550 Acta Physiologica Sinica, August 25, 2004, 56 (4): 550-557

http://www.actaps.com.cn

Experimental Technique

Visually guided patch-clamp recording of spinal dorsal horn neuron’s

postsynaptic current evoked by primary afferent fiber

WAN Ye-Hong, WANG Yu-Ying, DAI Fei, HU San-Jue*

Institute of Neuroscience, The Fourth Military Medical University, Xi’an, Shaanxi 710033, China

Abstract: The authors describe here the procedures for using the gelatin half-embedding method to obtain thin spinal cord slices with attached
dorsal roots and performing visually guided whole-cell patch-clamp recording of postsynaptic currents evoked by primary afferent fibers in
rat spinal dorsal horn. A segment of spinal cord with attached dorsal roots was prepared and half-embedded in an agar block with 20% (w/v)
gelatin. Thin spinal cord slices with attached dorsal roots were obtained with a vibratome and whole-cell patch-clamp configuration was
established under the infrared observation. At the holding potential of 70 mV, spontaneous excitatory postsynaptic currents (EPSCs) and
dorsal root stimulation-evoked EPSCs were recorded as inward currents. According to the conduction velocity of afferent fibers and stimulus
threshold, evoked EPSCs that are mediated by A-like or C-like fibers were distinguished. At the holding potential of 0 mV, spontaneous
inhibitory postsynaptic currents (IPSCs) and dorsal root stimulation-evoked IPSCs were recorded as outward currents. Using 5 µmol/L
strychnine or 20 µmol/L bicuculline, GABAergic or glycinergic evoked IPSCs could be isolated. Using visual patch-clamp method synaptic
transmission can be accurately assessed by measuring postsynaptic currents of the dorsal horn neurons. More importantly, with the aid of
infrared observation, the incidence of failure to establish a clamp configuration can be greatly reduced and it becomes easier to make recordings
from the neurons in deep dorsal horn laminae. Thus, the present research approach an effective approach to study the modulation of primary
afferent synaptic transmission.

Key words: patch-clamp technique; excitatory postsynaptic current (EPSC); inhibitory postsynaptic currents (IPSC); primary
afferent synapse; spine; dorsal horn; rat

