- Email: [email protected]
Changes in adrenergic and peptidergic nerves in the submucous plexus of streptozocin-diabetic rat ileum
GASTROENTEROLOGY
1990;98:1427-1436
Changes in Adrenergic and Peptidergic Nerves in the Submucous Plexus of Streptozocin-Diabetic Rat Ileum ABEBECH
BELA1 and GEOFFREY
Department of Anatomy and Developmental College London, London, England
The effect of streptozocin diabetes on the distribution of adrenergic and peptidergic nerves in the submucous plexus of rat ileum was investigated and compared with the changes in the myenteric plexus of the same region of ileum. There was an increase in the intensity of immunoreactivity in vasoactive intestinal polypeptide- and neuropeptide Y-like immunoreactive nerve fibers and neurons and a decrease in calcitonin gene-related peptide-like immunoreactivity but no change in substance P- and dopamine P-hydroxylase-like immunoreactivity in the nerve fibers and neurons of the submucous plexus of both 8- and 16-wk streptozocin-diabetic rat ileum. However, in the myenteric plexus of the diabetic rat ileum, there was enlargement of varicosities and an increase followed by a slight decrease in the intensity of immunoreactivity of vasoactive intestinal polypeptide- and dopamine &hydroxylase-like immunoreactive nerve fibers and neurons, increased substance P-like immunoreactivity in diabetes at 16 wk, and an initial decrease (at 8 wk) followed by a recovery of calcitonin gene-related peptide-like immunoreactivity at 16 wk, but no change in neuropeptide Y-like immunoreactivity. The markedly different changes in peptidergic and adrenergic nerves between the two enteric plexuses show that diabetic neuropathy induced by streptozocin is not selective and involves factors other than neurotransmitter types.
I
BURNSTOCK
Biology and Centre for Neuroscience,
mmunohistochemical studies have demonstrated the presence of nonadrenergic, noncholinergic neurons i,n the submucous plexus of many mammalian species, and the nerve fibers of most of these neurons have been traced to the mucosa of the small intestine (1,~). Nerve fibers in the mucosa showing immunoreactivity to vasoactive intestinal polypeptide (VIP), somatostatin, cholecystokinin, calcitonin gene-related peptide (CGRP), and neuropeptide Y (NPY) have been
University
found to arise from cell bodies in the submucous plexus, whereas substance P fibers arise partly from the myenteric plexus but mostly from neurons in the submucous plexus (12). Tapper (31 has reported that noradrenaline and somatostatin are able to increase absorptive fluxes of intestinal ions, while acetylcholine, VIP, and substance P enhance secretion in the intestinal mucosa. Recently, Miller (4) reported that addition of NPY to the serosal side of the ileal mucosa caused a rapid depression of transmural potential difference and short-circuit current in the guinea pig ileum. Functional and neuronal projection studies in the submucous plexus have implied that the majority of submucous neurons act as secretomotor neurons for the mucosa, although it is also known that enteric neurons, probably of submucous origin, also participate in the control of intestinal blood flow (1). Although recent structural and functional studies add to our understanding of the submucous plexus, the significance of the presence of several peptidecontaining neurons in relation to mucosal function and their complex projections and interconnections has not yet been fully established. In this report, we have conducted a comparative study of the effect of short- and long-term diabetes on VIP-, substance P-, dopamine p-hydroxylase (DBH)-, CGRP-, and NPYlike immunoreactivity (LI) in the submucous and myenteric plexuses of streptozocin-diabetic rat ileum using immunohistochemical techniques. Materials and Methods Male Wistar rats weighing 400-450 g were injected with buffered streptozocin (65 mg/kg body wt). The onset of Abbreviations used in this paper: CGRP, calcitonin generelated peptide; DBH, dopamine fl-hydroxylase; -LI, -like immunoreactivity: NPY, neuropeptide Y; PBS, phosphate-buffered saline; VIP, vasoactive intestinal polypeptide. @ 1990 by the American Gastroenterological Association 0016-5065/90/$3.00
1428 BELA1 AND BURNSTOCK
GASTROENTEROLOGY
diabetes was established by the presence of rapid weight loss, polyuria, and glycosuria (Clinitest reagent tablets; Miles, Stoke Poges, Slough, England]. Untreated controls were of the same initial weight range as the diabetic group and both groups were maintained under the same conditions, supplied with food and water ad libitum until death. One group each of control and treated rats was kept for 8 wk and another group of each for 16 wk after the streptozocin injection. Blood samples were taken under ether anesthesia from the posterior vena cava for blood glucose analysis and the animals were killed. Segments of ileum from both control and diabetic rats were dissected out, cleaned and stretched out onto strips of Sylgard silicon rubber (Dow Corning, Seneffe, Belgium] with the mucosal side down, and fixed in 4% paraformaldehyde in phosphate-buffered saline [PBS] for z h at 4°C. The whole-mount preparation for investigating the myenteric plexus was performed as described previously (5,6). The submucous plexus was exposed after peeling off the muscle layers together with the myenteric plexus, and the mucosa was removed with a pair of fine forceps. The submucous and myenteric preparations were then washed 10 times each for lo-min periods with 80% alcohol, dehydrated, rehydrated and washed 3 times for 5 min with PBS containing 0.1% Triton X-100 (Sigma, Poole, England]. The tissues were incubated with polyclonal antisera against VIP, substance P, DBH (a marker for noradrenergic nerves], CGRP, and NPY at a dilution of 1200 in PBS in a humid chamber for 18 hat room temperature. The tissues were then washed with PBS/Triton X-100 and incubated with fluorescein isothiocyanate-conjugated goat anti-rabbit immunoglobulin at a dilution of 1:40 (Nordic Immunological Laboratories, Tilburg, The Netherlands] at room temperature for 1 h.
