|Home | About | Journals | Submit | Contact Us | Français|
The mesocorticolimbic dopamine system, originating in the ventral tegmental area (VTA) are normally constrained by GABA-mediated synaptic inhibition. Accumulating evidence indicates that long-term potentiation of GABAergic synapses (LTPGABA) in VTA dopamine neurons plays an important role in the actions of drugs of abuse, including ethanol.
We previously showed that a single infusion of glycine into the VTA of rats strongly reduces ethanol intake for 24 hours. In the current study, we examined the effect of glycine on the electrophysiological activities of putative dopamine VTA neurons in midbrain slices from ethanol-naïve rats. We report here that a 15 min exposure to 10 μM glycine prevented trains of high frequency stimulation from producing LTPGABA, which was rescued by the glycine receptor antagonist strychnine. Glycine also concentration-dependently decreased the frequency of spontaneous excitatory postsynaptic currents. By contrast, glycine pretreatment did not prevent potentiation of inhibitory postsynaptic currents during a continuous exposure to the nitric oxide donor, SNAP (S-nitroso-N-acetylpenicillamine), or a brief exposure to 10 μM glycine and 10 μM NMDA (N-methyl-D-aspartate), an agonist of NMDA-type glutamate receptors. Thus, the blockade of LTPGABA by glycine is probably resulted from suppressing glutamate release by activating the glycine receptors on the glutamatergic terminals. This effect of glycine may contribute to the reduction in ethanol intake induced by intra-VTA glycine observed in vivo.
The mesocorticolimbic dopaminergic system, originating in the ventral tegmental area (VTA), plays an important role in alcohol use disorders (Gilpin and Koob, 2008). VTA dopamine neurons are normally constrained by powerful GABA-mediated synaptic inhibition (Johnson and North, 1992, Tepper et al., 1998). Both GABAA and GABAB receptor subtypes have been identified within the VTA. Activation of GABAA receptor (GABAAR) on GABAergic interneurons increases VTA dopamine neuron activity through disinhibition. Conversely, GABAergic neurons decreases spike output of dopamine neurons via GABAB receptors on dopamine neurons.
GABAergic signaling in the VTA has been linked to ethanol drinking behaviors. Intra-VTA administration of GABAA receptor (GABAAR) antagonists decreases ethanol intake in rodents (Nowak et al., 1998, Melon and Boehm, 2011). Importantly, VTA GABA neurons become hyperexcitable during ethanol withdrawal (Gallegos et al., 1999). Interestingly, rodents readily self-administer GABAAR antagonists into the VTA (Ikemoto et al., 1997, Gavello-Baudy et al., 2008) and this blockade of GABAARs in the VTA increases dopamine levels in the nucleus accumbens (Ikemoto et al., 1997), producing robust rewarding effects (Laviolette and van der Kooy, 2001). Changes in synaptic strength are important for persistent pathologies such as drug addiction. In particular, accumulating evidence indicates that long-term potentiation of GABAergic synapses (LTPGABA) in VTA dopamine neurons increases or decreases in response to the actions of drugs of abuse, including ethanol (Nugent et al., 2007, Nugent et al., 2009, Guan and Ye, 2010).
Like GABA, glycine is also a major inhibitory neurotransmitter in the CNS. Functional glycine receptors (GlyRs) are found throughout the mammalian CNS, including the mesocorticolimbic dopamine system (Zheng and Johnson, 2001, Ye et al., 2002, Lewis and O’Donnell, 2003, Wang et al., 2005). Our previous in vitro electrophysiological data indicate that GlyRs exist on the GABAergic terminals, which make synapses on VTA dopamine neurons, and activation of these GlyRs reduces GABAergic transmission and increases VTA dopamine neuron activity (Ye et al., 2004). In vivo studies have shown that GlyRs in the mesolimbic dopamine system are involved in ethanol drinking behaviors. Activation of GlyRs in the nucleus accumbens substantially decreases ethanol intake (Molander et al., 2005, Molander and Soderpalm, 2005b, a). Remarkably, we have shown that a single intra-VTA infusion of glycine substantially reduces ethanol intake for 24 hours in rats, in three different drinking models: intermittent access, continuous access, and operant self-administration (Li et al., 2012). However, the underlying neuronal basis has not been well explored. Glycine could alter the activity of dopamine neurons by activating the GlyRs on the GABAergic terminals, which make synapses on VTA dopamine neurons (Ye et al., 2004). However, given that extracellular glycine is probably removed rapidly by reuptake, additional mechanisms must be involved.
