SEARCH

SEARCH BY CITATION

Keywords:

  • computed image analysis;
  • duodenum;
  • endocrine cells;
  • immunocytochemistry;
  • myotonic dystrophy;
  • rectum

Abstract.

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Objectives. To study intestinal endocrine cell types in patients suffering from myotonic dystrophy (MD) and diarrhoea.

Design. Comparative study between MD patients and matched controls.

Setting. Departments of Medicine, Central Hospital, Boden, and University Hospital, Umeå, Sweden.

Subjects. Ten patients with MD (four males and six females) and suffering from diarrhoea. Ten healthy volunteers served as controls for the duodenal study and 13 patients under investigation for rectal bleeding and with endoscopically normal mucosa were controls for the rectal study.

Measurements. The duodenal and rectal endocrine cell types were identified by immunohistochemical investigation and quantified by computed image analysis.

Results. The total endocrine cell area in the duodenum as demonstrated by chromogranin A-immunoreactivity was significantly increased in MD as compared with the controls (126 ± 58 vs. 48 ± 22 × 103 µm2 mm−2 in crypts and 230 ± 183 vs. 28 ± 22 in villi, respectively, < 0.01). The increase included all types of endocrine cells studied, namely those positive for serotonin, cholecystokinin (CCK)/gastrin, secretin, gastric inhibitory peptide (GIP) and somatostatin. In the rectum, the total endocrine cell area as determined by chromogranin A-immunoreactivity was also significantly increased, but there was no statistical difference between the controls and patients with respect to the area of serotonin-, peptide YY (PYY)-, pancreatic polypeptide (PP)- or somatostatin-immunoreactive cells.

Conclusions. The increase in endocrine cell area indicates a disturbed endocrine regulation of the gastrointestinal tract that may contribute to the development of gastrointestinal symptoms encountered in MD patients.


Introduction

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Gastrointestinal (GI) symptoms such as dysphagia, abdominal pain or diarrhoea are common amongst individuals suffering from myotonic dystrophy (MD). In a recent study, it was demonstrated that one-quarter of the patients considered their gastrointestinal problems to be the most disabling consequence of the disease [ 1]. The mechanisms involved in causing GI symptoms in MD patients are not known. Since MD affects not only skeletal muscles but also smooth muscles of the GI tract, an impaired motility has been proposed to be of major importance [ 2]. The gastrointestinal neuroendocrine peptides play an important role in the regulation of gastrointestinal motility, secretion and ion transport [ 3–5]. It is reasonable therefore to assume that the gastrointestinal neuroendocrine system may be affected in MD patients.

The aim of the present study was to investigate the various endocrine cell types of the intestine in patients with MD suffering from diarrhoea. The duodenum and rectum were chosen as representative sites for the small and large intestine. The choice was based on the fact that they contain almost all the intestinal endocrine cell types and in large numbers [ 6]. Furthermore, mucosal biopsies of the duotenum and rectum are easily accessed with a gastroduodenoscope and sigmoidoscope, respectively.

Materials and methods

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Patients

Ten patients with MD suffering from diarrhoea were investigated (four males and six females; mean age 37, range 24–48 years). All patients had a positive family history and typical clinical findings. They represented different clinical severities, from subtle symptoms in a full-time worker to an immobilized man with the congenital form of the disease. The characteristics and clinical details of the patients are summarized in Table 1. The patients were investigated with established clinical methods (blood chemistry, light microscopy of duodenal and rectal biopsies, determination of fat excretion in faeces, stool culture, breath tests for detection of small bowel bacterial overgrowth and the SeHCAT-test) and the main results are presented in Table 1.

Table 1.   Clinical and demographic data of patients with myotonic dystrophy Thumbnail image of

Duodenal biopsies were obtained from 10 healthy volunteers (four males and six females; mean age 36, range 22–64 years), who had given their written consent to the study and served as controls. These volunteers were free from gastrointestinal symptoms. Control rectal biopsy specimens were obtained from 13 patients (seven females and six males; mean age 66 years, range 31–86 years). They were routinely taken from normal rectal mucosa under endoscopic examination for rectal bleeding. In these patients, haemorrhoids or polyps were identified as the source of bleeding.

The investigation was approved by the Ethics Committee, Umeå University.

Gastrointestinal endoscopy

After an overnight fast, a gastroduodenal endoscopy was performed in patients and controls. During the endoscopy procedure, two to four forceps biopsies were obtained from the descending part of the duodenum, distal to the papilla of Vateri. Rectosigmoidoscopy was performed without any bowel preparation and three to four biopsies were taken from the dorsal wall about 12 cm from the anus.