The tissues were subsequently washed with PBS/Triton X-100 and stained
with 0.05% Pontamine Sky Blue (BDH Ltd., Poole, England] in PBS and 1% dimethyl
Chemicals
sulphoxide for 30 min at room temperature to reduce background autofluorescence (7). The immunoreactive neurons and nerve fibres in the submucous and myenteric plexuses were viewed using a Zeiss microscope [Carl Zeiss Inc., Thornwood, N.Y.] equipped for viewing fluorescein isothiocyanate fluorescence. The specimens were coded and read blindly.
Vol. 98. No. 6
Results The general features of rats 8 wk after induction of diabetes with streptozocin have been reported previously (5,8), and there was no significant difference in the general features of 16-wk diabetic rats, except that two of the latter were blind (had developed cataract] at the time of death. Briefly, the diabetic rats were severly hyperglycemic [with an average blood glucose level of 492 * 15.4 mg/dl), and distension of the large and small intestines and the presence of pale watery stools were observed at the time of death. A summary of the detailed immunohistochemical observations is presented in Table 1.
Vasoactive Intestinal Polypeptide-Like lmmunoreactivity Vasoactive intestinal polypeptide-like immunoreactive neurons and nerve fibers were observed in the submucous plexus of the ileum of control rats, with an average of 5-i’ immunoreactive cell bodies in each ganglion (Figure 1A and C]. In both 8- and 16-wk diabetic preparations, the VIP-like immunoreactive neurons and some of their processes were intensely stained compared with age-matched controls (Figure 1B and D). The increased immunoreactivity was observed in all VIP-like immunoreactive cell bodies in the ganglia. There was no obvious change in the structure of the varicose nerve fibers. Vasoactive intestinal polypeptide-like immunoreactive nerve fibers in the primary and secondary ganglia and internodal fibers of the myenteric plexus of the control rat ileum were observed (Figure ZA). There was an increase in fluorescence intensity, the density of VIP-like immunoreactive fibers in the myenteric ganglia was reduced [Figure 2B), and more prominent enlarged varicosities were observed after 16 wk of diabetes [Figure 2C).
Table 1. Changes in the Intensity of the Different Neuroactive Substances in the Cell Bodies and Nerve Fibers of the Submucous and Myenteric Plexuses of 8- and 16-wk Diabetic Rat Ileum Compared With Age-Matched Controls Submucous plexus 8-wk diabetic Neuroactive substances DBH-LI VIP-L1 NPY-LI Substance CGRP-LI
16-wk diabetic
8-wk diabetic
Cell bodies
Fibers
Cell bodies
Fibers
i
i
7
i
7
I
5
5
Cell bodies
:r
16-wk diabetic
Fibers
Cell bodies
rft -
i
i
-
11
11
T 1
P-L1
No change: t, 1 increase
Zcrease.
Myenteric plexus
or decrease in the intensity of immunoreactivity.
11
Fibers
The number of arrows indicates the degree of increase or
June 19!30
DIABETES AND THE SLJBMUCOUS PLEXUS
1429
Figure 1. Immunofluorescence micrographs of VIP-L1 in the submucous plexus of rat ileum; bars = 30 pm. A.
Control (8 wk).
B. Diabetic (8 wk after induction with streptozocin). C. Control D. Diabetic
(16
of diabetes
wk).
(16
wk after induction of diabetes].
?\leuropeptide
Y-Like Immunoreactivity
In the submucous plexus of the control rat ileum, NPY-like immunoreactive neurons and nerve fibers were observed, and the number of immunoreactive neurons in each ganglion was less than that observed for VIP-L1 (Figure 3A and C). In both 8- and 16-wk diabetes, there was an increase in immunoreactivity in a subpopulation of submucous neurons and their processes containing NPY-LI (Figure 3B and D). In the myenteric plexus of the control rat ileum, the distribution of NPY-like immunoreactive nerve fibers was sparse, mostly with single cell bodies and few
varicose nerve fibers in the plexus (figures are not included in this study). There was no change in either fluorescence intensity or density of NPY-like immunoreactive nerve fibers in the myenteric plexus of 8- or 16-wk diabetic rat ileum (figures not included). Dopamine fl-Hydroxylase-Like Immunoreactivity Nerve fibers stained with anti-DBH antiserum were seen around unstained ganglion cells of the submucous plexus of control rat ileum (Figure 4A and C). No positive DBH-like immunoreactive cell bodies
GASTROENTEROLOGY
Vol. 98. No. 6
Figure 2. Immunofluorescence micrographs of VIP-L1(A-C) and DBH-LI (D-F) in the myenteric plexus of rat ileum; bars = 30 pm. A. Control. B. Diabetic (8 wk after induction of diabetes). C. Diabetic (16wk). D. Control. E. Diabetic (8 wk). F. Diabetic (16wk).
were observed, and there was no change in immunoreactivity of these nerve fibers in the submucous plexus of the 8- or 16-wk diabetic rat ileum (Figure 4B and D). Dense DBH-like immunoreactive nerve fibers were seen in the myenteric plexus of the control rat ileum [Figure ZD). Faintly stained cell bodies were present in the myenteric ganglia of some of the control preparations (Figure 20). Unlike in the submucous plexus, there was a change in the distribution of DBH-LI in the myenteric plexus of the streptozocin-diabetic rat ileum. A decrease in the density of DBH-like immunoreactive fibers with enlarged varicosities at 8 wk was followed by a further decrease in the density of DBH-like immunoreactive nerve fibers and more distorted varicose structures after 16 wk of diabetes [Figure 2E and F). Calcitonin Gene-Related Immunoreactivity
Peptide-Like
Calcitonin gene-related peptide-like immunoreactive neurons and nerve fibers were observed in the submucous plexus of the control rat ileum (Figure
5A and C). There was a decrease in the density of CGRP-like immunoreactive nerve fibers in the submucous ganglia, and immunoreactive neurons were absent in the submucous plexus of both 8- and 16-wk diabetic rat ileum [Figure 5B and D). Dense CGRP-like immunoreactive nerve fibers and intensely stained cell bodies were observed in the myenteric plexus of control rat ileum (Figure 6A). In the rat ileum, there was a significant decrease in the density of CGRP-like immunoreactive nerve fibers, mainly around the ganglion cells (Figure 6s). However, intensely stained single immunoreactive nerve fibers running over the myenteric ganglia and through the internodal fibers were observed, and there were no positively stained cell bodies in the myenteric plexus of either 8- or 16-wk diabetic rat ileum (Figure 6B). After 16 wk of diabetes, CGRP-LI reappeared around the ganglion cells, although it was not completely recovered (Figure SC). Substance P-Like Jmmunoreactivity Substance P-like immunoreactive nerve fibers around unstained ganglion cells were observed in the
lune 1!390
DIABETES AND THE SUBMUCOUS
PLEXUS
1431
Figure 3. Immunofluorescence micrographs of NPY-LI in the submucous plexus of rat ileum; bars = 30 pm. A. Control (8 wk).
