Am J Physiol Gastrointest Liver Physiol 289: G670 –G678, 2005.
First published May 26, 2005; doi:10.1152/ajpgi.00028.2005.
Morphological and electrophysiological changes in mouse dorsal root ganglia
after partial colonic obstruction
Tian-Ying Huang and Menachem Hanani
Laboratory of Experimental Surgery, Hebrew University-Hadassah Medical School, Mount Scopus, Jerusalem, Israel
Submitted 24 January 2005; accepted in final form 24 May 2005
dye coupling; neuronal excitability; hypertrophy; visceral pain
has provided evidence for a high
degree of plasticity in the innervation of visceral organs. It was
found that local injury or inflammation can alter both the
intrinsic and extrinsic ganglia innervating these organs. For
example, it was found that intestinal obstruction induced hypertrophy of myenteric neurons in the small intestine of experimental animals (7, 14).
An important component of the extrinsic sensory innervation
of visceral organs derives from dorsal root ganglia (DRG).
Whereas the physiology and pathophysiology of somatosensory innervation by DRG have been studied extensively (11),
little information is available on how local injury affects DRG
neurons that innervate the gastrointestinal (GI) tract and other
visceral organs. Bielefeldt et al. (4) found that gastric inflammation in rats increased the excitability of DRG neurons
RESEARCH IN RECENT YEARS
Address for reprint requests and other correspondence: M. Hanani, Hebrew
Univ.-Hadassah Medical School, Mount Scopus, Jerusalem 91240, Israel
(e-mail: [email protected]).
G670
projecting to the stomach. Similarly, inflammation in the
guinea pig small intestine (23) and mouse colon (3) sensitized
DRG neurons. These studies suggest that an important component of chronic visceral pain may originate at the sensory
ganglia, which is in accord with studies on somatic neuropathic
pain models showing that peripheral nerve injury induces
ectopic firing in DRG neurons (39).
Obstruction of hollow visceral organs is a common clinical
problem that may have severe consequences such as pain,
sepsis, and perforation. A previous study (15) showed that
obstruction of the bladder outlet caused hypertrophy of neurons
in the pelvic ganglia. Information on the effect of intestinal
obstruction on DRG is scarce. We are aware of only one
investigation of this question, done on a model of partial
obstruction of the rat small intestine (42). It was found that 3
wk after the induction of the partial obstruction, the crosssectional area of DRG neurons innervating the obstructed
region increased by 130% (42). No information is available
about physiological changes in the sensory neurons during
intestinal obstruction or on any changes in glial cells in these
ganglia. Glial cells are considered as important contributors to
pain states in the central nervous system (40). We found that
axotomy of the sciatic nerve in mice induced a sixfold increase
in the coupling among satellite glial cells (SGCs) surrounding
DRG neurons (17). This was correlated with an abnormal
growth of glial processes and with a 6.5-fold increase in the
number of gap junctions connecting these cells (17, 27).
Similar observations were made in mouse trigeminal ganglia
(9), and we hypothesized that this change might contribute to
chronic pain caused by the axotomy. In the present study, we
examined the effect of partial obstruction of the mouse colon
on dye coupling among SGCs, on electrophysiological properties of DRG neurons, and on behavioral pain responses.
MATERIALS AND METHODS
Subjects and surgery. Experiments were done on Balb/c mice of
either sex, 2–3 mo old, weighing 22–25 g. The experimental protocol
was approved by the Institutional Animal Care and Use Committee.
The animals were anesthetized by an intraperitoneal administration of
pentobarbital sodium (48 mg/kg). All procedures were done using
aseptic techniques.
To trace DRG neurons, the colon was externalized through a low
abdominal midline incision onto a mat of cotton soaked with saline,
and the distal colon wall was injected with 1,1⬘-dioctadecyl-3,3,3⬘,3⬘tetramethylindocarbocyanine perchlorate (DiI; Molecular Probes; Eugene, OR; 5% in methanol, 20 ␮l) (36). Eight injections with a
28-gauge needle were made circumferentially within an about 1-cmlong segment of the distal colon, 2 cm from the anus.
The costs of publication of this article were defrayed in part by the payment
of page charges. The article must therefore be hereby marked “advertisement”
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
0193-1857/05 $8.00 Copyright © 2005 the American Physiological Society
http://www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.5 on June 15, 2017
Huang, Tian-Ying, and Menachem Hanani. Morphological
and electrophysiological changes in mouse dorsal root ganglia
after partial colonic obstruction. Am J Physiol Gastrointest Liver
Physiol 289: G670 –G678, 2005. First published May 26, 2005;
doi:10.1152/ajpgi.00028.2005.—There is evidence that sensitization of neurons in dorsal root ganglia (DRG) may contribute to
pain induced by intestinal injury. We hypothesized that obstruction-induced pain is related to changes in DRG neurons and
satellite glial cells (SGCs). In this study, partial colonic obstruction
was induced by ligation. The neurons projecting to the colon were
traced by an injection of 1,1⬘-dioctadecyl-3,3,3⬘,3⬘-tetramethylindocarbocyanine perchlorate into the colon wall. The electrophysiological properties of DRG neurons were determined using intracellular electrodes. Dye coupling was examined with an intracellular injection of Lucifer yellow (LY). Morphological changes in
the colon and DRG were examined. Pain was assessed with von
Frey hairs. Partial colonic obstruction caused the following
changes. First, coupling between SGCs enveloping different neurons increased 18-fold when LY was injected into SGCs near
neurons projecting to the colon. Second, neurons were not coupled
to other neurons or SGCs. Third, the firing threshold of neurons
projecting to the colon decreased by more than 40% (P ⬍ 0.01),
and the resting potential was more positive by 4 – 6 mV (P ⬍ 0.05).
Finally, the number of neurons displaying spontaneous spikes
increased eightfold, and the number of neurons with subthreshold
voltage oscillations increased over threefold. These changes are
consistent with augmented neuronal excitability. The pain threshold to
abdominal stimulation decreased by 70.2%. Inflammatory responses
were found in the colon wall. We conclude that obstruction increased
neuronal excitability, which is likely to be a major factor in the pain
behavior observed. The augmented dye coupling between glial cells
may contribute to the neuronal hyperexcitability.
