The role of interleukin 17 in Crohn’s disease-associated intestinal fibrosis

We analyzed IL-17A and IL-17E levels by immunoblotting in tissue samples collected from uninflamed areas of strictured and non-strictured gut of 12 patients with fibrostenosing CD and from normal gut of 11 control subjects. IL-17A expression was significantly (P<0.005) upregulated in strictured CD areas compared with non-strictured CD areas and control gut (Figure 1A). No significant difference in IL-17A expression was found between non-strictured CD areas and control gut. IL-17E levels did not differ significantly between strictured and non-strictured CD areas and control gut.

In parallel, we analyzed IL-17A and IL-17E expression by ELISA in the same samples. IL-17A was significantly upregulated in strictured CD (mean 98.5 ± 13.7 pg/μg total protein) compared with non-strictured CD (mean 42.3 ± 11.0 pg/μg total protein, P<0.005) and control (mean 46.5 ± 8.3 pg/μg total protein, P<0.005) areas (Figure 1B). There was no significant difference in IL-17A expression between non-strictured CD areas and control gut. IL-17E did not significantly differ between strictured CD (mean 222.5 ± 32.4 pg/μg total protein), non-strictured CD (mean 178.1 ± 30.3 pg/μg total protein), and control areas (mean 188.2 ± 35.3 pg/μg total protein).

We then analyzed IL-17RC (receptor of IL-17A) and IL-17RB (receptor of IL-17E) expression by immunoblotting in the same tissue samples. IL-17RC expression was not significantly different between CD strictured, CD non-strictured, and control gut (Figure 2A). Similarly, IL-17RB expression did not differ between CD strictured, CD non-strictured and control gut (Figure 2B).

We cultured ex vivo for 24 hours tissue explants (1 mm³ in size) from uninflamed areas of strictured and non-strictured gut of six patients with fibrostenosing CD and from normal gut of seven control subjects, and measured IL-17A, IL-17E, IL-6, TNF-α, and collagen concentration in the culture supernatant and TGF-β1 transcripts in the cultured tissue (Figure 3). IL-17A was significantly higher in the organ culture supernatants of strictured CD (mean 110.0 ± 23.8 pg/ml) than in those of non-strictured CD (mean 55.9 ± 8.8 pg/ml, P<0.05) or normal gut (mean 51.1 ± 8.9 pg/ml, P<0.02). Concentrations of IL-17E, IL-6, and TNF-α did not differ significantly between the supernatants of strictured CD (mean 653.3 ± 283.5 pg/ml, 26,553 ± 6647 pg/ml, and 43.4 ± 7.6 pg/ml, respectively), non-strictured CD (mean 1,028.0 ± 202.5 pg/ml, 31,740 ± 3944 pg/ml, and 40.8 ± 7.7 pg/ml, respectively), and normal gut (mean 1068.0 ± 282.8 pg/ml, 36,784 ± 2516 pg/ml, and 30.0 ± 7.9 pg/ml, respectively). Collagen concentration was significantly higher in the supernatants of strictured CD (mean 979.0 ± 108.6 μg/ml, respectively) than in those of non-strictured CD (mean 604.6 ± 97.5 μg/ml, P<0.05) and normal gut areas (mean 523.6 ± 77.9 μg/ml, P=0.005). TGF-β1 transcripts were significantly higher in the cultured tissue explants from strictured CD than in those from non-strictured CD (P=0.0005) and normal gut (P=0.001).

We then analyzed IL-17RC and IL-17RB expression by immunoblotting in lysates of myofibroblasts isolated from uninflamed areas of strictured and non-strictured gut of six patients with fibrostenosing CD, and from normal gut of seven control subjects. Myofibroblasts isolated from strictured CD areas, non-strictured CD areas, and control gut expressed both IL-17RC (Figure 4A) and IL-17RB (Figure 4B). However, no significant difference in IL-17RC and IL-17RB expression was found between strictured, non-strictured CD, and control myofibroblasts.

Matrix metalloproteinase (MMP)-3, MMP-12, and TIMP-1 production was evaluated by immunoblotting in culture supernatants of myofibroblasts isolated from uninflamed areas of strictured or non-strictured gut of six patients with fibrostenosing CD, and from normal gut of six control subjects, and subsequently stimulated in vitro with recombinant human (rh)TNF-α, rhIL-17A, or rhIL-17E. rhIL-17A and rhIL-17E induced a significant (P<0.05) increase in both MMP-3 and MMP-12 production by myofibroblasts from strictured and non-strictured areas of patients with CD and from normal gut of control subjects (Figure 5A,B). rhTNF-α induced a significant (P<0.05) increase in both MMP-3 and MMP-12 production compared with control (myofibroblasts from the same group of patients cultured with medium alone). No significant difference in MMP-3 or MMP-12 production was found between rhTNF-α, rhIL-17A, and rhIL-17E stimulation in the three conditions studied. Myofibroblasts from CD strictured areas showed a significantly (P<0.05) lower spontaneous release of MMP-12 than myofibroblasts from CD non-strictured areas or control gut. Myofibroblast stimulation with rhIL-17A induced a significant (P<0.05) increase in TIMP-1 production compared with unstimulated cells from the same group of patients (Figure 5C). rhIL-17E did not induce any significant change in TIMP-1 production compared with myofibroblasts from the same group of patients cultured with medium alone. rhTNF-α induced a significant (P<0.05) increase in TIMP-1 production compared with myofibroblasts from the same group of patients cultured with medium alone. No significant difference in TIMP-1 production was found between rhTNF-α and rhIL-17A stimulation in all the three conditions studied.