B. Diabetic (8 wk after induction of diabetes]. C. Control (16wk). D. Diabetic (16wk after induction of diabetes].
submucous plexus of the control rats (Figure 7A and C]. No substance P-like immunoreactive cell bodies were observed in the present study, suggesting that colchicine treatment of the tissue would be necessary to visualize the distribution of the cell bodies. There was no change in either intensity of immunoreactivity or distribution of substance P-like immunoreactive nerve fibers in the submucous plexus of the diabetic rat ileum (Figure 7B and D). There was no change in either the density or fluorescence intensity of substance P-like immunoreactive nerve fibers in the myenteric plexus of 8-wk
diabetic preparations compared with controls (Figure 6D and E). However, after 16 wk of diabetes, there was an increase in fluorescence intensity in substance P-like immunoreactive nerve fibers in the myenteric plexus of the diabetic rat ileum (Figure 6F). Discussion The results of this study show increase in VIP- and NPY-LI and CGRP-LI, but no change in DBH- and in the nerve fibers and neurons of
that there is an a decrease in substance P-L1 the submucous
1432 BELA1 AND BURNSTOCK
GASTROENTEROLOGY
Vol. 98, No. 6
Figure 4. Immunofluorescence micrographs of DBH-LI in the submucous plexus of rat ileum; bars = 30 pm. A. Control (8 wk). B. Diabetic (8 wk after induction of diabetes). C. Control (16wk). D. Diabetic (16wk after induction of diabetes).
plexus of streptozocin-diabetic rat ileum. The pattern and extent of changes of the neuroactive substances in the submucous plexus are different from those observed in the myenteric plexus of the same region of the diabetic rat ileum. The differential effect of diabetes on the different neuroactive substances in the two plexuses further supports the evidence that the changes observed are caused by the state of diabetes rather than by a nonspecific action of streptozocin on gastrointestinal nerve fibers [5,6). The changes in peptidergic nerves observed in the submucous plexus of the diabetic rat ileum showed major differences from those seen in the myenteric
plexus. The increased VIP-L1 in the submucous plexus of 8- and 16-wk streptozocin-diabetic rat ileum was observed in almost all neurons of the submucous ganglia containing VIP-LI. However, increased NPYLI was seen only in subpopulations of submucous neurons containing NPY-LI. There was no change in either the size of varicosities or the density of nerve fibers containing VIP-L1 in the diabetic tissues. In contrast, the increase in VIP-L1 in the myenteric plexus was associated with enlarged varicosities; this was followed at a later stage of diabetes by decreased density and fluorescence intensity of VIP-like immunoreactive nerve fibers. On the other hand, although
June 1990
DIABETES AND THE SUBMCJCOUS PLEXUS
1433
Figure 5. Immunofluorescence micrographs of CGRP-LIin the submucous plexus of rat ileum; bars = 30 pm. A. Control (8 wk). B. Diabetic (8 wk after induction of diabetes]. C. Control (16wk). D. Diabetic (16wk).
there was no change in the nerve fibers and neurons containing NPY-LI in the myenteric plexus after 8 and 16 wk of diabetes, there was an increase in the intensity of immunoreactivity in the neurons and nerve fibers of the submucous plexus of the same region of the diabetic rat ileum. Substance P-L1 in the submulcous plexus was not affected at either stage of diabetes. In contrast, there was an increase in substance P-L1 in the nerve fibers of the myenteric plexus of the ileum after 16 wk of diabetes, although no obvious change was observed 8 wk after induction of diabetes (5). The effect of diabetes on CGRP-LI in the
submucous plexus was similar to that observed in the myenteric plexus (6). However, unlike in the myenteric plexus, it was not clear whether the effect of diabetes was on the submucous CGRP-like immunoreactive nerve fibers of only intrinsic or of both intrinsic and extrinsic origins. On the whole, the effects of streptozocin diabetes on a single neuroactive substance in the two enteric plexuses are far from identical, suggesting that the role of a given neuroactive substance can vary depending on the projection of its neurons and the function of the innervated tissue. The state of diabetes has also shown differential
GASTROENTEROLOGY
Vol. 98. No. 6
Figure 6. Immunofluorescence micrographs of CGRP-Lf (A-C) and substance P-LI (D-F) in the myenteric plexus of ileum; bars = 30 pm. A. Control (no difference in immunoreactivity between 8- and 16-wk controls). B. Diabetic (8 wk). C. Diabetic (16 wk). D. Control. E. Diabetic (8 wk). F. Diabetic (16 wk).
