- Email: [email protected]
Interstitial cells of Cajal mediate enteric inhibitory neurotransmission in the lower esophageal and pyloric sphincters
GASTROENTEROLOGY 1998;115:314–329
Interstitial Cells of Cajal Mediate Enteric Inhibitory Neurotransmission in the Lower Esophageal and Pyloric Sphincters SEAN M. WARD, GERARD MORRIS, LEE REESE, XUAN–YU WANG, and KENTON M. SANDERS Department of Physiology and Cell Biology, University of Nevada School of Medicine, Reno, Nevada
Background & Aims: Previous studies have suggested that a specific class of interstitial cells of Cajal (ICC) act as mediators in nitrergic inhibitory neurotransmission. The aim of this investigation was to examine the role of intramuscular ICC (IC-IM) in neurotransmission in the murine lower esophageal (LES) and pyloric sphincters (PS). Methods: Immunohistochemistry and electrophysiology were used to study the distribution and role of IC-IM. Results: The LES and PS contain spindle-shaped IC-IM, which form close relationships with nitric oxide synthase–containing nerve fibers. The PS contains ICC within the myenteric plexus and c-Kit immunopositive cells along the submucosal surface of the circular muscle. IC-IM were absent in the LES and PS of c-kit (W/W v) mutant mice. Using these mutants, we tested whether IC-IM mediate neural inputs in the LES and PS. Although the distribution of inhibitory nerves was normal in W/W v animals, NO-dependent inhibitory neurotransmission was reduced. Hyperpolarizations to sodium nitroprusside were also attenuated in W/W v animals. Conclusions: The data suggest that IC-IM play an important role in NO-dependent neurotransmission in the LES and PS. IC-IM may be the effectors that transduce NO signals into hyperpolarizing responses. Loss of IC-IM may interfere with relaxations and normal motility in these sphincters.
significant role of the lower esophageal sphincter (LES) and pyloric sphincter is to compartmentalize the gastric contents, which, due to their strong acidic nature, can be damaging to the contiguous organs. The LES stays tonically contracted until it reflexly opens after a swallow to allow ingested materials to pass from the esophagus into the stomach.1 The pylorus, although partially open between gastric peristaltic events, contracts as the gastric peristaltic wave reaches the terminal antrum, causing retention of most of the contents until particle size can be reduced. Feedback mechanisms sense the composition of chyme reaching the duodenum and regulate the motility of the antroduodenal canal to adjust the rate of gastric emptying.2 Control of emptying allows orderly delivery of the gastric contents to the small
A
intestine for digestion and absorption. Relaxation, and therefore control of luminal diameters and resistances, of both sphincters are regulated, in part by nonadrenergic, noncholinergic (enteric inhibitory) motor neurons.1,2 Neurons expressing nitric oxide synthase (NOS) are abundant in sphincteric regions of the gastrointestinal (GI) tract with up to 50% of myenteric neurons expressing NOS or reduced nicotinamide adenine dinucleotide phosphate (NADPH) diaphorase, a common histochemical label for NOS.3–9 A significant portion of NOSpositive neurons are motor neurons that innervate the musculature, and varicose fibers positive for NOS can be observed in whole-mount specimens of these tissues. Sphincteric regions of the GI tract, including the LES and pylorus, display prominent NO-dependent hyperpolarization of resting membrane potential and relaxation in response to activation of enteric inhibitory neurons, suggesting significant regulation by nitrergic mechanisms.10–12 Loss of NOS positive neurons has been associated with a variety of pathological conditions including achalasia and hypertrophic pyloric stenosis.13–15 However, the postjunctional mechanisms that mediate nitregic inhibitory responses are not fully understood. Enteric inhibitory neurotransmission has typically been conceptualized as direct regulation of GI muscles by nonadrenergic, noncholinergic (NANC) transmitter(s) released from nerve terminals. However, a century ago Cajal16 proposed an intermediary in autonomic neurotransmission involving interstitial cells (now referred to as interstitial cells of Cajal [ICC]). Recently, we have provided evidence that NO-dependent neurotransmission in the murine gastric fundus requires ICC for Abbreviations used in this paper: c-Kit–LI, c-Kit–like immunoreactivity; EFS, electrical field stimulation; FITC, fluorescein isothiocyanate; GI, gastrointestinal; ICC, interstitial cells of Cajal; IC-IM, intramuscular ICC; KRB, Krebs–Ringer’s buffer; L-NAME, N G-nitro-Larginine methyl ester; NADPH, reduced nicotinamide adenine dinucleotide phosphate; NANC, nonadrenergic, noncholinergic; NOS, nitric oxide synthase; SNP, sodium nitroprusside. r 1998 by the American Gastroenterological Association 0016-5085/98/$3.00
August 1998
ICC AND INHIBITORY NEUROTRANSMISSION
315
Figure 1. Isolation of muscles of the pyloric sphincter and LES. (A) A stomach with segments of the esophagus (E) and duodenum (D) attached. (B) A stomach opened along the lesser curvature from the gastric antrum (GA) through the duodenum segment. (C) The same preparation after removal of the mucosa. The pyloric sphincter (PS) is clearly visible as a thickened band of circular muscle at the junction between the stomach and duodenum. (D) A stomach opened along the greater curvature to reveal the junction between the esophagus and the stomach. (E ) The same preparation after opening the gastroesophageal junction through the esophagus. (F ) The same preparation after removal of the mucosa. The LES is visible as a band of thickened musculature. F, fundus.
hyperpolarization and relaxation responses to field stimulation of enteric neurons.17 In the present study we used c-kit mutant mice in which the intramuscular class of ICC fails to develop to investigate the role of ICC in the LES and pyloric sphincter. We have characterized the types of ICC present in normal mice and determined the
consequences of losing specific ICC on enteric inhibitory neural control.
Materials and Methods Wild-type (1/1; black coats) and W/Wv (pure white coats) mice,18 between the ages of 20 and 30 days postpartum,
316
WARD ET AL.
GASTROENTEROLOGY Vol. 115, No. 2
Figure 2. Distribution of ICC in the murine LES. (A ) A cryostat cross section through the circular (cm) and longitudinal (lm) muscle layers (section cut along the long axis of the longitudinal layer) of a wild-type (1/1) mouse. IC-IM are observed in both layers (arrows). Circular cells appear as points and longitudinal IC-IM as elongated cells with ovoid nuclear region and long, unbranching processes. No immunoreactivity was observed in the submucosal region(s). (B ) A confocal reconstruction of the entire tunica muscularis from a flattened whole mount. This shows the relative density of the IC-IM in the circular and longitudinal muscle layers. IC-IM were completely absent in W/W v animals. (C ) A cryostat section and (D ) a flattened whole-mount preparation from W/W v animals.
were anesthesized by chloroform inhalation and exsanguinated by cervical dislocation followed by decapitation. The entire stomach, including portions of the esophagus and duodenum (Figure 1A), was removed and placed in Krebs–Ringer’s buffer (KRB). To isolate the pyloric sphincter, stomachs were opened along the lesser curvature from the terminal antrum through the duodenum segment (Figure 1B). The mucosa was removed, revealing the underlying thickened band of circular muscle that constitutes the pyloric sphincter at the junction between the stomach and duodenum (Figure 1C). To isolate the LES, stomachs were opened along the greater curvature to reveal the junction between the esophagus and the stomach (Figure 1D). From the gastric aspect of the gastroesophageal junction, the esophagus was opened by sharp dissection (Figure 1E). The mucosa was removed revealing the thickened musculature of the LES (Figure 1F). Histological examination showed that the musculature immediately proximal to the LES was composed of skeletal muscle fibers (data not shown).
Morphological Studies Immunohistochemical studies were performed on tissues that were dehydrated in graded sucrose solutions, embedded in Tissue Tek (Miles, IL), and frozen in liquid nitrogen.
Cryostat sections were cut at 10-µm thickness, fixed in acetone, and preincubated in goat nonimmune serum for 1 hour (10% in phosphate buffered saline [PBS]) before being incubated with a monoclonal antibody against c-kit protein (ACK219; 5 µg/mL in PBS) at 4°C overnight. Immunoreactivity was detected with fluorescein isothiocyanate (FITC)-conjugated secondary antibody (goat anti-rat 1:100, 1 hour, room temperature; Vector Laboratories Inc., Burlingame CA). Control tissues were prepared in a similar manner, omitting ACK2 from the incubation solution. Whole-mount tissues were labeled with the ACK2 antibody using the same method as outlined above. The antibody was diluted in PBS (0.1 mol/L) containing 0.3% Triton X-100. For double immunostaining, tissues were first incubated with anti–c-Kit antibody for 48 hours at 4°C, as described above. This was followed by washing with PBS, and then tissues were incubated with secondary antibody (FITC)-conjugated goat anti-rat immunoglobulin G (IgG). Tissues were washed again before incubation with a second primary antibody to NOS (anti-sheep, 1:800, for 48 hours). After another wash with PBS, the tissues were washed again before being incubated with Texas Red–conjugated anti-sheep IgG raised in rabbit. Secondary antibodies were purchased from Vector Laboratories and
August 1998
ICC AND INHIBITORY NEUROTRANSMISSION
317
Figure 3. Distribution of ICC in the murine pyloric sphincter. (A ) A cryostat cross section featuring the various classes of ICC within the pyloric sphincter. ICC in the myenteric region (IC-MY) are brightly labeled in the plane between the circular (cm) and longitudinal (lm) muscle layers. As in the LES, IC-IM are found in both muscle layers, running parallel with the muscle fibers. We also observed cells with c-Kit–LI along the surface separating the submucosa (s) from the circular muscle, running perpendicular to the circular muscle layer (*). (B ) A confocal image featuring the IC-IM of the circular muscle layer. The cells are long and spindle-shaped and run parallel to the circular muscle fibers. (C ) A whole mount showing IC-IM (arrows) and IC-MY (arrowheads). (D ) A cryostat section revealing the absence of IC-IM and cells with c-Kit–LI at the submucosal surface in W/W v animals. IC-MY were apparently unaffected by the W/W v mutations. (E and F ) Confocal images of whole mounts featuring the circular muscle layer and the region of the myenteric plexus. Note the abundance of IC-MY (arrowheads).
diluted to 1:100 in PBS. Secondary incubations occurred for 1 hour at room temperature. Control tissues were prepared in a similar manner, either omitting primary or secondary antibodies from the incubation solutions. Labeled tissues were examined using a Bio-Rad MRC 600 confocal microscope (Bio-Rad Laboratories, Hercules, CA), with an excitation wavelength appropriate for FITC (496 nm) and Texas Red (592 nm), where appropriate. Confocal micrographs were obtained from digital composites of Z-series scans of 10–15 optical sections through a depth of
5–30 µm. Final images were constructed with Bio-Rad Comos software. NADPH-diaphorase histochemistry was performed on whole mounts fixed with paraformaldehyde, as described previously.19
Physiological Experiments After removing the mucosa, strips of LES and pyloric sphincter muscle (2 3 1 mm) were pinned to the floor of a recording chamber with the mucosal side of the circular muscle
318
WARD ET AL.