CHANGES IN SENSORY GANGLIA AFTER COLONIC OBSTRUCTION
AJP-Gastrointest Liver Physiol • VOL
a preamplifier (model IR 283, Neuro Data Instruments). Input resistance of the neurons was measured by passing hyperpolarizing currents (0.1 nA, 100 ms) and balancing the bridge. Electrophysiological
data were recorded with a video cassette recorder using a Neurocorder (model DR 390, Neuro Data Instruments). Neurons were
classified as displaying subthreshold oscillations when they showed
rhythmic changes in membrane potentials of at least 1 mV in amplitude (1). Membrane potentials were sampled for 2 s at 2 kHz. Spectral
analysis was done with the fast Fourier transform module of pCLAMP
9 (Axon Instruments; Foster City, CA). All intracellular recording
experiments were done on DRG harvested after 6 days of obstruction
or sham operation.
Morphological studies. Two weeks after the DiI injection into the
colon wall, DRG L1–L6 and S1 were harvested. The ganglia were
pinned onto the Sylgard bottom of a dish. DiI-labeled neurons were
counted under a microscope equipped with fluorescent illumination
using filters for TRITC. Six days after obstruction or sham operation,
DRG S1 and the colonic tissues, which were 1.5–3.5 cm proximal to
the anus, were harvested and fixed overnight at 4°C in 4% paraformaldehyde in PBS. The DRG and colon were then embedded in
paraffin using conventional histological techniques. Serial sections, 5
␮m thick, were cut, deparaffinized in xylene, rehydrated in a graded
series of ethanol, and stained with hematoxylin-eosin. DRG sections
were imaged at ⫻400. The cross-sectional areas of all nucleated
neurons encountered in the field were measured using Image-Pro Plus
software (Media Cybernetics; Silver Spring, MD) and divided by 0.85
to correct for 15% tissue shrinkage (10). Colonic sections were
examined and imaged in a similar manner. The thickness of the
muscle layers was measured. The external diameter of the empty
colon was evaluated using the external circumference of the freshly
dissected colon.
For myeloperoxide staining, the colon was opened along the
mesentery and pinned in a dish, and the mucosa and submucosa were
removed under microscopic observation. The segments were fixed in
100% ethanol for 15 min and then washed twice in PBS. Myeloperoxidase-positive cells were detected by incubating the tissues in 0.5
mg/ml Hanker-Yates reagent (Sigma) and 5 ␮l/ml of 3% H2O2 in PBS
at room temperature for 12 min (31). Tissues were air dried and
coveslipped. Stained cells were counted in seven randomly selected
areas in each specimen.
Assessment of visceral pain. Von Frey hairs were used to measure
the withdrawal responses in unrestrained mice to mechanical stimulation of unshaved skin in the low abdomen. Before the behavioral
tests, the animals were allowed to accustom to the new environment
for at least 30 min. Appearance of the following behaviors on
application of a hair was considered as a withdrawal response: 1)
sharp abdominal retraction, 2) immediate licking or scratching of the
site of application of the hair, or 3) jumping. Hairs calibrated for
forces of 0.5, 1.0, 2.0, 3.0, and 4.0 g were applied 10 times each in
ascending order of force. The probability and threshold of withdrawal
responses were recorded. The hair was applied for 1–2 s at 5- to 10-s
intervals. Care was taken not to stimulate the same point in succession. The threshold of the withdrawal response was defined as the
minimum force eliciting two subsequent withdrawal responses. The
use of this criterion allowed us to obtain reliable and reproducible
results.
Statistical analysis. Values are expressed as means ⫾ SE. Fisher’s
exact test, Mann-Whitney test, paired t-test, and ANOVA were used
for comparison. P ⬍ 0.05 was considered as statistically significant.
RESULTS
Retrograde labeling of DRG neurons. Two weeks after the
DiI injection into the distal colon, there were numerous DiIlabeled neurons evenly distributed in DRG L1 (17.6 ⫾ 2.6
cells/ganglion, n ⫽ 16) and S1 (17.5 ⫾ 2.8 cells/ganglion), few
in L2 (2.9 ⫾ 1.0 cells/ganglion) and L6 (3.1 ⫾ 1.0 cells/
289 • OCTOBER 2005 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.5 on June 15, 2017
To induce partial colonic obstruction, the distal colon was externalized as described above. A plastic tube (2 mm in diameter) was
placed along the longitudinal axis of the colon, and a silk thread was
tied around the colon and the tube. Partial obstruction was achieved
after the tube was withdrawn. The ligation was made 1.5 cm proximal
to the anus and distal to the DiI injection region. The colon was
replaced, and the incision was sutured in two layers. In sham operations, the colon was only externalized and replaced. Some control
animals were injected with DiI 8 days before the sham operation.
Tissue preparations. The animals were killed with CO2 and DRG
L1, L4, and S1 were removed and placed in cold (4°C) Krebs solution
(pH 7.4) containing (in mM) 118 NaCl, 4.7 KCl, 14.4 NaHCO3, 1.2
MgSO4, 1.2 NaH2PO4, 2.5 CaCl2, and 11.5 glucose. Ganglia for
intracellular recording or labeling were pinned onto the Sylgard
bottom of a chamber superfused with Krebs solution bubbled with
95% O2-5% CO2 at 23–24°C. For histological sections, the ganglia
and colon were fixed overnight at 4°C in 4% paraformaldehyde in
PBS (0.1 M, pH 7.4).
For mapping DRG neurons projecting to the colon, DRG L1–L6
and S1 were examined. For dye injection and intracellular recording,
we used DRG L1 and S1 from sham-operated (control) and obstructed
animals and DRG L4 from the same obstructed animals (negative
control). After identification of neurons projecting to the colon, both
dye injections and intracellular recordings were made on DRG L1 and
S1 from the sham-operated and obstructed mice.