effects on the adrenergic nerves of the two enteric plexuses. It is known that the myenteric and submucous plexuses, the mucosa, and the enteric arterioles are innervated by adrenergic nerve fibers of extrinsic origin (9). Although the adrenergic fibers are of extrinsic origin, the fibers innervating the two plexuses are reported to originate from different populations of neurons in the sympathetic ganglia (10). The adrenergic innervation of the submucous plexus in the diabetic rat ileum showed no sign of change either 8 or 16 wk after induction of diabetes. However, in the myenteric plexus of the same region of the diabetic rat ileum, there was a change in the adrenergic innervation in which the density of DBH-like immunoreactive nerve fibers was decreased and the structure of the varicosities was distorted after 8 and 16wk of diabetes. Although recent studies on the origins and projections of chemically identified enteric neurons have increased our understanding of the organization of the enteric plexuses, the physiological role of many of the neuroactive substances in the plexuses, especially in the submucous plexus, has not been established. The coexistence of VIP and NPY in single neurons in the submucous ganglia of rat intestine was reported by
Ekblad et al. (ll), and both VIP- and NPY-LI were shown to be increased in the diabetic submucous plexus in the present study. Thus the increased immunoreactivity in only a subpopulation of NPY-like immunoreactive submucous neurons of the diabetic rat ileum might be an indication that only those neurons containing NPY-LI that coexist with VIP are affected by the state of diabetes. Vasoactive intestinal polypeptide is reported to stimulate secretion across the mucosa (31,and the mucosal VIP-like immunoreactive nerve fibers are found to originate from submucous neurons (1). Furthermore, it is suggested that VIP is the likely noncholinergic enteric vasodilator in the intestine (12,131. Although the NPY-like immunoreactive nerve fibers around enteric arterioles are shown to be of extrinsic origin (14), the presence of intrinsic NPY-like immunoreactive nerve fibers in the submucous plexus has been reported previously (15). Miller (4) reported that addition of NPY to the serosal side of ileal mucosa caused a rapid depression of transmural potential difference and short-circuit current in guinea pig and rabbit ileum. Using experiments with radioactive tracers, he has also shown that NPY enhances NaCl absorption and reduces anion secretion in guinea
June 1990
Figure 7. Immunofl of substance P-LI inI the rat ileum; bars = 30 pm.
DIABETES AND THE SUBMLJCOUS PLEXUS
1435
ce mic:rograFIhs lU( :ous : of
A. Control (8 wk).
B. Diabetic (8 wk). C. Control (16 wk). D. Diabetic (16 wk).
pig and rabbit ileum (4). Because VIP and NPY are the two peptides suggested to be involved in the nonadrenergic, noncholinergic electrolyte transport across the mucosa (3,4), the next logical progression will be to investigate the functional implication of the increased VIP- and NPY-LI in the submucous plexus of the diabetic rat ileum. The lack of documented reports of the effect of diabetes on intestinal mucosa in diabetic patients and our limited knowledge of the role of the different neuroactive substances in the submucous plexus restrict the interpretation of the present results. However, the difference in the effects of streptozocin
diabetes on a single neuroactive substance in the two plexuses demonstrates that factors other than neurotransmitter types are involved in the manifestation of changes observed in diabetic neuropathy.
References 1. Keast JR, Furness
JB, Costa M. The origins of peptide and norepinephrine nerves in the mucosa of the guinea-pig small intestine. Gastroenterology 1984;86:637-644. 2. Ekblad E, Winther C. Ekman R, Hakanson R, Sundler F. Projections of peptide-containing neurons in rat small intestine. Neuroscience 1987;20:169-188.
GASTROENTEROLOGY
1436 BELA1 AND BURNSTOCK
3. Tapper EJ. Local modulation of intestinal ion transport by enteric neurons. Am J Physiol1983;244:G456-G468. 4. Miller RJ. Control of epithelial ion transport by neuropeptides. Regul Pept 1985;4(Suppl):203-208. 5. Belai A, Lincoln J, Milner P, Crowe R. Loesch A, Burnstock G. Enteric nerves in diabetic rats: increase in vasoactive intestinal polypeptide but not substance P. Gastroenterology 1985;89:967976. 6. Belai A, Burnstock G. Selective damage of intrinsic calcitonin gene-related peptide-like immunoreactive enteric nerve fibers in streptozotocin-induced diabetic rats. Gastroenterology 1987; 92:730-734. 7. Cowen T, Haven
8.
9.
10.
11.
AJ, Burnstock G. Pontamine sky blue: a counterstain for background autofluorescence in catecholamine fluorescence and immunofluorescence histochemistry. Histochemistry 1985;82:205-208. Lincoln J, Bokor JT, Crowe R, Griffith SG, Haven AJ, Burnstock G. Myenteric plexus in streptozotocin-treated rats. Neurochemical and histochemical evidence for diabetic neuropathy in the gut. Gastroenterology 1984;86:654-661. Costa M, Furness JB. Somatostatin is present in a subpopulation of noradrenergic nerve fibres supplying the intestine. Neuroscience 1984;13:911-920. Costa M, Furness JB. The origins of the adrenergic fibres which innervate the internal anal sphincter, the rectum and other tissues of the pelvic region in the guinea-pig. 2 Anat EntwicklGesch 1973;140:129-142. Ekblad E, Hakanson R, Sundler F. VIP and PHI coexist with an
12. 13.
14.
15.
Vol. 98, No. 6
NPY-like peptide in intramural neurons of the small intestine. Regul Pept 1984;10:47-55. Said SI, Mutt V. Potent peripheral and splanchnic vasodilator peptide from normal gut. Nature 1970;225:863-864. Eklund S, Jodal M, Lundgren 0, Sjaqvist A. Effectsof vasoactive intestinal polypeptide on blood flow, motility and fluid transport in the gastrointestinal tract of the cat. Acta Physiol Stand 1979;105:461-468. Sundler F, Moghimazadeh E, Hakanson R, Ekelund M, Emson PC. Nerve fibres in the gut and pancreas of the rat displaying neuropeptide Y immunoreactivity. Cell Tissue Res 1983;230:487496. Furness JB, Costa M, Emson PC, Hakanson R, Moghimzadeh E, Sundler F, Taylor IL, Chance RE. Distribution, pathways and reactions to drug treatment of nerves with neuropeptide Y- and pancreatic polypeptide-like immunoreactivity in the guinea-pig digestive tract. Cell Tissue Res 1983;234:71-92.
Received March 25,1988. Accepted November 15.1989. Address requests for reprints to: Professor G. Burnstock, Department of Anatomy and Developmental Biology, University College London, Gower Street, London WClE SBT, England. This research was supported in part by a grant from the Medical Research Council of Great Britain. The authors are grateful to Dr. S. Cornish and Ms. C. Sardon for assistance in the preparation of the manuscript.