GASTROENTEROLOGY Vol. 115, No. 2
Figure 4. Distribution of NOS-containing neurons in the LES and pyloric sphincter as determined by NADPH diaphorase histochemistry. (A, B, D, and E ) Whole-mount images of the myenteric plexus from 1/1 and W/W v animals at different levels of resolution. We could detect no differences between the number, size, or distribution of cell bodies within the myenteric plexus. (C and F ) Varicose fibers within the circular muscle layer. No difference in the density of innervation was detected. (G, H, J, and K ) Whole mounts of the pyloric sphincters of 1/1 and W/W v animals. (I–L) Varicose fibers within the circular muscle layer. No differences in the density of innervation of this region were detected in 1/1 and W/W v animals. Summaries of neuron and varicose fiber densities can be found in Table 1.
facing upward. Parallel platinum electrodes were placed on either side of the muscle strips. Circular muscle cells were impaled with glass microelectrodes, and transmembrane potentials were measured (World Precision Instruments, Sarasota, FL [WPI] S-7071). Electrical signals were recorded on magnetic tape (Racal 40S; Southampton, England). Neural responses were elicited by square wave pulses (0.5-millisecond duration,
supramaximal voltage; S48 Grass stimulator; Grass Medical Instruments, Quincy, MA), of electrical field stimulation (EFS; Grass S48). Separate mechanical experiments were performed using standard organ bath techniques. Strips of muscle (approximately 0.5 3 1.0 mm) were isolated from the LES and pyloric sphincters. The mucosa was removed by sharp dissec-
August 1998
ICC AND INHIBITORY NEUROTRANSMISSION
319
Table 1. Distribution of NADPH-Diaphorase–Positive Nerve Cell Bodies (Myenteric Plexus) and NADPH-Diaphorase–Positive Nerve Processes Running Parallel to the Circular Muscle Layer in the LES and Pylorus of 1/1 and W/W V Mice (1/1) LES
(W/W V ) LES
(1/1) Pylorus
(W/W V ) Pylorus
NADPH-d1 nerve cells 63.8 6 2.1 62.4 6 3.0 61.0 6 1.5 66.1 6 1.5 NADPH-d1 nerve trunks 14.0 6 0.8 15.2 6 0.6 13.6 6 1.5 15.0 6 0.7 NOTE. Numbers of cell bodies reflect their density per 1000 µm2. Numbers of nerve trunks were obtained from the number transecting an arbitrary 100-µm line. All analyses are from 20 fields of view from 4 wild-type and 4 W/W V mutant animals. NADPH-d1, NADPH-diaphorase positive.
tion, and muscles were attached to a fixed mount and to a Fort 10 isometric strain gauge (World Precision Instruments). The muscles were immersed in organ baths maintained at 37 6 0.5°C with oxygenated KRB. A resting force of 200 mg was applied, which was shown to set the muscles at optimum length (data not shown). This was followed by an equilibration period of 1 hour, during which time the bath was continuously perfused with oxygenated KRB. Signals were recorded onto a chart recorder (Gould RS 3600; Gould, Cleveland, OH). Data are expressed as means 6 SEM from 1/1 (wild-type) and W/Wv (mutants). Differences in the data were evaluated by Student’s t test. P values of ,0.05 were taken as a statistically significant difference. The n values reported in the text refer to the number of animals used for each protocol, and only one muscle of a given region could be prepared per animal. A total of 86 1/1 and 39 W/Wv animals were used for physiological experiments and 12 1/1 and 12 W/Wv animals used for morphological studies.
Solutions and Drugs Muscles were maintained in KRB (37.5 6 0.5°C; pH 7.3–7.4) containing (in mmol/L): Na1, 137.4; K1, 5.9; Ca21, 2.5; Mg, 1.2; Cl2, 134; HCO32, 15.5; H2PO42, 1.2; dextrose, 11.5, and bubbled with 97% O2–3% CO2. Solutions of BayK 8644 (RBI, Natick, MA) were prepared in ethanol at 1022 mol/L, and were diluted to 1026 mol/L in KRB. Atropine sulfate, guanethidine sulfate, sodium nitroprusside (SNP), and L-nitroarginine or NG-nitro-L-arginine methyl ester (L-NAME; Sigma Chemical Co., St. Louis, MO) were dissolved in distilled water at 1021 to 1022 mol/L and diluted in KRB to the stated final concentrations.
Results c-Kit–Positive Cells in the LES and Pyloric Sphincter and the Consequences of W Mutations on c-Kit–Expressing Cells Antibodies to c-Kit were used to determine the distribution of cells with c-Kit–like immunoreactivity (c-Kit–LI). In six LES examined, cells with c-Kit–LI were
Figure 5. Confocal micrographs of NOS-like immunoreactivity in the (A ) LES and (B ) pyloric sphincters. Whole mounts of the LES and pyloric sphincter fixed with formaldehyde showed NOS-like immunoreactivity in varicose nerve fibers (arrowheads) and cells with profiles suggestive of IC-IM (arrows). The varicose fibers were in close proximity to the IC-IM.
distributed within the circular and longitudinal muscle layers (Figure 2A). Cryosections of muscles from 3 animals showed a relatively uniform distribution throughout the thickness of both muscle layers. Cells with c-Kit–LI were spindle-shaped, with few or no lateral processes (Figure 2B). Cells with this morphology in the murine fundus are intramuscular interstitial cells of Cajal (IC-IM; Burns et al.17). Cells with c-Kit–LI were not found within the myenteric plexus region of the LES, and we observed no cells with multiple processes forming networks suggestive of myenteric ICC. This observation is also similar to the distribution of ICC in the murine fundus (Burns et al.17). In contrast to wild-type animals, cells with c-Kit–LI were absent from the circular and longitudinal muscle layers of the LES of 6 W/Wv animals examined (Figure 2C and D). Cells with c-Kit–LI were distributed in a different manner within the pyloric sphincter. c-Kit–positive cells were observed in several locations in the 6 wild-type mice examined (Figure 3A–C): (1) spindle-shaped cells were
320
WARD ET AL.
GASTROENTEROLOGY Vol. 115, No. 2
Figure 6. Double-labeling of cells with c-Kit–LI (IC-IM) and NOS-like immunoreactivity. Whole mounts of the (A–C ) LES and (D–F ) pyloric sphincter fixed with acetone showed clear labeling of IC-IM with both antibodies. A and D show immunoreactivity to NOS (FITC), and B and E show c-Kit–LI (Texas red). C and F show colocalization of both fluorophores (yellow) in IC-IM (arrows). The NOS immunoreactivity was reduced in varicose nerve fibers in acetone-fixed preparations. Scale bars apply to all figures.
found within the circular and longitudinal muscle layers (IC-IM); (2) a dense network of c-Kit–positive cells was found in the region of the myenteric plexus; and (3) occasional cells with c-Kit–LI were observed along the submucosal surface of the circular muscle layer. The latter had large oval nuclei and branching main processes extending perpendicular to the main axis of the circular muscle fibers (Figure 3A). This type of c-Kit–positive cell has not been previously described, and nothing is known about the function of these cells. IC-IM and cells with c-Kit–LI along the submucosal surface of the circular muscle layer were absent in muscles of 6
W/Wv mice examined, but the network of the myenteric plexus was unaffected by the W/Wv mutations (Figure 3D–F). Relationship Between Inhibitory Nerve Fibers and IC-IM in Murine LES and Pyloric Sphincter IC-IM have previously been shown to be a critical element in nitrergic neurotransmission.17 Therefore, we investigated the innervation of the tunica muscularis of the LES and pyloric sphincter by NOS-containing neurons and studied the relationship between NOS neurons
August 1998
ICC AND INHIBITORY NEUROTRANSMISSION
321
and IC-IM. The overall distributions of NOS neurons and varicose processes were characterized in the LES and pyloric sphincters of wild-type and W/Wv mutants using the NADPH-diaphorase technique, which labels NOS neurons in formaldehyde-fixed muscles.19 Figure 4 shows micrographs from whole mounts of the LES and pyloric sphincter. We could detect no differences in the number of NOS neurons or the distribution of diaphorase-positive varicose fibers in wild-type or mutant animals in tissues of either the LES or pyloric sphincter (Table 1). Previous ultrastructural studies have shown a close relationship between NOS-positive nerve terminals and ICC.20 We examined the distribution of NOS with immunohistochemistry in the LES and pylorus in paraformaldehyde-fixed muscles and found NOS-like immunoreactivity in varicose nerve fibers surrounding IC-IM (Figure 5). The latter point was further explored in double-labeling studies in which antibodies to c-Kit and neuronal NOS were used (see Materials and Methods). These studies were performed in acetone-fixed muscles because we have observed poor labeling of IC-IM with the ACK2 antibody in tissues fixed with formaldehyde. These experiments clearly showed that IC-IM (identified as c-Kit–positive cells) contained NOS-like immunoreactivity (Figure 6). Inhibitory Neurotransmission in Wild-type and W/W v Mutants Strips of LES muscles were stimulated with EFS in the presence of atropine and guanethidine (both 1 µmol/L) to reveal NANC inhibitory responses. In mechanical experiments, the tone of the muscle strips was increased by inclusion of BayK 8644 (1 µmol/L) to facilitate resolution of relaxation responses. EFS (1–20 Hz for 1 minute) caused frequency-dependent relaxation of LES strips (Figure 7A). After cessation of the stimulus, a pronounced rebound contraction was observed. These b Figure 7. Mechanical responses to EFS of the LES. (A ) Responses to 5- and 10-Hz stimulation (1-minute trains; 0.5-millisecond pulses; black bars). Responses under NANC conditions (see Materials and Methods) were characterized by a relaxation of the muscle during the period of stimulation and a rebound contraction upon cessation of the stimulation. L-NAME (200 µmol/L) blocked the relaxation phase and rebound contractions and converted the response during the stimulation at 10 Hz to a contraction. (C ) Responses to EFS of a strip from a W/W v animal. In all cases, EFS elicited a contractile response during the stimulation and the poststimulus rebound response did not occur. (D ) L-NAME had little or no effect in W/W v animals. (E ) A graph showing a summary of experiments in wild-type (n 5 5) and W/W v animals (n 5 5). Frequency-dependent inhibitory responses during stimulation of wild-type muscles (d) were decreased by L-NAME (s). The response during stimulation of muscles from W/W v animals was a frequency-dependent contraction during the period of stimulation (j), which was not significantly affected by L-NAME (h).
322
WARD ET AL.
GASTROENTEROLOGY Vol. 115, No. 2
responses are similar to the NANC responses reported for the LES of other species. Addition of L-NAME (200 µmol/L), inhibited the relaxation response during the stimulus, converting this response to a contraction, and blocked the poststimulation rebound response (Figure 7B). LES muscles of W/Wv animals were also stimulated with EFS using the same conditions and parameters. The responses of these muscles were considerably different from the control animals. EFS of muscles from W/Wv animals failed to produce relaxation or rebound contractions (Figure 7C). Addition of L-NAME either had no effect or produced a slight enhancement of the contractile responses during stimulation (Figure 7D). The increase failed to reach a level of significance in the series of experiments performed. These data suggest that nitrergic neurotransmission was reduced in the LES of W/Wv animals. These experiments are summarized in Figure 7E. The same protocols used for mechanical experiments on the LES were used on muscles from the pyloric sphincter of wild-type and W/Wv animals. As in the LES, EFS resulted in relaxation during the period of stimulation, and a poststimulus rebound contraction was observed upon cessation of the stimulus (Figure 8A). L-NAME (200 µmol/L) blocked both responses (Figure 8B). The relaxation phase and rebound contraction phase of the response to EFS were absent in W/Wv animals (Figure 8C), and L-NAME (200 µmol/L) produced a slight enhancement of the contractile responses during EFS (Figure 8D; P , 0.05). Responses of the pyloric sphincter are summarized in Figure 8E. To more closely examine nitrergic neurotransmission, a more detailed series of electrophysiological experiments was conducted on circular muscles of wild-type and mutant animals. LES cells had average resting potentials b Figure 8. Mechanical responses of the pyloric sphincter to EFS. (A ) Responses to 5- and 10-Hz stimulation (1-minute trains; 0.5millisecond pulses; black bars). (B ) Responses under NANC conditions (see Materials and Methods) were characterized by a relaxation of the muscle during the period of stimulation and a rebound contraction on cessation of the stimulation. L-NAME (200 µmol/L) blocked the relaxation phase and rebound contractions. (C ) Responses to EFS of a strip from a W/W v animal. In all cases, EFS elicited a contractile response during the stimulation and the poststimulus rebound response failed to occur. (D ) L-NAME slightly increased the contractions of muscles of W/W v animals during stimulation. (E ) A graph showing a summary of experiments in wild-type (n 5 8) and W/W v animals (n 5 5). Frequency-dependent inhibitory responses during stimulation of wild-type muscles (d) were decreased by L-NAME (s). The response during stimulation of muscles from W/W v animals was a frequency-dependent contraction during the period of stimulation (j), which was slightly increased by L-NAME (h).
