The present protocol describes the development of a Crohn’s-like colitis model in rodents. Transmural inflammation leads to stenosis at the TNBS instillation site, and mechanical enlargement is observed in the segment proximal to the stenosis. These changes allow studying mechanical stress in colitis.
Inflammatory bowel diseases (IBD) such as Crohn's disease (CD) are chronic inflammatory disorders of the gastrointestinal tract affecting approximately 20 per 1,000,000 in Europe and USA. CD is characterized by transmural inflammation, intestinal fibrosis, and luminal stenosis. Although anti-inflammatory therapies may help control inflammation, they have no efficacy on fibrosis and stenosis in CD. The pathogenesis of CD is not well understood. Current studies focus mainly on delineating dysregulated gut immune response mechanisms. While CD-associated transmural inflammation, intestinal fibrosis, and luminal stenosis all represent mechanical stress to the gut wall, the role of mechanical stress in CD is not well defined. To determine if mechanical stress plays an independent pathogenic role in CD, a protocol of TNBS-induced CD-like colitis model in rodents has been developed. This TNBS-induced transmural inflammation and fibrosis model resembles pathological hallmarks of CD in the colon. It is induced by intracolonic instillation of TNBS into the distal colon of adult Sprague-Dawley rats. In this model, transmural inflammation leads to stenosis at the TNBS instillation site (Site I). Mechanical distention is observed in the portion proximal to the instillation site (Site P), representing mechanical stress but not visible inflammation. Colonic portion distal to inflammation (Site D) presents neither inflammation nor mechanical stress. Distinctive changes of gene expression, immune response, fibrosis, and smooth muscle growth at different sites (P, I, and D) were observed, highlighting a profound impact of mechanical stress. Therefore, this model of CD-like colitis will help us better understand CD's pathogenic mechanisms, particularly the role of mechanical stress and mechanical stress-induced gene expression in immune dysregulation, intestinal fibrosis, and tissue remodeling in CD.
Inflammatory bowel disease (IBD), including ulcerative colitis (UC) and Crohn's disease (CD), is characterized by chronic inflammation in the gastrointestinal (GI) tract. It affects ~1-2 million Americans1. The estimated annual costs for IBD care in the US are $11.8 billion. Unlike UC, the CD is characterized by transmural inflammation and stricture formation2,3. Stricture formation (stenosis) occurs in up to 70% of CD patients3 and may be caused by transmural inflammation (inflammatory stenosis) or intestinal fibrosis (fibrotic stenosis)4,5. Intestinal fibrosis is characterized by excessive collagen deposition and other extracellular matrices (ECM) with smooth muscle cells (SMC) as one of the main mesenchymal cell types involved in the process3,4. Smooth muscle hyperplasia associated with hypertrophy is another significant histological change in fibrotic stenosis in CD6. Although stricture formation in CD is associated with chronic inflammation, no anti-inflammatory treatment is effective, except surgical treatment2,6. However, post-surgery recurrences are almost 100%, given sufficient time2,7. As an inflammatory response, fibrosis and SMC hyperplasia may also develop in non-inflammatory conditions (i.e., bowel obstruction) in the gut8,9; it is believed that both inflammation-dependent and independent mechanisms are involved in stricture formation3,4. Given that extensive research into the inflammation-dependent mechanisms has not translated into any effective therapy for stricture formation, studies into the possible role of inflammation-independent mechanisms in intestinal fibrosis are needed.
As a non-inflammatory factor, mechanical stress (MS) associated with edema, inflammatory cell infiltration, tissue deformation, fibrosis, and stenosis10,11,12,13 is commonly encountered in IBD, especially CD, which is characterized by transmural inflammation. Mechanical stress is most remarkable in stenotic CD, where stenosis (inflammatory or fibrotic) in the inflammation site presents mechanical stress in the local tissue and leads to lumen distention in the segment proximal to the obstruction site10,14. Previous in vitro studies have demonstrated that mechanical stress alters gene expression of specific inflammatory mediators (i.e., COX-2, IL-6)8,14,15 and growth factors (i.e., TGF-β) in the gastrointestinal tissues, especially gut smooth muscle cells (SMC)16. Recent studies also found that the expression of specific pro-fibrotic mediators such as connective tissue growth factor (CTGF) is highly sensitive to mechanical stress17,18. It was hypothesized that mechanical stress might play an independent pathogenic role in CD-associated inflammation, fibrosis, and tissue remodeling. However, the pathogenic significance of mechanical stress in gut inflammation, fibrosis, and smooth muscle hyperplasia in CD remains largely unexplored. This may be partly because inflammation is a more visible and better-studied process than mechanical stress. More importantly, there has been no well-defined animal model of IBD to distinguish the effect of mechanical stress from that of inflammation.
