SEARCH

SEARCH BY CITATION

Keywords:

  • cholinergic innervation;
  • deep muscular plexus;
  • enteric nerves;
  • human small intestine;
  • immuno-electronmicroscopy;
  • immunohistochemistry;
  • interstitial cells of Cajal;
  • nitrergic innervation;
  • nitric oxide synthase;
  • vesicular acetylcholine transporter

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Tissues
  6. Immunohistochemistry
  7. Quantification of Kit positive ICC-IM, VAChT and NOS positive nerves
  8. Conventional EM
  9. Immuno-EM
  10. Results
  11. ICC-DMP
  12. ICC-IM inside muscle bundles
  13. ICC-IM in and along septa
  14. Different distribution of ICC-IM and enteric nerves
  15. Innervation of smooth muscle not associated with ICC
  16. Presence of nNOS in ICC
  17. Discussion
  18. Innervation related to ICC-DMP
  19. No nNOS presence in ICC
  20. Innervation of ICC-IM
  21. Acknowledgments
  22. References

With functional evidence emerging that interstitial cells of Cajal (ICC) play a role in smooth muscle innervation, detailed knowledge is needed about the structural aspects of enteric innervation of the human gut. Conventional electronmicroscopy (EM), immunohistochemistry and immuno-EM were performed on the musculature of the distal human ileum focusing on ICC associated with the deep muscular plexus (ICC-DMP) and intramuscular ICC (ICC-IM). ICC-DMP could be identified by EM but not by c-Kit immunohistochemistry. Immuno-EM revealed that ICC-DMP were innervated by both cholinergic and nitrergic nerves, and were the only cells to possess specialized synapse-like junctions with nerve varicosities and gap junction contacts with smooth muscle cells. c-Kit positive ICC near the deep muscular plexus were not ICC-DMP, but ICC-IM located in septa. ICC-IM were innervated by both cholinergic and nitrergic nerves but without specialized contacts. Varicosities of both nerve types were also found scattered throughout the musculature without specialized contact with any ICC. No ICC showed immunoreactivity for neuronal nitric oxide synthase. As ICC-DMP form synapse-like junctions with cholinergic and nitrergic nerves and gap junction contacts with muscle cells, it is hypothfesized that ICC-DMP hold a specialized function related to innervation of smooth muscle of the human intestine.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Tissues
  6. Immunohistochemistry
  7. Quantification of Kit positive ICC-IM, VAChT and NOS positive nerves
  8. Conventional EM
  9. Immuno-EM
  10. Results
  11. ICC-DMP
  12. ICC-IM inside muscle bundles
  13. ICC-IM in and along septa
  14. Different distribution of ICC-IM and enteric nerves
  15. Innervation of smooth muscle not associated with ICC
  16. Presence of nNOS in ICC
  17. Discussion
  18. Innervation related to ICC-DMP
  19. No nNOS presence in ICC
  20. Innervation of ICC-IM
  21. Acknowledgments
  22. References

Interstitial cells of Cajal (ICC) have been implicated in the patho-physiology of several gastrointestinal motor disorders1–3 based on an abnormal distribution and/or morphological features.4 Most attention thus far has been given to ICC associated with the Auerbach's plexus (ICC-AP) of the stomach and small intestine, which function as pacemaker cells and are involved in controlling peristaltic activity. From animal studies it appears that ICC associated with the deep muscular plexus (ICC-DMP; a plexus just at the inside of the circular muscle adjacent to the submucosa of the small intestine) and the intramuscular ICC (ICC-IM; dispersed throughout the musculature in the oesophagus, stomach and colon) play a role in the innervation of the gut musculature.5–7 In the mouse, ICC-DMP in the small intestine8 and ICC-IM in the stomach9 and colon10 have synapse-like contacts with nerve varicosities and gap-junction contacts with smooth muscle cells. In vitro studies on innervation in W/Wv mice led to the conclusion that ICC-IM are essential for nitrergic and cholinergic innervation of the stomach9 and lower oesophagus sphincter (LES).6 Subsequent in vivo studies showed that in the absence of ICC-IM, normal oesophageal peristalsis and substantial swallow-induced LES relaxation occurred indicating that direct nitrergic innervation to smooth muscle cells is present in W/Wv mice.11 Hence, although the in vitro evidence and structural data indicate a role for ICC in innervation, its specific role is still to be elucidated. The first step in the evaluation of a possible role of ICC in innervation of the human small intestine is structural information on the relationships between enteric nerves and the different types of ICC. Such data are essential for the development of an understanding of the role of ICC in the physiology and pathophysiology of human gut innervation.

Studies on ICC of the human small intestine have been hampered by uncertainty about c-Kit immunoreactivity of ICC-DMP.4,12–18 We explored immuno-EM techniques to investigate differences in Kit immunoreactivity between the different subtypes of ICC and their relationships with cholinergic and nitrergic nerves. We observed that ICC-DMP form special synapse-like junctions with cholinergic and nitrergic nerves, and gap junctions with smooth muscle cells. ICC-IM, in contrast with animal models, does not form synapse like structures with enteric nerves. As nerve varicosities without contact with ICC were abundant, it is likely that different mechanisms of interaction between smooth muscle cells and enteric nerves exist.

Tissues

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Tissues
  6. Immunohistochemistry
  7. Quantification of Kit positive ICC-IM, VAChT and NOS positive nerves
  8. Conventional EM
  9. Immuno-EM
  10. Results
  11. ICC-DMP
  12. ICC-IM inside muscle bundles
  13. ICC-IM in and along septa
  14. Different distribution of ICC-IM and enteric nerves
  15. Innervation of smooth muscle not associated with ICC
  16. Presence of nNOS in ICC
  17. Discussion
  18. Innervation related to ICC-DMP
  19. No nNOS presence in ICC
  20. Innervation of ICC-IM
  21. Acknowledgments
  22. References

Freshly resected, uninvolved terminal ileum specimens were obtained from 12 adult patients (seven men, five women; aged 37–75 years) undergoing surgery because of colonic malignancies. Tissue was obtained from 5 to 10 cm oral to the ileocaecal junction. The patients had no other gastrointestinal disease. Tissues were handled according to institutional ethical guidelines.