August 1998
ICC AND INHIBITORY NEUROTRANSMISSION
323
Figure 9. Changes in membrane potential elicited by EFS of LES muscles. (A ) Responses to 1–20 Hz stimulation (1-second trains at arrows). The inhibitory junction potentials were characterized by a rapid hyperpolarization followed by a slow hyperpolarization phase. (B ) L-NAME (200 µmol/L) greatly reduced both phases of the inhibitory junction potential. Recordings in A and B were made from the same cell. (C ) In W/W v animals, inhibitory junction potentials were smaller in amplitude and consisted only of the rapid phase of hyperpolarization. (D ) These events were not significantly affected by L-NAME. Recordings in C and D were made from the same cell.
of 257 6 0.6 mV (n 5 24 animals) in the presence of atropine and guanethidine (both 1 µmol/L). Resting potential was not stable and was characterized by rapid, small-amplitude fluctuations (Figure 9A). EFS (1–20 Hz; 1-second trains) produced compound inhibitory junction potentials consisting of a rapid phase of hyperpolarization, followed by a second, slow phase of hyperpolarization (Figure 9A; most evident at 10 and 20 Hz). Addition of L-NAME (200 µmol/L) did not significantly alter resting membrane potential (256 6 0.7 mV; n 5 24 animals; P . 0.05), but it inhibited both phases of the inhibitory junction potential (Figure 9B), suggesting that the hyperpolarization responses strongly depended on the synthesis of NO. As in the case of mechanical responses, the electrical response to EFS was considerably different in the LES of W/Wv animals. Cells of W/Wv animals had average resting potentials of 259 6 0.8 mV (n 5 18 animals; resting potential not significantly differ-
ent from wild-type animals, P . 0.2), and resting membrane potential was always more stable in these tissues (Figure 9C). EFS (1–20 Hz; 1-second trains), caused inhibitory junction potentials of smaller amplitude and duration (Figure 9C). The slow, second phase of the inhibitory junction potentials were absent in the LES of W/Wv animals. The amplitude and duration of inhibitory junction potentials in W/Wv animals were approximately the same levels as the amplitude of inhibitory junction potentials in 1/1 animals after addition of L-NAME. Addition of L-NAME to W/Wv tissues caused no significant change in resting potential (i.e., 257 6 0.8 mV; P . 0.15) and did not change the amplitude or duration of inhibitory junction potentials (both P . 0.4; Figure 9D). Figure 10 shows a summary of these experiments performed on muscles of 24 1/1 and 18 W/Wv animals. Circular muscle cells of the pyloric sphincter had average resting potentials of 257 6 0.7 mV (n 5 49) in
324
WARD ET AL.
GASTROENTEROLOGY Vol. 115, No. 2
Figure 10. Summaries of electrical responses to EFS of wild-type and W/W v LES muscles. (A ) Amplitudes (fast component) of frequencydependent inhibitory junction potentials (IJPs) in wild-type animals before (d) and after (s) L-NAME. (B ) IJP duration in wild-type animals before (d) and after (s) L-NAME. The amplitudes and durations of IJPs significantly decreased at all frequencies (96 impalements from 24 animals). In W/W v animals (54 impalements from 18 animals), (C ) amplitudes and (D ) durations of IJPs were reduced (d) and unaffected (s) by L-NAME.
the presence of atropine and guanethidine (both 1 µmol/L). EFS (1–20 Hz; 1-second trains) produced compound junction potentials consisting of a rapid phase of hyperpolarization, followed by a rebound depolarization (Figure 11A). Addition of L-NAME (200 µmol/L) did not affect membrane potential (i.e., 257 6 0.6 mV; n 5 49; P . 0.5), but it reduced inhibitory junction potentials and the poststimulus rebound response, suggesting that both responses depended on the synthesis of NO (Figure 11B; P , 0.05). Circular muscle cells of the pyloric sphincter of W/Wv animals had average resting potentials of 259 6 0.9 mV (not significantly different from wild-type animals, P . 0.3). EFS (1–20 Hz; 1-second trains) caused inhibitory junction potentials of significantly smaller amplitude and duration (Figure 11C) than observed in wildtype animals (both P , 0.05). The amplitude of inhibitory junction potentials in W/Wv animals was approximately at the same levels as the events in wild-type animals in the presence of L-NAME (P . 0.4; analysis of variance). Addition of L-NAME to W/Wv tissues did not affect resting membrane potential (i.e., 257 6 0.9 mV; n 5 26; P . 0.3) and caused no significant change in the amplitude and duration of inhibitory junction potentials (both
P . 0.2; Figure 11D). Data from these experiments are summarized in Figure 12. Postjunctional Responses to SNP The hyperpolarizing responses to EFS, shown to be dependent on NO synthesis, were mimicked by SNP. In wild-type LES muscles, SNP caused a concentrationdependent hyperpolarization response (Figure 13A). For example, 1 µmol/L SNP caused an average 12.76 1.6 mV hyperpolarization (n 5 11). In LES of W/Wv animals, SNP had little effect on membrane potential (Figure 13B); 1 µmol/L SNP produced an average of 1.5 6 0.5 mV of hyperpolarization, which was significantly less than in wild-type animals (n 5 8 animals; P , 0.02). Similar results were observed in pyloric muscles. In wild-type muscles, SNP caused a concentration-dependent hyperpolarization (Figure 13C), averaging 8.5 6 0.7 mV in response to 1 µmol/L SNP (n 5 13 muscles). In the pyloric sphincter of W/Wv animals, the effect of SNP on membrane potential was also reduced (Figure 13D). SNP (1 µmol/L) caused an average hyperpolarization of 3.1 6 0.5 mV (n 5 7; P , 0.001).
August 1998
ICC AND INHIBITORY NEUROTRANSMISSION
325
Figure 11. Changes in membrane potential elicited by EFS of pyloric sphincter muscles. (A ) Responses to 1–20 Hz stimulation (1-second trains at arrows). The inhibitory junction potentials (IJPs) were characterized by a rapid hyperpolarization followed by a rebound depolarization. (B ) L-NAME (200 µmol/L) reduced both phases of the IJP. Recordings in A and B were made from the same cell. (C ) In W/W v animals, IJPs were smaller in amplitude. (D ) These events were not significantly affected by L-NAME. Recordings in C and D were made from the same cell.
Discussion The LES and pyloric sphincter of the mouse have spindle-shaped ICC similar to the gastric fundus.17 Similar cells have also been observed in the esophagus, LES, stomach, cecum, and internal anal sphincter of the guinea pig,21 but the physiological role of this class of ICC in sphincters has not been documented previously. Spindle-shaped ICC lie within the circular and longitudinal muscle layers in close apposition with smooth muscle cells and varicose processes of motor neurons. We have referred to this class of ICC as IC-IM.17,21,22 In the present study we found that IC-IM were missing in the LES and pyloric sphincters of c-kit mutants (W/Wv). By studying tissues of these mutants, we have obtained evidence that IC-IM are important mediators of enteric inhibitory neurotransmission in these sphincters. Daniel and Posey-Daniel23 characterized the relationship between ICC, smooth muscle cells, and nerve varicosities using morphometric techniques on tissues of the opossum esophagus. They found close association between nerve terminals and ICC (20 nm), but few extremely close associations could be found between varicosities and smooth muscle cells. ICC also formed frequent gap junctions with smooth muscle cells. Similar relationships were noted in the murine fundus,17 and the present study shows that IC-IM of the LES and pyloric
sphincter are surrounded by NOS-containing varicose nerve fibers. Thus, there seems to be a structural arrangement between nerve terminals, ICC, and smooth muscle cells that facilitates a role for ICC in neurotransmission. Comparisons of wild-type and W/Wv siblings provided an opportunity to test the role of IC-IM in neuromuscular transmission. Our data suggest that IC-IM facilitate NO-dependent inhibitory neurotransmission in the LES and pyloric sphincter. In the murine gastric fundus we could find no evidence for NO-dependent neurotransmission in the absence of IC-IM. The dependence on IC-IM for nitrergic input, although highly significant in the LES and pyloric sphincter, may not be as complete as in the fundus. There were no relaxation responses to EFS in the LES or pyloric sphincter of W/Wv animals; EFS elicited contractile responses in muscles of these animals. In some cases the contractions elicited by EFS were slightly enhanced by blocking NO synthesis. This indicates that a component of NO-dependent neurotransmission may have remained in W/Wv animals. Thus, in the absence of IC-IM, there may be some direct overflow of NO onto surrounding smooth muscle cells. This suggests parallel innervation of IC-IM and smooth muscle cells by enteric inhibitory neurons, as originally suggested by Daniel and PoseyDaniel23 from their structural analyses of the opossum
326
WARD ET AL.
GASTROENTEROLOGY Vol. 115, No. 2
Figure 12. Summaries of electrical responses to EFS in pyloric sphincters of wild-type and W/W v animals. (A ) Amplitudes of frequency-dependent inhibitory junction potentials (IJPs) in wild-type animals before (d) and after (s) L-NAME. (B ) IJP duration in wild-type animals before (d) and after (s) L-NAME. The amplitudes and durations of IJPs were significantly decreased at all frequencies (P , 0.05). In W/W v animals, the (C ) amplitude and (D ) duration of IJPs were significantly smaller than observed in wild-type animals (d; P , 0.05) and were unaffected by L-NAME (s; P . 0.05). The IJPs in W/W v animals were approximately the same amplitudes and durations as wild-type animals in the presence of L-NAME (P . 0.4; analysis of variance).
esophagus. The degree of influence exerted directly on smooth muscle cells seems to be far less, however, than the effects mediated by IC-IM. Brief periods of stimulation (1 second) resulted in NO-dependent, postjunctional electrical responses that were absent in the LES of W/Wv animals. However, NO-dependent responses to longer periods of stimulation (as used in mechanical experiments) were not totally blocked in W/Wv animals. Longer stimulus periods may result in the production of enough NO to overflow onto nearby smooth muscle and directly elicit mechanical responses in these cells. The relative contribution of nitrergic neurotransmission to regulation of the murine pyloric sphincter seemed to be less than in the LES, and it seemed that a significant portion of enteric inhibitory neurotransmission in the pylorus was mediated by a transmitter other than NO. The effects of this transmitter were evident in W/Wv animals, suggesting direct effects on smooth muscle cells. Future in vivo studies of mutant animals may help to evaluate the importance of IC-IM in mediating the reflex relaxation of the LES after a swallow and in regulating gastric emptying. The loss of the relaxation responses in the LES
and pyloric sphincter strongly suggest that IC-IM are important for normal motor function. Our data suggest that the transduction and/or effector mechanisms that convert NO to a hyperpolarization response in normal tissues may be expressed by IC-IM because the hyperpolarization response to exogenous NO (delivered via SNP) was lost in W/Wv mice. The ionic conductances activated by NO may be mainly expressed by IC-IM, and hyperpolarizing responses to NO normally observed in the LES and pyloric sphincter may be electrotonically conducted from IC-IM to neighboring smooth muscle cells. A complete test of this hypothesis will require comparative voltage-clamp studies on smooth muscle cells and IC-IM from these sphincters. It is important to note that IC-IM of the LES and pyloric sphincter expressed NOS-like immunoreactivity. Previous studies have shown expression of NOS-like immunoreactivity in ICC of the canine colon and rat small intestine,20,24 and the finding that NOS-like immunoreactivity is present in other classes of ICC (deep muscular plexus; IC-DMP) from a different species suggests that this enzyme could be a general feature of
August 1998
ICC AND INHIBITORY NEUROTRANSMISSION
327
Figure 13. Postjunctional responses to SNP. Traces in each panel are superimposed recordings from single cells showing responses to several concentrations of SNP (1028 to 1024 mol/L). The traces in each panel are lined up on the time of SNP application (arrow). SNP was applied at the concentrations noted throughout the period of each recording, and the wash periods between SNP doses were eliminated. (A and B ) Responses in LES muscles of wild-type (1/1) and W/W v animals, respectively. (C and D ) Responses in pyloric muscles wild-type (1/1) and W/W v animals, respectively.
