In addition to ethanol increasing the mIPSC frequency of cerebellar Purkinje neurons (Ming et al., 2006
), ethanol also increases the mIPSC frequency of central nucleus of the amygdala neurons (Roberto et al., 2003
), basolateral amygdala neurons (Zhu and Lovinger, 2006
), hippocampal CA1 pyramidal neurons (Sanna et al., 2004
; Li et al., 2006
), brain-stem motor neurons (Sebe et al., 2003
), and cerebellar granule cell neurons (Carta et al., 2004
). However, in cultured cortical neurons, ethanol actually decreases mIPSC frequency (Moriguchi et al., 2007
). Therefore, information regarding the mechanism of this ethanol-induced, brain region-specific increase in spontaneous GABA release could be useful in defining the GABAergic behavioral profile of ethanol.
Whereas ethanol increases mIPSC frequency at the interneuron-Purkinje cell synapse ( and ), it has no effect on mIPSC decay time or amplitude. These findings are consistent with many studies that have shown that ethanol increases mIPSC frequency while having no effect on mIPSC decay time (Sebe et al., 2003
; Li et al., 2006
; Ming et al., 2006
; Zhu and Lovinger, 2006
). The increase in mIPSC frequency indicates that ethanol is increasing spontaneous GABA release, whereas the lack of ethanol effect on mIPSC decay time indicates that ethanol is not altering the functioning of the postsynaptic GABAA
receptors in our recording conditions. In addition, ethanol did not alter the mIPSC amplitude, which further supports the conclusion that the change in mIPSC frequency by ethanol can be interpreted as ethanol having a presynaptic effect. Although we did not observe a postsynaptic action of ethanol in either the slice or the mechanically dissociated neuron preparation, the intracellular milieu of the postsynaptic neuron is altered during whole-cell voltage-clamp recordings (for review, see Sarantopoulos et al., 2004
); therefore, postsynaptic effects of ethanol cannot be ruled out at this synapse. Because ethanol increases the mIPSC frequency of mechanically dissociated cerebellar Purkinje neurons, it is not possible that ethanol is acting through glia to increase mIPSC frequency. Overall, our work and the work of others suggest that ethanol increases spontaneous GABA release in multiple brain regions through a presynaptic mechanism.
Voltage-dependent calcium channels (VDCCs), receptor-operated channels (ROCs), and store-operated channels (SOCs) increase intracellular calcium levels by allowing extracellular calcium to flow into the neuron (). A 0 mM Ca2+ext solution was used to eliminate the functionality of these channels to determine their importance in the mechanism of ethanol-enhanced spontaneous GABA release. Ethanol continued to increase mIPSC frequency in the presence of the 0 mM Ca2+ext solution (), which suggests that extracellular calcium influx does not play a role in this ethanol mechanism. Interestingly, the effect of ethanol on mIPSC frequency in the presence of the 0 mM Ca2+ext solution was actually enhanced compared with the effect of ethanol on mIPSC frequency in the control conditions. In the presence of the 0 mM Ca2+ext solution, only “extracellular-calcium insensitive” mIPSCs were present. We predict that ethanol is specifically increasing the frequency of these extracellular-calcium insensitive mIPSCs. Therefore, when the extracellular-calcium sensitive mIPSCs are eliminated in the presence of the 0 mM Ca2+ext solution, a larger ethanol effect is unmasked. In addition, inhibition of the VDCCs had no effect on the ability of ethanol to increase mIPSC frequency (). Therefore, overall, we conclude that extracellular calcium influx does not contribute to the mechanism of ethanol-enhanced spontaneous GABA release at the interneuron-Purkinje cell synapse.
Fig. 7 A summary of the antagonists used to inhibit pathways that result in increases in presynaptic intracellular calcium levels. The nominally calcium-free (0 mM Ca2+ext) solution tested for the involvement of all channels that increase intracellular calcium (more ...)
In and , the mechanically dissociated neuron preparation was used to explore the contribution of extracellular calcium influx to the mechanism of ethanol-enhanced spontaneous GABA release. This preparation was advantageous because it allowed instantaneous access of the 0 mM Ca2+ext
solution and CdCl2
to the neuron, which also dramatically reduced the possibility of decreasing levels of intracellular calcium. However, there are some things to consider when using the mechanically dissociated neuron preparation. It has been proposed that the mechanical dissociation procedure could alter presynaptic terminal excitability (Akaike and Moorhouse, 2003
), and the baseline mIPSC decay time, rise time, and amplitude are different in a mechanically dissociated neuron preparation compared with the slice (see ). However, baseline mIPSC frequency was not different between the two preparations, suggesting that the presynaptic components of both preparations are similar.
However, it is important to note that 50 mM ethanol increases mIPSC frequency to a higher degree in the mechanically dissociated neuron preparation. Previously, ethanol has been shown to increase GABA release in a mechanically dissociated neuron preparation from the basolateral amygdala (Zhu and Lovinger, 2006
). These investigators also saw a larger effect of ethanol on GABA release in the mechanically dissociated neuron preparation compared with the slice. Because the mechanically dissociated neuron preparation allows for instantaneous access of ethanol to the neuron, the effect of ethanol can be seen on a seconds timescale compared with the minutes required in the slice. Therefore, Zhu and Lovinger (2006)
hypothesize that because it takes a longer amount of time to see the ethanol effect in the slice, tolerance to the ethanol effect will start to develop, resulting in an overall smaller effect of ethanol in the slice compared with a mechanically dissociated neuron preparation.
