Role of transforming growth factor-β in inflammatory bowel disease and colitis-associated colon cancer

Abstract

Transforming growth factor-β (TGF-β) plays a central role in a wide array of cellular functions including control of cell growth and differentiation, embryonic development, wound healing, angiogenesis, and immune regulation. In the gastrointestinal tract, TGF-β can either promote or suppress inflammation and cancer formation. This report reviews recent data on the role of TGF-β in the pathogenesis of inflammatory bowel disease and how TGF-β might contribute to the cancer risk associated with chronic inflammation of the gut. (Inflamm Bowel Dis 2010)

Chronic inflammation has been linked to carcinogenesis in a number of organs. Within the gastrointestinal tract, reflux esophagitis, chronic gastritis, chronic pancreatitis, and chronic colitis have been implicated in the development of cancers in the esophagus, stomach, pancreas, and colon, respectively.1–3 In a retrospective case-control study, Rutter et al4 showed that histologic inflammation correlated significantly with the risk of colorectal cancer development in patients with ulcerative colitis (UC). In a cohort study of patients with UC, Gupta et al5 also found that the histologic degree of colitis correlated significantly with colorectal cancer risk. Moreover, in Crohn's disease (CD), when cancer risk is analyzed by bowel disease location, the relative risk of colorectal cancer development in patients who have colitis is 4.5 compared to that for the general population, whereas the risk of colorectal cancer is not increased significantly for patients who have CD confined to the terminal ileum. For patients with CD involving the small bowel, however, the relative risk of developing small bowel adenocarcinoma is increased 33.2-fold above that of the general population.6 These findings clearly illustrate the heightened risk of cancer in inflammatory states, although the mechanisms underlying this relationship are incompletely understood.

One mechanism to explain the contribution of inflammation to carcinogenesis is oxidant stress (the production of reactive oxygen and nitrogen species), which can cause damage to DNA, proteins, and lipids. The high levels of such reactive species in chronic inflammation appear to arise primarily from activated inflammatory cells, which produce oxidants via generators like NADPH oxidase, inducible nitric oxide synthase (iNOS), and myeloperoxidase.7 Moreover, the high demand for ATP production to fuel cell repair and regeneration in the tissues that are damaged by inflammation results in the production of additional reactive species through the electron transport chain in epithelial cells.7 Alterations in a number of genes important in repairing DNA damage from oxidative stress have been implicated in colitis-associated carcinogenesis.8 Ongoing oxidative damage from inflammation may be one explanation for the association of colorectal cancer with the extent, severity, and duration of colitis.

Another mechanism to explain the role of inflammation in cancer is that chemokines and cytokines produced by inflammatory cells (and by certain epithelial cells in response to inflammation) contribute to carcinogenesis. Cytokines are proteins produced mainly by immune cells to facilitate intercellular communication, and cytokines play an integral role in maintaining inflammation in inflammatory bowel disease (IBD).9 The important role of proinflammatory cytokines such as tumor necrosis factor alpha (TNF-α), interleukin-12 (IL-12), and interleukin-23 (IL-23) is evidenced by the success of treatments for IBD that use antibodies directed against these cytokines (e.g., infliximab, adalimumab, certolizomab pegol, ustakinumab). There is evidence to suggest that the cytokines also contribute directly to carcinogenesis. For example, in animal models of colitis-associated cancer, blockade of the IκB kinase, an inhibitor of nuclear factor-κB (NF-κB) translocation, which is downstream of TNF-α, decreases the rate of cancer formation, thereby implicating TNF-α in tumorigenesis.10 Just as important as proinflammatory cytokines are the antiinflammatory cytokines that function to keep the inflammatory state in balance. Two such important antiinflammatory cytokines are interleukin-10 (IL-10) and transforming growth factor-β (TGF-β). The pivotal role of IL-10 in IBD has been illustrated by the development of ileocolitis in animal knockout models of IL-10 and in recent genetic studies of patients with mutations in the IL-10 gene who develop severe Crohn's colitis.11, 12 Moreover, mice deficient in IL-10 not only develop colitis, but also develop colon cancer at a high rate.13 The role of TGF-β in inflammation and colitis-associated cancer is the focus of this review.

