Losartan reduces trinitrobenzene sulphonic acid-induced colorectal fibrosis in rats

1Institute of Gastroenterology, Hadassah University Hospital, Jerusalem, Israel; 2Gastroenterology Unit; 3Clinical Epidemiology Unit, Department of Internal Medicine and Public Health, University of L’Aquila, L’Aquila, Italy; 4Institute of Animal Science, The Volcani Center, Bet Dagan; 5Department of Pathology, Hadassah University Hospital, Jerusalem, Israel. *Contributed equally to this work Correspondence: Dr Dov Wengrower, Institute of Gastroenterology, Hadassah University Hospital, Hebrew University, PO Box 12000-Kyriat Hadassah, Jerusalem 91120, Israel. Telephone 972-2-6678417, fax 972-2-6437492, e-mail dovw@ekmd.huji.ac.il Received for publication March 5, 2011. Accepted March 25, 2011 Intestinal fibrosis, which may lead to stenosis and obstruction, may develop in several enteropathies including inflammatory bowel disease (IBD), radiation enteropathy, graft versus host disease, collagenous colitis, eosinophilic enteropathy, drug-induced enteropathy and postsurgical intestinal adhesions and stenosis (1-4). The most frequent and severe intestinal fibrosis occurs in patients with Crohn’s disease (CD), which invariably recurs after surgical intestinal resection (5,6). Despite advances in CD management over the past two decades, the incidence of strictures and the requirement for intestinal resection have not diminished (7). Currently, no effective preventive measures or medical therapies are available for intestinal fibrosis and, thus, surgery remains the only available antifibrotic strategy in the management of fibrostenosing CD (3). Intestinal strictures characteristically display thickening of all layers of the intestinal wall. Histological examination shows the presence of abnormal collagen deposition that disorganizes the submucosa, muscularis mucosa, muscularis propria and serosal layers (1,5). Collagen is one of the major component proteins of the extracellular matrix (ECM); several subtypes have been identified. Intestinal strictures are characterized by an increase in total collagen and also in the relative amount of type III collagen (8,9). While the pathogenesis of intestinal inflammation in IBD has been extensively investigated, knowledge of the pathogenesis of fibrosis remains limited. Intestinal fibrosis, as it occurs in other organs, develops following chronic inflammation, but may progress independently of inflammation originaL articLe

The activation of ECM-producing cells is regulated by several soluble factors including cytokines, chemokines, growth factors and products of oxidative stress (4,(10)(11)(12).Of the molecules regulating the activation of ECM-producing cells, transforming growth factor-beta (TGF-b) appears to play a pivotal role (4,16).TGF-b is a member of a large family of pleiotropic cytokines that include bone morphogenetic proteins, activins and other related factors.
TGF-b/Smad signalling plays a central role in tissue fibrogenesis, acting as a potent stimulus of ECM protein accumulation.The expression of TGF-b and its cell receptors is increased in intestinal fibrostenotic tissue and in the fibroblasts of the affected segments (16,(22)(23)(24)(25). Animal models of IBD provide additional compelling evidence for the fundamental role of TGF-b in triggering and sustaining intestinal fibrogenesis (21,(26)(27)(28).In particular, TGF-b stimulates the production of the ECM and suppresses matrix degradation by inhibiting MMPs, thereby enhancing the deposition of collagen and other components of the ECM (4,5,(10)(11)(12)29). Furthermore, TGF-b1 provides a strong stimulus for the synthesis of connective tissue growth factor (CTGF), a potent stimulator of fibroblast and myofibroblast proliferation, ECM production and for the autoinduction of TGF-b (13,19,30,31).TGF-b autoinduction is believed to be an important aspect in the maintenance of the fibrotic process over time (13).All of these responses to TGF-b contribute to the onset and maintenance of fibrosis in the target organs.
The production of TGF-b1 is strongly stimulated by the local activation of angiotensin (ANG) II (ANGII) (32)(33)(34)(35)(36)(37)(38), which is the principal effector of the renin-angiotensin system (RAS), the activity of which is increased in the colonic mucosa of CD patients (39).The role of the ANG cascade in the development of fibrosis has been widely described in different organs including kidney, heart, blood vessels, lung, pancreas and liver.It has been reported that the fibrogenic response to injury is strongly mediated through ANGII induction of TGF-b1 expression (34)(35)(36)(37)(38). Furthemore, fibrosis is significantly improved or even reversed by ANGII blockade by an angiotensin-converting enzyme (ACE) inhibitor or ANGII receptor antagonist: findings that are closely correlated to the reduction of TGF-b1 expression (40)(41)(42)(43)(44)(45)(46)(47)(48)(49)(50)(51)(52)(53)(54).These observations suggested that the neutralization of the fibrogenic ANGII could also be a beneficial therapeutic target in intestinal fibrosis.To test this hypothesis, we previously evaluated the efficacy of inhibition of ANGII by captopril, an ACE inhibitor, in the prevention of trinitrobenzene sulphonic acid (TNBS)-induced chronic colitis-associated fibrosis.For the first time, we demonstrated that captopril was effective in preventing colorectal fibrosis in 2,4,6-TNBS-induced chronic colitis, and showed that antifibrotic action was mediated by a downregulation of TGF-b1 overexpression (27).
Based on these data, we hypothesized that experimental colorectal fibrosis could be also prevented by administration of losartan -a specific ANGII receptor antagonist.In the present study, we evaluated the effects of losartan on the course of TNBS-induced chronic colitis with colorectal fibrosis, and on colonic TGF-b1 expression in rats.This experimental model of colitis resembles human CD with transmural lesions and serosal involvement.Intestinal lesions are observed from one to three weeks after intrarectal administration of TNBS.

