The causes leading to chronic intestinal inflammation in IBD are unknown. It has been suggested that, in genetically predisposed subjects, it could be the result of a dysregulated immune response to intra-luminal antigens, such as resident luminal bacteria, bacterial products, or dietary antigens: in a chronic defective mucosal barrier, the continuous exposure of lamina propria immune cells to the luminal antigens may trigger, amplify and maintain the local inflammation.

Genetic factors and luminal microbes, in addition to their role in triggering the IBD, are also directly, or indirectly, involved in the development of intestinal fibrosis in IBD.

The luminal bacteria express PAMPs, including lipopolysaccharide (LPS) components of bacterial cell walls, bacterial DNA, and double-stranded RNA, able to bind to pattern recognition receptors expressed by a wide variety of immune and non-immune cells, such as the extracellular TLRs and the intracellular nucleotide oligomerization domain/caspase recruitment domain (NOD/CARD)-like receptors[20].

TLRs are expressed on immune cells as well as non-immune cells, and are involved in innate immunity by recognizing specific patterns of microbial components.

The TLR family, comprised of 10 members, are able to act as primary sensors of microbial products triggering a host defence response against invading pathogens and activating signaling pathways that induce the expression of immune and pro-inflammatory genes[20].

Bacterial components induce the activation of an intracellular signaling cascade via TLRs that involves the preserved intra-cytoplasmic Toll/interleukin-1R (TIR) domain and TIR domain-containing adaptors, such as MyD88. A MyD88/interleukin-1 receptor-associated kinase signaling is thus generated, which regulates innate immunity and can strongly influence the course of the inflammatory response. An increased mucosal expression of TLRs in the intestine of patients with IBD has been reported[21,22]. The TLR activation can be considered a pivotal event in the immunological response in IBD patients[21], suggesting that over-expression of certain TLRs could be one of the underlying mechanisms leading to the abnormal host reaction to commensal bacteria, as seen in IBD[22]. Another important aspect of this defective immune tolerance to commensal bacteria seems to be TLR polymorphisms. Various studies have demonstrated the correlation between TLR4 polymorphisms, at Asp299Gly and Thr399Ile, and the development of IBD[23,24]. Indirect evidence also suggests the involvement of an aberrant innate immune response in intestinal fibrogenesis. Antibodies directed against microbial peptides represent good serological markers that could help in the identification of fibrostenosing CD. Patients with a stronger immune response to microbial peptides are more likely to develop earlier complicated CD, including the earlier occurrence of fibrostenotic CD[25,26].

Nevertheless, the exposure of subepithelial cells to bacterial ligands seems not only to lead to immune-mediated inflammation, but also to direct mesenchymal cell activation.

Studies performed on animal models of IBD have demonstrated not only the presence of TLRs in intestinal fibroblasts and the role of PAMPs in their activation, but have also revealed a close correlation between pathogen-mediated fibroblast activation and the progression of fibrosis. The TLR ligands could directly lead to differentiation of fibroblasts into activated myofibroblasts and TLR expression in non-immune cells can be considered another key event leading to tissue scarring and to the development of fibrosis[27-29].

The role of TLRs in the fibrosis-promoting signals has been frequently observed in the liver in which hepatic stellate cells (HSCs), the main precursors of ECM-producing cells, are sensitized by TLR4 activation; this receptor is functionally expressed by HSCs and directly stimulates myofibroblasts to enhance TGF-β signaling[30]. An increase of LPS during hepatic fibrogenesis induces a TLR4-mediated downregulation of bone morphogenetic protein and activin membrane-bound inhibitor (BAMBI), a pseudoreceptor that decoys TGF-β by lacking an intracellular kinase domain[30,31]. Thereby, the TLR4-MyD88-dependent downregulation of BAMBI induces an unrestricted activation of the TGF-β pathway in HSCs. The LPS-TLR4-MyD88 signaling cascade results are thus closely linked with TGF-β-mediated collagen deposition and promotion of liver fibrosis[29,30].

The role of TLRs has been investigated in the damaged tissue of several organs, including the intestine, even if their involvement in fibrogenic progression has not been completely elucidated.

An increased expression of TLR 2, 3, 4, 6 and 7 has been observed in the intestinal fibroblasts isolated from patients with CD[21,22,32]. These receptors are activated by their respective microbial ligands and promote the differentiation of fibroblasts into collagen-producing myofibroblasts to induce a fibrogenic response[32]. In addition, an increased production of fibronectin and upregulated levels of α-smooth muscle actin (α-SMA) in human intestinal myofibroblasts exposed to TLR ligand have been reported, thus confirming the pro-fibrogenic activity of TLRs and the direct link between bacterial innate immunity and intestinal fibrosis in IBD in man[33]. Furthermore, TLR activation of gut myofibroblasts induces an increased secretion of CXC chemokine ligand 8, a ligand that was initially characterized as a neutrophil chemotactic factor, but seems to exhibit both an angiogenic and angiostatic activity that can lead to the progression of intestinal fibrosis[34].

Another key factor able to activate an innate immune response by intracellular recognition of PAMPs of selected microorganisms is NOD2, an intracellular peptidoglycan receptor for muramyl dipeptide encoded by the NOD2/CARD15 gene present on chromosome 16. NOD2/CARD15 polymorphisms seem to be, in part, involved in the pathogenesis of CD. Considerable evidence confirms, indeed, a relationship between the mutation of NOD2/CARD15 genes and CD, either alone or in combination with mutation of TLR (especially TLR4) or ATG16L1 (an autophagy gene)[35].

The major NOD2/CARD15 polymorphisms associated with CD are Arg702Trp, Gly908Arg and Leu1007fsinsC and these appear to produce defects in the host defence response against invading bacteria that can lead to a persistent intracellular infection with chronic stimulation of inflammatory cells[36]. NOD2 is present in several cell types such as monocytes and macrophages where it seems to decrease the TLR-induced production of pro-inflammatory Th1 cytokines[37]. NOD2 is also expressed in Paneth cells. A correlation between NOD2/CARD15 mutations and the downregulation of mucosal α-defensin has been repeatedly proven[38,39]. Furthermore, NOD2 can induce autophagy by recruiting the critical autophagy protein ATG16L1 on the plasma membrane during bacterial internalization.

In addition, there is a growing body of evidence proving that the main NOD2/CARD15 variants are closely related to ileal disease, a stenosing phenotype and to a greater need for abdominal surgery in CD patients[26]. Cells carrying NOD2 variants show an enhanced pro-inflammatory response to various intestinal microbes and lead to an increased TGF-β production and collagen deposition by T cells[40]. All these findings provide evidence that may encourage the clinical use of NOD2/CARD15 genotyping, both as a marker of CD and as a prognostic factor of the need for early surgery due to stricturing and fibrostenosing disease[26].

IBD, however, is a polygenetic disease. The development of genome-wide association scanning technologies has led to the discovery of more than 100 confirmed IBD loci[41-43]. Some, such as the Th 17 pathway genes (IL23R, IL12B, Janus kinases 2 (JAK2), signal transducers and activators of the transcription 3 (STAT3), are shared between CD and UC, others are phenotype-specific (autophagy genes such as ATG16L1, IRGM and NOD2 for CD; epithelial barrier genes HNF4a, E-Cadherin, LAMB1 and IL-10 for UC). Variants of some of these genes would also be excellent candidates involved in the development of fibrosis in IBD[44-46].

Most studies have focused on the involvement of non-immune cell types in the progression of fibrosis in IBD, including mesenchymal cells (smooth muscle cells, fibroblasts, myofibroblasts), epithelial cells, nerve cells and platelets[47-49]. In addition, mucosal microvascular cells play a central role in the microcirculation regarding the onset and maintenance of IBD[47]. Much evidence indicates that neoangiogenesis, the growth of blood vessels from those already present, plays a crucial role in IBD[48]. The microcirculation and endothelial cells perform a crucial task in intestinal immune homeostasis: endothelial cells regulate the type and the number of leukocytes that migrate from the blood flow to the interstitial space and specialized vascular cells, such as those found in the endothelium of venules, control selectively the influx of a specific subset of T-cells[49].

Despite the evidence regarding the role of endothelial cells and neoangiogenesis in the inflammatory process, very limited data are available on their role in intestinal fibrosis.

Fibroblasts are a heterogeneous population of cells present in the interstitium of all normal tissues and organs where they play a pivotal role in maintaining structural integrity, regulating matrix homeostasis and taking part in healing and regenerative processes and/or in pathogenesis of scarring.

An acute injury or acute inflammation can launch activation of fibroblasts driving them towards a fibrogenic phenotype. This activated fibroblast can be engaged either in normal wound healing or fibrosis. Fibroblast activation is characterized by a post-transcriptional or post-translational up-regulation of ECM secretion.

Fibroblasts co-express vimentin, fibroblast-specific protein 1 (FSP-1, S100A4), N-cadherin, and prolyl 4-hydroxylase (Table 2).

The increase in the resident fibroblast population is a pivotal mechanism for the development of intestinal fibrosis. Several growth factors found in the inflamed gut, such as IGF-I, basic fibroblast growth factor (bFGF), EGF, CTGF, PDGF, and pro-inflammatory cytokines, such as interleukins (IL-1β, IL-6) and TNF-α, increase their proliferation rate[27].