Intracellular labeling and recording. Experiments were performed
using an upright microscope (Axioskop FS, Zeiss; Jena, Germany)
equipped with fluorescent illumination and a digital camera (Pixera
120e, Pixera) connected to a personal computer. A water-immersion
⫻40 objective (numerical aperture 0.8) was used in all experiments,
allowing us to visualize dye-labeled cells very clearly and to classify
them according to their morphology. Neurons and SGCs were singly
injected with the fluorescent dye Lucifer yellow (LY; Sigma Chemical; St. Louis, MO; 3% in 0.5 M LiCl solution) from a glass
microelectrode with a tip resistance of 80 –120 M⍀. After identification of neurons projecting to the colon with DiI retrograde labeling,
the SGCs near the DiI-labeled neurons were singly injected with LY.
The dye was injected by hyperpolarizing current pulses of 100 ms in
duration and 0.5 nA in amplitude at 10 Hz for 3–5 min. To determine
whether LY was injected into the DiI-labeled neurons or the SGCs
near DiI-labeled neurons, we photographed the same microscopic
field twice using the appropriate filter sets (TRITC for DiI and FITC
for LY) and then merged the two images to determine the spatial
relationship between DiI-labeled neurons and dye-injected cells. In a
series of control experiments, we found that in all cases (n ⫽ 35) when
the DiI-labeled neuron was injected with LY only the DiI-labeled
neuron was also stained with LY. As mentioned above, these observations were made with a high-power objective, which enabled
precise visual control in real time. During and after the dye injections,
living neurons and SGCs were photographed. The numbers of SGCs
coupled to the dye-injected cell were counted during dye injection by
changing the microscopic focus level. This allowed us to identify the
injected cells unequivocally. Also, we found that by using this
approach rather than imaging fixed tissues, we overcame the problem
of dye fading, which took place during the experiments that lasted
several hours. In some experiments, octanol (Sigma, 1.0 mM) was
added to the bathing solution. After the experiments, DRG were fixed
overnight at 4°C in 4% paraformaldehyde in PBS, washed with PBS,
and mounted in Gel/mount (Biomeda). Cells labeled with LY were
imaged with a Bio-Rad confocal microscope. For dye injection, DRG
were harvested after 4 or 6 – 8 days of obstruction or sham operation.
For intracellular recording, DRG were bathed in Krebs solution at
32°C. We recorded from DRG neurons that were impaled blindly and
also from the neurons projecting to the distal colon, as determined by
retrograde labeling with DiI. The microelectrodes were filled with 2 M
KCl, with tip resistances of 80 –120 M⍀. Transmembrane currents
were passed through the recording electrode using the bridge circuit of
G671
G672
CHANGES IN SENSORY GANGLIA AFTER COLONIC OBSTRUCTION
Mann-Whitney test) at 4 days and 4.15 ⫾ 0.32 SGCs/cell (n ⫽
133, P ⬍ 0.01) at 6 – 8 days, respectively (Fig. 1F). Dye
coupling in L4 from obstructed animals was very similar to that
observed in DRG L1 and S1 from sham-operated mice.
To test whether the increased dye coupling after obstruction
was due to gap junctions, we added octanol (1 mM) to the
bathing solution. In the presence of octanol, the incidence of
dye coupling in ganglia from obstructed animals at 6 – 8 days
decreased from 13.0% to 1.1% (n ⫽ 88, P ⬍ 0.001 by Fisher’s
exact test) between SGC envelopes and from 41.0% to 18.6%
(P ⬍ 0.001) between SGCs around the same neuron (Fig. 1, D–
F). The results indicate that dye coupling was mediated by gap
junctions.
The results described above were obtained when LY was
injected randomly into SGCs. When LY was injected into
SGCs near (⬍20 ␮m) DiI-labeled neurons, the interenvelope
dye coupling in the ganglia of obstructed animals was 44.1%
(P ⬍ 0.0001, n ⫽ 59) of the LY-injected cells versus 2.5%
(n ⫽ 40) in the control, a 18-fold increase. Coupling within
envelopes increased from 25% to 81.4% after 6 – 8 days of
obstruction (Fig. 2). These results indicated that the response to
obstruction occurred preferentially in SGCs located in the
vicinity of neurons innervating the obstructed colon.
Electrophysiological characteristics of DRG neurons. Intracellular recordings were made from 425 sensory neurons that
were impaled randomly and 99 DiI-labeled neurons that projected to the colon. The electrophysiological properties of the
Fig. 1. Dye coupling of dorsal root ganglia (DRG) cells. In control ganglia, a Lucifer yellow (LY)-injected neuron was not coupled to other cells (A), and dye
coupling was seen frequently (22.5%) between satellite glial cells (SGCs) ensheathing the same neuron (B), but was only seen in 2.3% of the cases between SGCs
of neighboring neurons. After obstruction, the incidence of coupling between SGCs around different neurons (C) increased to 13.0%. *LY-injected cells; arrows
indicate the cell body of SGCs coupled to the dye-injected cells. Scale bars ⫽ 20 ␮m. The histograms show changes after obstruction in coupling incidence
between SGC envelopes around different neurons (D), between SGCs around the same neuron (E), and the number of SGCs coupled to the LY-injected cells
(F). Fisher’s exact test (D and E) and Mann-Whitney test were used for comparison. *P ⬍ 0.05 and **P ⬍ 0.01 compared with the control.
AJP-Gastrointest Liver Physiol • VOL
289 • OCTOBER 2005 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.5 on June 15, 2017
ganglion), and none in L3, L4, and L5 ganglia. These results
are in accord with those obtained by Robinson et al. (30) and
indicated that the distal colon receives extrinsic sensory innervation mostly from DRG L1 and S1. We therefore focused on
DRG L1 and S1, and L4 ganglia from the same obstructed
animals were used as controls.