methoDs materials
Losartan was purchased from Hadassah Pharmacy (Jerusalem, Israel), as Cozaar (Merck Sharpe & Dohme, Isreal) (losartan potassium).All other reagents were purchased from Sigma Chemical Co (USA) unless otherwise indicated.2,4,6-TNBS was purchased from Sigma Chemical Co (USA).animals Thirty healthy male Sabra rats weighing 300 g to 325 g were used (Harlan Jerusalem, Israel).The animals were kept in a restrictedaccess room with controlled temperature (23°C) and 12 h light/12 h dark cycles.Standard rodent chow and tap water was allowed ad libitum.The Animal Research Committee of Hadassah School of Medicine and the Hebrew University (Jerusalem, Israel) approved the protocol.

induction of colitis
Under light ether anesthesia, colitis was induced by intrarectal instillation of TNBS (15 mg/mL) diluted in an equal volume of 50% ethanol (volume/volume) as described by Morris et al (55).The animals were deprived of food for 24 h before TNBS instillation.The solution of TNBS-ethanol was administered in a total volume of 1 mL through medical-grade, polyurethane 6 Fr enteral feeding tubes, the tips of which were positioned at 8 cm past the anus.The cannula was left in place for 1 min to ensure that the TNBS-solution would not be expelled immediately.Animals in the control groups were given an enema of the same volume containing 0.9% saline instead of TNBS.
Animals were observed daily for food and fluid intake.The rats were weighed before TNBS administration and on the day they were sacrificed.Major signs of colitis including body weight loss, diarrhea and hematochezia were evaluated and scored at 21 days after TNBS instillation.The severity of diarrhea was assessed according to a scale from 0 to 3 (0 = no diarrhea; 1 = mild; 2 = moderate; and 3 = severe diarrhea), and the degree of blood in the stool was evaluated on a scale from 0 to 2 (0 = negative; 1= hemoccult positive; and 2 = gross bleeding).
Twenty-one experimental days after intracolonic TNBS instillation, the animals of each group were sacrificed by cervical dislocation under deep ether anesthesia.
In performing the protocol, the authors adhered to the Israel National Research Council criteria for the care and use of laboratory animals.

experimental design
The following groups of rats were included in the present study: control (no treatment [n=10]) ; TNBS treated (TNBS + vehicle [n=10]); and TNBS + losartan treated (n=10).Losartan or vehicle were administered by oral gavage using a 5 cm catheter (7 mg/kg/day in 2.5 mL of saline, or 2.5 mL saline only, respectively), with daily administration starting on day 0 and continued for 21 days.

