Inflammation intersection: gp130 balances gut irritation and stomach cancer
Timothy C. Wang1
& James R. Goldenring2
1 Gastroenterology Division and Department of Medicine, University of Massachusetts Medical Center, Worcester, Massachusetts, USA timothy.wang@umassmed.edu
2 Vanderbilt-Ingram Cancer Center and Department of Surgery, Vanderbilt University School of Medicine, Nashville, Tennessee, USA
The stomach and the intestine respond differently to inflammation. Work on the cytokine receptor gp130 helps explain why cancer develops in the stomach and inflammatory bowel disease hits the intestine (pages 1089−1097).
The seemingly unrelated diseases gastric cancer and inflammatory bowel disease (IBD) share in common an origin in chronic inflammation. Often triggered by the bacterium Helicobacter pylori, inflammation in the stomach appears to nudge relatively undifferentiated epithelial cells into a program of unregulated growth. H. pylori-triggered gastric cancer ranks as the second leading cause of cancer-related mortality in the world. In IBD, disease results from an inability to maintain mucosal integrity in the large and/or small intestine. Patients with IBD, including ulcerative colitis and Crohn disease, rarely die from their disease. But they usually require lifelong medication or surgical intervention.
Given the close interplay between the immune response and epithelial cells, it is not surprising then that perturbations in signaling of immune system mediators play a key role in diseases of the gut. These include the gp130 receptor, which binds the IL6 family (IL6, IL11) of cytokines. While generally considered pro-inflammatory cytokines, the IL6 family also contributes to epithelial repair and healing. IL6 and IL11 bind to homodimers of gp130, a transmembrane receptor -chain that contains two discrete functional modules which signal through STAT1/3 and SHP-2/Erk, respectively.
In this issue, Tebbutt et al use gene "knock-ins" to elegantly demonstrate the "protective" nature of the signaling pathways downstream of gp 130 (Ref. 1). Mutations that block SHP-2/Erk signaling led to the development of gastric adenomas, while mutations that abrogated STAT1/3 signaling increased susceptibility to experimentally-induced colitis. The authors go on to provide evidence that the different behavior exhibited by the stomach and the intestine in response to inflammation can be explained by site-specific alterations in the expression of a family of genes called the trefoil factor (TFF) family.
Three small peptides represent the TFF family, each containing a characteristic three-looped structure known as the trefoil domain. They are expressed predominantly within certain mucous-secreting cells of the gastrointestinal tract. TFF1 and TFF2 are expressed at their highest levels in the antral (distal) region of the stomach. The surface cells produce TFF1, and the deeper cells produce TFF2. TFF3 (also known as intestinal trefoil factor or ITF) is expressed primarily in the mucous-secreting goblet cells of the small and large intestine. The precise function of these molecules has remained somewhat elusive. Trefoils are co-secreted with mucin proteins and are generally upregulated after injury to the gastrointestinal epithelium, suggesting that they play a role in healing or restitution. Restitution is a process of epithelial migration leading to the covering of wounds in the surface epithelium. Indeed, both TFF2 and TFF3 induce directed epithelial-cell migration and wound healing in vitro.
Gene deletion experiments have revealed more specific roles for each of the trefoil factors. Deletion of TFF2 in mice leads to decreased gastric proliferation, increased acid secretion and increased susceptibility to gastric injury2. In contrast, the TFF1 knock-out results in the development of hyperproliferation, adenomas and carcinoma in the antral (distal) portion of the stomach, suggesting that TFF1 is in fact a gastric-specific tumor suppressor gene3. In support of this notion, other studies show that TFF1 inhibits cellular proliferation and promotes cellular differentiation. Furthermore, mutation, deletion or down-regulation of TFF1 is fairly common in human gastric cancer4. Knockout of the TFF3 gene leads to an increased susceptibility to colitis after exposure to oral dextran sodium sulfate (DSS), a substance toxic to colonic epithelial cells leading to mucosal damage and chronic inflammation. Intraluminal infusion of TFF3 (Ref. 5) could rescue the TFF3-defect through coordinated promotion of intestinal cell migration and inhibition of intestinal cell apoptosis.