The current work describes a rodent model of Crohn's-like colitis induced by intracolonic injection of hapten reagent 2,4,6-trinitrobenzene sulfonic acid (TNBS)19,20, which may serve the purpose to study the role of mechanical stress in CD. It was found that TNBS instillation induced a localized (~2 cm in length) transmural inflammation with lumen narrowing (stenosis) in the distal colon. The stenosis leads to marked bowel distention (mechanical stress)14,15 but not visible inflammation in the colonic segment proximal to the instillation site. On the contrary, the colon segment distal to the stenosis site presents neither inflammation nor mechanical stress. Significant site-specific changes in gene expression, inflammation, fibrosis, and SMC hyperplasia were observed in the three different sites. The results suggest that mechanical stress, particularly mechanical stress-induced gene expression, may play a critical role in developing fibrosis and hyperplasia in Crohn's colitis.
All animal experiments were conducted according to the institutional animal care and use committee of the University of Texas Medical Branch (#0907051C). Male or female Sprague-Dawley rats, ~8-9 weeks old, were used for the study.
1. Animal preparation
2. Tissue preparations
3. Histopathologic assessment of gut inflammation and fibrosis
4. RNA extraction and quantitative RT-PCR
5. Statistical analysis
Macroscopic view of Crohn's-like colitis induced by intra-colonic instillation of TNBS
As shown in Figure 1, intracolonic instillation of TNBS in rats induced a localized transmural inflammation (~2 cm in length) with thickened bowel wall and narrowed lumen (stenosis) in the site of instillation in the distal colon (Figure 1A). The site of TNBS instillation is referred to as site I. As a result of transmural inflammation and stenosis, both inflammation and mechanical stress are present in site I. The stenosis in site I led to marked lumen distention in the segment proximal to the site of TNBS instillation (site P) (Figure 1). The colon circumference was significantly increased in sites P and I, compared to sham control colon (p < 0.05 vs. control) (Figure 1B). While mechanically distended, site P did not show visible inflammation. On the contrary, the colonic segment distal to the TNBS instillation site is site D and presents neither inflammation nor mechanical distention (Figure 1A,B).
To help distinguish the effect of mechanical stress from inflammation, we followed a unique design in collecting tissue samples from the colon in the model described in step 2.6. Site P is the primary focus of the study, as this part is mechanically distended in the CD model. Site D is self-control, as it does not present mechanical stress. Sites P and D do not demonstrate any visible inflammation (Figure 1). However, the site I experience inflammation and mechanical stress (Figure 1).
Site-specific changes of inflammation score in sites P, I, and D in colitis rats
Criteria were developed in grading inflammation based on the macroscopic score of live tissue (0-3)19 and the microscopic score of H&E stained specimens (0-3)23 as described with modifications. The results showed that the macroscopic score of inflammation in site I was 2.70 ± 0.20 in the TNBS treated rats (7 days after induction of inflammation), dramatically increased compared to that of sham controls (0.30 ± 0.22, p < 0.05) and that of sites P (0.80 ± 0.26) and D (0.50 ± 0.22) of colitis rats (Figure 1C). The inflammation scores in sites P and D are not significantly increased compared to sham (Figure 1C). Microscopic imaging showed TNBS treatment-induced transmural inflammation in rats (Figure 2A). The microscopic score of inflammation in site I was 2.80 ± 0.27 in the TNBS treated rats, again significantly (p < 0.05, n = 5 each group) increased compared to that in sham controls (0.3 ± 0.2) and in sites P (1.0 ± 0.31) and D (0.80 ± 0.38) of colitis rats. The inflammation scores in sites P and D were not significantly increased compared to sham (Figure 2C).