sample recovery and preparation
Under ether anesthesia, a laparotomy was performed, the colon was visualized and rapidly excised in its entirety and placed in a petridish containing sterile saline solution.The presence of adhesions between the colon and the adjacent organs was scored on a scale from 0 to 3 (0 = absent; 1= mild/focal; 2 = moderate/zonal; and 3= severe/diffuse) (56).The colon was opened longitudinally, rinsed with sterile saline, weighed, measured and subsequently attached to a wooden tongue depressor for the assessment of macroscopic lesions.It was subsequently cut longitudinally to obtain two equivalent strips.One of the colonic strips was fixed in 4% buffered formaldehyde and embedded in paraffin for histological studies.Sections were cut to a thickness of 5 μm and stained with hematoxylin and eosin (H&E) and Sirius red.The other colonic strip was stored at −80° C for determination of TGF-b1 levels at a later time.
assessment of macroscopic and microscopic colonic lesions Two independent observers (GZ and OP), who were blinded to the treatment groups, assessed and scored the individual macroscopic colonic lesions accrding to the following 0 to 3 scale: colon thickness (0 = normal; 1 = mild increase; 2 = moderate increase; and 3 = marked increased), mucosal edema/hyperemia (0 = absent; 1 = mild; 2 = moderate; and 3 = severe); mucosal ulcers (0 = absent; 1 = single ulceration; 2 = ulcerations at one site; and 3= ulcerations at more sites); and extension of the mucosal lesions (percentage of damaged mucosal surface) (56).The sum of the scores assessing colonic lesions, including adhesions, thickness, edema/hyperemia and ulcers, was expressed as a total macroscopic score.
Large bowel specimens obtained from all animals were washed and immediately immersed in 10% buffered formalin in phosphate-buffered saline (pH 7.4) for 3 h at room temperature followed by the standard procedure for paraffin embedding.Serial 3 μm sections were stained with H&E to assess the degree of inflammation, and with Sirius red to detect connective tissue and fibrosis.The stained sections were then observed under a light microscope (Olympus BX51 [Olympus Optical Co Ltd, Japan]).Two blinded pathologists (GZ and OP) examined and scored all histological sections of colonic samples according to the presence of ulcerations, degree of inflammation, depth of the lesions and fibrosis (56,57).The total microscopic score was obtained by adding the subscores of change in all of the different histological parameters.
The degree of intestinal inflammation was scored as absent, mild, moderate or severe, according to the density and extent of both the acute and chronic inflammatory infiltrate, loss of goblet cells and bowel wall thickening (57).Intestinal fibrosis was scored as mild, moderate or severe, according to the density and extent of Sirius redpositive connective tissue staining and disruption of tissue architecture, compared with the control rats, according to a method previously described (56,57).In this method, the severity of increased connective tissue deposition is assigned a score of 0 (normal) to 3 (the most severe fibrosis, representing progressive increased collagen deposition throughout all layers from mucosa to serosa).

measurement of colonic mucosal tGF-b1 levels
Determination of TGF-b1 levels was performed using ELISA (27).Briefly, tissue was homogenized in the presence of a mixture of protease inhibitors (for the inhibition of serine, cysteine, aspartic proteases and aminopeptidases [1 mL/20 g]) to improve the yield of fully functional proteins.The tissue homogenate was centrifuged at 1600 g at 0°C for 15 min, and the supernatant was transferred to a microfuge tube on ice.The supernatants were purified in 1% trifluoroacetic acid, centrifuged at 17,000 g for 20 min at 4°C, and eluted through a C18 column (Rik-Sepcol [Peninsula Laboratories, USA]) with 100% acetonitrile (Baker, USA) + 1% trifluoroacetic acid.For determination of TGF-b1 levels, an aliquot of the supernatents was treated with 1N HCl to activate TGF-b1, followed by neutralization with 1N NaOH, and a standard ELISA assay was performed (Quantikine [R&D Systems, USA]).

