctuation analysis: GABA currents were recorded in cerebellar granule cells following either a brief (5 ms) puff of muscimol (10 μM) at the soma or uncaging of RuBi-GABA (60 μM) using a brief (5 ms) flash from a 473 nm laser. For each sweep, amplitude and variance were measured from the peak current to the end of the decay. The average current was scaled to the peak of each individual sweep and subtracted from the current trace, producing a difference trace for each sweep which was then squared to produce a variance trace. Each sweep was then divided into 30 equivalent amplitude bins and the mean current amplitude and variance were measured for each bin. The mean current was plotted against the mean variance and fitted with the following function:

$${{\rm{\sigma }}}^{{\rm{2}}}={i}{{\rm{I}}}_{{\rm{m}}}-{{{\rm{I}}}_{{\rm{m}}}}^{{\rm{2}}}/{\rm{N}}$$

where σ2 us the variance, Im is the mean current, i is the single channel current, and N is the number of channels participating in the current. For all cells included in the final data set the peak current was at least 60% (mean: 78.2 ± 5.2%) of the theoretical maximal current (the x-intercept in the variance equation above). The maximum open probability (PO max) was calculated by dividing the peak current by (i*N), the x-intercept of the fitting function86.

Modelling

GABAAR-mediated currents evoked by GABA uncaging in cerebellar granule cells were modelled in NEURON87, based on a kinetic scheme first described by Jones and Westbrook35, consisting of single- and double-liganded closed, open, and desensitized states. The model was adapted from the granule cell model developed by Diwakar et al.88, and was obtained from ModelDB89 (accession number 116835). GABA uncaging was modelled as a 5 ms square pulse of 60 μm GABA followed by an exponential decay, to replicate the relatively low concentrations of GABA used in experiments. Transition rates were optimized to fit the rise-time and decay kinetics of the recorded currents.

Data availability

We will make all materials, data and associated protocols derived from this work promptly available to other researches without undue qualifications or delays.

References

1. Essrich, C., Lorez, M., Benson, J. A., Fritschy, J. M. & Luscher, B. Postsynaptic clustering of major GABAA receptor subtypes requires the gamma 2 subunit and gephyrin. Nature neuroscience 1, 563–571, https://doi.org/10.1038/2798 (1998).

2. Brickley, S. G., Cull-Candy, S. G. & Farrant, M. Single-channel properties of synaptic and extrasynaptic GABAA receptors suggest differential targeting of receptor subtypes. The Journal of neuroscience: the official journal of the Society for Neuroscience 19, 2960–2973 (1999).

3. Sun, M. Y. et al. Chemogenetic Isolation Reveals Synaptic Contribution of delta GABAA Receptors in Mouse Dentate Granule Neurons. The Journal of neuroscience: the official journal of the Society for Neuroscience 38, 8128–8145, https://doi.org/10.1523/JNEUROSCI.0799-18.2018 (2018).

4. Wei, W., Zhang, N., Peng, Z., Houser, C. R. & Mody, I. Perisynaptic localization of delta subunit-containing GABA(A) receptors and their activation by GABA spillover in the mouse dentate gyrus. The Journal of neuroscience: the official journal of the Society for Neuroscience 23, 10650–10661 (2003).

5. Bright, D. P. et al. Profound desensitization by ambient GABA limits activation of delta-containing GABAA receptors during spillover. The Journal of neuroscience: the official journal of the Society for Neuroscience 31, 753–763, https://doi.org/10.1523/JNEUROSCI.2996-10.2011 (2011).

6. Brickley, S. G., Cull-Candy, S. G. & Farrant, M. Development of a tonic form of synaptic inhibition in rat cerebellar granule cells resulting from persistent activation of GABAA receptors. The Journal of physiology 497(Pt 3), 753–759 (1996).

7. Mitchell, S. J. & Silver, R. A. Shunting inhibition modulates neuronal gain during synaptic excitation. Neuron 38, 433–445 (2003).