Site-specific changes of fibrosis and smooth muscle hyperplasia and hypertrophy in sites P, I, and D in colitis rats
The fibrosis score was determined based on Mason's trichrome stain (Figure 2B) in different sites (P, I, D) (Figure 2A). The grading system for fibrosis is described in Supplementary File 1. It was found that fibrosis score is significantly increased not only in site I (2.60 ± 0.25) but also in site P (1.60 ± 0.24) of the colitis rats, compared to sham control (0.40 ± 0.25. p < 0.05) (Figure 2D). The thickness and cell numbers of circular and longitudinal smooth muscle layers were measured in different sites in H&E stained specimens (4 views per specimen). The thickness and cell numbers of both circular and longitudinal smooth muscle layers were significantly increased in sites I and P (Figure 2E,F). The site D in the colitis rats does not show any significant increase in fibrosis score, smooth muscle cell numbers, or muscle thickness (Figure 2).
Site-specific expression of mechano-sensitive genes in sites P, I, and D in colitis rats
IL-6 plays a critical role in gut inflammation, as it promotes T cell differentiation, damages barrier function, and affects neuromuscular function8,24. CTGF is a well-recognized pro-fibrotic mediator, as its inhibition can reverse the process of fibrosis17. More importantly, recent studies found that gene expression of IL-6 and CTGF is highly responsive to mechanical stress8,14,18. The site-specific expression of IL-6 and CTGF mRNAs were determined in the full thickness tissue of sham control and TNBS colitis rats. The mRNA expression levels of IL-6 and CTGF were significantly increased in the inflammation site (site I) compared to sham controls. In site P, where there is mechanical distention but no visible inflammation, the mRNA expression of IL-6 and CTGF was also dramatically increased compared to control rats (Figure 3A,B). However, IL-6 and CTGF mRNA levels in site D of colitis rats were not significantly different from that in sham controls (Figure 3).
Figure 1: Rodent model of TNBS-induced CD-like colitis (7 day). (A) Outlook view of sham control and TNBS-treated colon (top) and macroscopic view of the mucosal surface of the distal colon (bottom). The yellow boxes indicate different sites of colon tissue. S, sham control; I, inflammation site; P, distended colon site proximal to inflammation site; D, non-distended site distal to inflammation. (B) Colon circumference in sham control and different sites (P, I, D) of TNBS-treated colon. (C) Macroscopic inflammation score of sham control colon and different sites of TNBS-treated colon. n = 5, *p < 0.05 vs. sham rats of the group. Bars represent SEM. Please click here to view a larger version of this figure.
Figure 2: Histopathologic assessment. Microscopic views in H&E (A) and Masson's trichrome stains (B) showing collagen distribution in sham and different sites of colitis rats. Quantitative analysis shows increased microscopic inflammation index (C), fibrosis (D) and muscular thickness (hypertrophy) (E), and smooth muscle cell number (hyperplasia) (F) in sites I and P, but not D, in TNBS treated rats. In (E) and (F), the open bars are for the circular smooth muscle layer, and streaked bars are for the longitudinal smooth muscle layer. n = 5 in each group, *p < 0.05 vs. sham rats of the group. Bars in graphs represent SEM. Bars in (A) and (B) = 100 µm. Please click here to view a larger version of this figure.
Figure 3: Site-specific expression of mechanosensitive genes (IL-6 and CTGF) in CD-like colitis. (A) Expression of IL-6 mRNA in sham control colon and different sites (P, I, and D) of TNBS-treated colitis colon. (B) Expression of CTGF mRNA in sham control colon and different sites (P, I, and D) of TNBS-treated colitis colon. n = 4 or 5, *p < 0.05 vs. S (sham control). Bars represent SEM. Please click here to view a larger version of this figure.
Supplementary File 1: Score of inflammation and fibrosis. Please click here to download this File.
TNBS-induced colitis was introduced in 1989 and has been used as an experimental model of Crohn's disease since then19,20,23. Significant features of this model in rodents include the development of a transmural inflammation that closely resembles the histopathological lesions developed in human Crohn's disease19,20. Previous studies on the model have focused mainly on aberrant immune response in the mucosa layer in the site of visible inflammation (the site I)19,20,23. Little attention has been paid to the bowel portions proximal and distal to inflammation. The present study on the inflammation site, as well as the distended proximal segment and the non-distended distal segment, reveals apparent site-specific changes in gene expression, inflammatory response, and histopathological features. The model has been revisited to address the potential pathogenic role of mechanical stress in fibrosis and tissue remodeling in Crohn's disease.