statistical analysis
Data are reported as mean ± SD.For ordinal variables (colorectal adhesions, thickness, mucosal edema/hyperemia, mucosal ulcers and extent of damage, total macroscopic and microscopic scores, degree of fibrosis, as well as diarrhea and hematochezia), results in all groups of rats were analyzed using Kruskal-Wallis nonparametric ANOVA.Post hoc comparisons between pairs of groups were assessed using the Wilcoxon rank-sum test, with a downward adjustment of the a level to compensate for multiple comparisons.To maintain the overall probability at a level of 0.05 in the three independent comparisons, the a value was divided by 3 to obtain a comparison wise a=0.017 (ie, 0.05/3); thus, each comparison was significant at the 0.017 level.
For quantitative variables (body weight, colorectal weight, length and TGF-b1 levels), results in all groups of rats were analyzed using one-way parametric ANOVA.Post hoc comparisons between pairs of groups were assessed using the honest significant differenceTukey's test, with P<0.05 considered to be statistically significant.
Data analysis was performed using SAS version 9.2 (SAS Institute, Inc, USA).

Clinical and macroscopic findings of colitis
The effects of losartan administration on the course and macroscopic findings of TNBS-induced colitis in rats are summarized in Table 1.
On day 21 after TNBS administration, patchy mild/moderate mucosal lesions were observed.However, all rats presented macroscopically visible strictures of the distal colon with proximal dilation, and the colon length was much shorter.The mortality rate of rats in the TNBS and TNBS + losartan groups was three in 10 (30%), and one in 10 (10%), respectively.Losartan administration resulted in a significant improvement in colon weight, length, adhesions, thickness, mucosal edema/hyperemia, ulcerations, and extent of damage.Furthermore, losartan administration improved the macroscopic severity of fibrosis of the colonic wall.The total macroscopic colonic damage score was significantly lower in the TNBS + losartan-treated rats compared with TNBS rats receiving only the vehicle (Figure 1A).
In a group of six rats receiving TNBS without vehicle, clinical, macroscopic and microscopic findings, similar to those observed in mice with TNBS plus vehicle, were found (data not shown).

microscopic findings
A representative microscopic histology panel (H&E and Sirius red staining) of the colon from control, TNBS-treated and TNBS + losartan-treated rats is shown in Figure 2. In the TNBS-treated rats, the affected colonic wall consisted of granulomatous tissue in which fibroblasts and fibrosis were seen in the submucosa together with regenerative changes in the overlying epithelium.The colon of the TNBS + losartan-treated rats showed nearly normal mucosal architecture and minimal granulation tissue and fibrosis in the submucosa (Figure 2).The degree of colonic fibrosis was significantly lower in the TNBS + losartan-treated rats compared with untreated TNBS rats (2.28±0.75versus 0.66±0.50,respectively [P<0.017]) (Figure 3).The total microscopic score was lower in the TNBS + losartan-treated group compared with the TNBS-treated group (Figure 1B).

DisCussion
In IBD, particularly in CD, and in other chronic enteropathies, chronic transmural inflammation triggers the activation of fibrogenetic mesenchymal cells, which leads to abnormal local deposition of ECM proteins, including collagen, and then to fibrosis and strictures (4,5).Despite recent advances in IBD therapy, the development of intestinal fibrosis and strictures remains a challenging complication of these diseases, particularly in CD (5).Existing therapies, predominantly aminosalicylates, steroids, immunomodulators and, more recently, biological therapies, can relieve the inflammatory symptoms