8. Meera, P., Wallner, M. & Otis, T. S. Molecular basis for the high THIP/gaboxadol sensitivity of extrasynaptic GABA(A) receptors. Journal of neurophysiology 106, 2057–2064, https://doi.org/10.1152/jn.00450.2011 (2011).

9. Houston, C. M. et al. Are extrasynaptic GABAA receptors important targets for sedative/hypnotic drugs? The Journal of neuroscience: the official journal of the Society for Neuroscience 32, 3887–3897, https://doi.org/10.1523/JNEUROSCI.5406-11.2012 (2012).

10. Hanchar, H. J., Dodson, P. D., Olsen, R. W., Otis, T. S. & Wallner, M. Alcohol-induced motor impairment caused by increased extrasynaptic GABA(A) receptor activity. Nature neuroscience 8, 339–345, https://doi.org/10.1038/nn1398 (2005).

11. Santhakumar, V., Wallner, M. & Otis, T. S. Ethanol acts directly on extrasynaptic subtypes of GABAA receptors to increase tonic inhibition. Alcohol 41, 211–221, https://doi.org/10.1016/j.alcohol.2007.04.011 (2007).

12. Diaz, M. R. & Valenzuela, C. F. Sensitivity of GABAergic Tonic Currents to Acute Ethanol in Cerebellar Granule Neurons is Not Age- or delta Subunit-Dependent in Developing Rats. Alcoholism, clinical and experimental research 40, 83–92, https://doi.org/10.1111/acer.12940 (2016).

13. Borghese, C. M. & Harris, R. A. Studies of ethanol actions on recombinant delta-containing gamma-aminobutyric acid type A receptors yield contradictory results. Alcohol 41, 155–162, https://doi.org/10.1016/j.alcohol.2007.03.006 (2007).

14. Brickley, S. G. & Mody, I. Extrasynaptic GABA(A) receptors: their function in the CNS and implications for disease. Neuron 73, 23–34, https://doi.org/10.1016/j.neuron.2011.12.012 (2012).

15. Maguire, J. L., Stell, B. M., Rafizadeh, M. & Mody, I. Ovarian cycle-linked changes in GABA(A) receptors mediating tonic inhibition alter seizure susceptibility and anxiety. Nature neuroscience 8, 797–804, https://doi.org/10.1038/nn1469 (2005).

16. Stell, B. M., Brickley, S. G., Tang, C. Y., Farrant, M. & Mody, I. Neuroactive steroids reduce neuronal excitability by selectively enhancing tonic inhibition mediated by delta subunit-containing GABAA receptors. Proceedings of the National Academy of Sciences of the United States of America 100, 14439–14444, https://doi.org/10.1073/pnas.2435457100 (2003).

17. Connelly, W. M. et al. GABAB Receptors Regulate Extrasynaptic GABAA Receptors. The Journal of neuroscience: the official journal of the Society for Neuroscience 33, 3780–3785, https://doi.org/10.1523/JNEUROSCI.4989-12.2013 (2013).

18. Tao, W., Higgs, M. H., Spain, W. J. & Ransom, C. B. Postsynaptic GABAB receptors enhance extrasynaptic GABAA receptor function in dentate gyrus granule cells. The Journal of neuroscience: the official journal of the Society for Neuroscience 33, 3738–3743, https://doi.org/10.1523/JNEUROSCI.4829-12.2013 (2013).

19. Howell, R. D. & Pugh, J. R. Biphasic modulation of parallel fibre synaptic transmission by co-activation of presynaptic GABAA and GABAB receptors in mice. The Journal of physiology 594, 3651–3666, https://doi.org/10.1113/JP272124 (2016).

20. Khatri, S. N., Wu, W. C., Yang, Y. & Pugh, J. R. Direction of action of presynaptic GABAA receptors is highly dependent on the level of receptor activation. Journal of neurophysiology, https://doi.org/10.1152/jn.00779.2018 (2019).