In an attempt to understand the mechanisms underlying the long-lasting effect of intra-VTA perfusion of glycine on ethanol drinking, we examined glycine’s effects on the electrophysiological activities of putative dopamine neurons in the VTA in midbrain slices from ethanol-naïve rats.
The experiments were carried out on Sprague–Dawley rats, in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals, and were approved by the Institutional Animal Care and Use Committee of the Rutgers, the State University of New Jersey, New Jersey Medical School.
DNQX (6, 7-dinitroquinoxaline-2, 3-dione), gabazine (SR-95531), glycine, NMDA (N-Methyl-D-aspartic acid), SNAP (S-nitroso-N-acetylpenicillamine), strychnine, and all common salts were obtained from Sigma-Aldrich Chemical Company (St Louis, MO). Drugs were added to the superfusate at final concentrations. Chemicals were applied to the recorded neurons at the stated concentrations through bath perfusion.
The midbrain slices were prepared as described previously (Guan and Ye, 2010). Briefly, Sprague–Dawley rats (21–35 d old) were anesthetized using ketamine/xylazine and then sacrificed. Coronal midbrain slices (200–250 mm thick) were cut using a VF-200 slicer (Precisionary Instruments, Greenville, NC) and prepared in ice-cold glycerol-based artificial cerebrospinal fluid (aCSF) containing (in mM): 250 glycerol, 1.6 KCl, 1.2 NaH2PO4, 1.2 MgCl2, 2.4 CaCl2, 25 NaHCO3, and 11 glucose, and saturated with 95%O2/5%CO2 (carbogen) (Ye et al., 2006). Slices were allowed to recover for at least 1 h in a holding chamber at 32 ° in carbogen-saturated regular aCSF, which had the same composition as glycerol-based aCSF, except that glycerol was replaced by 125 mM NaCl.
Electrical signals were obtained with a MultiClamp 700 A amplifier (Molecular Devices Co., Union City, CA, USA), and pCLAMP software (Molecular Devices Co.). Data were filtered at 1 kHz and sampled at 5 kHz, stored on a computer, and analyzed offline. The amplitudes of IPSCs were calculated by taking the mean of a 2–4 ms window around the peak and comparing it with the mean of a 2–8 ms window immediately before the stimulation artifact, using Clampfit (Molecular Devices).
Patch pipettes (4–6 MΩ) for recording IPSCs contained (in mM): 125 KCl, 2.8 NaCl, 2 MgCl2, 0.6 EGTA, 10 HEPES, 2 Mg ATP-Na and 0.2 GTP-Na; and for EPSCs contained (in mM): 140 Cs-methanesulfonate, 5 KCl, 2 MgCl2, 10 HEPES, 2 MgATP, 0.2 NaGTP, pH 7.2. The pH was adjusted to 7.2 with Tris base, and osmolality to 300 mOsmol/l with sucrose. A single slice was transferred to a 0.4 ml recording chamber, where a platinum ring held it down. Throughout the experiments, the bath was continually perfused with warm (32°C) carbogenated aCSF (1.5–2.0 ml/min).
Under infrared video microscopy (E600FN; Nikon Corporation, Tokyo, Japan), the VTA was identified medial to the accessory optic tract and lateral to the fasciculus retroflexus. Currents were recorded using whole-cell mode. Experiments were carried out only after series resistance had stabilized. Series resistance and input resistance were monitored continuously online with a −4 mV hyperpolarizing step (50 ms), which was given after every afferent stimulus, and experiments were discarded if these values changed by 20% during the experiment. Putative dopamine neurons in the current study were identified by the presence of a large Ih current (Johnson and North, 1992) that was assayed immediately after break-in, using a series of incremental 10 mV hyperpolarizing steps from a holding potential of −50 mV. Specifically, if the steady-state h-current was greater than 60 pA during a step from −50 to −100 mV, the neuron was considered a dopamine neuron. A recent study showed that expression of Ih alone is not sufficient to identify dopamine cells unequivocally (Margolis et al., 2006). Therefore, in each set of our experiments, a subset of the neurons recorded from and reported here are possibly non-dopaminergic neurons (Nugent et al., 2009).
