Insulin-like growth factors and inflammatory bowel disease

Insulin-like growth factors and inflammatory bowel disease P. KAY LUND ELLEN M. ZIMMERMANN

It has long been established that growth hormone exerts growth effects on the bowel and may modulate the immune system. Insulin-like growth factor I (IGF-I) mediates many of the actions of growth hormone. Recent evidence suggests that endogenous IGF-I may be up-regulated locally in involved bowel of animals with experimental enterocolitis or of patients with Crohn’s disease. Recombinant growth hormone or IGF-I is now used clinically in treatment of growth or nutritional disorders. This chapter reviews what is known about expression and actions of IGF-I in bowel during inflammation and associated complications. GROWTH Inflammatory

FACTORS

IN INFLAMMATORY

BOWEL

DISEASE

bowel disease

Ulcerative colitis and Crohn’s disease are common, immunologically mediated, inflammatory diseases of the gastrointestinal tract that are characterized by chronic inflammation with periods of clinical remission and reactivation (Fiocchi, 1991; Mayer, 1992; Owen, 1992; Sartor, 1994). Ulcerative colitis is limited to the colon and is characterized by acute and chronic inflammation of the lamina propria with or without epithelial cell destruction, epithelial ulceration and crypt abscesses (Owen, 1992). Crohn’s disease similarly involves mucosal damage, ulceration and inflammation, but differs from ulcerative colitis in that it may affect any region of the gastrointestinal tract and is characterized by transmural inflammation and fibrosis (Owen, 1992). Fibrosis involves disorganized hyperplasia and collagen deposition within the lamina propria, muscularis mucosa, submucosa and the external muscle layers (Graham, 1992; Owen, 1992). Fibrosis, fistulae and stricture formation, with subsequent bowel obstruction are frequent complications of Crohn’s disease that lead to surgical intervention and contribute significantly to symptoms and morbidity of the disease (Graham, 1992; Owen, 1992). Baillih’s Clinical GastroenterologyVol. 10, No. 1, March 1996 ISBN 0-7020-2002-S 095G3528/96/010083 + 14 $12.00/00

83 Copyright 0 1996, by Baillitre Tindall All rights of reproduction in any form reserved

84

P. K. LUND

AND

E. M. ZIMMERMANN

General considerations Growth factors may have multiple and potentially opposing roles in inflammatory bowel disease (IBD). Endogenous or exogenously administered growth factors may have beneficial effects in IBD due to trophic actions on intestinal epithelial or mesenchymal cells. Remission of symptoms in active IBD involves repair or regeneration of damaged mucosa to restore normal digestive or absorptive capabilities (Owen, 1992). Growth factors could potentially limit or protect against mucosal damage during active inflammation or could promote better or more rapid mucosal restitution/regeneration during disease remission. Small bowel resection is commonly used to treat complications of Crohn’s disease (Owen, 1992). By stimulating adaptive growth of remnant bowel after resection, growth factors could promote enhanced or optimal recovery. After mucosal or transmural damage during active inflammation, restoration of normal bowel architecture requires mesenchymal cell proliferation and collagen deposition to promote tissue repair. Growth factors could enhance these processes. Potentially beneficial actions of growth factors in IBD must be balanced against their possible detrimental effects. Growth factors could mediate or exacerbate inflammation by chemotactic or mitogenic effects on macrophages or immune cells or by stimulating the production of proinflammatory cytokines. In excess, growth factors could promote aberrant proliferation of mucosal epithelial cells and enhanced susceptibility to colon carcinoma, a problem in IBD (Owen, 1992). Growth factors could promote excessive mesenchymal cell responses that lead to complications of fibrosis associated with Crohn’s disease (Graham, 1992). Emerging information indicates altered insulin-like growth factor (IGF) production in bowel during inflammation (see below). Growth hormone (GH), which positively regulates IGF-I production, is used increasingly in clinical settings to correct growth abnormalities (Underwood, 1994). Clinical trials of IGF-I in children with growth hormone insensitivity syndrome are now well-established (Bondy et al, 1994; Clemmons and Underwood, 1994). A consideration of the possible roles of endogenous IGF-I in the pathophysiology of IBD and of exogenous GH or IGF-I in symptoms and complications of IBD is relevant to understanding the benefits or risks associated with GIVIGF-I therapy in patients with IBD or with a family history of IBD.

THE

IGF SYSTEM

Components IGFs were initially growth promoting (reviewed in Lund, and IGF-II (Lund,

discovered as factors in human serum that mediate the actions of growth hormone (GH) on the skeleton 1994). There are two insulin-like growth factors, IGF-I 1994). IGF-II is a predominantly fetal IGF. IGF-I is the

INSULIN-LIKE

GROWTH

85

FACTORS

predominant post-natal IGF that mediates many of the growth promoting actions of GH between birth and puberty (Lund, 1994). This chapter will focus primarily on IGF-I for which most information is available about actions in bowel and a potential role in IBD. The IGFs, two IGF receptors and a family of IGF binding proteins (IGFBPs) comprise the IGF system. Many actions of IGF-I and IGF-II are mediated by the same type 1 IGF receptor (Czech, 1989; Lund, 1994). The type 1 receptor binds both IGF-I and IGF-II with high affinity and is a tyrosine kinase (Czech, 1989; Lund, 1994). The type 2 receptor, which is identical to the cation-independent mannose-&phosphate receptor, is specific for IGF-II and may mediate specific growth promoting actions of IGF-II (Czech, 1989; Nishimoto et al, 1991). The IGFBPs, IGFBPl through IGFBP6 have been characterized and cloned (reviewed by Rechler, 1993). Traditionally, IGFBPs were thought to serve primarily as circulating carrier proteins that prolong the plasma halflife of the IGFs, limit insulin-like actions of the IGFs and limit the availability of free biologically active IGF-I or IGF-II for mitogenic actions (Rechler, 1993). More recent evidence suggests that the IGFBPs have other functions (Rechler, 1993). IGFBPs are widely expressed in many tissues including bowel (Lund, 1994; Rechler, 1993). Each IGFBP shows a distinct pattern of expression in different tissues during development and in response to hormone or nutrient status (Rechler, 1993). Secreted or soluble IGFBPs appear to inhibit IGF-mediated cell proliferation or amino acid uptake, probably by limiting the availability of free IGF that can interact with the type 1 receptor (Rechler, 1993). Conversely, cell surface/matrix associated IGFBPs may potentiate IGF action by increasing IGF-I or IGFII concentrations in proximity to their receptors, by preventing proteolytic degradation of the IGFs or by preventing receptor desensitization (Jones et al, 1990; McCusker et al, 1990; Rechler, 1993). Endocrine, paracrine and autocrine general considerations