TGF-β1, one of the most potent profibrogenic factors, shows a bifunctional role in fibroblast proliferation with either growth stimulation or growth inhibition and blocking of cell differentiation depending upon the concentration used or the organ from which the cells are derived[51,52]. TGF-β1 can also act in an indirect mode by up-regulation of the PDGF receptor, increasing synthesis of CTGF and promoting expression of IGF-I, all factors that directly affect proliferation[15,53,54].

Another mechanism that further increases the number of activated fibroblasts is the migration from non-affected tissue areas into and through the surrounding ECM, under the effect of a chemotactic gradient (chemotaxis) or with active random movement (chemokinesis) to the wound area. In the intestine, the cell migration is generally an essential factor for physiological wound repair, and is a fundamental process during the progression of pathological conditions such as fibrosis in IBD. Migration of fibroblasts can be induced by autocrine stimuli, like fibronectin, and paracrine factors, such as PDGF-AB, IGF-I, EGF and TGF-β1, all of which appear to be fibronectin-dependent[51,55,56]. Mediators of active inflammation, such as TNF-α, IFN-γ or PGE2, can inhibit fibroblast migration further indicating distinct mechanisms of inflammation and fibrogenesis[57,58].

Myofibroblasts represent a highly contractile cell type that is thought to be critical for tissue repair and the pathogenesis of fibrogenetic diseases[59,60]. These cells exhibit a “hybrid” phenotype between fibroblasts and smooth muscle cells and, when activated, synthesize high levels of ECM, particularly collagen, glycosaminoglycanes, tenascin and fibronectin[61]. Besides their normal activities in growth and differentiation of tissues, the myofibroblasts play a central role in wound healing. During normal wound repair of a tissue injury, a closely regulated sequence is induced, including activation and proliferation of myofibroblasts that are present in the wound bed where they play essential roles in wound contraction and connective tissue restoration, e.g., through the production of ECM and basement membrane molecules[1,15,17].

Myofibroblasts can also play a role in the up- or down-regulation of the inflammatory response by the secretion both of chemokines and cytokines[1,17]. When these processes are not controlled, deranged, or repeated, as occurs in several fibroproliferative diseases, the normal resolution stages are abrogated and the proliferation of myofibroblasts continues, inducing excessive accumulation of the ECM and leading to alterations in the tissue architecture and ultimately to organ failure. Therefore, fibrotic disease is a major pathological end point of activated and proliferating myofibroblasts in most, if not all, tissues[1,17].

Myofibroblastic cells contain smooth muscle cytoskeletal markers (in particular α-SMA) together with three filaments (vimentin, desmin or myosin), with variable expression depending on tissue, species and environmental factors[61,62]. Based on immunohistochemical staining of these filaments in a given tissue, a classification system has been proposed[61,62]. Myofibroblasts that express only vimentin (V-type myofibroblasts), those that express vimentin and desmin (VD-type), those that express vimentin, α-SMA and desmin (VAD-type), those that express vimentin and α-SMA (VA-type), and those that express vimentin and myosin (VM-type)[61,62].

Two types of myofibroblasts are present in the intestinal mucosa in physiological conditions, the interstitial cells of Cajal (ICC) and the intestinal sub-epithelial myofibroblasts (SEMFs)[61,62].

ICC are located in the submucosa and muscularis propria in association with the smooth muscle layer of the gut, while SEMFs are mainly located at the base of the intestinal crypts in the lamina propria. ICC are pacemaker cells which regulate gastrointestinal smooth muscle motility, facilitate the propagation of electrical events and modulate neurotransmission[63]. ICC express vimentin and the c-Kit receptor (Table 2).

SEMFs, also called pericryptal fibroblasts, are a syncytium of α-SMA-positive mesenchymal cells, which reside subjacent to the basement membrane of the small and large intestines[61,62]. SEMFs form a three-dimensional network and are in connection with each other by gap and adherent junctions, but also maintain connections with epithelial cells through fenestrations in the basement membrane; they also interact with intestinal macrophages. SEMFs express α-SMA and vimentin suggesting that they are members of the VA class of myofibroblasts. They may also express smooth muscle myosin (and thus may be referred to as VAM-type myofibroblasts), although the expression of myosin is less than that observed in the corresponding smooth muscle cells in the same tissue[61,62].

It is unknown whether SEMFs and ICC differentiate from a common precursor. Instead, it has been clearly demonstrated that mediators which act on myofibroblasts, promoting their proliferation and ECM production, are numerous and include: PDGF, EGF, IGF-1 and 2, CTGF, IL-1, IL-13, stem cell factor (SCF), endothelins (ET-1, -2, -3), ANG II, TGF-α, TGF-β, bFGF and PPARγ; these factors are all relevant for the transdifferentiation of fibroblasts to myofibroblasts[15].

In physiological conditions when there is a need for replacement of connective tissue following injury, the production of collagen is devolved to cells of mesodermal origin. Other cells do not synthesize collagen or may produce only minute quantities; however, under abnormal stimuli collagen expression may become considerably enhanced in these cells[61,62]. Airway smooth muscle cells (SMCs) appear to play an important role in controlling and perpetuating airway inflammation and fibrosis in chronic airway diseases[64].

SMCs are one of the three interrelated cell phenotypes into which intestinal mesenchymal cells can differentiate (the other two being fibroblasts and myofibroblasts) and are mainly identified by α-SMA, desmin and collagen type I[27] (Table 2).

Several studies have shown that SMCs are an important cellular component of IBD. In UC, their behavior leads to a considerable thickening of the muscularis mucosa and, in CD, to a remarkable thickening of the bowel wall, with subsequent stricture formation and obstruction. These cells actively contribute to the development of fibrosis in IBD by inducing production of collagen and MMPs in response to several inflammatory mediators, such as TGF-β and IL-1β. SMCs are also able to release significant amounts of IL-6, contributing to the inflammatory process[65].

Stellate cells were initially described in the liver as mesenchymal cell precursors important in retinoic acid metabolism, where they represent the site of fat or vitamin A storage. Thereafter, the important contribution of these cells to hepatic and pancreatic fibrosis has been shown[66,67].

These cells are identified through the expression of vitamin A, glial fibrillary acidic protein (GFAP) and α-SMA (Table 2).

In the liver, it has been clearly shown that HSCs are the main contributors to fibrogenesis. Acute or chronic inflammation can promote activation of HSCs through transdifferentiation from quiescent vitamin A-rich cells into myofibroblast-like cells with strong proliferative, fibrogenic and contractile ability. Moreover, the presence of stellate cells has been detected also in other organs, such as the pancreas, gut, lung, uterus, kidney and deferent duct, even if their functions remain to be fully elucidated[51,66,67].

Interestingly, there is very limited information regarding intestinal stellate cells, although recently these have been isolated and cultured from the human intestinal mucosa[51]. In IBD, these cells show a higher proliferation rate, faster differentiation into myofibroblasts, and an earlier and higher collagen production than those from the normal non-IBD mucosa[51].

Pericytes are cells derived from non-differentiated mesenchymal cells. They surround the endothelial cells of capillaries and small blood vessels.

Pericytes stain positive for α-SMA, desmin, melanoma chondroitin sulphate proteoglycan, platelet-derived growth factor receptor (PDGFR)-β, and regulator of G protein signaling-5. Pericytes also express costimulatory molecules such as CD80, CD86, CD13 and CD90, suggesting that they may be replenished by circulating fibrocytes. Furthermore, pericytes can express ANG I and II, endothelin-1 and collagen type I (Table 2). The pattern of these pericyte markers differs according to the type of vessel, organ and also pathological condition.

Pericytes control endothelial cell differentiation, endothelial signaling, angiogenesis and ECM degradation. These multiple functions can be exerted due to their location between the interstitium and endothelium[68]. Moreover, on account of their intermediate phenotype between vascular-SMC and fibroblasts, pericytes represent a useful reserve of fibroblasts during wound healing and are, therefore, considered an excellent contributor to inflammation-associated fibrosis on account of their ability to differentiate into collagen-producing fibroblast-like cells[51]. Pericytes increase the deposition of ECM proteins in proximity to the blood vessels during the initial phase of fibrotic processes. In addition, the involvement of pericytes in intestinal fibrogenesis has been recently highlighted, even if their contribution remains to be defined[51].

The main fibrogenic cells (fibroblasts, myofibroblasts) may also derive from non-mesenchymal cells, including epithelial and endothelial cells, via transformation.

Epithelial-to-mesenchymal transition (EMT) or endothelial-to-mesenchymal transition (EndoMT) is a key process in tissue development, carcinogenesis and organ fibrosis, and is characterized by dramatic changes in cell phenotype and function. Both are a consequence of disruption of the local basement membrane, loss of epithelial cell adhesion, reprogramming of the signaling machinery, de novo synthesis of α-SMA, rearrangement of cytoskeletal proteins and transmigration of the epithelial cells through the basement membrane into the interstitial space[69]. Epithelial or endothelial cells assume a spindle-shape morphology, lose classical cell markers and gain typical fibroblast or myofibroblast markers, such as FSP-1, α-SMA or vimentin, and show the capacity to produce interstitial collagens and fibronectin (Table 2). All these changes are due to the high plasticity of epithelial and endothelial cells. Therefore, these cell types can be considered a multipotent progenitor tissue, which can display alternative developmental pathways following injury.

