alterations in enteric nerve and smooth-muscle function in inflammatory bowel diseases

Inflammarory Bowel Direases@ 338-48 D 1997 Crohn’s & Colitis Foundation of America, Inc.

Basic Science Review

Alterations in Enteric Nerve and Smooth-Muscle Function in Inflammatory Bowel Diseases

Stephen M. Collins, Gert Van Assche, and Cory Hogaboam Gastroenterology Division and Intestinal Diseases Research Unit, McMaster University, Hamilton, Ontario, Canada

Summary: Although there is much focus on the factors that lead to the expression of human inflammatory bowel disease (IBD), with a view to identifying the cause(s) and cure, work is still required to better understand the patho- physiology of the disease, with views to developing innova- tive treatment approaches for those patients with established and often complicated disease. This review examines the nature and extent of enteric nerve and smooth-muscle involvement in IBD. Parallels are drawn with recent animal studies, which clearly demonstrate that the deeper neuromuscular tissues of the gut are rapidly and profoundly altered after even superficial degrees of inflammation in the mucosa, and that this can occur without penetration of these tissues by the inflammatory response. In the enteric nervous system, there are quantitative and qualitative changes in neurotransmitter content, and recent work has

shown that these agents (e.g., substance P) possess proinflammatory properties, whereas other (e.g., calcitonin gene-related peptide) possess antiinflammatory actions. Thus, the balance of neurotransmission clearly is important in determining the expression of diseases. With respect to muscle, inflammation-induced changes lead to alterations in contractility, which contribute to altered motility. There are also trophic changes and collagen formation that contribute to strictures of Crohn’s disease. More recent work has focused on the ability of muscle to influence immune function through cytokine production, antigen presentation, and adhesion molecule expression to activate T lymphocytes. Thus, the neuromuscular tissues act not just as “innocent bystanders” but also as “active participants” in the inflamed gut. Key Words: Smooth muscle-Nerves- Cytokines-Motility-Immune system-Lymphocytes.

Motility patterns in the gastrointestinal tract are the result of a continuous modulation of smooth- muscle contractility by highly organized neural circuits and hormonal influences. Alterations of the extrinsic or intrinsic neural circuits innervating the intestine will provoke profound motility disturbances. Human inflammatory bowel disease (IBD) is a mucosal disorder in origin, but transmural involvement and general immune activation during disease progression will affect the submucosal and myenteric neural plexus. Enteric nerves may have a dual role in the clinical presentation of bowel inflammation (1,2). Disruption of normal motility patterns will contribute to common symptoms such as diarrhea and abdominal cramps. Neurotransmitters,

Address correspondence and reprint requests to Dr. Stephen Collins, Room 4W8, Gastroenterology Division, McMaster Medi- cal Centre, 12000 Main Street West, Hamilton, Ontario, Canada L8N 325.

Manuscript received and accepted November 25, 1996.

particularly neuropeptides, released in the neuromuscular layers of the intestine may have a role in the generation or maintenance of inflammatory reac- tions in situ.

The autonomic nervous system innervates the gastrointestinal tract as it does most visceral organs. A unique feature of the intestinal innervation, however, is the existence of the highly organized enteric nervous system, which is entirely intrinsic (all nerve fibers travel within the bowel wall) (1). Nonetheless, the enteric nervous system connects with most of the extrinsic sympathetic and parasympathetic nerve fibers through synapses. The majority of extrinsic nerve fibers innervating the gut are thought to be afferent, carrying information back to the CNS. Therefore, the enteric nervous system probably con- tains the main hardwiring for intestinal motility patterns. Another interesting feature of the intestinal nerves is the high degree of neurotransmitter colocal- ization, with most fibers containing more than one neurotransmitter (2). Presynaptic nerve terminals

38

MUSCLE AND NERVES IN IBD 39

contain multiple neurotransmitters, including neuropeptides (3). Several of these peptides are considered putative mediators of inflammation. Involvement of enteric nerves, along with extrinsic nerves, has been widely documented in human and animal bowel inflammation. Both structural and functional alterations of the intestinal innervation occur in clinical IBD and in animal models of inflammation.

STRUCTURAL CHANGES IN THE INFLAMED NERVOUS SYSTEM

Observations in Human IBD

Structural alterations of enteric nerves have been reported in both ulcerative colitis and Crohn’s disease. The main histological findings include (a) ganglion cell hypertrophy and hyperplasia, (b) axonal necrosis, (c) alterations in neuropeptide innervation, and (d) neural damage in both inflamed and noninflamed areas.

A wide variety of proliferative and degenerative alterations of enteric nerves have been documented in resection specimens or biopsies from chronically inflamed intestine. Storsteen and co-workers (4) were the first to show an increase in the number of ganglion cells in the myenteric plexus of patients with ulcerative colitis. This increase in ganglion cell numbers is not related to local edema or connective tissue expansion during inflammation, because ganglion cell counts are independent of wall thickness (5). Adja- cent, noninflamed areas of the intestine also display a higher ganglion cell density. In addition, inflammatory disease of longer duration results in higher ganglion cell numbers, suggesting that the proliferative response is acquired during the course of disease. Centrioles were found to be more prominent in ganglion cells of Crohn’s disease by Siemers and Dobbins (6). However, these authors did not confirm the quantitative hyperplasia of ganglion cells. In the mucosa of ulcerative colitis biopsies, proliferation of adrenergic innervation is most predominant (7,8). This may be a result of increased density of perivascular nerves. In the deeper, neuromuscular layers of the intestinal wall, a similar preponderance of sympathetic nerve fibers has been observed (9-11). Little is known about growth factors mediating the hyperplastic response of ganglion cells to intestinal inflammation. Nerve growth factor immunostaining, however, is significantly increased in regional enteritis, suggesting that this peptide may be involved in nerve hyperplasia (12).

Damage to enteric ganglia was first reported in human parasitism, more particularly in the acquired megacolon of Chagas’ disease. Degeneration of intramural ganglion cells along with hyperplasia and glial cell proliferation was described by Oehmichen and Reifferscheid (13) in a variety of chronic intestinal disorders, including two cases of Crohn’s disease. This combination of proliferative and injurious changes in the autonomic nerves of Crohn’s disease was confirmed at the ultrastructural level with electron microscopy. Axonal necrosis was found in both diseased areas and grossly normal resection margins in accordance with previous observations (9,lO). Widespread, severe, axonal necrosis has been pro- posed by several authors as a histological marker discriminating between Crohn’s disease and other inflammatory bowel disorders (1 1). Axonal degeneration is not present in cases of radiation enteritis, suggesting that the transmural involvement of Crohn’s disease is not sufficient to explain the histological findings. Neurotoxicity may be a specific feature of the immune response in chronic intestinal inflammation.

Resident cells in the myenteric plexus are thought to be involved in the initiation of inflammation. Major histocompatibility complex (MHC) class I1 antigens are expressed by the enteroglial sheet surrounding the myenteric nerve fibers (14). MHC class I1 glyco- proteins are crucial in antigen presentation to CD4+ T cells. Therefore, immunoreactions in situ may be triggered by these molecules, causing axonal damage in myenteric neurons.