21. Otis, T. S., De Koninck, Y. & Mody, I. Characterization of synaptically elicited GABAB responses using patch-clamp recordings in rat hippocampal slices. The Journal of physiology 463, 391–407, https://doi.org/10.1113/jphysiol.1993.sp019600 (1993).

22. Alreja, M. & Aghajanian, G. K. QX-314 blocks the potassium but not the sodium-dependent component of the opiate response in locus coeruleus neurons. Brain research 639, 320–324 (1994).

23. Nusser, Z., Sieghart, W. & Somogyi, P. Segregation of different GABAA receptors to synaptic and extrasynaptic membranes of cerebellar granule cells. The Journal of neuroscience: the official journal of the Society for Neuroscience 18, 1693–1703 (1998).

24. Mapelli, L., Rossi, P., Nieus, T. & D’Angelo, E. Tonic activation of GABAB receptors reduces release probability at inhibitory connections in the cerebellar glomerulus. Journal of neurophysiology 101, 3089–3099, https://doi.org/10.1152/jn.91190.2008 (2009).

25. Chandra, D. et al. Prototypic GABA(A) receptor agonist muscimol acts preferentially through forebrain high-affinity binding sites. Neuropsychopharmacology: official publication of the American College of Neuropsychopharmacology 35, 999–1007, https://doi.org/10.1038/npp.2009.203 (2010).

26. Benkherouf, A. Y. et al. Extrasynaptic delta-GABAA receptors are high-affinity muscimol receptors. Journal of neurochemistry 149, 41–53, https://doi.org/10.1111/jnc.14646 (2019).

27. Kulik, A. et al. Distinct localization of GABA(B) receptors relative to synaptic sites in the rat cerebellum and ventrobasal thalamus. The European journal of neuroscience 15, 291–307 (2002).

28. Traynelis, S. F. & Jaramillo, F. Getting the most out of noise in the central nervous system. Trends in neurosciences 21, 137–145 (1998).

29. Alvarez, V. A., Chow, C. C., Van Bockstaele, E. J. & Williams, J. T. Frequency-dependent synchrony in locus ceruleus: role of electrotonic coupling. Proceedings of the National Academy of Sciences of the United States of America 99, 4032–4036, https://doi.org/10.1073/pnas.062716299 (2002).

30. Saviane, C. & Silver, R. A. Fast vesicle reloading and a large pool sustain high bandwidth transmission at a central synapse. Nature 439, 983–987, https://doi.org/10.1038/nature04509 (2006).

31. Overstreet, L. S., Jones, M. V. & Westbrook, G. L. Slow desensitization regulates the availability of synaptic GABA(A) receptors. The Journal of neuroscience: the official journal of the Society for Neuroscience 20, 7914–7921 (2000).

32. Saxena, N. C. & Macdonald, R. L. Assembly of GABAA receptor subunits: role of the delta subunit. The Journal of neuroscience: the official journal of the Society for Neuroscience 14, 7077–7086 (1994).

33. Wohlfarth, K. M., Bianchi, M. T. & Macdonald, R. L. Enhanced neurosteroid potentiation of ternary GABA(A) receptors containing the delta subunit. The Journal of neuroscience: the official journal of the Society for Neuroscience 22, 1541–1549 (2002).

34. Feng, H. J., Botzolakis, E. J. & Macdonald, R. L. Context-dependent modulation of alphabetagamma and alphabetadelta GABA A receptors by penicillin: implications for phasic and tonic inhibition. Neuropharmacology 56, 161–173, https://doi.org/10.1016/j.neuropharm.2008.08.010 (2009).

35. Jones, M. V. & Westbrook, G. L. Desensitized states prolong GABAA channel responses to brief agonist pulses. Neuron 15, 181–191 (1995).