Recent data on the contribution of EMT and EndoMT in intestinal fibrosis are now available. It has been shown in animal models and in human primary cells that EMT and EndoMT can contribute to intestinal fibrogenesis[70,71].

Stem cells are a non-differentiated cell type that do not increase in number but generate a range of differentiated progeny that may continue to be divided via a process of clonal growth. They are also defined as non-specialized cells that reside in a particular tissue and renew themselves thus becoming capable of producing all the specialized cell types of the tissue[72].

Reserves of adult stem cells include the bone marrow, blood stream, cornea and retina, dental pulp, liver, skin, gastrointestinal tract and pancreas. Bone marrow contains hematopoietic and mesenchymal stem cells, which are able to migrate to the majority of organs and differentiate into various cell types. These cells may represent a source of fibroblasts during inflammatory and fibrotic disease; indeed, stem cells can migrate from the bone marrow to sites of injury, during pathological conditions, and differentiate to mesenchymal cells able to express tissue connective proteins[73].

While HSCs give rise to 3 classes of blood cells (leukocytes, erythrocytes and thrombocytes), and are characterized by several specific markers, including CD45, CD34 and CD14, the mesenchymal stem cells (MSCs) are multipotent and can differentiate into several cell types such as osteoblasts, chondrocytes, myocytes and myofibroblasts. MSCs express CD105, CD73, CD44, CD71 and CD90. Both HSCs and MSCs express vimentin, chemokines and chemokine receptors, adhesion molecules and integrins[74].

A class of bone marrow-derived cells that become progenitors for mesenchymal cells is represented by fibrocytes that circulate in the peripheral blood. These cells appear to be involved in intestinal repair and fibrosis in IBD[50,51,73,74]. Fibrocytes express hematopoietic markers (CD45, CD34, CD11, CD13, CD14, CD80, CD86), as well as collagen I and α-SMA[27,51,74-76] (Table 2).

The ability of fibrocytes to differentiate into fibroblasts and contractile myofibroblasts has been highlighted. This leads to the production of ECM components and ECM-modifying enzymes by differentiated fibrocytes that migrate to the affected tissues during both normal tissue repair and inflammatory fibrotic processes. Under physiological conditions, fibrocytes, after their maturation in the bloodstream, may contribute to the local population of macrophages and dendritic cells, whereas during an inflammatory process, numerous fibrocytes differentiate into mesenchymal cells synthesizing ECM. This behavior demonstrates the close relationship between fibrocyte proliferation, migration and differentiation, as well as tissue fibrogenesis[27].

Fibrocyte functions that lead to tissue fibrosis are modulated by IL-1, TGF-β and serum amyloid P (SAP), a serum protein that inhibits the maturation process in the circulation but promotes differentiation of fibrocytes within the tissue[75]. Furthermore, fibrocytes themselves produce growth factors, such as CTGF and TGF-β, inflammatory cytokines and chemokines, that in turn promote the proliferation of resident fibroblasts and their differentiation into myofibroblasts[75].

The contribution of fibrocytes to tissue damage and progression of fibrosis has been shown in several pathological conditions, including asthma, nephrogenic fibrosis, systemic sclerosis, atherosclerosis, chronic pancreatitis, chronic cystitis and tumor-associated stromal reactions[77].

Another aspect of intestinal fibrosis under investigation is the presence and the involvement of local stem cells in the fibrotic process. Undifferentiated intestinal stem cells (ISCs) give rise to daughter or progenitor cells, which can subsequently differentiate into mature cell types such as columnar cells, goblet cells, neuroendocrine cells and Paneth cells[72,74,76].

Differentiation of ISCs is induced by various transduction pathways and appears to be involved in the pathogenesis of IBD[78]. It has been highlighted that a potential mechanism linked to the development and progression of Crohn’s disease is the differentiation of ISCs into Paneth cells[78]. These cells produce different broad spectrum antimicrobial peptides, principally the α-defensins HD-5 and HD-6. In small intestinal Crohn’s disease, both these Paneth cell products are specifically reduced. Mechanisms for defective antimicrobial Paneth cell function are complex due to the impaired functions of various pathways including the NOD2, Wnt pathway transcription factor TCF7L2 (also known as TCF4), autophagy factor ATG16L1, endosomal stress protein XBP1, toll-like receptor TLR9, calcium-mediated potassium channel KCNN4, as well as to the inactivation of HD-5.

A better understanding of all these mechanisms may lead to the development of new therapeutic approaches for the prevention and/or treatment of intestinal fibrosis progression.

Studies on transgenic mice over-expressing TGF-β have revealed the development of fibrosis in several organs, including skin, kidney, lung, heart, blood vessels, liver, pancreas and intestine[15].

TGF-β is a multifunctional polypeptide hormone acting in essentially all cells, influencing distinct functions including proliferation, differentiation, apoptosis, immunoregulation, regulation of the inflammatory response, restitution and healing, as well as fibrosis[79]. At a cellular level, TGF-β affects virtually all stages of the chronic inflammatory and fibrotic disease processes.

The profibrotic effects of TGF-β are numerous, including influx and activation of ECM-producing cells, as well as promoting EMT and EndoMT. The biological action of TGF-β that contributes to several fibrotic diseases is the regulation not only of the synthesis but also the breakdown of ECM proteins, including collagens, fibronectins and proteoglycans[27].

Multicellular organisms present more than 60 TGF-β family members. These include 3 TGF-βs, 5 activins and at least 8 bone morphogenetic proteins (BMPs), all encoded by distinct genes. The three mammalian TGF-β isoforms, TGF-β1, 2 and 3, are secreted as latent precursor molecules (LTGF-β) containing an amino-terminal hydrophobic signal peptide region, the latency associated peptide (LAP) region and the C-terminal potentially bioactive region. The LTGF-β is usually linked to latent TGF-β-binding proteins (LTBP), requiring activation into a mature form for receptor binding and subsequent activation of signal transduction pathways. The LTBP is removed extracellularly either by proteolytic cleavage by various proteases such as plasmin, thrombin, plasma transglutaminase, or endoglycosidases, or by physical interactions of the LAP with other proteins, such as thrombospondin-1[80].

The actions of the TGF-β family are mediated through at least three types of TGF-β receptors, TGF-β RI, RII and RIII. TGF-β RI and RII are two different transmembrane serine/threonine kinase receptors with single transmembrane domains, which form homo- and heterodimer complexes that bind TGF-β, inducing phosphorylation of a family of proteins designated as Smads that transduce the ligand signal from the cell surface to the nucleus[80] (Figure 2).

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Transforming growth factor-α signaling and induction of gene transcription. Smad: Small mother against decapentaplegic; SARA: Smad anchor for receptor activation; TGF-β: Transforming growth factor β.

Smads are a family of 8 related proteins, which function as signaling intermediates for the TGF-β superfamily of ligands. Type I receptors specifically recognize and phosphorylate the ligand-specific receptor-activated Smad (R-Smad) and include Smad1, Smad5 and Smad8 (downstream of the BMP), and Smad2 and Smad3 (downstream of TGF-β and activin), that are recruited to activated TβRI by a membrane-bound cytoplasmic protein called Smad anchor for receptor activation (SARA). Upon ligation of TGF-β to its receptors, the phosphorylated Smad2 and 3 form a complex with the common mediator Smad (Co-Smad), such as Smad4. Co-Smad acts as a convergent node in the Smad pathways downstream of the TGF-β superfamily receptors, complexing R-Smad. R-Smad/Smad4 complexes are then translocated into the nucleus by a mechanism involving the cytoplasmic protein importin; they may then function as transcription factors, binding DNA either directly or in association with other DNA-binding proteins[80] (Figure 2).

A third group of Smad proteins, the inhibitory Smads such as Smad6 or Smad7, antagonize TGF-β signaling by interfering with the ligation of Smad2/3 to the activated receptor complex[80].

Using a combined cDNA microarray/promoter transactivation approach, several new Smad gene targets have been identified, among which are COL1A1, COL3A1, COL5A2, COL6A1, COL6A3, and TIMP-1, indicating that the Smad signaling pathway is crucial for the simultaneous activation of several fibrillar collagen genes by TGF-β[80]. About 60 other ECM-related genes were also identified as immediate-early gene targets downstream of TGF-β[81].

Several studies suggest that disruption of the TGF-β/Smad signaling pathway, either by the loss of Smad3 or the increase of Smad7 expression, prevents the development of tissue fibrosis in various organs, including the skin, kidney, lungs, liver and intestine[82-84].

Differences in small bowel and colonic morphology, as well as in the expression of collagens I-VI, α-SMA, TGF-β1, Smad7, and other factors, have been reported in Smad3 knockout mice compared to the littermate wild-type controls[85]. Using this tool made it possible to define the key role of the TGF-β/Smad signaling pathway in chronic intestinal inflammation and fibrosis[85]. In fact, a significant reduction of trinitrobenzene sulfonic acid (TNBS)-induced intestinal fibrosis has been reported in knockout mice as compared to wild-type mice[84]. Smad3 may take part in recruitment of fibroblasts to the site of injury, as well as in differentiation of fibroblasts to myofibroblasts and regulation of collagen synthesis. Furthermore, the loss of Smad3 interferes with the effects of TGF-β on chemotaxis and auto-induction of inflammatory cells[84,86]. These findings indicate that the TGF-β/Smad signaling pathway is the key element in determining the progressive nature of intestinal fibrosis.

