Cellular, molecular and developmental neuroscience 313
Accumulated GABA activates presynaptic GABAB receptors
and inhibits both excitatory and inhibitory synaptic
transmission in rat midbrain periaqueductal gray
Guangying Lia,e,*, Caifeng Shaob,d,*, Qian Chenb, Qiang Wangc and
Kun Yangb,f
γ-Aminobutyric acid (GABA), the major inhibitory
neurotransmitter, activates ionotropic GABAA receptors and
metabotropic GABAB receptors in the central nervous
system, respectively. In ventrolateral division of the midbrain
periaqueductal gray (PAG), GABAB receptors play important
roles in pain modulation. However, the role of endogenous
GABA action in presynaptic GABAB receptors remains
elusive. Using whole-cell recordings on acute PAG slices
from adult rats, here, we show that ambient GABA exerts a
tonic inhibition on presynaptic terminals by binding GABAB
receptors. Extracellular GABA accumulated by nipecotic
acid, which blocks GABA transporters, strengthened GABAB
receptor-mediated presynaptic inhibition on both excitatory
and inhibitory synapses. Our results indicate that PAG
neurons experience GABAB receptor-mediated inhibition
determined by GABA transporters. The accumulated GABA-
mediated actions may indicate a therapeutic way.
Introduction
γ-Aminobutyric acid (GABA) is the principal inhibitory
neurotransmitter in the central nervous system (CNS) [1].
Through the activation of ionotropic GABAA receptors
and metabotropic GABAB receptors, GABA exerts ‘fast’
and ‘slow’ inhibitions, respectively [2]. At the subcellular
level, in contrast to the intrasynaptic distribution of
GABAA receptors, GABAB receptors are primarily dis-
tributed at extrasynaptic structures [3,4], making GABAB
receptors accessible to ambient GABA. Therefore, pre-
synaptic GABAB receptors mediate tonic inhibition by
extracellular GABA in many CNS structures [5–8],
although this is not the case in all brain regions [9].
The periaqueductal gray matter (PAG) in the midbrain is
an important area in integrating autonomic, somatomotor,
and behavioral responses to stress, threat, and pain [10].
The neurons in ventrolateral periaqueductal gray matter
(vlPAG) are particularly involved in the descending pain
modulation system [10]. Following the discovery of
cannabinoid type 1 receptors and μ-opioid receptors in
modulating this descending circuit [10,11], GABAB
receptors have been suggested to also participate in pain
modulation [10,12]. It is most likely that GABAB receptor
activation in vlPAG affects the neuronal excitability
through a combination of presynaptic and postsynaptic
mechanisms [13–17].
0959-4965 Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.
Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.
NeuroReport 28:313–318 Copyright © 2017 Wolters Kluwer
Health, Inc. All rights reserved.
NeuroReport 2017, 28:313–318
Keywords: CGP52432, GABAB receptors, nipecotic acid, periaqueductal gray,
whole-cell patch-clamp recordings
a
Department of Psychology, the Affiliated Hospital to Changchun University of
Chinese Medicine, Changchun, Jilin, Departments of bAnatomy, cPreventive
Medicine and Public Health Laboratory Science, School of Medicine, Jiangsu
University, dDepartment of Anesthesiology, Affiliated People’s Hospital of Jiangsu
University, Zhenjiang, Jiangsu, China, eDepartment of Psychiatry and fDepartment
of Physiology, Saga Medical School, Saga, Japan
Correspondence to Kun Yang, MD, PhD, Department of Anatomy, School of
Medicine, Jiangsu University, 301 Xuefu Road, Zhenjiang, Jiangsu 212013, China
Tel: + 86 511 8595 9086; fax: + 86 511 8503 8483; e-mail: yangk@ujs.edu.cn
*Guangying Li and Caifeng Shao contributed equally to the writing of this article.
Received 20 December 2016 accepted 31 January 2017
GABAB receptors are activated by endogenous GABA [8]
and in response to pharmacological activation by exo-
genous GABAB receptor agonists. One determinant of
the GABAB receptor function is the concentration of
extrasynaptic (ambient) GABA [18]. In the vlPAG, it is
not clear whether elevated GABA affects GABAB
receptor function in its modulation on synaptic trans-
mission. In the present study, we addressed this issue by
performing whole-cell recordings in acute PAG slices
from adult rats to test the hypotheses that blockade of
GABA uptake modulates neurotransmission by pre-
synaptic GABAB receptors.
Materials and methods
The Animal Care and Use Committees of the
Changchun University of Chinese Medicine and Jiangsu
University School of Medicine approved the experi-
mental procedures. The methods for PAG slice pre-
paration have been described elsewhere [15–17]. Briefly,
male Sprague-Dawley rats (6–8 weeks old) were deeply
anesthetized by isoflurane and decapitated. The whole
brain was quickly removed into ice-cold oxygenated
(95% O2/5% CO2) artificial cerebrospinal fluid (aCSF)
containing (in mM) NaCl 124, KCl 3, CaCl2 2.5,
NaH2PO4 1.2, MgCl2 1.2, NaHCO3 25, and glucose 11. A
brain block containing the PAG was mounted on the
DOI: 10.1097/WNR.0000000000000756
314 NeuroReport 2017, Vol 28 No 6
cutting stage and transverse slices (400 μm in thickness)
were excised by a vibratome (7000smz-2; Campden
Instruments Ltd, Loughborough, Leics., UK). The slices
were quickly transferred to warm oxygenated aCSF
(∼35°C) in an incubation chamber for 40 min recovery
before recordings.
A single slice was placed in a custom-made recording
chamber (∼1 ml) under a differential interference con-
trast (DIC) microscope (Olympus BX51WI; Olympus,