TGF-β: UNDERSTANDING THE PATHWAY

In the gastrointestinal tract, communication between the intestinal epithelial cells and the mucosal immune system is essential to maintain a normal balance between proinflammatory and antiinflammatory factors. There are a number of ways that the epithelial cells communicate with immune cells, including antigen presentation to T cells and signaling to immune cells via chemokines or cytokines. TGF-β, a regulatory cytokine secreted both by intestinal epithelial cells and T cells, appears to serve several functions, including suppressing local immune responses to luminal antigens, promoting the production of the mucosal immunoglobulin secretory IgA, and enhancing barrier function.14 TGF-β exerts its effects via activation of cell surface receptors, of which there are multiple types. After binding of the TGF-β ligand to the cell surface receptors, an active complex containing both type I and II receptors is formed.15 Activation of the receptor complex will then lead to intracellular signaling via the phosphorylation of SMADs, which are TGF-β-associated signaling molecules. SMADs comprise three main types: 1) receptor-regulated SMADs (R-SMADs, which include SMADs 1, 2, 3, 5, and 8) that are activated by the TGF-β receptor; 2) comediator SMADs (co-SMADs, which include SMAD4) that are necessary to carry out signal transduction; and 3) inhibitory-SMADs (I-SMADs, which include SMADs 6 and 7) that negatively regulate the activity of the other SMADs. An activated SMAD-complex (comprised of an R-SMAD and a co-SMAD) translocates to the nucleus where it acts as a transcription factor to activate or repress target genes (Fig. 1). Which target genes are activated or repressed varies with the cell type and with conditions affecting the cell at the time of TGF-β stimulation.16 There are a multitude of target genes for TGF-β signaling, and the pathway is intricately regulated at multiple points both pre- and postreceptor binding.17

Figure 1. Classical TGF-β pathway.

ROLE OF TGF-β IN INFLAMMATION

Regulatory Function

A derangement in the normal balance between proinflammatory and antiinflammatory factors in the gut is central in the pathogenesis of IBD. Important proinflammatory cytokines include TNF-α, interleukin-1 (IL-1), and interleukin-6 (IL-6), while important antiinflammatory mediators and regulatory cytokines include IL-10 and TGF-β.9 Evidence for the integral role of TGF-β in the suppression of inflammation is provided by several animal studies. Mice that have knockout of the TGF-β1 gene develop inflammation in multiple organs, including the gut.18, 19 Animals with blockade of the type II T-cell receptor for TGF-β (TGFβRII) also develop colitis and lung inflammation.20 Furthermore, animals with disruption of SMAD3 (an R-SMAD) exhibit abnormal immune function and develop inflammation in numerous organs, including the intestine.21

In humans, studies using normal colonic biopsies also suggest that TGF-β plays an important role in downregulating inflammation. In one such study, biopsy specimens from normal colons that were exposed to anti-TGF-β antibodies exhibited downregulation of T-cell apoptosis and an increase in the production of proinflammatory cytokines including interferon-γ, TNF-α, IL-2, IL-6, IL-8, and IL-17.22 However, it seems paradoxical that colonic biopsies from areas of active inflammation in patients with UC or CD have shown increased levels of TGF-β expression.23 Subsequent studies have elucidated that this seemingly paradoxical increase in the antiinflammatory TGF-β in areas of colitis is due to overexpression of SMAD7, an inhibitory SMAD that negatively regulates TGF-β signaling by competing for SMAD3's binding site on the TGF-β receptor, thereby blocking SMAD3 activation (Fig. 1). The same investigators found that they were able to restore TGF-β signaling by blocking SMAD7.24 Of note, the regulation of SMAD7 is posttranscriptional and appears to be controlled, in part, by activators of NF-κB and STAT-1 through the proinflammatory mediators TNF-α and interferon-γ, respectively.25

T_H17 Effector Cell Differentiation

In contrast to the aforementioned immunosuppressive functions of TGF-β, this regulatory cytokine also plays an integral role in the differentiation of naïve T-cell precursors into proinflammatory T_H17 cells in chronic inflammatory conditions like IBD (Fig. 2). CD4+ effector (or helper) T cells traditionally have been categorized into two subsets, T helper type 1 (T_H1) and T helper type 2 (T_H2) cells, based on their functions and cytokine production profiles. These proinflammatory helper T cells are kept in check by regulatory T cells (T_reg). Recently, investigators have discovered an additional distinctive set of T helper cells called T_H17 cells, which produce IL-17, a potent proinflammatory cytokine.26 T_H17 cells appear to play a central role in maintaining the bowel inflammation characteristic of IBD.27

Figure 2. The differentiation endpoint of T-cell precursors varies depending on the cytokine milieu. In the presence of TGF-β they can differentiate into regulatory T cells, which control and dampen inflammation in addition to promoting tolerance. On the other hand, in the presence of inflammatory cytokines like IL-12 or IL-4, they differentiate into proinflammatory effector T cells. More recently recognized is the necessary presence of TGF-β in addition to IL-6 to generate the T_H17 cells that seem to play a critical role in the chronic inflammation in inflammatory disorders like IBD.

