- Email: [email protected]
Age-induced changes in the enteric nervous system in the mouse
Mechanisms of Ageing and Development 107 (1999) 93 – 103
Age-induced changes in the enteric nervous system in the mouse Magdy El-Salhy *, Olof Sandstro¨m, Frank Holmlund Section for Gastroenterology and Hepatology, Department of Medicine, Uni6ersity Hospital, S-901 85 Umea˚, Sweden Received 27 July 1998; received in revised form 20 October 1998; accepted 3 November 1998
Abstract The enteric nervous system of the murine gut was investigated by immunocytochemistry in 1-, 3-, 12- and 24-month-old mice, using protein gene product 9.5, a general marker for nerve elements. Myenteric and submucosal plexi were quantified by computerized image analysis. In antrum, there were significantly fewer neurones per ganglion in both myenteric and submucosal ganglia of 12- and 24-month-old mice than in 3-month-old animals. The same was true of duodenum and colon. The relative volume density of nerve fibres in antral muscularis propria was significantly greater in the 1-, 12- and 24-month-old mice than in the 3-month-old mice. In colon, there were fewer submucosal ganglia per millimetre baseline in 1-month-old mice than in 3-month-old mice. The colonic myenteric ganglion in 1-, 12- and 24-month-old mice was smaller than in 3-month-old mice. There was no statistical difference between females and males regarding the number of ganglia per millimetre baseline, ganglionic area, number of neurones per ganglion or the relative volume density of nerve fibres in either the myenteric or submucosal plexi. As the enteric nervous system is responsible for coordinating and integrating the motility of the gut, the ageing-related changes reported here may well be of some relevance for the increased gastrointestinal motility dysfunction in the elderly persons. © 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Aging; Computerized image analysis; Enteric nervous system; Gastrointestinal tract; Immunocytochemistry; Mice
* Corresponding author. Fax: + 46-90-143986; e-mail: [email protected]. 0047-6374/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 0 4 7 - 6 3 7 4 ( 9 8 ) 0 0 1 4 2 - 0
94
M. El-Salhy et al. / Mechanisms of Ageing and De6elopment 107 (1999) 93–103
1. Introduction Motility disorders of the gastrointestinal tract are known to become more common with advanced age (Burkitt et al., 1972; Hollis and Castell, 1974; Moore et al., 1983; McDougal et al., 1984; Bannister et al., 1987; Madsen, 1992). For instance, it has been reported that colonic transit is protracted in elderly humans (Madsen, 1992) and emptying of liquid of the stomach is tardy in the elderly (Moore et al., 1983). Delayed colonic transit has been also reported in old rats (McDougal et al., 1984). As it is the enteric nervous system of the gastrointestinal tract that coordinates and integrates the motility activities of the gut (Smout and Akkermans, 1992), several studies have been performed on this regulatory system to establish a plausible aetiology for the motility disorders in the elderly (Santer and Baker, 1988; Gabella, 1989; de Souza et al., 1993 Meciano et al., 1995; Gomes et al., 1997). The number of myenteric neurones has been reported to decrease with age in the human oesophagus and in small and large intestines (de Souza et al., 1993; Meciano et al., 1995; Gomes et al., 1997) The same age-related decrease in neurones has been found in the small intestine of the rat (Santer and Baker, 1988) and guinea-pig (Gabella, 1989). No data are available, however, concerning the effect of ageing on the submucosal plexus, which is also an important regulator of gastrointestinal motility by means of tactile or chemical stimulus of the mucosa (Materia et al., 1988; Tomlin and Read, 1988; Frexinos et al., 1989; Leng-Pechlow, 1989). Furthermore, disturbances in the enteric nervous system in the stomach are unknown. In order to fill these gaps in our knowledge, the present investigation was undertaken, in which conceivable age-related changes in both myenteric and submucosal plexi of the enteric nervous system were studied in different parts of the gastrointestinal tract of an animal model, namely the mouse. For this purpose, a general marker for nerve elements, protein gene product 9.5 (PGP 9.5), was used (Krammer et al., 1993, 1994). 2. Materials and methods
2.1. Animals NMRI/Bom mice (Bomholtga˚rd Breeding and Research Centre, Denmark) were used. Males and females were housed separately in cages, five to a cage, in a room with a 12/12-h light/dark cycle, and fed a standard pellet diet (Astra-Ewos AB, So¨derta¨lje, Sweden) with free access to water. Ten mice (five males, five females) were sacrificed when 1 month old, ten more at 3 months and another ten at 12 months. Six 24-month-old males were then also killed. The animals were starved overnight, before sacrifice by cervical dislocation. Antrum, proximal duodenum and distal colon were excised. The tissue specimens were fixed overnight in 4% buffered formaldehyde, before embedding in paraffin wax and sectioning at 10 mm thick. The investigation was approved by the local committee on animal ethics at Umea˚ University.
M. El-Salhy et al. / Mechanisms of Ageing and De6elopment 107 (1999) 93–103
95
2.2. Microwa6e antigen retrie6al The sections were treated as described in detail elsewhere (Nyhlin et al., 1997). Briefly, they were hydrated and immersed in 0.01 M citrate buffer, pH 6, in plastic Coplin jars, which were placed in a microwave oven for three boiling cycles of 5 min each, at maximum power (650 W). For each run, the sections were quickly transferred to a new jar with fresh retrieval solution, cooled to room temperature. After the last run, the slides were allowed to cool to room temperature for 20 min, rinsed in Tris buffer, pH 7.6, and then immunostained. In a preliminary study the sections were immunostained with or without antigen retrieval. Antigen retrieval was found to increase the number of nerve fibres by allowing small branches of the nerve fibres to be detected.