many types of ICC. We have previously suggested that ICC could act as amplifiers of enteric inhibitory neurotransmission because we observed that NO enhanced Ca21 in ICC and enhancement in Ca21 led to production and release of NO by ICC.25 Loss of IC-IM in the LES and pyloric sphincters of W/Wv mice and the amplification of enteric inhibitory neurotransmission provided by these cells could also possibly contribute to the reduction in NO-dependent neurotransmission observed in these animals. Loss of the amplification function hypothesized for ICC, however, does not fully explain the functional defect we observed in W/Wv tissues. Hyperpolarizing responses to exogenous SNP were greatly reduced or abolished in the LES and pyloric muscles of these animals, suggesting that IC-IM may express a critical transduction mechanism (i.e., second messengers or ion channels) that mediates the electrical responses to nitrergic stimulation. All classes of ICC were not lost in W/Wv mice. The spindle-shaped IC-IM were gone, but the multiprocessed cells in the myenteric region of the pyloric canal seemed
to be unaffected in the mutant animals. Similar observations regarding the myenteric region in W/Wv mice were made previously in tissues of the gastric corpus and antrum.17 Others have concluded that c-Kit is not important for the development of some classes of ICC,26 but results with neutralizing antibodies against c-Kit do not support this view. For example, we have found that all classes of ICC are reduced in the colon and small intestine in animals treated with anti–c-Kit antibodies.27 Cells with c-Kit–LI were also missing from the small intestine and pyloric sphincter in another study of similarly treated animals.28 Recent work using organ cultures confirms the observation that treatment with a neutralizing anti–c-Kit antibody early in development or immediately following birth causes loss of all classes of ICC.29 Thus, we suggest that the development of some classes of ICC (e.g., the myenteric region of the intestine and IC-IM of sphincteric regions and stomach) is susceptible to the nontotal lesion in tyrosine kinase activity in W/Wv compound heterozygotes. Other classes of ICC (e.g., the myenteric
328
WARD ET AL.
region of the stomach and colon, the deep muscular plexus of the small intestine, IC-IM in the circular and longitudinal muscle layers of the colon, and the submucosal surface of the colon) are resistant to mutations in c-kit that do not totally abolish the function of the receptor tyrosine kinase. Loss of nitrergic neurons from sphincteric regions has been reported in clinical disorders such as achalasia,14 congenital esophageal stenosis,30 and hypertrophic pyloric stenosis.13,15,31–33 Results of the present study suggest that defective nitrergic neurotransmission could also arise from lesions in pyloric and LES IC-IM. Circumstances involving loss of ICC have been noted in tissues from cases of pyloric stenosis34 and achalasia.35 The parallelism between loss of nitrergic neurons and ICC in these clinical conditions raises the question whether there is a relationship tying the phenotype of ICC to nitrergic motor neurons or vice versa. It is possible, for example, that nitrergic neurons and ICC could express growth factors that regulate the phenotype of the other cell type. It has been reported, for example, that enteric neurons express stem cell factor,36 the natural ligand for c-Kit receptors (expressed by ICC). Expression of mutant forms of stem cell factor, such as in steel mutants (i.e., Sl/Sld) causes lesions in some types of ICC, including the spindle-shaped IC-IM of the fundus and sphincteric regions.37 If tonic stimulation by stem cell factor is important in adults to maintain the phenotype of ICC, then it is possible that loss of ICC could result from damage to nitrergic neurons. The specific defects, genetic or environmental, that cause loss of nitrergic neurons and ICC are unknown. Understanding the factors that regulate the plasticity of ICC and enteric nerve/ICC interactions seems to be a very important new direction for motility research. In summary, IC-IM are distributed within the tunica muscularis of the LES and pyloric sphincters. Like in the murine fundus, cells of this morphology play a significant role in the neuroregulation of these sphincters by mediating nitrergic neurotransmission. Developmental defects that block the formation of mature IC-IM in the LES and pyloric sphincter or loss of these cells from adults might seriously jeopardize the normal functions of these sphincters. It will be interesting to evaluate the status of IC-IM in human disorders associated with sphincter dysfunction.
GASTROENTEROLOGY Vol. 115, No. 2
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16. 17.
18.
19.
20.
References 1. Conklin JL, Christensen J. Motor functions of the pharynx and esophagus. In: Johnson LR, ed. Physiology of the gastrointestinal tract. 3rd ed. New York: Raven, 1994:903–928. 2. Mayer EA. The physiology of gastric storage and emptying. In:
21.
22.
Johnson LR, ed. Physiology of the gastrointestinal tract. 3rd ed. New York: Raven, 1994:929–976. Aimi Y, Kimura H, Kinoshita T, Minami Y, Fujimura M, Vincent SR. Histochemical localization of nitric oxide synthase in rat enteric nervous system. Neuroscience 1993;53:553–560. Furness JB, Li ZS, Young HM, Fo¨rstermann U. Nitric oxide synthase in the enteric nervous system of the guinea pig: a quantitative description. Cell Tissue Res 1994;277:139–149. Ward SM, Xue C, Sanders KM. Localization of nitric oxide synthase immunoreactivity in ileocolonic and pyloric sphincters. Cell Tissue Res 1994;275:513–527. Singaram C, Sengupta A, Sweet MA, Sugarbaker DJ, Goyal RK. Nitrinergic and peptidergic innervation of the human oesophagus. Gut 1994;35:1690–1696. Christensen J, Fang S, Rick GA. NADPH-diaphorase-positive nerve fibers in smooth muscle layers of opossum esophagus: gradients in density. J Auton Nerv Syst 1995;52:99–105. Lynn RB, Sankey SL, Chakder S, Rattan S. Colocalization of NADPH-diaphorase staining and VIP immunoreactivity in neurons in opossum internal anal sphincter. Dig Dis Sci 1995;40:781– 791. Wang XY, Wong WC, Ling EA. Localization of nicotinamide adenine dinucleotide phosphate diaphorase (NADPH-d) activity in the gastrointestinal sphincters in the guinea pig. J Auton Nerv Syst 1996;58:51–55. Conklin JL, Du C, Murray JA, Bates JN. Characterization and mediation of inhibitory junction potentials from opossum lower esophageal sphincter. Gastroenterology 1993;104:1439–1444. Bayguinov O, Sanders KM. Role of nitric oxide as an inhibitory neurotransmitter in the canine pyloric sphincter. Am J Physiol 1993;264:G975–G983. Tottrup A, Ny L, Alm P, Larsson B, Forman A, Andersson KE. The role of the L-arginine/nitric oxide pathway for relaxation of the human lower oesophageal sphincter. Acta Physiol Scand 1993; 149:451–459. Huang PL, Dawson TM, Bredt DS, Snyder SH, Fishman MC. Targeted disruption of the neuronal nitric oxide synthase gene. Cell 1993;75:1273–1286. Mearin F, Mourelle M, Guarner F, Salas A, Riveros-Moreno V, Moncada S, Malagelada JR. Patients with achalasia lack nitric oxide synthase in the gastro-oesophageal junction. Eur J Clin Invest 1993;23:724–728. Vanderwinden JM, Mailleux P, Schiffmann SN, Vanderhaeghen JJ, De Laet MH. Nitric oxide synthase activity in infantile hypertrophic pyloric stenosis. N Engl J Med 1992;327:511–515. Cajal SR. Histologie du syste`me nerveux de l’homme et des verte´bre´s. Volume 2. Paris: Maloine, 1911:891–942. Burns AJ, Lomax AEJ, Torihashi S, Sanders KM, Ward SM. Interstitial cells of Cajal mediate inhibitory neurotransmission in the stomach. Proc Natl Acad Sci USA 1996;93:12008–12013. Nishikawa S, Kusakabe M, Yoshinaga K, Ogawa M, Hayashi S-I, Kunisada T, Era T, Sakakura T, Nishikawa S-I. In utero manipulation of coat color formation by a monoclonal anti–c-kit antibody: two distinct waves of c-kit dependency during melanocyte development. EMBO J 1991;10:2111–2118. Ward SM, Xue C, Shuttleworth CWR, Bredt DS, Snyder SH, Sanders KM. NADPH diaphorase and nitric oxide synthase colocalization in enteric neurons of the canine proximal colon. Am J Physiol 1992;263:G277–G284. Matini P, Faussone-Pellegrini M-S. Ultrastructural localization of neuronal nitric oxide synthase-immunoreactivity in the rat ileum. Neurosci Lett 1997;229:45–48. Burns AJ, Herbert TM, Ward SM, Sanders KM. Interstitial cells of Cajal in the guinea pig gastrointestinal tract as revealed by c-Kit immunohistochemistry. Cell Tissue Res 1997;290:11–20. Sanders KM. A case for interstitial cells of Cajal as pacemakers
August 1998
23.
24.
25.
26.
27.
28.
29.
30.
31.
and mediators of neurotransmission in the gastrointestinal tract. Gastroenterology 1996;111:492–515. Daniel EE, Posey-Daniel V. Neuromuscular structures in opossum esophagus: role of interstitial cells of Cajal. Am J Physiol 1984;246:G305–G315. Xue C, Pollock J, Schmidt HHHW, Ward SM, Sanders KM. Expression of nitric oxide synthase by interstitial cells of the canine proximal colon. J Auton Nerv Syst 1994;49:1–14. Publicover NG, Hammond EM, Sanders KM. Amplification of nitric oxide signalling by interstitial cells. Proc Natl Acad Sci USA 1993;90:2087–2091. Malysz J, Thuneberg L, Mikkelsen HB, Huizinga JD. Action potential generation in the small intestine of W mutant mice that lack interstitial cells of Cajal. Am J Physiol 1996;271:G387– G399. Torihashi S, Ward SM, Nishikawa S-I, Nishi K, Kobayashi S, Sanders KM. c-kit–dependent development of interstitial cells and electrical activity in the murine gastrointestinal tract. Cell Tissue Res 1995;280:97–111. Maeda H, Yamagata A, Nishikawa S, Yoshinaga K, Kobayashi S, Nishi K, Nishikawa S. Requirement of c-kit for development of intestinal pacemaker system. Development 1992;116:369– 375. Ward SM, Harney SH, Bayguinov JR, Sanders KM. Development of electrical rhythmicity in the murine gastrointestinal tract is specifically encoded in the tunica muscularis. J Physiol 1997;505.1:241– 258. Singaram C, Sweet MA, Gaumnitz EA, Cameron AJ, Camilleri M. Peptidergic and nitrinergic denervation in congenital esophageal stenosis. Gastroenterology 1995;109:275–281. Kobayashi H, O’Briain DS, Puri P. Immunochemical characterization of neural cell adhesion molecule (NCAM), nitric oxide synthase, and neurofilament protein expression in pyloric muscle of patients with pyloric stenosis. J Pediatr Gastroenterol Nutr 1995;20:319–325.
ICC AND INHIBITORY NEUROTRANSMISSION
329
32. Abel RM. The ontogeny of the peptide innervation of the human pyloric sphincter, with special reference to understanding the aetiology and pathogenesis of infantile hypertrophic pyloric stenosis. J Pediatr Surg 1996;31:490–497. 33. Lefebvre RA. Nitric oxide in the peripheral nervous system. Ann Med 1995;27:379–388. 34. Vanderwinden JM, Liu H, De Laet MH, Vanderhaeghen JJ. Study of the interstitial cells of Cajal in infantile hypertrophic pyloric stenosis. Gastroenterology 1996;111:279–288. 35. Faussone-Pellegrini MS, Cortesini C. The muscle coat of the lower esophageal sphincter in patients with achalasia and hypertensive sphincter. An electron microscopic study. J Submicrosc Cytol Pathol 1985;17:673–685. 36. Torihashi S, Yoshida H, Nishikawa S-I, Kunisada T, Sanders KM. Enteric neurons express steel factor-lacZ transgene in the murine gastrointestinal tract. Brain Res 1996;738:323–328. 37. Ward SM, Burns AJ, Torihashi S, Harney SC, Sanders KM. Impaired development of interstitial cells and intestinal electrical rhythmicity in steel mutants. Am J Physiol 1995;269:C1577– C1585.
Received December 3, 1997. Accepted May 4, 1998. Address requests for reprints to: Kenton M. Sanders, Ph.D., Department of Physiology and Cell Biology, University of Nevada School of Medicine, Reno, Nevada 89557. e-mail: [email protected]; fax: (702) 784-6903. Supported by grant DK 40569 from the National Institute of Diabetes and Digestive and Kidney Diseases. Morphology was provided by a core laboratory supported by PO1 DK 41315. The authors thank Julia Bayguinov for technical assistance and Piers Emson at the Molecular Neuroscience Group in Cambridge, England, for the gift of anti-nNOS antibody.