, 710033

: ,
, 20% ,
, 70 mV ,
, A C
0 mV , 5 µmol/L 20 µmol/L γ-
,
, , ,

: ; ; ; ; ; ;
: Q424; R338
Received 2003-12-01 Accepted 2004-02-17
This work was supported by the National Natural Science Foundation of China (No.30030040) and the National Basic Research
Priorities Programme of China (G1999054000).
*
Corresponding author. Tel: +86-29-3374590; Fax: +86-29-3246270; E-mail: sjhu@fmmu.edu.cn
WAN Ye-Hong et al: Visual Patch-Clamp Recording of Spinal DH Neuron’s Postsynaptic Current
The application of infrared videomicroscopy in neu-
rophysiological experiments has successfully solved the
difficulty to visualize individual neurons in standard brain
slices[1,2]. Moreover, the improved resolution afforded by
infrared differential-interference-contrast (IR-DIC)
videomicroscopy greatly increases the visibility of the fine
structure and aids the identification of neurons. Although
the visualization of spinal DH neurons has been carried out
in experimental studies[3], the difficulty still remains in study-
ing the synaptic transmission between primary afferent fi-
bers and DH neurons by infrared microscopy. On the one
hand, the brain slices are generally required to be rather
thin (300~400 µm in thickness) in order to obtain a clear
observation with infrared microscope. This is of more im-
portance for the visualization of neurons in spinal cord
slices, since under the same conditions, the infrared im-
ages collected from spinal cord slices usually have a worse
resolution than those from common brain slices, probably
owing to the fact that a large number of fibers course
through the gray matter of spinal cord. On the other hand,
thick spinal cord slices, which are more than 500 µm in
thickness or even about 1000 µm in some cases, are needed
in conventional slice-making methods to retain the attached
dorsal root[4-6]. Thus, the main problem is how to obtain
the dorsal root-attached slices that are thin enough to al-
low a satisfactory infrared visualization. In the present
study, we employed the gelatin half-embedding method to
obtain thin spinal cord slices with an attached dorsal root
and made patch-clamp recording of postsynaptic currents
evoked by primary afferent fibers under the infrared visual
guidance.
1 MATERIALS AND METHODS
1.1 Animals and materials. Sprague-Dawley rats (18~23
days old) of both sexes were used in our experiments.
Pentobarbital sodium solution for anesthesia was prepared
with distilled water at the concentration of 1% (w/v). The
procedure for making the agar blocks is as follows. 20 g
of agar powder and 4.25 g of NaCl were put into 500 ml of
distilled water in a beaker. The mixture was heated and
stirred until the agar powder was dissolved and the mix-
ture became clear yellowish liquid. Then the hot liquid was
poured into a 150-mm Petri dish, and tens of solid cylin-
ders whose diameter is about 2.5 mm was vertically put
into the liquid. After the cooling and solidification of agar
liquid, the cylinders were pulled out and the agar block
was stored at 4°C. 20% (w/v) gelatin was prepared to
half-embed the spinal cord. 2 g of gelatin powder was
551
added into 10 ml of distilled water. Then, the mixture was
heated and stirred until the gelatin powder was dissolved.
After the bubbles in gelatin solution were gone, the gelatin
solution was drawn into 1-ml syringes and stored at 4°C.
Ringer solution used for slice preparation and perfusion
contained (in mol/L) NaCl 124, KCl 1.9, KH2PO4 1.2,
MgSO4 1.3, CaCl2 2.4, NaHCO3 26 and glucose 10. Mi-
cropipette solution contained (in mol/L) potassium gluconate
135, KCl 5, CaCl2 0.5, MgCl2 2, EGTA 5, HEPES 5 and
Mg-ATP 5. The pH value of micropipette solution was ad-
justed to 7.2~7.3 with 1 mol/L HCl and KOH. At the begin-
ning of each experiment, Ringer solution was equilibrated
with 95% O2 and 5% CO2 for at least 1 h. Using the previ-
ously built agar, we cut out a block with a shallow groove
on the surface. The syringe containing 20% gelatin was
preheated in a thermostat at 37~39°C.
1.2 Preparation of spinal cord slices. Animals were intra-
peritoneally anesthetized with pentobarbital sodium (30~40
mg/kg), and the depth of anesthesia was assessed by check-
ing the palpebral reflex of animals. Then the animal was
placed prostrate on an ice brick to be cooled, with close
attention paid to respiration during the cooling. Usually 5~10
min after the onset of cooling, respiration of the animal
became shallow and regular and the skin temperature fell
to 20~23°C. A 6-cm longitudinal incision was made on the
dorsal skin along the midline, and connective tissue and
longitudinal muscles were quickly divided to expose spi-
nose processes, transverse processes and bilateral verte-
bral foramina. A laminectomy was performed from the
low-thoracic to sacral segment. The surface of exposed
spinal cord was immediately covered with ice-cold Ringer
solution (1~4°C), and about 2 cm long segment of spinal
cord with attached dorsal roots (8~12 mm) was excised
and immersed into cold Ringer solution in a 100-mm Petri
dish. The animal was then killed by exsanguination.
Under a dissecting microscope, the pia mater on the
spinal cord was removed with microforceps. Except for
the L4 and/or L5 dorsal roots on one side, all the dorsal
and ventral roots were carefully cut near their entry zones
with microscissors. Then, the spinal cord was placed into
the shallow groove on the agar block, and several drops of
gelatin were administrated on the low-thoracic and high-
lumbar segment to half-embed the spinal cord (Fig. 1A).
Note that special care should be taken to avoid putting the
gelatin on the entry zones of dorsal roots. The agar block,
together with the half-embedded spinal cord, was glued on
the stage of a vibratome (753 Vibroslice, Campden, UK)
with cyanoacrylate adhesive, and was immersed in the ice-
cold Ringer solution. Then, the process of transverse slic-
552
Fig. 1. Preparation of spinal cord slices. A: A segment of spinal cord
with an attached dorsal root is placed in the shallow groove on the
agar block, and several drops of gelatin are administrated onto the
low-thoracic and high-lumbar segment to half-embed the spinal
cord. B: The agar block, together with the half-embedded spinal
cord, is glued on the stage of the vibratome. The spinal cord stands
vertically with its rostral side downwards and the root entry facing
