- Email: [email protected]
GABAB receptor-mediated presynaptic inhibition of glycinergic transmission onto substantia gelatinosa neurons in the rat spinal cord
Pain 138 (2008) 330–342 www.elsevier.com/locate/pain
GABAB receptor-mediated presynaptic inhibition of glycinergic transmission onto substantia gelatinosa neurons in the rat spinal cord In-Sun Choi a, Jin-Hwa Cho a, Seok-Gwon Jeong a, Jung-Soo Hong a, Sang-Jung Kim b, Jun Kim b, Maan-Gee Lee c,d, Byung-Ju Choi a, Il-Sung Jang a,d,* a
Department of Pharmacology, School of Dentistry, Kyungpook National University, 188-1 Samduk 2 ga-dong, Jung-gu, Daegu 700-412, Republic of Korea b Department of Physiology, School of Medicine, Seoul National University, Seoul 110-799, Republic of Korea c Department of Pharmacology, School of Medicine, Kyungpook National University, Daegu 700-412, Republic of Korea d Brain Science & Engineering Institute, Kyungpook National University, Daegu 700-412, Republic of Korea Received 26 October 2007; received in revised form 17 December 2007; accepted 8 January 2008
Abstract The GABAB receptor-mediated presynaptic inhibition of glycinergic transmission was studied from young rat substantia gelatinosa (SG) neurons using a conventional whole-cell patch clamp technique. Action potential-dependent glycinergic inhibitory postsynaptic currents (IPSCs) were recorded from SG neurons in the presence of 3 mM kynurenic acid and 10 lM SR95531. In these conditions, baclofen (30 lM), a selective GABAB receptor agonist, greatly reduced the amplitude of glycinergic IPSCs and increased the paired-pulse ratio. Such effects were completely blocked by 3 lM CGP55845, a selective GABAB receptor antagonist, indicating that the activation of presynaptic GABAB receptors decreases glycinergic synaptic transmission. Glycinergic IPSCs were largely dependent on Ca2+ influxes passing through presynaptic N- and P/Q-type Ca2+ channels, and these channels contributed equally to the baclofen-induced inhibition of glycinergic IPSCs. However, the baclofen-induced inhibition of glycinergic IPSCs was not affected by either 100 lM SQ22536, an adenylyl cyclase inhibitor, or 1 mM Ba2+, a G-protein coupled inwardly rectifying K+ channel blocker. During the train stimulation (10 pulses at 20 Hz), which caused a marked synaptic depression of glycinergic IPSCs, baclofen at a 30 lM concentration completely blocked glycinergic synaptic depression, but at a 3 lM concentration it largely preserved glycinergic synaptic depression. Such GABAB receptor-mediated dynamic changes in short-term synaptic plasticity of glycinergic transmission onto SG neurons might contribute to the central processing of sensory signals. Ó 2008 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved. Keywords: GABAB receptor; Substantia gelatinosa; Glycinergic IPSCs; Presynaptic inhibition; Pain
1. Introduction Glycine is the primary inhibitory neurotransmitter in the brain stem as well as the spinal cord, and exerts its *
Corresponding author. Address: Department of Pharmacology, School of Dentistry, Kyungpook National University, 188-1 Samduk 2 ga-dong, Jung-gu, Daegu 700-412, Republic of Korea. Tel.: +82 53 660 6887; fax: +82 53 424 5130. E-mail address: [email protected] (I.-S. Jang).
inhibitory actions by activating strychnine-sensitive glycine receptor-Cl channel complexes. In the spinal cord including the substantia gelatinosa (SG), glycine is often co-localized with GABA [43] and the corelease of glycine and GABA has been demonstrated [11,22,26]. Glycinergic, in addition to GABAergic, inhibitory neurotransmission also plays a pivotal role in the modulation of pain signals as well as neuronal excitability (for review, [15]). For example, coadministration of
0304-3959/$34.00 Ó 2008 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.pain.2008.01.005
I.-S. Choi et al. / Pain 138 (2008) 330–342
intrathecal strychnine and bicuculline, selective glycine and GABAA receptor antagonists, elicits synergistic allodynia in the rat [30], while an intrathecal administration of glycine and related compounds reduces the mechanonociceptive response or thermal hyperalgesia [38,39]. In addition, the long-lasting inhibition of glycine receptors rather than GABAA receptors has been implicated in prostaglandin E2-mediated central sensitization [2]. These results suggest that glycinergic inhibitory transmission within the dorsal horn plays a pivotal role in the process of pain signals. GABAB receptors belong to the super family of G-protein coupled seven transmembrane receptors, and are coupled to Gi/o proteins. The activation of GABAB receptors mediates a variety of intracellular signal transduction pathways including adenylyl cyclase (AC), G-protein coupled inwardly rectifying K+ (GIRK) channels and voltage-dependent Ca2+ channels (VDCCs) (for review, [35]). GABAB receptors are widely expressed on the CNS including the spinal cord as well as dorsal root ganglia [6,31,33,45]. In the spinal cord, GABAB receptors play an important role in the modulation of nociception [16], as demonstrated by previous studies showing that baclofen, a selective GABAB receptor agonist, induces antinociception against formalin-induced pain behaviors [12,41]. The baclofen-induced antinociceptive actions seem to be mediated by either postsynaptic or presynaptic GABAB receptors. For example, baclofen induces a membrane hyperpolarization by activating K+ channels in spinal dorsal horn neurons [25]. In addition, baclofen inhibits the expression of Fos protein in the spinal dorsal horn evoked by innocuous stimuli [9]. The presynaptic GABAB receptor-mediated presynaptic inhibition of both glutamatergic and GABAergic transmissions onto SG neurons also contributes to the baclofen-induced antinociceptive actions [3,19]. However, it is still unknown whether presynaptic GABAB receptors regulate glycinergic synaptic transmission in the spinal dorsal horn area. The present study, therefore, has investigated the functional roles of presynaptic GABAB receptors expressed on glycinergic nerve terminals innervating to SG neurons.
331
den, CT, USA) in a cold low-Na+ medium (in mM; 230 sucrose, 2 KCl, 1 KH2PO4, 1 MgCl2, 0.5 CaCl2, 26 NaHCO3 and 10 glucose) saturated with 95% O2 and 5% CO2. Slices were kept in artificial cerebrospinal fluid (ACSF; 120 NaCl, 2 KCl, 1 KH2PO4, 26 NaHCO3, 2 CaCl2, 1 MgCl2 and 10 glucose) saturated with 95% O2 and 5% CO2 at room temperature for at least 1 h before electrophysiological recording. Thereafter the slices were transferred into a recording chamber, and the superficial dorsal horn was identified under an upright microscope (E600FN, Nikon, Tokyo, Japan) with a waterimmersion objective (40). The ACSF routinely contained 3 mM kynurenic acid and 10 lM SR95531 to block ionotropic glutamate and GABAA receptors, respectively, except where indicated. The bath was perfused with ACSF at 2 ml/min by the use of a peristaltic pump (MP-1000, EYELA, Tokyo, Japan). 2.2. Electrical measurements All electrical measurements were performed by use of a computer-controlled patch clamp amplifier (MultiClamp 700B; Molecular Devices; Union City, CA, USA). For whole-cell recording, patch pipettes were made from borosilicate capillary glass (1.5 mm outer diameter, 0.9 mm inner diameter; G-1.5; Narishige, Tokyo, Japan) by use of a pipette puller (P-97; Shutter Instrument Co., Novato, CA, USA). The resistance of the recording pipettes filled with internal solution (in mM; 140 CsMeHSO3, 5 TEA-Cl, 5 CsCl, 2 EGTA, 2 Mg-ATP and 10 Hepes, pH 7.2 with Tris-base) was 4–6 MX. Membrane currents were filtered at 1 kHz (MultiClamp Commander; Molecular Devices), digitized at 4 kHz (Digidata 1322A, Molecular Devices), and stored on a computer equipped with pCLAMP 10.0 (Molecular Devices). In whole-cell recordings, 10 mV hyperpolarizing step pulses (30 ms in duration) were periodically delivered to monitor the access resistance, and recordings were discontinued if access resistance changed by more than 15%. All slice experiments were performed at room temperature (22–25 °C). To record action potential-dependent glycinergic IPSCs, a glass stimulation pipette (10 lm diameter), filled with a bath solution, was positioned around the lamina II region. Brief paired-pulses (500 ls, 100–200 lA, 10 Hz), except where indicated, were applied by the pipette at a stimulation frequency of 0.1 Hz using a stimulator (SEN-7203, Nihon Kohden, Tokyo, Japan) equipped with an isolator unit (SS-701 J, Nihon Kohden).
2. Materials and methods 2.3. Data analysis 2.1. Preparations All experiments complied with the guiding principles for the care and use of animals approved by the Council of the Physiological Society of Korea and the National Institutes of Health Guide for the Care and Use of Laboratory Animals, and every effort was made to minimize both the number of animals used and their suffering. Sprague Dawley rats (12–15 d-old) were decapitated under ketamine anesthesia (100 mg/kg, i.p.). The spinal cord was dissected and longitudinally sliced at a thickness of 300 lm by use of a microslicer (VibratomeÒ 1000; Warner Instruments, Ham-
The amplitudes of action potential-dependent IPSCs were calculated by subtracting the baseline from the peak amplitude. The effect of baclofen was quantified as a percentage change in IPSC amplitude compared to the control values. Spontaneous mIPSCs (mIPSCs) were counted and analyzed using the MiniAnalysis program (Synaptosoft, Inc., Decatur, GA), as described previously [20]. Briefly, mIPSCs were screened automatically using an amplitude threshold of 10 pA and then visually accepted or rejected based upon the rise and decay times. The average values of both the frequency and amplitude of mIPSCs during the control period (5–10 min)
332
I.-S. Choi et al. / Pain 138 (2008) 330–342
were calculated for each recording, and the frequency and amplitude of all the events during different experimental conditions were normalized to these values. The effect of baclofen was quantified as a percentage change in mIPSC frequency compared to the control values. The inter-event intervals and amplitudes of a large number of mIPSCs obtained from the same neuron were examined by constructing cumulative probability distributions and these distributions were compared under different conditions using the Kolmogorov–Smirnov (K–S) test with Stat View software (SAS Institute, Inc.). The continuous curves for the concentration–inhibition relationship of baclofen were fitted using a least-squares fit to the following equation: I ¼ 1 ðC n =ðC n þ ICn50 ÞÞ; where I is the inhibition ratio of baclofen-induced IPSC amplitude, C is the concentration of baclofen, IC50 is the concentration for the half-inhibitory response and n is the Hill coefficient. Numerical values are provided as the means ± standard error of the mean (SEM) using values normalized to the control. Significant differences in the mean amplitude and frequency were tested using the Student’s two-tailed paired t-test, using absolute values rather than normalized ones. Values of p < 0.05 were considered significant. 2.4. Drugs The drugs used in the present study were baclofen, kynurenic acid, strychnine, 9-(tetrahydro-2-furanyl)-9H-purin-6amine (SQ22536), forskolin (from Sigma, St. Louis, MO, USA) and (2S)-3-[[(1S)-1-(3,4-dichlorophenyl)ethyl]amino-2hydroxypropyl](phenylmethyl) phosphinic acid (CGP55845), 6-imino-3-(4-methoxyphenyl)-1(6H)-pyridazinebutanoic acid HBr (SR95531), tetrodotoxin (TTX) (from Tocris, UK), and x-agatoxin IVA (x-AgTx), x-conotoxin GVIA (x-CgTx) (from Peptide institute, Osaka, Japan). All drugs were applied by bath application (2 ml/min).