36. Pugh, J. R. & Raman, I. M. GABAA receptor kinetics in the cerebellar nuclei: evidence for detection of transmitter from distant release sites. Biophysical journal 88, 1740–1754, https://doi.org/10.1529/biophysj.104.055814 (2005).

37. Bettler, B., Kaupmann, K., Mosbacher, J. & Gassmann, M. Molecular structure and physiological functions of GABA(B) receptors. Physiological reviews 84, 835–867, https://doi.org/10.1152/physrev.00036.2003 (2004).

38. Bowery, N. G. et al. International Union of Pharmacology. XXXIII. Mammalian gamma-aminobutyric acid(B) receptors: structure and function. Pharmacological reviews 54, 247–264 (2002).

39. Knight, A. R. & Bowery, N. G. The pharmacology of adenylyl cyclase modulation by GABAB receptors in rat brain slices. Neuropharmacology 35, 703–712 (1996).

40. Xu, J. & Wojcik, W. J. Gamma aminobutyric acid B receptor-mediated inhibition of adenylate cyclase in cultured cerebellar granule cells: blockade by islet-activating protein. The Journal of pharmacology and experimental therapeutics 239, 568–573 (1986).

41. Saliba, R. S., Kretschmannova, K. & Moss, S. J. Activity-dependent phosphorylation of GABAA receptors regulates receptor insertion and tonic current. The EMBO journal 31, 2937–2951, https://doi.org/10.1038/emboj.2012.109 (2012).

42. Brandon, N. J. et al. GABAA receptor phosphorylation and functional modulation in cortical neurons by a protein kinase C-dependent pathway. The Journal of biological chemistry 275, 38856–38862, https://doi.org/10.1074/jbc.M004910200 (2000).

43. McDonald, B. J. et al. Adjacent phosphorylation sites on GABAA receptor beta subunits determine regulation by cAMP-dependent protein kinase. Nature neuroscience 1, 23–28, https://doi.org/10.1038/223 (1998).

44. McDonald, B. J. & Moss, S. J. Conserved phosphorylation of the intracellular domains of GABA(A) receptor beta2 and beta3 subunits by cAMP-dependent protein kinase, cGMP-dependent protein kinase protein kinase C and Ca2+/calmodulin type II-dependent protein kinase. Neuropharmacology 36, 1377–1385 (1997).

45. Kittler, J. T. et al. Regulation of synaptic inhibition by phospho-dependent binding of the AP2 complex to a YECL motif in the GABAA receptor gamma2 subunit. Proceedings of the National Academy of Sciences of the United States of America 105, 3616–3621, https://doi.org/10.1073/pnas.0707920105 (2008).

46. Kittler, J. T. et al. Phospho-dependent binding of the clathrin AP2 adaptor complex to GABAA receptors regulates the efficacy of inhibitory synaptic transmission. Proceedings of the National Academy of Sciences of the United States of America 102, 14871–14876, https://doi.org/10.1073/pnas.0506653102 (2005).

47. Houston, C. M. & Smart, T. G. CaMK-II modulation of GABA(A) receptors expressed in HEK293, NG108-15 and rat cerebellar granule neurons. The European journal of neuroscience 24, 2504–2514, https://doi.org/10.1111/j.1460-9568.2006.05145.x (2006).

48. Fink, C. C. & Meyer, T. Molecular mechanisms of CaMKII activation in neuronal plasticity. Current opinion in neurobiology 12, 293–299 (2002).

49. Koschak, A. et al. Molecular nature of anomalous L-type calcium channels in mouse cerebellar granule cells. The Journal of neuroscience: the official journal of the Society for Neuroscience 27, 3855–3863, https://doi.org/10.1523/JNEUROSCI.4028-06.2007 (2007).

50. Rossi, D. J., Hamann, M. & Attwell, D. Multiple modes of GABAergic inhibition of rat cerebellar granule cells. The Journal of physiology 548, 97–110, https://doi.org/10.1113/jphysiol.2002.036459 (2003).