Peptidergic neurotransmitters are probably more predominant in the enteric nervous system than in any other neural tissue. Neuropeptides are known to be of crucial importance in normal intestinal motility. Substance P (SP) and vasoactive intestinal peptide (VIP) have been studied most extensively in human IBD because of their established role in inflammation at other sites in the body (15). More recently, calcitonin gene-related peptide and somatostatin containing nerves have been shown to be affected by intestinal inflammation.

Enterochromaffin cells, producing intestinal peptides, were reported to be decreased in rectal biopsies of ulcerative colitis by Ahonen et al. (16). Subsequently, VIP-encoding nerves were shown to be increased and structurally altered in resection specimens and rectal biopsies of Crohn’s disease by Bishop et al. (17) and O’Morain et al. (18). The VIP- immunoreactive content of the intestinal tissue as measured by radioimmunoassay was found to be sub-

Inflammatory Bowel Diseases@, Vol. 3, No. I , 1997

40 S. M. COLLINS ET AL.

stantially elevated in Crohn's disease. Changes in VIP innervation were present in inflamed and noninflamed specimens. Duffy et al. (19) even suggested VIP plasma levels to be used as a laboratory supplement to score disease activity in IBD. Other studies, however, have shown unchanged and even decreased levels of VIP in Crohn's disease (20-22). The reports on elevated SP in human IBD have been more consis- tent. Substance P levels and SP receptor density are increased in both Crohn's disease and ulcerative colitis (21,23,24). The somatostatin immunoreactivity in the descending colon of IBD patients is significantly decreased in both endocrine cells and ganglion cells (23,25). Taken together, these studies provide sub- stantial evidence of alterations in neuropeptide contents in the enteric nervous plexus of inflamed intestine. If damage to peptidergic innervation is sec- ondary to local inflammation, disease activity and duration will determine the histological findings. The conflicting reports on VIP contents in inflamed intestine might be explained by heterogeneity of disease activity or drug regimens in the patients studied.

Studies in Animal Models of Intestinal Inflammation

Neuropeptides have recently been studied in experimental inflammation. SP levels were found to be increased in both mucosa and neuromuscular layers of rat jejunum after infection with the enteric parasite, Trichinella spiralis (26). This increase in SP immunoreactivity seems to be localized in enteric nerves, because treatment of tissues with scorpion venom and capsaicin, both known to deplete SP from nerve endings, virtually abolished the neuropeptide changes in inflamed tissue. The corticosteroid com- pound, betamethasone, suppresses the changes in SP in vitro, suggesting that inflammatory processes are involved in neuropeptide alterations. Furthermore, it was shown that T lymphocytes are involved in these alterations because nematode infection did not cause SP upregulation in athymic rats. Interleukin-10 (IL- 10) might be the endogenous mediator of this inflammatory response because IL-1 receptor antago- nist and IL-lP neutralizing antibody virtually abolish the increase in SP immunoreactivity (27). Neuropep- tide content of intestinal tissue can be altered by both alterations in peptide synthesis or metabolism. Interestingly, it has been shown that neutral endopeptidase (EC 3.4.24.11), a crucial enzyme in peptide metabolism, is downregulated in inflamed rat intestine (28). Contrary to the observations in the rat,

significant reductions in SP content were reported in the parasitized ferret, guinea pig experimental ileitis, and rabbit immune complex colitis (29-31). VIP content and immunostaining nerves are both significantly increased in rat chemical colitis (32). The conflicting reports on neuropeptide contents might be explained in part by different inflammatory mechanisms underlying animal models of inflammation.

Nitric oxide synthase (NOS) activity in neuromuscular layers is significantly increased in experimental colitis of the rat (33). Muscularis externa expresses NOS constitutively, However, inflammation greatly increases NOS activity, which is mostly calcium independent as opposed to the predominantly calcium- dependent NOS activity in normal tissue (34). Nitric oxide is a crucial inhibitory neurotransmitter in the gastrointestinal tract. Upregulation of nitric oxide synthase may responsible for some of the motility changes observed in intestinal inflammation.

Alterations of adrenergic receptor expression in experimental ileitis have been demonstrated in the guinea pig (35). Upregulation of a- and downregulation of p-adrenergic receptors in neuromuscular tissue in this model was interpreted by the authors as a consequence of autonomic denervation. However, previous studies have shown that catecholamine content is increased in human IBD tissue, suggesting that sympathetic innervation is hyperplastic (8).

FUNCTIONAL ALTERATIONS IN ENTERIC NERVES

Studies in Human 1BD

Changes in enteric nerve function are thought to have a role in both aberrant motility patterns and altered reflex contractility to luminal stimuli observed in IBD. Most studies, however, have focused entirely on alterations in colonic contractility. Farthing and Lennard-Jones have shown that the rectum of patients with ulcerative colitis is more sensitive to distension (36). Changes in perception because of hyperexcitability of primary afferent neurons might explain these data. However, a decreased compliance of the inflamed rectum could generate similar changes in rectal contractility (37,38). Further evidence of altered anorectal contractility in ulcerative colitis was provided when Rao et al. showed that resting rectal motor activity is diminished in active colitis patients, whereas saline infusion provokes abnormally strong contractions. As a consequence, patients with IBD had a significantly lower rectal fluid capacity than controls did (39). Although grossly normal myoelec-

Inflammatorv Bowel Diseases", Vol. 3, No. I , 1997

MUSCLE A N D NERVES IN IBD 41

trical Patterns are present in diseased colon, major motility disturbances such as abnormal colonic response to eating and proximal colonic stasis have been observed (39,40). The absence of deviant slow- wave or electrical spiking activity in these studies suggests that the underlying damage resides in the enteriC nervous system rather than in the smooth- muscle cells. Evidence of altered inhibitory innervation in Crohn’s colitis has been provided in a siudy by Koch et al. (41). Inhibitory junction potentials generated by the enteric nervous system in vitro are decreased along with decreased VIP levels in the myenteric plexus (41). VIP is known to have a crucial role in nonadrenergic inhibitory transmission in the colon (42).

Although IBD, more particularly Crohn’s disease, is considered a systemic disease, very limited data exist on neuronal damage outside the intestine. Gen- eral autonomic nerve dysfunction has been shown in Crohn’s patients in one study by Lindgren et al. (43). Noninvasive tests of autonomic nerve integrity (heart rate variability and tilt test) are abnormal in 48% of patients. This autonomic dysfunction is not related to disease duration and does not coincide with pe- ripheral, somatic neuropathy.