actions of IGF-I:

Liver is a major site of IGF synthesis and major source of circulating IGFI for endocrine actions (Lund et al, 1986; Lund, 1994). IGF-I is expressed in most if not all tissues (Lund et al, 1986; Lund, 1994). Locally expressed IGF-I may have paracrine or autocrine actions to regulate tissue-specific growth events. To date, there is no clear cut information about distinct endocrine versus paracrine/endocrine actions of IGF-I in bowel or other tissues. Circulating

IGFBPs and endocrine actions of IGF-I

IGF-I exists primarily in the circulation as two major complexes. A 150 kDa complex comprises IGF-I, IGFBP3 and an acid labile subunit (ALS), which does not bind IGF-I directly but binds to IGFBP3-IGF-I complexes (Rechler, 1993). The IGF-I associated with this high molecular mass complex cannot cross capillary membranes and is not readily

86

P. K. LUND

AND

E. M.

ZIMMERMANN

available for endocrine actions. IGF-I also circulates as a heterogenous 35-50 kDa complex of IGF-I bound to the six IGFBPs (Rechler, 1993). These smaller IGF-IGFBP complexes appear able to cross capillary membranes and thus may serve as a readily available source of IGF-I for the endocrine actions (Bar et al, 1990a,b). Meaningful interpretations of altered circulating levels of IGF-I in IBD or other disease states must take into account relative levels of IGF-I, ALS and the different IGFBPs. Truncated or des-IGF-I The amino terminus of IGF-I is essential for high-affinity binding of IGF-I to the major IGFBPs (Clemmons et al, 1990b). IGF-I analogues that are truncated or extended at the amino terminus have a low affinity for IGFBPs (Clemmons et al, 1990a). Such analogues are more biologically potent than intact IGF-I when tested for bioactivity in cells in culture or in vivo (Clemmons et al, 1990b; Lemmey et al, 1991). A naturally occurring truncated IGF-I variant, des-IGF-I, that lacks the first three amino acid residues of IGF-I was first isolated from human brain (Carlsson-Swkirut et al, 1986). Recent studies have characterized an enzymatic activity in rat plasma that cleaves intact IGF-I into des-IGF-I (Yamamoto and Murphy, 1994). One level of regulation of the endocrine actions of IGF-I may lie in enzymatic conversion of circulating IGF-I to des-IGF-I. Multiple IGF-I precursors may dictate endocrine, paracrine autocrine actions

or

In mammals, a single large IGF-I gene gives rise to multiple IGF-I mRNAs that specify at least four IGF-I precursors (Lund, 1994). Within a given species, all four precursors encode the same mature IGF-I sequence but have distinct amino-terminal signal peptides and distinct carboxyl-terminal precursor peptides or E domains (Simmons et al, 1993; Lund, 1994). Tissue and development specific patterns of expression of different IGF precursors and evidence for differential processing has led to the hypothesis that distinct signal peptides or E domains may determine targeting of IGF-I for endocrine or local paracrine and autocrine actions (Simmons et al, 1993; Lund, 1994). Local expression and paracrine/autocrine

actions of ZGF-I in vivo

Studies of local IGF synthesis in tissues are problematical (Lund, 1994). Tissue levels of IGF-I are very low relative to plasma levels (D’Ercole et al, 1984). This and the presence of IGFBPs in tissue create difficulties in measurement and interpretation of data on concentrations of immunoreactive IGF-I in tissue extracts. Localization of IGF-I in tissues by immunohistochemistry is problematical. This may be due to low tissue levels, rapid secretion of IGF-I after synthesis and association of tissue IGF-I with IGFBPs. Immunohistochemical localization of IGF-I in tissues in vivo probably does not distinguish IGF-I that is taken up into tissues

INSULIN-LIKE

GROWTH

87

FACTORS

from the circulation and IGF-I that is synthesized locally. These problems have led to an emphasis on quantifying and localizing IGF-I mRNAs as primary measures of the levels and cellular sites of IGF synthesis in tissues in vivo. Such data provide primarily indirect or correlative indices of local IGF-I action, but probably represent the best available approach at present. IGF-I Mucosal