2.3. Immunocytochemistry The sections were immunostained with the avidin–biotin complex (ABC) method (Dakopatts, Glostrup, Denmark) as described earlier in detail (El-Salhy et al., 1993). Briefly, following microwave antigen retrieval, the sections were incubated first for 10 min with 0.5% H2O2 in Tris buffer to inhibit endogenous peroxidase, then with 1% bovine serum albumin for 10 min to block the non-specific binding sites, and then with rabbit anti-PGP 9.5 (diluted 1:600; code no. RA 95101, Ultra Clone Ltd, Isle of Wight, UK) overnight at room temperature. Next, the sections were incubated for 30 min at room temperature with biotinylated swine anti-rabbit IgG, diluted 1:100 and finally with the avidin–biotin–peroxidase complex, diluted 1:50 for 30 min at room temperature. The sections were developed in 50 ml of Tris buffer containing 10 ml of 30% H2O2 and 25 mg of diaminobenzidine tetrahydrochloride (DAB) and lightly counterstained with Mayer’s haemotoxylin. Negative controls were performed by replacing the anti-PGP 9.5 with non-immune rabbit serum. Positive controls were obtained by staining sections from rat brain and human colon. An identical immunostaining procedure was followed in the control experiments.
2.4. Computerized image analysis In order to quantify the enteric nervous system, the video image analysis Quantimet 500MC image processing and analysis system (Leica, Cambridge, UK) linked to an Olympus microscope (type BX50) was used. The software was QWIN (version 1.02), a Windows-based image analysis program from Leica. Leica’s interactive program, QUIPS (version 1.02), was also included in the system. Quantification was performed using × 4, × 20 and × 40 objectives. At these magnifications, each pixel of the image corresponded to 2.12, 0.414 and 0.21 mm, respectively, and each field observed in the monitor represented a tissue area of 1.3, 0.04 and 0.009 mm2, respectively. The number of ganglia in the myenteric and submucosal plexus per millimeter baseline was determined in ten sites from three sections from each mouse, using the
96
M. El-Salhy et al. / Mechanisms of Ageing and De6elopment 107 (1999) 93–103
× 4 objective, and by interactive measurements from the manual menu. This was done by drawing a line parallel to the muscle fibres (baseline) and counting the number of ganglia. The area of each ganglion was measured again by using the × 20 objective and interactive measurements from the manual menu. This was carried out by drawing a line around 15 randomly chosen ganglia from each mouse. The numbers of neurones in 15 myenteric and 15 submucosal ganglia in each animal were determined with the × 40 objective and again by using the interactive measurements from the manual menu. In order to determine the relative volume density of the nerve fibres in muscularis propria and submucosa, the classical stereological point-counting method (Weibel and Elias, 1967; Weibel et al., 1969) was used as adapted for computerized image analysis (El-Salhy et al., 1997). Briefly, an automated standard sequence analysis procedure was carried out, in which a regular 400-point lattice was superimposed on the frame containing the tissue. Points covering tissue other than muscularis propria were erased, while those covering the immunoreactive nerve fibres were targeted with the computer ‘mouse’. When the mouse was clicked, a series of blue highlight points appeared. A step was introduced at the end of the sequence whereby the editor menu was used to add missed points, or to erase errors. In each field, the ratio of points overlying nerve fibres to those lying on muscularis propria or submucosa was tabulated. The sum of all fields in the specimen was computed and statistically analysed automatically. Twenty fields from four different sections, 50 mm apart, were randomly chosen from each individual and analysed with the ×40 objective. The thickness of muscularis propria and submucosa in the antrum, duodenum and colon was determined in four randomly chosen fields from two perpendicularly cut sections, 50 mm apart, taken from each mouse. Fifteen measurements in each field were performed with the × 20 objective and applying the interactive measurements from the manual menu.
2.5. Statistical analysis A pilot study was performed on the antrum, duodenum and colon to determine the number of fields required for measurements. For this, the number of ganglia per millimetre baseline was determined in ten sites, and the area of ten ganglia and the number of neurones were determined in the same ganglia in the myenteric and submucosal plexi. Furthermore, the volume density of nerve fibres was determined in ten fields in both muscularis propria and submucosa. In the same pilot study, the thickness of muscularis propria and submucosa was measured in four fields, ten measurements in each field. The standard deviation was calculated, and the number of fields required in order to obtain an a corresponding to 0.05, b to 0.05 and power to 0.95 was calculated using the GraphPad Instat computer program (version 2.1). In this calculation the minimum difference wished to be detected as significant was chosen to be 50% of the standard deviation. In antrum, the number of fields required were 6, 5, 8, 11, 12, 10, 15 and 14, respectively. The required total measurements of the thickness of muscularis propria and submucosa were 20 and
M. El-Salhy et al. / Mechanisms of Ageing and De6elopment 107 (1999) 93–103
97
Fig. 1. Duodenal myenteric ganglia in a 3-month-old mouse (A) and in a 24-month-old mouse (B). Note the immunoreactive nerve fibres in the muscularis propria (arrows). Colonic submucosal ganglion in a 3-month-old mouse (C) and a 24-month-old mouse (D). ×200.
98
M. El-Salhy et al. / Mechanisms of Ageing and De6elopment 107 (1999) 93–103
Fig. 1. (Continued)
30, respectively. The corresponding figures for duodenum were 7, 9, 8, 6, 13, 14, 15 and 17, as well as 30 and 32; and for colon 9, 8, 11, 14, 13, 13, 14 and 18, as well as 25 and 27. Comparisons between the 3-month-old mice and the other age groups (1-month, 12-months and 24-months old) were made with the non-parametric Wilcoxon test. P-values below 0.05 were regarded as significant.
3. Results
3.1. Immunocytochemistry PGP 9.5-immunoreactive nerve fibres were found in muscularis propria, in submucosa and lamina propria, as well as neurones of both myenteric and submucosal plexi, in all age groups (Fig. 1). Specificity controls showed that replacement of the primary antiserum by non-immune rabbit serum failed to produce immunostaining. The antiserum used immunostained nerve elements in rat brain and the enteric nervous system of human colon.