Studies in Animal Models

Functional changes in enteric nerves have been reported in a wide variety of experimental models including nematode parasitism, hapten-induced inflammation, and immunocomplex colitis. Impaired neurotransmitter release from myenteric plexus was first reported in the parasitized jejunum. T. spiralis infection of the rat suppresses both acetylcholine and norepinephrine release from the rat intestine in vitro (44,45). Interestingly, this suppression of norepinephrine and acetylcholine release is still present up to 3 weeks postinfection. By then, the worm expulsion from the gut has been completed, and smooth-muscle function has returned to normal. Betamethasone treatment of Trichinellu-infected rats prevents the suppression of norepinephrine release (45). Cortico- steroid treatment also abolishes the increase in myeloperoxidase activity, reflecting neutrophil infiltration, in the infected intestine (44). The inflammatory reaction of the host animal is therefore most likely responsible for the impaired sympathetic nerve function in rat trichinosis. Experimental colitis in the rat induces a similar suppression of norepinephrine release. Functional alterations in colonic enteric nerves are independent of the noxious agent, because both

T. spiralis infection and trinitrobenzene sulfonic acid enema cause impaired neurotransmitter release. Contrary to previous observations in the intestine, sympathetic nerve dysfunction occurs in noninflamed segments of the rat colon, suggesting a role for the systemic inflammatory response in generating these changes (46). Contractility studies in the intestine of nematode-infected rats have confirmed a decrease in inhibitory neural function (47). In the rabbit, however, experimental terminal ileitis results in enhanced inhibitory junction potentials by a nitric oxide- dependent mechanism (48).

Local and systemic inflammatory effects on enteric nerves may be mediated by cytokines, eicosanoids, free-radical-bearing molecules, and many other putative proinflammatory agents (Table 2). Experimental evidence has been found to support a role for several cytokines as modulators of enteric nerve function. Exogenous IL-lj3, tumour necrosis factor-a (TNF- a) , and 1L-6 alter neurotransmitter release from rat jejunal neuromuscular preparations (49-51). Impair- ment of neuronal function by these cytokines seems to require the synthesis of a protein intermediate. Whether a single protein is induced by IL-1, IL-6, and TNF-a in the myenteric plexus to modulate neurotransmitter release has not been established so far. Messenger RNA expression of proinflammatory cytokines in the neuromuscular layers occurs at an early stage of T. spiralis enteritis in rats. It is therefore most likely that endogenous cytokines are present in situ during intestinal inflammation to modulate enteric nerve function (52). More extensive identifi- cation of inflammatory mediators and cellular interactions involved in the response of intestinal neuromuscular layers to inflammation is required to further elucidate the nature of neuroimmune interactions in the myenteric plexus.

EFFECTS OF INFLAMMATION ON INTESTINAL SMOOTH-MUSCLE CELLS

Human IBD Is Associated with Altered Smooth-Muscle Contractility

Abnormal motility was first reported in patients with active ulcerative colitis in the early 1950s, and it was suggested then that disruptions to normal motility resulted in large part from the augmented immune or inflammatory process in the colon (5354). Evidence to support this postdate came from observations that colonic motility returned toward normal during remission of the inflammation in the colon (54). Indeed, numerous investigators have since re-

Inflammatory Bowel DLemes@, Vol. 3, No. 1, 1997

42 S. M . COLLINS ET AL.

ported that active ulcerative Colitis is accompanied by a reduction in motor activity in the distal colon (55,56). In addition, these more recent studies have suggested that the altered motility is a consequence, in part, of the inflammatory process directly affecting the contractile properties of colonic smooth-muscle cells, This has been confirmed in an in vitro study of human colonic circular muscle in which contraction induced by either bethanechol or potassium chloride was reduced significantly when compared with muscle obtained from patients without IBD (57). These observations suggest that the underlying mechanism is located at the postreceptor level and involves some aspect of the excitation-contraction coupling of the intestinal muscle cell. However, conflicting results have been reported by Koch et al. (58) , who showed that there was a weak association between the duration of symptoms and a decreased frequency of spontaneous summation contractions in colonic circular muscle from patients with ulcerative colitis compared with noninflamed controls. The reason for the dis- crepancies between these studies is not presently ap- parent, but taken together these findings support the notion that inflammation in the bowel alters the function of the gut motor system.

Unfortunately, there are no direct studies of small bowel motility changes in Crohn’s disease patients from which to speculate about the nature of any underlying alteration in intestinal smooth-muscle contractility. However, an in vitro study conducted in this laboratory demonstrated that both circular and longitudinal muscles from the inflamed ileum of Crohn’s disease patients showed increased contractility when compared to muscle from patients without IBD (59). Further, the passive properties, basal tone, and spontaneous activity were unchanged in these tissues as compared to control. Stimulation with either carbamylcholine or histamine increased contractility, intimated that the alteration in response was not exclusively localized at the receptor level. Abso- lute differences in the dose-response curves between control and inflamed tissues suggested that an alteration in ligand-recognition properties of receptors may promote the observed changes. Taken together with findings in intestinal muscle from patients with ulcerative colitis, these observations in muscle from Crohn’s patients serve to emphasize that gastrointestinal muscle responds to inflammation in a region- sensitive and/or specific manner. Interestingly, the regional differences in muscle response observed in human intestinal inflammation hold true for the ani-

mal models of intestinal inflammation described in the next section.

Studies in Animal Models of Intestinal Inflammation

Data regarding the mechanisms underlying altered smooth-muscle function in the inflamed gut are pri- marily derived from studies in animal models of acute intestinal inflammation. However, as is the case for many animal models of human disease, it is necessary to exercise caution when extrapolating from the animal models to the human condition. Thus, the models described herein are not replicas of IBD per se, but are prototypes for assessing the impact of inflammation on intestinal muscle function and, on a broader scale, to explore underlying mechanisms leading to alterations in the gut motor system.

The hypomotility and intestinal muscle hypocontractility observed in patients with ulcerative colitis have also been reported in models of colitis involving the rat (60) and dog (61). In a study from this laboratory (60), a similar decrease in colonic smooth-muscle contractility was observed in colitis induced by chemical injury (acetic acid or trinitrobenzene sulfonic acid, TNB), as well as intrarectal administration of T. spi- rulis (NB: the colon is not the normal enteric habitat of this nematode). I t should be emphasized that these results illustrate that inflammation-induced changes in smooth muscle are nonspecific, that is, effects on contractility do not seem to be influenced by the manner in which colitis is induced. More recent studies in our laboratory (62) using the TNB-colitis model suggest that part of the hypocontractility response in this modcl is caused by the induction of NOS in the mucosal and neuromuscular layers. Studies are currently under way to elucidate the direct effect of excessive nitric oxide synthesis on motility/contractility disturbances in the inflamed colon.