Tokyo, Japan) and perfused continuously with oxyge-
nated aCSF (5–6 ml/min). Conventional whole-cell
recordings were performed at room temperature
(22–24°C). To record inhibitory postsynaptic currents
(IPSCs), the internal solution was (in mM) Cs-
methanesulfonate 85, CsCl 50, Hepes 10, MgCl2 2.5,
Na2-ATP 4, Na3-GTP 0.4, Na-phosphocreatine 10, and
EGTA 0.6 (pH 7.2 modulated by CsOH, mOsm ∼ 290).
For recording excitatory postsynaptic currents (EPSCs)
and postsynaptic potentials (PSPs), the glass pipettes
were filled with a potassium-based internal solution
containing (in mM) K-gluconate 135, KCl 5, CaCl2 0.5,
MgCl2 2, EGTA 5, Hepes 5, and Mg-ATP 3.6 (pH 7.2
modulated by KOH, mOsm ∼ 290). Guanosine 5′-[β-thio]
diphosphate trilithium salt (GDP-β-S; 1 mM) was inclu-
ded routinely in the internal solutions for intracellular
dialysis to inactivate putative G-protein-coupled recep-
tors including GABAB receptors [15,19]. The recording
pipette had a resistance of 4–6 MΩ when back filled with
internal solution. The holding potential was − 70 mV for
both IPSCs and EPSCs, which were pharmacologically
isolated by antagonists [15]. IPSCs were pharmacologi-
cally isolated by 6,7-dinitroquinoxaline-2,3(1H,4H)-dione
(DNQX; 10 μM) and D-2-amino-5-phosphonopentanoic
acid (100 μM) to block non-N-methyl-D-aspartate and
N-methyl-D-aspartate glutamate receptors, respectively.
EPSCs were recorded in the presence of picrotoxin (100 μM),
which abolished GABAA receptor-mediated GABAergic
components. Under the present recording conditions, a
high Cl− concentration in the recording pipettes yielded
GABAergic IPSCs as inward events [17]. It was reported
in some electrophysiological studies that simultaneous
IPSCs and EPSCs recordings could be obtained at a
holding potential of − 40 mV with a K+-based internal
solution. However, both EPSCs and IPSCs show only
relative ‘small’ amplitudes because of the respective
equilibrium potentials to − 40 mV. Also, some whole-cell
patch-clamp experiments used a cesium methane-
sulfonate internal solution and maintained holding
potential at 0 mV to obtain IPSCs [15]. In the present
study, we used a ‘high Cl−’ (50 mM Cl−) internal
solution and obtained ‘large amplitude’ IPSCs at a
holding potential of − 70 mV with a blocker cocktail. The
relatively large amplitudes responses facilitated IPSCs
analysis as we described before [8]. The miniature inhi-
bitory postsynaptic currents (mIPSCs) and miniature
excitatory postsynaptic currents (mEPSCs) were
Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.
recorded in the aCSF with 0.5 μM tetrodotoxin. The
evoked inhibitory postsynaptic currents (eIPSCs) and
evoked excitatory postsynaptic currents (eEPSCs) were
elicited by focal stimulation (0.1 ms square waveform)
using a bipolar stimulating electrode [17]. PSPs
were recorded under the current-clamp mode
with 0 pA current injection. Signals were amplified
by Axopatch 200B with Clampex 9 (Molecular
Devices, Sunnyvale, California, USA) run on a PC
(Windows 7) as we described previously [8]. Analysis of a
single neuron’s mIPSCs and mEPSCs frequency and
amplitude distributions were carried out using the
Kolmogorov–Smirnov test as we described elsewhere
[15,19]. The PSPs were analyzed by comparing the area
under the curve, which was calculated during the length
of 500 ms long epoch of the evoked response from the
stimulation artifacts. The results were expressed as
mean ± SEM. Statistics were calculated with GraphPad
Prism 6 (GraphPad Software, La Jolla, California, USA)
and Excel 2013 (Microsoft Co., Redmond, Washington,
USA) using Student’s t-test and two-way analysis of
variance. Significant differences were set at P value less
than 0.05.
Results
To analyze ambient GABA action on presynaptic trans-
mission, postsynaptic GABAB receptors were nulled by
intracellular GDP-β-S dialysis, leaving only presynaptic
GABAB receptors available for GABA or pharmacological
agents [15,19]. The eIPSCs were induced by low-
frequency (0.033 Hz) stimulation. The additional
perfusion of the selective GABAA receptor antagonist
picrotoxin (100 μM) almost abolished the eIPSCs, con-
firming that all eIPSCs were the result of GABAergic
transmission [15]. Bath application of nipecotic acid
(NPA; 10 mM), a transportable blocker for all types of
γ-aminobutyric acid transporters (GATs) [20], decreased
the peak amplitude of eIPSCs to 30 ± 9% of the control at
room temperature (305 ± 68 pA in control vs. 77 ± 19 pA in
NPA; n = 10, P = 0.003, paired t-test). The half-decay
time of eIPSCs was increased to 271 ± 51% of the control
(77 ± 9 ms in control vs. 194 ± 28 ms in NPA; n = 10,
P < 0.01, paired t-test) (Fig. 1a). Next, we investigated
the role of GABAB receptors in presynaptic GABAergic
terminals. The amplitude of eIPSCs was facilitated by
perfusing CGP52432 (1 μM), a selective GABAB receptor
antagonist, indicating a tonic inhibition of GABA release
by presynaptic GABAB receptors. In the presence of
CGP52432, the NPA-induced depression on eIPSCs was
11 ± 4% (443 ± 53 pA in the presence of CGP52432 vs.
381 ± 40 pA in the presence of CGP52432 + NPA; n = 9,
P < 0.05, paired t-test), confirming that accumulated
GABA was acting on GABAB receptors (Fig. 1b).
However, in the presence of CGP52432, the extent of
inhibition was lower than that without CGP52432 pre-
treatment (unpaired t-test; Fig. 1c). The data suggest that
 Accumulated GABA modulates PAG neurons Li et al. 315