Histopathological and immunocytochemical methods

Biopsy specimens were fixed in 4% buffered formaldehyde overnight, embedded in paraffin and cut in 5 μm sections. The sections were stained by haematoxylin–eosin, van Gieson stain and by immunocytochemical methods.

For immunocytochemical demonstration of different endocrine cell types, the avidin–biotin complex (ABC) method (Dako A/S, Glostrup, Denmark) was used. Briefly, each section was deparaffinized, hydrated, immersed in 0.01% hydrogen peroxide in tris-HCl buffer, pH 7.4, for 10 min to inhibit endogenous peroxidase activity. The section was washed three times with Tris buffer and treated with 1% bovine serum albumin for 30 min to block the non-specific binding sites. The specimens were incubated overnight with one of the following antisera: antichromogranin A, antiserotonin, antisomatostatin, and antiglucagon. In addition, the duodenal specimens were incubated with antigastrin/CCK (cholecystokinin), antisecretin, and anti-GIP (gastric inhibitory polypetide), and the rectal specimens with anti-PYY (peptide YY) and anti-PP (pancreatic polypeptide) (for details concerning the antisera used see Table 2). Sections were incubated for 30 min with biotinylated swine antirabbit IgG or biotinylated antimouse IgG diluted 1:200. They were then incubated with the avidin–biotin–peroxidase complex diluted 1:100 for 30 min. Development of the section was performed in 50 mL Tris buffer containing 10 μL of 30% H2O2 and 25 mg diaminobenzidine hydrochloride (DAB).

Table 2.   Detailed account of the antisera used Thumbnail image of

The specificity controls were similar to those described previously [ 7]. Briefly, controls were obtained by (1) replacing the primary antiserum/ antibody by normal rabbit serum (Tris buffer in the case of monoclonal antibodies), and (2) preincubating the antiserum/antibody with excess of the corresponding or structurally related antigen (75– 100 μg mL−1 diluted antiserum) for 24 h at 4 °C.

Morphometry

Mucosa. In the duodenum, the number of crypts and villi in controls and patients was estimated using a 10 × 10 mm square grid which was inserted in a 12.5× eyepiece. The baseline of the grid was aligned with the base of the crypts in three sections cut perpendicular to the mucosa from each individual. The number of crypts and villi in an area of 1 mm2 was counted using the 10× objective. The villus height was measured with a 1 mm scale inserted in a 12.5× eyepiece and a 10× objective.

In the rectum, the area of the mucosa in both the controls and patients was measured by Quantimet 500MC image processing and analysis system (Leica, Cambridge), connected to Olympus microscope type BX50. A 420 × 420 μm square frame and 10× objective was used. The baseline of the frame was aligned with the base of the crypts in three sections cut perpendicular to the mucosa from each individual. The area of the epithelial cells was determined as described above by using the interactive image editor.

Statistical analysis

Differences between the groups were analysed by means of the Mann–Whitney U-test. < 0.05 was regarded as significant.

Results

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Endoscopy and routine histology

The patients and the controls had normal endoscopic findings with respect to both stomach and duodenum. Histopathological examination of duodenal biopsies from patients and healthy controls revealed normal structure of villi and mucosa. The rectal biopsies were normal except for three patients and two controls where the mucosa displayed a slight inflammation.

Immunocytochemistry

Chromogranin A, serotonin- ( Fig. 1), gastrin/CCK C-terminus-, somatostatin-, secretin- and GIP-immunoreactive cells were detected in the descending part of the duodenum in both patients and controls. These cells showed the same pattern of distribution as described previously [ 6]. Thus, chromogranin A- and serotonin-immunoreactive cells were found both in villi and crypts. Gastrin/CCK and secretin-immunoreactive cells were confined almost exclusively to villous epithelium, whereas somatostatin- and GIP-immunoreactive cells were mainly localized within the crypts. The shape of these cells varied from flask-shaped with a thin luminal process to basket-shaped with a basal process running parallel to the basement membrane.

image

Figure 1.  Serotonin-immunoreactive cells in the duodenum of a patient with myotonic dystrophy (a) and of a control (b) (× 105).

Download figure to PowerPoint

In the rectum, chromogranin A, serotonin-, PYY-, PP-, somatostatin- and enteroglucagon-immunoreactive cells were found. They occurred mostly in the middle part of the crypts of Lieberkühn. Their shape also varied from basket-shaped to flask-shaped.

Specificity controls showed that the immunostaining was completely abolished after preincubation of the antisera with the corresponding peptide. Preincubation with structurally related peptides had no effect on the immunostaining. Replacing the antisera/antibodies with normal rabbit serum or Tris buffer gave no staining.