1990;98:1427-1436
Changes in Adrenergic and Peptidergic Nerves in the Submucous Plexus of Streptozocin-Diabetic Rat Ileum ABEBECH
BELA1 and GEOFFREY
Department of Anatomy and Developmental College London, London, England
The effect of streptozocin diabetes on the distribution of adrenergic and peptidergic nerves in the submucous plexus of rat ileum was investigated and compared with the changes in the myenteric plexus of the same region of ileum. There was an increase in the intensity of immunoreactivity in vasoactive intestinal polypeptide- and neuropeptide Y-like immunoreactive nerve fibers and neurons and a decrease in calcitonin gene-related peptide-like immunoreactivity but no change in substance P- and dopamine P-hydroxylase-like immunoreactivity in the nerve fibers and neurons of the submucous plexus of both 8- and 16-wk streptozocin-diabetic rat ileum. However, in the myenteric plexus of the diabetic rat ileum, there was enlargement of varicosities and an increase followed by a slight decrease in the intensity of immunoreactivity of vasoactive intestinal polypeptide- and dopamine &hydroxylase-like immunoreactive nerve fibers and neurons, increased substance P-like immunoreactivity in diabetes at 16 wk, and an initial decrease (at 8 wk) followed by a recovery of calcitonin gene-related peptide-like immunoreactivity at 16 wk, but no change in neuropeptide Y-like immunoreactivity. The markedly different changes in peptidergic and adrenergic nerves between the two enteric plexuses show that diabetic neuropathy induced by streptozocin is not selective and involves factors other than neurotransmitter types.
I
BURNSTOCK
Biology and Centre for Neuroscience,
mmunohistochemical studies have demonstrated the presence of nonadrenergic, noncholinergic neurons i,n the submucous plexus of many mammalian species, and the nerve fibers of most of these neurons have been traced to the mucosa of the small intestine (1,~). Nerve fibers in the mucosa showing immunoreactivity to vasoactive intestinal polypeptide (VIP), somatostatin, cholecystokinin, calcitonin gene-related peptide (CGRP), and neuropeptide Y (NPY) have been
University
found to arise from cell bodies in the submucous plexus, whereas substance P fibers arise partly from the myenteric plexus but mostly from neurons in the submucous plexus (12). Tapper (31 has reported that noradrenaline and somatostatin are able to increase absorptive fluxes of intestinal ions, while acetylcholine, VIP, and substance P enhance secretion in the intestinal mucosa. Recently, Miller (4) reported that addition of NPY to the serosal side of the ileal mucosa caused a rapid depression of transmural potential difference and short-circuit current in the guinea pig ileum. Functional and neuronal projection studies in the submucous plexus have implied that the majority of submucous neurons act as secretomotor neurons for the mucosa, although it is also known that enteric neurons, probably of submucous origin, also participate in the control of intestinal blood flow (1). Although recent structural and functional studies add to our understanding of the submucous plexus, the significance of the presence of several peptidecontaining neurons in relation to mucosal function and their complex projections and interconnections has not yet been fully established. In this report, we have conducted a comparative study of the effect of short- and long-term diabetes on VIP-, substance P-, dopamine p-hydroxylase (DBH)-, CGRP-, and NPYlike immunoreactivity (LI) in the submucous and myenteric plexuses of streptozocin-diabetic rat ileum using immunohistochemical techniques. Materials and Methods Male Wistar rats weighing 400-450 g were injected with buffered streptozocin (65 mg/kg body wt). The onset of Abbreviations used in this paper: CGRP, calcitonin generelated peptide; DBH, dopamine fl-hydroxylase; -LI, -like immunoreactivity: NPY, neuropeptide Y; PBS, phosphate-buffered saline; VIP, vasoactive intestinal polypeptide. @ 1990 by the American Gastroenterological Association 0016-5065/90/$3.00
1428 BELA1 AND BURNSTOCK
GASTROENTEROLOGY
diabetes was established by the presence of rapid weight loss, polyuria, and glycosuria (Clinitest reagent tablets; Miles, Stoke Poges, Slough, England]. Untreated controls were of the same initial weight range as the diabetic group and both groups were maintained under the same conditions, supplied with food and water ad libitum until death. One group each of control and treated rats was kept for 8 wk and another group of each for 16 wk after the streptozocin injection. Blood samples were taken under ether anesthesia from the posterior vena cava for blood glucose analysis and the animals were killed. Segments of ileum from both control and diabetic rats were dissected out, cleaned and stretched out onto strips of Sylgard silicon rubber (Dow Corning, Seneffe, Belgium] with the mucosal side down, and fixed in 4% paraformaldehyde in phosphate-buffered saline [PBS] for z h at 4°C. The whole-mount preparation for investigating the myenteric plexus was performed as described previously (5,6). The submucous plexus was exposed after peeling off the muscle layers together with the myenteric plexus, and the mucosa was removed with a pair of fine forceps. The submucous and myenteric preparations were then washed 10 times each for lo-min periods with 80% alcohol, dehydrated, rehydrated and washed 3 times for 5 min with PBS containing 0.1% Triton X-100 (Sigma, Poole, England]. The tissues were incubated with polyclonal antisera against VIP, substance P, DBH (a marker for noradrenergic nerves], CGRP, and NPY at a dilution of 1200 in PBS in a humid chamber for 18 hat room temperature. The tissues were then washed with PBS/Triton X-100 and incubated with fluorescein isothiocyanate-conjugated goat anti-rabbit immunoglobulin at a dilution of 1:40 (Nordic Immunological Laboratories, Tilburg, The Netherlands] at room temperature for 1 h.
The tissues were subsequently washed with PBS/Triton X-100 and stained
with 0.05% Pontamine Sky Blue (BDH Ltd., Poole, England] in PBS and 1% dimethyl
Chemicals
sulphoxide for 30 min at room temperature to reduce background autofluorescence (7). The immunoreactive neurons and nerve fibres in the submucous and myenteric plexuses were viewed using a Zeiss microscope [Carl Zeiss Inc., Thornwood, N.Y.] equipped for viewing fluorescein isothiocyanate fluorescence. The specimens were coded and read blindly.