3.2. Computerized image analysis The results of the morphometric measurements are shown in Fig. 2. In antrum, there was no statistical difference between the different age groups, in either myenteric or submucosal plexi as regards the number of ganglia per millimetre baseline, or in ganglion area in the 12-month and 24-month-old groups. There were significantly fewer neurones per ganglion in both myenteric and submucosal ganglia than in 3-month-old mice. The relative volume density of nerve fibres in muscularis
M. El-Salhy et al. / Mechanisms of Ageing and De6elopment 107 (1999) 93–103
99
Fig. 2. Results of morphometric measurements of the myenteric and submucosal plexi in different age groups. *P B 0.05; **PB0.01; ***PB 0.001.
100
M. El-Salhy et al. / Mechanisms of Ageing and De6elopment 107 (1999) 93–103
propria in the 1-, 12- and 24-month-old mice was significantly greater than in the 3-month-old animals. There was no statistical difference between the different age groups as regards the relative volume density of nerve fibres in the submucosa. In duodenum, neither the number of ganglia per millimetre baseline, the ganglion area, nor the relative volume density of nerve fibres varied between different age groups in either myenteric or submucosal plexi. There were fewer neurones per ganglion in the myenteric and submucosal ganglia in 12- and 24-month-old mice than in 3-month-old mice. In colon, there were fewer myenteric ganglia per millimetre baseline in 1-monthold mice than in the 3-month-old mice. Between the other age groups, there was no difference as regards the number of myenteric ganglia per millimetre baseline, nor was any difference observed between the different age groups regarding numbers of submucosal ganglia per millimetre baseline. Myenteric ganglion area in the 1-, 12-, and 24-month-old mice was smaller than that in 3-month-old mice, but there was no difference between the different age groups regarding submucosal ganglion area. In both myenteric and submucosal ganglia, there were fewer neurones per ganglion in 12- and 24-month-old mice than in 3-month-old mice. There was no statistical difference in the relative volume density of nerve fibres in either the muscularis propria or submucosa, between the different age groups. When the females and males from all age groups were pooled (Table 1), no statistical difference was seen regarding the number of ganglia per millimetre baseline, the ganglion area, the number of neurones per ganglion or the relative volume density of nerve fibres in both myenteric or submucosal plexi. The results of measuring the thickness of muscularis propria and submucosa in various parts of the gastrointestinal tract investigated in different age groups are reported in Table 2.
4. Discussion The present findings confirm earlier observations in human oesophagus and both small and large intestines (de Souza et al., 1993 Meciano et al., 1995 Gomes et al., 1997), as well as in the small intestine of rats (Santer and Baker, 1988) and of guinea-pigs (Gabella, 1989), that ageing is accompanied by changes in the myenteric plexus. Moreover, the present study showed that the enteric nervous system in the stomach was also affected by ageing and that changes also occurred in the gastrointestinal submucosal plexus. As the thickness of the muscularis propria and submucosa of various parts of the gastrointestinal tract investigated did not vary between different age groups, with the exception of the duodenum of 1-month-old mice, these changes are not related to changes in these structures. In all the gastrointestinal segments investigated, there was nerve cell loss in 12and 24-month-old animals, from both the myenteric and submucosal plexi, as compared with 3-month-old mice. These observations are consistent with earlier findings in the myenteric plexus of the human gastrointestinal tract (de Souza et al., 1993 Meciano et al., 1995 Gomes et al., 1997) and in the small intestine of rats
3.78+ 6 0.33 4.11+ 6 0.29
Colon Females Males a
4.12+ 6 0.25 4.21+ 6 0.21
Duodenum Females 0.64+ 6 0.17 Males 0.51+ 6 0.13
All values are expressed as mean+ 6 S.E.
0.9+ 6 0.24 1.61+ 6 0.36
8.25+ 6 0.41 9.08+ 6 0.51
6.2+ 6 0.27 7.04+ 6 0.36
Antrum Females Males
800+ 6 102 724+ 6 124
308+ 6 32 357+ 6 39
870+ 6 128 901+ 6 139
137+ 6 23 132+ 6 11
203+ 6 14 182+ 6 11
152+ 6 37 161+ 6 40
Submucosal plexus
Myenteric plexus
Myenteric plexus
Submucosal plexus
Area of ganglion (mm2)
Number of ganglia per mm baseline
Table 1 Comparison between females and males pooled from different age groupsa
7.4+ 6 0.9 6.9+ 6 1
7.6+ 6 2 8+ 6 1.7
10.2+ 6 0.8 9.3+ 6 1.7
Myenteric plexus
1.8+ 6 0.29 2+6 0.4
3.5+ 6 0.4 3.1+ 6 0.4
2.5+ 6 0.4 3+6 0.5
Submucosal plexus
Number of neurons per ganglion
0.81+ 6 0.05 0.9+ 6 0.26
1.34+ 6 0.15 1.25+ 6 0.11
0.76+ 6 0.08 0.8+ 6 0.03
Myenteric plexus
0.9+ 6 0.08 0.8+ 6 0.07
1.2+ 6 0.2 1.3+ 6 0.4
0.5+ 6 0.05 0.6+ 6 0.09
Submucosal plexus
Relative volume density of nerve fibres
M. El-Salhy et al. / Mechanisms of Ageing and De6elopment 107 (1999) 93–103 101
102
M. El-Salhy et al. / Mechanisms of Ageing and De6elopment 107 (1999) 93–103
(Santer and Baker, 1988) and guinea-pigs (Gabella, 1989). This neurone loss was not accompanied by a change in the number of ganglia, or of ganglion area, except for myenteric ganglia in the colon. It has been reported that both collagen and elastic systems are more numerous in the myenteric ganglia of old than of young subjects (Gomes et al., 1997), which would explain the unchanged ganglionic area in old mice observed here (except in the colon) despite the neurone loss. There was no difference between 3-month-old and 12- and 24-month-old mice regarding the relative volume density of nerve fibres in either myenteric or submucosal plexus in various parts of the gastrointestinal tract investigated—except in the antrum, where they increased. This could be partially explained by the possibility that some of the nerve fibres detected were of extrinsic origin and/or by the possibility that the reduced neurone number is accompanied by structural changes and reorganization of the remaining neurones, as described earlier in the guinea-pig (Gomes et al., 1997). In 1-month-old mice, which had not yet reached maturity, the myenteric ganglion area and the number of submucosal ganglia per millimetre baseline in the colon were both smaller than in 3-month-old mice. Moreover, in the antral myenteric plexus, the relative volume density of nerve fibres in this age group was greater than that of 3-month-old mice. These differences may reflect a developing process that these animals pass through as they approach maturity. The present finding that the enteric nervous system changes with ageing may well be of some relevance for the increased gastrointestinal motility dysfunction in the elderly. On the other hand, the present investigation showed that there is no sex-related difference regarding this regulatory system.