Immunohistochemistry

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Tissues
  6. Immunohistochemistry
  7. Quantification of Kit positive ICC-IM, VAChT and NOS positive nerves
  8. Conventional EM
  9. Immuno-EM
  10. Results
  11. ICC-DMP
  12. ICC-IM inside muscle bundles
  13. ICC-IM in and along septa
  14. Different distribution of ICC-IM and enteric nerves
  15. Innervation of smooth muscle not associated with ICC
  16. Presence of nNOS in ICC
  17. Discussion
  18. Innervation related to ICC-DMP
  19. No nNOS presence in ICC
  20. Innervation of ICC-IM
  21. Acknowledgments
  22. References

After resection, 1–2 cm pieces of ileum were cut and washed in Krebs Ringer bicarbonate solution (KRB) and pinned onto the base of a Sylgard dish with the serosal side facing up. After immersion in modified Zamboni's fixative containing 4% paraformaldehyde and 0.2% picric acid in 0.1 mol L−1 phosphate buffer (PB, pH 7.4) for 2–3 h at 4 °C, tissues were rinsed in 0.1 mol L−1 PB and dehydrated in 20% sucrose in PB overnight. On the second day tissues were embedded in Tissue-Tek (Miles Lab., Haperville, Illinois, USA) and frozen in isopentane, in a beaker submerged in liquid nitrogen. Frozen sections of 6–8 μm were cut with a cryostat and mounted on coated slides. Frozen sections were washed for 30 min in phosphate buffered saline (PBS; 0.05 mol L−1, pH 7.4 with 0.3% Triton X 100). Non-specific antibody binding was reduced by incubating the tissues in 1% bovine serum albumin (BSA) for 1 h at room temperature before addition of the primary antibodies. Tissues were then incubated with anti-c-Kit, anti-nitric oxide synthase (NOS) or anti-vesicular acetylcholine transporter (VAChT) for 48 h at 4 °C (Table 1). Secondary immunoreactions were carried out with Vectastain ABC kits with biotinylated anti-rabbit or anti-goat IgGs. All the secondary antibodies were from Vector Laboratories (Burlingame, CA, USA). 3,3′ diaminobenzidine (0.05% DAB plus 0.01% H2O2 in 0.05 mol L−1 Tris buffer saline, pH 7.6) was used as a peroxidase substrate.

Table 1.  Details of antibodies used for immunohistochemistry
AntibodiesSourceMono/poly-clonal antibodiesHostDilution
Anti-Kit (C-19)Santa Cruz Biotech, Santa Cruz, CA, USAPoly-Rabbit1 : 500
MBL, Nagoya, JapanPoly-Rabbit1 : 300
DAKO, Glostrup, DenmarkPoly-Rabbit1 : 100
Anti-NOSChemicon, Temecula, CA, USAPoly-Rabbit1 : 2500
Anti-VAChTTransduction Lab, Lexington, KY, USAPoly-Goat1 : 50

Control tissues were prepared by omitting primary antibodies from the incubation solutions. All the antisera were diluted with 0.3% Triton X 100 in 0.05 mol L−1 PBS (PBS–TX, pH 7.4) except for the c-Kit antibody, which was diluted with 1% normal goat serum and 10% human serum in PBS–TX. Tissues were examined with a conventional microscope with an attached digital camera (Sony 3CCD, Model no. DXC-930, Japan).

Quantification of Kit positive ICC-IM, VAChT and NOS positive nerves

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Tissues
  6. Immunohistochemistry
  7. Quantification of Kit positive ICC-IM, VAChT and NOS positive nerves
  8. Conventional EM
  9. Immuno-EM
  10. Results
  11. ICC-DMP
  12. ICC-IM inside muscle bundles
  13. ICC-IM in and along septa
  14. Different distribution of ICC-IM and enteric nerves
  15. Innervation of smooth muscle not associated with ICC
  16. Presence of nNOS in ICC
  17. Discussion
  18. Innervation related to ICC-DMP
  19. No nNOS presence in ICC
  20. Innervation of ICC-IM
  21. Acknowledgments
  22. References

To compare Kit immunoreactivity in longitudinal and circular muscle layers, Kit positive ICC-IM were quantified using tissues from 10 individuals. In addition, the distribution of ICC within the longitudinal muscle layer was examined. Quantification was performed using the KS400 program (Zeiss, Jena, Germany). Kit immunopositive cells were identified and highlighted using density slicing on colour scale images. The area of immunopositive cells on each picture was measured and expressed as percentage of total area. Regions with increased background staining, occasionally found at the borders of the image, were excluded from the analysis. Due to stronger staining and different shape, mast cells were easily distinguished from ICC and excluded before the analysis.

Conventional EM

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Tissues
  6. Immunohistochemistry
  7. Quantification of Kit positive ICC-IM, VAChT and NOS positive nerves
  8. Conventional EM
  9. Immuno-EM
  10. Results
  11. ICC-DMP
  12. ICC-IM inside muscle bundles
  13. ICC-IM in and along septa
  14. Different distribution of ICC-IM and enteric nerves
  15. Innervation of smooth muscle not associated with ICC
  16. Presence of nNOS in ICC
  17. Discussion
  18. Innervation related to ICC-DMP
  19. No nNOS presence in ICC
  20. Innervation of ICC-IM
  21. Acknowledgments
  22. References

Tissues were fixed with 2% paraformaldehyde, 2.5% glutaraldehyde, 3% sucrose and 1.25 mmol L−1 CaCl2 in 0.05 mol L−1 cacodylate buffer (pH 7.4) at 4 °C for 2 h. They were then postfixed in 2% osmium tetroxide (OsO4) for 1 h, stained en bloc with 2% aqueous uranyl acetate for 40 min, dehydrated, infiltrated and embedded in Epon-Araldite resin. Ultrathin sections were cut parallel to the circular muscle layer and stained with lead citrate for 5 min before viewing with a transmission electron microscope (Jeol 1200EX Biosystem, Tokyo, Japan).