Considerable research attention has focused on primary nematode infections of rodents because these models are reproducible and serve as an important tool for the study of immunophysiological associations in the gut (63). A characteristic finding in the T. spiralis infection model in thc rat is the development of increased tension by longitudinal intestinal muscle from the inflamed jejunum. The tension development increased until day 6 postinfection, where it peaks, and then returned to normal by day 23 (64). In keeping with the postulate that the mechanisms underlying changes in smooth-muscle contractility in the inflamed intestine are, for the most part,

Inflammatory Bowel Diseases? Vol. 3. No. I, 1997


receptor independent is the finding of suppressed sodium pump activity in muscle from the inflamed jejunum of 7'. spiralis-infected rats (65). In that study, ouabain-sensitive &Rb uptake by longitudinal muscle was reduced by >80% compared with controls, and this was accompanied by a corresponding decrease in the activity of p-nitrophenylphosphatase, an enzyme marker of sodium-pump activity. An explanation for this decrease may be that sodium pump is inhibited at the level of gene transcription of the a-1 isoform of the sodium pump protein (66). Considering that the sodium pump is electrogenic, its suppression during inflammation could lead to hyperexcitability of the intestinal muscle because the membrane potential would be reduced to threshold for contraction. How- ever, it is conceivable that the increased contractility of muscle in the inflamed gut is multifactorial, because it has been shown in this model that contractile protein content of muscle is increased (67).

There is also growing evidence that the immune system is also directly involved in the altered muscle contractility observed during T. spiralis infection. Alterations in muscle contraction are absent from nematode-infected rats in which the inflammatory response was suppressed by corticosteroid treatment (68). Muscle contractility changes are also absent from congenitally athymic rats during T. spirafis infection, despite the presence of detectable mucosal inflammation (as determined by myeloperoxidase activity) in the jejunum (69). Moreover, the changes in intestinal muscle contraction are restored through the successful reconstitution of T-lymphocyte numbers in these rats before infection (69). Taken in conjunction with the immunohistochemical evidence that T-cell infiltration of the neuromuscular layer occurs within the first 48 h of infection (70), these findings bolster the hypothesis that increased contraction of intestinal smooth muscle from the inflamed intestine of 7'. spiralis-infected rats is T lymphocyte dependent. We have expanded on these initial observations in the rat, and now have made similar observations in nematode-infected mice (71). It is well known that certain inbred mouse strains develop a greater or more effective immune responsiveness to nematode infections than other strains do (72). Studies in this laboratory using inbred mice indicate that changes in muscle contractility may also be genetically determined, and that there is a positive correlation between the increased contractility of muscle and the ability of the animal to expel the parasite from the gut. This correlation is potentially important since worm expulsion is T lymphocyte dependent, particu-

larly CD4+ T-helper cells [for review see (73)]. Cur- rent studies include ongoing exploration of the role of CD4' T-helper cell subpopulations and their cytokine products as mediators of the observed changes in muscle contraction.

TROPHIC CHANGES IN SMOOTH MUSCLE

IBD Is Associated with Increased Intestinal Smooth-Muscle Growth

Stricture formation is a common complication in Crohn's disease and often requires surgical interven- tion. Crohn's strictures are characterized, in part, by a marked thickening of both the muscularis mucosae and propria and marked muscle cell proliferation and hypertrophy (10,74). Furthermore, there are preliminary data suggesting that intestinal muscle cells from patients with Crohn's disease and ulcerative colitis exhibit altered growth patterns compared with those of controls (75). The mechanism(s) underlying the hyperplasia of smooth muscle in the inflamed intestine remain to be determined, but recent studies demonstrate that IL-10 and platelet-derived growth factor (PDGF-BB) have calcium-dependent mitogenic effects on humans in vitro (76).

Insights Obtained from Animal Models of Intestinal Inflammation

The mechanisms underlying the hyperplasia of smooth muscle in the inflamed intestine remain to be determined, but studies in animal models indicate that both hyperplasia and hypertrophy of the muscularis externa occur during intestinal and colonic inflammation (77). Subsequent studies have indicated that these proliferative changes are determined by cells and mediators potentially involved in the inflammatory response. For example, Blennerhassett et al. (78) have preliminary evidence that intestinal muscle proliferation during T. spiralis infection is T cell dependent, as evidenced by the paucity of muscle proliferation in congenitally athymic rats infected concurrently with normal littermates. Mediators involved in intestinal muscle proliferation include IL- 10 and PDGF-BB. Both appear to have a distinct calcium-dependent mitogenic effect on rodent intestinal smooth-muscle growth in vitro. Furthermore, in a similar system, we have preliminary data suggesting that the constitutive release of nitric oxide modulates intestinal smooth-muscle proliferation. Even less is known about mechanism(s) leading to an increase in intestinal smooth-muscle mass or hypertrophy. Once

Inframmatarv Bowel Diseases@, Val. 3, No. 1, 1997


again, we have preliminary evidence that endogenous nitric oxide regulates, in part, the size of intestinal smooth-muscle cells. Studies are ongoing to explore the relative impact of proinflammatory and antiinflammatory substances on intestinal smooth-muscle proliferation.

SYNTHETIC FUNCTION OF SMOOTH MUSCLE

IBD Is Associated with Alterations in the Synthetic Capacity of Intestinal

Smooth-Muscle Cells

Several observations prompt consideration of an active role by muscle in the inflammatory process in the gut. In IBD, ultrastructural studies have shown changes in muscle suggestive of active protein synthesis (9). such as muscle cells surrounded by collagen. Human intestinal smooth-muscle cells in culture syn- thesize and secrete collagen types I, 111. and V (79). One important stimulus for collagen synthesis by human intestinal smooth-muscle cells is transforming growth factor-p (TGF-p) (80). Collagen synthesis by muscle is not only clinically important in the context ol stricture formation (see earlier section entitled IBD Is Associated with Increased Intestinal Smooth- Muscle Growth), but its synthesis by intestinal smooth muscle exemplifies two important new per- ceptions about these cells. The first is that muscle cells engage in noncontractile activities that contribute to the inflammatory process, and the second is that muscle cells are receptive to immune modulation, as illus- trated by their responsiveness to the cytokine TGF-p. Further exploration of these two concepts in animal models is discussed in the next section.

Insights Obtained from Animal Models of Intestinal Inflammation

Mediator and Cytokine Production by Muscle

A previous study by Kao and Zipser (81) demonstrated that inflammatory mediator production in the gut is not restricted to cells of the mucosa and lamina propria; these authors showed that there was exaggerated production of prostaglandin E2 (PGE2) in thc muscularis externa of the colon, inflamed after the administration of formalin and immune complexes to rabbits (81). In the absence of a discernible inflammatory cell infiltrate in the muscularis externa, the production of PGEz was attributed to intestinal smooth-muscle cells, which have been known for some time to produce prostaglandins. Accordingly,

Bortolami et al. (82) have preliminary evidence that rabbit colonic smooth-muscle cells possess type I IL- 1 receptor, which promotes PGE2 synthesis when bound by IL-1. Although the idea that cytokine production is possible by cells other than those of bone marrow origin is not new, the application of this to intestinal muscle is novel. Indeed, it was two observations in the nematode-infected rat that inspired inves- tigation of this in our laboratory. First. ultrastructural changes in muscle cells of the inflamed rat jejunum are similar to those observed in muscle from patients with IBD. These changes include enhancement of the Golgi apparatus and prominence of the rough endoplasmic reticulum, which is suggestive of active protein synthesis. Second, cytokine gene expression and protein production in the muscularis externa of the inflamed jejunum is elevated after T. spiralis infection in rats (83). Specifically, there is constitutive expression of IL-lp mRNA and protein in the muscu- lark extcrna within 12 h of infection. The increased expression of IL-16 is followed by the expression of other cytokines. including IL-6 and TNF-a. Because the expression of IL-1p was enhanced earliest in the muscularis externa, it was postulated that it might be the stimulus for the induction of other cytokine genes in smooth muscle. Our present results provide clear confirmation that intestinal muscle cells express cytokine genes and secrete the corresponding proteins. First, IL- ID induces its own gene expression in muscle cells, and this is accompanied by protein production (84). Second, IL-16 also induces IL-6 gene expression, and this is also accompanied by protein secre- tion (85). Ongoing studies are evaluating the ability of muscle cells to produce other cytokines. including TNF-a and TGF-P.