Morphometry

Mucosa. No statistically significant differences were found in the duodenal mucosa between controls and MD patients with respect to number of crypts, villi or villous height. The average number of crypts was 7.5 (range 7–10) and 8.0 mm−2 (range 7–9) for controls and patients, respectively. The corresponding figures for villi were 3.7 (range 3–5) and 3.6 mm−2 (range 3–4). The average height of the villi in the controls was 0.9 mm (range 0.7–1.1) and in the patients 0.9 mm (range 0.8–1.0). In the rectum there was no statistical difference in the epithelial cell area between controls and MD patients. The area of epithelial cells in the controls was 0.21 ± 0.034 (mean ± SD) mm2 mm−1 baseline and in patients 0.21 ± 0.036.

Endocrine cells. The duodenal endocrine cell area, as demonstrated by chromogranin A-immunoreactivity, was significantly increased in MD patients compared with controls. The increase included all the endocrine cell types studied, namely serotonin-, secretin-, gastrin/CCK-, GIP- and somatostatin-immunoreactive cells ( Fig. 2).

image

Figure 2.  Endocrine cell area expressed as 103 µm2 mm−2 of epithelial cell area mean ± SEM. Statistical difference between MD patients and controls; *< 0.05, **< 0.01, ***< 0.001. n, number of individuals; GIP, gastric inhibitory polypeptide; Somatostat, somatostatin; CCK, cholecystokinin; Chrom, chromogranin A; c, crypt; v, villus.

Download figure to PowerPoint

In the rectum the total endocrine cell area, as demonstrated by chromogranin A-immunoreactive cells, was also significantly increased in the MD patients compared with the controls. However, no statistical difference could be found regarding the individual peptides, i.e. serotonin-, PYY-, PP- or somatostatin-immunoreactive cells (Table 3). Enteroglucagon-immunoreactive cells were few in the randomly chosen fields in the rectal biopsies of both the controls and patients, which made it difficult to perform reliable statistical analysis.

Discussion

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Chromogranin A is used as a common marker for endocrine cells [ 8]. The area of the chromogranin A-immunoreactive cells was significantly increased in the duodenum and rectum of patients suffering from MD and diarrhoea compared with healthy controls, suggesting an increase in intestinal endocrine cells. An increase in the duodenal endocrine cell area could be demonstrated for all cell types studied, namely serotonin-, gastrin/CCK-, secretin-, GIP- and somatostatin-reactive cells. In the rectum, the cell types studied displayed a non-significant trend to be reduced compared with the controls, whereas the chromogranin A-immunoreactive cells were more common in the MD patients. This finding may indicate the presence of other endocrine cells that were not identified, thereby supporting the objections to using chromogranin A as a general marker for endocrine cells [ 9–11].

Several diseases have been found to be accompanied by impairment of the neuroendocrine system; thus endocrine cells have been investigated in both systemic diseases with gastrointestinal symptoms, i.e. familial amyloidic polyneuropathy (FAP) [ 12], and gastrointestinal diseases, i.e. coeliac disease [ 13] and inflammatory bowel disease (IBD) [ 14]. In FAP, a reduction of the intestinal endocrine cell content has been reported, and it has been proposed that this could contribute to the dysfunction of intestinal motility and maldigestion/malnutrition in these patients [ 12]. In patients with coeliac disease, increased levels of plasma PYY have been reported [ 15], a finding that may be explained by PYY acting as a mediator in fat-induced delay of gastric emptying and gut motility.

The finding of increased duodenal endocrine cells in patients with myotonic dystrophy and diarrhoea leads to two questions: what is the reason for this abnormality; and what is the consequence? The increase in the endocrine cell area could be either primary or secondary to gastrointestinal changes induced by MD. The increase in endocrine cell area may be caused by defect peptide synthesis, resulting in non-biological active peptides, reduced receptor sensitivity of the effector cells, or a defect or malfunctioning effector organ. In all these situations the feedback mechanism would lead to an increased production of the hormones. The assumptions of reduced receptor sensitivity and malfunctioning effector organ may be supported by previous findings that testicular atrophy in men with MD is associated with high FSH levels [ 16]. Moreover, MD patients have an increased insulin response following a glucose load [ 17], which has been interpreted as peripheral insulin resistance at the receptor level [ 18]. It has also been demonstrated that patients with MD have ACTH hypersecretion after CRH-mediated stimuli, as well as basal ACTH levels twice those of controls. However, neither the cortisol response nor the basal cortisol levels differed significantly [ 19]. The proposed mechanism was a defect in the intracellular pathway initiated by CRH interaction as a result of abnormal levels of cAMP-dependent kinase.