We cultured myofibroblasts isolated from uninflamed areas of strictured and non-strictured gut of six patients with fibrostenosing CD, and from normal gut of six control subjects, in the presence or absence of rhTNF-α, rhIL-17A, or rhIL-17E, and measured collagen in the culture supernatants (Figure 6). Myofibroblast stimulation with rhIL-17A induced a significant (P<0.05) increase in mean collagen production (strictured CD myofibroblasts: 412 ± 57 μg/ml; non-strictured CD myofibroblasts: 154 ± 17 μg/ml; control myofibroblasts: 114 ± 12 μg/ml) compared with cells from the same group of patients cultured with medium alone (strictured CD myofibroblasts: 155 ± 23 μg/ml; non-strictured CD myofibroblasts: 76 ± 15 μg/ml; control myofibroblasts: 67 ± 7 μg/ml). rhIL-17E did not induce any significant change in mean collagen production by strictured CD myofibroblasts (181 ± 37 μg/ml), non-strictured CD myofibroblasts (83 ± 7 μg/ml) or control myofibroblasts (61 ± 10 μg/ml) compared with unstimulated cells from the same group of patients. rhTNF-α induced a significant (P<0.05) increase in mean collagen production by strictured CD myofibroblasts (399 ± 62 μg/ml), non-strictured CD myofibroblasts (175 ± 22 μg/ml) and control myofibroblasts (124 ± 18 μg/ml) compared with unstimulated cells from the same group of patients. No significant difference in collagen production was found between cells cultured with rhTNF-α and rhIL-17A. Strictured CD myofibroblasts produced significantly (P<0.05) higher amounts of collagen than did non-strictured CD and control myofibroblasts cultured under the same conditions. No significant difference in collagen production was found between non-strictured CD and control myofibroblasts cultured under the same conditions.

To evaluate the role of IL-17A and IL-17E on myofibroblast migration, a wound-healing scratch assay was performed using subconfluent monolayers of myofibroblasts isolated from uninflamed areas of strictured and non-strictured gut of six patients with fibrostenosing CD, and from normal gut of seven control subjects. Cell migration was measured as the percentage of wound repair and results were expressed as mean percentage of wound repair (see Methods section). rhIL-17A significantly (P<0.05) reduced migration of strictured CD, non-strictured CD and control myofibroblasts at 8 hours (3.2 ± 1.1%, 4.6 ± 1.0%, and 6.4 ± 1.6%, respectively), 16 hours (5.8 ± 1.7%, 8.4 ± 1.8%, and 10.6 ± 1.9%, respectively) and 24 hours (8.8 ± 1.9%, 23.2 ± 2.3%, and 29.7 ± 2.0%, respectively) compared with myofibroblasts from the same three groups cultured with medium alone and evaluated at the same time points (8 hours 9.8 ± 1.3%, 13.7 ± 1.8%, and 14.6 ± 1.7%, respectively; 16 hours: 15.6 ± 1.6%, 25.3 ± 2.3%, and 20.3 ± 1.1%, respectively; and 24 hours: 30.6 ± 2.3%, 49.1 ± 1.9%,and 48.4 ± 1.8%, respectively) (Figure 7). Compared with medium alone, rhIL-17E did not induce any significant change in strictured CD, non-strictured CD, and control myofibroblast migration (8 hours: 11.2 ± 1.2%, 15.4 ± 1.4%, 12.9 ± 1.5%, respectively; 16 hours: 16.2 ± 1.4%, 28.2 ± 1.7%, 25.1 ± 1.1%, respectively; and 24 hours: 26.3 ± 1.9%, 41.3 ± 2.2%, 43.4 ± 2.2%, respectively).

This study was approved by the local Ethics Committee (NRES Committee London - City & East). Each patient who took part in the study gave informed consent to participate in the study and to publish the results deriving from this research. All the experiments reported in this study are in compliance with the Helsinki Declaration.

Surgical specimens were taken from uninflamed areas of strictured and non-strictured ileum or colon of 29 patients with fibrostenosing CD (Table 1). Because the excessive collagen deposition occurs in the sub-mucosa and outer muscle layers, we discarded the mucosa upon dissection, and used the muscle-enriched fraction. Diagnosis of CD was ascertained by standard clinical criteria [34], and strictured areas were identified by enteroclysis [35]. None of the patients with CD had been treated previously with ciclosporin, tacrolimus, methotrexate, or anti-TNF-α antibodies. Intestinal samples were also collected from macroscopically and microscopically unaffected (at least 500 mm from the tumoral mass) ileum or colon of 27 patients undergoing intestinal resection for colon cancer (mean age 57 years, range 45–71 years). The intestinal tissue was homogenized and used in immunoblotting experiments, or processed to isolate myofibroblasts, or cultured ex vivo.