It was found that acute inflammation is immediately developed after mucosal exposure of TNBS in ethanol and is peaked at day 319. Transmural inflammation and inflammatory stenosis are present, which is associated with lumen distention in site P in every rat treated with TNBS. By day 7, chronic transmural inflammation is well developed in site I, as found in the current study and reported elsewhere20. Meanwhile, fibrotic changes are apparent in the site, characterized by excessive collagen deposition, as seen in the present study and elsewhere20. TNBS instillation in ethanol in the distal colon injures the local mucosa tissue23 and leads to transmural inflammation in a localized area in the colon. The transmural inflammation with inflammatory infiltration, edema, and tissue deformation10,12,13 in the instillation site present inflammation and mechanical stress14 to the localized area (Site I). Moreover, transmural inflammation also causes luminal stenosis in site I10. It was found that stenosis, a partial bowel obstruction, causes mechanical distention in the proximal segment (site P). However, the part distal to the instillation site is not distended (site D). As found in the macroscopic and histological scoring systems, while site I presents both inflammation and mechanical stress, sites P and D do not present inflammation. Furthermore, site P shows significantly increased circumference (and thus mechanical stress, according to the law of Laplace14), but site D does not. Therefore, a study on the site-specific changes, especially the site P, will explore the pathogenic importance of mechanical stress in the inflammation model in the gut.
It was observed that the expression of mechano-sensitive genes IL-6 and CTGF was increased in site I and in mechanically stretched site P, but not in neighboring site D, where there is no mechanical distention. To examine the hypothesis that mechanical stress may contribute to the development of fibrosis and smooth muscle hyperplasia, fibrosis scores and measured smooth muscle cell numbers and muscle thickness in sites P, I, and D were determined separately. It was found that fibrosis and SMC hyperplasia are present in sites I and P. However, fibrosis and smooth muscle hyperplasia are not detected in site D. These findings indicate that mechanical stress may play an independent pathogenic role in collagen synthesis and cell proliferation in gut inflammation. Further studies are warranted to determine if this effect may be mediated by mechanical stress-induced expression of pro-fibrotic and growth factors such as CTGF in the sites P and I.
Although the described animal model itself can be used to address mechanical stress in stenotic CD-like colitis, it has limitations in the long-term goal to fully define the pathogenic role of mechanical stress in inflammation. For example, the effects of mechanical stress and inflammation in site I of the described model cannot be differentiated. Although it is assumed that mechanical stress is present in site I because of inflammatory infiltration, tissue deformation, stenosis, and distention, how much mechanical stress contributes to the pathological changes there cannot be determined. Further comprehensive studies using in vitro, in vivo, and ex vivo approaches may be needed. For example, preventing mechanical distention by feeding the colitis animals exclusively with a clear liquid diet25 may help create a loss-of-mechanical distention status in the colitis model. On the other hand, induction of pure mechanical distention by obstruction band15 may help create a gain-of-mechanical stress model. Moreover, the in vitro mechanical stretch model in cultured cells14,15 assists in the quantitative determination of the effects of mechanical stress on gene expression and function, as the mode and extent of mechanical stress can be finely controlled in the in vitro setting.
To prepare a reproducible model of transmural and stenotic inflammation in CD-like colitis as described in the study, TNBS at 65 mg/kg in 250 µL of 40% ethanol must be used. The model was tested mainly in rats, male or female, 8-9 weeks old. After instillation of TNBS at the dose, transmural inflammation is consistently developed at the local instillation site. The inflammation is associated with inflammatory cells infiltration, edema, and bowel wall thickening, leading to lumen narrowing at the instillation site, resembling pathological characteristics of Crohn's disease20. Pilot studies in the lab showed that TNBS at doses lower than 50 mg/kg in the same volume of 40% ethanol might cause gut inflammation but not reliable stenosis in site I. Thus, there would be no apparent mechanical distention in site P. On the other hand, TNBS at doses greater than 80 mg/kg may cause severe inflammation and fatalities. The current protocol using TNBS at 65 mg/kg in 250 µL of 40% ethanol hardly causes fatalities (1 in 16 rats of TNBS colitis).