TGF-β plays an important role in mediating the differentiation of naïve T cells into T_reg cells, which express a protein called Foxp3. However, TGF-β in the presence of IL-6, which is produced during active infections or other inflammatory states, does not lead to the differentiation of naïve T cells into T_reg cells, but rather causes the naïve cells to differentiate into T_H17 cells.28 The resulting IL-17 production by the T_H17 cells leads to expression of chemokines, other proinflammatory cytokines (like IL-6 and TNF), and matrix metalloproteases, which meditate tissue infiltration and destruction by inflammatory cells.26 While the combination of TGF-β and IL-6 appears to be necessary for the differentiation of naïve T cells into T_H17 cells, sustained T_H17 cell function is driven by IL-23. TGF-β seems to initially upregulate the IL-23 receptor to enhance T_H17 cell responsiveness to IL-23.29

Strictures and Fibrosis

In addition to its roles in the regulation of inflammation and T_H17 cell differentiation, TGF-β also plays an integral role in the process of healing of the intestinal mucosa and in the development of fibrosis and strictures.30 Fibrosis and stricturing appear to occur due to failure of termination of the wound-healing process. Collagen deposition and contraction can be a normal part of wound healing, but excessive collagen contractile activity leads to stricture formation. Fibroblasts from strictures in CD patients both overexpress TGF-β31 and exhibit an enhanced ability to contract and reorganize collagen, a feature that appears to be promoted by TGF-β.32, 33 In animal models of colitis, blockade of the TGF-β pathway has been effective in preventing stricture formation.34–38 In these studies the TGF-β pathway has been blocked with angiotensin-converting enzyme inhibitors (angiotensin II is an inducer of TGF-β), with Boswellia and Scutellaria (botanicals) extracts, and with TGF-β1 recombinant vaccines. Similar studies have not been done in humans; however, and it is conceivable TGF-β blockade might even exacerbate bowel inflammation and stricture formation, considering TGF-β's important role in immune regulation.

TGF-β IN CARCINOGENESIS OF COLITIS-INDUCED COLORECTAL CANCER

There are a series of physiologic abilities or “hallmarks” that cells must acquire to become malignant. These hallmarks include the ability to proliferate without exogenous stimulation, to ignore growth inhibitory signals, to avoid triggering the mechanisms for programmed cell death (apoptosis), to resist cell senescence, to develop new vascular supplies (angiogenesis), and to invade and metastasize.39 TGF-β clearly plays an important role in carcinogenesis for a number of tumors including colitis associated-colon cancer, and it appears that TGF-β contributes to the development of many, if not all, of the physiologic hallmarks of cancer cells.40 Interestingly, there are data to suggest that TGF-β can be either a tumor promoter or a tumor suppressor, depending on the timing of its activation in the process of carcinogenesis (Figure 3). Early on, TGF-β seems to serve mostly a tumor-suppressor function by inhibiting the proliferation of epithelial cells. During later stages of carcinogenesis, however, TGF-β can promote tumor growth by creating an immunotolerant tumor environment.

Figure 3. Roles of TGF-β as both a tumor suppressor and a tumor promoter. It remains unclear if it plays both roles simultaneously by exerting different effects on different cells, or if at some point during the progression of carcinogenesis it switches functions from being antitumorigenic to protumorigenic.

Antiproliferative Effects of TGF-β

In a number of epithelial cell types, TGF-β inhibits cell proliferation.41 This inhibition appears to result predominately through TGF-β effects on the G1 checkpoint of the cell cycle, which cause growth arrest. In this regard, TGF-β inhibits the cyclin-dependent kinases (cdks) by inducing the expression of p15Ink4b (a cdk inhibitor) and by downregulating c-myc, a transcription factor that promotes cell growth and proliferation.41 In a mouse model of colitis and colon cancer, TGF-β has been found to exert antiproliferative effects by inhibiting the production of IL-6 and its activation of STAT3, a pathway known to promote cancer growth in this model.42 Taken together, these data suggest that when antiproliferative functions of TGF-β are impaired via mutations or dysregulation, cell proliferation and growth may proceed unchecked.

Mutations in TGF-β and its signaling pathway members are associated with a number of cancers, including colitis-associated colon cancer. The most frequent genetic alterations have been found in the TGF-β type II receptor (TGFβRII), which contains two microsatellite sequences that seem to be the preferred targets for these alterations.43 TGFβRII mutations, often associated with microsatellite instability, have been found in 6%–36% of colitis-associated cancers.44, 45 Mutations in the TGF-β type I receptor, which are prominent in ovarian and certain other types cancer,46, 47 have not been studied in colitis-associated cancers. However, a polymorphism involving the TGF-β receptor I (TGFBR1*6A), which leads to a difference in the receptor protein's signaling sequence, has been associated with increased susceptibility to a number of cancers, including colorectal cancer.48 To date, this polymorphism has not been studied specifically in colitis-associated colon cancer.