AND

TROPHIC

ACTIONS

IN THE

NORMAL

BOWEL

growth

The type 1 IGF receptor is expressed throughout the gastrointestinal tract and throughout the bowel wall (Lab&he et al, 1989; Termanini et al, 1990). IGF receptors appear to be enriched in intestinal crypt cells compared with villus cells or other cell types (Laburthe et al, 1989) providing indirect evidence that IGF-I may have a role in regulating crypt cell proliferation. Systemically administered IGF-I increases the mass of small bowel mucosa in rats after proximal small bowel resection or during catabolic weight loss induced by dexamethasone (Lemmey et al, 1991; Read et al, 1991; Vanderhoof et al, 1992) indicating that IGF-I exerts endocrine actions to promote mucosal growth. IGF-I has mitogenic effects on intestinal epithelial cells in vitro indicating that IGF-I can act directly to stimulate proliferation of intestinal epithelium (Park et al, 1990, 1992; Simmons et al, 1995). Local expression of IGF-I mRNA in bowel provides indirect evidence that IGF-I exerts local, paracrine or autocrine actions to regulate bowel growth (Han et al, 1987a; Lund, 1994; Lund et al, 1986). Identification of cells that synthesize IGF-I represents an important step towards understanding paracrine or autocrine actions. In human fetal stomach and small intestine, IGF-I and IGF-II mRNAs are expressed primarily in mesenchymal cells of lamina propria and submucosa (Han et al, 1987a). This finding, and observations that IGFs are expressed primarily in mesenchymal cells in other human fetal tissues, led to the hypothesis that mesenchymal cells may represent a source of IGFs for paracrine effects on other adjacent cell types (Han et al, 1987a). IGF immunoreactivity was localized to epithelial cells in human fetal bowel (Han et al, 1987b), contrasting with the sites of IGF mRNA expression in mesenchymal cells (Han et al, 1987a). These data suggest that cells which contain immunoreactive IGF-I may represent target cells that have undergone receptormediated internalization of IGF-I (Han et al, 1987a,b). Little is known about cellular sites of IGF synthesis in normal bowel of any mammalian species post-natally. Even though IGF-I mRNAs are readily detected by quantitative analyses on mRNA isolated from adult rat or human bowel (Lund et al, 1986; Cohen et al, 1993) attempts to localize cellular sites of IGF-I expression by multiple different in situ hybridization histochemistry methods have failed to yield definitive information. Inconclusive data indicate that cells within the lamina propria of adult rat bowel express IGF-I (Winesett et al, 1995).

88

P. K.

LUND

AND

E. M.

ZIMMERMANN

Despite uncertainties about the cellular origin of locally expressed IGF-I in normal bowel, there is evidence that altered expression of IGF-I in normal bowel correlates with altered growth of normal bowel in some circumstances. Changes in levels of IGF-I mRNA in small intestine in response to fasting and refeeding correlate with changes in bowel mass (Winesett et al, 1995). Such correlation occurs primarily in regions of bowel that are exposed to unabsorbed luminal nutrient, indicating that luminal exposure to or assimilation of ingested nutrient may regulate local IGF-I expression in bowel. Adaptive hyperplasia of small bowel mucosa in response to proximal small bowel resection appears not to be associated with altered or elevated IGF-I expression in bowel relative to levels observed in transected controls (Albiston et al, 1992). IGF-I may, however, play a role in regulating adaptive growth of bowel after resection. Levels of IGFBP3 mRNA are reduced in small bowel after resection and if locally expressed IGFBP3 inhibits or limits trophic actions of IGF-I on bowel, this may serve to amplify the trophic actions of IGF-I (Albiston et al, 1992). In refed animals there are also discordant changes in IGF-I and IGFBP3 such that the ratio of IGF-I : IGFBP3 expression is higher than in fasted animals (Winesett et al, 1995). These observations underscore the need to consider both IGF-I and IGFBP expression in order to define trophic effects of IGF-I in different physiological states. Growth

of subepitheliaVsubmucosa1

layers of bowel

To date, little attention has focused on IGF-I action in bowel-derived cells other than mucosal epithelial cells. Since macrophages and lymphocytes mediate inflammatory processes in IBD (Mayer, 1992) and mesenchymal cells such as fibroblasts or intestinal smooth muscle cells are likely cellular mediators of tissue repair and fibrogenic complications in IBD (Graham, 1992), this area warrants further study. Recent in vitro studies indicate that IGF-I has mitogenic actions on intestinal fibroblasts and smooth muscle cells in culture (PK Lund, unpublished results). Other studies demonstrate that IGF expression is induced in enteric smooth muscle and lamina propria cells during adaptive hyperplasia of enteric smooth muscle and bowel mucosa in response to myenteric denervation (Mohapatra et al, 1995). While preliminary, such observations indicate that IGF-I may have widespread actions on multiple target cells within the bowel. IGF-I

AND

Macrophages,

INTESTINAL immune

INFLAMMATION

AND

INJURY

cells and cytokines

GH and IGF-I regulate immune function (Kelley, 1989). Both GH and IGF-I prime neutrophils for superoxide anion secretion, which is increased in active Crohn’s disease (Fu et al, 1991). IGF-I is expressed in lympho-

INSULIN-LIKE

GROWTH

89

FACTORS

cytes and stimulates proliferation of immature T lymphocytes in an autocrine manner (Gjerset et al, 1990). Increasing evidence suggests interactions between GH/IGF-I and cytokines derived from macrophages or immune cells, GH augments production of tumour necrosis factor a (TNFa) by lipopolysaccharide (LPS) stimulated macrophages (Edwards et al, 1991). TNFa decreases GH secretion by the pituitary (Milenkovic et al, 1989). IGF-I decreases degradation of cartilage proteoglycan induced by interleukin-1 (IL-l) or TNFa (Tyler, 1989). Recent studies provide evidence that pro-inflammatory cytokines induce IGF-I expression. Macrophages that migrate to sites of tissue injury express IGF-I (Rappolee et al, 1988; Rom et al, 1988; Nagaoka et al, 199 1). Receptor-mediated uptake of advanced glycosylation end products (AGES) found in diabetes and ageing induces macrophages to synthesize and secrete pro-IGF-I (Kirstein et al, 1992). IL-l is the mediator of AGEstimulated expression and secretion of IGF-1 precursor (Kirstein et al, 1992). Hyaluronate activation of CD44 induces IGF-I expression in murine macrophages by a TNFa-dependent mechanism (Noble et al, 1993). In contrast, interferon-P or y suppress TNFa induction of IGF-I in macrophages (Lake et al, 1994). Together, these findings suggest that IGF-I may play a role in the immunologically mediated pro-inflammatory processes or in the wound healing that is induced in response to inflammation. Wound healing and mesenchymal