3. Results 3.1. Activation of presynaptic GABAB receptors inhibits action potential-dependent glycine release In the presence of 3 mM kynurenic acid, which blocks ionotropic glutamate receptors, IPSCs were recorded at a VH of 0 mV by an electrical stimulation through a glass pipette placed near to the lamina II area. In most of the SG neurons recorded, the IPSCs were completely blocked by the cumulative application of both 10 lM SR95531, a selective GABAA receptor antagonist, and 1 lM strychnine, a glycine receptor antagonist, indicating that the recorded IPSCs were a mixture of GABAergic and glycinergic IPSCs (Fig. 1A). In order to investigate whether presynaptic GABAB receptors modulate action potential-dependent glycinergic synaptic transmission on to lamina II neurons, all the following experiments were performed in the presence of 10 lM SR95531. In these conditions, the effect of baclofen on
glycinergic IPSCs evoked by pairing stimulation at an interval of 100 ms (10 Hz) was observed. As shown in Fig. 1B and C, bath applied baclofen (30 lM) reversibly decreased the first IPSC (IPSC1) amplitude to 28.1 ± 2.7% of the control (n = 24, p < 0.01), and increased the paired-pulse ratio (PPR) from 1.05 ± 0.03 to 2.02 ± 0.22 (n = 24, p < 0.01), suggesting that baclofen acts presynaptically to decrease the probability of glycine release. The inhibitory effect of baclofen on glycinergic IPSCs was completely attenuated in the presence of 3 lM CGP55845, a selective GABAB receptor antagonist (99.4 ± 2.5% of the CGP55845 condition, n = 6, p = 0.83, Fig. 2A and B). In a subset of experiments, GDP-bS (0.3 mM) was added to the pipette solution in order to exclude, if any, an influence of postsynaptic GABAB receptor activation. Bath applied baclofen again decreased the IPSC1 amplitude to 30.0 ± 2.3% of the control (n = 5, p < 0.05), even with a pipette solution containing GDP-bS. The inhibition ratios were not significantly different between the absence and presence of GDP-bS. In addition, baclofen inhibited glycinergic IPSCs in a concentration-dependent manner with an IC50 value of 6.0 lM (Fig. 2C). The results suggest that activation of presynaptic GABAB receptors inhibits action potential-dependent glycine release onto lamina II neurons. On the other hand, CGP55845 had no effect on either the IPSC1 amplitude or the PPR (Fig. 2A and B), indicating that there is little tonic activation of presynaptic GABAB receptors. 3.2. Endogenous GABA modulates glycine release by acting on presynaptic GABAB receptors In the spinal cord, the co-release of both GABA and glycine from a subset of inhibitory nerve terminals has been demonstrated [26]. This suggests that presynaptic GABAB receptors on co-releasing nerve terminals might act as autoreceptors to regulate their own release. To test this possibility, brief train stimuli (five stimuli at 20 Hz) at an interval of 30 s were applied, and the effect of CGP55845 on glycinergic IPSCs was observed during the train. In 6 of 19 neurons tested, CGP55845 (3 lM) had no effect on IPSC1 amplitude (102.2 ± 0.5% of the control, n = 6, p = 0.88), but it increased IPSC5 amplitude to 125.2 ± 3.3% of the control (n = 6, p < 0.01) and the IPSC5/IPSC1 ratio (Fig. 3A). The extent of CGP55845-induced increase in IPSC amplitude was larger in later IPSCs during the train (Fig. 3B). The results suggest that synaptically released GABA during the repetitive stimulation could act presynaptic GABAB receptors on glycinergic nerve terminals to decrease glycine release. In the remaining 13 neurons, however, CGP55845 had no apparent effect on glycinergic IPSCs within the train (data not shown).
I.-S. Choi et al. / Pain 138 (2008) 330–342
333
Fig. 1. Effects of baclofen on glycinergic IPSCs. (A) A typical time course of IPSC amplitude before and during the application of 10 lM SR95531 (SR) and SR95531 plus 1 lM strychnine (STR). Insets represent typical traces of the numbered region. Note that the cumulative application of both SR95531 and strychnine eliminated all IPSCs. (B) A typical time course of the IPSC1 amplitude (a) and paired-pulse ratio (IPSC2/IPSC1; (b) before, during and after the application of 30 lM baclofen. Insets represent typical traces of the numbered region. (C) Baclofen-induced changes in the IPSC1 amplitude (a) and paired-pulse ratio (b). Each column was normalized to the control and represents the mean and SEM from 24 experiments. ** p < 0.01.
3.3. Mechanisms underlying GABAB receptor-mediated presynaptic inhibition of action potential-dependent glycine release Activation of presynaptic GABAB receptors inhibits multiple types of voltage-dependent Ca2+ channels (VDCCs) to decrease neurotransmitter release at various synapses [10,13,18,28]. Since multiple types of voltage-dependent Ca2+ channels (VDCCs) control the neurotransmitter release, and especially, N- and P/Qtypes are closely involved in the action potential-dependent transmitter release [47], the effects of x-conotoxin GVIA (x-CgTx; a selective N-type VDCC blocker) and x-agatoxin IVA (x-AgTx; a selective P/Q-type VDCC blocker) on the baclofen-induced inhibition of glycinergic IPSCs were observed. As expected, the cumulative application of both 2 lM x-CgTx and 200 nM x-AgTx
decreased the glycinergic IPSC amplitude to 21.0 ± 6.4% of the control (n = 5, p < 0.01, Fig. 4A). The results indicate that action potential-dependent glycine release onto SG neurons is largely mediated by presynaptic Ca2+ influxes via both N- and P/Q-type VDCCs, and that the remaining glycinergic IPSC might be mediated by other VDCC subtypes such as R-type VDCCs. As shown in Fig. 4B, 30 lM baclofen alone decreased IPSC amplitude by 78.7 ± 5.7% of the control (n = 4, p < 0.05). In the same neurons, 200 nM x-AgTx decreased IPSC amplitude by 53.5 ± 7.2% of the control (n = 4, p < 0.05), and the subsequent application of baclofen further decreased the remaining IPSCs by 31.2 ± 7.0% of the initial IPSC amplitude (n = 4, p < 0.05). The extent of IPSC inhibition by baclofen alone was not significantly different from that by baclofen plus x-AgTx (84.8 ± 1.4%). Similarly, as shown
334
I.-S. Choi et al. / Pain 138 (2008) 330–342
Fig. 2. Presynaptic GABAB receptor-mediated inhibition of glycinergic IPSCs. (A) A typical time course of the IPSC1 amplitude before, during and after application of 30 lM baclofen in the absence or presence of 3 lM CGP55845. Insets represent typical traces of the numbered region. (B) Baclofen-induced changes in the IPSC1 amplitude (a) and paired-pulse ratio (b). Each column was normalized to the control and represents the mean and SEM from six experiments. **p < 0.01, n.s, not significant. (C) Concentration–response relationship of baclofen. IC50 was 6.0 lM. Each point and error bar represents the mean and SEM from 6 to 12 experiments.
in Fig. 4C, 30 lM baclofen alone decreased IPSC amplitude by 49.7 ± 5.5% of the control (n = 4, p < 0.05). In the same neurons, 2 lM x-CgTx decreased IPSC amplitude by 37.9 ± 4.3% of the control (n = 5, p < 0.05), and the subsequent application of baclofen further decreased the remaining IPSCs by 14.3 ± 5.8% of the initial IPSC amplitude (n = 5, p < 0.05). The extent of IPSC inhibition by baclofen alone was not significantly different from that by baclofen plus x-CgTx (52.2 ± 2.9%). The results suggest that both N- and P/Q-type VDCCs are approximately equally involved in the baclofen-induced inhibition of glycine release. GABAB receptors are seven-transmembrane proteins coupled to Gi/o proteins and negatively coupled to adenylyl cyclase (AC), which increases intracellular cAMP concentration. cAMP within presynaptic nerve terminals is known to regulate neurotransmitter release via PKA-dependent and/or PKA-independent signal transduction pathways [5,24,37]. If presynaptic GABAB receptors decrease glycinergic IPSCs by inhibiting AC, the direct blockade of AC should decrease the glycinergic IPSC amplitude. However, this was not the case because SQ22536, a selective AC inhibitor, even at a 100 lM concentration, had no effect on either the IPSC1 amplitude (101 ± 9.2% of the control, n = 5, p = 0.92) or the PPR (95.2 ± 2.3% of the control, n = 5,
p = 0.16, Fig. 5A and B). In addition, the baclofeninduced inhibition of glycinergic IPSCs was not affected by 100 lM SQ22536 (Fig. 5C and D). The activation of GABAB receptors is also known to open G-protein coupled inwardly rectifying K+ (GIRK) channels to hyperpolarize presynaptic nerve terminals [42]. Therefore, the effect of Ba2+, a GIRK channel blocker [40], on baclofen-induced inhibition of glycinergic IPSCs was observed. Ba2+ at a 1 mM concentration did not change either glycinergic IPSC amplitude (100.0 ± 5.4% of the control, n = 5, p = 0.83) or the PPR (99.5 ± 1.0% of the control, n = 5, p = 0.64) (Fig. 6A and B). The baclofen-induced inhibition of glycinergic IPSC amplitude was not affected in the presence of Ba2+ (25.4 ± 4.2% of the control and 30.3 ± 2.9% of the Ba2+ condition, n = 5, p = 0.08) (Fig. 6A and B). The results suggest that the baclofen-induced inhibition of glycinergic IPSCs is not related to the activation of presynaptic GIRK channels. 3.4. GABAB receptor-mediated presynaptic inhibition of action potential-independent glycine release The activation of GABAB receptors inhibits spontaneous miniature neurotransmitter release, which is independent on the Ca2+ influx from extracellular space
I.-S. Choi et al. / Pain 138 (2008) 330–342
335
out affecting the mean mIPSC amplitude (93.9 ± 4.2% of the control, n = 10, p = 0.31) (Fig. 7A and B insets). In addition, as shown in Fig. 7B, baclofen significantly shifted the distribution of the inter-event interval to the right, indicating a decrease in mIPSC frequency. However, baclofen did not change the distribution of the current amplitude. The results again suggest that baclofen acts presynaptically to reduce the probability of spontaneous glycine release. Such an inhibitory action of baclofen on glycinergic mIPSC frequency was completely blocked by 3 lM CGP55845 (n = 5) (Fig. 7B). However, the baclofen-induced decrease in glycinergic mIPSC frequency was not affected by SQ22536 (100 lM, n = 6), forskolin (10 lM, n = 7), Cd2+ (200 lM, n = 7), or Ba2+ (1 mM, n = 6) (Fig. 7C). Taken together, the results suggest that the activation of presynaptic GABAB receptors affects the downstream of a Ca2+ influx to decrease the action potential-independent spontaneous glycine release. 3.5. Activation of presynaptic GABAB receptors reduces glycinergic synaptic depression
Fig. 3. Effect of CGP55845 on glycinergic IPSC amplitude during the train stimulation. (A) Typical time courses of the first IPSC (IPSC1) and 5th IPSC (IPSC5) amplitudes (a) and the IPSC5/IPSC1 ratio (b) before, during and after the application of 3 lM CGP55845. Insets represent typical traces of the numbered region. (B) CGP55845induced changes in IPSC amplitude during the train stimulation. Each point represents the mean and SEM from 6 experiments. *p < 0.05, ** p < 0.01. Note that CGP55845 had no effect on the first three IPSCs but increased the last two IPSCs within the train.