Table 2 The thicknessa of muscularis propria and submucosa in different age groups in various parts of the gastrointestinal tract investigated Age group 1 month
3 months
12 months
24 months
Antrum Muscularis propria Submucosa
266+ 6 17 280+ 6 20
322+ 6 28 280+ 6 20
419+ 6 47 310+ 6 40
2343+ 6 24 370+ 6 20
Duodenum Muscularis propria Submucosa
35+ 6 2b 73+ 6 3c
49+ 6 5 115+ 6 6
49+ 6 5 117+ 6 8
55+ 6 6 104+ 6 8
Colon Muscularis propria Submucosa
89+ 6 4 262+ 6 14
90+ 6 9 293+ 6 20
110+ 6 8 289+ 6 9
108+ 6 11 331+ 6 28
Data are expressed in mm (mean+ 6 S.E.). PB0.01. c PB0.001. a
b
M. El-Salhy et al. / Mechanisms of Ageing and De6elopment 107 (1999) 93–103
103
Acknowledgements This study was supported by grants from the Medical Faculty, Umea˚ University. References Bannister, J.J., Abouzekry, L., Read, N.W., 1987. Effect of aging on anorectal function. Gut 28, 353 – 357. Burkitt, D.P., Walker, A.R.P., Painter, N.S., 1972. Effect of dietary fibre on stools and transit times and its role in the causation of disease. Lancet 2, 1408 – 1411. de Souza, R.R., Moratelli, H.B., Borges, N., Liberti, E.A., 1993. Age-induced nerve cell loss in the myenteric plexus of the small intestine in man. Gerontology 39. El-Salhy, M., Stenling, R., Grimelius, L., 1993. Peptidergic nerves and endocrine cells in human liver. Scand. J. Gastroenterol. 28, 809–815. El-Salhy, M., Sandstro¨m, O., Na¨sstro¨m, E., Mustajbasic, M., Zachrisson, S., 1997. Application of computer image analysis in endocrine cell quantification. Histochem. J. 29, 249 – 256. Frexinos, J., Staumont, G., Fioramonti, J., Bueno, L., 1989. Effects of sennosides on colonic myoelectrical activity in man. Dig. Dis. Sci. 34, 214 – 218. Gabella, G., 1989. Fall in the number of myenteric neurons in aging guinea-pigs. Gastroenterology 96, 1487–1493. Gomes, O.A., de-Souza, R.R., Liberti, E.A., 1997. A preliminary investigation of the effects of aging on the nerve cell number in the myenteric ganglia of the human colon. Gerontology 43, 210 – 217. Hollis, J., Castell, D.O., 1974. Esophageal function in elderly men. Ann. Intern. Med. 80, 371 – 374. Krammer, H.J., Karahan, S.T., Rumpel, E., Klinger, M., Kuhnel, W., 1993. Immunohistochemical visualization of the enteric nervous system using antibodies against protein gene product (PGP) 9.5. Ann. Anat. 175, 321–325. Krammer, H.J., Karahan, S.T., Sigge, W., Kuhnel, W., 1994. Immunohistochemistry of markers of the enteric nervous system in whole-mount preparations of the human colon. Eur. J. Pediatr. Surg. 4, 274–278. Leng-Pechlow, E., 1989. Effects of sennosides A + B and bisacody on rat large intestine. Pharmacology 38, 310–318. Madsen, J.L., 1992. Effects of gender, age and body mass index on gastrointestinal transit time. Dig. Dis. Sci. 37, 1548–1553. Materia, E., Corraziari, E., Bausano, G., Anzini, F.A., Torsoli, A., 1988. Effect of bisacody on motor activity on sigmoid colon. Ital. J. Gastroenterol. 20, 246 – 249. McDougal, J.N., Miller, M.S., Burks, T.F., Kreulen, D.L., 1984. Age-related changes in colonic functions in rats. Am. J. Physiol. 247, 542–546. Meciano, J.F., Carvalho, V.C., de Souza, R.R., 1995. Nerve cell loss in the myenteric plexus of human esophagus in relation to age: a preliminary investigation. Gerontology 41, 18 – 21. Moore, J.G., Tweedy, C., Christian, P.E., Datz, F.L., 1983. Effect of age on gastric emptying of liquid-solid meals in man. Dig. Dis. Sci. 28, 340– 344. Nyhlin, N., El-Salhy, M., Sandstro¨m, O., Suhr, O., 1997. Evaluation of immunohistochemical staining of human duodenal endocrine cells after microwave antigen retrieval. Histochem. J. 29, 177 – 181. Santer, R.M., Baker, D.M., 1988. Enteric neuron numbers and sizes in Auerbach%s plexus in the small and large intestine of adult and aged rats. J. Auton. Nerv. Syst. 25, 59 – 67. Smout, A.J.P., Akkermans, L.M., 1992. Normal and Disturbed Motility of the Gastrointestinal Tract. Wrightson Biomedical Publishing, Petersfield, pp. 25 – 37. Tomlin, J., Read, N.W., 1988. Laxative properties of indigestible plastic particles. Br. Med. J. 297, 1175–1176. Weibel, E., Elias, H., 1967. Introduction to stereologic principles. In: Weibel, E., Elias, H. (Eds.), Quantitative Methods in Morphology. Springer-Verlag, Berlin, pp. 89 – 98. Weibel, E., Sta¨ubli, R.H., Hess, F.A., 1969. Correlated morphometric and biochemical studies on the liver cell. I. Morphometric model, stereologic methods, and normal morphometric data for rat liver. J. Cell Biol. 42, 68–91.