Mutations in members of the intracellular signaling pathway activated by TGF-β, particularly in SMAD2 and SMAD4, have also been noted in some cancers. Mutations in SMAD4 are known to be among the genetic alterations that underlie the juvenile polyposis syndrome,49 and SMAD4 mutations are often found in sporadic colorectal cancers as well.50 However, these genes have not been well studied in colitis-associated colon cancer. Interestingly, while SMAD3 mutations are found only rarely in sporadic colorectal cancer,47 such mutations may contribute to colitis-associated colon cancer. SMAD3 has two phosphoisoforms, one phosphorylated at the C-terminus (pSMAD3C) by TGF-β1 and the other phosphorylated at the linker region (pSMAD3L) by JNK (a member of the mitogen activated protein kinase, or MAPK, family). The pSMAD3C form is expressed in normal epithelial cells, where it mediates TGF-β's antiproliferative effect, while the pro-proliferative pSMAD3L form has been found to be upregulated in dysplasia and cancer in an IL-10-deficient mouse model of colitis.51 It has been suggested that, during carcinogenesis, a shift from pSmad3C/TGF-β1 signaling to pSmad3L/JNK signaling may endow tumor cells with a growth advantage by allowing a switch from inhibitive to pro-proliferative TGF-β signaling.48, 51

Protumorigenic Effects of TGF-β

In contrast to the antiproliferative effects of TGF-β just discussed, there are data to suggest that TGF-β also can exert pro-proliferative effects. Overexpression of TGF-β has been noted in several cancers, including colorectal, breast, hepatocellular, and malignant melanoma.52–57 In these studies the overexpression of TGF-β appears to increase as the cancer progresses and, when present, TGF-β overexpression portends a worse clinical course. Moreover, in an IL-10-deficient model of colitis and colon cancer (in which 65% of animals develop cancer by 31 weeks of age), plasma TGF-β levels are associated with a higher incidence of dysplasia and cancer, and these levels are significantly higher in affected animals than in wildtype littermates.13 Furthermore, while not yet specifically studied in colitis-associated cancer, in sporadic colon cancer cell lines and animal models, inhibitors of TGF-β types I and II receptors decrease tumor cell migration, invasion, and tumorigenicity.58

Under normal circumstances, it is not clear what causes TGF-β to exert antiproliferative or pro-proliferative effects. However, the mechanisms by which high levels of TGF-β promote tumor growth are better defined. One way in which high levels of TGF-β may promote tumor growth is by increasing tissue invasion and metastasis. One study found that more intense staining for TGF-β in sporadic colon cancers (independent of lymph node status and tumor differentiation) increased the risk of progression to metastatic disease 18-fold.52 While these mechanisms have not been studied in colitis-associated cancers, other models have demonstrated that TGF-β decreases cell adhesiveness by decreasing the expression of cell-linking E-cadherins and by increasing the expression of invasion-associated integrins. In addition, high levels of TGF-β increase the motility and proteolytic activity of cancer cells.40

TGF-β also has potent effects on T cells that play a role in tumor immunosurveillance. TGF-β is well known to play an important role in the maintenance of immune tolerance either by downregulating the differentiation and function of autoreactive T cells or by inducing the production of regulatory T cells.59 Therefore, at increased levels, TGF-β dampens the immune system's response to tumor growth. While this has not been specifically studied in colitis models, TGF-β knockout or inhibition in other animal models restores the effector functions of cytotoxic T cells, which causes tumor rejection.60, 61

CONCLUSIONS

Chronic inflammation is pivotal in colitis-associated colon cancer, and TGF-β seems to play important roles in both the pathogenesis of the chronic inflammation and in carcinogenesis. Animal studies clearly demonstrate that TGF-β plays a crucial regulatory role in inflammation. TGF-β is necessary for the differentiation and maturation of the regulatory T cells that control the immune response and allow tolerance. Alternatively, however, TGF-β in conjunction with IL-6 leads to the production of the T_H17 cells, which appear to play an important role in maintaining chronic inflammatory conditions like IBD.

Given these disparate roles in stimulating and suppressing the immune system, it is not surprising that TGF-β may either inhibit or promote carcinogenesis in colitis-associated colon cancer. The numerous mutations that have been found in the signaling pathway of TGF-β in colitis-associated cancer suggest a tumor-suppressive role for the cytokine. In contrast, high levels of TGF-β promote an immunotolerant environment that might promote uncensored tumor growth. The fact that, under certain conditions, TGF-β might contribute to chronic inflammation and carcinogenesis in IBD make TGF-β an interesting target for treatment. However, given its disparate roles in promoting and suppressing inflammation and carcinogenesis, further study is needed to ascertain the outcomes of such treatment.

Acknowledgements

I thank Dr. Stuart Spechler for thoughtful review and helpful suggestions in the preparation of the article.