51. Semyanov, A., Walker, M. C. & Kullmann, D. M. GABA uptake regulates cortical excitability via cell type-specific tonic inhibition. Nature neuroscience 6, 484–490, https://doi.org/10.1038/nn1043 (2003).

52. Bowery, N. G. GABAB receptor pharmacology. Annual review of pharmacology and toxicology 33, 109–147, https://doi.org/10.1146/annurev.pa.33.040193.000545 (1993).

53. Froestl, W. & Mickel, S. J. In The GABA receptors 271–296 (Springer, 1997).

54. Santhakumar, V., Hanchar, H. J., Wallner, M., Olsen, R. W. & Otis, T. S. Contributions of the GABAA receptor alpha6 subunit to phasic and tonic inhibition revealed by a naturally occurring polymorphism in the alpha6 gene. The Journal of neuroscience: the official journal of the Society for Neuroscience 26, 3357–3364, https://doi.org/10.1523/JNEUROSCI.4799-05.2006 (2006).

55. Wall, M. J. & Usowicz, M. M. Development of action potential-dependent and independent spontaneous GABAA receptor-mediated currents in granule cells of postnatal rat cerebellum. The European journal of neuroscience 9, 533–548 (1997).

56. Wright, R. et al. Neuronal Chloride Regulation via KCC2 Is Modulated through a GABAB Receptor Protein Complex. The Journal of neuroscience: the official journal of the Society for Neuroscience 37, 5447–5462, https://doi.org/10.1523/JNEUROSCI.2164-16.2017 (2017).

57. Moss, S. J. & Smart, T. G. Modulation of amino acid-gated ion channels by protein phosphorylation. International review of neurobiology 39, 1–52 (1996).

58. Wisden, W., Korpi, E. R. & Bahn, S. The cerebellum: a model system for studying GABAA receptor diversity. Neuropharmacology 35, 1139–1160 (1996).

59. Nusser, Z., Roberts, J. D., Baude, A., Richards, J. G. & Somogyi, P. Relative densities of synaptic and extrasynaptic GABAA receptors on cerebellar granule cells as determined by a quantitative immunogold method. The Journal of neuroscience: the official journal of the Society for Neuroscience 15, 2948–2960 (1995).

60. Houston, C. M., Hosie, A. M. & Smart, T. G. Distinct regulation of beta2 and beta3 subunit-containing cerebellar synaptic GABAA receptors by calcium/calmodulin-dependent protein kinase II. The Journal of neuroscience: the official journal of the Society for Neuroscience 28, 7574–7584, https://doi.org/10.1523/JNEUROSCI.5531-07.2008 (2008).

61. Lin, Y. F., Angelotti, T. P., Dudek, E. M., Browning, M. D. & Macdonald, R. L. Enhancement of recombinant alpha 1 beta 1 gamma 2L gamma-aminobutyric acidA receptor whole-cell currents by protein kinase C is mediated through phosphorylation of both beta 1 and gamma 2L subunits. Molecular pharmacology 50, 185–195 (1996).

62. Poisbeau, P., Cheney, M. C., Browning, M. D. & Mody, I. Modulation of synaptic GABAA receptor function by PKA and PKC in adult hippocampal neurons. The Journal of neuroscience: the official journal of the Society for Neuroscience 19, 674–683 (1999).

63. Connolly, C. N., Wooltorton, J. R., Smart, T. G. & Moss, S. J. Subcellular localization of gamma-aminobutyric acid type A receptors is determined by receptor beta subunits. Proceedings of the National Academy of Sciences of the United States of America 93, 9899–9904, https://doi.org/10.1073/pnas.93.18.9899 (1996).

64. Mhyre, T. R., Maine, D. N. & Holliday, J. Calcium-induced calcium release from intracellular stores is developmentally regulated in primary cultures of cerebellar granule neurons. Journal of neurobiology 42, 134–147 (2000).