The effects of TGF-β on ECM gene expression and subsequent development of tissue fibrosis may be related to additional mechanisms, such as CTGF, the expression of which is controlled by TGF-β in a Smad-dependent manner[87].

TGF-β preferentially transduces intracellular signaling through Smad proteins, but signal mechanisms that stimulate ECM accumulation can also be activated by other molecules such as members of the mitogen-activated protein kinase (MAPK) family including extracellular signal-regulated kinase (ERK), c-Jun N terminal kinase and p38 kinases, as well as members of the JAK and STAT protein family[88].

Studies on liver fibrosis demonstrate that both MAPK and Smad signaling, independently and in synergy, stimulate HSC activation, thus increasing collagen and α-SMA gene expression[88,89].

The JAK-STAT signaling pathway, activated by TGF-β, is involved in fibrogenesis and can be negatively regulated on multiple levels[90]. Suppressors of cytokine signaling may inhibit STAT phosphorylation by binding and inhibiting JAKs or competing with STATs for phosphotyrosine binding sites[91]. Protein inhibitors of activated STATs (PIAS) may act directly in the nucleus on STAT proteins. PIAS1 and PIAS3 inhibit transcriptional activation by STAT1 and STAT3, respectively, by binding and blocking their access to the DNA sequences[92].

TGF-β1 may also influence the imbalance between enhanced production and deposition and impaired degradation of ECM components. In IBD, the tissue expression of MMP-1, 2, 3 and 9 is increased in relation to that of TIMP-1 and 2, compared with controls[92]. How the balance of MMPs and TIMPs is regulated is still not clear, but blocking TGF-β1 in cultures of intestinal biopsies upregulated the expression of MMP-3 but not of TIMP-1, thus suggesting a new role for TGF-β1 in IBD tissue remodeling[93].

The activins are members of the TGF-β superfamily with broad and complex effects on cell growth and differentiation. The activins interact with heterodimeric serine/threonine kinase receptor complexes to activate Smad transcription factors and the MAP kinase signaling pathways. Important functions of activins, in particular of activin A, in tissue inflammation, repair and fibrosis of several organs, including the intestine, have been reported[94-96].

Activin is a dimeric protein consisting of β_A and β_B subunits connected by disulfide linkages resulting in three different configurations with similar functions: the homodimeric activin A (β_Aβ_A) and activin B (β_Bβ_B), and the heterodimeric activin AB (β_Aβ_B).

Their action is mediated by two heteromeric receptor complexes consisting of two transmembrane receptors (type I or II) with intrinsic serine/threonine kinase activities, similar to the TGF-β mechanism of signal transduction, while its biological effects seem to be inhibited by another activin-binding protein, namely follistatin[94].

Several studies have shown an increase in activin levels in IBD and in many other inflammatory diseases, thus giving rise to the hypothesis that it plays a significant role in the inflammatory response as well as in fibrosis[95].

Munz et al[96] reported an increase in the expression of activin, particularly the activin β_A-subunit, in surgical specimens from the gut of patients with IBD, showing a strong correlation with the degree of inflammation. In situ hybridization studies revealed the highest levels of activin mRNA in the mucosa and submucosa of highly inflamed areas. These findings suggest that activins play a novel and important role during the inflammatory processes of the gut[96].

TGF-β has been regarded as a pivotal growth factor in the formation and maintenance of connective tissues but it has been clearly proven that the effects on ECM gene expression and subsequent development of tissue fibrosis are related to other factors, such as CTGF, the expression of which is controlled by TGF-β in a Smad-dependent manner. Northern blot and in situ hybridization studies have demonstrated that CTGF is co-expressed with TGF-β in principally every fibrotic disorder and CTGF has been considered a possible key determinant of progressive fibrosis[97].

CTGF is a cysteine-rich mitogenic peptide that binds heparin and is secreted by fibroblasts after activation with TGF-β. In the adult mammal, CTGF acts as a downstream mediator of TGF-β action on connective tissue cells, where it stimulates cell proliferation and ECM synthesis. It does not appear to act on epithelial cells or immune cells. Since the biological actions of TGF-β are complex and affect several different cell types, CTGF may serve as a more specific target for selective intervention in processes involving connective tissue formation during wound repair or fibrotic disorders[97].

CTGF is an interesting molecule for future anti-fibrotic therapies as it is possible that inhibition of CTGF might block the pro-fibrotic effects of TGF-β, without affecting TGF-β’s immunosuppressive and anti-inflammatory effects. In addition to TGF-β, a number of other regulators of CTGF expression have been identified, including VEGF, TNF-α, shear stress, cell stretch and static pressure, hydrogen peroxide (H₂O₂), superoxide (O₂^-), hydroxyl radical (HO·) and nitric oxide (NO)[98].

Outlining the mechanisms that underlie CTGF gene regulation in normal and fibrotic cells might help in the design of future intervention strategies aimed at targeting specific interference with CTGF expression at sites of progressive fibrosis. In addition, forms of treatment targeting CTGF effects have been proposed which might lead to a favorable outcome of wound repair.

PDGF consists of two related peptide chains, PDGF-A or PDGF-1 (16 kDa, 124 amino acids), and PDGF-B or PDGF-2 (14 kDa, 140 amino acids) linked by disulfide bonds. All possible isoforms, i.e., PDGF-AA, PDGF-BB and PDGF-AB, are biologically active. More recently, two additional PDGF genes and proteins have been identified, namely PDGF-C and PDGF-D[99].

PDGF is synthesized mainly by megakaryocytes. It is stored in the α granules of platelets from which it is released after platelet activation. Other cell types also synthesize PDGF, including macrophages, endothelial cells, fibroblasts, glial cells, astrocytes, myoblasts, smooth muscle cells, and a number of tumor cell lines. Several factors induce PDGF synthesis, including hypoxia, thrombin and several cytokines and growth factors such as IL-1, IL-6, TNF-α, TGF-β, FGF-β and EGF[100].

Two tyrosine kinase receptors, called PDGFRα and PDGFRβ, have been described. Binding of the ligand leads to receptor dimerization, which initiates signaling. Ligand configuration and pattern of receptor expression influence the formation of different receptor dimers. Generally, mesenchymal cell expression of PDGFRs is low in normal conditions, but increases dramatically during inflammation.

PDGF plays crucial roles during development. Increased PDGF activity has been linked with several diseases and pathological conditions. Its expression is significantly increased in the inflamed intestine of patients with IBD. Intestinal fibroblasts, intestinal subepithelial myofibroblasts and the interstitial cells of Cajal are activated and proliferate in response to the PDGF family (mainly PDGF-BB)[51]. PDGF also enhances migration of fibroblasts, and its effects seem to be fibronectin-dependent. Increased activity of PDGF is also responsible for an excessive deposition of ECM in fibrotic processes within several organs, including the intestine.

IGFs (IGF-I and -II) are polypeptides composed of a single chain of 70 and 67 amino acids, respectively. They are potent mitogens since they have stimulatory effects on proliferation and inhibitory effects on apoptosis in epithelial cells. IGF-I is the predominant post-natal IGF, whereas IGF-II is a predominantly fetal IGF. IGF-I interacts with IGF-I receptor type I (IGF-IR) on the cell surface. Subsequent phosphorylation by a receptor tyrosine kinase initiates the intra-cellular signaling process. IGF-II also interacts with a receptor type II (mannose-6-phosphate IGF-IIR)[101].

In the intestine, IGF-I interacts principally with fibroblasts and epithelial and endothelial cells. The entire GI tract expresses IGF-I and -II and their respective receptors, the latter being localized in both the mucosal and muscularis layers. Expression of insulin-like growth factor binding protein (IGFBP)-6 in the intestine has also been described; IGFBPs are proteins that, by binding to IGF, are able to modulate its bioavailability and activity.

Expression of IGF-I is locally increased in several bowel diseases. During bowel inflammation, pro-inflammatory cytokines are able to induce IGF-I expression by mesenchymal cells. The IGF-I, in turn, could regulate the proliferation of these or other cell types acting in a paracrine manner on epithelial cells or in an autocrine manner on intestinal mesenchymal cells[102]. IGF-I plays a relevant role in the deposition of collagen and fibrosis. It has been shown to be up-regulated in the bowel of animals with experimental intestinal fibrosis and of patients with CD[102]. In an experimental model of rat colitis, an up-regulated IGFBP and collagen expression and down-regulated collagenase expression were shown, confirming the important role of IGF-I in collagen synthesis in colitis, mediated by IGFBPs[103]. It has been hypothesized, moreover, that IGF-I, through IGFBP-5, is able to modulate proliferation of fibroblasts/myofibroblasts and collagen synthesis[104].

In addition, in contrast with the local expression, the circulating levels of IGF-I and IGFBP-3 seem to be reduced in patients with IBD[105].