Immuno-EM

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Tissues
  6. Immunohistochemistry
  7. Quantification of Kit positive ICC-IM, VAChT and NOS positive nerves
  8. Conventional EM
  9. Immuno-EM
  10. Results
  11. ICC-DMP
  12. ICC-IM inside muscle bundles
  13. ICC-IM in and along septa
  14. Different distribution of ICC-IM and enteric nerves
  15. Innervation of smooth muscle not associated with ICC
  16. Presence of nNOS in ICC
  17. Discussion
  18. Innervation related to ICC-DMP
  19. No nNOS presence in ICC
  20. Innervation of ICC-IM
  21. Acknowledgments
  22. References

Tissues prepared for immunocytochemistry were cut into small pieces and fixed by immersion in 4% paraformaldehyde, 0.1% glutaraldehyde and 0.2% picric acid in 0.1 mol L−1 PB (pH 7.4) for 2–3 h at room temperature. After a brief rinse in 0.1 mol L−1 PB, tissues was washed vigorously at room temperature in several changes of 50% ethanol until the picric acid staining of the tissue had disappeared (about 20–30 min). The tissue was then washed in 0.1 mol L−1 PB and incubated in 0.1% NaBH3 CN (Aldrich Chemical Co., Milwaukee, WI, USA) in 0.1 mol L−1 PB for 30 min at room temperature. After washing in PB several times, small pieces of tissues were glued on wax blocks. Cross vibra-sections of 150 μm were cut with Lancer vibratome (series 1000) (Lancer, St. Louis, Missouri, USA). After blocking non-specific binding with 1% BSA for 1 h at room temperature, all the vibra-sections were incubated with the same primary antibodies (Table 1) as those for immunohistochemistry with the same time and dilution. On the third day, after washing in PBS several times, the same secondary antibody, ABC reagent and peroxidase substrate were used for the subsequent immunoreaction. All the antisera for immuno-EM were diluted with 0.05 mol L−1 PBS (without Triton X-100, pH 7.4). Tissues were continuously checked under the light microscope for suitable reaction, prior to being postfixed, block-stained, dehydrated, embedded and grid-stained for EM studies.

ICC-DMP

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Tissues
  6. Immunohistochemistry
  7. Quantification of Kit positive ICC-IM, VAChT and NOS positive nerves
  8. Conventional EM
  9. Immuno-EM
  10. Results
  11. ICC-DMP
  12. ICC-IM inside muscle bundles
  13. ICC-IM in and along septa
  14. Different distribution of ICC-IM and enteric nerves
  15. Innervation of smooth muscle not associated with ICC
  16. Presence of nNOS in ICC
  17. Discussion
  18. Innervation related to ICC-DMP
  19. No nNOS presence in ICC
  20. Innervation of ICC-IM
  21. Acknowledgments
  22. References

ICC-DMP could not be identified by Kit immunohistochemistry using antibodies from Santa Cruz Biotechnology (Santa Cruz, CA, USA), MBL (Nagoya, Japan) and Dako (Glostrup, Denmark). Some positive cells in this area were actually ICC-IM within the septa of circular muscle bundles (Fig. 1). Mast cells were present occasionally at any level of the intestinal musculature and easily identified as small round dark c-Kit positive cells, with a very dark smooth thin membrane contour.

image

Figure 1. Kit immunopositive cells at the submucosal border. c-Kit immunostaining on sections transversely cut through the circular muscle (CM) layer reveal ICC-IM in the longitudinal muscle (LM) (lengthwise), circular muscle (cross sectioned) and ICC-IM in the septa (lengthwise). ICC-IM are also observed at the inner border of the circular muscle layer adjacent to the submucosa (arrows) where ICC-DMP would be expected to reside. Arrowheads: Kit positive ICC-IM within CM and LM.

Download figure to PowerPoint

As ICC-DMP were Kit negative, immuno-EM was applied to identify ICC by their ultrastructural features and to study associated nerve structures by immuno-staining with different neurotransmitter markers. Immuno-EM revealed an abundance of ICC-DMP that were electron-microscopically identifiable but c-Kit negative. Fig. 2A shows a c-Kit negative ICC-DMP together with an intensely stained ICC-IM outside the circular muscle bundle. The ICC-DMP typically surrounds a nerve fiber in the deep muscular plexus. Although tissue prepared for immuno-EM may give poor resolution, with the modified method utilized in this study, the ICC displayed all the structural features necessary for cell identification (Fig. 2B). Immuno-EM revealed abundant intimate contacts between ICC-DMP and enteric nerves that were either VAChT or NOS immunoreactive (Figs 3A,C and 4A,B). Most of the contacts were between ICC-DMP and nerve fibers with multiple varicosities (Fig. 4A,B) but contacts with individual nerve varicosities were also prominent (Fig. 3A,B). Many special synapse-like junctions were present between the membranes of ICC-DMP and neighbouring enteric nerves (Figs 5A,B and 6A,B), with increased membrane density in the pre-junctional area (Figs 5A and 6B). The distance between the two adjacent membranes was as close as 25 nm, with an average of 26.9 ± 0.43 nm (n = 14). Numerous gap-junction contacts were identified between multiple processes of ICC-DMP and smooth muscle cells of the outer circular muscle layer (Figs 5A,B and 7).

image

Figure 2. Distinguishing ICC-IM and ICC-DMP at the inner border of the circular muscle layer. (A) Electron microscopy studies on tissue subjected to c-Kit immunostaining distinguished ICC-IM from ICC-DMP. The c-Kit positive ICC (white*) in the region of the deep muscular plexus were ICC-IM without close association with the deep muscular plexus nerve structures. ICC-DMP (black*) were c-Kit negative and closely associated with nerves. (B) Enlarged picture showing the ultrastructure of ICC-DMP in figure (A). The ICC-DMP were characterized by the presence of many mitochondria (m) and caveoli (small arrows) and a distinct basal lamina (small arrow heads).

Download figure to PowerPoint

image

Figure 3. Intimate contact between ICC-DMP and single cholinergic and nitrergic nerve varicosities. An ICC-DMP (A,*) and a process of an ICC-DMP (B,*) were closely associated with an isolated cholinergic (A, white arrow) and nitrergic (B, large black arrow) nerve varicosity. In Fig. B, another ICC-DMP cell body (**) was nearby. All small arrows, caveoli; f, high density of intermediate filaments; m, mitochondria; all small arrowheads, the basal lamina.

Download figure to PowerPoint

image

Figure 4. Interstitial cells of Cajal-DMP closely connected with VAChT and NOS positive varicosities within the enteric nerve bundles. An ICC-DMP (*) between the inner (ICM) and outer circular muscle (OCM) layer was intimately contacted by individual cholinergic (A, white arrow) or nitrergic (B, white arrow) varicosities within nerve bundles. Small arrows, caveoli; f, high density of intermediate filaments.