to presynaptic nerve terminals [8,19]. In a subset of experiments, therefore, the effect of baclofen on glycinergic mIPSCs was observed. Glycinergic mIPSCs were recorded from SG neurons in the presence of 300 nM TTX, which completely blocks voltage-dependent Na+ channels, 3 mM kynurenic acid and 10 lM SR95531. In 10 SG neurons, in which the effect of baclofen was fully analyzed, baclofen decreased the mean mIPSC frequency to 57.1 ± 2.5% of the control (p < 0.01), with-
Finally, this study observed whether the activation of presynaptic GABAB receptors affects the glycinergic synaptic depression. To test this, brief train stimuli (10 stimuli at 20 Hz) were applied at an interval of 30 s, which allowed for the complete recovery of the IPSC1 amplitude without depression in between the trains, and the effect of baclofen on glycinergic IPSC amplitudes was observed during the train. As shown in Fig. 8A, 3 lM baclofen decreased IPSC1 to 78.0 ± 3.2% of the control (n = 8, p < 0.05), but it significantly increased the IPSC10 amplitude to 133.9 ± 9.8% of the control (n = 8, p < 0.05) (Fig. 8A). The IPSC10/IPSC1 ratio in the 3 lM baclofen condition was 0.51 ± 0.06 significantly larger that that in the control condition (0.30 ± 0.03, p < 0.05). Whereas, 30 lM baclofen decreased IPSC1 to 34.0 ± 2.4% of the control (n = 8, p < 0.01), but it hardly affected the IPSC10 amplitude (107.3 ± 00% of the control, n = 8, p = 0.50, Fig. 8A). However, 30 lM baclofen greatly increased the IPSC10/IPSC1 ratio (0.30 ± 0.03 in the control and 0.95 ± 0.09 in 30 lM baclofen, n = 8, p < 0.01) (Fig. 8Ac). That is, while baclofen at a 30 lM concentration completely blocked glycinergic synaptic depression, it at a 3 lM concentration somewhat preserved glycinergic synaptic depression induced by the train stimulation (Fig. 8B). The effects of baclofen on the absolute amplitudes of glycinergic IPSCs were also observed during the train. As summarized in Fig. 8C, baclofen at a 30 lM concentration decreased the amplitudes of the initial four IPSCs, but it had no effect on the amplitudes of remaining IPSCs. On the contrary, while baclofen at a 3 lM concentration decreased the amplitudes of the initial three
336
I.-S. Choi et al. / Pain 138 (2008) 330–342
Fig. 4. Effects of specific VDCC antagonists on baclofen-induced decrease in glycinergic IPSCs. (Aa) A typical time course of the IPSC1 amplitude before and during the application of 200 nM x-AgTx and 2 lM x-CgTx. (b) VDCC blocker-induced changes in the IPSC1 amplitude. Each column was normalized to the control and represents the mean and SEM from five experiments. *p < 0.05, **p < 0.01. (Ba) A typical time course of the IPSC1 amplitude before, during and after application of 30 lM baclofen in the absence or presence of 200 nM x-AgTx. (b) The extent of the IPSC1 amplitude inhibition in each condition. Each column represents the mean and SEM from four experiments. n.s, not significant. (Ca) A typical time course of IPSC1 amplitude before, during and after application of 30 lM baclofen in the absence or presence of 2 lM x-CgTx. (b) The extent of IPSC1 amplitude inhibition in each condition. Each column represents the mean and SEM from five experiments. n.s, not significant.
IPSCs, it did significantly increase the last three IPSCs (Fig. 8C). The results suggest that while the inhibitory action of 30 lM baclofen on glycinergic IPSCs during
the train stimulation disappears at later IPSCs, the action of 3 lM baclofen shifts from the inhibition to potentiation.
I.-S. Choi et al. / Pain 138 (2008) 330–342
337
Fig. 5. Effect of SQ22536 on baclofen-induced decrease in glycinergic IPSCs. (A) Typical traces of glycinergic IPSCs observed before and during application of 100 lM SQ22536, a specific adenylyl cyclase inhibitor. (B) SQ22536-induced changes in the IPSC1 amplitude (a) and paired-pulse ratio (b). Each column was normalized to the control and represents the mean and SEM from 6 neurons. Note the poor effect of SQ22536 on glycinergic IPSCs. n.s; not significant. (C) A typical time course of the IPSC1 amplitude before, during and after application of 30 lM baclofen in the absence or presence of 100 lM SQ22536. Insets represent typical traces of the numbered region. (D) Baclofen-induced changes in the IPSC1 amplitude (a) and paired-pulse ratio (b). Each column was normalized to the control and represents the mean and SEM from five experiments. **p < 0.01.
4. Discussion 4.1. GABAB receptor-mediated presynaptic inhibition of glycine release onto SG neurons GABAB receptors are known to modulate neurotransmitter release as heteroreceptors or autoreceptors at a number of central synapses. In the spinal dorsal horn area, the activation of presynaptic GABAB receptors inhibits not only glutamate release from primary afferent terminals but also GABA release from inhibitory nerve terminals [19,25]. The present results also provide evidence that presynaptic GABAB receptors are expressed on glycinergic nerve terminals projecting to SG neurons, and that their activation decreases action potential-dependent and -independent glycine
release onto SG neurons. Several lines of evidence support the conclusion that presynaptic GABAB receptors are responsible for the baclofen-induced inhibition of glycinergic IPSCs. First, exogenously applied baclofen simultaneously decreased glycinergic IPSC amplitude and increased the PPR, consistent with baclofen acting presynaptically to decrease the probability of glycine release from inhibitory nerve terminals. In addition, baclofen decreased glycinergic mIPSC frequency without affecting the current amplitude. On the other hand, GABAB receptors are also expressed on postsynaptic SG neurons, and their activation increases K+ conductance [25], which would be expected to decrease incoming synaptic currents by shunting mechanisms. However, the involvement of postsynaptic GABAB receptors in the baclofen-induced inhibition of glycinergic IPSCs should
338
I.-S. Choi et al. / Pain 138 (2008) 330–342
in a subset of SG neurons (6 of 19 neurons tested; 32%) although the inhibitory synaptic currents recorded from SG neurons were a mixture of both glycinergic and GABAergic IPSCs in most of the recordings (data not shown). One possibility to explain this difference would be that GABAB receptors on a subset of presynaptic terminals, which release both GABA and glycine simultaneously, act as autoreceptors to regulate their own release, as shown in auditory brainstem neurons [29]. In fact, an electrophysiological study has demonstrated that a small portion (20%) of mIPSCs recorded from SG neurons are mixed mIPSCs mediated by corelease of both GABA and glycine [26], although most of glycinergic neurons in the spinal dorsal horn area also contain GABA [34,44]. Therefore, it is highly feasible that presynaptic GABAB receptors might act as autoreceptors to regulate both GABA and glycine release from a subset of inhibitory nerve terminals. 4.2. Mechanisms underlying GABAB receptor-mediated presynaptic inhibition of glycine release onto SG neurons
Fig. 6. Effect of Ba2+ on baclofen-induced decrease in glycinergic IPSCs. (A) A typical time course of the IPSC1 amplitude before, during and after application of 30 lM baclofen in the absence or presence of 1 mM Ba2+, a GIRK channel inhibitor. Insets represent typical traces of the numbered region. (B) Baclofen-induced changes in the IPSC1 amplitude (a) and paired-pulse ratio (b). Each column was normalized to the control and represents the mean and SEM from five experiments. **p < 0.01, n.s, not significant.
be negligible because baclofen also decreased IPSC amplitude even with the pipette solution containing GDP-bS. On the other hand, CGP55845 had no effect on either the IPSC1 amplitude or the PPR during the paired stimulation. The results indicate that there might be little tonic activation of GABAB receptors on most of glycinergic nerve terminals innervating to SG neurons. However, CGP55845 did increase the later but not initial IPSCs during the repetitive stimulation in a subset of SG neurons, suggesting that an accumulation of synaptically released GABA during the intense neural activity should be needed to activate presynaptic GABAB receptors and thus inhibit glycine release from a subpopulation of glycinergic nerve terminals. However, the facilitatory effect of CGP55845 on the later IPSCs recorded during the train stimulation was only observed
The mechanisms underlying the presynaptic inhibition of action potential-dependent neurotransmitter release mediated by G-protein coupled receptors including GABAB receptors are diverse [47]. For example, the activation of presynaptic GABAB receptors decrease neurotransmitter release at various central synapses by inhibiting multiple types of presynaptic VDCCs, mainly N- and P/Q-type Ca2+ channels [13,18,28], although presynaptic GABAB receptors selectively inhibit P/Qtype Ca2+ channels in suprachiasmatic nucleus neurons [10]. The present results also suggest that GABAB receptors expressed on glycinergic nerve terminals projecting to SG neurons might inhibit both N- and P/Q-type Ca2+ channels because baclofen still decreased glycinergic IPSCs in the presence of either x-AgTx (a P/Q-type Ca2+ channel blocker) or x-CgTx (an N-type Ca2+ channel blocker), and the baclofen-induced decrease in glycinergic IPSCs was largely occluded in the presence of both toxins (data not shown). However, the involvement of other VDCC subtypes, such as R-type, in the GABAB receptor-mediated presynaptic inhibition of glycine release cannot be ruled out. On the other hand, GABAB receptors are negatively coupled to AC to decrease the cAMP production [23], or open GIRK channels to hyperpolarize presynaptic nerve terminals [42]. In the present study, however, the contribution of either AC or GIRK channels to the baclofen-induced inhibition of glycinergic IPSCs would be negligible because the blockade of AC or GIRK channels by SQ22536 or Ba2+ did not affect the inhibitory action of baclofen on glycinergic IPSCs. The activation of presynaptic GABAB receptors also inhibits spontaneous miniature neurotransmitter release, which is independent on the Ca2+ influx from
I.-S. Choi et al. / Pain 138 (2008) 330–342
339
Fig. 7. Effect of baclofen on glycinergic mIPSCs. (Aa) A typical trace of glycinergic mIPSCs recorded before, during and after the application of 30 lM baclofen. Insets represent typical traces with an expanded time scale. (b) An all-point scatter plot of mIPSC amplitude as shown in (Aa). 752 events were plotted. (B) Cumulative probability distributions for the inter-event interval (a; p < 0.01, K–S test) and current amplitude (b; p = 0.63, K–S test) of glycinergic mIPSCs shown in (Aa). 203 for the control and 139 events for baclofen were plotted. Insets, 30 lM baclofen-induced changes in mIPSC frequency (left) and amplitude (right). Each column was the mean and SEM from 10 experiments. **p < 0.01. (C) 30 lM baclofen-induced changes of mIPSC frequency in the control condition and in the presence of 3 lM CGP55845 (n = 5), 100 lM SQ22536 (n = 6), 10 lM forskolin (n = 7), 1 mM Ba2+ (n = 6) and 200 lM Cd2+ (n = 7).
extracellular space to presynaptic nerve terminals [8]. In the present study, the baclofen significantly decreased glycinergic mIPSC frequency without affecting the current amplitude, suggesting the presynaptic locus of baclofen action. Since the baclofen-induced decrease in glycinergic mIPSC frequency was not affected by SQ22536, forskolin, Cd2+, or Ba2+, GABAB receptors expressed on glycinergic nerve terminals projecting to SG neurons might inhibit Ca2+-independent glycine release by modulating some presynaptic kinases, such as PKC and/or PKA, as demonstrated by previous reports [21,27]. However, further studies are needed to reveal the exact mechanisms underlying GABAB receptor-mediated inhibition of glycinergic mIPSCs in SG neurons. 4.3. Physiological implications Activity-dependent synaptic depression is a major form of short-term plasticity of intrinsic synaptic transmission. This synaptic depression has been suggested to equalize the synaptic strength of neocortical synapses having different initial release probabilities [1,32]. A
potential mechanism underlying synaptic depression may include the depletion of readily releasable vesicles at the synapses with high release probability [7,14,28], although other mechanisms, such as the activation of presynaptic autoreceptors [4,46], and the desensitization of postsynaptic receptors [36], have also been suggested. Therefore, presynaptic GABAB receptors may affect synaptic depression by regulating the release probability. For example, the activation of presynaptic GABAB receptors is known to block frequency-dependent synaptic depression by reducing the exhaustion of the readily releasable pool at GABAergic synapses of hippocampal CA3 interneurons [28], although synaptic depression is likely to be largely preserved during the GABAB receptor activation at basket cell-granule cell GABAergic synapses of the hippocampal dentate gyrus [17]. At glutamatergic synapses of the nucleus magnocellularis, the activation of presynaptic GABAB receptors shifts from the synaptic depression even to facilitation, resulting in the enhancement of synaptic efficacy [7]. In the present study, the strong activation of presynaptic GABAB receptors was found to completely block
340
I.-S. Choi et al. / Pain 138 (2008) 330–342
Fig. 8. Baclofen-induced reduction in synaptic depression of glycinergic IPSCs during train stimulation. (A) Typical time courses of the first IPSC (IPSC1) and 10th IPSC (IPSC10) amplitudes (a) and the IPSC10/IPSC1 ratio (b) before, during and after the application of 3 lM or 30 lM baclofen. Insets represent typical traces of the numbered region. (B) Baclofen (a; 30 lM; b, 3 lM)-induced changes in the mean ratio of IPSCn/IPSC1 plotted against the number within the train (n). Each point represents the mean and SEM from eight experiments. All points were normalized to the IPSC1 amplitude of the control condition, but diamond symbols were normalized to IPSC1 amplitude of the baclofen condition. (C) Baclofen-induced changes in the IPSC amplitude within the train. Each point represents the mean and SEM from eight experiments. *p < 0.05, **p < 0.01. Note that the action of 3 lM baclofen shifts from the inhibition to potentiation within the train.