Surface Immune Molecule Expression by Muscle

As already mentioned, lymphocytic infiltration of the muscularis externa in the intestine is characteris- tie of T. spiralis infection in rats (70). Concomitant to the increasing T-cell infiltration into the neuromuscular tissue is the appearance of and gradual increase in expression of MHC I1 in this tissue during infection (70). Similar observations have been made in T. spiralis-infected mice; in addition to MHC I1 expression, we have also documented the expression of the adhesion molecule ICAM-1 in the muscularis externa of infected mice (86). The localization of the MHC I1 complexes and ICAM-1 to a particular cell(s) has yet to be determined, but the distribution was suggestive of intestinal smooth-muscle-cell origin. Taken to-


gether, these findings raise the possibility that immune activation and/or modulation could occur in the neuromuscular layers during intestinal inflammation. Accordingly, the increased intestinal permeability as well as changes in vascular permeability in the gut wall during nematode infection would permit access of luminal as well as parasite antigens to the muscularis externa (87). Then, cytokines released by infil- trating lymphocytes and monocytes may induce MHC I1 and ICAM-1 expression by muscle cells. The expression of these immune molecules on muscle would allow for the direct interaction of immune cells with muscle. If this is accompanied by antigen processing and presentation, together with the elabo- ration of costimulatory factors such as IL-16, lymphocyte activation, including cytokine release, may occur.

T-cell Activation by Muscle

Although MHC I1 expressing antigen-presenting cells (APCs) are necessary for the initiation or ampli- fication of antigen-specific immune processes, there is abundant evidence that tissue-resident cells acquire this ability after activation by interferon (1FN)-y (88). These cells, then, may in turn amplify or maintain the local immune response (89) after exposure to inflammatory cytokines and/or foreign or autoanti- gens during pathological conditions. Although it is accepted that surface expression of MHC I1 is a necessary component for antigen presentation to CD4+ T lymphocytes (90), it has previously been shown that presence of accessory signals on APCs are required for antigen presentation and subsequent proliferation of CD4' T lymphocytes (88). Studies using cultured muscle cells isolated from the mouse intestine have shown that MHC I1 and ICAM-1 expression is induced after exposure to a T-lymphocyte- derived cytokine, IFN-.)I (86). These in vitro findings raised the possibility that muscle may contribute to immune activation via MHC 11-linked antigen presentation, a speculation based on other studies dem- onstrating that antigen presentation is possible in a variety of cell types including human myoblasts (91) and vascular smooth muscle (92). We observed that recombinant murine IFN-y induced the expression of MHC I1 and ICAM-1 in murine intestinal smooth- muscle cells, although these cells did not constitutively express MHC 11. Unlike other tissue-resident cells that express MHC I1 and ICAM-1 but fail to activate sensitized syngeneic T cells, we observed that IFN-y activation of smooth muscle facilitated their

TABLE 1. Structural changes of enteric nerves in intestinal inflammation

Structural changes documented in human inflammatory bowel disease Hyperplasia of myenteric ganglion cells Proliferation of sympathetic nerve fibers Axonal necrosis in inflamed and noninflamed tissue Expression of MHC class I1 antigens on enteric glial sheet Alterations in peptidergic innervation (VIP, SP, SOM,

CGRP) Structural and functional changes in nerves in animal models of

intestinal inflammation Alterations in peptidergic innervation (SP, VIP, CGRP) Incrcased nitric oxidc synthase Altered adrenergic receptor expression

MHC, major histocompatibility complex; VIP, vasoactive intestinal peptide; SP, substance P; SOM, somatostatin; CGRP, calcitonin gene-related pcptide.

ability to stimulate T lymphocytes (from mesenteric lymph node) to proliferate in an MHC 11- and, in part, antigen-dependent manner (93). Indeed, although MHC I1 expression by intestinal muscle was absolutely required for significant T-cell proliferation, we observed that the absence of Ovalbumin (OVA) during activation of intestinal smooth-muscle cells with IFN-y for 72 h did not prevent T-cell proliferation. Mitomycin C, a potent inhibitor of intestinal muscle-cell proliferation, inhibited this response and, as demonstrated using anti-ICAM-1 antibody, the proliferative response in the absence of OVA was dependent on ICAM-1. This last finding raises the intriguing possibility that exposure of intestinal muscle to IFN--)I permits T-cell activation via a soluble or surface adherent factor produced by intestinal smooth-muscle cells (i.e., IL-10) regardless of T-cell antigen specificity.

More recent studies of intestinal smooth-muscle cell and T-cell cocultures have demonstrated that the surface molecule expression on intestinal muscle and the subsequent T-cell proliferative response are dependent on the cytokine concentration or profile that the muscle cells are exposed to. The findings from these studies are summarized in Table 1. Important observations from these studies include the following: IL-1p inhibits MHC I1 and ICAM-1 expression in-

TABLE 2. Putative mediators of altered enteric nerve function in inflammation

Cytokines Eicosanoids Growth factors

Interleukin-1 Interleukin-6 Prostaglandin Ez Nerve growth factor Tumor necrosis

factor-a

Inflammatory Bowel Diseases'? Vol. 3, No. 1, 1997

46 S. M . COLLINS ET AL.

TABLE 3. T-cell responses after a 72-h coculture with cultured murine intestinal smooth-muscle cells"

Intestinal muscle cytokine pretreatment (72 h)

T-cell cytokine profile Surface immune molecules T-cell proliferative after 72 h of

induced on muscle response coculture

IFN-y (1.000 Uiml) TFN-y (100 U/ml) IL-4 (10 or 100 ngiml) IL-lp (10 ngiml) TNF-a (5 ngiml) IL-lp + IFN-y (100 or 1,000 Uiml) TNF-a + IFN-y (100 Uiml) TNF-a + IFN-y (1,000 Ulml)

MHC 11, ICAM-1 ICAM-1 ICAM-I

-

MHC I1 -

+++ +i- -

IFN-y, IL-4 IFN-7, IL-4 IFN-y. IL-4 N.D. N.D. N.D. N.D. N.D.