Could the increase in endocrine cells contribute to the GI symptoms? This is evident if the increase is a primary disturbance; however, if the increase is secondary to a malfunctioning effector organ, a cause–effect relationship could still exist since many of these peptides have diverse effects. An increase of the peptide activity because of impaired muscle function could possibly induce, for example, increased secretion at another effector organ. The patients have been investigated with respect to intestinal function ( Table 1), but no correlation could be found between the endocrine cell content and results of these tests or severity of disease.

Acknowledgement

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

This study was supported by grants from the county council of Norrbotten, Sweden, and the Swedish Medical Research Council (project no. 19X-11240).

References

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References
  • 1
    Rönnblom A, Forsberg H, Danielsson Å. Gastrointestinal symptoms in myotonic dystrophy. Scand J Gastroenterol 1996; 31: 654 7.
  • 2
    Nowak TW, Anuras S, Brown BP. Small intestinal motility in myotonic dystrophy. Gastroenterology 1984; 86: 808 13.
  • 3
    Debas TH & Mulvihill SJ. Neuroendocrine design of the gut. Am J Surg 1991; 161: 243 9.
  • 4
    Ekblad E, Håkansson R, Sundler F. Microanatomy and chemical coding of peptides-containing neurons in the digestive tract. In: Daniel E, ed. Neuropeptides Function in Gastrointestinal Tract. Boca Raton: CRC Press, 1991; 131 80.
  • 5
    Harvey RF. Hormonal control of gastrointestinal motility. Am J Dig Dis 1975; 20: 523 39.
  • 6
    Sjölund K, Sandén G, Håkansson R, Sundler F. Endocrine Cells in Human Intestine. An immunocytochemical study. Gastroenterology 1983; 85: 1120 30.
  • 7
    El-Salhy M, Stenling R, Grimelius L. Peptidergic innervation and endocrine cells in the human liver. Scand J Gastroenterol 1993; 28: 809 15.
  • 8
    Cetin Y. Chromogranin A immunoreactivity and Grimelius' argyrophilia. A correlative study in mammalian endocrine cells. Anat Embryol 1992; 185: 207 15.
  • 9
    Cetin Y, Muller-Koppel L, Aunis D, Bader MF, Grube D. Chromogranin A (Cg A) in the gastro-entero-pancreatic (GEP) endocrine system. II. Histochem 1989; 92: 265 75.
  • 10
    Bargsten G & Grube D. Serotonin storage in chromogranins: an experimental study in rat gastric endocrine cells. J Histochem Cytochem 1992; 40: 1147 55.
  • 11
    Hörsch D, Wiehe E, Müller S, Hancke E. Distribution and coexistence of chromogranin A-, serotonin- and pancreastatin-like immunoreactivity in endocrine-like cells of the human anal canal. Cell Tissue Res 1992; 268: 109 16.
  • 12
    El-Salhy M, Suhr O, Stenling R, Wilander E, Grimelius L. Impact of familial amyloid associated polyneuropathy on duodenal endocrine cells. Gut 1994; 35: 1413 8.
  • 13
    Pietroletti R, Bishop AE, Carlei F, Bonamico M, Lloyd RV, Wilson B et al Gut endocrine cell population in coeliac disease estimated by immunocytochemistry using a monoclonal antibody to chromogranin. Gut 1986; 27: 838 43.
  • 14
    Koch TR, Carney JA, Go VL. Distribution and quantitation of gut neuropeptides in normal intestine and inflammatory bowel diseases. Dig Dis Sci 1987; 32: 369 76.
  • 15
    Sjölund K & Ekman R. Increased plasma levels of peptide YY in coeliac disease. Scand J Gastroenterol 1988; 23: 297 300.
  • 16
    Harper PS, Penny R, Foley JRTP, Mignon CJ, Blizzard RM. Gonadal function in males with myotonic dystrophy. J Clin Endocrinol Metab 1972; 35: 852 6.
  • 17
    Huff TA, Horton ES, Lebowitz HE. Abnormal insulin secretion in myotonic dystrophy. N Engl J Med 1967; 277: 837 41.
  • 18
    Mishra SK, Chetty S, Kataoka M. Abnormal insulin receptor binding in cultured monocytes in myotonic dystrophy. Biochem Med Metab Biol 1992; 47: 161 7.
  • 19
    Grice JE, Jackson RV, Hockings GI, Torpy DJ, Walters MM, Crosbie GV. Adrenocorticotropin hyperresponse to the corticotropin-releasing hormone-mediated stimulus of naloxone in patients with myotonic dystrophy. J Clin Endocrinol Metab 1995; 80: 179 84.

Received 11 February 1998; accepted 9 June 1998.