Download figure to PowerPoint

image

Figure 5. Interstitial cells of Cajal-DMP form special close apposition junctions (25 nm) with nerves including increased pre-junctional membrane density in the varicosities. (A) An ICC-DMP (white*) intercalated between an individual nerve varicosity (N, enlarged in left bottom square) and an outer circular smooth muscle cell (OCM, enlarged in right top square). The ICC-nerve junction is 25 nm wide (measured at larger magnification) and shows increased pre-junctional membrane density at the varicosity side (inset white arrow). Gap junctions occurred between ICC-DMP and the outer circular muscle cells (inset arrow). (B) A process of an ICC-DMP (white*) was closely contacted by both a nerve varicosity (N) and OCM cells. Gap junctions connect smooth muscle cells with ICC-DMP (arrows). ICM, inner circular muscle; small arrows, caveoli; m, mitochondria; small arrowheads, basal lamina.

Download figure to PowerPoint

image

Figure 6. Special synapse-like junctions between ICC-DMP and enteric nerves in the deep muscular plexus. (A) An ICC-DMP (*) between the inner (ICM) and outer circular muscle (OCM) layers was close to a nerve varicosity (N) in the deep muscular plexus. (B) Enlarged square of (A). Arrow shows increased pre-junctional membrane density in the nerve varicosity. f, abundant intermediate filaments; ER, endoplasmic reticulum; m, mitochondria; small arrows, caveoli; small arrowheads, basal lamina.

Download figure to PowerPoint

image

Figure 7. Gap-junctional connection between ICC-DMP and outer circular muscle cells. An ICC-DMP (*) located between enteric nerves (N) in the deep muscular plexus and outer circular muscle cells (OCM). Insert photo: enlarged area from the square shows the gap-junctional contact between ICC-DMP and smooth muscle cells. ICM, inner circular muscle cells; SM, submucosa; small arrows, caveoli.

Download figure to PowerPoint

ICC-IM inside muscle bundles

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Tissues
  6. Immunohistochemistry
  7. Quantification of Kit positive ICC-IM, VAChT and NOS positive nerves
  8. Conventional EM
  9. Immuno-EM
  10. Results
  11. ICC-DMP
  12. ICC-IM inside muscle bundles
  13. ICC-IM in and along septa
  14. Different distribution of ICC-IM and enteric nerves
  15. Innervation of smooth muscle not associated with ICC
  16. Presence of nNOS in ICC
  17. Discussion
  18. Innervation related to ICC-DMP
  19. No nNOS presence in ICC
  20. Innervation of ICC-IM
  21. Acknowledgments
  22. References

ICC-IM were found within both circular (Fig. 8A,C) and longitudinal muscle layers (Fig. 8D,E) concentrated in the region bordered by Auerbach's plexus. ICC-IM were bipolar in shape and aligned with surrounding smooth muscle cells. Some of the processes may have been part of the ICC-AP network, but numerous independent c-Kit positive ICC cell bodies were present, separate from Auerbach's plexus (Fig. 8A,D). Immuno EM revealed that ICC-IM cell bodies ultrastructurally resembled ICC-AP (Fig. 8B,C,E) but that many c-Kit positive processes were ultrastructurally heterogeneous on ultra thin profiles (Fig. 9A,B). Some of them were similar to nearby smooth muscle cells (Fig. 9B). A low density of organelles and difficulty in identifying caveoli because of immune reaction allowed many ICC, especially ICC processes, to be identified only by Kit staining and not by ultrastructural criteria. Many ICC-IM were found surrounding nerve structures within the muscle layer (Fig. 9C). No specialized contacts were observed between nerves and these ICC-IM. The distance between nerves and these ICC was always >100 nm. Nitrergic and cholinergic nerve fibres and varicosities were scattered throughout the muscle layers with equal density in the whole musculature (Fig. 10A,D), hence there was no special distributional relationship with ICC (Table 2). Specialized contacts between nitrergic or cholinergic nerves and ICC-IM (Fig. 10B,E) or between nerves and smooth muscle (Fig. 10C,F) were not found.

image

Figure 8. Abundant ICC-IM in circular (A, B and C) longitudinal muscle (D and E) identified by EM features were c-Kit positive. A–C: Two c-Kit positive ICC (in rectangular boxes), one located within the circular muscle layer (CM), another near Auerbach plexus (AP). B and C showed enlarged ICC-IM (B) and ICC-AP (C). They both share similar ultrastructural features, such as a limited perinuclear region and more abundant subcellular organelles than the neighbouring smooth muscle cells. D and E: An ICC-IM located in the septa between two longitudinal muscle bundles (LM). Nearby is a process of a fibroblast-like cell (F). Enlarged section (D) shows subcellular organelles in the cytoplasm of ICC-IM. m, mitochondria; small arrow, caveoli; arrows, endoplasmic reticulum.

Download figure to PowerPoint

image

Figure 9. Many ICC-IM or their processes can be identified by c-Kit immuno- reactivity but not by EM features. (A) A c-Kit positive process (white*) in the circular muscle (CM) layer showing typical ICC features: a rich complement of subcellular organelles including mitochondria and endoplasmic reticulum. (B) A c-Kit positive process (white*) with a filament structure similar to nearby smooth muscle cells. (C) A nerve bundle (N) in the circular muscle layer surrounded by c-Kit positive processes (arrows), one of which was closely contacted by a macrophage (MC).

Download figure to PowerPoint

image

Figure 10. Distribution of cholinergic and nitrergic nerves related to ICC-IM and smooth muscle cells. A and D, light microscopy. NOS positive (A) and VAChT positive (D) nerves were seen to be present in the Auerbach's plexus (AP), and evenly dispersed throughout the longitudinal (LM, arrows) and circular (CM, arrowheads) muscle layer with many lining the border between lamella and septa. B, C, E and F: immuno-electron microscopy. ICC-IM (*) were close to the nerve bundles (N) with either NOS (B) and VAChT (E) immunoreactivities, but no specialized junctions were observed. Scattered throughout the muscle layers, without ICC contact, were nerves with either NOS (C) or VAChT (F) immunoreactivities. No specialized connections were found between smooth muscle cells (CM) and nerves (N).

Download figure to PowerPoint

Table 2.  NOS and VAChT are not distributed equally to kit
Antibodiesc-KitNOSVAChT
  1. Percentage of immuno-positive areas of Kit, NOS and VAChT on sections of the inner and outer parts of the longitudinal muscle layer (n = 10).

  2. *Significantly different from inner LM (P < 0.0006).