synaptic depression of glycinergic transmission onto SG neurons. The decrease in initial IPSC amplitude during the strong activation of presynaptic GABAB receptors would further contribute to the blockade of synaptic depression by decreasing the desensitization of postsynaptic glycine receptors. The resultant blockade of glycinergic synaptic depression during the strong activation of presynaptic GABAB receptors would sustain the minimal glycinergic inhibition during the high neuronal activity, as baclofen did not decrease the absolute amplitudes of later IPSCs during the train. On the contrary, the weak activation of presynaptic GABAB receptors largely preserved synaptic depression of glycinergic transmission onto SG neurons, suggesting that the extent of changes in initial release probability indeed affects the extent of synaptic depression. However, it should be noted that the weak activation of presynaptic
GABAB receptors increased the absolute amplitudes of later glycinergic IPSCs during the train. The results suggest that a small decrease in initial release probability could increase inhibitory synaptic efficacy during the high neuronal activity. A potential mechanism underlying the baclofeninduced antinociception involves presynaptic inhibition of glutamate release from primary afferent terminals [3,19], although baclofen also induces a membrane hyperpolarization by activating K+ channels in spinal dorsal horn neurons [25]. However, the activation of presynaptic GABAB receptors also decreases both GABAergic [19] and glycinergic synaptic transmission onto SG neurons (present results). This GABAB receptor-mediated reduction of inhibitory synaptic transmission would complicate the baclofen-induced antinociceptive action, because baclofen has a similar affinity to GABA (IC50 = 4.3 lM),
I.-S. Choi et al. / Pain 138 (2008) 330–342
glutamate (IC50 = 4.4 lM) [19], and glycine release (IC50 = 6.0 lM). In the present study, it has been shown that baclofen could affect short-term synaptic depression of glycinergic transmission during the high neuronal activity in a concentration-dependent manner. Baclofen at higher concentrations would warrant the minimal glycinergic inhibition, and at lower concentration it would increase the efficacy of glycinergic inhibition during the high neuronal activity. Such effects of baclofen on short-term plasticity of glycinergic transmission would further contribute to the baclofen-induced antinociceptive action. However, more studies such as the baclofen actions on GABAergic and glutamatergic short-term plasticity are needed to understand the baclofen-induced antinociceptive action.
Acknowledgements We thank Prof. Patria Morris (Kyungpook National University) for correcting the English. This work was supported by the Korea Science and Engineering Foundation (KOSEF) Grant (No. R01-2006-000-10406-0). I.-S. Choi was supported by Brain Korea 21 Project.
References [1] Abbott LF, Varela JA, Sen K, Nelson SB. Synaptic depression and cortical gain control. Science 1997;275:220–4. [2] Ahmadi S, Lippross S, Neuhuber WL, Zeilhofer HU. PGE2 selectively blocks inhibitory glycinergic neurotransmission onto rat superficial dorsal horn neurons. Nat Neurosci 2002;5:34–40. [3] Ataka T, Kumamoto E, Shimoji K, Yoshimura M. Baclofen inhibits more effectively C-afferent than Ad-afferent glutamatergic transmission in substantia gelatinosa neurons of adult rat spinal cord slices. Pain 2000;86:273–82. [4] Barnes-Davies M, Forsythe ID. Pre- and postsynaptic glutamate receptors at a giant excitatory synapse in rat auditory brainstem slices. J Physiol (Lond) 1995;488:387–406. [5] Beaumont V, Zucker RS. Enhancement of synaptic transmission by cyclic AMP modulation of presynaptic Ih channels. Nat Neurosci 2000;3:133–41. [6] Bischoff S, Leonhard S, Reymann N, Schuler V, Shigemoto R, Kaupmann K, et al. Spatial distribution of GABAB R1 receptor mRNA and binding sites in the rat brain. J Comp Neurol 1999;412:1–16. [7] Brenowitz S, David J, Trussell L. Enhancement of synaptic efficacy by presynaptic GABAB receptors. Neuron 1998;20:135–41. [8] Capogna M, Gahwiler BH, Thompson SM. Presynaptic inhibition of calcium-dependent and -independent release elicited with ionomycin, gadolinium, and a-latrotoxin in the hippocampus. J Neurophysiol 1996;75:2017–28. [9] Castro AR, Bowery N, Castro-Lopes JM. Baclofen and midazolam alter c-fos induction by peripheral noxious or innocuous stimulation in the spinal cord of normal and monoarthritic rats. Neuropharmacology 1999;38:1775–88. [10] Chen G, van den Pol AN. Presynaptic GABAB autoreceptor modulation of P/Q-type calcium channels and GABA release in rat suprachiasmatic nucleus neurons. J Neurosci 1998;18:1913–22.
341
[11] Chery N, De Koninck Y. Junctional versus extrajunctional glycine and GABAA receptor-mediated IPSCs in identified lamina I neurons of the adult rat spinal cord. J Neurosci 1999;17:7342–55. [12] Dirig DM, Yaksh TL. Intrathecal baclofen and muscimol, but not midazolam, are antinociceptive using the rat-formalin model. J Pharmacol Exp Ther 1995;275:219–27. [13] Doze VA, Cohen GA, Madison DV. Calcium channel involvement in GABAB receptor-mediated inhibition of GABA release in area CA1 of the rat hippocampus. J Neurophysiol 1995;74:43–53. [14] Elmquist D, Quastel D. A quantitative study of end-plate potentials in isolated human muscle. J Physiol (Lond) 1965;178:505–29. [15] Furue H, Katafuchi T, Yoshimura M. Sensory processing and functional reorganization of sensory transmission under pathological conditions in the spinal dorsal horn. Neurosci Res 2004;48:361–8. [16] Hammond DL. GABAB receptors: new tricks by an old dog. Curr Opin Pharmacol 2001;1:26–30. [17] Hefft S, Kraushaar U, Geiger JR, Jonas P. Presynaptic short-term depression is maintained during regulation of transmitter release at a GABAergic synapse in rat hippocampus. J Physiol (Lond) 2002;539:201–8. [18] Huston E, Cullen GP, Burley JR, Dolphin AC. The involvement of multiple calcium channel sub-types in glutamate release from cerebellar granule cells and its modulation by GABAB receptor activation. Neuroscience 1995;68:465–78. [19] Iyadomi M, Iyadomi I, Kumamoto E, Tomokuni K, Yoshimura M. Presynaptic inhibition by baclofen of miniature EPSCs and IPSCs in substantia gelatinosa neurons of the adult rat spinal dorsal horn. Pain 2000;85:385–93. [20] Jang IS, Jeong HJ, Katsurabayashi S, Akaike N. Functional roles of presynaptic GABAA receptors on glycinergic nerve terminals in the rat spinal cord. J Physiol (Lond) 2002;541:423–34. [21] Jarolimek W, Misgeld U. GABAB receptor-mediated inhibition of tetrodotoxin-resistant GABA release in rodent hippocampal CA1 pyramidal cells. J Neurosci 1997;17:1025–32. [22] Jonas P, Bischofberger J, Sandkuhler J. Corelease of two fast neurotransmitters at a central synapse. Science 1998;281: 419–24. [23] Kamatchi GL, Ticku MK. Functional coupling of presynaptic GABAB receptors with voltage-gated Ca2+ channel: regulation by protein kinases A and C in cultured spinal cord neurons. Mol Pharmacol 1990;38:342–7. [24] Kaneko M, Takahashi T. Presynaptic mechanism underlying cAMP-dependent synaptic potentiation. J Neurosci 2004;24:5202–8. [25] Kangrga I, Jiang MC, Randic M. Actions of ()-baclofen on rat dorsal horn neurons. Brain Res 1991;562:265–75. [26] Keller AF, Coull JA, Chery N, Poisbeau P, De Koninck Y. Region-specific developmental specialization of GABA-glycine cosynapses in laminas I–II of the rat spinal dorsal horn. J Neurosci 2001;21:7871–80. [27] Kubota H, Katsurabayashi S, Moorhouse AJ, Murakami N, Koga H, Akaike N. GABAB receptor transduction mechanisms, and cross-talk between protein kinases A and C, in GABAergic terminals synapsing onto neurons of the rat nucleus basalis of Meynert. J Physiol (Lond) 2003;551:263–76. [28] Lei S, McBain CJ. GABAB receptor modulation of excitatory and inhibitory synaptic transmission onto rat CA3 hippocampal interneurons. J Physiol (Lond) 2003;546:439–53. [29] Lim R, Alvarez FJ, Walmsley B. GABA mediates presynaptic inhibition at glycinergic synapses in a rat auditory brainstem nucleus. J Physiol (Lond) 2000;525:447–59. [30] Loomis CW, Khandwala H, Osmond G, Hefferan MP. Coadministration of intrathecal strychnine and bicuculline effects synergistic allodynia in the rat: an isobolographic analysis. J Pharmacol Exp Ther 2001;296:756–61.
342
I.-S. Choi et al. / Pain 138 (2008) 330–342
[31] Margeta-Mitrovic M, Mitrovic I, Riley RC, Jan LY, Basbaum AI. Immunohistochemical localization of GABAB receptors in the rat central nervous system. J Comp Neurol 1999;405:299–321. [32] Markram H, Tsodyks M. Redistribution of synaptic efficacy between neocortical pyramidal neurons. Nature 1996;382:807–10. [33] McCarson KE, Enna SJ. Nociceptive regulation of GABAB receptor gene expression in rat spinal cord. Neuropharmacology 1999;38:1767–73. [34] Mitchell K, Spike RC, Todd AJ. An immunocytochemical study of glycine receptor and GABA in laminae I–III of rat spinal dorsal horn. J Neurosci 1993;13:2371–81. [35] Mott DD, Lewis DV. The pharmacology and function of central GABAB receptors. Int Rev Neurobiol 1994;36:97–223. [36] Otis T, Zhang S, Trussell LO. Direct measurement of AMPA receptor desensitization induced by glutamatergic synaptic transmission. J Neurosci 1996;16:7496–504. [37] Salin PA, Malenka RC, Nicoll RA. Cyclic AMP mediates a presynaptic form of LTP at cerebellar parallel fiber synapses. Neuron 1996;16:797–803. [38] Simpson Jr RK, Gondo M, Robertson CS, Goodman JC. Reduction in the mechanonociceptive response by intrathecal administration of glycine and related compounds. Neurochem Res 1996;21:1221–6. [39] Simpson Jr RK, Gondo M, Robertson CS, Goodman JC. Reduction in thermal hyperalgesia by intrathecal administration of glycine and related compounds. Neurochem Res 1997;22:75–9.
[40] Takigawa T, Alzheimer C. Phasic and tonic attenuation of EPSPs by inward rectifier K+ channels in rat hippocampal pyramidal cells. J Physiol (Lond) 2002;539:67–75. [41] Thomas DA, Navarrete IM, Graham BA, McGowan MK, Hammond DL. Antinociception produced by systemic R(+)baclofen hydrochloride is attenuated by CGP 35348 administered to the spinal cord or ventromedial medulla of rats. Brain Res 1996;718:129–37. [42] Thompson SM, Gahwiler BH. Comparison of the actions of baclofen at pre- and postsynaptic receptors in the rat hippocampus in vitro. J Physiol (Lond) 1992;451:329–45. [43] Todd AJ, Spike RC. The localization of classical transmitters and neuropeptides within neurons in laminae I–III of the mammalian spinal dorsal horn. Prog Neurobiol 1993;41:609–45. [44] Todd AJ, Sullivan AC. Light microscope study of the coexistence of GABA-like and glycine-like immunoreactivities in the spinal cord of the rat. J Comp Neurol 1990;296:496–505. [45] Towers S, Princivalle A, Billinton A, Edmunds M, Bettler B, Urban L, et al. GABAB receptor protein and mRNA distribution in rat spinal cord and dorsal root ganglia. Eur J Neurosci 2000;12:3201–10. [46] von Gersdorff H, Schneggenburger R, Weis S, Neher E. Presynaptic depression at a calyx synapse: the small contribution of metabotropic glutamate receptors. J Neurosci 1997;17:8137–46. [47] Wu LG, Saggau P. Presynaptic inhibition of elicited neurotransmitter release. Trends Neurosci 1997;20:204–12.