cell responses to inflammation

There are precedents for a role for IGF-I in wound healing and collagen deposition. IGF-I stimulates proliferation and collagen synthesis in cultured skin fibroblasts and aortic smooth muscle cells (Clemmons and Van Wyk, 1985; Clemmons and Shaw, 1986; Mueller et al, 1991). Insulin at concentrations that would activate type 1 IGF receptor stimulates collagen synthesis in intestinal fibroblasts from patients with Crohn’s Disease (Stallmach et al, 1992). IGF-I acts in an autocrine manner to stimulate wound repair in cultured endothelial cells (Taylor and Alexander, 1993). Recent studies in our laboratories provide direct evidence that IGF-I stimulates proliferation of intestinal fibroblasts and smooth muscle cells in culture (PK Lund, unpublished results). In a model of freeze-thaw injury, IGF-I is produced in macrophages during initial stages of injury, but later it is increased in the endothelial cells of newly formed blood vessels and in chondroblasts and chondrocytes (Jennische et al, 1987). By analogy with this model, during acute phase bowel inflammation, cytokines may regulate IGF-I or IGF-II production by macrophages or lymphocytes and the IGFs, in turn, could regulate proliferation of these or other cell types. During the progression from acute to chronic inflammation, cytokines from immune cells may induce IGF production by mesenchymal cells and trigger cell proliferation, phenotypic modification, collagen deposition and fibrosis. To date there is little information about the effects of IGF-I or expression of IGFs during acute intestinal inflammation. Available data support a hypothesis that IGF-I

90

P. K. LUND

AND

E. M.

ZIMMERMANN

plays a role in chronic intestinal inflammation and in particular in the mesenchymal cell responses/fibrosis associated with chronic bowel inflammation and Crohn’s disease (see below). Altered IGF expression and fibrosis

associated with chronic bowel inflammation

Analyses of IGF-I expression in bowel of rats with chronic granulomatous inflammation induced by intramural injection of peptidoglycan polysaccharides (PG-PS) provided the first indication that there are dramatic local increases in IGF-I expression in bowel during chronic inflammation (Zimmermann et al, 1993b). Sites of increased IGF-I rr&NA expression are submucosal mesenchymal cells, that lie adjacent to IL-l expressing cells at the core of granulomas (Zimmermann et al, 1993b). These findings and data in other systems demonstrating that IL-l or TNFa induce IGF-I expression in macrophages (Kirstein et al, 1992; Noble et al, 1993) support a hypothesis that pro-inflammatory cytokines induce IGF-I expression in intestinal mesenchymal cells during chronic inflammation or during the progression from acute to chronic inflammation (Zimmermann et al, 1993b). The PG-PS model represents a useful model that permits analyses during acute remission, reactivation and chronic phases of intestinal inflammation (Sartor et al, 1985). Further studies in this model will determine whether IGF-I expression is induced in macrophages or immune cells during acute inflammation, whether cellular sites of IGF-I expression alter during the progression from acute to chronic phase intestinal inflammation and whether specific inhibitors of the inflammatory response such as cyclosporin or IL-l receptor antagonist can alter the levels or cellular sites of IGF-I expression. The latter studies will be of particular interest as they should provide insights into the cellular or molecular mediators of increased IGF-I expression in mesenchymal cells during chronic inflammation. During chronic PG-PS induced intestinal inflammation, the cells that show increased IGF-I expression have fibroblast/smooth muscle-like morphology and represent sites of increased collagen deposition and fibrosis (Zimmermann et al, 1993a,b). These IGF-I expressing cells also show positive immunostaining for a-smooth muscle actin (Zimmermann et al, 1993a,b) indicating that IGF-I expression is increased in modified fibroblasts or smooth muscle cells in areas of fibrosis during granulomatous inflammation. Other more recent data indicate that increased IGF-I expression is a characteristic feature of chronic bowel inflammation associated with fibrosis. In a different animal model of chronic inflammation induced by intracolonic instillation of ethanol/trinitrobenzene sulphonic acid (TNBS), IGF-I expression is increased in the lamina propria and in smooth muscle layers, particularly in areas of transmural inflammation and disorganization of smooth muscle (Zhee et al, 1995). Observations in TNBS and PG-PS treated animals indicate that IGF-I may mediate mesenchymal cell repair responses in at least two different models of chronic bowel inflammation. Additional studies suggest that