Fig. 1

(a) (c)
150

eIPSCs amplitude (% baseline)

**

100
NPA

50
Ctrl.

(b) CGP+ (d) *

Control 6
NPA

Freq. (Hz)
0
NPA

30 n.s.

Ampl. (pA)
Washout
Ctrl.
CGP 0

Effects of nipecotic acid (NPA) on inhibitory synaptic transmission. (a) A representative neuron shows that NPA (10 mM) decreased the amplitudes of
evoked inhibitory postsynaptic currents (eIPSCs) and increased their half-decay times. Scale bar: 100 pA, 100 ms. (b) The amplitude of eIPSCs was
increased by CGP52432, which blocked GABAB receptors. In the presence of CGP52432, NPA (10 mM) still depressed the amplitude. Scale bar:
80 pA, 100 ms. (c) Pooled data show NPA effects on peak amplitudes of eIPSCs in artificial cerebrospinal fluid (NPA; n = 10) and in the presence of
CGP52432 (CGP + NPA; n = 9) to different extents (unpaired t-test). The broken line indicates the baseline (100%). (d) Left: example traces of
miniature inhibitory postsynaptic currents (mIPSCs) before (upper), during (middle), and after (bottom) NPA (10 mM) bath application. Right: pooled
data show that NPA decreased the relative frequency (upper), but not the relative amplitude (bottom) of mIPSCs (n = 6). Scale bar: 20 pA, 1 s. Data
are shown as mean ± SEM. *P < 0.05; **P < 0.01. Ampl., amplitude; Ctrl., control; Freq., frequency.