N.D.. not done; -, n o expression or proliferation; +, proliferation; IFN, interferon; MHC, major histocompatibility complex; IL, interleukin; TNF, tumor necrosis factor.

Intestinal smooth muscle cells were exposed to cytokine(s) and ovalbumin for 72 h before washing and the addition of T cells purified from mesenteric lymph nodes. The T cells were removed from ovalbumin-sensitized mice. The T-cell cytokine profile was determined using the enzyme-linked immunosorbent assay.

duced by IFN-y, TNF-a augments MHC I1 expression in the presence of 100 U/ml IF"-y but inhibits class I1 expression in the presence of 1,000 U/ml IFN-y, and T-cell cytokine release is affected by the intestinal smooth-muscle pretreatment (Table 3) . Ongoing studies will further address the effect of cytokines on intestinal smooth-muscle/T-cell interactions so as to more fully define the impact of this cellkell interaction on immune regulation in the gut.

Acknowledgmenk This work was made possible by grants from MRC to S. M. Collins and a fellowship to C. Hogaboam. G. Van Assche was supported by the Research Council of Belgium.

REFERENCES

1. Furness JB, Costa M. The enteric nervous system. Edinburgh: Churchill-Livingstone, 1987.

2. Costa M, Furness JB, Gibbins IL. Chemical coding of enteric neurons. Prog Bruin Res 1986;68:217-40.

3. Watchoff DA, Furness JB, Costa M. Distribution and coexis- tence of peptides in nerve fibers of external muscle of the human gastrointestinal tract. Gastroenterology 1988;95:32-41.

4. Storsteen KA, Kernohan JW, Bargen JA. The myenteric plexus in chronic ulcerative colitis. Surg Gynecol Obstef 1953;97:335-43.

5. Davis DR, Dockerty MB, Mayo CW. The myenteric plexus in regional enteritis: a study of the number of ganglion cells in the ileum in 24 cases. Surg Gynecol Obsret 1953;101:208-16.

6. Siemers PT, Dobbins WO. The Meissner plexus in Crohn's disease of the colon. Surg Gynecol Obstet 1974138:39-42.

7. Kyosola K, Pentila 0, Salaspuro M. Rectal mucosal adrenergic innervation and enterochromaffin cells in ulcerative colitis and irritable colon. Scund J Gustroenterol 1977;12:363-7.

8. Pentilla 0, Kyosola K, Klinge E, Ahonen A, Tallqvist G. Studies of rectal mucosal catecholamines in ulcerative colitis. Ann Clin Res 1975;7:32-6.

9. Dvorak AM, Osage JE, Monahan RA, Dickersin GR. Crohn's disease: transmission electron microscopic studies 111. Target tissues. Proliferation of and injury to smooth muscle and the autonomic nervous system. Hum Puthol 1980;11:620-34.

10. Dvorak AM, Connell AB, Dickersin GK. Crohn's disease: a

scanning electron microscopic study. IJum Parhol 1979:

11. Dvorak AM, Silen WS. Differentiation between Crohn's disease and other inflammatory conditions hy electron microscopy. Ann Surg 1985;201:53-63.

12. Strobach SS, Ross AH, Markin RS, Zetterman RK. Linder J. Neural patterns in inflammatory bowel disease: an imrnunohis- tochemical survey. Mod Parhol 1990;3:488-93.

13. Oehmichen M, Reifferscheid P. Intramural ganglion cell degeneration in inflammatory bowel disease. Dige.rtion 1977;15:482-96.

14. Gehoes K, Rutgeerts P, Ectors N, Mebis J, Penninckx F. Van- trappen G, Desmet VJ. Major histocompatibility class 11 cx- pression on the small intestinal nervous system in Crohn's disease. Gastroenterology 1992;103:439-47.

IS. Payan DG. Neuropeptides and inflammation: the role of substance P. Ann Rev Med 1989;40:341-52.

16. Ahoncn A, Kyosola K, Pentilla 0. Enterochromaffin cells and macrophages in ulcerative colitis and irritable colon. Ann Clin Res 1976;8:1-7.

17. Bishop AE, Polak JM, Bryant MG, Bloom SR, Hamilton S. Abnormalities of vasoactive intestinal polypeptide-containing nerves in Crohn's disease. Gastroenterology 1980:79:853-60.

18. O'Morain C, Bishop AE, McGregor GP, Levi AJ, Bloom SR, Polak JM. Peters J. Vasoactivc intestinal peptide concentra- tions and immunocytochemical studies in rectal biopsies from patients with inflammatory bowel disease. Cur 1984;25:57-61.

19. Duffy LC, Zielezny MA, Riepenhoff-Talty M, Byers TE, Mar- shall J, Wciser MM, Graham S, Ogra PL. Vasoactive intestinal peptide as a laboratory supplement to clinical activity index in inflammatory bowel disease. Dig Dis Sci 1989;10:1528-35.

20. Sjolund K. Schaffalitzky De Muckadel OB, Fahrcnkrug J. Hakanson K, Peterson BG, Sundler F. Peptide-containing nerve fibers in thc gut wall in Crohn's disease. Gut 1983; 24:724-33.

21. Mazumdar S, Das KM. lmmunocytochernical localization of vasoactive intestinal peptide and substance P in the colon from normal subjects and patients with inflammatory bowel disease. A m J Gustroenterol 1992;87:176-81.

22. Kuhota Y, Petras RE, Ottaway CA, Tuhbs KR, Farmer RG, Fiocchi C. Colonic vasoactive intestinal peptide nerves in inflammatory bowel disease. Gastroenterology 1992; 102:1242-51.

23. Koch TR, Carney JA, Go VLW. Distribution and quantification of gut neuropeptides in normal intestine and inflammatory bowel disease. Dig Dis Sci 1987;32:369-76.

24. Mantyh PW, Catton MD, Boehmer CG, Welton ML, Passaro

10: 165-77.

irtj7ammutory Bowel Diseases@', Vol. 3, No. I, 1997

MUSCLE AND NERVES IN IBD 47

EP, Maggio JE, Vigna SR. Receptors for sensory neuropeptides in human inflammatory diseases: implications for the effector role of sensory neurons. Peptides 1989;10:627-45.

25. Watanabe T, Kubota Y , Sawada T, Muto T. Distribution and quantification of somatostatin in inflammatroy disease. Dis Colon Rectum 1992;35:488-94.

26. Swain MG, Agro A, Blennerhassett P, Stanisz A, Collins SM. Increased levels of substance P in the myenteric plexus of Trichinella-infected rats. Gasrroenterology 1992;102:1913-9,

27. Hurst SM, Stepien H, Stanisz A, Blennerhassett P, Sharky K, Bunnett N, Collins SM. The relationship between the proinflammatory peptides interleukin 1/3 and substance P in the inflamed rat intestine [Abstract]. Gastroenterology 19933 CQ:A640.

28. Hwang L, Leichter R, Okamoto A, Payan D, Collins SM, Bunnett NW. Downregulation of neutral endopeptidase (EC3.4.24.11) in the inflamed rat intestine. Am J Physiol 1993;264:G735-43.