Dye coupling. The dye LY, which crosses gap junctions, was
injected into single DRG neurons or SGCs. The labeling
process was observed in real time under a fluorescence microscope. All LY-injected neurons (n ⫽ 237) were not coupled to
other cells, neurons, or SGCs (Fig. 1A), and, also, all LYinjected SGCs were not coupled to any neurons. Glial cells
displayed two types of dye coupling: between SGCs enveloping the same neuron (Fig. 1B) and between SGC envelopes
around different neurons (Fig. 1C). In control ganglia, the dye
coupling incidence was 2.3% (n ⫽ 258) between SGC envelopes and 22.5% within the SGC envelope around a given
neuron. After the induction of obstruction, the incidence of
coupling between SGC envelopes was 7.4% (n ⫽ 231, P ⬍
0.05 by Fisher’s exact test) at 4 days and 13.0% (n ⫽ 324, P ⬍
0.001) at 6 – 8 days (Fig. 1D), i.e., the coupling incidence
increased 3.2- and 5.6-fold compared with controls. The incidence of coupling between SGCs around the same neuron was
32.0% (P ⬍ 0.01) and 41.0% (P ⬍ 0.001) at 4 and 6 – 8 days,
i.e., increased by 42% and 82%, respectively (Fig. 1E). The
number of SGCs coupled to LY-injected cells increased from
2.15 ⫾ 0.26 to 3.16 ⫾ 0.30 SGCs/cell (n ⫽ 74, P ⬍ 0.05 by
CHANGES IN SENSORY GANGLIA AFTER COLONIC OBSTRUCTION
G673
determined by DiI labeling (see Table 1). Thus partial colonic
obstruction augmented neuronal excitability markedly.
In control ganglia, the majority of neurons were electrically
quiescent (Fig. 3A); 20.4% (n ⫽ 98) of A-type cells and 22.7%
(n ⫽ 95) of C-type cells displayed subthreshold membrane
potential oscillations (Fig. 3B and C). In the obstructed animals, the proportion of neurons displaying subthreshold membrane potential oscillations was more than twofold greater than
in controls (P ⬍ 0.05 by Fisher’s exact test) in both A- and
C-type cells when the neurons were impaled blindly. In neurons projecting into the colon, the proportion was over threefold greater and reached 54.2% and 68.8% in A- and C-type
neurons, respectively (P ⬍ 0.01). Spectral analysis of the
oscillations showed that the peak frequency of A-type neurons
was greater than that of C-type cells in control as well as
obstructed animals. In control animals, the oscillation frequency was 35.2 ⫾ 3.1 Hz (n ⫽ 5) in A-type cells versus
neurons were compared between experimental animals (6 days
after the induction of partial obstruction) and controls (ganglia
L1 and S1 from sham-operated animals and L4 ganglia in
obstructed animals). DRG neurons were classified as A-type
and C-type cells, based on the duration of their action potentials, measured at their base. A-type neurons were defined by a
duration of ⬍10 ms, and the rest were classified as C-type
cells (8).
After obstruction, the resting potential was less negative and
the membrane input resistance was lower in both A- and
C-type neurons compared with controls (Table 1). We used the
current threshold for firing an action potential as a measure for
neuronal excitability and found that this current was about 30%
lower in the obstructed animals (in both cell types, impaled
randomly). The threshold current was considerably lower in
obstructed animals (46% and 42% of controls in A- and C-type
cells, respectively) in neurons projecting to the distal colon, as
Table 1. Electrophysiological properties of DRG neurons after partial colonic obstruction
A-type neurons
n
RMP, mV
APA, mV
APD, ms
It, nA
Ri, M⍀
SPS
SPO
C-type neurons
n
RMP, mV
APA, mV
APD, ms
It, nA
Ri, M⍀
SPS
SPO
Control (L1/S1)
Obstructed (L4)
Obstructed (L1/S1)
Control (DiI, L1/S1)
Obstructed (DiI, L1/S1)
98
53.0⫾1.2
61.8⫾1.7
6.4⫾0.2
0.47⫾0.03
45.5⫾0.8
3 (3.1%)
20 (20.4%)
47
54.0⫾1.4
61.3⫾1.7
6.3⫾0.2
0.48⫾0.03
47.5⫾1.2
1 (2.1%)
10 (21.3%)
93
48.2⫾1.4*
59.5⫾1.8
6.4⫾0.2
0.32⫾0.03†
41.5⫾0.8*
14 (15.1%)†
40 (43.0%)†
33
53.8⫾1.2
64.2⫾1.9
6.2⫾0.3
0.48⫾0.04
44.6⫾1.2
1 (3.0%)
5 (15.1%)
24
47.5⫾1.5*
61.9⫾2.0
6.3⫾0.3
0.26⫾0.03†
40.2⫾0.8*
6 (25%)†
13 (54.2%)†
75
53.5⫾1.5
62.8⫾1.9
11.6⫾0.3
0.42⫾0.03
47.8⫾0.9
2 (2.7%)
17 (22.7%)
35
53.6⫾1.4
61.4⫾1.7
11.8⫾0.5
0.41⫾0.04
47.3⫾1.2
1 (2.9%)
7 (20%)
77
48.0⫾1.4*
59.8⫾1.8
11.6⫾0.3
0.29⫾0.02†
42.4⫾0.8*
14 (18.2%)†
36 (46.8%)†
29
54.5⫾1.1
63.2⫾1.8
11.7⫾0.3
0.38⫾0.03
45.8⫾1.2
1 (3.4%)
6 (20.7%)
32
49.1⫾1.2*
62.6⫾2.0
11.9⫾0.4
0.22⫾0.03†
42.6⫾1.1*
9 (28.1%)†
22 (68.8%)†
Values are expressed as means ⫾ SE; n is the number of neurons examined. RMP, resting membrane potential; APA, action potential amplitude; APD, action
potential duration; It, current threshold for firing an action potential; Ri, membrane input resistance; SPS, spontaneous spikes; SPO, spontaneous membrane
potential oscillations. *P ⬍ 0.05 and †P ⬍ 0.01 compared with controls. Fisher’s exact test and Mann-Whitney test were used for comparison.
AJP-Gastrointest Liver Physiol • VOL
289 • OCTOBER 2005 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.5 on June 15, 2017
Fig. 2. SGC dye coupling in the vicinity (within 20 ␮m) of DRG neurons projecting to the colon. Neurons projecting to the colon were traced by injecting
1,1⬘-dioctadecyl-3,3,3⬘,3⬘-tetramethylindocarbocyanine perchlorate (DiI) into the colon wall. A: a neuron (N1) was labeled by DiI, indicating that it projects to
the distal colon. B: a single SGC (*) near this neuron was injected with LY. The arrows indicate cell bodies of SGCs coupled to the dye-injected cell, and N2
marks an adjacent neuron not labeled with DiI. This SGC was coupled to SGCs ensheathing two adjacent neurons. Scale bars ⫽ 20 ␮m. C: histogram comparing
the coupling incidence of SGCs near neurons projecting to the distal colon in controls and after obstruction. After obstruction, coupling incidence increased for
both SGCs around different neurons (open bars) and those around the same neuron (solid bars). Fisher’s exact test was used for comparison. *P ⬍ 0.0001
compared with the control.