Interstitial Cells of Cajal Mediate Enteric Inhibitory Neurotransmission in the Lower Esophageal and Pyloric Sphincters SEAN M. WARD, GERARD MORRIS, LEE REESE, XUAN–YU WANG, and KENTON M. SANDERS Department of Physiology and Cell Biology, University of Nevada School of Medicine, Reno, Nevada
Background & Aims: Previous studies have suggested that a specific class of interstitial cells of Cajal (ICC) act as mediators in nitrergic inhibitory neurotransmission. The aim of this investigation was to examine the role of intramuscular ICC (IC-IM) in neurotransmission in the murine lower esophageal (LES) and pyloric sphincters (PS). Methods: Immunohistochemistry and electrophysiology were used to study the distribution and role of IC-IM. Results: The LES and PS contain spindle-shaped IC-IM, which form close relationships with nitric oxide synthase–containing nerve fibers. The PS contains ICC within the myenteric plexus and c-Kit immunopositive cells along the submucosal surface of the circular muscle. IC-IM were absent in the LES and PS of c-kit (W/W v) mutant mice. Using these mutants, we tested whether IC-IM mediate neural inputs in the LES and PS. Although the distribution of inhibitory nerves was normal in W/W v animals, NO-dependent inhibitory neurotransmission was reduced. Hyperpolarizations to sodium nitroprusside were also attenuated in W/W v animals. Conclusions: The data suggest that IC-IM play an important role in NO-dependent neurotransmission in the LES and PS. IC-IM may be the effectors that transduce NO signals into hyperpolarizing responses. Loss of IC-IM may interfere with relaxations and normal motility in these sphincters.
significant role of the lower esophageal sphincter (LES) and pyloric sphincter is to compartmentalize the gastric contents, which, due to their strong acidic nature, can be damaging to the contiguous organs. The LES stays tonically contracted until it reflexly opens after a swallow to allow ingested materials to pass from the esophagus into the stomach.1 The pylorus, although partially open between gastric peristaltic events, contracts as the gastric peristaltic wave reaches the terminal antrum, causing retention of most of the contents until particle size can be reduced. Feedback mechanisms sense the composition of chyme reaching the duodenum and regulate the motility of the antroduodenal canal to adjust the rate of gastric emptying.2 Control of emptying allows orderly delivery of the gastric contents to the small
A
intestine for digestion and absorption. Relaxation, and therefore control of luminal diameters and resistances, of both sphincters are regulated, in part by nonadrenergic, noncholinergic (enteric inhibitory) motor neurons.1,2 Neurons expressing nitric oxide synthase (NOS) are abundant in sphincteric regions of the gastrointestinal (GI) tract with up to 50% of myenteric neurons expressing NOS or reduced nicotinamide adenine dinucleotide phosphate (NADPH) diaphorase, a common histochemical label for NOS.3–9 A significant portion of NOSpositive neurons are motor neurons that innervate the musculature, and varicose fibers positive for NOS can be observed in whole-mount specimens of these tissues. Sphincteric regions of the GI tract, including the LES and pylorus, display prominent NO-dependent hyperpolarization of resting membrane potential and relaxation in response to activation of enteric inhibitory neurons, suggesting significant regulation by nitrergic mechanisms.10–12 Loss of NOS positive neurons has been associated with a variety of pathological conditions including achalasia and hypertrophic pyloric stenosis.13–15 However, the postjunctional mechanisms that mediate nitregic inhibitory responses are not fully understood. Enteric inhibitory neurotransmission has typically been conceptualized as direct regulation of GI muscles by nonadrenergic, noncholinergic (NANC) transmitter(s) released from nerve terminals. However, a century ago Cajal16 proposed an intermediary in autonomic neurotransmission involving interstitial cells (now referred to as interstitial cells of Cajal [ICC]). Recently, we have provided evidence that NO-dependent neurotransmission in the murine gastric fundus requires ICC for Abbreviations used in this paper: c-Kit–LI, c-Kit–like immunoreactivity; EFS, electrical field stimulation; FITC, fluorescein isothiocyanate; GI, gastrointestinal; ICC, interstitial cells of Cajal; IC-IM, intramuscular ICC; KRB, Krebs–Ringer’s buffer; L-NAME, N G-nitro-Larginine methyl ester; NADPH, reduced nicotinamide adenine dinucleotide phosphate; NANC, nonadrenergic, noncholinergic; NOS, nitric oxide synthase; SNP, sodium nitroprusside. r 1998 by the American Gastroenterological Association 0016-5085/98/$3.00
August 1998
ICC AND INHIBITORY NEUROTRANSMISSION
315
Figure 1. Isolation of muscles of the pyloric sphincter and LES. (A) A stomach with segments of the esophagus (E) and duodenum (D) attached. (B) A stomach opened along the lesser curvature from the gastric antrum (GA) through the duodenum segment. (C) The same preparation after removal of the mucosa. The pyloric sphincter (PS) is clearly visible as a thickened band of circular muscle at the junction between the stomach and duodenum. (D) A stomach opened along the greater curvature to reveal the junction between the esophagus and the stomach. (E ) The same preparation after opening the gastroesophageal junction through the esophagus. (F ) The same preparation after removal of the mucosa. The LES is visible as a band of thickened musculature. F, fundus.
hyperpolarization and relaxation responses to field stimulation of enteric neurons.17 In the present study we used c-kit mutant mice in which the intramuscular class of ICC fails to develop to investigate the role of ICC in the LES and pyloric sphincter. We have characterized the types of ICC present in normal mice and determined the
consequences of losing specific ICC on enteric inhibitory neural control.
Materials and Methods Wild-type (1/1; black coats) and W/Wv (pure white coats) mice,18 between the ages of 20 and 30 days postpartum,
316
WARD ET AL.
GASTROENTEROLOGY Vol. 115, No. 2
Figure 2. Distribution of ICC in the murine LES. (A ) A cryostat cross section through the circular (cm) and longitudinal (lm) muscle layers (section cut along the long axis of the longitudinal layer) of a wild-type (1/1) mouse. IC-IM are observed in both layers (arrows). Circular cells appear as points and longitudinal IC-IM as elongated cells with ovoid nuclear region and long, unbranching processes. No immunoreactivity was observed in the submucosal region(s). (B ) A confocal reconstruction of the entire tunica muscularis from a flattened whole mount. This shows the relative density of the IC-IM in the circular and longitudinal muscle layers. IC-IM were completely absent in W/W v animals. (C ) A cryostat section and (D ) a flattened whole-mount preparation from W/W v animals.
were anesthesized by chloroform inhalation and exsanguinated by cervical dislocation followed by decapitation. The entire stomach, including portions of the esophagus and duodenum (Figure 1A), was removed and placed in Krebs–Ringer’s buffer (KRB). To isolate the pyloric sphincter, stomachs were opened along the lesser curvature from the terminal antrum through the duodenum segment (Figure 1B). The mucosa was removed, revealing the underlying thickened band of circular muscle that constitutes the pyloric sphincter at the junction between the stomach and duodenum (Figure 1C). To isolate the LES, stomachs were opened along the greater curvature to reveal the junction between the esophagus and the stomach (Figure 1D). From the gastric aspect of the gastroesophageal junction, the esophagus was opened by sharp dissection (Figure 1E). The mucosa was removed revealing the thickened musculature of the LES (Figure 1F). Histological examination showed that the musculature immediately proximal to the LES was composed of skeletal muscle fibers (data not shown).
Morphological Studies Immunohistochemical studies were performed on tissues that were dehydrated in graded sucrose solutions, embedded in Tissue Tek (Miles, IL), and frozen in liquid nitrogen.
Cryostat sections were cut at 10-µm thickness, fixed in acetone, and preincubated in goat nonimmune serum for 1 hour (10% in phosphate buffered saline [PBS]) before being incubated with a monoclonal antibody against c-kit protein (ACK219; 5 µg/mL in PBS) at 4°C overnight. Immunoreactivity was detected with fluorescein isothiocyanate (FITC)-conjugated secondary antibody (goat anti-rat 1:100, 1 hour, room temperature; Vector Laboratories Inc., Burlingame CA). Control tissues were prepared in a similar manner, omitting ACK2 from the incubation solution. Whole-mount tissues were labeled with the ACK2 antibody using the same method as outlined above. The antibody was diluted in PBS (0.1 mol/L) containing 0.3% Triton X-100. For double immunostaining, tissues were first incubated with anti–c-Kit antibody for 48 hours at 4°C, as described above. This was followed by washing with PBS, and then tissues were incubated with secondary antibody (FITC)-conjugated goat anti-rat immunoglobulin G (IgG). Tissues were washed again before incubation with a second primary antibody to NOS (anti-sheep, 1:800, for 48 hours). After another wash with PBS, the tissues were washed again before being incubated with Texas Red–conjugated anti-sheep IgG raised in rabbit. Secondary antibodies were purchased from Vector Laboratories and
August 1998
ICC AND INHIBITORY NEUROTRANSMISSION
317
Figure 3. Distribution of ICC in the murine pyloric sphincter. (A ) A cryostat cross section featuring the various classes of ICC within the pyloric sphincter. ICC in the myenteric region (IC-MY) are brightly labeled in the plane between the circular (cm) and longitudinal (lm) muscle layers. As in the LES, IC-IM are found in both muscle layers, running parallel with the muscle fibers. We also observed cells with c-Kit–LI along the surface separating the submucosa (s) from the circular muscle, running perpendicular to the circular muscle layer (*). (B ) A confocal image featuring the IC-IM of the circular muscle layer. The cells are long and spindle-shaped and run parallel to the circular muscle fibers. (C ) A whole mount showing IC-IM (arrows) and IC-MY (arrowheads). (D ) A cryostat section revealing the absence of IC-IM and cells with c-Kit–LI at the submucosal surface in W/W v animals. IC-MY were apparently unaffected by the W/W v mutations. (E and F ) Confocal images of whole mounts featuring the circular muscle layer and the region of the myenteric plexus. Note the abundance of IC-MY (arrowheads).
diluted to 1:100 in PBS. Secondary incubations occurred for 1 hour at room temperature. Control tissues were prepared in a similar manner, either omitting primary or secondary antibodies from the incubation solutions. Labeled tissues were examined using a Bio-Rad MRC 600 confocal microscope (Bio-Rad Laboratories, Hercules, CA), with an excitation wavelength appropriate for FITC (496 nm) and Texas Red (592 nm), where appropriate. Confocal micrographs were obtained from digital composites of Z-series scans of 10–15 optical sections through a depth of
5–30 µm. Final images were constructed with Bio-Rad Comos software. NADPH-diaphorase histochemistry was performed on whole mounts fixed with paraformaldehyde, as described previously.19
Physiological Experiments After removing the mucosa, strips of LES and pyloric sphincter muscle (2 3 1 mm) were pinned to the floor of a recording chamber with the mucosal side of the circular muscle
318
WARD ET AL.