Age-induced changes in the enteric nervous system in the mouse Magdy El-Salhy *, Olof Sandstro¨m, Frank Holmlund Section for Gastroenterology and Hepatology, Department of Medicine, Uni6ersity Hospital, S-901 85 Umea˚, Sweden Received 27 July 1998; received in revised form 20 October 1998; accepted 3 November 1998
Abstract The enteric nervous system of the murine gut was investigated by immunocytochemistry in 1-, 3-, 12- and 24-month-old mice, using protein gene product 9.5, a general marker for nerve elements. Myenteric and submucosal plexi were quantified by computerized image analysis. In antrum, there were significantly fewer neurones per ganglion in both myenteric and submucosal ganglia of 12- and 24-month-old mice than in 3-month-old animals. The same was true of duodenum and colon. The relative volume density of nerve fibres in antral muscularis propria was significantly greater in the 1-, 12- and 24-month-old mice than in the 3-month-old mice. In colon, there were fewer submucosal ganglia per millimetre baseline in 1-month-old mice than in 3-month-old mice. The colonic myenteric ganglion in 1-, 12- and 24-month-old mice was smaller than in 3-month-old mice. There was no statistical difference between females and males regarding the number of ganglia per millimetre baseline, ganglionic area, number of neurones per ganglion or the relative volume density of nerve fibres in either the myenteric or submucosal plexi. As the enteric nervous system is responsible for coordinating and integrating the motility of the gut, the ageing-related changes reported here may well be of some relevance for the increased gastrointestinal motility dysfunction in the elderly persons. © 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Aging; Computerized image analysis; Enteric nervous system; Gastrointestinal tract; Immunocytochemistry; Mice
* Corresponding author. Fax: + 46-90-143986; e-mail: [email protected]. 0047-6374/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 0 4 7 - 6 3 7 4 ( 9 8 ) 0 0 1 4 2 - 0
94
M. El-Salhy et al. / Mechanisms of Ageing and De6elopment 107 (1999) 93–103
1. Introduction Motility disorders of the gastrointestinal tract are known to become more common with advanced age (Burkitt et al., 1972; Hollis and Castell, 1974; Moore et al., 1983; McDougal et al., 1984; Bannister et al., 1987; Madsen, 1992). For instance, it has been reported that colonic transit is protracted in elderly humans (Madsen, 1992) and emptying of liquid of the stomach is tardy in the elderly (Moore et al., 1983). Delayed colonic transit has been also reported in old rats (McDougal et al., 1984). As it is the enteric nervous system of the gastrointestinal tract that coordinates and integrates the motility activities of the gut (Smout and Akkermans, 1992), several studies have been performed on this regulatory system to establish a plausible aetiology for the motility disorders in the elderly (Santer and Baker, 1988; Gabella, 1989; de Souza et al., 1993 Meciano et al., 1995; Gomes et al., 1997). The number of myenteric neurones has been reported to decrease with age in the human oesophagus and in small and large intestines (de Souza et al., 1993; Meciano et al., 1995; Gomes et al., 1997) The same age-related decrease in neurones has been found in the small intestine of the rat (Santer and Baker, 1988) and guinea-pig (Gabella, 1989). No data are available, however, concerning the effect of ageing on the submucosal plexus, which is also an important regulator of gastrointestinal motility by means of tactile or chemical stimulus of the mucosa (Materia et al., 1988; Tomlin and Read, 1988; Frexinos et al., 1989; Leng-Pechlow, 1989). Furthermore, disturbances in the enteric nervous system in the stomach are unknown. In order to fill these gaps in our knowledge, the present investigation was undertaken, in which conceivable age-related changes in both myenteric and submucosal plexi of the enteric nervous system were studied in different parts of the gastrointestinal tract of an animal model, namely the mouse. For this purpose, a general marker for nerve elements, protein gene product 9.5 (PGP 9.5), was used (Krammer et al., 1993, 1994). 2. Materials and methods
2.1. Animals NMRI/Bom mice (Bomholtga˚rd Breeding and Research Centre, Denmark) were used. Males and females were housed separately in cages, five to a cage, in a room with a 12/12-h light/dark cycle, and fed a standard pellet diet (Astra-Ewos AB, So¨derta¨lje, Sweden) with free access to water. Ten mice (five males, five females) were sacrificed when 1 month old, ten more at 3 months and another ten at 12 months. Six 24-month-old males were then also killed. The animals were starved overnight, before sacrifice by cervical dislocation. Antrum, proximal duodenum and distal colon were excised. The tissue specimens were fixed overnight in 4% buffered formaldehyde, before embedding in paraffin wax and sectioning at 10 mm thick. The investigation was approved by the local committee on animal ethics at Umea˚ University.
M. El-Salhy et al. / Mechanisms of Ageing and De6elopment 107 (1999) 93–103
95
2.2. Microwa6e antigen retrie6al The sections were treated as described in detail elsewhere (Nyhlin et al., 1997). Briefly, they were hydrated and immersed in 0.01 M citrate buffer, pH 6, in plastic Coplin jars, which were placed in a microwave oven for three boiling cycles of 5 min each, at maximum power (650 W). For each run, the sections were quickly transferred to a new jar with fresh retrieval solution, cooled to room temperature. After the last run, the slides were allowed to cool to room temperature for 20 min, rinsed in Tris buffer, pH 7.6, and then immunostained. In a preliminary study the sections were immunostained with or without antigen retrieval. Antigen retrieval was found to increase the number of nerve fibres by allowing small branches of the nerve fibres to be detected.