histology (hematoxylin and eosin [H&E], and Sirius red staining) of the colon from control, trinitrobenzene sulphonic acid (TNBS) + vehicle-and TNBS + losartan-treated rats at day 21. H&E staining of the colon shows a normal colonic mucosa in the control group. In the TNBS + vehicle rats, a dense cellular fibrosis was observed in the colonic submucosa together with regenerative changes in the overlying epithelium. The colon of TNBS + losartan-treated rats shows a normal mucosal architecture and a mild fibrosis in the submucosa. Sirius red staining of the colon shows a marked fibrosis in the TNBS + vehicle-treated rats, a mild fibrosis in the TNBS + losartan-treated rats, and a normal finding in the control rats
Figure 3) Degree of colonic fibrosis in control, trinitrobenzene sulphonic acid (TNBS)-treated and TNBS + losartan-treated rats.Values are expressed as mean ± SD. *P<0.017versus controls and TNBS + vehicle, by Wilcoxon rank-sum test after Kruskal-Wallis nonparametric ANOVA of the intestinal fibrostenotic diseases such as CD, but they do not significantly improve stricture lesions of the bowel (2,3).Therefore, intestinal fibrosis and its associated complications remain the major causes of surgical intervention for which there are no approved pharmacological approaches.
Fibrosis is a chronic and progressive process mediated by complex cell/ECM/cytokine and growth factor interactions, but it may be a reversible phenomenon -even in advanced stages (4,5,(10)(11)(12)(13)(14)(15).This has stimulated research on the development of antifibrotic drugs that inhibit the accumulation of activated ECM-producing cells and their profibrogenic actions, or promote ECM degradation.Nevertheless, efficient and well-tolerated antifibrotic drugs are still lacking because the main molecular pathway of fibrogenesis has yet to be elucidated.
In the present study, we showed that the prophylactic oral administration of losartan, an ANGII type 1 receptor antagonist, attenuates the development and progression of colonic fibrosis in TNBS-induced chronic colitis in rats, which appears to be related to the inhibition of the colonic expression of TGF-b1.Daily losartan administration for 21 days significantly decreased the macroscopic and microscopic pattern of both colonic inflammation and fibrosis and re-established normal colonic TGF-b1 levels.The production of TGF-b1 has been shown to be stimulated by the local activation of ANGII, which represents the primary effector molecule of the RAS (32)(33)(34)(35)(36)(37)(38).Renin converts ANG into ANGI, and ACEs convert ANGI into ANGII.To date, four ANG receptors, ANG types 1 to 4 (AT1-4), have been identified.AT1 has further been subdivided into AT1a and AT1b.ANGII binds to both AT1 and AT2.AT1 is responsible for most of the physiological and pathophysiological actions of ANGII.ANG receptor blockers, currently used as antihypertensive drugs, are selective AT1 antagonists.
The reported role of RAS, usually regarded as an endocrine system that regulates physiological and vital cardiovascular processes, has expanded in the past decade to include independently regulated local systems in several tissues, new functions and new active products of ANGII (58,59).It has been reported that all components of the RAS exist in the human large bowel (60).In the colon, many cells such as epithelial, vascular endothelial, mesenchymal and inflammatory cells express ANGII receptors.ANGII participates in the pathogenesis of chronic fibrogenetic diseases of several organs, including the intestine, through the regulation of both inflammatory and fibrotic processes (61,62).The direct effect of ANGII on the pathogenesis of immunemediated colitis was assessed using mice genetically deficient in angiotensinogen (Atg −/− ), which is the precursor of ANGII (63).The degree of TNBS-induced acute colitis was less severe in the Atg −/− mice than that in wild-type (Atg +/+ ) mice.ANGII contributes to the recruitment of infiltrating inflammatory cells into the affected organ by the upregulation of adhesion molecules, chemokines and many proinflammatory cytokines.ACE inhibitors and AT1 receptor antagonists have been shown to diminish inflammatory markers and inflammatory cell infiltration (61)(62)(63).Through the AT1 receptor, ANGII promotes ECM accumulation and fibrosis by the local production of profibrotic molecules such as TGF-b and CTGF.ACE inhibitors and AT1 antagonists reduce tissue fibrosis and TGF-b expression (62).Furthermore, blockade of TGF-b decreases ECM production induced by ANGII.We previously reported that prophylactic oral admistration of the ACE inhibitor captopril prevented the development of colorectal fibrosis in TNBSinduced colitis and reduced colorectal TGF-b levels (27).The same results have been reported with losartan in the present study.The rescue effects of losartan on established colonic fibrosis was not evaluated; however, the therapeutic antifibrotic effect