GASTROENTEROLOGY Vol. 115, No. 2
Figure 4. Distribution of NOS-containing neurons in the LES and pyloric sphincter as determined by NADPH diaphorase histochemistry. (A, B, D, and E ) Whole-mount images of the myenteric plexus from 1/1 and W/W v animals at different levels of resolution. We could detect no differences between the number, size, or distribution of cell bodies within the myenteric plexus. (C and F ) Varicose fibers within the circular muscle layer. No difference in the density of innervation was detected. (G, H, J, and K ) Whole mounts of the pyloric sphincters of 1/1 and W/W v animals. (I–L) Varicose fibers within the circular muscle layer. No differences in the density of innervation of this region were detected in 1/1 and W/W v animals. Summaries of neuron and varicose fiber densities can be found in Table 1.
facing upward. Parallel platinum electrodes were placed on either side of the muscle strips. Circular muscle cells were impaled with glass microelectrodes, and transmembrane potentials were measured (World Precision Instruments, Sarasota, FL [WPI] S-7071). Electrical signals were recorded on magnetic tape (Racal 40S; Southampton, England). Neural responses were elicited by square wave pulses (0.5-millisecond duration,
supramaximal voltage; S48 Grass stimulator; Grass Medical Instruments, Quincy, MA), of electrical field stimulation (EFS; Grass S48). Separate mechanical experiments were performed using standard organ bath techniques. Strips of muscle (approximately 0.5 3 1.0 mm) were isolated from the LES and pyloric sphincters. The mucosa was removed by sharp dissec-
August 1998
ICC AND INHIBITORY NEUROTRANSMISSION
319
Table 1. Distribution of NADPH-Diaphorase–Positive Nerve Cell Bodies (Myenteric Plexus) and NADPH-Diaphorase–Positive Nerve Processes Running Parallel to the Circular Muscle Layer in the LES and Pylorus of 1/1 and W/W V Mice (1/1) LES
(W/W V ) LES
(1/1) Pylorus
(W/W V ) Pylorus
NADPH-d1 nerve cells 63.8 6 2.1 62.4 6 3.0 61.0 6 1.5 66.1 6 1.5 NADPH-d1 nerve trunks 14.0 6 0.8 15.2 6 0.6 13.6 6 1.5 15.0 6 0.7 NOTE. Numbers of cell bodies reflect their density per 1000 µm2. Numbers of nerve trunks were obtained from the number transecting an arbitrary 100-µm line. All analyses are from 20 fields of view from 4 wild-type and 4 W/W V mutant animals. NADPH-d1, NADPH-diaphorase positive.
tion, and muscles were attached to a fixed mount and to a Fort 10 isometric strain gauge (World Precision Instruments). The muscles were immersed in organ baths maintained at 37 6 0.5°C with oxygenated KRB. A resting force of 200 mg was applied, which was shown to set the muscles at optimum length (data not shown). This was followed by an equilibration period of 1 hour, during which time the bath was continuously perfused with oxygenated KRB. Signals were recorded onto a chart recorder (Gould RS 3600; Gould, Cleveland, OH). Data are expressed as means 6 SEM from 1/1 (wild-type) and W/Wv (mutants). Differences in the data were evaluated by Student’s t test. P values of ,0.05 were taken as a statistically significant difference. The n values reported in the text refer to the number of animals used for each protocol, and only one muscle of a given region could be prepared per animal. A total of 86 1/1 and 39 W/Wv animals were used for physiological experiments and 12 1/1 and 12 W/Wv animals used for morphological studies.
Solutions and Drugs Muscles were maintained in KRB (37.5 6 0.5°C; pH 7.3–7.4) containing (in mmol/L): Na1, 137.4; K1, 5.9; Ca21, 2.5; Mg, 1.2; Cl2, 134; HCO32, 15.5; H2PO42, 1.2; dextrose, 11.5, and bubbled with 97% O2–3% CO2. Solutions of BayK 8644 (RBI, Natick, MA) were prepared in ethanol at 1022 mol/L, and were diluted to 1026 mol/L in KRB. Atropine sulfate, guanethidine sulfate, sodium nitroprusside (SNP), and L-nitroarginine or NG-nitro-L-arginine methyl ester (L-NAME; Sigma Chemical Co., St. Louis, MO) were dissolved in distilled water at 1021 to 1022 mol/L and diluted in KRB to the stated final concentrations.
Results c-Kit–Positive Cells in the LES and Pyloric Sphincter and the Consequences of W Mutations on c-Kit–Expressing Cells Antibodies to c-Kit were used to determine the distribution of cells with c-Kit–like immunoreactivity (c-Kit–LI). In six LES examined, cells with c-Kit–LI were
Figure 5. Confocal micrographs of NOS-like immunoreactivity in the (A ) LES and (B ) pyloric sphincters. Whole mounts of the LES and pyloric sphincter fixed with formaldehyde showed NOS-like immunoreactivity in varicose nerve fibers (arrowheads) and cells with profiles suggestive of IC-IM (arrows). The varicose fibers were in close proximity to the IC-IM.
distributed within the circular and longitudinal muscle layers (Figure 2A). Cryosections of muscles from 3 animals showed a relatively uniform distribution throughout the thickness of both muscle layers. Cells with c-Kit–LI were spindle-shaped, with few or no lateral processes (Figure 2B). Cells with this morphology in the murine fundus are intramuscular interstitial cells of Cajal (IC-IM; Burns et al.17). Cells with c-Kit–LI were not found within the myenteric plexus region of the LES, and we observed no cells with multiple processes forming networks suggestive of myenteric ICC. This observation is also similar to the distribution of ICC in the murine fundus (Burns et al.17). In contrast to wild-type animals, cells with c-Kit–LI were absent from the circular and longitudinal muscle layers of the LES of 6 W/Wv animals examined (Figure 2C and D). Cells with c-Kit–LI were distributed in a different manner within the pyloric sphincter. c-Kit–positive cells were observed in several locations in the 6 wild-type mice examined (Figure 3A–C): (1) spindle-shaped cells were
320
WARD ET AL.
GASTROENTEROLOGY Vol. 115, No. 2
Figure 6. Double-labeling of cells with c-Kit–LI (IC-IM) and NOS-like immunoreactivity. Whole mounts of the (A–C ) LES and (D–F ) pyloric sphincter fixed with acetone showed clear labeling of IC-IM with both antibodies. A and D show immunoreactivity to NOS (FITC), and B and E show c-Kit–LI (Texas red). C and F show colocalization of both fluorophores (yellow) in IC-IM (arrows). The NOS immunoreactivity was reduced in varicose nerve fibers in acetone-fixed preparations. Scale bars apply to all figures.
found within the circular and longitudinal muscle layers (IC-IM); (2) a dense network of c-Kit–positive cells was found in the region of the myenteric plexus; and (3) occasional cells with c-Kit–LI were observed along the submucosal surface of the circular muscle layer. The latter had large oval nuclei and branching main processes extending perpendicular to the main axis of the circular muscle fibers (Figure 3A). This type of c-Kit–positive cell has not been previously described, and nothing is known about the function of these cells. IC-IM and cells with c-Kit–LI along the submucosal surface of the circular muscle layer were absent in muscles of 6
W/Wv mice examined, but the network of the myenteric plexus was unaffected by the W/Wv mutations (Figure 3D–F). Relationship Between Inhibitory Nerve Fibers and IC-IM in Murine LES and Pyloric Sphincter IC-IM have previously been shown to be a critical element in nitrergic neurotransmission.17 Therefore, we investigated the innervation of the tunica muscularis of the LES and pyloric sphincter by NOS-containing neurons and studied the relationship between NOS neurons
August 1998
ICC AND INHIBITORY NEUROTRANSMISSION
321
and IC-IM. The overall distributions of NOS neurons and varicose processes were characterized in the LES and pyloric sphincters of wild-type and W/Wv mutants using the NADPH-diaphorase technique, which labels NOS neurons in formaldehyde-fixed muscles.19 Figure 4 shows micrographs from whole mounts of the LES and pyloric sphincter. We could detect no differences in the number of NOS neurons or the distribution of diaphorase-positive varicose fibers in wild-type or mutant animals in tissues of either the LES or pyloric sphincter (Table 1). Previous ultrastructural studies have shown a close relationship between NOS-positive nerve terminals and ICC.20 We examined the distribution of NOS with immunohistochemistry in the LES and pylorus in paraformaldehyde-fixed muscles and found NOS-like immunoreactivity in varicose nerve fibers surrounding IC-IM (Figure 5). The latter point was further explored in double-labeling studies in which antibodies to c-Kit and neuronal NOS were used (see Materials and Methods). These studies were performed in acetone-fixed muscles because we have observed poor labeling of IC-IM with the ACK2 antibody in tissues fixed with formaldehyde. These experiments clearly showed that IC-IM (identified as c-Kit–positive cells) contained NOS-like immunoreactivity (Figure 6). Inhibitory Neurotransmission in Wild-type and W/W v Mutants Strips of LES muscles were stimulated with EFS in the presence of atropine and guanethidine (both 1 µmol/L) to reveal NANC inhibitory responses. In mechanical experiments, the tone of the muscle strips was increased by inclusion of BayK 8644 (1 µmol/L) to facilitate resolution of relaxation responses. EFS (1–20 Hz for 1 minute) caused frequency-dependent relaxation of LES strips (Figure 7A). After cessation of the stimulus, a pronounced rebound contraction was observed. These b Figure 7. Mechanical responses to EFS of the LES. (A ) Responses to 5- and 10-Hz stimulation (1-minute trains; 0.5-millisecond pulses; black bars). Responses under NANC conditions (see Materials and Methods) were characterized by a relaxation of the muscle during the period of stimulation and a rebound contraction upon cessation of the stimulation. L-NAME (200 µmol/L) blocked the relaxation phase and rebound contractions and converted the response during the stimulation at 10 Hz to a contraction. (C ) Responses to EFS of a strip from a W/W v animal. In all cases, EFS elicited a contractile response during the stimulation and the poststimulus rebound response did not occur. (D ) L-NAME had little or no effect in W/W v animals. (E ) A graph showing a summary of experiments in wild-type (n 5 5) and W/W v animals (n 5 5). Frequency-dependent inhibitory responses during stimulation of wild-type muscles (d) were decreased by L-NAME (s). The response during stimulation of muscles from W/W v animals was a frequency-dependent contraction during the period of stimulation (j), which was not significantly affected by L-NAME (h).
322
WARD ET AL.
GASTROENTEROLOGY Vol. 115, No. 2
responses are similar to the NANC responses reported for the LES of other species. Addition of L-NAME (200 µmol/L), inhibited the relaxation response during the stimulus, converting this response to a contraction, and blocked the poststimulation rebound response (Figure 7B). LES muscles of W/Wv animals were also stimulated with EFS using the same conditions and parameters. The responses of these muscles were considerably different from the control animals. EFS of muscles from W/Wv animals failed to produce relaxation or rebound contractions (Figure 7C). Addition of L-NAME either had no effect or produced a slight enhancement of the contractile responses during stimulation (Figure 7D). The increase failed to reach a level of significance in the series of experiments performed. These data suggest that nitrergic neurotransmission was reduced in the LES of W/Wv animals. These experiments are summarized in Figure 7E. The same protocols used for mechanical experiments on the LES were used on muscles from the pyloric sphincter of wild-type and W/Wv animals. As in the LES, EFS resulted in relaxation during the period of stimulation, and a poststimulus rebound contraction was observed upon cessation of the stimulus (Figure 8A). L-NAME (200 µmol/L) blocked both responses (Figure 8B). The relaxation phase and rebound contraction phase of the response to EFS were absent in W/Wv animals (Figure 8C), and L-NAME (200 µmol/L) produced a slight enhancement of the contractile responses during EFS (Figure 8D; P , 0.05). Responses of the pyloric sphincter are summarized in Figure 8E. To more closely examine nitrergic neurotransmission, a more detailed series of electrophysiological experiments was conducted on circular muscles of wild-type and mutant animals. LES cells had average resting potentials b Figure 8. Mechanical responses of the pyloric sphincter to EFS. (A ) Responses to 5- and 10-Hz stimulation (1-minute trains; 0.5millisecond pulses; black bars). (B ) Responses under NANC conditions (see Materials and Methods) were characterized by a relaxation of the muscle during the period of stimulation and a rebound contraction on cessation of the stimulation. L-NAME (200 µmol/L) blocked the relaxation phase and rebound contractions. (C ) Responses to EFS of a strip from a W/W v animal. In all cases, EFS elicited a contractile response during the stimulation and the poststimulus rebound response failed to occur. (D ) L-NAME slightly increased the contractions of muscles of W/W v animals during stimulation. (E ) A graph showing a summary of experiments in wild-type (n 5 8) and W/W v animals (n 5 5). Frequency-dependent inhibitory responses during stimulation of wild-type muscles (d) were decreased by L-NAME (s). The response during stimulation of muscles from W/W v animals was a frequency-dependent contraction during the period of stimulation (j), which was slightly increased by L-NAME (h).