RegionInner LMOuter LMInner LMOuter LMInner LMOuter LM
Percentage ± Standard error7.76 ± 1.023.41 ± 0.48*9.15 ± 1.258.43 ± 1.268.13 ± 1.207.83 ± 1.19

ICC-IM in and along septa

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Tissues
  6. Immunohistochemistry
  7. Quantification of Kit positive ICC-IM, VAChT and NOS positive nerves
  8. Conventional EM
  9. Immuno-EM
  10. Results
  11. ICC-DMP
  12. ICC-IM inside muscle bundles
  13. ICC-IM in and along septa
  14. Different distribution of ICC-IM and enteric nerves
  15. Innervation of smooth muscle not associated with ICC
  16. Presence of nNOS in ICC
  17. Discussion
  18. Innervation related to ICC-DMP
  19. No nNOS presence in ICC
  20. Innervation of ICC-IM
  21. Acknowledgments
  22. References

ICC-IM were also found within and lining the septa, usually positioned perpendicular to the surrounding muscle cells, which is in contrast to the ICC-IM within the muscle layers that were running parallel to the adjacent smooth muscle cells (Fig. 11A,B). ICC-IM are almost always located near enteric nerves, but no specialized contacts were observed with either nitrergic or cholinergic nerves, with distances always exceeding 100 nm (Fig. 11B). Septa-associated ICC-IM were also observed at the level of the deep muscular plexus (Figs 1 and 2A); their location (within the septa), running orientation (perpendicular to the circular muscle cells) and EM features (very few caveoli) indicated that they were not ICC-DMP. No specialized contacts occurred between these c-Kit positive ICC-IM and neighbouring nerves in the deep muscular plexus. Some ICC-IM formed close membrane to membrane contacts with macrophages (Fig. 9C) or fibroblasts (not shown).

image

Figure 11. Immuno-EM identifying the orientation of cell bodies and processes of ICC-IM in muscle and septa. (A) ICC-IM cell bodies; (B) ICC-IM processes. Kit positive ICC-IM inside the muscle bundles ran parallel to circular muscle cells (*). ICC-IM situated in septa between muscle bundles ran perpendicular to the muscle cells (**). They were both close to enteric nerves (N) but without specialized contacts.

Download figure to PowerPoint

Different distribution of ICC-IM and enteric nerves

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Tissues
  6. Immunohistochemistry
  7. Quantification of Kit positive ICC-IM, VAChT and NOS positive nerves
  8. Conventional EM
  9. Immuno-EM
  10. Results
  11. ICC-DMP
  12. ICC-IM inside muscle bundles
  13. ICC-IM in and along septa
  14. Different distribution of ICC-IM and enteric nerves
  15. Innervation of smooth muscle not associated with ICC
  16. Presence of nNOS in ICC
  17. Discussion
  18. Innervation related to ICC-DMP
  19. No nNOS presence in ICC
  20. Innervation of ICC-IM
  21. Acknowledgments
  22. References

The densities of ICC-IM in circular and longitudinal muscle layers were similar. In the circular muscle layer, ICC-IM occupied 6.06 ± 0.70% of the total area (mean ± SE, n = 10), while in longitudinal muscle layer this was 5.92 ± 0.67% (mean ± SE, n = 10). These values were not significantly different (P = 0.44). The distribution of ICC-IM showed a different pattern compared with that of cholinergic and nitrergic nerve varicosities. Interstitial cells of Cajal-IM (those inside the muscle bundle) concentrated in the outer part of the circular and inner part of the longitudinal muscle layers, while both cholinergic and nitrergic nerves were evenly scattered throughout the whole musculature. Table 2 shows the percentages of reactive areas of c-Kit, NOS and VAChT on the sections of both inner and outer parts of the longitudinal muscle layer. The density of ICC-IM in the inner part was significantly higher than in the outer part of the longitudinal muscle layer (P < 0.0006), while there were no significant differences between the distribution of both cholinergic (P = 0.35) and nitrergic nerves (P = 0.43) in the inner and outer parts of longitudinal muscle layer.

Innervation of smooth muscle not associated with ICC

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Tissues
  6. Immunohistochemistry
  7. Quantification of Kit positive ICC-IM, VAChT and NOS positive nerves
  8. Conventional EM
  9. Immuno-EM
  10. Results
  11. ICC-DMP
  12. ICC-IM inside muscle bundles
  13. ICC-IM in and along septa
  14. Different distribution of ICC-IM and enteric nerves
  15. Innervation of smooth muscle not associated with ICC
  16. Presence of nNOS in ICC
  17. Discussion
  18. Innervation related to ICC-DMP
  19. No nNOS presence in ICC
  20. Innervation of ICC-IM
  21. Acknowledgments
  22. References

In the inner circular muscle and outer longitudinal muscle, where ICC density was low, the density of cholinergic and nitrergic nerves was similar to that in the outer circular and inner longitudinal muscle layer, where density of ICC-IM is high. Nerve varicosities of both types of nerves were ending in between muscle cells (Fig. 10C,F), but no specialized junctions were observed between either type of nerve endings and smooth muscle cells.

Innervation related to ICC-DMP

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Tissues
  6. Immunohistochemistry
  7. Quantification of Kit positive ICC-IM, VAChT and NOS positive nerves
  8. Conventional EM
  9. Immuno-EM
  10. Results
  11. ICC-DMP
  12. ICC-IM inside muscle bundles
  13. ICC-IM in and along septa
  14. Different distribution of ICC-IM and enteric nerves
  15. Innervation of smooth muscle not associated with ICC
  16. Presence of nNOS in ICC
  17. Discussion
  18. Innervation related to ICC-DMP
  19. No nNOS presence in ICC
  20. Innervation of ICC-IM
  21. Acknowledgments
  22. References

The present study reports on special synapse-like junctions between the ICC-DMP and cholinergic and nitrergic enteric nerves, i.e. close apposition (as close as 25 nm between pre- and post-junctional membrane) and increased density of pre-junctional membranes in the nerve varicosities. Such specialized contacts between ICC-DMP and enteric nerves, together with the fact that ICC-DMP make gap-junction contacts with neighbouring smooth muscle cells, makes the ‘nerve – ICC-DMP – smooth muscle unit’ unique in that similar structures were not observed with ICC-AP and ICC-IM. Such an arrangement is suggestive of ICC-DMP playing a role in cholinergic and nitrergic muscle innervation of ICC. Whether this is primarily to innervate ICC or whether this is a major pathway to innervation of smooth muscle is yet to be elucidated. The ‘bridge’ feature of ICC-DMP in the human small intestine is similar to corresponding structures in the murine small intestine, where ICC-DMP also interpose between enteric nerve varicosities and smooth muscle cells.8,21–23 Several studies have contributed to the ultrastructural characterization of human ICC-DMP.13,24,25 With respect to contact with nerves, Faussone-Pellegrini24 and Torihashi et al.13 reported close apposition and Rumessen et al.25 observed presynaptic densities in nerve varicosities facing the ICC-DMP. The present study shows structural data that combine both features to show the existence of specialized junctions. Furthermore, it shows that ICC-DMP form synapse-like structures with both cholinergic and nitrergic nerves.