65. Alonso, M. T., Chamero, P., Villalobos, C. & Garcia-Sancho, J. Fura-2 antagonises calcium-induced calcium release. Cell calcium 33, 27–35, https://doi.org/10.1016/s0143-4160(02)00179-3 (2003).

66. Carter, A. G., Vogt, K. E., Foster, K. A. & Regehr, W. G. Assessing the role of calcium-induced calcium release in short-term presynaptic plasticity at excitatory central synapses. The Journal of neuroscience: the official journal of the Society for Neuroscience 22, 21–28 (2002).

67. Meier, S. D., Kafitz, K. W. & Rose, C. R. Developmental profile and mechanisms of GABA-induced calcium signaling in hippocampal astrocytes. Glia 56, 1127–1137, https://doi.org/10.1002/glia.20684 (2008).

68. Kang, J., Jiang, L., Goldman, S. A. & Nedergaard, M. Astrocyte-mediated potentiation of inhibitory synaptic transmission. Nature neuroscience 1, 683–692, https://doi.org/10.1038/3684 (1998).

69. Serrano, A., Haddjeri, N., Lacaille, J. C. & Robitaille, R. GABAergic network activation of glial cells underlies hippocampal heterosynaptic depression. The Journal of neuroscience: the official journal of the Society for Neuroscience 26, 5370–5382, https://doi.org/10.1523/JNEUROSCI.5255-05.2006 (2006).

70. De Erausquin, G., Brooker, G., Costa, E. & Wojcik, W. J. Stimulation of high affinity gamma-aminobutyric acidB receptors potentiates the depolarization-induced increase of intraneuronal ionized calcium content in cerebellar granule neurons. Molecular pharmacology 42, 407–414 (1992).

71. Mariotti, L., Losi, G., Sessolo, M., Marcon, I. & Carmignoto, G. The inhibitory neurotransmitter GABA evokes long-lasting Ca(2+) oscillations in cortical astrocytes. Glia 64, 363–373, https://doi.org/10.1002/glia.22933 (2016).

72. Pierce, K. L., Premont, R. T. & Lefkowitz, R. J. Seven-transmembrane receptors. Nature reviews. Molecular cell biology 3, 639–650, https://doi.org/10.1038/nrm908 (2002).

73. Zeng, W. et al. A new mode of Ca2+ signaling by G protein-coupled receptors: gating of IP3 receptor Ca2+ release channels by Gbetagamma. Current biology: CB 13, 872–876, https://doi.org/10.1016/s0960-9822(03)00330-0 (2003).

74. Smrcka, A. V. & Sternweis, P. C. Regulation of purified subtypes of phosphatidylinositol-specific phospholipase C beta by G protein alpha and beta gamma subunits. The Journal of biological chemistry 268, 9667–9674 (1993).

75. Bugrim, A. Regulation of Ca2+ release by cAMP-dependent protein kinase A mechanism for agonist-specific calcium signaling? Cell calcium 25, 219–226 (1999).

76. Eccles, J. C., Ito, M. & Szentágothai, J. The Cerebellum as a Neuronal Machine (1967).

77. Berglund, K., Wen, L., Dunbar, R. L., Feng, G. & Augustine, G. J. Optogenetic Visualization of Presynaptic Tonic Inhibition of Cerebellar Parallel Fibers. The Journal of neuroscience: the official journal of the Society for Neuroscience 36, 5709–5723, https://doi.org/10.1523/JNEUROSCI.4366-15.2016 (2016).

78. Silver, R. A., Colquhoun, D., Cull-Candy, S. G. & Edmonds, B. Deactivation and desensitization of non-NMDA receptors in patches and the time course of EPSCs in rat cerebellar granule cells. The Journal of physiology 493(Pt 1), 167–173 (1996).

79. Hernandez-Rabaza, V. et al. Neuroinflammation increases GABAergic tone and impairs cognitive and motor function in hyperammonemia by increasing GAT-3 membrane expression. Reversal by sulforaphane by promoting M2 polarization of microglia. Journal of neuroinflammation 13, 83, https://doi.org/10.1186/s12974-016-0549-z (2016).