EGF is the prototype member of a family comprising different peptides with a similar primary structure that bind to a family of EGF receptors and have similar biological effects: they are potent mitogens and are also able to modify several properties of non-proliferating cells[106].

Several tyrosine kinase receptors are now known for EGF family members, such as human EGF receptors 1-4 (HER1, HER2, HER3 and HER4). Since these receptors can heterodimerize, they can form at least 10 different dimers that likely bind to distinct EGF-like molecules and deliver different signaling when activated[106].

Binding of EGF induces the phosphorylation of several intracellular proteins regulating transcription, translation, cell architecture, cell proliferation, and the production of inflammatory mediators.

EGF is a 53-residue peptide and is also known as urogastrone, since it was initially isolated from urine and was found to inhibit gastric acid secretion[107]; it is expressed in exocrine pancreas, duodenum, breast milk, colostrum and gastric juice[108].

EGF can be isolated from the lumen of the intestine; its receptors have been detected in the bowel, especially on monocytes and myofibroblasts.

EGF has numerous functions within the GI tract: it stimulates cell proliferation, inhibits gastric acid secretion, up-regulates intestinal electrolyte and nutrient transport, induces expression of several enzymes, enhances epithelial restitution and stimulates angiogenesis.

EGF stimulates fibroblast proliferation and ECM production in idiopathic pulmonary fibrosis; moreover, it stimulates a more pronounced ECM production in interstitial pneumonia fibroblasts than in normal fibroblasts, thus suggesting an important role of this growth factor in fibrotic processes[109].

The role of EGF in IBD is not clear. EGFR expression is generally upregulated in TNBS-induced colitis[110]. EGF administered intraperitoneally and subcutaneously reduces the severity of TNBS-induced colitis in rats[110]. EGF appears to be involved in the regulation of migration of human colonic fibroblasts and myofibroblasts[111,112]. Further studies are needed to fully elucidate the role of EGF and EGFR in IBD and intestinal fibrosis.

Endothelins are a group of polyfunctional cytokines. So far, four endothelins [(ET)-1, ET-2, ET-3 and ET-4)] and two endothelin receptors (ET_A and ET_B) have been recognized.

Endothelins strongly induce vasoconstriction and may provoke local ischemia. They can also induce the release of pro-inflammatory cytokines, alter the intestinal permeability and stimulate leukocyte adhesion in submucosal venules of the intestine, probably through up-regulation of cell adhesion molecules on the endothelium and leukocytes[113-115].

Adhesion of circulating leukocytes to the endothelium could be one of the initial steps in the pathogenesis of IBD, since adhesion is a crucial event in the recruitment of leukocytes to the inflamed tissue. In addition, several of the pro-inflammatory cytokines playing a key role in IBD, such as IL-1β and TNF-α, can remarkably increase endothelin production[116,117].

It has been shown that bosentan, a non-selective endothelin receptor antagonist, reduces intestinal inflammation in a murine model of IBD[118]. Probably the reduction of inflammation induced by this drug was due to a reduction in adherent leukocytes leading to a reduction of the mucosa infiltrate of inflammatory cells.

In vivo studies performed on cardiac, renal and pulmonary tissue have shown the role of endothelins in fibrosis[119]. ET1 stimulates colonic myofibroblast activation (differentiation, migration and contraction), effects prevented using an ET-1 antagonist[120]. These findings suggest a possible role for ET1 also in the development of intestinal fibrosis.

The relevance of the renin-ANG system, usually viewed as an endocrine system regulating physiological and vital cardiovascular processes, has expanded in the last decade to include independently regulated local systems in several tissues, new functions of the RAS and new active products of ANG II[121,122] (Figure 3).

Open in New Tab   Full Size Figure   Download Figure

Figure 3 Components and effects of the renin-angiotensin system. ACE: Angiotensin-converting enzyme; AT1: Angiotensin II type 1 receptor; AT2: Angiotensin II type 2 receptor; AT4: Angiotensin II type 4 receptor; MAS: Mas receptor; RAS: Renin-angiotensin system.

Besides circulating RAS, local RASs also exist in various organs and tissues, such as the heart, blood vessels, kidney, liver, pancreas, intestine, nervous system, reproductive system, as well as lymphatic and adipose tissue[123-125]. It has been reported that the ANG converting enzyme (ACE) produces the decapeptide ANG I and octapeptide ANG II, respectively. In addition, alternative enzymes to renin and ACE generate a number of bioactive peptides from ANG I and/or ANG II, such as ANG III, ANG IV and ANG-(1-7). ANG II, together with these bioactive peptides, mediates their specific functions via the respective cellular receptors of target tissues and organs[123]. In the last few years, the understanding of the RAS function has changed considerably, following the discovery of other ANG forms [such as ANG III, IV or (1-7)], additional enzymes involved in their generation (such as ACE2, aminopeptidase A or prolyl endopeptidase) and receptors (such as ANG II type 4 receptor and Mas). RAS, therefore, is not represented by a simple linear proteolytic cascade, but by a series of extremely complex reactions (Figure 3). All components of the RAS exist in the large bowel in adults[126]. In the colon, many cells, such as epithelial cells, vascular endothelial cells, mesenchymal cells and inflammatory cells, express ANG II receptors.

Local RAS has novel functions including the regulation of cell growth, differentiation, proliferation and apoptosis, generation of ROS, expression of cytokines (such as IL-6, TNF-α), activation of endothelial cells, as well as tissue inflammation, ECM production and fibrosis[123].

These different roles make some of the RAS components attractive therapeutic targets in various chronic diseases, including fibrosis.

ANG II, moreover, plays a role in the pathogenesis of chronic fibrogenetic diseases of various organs, including kidney, heart, lung, pancreas, liver and intestine, through the regulation of both inflammatory and fibrotic processes[127,128]. In such conditions, fibrosis is mediated, at least in part, through ANG II induction of TGF-β1[13].

Several components of RAS are over-expressed in the fibrotic liver, both in human and animal models[129]. In particular, it would appear that ANG II is able to induce proliferation of myofibroblasts and stellate cells, to stimulate inflammatory cells and to induce the release of several profibrotic molecules, such as TGF-β, CTGF and IL-1β[130]. Blocking RAS, by using ACE inhibitors or ANG II type I (AT1) receptor antagonists, reduces hepatic fibrosis[131].

In human lung fibroblasts, ANG II is able to modify PDGF-D, IL-4 and IL-7 expression and to induce higher levels of collagen and elastin. Using candesartan, an inhibitor of AT1 receptor, these effects have been shown to be suppressed[132].

In experimental models of kidney damage, RAS inhibitors (ACE inhibitors and AT1 antagonists) have shown beneficial effects on proteinuria, cell growth, inflammation and fibrosis, thus suggesting that ANG II could be involved in the fibrotic process activating mononuclear cells, increasing proinflammatory mediators and regulating matrix degradation[132].

New insights into the role of RAS in the development and modulation of chronic intestinal inflammation and related fibrosis have been reported[133]. The direct effect of ANG II on the pathogenesis of immune-mediated colitis was assessed using mice genetically deficient in angiotensinogen (Atg-/-), which is the precursor of ANGII[133]. TNBS-induced acute colitis was less severe in the Atg-/- mice compared to wild-type (Atg+/+) mice.

Both ANG II and TGF-β1 are overexpressed in intestinal fibrosis and stenosis, particularly in Crohn’s disease[134,135]. The production of TGF-β1, indeed, is strongly stimulated by the local activation of ANG II[13]. An up-regulation of ANG II in the colon of rats and mice with experimentally induced colitis also supports the role of RAS in intestinal inflammation and fibrosis[13,136].

Daily administration of the ACE inhibitor captopril in rats with chronic TNBS-induced colitis significantly reduced the macroscopic and microscopic pattern of both colonic inflammation and fibrosis, decreased the colon collagen content, and reduced TGF-β1 mRNA levels by about 60%. The antifibrotic mechanism of captopril could be related to the inhibition of ANG II-mediated TGF-β1 overexpression, and/or to a direct downregulation of TGF-β1 transcripts[13]. Likewise, the use of losartan, a specific AT1 receptor antagonist, significantly improved the macro- and microscopic scores of experimentally-induced colorectal fibrosis and reduced TGF-β1 concentration, thus suggesting that this drug has a preventive effect on colorectal fibrosis complicating TNBS-induced chronic colitis by a downregulation of TGF-β1 expression[137].

In view of these data, RAS could be considered as a future target for new antifibrotics in IBDs.

Cytokines regulate the inflammatory process, mediating the interactions between activated immune cells and non-immune cells. They can enhance an inflammatory response or reduce inflammation and promote healing.

IBDs are mainly characterized by an enhanced CD4+ T cell proliferation and trafficking into the intestinal mucosa and by alterations in the cytokine profile of the main activated Th cells, resulting in a disturbed balance between pro- and anti-inflammatory cytokines. The abnormal lamina propria T-cell activation and the consequent resistance to apoptosis are, indeed, considered as key events in the pathogenesis of IBD.

Upon activation, naïve CD4+ T cells can differentiate into different subsets depending on the surrounding cytokine milieu. CD4+ Th cells are currently divided into four major subsets, based on their expression profile of transcription factors and secreted cytokines: Th1, Th2, Th17 and Treg.