the blade edge. The blade is carefully adjusted to the position just
above the root entry, and slicing is made to remove the upper part
of the spinal cord. C: The dorsal root is brushed upwards with a
small paintbrush, the blade is adjusted downwards for 300~350
µm, and slicing is made to produce a spinal cord slice with an
attached dorsal root.
Acta Physiologica Sinica, August 25, 2004, 56 (4): 550-557
ing was carried out under an operation microscope. The
spinal cord stood vertically with its rostral side downwards
and the root entry facing the blade edge. The blade was
carefully adjusted to the position just above the root entry
and the slicing was made to remove the upper part of the
spinal cord (Fig. 1B). The dorsal root was brushed up-
wards with a small paintbrush, the blade was adjusted down-
wards for 300~350 µm, and the slicing was made to pro-
duce a spinal cord slice with an attached dorsal root (Fig.
1C). If two dorsal roots were kept intact in the preceding
operation, the above steps were repeated to obtain an other
slice. Then, the obtained slice was incubated in Ringer so-
lution at 24~26°C for at least 1 hour before the patch-
clamp recording.
1.3 Infrared imaging. Video-enhanced infrared micros-
copy was performed according to the previously described
technique[1,2]. The slice was transferred from the incuba-
tion chamber into a recording chamber, where it was held
in position with a “grid” (Fig. 2A), a U-shaped piece of
flattened platinum wire (0.5 mm in diameter) with a paral-
lel array of fine nylon threads. The dorsal root was gently
sucked into a suction electrode. The recoding chamber
was placed on the stage of an upright microscope (BX51WI,
Olympus, Japan) equipped with a 5×objective with nu-
merical aperture (NA) 0.10, 40× water-immersion objec-
tive with NA 0.80, 900-nm infrared bypass filter, and IR-
DIC optics. Lamina regions were identified with a 5×
objective, and individual neurons were visualized with a
40× objective (Fig. 2B and C). The image was enhanced
further with a CCD camera (COHU, USA) and was dis-
played on a computer monitor. The combination of infra-
red illumination, contrast optics and contrast enhancement
by video, allowed the visualization of neurons at a depth of
80~100 µm in the spinal cord slices.
1.4 Patch-clamp recording and dorsal root stimulation.
During the recording, the slice was perfused at a flow rate
of 2~3 ml/min with the equilibrated Ringer solution at
24~26°C. Recording pipettes were made from thin-walled
borosilicate glass and had a resistance of 8~12 MΩ when
filled with pipette solution. The clamp configuration could
be established on DH neurons according to conventional
whole-cell methods. Briefly, the neuron of interest was
approached while applying a slight positive pressure to the
micropipette. After touching the neuron, as seen by a de-
pression in the surface and an increase in resistance, the
positive pressure was released and slight negative pressure
was then applied until a high seal resistance (2~8 GΩ) was
obtained. Then, brief pulses of suction were applied, until
WAN Ye-Hong et al: Visual Patch-Clamp Recording of Spinal DH Neuron’s Postsynaptic Current
Fig. 2. Experimental setup and infrared-visualized patch-clamp
recording. A: Schematic graph of experimental arrangement for the
patch-clamp recording. The spinal cord slice is held in position using
a U-shaped piece of flattened platinum wire with a parallel array of
fine nylon threads. The dorsal root is sucked into a suction electrode.
B and C: Micropipette and dorsal horn neuron observed under the
microscope. The visualization of superficial and deep DH neurons is
shown in (B) and (C), respectively. Note that the image visuallized in
superficial dorsal horn is clearer than that in deep dorsal horn.
553
a large capacitance transient suddenly appeared, which in-
dicates the establishment of whole-cell configuration. Mem-
brane properties were measured under current-clamp
condition, and the statistics were shown as Mean ± SD.
The recordings were made using an amplifier (Multiclamp
700A, Axon Instruments, USA). Signals were low-pass
filtered at 2 kHz and sampled at 20~100 kHz with an
analog-to-digital converter (Digidata 1322A, Axon In-
struments ), and data were stored on a computer for sub-
sequent off-line analysis.
Dorsal roots were orthodromically stimulated through
the suction electrode with an electronic isolated current
stimulator (Nihon Kohden, Japan), as shown in Fig. 2A.
To avoid direct activation of DH neurons or central fibers
in the spinal cord, current stimuli had a short duration
(100~200 µs) and low intensities (1.2~1.5 times the thresh-
old required to evoke the EPSCs or IPSCs on DH neurons).
The evoked synaptic responses were considered as mono-
synaptic ones if their poststimulus latency remained con-
stant and no failures were observed during high-frequency
(20 Hz for A-like fibers and 2 Hz for C-like fibers) repeti-
tive stimulation[4,7,8].
2 RESULTS
Spinal cord slices could be routinely worked on for
6~8 h after their preparation, and it was possible for them
to remain viable for 10 h. Stable whole-cell recording could
be maintained for up to 7 h. Neurons in superficial (lamina
) and deep (lamina - ) dorsal horn could be visual-
ized under the infrared microscope (Fig. 2B and C). We
found that the superficial DH neurons could be more clearly
observed than the deep DH neurons, but the visualization
is good enough to guide us in establishing the clamp con-
figuration on deep DH neurons. The neurons that were
further studied satisfied the following criteria: resting mem-
brane potential more negative than –55 mV and the ampli-
tude of action potential greater than 65 mV. In this research,
postsynaptic currents were recorded from 89 DH neurons.
The mean resting potential of these neurons was –58±3.4
mV, the mean membrane resistance was 537 ± 57 MΩ
and the mean membrane capacitance was 38 ± 5.6 pF.
The mean amplitude of action potentials was 73.2 ± 7.8
mV and the mean overshot was 13.4 ± 6.7 mV. Dorsal
root stimulation under the given experimental conditions
was assessed quantitatively. The evoked EPSCs that are
mediated by A-like or C-like fibers were distinguished with
two measurements: the conduction velocity of afferent fi-
bers and stimulus threshold[4,5,7,8]. The conduction velocity
554 Acta Physiologica Sinica, August 25, 2004, 56 (4): 550-557