GABAB receptor-mediated presynaptic inhibition of glycinergic transmission onto substantia gelatinosa neurons in the rat spinal cord In-Sun Choi a, Jin-Hwa Cho a, Seok-Gwon Jeong a, Jung-Soo Hong a, Sang-Jung Kim b, Jun Kim b, Maan-Gee Lee c,d, Byung-Ju Choi a, Il-Sung Jang a,d,* a
Department of Pharmacology, School of Dentistry, Kyungpook National University, 188-1 Samduk 2 ga-dong, Jung-gu, Daegu 700-412, Republic of Korea b Department of Physiology, School of Medicine, Seoul National University, Seoul 110-799, Republic of Korea c Department of Pharmacology, School of Medicine, Kyungpook National University, Daegu 700-412, Republic of Korea d Brain Science & Engineering Institute, Kyungpook National University, Daegu 700-412, Republic of Korea Received 26 October 2007; received in revised form 17 December 2007; accepted 8 January 2008
Abstract The GABAB receptor-mediated presynaptic inhibition of glycinergic transmission was studied from young rat substantia gelatinosa (SG) neurons using a conventional whole-cell patch clamp technique. Action potential-dependent glycinergic inhibitory postsynaptic currents (IPSCs) were recorded from SG neurons in the presence of 3 mM kynurenic acid and 10 lM SR95531. In these conditions, baclofen (30 lM), a selective GABAB receptor agonist, greatly reduced the amplitude of glycinergic IPSCs and increased the paired-pulse ratio. Such effects were completely blocked by 3 lM CGP55845, a selective GABAB receptor antagonist, indicating that the activation of presynaptic GABAB receptors decreases glycinergic synaptic transmission. Glycinergic IPSCs were largely dependent on Ca2+ influxes passing through presynaptic N- and P/Q-type Ca2+ channels, and these channels contributed equally to the baclofen-induced inhibition of glycinergic IPSCs. However, the baclofen-induced inhibition of glycinergic IPSCs was not affected by either 100 lM SQ22536, an adenylyl cyclase inhibitor, or 1 mM Ba2+, a G-protein coupled inwardly rectifying K+ channel blocker. During the train stimulation (10 pulses at 20 Hz), which caused a marked synaptic depression of glycinergic IPSCs, baclofen at a 30 lM concentration completely blocked glycinergic synaptic depression, but at a 3 lM concentration it largely preserved glycinergic synaptic depression. Such GABAB receptor-mediated dynamic changes in short-term synaptic plasticity of glycinergic transmission onto SG neurons might contribute to the central processing of sensory signals. Ó 2008 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved. Keywords: GABAB receptor; Substantia gelatinosa; Glycinergic IPSCs; Presynaptic inhibition; Pain
1. Introduction Glycine is the primary inhibitory neurotransmitter in the brain stem as well as the spinal cord, and exerts its *
Corresponding author. Address: Department of Pharmacology, School of Dentistry, Kyungpook National University, 188-1 Samduk 2 ga-dong, Jung-gu, Daegu 700-412, Republic of Korea. Tel.: +82 53 660 6887; fax: +82 53 424 5130. E-mail address: [email protected] (I.-S. Jang).
inhibitory actions by activating strychnine-sensitive glycine receptor-Cl channel complexes. In the spinal cord including the substantia gelatinosa (SG), glycine is often co-localized with GABA [43] and the corelease of glycine and GABA has been demonstrated [11,22,26]. Glycinergic, in addition to GABAergic, inhibitory neurotransmission also plays a pivotal role in the modulation of pain signals as well as neuronal excitability (for review, [15]). For example, coadministration of
0304-3959/$34.00 Ó 2008 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.pain.2008.01.005
I.-S. Choi et al. / Pain 138 (2008) 330–342
intrathecal strychnine and bicuculline, selective glycine and GABAA receptor antagonists, elicits synergistic allodynia in the rat [30], while an intrathecal administration of glycine and related compounds reduces the mechanonociceptive response or thermal hyperalgesia [38,39]. In addition, the long-lasting inhibition of glycine receptors rather than GABAA receptors has been implicated in prostaglandin E2-mediated central sensitization [2]. These results suggest that glycinergic inhibitory transmission within the dorsal horn plays a pivotal role in the process of pain signals. GABAB receptors belong to the super family of G-protein coupled seven transmembrane receptors, and are coupled to Gi/o proteins. The activation of GABAB receptors mediates a variety of intracellular signal transduction pathways including adenylyl cyclase (AC), G-protein coupled inwardly rectifying K+ (GIRK) channels and voltage-dependent Ca2+ channels (VDCCs) (for review, [35]). GABAB receptors are widely expressed on the CNS including the spinal cord as well as dorsal root ganglia [6,31,33,45]. In the spinal cord, GABAB receptors play an important role in the modulation of nociception [16], as demonstrated by previous studies showing that baclofen, a selective GABAB receptor agonist, induces antinociception against formalin-induced pain behaviors [12,41]. The baclofen-induced antinociceptive actions seem to be mediated by either postsynaptic or presynaptic GABAB receptors. For example, baclofen induces a membrane hyperpolarization by activating K+ channels in spinal dorsal horn neurons [25]. In addition, baclofen inhibits the expression of Fos protein in the spinal dorsal horn evoked by innocuous stimuli [9]. The presynaptic GABAB receptor-mediated presynaptic inhibition of both glutamatergic and GABAergic transmissions onto SG neurons also contributes to the baclofen-induced antinociceptive actions [3,19]. However, it is still unknown whether presynaptic GABAB receptors regulate glycinergic synaptic transmission in the spinal dorsal horn area. The present study, therefore, has investigated the functional roles of presynaptic GABAB receptors expressed on glycinergic nerve terminals innervating to SG neurons.
331
den, CT, USA) in a cold low-Na+ medium (in mM; 230 sucrose, 2 KCl, 1 KH2PO4, 1 MgCl2, 0.5 CaCl2, 26 NaHCO3 and 10 glucose) saturated with 95% O2 and 5% CO2. Slices were kept in artificial cerebrospinal fluid (ACSF; 120 NaCl, 2 KCl, 1 KH2PO4, 26 NaHCO3, 2 CaCl2, 1 MgCl2 and 10 glucose) saturated with 95% O2 and 5% CO2 at room temperature for at least 1 h before electrophysiological recording. Thereafter the slices were transferred into a recording chamber, and the superficial dorsal horn was identified under an upright microscope (E600FN, Nikon, Tokyo, Japan) with a waterimmersion objective (40). The ACSF routinely contained 3 mM kynurenic acid and 10 lM SR95531 to block ionotropic glutamate and GABAA receptors, respectively, except where indicated. The bath was perfused with ACSF at 2 ml/min by the use of a peristaltic pump (MP-1000, EYELA, Tokyo, Japan). 2.2. Electrical measurements All electrical measurements were performed by use of a computer-controlled patch clamp amplifier (MultiClamp 700B; Molecular Devices; Union City, CA, USA). For whole-cell recording, patch pipettes were made from borosilicate capillary glass (1.5 mm outer diameter, 0.9 mm inner diameter; G-1.5; Narishige, Tokyo, Japan) by use of a pipette puller (P-97; Shutter Instrument Co., Novato, CA, USA). The resistance of the recording pipettes filled with internal solution (in mM; 140 CsMeHSO3, 5 TEA-Cl, 5 CsCl, 2 EGTA, 2 Mg-ATP and 10 Hepes, pH 7.2 with Tris-base) was 4–6 MX. Membrane currents were filtered at 1 kHz (MultiClamp Commander; Molecular Devices), digitized at 4 kHz (Digidata 1322A, Molecular Devices), and stored on a computer equipped with pCLAMP 10.0 (Molecular Devices). In whole-cell recordings, 10 mV hyperpolarizing step pulses (30 ms in duration) were periodically delivered to monitor the access resistance, and recordings were discontinued if access resistance changed by more than 15%. All slice experiments were performed at room temperature (22–25 °C). To record action potential-dependent glycinergic IPSCs, a glass stimulation pipette (10 lm diameter), filled with a bath solution, was positioned around the lamina II region. Brief paired-pulses (500 ls, 100–200 lA, 10 Hz), except where indicated, were applied by the pipette at a stimulation frequency of 0.1 Hz using a stimulator (SEN-7203, Nihon Kohden, Tokyo, Japan) equipped with an isolator unit (SS-701 J, Nihon Kohden).
2. Materials and methods 2.3. Data analysis 2.1. Preparations All experiments complied with the guiding principles for the care and use of animals approved by the Council of the Physiological Society of Korea and the National Institutes of Health Guide for the Care and Use of Laboratory Animals, and every effort was made to minimize both the number of animals used and their suffering. Sprague Dawley rats (12–15 d-old) were decapitated under ketamine anesthesia (100 mg/kg, i.p.). The spinal cord was dissected and longitudinally sliced at a thickness of 300 lm by use of a microslicer (VibratomeÒ 1000; Warner Instruments, Ham-
The amplitudes of action potential-dependent IPSCs were calculated by subtracting the baseline from the peak amplitude. The effect of baclofen was quantified as a percentage change in IPSC amplitude compared to the control values. Spontaneous mIPSCs (mIPSCs) were counted and analyzed using the MiniAnalysis program (Synaptosoft, Inc., Decatur, GA), as described previously [20]. Briefly, mIPSCs were screened automatically using an amplitude threshold of 10 pA and then visually accepted or rejected based upon the rise and decay times. The average values of both the frequency and amplitude of mIPSCs during the control period (5–10 min)
332
I.-S. Choi et al. / Pain 138 (2008) 330–342
were calculated for each recording, and the frequency and amplitude of all the events during different experimental conditions were normalized to these values. The effect of baclofen was quantified as a percentage change in mIPSC frequency compared to the control values. The inter-event intervals and amplitudes of a large number of mIPSCs obtained from the same neuron were examined by constructing cumulative probability distributions and these distributions were compared under different conditions using the Kolmogorov–Smirnov (K–S) test with Stat View software (SAS Institute, Inc.). The continuous curves for the concentration–inhibition relationship of baclofen were fitted using a least-squares fit to the following equation: I ¼ 1 ðC n =ðC n þ ICn50 ÞÞ; where I is the inhibition ratio of baclofen-induced IPSC amplitude, C is the concentration of baclofen, IC50 is the concentration for the half-inhibitory response and n is the Hill coefficient. Numerical values are provided as the means ± standard error of the mean (SEM) using values normalized to the control. Significant differences in the mean amplitude and frequency were tested using the Student’s two-tailed paired t-test, using absolute values rather than normalized ones. Values of p < 0.05 were considered significant. 2.4. Drugs The drugs used in the present study were baclofen, kynurenic acid, strychnine, 9-(tetrahydro-2-furanyl)-9H-purin-6amine (SQ22536), forskolin (from Sigma, St. Louis, MO, USA) and (2S)-3-[[(1S)-1-(3,4-dichlorophenyl)ethyl]amino-2hydroxypropyl](phenylmethyl) phosphinic acid (CGP55845), 6-imino-3-(4-methoxyphenyl)-1(6H)-pyridazinebutanoic acid HBr (SR95531), tetrodotoxin (TTX) (from Tocris, UK), and x-agatoxin IVA (x-AgTx), x-conotoxin GVIA (x-CgTx) (from Peptide institute, Osaka, Japan). All drugs were applied by bath application (2 ml/min).