INSULIN-LIKE

GROWTH

FACTORS

91

findings in the animal models are relevant to IBD. In ulcerated or strictured bowel of patients with Crohn’s disease, levels of IGF-I and IGF-II mRNAs obtained at surgical resection are increased relative to levels in uninvolved bowel (Cohen et al, 1993). In patients with ulcerative colitis levels of IGF-I and IGF-II mRNAs are similar in grossly involved and uninvolved bowel (Cohen et al, 1993). Additional numbers of patients must be analyzed to verify these observations. Nonetheless, they provide evidence that increased IGF expression may be a distinguishing feature of Crohn’s disease and ulcerative colitis. If, as in animal models of chronic inflammation, the sites of increased IGF expression correspond to regions of fibrosis, this will add strong support to our hypothesis that IGFs mediate or contribute to fibrosis in Crohn’s disease. In order to definitely test this hypothesis, it will be essential to define the precise cell types that are responsible for fibrosis in Crohn’s disease and to develop in vitro systems to allow direct tests of the actions of IGF-I (or IGF-II) on intestinal mesenchymal cells. The precise cell type(s) mediating the fibrotic response in intestinal inflammation are not conclusively defined. Graham and colleagues have obtained strong evidence to suggest that phenotypically modified intestinal smooth muscle cells underlie stricture and fibrosis of Crohn’s disease (Graham et al, 1987; Graham, 1992). Other groups have implicated subepithelial and/or submucosal fibroblasts as sites of increased collagen deposition in Crohn’s disease (Matthes et al, 1992; Stallmach et al, 1992). A third possibility is that intestinal myofibroblasts underlie the fibrosis of Crohn’s disease. a-Smooth muscle actin expression is characteristic of cells in strictures associated with Crohn’s disease (Graham, 1992; Graham et al, 1994) and could reflect smooth muscle or myofibroblast phenotype. Myofibroblasts play a role in many fibrotic conditions that involve reorganization of connective tissues including wound healing, hepatic fibrosis, vascular injury, pulmonary fibrosis and stromal cell responses to neoplasia (Sappino et al, 1990). During a number of situations of wound repair/connective tissue reorganization, cytokines, growth factors and/or specific extracellular matrix components appear to induce preferential ‘selection’ or ‘clonal expansion’ of cells with a myofibroblast phenotype that are characterized by expression of vimentin and a-smooth muscle actin but lack of expression of desmin and smooth muscle myosin (Sappino et al, 1990; Desmouliere et al, 1992a,b). Other studies indicate that smooth muscle cells or fibroblasts can assume myofibroblast-like phenotype during injury or proliferation (Sappino et al, 1990). Meaningful evaluation of the role of IGF-I in inflammation-induced fibrosis will require definition of the precise mesenchymal cells types that overexpress IGF-I in the bowel of animal models of chronic bowel inflammation and fibrosis such as the PGPS model and in the bowel of patients with Crohn’s disease. In addition, suitable in vitro systems are required to study the direct actions of IGF-I on mesenchymal cell proliferation and fibrosis. Recent progress in isolation and culture of intestinal smooth muscle cells and fibroblasts should facilitate progress in this area (Graham, 1992; Stallmach et al, 1992; Graham et al, 1994).

92 The IGF system and therapy

P. K. LUND

AND

E. M.

ZIMMERMANN

in IBD

Growth failure in children with IBD that results from malabsorption, malnutrition or steroid therapy has been linked to decreases in circulating IGF-I (Kirschner and Sutton, 1986). Nutritional therapy to normalize circulating IGF-I appears to reverse growth failure in children with IBD (Kirschner and Sutton, 1986). With the availability of recombinant GH and IGF-I, replacement therapy in children with IBD to reverse growth failure is an obvious future direction. Tests of recombinant GH therapy in adults after bowel surgery for IBD or during parenteral nutrition are already in progress (Ziegler et al, 1988; Mueller et al, 1991; Ward et al, 1987). Shortterm treatment with recombinant GH improved the symptoms of malabsorption, nutritional insufficiency or negative nitrogen balance in patients after bowel resection for IBD or surgery for other bowel diseases (Ward et al, 1987). In patients receiving parenteral nutrition due to gastrointestinal disease that precluded enteral nutrition, 1-2 weeks of GH therapy significantly improved nitrogen balance and parenteral nutrient utilization and effects were associated with increases in circulating IGF-I (Ward et al, 1987; Ziegler et al, 1988; Mueller et al, 1991). As indicated at the outset of this chapter, GH/IGF-I may promote mucosal growth, restitution or digestive and absorptive capabilities in patients with IBD as well as having general effects on nutrient status. The beneficial effects of GH or IGF-I therapy may, however, need to be balanced against the possibility that GH, or IGF-I produced locally in the bowel in response to GH, may promote fibrogenic complications associated with IBD, especially Crohn’s disease. With this in mind, caution should probably be exercised in the use of GH and IGF-I to correct growth disorders in children with IBD or a family history of IBD. Recombinant IGFBPs are increasingly available and are being used in conjunction with IGFs in other systems such as burn injury (Mueller et al, 1991) to promote an optimal balance between tissue repair and scarring. While speculative at present, inhibitory IGFBPs could represent a potential therapeutic approach to control or limit fibrogenic complications of Crohn’s disease.

SUMMARY

Hallmarks of IGF-I action include synergy with other hormones and growth factors and the ability to stimulate proliferation or differentiated cell function dependent on physiological or pathophysiological context. A complete understanding of IGF action in IBD will require analyses of mechanisms of IGF interaction with other growth factors, hormones and cytokines. GH and IGF-I may be administered to children over prolonged periods to correct growth disorders. The definition of the benefits and problems of GIUIGF-I therapy in IBD needs to distinguish between longterm and short-term effects. Short-term administration of GH and IGF-I to animal models of IBD such as the PG-PS and TNBS models, which share

INSULIN-LIKE

GROWTH

FACTORS

93

features of Crohn’s disease (Sartor, 1992), and a recently developed murine model of ulcerative colitis induced by ingestion of dextran sulphate (Okayasu et al, 1990; Sartor, 1992; Cooper et al, 1993) could address the beneficial or detrimental consequences of short-term GH/IGF-I therapy. Adaptation of the PG-PS, TNBS and dextran sulphate models of inflammation to available transgenic mouse lines that over-express GH and IGF-I (Behringer et al, 1990; Ulshen et al, 1993), especially if over-expression is inducible, could help to define the potential benefits and problems of longterm GH/IGF-I therapy or the effects of GH/IGF-I on immune cell function and cytokine production during intestinal inflammation. It will be useful to study intestinal inflammation and complication in animal models of GH or IGF-I deficiency. In this regard, mice with targeted ablation of the IGF-I gene could be useful (Liu et al, 1993) although neonatal mortality in these models currently poses problems for in vivo studies. Development of mesenchymal cell lines from such animals could, however, provide a useful in vitro system to study the role of IGF-I in altered cell function in response to pro-inflammatory cytokines. Acknowledgements This work was supported by NIH grant DK01022, DK40247 and a grant from the Crohn’s and Colitis Foundation of America (PKL) and NIH grant DK02013 (EMZ). Assistance from the core services of the Center for Gastrointestinal Biology and Disease is gratefully acknowledged (DK34987). The authors thank Dr R.B. Sartor for invaluable input.