It was found that bowel cleansing the day before instillation of TNBS is an important step to ensure a relatively clean colon for a reliable model of stenotic colitis. For that, rats need to fast for 24 h and give bowel cleanser overnight before TNBS treatment. It is also important to keep rats in a supine and slightly head-down position with the anus closed for 2 min after instillation of TNBS. This helps to ensure a good distribution of TNBS inside the distal colon.
In summary, it was found that intracolonic instillation of TNBS at 65 mg/kg in 250 µL of 40% ethanol consistently induced CD-like colitis in rats. Transmural inflammation in the model is associated with stenosis at the TNBS instillation site. Mechanical distention, but not inflammation, is observed in the segment proximal to the stenosis. Neither inflammation nor mechanical stress is present in the segment distal to the stenosis. With these changes in different colon sites in a colitis rat, mechanical stress from inflammation in CD-like colitis can be distinguished.
The authors have nothing to disclose.
This work is supported in part by grants from NIH (R01 DK124611 to XZS) and the US Department of Defense (W81XWH-20-1-0681 to XZS). The histology work was done with the help of the UTMB Surgical Pathology Lab.
ACT-1 Control Software Ver2.63 | Nikon | DXM1200F | |
C1000 Touch Thermal Cycler with 96-Well Fast Reaction Module | BIO-RAD | 1851196 | |
CFX96 Optical Reaction Module for Real-Time PCR Systems | BIO-RAD | 1845097 | |
Dako Agilent Artisan Link Pro Special stainer | Dako | AR310 | |
Dako-Agilent Masson's Trichrome Kit ref# AR173 | Dako | AR173 | |
DXM1200 Digital Color HR Camera | Nikon | DXM1200 | |
Eukaryotic 18S rRNA Endogenous Control | ThermoFisher Scientific | 4352930E | |
E-Z Anesthesia | E-Z Systems Inc. | EZ-155 | |
GraphPad Prism 9 | GraphPad | 9.0.2 (161) | |
Hard-Shell 96-Well PCR Plates, low profile, thin wall, skirted, white/clear | BIO-RAD | HSP9601 | |
HBSS (Corning Hank's Balanced Salt Solution, 1x without calcium and magnesium) | CORNING | 21-021-CV | |
HM 325 Microtome | Thermo Scientific | 23-900-667 | |
Isoflurane | Piramal | NDC 66794-017-10 | |
LI-COR Odyssey Digital Imaging System | LI-COR | 9120 | |
Mastercycler epGradient Thermal Cycler with Control Panel 5340 Thermal Cycler | Eppendorf | 5341 | |
Medical grade open end polyurethane catheter | Covidien | 8890703013 | |
NanoDrop 2000/2000c Spectrophotometers | Thermo Fisher Scientific | ND2000CLAPTOP | |
Nikon Eclipse E800 Upright Microscope | Nikon | E800 | |
Nitrocellulose/Filter Paper Sandwiches Pkg of 50, 0.45 μm, 7 x 8.5 cm | BIO-RAD | 1620215 | |
Polyethylene Glycol 3350, Osmotic Laxative | Miralax | C8175 | Dose: 17g in 226 mL of water |
RNeasy Mini Kit (250) 250 RNeasy Mini Spin Columns, Collection Tubes (1.5 mL and 2 mL), RNase-free Reagents and Buffers |
QIAGEN | 74106 | |
SuperScript III First-Strand Synthesis System | ThermoFisher Scientific | 18080051 | |
TaqMan Gene Expression Assays Rn00573960_g1 CTGF Probe | ThermoFisher Scientific | 4331182 | |
TaqMan Gene Expression Assays Rn99999011_m1 IL6 Probe | ThermoFisher Scientific | 4331182 | |
TaqMan Fast Advanced Master Mix | ThermoFisher Scientific | 4444557 | |
Tissue-Tek Prisma H&E Stain Kit #1 | Sakura | 6190 | |
Tissue-Tek Prisma Plus Automated Slide Stainer | Sakura | 6171 | |
TNBS (Picrylsulfonic acid solution) | SIGMA-ALDRICH | 92822 |