3. Results 3.1. Activation of presynaptic GABAB receptors inhibits action potential-dependent glycine release In the presence of 3 mM kynurenic acid, which blocks ionotropic glutamate receptors, IPSCs were recorded at a VH of 0 mV by an electrical stimulation through a glass pipette placed near to the lamina II area. In most of the SG neurons recorded, the IPSCs were completely blocked by the cumulative application of both 10 lM SR95531, a selective GABAA receptor antagonist, and 1 lM strychnine, a glycine receptor antagonist, indicating that the recorded IPSCs were a mixture of GABAergic and glycinergic IPSCs (Fig. 1A). In order to investigate whether presynaptic GABAB receptors modulate action potential-dependent glycinergic synaptic transmission on to lamina II neurons, all the following experiments were performed in the presence of 10 lM SR95531. In these conditions, the effect of baclofen on
glycinergic IPSCs evoked by pairing stimulation at an interval of 100 ms (10 Hz) was observed. As shown in Fig. 1B and C, bath applied baclofen (30 lM) reversibly decreased the first IPSC (IPSC1) amplitude to 28.1 ± 2.7% of the control (n = 24, p < 0.01), and increased the paired-pulse ratio (PPR) from 1.05 ± 0.03 to 2.02 ± 0.22 (n = 24, p < 0.01), suggesting that baclofen acts presynaptically to decrease the probability of glycine release. The inhibitory effect of baclofen on glycinergic IPSCs was completely attenuated in the presence of 3 lM CGP55845, a selective GABAB receptor antagonist (99.4 ± 2.5% of the CGP55845 condition, n = 6, p = 0.83, Fig. 2A and B). In a subset of experiments, GDP-bS (0.3 mM) was added to the pipette solution in order to exclude, if any, an influence of postsynaptic GABAB receptor activation. Bath applied baclofen again decreased the IPSC1 amplitude to 30.0 ± 2.3% of the control (n = 5, p < 0.05), even with a pipette solution containing GDP-bS. The inhibition ratios were not significantly different between the absence and presence of GDP-bS. In addition, baclofen inhibited glycinergic IPSCs in a concentration-dependent manner with an IC50 value of 6.0 lM (Fig. 2C). The results suggest that activation of presynaptic GABAB receptors inhibits action potential-dependent glycine release onto lamina II neurons. On the other hand, CGP55845 had no effect on either the IPSC1 amplitude or the PPR (Fig. 2A and B), indicating that there is little tonic activation of presynaptic GABAB receptors. 3.2. Endogenous GABA modulates glycine release by acting on presynaptic GABAB receptors In the spinal cord, the co-release of both GABA and glycine from a subset of inhibitory nerve terminals has been demonstrated [26]. This suggests that presynaptic GABAB receptors on co-releasing nerve terminals might act as autoreceptors to regulate their own release. To test this possibility, brief train stimuli (five stimuli at 20 Hz) at an interval of 30 s were applied, and the effect of CGP55845 on glycinergic IPSCs was observed during the train. In 6 of 19 neurons tested, CGP55845 (3 lM) had no effect on IPSC1 amplitude (102.2 ± 0.5% of the control, n = 6, p = 0.88), but it increased IPSC5 amplitude to 125.2 ± 3.3% of the control (n = 6, p < 0.01) and the IPSC5/IPSC1 ratio (Fig. 3A). The extent of CGP55845-induced increase in IPSC amplitude was larger in later IPSCs during the train (Fig. 3B). The results suggest that synaptically released GABA during the repetitive stimulation could act presynaptic GABAB receptors on glycinergic nerve terminals to decrease glycine release. In the remaining 13 neurons, however, CGP55845 had no apparent effect on glycinergic IPSCs within the train (data not shown).
I.-S. Choi et al. / Pain 138 (2008) 330–342
333
Fig. 1. Effects of baclofen on glycinergic IPSCs. (A) A typical time course of IPSC amplitude before and during the application of 10 lM SR95531 (SR) and SR95531 plus 1 lM strychnine (STR). Insets represent typical traces of the numbered region. Note that the cumulative application of both SR95531 and strychnine eliminated all IPSCs. (B) A typical time course of the IPSC1 amplitude (a) and paired-pulse ratio (IPSC2/IPSC1; (b) before, during and after the application of 30 lM baclofen. Insets represent typical traces of the numbered region. (C) Baclofen-induced changes in the IPSC1 amplitude (a) and paired-pulse ratio (b). Each column was normalized to the control and represents the mean and SEM from 24 experiments. ** p < 0.01.
3.3. Mechanisms underlying GABAB receptor-mediated presynaptic inhibition of action potential-dependent glycine release Activation of presynaptic GABAB receptors inhibits multiple types of voltage-dependent Ca2+ channels (VDCCs) to decrease neurotransmitter release at various synapses [10,13,18,28]. Since multiple types of voltage-dependent Ca2+ channels (VDCCs) control the neurotransmitter release, and especially, N- and P/Qtypes are closely involved in the action potential-dependent transmitter release [47], the effects of x-conotoxin GVIA (x-CgTx; a selective N-type VDCC blocker) and x-agatoxin IVA (x-AgTx; a selective P/Q-type VDCC blocker) on the baclofen-induced inhibition of glycinergic IPSCs were observed. As expected, the cumulative application of both 2 lM x-CgTx and 200 nM x-AgTx
decreased the glycinergic IPSC amplitude to 21.0 ± 6.4% of the control (n = 5, p < 0.01, Fig. 4A). The results indicate that action potential-dependent glycine release onto SG neurons is largely mediated by presynaptic Ca2+ influxes via both N- and P/Q-type VDCCs, and that the remaining glycinergic IPSC might be mediated by other VDCC subtypes such as R-type VDCCs. As shown in Fig. 4B, 30 lM baclofen alone decreased IPSC amplitude by 78.7 ± 5.7% of the control (n = 4, p < 0.05). In the same neurons, 200 nM x-AgTx decreased IPSC amplitude by 53.5 ± 7.2% of the control (n = 4, p < 0.05), and the subsequent application of baclofen further decreased the remaining IPSCs by 31.2 ± 7.0% of the initial IPSC amplitude (n = 4, p < 0.05). The extent of IPSC inhibition by baclofen alone was not significantly different from that by baclofen plus x-AgTx (84.8 ± 1.4%). Similarly, as shown
334
I.-S. Choi et al. / Pain 138 (2008) 330–342
Fig. 2. Presynaptic GABAB receptor-mediated inhibition of glycinergic IPSCs. (A) A typical time course of the IPSC1 amplitude before, during and after application of 30 lM baclofen in the absence or presence of 3 lM CGP55845. Insets represent typical traces of the numbered region. (B) Baclofen-induced changes in the IPSC1 amplitude (a) and paired-pulse ratio (b). Each column was normalized to the control and represents the mean and SEM from six experiments. **p < 0.01, n.s, not significant. (C) Concentration–response relationship of baclofen. IC50 was 6.0 lM. Each point and error bar represents the mean and SEM from 6 to 12 experiments.
in Fig. 4C, 30 lM baclofen alone decreased IPSC amplitude by 49.7 ± 5.5% of the control (n = 4, p < 0.05). In the same neurons, 2 lM x-CgTx decreased IPSC amplitude by 37.9 ± 4.3% of the control (n = 5, p < 0.05), and the subsequent application of baclofen further decreased the remaining IPSCs by 14.3 ± 5.8% of the initial IPSC amplitude (n = 5, p < 0.05). The extent of IPSC inhibition by baclofen alone was not significantly different from that by baclofen plus x-CgTx (52.2 ± 2.9%). The results suggest that both N- and P/Q-type VDCCs are approximately equally involved in the baclofen-induced inhibition of glycine release. GABAB receptors are seven-transmembrane proteins coupled to Gi/o proteins and negatively coupled to adenylyl cyclase (AC), which increases intracellular cAMP concentration. cAMP within presynaptic nerve terminals is known to regulate neurotransmitter release via PKA-dependent and/or PKA-independent signal transduction pathways [5,24,37]. If presynaptic GABAB receptors decrease glycinergic IPSCs by inhibiting AC, the direct blockade of AC should decrease the glycinergic IPSC amplitude. However, this was not the case because SQ22536, a selective AC inhibitor, even at a 100 lM concentration, had no effect on either the IPSC1 amplitude (101 ± 9.2% of the control, n = 5, p = 0.92) or the PPR (95.2 ± 2.3% of the control, n = 5,
p = 0.16, Fig. 5A and B). In addition, the baclofeninduced inhibition of glycinergic IPSCs was not affected by 100 lM SQ22536 (Fig. 5C and D). The activation of GABAB receptors is also known to open G-protein coupled inwardly rectifying K+ (GIRK) channels to hyperpolarize presynaptic nerve terminals [42]. Therefore, the effect of Ba2+, a GIRK channel blocker [40], on baclofen-induced inhibition of glycinergic IPSCs was observed. Ba2+ at a 1 mM concentration did not change either glycinergic IPSC amplitude (100.0 ± 5.4% of the control, n = 5, p = 0.83) or the PPR (99.5 ± 1.0% of the control, n = 5, p = 0.64) (Fig. 6A and B). The baclofen-induced inhibition of glycinergic IPSC amplitude was not affected in the presence of Ba2+ (25.4 ± 4.2% of the control and 30.3 ± 2.9% of the Ba2+ condition, n = 5, p = 0.08) (Fig. 6A and B). The results suggest that the baclofen-induced inhibition of glycinergic IPSCs is not related to the activation of presynaptic GIRK channels. 3.4. GABAB receptor-mediated presynaptic inhibition of action potential-independent glycine release The activation of GABAB receptors inhibits spontaneous miniature neurotransmitter release, which is independent on the Ca2+ influx from extracellular space
I.-S. Choi et al. / Pain 138 (2008) 330–342
335
out affecting the mean mIPSC amplitude (93.9 ± 4.2% of the control, n = 10, p = 0.31) (Fig. 7A and B insets). In addition, as shown in Fig. 7B, baclofen significantly shifted the distribution of the inter-event interval to the right, indicating a decrease in mIPSC frequency. However, baclofen did not change the distribution of the current amplitude. The results again suggest that baclofen acts presynaptically to reduce the probability of spontaneous glycine release. Such an inhibitory action of baclofen on glycinergic mIPSC frequency was completely blocked by 3 lM CGP55845 (n = 5) (Fig. 7B). However, the baclofen-induced decrease in glycinergic mIPSC frequency was not affected by SQ22536 (100 lM, n = 6), forskolin (10 lM, n = 7), Cd2+ (200 lM, n = 7), or Ba2+ (1 mM, n = 6) (Fig. 7C). Taken together, the results suggest that the activation of presynaptic GABAB receptors affects the downstream of a Ca2+ influx to decrease the action potential-independent spontaneous glycine release. 3.5. Activation of presynaptic GABAB receptors reduces glycinergic synaptic depression
Fig. 3. Effect of CGP55845 on glycinergic IPSC amplitude during the train stimulation. (A) Typical time courses of the first IPSC (IPSC1) and 5th IPSC (IPSC5) amplitudes (a) and the IPSC5/IPSC1 ratio (b) before, during and after the application of 3 lM CGP55845. Insets represent typical traces of the numbered region. (B) CGP55845induced changes in IPSC amplitude during the train stimulation. Each point represents the mean and SEM from 6 experiments. *p < 0.05, ** p < 0.01. Note that CGP55845 had no effect on the first three IPSCs but increased the last two IPSCs within the train.