G674
CHANGES IN SENSORY GANGLIA AFTER COLONIC OBSTRUCTION
24.5 ⫾ 3.3 Hz (n ⫽ 6) in C-type cells (P ⬍ 0.05 by MannWhitney test). After obstruction, the frequency was 29.7 ⫾ 2.1
Hz (n ⫽ 13) in A-type neurons and 24.2 ⫾ 1.7 Hz (n ⫽ 22) in
C-type neurons (P ⬍ 0.05; see Fig. 3, F and G). Qualitatively
similar results were obtained by Amir et al. (1) in rat DRG
neurons after axotomy.
Neurons with spontaneous action potentials (Fig. 3, D and E)
were rare in control ganglia (3.1% of A-type cells and 2.7% in
C-type cells). After obstruction, the proportion of neurons
firing spontaneously increased by over fivefold in randomly
impaled A- and C-type neurons and by over eightfold in
neurons of both types projecting into the distal colon (P ⬍ 0.01
compared with the control).
Change in neuronal size. Figure 4A shows a cross section of
a control ganglion, and Fig. 4B shows a ganglion from an
obstructed animal. The neurons appeared to be larger after
obstruction, and this was demonstrated quantitatively in the
size distributions shown in Fig. 4C. The mean cross-sectional
AJP-Gastrointest Liver Physiol • VOL
area of neurons in obstructed animals was 35.3% greater than
the control (control, 1,122.1 ⫾ 33.9 ␮m2, n ⫽ 318; obstruction,
1,518.1 ⫾ 34.3 ␮m2, n ⫽ 337; P ⬍ 0.0001 by Mann-Whitney
test; see Fig. 4D).
Changes in the colon. The thickness of the muscle layers in
the obstructed colon was greater than that in the control by
60% (P ⬍ 0.0001 by Mann-Whitney test; Fig. 5), and the
diameter of the obstructed colon was greater than that in the
control by 69.8% (P ⬍ 0.0001 by Mann-Whitney test; Fig.
5D). Myeloperoxide staining showed that the density of neutrophils in the colonic muscle layers was 89.2 ⫾ 6.4 cells/mm2
in the controls versus 406.9 ⫾ 36.0 cells/mm2 after obstruction
(n ⫽ 4 for each group, P ⬍ 0.0001 by Mann-Whitney test; see
Fig. 6). These results show that obstruction led to hypertrophy
and inflammation of the colon.
Change in pain threshold. Before induction of the partial
obstruction, the threshold for the withdrawal response was
3.83 ⫾ 0.12 g (n ⫽ 18) compared with 1.14 ⫾ 0.12 g after 6
289 • OCTOBER 2005 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.5 on June 15, 2017
Fig. 3. Partial obstruction-induced changes in spontaneous electrical activity in DiI-labeled neurons. A: recording from control ganglia showed that a C-type
neuron was typically electrically quiescent. B: an A-type neuron from obstructed animal displayed spontaneous subthreshold membrane potential oscillations at
39 Hz. C: another recording from a C-type cell in an obstructed mouse showed subthreshold oscillations with lower frequency (17 Hz). D: an A-type neuron
from an obstructed mouse displaying spontaneous action potentials at 35 Hz. E: another example of spontaneous action potentials at lower frequency (15 Hz)
was recorded from a C-type cell in an obstructed animal. F: spectral analysis (power spectrum) of spontaneous subthreshold membrane potential oscillations
recorded from 13 A-type cells in obstructed animals and 5 cells in control animals. The peaks were observed at 30 Hz for neurons in obstructed animals and
at 35 Hz in controls. G: spectral analysis of C-type spontaneous membrane potential oscillations obtained from 22 cells in obstructed and 6 cells in controls. The
peak frequency was about 25 Hz for both obstructed and control animals. The reference power spectrum (without oscillations) in F and G was obtained from
data of 10 quiescent DRG neurons (5 A-type and 5 C-type) in control animals.
CHANGES IN SENSORY GANGLIA AFTER COLONIC OBSTRUCTION
G675
days of obstruction (P ⬍ 0.0001 by paired t-test). Thus the
threshold was 70% lower. The probability (in %; ⫾SE) of the
withdrawal response to the mechanical stimulation to the low
abdominal skin is summarized in Fig. 7. The probability was significantly higher in the obstructed than the control animals. These
findings indicate that obstruction induced hyperalgesia of the
somatic area to which visceral pain from the colon referred.
DISCUSSION
Our results demonstrate that partial colonic obstruction induces profound physiological and morphological changes in
neurons and SGCs in DRG that innervate the colon. Thus
intestinal lesion can lead to changes not only locally but also at
far removed sites, as found for DRG neurons after intestinal (3,
Fig. 5. Hypertrophy in the colon wall after
partial obstruction. Compared with the colon
section of control animals (A), the muscle
layers and mucosa were much thicker in the
section of obstructed colon (B) after 6 days
of obstruction (scale bar ⫽ 100 ␮m). Histograms show an increase in the thickness of
muscle layers (C) and in the diameter (D) of
obstructed colons; n ⫽ 18 for controls and
28 for obstruction. Mann-Whitney test was
used for comparison. *P ⬍ 0.0001 compared
with the control.
AJP-Gastrointest Liver Physiol • VOL
289 • OCTOBER 2005 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.5 on June 15, 2017
Fig. 4. Neuronal hypertrophy in DRG after
6 days of partial colonic obstruction. Compared with the control section (A), the neuronal profiles in the section from the obstructed animal (B) appear enlarged (scale
bar ⫽ 50 ␮m). C: frequency histograms
showing the size distributions of neurons
after obstruction vs. controls. Obstruction
shifted the peak of the histogram to the right.
D: histogram showing an increase in crosssectional area of the neurons. *P ⬍ 0.0001
by Mann-Whitney test.