In the current study, the investigators generated a gp130 'knock-in' mutation (gp130757F) that rendered the receptor incapable of activating the SHP2/Erk pathway. The mice closely resembled the TFF1 knockout mice (hyperplasia and adenomas in the gastric antrum). The group then examined a second gp130 knock-in mutation. This knock-in (gp130STAT) deleted all STAT-binding sites, and phenocopied the intestinal defect of mice deficient for TFF3 (exaggerated colitis in response to DSS). Through a series of transient transfection studies with a gastric cell line, the authors then provide convincing evidence that gp130 directly regulates the TFF genes. TFF1 regulation was found to require Ras/Erk signaling and TFF3 required STAT1/3.
The study falls somewhat short of definitively proving that the Gp130-knockin phenotypes are due to altered trefoil expression, a conclusion that would likely require TFF reconstitution studies. Nevertheless, this study strongly suggests that the trefoils are key downstream targets of the SHP2/Erk and STAT signaling pathways. The data also add to TFF1's credibility as a gastric-specific tumor suppressor gene and TFF3's as a mediator of intestinal homeostasis. With these new findings, we can begin to appreciate that maintaining the gastrointestinal mucosa may require a critical balance between SHP2/Erk and STAT signaling (Fig. 1). According to this model, an imbalance in the ratio of SHP2/Erk versus STAT signaling would lead either to excessive proliferation or to mucosal healing defects. While standard dogma views SHP2/Erk signaling primarily as a proliferative pathway, it now seems that Erk signaling may play more of a role in suppressing proliferation and inducing cellular differentiation—at least in the gastric mucosa. Decreased SHP2/Erk signaling in this model produces massive antral proliferation, likely driven by the unopposed effects of STAT signaling. However, in contrast with the TFF1 knockout, no dysplasia or neoplasia has yet been observed. The lack of neoplasia may be due to the short duration of observation in these mice or to incomplete down-regulation of TFF1.
Figure 1. Model for SHP2/Erk, STAT signaling and trefoil (TFF) gene expression in gastric neoplasia and inflammatory bowel disease.
Binding of IL6 family members to the gp130 receptor normally leads to balanced signaling through SHP2/Erk and STAT1/3 pathways. Shown are the consequences of disrupting this balance. a, Chronic H. pylori infection in susceptible individuals leads to a strong Th1 immune response as well as increases in circulating hormones and growth factors (such as gastrin); these may potentially alter the balance between SHP2/Erk and STAT3. Increased STAT3 signaling (relative to SHP2/Erk signaling) is accompanied by downregulation of TFF1 gene expression and predisposition to neoplastic transformation of the gastric antrum (the distal portion of the stomach). This pattern of events has been produced by the gp130757F knockin mutation, which prevents gp130 mediated SHP2 signaling. b, Decreased STAT3 signaling (relative to SHP2/Erk signaling) may lead to downregulation of TFF3 gene expression in intestinal goblet cells and predisposition to IBD. This pattern has been produced by the gp130STAT knockin. Increased IL6/gp130 signaling can potentially lead to decreased T cell apoptosis, which theoretically could worsen inflammatory bowel disease.
The STAT molecules responsible in the gp130757F model for gastric hyperplasia have not been precisely identified. Nevertheless, accumulating evidence points at STAT3, given STAT3's status as an oncogene6 and the link between gp130 and STAT3 signaling. The present study also supports a function for STAT3 signaling in mucosal restitution and repair in the gastrointestinal tract, particularly in the intestine. But this latter interpretion requires caution, as DSS-induced colitis is an acute injury model, while human IBD represents a chronic, sustained form of inflammation. Indeed, elevated levels of IL6 have been observed in the serum and mucosa of IBD patients, and several studies have suggested that elevated IL6 or hyperactive STAT3 can inhibit apoptosis in T cells in the intestine leading to prolongation of intestinal inflammation. Nevertheless, while STAT3 likely has dual roles with respect to intestinal homeostasis, intestinal deficiency in activated STAT 3 (leading to down-regulation of TFF3) might be expected to predispose to ulceration and inflammatory bowel disease (Fig. 1).