of losartan on established fibrosis in different organs was previously reported (41,45,51).It is worth noting that the beneficial effect of both captopril and losartan in the prevention of colonic fibrosis occurs with a partial downregulation of TGF-b, as reported in other organs (41,42,49).Accordingly, blockade of ANGII reduced renal, cardiac and hepatic fibrosis by inhibiting TGF-b expression by 40% to 60%.This finding could be explained by the different role of TGF-b in the wound healing response.At low femtomolar concentrations, TGF-b1 promotes the healing of acutely damaged tissue, and fibrosis does not occur.In chronic diseases, repeated injury sustains the overproduction of TGF-b1, leading to progressive activation of monocytes that, in turn, release growth factors that stimulate fibroblasts to produce ECM resulting in tissue fibrosis (64).These findings are consistent with the observation that gene transfer of TGF-b1 to the mouse colon leads to intestinal fibrosis (28).On the other hand, suppresing excessive TGF-b1 bioactivity by using a TGF-b1 peptidebased vaccine ameliorates the development of intestinal fibrosis and associated complications in a mouse model of TNBS-induced chronic colitis (65).Nevertheless, although TGF-b is one of the main regulators of fibrosis, therapeutic strategies aimed at blocking TGF-b have not demonstrated the expected beneficial effects, probably due to its anti-inflammatory actions (20,66).For these reasons, novel antifibrotic targets including Smad proteins and CTGF are under consideration.Smad proteins, which are important components of intracellular signalling pathways, act as transcription factors in the majority of TGF-b-mediated responses including fibrosis (17)(18)(19)64).It has been reported that ANGII, through the AT1 receptor, activates the Smad2 and Smad4 signalling system, and Smad-dependent gene transcription independently of TGF-b (62).Smad7 functions as a negative regulator of TGF-b receptor signalling.Disruption of the Smad signalling pathway, either through the loss of Smad3 or the increase of Smad7 expression, confers resistance to tissue fibrosis in several organs including skin, kidney, lung and liver (19,(67)(68)(69)(70)(71)(72)(73)(74)(75).In previous studies, we also recognized the important role of Smad proteins in the development of intestinal fibrosis, and demonstrated that the beneficial effects of antifibrotics is associated with a downregulation of both Smad and CTGF proteins (56,76).CTGF is a pivotal profibrotic factor that is upregulated in the fibrotic process of several organs (30,31,62).CTGF is a potent stimulator of activated myofibroblasts, and ECM production and deposition.ANGII, via AT1, upregulates CTGF production and fibrosis by the activation of the Smad signalling system and the Rho/Rho kinase pathway independendly of TGF-b (62,77).ACE inhibitors and AT1 antagonists reduced CTGF expression and fibrosis.Furthemore, blockade of CTGF reduced ANGII-induced fibrosis.Collectively, these data suggest that drugs that inhibit RAS and, in particular, AT1 antagonists such as losartan, could lead to a selective blockade of intracellular fibrogenic TGF-b-mediated transduction signalling pathways without involving physiologically important -or even vital -TGF-b activities.Additional studies are necessary to evaluate their potential therapeutic effects in fibrostenosing chronic enteropathies, particularly CD.The association of the angiotensinogen-6 variant with CD supports a potential role for ACE inhibitors and ANGII receptor antagonists in the treatment of this disease (78). in memoriam: All authors will remember with great esteem and affection Giuliana Zanninelli, who suddenly, and at an early age, departed this world.Giuliana was the most brilliant member of the research group and will be sadly missed by all who knew her.

ConCLusion
The present study demonstrated that the prophylactic oral administration of losartan, an ANGII receptor antagonist, attenuates colorectal fibrosis in experimental TNBS-induced colitis.This effect of losartan appears to be mediated by a downregulation of TGF-b1 expression.An indirect antifibrotic effect of losartan may also persist through its anti-inflammatory action.Thus, losartan may represent an effective prophylactic agent in reducing the risk of intestinal fibrosis and strictures in patients with CD.

Figure 1 )Figure 2 )
Figure 1) A Total macroscopic colonic damage score at day 21 in control, trinitrobenzene sulphonic acid (TNBS) + vehicle-and TNBS + losartan-treated rats.Values are expressed as mean ± SD. *P<0.017versus controls and TNBS + vehicle, by Wilcoxon rank-sum test after Kruskal-Wallis nonparametric ANOVA.B Total microscopic colonic damage score at day 21 in control, TNBS + vehicle and TNBS + losartan-treated rats.Values are expressed as mean ± SD. *P<0.017versus controls and TNBS + vehicle, by Wilcoxon rank-sum test after Kruskal-Wallis nonparametric ANOVA