Vol. 98. No. 6
Results The general features of rats 8 wk after induction of diabetes with streptozocin have been reported previously (5,8), and there was no significant difference in the general features of 16-wk diabetic rats, except that two of the latter were blind (had developed cataract] at the time of death. Briefly, the diabetic rats were severly hyperglycemic [with an average blood glucose level of 492 * 15.4 mg/dl), and distension of the large and small intestines and the presence of pale watery stools were observed at the time of death. A summary of the detailed immunohistochemical observations is presented in Table 1.
Vasoactive Intestinal Polypeptide-Like lmmunoreactivity Vasoactive intestinal polypeptide-like immunoreactive neurons and nerve fibers were observed in the submucous plexus of the ileum of control rats, with an average of 5-i’ immunoreactive cell bodies in each ganglion (Figure 1A and C]. In both 8- and 16-wk diabetic preparations, the VIP-like immunoreactive neurons and some of their processes were intensely stained compared with age-matched controls (Figure 1B and D). The increased immunoreactivity was observed in all VIP-like immunoreactive cell bodies in the ganglia. There was no obvious change in the structure of the varicose nerve fibers. Vasoactive intestinal polypeptide-like immunoreactive nerve fibers in the primary and secondary ganglia and internodal fibers of the myenteric plexus of the control rat ileum were observed (Figure ZA). There was an increase in fluorescence intensity, the density of VIP-like immunoreactive fibers in the myenteric ganglia was reduced [Figure 2B), and more prominent enlarged varicosities were observed after 16 wk of diabetes [Figure 2C).
Table 1. Changes in the Intensity of the Different Neuroactive Substances in the Cell Bodies and Nerve Fibers of the Submucous and Myenteric Plexuses of 8- and 16-wk Diabetic Rat Ileum Compared With Age-Matched Controls Submucous plexus 8-wk diabetic Neuroactive substances DBH-LI VIP-L1 NPY-LI Substance CGRP-LI
16-wk diabetic
8-wk diabetic
Cell bodies
Fibers
Cell bodies
Fibers
i
i
7
i
7
I
5
5
Cell bodies
:r
16-wk diabetic
Fibers
Cell bodies
rft -
i
i
-
11
11
T 1
P-L1
No change: t, 1 increase
Zcrease.
Myenteric plexus
or decrease in the intensity of immunoreactivity.
11
Fibers
The number of arrows indicates the degree of increase or
June 19!30
DIABETES AND THE SLJBMUCOUS PLEXUS
1429
Figure 1. Immunofluorescence micrographs of VIP-L1 in the submucous plexus of rat ileum; bars = 30 pm. A.
Control (8 wk).
B. Diabetic (8 wk after induction with streptozocin). C. Control D. Diabetic
(16
of diabetes
wk).
(16
wk after induction of diabetes].
?\leuropeptide
Y-Like Immunoreactivity
In the submucous plexus of the control rat ileum, NPY-like immunoreactive neurons and nerve fibers were observed, and the number of immunoreactive neurons in each ganglion was less than that observed for VIP-L1 (Figure 3A and C). In both 8- and 16-wk diabetes, there was an increase in immunoreactivity in a subpopulation of submucous neurons and their processes containing NPY-LI (Figure 3B and D). In the myenteric plexus of the control rat ileum, the distribution of NPY-like immunoreactive nerve fibers was sparse, mostly with single cell bodies and few
varicose nerve fibers in the plexus (figures are not included in this study). There was no change in either fluorescence intensity or density of NPY-like immunoreactive nerve fibers in the myenteric plexus of 8- or 16-wk diabetic rat ileum (figures not included). Dopamine fl-Hydroxylase-Like Immunoreactivity Nerve fibers stained with anti-DBH antiserum were seen around unstained ganglion cells of the submucous plexus of control rat ileum (Figure 4A and C). No positive DBH-like immunoreactive cell bodies
GASTROENTEROLOGY
Vol. 98. No. 6
Figure 2. Immunofluorescence micrographs of VIP-L1(A-C) and DBH-LI (D-F) in the myenteric plexus of rat ileum; bars = 30 pm. A. Control. B. Diabetic (8 wk after induction of diabetes). C. Diabetic (16wk). D. Control. E. Diabetic (8 wk). F. Diabetic (16wk).
were observed, and there was no change in immunoreactivity of these nerve fibers in the submucous plexus of the 8- or 16-wk diabetic rat ileum (Figure 4B and D). Dense DBH-like immunoreactive nerve fibers were seen in the myenteric plexus of the control rat ileum [Figure ZD). Faintly stained cell bodies were present in the myenteric ganglia of some of the control preparations (Figure 20). Unlike in the submucous plexus, there was a change in the distribution of DBH-LI in the myenteric plexus of the streptozocin-diabetic rat ileum. A decrease in the density of DBH-like immunoreactive fibers with enlarged varicosities at 8 wk was followed by a further decrease in the density of DBH-like immunoreactive nerve fibers and more distorted varicose structures after 16 wk of diabetes [Figure 2E and F). Calcitonin Gene-Related Immunoreactivity
Peptide-Like
Calcitonin gene-related peptide-like immunoreactive neurons and nerve fibers were observed in the submucous plexus of the control rat ileum (Figure
5A and C). There was a decrease in the density of CGRP-like immunoreactive nerve fibers in the submucous ganglia, and immunoreactive neurons were absent in the submucous plexus of both 8- and 16-wk diabetic rat ileum [Figure 5B and D). Dense CGRP-like immunoreactive nerve fibers and intensely stained cell bodies were observed in the myenteric plexus of control rat ileum (Figure 6A). In the rat ileum, there was a significant decrease in the density of CGRP-like immunoreactive nerve fibers, mainly around the ganglion cells (Figure 6s). However, intensely stained single immunoreactive nerve fibers running over the myenteric ganglia and through the internodal fibers were observed, and there were no positively stained cell bodies in the myenteric plexus of either 8- or 16-wk diabetic rat ileum (Figure 6B). After 16 wk of diabetes, CGRP-LI reappeared around the ganglion cells, although it was not completely recovered (Figure SC). Substance P-Like Jmmunoreactivity Substance P-like immunoreactive nerve fibers around unstained ganglion cells were observed in the
lune 1!390
DIABETES AND THE SUBMUCOUS
PLEXUS
1431
Figure 3. Immunofluorescence micrographs of NPY-LI in the submucous plexus of rat ileum; bars = 30 pm. A. Control (8 wk).