to presynaptic nerve terminals [8,19]. In a subset of experiments, therefore, the effect of baclofen on glycinergic mIPSCs was observed. Glycinergic mIPSCs were recorded from SG neurons in the presence of 300 nM TTX, which completely blocks voltage-dependent Na+ channels, 3 mM kynurenic acid and 10 lM SR95531. In 10 SG neurons, in which the effect of baclofen was fully analyzed, baclofen decreased the mean mIPSC frequency to 57.1 ± 2.5% of the control (p < 0.01), with-
Finally, this study observed whether the activation of presynaptic GABAB receptors affects the glycinergic synaptic depression. To test this, brief train stimuli (10 stimuli at 20 Hz) were applied at an interval of 30 s, which allowed for the complete recovery of the IPSC1 amplitude without depression in between the trains, and the effect of baclofen on glycinergic IPSC amplitudes was observed during the train. As shown in Fig. 8A, 3 lM baclofen decreased IPSC1 to 78.0 ± 3.2% of the control (n = 8, p < 0.05), but it significantly increased the IPSC10 amplitude to 133.9 ± 9.8% of the control (n = 8, p < 0.05) (Fig. 8A). The IPSC10/IPSC1 ratio in the 3 lM baclofen condition was 0.51 ± 0.06 significantly larger that that in the control condition (0.30 ± 0.03, p < 0.05). Whereas, 30 lM baclofen decreased IPSC1 to 34.0 ± 2.4% of the control (n = 8, p < 0.01), but it hardly affected the IPSC10 amplitude (107.3 ± 00% of the control, n = 8, p = 0.50, Fig. 8A). However, 30 lM baclofen greatly increased the IPSC10/IPSC1 ratio (0.30 ± 0.03 in the control and 0.95 ± 0.09 in 30 lM baclofen, n = 8, p < 0.01) (Fig. 8Ac). That is, while baclofen at a 30 lM concentration completely blocked glycinergic synaptic depression, it at a 3 lM concentration somewhat preserved glycinergic synaptic depression induced by the train stimulation (Fig. 8B). The effects of baclofen on the absolute amplitudes of glycinergic IPSCs were also observed during the train. As summarized in Fig. 8C, baclofen at a 30 lM concentration decreased the amplitudes of the initial four IPSCs, but it had no effect on the amplitudes of remaining IPSCs. On the contrary, while baclofen at a 3 lM concentration decreased the amplitudes of the initial three
336
I.-S. Choi et al. / Pain 138 (2008) 330–342
Fig. 4. Effects of specific VDCC antagonists on baclofen-induced decrease in glycinergic IPSCs. (Aa) A typical time course of the IPSC1 amplitude before and during the application of 200 nM x-AgTx and 2 lM x-CgTx. (b) VDCC blocker-induced changes in the IPSC1 amplitude. Each column was normalized to the control and represents the mean and SEM from five experiments. *p < 0.05, **p < 0.01. (Ba) A typical time course of the IPSC1 amplitude before, during and after application of 30 lM baclofen in the absence or presence of 200 nM x-AgTx. (b) The extent of the IPSC1 amplitude inhibition in each condition. Each column represents the mean and SEM from four experiments. n.s, not significant. (Ca) A typical time course of IPSC1 amplitude before, during and after application of 30 lM baclofen in the absence or presence of 2 lM x-CgTx. (b) The extent of IPSC1 amplitude inhibition in each condition. Each column represents the mean and SEM from five experiments. n.s, not significant.
IPSCs, it did significantly increase the last three IPSCs (Fig. 8C). The results suggest that while the inhibitory action of 30 lM baclofen on glycinergic IPSCs during
the train stimulation disappears at later IPSCs, the action of 3 lM baclofen shifts from the inhibition to potentiation.
I.-S. Choi et al. / Pain 138 (2008) 330–342
337
Fig. 5. Effect of SQ22536 on baclofen-induced decrease in glycinergic IPSCs. (A) Typical traces of glycinergic IPSCs observed before and during application of 100 lM SQ22536, a specific adenylyl cyclase inhibitor. (B) SQ22536-induced changes in the IPSC1 amplitude (a) and paired-pulse ratio (b). Each column was normalized to the control and represents the mean and SEM from 6 neurons. Note the poor effect of SQ22536 on glycinergic IPSCs. n.s; not significant. (C) A typical time course of the IPSC1 amplitude before, during and after application of 30 lM baclofen in the absence or presence of 100 lM SQ22536. Insets represent typical traces of the numbered region. (D) Baclofen-induced changes in the IPSC1 amplitude (a) and paired-pulse ratio (b). Each column was normalized to the control and represents the mean and SEM from five experiments. **p < 0.01.
4. Discussion 4.1. GABAB receptor-mediated presynaptic inhibition of glycine release onto SG neurons GABAB receptors are known to modulate neurotransmitter release as heteroreceptors or autoreceptors at a number of central synapses. In the spinal dorsal horn area, the activation of presynaptic GABAB receptors inhibits not only glutamate release from primary afferent terminals but also GABA release from inhibitory nerve terminals [19,25]. The present results also provide evidence that presynaptic GABAB receptors are expressed on glycinergic nerve terminals projecting to SG neurons, and that their activation decreases action potential-dependent and -independent glycine
release onto SG neurons. Several lines of evidence support the conclusion that presynaptic GABAB receptors are responsible for the baclofen-induced inhibition of glycinergic IPSCs. First, exogenously applied baclofen simultaneously decreased glycinergic IPSC amplitude and increased the PPR, consistent with baclofen acting presynaptically to decrease the probability of glycine release from inhibitory nerve terminals. In addition, baclofen decreased glycinergic mIPSC frequency without affecting the current amplitude. On the other hand, GABAB receptors are also expressed on postsynaptic SG neurons, and their activation increases K+ conductance [25], which would be expected to decrease incoming synaptic currents by shunting mechanisms. However, the involvement of postsynaptic GABAB receptors in the baclofen-induced inhibition of glycinergic IPSCs should
338
I.-S. Choi et al. / Pain 138 (2008) 330–342
in a subset of SG neurons (6 of 19 neurons tested; 32%) although the inhibitory synaptic currents recorded from SG neurons were a mixture of both glycinergic and GABAergic IPSCs in most of the recordings (data not shown). One possibility to explain this difference would be that GABAB receptors on a subset of presynaptic terminals, which release both GABA and glycine simultaneously, act as autoreceptors to regulate their own release, as shown in auditory brainstem neurons [29]. In fact, an electrophysiological study has demonstrated that a small portion (20%) of mIPSCs recorded from SG neurons are mixed mIPSCs mediated by corelease of both GABA and glycine [26], although most of glycinergic neurons in the spinal dorsal horn area also contain GABA [34,44]. Therefore, it is highly feasible that presynaptic GABAB receptors might act as autoreceptors to regulate both GABA and glycine release from a subset of inhibitory nerve terminals. 4.2. Mechanisms underlying GABAB receptor-mediated presynaptic inhibition of glycine release onto SG neurons
Fig. 6. Effect of Ba2+ on baclofen-induced decrease in glycinergic IPSCs. (A) A typical time course of the IPSC1 amplitude before, during and after application of 30 lM baclofen in the absence or presence of 1 mM Ba2+, a GIRK channel inhibitor. Insets represent typical traces of the numbered region. (B) Baclofen-induced changes in the IPSC1 amplitude (a) and paired-pulse ratio (b). Each column was normalized to the control and represents the mean and SEM from five experiments. **p < 0.01, n.s, not significant.
be negligible because baclofen also decreased IPSC amplitude even with the pipette solution containing GDP-bS. On the other hand, CGP55845 had no effect on either the IPSC1 amplitude or the PPR during the paired stimulation. The results indicate that there might be little tonic activation of GABAB receptors on most of glycinergic nerve terminals innervating to SG neurons. However, CGP55845 did increase the later but not initial IPSCs during the repetitive stimulation in a subset of SG neurons, suggesting that an accumulation of synaptically released GABA during the intense neural activity should be needed to activate presynaptic GABAB receptors and thus inhibit glycine release from a subpopulation of glycinergic nerve terminals. However, the facilitatory effect of CGP55845 on the later IPSCs recorded during the train stimulation was only observed
The mechanisms underlying the presynaptic inhibition of action potential-dependent neurotransmitter release mediated by G-protein coupled receptors including GABAB receptors are diverse [47]. For example, the activation of presynaptic GABAB receptors decrease neurotransmitter release at various central synapses by inhibiting multiple types of presynaptic VDCCs, mainly N- and P/Q-type Ca2+ channels [13,18,28], although presynaptic GABAB receptors selectively inhibit P/Qtype Ca2+ channels in suprachiasmatic nucleus neurons [10]. The present results also suggest that GABAB receptors expressed on glycinergic nerve terminals projecting to SG neurons might inhibit both N- and P/Q-type Ca2+ channels because baclofen still decreased glycinergic IPSCs in the presence of either x-AgTx (a P/Q-type Ca2+ channel blocker) or x-CgTx (an N-type Ca2+ channel blocker), and the baclofen-induced decrease in glycinergic IPSCs was largely occluded in the presence of both toxins (data not shown). However, the involvement of other VDCC subtypes, such as R-type, in the GABAB receptor-mediated presynaptic inhibition of glycine release cannot be ruled out. On the other hand, GABAB receptors are negatively coupled to AC to decrease the cAMP production [23], or open GIRK channels to hyperpolarize presynaptic nerve terminals [42]. In the present study, however, the contribution of either AC or GIRK channels to the baclofen-induced inhibition of glycinergic IPSCs would be negligible because the blockade of AC or GIRK channels by SQ22536 or Ba2+ did not affect the inhibitory action of baclofen on glycinergic IPSCs. The activation of presynaptic GABAB receptors also inhibits spontaneous miniature neurotransmitter release, which is independent on the Ca2+ influx from
I.-S. Choi et al. / Pain 138 (2008) 330–342
339
Fig. 7. Effect of baclofen on glycinergic mIPSCs. (Aa) A typical trace of glycinergic mIPSCs recorded before, during and after the application of 30 lM baclofen. Insets represent typical traces with an expanded time scale. (b) An all-point scatter plot of mIPSC amplitude as shown in (Aa). 752 events were plotted. (B) Cumulative probability distributions for the inter-event interval (a; p < 0.01, K–S test) and current amplitude (b; p = 0.63, K–S test) of glycinergic mIPSCs shown in (Aa). 203 for the control and 139 events for baclofen were plotted. Insets, 30 lM baclofen-induced changes in mIPSC frequency (left) and amplitude (right). Each column was the mean and SEM from 10 experiments. **p < 0.01. (C) 30 lM baclofen-induced changes of mIPSC frequency in the control condition and in the presence of 3 lM CGP55845 (n = 5), 100 lM SQ22536 (n = 6), 10 lM forskolin (n = 7), 1 mM Ba2+ (n = 6) and 200 lM Cd2+ (n = 7).
extracellular space to presynaptic nerve terminals [8]. In the present study, the baclofen significantly decreased glycinergic mIPSC frequency without affecting the current amplitude, suggesting the presynaptic locus of baclofen action. Since the baclofen-induced decrease in glycinergic mIPSC frequency was not affected by SQ22536, forskolin, Cd2+, or Ba2+, GABAB receptors expressed on glycinergic nerve terminals projecting to SG neurons might inhibit Ca2+-independent glycine release by modulating some presynaptic kinases, such as PKC and/or PKA, as demonstrated by previous reports [21,27]. However, further studies are needed to reveal the exact mechanisms underlying GABAB receptor-mediated inhibition of glycinergic mIPSCs in SG neurons. 4.3. Physiological implications Activity-dependent synaptic depression is a major form of short-term plasticity of intrinsic synaptic transmission. This synaptic depression has been suggested to equalize the synaptic strength of neocortical synapses having different initial release probabilities [1,32]. A
potential mechanism underlying synaptic depression may include the depletion of readily releasable vesicles at the synapses with high release probability [7,14,28], although other mechanisms, such as the activation of presynaptic autoreceptors [4,46], and the desensitization of postsynaptic receptors [36], have also been suggested. Therefore, presynaptic GABAB receptors may affect synaptic depression by regulating the release probability. For example, the activation of presynaptic GABAB receptors is known to block frequency-dependent synaptic depression by reducing the exhaustion of the readily releasable pool at GABAergic synapses of hippocampal CA3 interneurons [28], although synaptic depression is likely to be largely preserved during the GABAB receptor activation at basket cell-granule cell GABAergic synapses of the hippocampal dentate gyrus [17]. At glutamatergic synapses of the nucleus magnocellularis, the activation of presynaptic GABAB receptors shifts from the synaptic depression even to facilitation, resulting in the enhancement of synaptic efficacy [7]. In the present study, the strong activation of presynaptic GABAB receptors was found to completely block
340
I.-S. Choi et al. / Pain 138 (2008) 330–342
Fig. 8. Baclofen-induced reduction in synaptic depression of glycinergic IPSCs during train stimulation. (A) Typical time courses of the first IPSC (IPSC1) and 10th IPSC (IPSC10) amplitudes (a) and the IPSC10/IPSC1 ratio (b) before, during and after the application of 3 lM or 30 lM baclofen. Insets represent typical traces of the numbered region. (B) Baclofen (a; 30 lM; b, 3 lM)-induced changes in the mean ratio of IPSCn/IPSC1 plotted against the number within the train (n). Each point represents the mean and SEM from eight experiments. All points were normalized to the IPSC1 amplitude of the control condition, but diamond symbols were normalized to IPSC1 amplitude of the baclofen condition. (C) Baclofen-induced changes in the IPSC amplitude within the train. Each point represents the mean and SEM from eight experiments. *p < 0.05, **p < 0.01. Note that the action of 3 lM baclofen shifts from the inhibition to potentiation within the train.