29. Miller MJS, Sadowska-Krowicka H, Jeng AY, Chotinaruemol S, Wong M, Clark DA, Ho W, Sharkey KA. Substance P levels in experimental ileitis in guinea pigs: effects of misoprostol. Am J Physiol 1993;265:G321-30.

30. Palmer JM, Greenwood B. Regional content of enteric substance P and vasoactive intestinal peptide during intestinal inflammation in the parasitized ferret. Neuropeptides 1993;

31. Eysselein VE, Reinshagen M, Cominelli F, Sternini C, Davis W, Patel A, Nast CC, Bernstein D, Anderson K, Khan H, Snape W. Calcitonin gene related peptide and substance P decrease in the rabbit colon during colitis. Gastroenferology

32. Kishimoto S, Kobayashi H, Shimizu S, Haruma K, Tamaru T, Kajiyama G, Miyoshi A. Changes of colonic vasoactive intestinal peptide and cholinergic activity in rats with chemical colitis. Dig Dis Sci 1992;37:1729-37.

33. Boughton-Smith NK, Evans SM, Whittle BJR. Characterisa- tion of nitric oxide synthase activity in the rat colonic mucosa and muscle after endotoxin and in a model of colitis. Agents Actions 1994;41:C223-5.

34. Vizzard MA, Erdman SL, de Groat WC. Increased expression of neuronal nitric oxide synthase (NOS) in visceral neurons after nerve injury. J Neurosci 1995;15:4033-45.

35. Martinolle JP, More J, Dubech N, Garcia-Villar R. Inverse regulation of alpha- and beta-adrenoreceptors during TNB- induced inflammation in guinea pig small intestine. Life Sci

36. Farthing MJG, Lennard-Jones JE. Sensibility of the rectum to distension and the anorectal distension reflex in ulcerative colitis. Gut 1978;19:64-9.

37. Denis P, Collin R, Galmiche JP. Elastic properties of the rectal wall in normal adults and in patients with ulcerative colitis. Gastroenterology I97979:45-8.

38. Rao SSC, Read NW, Brown C, Bruce C, Holdsworth CD. Studies on the mechanism of bowel disturbance in ulcerative colitis. Gastroenterology 1987;93:934-40.

39. Snape WJ, Matarazzo SA, Cohen S. Abnormal gastrocolonic response in patients with ulcerative colitis. Gut 1980;21:392-6.

40. Loening-Baucke V, Metclaf AM, Shirazi S. Rectosigmoid motility in patients with quiescent and active ulcerative colitis. Am J Gastroenterol 1989;84:34-9.

41. Koch TR, Carney JA, Go VLW, Szurszewski JH. Altered inhibitory innervation of circular smooth muscle in Crohn’s colitis. Gastroenterology 199098: 1437-44.

42. Makhlouf GM. Vasoactive intestinal peptide: transmitter of inhibitorv motor neurons of the gut. Ann N Y Acad Sci

25:95-103.

1991;101:1211-9.

1993;52:1499-508.

1988;527:369-77. 43. Lindgren S, Lilja B, Rosen I, Sundkvist G. Disturbed auto-

nomic nerve function in patients with Crohn’s disease. Scand

-

J Gastroenterol 1991;26:361-6.

44. C M h s SM, Blennerhassett PA, Blennerhassett MG, Vermil- lion DL. Impaired acetylcholine release from the myenteric Plexus of Trichinella-infected rats. Am J Physiol 1989; 257:G898-903.

45. Swain MG, Blennerhassett PA, Collins SM. Impaired sympathetic nerve function in the inflamed rat intestine. Gastroenter- ology 1991;100:675-82.

46. Jacobson K, McHugh K, Collins SM. Experimental colitis alters myenteric nerve function at inflamed and non-inflamed sites in the rat. Gastroenterology 1995;109:718-22.

47. Crosthwaite AIP, Huizinga JD, Fox JET. Jejunal circular muscle motility is decreased in nematode-infectedrat. Gastroenter- ology 1990;98:59-65.

48. Goldhill JM, Sanders KM, Sjogren R, Shea-Donohue T. Changes in enteric neural regulation of smooth muscle in a rabbit model of small intestinal inflammation. Am J Physiol 1995;268:G823-30.

49. Hunt SM, Collins SM. Mechanism underlying tumour necrosis factor-a suppression of norepinephrine release from rat myenteric plexus. Am J Physiol 199q266:GI 123-9.

50. Main C, Blennerhassett P. Collins SM. Human recombinant interleukin-lp suppresses acetylcholine release from rat myenteric plexus. Gastroenterology 1993;104:1648-54.

51. Riihl A, Hunt S, Collins SM. Synergism between interleukins 10 and 6 on noradrenergic nerves in rat myenteric plexus. Gasfroenferology 1994;107:993-1001.

52. Kahn I, Collins SM. Expression of cytokines in the longitudinal muscle myenteric plexus of the inflamed intestine of rat. Gas- froenferology 1994;107:691-700.

53. Spriggs EA, Code CF, Bargen JA, Curtiss RK, Hightower NC, Jr. Motility of the pelvic colon and rectum of normal persons and patients with ulcerative colitis. Gastroenterol-

54. Kern FJ, Almy TP, Abbot FK, Bogdonoff MD. Motility of the distal colon in nonspecific ulcerative colitis. Gasfroenterol-

55. Rao SSC, Read NW, Brown C, Bruce C, Holdsworth CD. Studies on the mechanism of bowel disturbance in ulcerative colitis. Gastroenterology 1987;93:934-40.

56. Snape WJ, Matarazzo SA, Cohen S. Abnormal gastrocolonic response in patients with ulcerative colitis. Gut 1980;21:392-6.

57. Snape WJ, Williams R, Hyman PE. Defect in colonic muscle contraction in patients with ulcerative colitis. Am J Physiol

58. Koch TR, Carney JA, Go VLW, Szurszewski JH. Spontaneous contractions and some electrophysiological properties of circular muscle from normal sigmoid colon and ulcerative colitis. Gastroenterology 1988;95:77-84.

59. Vermillion DL, Huizinga JD, Riddell RH, Collins SM. Altered small intestinal smooth muscle function in Crohn’s disease. Gastroenterology 1993;104:1692-700.

60. Grossi L, McHugh K, Collins SM. On the specificity of altered muscle function in experimental colitis in rats. Gastroenterol- ogy 1993;104:1049-56.

61. Sethi AK, Sarna SK. Colonic motility in acute colitis in con- scious dogs. Gastroenterology 1991;100:954-63.

62. Hogaboam CM, Jacobson K, Collins SM, Blennerhassett MG. The selective beneficial effects of nitric oxide inhibition in experimental colitis. Am J Physiol 1995;268:G673-84.

63. Russell DA, Castro GA. Physiology of the gastrointestinal tract in the parasitized host. In: Johnson LR, ed. Physiology of the gastrointestinal tract, 2nd ed. New York: Raven Press,

64. Vermillion DL, Collins SM. Increased responsiveness of jejunal longitudinal muscle in Trichinella-infected rats. Am J Phys- iol 1988;254:G124-9.