B. Diabetic (8 wk after induction of diabetes]. C. Control (16wk). D. Diabetic (16wk after induction of diabetes].
submucous plexus of the control rats (Figure 7A and C]. No substance P-like immunoreactive cell bodies were observed in the present study, suggesting that colchicine treatment of the tissue would be necessary to visualize the distribution of the cell bodies. There was no change in either intensity of immunoreactivity or distribution of substance P-like immunoreactive nerve fibers in the submucous plexus of the diabetic rat ileum (Figure 7B and D). There was no change in either the density or fluorescence intensity of substance P-like immunoreactive nerve fibers in the myenteric plexus of 8-wk
diabetic preparations compared with controls (Figure 6D and E). However, after 16 wk of diabetes, there was an increase in fluorescence intensity in substance P-like immunoreactive nerve fibers in the myenteric plexus of the diabetic rat ileum (Figure 6F). Discussion The results of this study show increase in VIP- and NPY-LI and CGRP-LI, but no change in DBH- and in the nerve fibers and neurons of
that there is an a decrease in substance P-L1 the submucous
1432 BELA1 AND BURNSTOCK
GASTROENTEROLOGY
Vol. 98, No. 6
Figure 4. Immunofluorescence micrographs of DBH-LI in the submucous plexus of rat ileum; bars = 30 pm. A. Control (8 wk). B. Diabetic (8 wk after induction of diabetes). C. Control (16wk). D. Diabetic (16wk after induction of diabetes).
plexus of streptozocin-diabetic rat ileum. The pattern and extent of changes of the neuroactive substances in the submucous plexus are different from those observed in the myenteric plexus of the same region of the diabetic rat ileum. The differential effect of diabetes on the different neuroactive substances in the two plexuses further supports the evidence that the changes observed are caused by the state of diabetes rather than by a nonspecific action of streptozocin on gastrointestinal nerve fibers [5,6). The changes in peptidergic nerves observed in the submucous plexus of the diabetic rat ileum showed major differences from those seen in the myenteric
plexus. The increased VIP-L1 in the submucous plexus of 8- and 16-wk streptozocin-diabetic rat ileum was observed in almost all neurons of the submucous ganglia containing VIP-LI. However, increased NPYLI was seen only in subpopulations of submucous neurons containing NPY-LI. There was no change in either the size of varicosities or the density of nerve fibers containing VIP-L1 in the diabetic tissues. In contrast, the increase in VIP-L1 in the myenteric plexus was associated with enlarged varicosities; this was followed at a later stage of diabetes by decreased density and fluorescence intensity of VIP-like immunoreactive nerve fibers. On the other hand, although
June 1990
DIABETES AND THE SUBMCJCOUS PLEXUS
1433
Figure 5. Immunofluorescence micrographs of CGRP-LIin the submucous plexus of rat ileum; bars = 30 pm. A. Control (8 wk). B. Diabetic (8 wk after induction of diabetes]. C. Control (16wk). D. Diabetic (16wk).
there was no change in the nerve fibers and neurons containing NPY-LI in the myenteric plexus after 8 and 16 wk of diabetes, there was an increase in the intensity of immunoreactivity in the neurons and nerve fibers of the submucous plexus of the same region of the diabetic rat ileum. Substance P-L1 in the submulcous plexus was not affected at either stage of diabetes. In contrast, there was an increase in substance P-L1 in the nerve fibers of the myenteric plexus of the ileum after 16 wk of diabetes, although no obvious change was observed 8 wk after induction of diabetes (5). The effect of diabetes on CGRP-LI in the
submucous plexus was similar to that observed in the myenteric plexus (6). However, unlike in the myenteric plexus, it was not clear whether the effect of diabetes was on the submucous CGRP-like immunoreactive nerve fibers of only intrinsic or of both intrinsic and extrinsic origins. On the whole, the effects of streptozocin diabetes on a single neuroactive substance in the two enteric plexuses are far from identical, suggesting that the role of a given neuroactive substance can vary depending on the projection of its neurons and the function of the innervated tissue. The state of diabetes has also shown differential
GASTROENTEROLOGY
Vol. 98. No. 6
Figure 6. Immunofluorescence micrographs of CGRP-Lf (A-C) and substance P-LI (D-F) in the myenteric plexus of ileum; bars = 30 pm. A. Control (no difference in immunoreactivity between 8- and 16-wk controls). B. Diabetic (8 wk). C. Diabetic (16 wk). D. Control. E. Diabetic (8 wk). F. Diabetic (16 wk).