Th1 cells are characterized by the secretion of interferon γ (INF-γ). The differentiation of Th1 cells is mainly induced by IL-12 and can be further enhanced by INF-γ. Th1 cells express the transcription factors T-bet and STAT4. IFN-γ is a pro-inflammatory cytokine which together with IL-12 induces the differentiation of macrophages and the production of other pro-inflammatory cytokines, such as IL-1β, IL-6, IL-8, granulocyte-macrophage colony-stimulating factor (GM-CSF) and TNF-α[138].

Th2 cells produce primarily IL-4, IL-5 and IL-13. Th2 cells develop in the presence of IL-4 and require the transcription factors GATA3 and STAT6.

Th1 and Th2 negatively regulate each other through their specific cytokine actions: IL-12 represses the induction of Th2 cells, whereas IL-4 inhibits Th1 cell development.

Microarray experiments have established that the various gene-expression profiles associated with different pathological conditions are strongly influenced by Th1- and Th2-polarization of the chronic inflammatory responses[139].

The Th1 response seems to be involved mainly in CD, whereas the Th2 response occurs mainly in UC[140]. However, during the late stages of IBD, both Th1 and Th2 cytokines play a pivotal role in the progression of intestinal fibrosis. Several findings have demonstrated that IL-1β and TNF-α, two Th1 pro-inflammatory cytokines, stimulate fibroblast proliferation and collagen deposition, whereas the development of a Th2 cell response, and particularly IL-4 and IL-13, upregulates the expression of several genes including those that encode procollagen-I, procollagen-III, MMP-2 and MMP-9 as well as TIMP-1[139,141].

In addition, the other two subsets of CD4+ T cells, represented by the Th17 and Treg cells, are involved in the development of intestinal fibrosis in alternate ways[142].

Th17 cells, which are strongly associated with autoimmune pathology, are characterized by the production of IL-17A, IL-17F, IL-21, IL-22 and TNF-α[142]. Differentiation of Th17 cells needs the combined action of TGF-β, IL-6 and IL-21 in mice, whereas IL-6 and IL-21 can be replaced by IL-23 or IL-1β in humans. These cytokines induce the transcription factors RORγt (mice) or RORc (human) and STAT3. Development of Th17 cells is suppressed by INF-γ and IL-4, which promote Th1 and Th2 cells, respectively. Once Th17 cells have developed, IL-23 is needed for stabilization and further expansion of these cells. Also, IL-1β and IL-6 can act to enhance the development and expansion of these cells. The intestine of patients with IBD presents higher IL-17 levels than in healthy subjects. There is emerging evidence that IL-17 strongly interacts with IL-23, the expression of which is considerably increased in the inflamed intestine; its selective depletion with monoclonal antibodies has been demonstrated to greatly attenuate T-cell mediated colitis in mice[143].

Treg cells, which control effector T-cell responses, produce the anti-inflammatory cytokines IL-10 and TGF-β. The differentiation of Treg is induced by TGF-β but inhibited in the presence of pro-inflammatory cytokines. Treg are characterized by the expression of the transcription factor Foxp3 and STAT5 and the expression of CD25 on their surface.

There is a growing body of evidence proving that the inflammatory cascade which influences the progression of intestinal fibrosis can involve many Th1, Th2, Th17 and Treg cytokines, such as IL-1β, IL-4, IL-6, IL-13, IL-17, IL-21, IL-22, IL-23, IL-33 and TNF-α (Table 3). Consequently, the use of biological therapies antagonizing these cytokines could be of value in the treatment of intestinal fibrosis and, therefore, may represent the targets of new antifibrotic drugs.

IL-1 and IL-33: IL-1 belongs to a family of proinflammatory cytokines that are rapidly expressed following tissue damage. Besides its role in acute and chronic inflammation, IL-1 is involved in tissue remodeling following chronic inflammation in various organs.

IL-1 contributes to the development of fibrosis during chronic intestinal inflammation through various mechanisms, such as regulating intestinal myofibroblast activation, the induction of chemokines (e.g., IL-8, MCP-1) and MMP secretion, and the turnover of ECM[144]. Furthermore, IL-1, in combination with TNF and INF-γ, is able to increase the TGF-β-induced epithelial-mesenchymal transition (EMT), an important cellular process of fibrogenesis[145].

Recently, it has been reported that IL-33, a novel member of the IL-1 family, induces mucosal pathology in vivo and may lead to the development of fibrosis and angiogenesis[146]. TLR-3 is one of the strongest promoters inducing IL-33, which activates myofibroblasts and pericytes.

IL-4 and IL-13: Some Th-2 cytokines (IL-4 and IL-13) are present in elevated levels in fibrotic processes and induce activation and differentiation of fibroblasts to myofibroblasts and also induce production of collagens[1,17,139].

IL-4 appears to be involved in the development of pulmonary fibrosis; in particular, activation of the IL-4Rα pathway in macrophages seems to be fundamental in silica-induced pulmonary fibrosis[147]. Furthermore, it was demonstrated that IL-4 increases the mRNA expression of collagen I, III and IV in cultured human hepatic fibroblasts; these effects require STAT-6 activation[148]. IL-4 level, finally, correlated with cardiac fibrosis in patients with heart failure[149].

Inhibition of IL-13 reduces hepatic and skin fibrosis[150,151]. In experimental studies a marked reduction of hepatic fibrosis by IL-13 blockade has been reported[150,151]. IL-13 signaling through the corresponding receptor IL-13Rα induces production of latent TGF-β (TGF-β bound to LAP) in macrophages and, indirectly, contributes to its activation by stimulating the synthesis of enzymes, such as MMPs and cathepsins, able to remove LAP[152]. In TNBS-induced colitis, fibrosis development is dependent upon IL-13 binding to the IL-13 receptor to induce TGF-β[153]. In the same way, if IL-13 signaling is inhibited TGF-β is produced in reduced amounts and fibrosis does not occur[153]. Soluble IL-13Rα2-Fc is a highly effective decoy receptor of IL-13 which can reduce the progression of established fibrotic disease. IL-10 has also been shown to inhibit fibrosis in numerous experimental models[1]. The IL-13 decoy receptor and IL-10, by suppressing collagen deposition, act as endogenous factors that slow the progression of fibrosis.

Both IL-4 and IL-13 appear to be responsible for promoting fibrocyte differentiation from CD14+ peripheral blood monocytes, without induction of their proliferation. In dermal fibroblasts, IL-4 and IL-13 stimulate ERK1/2 pathways, which are involved in the modulation of collagen gene expression[154].

IL-6: In addition to a potent proinflammatory action, IL-6 may also have a relevant role in the development of fibrosis. In the cardiac allograft model, IL-6 neutralization improved graft fibrosis[155]. The role of IL-6 in fibrosis has been further confirmed by the finding of its ability to modulate TGF-β and TGF-βRII mRNA and protein levels in mouse skin[156] and to strongly stimulate proliferation in normal and keloid fibroblasts, as well as increase levels of STAT-3, which contributes to several processes including collagen production[157]. IL-6 is markedly increased in CD where it appears to stimulate fibrogenetic mesenchymal cells[158].

IL-21 and IL-22: IL-21 appears to be related to IBD-associated intestinal fibrosis. IL-21 promotes fibrosis by enhancing the development, survival and migration of Th2 cells. Moreover, it stimulates the expression of IL-4 and IL-13 receptors in macrophages, inducing their activation[159], stimulates the secretion of ECM-degrading enzymes by fibroblasts and the secretion of the T cell chemoattractants by epithelial cells[160]. IL-21 is produced in excess in CD compared to controls. IL-21 together with IL-6 is critical in the development of Th17 cells.

IL-22, primarily produced by Th17 cells, presents bifunctional characteristics, pro- and anti-inflammatory, depending on the local milieu. In the intestine, it is involved both in mucosal defense and in wound healing processes. IL-22 plays a protective role in IBD by enhancing barrier integrity and epithelial immunity of the intestinal tract[161]. A mouse model of UC has, in addition, demonstrated that IL-22 is able to activate the innate immune pathway by enhancing STAT3 activation, particularly within colonic epithelial cells, and by inducing both the restitution of mucus-producing goblet cells and the STAT3-dependent expression of mucus-associated molecules. These studies might, therefore, suggest a protective role for this cytokine in intestinal inflammation[162]. The involvement of IL-22 in the regulation of fibrotic processes has been shown in a mouse model of hypersensitivity pneumonitis (that generally progresses to lung fibrosis), in which direct blockade of IL-22 enhanced the deposition of collagen in the lung. These findings reveal a protective pathway of this cytokine in the development of lung fibrosis and could suggest an analogous role also in the intestine[163].

IL-17 and IL-23: The IL-23/IL-17 axis plays a role in normal intestinal homeostasis, although the precise actions of these cytokines remain to be fully elucidated. In the normal intestine, constitutive production of small amounts of both IL-23 and IL-17 may act to protect the epithelial layer fortifying tight junction formation between epithelial cells and inhibiting bacterial colonization. In inflammation, activated dendritic cells produce large amounts of IL-23, which could activate innate immune cells to produce pro-inflammatory cytokines such as IL-1β, TNF-α and IL-6. IL-23 also induces an increased production of IL-17 and INFα, both by T cells and non-T cells. Recently, it has been shown that targeting IL-23 by employing a p40 peptide-based vaccine improves TNBS-induced acute and chronic murine colitis with a significant decrease in collagen deposition[164]. The main action of IL-17 may be to promote the production of chemokines that recruit and activate granulocytes, further increasing inflammation[143]. IL-17 was also found to be a potent activator of mesenchymal cells, especially of myofibroblasts[165].