2.3. Immunocytochemistry The sections were immunostained with the avidin–biotin complex (ABC) method (Dakopatts, Glostrup, Denmark) as described earlier in detail (El-Salhy et al., 1993). Briefly, following microwave antigen retrieval, the sections were incubated first for 10 min with 0.5% H2O2 in Tris buffer to inhibit endogenous peroxidase, then with 1% bovine serum albumin for 10 min to block the non-specific binding sites, and then with rabbit anti-PGP 9.5 (diluted 1:600; code no. RA 95101, Ultra Clone Ltd, Isle of Wight, UK) overnight at room temperature. Next, the sections were incubated for 30 min at room temperature with biotinylated swine anti-rabbit IgG, diluted 1:100 and finally with the avidin–biotin–peroxidase complex, diluted 1:50 for 30 min at room temperature. The sections were developed in 50 ml of Tris buffer containing 10 ml of 30% H2O2 and 25 mg of diaminobenzidine tetrahydrochloride (DAB) and lightly counterstained with Mayer’s haemotoxylin. Negative controls were performed by replacing the anti-PGP 9.5 with non-immune rabbit serum. Positive controls were obtained by staining sections from rat brain and human colon. An identical immunostaining procedure was followed in the control experiments.
2.4. Computerized image analysis In order to quantify the enteric nervous system, the video image analysis Quantimet 500MC image processing and analysis system (Leica, Cambridge, UK) linked to an Olympus microscope (type BX50) was used. The software was QWIN (version 1.02), a Windows-based image analysis program from Leica. Leica’s interactive program, QUIPS (version 1.02), was also included in the system. Quantification was performed using × 4, × 20 and × 40 objectives. At these magnifications, each pixel of the image corresponded to 2.12, 0.414 and 0.21 mm, respectively, and each field observed in the monitor represented a tissue area of 1.3, 0.04 and 0.009 mm2, respectively. The number of ganglia in the myenteric and submucosal plexus per millimeter baseline was determined in ten sites from three sections from each mouse, using the
96
M. El-Salhy et al. / Mechanisms of Ageing and De6elopment 107 (1999) 93–103
× 4 objective, and by interactive measurements from the manual menu. This was done by drawing a line parallel to the muscle fibres (baseline) and counting the number of ganglia. The area of each ganglion was measured again by using the × 20 objective and interactive measurements from the manual menu. This was carried out by drawing a line around 15 randomly chosen ganglia from each mouse. The numbers of neurones in 15 myenteric and 15 submucosal ganglia in each animal were determined with the × 40 objective and again by using the interactive measurements from the manual menu. In order to determine the relative volume density of the nerve fibres in muscularis propria and submucosa, the classical stereological point-counting method (Weibel and Elias, 1967; Weibel et al., 1969) was used as adapted for computerized image analysis (El-Salhy et al., 1997). Briefly, an automated standard sequence analysis procedure was carried out, in which a regular 400-point lattice was superimposed on the frame containing the tissue. Points covering tissue other than muscularis propria were erased, while those covering the immunoreactive nerve fibres were targeted with the computer ‘mouse’. When the mouse was clicked, a series of blue highlight points appeared. A step was introduced at the end of the sequence whereby the editor menu was used to add missed points, or to erase errors. In each field, the ratio of points overlying nerve fibres to those lying on muscularis propria or submucosa was tabulated. The sum of all fields in the specimen was computed and statistically analysed automatically. Twenty fields from four different sections, 50 mm apart, were randomly chosen from each individual and analysed with the ×40 objective. The thickness of muscularis propria and submucosa in the antrum, duodenum and colon was determined in four randomly chosen fields from two perpendicularly cut sections, 50 mm apart, taken from each mouse. Fifteen measurements in each field were performed with the × 20 objective and applying the interactive measurements from the manual menu.
2.5. Statistical analysis A pilot study was performed on the antrum, duodenum and colon to determine the number of fields required for measurements. For this, the number of ganglia per millimetre baseline was determined in ten sites, and the area of ten ganglia and the number of neurones were determined in the same ganglia in the myenteric and submucosal plexi. Furthermore, the volume density of nerve fibres was determined in ten fields in both muscularis propria and submucosa. In the same pilot study, the thickness of muscularis propria and submucosa was measured in four fields, ten measurements in each field. The standard deviation was calculated, and the number of fields required in order to obtain an a corresponding to 0.05, b to 0.05 and power to 0.95 was calculated using the GraphPad Instat computer program (version 2.1). In this calculation the minimum difference wished to be detected as significant was chosen to be 50% of the standard deviation. In antrum, the number of fields required were 6, 5, 8, 11, 12, 10, 15 and 14, respectively. The required total measurements of the thickness of muscularis propria and submucosa were 20 and
M. El-Salhy et al. / Mechanisms of Ageing and De6elopment 107 (1999) 93–103
97
Fig. 1. Duodenal myenteric ganglia in a 3-month-old mouse (A) and in a 24-month-old mouse (B). Note the immunoreactive nerve fibres in the muscularis propria (arrows). Colonic submucosal ganglion in a 3-month-old mouse (C) and a 24-month-old mouse (D). ×200.
98
M. El-Salhy et al. / Mechanisms of Ageing and De6elopment 107 (1999) 93–103
Fig. 1. (Continued)
30, respectively. The corresponding figures for duodenum were 7, 9, 8, 6, 13, 14, 15 and 17, as well as 30 and 32; and for colon 9, 8, 11, 14, 13, 13, 14 and 18, as well as 25 and 27. Comparisons between the 3-month-old mice and the other age groups (1-month, 12-months and 24-months old) were made with the non-parametric Wilcoxon test. P-values below 0.05 were regarded as significant.
3. Results
3.1. Immunocytochemistry PGP 9.5-immunoreactive nerve fibres were found in muscularis propria, in submucosa and lamina propria, as well as neurones of both myenteric and submucosal plexi, in all age groups (Fig. 1). Specificity controls showed that replacement of the primary antiserum by non-immune rabbit serum failed to produce immunostaining. The antiserum used immunostained nerve elements in rat brain and the enteric nervous system of human colon.
3.2. Computerized image analysis The results of the morphometric measurements are shown in Fig. 2. In antrum, there was no statistical difference between the different age groups, in either myenteric or submucosal plexi as regards the number of ganglia per millimetre baseline, or in ganglion area in the 12-month and 24-month-old groups. There were significantly fewer neurones per ganglion in both myenteric and submucosal ganglia than in 3-month-old mice. The relative volume density of nerve fibres in muscularis
M. El-Salhy et al. / Mechanisms of Ageing and De6elopment 107 (1999) 93–103
99
Fig. 2. Results of morphometric measurements of the myenteric and submucosal plexi in different age groups. *P B 0.05; **PB0.01; ***PB 0.001.