GABAAR-mediated IPSCs were stimulated at 0.1 Hz using a bipolar stainless steel-stimulating electrode placed 200–400 μm away from the recording site in the VTA. As previously described (Guan and Ye, 2010), LTPGABA was induced by stimulating afferents at 100 Hz for 1 s, and the train was repeated twice, 20 s apart (high-frequency stimulation; HFS). After recording the baseline currents, during the drug application and washout, synaptic stimulation was stopped and the recorded neuron was taken from voltage-clamp into bridge mode. The HFS trains were also delivered under bridge mode.
Data are presented, as means ± SEM. Significance was determined using a Student’s unpaired t-test with a significance level of P <0.05. Levels of LTP are reported as averaged IPSC amplitudes for 5 min just before LTP induction compared with averaged IPSC amplitudes during the 5 min period from 20 to 25 min after HFS. Paired-pulse ratios (PPR=IPSC2/IPSC1) (50 ms inter-stimulus interval) were measured over 5 min epochs of 30 IPSCs each, as previously described (Guan and Ye, 2010, Guan et al., 2012).
Currents evoked by electrical stimuli were recorded from putative dopamine neurons of the VTA in midbrain slices in the presence of DNQX (10 μM), an AMPA receptor antagonist, and at a holding potential of −70 mV. These currents were blocked by 10 μM gabazine, a GABAA receptor (GABAAR) antagonist (not illustrated), indicating that they were IPSCs mediated by GABAARs. The amplitude of IPSCs was greatly increased and maintained after high frequency stimulation (HFS, see Method). Thus, HFS induced long-term potentiation of IPSC (LTPGABA, Fig. 1A), in keeping with previous reports (Nugent et al., 2007, Guan and Ye, 2010). LTPGABA was associated with a decrease in the paired pulse ratio (PPR; Fig. 1a, insets), suggesting that LTPGABA is maintained by persistently increased GABA release, which is probably caused by high-frequency firing of the presynaptic GABAergic afferents (Nugent et al., 2007, Guan and Ye, 2010).
We then examined the effect of glycine on LTPGABA. Glycine (10 μM) was bath perfused for 15 min and then washed out for 6 min before HFS. HFS failed to induce LTPGABA under such conditions (Fig. 1B and C). In fifteen experiments, in the absence of glycine, the averaged peak amplitude of IPSC (IPSC1) elicited by the first stimulus of the paired pulse measured 20–25 min after HFS was 198.25 ± 1.37 % of pre-HFS values (Fig. 1C, filled circles). After exposure to glycine, it was 101.43 ± 2.18 % of pre-HFS values (Fig. 1C, open circles). Glycine also reversed the changes in PPR (in aCSF, PPR = 0.68 ± 0.15 of pre-HFS values; in glycine, PPR = 1.06 ± 0.18 of pre-HFS values, P < 0.05; Fig. 1D). To determine whether the effects of glycine were mediated by strychnine sensitive GlyRs, strychnine (1 μM) was bath applied for 5 min before the application of the mixture containing strychnine (1 μM) and glycine (10 μM) for 15 min. HFS was delivered 6 min after washout of the mixture. Under these experimental conditions, the cells exhibited normal LTPGABA (Fig. 1E). In ten experiments, the averaged peak amplitude of IPSC1 measured 20–25 min after HFS was 196.29 ± 10.47 % of pre-HFS values in cells exposed to glycine plus strychnine (Fig. 1F, filled circles) vs 100.33 ± 4.26% of pre-HFS values in glycine alone (Fig. 1F, open circles). The PPR was 0.70 ± 0.10 of pre-HFS values in cells exposed to glycine plus strychnine. These results suggest that glycine-induced suppression of LTPGABA was mediated by strychnine-sensitive GlyRs, which prevents HFS-induced increase of GABA release.