accumulated GABA inhibits GABA release from pre- in NPA; n = 8, P < 0.01, paired t-test) and the half-decay
synaptic terminals by activating GABAB receptors. time was increased to 139 ± 12% of the control (48 ± 7 ms
in control vs. 69 ± 11 ms in NPA; n = 8, P < 0.05, paired
In the presence of tetrodotoxin, which blocked action t-test) (Fig. 2a). The effect of NPA on eEPSCs amplitude
potentials, the GABAergic mIPSCs occurred at a mean was largely prevented by pretreatment with CGP52432
frequency of 2.2 ± 0.7 Hz with a mean amplitude of (to 68 ± 9% of that with CGP52432), suggesting an
16 ± 3 pA at − 70 mV holding potential (n = 6). As shown accumulation of GABA action on presynaptic GABAB
in Fig. 1d, application of NPA (10 mM) caused a sig- receptors (Fig. 2b and c). NPA also decreased the fre-
nificant reduction in the mean frequency of spontaneous quency, but not the average amplitude of mEPSCs
mIPSCs to 1.3 ± 0.4 Hz (P = 0.002) without a significant (11.9 ± 1.7 pA and 3.8 ± 0.6 Hz in control vs. 9.7 ± 1.5 pA
change in the mean amplitude (15 ± 2 pA; P = 0.20). and 2.0 ± 0.7 Hz in NPA; n = 8, P = 0.30 and P < 0.05,
Taken together, the inhibition of mIPSCs frequency by respectively; paired t-test) (Fig. 2d). Altogether, the
NPA application appears to be a presynaptic GABA results indicate that presynaptic GABAB receptors acti-
action. vated by accumulated GABA are involved in regulating
glutamate release from presynaptic terminals.
GABAB receptors are expressed on glutamatergic term-
inals in vlPAG as well [15,21]. We therefore investigated The coordinated excitatory and inhibitory inputs shape
the effects of NPA on glutamatergic synapses. The neuronal excitability and network activity. The PAG
amplitude of eEPSCs was largely attenuated by NPA neurons receive both excitatory and inhibitory synapses
(10 mM) to 25 ± 5% (198 ± 34 pA in control vs. 46 ± 10 pA [15]; it is conceivable that the NPA action on both

Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.

316 NeuroReport 2017, Vol 28 No 6

Fig. 2

(a) (c)
1.5

eEPSCs amplitude (% baseline)

**

NPA 1.0

Ctrl.
0.5

0.0

(b) (d)
8 *
CGP+ Control

Freq. (Hz)
NPA
4

0
NPA

Ctrl.
n.s.
CGP 20

Ampl. (pA)
Washout
10

Effects of nipecotic acid (NPA) on excitatory synaptic transmission. (a) A representative neuron shows that NPA (10 mM) decreased the amplitudes of
evoked excitatory postsynaptic currents (eEPSCs) and increased their half-decay times. Scale bar: 100 pA, 80 ms. (b) The amplitude of eEPSCs was
increased by CGP52432, which blocked GABAB receptors. In the presence of CGP52432, NPA (10 mM) still depressed the amplitude. Scale bar:
60 pA, 80 ms. (c) Pooled data show NPA effects on eEPSC peak amplitudes in artificial cerebrospinal fluid (n = 8) and in the presence of CGP52432
(n = 7) to different extents (unpaired t-test). The broken line indicates the baseline (100%). (d) Left: example traces of miniature excitatory
postsynaptic currents (mEPSCs) before (upper), during (middle), and after (bottom) NPA (10 mM) bath application. Right: compiled data show that
NPA decreased the relative frequency (upper), but not the relative amplitude (bottom) of mEPSCs (n = 8). Scale bar: 20 pA, 1 s. Data are shown as
mean ± SEM. *P < 0.05; **P < 0.01. Ampl., amplitude; Ctrl., control; Freq., frequency.