REFERENCES Albiston AL, Taylor RG, Herington AC et al (1992) Divergent ileal IGF-I and IGFBP3 gene expression after small bowel resection: a novel mechanism to amplify IGF action. Molecular and Cell Endocrinology 83: Rl7-R20. Bar RS, Boes M, Clemmons DR et al (1990a) Insulin differentially alters transcapillary movement of intravascular IGFBP-1, IGFBP-2, and endothelial cell IGF-binding proteins in the rat heart. Endocrinology 127: 497-499. Bar RS, Boes M, Dake BL et al (1990b) Tissue localization of perfused endothelial cell IGF binding protein is markedly altered by association with IGF-I. Endocrinology 127: 3243-3244. Behringer R, Lewin T, Quaife C et al (1990) Expression of insulin-like growth factor I stimulates normal somatic growth in growth hormone-deficient transgenic mice. Endocrinology 127: 1033-1040. Bondy CA, Underwood LE, Clemmons DR et al (1994) Clinical uses of insulin-like growth factor I. Annals of Internal Medicine 120: 593-601. Carlsson-Skwirut C, Jomvail H, Holmgren A et al (1986) Isolation and characterization of variant IGF-I as we11 as IGF-2 from adult human brain. FEBS Letters 201: ‘46-50. Clemmons DR & Van Wyk JJ (1985) Evidence for a functional role of endogenously produced somatomedin-like peptides in regulation of DNA synthesis in cultured human fibroblasts and paracrine smooth muscle cells. Journal of Clinical Investigation 75: 1914-1918. Clemmons DR & Shaw DS (1986) Purification and biological properties of fibroblast somatomedin. Journal of Biological Chemistry 261: 10 293-10 298. Clemmons DR & Underwood LE (1994) Clinical review 59: uses of human insulin-like growth factors in clinical conditions. Journal of Clinical and Endocrinological Metabolism 79: 4-6. Clemmons DR, Thrailkill KM, Handwerger S & Busby WH (1990a) Three distinct forms of insulinlike growth factor binding proteins are released from decidual cells in culture. Endocrinology 127: 643-645.

94

P. K. LUND

AND

E. M. ZIMMERMANN

Clemmons DR, Cascieri MA, Camacho-Hubner C et al (1990b) Discrete alterations of the insulin-like growth factor I molecule which after its affinity for insulin-like growth factor binding proteins result in changes in bioactivity. Journal ofBio~ogica1 Chemistry 265: 12 210-12 216.- Cohen JA. Zimmermann EM, Sartor RB & Lund PK (1993) IGF-I and IGF-II are overexuressed in inflamed and strictured intestine in Crohn’s Disease. Gkstroenterology 104(4): A683 iabstract). Cooper HS, Murthy SNS, Shah RS & Sedergran DJ (1993) Clinicopathologic study of dextran sulfate sodium experimental murine colitis. Laboratory Investigation 69(2): 238-245. Czech MP (1989) Signal transmission by the insulin-like growth factors. Cell 59: 235-238. D’Ercole A, Stiles A & Underwood L (1984) Tissue concentrations of somatomedin C: further evidence for multiple sites of synthesis, and paracrine or autocrine mechanisms of action. Proceedings of the Nationnl Academy of Sciences of the USA 81: 935-939. Desmouliere A, Rubbia-Brandt L, Grau G & Gabbiani G (1992a) Heparin induces a-smooth muscle actin expression in cultured fibroblasts and in granulation tissue myofibroblasts. Laboratory Investigation 67(6): 716-732. Desmouliere A, Rubbia-Brandt L, Abdiu A et al (1992b) &Smooth muscle actin is expressed in a subpopulation of cultured and cloned fibroblasts and is modulated by ‘y-interferon. Experimental Cell Research 201: 64-73. Edwards CK, Lorence RM, Dunham DM et al (1991) Hypophysectomy inhibits the synthesis of tumor necrosis factor a by rat macrophages: partial restoration by exogenous growth hormone or interferon. Endocrinology 128: 989-996. Fiocchi C (1992) In MacDermott RP & Stenson WF (eds) Injlumnzatory Bowel Disease, pp 137-162. New York: Elsevier Press. Fu YK, Arkins S, Wnag BS & Kelley KW (1991) A novel role of growth hormone and insulin-like growth factor I priming neutrophils for superoxide anion secretion. Journal of Immunology 146: 1602-1608. Gjerset RA, Yeargin J, Volkman SK et al (1990) Insulin-like growth factor-I supports proliferation of autocrine thymic lymphoma cells with a pre-T cell phenotype. Journal of Immunology 145: 3497-3501. Graham MF (1992) Stricture formation: pathophysiologic and therapeutic concepts. In MacDermott RP & Seerson WF (eds) Injhzmmatory Bowel Disease, pp 323-336. New York: Elsevier Press. Graham MF, Drucker DEM, Diegelmann RF & Elson CO (1987) Collagen synthesis by human intestinal smooth muscle cells in culture. Gastroenterology 92: 400405. Graham MF, Gluck UM & Shah BV (1994) B-Tropomyosin and a-actin are phenotypic markers for human intestinal smooth muscle cells in vitro. Molecular and Cellular Diflerentiation 2: 45-60. Han VKM, D’Ercole AL & Lund PK (1987a) Cellular localization of somatomedin (insulin-like growth factor) messenger RNA in the human fetus. Science 236: 193-197. Han VK, Hill DJ, Strain AJ et al (1987b) Identification of somatomedin/insulin-like growth factor immunoreactive cells in the human fetus. Pediatric Research 22: 245-249. Hoyt EC, Van Wyk JJ & Lund PK (1988) Tissue and development specific regulation of a complex family of rat IGF-I mRNAs. Molecular Endocrinology 2: 1077-1086. Jennische E, Skottner A & Hansson HA (1987) Dynamic changes in insulin-like growth factor I immunoreactivity correlate to repair events in rat ear after freeze-thaw injury. Experimental Molecular Pathology 47: 193-201. Jones JI, Gockerman A, Busby WH et al (1990) Extracellular matrix contains insulin-like growth factor binding protein-5: potentiation of the effects of IGF-I. Journal of Cellular Biology 121: 679687. Kelley KW (1989) Growth hormone, lymphocytes and macrophages. Biochemistry and Pharmacology 38: 705-7 13 (commentary). Kirschner BS & Sutton MM (1986) Somatomedin-C levels in growth-impaired children and adolescents with chronic inflammatory bowel disease. Gastroenterology 91: 830-836. Kirstein M, Aston C, Hintz R & Vlassara H (1992) Receptor specific induction of insulin-like growth factor-I (IGF-I) in human monocytes by advanced glycosylation and product-modified proteins. Journal of Clinical Investigation 90: 439-446. Laburthe M, Rouyer-Fessard C & Gammeltoft S (1989) Receptors for insulin-like growth factors I and II in rat gastrointestinal epithelium. American Journal of Physiology 254: G457-G462. Lake FR, Noble PW, Henson PM & Riches DWH (1994) Functional switching of macrophage responses to tumor necrosis factor alpha (TNF) by interferons. Journal of Clinical Investigation 93: 1661-1669.