LTPGABA is characterized as heterosynaptic plasticity, initiated by glutamate release onto NMDARs on the postsynaptic dopamine neurons. There is previous evidence that nitric oxide (NO) is produced in dopamine neurons and released extracellularly to enhance GABA release from GABAergic terminals, and application of the NO donor, SNAP, rapidly enhances GABAergic IPSCs (Nugent et al., 2007). In the current study, we perfused SNAP (400 μM) in the aCSF in the presence of glycine (10 μM) throughout recording. Thirty minutes after SNAP application, eIPSC amplitudes potentiated to 147.98 ± 4.26% of pre-SNAP values (n = 11) (Fig. 2). These data indicate that glycine blocks LTPGABA in the upstream of nitric oxide synthase (NOS) in dopamine neurons.
We next tested whether glycine’s prevention of LTPGABA involves NMDARs in VTA-dopamine neurons. To inhibit evoked IPSCs, we first bath applied glycine (10 μM, for 5 min), and then to activate NMDARs, we bath applied a mixture containing 10 μM NMDA and 10 μM glycine, for 10 min. After washout, IPSCs recovered to an enhanced level (Fig. 3). Eighty minutes after washout, eIPSC amplitude was 142.14±6.22 % (n =5) of pre-treatment values. Bath application of glycine (10 μM) alone did not produce potentiation (not illustrated). These data support previous observations that NMDAR activation alone (without HFS) is sufficient to induce LTPGABA (Nugent et al., 2007).
The above result showed that glycine could not inhibit LTPGABA that was induced by NMDAR activation; glycine’s inhibition of LTPGABA thus may result from targeting upstream events of NMDARs (e.g. glutamate release). To test this possibility, we examined the effects of glycine on spontaneous EPSCs (sEPSCs) in VTA-dopamine neurons. All sEPSCs were recorded in the presence of 10 μM gabazine and at a holding potential of −70 mV with Cs-methanesulfonate based pipette solution. These events were eliminated by DNQX (20 μM, data not show), indicating that they are AMPAR-mediated EPSCs. As illustrated in Fig. 4, glycine (10 μM) decreased sEPSC frequency, and the inhibition was blocked by strychnine (1 μM). Glycine-induced inhibition of sEPSC frequency depended on its concentration. Specifically, glycine (1, 3 and 10 μM) decreased sEPSC frequency by 9.15 ± 11.05%, 26.31 ± 7.42 %, and 51.38 ± 7.19% (*p < 0.05, n = 15, **p<0.01 (Fig. 4)), respectively, without a significant effect on sEPSC amplitude (7.24 ± 9.10%, 8.18 ± 11.27%, 11.41 ± 14.27%, respectively, all p > 0.05, n = 15).
Our major finding is that a short (15 minute) exposure to 10 μM glycine prevented trains of HFS from producing LTPGABA on VTA-dopamine neurons in midbrain slices of rats, which was paralleled with the reduction of glutamate release. By contrast, glycine pretreatment did not prevent potentiation of eIPSCs during a continuous exposure to the NO donor SNAP. Similarly, glycine pretreatment did not prevent an increase of IPSC amplitude in response to a brief exposure to 10 μM NMDA and 10 μM glycine. The effects of glycine could conceivably be mediated by strychnine-sensitive GlyRs, because co-application of glycine and strychnine failed to affect LTPGABA. Our data suggest that glycine-induced change of LTPGABA may contribute to the reduction of ethanol drinking by intra-VTA infusion of glycine observed in vivo (Li et al., 2012).