August 1998
ICC AND INHIBITORY NEUROTRANSMISSION
323
Figure 9. Changes in membrane potential elicited by EFS of LES muscles. (A ) Responses to 1–20 Hz stimulation (1-second trains at arrows). The inhibitory junction potentials were characterized by a rapid hyperpolarization followed by a slow hyperpolarization phase. (B ) L-NAME (200 µmol/L) greatly reduced both phases of the inhibitory junction potential. Recordings in A and B were made from the same cell. (C ) In W/W v animals, inhibitory junction potentials were smaller in amplitude and consisted only of the rapid phase of hyperpolarization. (D ) These events were not significantly affected by L-NAME. Recordings in C and D were made from the same cell.
of 257 6 0.6 mV (n 5 24 animals) in the presence of atropine and guanethidine (both 1 µmol/L). Resting potential was not stable and was characterized by rapid, small-amplitude fluctuations (Figure 9A). EFS (1–20 Hz; 1-second trains) produced compound inhibitory junction potentials consisting of a rapid phase of hyperpolarization, followed by a second, slow phase of hyperpolarization (Figure 9A; most evident at 10 and 20 Hz). Addition of L-NAME (200 µmol/L) did not significantly alter resting membrane potential (256 6 0.7 mV; n 5 24 animals; P . 0.05), but it inhibited both phases of the inhibitory junction potential (Figure 9B), suggesting that the hyperpolarization responses strongly depended on the synthesis of NO. As in the case of mechanical responses, the electrical response to EFS was considerably different in the LES of W/Wv animals. Cells of W/Wv animals had average resting potentials of 259 6 0.8 mV (n 5 18 animals; resting potential not significantly differ-
ent from wild-type animals, P . 0.2), and resting membrane potential was always more stable in these tissues (Figure 9C). EFS (1–20 Hz; 1-second trains), caused inhibitory junction potentials of smaller amplitude and duration (Figure 9C). The slow, second phase of the inhibitory junction potentials were absent in the LES of W/Wv animals. The amplitude and duration of inhibitory junction potentials in W/Wv animals were approximately the same levels as the amplitude of inhibitory junction potentials in 1/1 animals after addition of L-NAME. Addition of L-NAME to W/Wv tissues caused no significant change in resting potential (i.e., 257 6 0.8 mV; P . 0.15) and did not change the amplitude or duration of inhibitory junction potentials (both P . 0.4; Figure 9D). Figure 10 shows a summary of these experiments performed on muscles of 24 1/1 and 18 W/Wv animals. Circular muscle cells of the pyloric sphincter had average resting potentials of 257 6 0.7 mV (n 5 49) in
324
WARD ET AL.
GASTROENTEROLOGY Vol. 115, No. 2
Figure 10. Summaries of electrical responses to EFS of wild-type and W/W v LES muscles. (A ) Amplitudes (fast component) of frequencydependent inhibitory junction potentials (IJPs) in wild-type animals before (d) and after (s) L-NAME. (B ) IJP duration in wild-type animals before (d) and after (s) L-NAME. The amplitudes and durations of IJPs significantly decreased at all frequencies (96 impalements from 24 animals). In W/W v animals (54 impalements from 18 animals), (C ) amplitudes and (D ) durations of IJPs were reduced (d) and unaffected (s) by L-NAME.
the presence of atropine and guanethidine (both 1 µmol/L). EFS (1–20 Hz; 1-second trains) produced compound junction potentials consisting of a rapid phase of hyperpolarization, followed by a rebound depolarization (Figure 11A). Addition of L-NAME (200 µmol/L) did not affect membrane potential (i.e., 257 6 0.6 mV; n 5 49; P . 0.5), but it reduced inhibitory junction potentials and the poststimulus rebound response, suggesting that both responses depended on the synthesis of NO (Figure 11B; P , 0.05). Circular muscle cells of the pyloric sphincter of W/Wv animals had average resting potentials of 259 6 0.9 mV (not significantly different from wild-type animals, P . 0.3). EFS (1–20 Hz; 1-second trains) caused inhibitory junction potentials of significantly smaller amplitude and duration (Figure 11C) than observed in wildtype animals (both P , 0.05). The amplitude of inhibitory junction potentials in W/Wv animals was approximately at the same levels as the events in wild-type animals in the presence of L-NAME (P . 0.4; analysis of variance). Addition of L-NAME to W/Wv tissues did not affect resting membrane potential (i.e., 257 6 0.9 mV; n 5 26; P . 0.3) and caused no significant change in the amplitude and duration of inhibitory junction potentials (both
P . 0.2; Figure 11D). Data from these experiments are summarized in Figure 12. Postjunctional Responses to SNP The hyperpolarizing responses to EFS, shown to be dependent on NO synthesis, were mimicked by SNP. In wild-type LES muscles, SNP caused a concentrationdependent hyperpolarization response (Figure 13A). For example, 1 µmol/L SNP caused an average 12.76 1.6 mV hyperpolarization (n 5 11). In LES of W/Wv animals, SNP had little effect on membrane potential (Figure 13B); 1 µmol/L SNP produced an average of 1.5 6 0.5 mV of hyperpolarization, which was significantly less than in wild-type animals (n 5 8 animals; P , 0.02). Similar results were observed in pyloric muscles. In wild-type muscles, SNP caused a concentration-dependent hyperpolarization (Figure 13C), averaging 8.5 6 0.7 mV in response to 1 µmol/L SNP (n 5 13 muscles). In the pyloric sphincter of W/Wv animals, the effect of SNP on membrane potential was also reduced (Figure 13D). SNP (1 µmol/L) caused an average hyperpolarization of 3.1 6 0.5 mV (n 5 7; P , 0.001).
August 1998
ICC AND INHIBITORY NEUROTRANSMISSION
325
Figure 11. Changes in membrane potential elicited by EFS of pyloric sphincter muscles. (A ) Responses to 1–20 Hz stimulation (1-second trains at arrows). The inhibitory junction potentials (IJPs) were characterized by a rapid hyperpolarization followed by a rebound depolarization. (B ) L-NAME (200 µmol/L) reduced both phases of the IJP. Recordings in A and B were made from the same cell. (C ) In W/W v animals, IJPs were smaller in amplitude. (D ) These events were not significantly affected by L-NAME. Recordings in C and D were made from the same cell.
Discussion The LES and pyloric sphincter of the mouse have spindle-shaped ICC similar to the gastric fundus.17 Similar cells have also been observed in the esophagus, LES, stomach, cecum, and internal anal sphincter of the guinea pig,21 but the physiological role of this class of ICC in sphincters has not been documented previously. Spindle-shaped ICC lie within the circular and longitudinal muscle layers in close apposition with smooth muscle cells and varicose processes of motor neurons. We have referred to this class of ICC as IC-IM.17,21,22 In the present study we found that IC-IM were missing in the LES and pyloric sphincters of c-kit mutants (W/Wv). By studying tissues of these mutants, we have obtained evidence that IC-IM are important mediators of enteric inhibitory neurotransmission in these sphincters. Daniel and Posey-Daniel23 characterized the relationship between ICC, smooth muscle cells, and nerve varicosities using morphometric techniques on tissues of the opossum esophagus. They found close association between nerve terminals and ICC (20 nm), but few extremely close associations could be found between varicosities and smooth muscle cells. ICC also formed frequent gap junctions with smooth muscle cells. Similar relationships were noted in the murine fundus,17 and the present study shows that IC-IM of the LES and pyloric
sphincter are surrounded by NOS-containing varicose nerve fibers. Thus, there seems to be a structural arrangement between nerve terminals, ICC, and smooth muscle cells that facilitates a role for ICC in neurotransmission. Comparisons of wild-type and W/Wv siblings provided an opportunity to test the role of IC-IM in neuromuscular transmission. Our data suggest that IC-IM facilitate NO-dependent inhibitory neurotransmission in the LES and pyloric sphincter. In the murine gastric fundus we could find no evidence for NO-dependent neurotransmission in the absence of IC-IM. The dependence on IC-IM for nitrergic input, although highly significant in the LES and pyloric sphincter, may not be as complete as in the fundus. There were no relaxation responses to EFS in the LES or pyloric sphincter of W/Wv animals; EFS elicited contractile responses in muscles of these animals. In some cases the contractions elicited by EFS were slightly enhanced by blocking NO synthesis. This indicates that a component of NO-dependent neurotransmission may have remained in W/Wv animals. Thus, in the absence of IC-IM, there may be some direct overflow of NO onto surrounding smooth muscle cells. This suggests parallel innervation of IC-IM and smooth muscle cells by enteric inhibitory neurons, as originally suggested by Daniel and PoseyDaniel23 from their structural analyses of the opossum
326
WARD ET AL.
GASTROENTEROLOGY Vol. 115, No. 2
Figure 12. Summaries of electrical responses to EFS in pyloric sphincters of wild-type and W/W v animals. (A ) Amplitudes of frequency-dependent inhibitory junction potentials (IJPs) in wild-type animals before (d) and after (s) L-NAME. (B ) IJP duration in wild-type animals before (d) and after (s) L-NAME. The amplitudes and durations of IJPs were significantly decreased at all frequencies (P , 0.05). In W/W v animals, the (C ) amplitude and (D ) duration of IJPs were significantly smaller than observed in wild-type animals (d; P , 0.05) and were unaffected by L-NAME (s; P . 0.05). The IJPs in W/W v animals were approximately the same amplitudes and durations as wild-type animals in the presence of L-NAME (P . 0.4; analysis of variance).
esophagus. The degree of influence exerted directly on smooth muscle cells seems to be far less, however, than the effects mediated by IC-IM. Brief periods of stimulation (1 second) resulted in NO-dependent, postjunctional electrical responses that were absent in the LES of W/Wv animals. However, NO-dependent responses to longer periods of stimulation (as used in mechanical experiments) were not totally blocked in W/Wv animals. Longer stimulus periods may result in the production of enough NO to overflow onto nearby smooth muscle and directly elicit mechanical responses in these cells. The relative contribution of nitrergic neurotransmission to regulation of the murine pyloric sphincter seemed to be less than in the LES, and it seemed that a significant portion of enteric inhibitory neurotransmission in the pylorus was mediated by a transmitter other than NO. The effects of this transmitter were evident in W/Wv animals, suggesting direct effects on smooth muscle cells. Future in vivo studies of mutant animals may help to evaluate the importance of IC-IM in mediating the reflex relaxation of the LES after a swallow and in regulating gastric emptying. The loss of the relaxation responses in the LES
and pyloric sphincter strongly suggest that IC-IM are important for normal motor function. Our data suggest that the transduction and/or effector mechanisms that convert NO to a hyperpolarization response in normal tissues may be expressed by IC-IM because the hyperpolarization response to exogenous NO (delivered via SNP) was lost in W/Wv mice. The ionic conductances activated by NO may be mainly expressed by IC-IM, and hyperpolarizing responses to NO normally observed in the LES and pyloric sphincter may be electrotonically conducted from IC-IM to neighboring smooth muscle cells. A complete test of this hypothesis will require comparative voltage-clamp studies on smooth muscle cells and IC-IM from these sphincters. It is important to note that IC-IM of the LES and pyloric sphincter expressed NOS-like immunoreactivity. Previous studies have shown expression of NOS-like immunoreactivity in ICC of the canine colon and rat small intestine,20,24 and the finding that NOS-like immunoreactivity is present in other classes of ICC (deep muscular plexus; IC-DMP) from a different species suggests that this enzyme could be a general feature of
August 1998
ICC AND INHIBITORY NEUROTRANSMISSION
327
Figure 13. Postjunctional responses to SNP. Traces in each panel are superimposed recordings from single cells showing responses to several concentrations of SNP (1028 to 1024 mol/L). The traces in each panel are lined up on the time of SNP application (arrow). SNP was applied at the concentrations noted throughout the period of each recording, and the wash periods between SNP doses were eliminated. (A and B ) Responses in LES muscles of wild-type (1/1) and W/W v animals, respectively. (C and D ) Responses in pyloric muscles wild-type (1/1) and W/W v animals, respectively.