effects on the adrenergic nerves of the two enteric plexuses. It is known that the myenteric and submucous plexuses, the mucosa, and the enteric arterioles are innervated by adrenergic nerve fibers of extrinsic origin (9). Although the adrenergic fibers are of extrinsic origin, the fibers innervating the two plexuses are reported to originate from different populations of neurons in the sympathetic ganglia (10). The adrenergic innervation of the submucous plexus in the diabetic rat ileum showed no sign of change either 8 or 16 wk after induction of diabetes. However, in the myenteric plexus of the same region of the diabetic rat ileum, there was a change in the adrenergic innervation in which the density of DBH-like immunoreactive nerve fibers was decreased and the structure of the varicosities was distorted after 8 and 16wk of diabetes. Although recent studies on the origins and projections of chemically identified enteric neurons have increased our understanding of the organization of the enteric plexuses, the physiological role of many of the neuroactive substances in the plexuses, especially in the submucous plexus, has not been established. The coexistence of VIP and NPY in single neurons in the submucous ganglia of rat intestine was reported by
Ekblad et al. (ll), and both VIP- and NPY-LI were shown to be increased in the diabetic submucous plexus in the present study. Thus the increased immunoreactivity in only a subpopulation of NPY-like immunoreactive submucous neurons of the diabetic rat ileum might be an indication that only those neurons containing NPY-LI that coexist with VIP are affected by the state of diabetes. Vasoactive intestinal polypeptide is reported to stimulate secretion across the mucosa (31,and the mucosal VIP-like immunoreactive nerve fibers are found to originate from submucous neurons (1). Furthermore, it is suggested that VIP is the likely noncholinergic enteric vasodilator in the intestine (12,131. Although the NPY-like immunoreactive nerve fibers around enteric arterioles are shown to be of extrinsic origin (14), the presence of intrinsic NPY-like immunoreactive nerve fibers in the submucous plexus has been reported previously (15). Miller (4) reported that addition of NPY to the serosal side of ileal mucosa caused a rapid depression of transmural potential difference and short-circuit current in guinea pig and rabbit ileum. Using experiments with radioactive tracers, he has also shown that NPY enhances NaCl absorption and reduces anion secretion in guinea
June 1990
Figure 7. Immunofl of substance P-LI inI the rat ileum; bars = 30 pm.
DIABETES AND THE SUBMLJCOUS PLEXUS
1435
ce mic:rograFIhs lU( :ous : of
A. Control (8 wk).
B. Diabetic (8 wk). C. Control (16 wk). D. Diabetic (16 wk).
pig and rabbit ileum (4). Because VIP and NPY are the two peptides suggested to be involved in the nonadrenergic, noncholinergic electrolyte transport across the mucosa (3,4), the next logical progression will be to investigate the functional implication of the increased VIP- and NPY-LI in the submucous plexus of the diabetic rat ileum. The lack of documented reports of the effect of diabetes on intestinal mucosa in diabetic patients and our limited knowledge of the role of the different neuroactive substances in the submucous plexus restrict the interpretation of the present results. However, the difference in the effects of streptozocin
diabetes on a single neuroactive substance in the two plexuses demonstrates that factors other than neurotransmitter types are involved in the manifestation of changes observed in diabetic neuropathy.
References 1. Keast JR, Furness
JB, Costa M. The origins of peptide and norepinephrine nerves in the mucosa of the guinea-pig small intestine. Gastroenterology 1984;86:637-644. 2. Ekblad E, Winther C. Ekman R, Hakanson R, Sundler F. Projections of peptide-containing neurons in rat small intestine. Neuroscience 1987;20:169-188.
GASTROENTEROLOGY
1436 BELA1 AND BURNSTOCK
3. Tapper EJ. Local modulation of intestinal ion transport by enteric neurons. Am J Physiol1983;244:G456-G468. 4. Miller RJ. Control of epithelial ion transport by neuropeptides. Regul Pept 1985;4(Suppl):203-208. 5. Belai A, Lincoln J, Milner P, Crowe R. Loesch A, Burnstock G. Enteric nerves in diabetic rats: increase in vasoactive intestinal polypeptide but not substance P. Gastroenterology 1985;89:967976. 6. Belai A, Burnstock G. Selective damage of intrinsic calcitonin gene-related peptide-like immunoreactive enteric nerve fibers in streptozotocin-induced diabetic rats. Gastroenterology 1987; 92:730-734. 7. Cowen T, Haven
8.
9.
10.
11.
AJ, Burnstock G. Pontamine sky blue: a counterstain for background autofluorescence in catecholamine fluorescence and immunofluorescence histochemistry. Histochemistry 1985;82:205-208. Lincoln J, Bokor JT, Crowe R, Griffith SG, Haven AJ, Burnstock G. Myenteric plexus in streptozotocin-treated rats. Neurochemical and histochemical evidence for diabetic neuropathy in the gut. Gastroenterology 1984;86:654-661. Costa M, Furness JB. Somatostatin is present in a subpopulation of noradrenergic nerve fibres supplying the intestine. Neuroscience 1984;13:911-920. Costa M, Furness JB. The origins of the adrenergic fibres which innervate the internal anal sphincter, the rectum and other tissues of the pelvic region in the guinea-pig. 2 Anat EntwicklGesch 1973;140:129-142. Ekblad E, Hakanson R, Sundler F. VIP and PHI coexist with an
12. 13.
14.
15.
Vol. 98, No. 6
NPY-like peptide in intramural neurons of the small intestine. Regul Pept 1984;10:47-55. Said SI, Mutt V. Potent peripheral and splanchnic vasodilator peptide from normal gut. Nature 1970;225:863-864. Eklund S, Jodal M, Lundgren 0, Sjaqvist A. Effectsof vasoactive intestinal polypeptide on blood flow, motility and fluid transport in the gastrointestinal tract of the cat. Acta Physiol Stand 1979;105:461-468. Sundler F, Moghimazadeh E, Hakanson R, Ekelund M, Emson PC. Nerve fibres in the gut and pancreas of the rat displaying neuropeptide Y immunoreactivity. Cell Tissue Res 1983;230:487496. Furness JB, Costa M, Emson PC, Hakanson R, Moghimzadeh E, Sundler F, Taylor IL, Chance RE. Distribution, pathways and reactions to drug treatment of nerves with neuropeptide Y- and pancreatic polypeptide-like immunoreactivity in the guinea-pig digestive tract. Cell Tissue Res 1983;234:71-92.
Received March 25,1988. Accepted November 15.1989. Address requests for reprints to: Professor G. Burnstock, Department of Anatomy and Developmental Biology, University College London, Gower Street, London WClE SBT, England. This research was supported in part by a grant from the Medical Research Council of Great Britain. The authors are grateful to Dr. S. Cornish and Ms. C. Sardon for assistance in the preparation of the manuscript.