LTPGABA is known to be produced by a heterosynaptic mechanism, requiring the activation of postsynaptic NMDARs at glutamate synapses, resulting from increased GABA release at neighboring inhibitory nerve terminals (Nugent et al., 2007). We recorded HFS-induced LTPGABA on VTA-dopamine neurons and found that LTPGABA was coupled to a change in PPR, an electrophysiological parameter typically used to identify changes in probability of transmitter release (Nugent et al., 2007, Guan and Ye, 2010). An increased presynaptic GABA release in response to HFS was indicated by the significant reduction in the mean PPR of GABAergic synaptic currents associated with LTPGABA (Nugent et al., 2007, Guan and Ye, 2010), arguing against a potential role of increased sensitivity of postsynaptic GABAARs in this process. In the present study, we demonstrated that bath perfusion of 10 μM glycine blocks the change of PPR induced by HFS, indicating that glycine prevents the increase in presynaptic GABA release induced by HFS.
Our previous in vivo study showed that a single intra-VTA infusion of glycine (5–10 μM, 500 μl/side) reduces alcohol drinking for 24 hours, and the effect is mediated by strychnine-sensitive GlyRs (Li et al., 2012). We speculate that this glycine may quickly disappear by reuptake, and the activation of GlyRs must trigger a process that outlasts acute receptor activation. In an attempt to mimic the in vivo experimental conditions, we applied HFS 6 mins after wash-out of glycine. Future studies are necessary to determine how long it takes the system to recover from glycine-induced suppression of LTPGABA, and whether the results are different if glycine were continually present.
Remarkably, glycine-induced blockade of LTPGABA is prevented by strychnine, indicating that glycine’s effect is mediated by strychnine-sensitive GlyRs, instead of the NMDARs, where it is a co-agonist. The strychnine-sensitive GlyR conducts Cl−. In adult mammals, it is inhibitory because it enhances Cl− influx (Werman et al., 1968, Nicoll et al., 1990, Kuhse et al., 1995). Previous studies have shown that GlyRs exist in the VTA (Ye et al., 1998, Ye, 2000, Zheng and Johnson, 2001). Notably, in the VTA, functional GlyRs are not only found in the postsynaptic membrane, but also in the presynaptic membrane, including the GABAergic terminals, which make synapses onto VTA-dopamine neurons. Activation of these receptors modulates the release of transmitters and the activity of VTA-dopamine neurons (Ye et al., 2004). Glycine at 3 and 10 μM substantially suppressed sEPSC frequency but not their amplitude (Fig. 4), suggesting that strychnine-sensitive GlyRs exist on the glutamatergic terminals. By activating these GlyRs, glycine reduces glutamate release, and thus the activation of NMDARs, resulting in the prevention of LTPGABA. These data also indicate that the GlyRs on the glutamatergic terminals are very sensitive to glycine, in keeping with our previous report that GlyRs on the GABAergic terminals are very sensitive to glycine (Ye et al., 2004). Specifically, whereas the presynaptic GlyRs have an EC50 of ~1.5 μM (Ye et al., 2004), which is much lower than that for the postsynaptic GlyRs, which have an EC50 of ~25 μM (Ye, 2000). Currently, it remains unclear why the GlyRs on presynaptic axons in the VTA are more sensitive to glycine than GlyRs on DA neurons, although receptor density, subunit composition, and structure difference may play a role. In the current study, we did not investigate the mechanisms of glycine-induced reduction of glutamate release. However, previous studies have shown that calcium channels are involved in glycine-induced changes of GABAergic IPSCs in the VTA (Ye et al., 2004), and in glycine-induced changes of glutamatergic EPSCs in other CNS areas (Turecek and Trussell, 2001).
The sources of endogenous GlyR agonists in the VTA are unclear. Glycine-releasing neurons are sparse in rostral areas of the brain (Danglot et al., 2004), including the VTA. However, glycine could be released by glial cells (Legendre, 2001). In human cerebrospinal fluid and in the frontal cortex, free glycine is normally at levels between 7 to 10 μM, however, higher levels (18 to 28 μM) have been detected in the cerebellum (Ferraro and Hare, 1985) (Matsui et al., 1995) (Fabricius et al., 1993). The extracellular concentration of glycine is regulated by at least two glycine transporters (GLYT1 and GLTY2). These GLYTs are likely to play a role both at inhibitory glycinergic synapses, and at glutamatergic synapses (Ascher, 1990); (Attwell et al., 1993).