G676
CHANGES IN SENSORY GANGLIA AFTER COLONIC OBSTRUCTION
Fig. 6. Histochemical staining for myeloperoxidase to detect neutrophils in the colon
musculature after obstruction. A: control;
neutrophils were rare. B: after obstruction,
numerous neutrophils were present. Scale
bar ⫽ 50 ␮m.
Fig. 7. Change in pain behavior after partial colonic obstruction in mice. The
withdrawal responses to the mechanical stimulation of low abdominal skin are
plotted as a function of the applied force, and the curves show that the
probability of withdrawal response was much higher in obstructed animals
than control; n ⫽ 18. *P ⬍ 0.0001 by ANOVA.
AJP-Gastrointest Liver Physiol • VOL
Dye coupling was largely inhibited by the gap junction
blocker octanol, indicating that it was due to gap junctions
connecting the SGCs. The morphological basis for the increased coupling is not yet known, but it can be proposed that
it is at least partly the result of growth of glial processes
bridging SGCs around different neurons and the formation of
new gap junctions between them, as found in our electron
microscopic study on DRG after axotomy (17, 27). The functional significance of the augmented glial coupling is not clear,
but two possible ideas can be put forward. First, it is established that in the central nervous system, glial coupling contributes to the buffering of K⫹ that accumulate in the extracellular space during intense neural activity (25). There is evidence for ectopic electrical activity in DRG neurons after
peripheral nerve injury (11), as also found in the present study
(Fig. 3), which may lead to local K⫹ accumulation. It can be
proposed that the augmented glial coupling after obstruction
prevents the build up of harmful levels of K⫹ in the ganglia.
Interestingly, increasing extracellular K⫹ was found to promote dye coupling between SGCs (20). Second, it is known
that damage causes tissues to revert to an embryonic-like state,
which is characterized by extensive intercellular coupling via
gap junctions (13, 22). It is conceivable that during obstruction-induced damage, the same mechanism is activated. We
proposed previously that the increased glial coupling after
nerve injury might contribute to the symptoms of neuropathic
pain, as it has the potential to mediate long-distance communications between neurons (9, 17). A similar mechanism may
operate during obstruction.
Changes in DRG neurons. We found that the cross-sectional
area of neurons in DRG that innervate the partly obstructed
colon was 35.3% greater than that in controls. Similar results
were reported for rat DRG after partial obstruction of the ileum
(42), rat pelvic ganglion after partial obstruction of the urinary
bladder outlet (15), and in rat myenteric plexus after partial
obstruction of the ileum (14). In all these cases, there was also
hypertrophy of smooth muscle in the obstructed organs. Purves
(28) proposed that the size of neurons increases when their
target organs undergo hypertrophy, and apparently the same
principle applies to the cases mentioned above as well as to the
present results. A possible explanation for this phenomenon is
the greater metabolic demand on the neurons innervating the
hypertrophic muscles (26). Another factor that probably con289 • OCTOBER 2005 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.5 on June 15, 2017
23) or gastric inflammation (4) and ulcers (5, 6). We investigated only the DRG, but there is evidence that GI injury may
have long-term effects on the central nervous system as well
(24). Therefore, the behavioral changes we observed may be
due to sensitization at the DRG level combined with a central
contribution. The DISCUSSION below is an attempt to integrate
the morphological, physiological, and behavioral observations.
Dye coupling among SGCs. In control ganglia, dye coupling
was rather common (22.5%) among SGCs ensheathing a given
neuron and rare among SGCs forming envelopes around different neurons. After obstruction, coupling between glial envelopes increased about 4-fold when the cells were injected
randomly and 18-fold when the injected SGCs were close to
neurons projecting to the distal colon. These changes were
observed only in DRG L1 and S1, which were found by us and
others (30) to project to the colon. These observations indicate
that the cellular changes in DRG are related to events occurring
in the obstructed colon.
CHANGES IN SENSORY GANGLIA AFTER COLONIC OBSTRUCTION
AJP-Gastrointest Liver Physiol • VOL
release of inflammatory mediators and/or neurotrophins. Because the release of these agents appears to be a general
phenomenon in inflamed visceral organs in both humans (12)
and animals (6, 34), it can be proposed that retrograde transport
of these substances into DRG contributes to the hypertrophy of
DRG neurons. Thus both the presence of inflammation and the
thickening of the colonic wall that resulted from partial colonic
obstruction may have contributed to the neuronal hypertrophy
reported here. We propose that colonic inflammation served as
a major trigger for the other observed changes in the DRG
cells. This hypothesis is supported by our recent experiments
(18), which showed that chemically induced colonic inflammation led to similar effects as partial colonic obstruction.
Partial colonic obstruction and abdominal pain. Ectopic
firing of sensory neurons was proposed as an important contributing factor in neuropathic pain (11). Similarly, the augmented excitability and spontaneous electrical activity observed in sensory neurons innervating the obstructed colon
might contribute to visceral pain. GI inflammation (3, 4, 23)
augmented neuronal excitability, which was proposed to contribute to abdominal pain. Previous studies did not consider the
role of SGCs in visceral sensation, but our observations on the
increased coupling among these cells indicate that inflammation-induced changes in SGCs should be taken into account
along with altered neuronal excitability when neuropathic pain
is discussed. It can be hypothesized that glial changes are
involved in events leading to neuronal sensitization. Because
this coupling is mediated by gap junctions, it can be proposed
that drugs that block gap junctions may have a therapeutic
potential as analgesics for visceral pain, especially as there is
recent progress in developing such drugs (33). Furthermore,
SGCs were found to express a variety of receptors (19), which
may serve as drug targets; we have shown that SGCs possess
receptors for ATP (41), which has a role in pain signaling (16).
This and other mechanisms are currently under research in our
laboratory.
ACKNOWLEGMENTS
We thank Dr. Naomi Melamed-Book for helping with the confocal imaging.
GRANTS
This work was supported by the Israel Science Foundation, by United
States-Israel Binational Science Foundation Grant 2003262, and by the Hebrew University Center for Pain Research.
REFERENCES
1. Amir R, Michaelis M, and Devor M. Membrane potential oscillations in
dorsal root ganglion neurons: role in normal electrogenesis and neuropathic pain. J Neurosci 19: 8589 – 8596, 1999.