In the present study, electron-microscopically identified ICC-DMP were Kit negative using immunostaining with the c-Kit antibody, which has been most often used to detect ICC in human tissue (C-19, Cat no. SC168, Santa Cruz Biotechnology). We also tested the c-Kit antibody from MBL and Dako on frozen sections. While all the three antibodies resulted in intense staining of ICC-AP and ICC-IM, neighbouring ICC-DMP did not show any staining. This is consistent with observations by Romert and Mikkelsen12 and Torihashi13 and earlier observations of Vanderwinden and co-workers.14 There is however discrepancy in the literature. ICC-DMP in one case of foetal tissue were seen positive13 using the monoclonal antibodies from Boeringer (Mannheim, Germany). Foetal ICC-DMP were seen negative with Santa Cruz SC-168 antibodies16 (Santa Cruz Biotech) and with IBL antibodies13 (Immuno Biological Lab, Fujioka, Japan). ICC-DMP were seen positive in tissue from 3 to 12-year old26 using Kit antibody (no. SC39, Santa Cruz). A case study on the human small intestine reported Kit positive ICC-DMP using the MBL antibody17 and ICC-DMP were also reported to be positive using the antibodies from Santa Cruz.18 Although there is no reason to assume that these are incorrect observations, it is difficult to discriminate between ICC-DMP and ICC-IM from the figures and data provided because of the lack of counter-staining. In any case, discrimination is best done by immuno-electron microscopy. At this moment it cannot be stated that ICC-DMP does not possess the Kit protein; it is possible that the receptor density is very low or that the Kit protein is of a different phenotype compared with that of ICC-IM and ICC-AP. In animal models, there is no discrepancy between staining of ICC-AP and ICC-DMP.21,23,27–29 Kit negativity should not be a reason for doubting the classification of the cells as ICC when electron microscopic characteristics are fully consistent with such a classification. The present study shows that in the human intestine, when ICC-DMP are thought to be Kit positive, evidence should be provided that Kit staining in the DMP area is indeed associated with ICC-DMP and not with closely situated ICC-IM.

Our fixation protocol for immuno-EM provided sufficient resolution to be able to positively identify different ICC subtypes based on their ultrastructural features. We observed strongly stained c-Kit positive cells near the deep muscular plexus and they would probably be classified as ICC-DMP with immuno-histochemical studies under the light microscope, but electron-microscopically they could be identified as ICC-IM associated with septa (see below).

No nNOS presence in ICC

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Tissues
  6. Immunohistochemistry
  7. Quantification of Kit positive ICC-IM, VAChT and NOS positive nerves
  8. Conventional EM
  9. Immuno-EM
  10. Results
  11. ICC-DMP
  12. ICC-IM inside muscle bundles
  13. ICC-IM in and along septa
  14. Different distribution of ICC-IM and enteric nerves
  15. Innervation of smooth muscle not associated with ICC
  16. Presence of nNOS in ICC
  17. Discussion
  18. Innervation related to ICC-DMP
  19. No nNOS presence in ICC
  20. Innervation of ICC-IM
  21. Acknowledgments
  22. References

In the mouse colon, ICC harbours nNOS and eNOS but not iNOS.30 The nNOS was also found in ICC of the rat ileum19 and dog colon.20 In the present study, nNOS was not observed in ICC of the human small intestine. In the canine colon, nitric oxide produced by ICC in response to increased intracellular calcium was shown to amplify the response to nitric oxide released from enteric inhibitory neurons.31 As nNOS was not observed in any of the ICC of the human small intestine, the mechanisms by which ICC transmit information from nerves to smooth muscle cells or vise versa are still to be elucidated.

Innervation of ICC-IM

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Tissues
  6. Immunohistochemistry
  7. Quantification of Kit positive ICC-IM, VAChT and NOS positive nerves
  8. Conventional EM
  9. Immuno-EM
  10. Results
  11. ICC-DMP
  12. ICC-IM inside muscle bundles
  13. ICC-IM in and along septa
  14. Different distribution of ICC-IM and enteric nerves
  15. Innervation of smooth muscle not associated with ICC
  16. Presence of nNOS in ICC
  17. Discussion
  18. Innervation related to ICC-DMP
  19. No nNOS presence in ICC
  20. Innervation of ICC-IM
  21. Acknowledgments
  22. References

ICC-IM were abundantly present in the circular muscle and in agreement with Rumessen et al.32 running parallel to smooth muscle cells and resembling ultrastructurally ICC-AP. In addition, we found ICC-IM to be abundantly present in the longitudinal muscle in contrast to previous suggestions that they were rare in this layer.12 ICC-IM were close to cholinergic and nitrergic nerves, although always separated by a gap of at least 100 nm. No specialized junctions formed with individual nerve varicosities. No junctions were found either between neighbouring ICC-IM nor between ICC-IM and smooth muscle cells. The presence of Kit positive ICC-IM within the muscle layers of the human small intestine, contrasts with most animal models where no ICC-IM are found in the small intestine.23,28,29 ICC-IM are common in the stomach9,33 and colon10,12, where they are evenly distributed in the muscle layers, make gap-junction contact with smooth muscle cells and have synapse-like communication with nerves.9,10,34 Because of these special structural features and because of emerging functional evidence, a special role for ICC-IM in stomach and colon innervation is proposed as discussed above. The present study suggests that ICC-IM in the human small intestine may not have a similar role in innervation because it lacks the typical structural feature of synapse-like contacts. Nevertheless, the ICC-IM are close to nerves. Functional studies will have to reveal the role of ICC-IM in the human small intestine motor control. If ICC-IM are intermediaries in innervating smooth muscle, then the innervation kinetics involving ICC-IM are probably slower compared to that of ICC-DMP.