80. Kaupmann, K. et al. Expression cloning of GABA(B) receptors uncovers similarity to metabotropic glutamate receptors. Nature 386, 239–246, https://doi.org/10.1038/386239a0 (1997).

81. Courjaret, R., Troger, M. & Deitmer, J. W. Suppression of GABA input by A1 adenosine receptor activation in rat cerebellar granule cells. Neuroscience 162, 946–958, https://doi.org/10.1016/j.neuroscience.2009.05.045 (2009).

82. Sachidanandan, D., Reddy, H. P., Mani, A., Hyde, G. J. & Bera, A. K. The Neuropeptide Orexin-A Inhibits the GABAA Receptor by PKC and Ca(2+)/CaMKII-Dependent Phosphorylation of Its beta1 Subunit. Journal of molecular neuroscience: MN 61, 459–467, https://doi.org/10.1007/s12031-017-0886-0 (2017).

83. Chavis, P., Shinozaki, H., Bockaert, J. & Fagni, L. The metabotropic glutamate receptor types 2/3 inhibit L-type calcium channels via a pertussis toxin-sensitive G-protein in cultured cerebellar granule cells. The Journal of neuroscience: the official journal of the Society for Neuroscience 14, 7067–7076 (1994).

84. Thompson, M. D., Xhaard, H., Sakurai, T., Rainero, I. & Kukkonen, J. P. OX1 and OX2 orexin/hypocretin receptor pharmacogenetics. Frontiers in neuroscience 8, 57, https://doi.org/10.3389/fnins.2014.00057 (2014).

85. Rothman, J. S. & Silver, R. A. NeuroMatic: An Integrated Open-Source Software Toolkit for Acquisition, Analysis and Simulation of Electrophysiological Data. Frontiers in neuroinformatics 12, 14, https://doi.org/10.3389/fninf.2018.00014 (2018).

86. Sigworth, F. J. The variance of sodium current fluctuations at the node of Ranvier. The Journal of physiology 307, 97–129, https://doi.org/10.1113/jphysiol.1980.sp013426 (1980).

87. Hines, M. L. & Carnevale, N. T. The NEURON simulation environment. Neural computation 9, 1179–1209 (1997).

88. Diwakar, S., Magistretti, J., Goldfarb, M., Naldi, G. & D’Angelo, E. Axonal Na+ channels ensure fast spike activation and back-propagation in cerebellar granule cells. Journal of neurophysiology 101, 519–532, https://doi.org/10.1152/jn.90382.2008 (2009).

89. McDougal, R. A. et al. Twenty years of ModelDB and beyond: building essential modeling tools for the future of neuroscience. Journal of computational neuroscience 42, 1–10, https://doi.org/10.1007/s10827-016-0623-7 (2017).

Acknowledgements

We are grateful to members of the Pugh lab for helpful discussion and comments on data collection and analysis. We would particularly like to thank Rebecca Howell for her contributions to this work. This work was funded by National Institutes of Health, Grant # R01 NS092809.

Author information

Authors

Contributions

S.N.K., W.C.W. and Y.Y. collected and analyzed data. S.N.K. and J.R.P. designed experiments and interpreted the data. S.N.K. created the figures and wrote the first draft of the manuscript. J.R.P. edited the final draft of the manuscript.

Corresponding author

Correspondence to Jason R. Pugh.

Ethics declarations

Competing interests

The authors declare no competing interests.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

Khatri, S.N., Wu, WC., Yang, Y. et al. Mechanisms of GABAB receptor enhancement of extrasynaptic GABAA receptor currents in cerebellar granule cells. Sci Rep 9, 16683 (2019). https://doi.org/10.1038/s41598-019-53087-4

• Accepted:

• Published:

• DOI: https://doi.org/10.1038/s41598-019-53087-4