INSULIN-LIKE

GROWTH

FACTORS

95

Lemmey ABL, Martin AA, Read LC et al (1991) IGF-I and the truncated analogue des-(l-3) IGF-I enhance growth in rat after gut resection. American Journal of Physiology 260: E213-E219. Liu JP, Baker J, Perkins AS et al (1993) Mice carrying null mutations of the genes encoding insulinlike growth factor I (IGF-I) and type 1 IGF receptor (IGFlr). Cell 75: 59-72. Lund PK (1994) Insulin-like growth factors. In Dockray G & Walsh JH (eds) Gut Peptides: Biochemistry and Physiology, pp 587-613. New York: Raven Press. Lund PK, Moats-Staats BM, Hynes MA et al (1986) Somatomedin-C/IGF-I and IGF-II mRNAs in rat fetal and adult tissues. Journal of Biological Chemistry 261: 14 539-14 544. Matthes H, Herbst H, Schuppan D et al (1992) Cellular localization of procollagen gene transcripts in inflammatory bowel diseases. Gastroenterology 102: 431-442. Mayer L (1992) Mucosal immune system in inflammatory bowel disease. In MacDermott RF & Stenson WF (eds) In$unzmafory Bowel Disease, pp 53-76. New York: Elsevier Press. McCusker RH, Camacho-Hubner C, Bayne ML et al (1990) Insulin-like growth factor (IGF) binding to human fibroblast and glioblastoma cells: the modulating effects of cell released IGF binding proteins (IGFBPs). Journal of Cell Physiology 144: 244-253. Milenkovic L, Rettori V, Snyder GD et al (1989) Cachectin alters anterior pituitary hormone release by a direct action in vitro. Proceedings of the National Academy of Sciences of the USA 86: 2418-2422. Mohapatra NK, Ulshen MH, Fuller CR et al (1995) Cell specific expression and regulation of insulinlike growth factor I (IGF-I) and IGF binding proteins (IGFBPs) during adaptive growth of small intestine. The 77th Annual Endocrine Society Meeting, Washington, DC, p. 96 abstract no. OR 37.5. Mueller RV, Spencer EM, Sommer A et al (199 1) The role of IGF-I and IGFBP-3 in wound healing. In Spencer E (ed.) Modern Concepts of Insulin-like Growth Factors, pp 185-192. New York: Elsevier Press. Nagaoka J, Someya A, Iwabuchi K & Yamashita T (1991) Expression of insulin-like growth factorIA and factor IB mRNA in human liver, hepatoma cells, macrophage-like cells, and tibroblasts. FEBS Letters 280: 79-83. Nishimoto I, Murayama Y & Okamoto T (1991) Signal transduction mechanism of IGF-II/Man-6-P receptor. In Spencer E (ed.) Modern Concepts of Insulin-like Growth Factors, pp 517-528. New York: Elsevier Press. Noble PW, Lake FR, Henson PM & Riches DWH (1993) Hyaluronate activation of CD44 induced insulin-like growth factor-l expression by a tumor necrosis factor-a-dependent mechanism in murine macrophages. Journal of Clinical Investigation 91: 2368-2377. Okayasu I, Hatakeyama S, Yamada M et al (1990) A novel method in the induction of reliable experimental acute and chronic ulcerative colitis in mice. Gustroenterology 98: 694-702. Owen DA (1992) Pathology of inflammatory bowel disease. In MacDermott RP & Stenson WF (eds) Infamrnatory Bowel Disease, pp 493-524. New York: Elsevier Press. Park JHY Vanderhoof JA, Blackwood D & MacDonald RG (1990) Characterization of type I and type II insulin-like growth factor receptors in an intestinal epithelial cell line. Endocrinology 126:

2998-3005. Park JHY, McCusker RH, Vanderhoof JA et al (1992) Secretion of insulin-like growth factor II (IGFII) and IGF-binding protein-2 by intestinal epithelial (IEC-6) cells: implications for autocrine growth regulation. Endocrinology 131: 1359-1368. Rappolee DA, Mark D, Banda MJ & Werb Z (1988) Wound macrophages express TGF-IX and other growth factors in vivo: analysis by mRNA phenotyping. Science 241: 708-712. Read LC, Lemmey AB, Howarth GS et al (1991) The gastrointestinal tract is one of the most responsive target tissues for IGF-I and its potential analogs. In Spencer E (ed.) Modern Concepts of Insulin-like Growth Factors, pp 225-231. New York: Elsevier Press. Rechler MM (1993) Insulin-like growth factor binding proteins. Vitamins and Hormones. 47: l114. Rom WM, Basset P, Fells GA et al (1988) Alveolar macrophages release an insulin-like growth factor I-type molecule. Journal of Clinical Investigation 82: 1685-1693. Sappino AP, Schurch W & Gabbiani G (1990) Biology of disease: differentiation repertoire of fibroblastic cells: expression of cytoskeletal proteins as marker of phenotypic modulations. Laboratory Investigation 63(2): 144-158. Sartor BF (1992) Animal models of intestinal inflammation: relevance to inflammatory bowel disease. In MacDermott RP & Stenson WF (eds) Injlummatory Bowel Disease, pp 493-524. New York: Elsevier Press.

96

I’, K. LUND AND E. M. ZIMMERMANN

Sartor BF (1994) Cytokines in intestinal inflammation: pathophysiological and clinical considerations. Gastroenterology 106: 533-539. Sartor RB, Cromartie WJ, Powell DW & Schwab JH (1985) Granulomatous enterocolitis induced by purified bacterial wall fragments. Gustroenterology 89: 587-595. Simmons JG, Van Wyk JJ, Hoyt EC & Lund PK (1993) Multiple transcription sites in the rat IGF-I gene give rise to IGF-I mRNAs that encode different IGF-I precursors in cell-free systems. Growth Factors 9: 205-221. Simmons JG, Hoyt EC, Westwick JK et al (1995) Insulin-like growth factor-I (IGF-I) and epidermal growth factor (EGF) interact to regulate growth and gene expression in IEC-6 intestinal epithelial cells. Molecular Endocrinology 9: 1157-l 165. Stallmach A, Schuppan D, Riese HH et al (1992) Increased collagen type III synthesis by fibroblasts isolated from strictures of patients with Crohn’s disease. Gastroenferology 102: 1920-1929. Taylor WR & Alexander RW (1993) Autocrine control of wound repair by insulin-like growth factor I in cultured endothelial cells. American Journal of Physiology 265: C81tXC805. Termanini B, Nardi RV, Finan TM et al (1990) Insulin-like growth factor I receptors in rabbit gastrointestinal tract. Gastroenterology 98: 703-707. Tyler JA (1989) Insulin-like growth factor I can decrease degradation and promote synthesis of proteoglycan in cartilage exposed to cytokines. Journal of Biochemistry 260: 543-548. Ulshen MH, Dowling RH, Fuller CR et al (1993) Enhanced growth of small bowel in transgenic mice overexpressing bovine growth hormone. Gastroenterology 104: 973-980. Underwood LE (1992) Therapeutic applications of human growth hormone. In Trends in Biotechnology, vol. l(4). Philadelphia College of Pharmacy and Science: Office of Professional Programs. Underwood LE (1994) Growth hormone treatment, acromegaly, and relationship to cancer and leukemia. In Bercu BB & Walker RF (eds) Growth Hormone II: Basic and Clinical Aspects, pp 259-268. New York: Springer Verlag. Vanderhoof JA, Clark R, McCusker RH et al (1992) Effects of IGF-I and des (l-3) IGF-I on mucosal adaptation after jejunoileal resection in rats. Gastroenterology 102: 1949-1956. Wahl SM (1991) Transforming growth factor beta (TGF-/3) in inflammation: a cause and a cure. Journal of Clinical Immunology 12(2): 61-74. Ward HC, Halliday D & Sim JW (1987) Protein and energy metabolism with biosynthetic human growth hormone after gastrointestinal surgery. Annuls ofSurgery 206: 56-61. Winesett DE, Ulshen MH, Hoyt EC et al (1995) Regulation and localization of the insulin-like growth factor system in small bowel during altered nutrient status. American Journal of Physiology G63 l-G640. Yamamoto H & Murphy LJ (1994) Generation of des-(l-3) insulin-like growth factor-I in serum by an acid protease. Endocrinology 135(6): 2432-2439. Zhee JM, Mohapatra N, Lund PK et al (1995) Differential expression and localization of IGF-I and IGF binding protein mRNA in inflamed rat colon. Gustroenferology lOS(4): A948. Ziegler RT, Young LS, Manson JM & Wilmore DW (1988) Metabolic effects of recombinant human growth hormone in patients receiving parenteral nutrition. Annuls ofSurgery 208: 6-16. Zimmermann EM, McNaughton K, Sartor RB & Lund PK (1993a) IGF-I is overexpressed in cells with a smooth muscle phenotype in peptidoglycan-polysaccharide induced chronic enterocolitis in the rat. Gustroenferology 104(4): A808 (abstract). Zimmermann E, Sartor B, McCall RD et al (1993b) IGF-I and IG-ID in a rat model of granulomatous enterocolitis and hepatitis. Gustroenterology 105: 399409.