65. Muller MJ, Huizinga JD, Collins SM. Altered smooth muscle contraction and sodiumpump activity in the inflamed rat intestine. Am J Physiol 1989;257:G570-7.

ogy 1951;19:480-91.

Ogy 1951;19:492-503.

1991;261:G987-91.

198711749-80.

Inflammatory Bowel Diseasesa, Vol. 3, No. 1, 1997


66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

SO.

Khan 1. Collins SM. Altered sodium pump gene expression in the inflamed intestine of the ncmatode-infected rat. Am J Physiol 1993;264:G 11 60-8. Bowers RL, Castro GA, Lai M, Harari Y. Weisbrodt NW. Actin mRNA expression in intcstinal smooth muscle of rats infected with Trichinella ,spiralis. [Abstract]. Gastroenterol- ogy 1990;99:1236. Marrrio L, Blenncrhassett P, Chiverton S, Vcrmillion DL, Langcr J, Collins SM. Altered smooth muscle function in worm-free gut regions of Trichinellu-infected rats. Am J Phys- iol 1990:259:G306-13. Vcrmillion DL, Ernst PB, Collins SM. T-lymphocyte modulation of intestinal muscle function in the TrichineNu-infected rat. Gustroenterology 1991;101:31-~8. Dzwonkowksi P, Stead RH, Blennerhassett MG, Collins SM. Induction of class I1 major histocompatibility complex (MHCII) in entcric smooth muscle [Abstract]. Gustroenterol- ogy 1991;100:A577. Vallance BA, Rlennerhassett PA, Collins SM. T lymphocyte

growth factor beta 1 selectively augments collagen synthesis by human intestinal smooth musclc cells. Gastroenterology

81. Kao HW. Zipser RD. Exaggerated prostaglandin production by colonic smooth muscle in rabbit colitis. Dig Dis Sci 1988; 33697-704.

82. Bortolami M, Ennes HS, McRoberts JA, Snape WJ Jr, Cornin- elli F. 1L-1 enhances the production of prostaglandin EZ in rabbit colonic smooth muscle cells (SMC) through a type 1 IL-1 receptor (Abstract]. Gastroenterology 1993;104:A673.

83. Khan I, Collins SM. Expression of cytokines in the longitudinal muscle myenteric plexus of the inflamcd intesiine of rat. Gas- troenterology 1994:107:691-700.

84. Khan I, Kataeva Cr, Blennerhassett MG, Collins SM. Auto- induction of interlcukin-lp gene expression in cnteric smooth muscle cells [Abstract]. Gastroenterology 1993;104:A534.

85. Khan 1, Blennerhassett MG, Kataeva GV, Collins SM. Interleukin-10 induces the expression of interleukin-6 in rat intestinal ,7.l,\ 0 smooth muscle cells. Gastroenterology 1995;lOS:

1990;99:447-53.

I ILU-0. dependence of persistent intestinal muscle function post infection by 'Tricinellu spiralis in the mousc [Abstract]. Gastroenter- ology 1994:106:A1054. Zhu DH. Bell RG. IL-2 production, IL-2 receptor expression, and 1L-2 responsiveness of spleen and mesenteric lymph node cells from inbred mice infected with Trichinella spiralis. J I m - munol 1989;142:3262-7. Wakelin D. Allergic inflammation as a hypothesis for the expulsion of worms from tissues. Parasitol Today 1993;9:115-6.

schalk N, Gay S, Gay R. Collagen content and types in the intestinal strictures of Crohn's diseasc. Gastroenterology

86. Hogaboam CM, Hewlett B, Stead RH, Snider DP, Collins SM. Cytokine-induced class 11 major histocompatibility antigen and intracellular adhesion molecule expression on murine intestinal smooth muscle cells [Abstract]. Gostroenterology 1994;106:A1965.

87. Blennerhassett MG, Catallo D. Dcxamethasone blocks vascular leakage, but not endothelial activation, in smooth muscle of inflamed rat jejunum [Abstract]. Gasrroenterology 1993; 104:A670.

professional antigen-presenting cells: insights from studies of T cell interactions with keratinocytes. Immiinol Today 1994;

Graham MF' Diegelmann RF, co7 Lindblad WJ. Got- 88, Nickoloff BJ, Turks LA, Immunological functions of

I988;94:2S7- 65. i z . n * n 0 Shah M, Willey A. Graham MP. Inflammatory bowel disease induces changes in the in vitro phenotype of human intestinal muscle cells [Abstract]. Gastroenterology 1993;104:A779. Scheinfeld BA, Collins SM, Blennerhassett MG. Verapaniil inhibits interleukin-lp-mediated hyperplasia o f human intestinal smooth muscle [Abstract]. Gastroenterology 1994;106:A. Blennerhassett MG, Vignjevic P, Vermillion DL. Collins SM. Inflammation causes hyperplasia and hypcrtrophy in smooth muscle of rat small intestine. Am J Physiol 1992;262:G1041-6. Hlcnnerhassett MG, Vignjevic P, Vermillion DL, Ernst PH. Collins SM. Intestinal inflammation induces 'I-lymphocyte- dependent hyperplasia of jejunal smooth muscle (Abstract]. Gastroenterology 1990;98:A328. Graham MF, Drucker DE, Diegelmann RF. Elson CO. Colla- gen synthesis by human intcstinal smooth muscle cells in culture. GartroenteroIogy 1987;92:400-5. Graham MF, Bryson GR. Diegclmann RF. Transforming

89.

90.

91.

92.

93.

IJ . - t l , - f - l .

Maycr L, Eisenhardt D. Salomon P, Baucr W, PIOUS R, Piccin- ini L. Expression of class I1 molecules on intestinal epithelial cells in humans: differences between normal and inflammatory bowel disease. Gasfroenterology 1991;100:3-12. Bcrzofsky JA, Brett SJ. Streicher HZ, Takahashi H. Antigen processing for presentation to Tlymphocytes: function, mechanisms, and implications for the 1 cell rcpertoire. Immunol Rev 1988;106:5-13. Gocbcls N , Michaelis D. Wekerle H, Hohfeld R. Human myoblasts as antigen presenting cells. J ImmiinoI1992;149:661-7. Fabry Z, Sandor M, Gajewski TF, Herlein JA, Waldschmidt MM, Lynch RG, Hart MN. Differential activation of Thl and Th2 CD4 t cells by murine brain microvessel endothclial cclls and smooth muscle/pericytes. .I Immunol 1993;151:38-47. Hogaboam CM, Snider DP, Collins SM. Activation of T lymphocytes by syngeneic murine intestinal smooth muscle cells. Gastroenterology 1996;110:1456-66.

Infiamrnafory Bowei Diseasesm, Vol. 3, No. I, 1997

alterations in enteric nerve and smooth-muscle function in inflammatory bowel diseases

Documents