excitatory and inhibitory synapses affects functional Discussion

repercussions of neuronal output [22]. To address this The main findings of accumulated GABA in the vlPAG
issue, we used current-clamp mode to record PSPs with a neurons are as follows: (a) ambient GABA tonically
K+-based internal solution (see the Materials and methods inhibits GABAergic and glutamatergic synapses; (b) the
section) in the aCSF without antagonists for glutamate or accumulation of GABA by NPA (a GATs blocker) exerts
GABA [22,23]. NPA (10 mM) application depressed the inhibition on both inhibitory and excitatory synapses; and
area under the curve to 32 ± 4% of the control (c) the accumulated GABA activates presynaptic GABAB
(3.4 ± 0.2 ms × mV in control vs. 1.1 ± 0.1 ms × mV in NPA; receptors.
n = 8, P < 0.01, paired t-test) (Fig. 3a). In the presence of GABAB receptors are distributed throughout the CNS
CGP52432 (1 µM), the area was 53 ± 7% of that with fur- [24], including systems involved in the transmission and
ther NPA (10 mM) treatment (3.4 ± 0.2 ms × mV in modulation of pain [4,21,25]. In the PAG, GABAB
CGP52432 vs. 1.1 ± 0.1 ms × mV in CGP52432 + NPA; receptors are expressed in both presynaptic and post-
n = 6, P < 0.05, paired t-test) (Fig. 3a and b). The data synaptic compartments, the activation of which yields
suggest that the ‘net effect’ of blocking GATs is inhibition both presynaptic and postsynaptic inhibition [16,17,19].
on neuronal postsynaptic responses and presynaptic Because GABAB receptors are primarily found in extra-
GABAB receptors are involved in this process (Fig. 3b synaptic structures, ambient GABA could tonically con-
and c). strain the synapses. In the present study, we show

Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.

 Accumulated GABA modulates PAG neurons Li et al. 317

Fig. 3

(b) * **
6
(a)

under curve area (ms·mV)

**

Control 4

NPA

(c)
CGP
80 **
Control
Relative area (%)

CGP+NPA 40

The ‘net effects’ of nipecotic acid (NPA) on postsynaptic potentials under different conditions. (a) NPA (10 mM) decreased the area under the curve
of evoked postsynaptic potentials in normal artificial cerebrospinal fluid (upper). Pretreatment of CGP52432 (10 µM) increased the area; in the
presence of CGP52432, NPA (10 mM) decreased the area (lower). Scale bar: 10 mV, 200 ms. (b) Pooled data show the change of area of
postsynaptic potentials under different conditions. (c) With CGP52432 pretreatment, NPA decreased area with a less extent. Data were shown as
mean ± SEM. *P < 0.05; **P < 0.01. CGP, CGP52432.

evidence that blocking GABAB receptors with the spe-
cific antagonist CGP52432 shows a tonic inhibition.
Furthermore, elevated GABA exerts inhibition by acti-
vating presynaptic GABAB receptors, thereby decreasing
the amplitude of eIPSCs and eEPSCs. The eIPSCs and
eEPSCs elongation can be attributed to slowed down
GABA clearance [20].
Our present data lead to a conclusion that the accumu-
lated GABA acts on presynaptic GABAB receptors. It is
worth noting that in electrophysiological data analysis,
the paired-pulse ratio provides a simple and reliable
strategy for analyzing presynaptic action. In the present
study, we did not use this strategy for two reasons. First,
paired-pulse ratio is the ratio between the amplitude of
the second response over the first one. The decay time of
IPSCs or EPSCs changed to NPA treatment (Figs 1a and
2a), altering the baseline for the second IPSCs and
EPSCs. Therefore, it was not accurate to measure the
amplitude of the second responses. Second, as classic
G-protein-coupled receptors, the postsynaptic GABAB
receptors were nulled by intracellular GDP-β-S (1 mM)
dialysis, leaving only presynaptic GABAB receptors
available for their ligands. Therefore, the results
observed are the actions of the presynaptic receptors as
we suggested in our previous study [15].
In the present study, both eIPSCs and eEPSCs were
inhibited by ambient GABA. This inhibition was
enforced by NPA, which blocks all types of GATs,
indicating that elevated GABA exerts a robust inhibition
on release [26]. Because blocking GABAB receptors by
CGP52432 impaired the NPA action, it is concluded that
NPA action involved the activation of GABAB receptors.
It is worth noting that in the present study, we only
investigated the effect of elevated GABA on presynaptic

Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.