The current study demonstrated that glycine blockage of LTPGABA is paralleled with the inhibition of sEPSCs. These data support the idea that presynaptic glutamate release is necessary for the induction of LTPGABA (Nugent et al., 2007). By contrast, glycine failed to inhibit LTPGABA either induced by a short co-application of NMDA and glycine, or by the continuous perfusion of the NO donor SNAP. Previous studies by Kauer’s group have shown that NO, a retrograde signal generated by an NMDAR-mediated rise in postsynaptic Ca2+ acting at a presynaptic site, is produced in dopamine neurons and released extracellularly to enhance GABA release from GABAergic terminals (Nugent et al., 2007). These investigators also showed that NO scavengers (Hemoglobin and 2-pheyl-4, 4, 5, 5,-tetramethylimidazoline-1-oxyl 3-oxide) prevent HFS-induced LTPGABA. On the basis of their findings, our results suggest that glycine is probably acting upstream of NMDAR and NO production. We propose that glycine inhibition of LTPGABA is mainly resulted from its inhibition of glutamate release, since the induction of LTPGABA requires the activation of NMDARs (Nugent et al., 2007). By activating the GlyRs on the GABAergic terminals on VTA-dopamine neurons, glycine could directly suppress GABAergic IPSCs (Ye et al., 2004).
Interestingly, 10 μM glycine did not alter the eIPSC amplitude as shown in Fig. 1B, but decreased it as seen in Fig. 3. The mechanism underlying the difference is unclear, but may be resulted from the difference in the conditions when the data were obtained. In Fig. 1, glycine was applied for 15 min and washed out for 6 min before HFS. During this period (at least 21 min), we did not record IPSCs since our focus was on the changes of IPSCs after HFS, therefore it is not clear whether IPSCs have been changed or not. Conversely, in Fig. 3, a 5 min exposure to 10 μM glycine decreased IPSCs, which is consistent with our previous report, that activation of the GlyRs on the GABAergic terminals reduces GABA release (Ye et al., 2004).
According to the model circuit, LTPGABA should decrease activity of dopamine neurons and the release of dopamine. The hypo-dopamine state has been associated with the motivation for ethanol drinking (Gonzales et al., 2004, Chen et al., 2011). Interestingly, as described earlier, GABAA antagonists are rewarding at the same time as they reduce ethanol intake. The underlying mechanism needed to be determined. The rewarding effect of GABAA antagonists may be resulted from the ability of these agents to increase activity of dopamine neurons and dopamine levels in the target areas. Higher dopamine levels in the nucleus accumbens may cause rewarding effects. Under these conditions, the animals would not need to drink as much ethanol as that in normal conditions to achieve the wanted effect, thereby resulting in reduced ethanol consumption. Although the current study demonstrates that glycine blocks LTPGABA on dopamine neurons in the VTA of ethanol-naïve rats, it remains to be determined whether glycine has the same effect on these neurons from rats chronically exposed to ethanol. The effects of glycine on GABA transmission under such conditions may have been changed since GABAergic transmission has been changed after chronic ethanol exposure (Gallegos et al., 1999). Additional limitation of current study is that we used 11 mM glucose in aCSF that is far beyond the physiological level of 2.5 mM. Future study using 2.5 mM glucose is needed to determine whether this will affect the outcome.
Collectively, our data support an important role for glutamate release in the maintenance of heterosynaptic efficacy in the VTA. A reduction in glutamate release by activation of strychnine sensitive GlyRs on glutamate terminals in VTA-dopamine neurons triggers a rapid decrease in heterosynaptic efficacy (GABAergic synapse). Together, the current study reports a previous unrecognized effect of glycine on the LTPGABA recorded from VTA-dopamine neurons. These data provide new insight into glycine’s effects on brain reward pathways and could lead to the development of new therapies against alcohol use disorders.
This work was made possible by NIH grants (AA01694; AA 021657; AA 022292, JHY), New Jersey Health foundation (PC109-13, JHY), and National Natural Science Foundation of China (grant 81371463, YZG); Graduate innovative research programs of Mudanjiang Medical University, China. (No. 2014YJSCX-04 & YJSCX2012-380HLJ, YZG)
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
The authors declare no conflict of interest.