2. Amir R, Michaelis M, and Devor M. Burst discharge in primary sensory
neurons: triggered by subthreshold oscillations maintained by depolarizing
afterpotentials. J Neurosci 22: 1187–1198, 2002.
3. Beyak MJ, Ramji N, Krol KM, Kawaja MD, and Vanner SJ. Two
TTX-resistant Na⫹ currents in mouse colonic dorsal root ganglia neurons
and their role in colitis-induced hyperexcitability. Am J Physiol Gastrointest Liver Physiol 287: G845–G855, 2004.
4. Bielefeldt K, Ozaki N, and Gebhart GF. Mild gastritis alters voltagesensitive sodium currents in gastric sensory neurons in rats. Gastroenterology 122: 752–761, 2002.
5. Bielefeldt K, Ozaki N, and Gebhart GF. Experimental ulcers alter
voltage-sensitive sodium currents in rat gastric sensory neurons. Gastroenterology 122: 394 – 405, 2002.
6. Bielefeldt K, Ozaki N, and Gebhart GF. Role of nerve growth factor in
modulation of gastric afferent neurons in the rat. Am J Physiol Gastrointest Liver Physiol 284: G499 –G507, 2003.
289 • OCTOBER 2005 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.5 on June 15, 2017
tributed to the neuronal changes is the inflammation in the
colon that was associated with the partial obstruction.
We believe that this is the first study of changes induced by
obstruction on the physiological properties of DRG neurons.
Using intracellular recordings, we showed that the threshold
for firing a spike was lower in neurons in DRG L1 and S1 after
obstruction. Also, the number of neurons displaying spontaneous subthreshold potential oscillations and the number of
neurons firing spontaneous action potentials were significantly
higher in the obstructed animals. All these effects are consistent with increased excitability of the neurons. Augmented
excitability of sensory neurons was observed after inflammation in the stomach (4), ileum (23), and colon (3). These reports
provided evidence that increased expression of TTX-resistant
Na⫹ channels may underlie this effect. Similar physiological
changes were observed in DRG after axotomy of the sciatic
nerve (44), and Amir et al. (1) proposed that the excitability
changes were associated with an increase in the amplitude of
spontaneous potential oscillations. They also found that the
amplitude of these oscillations was enhanced by depolarization
(2), which may be correlated with our observation that in the
obstructed animals DRG neurons were depolarized by an
average of 4 – 6 mV compared with controls.
The proportion of DRG neurons affected by obstruction is
very likely to be greater than the number of DRG neurons
projecting to the distal colon. There is no accurate information
on the number of neurons innervating the colon, but an estimate can be obtained from retrograde labeling studies. In the
rat colon, 1–1.2% of DRG S1 neurons were found to innervate
the colon (35, 36), and the numbers are expected to be similar
in the mouse colon. We found in both dye coupling and
electrical recordings that over 20% of the DRG neurons and
SGCs were affected by the obstruction, indicating that the
signals coming from the affected colon spread in the ganglion.
This implies that the affected neurons influence a much larger
population of neurons in the ganglion. As found in the dye
coupling experiments, glial cells were also affected by the
colonic obstruction. This includes not only SGCs around affected neurons but also glia around many of the unaffected
ones, again indicating the presence of long-range influences.
The mechanisms underlying these cellular changes are still
unknown. We propose that augmented glial coupling has an
important role in the spread of the augmented neuronal excitability. Evidence for long-range effects within sensory ganglia
after injury has been obtained by several authors (21, 29, 32).
The nature of the signals mediating this spread is not known,
but it can be suggested that chemical messengers released from
the affected neurons travel within the ganglia, thus altering
cellular properties. Possible candidates can be nitric oxide or
nerve growth factor, which are produced at increased rates
after nerve damage (6, 37, 38). It is established that in neuropathic pain states, there is a spread of the sensation beyond the
injured neuron (11). The present results are consistent with this
concept and provide ideas for possible mechanisms that can
account for it.
What are the mechanisms underlying the cellular changes?
We consider the evidence for inflammatory processes in the
partly obstructed colon as a key observation because it appears
to account for most of the present findings. In a study (43) on
the inflamed urinary bladder, it was found that the respective
DRG neurons were hypertrophied, which was explained by the
G677
G678
CHANGES IN SENSORY GANGLIA AFTER COLONIC OBSTRUCTION
AJP-Gastrointest Liver Physiol • VOL
27. Pannese E, Ledda M, Cherkas PS, Huang TY, and Hanani M. Satellite
cell reactions to axon injury of sensory ganglion neurons: increase in
number of gap junctions and formation of bridges connecting previously
separate perineuronal sheaths. Anat Embryol (Berl) 206: 337–347, 2003.
28. Purves D. Neuronal form and its consequences. In: Body and Brain.
Cambridge, MA: Harvard Univ. Press, 1990.
29. Ramer MS, Kawaja MD, Henderson JT, Roder JC, and Bisby MA.
Glial overexpression of NGF enhances neuropathic pain and adrenergic
sprouting into DRG following chronic sciatic constriction in mice. Neurosci Lett 251: 53–56, 1998.
30. Robinson DR, Mcnaughton PA, Evans ML, and Hicks GA. Characterization of the primary spinal afferent innervation of the mouse colon
using retrograde labeling. Neurogastroenterol Motil 16: 113–124, 2004.
31. Schwarz NT, Kalff JC, Türler A, Speidel N, Grandis JR, Billiar TR,
and Bauer AJ. Selective jejunal manipulation causes postoperative panenteric inflammation and dysmotility. Gastroenterology 126: 159 –169,
2004.
32. Soares HD, Chen SC, and Morgan JI. Differential and prolonged
expression of Fos-lacZ and Jun-lacZ in neurons, glia, and muscle following sciatic nerve damage. Exp Neurol 167: 1–14, 2001.
33. Srinivas M and Spray DC. Closure of gap junction channels by arylaminobenzoates. Mol Pharmacol 63: 1389 –1397, 2003.
34. Steers WD, Kolbeck S, Creedon D, and Tuttle JB. Nerve growth factor
in the urinary bladder of the adult regulates neuronal form and function.
J Clin Invest 88: 1709 –1715, 1991.