318 NeuroReport 2017, Vol 28 No 6
GABAB receptors. It is possible that GABA also binds to
GABAA receptors. Further experiments are needed to
determine the response of GABAA receptors to accu-
mulated GABA. It should be noted that NPA action is
temperature sensitive [27]; further experiments should
be conducted to test NPA action on either IPSCs or
EPSCs at physiological temperature.
The vlPAG is involved in the descending pain modula-
tion system. Therefore, changes in vlPAG neuronal
excitability would determine the output and influence to
the spinal cord. As both inhibitory and excitatory inputs
upon vlPAG neuron are weakened, the ‘net effect’ on
this neuron depends on the sum of excitatory and inhi-
bitory inputs’ impairment [14,15]. In the present study,
the recorded PSPs evoked in the vlPAG reflect the
summation of postsynaptic responses from both excita-
tory and inhibitory inputs. The results suggest that the
‘net effect’ of NPA is a depression on PSPs. Our present
study provides additional evidence that elevated GABA
modulates PAG neurons synaptic transmission and
excitability. The action of accumulated GABA should be
taken into account when the vlPAG is considered to be a
pivotal site for modulating pain [28,29].
Acknowledgements
K.Y. designed research; G.L., C.S., Q.C. and K.Y. per-
formed research; G.L., C.S., Q.C., Q.W., and K.Y. ana-
lyzed data; K.Y. wrote the paper. All authors agreed on
the final version of the manuscript.
This work was supported in part by the National Natural
Science Foundation of China (no. 81400911), the Natural
Science Foundation of Jiangsu Province (no. BK20140573)
to Q.W., and the Jiangsu Distinguished Professor Program
to K.Y.
Conflicts of interest
There are no conflicts of interest.
References
1 Bowery NG, Smart TG. GABA and glycine as neurotransmitters: a brief
history. Br J Pharmacol 2006; 147 (Suppl 1):S109–S119.
2 Bowery NG, Bettler B, Froestl W, Gallagher JP, Marshall F, Raiteri M, et al.
International Union of Pharmacology. XXXIII. Mammalian γ-aminobutyric acid
(B) receptors: structure and function. Pharmacol Rev 2002; 54:247–264.
3 Goudet C, Magnaghi V, Landry M, Nagy F, Gereau RW IV, Pin JP.
Metabotropic receptors for glutamate and GABA in pain. Brain Res Rev
2009; 60:43–56.
4 Gassmann M, Bettler B. Regulation of neuronal GABAB receptor functions
by subunit composition. Nat Rev Neurosci 2012; 13:380–394.
5 Fearon IM, Zhang M, Vollmer C, Nurse CA. GABA mediates autoreceptor
feedback inhibition in the rat carotid body via presynaptic GABAB receptors
and TASK-1. J Physiol 2003; 553 (Pt 1):83–94.
6 Lei S, McBain CJ. GABAB receptor modulation of excitatory and inhibitory
synaptic transmission onto rat CA3 hippocampal interneurons. J Physiol
2003; 546 (Pt 2):439–453.
Copyright r 2017 Wolters Kluwer Health, Inc. All rights reserved.
7 Wu Y, Wang HY, Lin CC, Lu HC, Cheng SJ, Chen CC, et al. GABAB
receptor-mediated tonic inhibition of noradrenergic A7 neurons in the rat.
J Neurophysiol 2011; 105:2715–2728.
8 Yang K, Ma H. Blockade of GABAB receptors facilitates evoked
neurotransmitter release at spinal dorsal horn synapse. Neuroscience 2011;
193:411–420.
9 Guetg N, Seddik R, Vigot R, Turecek R, Gassmann M, Vogt KE, et al. The
GABAB1a isoform mediates heterosynaptic depression at hippocampal
mossy fiber synapses. J Neurosci 2009; 29:1414–1423.
10 Millan MJ. Descending control of pain. Prog Neurobiol 2002; 66:355–474.
11 Chen T, Li J, Feng B, Hui R, Dong YL, Huo FQ, et al. Mechanism underlying
the analgesic effect exerted by endomorphin-1 in the rat ventrolateral