many types of ICC. We have previously suggested that ICC could act as amplifiers of enteric inhibitory neurotransmission because we observed that NO enhanced Ca21 in ICC and enhancement in Ca21 led to production and release of NO by ICC.25 Loss of IC-IM in the LES and pyloric sphincters of W/Wv mice and the amplification of enteric inhibitory neurotransmission provided by these cells could also possibly contribute to the reduction in NO-dependent neurotransmission observed in these animals. Loss of the amplification function hypothesized for ICC, however, does not fully explain the functional defect we observed in W/Wv tissues. Hyperpolarizing responses to exogenous SNP were greatly reduced or abolished in the LES and pyloric muscles of these animals, suggesting that IC-IM may express a critical transduction mechanism (i.e., second messengers or ion channels) that mediates the electrical responses to nitrergic stimulation. All classes of ICC were not lost in W/Wv mice. The spindle-shaped IC-IM were gone, but the multiprocessed cells in the myenteric region of the pyloric canal seemed
to be unaffected in the mutant animals. Similar observations regarding the myenteric region in W/Wv mice were made previously in tissues of the gastric corpus and antrum.17 Others have concluded that c-Kit is not important for the development of some classes of ICC,26 but results with neutralizing antibodies against c-Kit do not support this view. For example, we have found that all classes of ICC are reduced in the colon and small intestine in animals treated with anti–c-Kit antibodies.27 Cells with c-Kit–LI were also missing from the small intestine and pyloric sphincter in another study of similarly treated animals.28 Recent work using organ cultures confirms the observation that treatment with a neutralizing anti–c-Kit antibody early in development or immediately following birth causes loss of all classes of ICC.29 Thus, we suggest that the development of some classes of ICC (e.g., the myenteric region of the intestine and IC-IM of sphincteric regions and stomach) is susceptible to the nontotal lesion in tyrosine kinase activity in W/Wv compound heterozygotes. Other classes of ICC (e.g., the myenteric
328
WARD ET AL.
region of the stomach and colon, the deep muscular plexus of the small intestine, IC-IM in the circular and longitudinal muscle layers of the colon, and the submucosal surface of the colon) are resistant to mutations in c-kit that do not totally abolish the function of the receptor tyrosine kinase. Loss of nitrergic neurons from sphincteric regions has been reported in clinical disorders such as achalasia,14 congenital esophageal stenosis,30 and hypertrophic pyloric stenosis.13,15,31–33 Results of the present study suggest that defective nitrergic neurotransmission could also arise from lesions in pyloric and LES IC-IM. Circumstances involving loss of ICC have been noted in tissues from cases of pyloric stenosis34 and achalasia.35 The parallelism between loss of nitrergic neurons and ICC in these clinical conditions raises the question whether there is a relationship tying the phenotype of ICC to nitrergic motor neurons or vice versa. It is possible, for example, that nitrergic neurons and ICC could express growth factors that regulate the phenotype of the other cell type. It has been reported, for example, that enteric neurons express stem cell factor,36 the natural ligand for c-Kit receptors (expressed by ICC). Expression of mutant forms of stem cell factor, such as in steel mutants (i.e., Sl/Sld) causes lesions in some types of ICC, including the spindle-shaped IC-IM of the fundus and sphincteric regions.37 If tonic stimulation by stem cell factor is important in adults to maintain the phenotype of ICC, then it is possible that loss of ICC could result from damage to nitrergic neurons. The specific defects, genetic or environmental, that cause loss of nitrergic neurons and ICC are unknown. Understanding the factors that regulate the plasticity of ICC and enteric nerve/ICC interactions seems to be a very important new direction for motility research. In summary, IC-IM are distributed within the tunica muscularis of the LES and pyloric sphincters. Like in the murine fundus, cells of this morphology play a significant role in the neuroregulation of these sphincters by mediating nitrergic neurotransmission. Developmental defects that block the formation of mature IC-IM in the LES and pyloric sphincter or loss of these cells from adults might seriously jeopardize the normal functions of these sphincters. It will be interesting to evaluate the status of IC-IM in human disorders associated with sphincter dysfunction.
GASTROENTEROLOGY Vol. 115, No. 2
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16. 17.
18.
19.
20.
References 1. Conklin JL, Christensen J. Motor functions of the pharynx and esophagus. In: Johnson LR, ed. Physiology of the gastrointestinal tract. 3rd ed. New York: Raven, 1994:903–928. 2. Mayer EA. The physiology of gastric storage and emptying. In:
21.
22.
Johnson LR, ed. Physiology of the gastrointestinal tract. 3rd ed. New York: Raven, 1994:929–976. Aimi Y, Kimura H, Kinoshita T, Minami Y, Fujimura M, Vincent SR. Histochemical localization of nitric oxide synthase in rat enteric nervous system. Neuroscience 1993;53:553–560. Furness JB, Li ZS, Young HM, Fo¨rstermann U. Nitric oxide synthase in the enteric nervous system of the guinea pig: a quantitative description. Cell Tissue Res 1994;277:139–149. Ward SM, Xue C, Sanders KM. Localization of nitric oxide synthase immunoreactivity in ileocolonic and pyloric sphincters. Cell Tissue Res 1994;275:513–527. Singaram C, Sengupta A, Sweet MA, Sugarbaker DJ, Goyal RK. Nitrinergic and peptidergic innervation of the human oesophagus. Gut 1994;35:1690–1696. Christensen J, Fang S, Rick GA. NADPH-diaphorase-positive nerve fibers in smooth muscle layers of opossum esophagus: gradients in density. J Auton Nerv Syst 1995;52:99–105. Lynn RB, Sankey SL, Chakder S, Rattan S. Colocalization of NADPH-diaphorase staining and VIP immunoreactivity in neurons in opossum internal anal sphincter. Dig Dis Sci 1995;40:781– 791. Wang XY, Wong WC, Ling EA. Localization of nicotinamide adenine dinucleotide phosphate diaphorase (NADPH-d) activity in the gastrointestinal sphincters in the guinea pig. J Auton Nerv Syst 1996;58:51–55. Conklin JL, Du C, Murray JA, Bates JN. Characterization and mediation of inhibitory junction potentials from opossum lower esophageal sphincter. Gastroenterology 1993;104:1439–1444. Bayguinov O, Sanders KM. Role of nitric oxide as an inhibitory neurotransmitter in the canine pyloric sphincter. Am J Physiol 1993;264:G975–G983. Tottrup A, Ny L, Alm P, Larsson B, Forman A, Andersson KE. The role of the L-arginine/nitric oxide pathway for relaxation of the human lower oesophageal sphincter. Acta Physiol Scand 1993; 149:451–459. Huang PL, Dawson TM, Bredt DS, Snyder SH, Fishman MC. Targeted disruption of the neuronal nitric oxide synthase gene. Cell 1993;75:1273–1286. Mearin F, Mourelle M, Guarner F, Salas A, Riveros-Moreno V, Moncada S, Malagelada JR. Patients with achalasia lack nitric oxide synthase in the gastro-oesophageal junction. Eur J Clin Invest 1993;23:724–728. Vanderwinden JM, Mailleux P, Schiffmann SN, Vanderhaeghen JJ, De Laet MH. Nitric oxide synthase activity in infantile hypertrophic pyloric stenosis. N Engl J Med 1992;327:511–515. Cajal SR. Histologie du syste`me nerveux de l’homme et des verte´bre´s. Volume 2. Paris: Maloine, 1911:891–942. Burns AJ, Lomax AEJ, Torihashi S, Sanders KM, Ward SM. Interstitial cells of Cajal mediate inhibitory neurotransmission in the stomach. Proc Natl Acad Sci USA 1996;93:12008–12013. Nishikawa S, Kusakabe M, Yoshinaga K, Ogawa M, Hayashi S-I, Kunisada T, Era T, Sakakura T, Nishikawa S-I. In utero manipulation of coat color formation by a monoclonal anti–c-kit antibody: two distinct waves of c-kit dependency during melanocyte development. EMBO J 1991;10:2111–2118. Ward SM, Xue C, Shuttleworth CWR, Bredt DS, Snyder SH, Sanders KM. NADPH diaphorase and nitric oxide synthase colocalization in enteric neurons of the canine proximal colon. Am J Physiol 1992;263:G277–G284. Matini P, Faussone-Pellegrini M-S. Ultrastructural localization of neuronal nitric oxide synthase-immunoreactivity in the rat ileum. Neurosci Lett 1997;229:45–48. Burns AJ, Herbert TM, Ward SM, Sanders KM. Interstitial cells of Cajal in the guinea pig gastrointestinal tract as revealed by c-Kit immunohistochemistry. Cell Tissue Res 1997;290:11–20. Sanders KM. A case for interstitial cells of Cajal as pacemakers
August 1998
23.
24.
25.
26.
27.
28.
29.
30.
31.
and mediators of neurotransmission in the gastrointestinal tract. Gastroenterology 1996;111:492–515. Daniel EE, Posey-Daniel V. Neuromuscular structures in opossum esophagus: role of interstitial cells of Cajal. Am J Physiol 1984;246:G305–G315. Xue C, Pollock J, Schmidt HHHW, Ward SM, Sanders KM. Expression of nitric oxide synthase by interstitial cells of the canine proximal colon. J Auton Nerv Syst 1994;49:1–14. Publicover NG, Hammond EM, Sanders KM. Amplification of nitric oxide signalling by interstitial cells. Proc Natl Acad Sci USA 1993;90:2087–2091. Malysz J, Thuneberg L, Mikkelsen HB, Huizinga JD. Action potential generation in the small intestine of W mutant mice that lack interstitial cells of Cajal. Am J Physiol 1996;271:G387– G399. Torihashi S, Ward SM, Nishikawa S-I, Nishi K, Kobayashi S, Sanders KM. c-kit–dependent development of interstitial cells and electrical activity in the murine gastrointestinal tract. Cell Tissue Res 1995;280:97–111. Maeda H, Yamagata A, Nishikawa S, Yoshinaga K, Kobayashi S, Nishi K, Nishikawa S. Requirement of c-kit for development of intestinal pacemaker system. Development 1992;116:369– 375. Ward SM, Harney SH, Bayguinov JR, Sanders KM. Development of electrical rhythmicity in the murine gastrointestinal tract is specifically encoded in the tunica muscularis. J Physiol 1997;505.1:241– 258. Singaram C, Sweet MA, Gaumnitz EA, Cameron AJ, Camilleri M. Peptidergic and nitrinergic denervation in congenital esophageal stenosis. Gastroenterology 1995;109:275–281. Kobayashi H, O’Briain DS, Puri P. Immunochemical characterization of neural cell adhesion molecule (NCAM), nitric oxide synthase, and neurofilament protein expression in pyloric muscle of patients with pyloric stenosis. J Pediatr Gastroenterol Nutr 1995;20:319–325.
ICC AND INHIBITORY NEUROTRANSMISSION
329
32. Abel RM. The ontogeny of the peptide innervation of the human pyloric sphincter, with special reference to understanding the aetiology and pathogenesis of infantile hypertrophic pyloric stenosis. J Pediatr Surg 1996;31:490–497. 33. Lefebvre RA. Nitric oxide in the peripheral nervous system. Ann Med 1995;27:379–388. 34. Vanderwinden JM, Liu H, De Laet MH, Vanderhaeghen JJ. Study of the interstitial cells of Cajal in infantile hypertrophic pyloric stenosis. Gastroenterology 1996;111:279–288. 35. Faussone-Pellegrini MS, Cortesini C. The muscle coat of the lower esophageal sphincter in patients with achalasia and hypertensive sphincter. An electron microscopic study. J Submicrosc Cytol Pathol 1985;17:673–685. 36. Torihashi S, Yoshida H, Nishikawa S-I, Kunisada T, Sanders KM. Enteric neurons express steel factor-lacZ transgene in the murine gastrointestinal tract. Brain Res 1996;738:323–328. 37. Ward SM, Burns AJ, Torihashi S, Harney SC, Sanders KM. Impaired development of interstitial cells and intestinal electrical rhythmicity in steel mutants. Am J Physiol 1995;269:C1577– C1585.
Received December 3, 1997. Accepted May 4, 1998. Address requests for reprints to: Kenton M. Sanders, Ph.D., Department of Physiology and Cell Biology, University of Nevada School of Medicine, Reno, Nevada 89557. e-mail: [email protected]; fax: (702) 784-6903. Supported by grant DK 40569 from the National Institute of Diabetes and Digestive and Kidney Diseases. Morphology was provided by a core laboratory supported by PO1 DK 41315. The authors thank Julia Bayguinov for technical assistance and Piers Emson at the Molecular Neuroscience Group in Cambridge, England, for the gift of anti-nNOS antibody.