ICC-IM were also present in the septal structures of both circular and longitudinal muscle layers. They were c-Kit positive and ran perpendicular to neighbouring smooth muscle cells. Rumessen and coworkers described these ICC using conventional EM.32 ICC-IM in septa were not associated with the enteric nerves in the deep muscular plexus. They were close to cholinergic and nitrergic nerves in the septa but without forming specialized junctions.

The structural characteristics of innervation are markedly heterogeneous in the gut musculature suggesting that several mechanisms exist by which smooth muscle–nerve communication occurs. Nerve varicosities can be found to have synapse-like contacts with ICC that have gap-junction contacts with smooth muscle cells at the same time (e.g. ICC-IM in mouse stomach and colon, ICC-DMP in human small intestine). Enteric nerve varicosities can be close to ICC without any specialized junctions (e.g. ICC-AP and ICC-IM in the human gut). We hypothesize that the ICC-DMP mediates a specialized function in neurotransmission. It is possible that ICC-DMP are a preferred pathway for cholinergic and nitrergic neurotransmission consistent with the interpretation of functional studies in the mouse where ICC-IM have similar ultrastructural features.9 Interstitial cells of Cajal are also proposed to play a role in stretch sensation,35 hence facilitating information transfer from smooth muscle cells to nerve structures. It is possible that the smooth muscle – ICC-DMP – nerve structure is designed to facilitate that process, because similar structural arrangements are seen for ICC-IM in the cat oesophagus36 and ICC-IM in the rat stomach37,38 where such a role is proposed.

Enteric nerve varicosities without special contact with any particular cell are abundant within the musculature. This is particularly evident in the human small intestine where the inner part of the CM and the outer part of the LM have few ICC yet abundant nerve varicosities (Table 2). There is no reason to assume that direct innervation of smooth muscle is not a prominent feature in the gut. Indeed, in vivo studies on mice that lack ICC-IM showed nitrergic involvement in oesophageal peristalsis and LES relaxation.11 Furthermore, although synapse-like structures between nerve varicosities and smooth muscle cells are not often described, they are found in the human stomach,39 the rat stomach,40 and the guinea pig ileum.41 Hence they may be more common than generally assumed, but far less abundant than synapse-like contact between nerve varicosities and ICC.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Tissues
  6. Immunohistochemistry
  7. Quantification of Kit positive ICC-IM, VAChT and NOS positive nerves
  8. Conventional EM
  9. Immuno-EM
  10. Results
  11. ICC-DMP
  12. ICC-IM inside muscle bundles
  13. ICC-IM in and along septa
  14. Different distribution of ICC-IM and enteric nerves
  15. Innervation of smooth muscle not associated with ICC
  16. Presence of nNOS in ICC
  17. Discussion
  18. Innervation related to ICC-DMP
  19. No nNOS presence in ICC
  20. Innervation of ICC-IM
  21. Acknowledgments
  22. References