Because influx of extracellular calcium was not required for ethanol to increase spontaneous GABA release, our focus shifted next to calcium release from internal stores. After using the thapsigargin protocol to deplete internal stores of calcium, there was no significant change in baseline mIPSC frequency. It has been reported that there is an increase in baseline mIPSC frequency when thapsigargin is applied for 20 min or less (Bardo et al., 2002
; Li et al., 2004
). These latter data are consistent with calcium still being released from internal stores while thapsigargin is blocking calcium reuptake through the SERCA pump (). However, after a period of time, the effect of thapsigargin on mIPSC frequency subsides when even more calcium is depleted from internal stores (Li et al., 2004
), which is consistent with our current data. When the internal stores had been depleted of calcium, ethanol was not able to increase spontaneous GABA release (). Therefore, these data are consistent with calcium release from internal stores playing a vital role in the mechanism of ethanol-enhanced spontaneous GABA release.
After determining that calcium release from internal stores plays an imperative role in ethanol-enhanced spontaneous GABA release, we wanted to investigate whether the IP3
Rs and RyRs were also involved in this ethanol mechanism. The IP3
R antagonist 2-APB significantly blocked the ethanol-induced increase in mIPSC frequency (). Even though 2-APB is the most widely used membrane-permeable IP3
R antagonist, it has selectivity issues with respect to intracellular calcium signaling that needed to be considered before we interpreted these results. When 2-APB concentrations higher than 90 µM are used, there is a nonspecific calcium leak from internal stores and slight inhibition of the SERCA pump (Missiaen et al., 2001
). This nonspecific effect of 2-APB offers an explanation for the large increase in miniature excitatory postsynaptic current frequency (Simkus and Stricker, 2002
) and mIPSC frequency (unpublished results) seen with 2-APB concentrations higher than 80 µM. To circumvent these nonspecific effects of 2-APB, a low concentration of 2-APB (14 µM) was used that does not increase baseline mIPSC frequency. However, 2-APB has additional nonspecific effects that could occur at any concentration, including inhibition of SOCs as well as transient receptor potential (TRP) channels (Lievremont et al., 2005
). Entry of calcium through SOCs is activated when internal stores are depleted of calcium; therefore, inhibition of the IP3
Rs would reduce the normal functioning of the SOCs because inhibition of the IP3
Rs prevents depletion of internal calcium stores. In addition, TRP channels have been implicated in this store-operated calcium entry mechanism (Zhu et al., 1996
). Because IP3
Rs, SOCs, and TRP channels are mechanistically linked, defining the selectivity of 2-APB has been controversial (Boulay et al., 1999
; Lievremont et al., 2005
). However, in the present experiments, the possibility that 2-APB was acting through the SOCs and the TRP channels to block the effect of ethanol on spontaneous GABA release can be ruled out because removal of extracellular calcium did not prevent ethanol from increasing spontaneous GABA release (). Therefore, these data suggest that calcium release from the IP3
Rs is playing a role in ethanol-enhanced spontaneous GABA release. The fact that depletion of internal calcium stores was sufficient to block the effect of ethanol on GABA release further supports this view ().
Inhibition of RyRs also prevented ethanol from increasing spontaneous GABA release (). This result is not surprising considering that there is a high density of RyRs present at the presynaptic component of the interneuron-Purkinje cell synapse (Llano et al., 2000
). Calcium release from the RyRs is affected by calcium release from the IP3
Rs, and vice versa, because the amount of calcium in close proximity to each receptor affects its ability release calcium (Berridge, 1998
). Therefore, at this time, we cannot distinguish between the relative importance of the IP3
Rs and RyRs; however, the results strongly suggest that internal calcium stores play a central role in ethanol-enhanced spontaneous GABA release.
Because thapsigargin, 2-APB, and ryanodine were applied to the bath, these drugs could affect both presynaptic and postsynaptic calcium levels. A change in postsynaptic calcium could alter release of calcium-dependent retrograde messengers, which could account for the effect of ethanol on mIPSC frequency. Ethanol was still able to increase mIPSC frequency with BAPTA in the internal solution (), which suggests that calcium-dependent retrograde messengers are not responsible for the effect of ethanol on spontaneous GABA release and is consistent with previous results (Zhu and Lovinger, 2006
). However, inclusion of BAPTA in the internal solution would not necessarily prevent CB release because it has been proposed that CB release does not always require the presence of postsynaptic calcium (Hashimotodani et al., 2007
). As a result, a separate control was conducted to test for the contribution of CBs to the mechanism of ethanol-enhanced spontaneous GABA release. Because the CB receptor antagonist, AM-251, did not alter the ability of ethanol to increase spontaneous GABA release, we concluded that CBs are not playing a major role in the mechanism of ethanol-enhanced spontaneous GABA release ().
In conclusion, the present study suggests that ethanol is acting presynaptically to increase spontaneous GABA release and that this mechanism involves calcium release from internal stores. To the best of our knowledge, this is the first evidence of calcium release from internal stores being necessary for ethanol-enhanced spontaneous GABA release. However, the exact interaction occurring between ethanol and internal calcium stores has yet to be elucidated. Further knowledge of the mechanism of ethanol-enhanced GABA release will provide information that can be applied toward delineating the mechanisms contributing to the GABAergic behavioral profile of ethanol.