How does this study add to our current model of Helicobacter-mediated gastric cancer (Fig. 1)? H. pylori infection leads to chronic inflammation and a strong Th1-skewed immune response, boosting production of IL-6 as well as other cytokines. This chronic inflammatory process accompanies alteration in the mucosa, including the loss of acid-secreting parietal cells and the development of mucous cell metaplasias. Interestingly, while TFF1 is down-regulated during progression to gastric cancer in several Helicobacter-mouse models (our unpublished observations), there is a concurrent expansion of a TFF2-expressing metaplasia into proximal stomach regions7. Importantly, the TFF2-expressing metaplasia resembles the TFF2-expressing cells of the deep antral glands8. Indeed, the antral gland lineage appears responsible for hyperplasia and dysplasia in the TFF1 knockout mouse9. Like the TFF1 knockout mice, the gp130757F knockout shows a relatively normal mucosa in proximal regions normally lacking high TFF expression. (Figure 2f and h in Ref 1). The hyperplasia in the gp130757F mouse may also evolve from the TFF2-expressing deep antral cell lineage (Fig. 1).
While certain single nucleotide polymorphisms (SNPs) and increased expression of Th1 cytokines are associated with an increased gastric cancer risk10, no genetic polymorphisms linked to gastric cancer have been reported for either gp130 or IL-6. However, the paradigm suggested by the current study does not necessarily require alterations in the gp130/IL-6 machinery. Other factors upregulated in response to H. pylori infection could also modulate SHP2/Erk or STAT1/3 signaling. Indeed, preneoplasia is strongly associated with the induction of a strong Th1 response, with high levels of INF- (Ref. 11) and circulating growth factors such as gastrin7. Thus, in the current model (Fig. 1), factors that tip the balance by activating STAT3 or decreasing SHP2/Erk may hasten progression to gastric neoplasia. Further work will be needed to elucidate these intracellular signaling cascades and determine their role in both regulating trefoil expression and maintaining gastrointestinal homeostasis.
Tebbutt, N.C. et al. Reciprocal regulation of gastrointestinal homeostasis by Shp2 and STAT-mediated trefoil gene activation in gp130 mice. Nature Med.8, 1089–1097 (2002). | Article | PubMed | ISI | ChemPort |
Farrell, J.J. et al. TFF2/SP-deficient mice show decreased gastric proliferation, increased acid secretion, and increased susceptibility to NSAID injury. J. Clin. Invest.109, 193–204 (2002). | Article | PubMed | ISI | ChemPort |
Lefebvre, O. et al. Gastric mucosa abnormalities and tumorigenesis in mice lacking the pS2 trefoil protein. Science274, 259–262. (1996). | Article | PubMed | ISI | ChemPort |
Park, W.S. et al. Somatic mutations of the trefoil factor family 1 gene in gastric cancer. Gastroenterology119, 691–698 (2000). | PubMed | ISI | ChemPort |
Bromberg, J.F. et al. Stat3 as an oncogene. Cell98, 295–303 (1999). | Article | PubMed | ISI | ChemPort |
Wang, T.C. et al. Synergistic interaction between hypergastrinemia and Helicobacter infection in a mouse model of gastric carcinoma. Gastroenterology118, 36–47 (2000). | PubMed | ISI | ChemPort |
Wang, T.C. et al. C57BL/6 mice deficient in secretory phospholipase A2 show increased apoptosis and altered cellular differentiation after Helicobacter felis infection. Gastroenterology114, 675–689 (1998). | PubMed | ISI | ChemPort |
El-Omar, E.M. et al. The role of interleukin-1 polymorphisms in the pathogenesis of gastric cancer. Nature404, 398–402 (2000). | Article | PubMed | ISI | ChemPort |
-chain that contains two discrete functional modules which signal through STAT1/3 and SHP-2/Erk, respectively.
STAT) deleted all STAT-binding sites, and phenocopied the intestinal defect of mice deficient for TFF3 (exaggerated colitis in response to DSS). Through a series of transient transfection studies with a gastric cell line, the authors then provide convincing evidence that gp130 directly regulates the TFF genes. TFF1 regulation was found to require Ras/Erk signaling and TFF3 required STAT1/3.
(Ref. 