In vivo, fibrotic models show enhanced IL-17A expression and activation of IL-17A-associated signaling pathways. Studies performed on pulmonary fibrosis models showed evidence that IL-17A increased synthesis and secretion of collagen and induced epithelial-mesenchymal transition in alveolar epithelial cells in a TGF-β1-dependent manner[166].

Moreover, it has been demonstrated that IL-1β-mediated pulmonary fibrosis is IL-17A-dependent[167] and that IL-17A blocking is able to attenuate myocarditis-induced cardiac fibrosis and ameliorate ventricular function; further confirming its role in the fibrogenetic process[168].

A study on Th17 cells in human peripheral blood demonstrated that the expansion of IL-23R+ CD4+ cells was promoted by the combination of IL-7 and IL-12, which are also able to increase the secretion of IFN-γ, while IL-12 alone stimulated these cells to secrete predominately IL-17[169]. On the basis of these data, IL-7 and IL-12 have similarly been identified as anti-fibrotic cytokines, even if their effective role in the fibrotic process remains to be established.

Another central mediator of the fibrotic process in IBD is TNF-α, a proinflammatory cytokine with important immunomodulatory properties, abundantly expressed in the intestine of patients with CD and UC. This cytokine is produced from various cells such as monocytes/macrophages, adipocytes and T cells, and has different effects including further macrophage and T cell activation, stimulation of granuloma formation and expression of adhesion molecules in endothelial cells with consequent recruitment of neutrophils and other immune cells at the site of inflammation.

TNF-α is able to induce intestinal fibrosis by up-regulating collagen accumulation; moreover, it has mitogenic effects in intestinal myofibroblasts and is able to extend the inflammatory state thus increasing the expression of other inflammation mediators, such as IL-6 and IL-1β, and activating NFκB-dependent pathways, a transcription factor that increases the expression of cytokines, enzymes and adhesion molecules[170].

TNF-α acts by binding to two immunologically distinct TNF-α receptors of approximately 55 kDa (TNF-R1) and 75 kDa (TNF-R2). Studies performed on intestinal myofibroblasts have shown that TNFR2 is essential for TNF-α-induced cell proliferation and collagen synthesis[171].

Furthermore, TNF-α induces TIMP-1 expression and reduces MMP-2 activity and collagen degradation. Inhibition of collagen degradation, through TIMP-1, and induction of collagen gene transcription are responsible for the effects of TNF-α on total collagen accumulation; moreover, TNF-α also appears to have additional effects on collagen synthesis when combined with IGF-I, a growth factor implicated as a key pro-fibrogenic mediator during intestinal inflammation in vivo. IGF-I and TNF-α synergistically stimulate intestinal myofibroblast proliferation and collagen production[172,173].

The role of TNF-α in IBD was confirmed by the therapeutic benefits of anti-TNF-α monoclonal antibodies. In several clinical trials, anti-TNF-α drugs, such as infliximab, adalimumab and certolizumab pegol, induced clinical remission and healing of mucosal lesions in about one third of patients with CD[174-176].

There is not strong evidence regarding the effectiveness of anti-TNF-α drugs to reduce intestinal fibrosis in IBD.

Infiltrating leucocytes in the inflamed mucosa produce a large amount of ROS. ROS include various species (O^2-, H₂O₂, HO·) and are involved in several processes, such as regulation of signal transduction, activity of phagocytes, oxidative damage of proteins, lipids and nucleic acids[177].

Superoxide is mainly generated from the uncoupling of the mitochondrial electron-transport system and from several enzymes including NADPH oxidases, cyclooxygenases, lipoxygenases and nitric oxide synthases. Hydrogen peroxide is directly obtained from oxygen, spontaneously or enzymatically (through the action of the enzyme superoxide dismutase); it can also be the result of lipid metabolism in peroxisomes. In the presence of metals, hydrogen peroxide can be involved in HO· formation[178].

ROS, moreover, can negatively affect cell signaling by altering the transcription process and phosphorylation of proteins, including transcription factors.

ROS are involved in acute and chronic inflammatory processes that include a neutrophil or macrophage infiltrate, such as acute and chronic infections, autoimmune conditions such as IBD, arthritis and other inflammatory conditions[179]. In several pathological states the increase of ROS depends on an over-expression or increased activity of enzymes involved in their production. Increased activity of the NADPH oxidase enzyme is associated with pulmonary, cardiac, hepatic and intestinal fibrosis[179].

Patients with idiopathic pulmonary fibrosis have a lower antioxidant capacity compared to healthy subjects[180-182]. It has been reported that antioxidants protect rats against experimental pulmonary fibrosis[183,184]. It is thus likely that attenuating oxidative stress in tissues could prevent fibrosis caused by ROS.

A recent study showed that mice with NADPH oxidase deficiency had phagocytes which did not produce ROS and they did not develop lung fibrosis induced by bleomycin[185]. Inhibition of ROS production in this experimental model was associated with alterations in IL-6 levels and in the MMP-9/TIMP-1 ratio, molecules involved in pulmonary fibrosis and remodeling[185].

The role of ROS has been widely studied in liver fibrosis and substantial data suggest its importance in the transformation of HSCs into activated collagen-producing cells[186]; ROS appear to be a key mediator in collagen gene regulation[187].

ROS could also be involved in intestinal fibrosis. It has been shown that inhibition of oxygen radical secretion improves experimental colitis in mice[188].

Novel molecules involved in IBD and in tissue fibrogenesis appears to be PPARs[189]. Three different isoforms of PPARs have been identified, termed PPAR-α, PPAR-γ and PPAR-δ, each one encoded by specific genes and with distinct patterns of tissue distribution but with similar structure and function. PPARs are nuclear receptors, which regulate gene transcription by binding to retinoid X receptors (RXR) as functional heterodimers in response to a variety of endogenous and exogenous ligands. The PPAR/RXR complex modulates gene target expression by binding to the PPAR response elements (PPREs) in the gene promoter[190]. These nuclear receptors present three main domains: the N-terminal domain containing a ligand-independent transactivation function; the DNA-binding domain containing two zinc-finger motifs responsible for binding on PPREs to regulate transcription of PPAR-responsive genes; the ligand-binding domain (LBD) containing a ligand-dependent activation function. The LBD shows the presence of a wide hydrophobic region which could explain the increased ability of PPARs to bind a wide variety of natural and synthetic ligands[191,192] (Figure 4).

Open in New Tab   Full Size Figure   Download Figure

Figure 4 Activation of the peroxisome proliferator activator receptor-γ pathway. Regulation of several genes involved in fibrogenesis. PPAR-γ: Peroxisome proliferator activator receptor-γ; PPRE: PPAR response elements; RXR: Retinoid X receptors; L: Ligand; FA: Fatty acids.

In particular, the PPAR-γ isoform, identified mainly in the colorectal mucosa but also in adipocytes, the liver, vascular tissue and several inflammatory cells (monocytes and macrophages, dendritic cells, B and T cells), seems to be involved in several physiological processes, such as differentiation of adipocytes, glucose homeostasis, lipid metabolism, inflammatory and immune processes, as well as fibrosis[189,190]. Several ligands, either natural, such as arachidonic acid metabolites (15-d-PGJ2) and some eicosanoids, or synthetic, such as thiazolidinediones and some non-steroidal anti-inflammatory drugs, appear to be responsible for the activation of PPAR-γ (Figure 4).

PPAR-γ activation seems to be strongly related to the TGF-β/Smads pathway. The stimulation of PPAR-γ with specific ligands, indeed, interferes with the Smad3 pathway by directly antagonizing Smad3 or downregulating the CTGF expression that promotes TGF-induced synthesis of collagen[189,193]. PPAR-γ agonists inhibit fibroblast migration and proliferation[194] as well as the transdifferentiation of epithelial and mesenchymal cells in activated myofibroblasts[195], one of the key points in fibrosis development. PPAR-γ ligands repress TGF-β-induced myofibroblast differentiation and activation by targeting the PI3K/Akt and Smad3 pathways, respectively[196,197]. Overexpression of PPAR-γ prevents the development of tissue fibrosis, whereas its loss increases susceptibility to fibrosis[198,199]. All these findings could explain the ability of PPAR-γ to interfere in multiple phases of the tissue fibrotic processes. Therefore, PPAR-γ should be regarded as providing innate protection from excessive fibrogenesis and as a potential new target for the development of novel compounds with anti-fibrotic properties[200]. Several PPAR-γ ligands with selective activity are under development. Experimental studies have shown that PPAR-γ agonists attenuate fibrosis in various organs including lung, kidney, pancreas, liver and intestine, antifibrotic effects that are abolished by the use of PPAR-γ selective antagonists[201-206]. The PPAR-γ coding gene is considered a potential susceptibility gene for IBD[189].