periaqueductal gray. Mol Neurobiol 2016; 53:2036–2053.
12 Levy RA, Proudfit HK. Analgesia produced by microinjection of baclofen and
morphine at brain stem sites. Eur J Pharmacol 1979; 57:43–55.
13 Chieng B, Christie MJ. Hyperpolarization by GABAB receptor agonists in
mid-brain periaqueductal gray neurones in vitro. Br J Pharmacol 1995;
116:1583–1588.
14 Chiou LC, Huang LYM. Mechanism underlying increased neuronal activity in
the rat ventrolateral periaqueductal grey by a µ-opioid. J Physiol 1999; 518
(Pt 2):551–559.
15 Yang K, Furue H, Kumamoto E, Dong YX, Yoshimura M. Pre- and
postsynaptic inhibition mediated by GABAB receptors in rat ventrolateral
periaqueductal gray neurons. Biochem Biophys Res Commun 2003;
302:233–237.
16 Liu J, Ren Y, Li G, Liu ZL, Liu R, Tong Y, et al. GABAB receptors resist acute
desensitization in both postsynaptic and presynaptic compartments of
periaqueductal gray neurons. Neurosci Lett 2013; 543:146–151.
17 Li G, Liu ZL, Zhang WN, Yang K. Blockade of presynaptic 4-aminopyridine-
sensitive potassium channels increases initial neurotransmitter release
probability, reinstates synaptic transmission altered by GABAB receptor
activation in rat midbrain periaqueductal gray. NeuroReport 2016; 27:50–55.
18 Coull JA, Boudreau D, Bachand K, Prescott SA, Nault F, Sík A, et al. Trans-
synaptic shift in anion gradient in spinal lamina I neurons as a mechanism of
neuropathic pain. Nature 2003; 424:938–942.
19 Yang K, Li YQ, Kumamoto E, Furue H, Yoshimura M. Voltage-clamp
recordings of postsynaptic currents in substantia gelatinosa neurons in vitro
and its applications to assess synaptic transmission. Brain Res Brain Res
Protoc 2001; 7:235–240.
20 Johnston GA, Stephanson AL, Twitchin B. Uptake and release of nipecotic
acid by rat brain slices. J Neurochem 1976; 26:83–87.
21 Yang K, Ma WL, Feng YP, Dong YX, Li YQ. Origins of GABAB receptor-like
immunoreactive terminals in the rat spinal dorsal horn. Brain Res Bull 2002;
58:499–507.
22 Yang K. Regulation of excitability in tonic firing substantia gelatinosa neurons
of the spinal cord by small-conductance Ca2 + -activated K + channels.
Neuropharmacology 2016; 105:15–24.
23 Yang K. Postnatal excitability development and innervation by functional
transient receptor potential vanilloid 1 (TRPV1) terminals in neurons of the
rat spinal sacral dorsal commissural nucleus: an electrophysiological study.
Mol Neurobiol 2016; 53:6033–6042.
24 Margeta-Mitrovic M, Mitrovic I, Riley RC, Jan LY, Basbaum AI.
Immunohistochemical localization of GABA(B) receptors in the rat central
nervous system. J Comp Neurol 1999; 405:299–321.
25 Yang K, Wang D, Li YQ. Distribution and depression of the GABAB receptor
in the spinal dorsal horn of adult rat. Brain Res Bull 2001; 55:279–285.
26 Smith CG, Bowery NG, Whitehead KJ. GABA transporter type 1 (GAT-1)
uptake inhibition reduces stimulated aspartate and glutamate release in the
dorsal spinal cord in vivo via different GABAergic mechanisms.
Neuropharmacology 2007; 53:975–981.
27 Keros S, Hablitz JJ. Subtype-specific GABA transporter antagonists
synergistically modulate phasic and tonic GABAA conductances in rat
neocortex. J Neurophysiol 2005; 94:2073–2085.
28 Fields HL, Heinricher MM, Mason P. Neurotransmitters in nociceptive
modulatory circuits. Annu Rev Neurosci 1991; 14:219–245.
29 Li YQ, Rao ZR, Shi JW. Collateral projections from the midbrain
periaqueductal gray to the nucleus raphe magnus and nucleus accumbens in
the rat. A fluorescent retrograde double-labelling study. Neurosci Lett 1990;
117:285–288.