100
M. El-Salhy et al. / Mechanisms of Ageing and De6elopment 107 (1999) 93–103
propria in the 1-, 12- and 24-month-old mice was significantly greater than in the 3-month-old animals. There was no statistical difference between the different age groups as regards the relative volume density of nerve fibres in the submucosa. In duodenum, neither the number of ganglia per millimetre baseline, the ganglion area, nor the relative volume density of nerve fibres varied between different age groups in either myenteric or submucosal plexi. There were fewer neurones per ganglion in the myenteric and submucosal ganglia in 12- and 24-month-old mice than in 3-month-old mice. In colon, there were fewer myenteric ganglia per millimetre baseline in 1-monthold mice than in the 3-month-old mice. Between the other age groups, there was no difference as regards the number of myenteric ganglia per millimetre baseline, nor was any difference observed between the different age groups regarding numbers of submucosal ganglia per millimetre baseline. Myenteric ganglion area in the 1-, 12-, and 24-month-old mice was smaller than that in 3-month-old mice, but there was no difference between the different age groups regarding submucosal ganglion area. In both myenteric and submucosal ganglia, there were fewer neurones per ganglion in 12- and 24-month-old mice than in 3-month-old mice. There was no statistical difference in the relative volume density of nerve fibres in either the muscularis propria or submucosa, between the different age groups. When the females and males from all age groups were pooled (Table 1), no statistical difference was seen regarding the number of ganglia per millimetre baseline, the ganglion area, the number of neurones per ganglion or the relative volume density of nerve fibres in both myenteric or submucosal plexi. The results of measuring the thickness of muscularis propria and submucosa in various parts of the gastrointestinal tract investigated in different age groups are reported in Table 2.
4. Discussion The present findings confirm earlier observations in human oesophagus and both small and large intestines (de Souza et al., 1993 Meciano et al., 1995 Gomes et al., 1997), as well as in the small intestine of rats (Santer and Baker, 1988) and of guinea-pigs (Gabella, 1989), that ageing is accompanied by changes in the myenteric plexus. Moreover, the present study showed that the enteric nervous system in the stomach was also affected by ageing and that changes also occurred in the gastrointestinal submucosal plexus. As the thickness of the muscularis propria and submucosa of various parts of the gastrointestinal tract investigated did not vary between different age groups, with the exception of the duodenum of 1-month-old mice, these changes are not related to changes in these structures. In all the gastrointestinal segments investigated, there was nerve cell loss in 12and 24-month-old animals, from both the myenteric and submucosal plexi, as compared with 3-month-old mice. These observations are consistent with earlier findings in the myenteric plexus of the human gastrointestinal tract (de Souza et al., 1993 Meciano et al., 1995 Gomes et al., 1997) and in the small intestine of rats
3.78+ 6 0.33 4.11+ 6 0.29
Colon Females Males a
4.12+ 6 0.25 4.21+ 6 0.21
Duodenum Females 0.64+ 6 0.17 Males 0.51+ 6 0.13
All values are expressed as mean+ 6 S.E.
0.9+ 6 0.24 1.61+ 6 0.36
8.25+ 6 0.41 9.08+ 6 0.51
6.2+ 6 0.27 7.04+ 6 0.36
Antrum Females Males
800+ 6 102 724+ 6 124
308+ 6 32 357+ 6 39
870+ 6 128 901+ 6 139
137+ 6 23 132+ 6 11
203+ 6 14 182+ 6 11
152+ 6 37 161+ 6 40
Submucosal plexus
Myenteric plexus
Myenteric plexus
Submucosal plexus
Area of ganglion (mm2)
Number of ganglia per mm baseline
Table 1 Comparison between females and males pooled from different age groupsa
7.4+ 6 0.9 6.9+ 6 1
7.6+ 6 2 8+ 6 1.7
10.2+ 6 0.8 9.3+ 6 1.7
Myenteric plexus
1.8+ 6 0.29 2+6 0.4
3.5+ 6 0.4 3.1+ 6 0.4
2.5+ 6 0.4 3+6 0.5
Submucosal plexus
Number of neurons per ganglion
0.81+ 6 0.05 0.9+ 6 0.26
1.34+ 6 0.15 1.25+ 6 0.11
0.76+ 6 0.08 0.8+ 6 0.03
Myenteric plexus
0.9+ 6 0.08 0.8+ 6 0.07
1.2+ 6 0.2 1.3+ 6 0.4
0.5+ 6 0.05 0.6+ 6 0.09
Submucosal plexus
Relative volume density of nerve fibres
M. El-Salhy et al. / Mechanisms of Ageing and De6elopment 107 (1999) 93–103 101
102
M. El-Salhy et al. / Mechanisms of Ageing and De6elopment 107 (1999) 93–103
(Santer and Baker, 1988) and guinea-pigs (Gabella, 1989). This neurone loss was not accompanied by a change in the number of ganglia, or of ganglion area, except for myenteric ganglia in the colon. It has been reported that both collagen and elastic systems are more numerous in the myenteric ganglia of old than of young subjects (Gomes et al., 1997), which would explain the unchanged ganglionic area in old mice observed here (except in the colon) despite the neurone loss. There was no difference between 3-month-old and 12- and 24-month-old mice regarding the relative volume density of nerve fibres in either myenteric or submucosal plexus in various parts of the gastrointestinal tract investigated—except in the antrum, where they increased. This could be partially explained by the possibility that some of the nerve fibres detected were of extrinsic origin and/or by the possibility that the reduced neurone number is accompanied by structural changes and reorganization of the remaining neurones, as described earlier in the guinea-pig (Gomes et al., 1997). In 1-month-old mice, which had not yet reached maturity, the myenteric ganglion area and the number of submucosal ganglia per millimetre baseline in the colon were both smaller than in 3-month-old mice. Moreover, in the antral myenteric plexus, the relative volume density of nerve fibres in this age group was greater than that of 3-month-old mice. These differences may reflect a developing process that these animals pass through as they approach maturity. The present finding that the enteric nervous system changes with ageing may well be of some relevance for the increased gastrointestinal motility dysfunction in the elderly. On the other hand, the present investigation showed that there is no sex-related difference regarding this regulatory system.