was calculated from the length of dorsal root and the la-
tency between postsynaptic current and the preceding
stimulus artifact, with the central delay considered to be 1
ms. Generally, Aβ-like fibers had the fast conduction ve-
locities of 7.5~10 m/s and low stimulus thresholds of 0.
1~0.2 mA, Aδ-like fibers had the velocities of 2~7 m/s and
thresholds of 0.3~0.5 mA, and C-like fibers had the slow
velocities of less than 1 m/s and high thresholds of 0.7~2.
0 mA.
After establishing the whole-cell configuration, spon-
taneous EPSCs and dorsal root stimulation-evoked EPSCs
were recorded at the holding potential of 70 mV (Fig. 3).
Most of the neurons recorded in our experiments were
located in lamina - , and the evoked EPSCs that are
mediated by Aβ-like, Aδ-like or C-like fibers could be ob-
served (Fig. 3B). In some cases, two types of evoked
EPSC could be recorded on one DH neuron (Fig. 3B).
Similar with previous research[4,6], all spontaneous and dor-
sal root stimulation-evoked EPSCs could be blocked by an
AMPA receptor antagonist, CNQX (5 µmol/L), indicating
that these EPSCs were mediated by non-NMDA receptor
(data are not shown). Spontaneous IPSCs and dorsal root
stimulation-evoked IPSCs were recorded at the holding
potential of 0 mV (Fig. 4). Using 5 µmol/L strychnine or
20 µmol/L bicuculline, GABAergic or glycinergic evoked
IPSCs could be isolated (Fig. 4B). The evoked IPSCs in
spinal DH are known to be caused by the activation of
GABAergic or glycinergic interneurons that receive pri-
mary afferent inputs from the periphery[9]. Thus, the IPSCs
evoked by primary afferent fibers are polysynaptic and the