35. Su X, Wachtel RE, and Gebhart GF. Inhibition of calcium currents in
rat colon sensory neurons by ␬- but not ␮- or ␦-opioids. J Neurophysiol
80: 3112–3119, 1998.
36. Su X, Wachtel RE, and Gebhart GF. Capsaicin sensitivity and voltagegated sodium currents in colon sensory neurons from dorsal root ganglia.
Am J Physiol Gastrointest Liver Physiol 277: G1180 –G1188, 1999.
37. Thippeswamy T and Morris R. The roles of nitric oxide in dorsal root
ganglion neurons. Ann NY Acad Sci 962: 103–110, 2002.
38. Vizzard MA, Erdman SL, and de Groat WC. Increased expression of
neuronal nitric oxide synthase (NOS) in visceral neurons after nerve
injury. J Neurosci 15: 4033– 4045, 1995.
39. Wall PD and Devor M. Sensory afferent impulse originate from dorsal
root ganglia as well as from the periphery in normal and nerve injury rats.
Pain 17: 321–339, 1983.
40. Watkins LR and Maier SF. Beyond neurons: evidence that immune and
glial cells contribute to pathological pain states. Physiol Rev 82: 981–
1011, 2002.
41. Weick M, Cherkas PS, Härtig W, Pannicke T, Uckermann O, Bringmann A, Tal M, Reichenbach A, and Hanani M. P2 receptors in satellite
glial cells in trigeminal ganglia of mice. Neuroscience 120: 969 –977,
2003.
42. Williams TH, Zhang MQ, and Jew JY. Hypertrophy of rat sensory
ganglion neurons following intestinal obstruction. Gastroenterology 105:
8 –14, 1993.
43. Yoshimura N and de Groat WC. Increased excitability of afferent
neurons innervating rat urinary bladder after chronic bladder inflammation. J Neurosci 19: 4644 – 4653, 1999.
44. Zhang JM, Donnelly DF, Song XJ, and LaMotte RH. Axotomy increases the excitability of dorsal root ganglion cells with unmyelinted
axons. J Neurophysiol 78: 2790 –2794, 1997.
289 • OCTOBER 2005 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.5 on June 15, 2017
7. Brehmer A, Gobel D, Frieser M, Graf M, Radespirl-Troger M, and
Neuhuber W. Experimental hypertrophy of myenteric neurons in the pig:
morphometric study. Neurogastroenterol Motil 12: 155–162, 2000.
8. Cabanes C, López de Armentia M, Viana F, and Belmonte C. Postnatal
changes in membrane properties of mice trigeminal ganglion neurons.
J Neurophysiol 87: 2398 –2407, 2002.
9. Cherkas PS, Huang TY, Pannicke T, Tal M, Reichenbach A, and
Hanani M. The effects of axotomy on neurons and satellite glial cells in
mouse trigeminal ganglion. Pain 110: 290 –298, 2004.
10. Devor M, Govrin-Lippmann R, Frank I, and Raber P. Proliferation of
primary sensory neurons in adult rat dorsal root ganglion and the kinetics
of retrograde cell loss after sciatic nerve section. Somatosensory Res 3:
139 –167, 1985.
11. Devor M and Seltzer Z. The pathophysiology of damaged nerves in
relation to chronic pain. In: Textbook of Pain, edited by Wall PD and
Melzack R. London: Churchill-Livingston, 1999, p.129 –164.
12. Di Mola FF, Friess H, Zhu ZW, Koliopnos A, Bley T, Di Sebastiano P,
Innocenti P, Zimmermann A, and Buchler MW. Nerve growth factor
and Trk high affinity receptor (TrkA) gene expression in inflammatory
bowel desease. Gut 46: 670 – 679, 2000.
13. Fulton BP. Gap junctions in the developing nervous system. Perspect Dev
Neurobiol 2: 327–334, 1995.
14. Gabella G. Hypertrophy of visceral smooth muscle. Anat Embryol (Berl)
182: 409 – 424, 1990.
15. Gabella G, Berggren T, and Uvelius B. Hypertrophy and reversal of
hypertrophy in rat pelvic ganglion neurons. J Neurocytol 21: 649 – 622,
1992.
16. Hamilton SG and McMahon SB. ATP as a peripheral mediator of pain.
J Auton Nerv Syst 81: 187–194, 2000.
17. Hanani M, Huang TY, Cherkas PS, Ledda M, and Pannese E. Glial
cell plasticity in sensory ganglia induced by nerve damage. Neuroscience
114: 279 –283, 2002.
18. Hanani M and Huang TY. Contribution of gap junctions in sensory
ganglia to inflammatory pain (Abstract). FENS Abstr 2: A121.7, 2004.
19. Hanani M. Satellite glial cells in sensory ganglia: from form to
function. Brain Res Rev 48: 457– 476, 2005.
20. Huang TY, Cherkas PS, Rosenthal DW, and Hanani M. Dye coupling
among satellite glial cells in mammalian dorsal root ganglia. Brain Res
1036: 42– 49, 2005.
21. Hudson LJ, Bevan S, Wotherspoon G, Gentry C, Fox A, and Winter
J. VR1 protein expression increases in undamaged DRG neurons after
partial nerve injury. Eur J Neurosci 13: 2105–2114, 2001.
22. Mathews GA and Ffrench-Constant C. Embryonic fibronectins are
up-regulated following peripheral nerve injury in rats. J Neurobiol 26:
171–188, 1995.
23. Moore BA, Stewart TMR, Hill C, and Vanner SJ. TNBS ileitis evokes
hyperexcitability and changes in ionic membrane properties of nociceptive
DRG neurons. Am J Physiol Gastrointest Liver Physiol 282: G1045–
G1051, 2002.
24. Ness TJ and Gebhart GF. Inflammation enhances reflex and spinal
neuron responses to noxious visceral stimulation on rats. Am J Physiol
Gastrointest Liver Physiol 280: G649 –G657, 2001.
25. Newman EA. Glial cell regulation of extracellular potassium. In: Neuroglia, edited by Kettenmann H and Ransom BR. New York: Oxford Univ.
Press, 1995, p. 717–731.
26. Pannese E. The satellite cells of sensory ganglia. Adv Anat Embryol Cell
Biol 65: 1–111, 1981.