synaptic depression of glycinergic transmission onto SG neurons. The decrease in initial IPSC amplitude during the strong activation of presynaptic GABAB receptors would further contribute to the blockade of synaptic depression by decreasing the desensitization of postsynaptic glycine receptors. The resultant blockade of glycinergic synaptic depression during the strong activation of presynaptic GABAB receptors would sustain the minimal glycinergic inhibition during the high neuronal activity, as baclofen did not decrease the absolute amplitudes of later IPSCs during the train. On the contrary, the weak activation of presynaptic GABAB receptors largely preserved synaptic depression of glycinergic transmission onto SG neurons, suggesting that the extent of changes in initial release probability indeed affects the extent of synaptic depression. However, it should be noted that the weak activation of presynaptic
GABAB receptors increased the absolute amplitudes of later glycinergic IPSCs during the train. The results suggest that a small decrease in initial release probability could increase inhibitory synaptic efficacy during the high neuronal activity. A potential mechanism underlying the baclofeninduced antinociception involves presynaptic inhibition of glutamate release from primary afferent terminals [3,19], although baclofen also induces a membrane hyperpolarization by activating K+ channels in spinal dorsal horn neurons [25]. However, the activation of presynaptic GABAB receptors also decreases both GABAergic [19] and glycinergic synaptic transmission onto SG neurons (present results). This GABAB receptor-mediated reduction of inhibitory synaptic transmission would complicate the baclofen-induced antinociceptive action, because baclofen has a similar affinity to GABA (IC50 = 4.3 lM),
I.-S. Choi et al. / Pain 138 (2008) 330–342
glutamate (IC50 = 4.4 lM) [19], and glycine release (IC50 = 6.0 lM). In the present study, it has been shown that baclofen could affect short-term synaptic depression of glycinergic transmission during the high neuronal activity in a concentration-dependent manner. Baclofen at higher concentrations would warrant the minimal glycinergic inhibition, and at lower concentration it would increase the efficacy of glycinergic inhibition during the high neuronal activity. Such effects of baclofen on short-term plasticity of glycinergic transmission would further contribute to the baclofen-induced antinociceptive action. However, more studies such as the baclofen actions on GABAergic and glutamatergic short-term plasticity are needed to understand the baclofen-induced antinociceptive action.
Acknowledgements We thank Prof. Patria Morris (Kyungpook National University) for correcting the English. This work was supported by the Korea Science and Engineering Foundation (KOSEF) Grant (No. R01-2006-000-10406-0). I.-S. Choi was supported by Brain Korea 21 Project.
References [1] Abbott LF, Varela JA, Sen K, Nelson SB. Synaptic depression and cortical gain control. Science 1997;275:220–4. [2] Ahmadi S, Lippross S, Neuhuber WL, Zeilhofer HU. PGE2 selectively blocks inhibitory glycinergic neurotransmission onto rat superficial dorsal horn neurons. Nat Neurosci 2002;5:34–40. [3] Ataka T, Kumamoto E, Shimoji K, Yoshimura M. Baclofen inhibits more effectively C-afferent than Ad-afferent glutamatergic transmission in substantia gelatinosa neurons of adult rat spinal cord slices. Pain 2000;86:273–82. [4] Barnes-Davies M, Forsythe ID. Pre- and postsynaptic glutamate receptors at a giant excitatory synapse in rat auditory brainstem slices. J Physiol (Lond) 1995;488:387–406. [5] Beaumont V, Zucker RS. Enhancement of synaptic transmission by cyclic AMP modulation of presynaptic Ih channels. Nat Neurosci 2000;3:133–41. [6] Bischoff S, Leonhard S, Reymann N, Schuler V, Shigemoto R, Kaupmann K, et al. Spatial distribution of GABAB R1 receptor mRNA and binding sites in the rat brain. J Comp Neurol 1999;412:1–16. [7] Brenowitz S, David J, Trussell L. Enhancement of synaptic efficacy by presynaptic GABAB receptors. Neuron 1998;20:135–41. [8] Capogna M, Gahwiler BH, Thompson SM. Presynaptic inhibition of calcium-dependent and -independent release elicited with ionomycin, gadolinium, and a-latrotoxin in the hippocampus. J Neurophysiol 1996;75:2017–28. [9] Castro AR, Bowery N, Castro-Lopes JM. Baclofen and midazolam alter c-fos induction by peripheral noxious or innocuous stimulation in the spinal cord of normal and monoarthritic rats. Neuropharmacology 1999;38:1775–88. [10] Chen G, van den Pol AN. Presynaptic GABAB autoreceptor modulation of P/Q-type calcium channels and GABA release in rat suprachiasmatic nucleus neurons. J Neurosci 1998;18:1913–22.
341
[11] Chery N, De Koninck Y. Junctional versus extrajunctional glycine and GABAA receptor-mediated IPSCs in identified lamina I neurons of the adult rat spinal cord. J Neurosci 1999;17:7342–55. [12] Dirig DM, Yaksh TL. Intrathecal baclofen and muscimol, but not midazolam, are antinociceptive using the rat-formalin model. J Pharmacol Exp Ther 1995;275:219–27. [13] Doze VA, Cohen GA, Madison DV. Calcium channel involvement in GABAB receptor-mediated inhibition of GABA release in area CA1 of the rat hippocampus. J Neurophysiol 1995;74:43–53. [14] Elmquist D, Quastel D. A quantitative study of end-plate potentials in isolated human muscle. J Physiol (Lond) 1965;178:505–29. [15] Furue H, Katafuchi T, Yoshimura M. Sensory processing and functional reorganization of sensory transmission under pathological conditions in the spinal dorsal horn. Neurosci Res 2004;48:361–8. [16] Hammond DL. GABAB receptors: new tricks by an old dog. Curr Opin Pharmacol 2001;1:26–30. [17] Hefft S, Kraushaar U, Geiger JR, Jonas P. Presynaptic short-term depression is maintained during regulation of transmitter release at a GABAergic synapse in rat hippocampus. J Physiol (Lond) 2002;539:201–8. [18] Huston E, Cullen GP, Burley JR, Dolphin AC. The involvement of multiple calcium channel sub-types in glutamate release from cerebellar granule cells and its modulation by GABAB receptor activation. Neuroscience 1995;68:465–78. [19] Iyadomi M, Iyadomi I, Kumamoto E, Tomokuni K, Yoshimura M. Presynaptic inhibition by baclofen of miniature EPSCs and IPSCs in substantia gelatinosa neurons of the adult rat spinal dorsal horn. Pain 2000;85:385–93. [20] Jang IS, Jeong HJ, Katsurabayashi S, Akaike N. Functional roles of presynaptic GABAA receptors on glycinergic nerve terminals in the rat spinal cord. J Physiol (Lond) 2002;541:423–34. [21] Jarolimek W, Misgeld U. GABAB receptor-mediated inhibition of tetrodotoxin-resistant GABA release in rodent hippocampal CA1 pyramidal cells. J Neurosci 1997;17:1025–32. [22] Jonas P, Bischofberger J, Sandkuhler J. Corelease of two fast neurotransmitters at a central synapse. Science 1998;281: 419–24. [23] Kamatchi GL, Ticku MK. Functional coupling of presynaptic GABAB receptors with voltage-gated Ca2+ channel: regulation by protein kinases A and C in cultured spinal cord neurons. Mol Pharmacol 1990;38:342–7. [24] Kaneko M, Takahashi T. Presynaptic mechanism underlying cAMP-dependent synaptic potentiation. J Neurosci 2004;24:5202–8. [25] Kangrga I, Jiang MC, Randic M. Actions of ()-baclofen on rat dorsal horn neurons. Brain Res 1991;562:265–75. [26] Keller AF, Coull JA, Chery N, Poisbeau P, De Koninck Y. Region-specific developmental specialization of GABA-glycine cosynapses in laminas I–II of the rat spinal dorsal horn. J Neurosci 2001;21:7871–80. [27] Kubota H, Katsurabayashi S, Moorhouse AJ, Murakami N, Koga H, Akaike N. GABAB receptor transduction mechanisms, and cross-talk between protein kinases A and C, in GABAergic terminals synapsing onto neurons of the rat nucleus basalis of Meynert. J Physiol (Lond) 2003;551:263–76. [28] Lei S, McBain CJ. GABAB receptor modulation of excitatory and inhibitory synaptic transmission onto rat CA3 hippocampal interneurons. J Physiol (Lond) 2003;546:439–53. [29] Lim R, Alvarez FJ, Walmsley B. GABA mediates presynaptic inhibition at glycinergic synapses in a rat auditory brainstem nucleus. J Physiol (Lond) 2000;525:447–59. [30] Loomis CW, Khandwala H, Osmond G, Hefferan MP. Coadministration of intrathecal strychnine and bicuculline effects synergistic allodynia in the rat: an isobolographic analysis. J Pharmacol Exp Ther 2001;296:756–61.
342
I.-S. Choi et al. / Pain 138 (2008) 330–342
[31] Margeta-Mitrovic M, Mitrovic I, Riley RC, Jan LY, Basbaum AI. Immunohistochemical localization of GABAB receptors in the rat central nervous system. J Comp Neurol 1999;405:299–321. [32] Markram H, Tsodyks M. Redistribution of synaptic efficacy between neocortical pyramidal neurons. Nature 1996;382:807–10. [33] McCarson KE, Enna SJ. Nociceptive regulation of GABAB receptor gene expression in rat spinal cord. Neuropharmacology 1999;38:1767–73. [34] Mitchell K, Spike RC, Todd AJ. An immunocytochemical study of glycine receptor and GABA in laminae I–III of rat spinal dorsal horn. J Neurosci 1993;13:2371–81. [35] Mott DD, Lewis DV. The pharmacology and function of central GABAB receptors. Int Rev Neurobiol 1994;36:97–223. [36] Otis T, Zhang S, Trussell LO. Direct measurement of AMPA receptor desensitization induced by glutamatergic synaptic transmission. J Neurosci 1996;16:7496–504. [37] Salin PA, Malenka RC, Nicoll RA. Cyclic AMP mediates a presynaptic form of LTP at cerebellar parallel fiber synapses. Neuron 1996;16:797–803. [38] Simpson Jr RK, Gondo M, Robertson CS, Goodman JC. Reduction in the mechanonociceptive response by intrathecal administration of glycine and related compounds. Neurochem Res 1996;21:1221–6. [39] Simpson Jr RK, Gondo M, Robertson CS, Goodman JC. Reduction in thermal hyperalgesia by intrathecal administration of glycine and related compounds. Neurochem Res 1997;22:75–9.
[40] Takigawa T, Alzheimer C. Phasic and tonic attenuation of EPSPs by inward rectifier K+ channels in rat hippocampal pyramidal cells. J Physiol (Lond) 2002;539:67–75. [41] Thomas DA, Navarrete IM, Graham BA, McGowan MK, Hammond DL. Antinociception produced by systemic R(+)baclofen hydrochloride is attenuated by CGP 35348 administered to the spinal cord or ventromedial medulla of rats. Brain Res 1996;718:129–37. [42] Thompson SM, Gahwiler BH. Comparison of the actions of baclofen at pre- and postsynaptic receptors in the rat hippocampus in vitro. J Physiol (Lond) 1992;451:329–45. [43] Todd AJ, Spike RC. The localization of classical transmitters and neuropeptides within neurons in laminae I–III of the mammalian spinal dorsal horn. Prog Neurobiol 1993;41:609–45. [44] Todd AJ, Sullivan AC. Light microscope study of the coexistence of GABA-like and glycine-like immunoreactivities in the spinal cord of the rat. J Comp Neurol 1990;296:496–505. [45] Towers S, Princivalle A, Billinton A, Edmunds M, Bettler B, Urban L, et al. GABAB receptor protein and mRNA distribution in rat spinal cord and dorsal root ganglia. Eur J Neurosci 2000;12:3201–10. [46] von Gersdorff H, Schneggenburger R, Weis S, Neher E. Presynaptic depression at a calyx synapse: the small contribution of metabotropic glutamate receptors. J Neurosci 1997;17:8137–46. [47] Wu LG, Saggau P. Presynaptic inhibition of elicited neurotransmitter release. Trends Neurosci 1997;20:204–12.