Fig. 3. Spontaneous and evoked EPSCs recorded from dorsal horn neurons. A: The upper trace shows continuous recording of spontaneous
EPSCs at the holding potential of –70 mV, with EPSCs shown as downward deflections. The lower panel shows consecutive traces of
spontaneous EPSCs with an expanded time scale for the period marked by the horizontal bar below the upper record. B: Dorsal root
stimulation-evoked EPSCs mediated by different types of primary afferent fibers. The recordings were made at the holding potential of –70
mV under the stimulation of 0.5 Hz.
WAN Ye-Hong et al: Visual Patch-Clamp Recording of Spinal DH Neuron’s Postsynaptic Current 555

Fig. 4. Spontaneous and evoked IPSCs recorded from dorsal horn neurons. A: The upper trace shows the continuous recording of spontaneous
IPSCs at the holding potential of 0 mV, with IPSCs shown as upward deflections. The lower panel shows consecutive traces of spontaneous
IPSCs with an expanded time scale for the period marked by the horizontal bar below the upper record. B: Evoked IPSCs recorded at the
holding potential of 0 mV under the stimulation of 0.5 Hz. To isolate GABAergic or glycinergic IPSCs, 5 µmol/L strychnine or 20 µmol/L
bicuculline was administrated, respectively.

conduction velocities could not be accurately calculated.
Consistent with other research[8, 10], the time course of
GABAergi IPSCs is longer than that of glycinergic IPSCs.
Spontaneous EPSCs and IPSCs could be observed from
all the recorded neurons. By stimulating the dorsal root,
Aβ-like fiber evoked EPCSs were observed from 12 cells
(13%), Aδ-like fiber evoked EPCSs 53 cells (60%) and C-
like fiber evoked EPCSs 29 cells (33%). The evoked IPSCs
were observed from 9 cells (10%).
3 DISCUSSION
The dorsal horn (DH) of spinal cord is a region where
the central terminals of primary sensory neurons terminate
to form the first synaptic relay with the neurons of central
nervous system. Comprehensive research into synaptic
transmission between primary afferent fibers and DH neu-
rons is of great significance for fully understanding the
relay and modulation of somatosensory information. In-
tracellular recording method has been extensively applied
in the early research on synaptic transmission at primary
afferent synapses in spinal DH[11-13]. Since the intracellular
recordings are made with high-resistance sharp microelec-
trodes , the tip potential of microelectrodes is often un-
stable and can change erratically as the cell is penetrated.
Furthermore, it is difficult to measure quantal events of
the postsynaptic response evoked by presynaptically re-
leased neurotransmitters[14,15]. Afterwards, the blind patch-
clamp method was developed and widely used in electro-
physiological experiments[16,17]. This method was also ap-
plied to make recordings from substantia gelatinosa (SG)
neurons in the superficial dorsal horn[4]. The synaptic func-
tion of primary afferent fibers in SG region was assessed
using the blind method in combination with spinal cord
slice with attached dorsal root[4-6], which greatly advances
research on synaptic transmission in the spinal DH. Such
blind methods can measure miniature postsynaptic cur-
rents and allow the introduction of drugs or proteins into
the cytoplasm through the exchange between electrode
solution and intracellular plasm. But with the blind method,
556
researchers cannot directly observe the condition of neu-
rons in the slices, so that the failure to establish a clamp
configuration often occurs, largely owing to the poor con-
dition of neurons[6]. In addition, seals are often formed on
glial cells or unidentified debris[4]. More importantly, the
blind patch-clamp method is applicable to clamp the neu-
rons that are densely packed, as those in SG region. With
the blind method, it becomes inefficient to record neurons
that are scattered in deep dorsal horn (lamina - ) with
a rather low density.
We described in this paper the procedures for perform-
ing visually guided patch-clamp recording of postsynaptic
currents evoked by primary afferent fibers in rat spinal
dorsal horn. Unlike the brain, the spinal cord is a long, thin
column, which makes it difficult to fix the preparation of
spinal cord on the stage. Consequently, the motion of the
preparation seriously influences the making of thin spinal
cord slices with attached dorsal root. By employing gelatin
half-embedding method, we modified the conventional way
of preparing the spinal cord slices. This method can firmly
fix the preparation on the agar block, which makes it much
easier to successfully obtain thin spinal cord slices retain-
ing primary afferent input from dorsal root. The satisfac-
tory infrared visualization could be made from the thin spi-
nal cord slices prepared according to our method. Like the
blind patch-clamp method, the visual patch-clamp method
overcomes the disadvantage of sharp-microelectrode in-