Table 2 The thicknessa of muscularis propria and submucosa in different age groups in various parts of the gastrointestinal tract investigated Age group 1 month
3 months
12 months
24 months
Antrum Muscularis propria Submucosa
266+ 6 17 280+ 6 20
322+ 6 28 280+ 6 20
419+ 6 47 310+ 6 40
2343+ 6 24 370+ 6 20
Duodenum Muscularis propria Submucosa
35+ 6 2b 73+ 6 3c
49+ 6 5 115+ 6 6
49+ 6 5 117+ 6 8
55+ 6 6 104+ 6 8
Colon Muscularis propria Submucosa
89+ 6 4 262+ 6 14
90+ 6 9 293+ 6 20
110+ 6 8 289+ 6 9
108+ 6 11 331+ 6 28
Data are expressed in mm (mean+ 6 S.E.). PB0.01. c PB0.001. a
b
M. El-Salhy et al. / Mechanisms of Ageing and De6elopment 107 (1999) 93–103
103
Acknowledgements This study was supported by grants from the Medical Faculty, Umea˚ University. References Bannister, J.J., Abouzekry, L., Read, N.W., 1987. Effect of aging on anorectal function. Gut 28, 353 – 357. Burkitt, D.P., Walker, A.R.P., Painter, N.S., 1972. Effect of dietary fibre on stools and transit times and its role in the causation of disease. Lancet 2, 1408 – 1411. de Souza, R.R., Moratelli, H.B., Borges, N., Liberti, E.A., 1993. Age-induced nerve cell loss in the myenteric plexus of the small intestine in man. Gerontology 39. El-Salhy, M., Stenling, R., Grimelius, L., 1993. Peptidergic nerves and endocrine cells in human liver. Scand. J. Gastroenterol. 28, 809–815. El-Salhy, M., Sandstro¨m, O., Na¨sstro¨m, E., Mustajbasic, M., Zachrisson, S., 1997. Application of computer image analysis in endocrine cell quantification. Histochem. J. 29, 249 – 256. Frexinos, J., Staumont, G., Fioramonti, J., Bueno, L., 1989. Effects of sennosides on colonic myoelectrical activity in man. Dig. Dis. Sci. 34, 214 – 218. Gabella, G., 1989. Fall in the number of myenteric neurons in aging guinea-pigs. Gastroenterology 96, 1487–1493. Gomes, O.A., de-Souza, R.R., Liberti, E.A., 1997. A preliminary investigation of the effects of aging on the nerve cell number in the myenteric ganglia of the human colon. Gerontology 43, 210 – 217. Hollis, J., Castell, D.O., 1974. Esophageal function in elderly men. Ann. Intern. Med. 80, 371 – 374. Krammer, H.J., Karahan, S.T., Rumpel, E., Klinger, M., Kuhnel, W., 1993. Immunohistochemical visualization of the enteric nervous system using antibodies against protein gene product (PGP) 9.5. Ann. Anat. 175, 321–325. Krammer, H.J., Karahan, S.T., Sigge, W., Kuhnel, W., 1994. Immunohistochemistry of markers of the enteric nervous system in whole-mount preparations of the human colon. Eur. J. Pediatr. Surg. 4, 274–278. Leng-Pechlow, E., 1989. Effects of sennosides A + B and bisacody on rat large intestine. Pharmacology 38, 310–318. Madsen, J.L., 1992. Effects of gender, age and body mass index on gastrointestinal transit time. Dig. Dis. Sci. 37, 1548–1553. Materia, E., Corraziari, E., Bausano, G., Anzini, F.A., Torsoli, A., 1988. Effect of bisacody on motor activity on sigmoid colon. Ital. J. Gastroenterol. 20, 246 – 249. McDougal, J.N., Miller, M.S., Burks, T.F., Kreulen, D.L., 1984. Age-related changes in colonic functions in rats. Am. J. Physiol. 247, 542–546. Meciano, J.F., Carvalho, V.C., de Souza, R.R., 1995. Nerve cell loss in the myenteric plexus of human esophagus in relation to age: a preliminary investigation. Gerontology 41, 18 – 21. Moore, J.G., Tweedy, C., Christian, P.E., Datz, F.L., 1983. Effect of age on gastric emptying of liquid-solid meals in man. Dig. Dis. Sci. 28, 340– 344. Nyhlin, N., El-Salhy, M., Sandstro¨m, O., Suhr, O., 1997. Evaluation of immunohistochemical staining of human duodenal endocrine cells after microwave antigen retrieval. Histochem. J. 29, 177 – 181. Santer, R.M., Baker, D.M., 1988. Enteric neuron numbers and sizes in Auerbach%s plexus in the small and large intestine of adult and aged rats. J. Auton. Nerv. Syst. 25, 59 – 67. Smout, A.J.P., Akkermans, L.M., 1992. Normal and Disturbed Motility of the Gastrointestinal Tract. Wrightson Biomedical Publishing, Petersfield, pp. 25 – 37. Tomlin, J., Read, N.W., 1988. Laxative properties of indigestible plastic particles. Br. Med. J. 297, 1175–1176. Weibel, E., Elias, H., 1967. Introduction to stereologic principles. In: Weibel, E., Elias, H. (Eds.), Quantitative Methods in Morphology. Springer-Verlag, Berlin, pp. 89 – 98. Weibel, E., Sta¨ubli, R.H., Hess, F.A., 1969. Correlated morphometric and biochemical studies on the liver cell. I. Morphometric model, stereologic methods, and normal morphometric data for rat liver. J. Cell Biol. 42, 68–91.