This study was supported by an operating grant from the Canadian Institutes of Health Research (CIHR). Dr Wang was supported by a fellowship from the Canadian Association of Gastroenterology sponsored by the CIHR and Janssen-Ortho. The authors want to thank Dr Johan Söderholm for collaboration in obtaining human tissues and Dr Natalia Zárate for discussing the manuscript.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Tissues
  6. Immunohistochemistry
  7. Quantification of Kit positive ICC-IM, VAChT and NOS positive nerves
  8. Conventional EM
  9. Immuno-EM
  10. Results
  11. ICC-DMP
  12. ICC-IM inside muscle bundles
  13. ICC-IM in and along septa
  14. Different distribution of ICC-IM and enteric nerves
  15. Innervation of smooth muscle not associated with ICC
  16. Presence of nNOS in ICC
  17. Discussion
  18. Innervation related to ICC-DMP
  19. No nNOS presence in ICC
  20. Innervation of ICC-IM
  21. Acknowledgments
  22. References
  • 1
    Sanders KM. A case for interstitial cells of Cajal as pacemakers and mediators of neurotransmission in the gastrointestinal tract. Gastroenterology 1996; 111: 492515.
  • 2
    Huizinga JD. Neural injury, repair, and adaptation in the GI tract. IV. Pathophysiology of gastrointestinal motility related to interstitial cells of Cajal. Am J Physiol 1998; 275: G3816.
  • 3
    Timmermans JP. Interstitial cells of Cajal: is their role in gastrointestinal function in view of therapeutic perspectives underestimated or exaggerated? Folia Morphol (Warsz) 2001; 60: 19.
  • 4
    Vanderwinden JM, Rumessen JJ. Interstitial cells of Cajal in human gut and gastrointestinal disease. Microsc Res Tech 1999; 47: 34460.
  • 5
    Daniel EE, Posey Daniel V. Neuromuscular structures in opossum esophagus: role of interstitial cells of Cajal. Am J Physiol 1984; 246: G30515.
  • 6
    Ward SM, Morris G, Reese L, Wang XY, Sanders KM. Interstitial cells of Cajal mediate enteric inhibitory neurotransmission in the lower esophageal and pyloric sphincters. Gastroenterology 1998; 115: 31429.
  • 7
    Ward SM, Sanders KM. Interstitial cells of Cajal: primary targets of enteric motor innervation. Anat Rec 2001; 262: 12535.
  • 8
    Rumessen JJ, Thuneberg L, Mikkelsen HB. Plexus muscularis profundus and associated interstitial cells. II. Ultrastructural studies of mouse small intestine. Anat Rec 1982; 203: 12946.
  • 9
    Ward SM, Beckett EA, Wang X, Baker F, Khoyi M, Sanders KM. Interstitial cells of Cajal mediate cholinergic neurotransmission from enteric motor neurons. J Neurosci 2000; 20: 1393403.
  • 10
    Wang XY, Sanders KM, Ward SM. Relationship between interstitial cells of Cajal and enteric motor neurons in the murine proximal colon. Cell Tissue Res 2000; 302: 33142.
  • 11
    Sivarao DV, Mashimo HL, Thatte HS, Goyal RK. Lower esophageal sphincter is achalasic in nNOS−/− and hypotensive in W/Wv mutant mice. Gastroenterology 2001; 121: 3442.
  • 12
    Romert P, Mikkelsen HB. c-Kit immunoreactive interstitial cells of Cajal in the human small and large intestine. Histochem Cell Biol 1998; 109: 195202.
  • 13
    Torihashi S, Horisawa M, Watanabe Y. c-Kit immunoreactive interstitial cells in the human gastrointestinal tract. J Auton Nerv Sys 1999; 75: 3850.
  • 14
    Vanderwinden JM, Rumessen JJ, Liu H, Descamps D, De Laet MH, Vanderhaeghen JJ. Interstitial cells of Cajal in human colon and in Hirschsprung's disease. Gastroenterology 1996; 111: 90110.
  • 15
    Isozaki K, Hirota S, Miyagawa J, Taniguchi M, Shinomura Y, Matsuzawa Y. Deficiency of c-Kit+ cells in patients with a myopathic form of chronic idiopathic intestinal pseudo-obstruction. Am J Gastroenterol 1997; 97: 3324.
  • 16
    Wester T, Eriksson L, Olsson Y, Olsen L. Interstitial cells of Cajal in the human fetal small bowel as shown by c-Kit immunohistochemistry. Gut 1999; 44: 6571.
  • 17
    He CL, Soffer EE, Ferris CD, Walsh RM, Szurszewski JH, Farrugia G. Loss of interstitial cells of Cajal and inhibitory innervation in insulin-dependent diabetes. Gastroenterology 2001; 121: 42734.
  • 18
    Porcher C, Baldo M, Henry M, Orsoni P, Jule Y, Ward SM. Deficiency of interstitial cells of Cajal in the small intestine of patients with Crohn's disease. Am J Gastroenterol 2002; 97: 11825.
    Direct Link:
  • 19
    Matini P, Faussone-Pellegrini MS. Ultrastructural localization of neuronal nitric oxide synthase-immunoreactivity in the rat ileum. Neurosci Lett 1997; 229: 458.
  • 20
    Xue C, Pollock J, Schmidt HH, Ward SM, Sanders KM. Expression of nitric oxide synthase immunoreactivity by interstitial cells of the canine proximal colon. J Auton Nerv Syst 1994; 49: 114.
  • 21
    Zhou DS, Komuro T. Interstitial cells associated with the deep muscular plexus of the guinea-pig small intestine, with special reference to the interstitial cells of Cajal. Cell Tissue Res 1992; 268: 20516.
  • 22
    Komuro T, Seki K. Fine structural study of interstitial cells associated with the deep muscular plexus of the rat small intestine, with special reference to the intestinal pacemaker cells. Cell Tissue Res 1995; 282: 12934.
  • 23
    Wang XY, Sanders KM, Ward SM. Intimate relationship between interstitial cells of cajal and enteric nerves in the guinea-pig small intestine. Cell Tissue Res 1999; 295: 24756.
  • 24
    Faussone Pellegrini MS, Cortesini C. Some ultrastructural features of the muscular coat of human small intestine. Acta Anat (Basel) 1983; 115: 4768.
  • 25
    Rumessen JJ, Mikkelsen HB, Thuneberg L. Ultrastructure of interstitial cells of Cajal associated with deep muscular plexus of human small intestine. Gastroenterology 1992; 102: 5668.
  • 26
    Vanderwinden JM, Rumessen JJ, De Laet MH, Vanderhaeghen JJ, Schiffmann SN. CD34+ cells in human intestine are fibroblasts adjacent to, but distinct from, interstitial cells of Cajal. Lab Invest 1999; 79: 5965.
  • 27
    Seki K, Zhou DS, Komuro T. Immunohistochemical study of the c-kit expressing cells and connexin 43 in the guinea-pig digestive tract. J Auton Nerv Sys 1998; 68: 1827.
  • 28
    Takeda M, Takayama I, Terada N, Baba T, Ward SM, Ohno S, Fujino MA. Immunoelectron-microscopic study of Kit-expressing cells in the jejunum of wildtype and Ws/Ws rats. Cell Tissue Res 2001; 304: 2130.
  • 29
    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: 1120.
  • 30
    Vannucchi MG, Corsani L, Bani D, Faussone-Pellegrini MS. Myenteric neurons and interstitial cells of Cajal of mouse colon express several nitric oxide synthase isoforms. Neurosci Lett 2002; 326: 1915.
  • 31
    Publicover NG, Hammond EM, Sanders KM. Amplification of nitric oxide signaling by interstitial cells isolated from canine colon. Proc Natl Acad Sci U S A 1993; 90: 208791.
  • 32
    Rumessen JJ, Mikkelsen HB, Qvortrup K, Thuneberg L. Ultrastructure of interstitial cells of Cajal in circular muscle of human small intestine. Gastroenterology 1993; 104: 34350.
  • 33
    Burns AJ, Lomax AE, Torihashi S, Sanders KM, Ward SM. Interstitial cells of Cajal mediate inhibitory neurotransmission in the stomach. Proc Natl Acad Sci USA 1996; 93: 1200813.
  • 34
    Faussone-Pellegrini MS, Thuneberg L. Guide to the identification of interstitial cells of Cajal. Microsc Res Tech 1999; 47: 24866.
  • 35
    Rumessen JJ. Identification of interstitial cells of Cajal. Significance for studies of human small intestine and colon. Danish Medical Bulletin 1994; 94: 27593.
  • 36
    Huizinga JD, Berezin I, Valdez DT, Xue S, Diamant NE. Interstitial cells of Cajal as sensory receptors for vagal afferents in the cat esophagus. Neurogastroenterol Motil 1999; 11: 266.
  • 37
    Berthoud HR, Powley TL. Vagal afferent innervation of the rat fundic stomach: morphological characterization of the gastric tension receptor. J Comp Neurol 1992; 319: 26176.
  • 38
    Liu LWC, Ye J, Wang L, Tougas G, Perdue MH, Huizinga JD. Role of interstitial cells of Cajal as mechano-sensory receptors in the activation of gastric vagal afferent pathways. Gastroenterology 2002; 122: A158.
  • 39
    Faussone-Pellegrini MS, Pantalone D, Cortesini C. An ultrastructural study of the smooth muscle cells and nerve endings of the human stomach. J Submicrosc Cytol Pathol 1989; 21: 42137.
  • 40
    Mitsui R, Komuro T. Direct and indirect innervation of smooth muscle cells of rat stomach, with special reference to the interstitial cells of Cajal. Cell Tissue Res 2002; 309: 21927.
  • 41
    Klemm MF. Neuromuscular junctions made by nerve fibres supplying the longitudinal muscle of the guinea-pig ileum. J Auton Nerv Syst 1995; 55: 15564.