tracellular recording method. Moreover, with the aid of
infrared visualization, the incidence of failure to establish a
clamp configuration can be greatly reduced and it becomes
easier to make recordings of the neurons in deep dorsal
horn laminas.
The present research provides an effective method to
study the modulation of primary afferent synaptic
transmission, especially the synaptic transmission in deep
dorsal horn. Using the visual method, the condition of neu-
rons can be observed and the spontaneous and evoked
EPSCs or IPSCs can be stably recorded. Primary afferent
mediated by different types of fiber can be distinguished
by measuring the conduction velocity and stimulus
threshold. Moreover, under the condition of whole-cell
voltage-clamp recording, spontaneous EPSCs or IPSCs
(including miniature EPSCs or IPSCs) can be accurately
measured and distinguished by switching the holding
potential. However, the present method has its own
shortcomings. Using the present preparation, it is impos-
sible to decide different somatosensory information medi-
ated by the fibers of same type. For example, we cannot
tell the thermal and mechanical nociceptive information
Acta Physiologica Sinica, August 25, 2004, 56 (4): 550-557
mediated by Aδ- or C- fibers. To solve this problem, in
vivo methods are required. Another problem is that the
whole-cell patch-clamp method used here can lead to the
dialysis between intracellular plasm and micropipette
solution, thus causing the “rundown” of postsynaptic
response. Sometimes, the serious rundown could be ob-
served in our experiments. Perforated patch-clamp tech-
nique is needed to solve this problem.
Acknowledgement:
We thank Dr. Victor Z. Han for his candid and helpful
advice on our research.
REFERENCES
1 Stuart GJ, Dodt HU, Sakmann B. Patch-clamp recordings from
the soma and dendrites of neurons in brain slices using infrared
video microscopy. Pflugers Arch 1993; 423: 511-518.
2 Dodt HU, Zieglgansberger W. Infrared videomicroscopy: a new
look at neuronal structure and function. Trends Neurosci 1994;
17: 453-458.
3 Chery N, Yu XH, de Koninck Y. Visualization of lamina I of the
dorsal horn in live adult rat spinal cord slices. J Neurosci. Meth-
ods 2000; 96: 133-142.
4 Yoshimura M, Nishi S. Blind patch-clamp recordings from
substantia gelatinosa neurons in adult rat spinal cord slices: phar-
macological properties of synaptic currents. Neuroscience 1993;
53: 519-526.
5 Nakatsuka T, Ataka T, Kumamoto E, Tamaki T, Yoshimura M.
Alteration in synaptic inputs through C-afferent fibers to sub-
stantia gelatinosa neurons of the rat spinal dorsal horn during
postnatal development. Neuroscience 2000; 99: 549-556.
6 Yang K, Li Y, 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.
7 Randic M, Jiang MC, Cerne R. Long-term potentiation and long-
term depression of primary afferent neurotransmission in the rat
spinal cord. J Neurosci 1993; 13: 5228-5241.
8 Luo C, Kumamoto E, Furue H, Chen J, Yoshimura M. Nociceptin
inhibits excitatory but not inhibitory transmission to substantia
gelatinosa neurones of adult rat spinal cord. Neuroscience 2002;
109: 349-358.
[9] Narikawa K, Furue H, Kumamoto E, Yoshimura M. In vivo patch-
clamp analysis of IPSCs elicited in rat substantia gelatinosa neu-
rons by cutaneous mechanical stimulation. J Neurophysiol 2000;
84: 2171-2174.
10 Yang K, Feng Y, Li Y. Baclofen inhibition of dorsal root-evoked
inhibitory postsynaptic currents in substantia gelatinosa neu-
rons of rat spinal cord slice. Brain Res 2001; 900: 320-323.
11 King AE, Thompson SW, Urban L, Woolf CJ. The responses
WAN Ye-Hong et al: Visual Patch-Clamp Recording of Spinal DH Neuron’s Postsynaptic Current
recorded in vitro of deep dorsal horn neurons to direct and
orthodromic stimulation in the young rat spinal cord. Neuro-
science 1988; 27: 231-242.
12 Yoshimura M, Jessell TM. Primary afferent-evoked synaptic
responses and slow potential generation in rat substantia
gelatinosa neurons in vitro. J Neurophysiol 1989; 62: 96-108.
13 Gerber G, Cerne R, Randic M. Participation of excitatory amino
acid receptors in the slow excitatory synaptic transmission in rat
spinal dorsal horn. Brain Res 1991; 561: 236-251.
14 Yoshimura M, North RA. Substantia gelatinosa neurones hyper-
polarized in vitro by enkephalin. Nature 1983; 305: 529-530.
557
15 Yoshimura M, Jessell TM. Amino acid-mediated EPSPs at
primary afferent synapses with substantia gelatinosa neurones in
the rat spinal cord. J Physiol 1990; 430: 315-335.
16 Liu ZW ( ), Li LJ, Liu CG. Blind patch clamp whole-cell
recording technique for neurons in hippocampal slices. Acta
Physiol Sin ( ) 2001; 53(5): 405-408 (Chinese, English
abstract).
17 Tu YL ( ), Liu YB, Zhang L, Zhao YJ, Wang L, Hu ZA.
Changes in electrophysiological and morphological properties of
neurons during the development of the visual cortex in the rat.
Acta Physiol Sin ( ) 2003; 55(2): 206-212 (Chinese,
English abstract).