In the 1970s, during a discovery programme for novel anti-microbial agents from natural sources, a soil sample from Easter Island (known as Rapa Nui) was found to contain Streptomyces hygroscopicus, a bacterium that produces a potent antifungal metabolite; the antibiotic was isolated and called rapamycin (Rapa + mycin). TOR was originally identified in Saccharomyces cerevisiae and, in subsequent biochemical studies, also in mammals and therefore was named the mTOR.

mTOR is a 289 kDa phosphatidylinositol 3-kinase-related kinase[207]. mTOR signaling is activated by hormones, growth factors, amino acid levels, stress and alterations in cellular energy status[208] (Figure 5). Among the signal inputs, growth factor- and hormone-induced mTOR activation is the best characterized and is believed to be mediated through the PI3K pathway and PKB/Akt.

Open in New Tab   Full Size Figure   Download Figure

Figure 5 Features of mammalian target of rapamycin. mTORC1: Mammalian target of rapamycin complex1; mTORC2: Mammalian target of rapamycin complex 2; 4EBP1: 4E binding protein; S6K: S6 kinase; PKB/Akt: Protein kinase B; SGK: Serum- and glucocorticoid-inducible kinase; PKC: Protein kinase C.

mTOR forms at least two distinct complexes. mTOR complex 1 (mTORC1) is composed of mTOR, GβL and Raptor and is responsible for sensing the nutrient; mTORC2 consists of mTOR, GβL and Rictor and is involved in the organization of actin. mTORC1 is sensitive to rapamycin whereas mTORC2 is insensitive to rapamycin.

The best characterized downstream targets of mTORC1 are S6K1 and 4E-BPI which control protein synthesis and cell growth and proliferation, respectively, as well as autophagy, angiogenesis and fibrosis[208] (Figure 5). Rapamycin induces a translational arrest by preventing phosphorylation of S6K-1 and 4E-BPI by mTOR[209].

It has been demonstrated that mTOR strongly enhances expression of hypoxia-inducible factor (HIF)1-α, a subunit of HIF which is a transcription factor that mediates expression of several genes, the products of which play a considerable role in inducing angiogenesis[210]. Two of the key gene products induced by HIF are VEGF and angiopoietin-2, which represent the main driving factors of neo-angiogenesis.

It has been shown that mTOR signaling is required for angiogenesis and plays a key role in endothelial cell proliferation in response to hypoxia. Hypoxia, indeed, rapidly promotes and sustains mTOR phosphorylation, whereas mTOR inhibition by rapamycin specifically abrogates hypoxia-mediated amplification of endothelial proliferation and angiogenesis[211,212]. Moreover, it has been recognized that mTOR-dependent HIF1-α expression is sensitive to rapamycin. Inhibition of mTOR with rapamycin can reduce the process of angiogenesis by blocking VEGF, TNF-α and PDGFα production through inhibition of HIF1-α expression and its transcriptional activation[213-215]. mTOR might, therefore, play an important role in the abnormal angiogenesis associated with IBD. It has been reported that mTOR is an effector of EGFR- and ANG type 1 receptor-induced fibrosis[216,217].

mTOR signaling is considered an attractive target for antifibrotic intervention. mTOR inhibitors constitute a relatively new category of immunosuppressive and antineoplastic drugs[218]. These share a unique mechanism of action that is focused on the inhibition of the mTOR. Their clinical applications have recently expanded significantly to cover a wide spectrum of immune and non-immune-mediated disorders, including IBD as well as solid organ transplantation, various solid organ and hematological malignancies, metabolic problems such as diabetes mellitus and obesity, and even fibrotic conditions, including skin, pulmonary, renal, hepatic and intestinal fibrosis[215-217,219-223].

It has been reported that rapamycin exerts direct antifibrotic activities both by reducing the number of fibroblasts and myofibroblasts and by down-regulating the production of fibrogenic cytokines, such as IL-4, IL-6, IL-17 and TGF-β1, and the synthesis of type I and III collagen[224-226]. Application of rapamycin to cultures of fibroblasts dose- and time-dependently downregulates the expression of cytoplasmic PCNA, cyclin D1, α-SMA, fibronectin and collagen[227]. Rapamycin may improve the defective autophagy associated with fibrostenotic CD[26,35,42,207].

mTOR might, therefore, also play important roles in IBD-associated fibrosis. Combined immunosuppressive and anti-fibrotic actions of rapamycin and its analogues may result in a promising treatment approach to fibrotic chronic enteropathies such as CD. This has been confirmed by two case reports in which two patients with severe refractory CD were successfully treated with two different analogues of rapamycin: sirolimus and everolimus[228,229]. In addition, a recent clinical trial demonstrated the effectiveness of everolimus to maintain steroid-induced remission in patients with moderate-to-severe active CD[230].

An increased turnover of ECM components leads to an intensive remodeling of connective tissue. A delicate balance exists between synthesis and degradation of ECM components. Disturbance of this balance may result either in progressive organ destruction, as seen in formation of ulcers and fistulae, or excessive deposition of collagen, resulting in fibrosis[231-233]. Degradation of all ECM components is regulated by the enzymatic activity of the predominant and large family of MMPs. In addition to playing a central role in ECM turnover, MMPs proteolytically activate or degrade a variety of non-matrix substrates including chemokines, cytokines, growth factors and junctional proteins. Thus, they are increasingly recognized as critical players in inflammatory response and fibrogenesis[231-233].

MMPs are a group of calcium-activated and zinc-dependent endopeptidases that are secreted as proforms (inactive zymogens)[231-233]. They are produced by various cell types, including mesenchymal cells, T-cells, monocytes, macrophages and neutrophils, in response to inflammatory stimuli, such as IL-1, TNF-α, or TGF-β1[233]. The MMP family consists of at least 25 distinct members[234], which can be subclassified according to their substrate specificities: collagenases (e.g., MMP-1, -8, -13, -18), gelatinases (MMP-2, -9), stromelysins (MMP-3, -7, -10, -11), elastase (MMP-12), membrane types (MMP-14, -15, -16, -17, -24, -25) and others (MMP-19, -20, -23, -26, -27, -28). All MMPs become activated via proteolytic cleavage and are regarded as the major enzyme group capable of degrading ECM components such as collagens, laminins and fibronectins, including basement membranes. Once one type of MMP is activated it can catalyze activation of other MMP proforms.

MMP activity is controlled by specific and non-specific inhibitors such as TIMPs and α2-macroglobulin, respectively[16,232,233]. TIMPs are produced by the same cell types as MMPs and form a 1:1 complex with activated MMPs. The fine balance between MMPs and TIMPs regulates the turnover of ECM both under physiological conditions and in tissue remodeling during inflammation and wound healing. Thereby, the imbalance, due to reduced MMP activity and/or increased expression of TIMPs, may lead to excessive deposition of ECM proteins driving fibrogenesis[15,50,51].

In IBD, overexpression of several MMPs has been reported, which might lead to inflammation and tissue injury, facilitating the migration and invasion of fibroblasts into the bowel wall, ultimately leading to fibrosis[235-238]. Addition of recombinant MMP-3 to an ex vivo gut model caused extensive tissue injury, and inhibition of MMPs almost completely abolished tissue destruction[239]. The expression of MMP-1, -3, -7 and -13, as well as TIMP-3, was increased in the wound edges of ulcerated tissue in UC and CD[240]. Levels of MMP-1, -2, -3 and -9 relative to TIMP-1 and -2 are increased in inflamed, compared to non-inflamed, IBD tissue homogenates, regardless of the presence of fibrosis[92,231].

TGF-β1 may lead to fibrosis by creating an imbalance of TIMP1/MMP expression in favor of TIMP-1[241]. Stricture-derived myofibroblasts in CD overexpress TIMP-1[242] and the expression of TGF-β1 and TIMP-1 is increased in the mucosa overlying strictured compared to non-strictured intestine, while MMP-3 and -12 are down-regulated[243]. In support of this notion, blocking TGF-β1 in cultures of intestinal biopsies upregulates the expression of MMP-3, but not of TIMP-1[93]. In a murine model of chronic inflammation, fibrosis is associated with an increase in TIMP-1[244].

A possible genetic basis has been explored: SNPs in genes encoding MMPs and TIMPs have been reported. An SNP at the TIMP-1 site is associated with increased susceptibility for CD, and an SNP at the MMP-3 site may increase the chance of stenotic complications[245].

An imbalance in MMP:TIMP expression and enhanced levels of the messages for fibrogenic cytokines and ECM proteins were also reported in late radiation enteritis and Schistosoma mansoni-induced chronic colitis[246,247]. Despite the fact that the expression of collagens, MMPs and TIMPs simultaneously increased, quantification of net collagen deposition showed an overall accumulation of collagen. These findings indicate that the intestine affected by chronic inflammation is subjected to an active process of fibrogenesis as well as fibrolysis, with a balance toward fibrogenesis. This demonstrates that established fibrotic tissue is not scarred fixed tissue but is subjected to a dynamic remodeling process.

Taken together, all these data strongly support the hypothesis that an imbalance of tissue-degrading enzymes and their inhibitors may cause intestinal fibrosis[232,233,248]. MMPs and TIMPs may, therefore, be considered as targets for new anti-fibrotic treatment approaches.