Intestinal tissue explants (1 mm³ in size) were placed in 12-well tissue culture plates (BD Biosciences, Oxford, UK; one explant per well) and cultured at 37°C and 5%CO₂ in 800 μl serum-free HL-1 medium (Cambrex Bio Science, Wokingham, UK) supplemented with 100 U/ml penicillin and 100 μg/ml streptomycin, as previously described [36]. After 24 hours ex vivo culture, supernatants were collected and stored at −70°C until used for cytokine measurement by ELISA, and intestinal explants were homogenized for subsequent evaluation of TGF-β1 expression by quantitative reverse transcription (qRT)-PCR.

The mucosa was dissected away from full thickness samples of bowel and discarded. The remaining tissue was cut into 1 mm-sized cubes and cultured at 37°C in a humidified CO₂ incubator in Dulbecco’s modified Eagle’s medium (DMEM; Sigma-Aldrich, Poole, UK) supplemented with 20% fetal bovine serum (FBS), 1% non-essential amino acids (Invitrogen Ltd., Paisley, UK), 100 U/ml penicillin, 100 μg/ml streptomycin, 50 μg/ml gentamycin, and 1 μg/ml amphotericin (Sigma-Aldrich). Myofibroblasts migrated out of the tissue after a few days. Established colonies of myofibroblasts were seeded into 25-cm² culture flasks and cultured in DMEM supplemented with 20% FBS and antibiotics. At confluence, the cells were passaged using trypsin-EDTA in a 1:2 to 1:3 split ratio. Cells were grown to at least passage 4 before there were enough to use in stimulation experiments, and were characterized by immunocytochemical staining, as previously described [37]. Subconfluent monolayers of myofibroblasts seeded in 12-well plates at 3×10⁵ cells per well were starved in serum-free medium for 24 hours at 37°C and 5%CO₂ before culture for 24 hours at 37°C and 5%CO₂ with serum-free DMEM containing antibiotics in the absence or presence of 10 ng/ml of rhIL-17A, or rhIL-17E, or rhTNF-α (all R&D Systems, Abingdon, Oxfordshire, UK).

Myofibroblast migration was assessed in accordance with the modified method of Rodriguez et al. [16, 38]. Briefly, cells (2×10⁵) were seeded into cell culture dishes (Nunc; Nalge Nunc International, Rochester, NY, USA) with 2 mm grids, size 35×10 mm, in 2 ml of DMEM supplemented with 20% FBS and antibiotics. The cells were maintained at 37°C and 5% CO₂ until confluent. Once confluent, each dish of monolayer cells was given a mechanical wound by scoring with a 200 μl pipette tip, parallel to the grid bars along the central grid line. This permitted easy viewing of the cells growing back together, and ensured that the 2 mm grid could be used as a reference, so that the wound areas could be measured and compared. Wound placement was checked with an inverted microscope (CK2; Olympus UK Ltd, London, UK). The medium was then removed, and the cells were washed five times with HL-1 serum-free medium (Cambrex Bio Science) supplemented with antibiotics, and then replaced with 1.5 ml HL-1 medium with or without 10 ng/ml rhIL-17A or rhIL-17E (both R&D Systems). Photographs of the cells in each grid along the induced wound were taken at 0, 2, 4, 8, 16 and 24 hours, using a digital camera (Camedia; Olympus UK Ltd) with 34 to 40 zoom, and 20× magnification, attached to a light microscope. The computer program Image J was used to measure the area of initial damage (images taken at time 0) and of the remaining damage at subsequent time points. Each grid image was observed separately, and two points per grid at the same position at every time point were measured using imaging software at the same magnification. The percentage of wound repair was then calculated.

Concentrations of IL-17A and IL-17E in tissue sample homogenates were measured using specific ELISA kits (R&D Systems), in accordance with the manufacturer’s instructions, and were normalized to the total protein concentration, determined by a protein assay (Bio-Rad Laboratories, Hemel Hempstead, UK). Concentrations of IL-17A, IL-17E, IL-6, and TNF-α in organ-culture supernatants were assessed using specific ELISA kits (R&D Systems), in accordance with the manufacturer’s instructions.

RNA was extracted from cultured tissue explants. cDNAs were synthesized (Improm-II RT system; Promega, Southampton, Hampshire, UK) using random hexamers and 1 μg of RNA in a final volume of 20 μl. RT reactions were performed using the Improm-II reverse transcriptase enzyme from the kit. An RT reaction without reverse transcriptase enzyme was performed for each tissue type as a negative control for qPCR. TGF-β1 primers and probe sets were validated for use with the ^∆∆C_t method of quantification. The probe was labeled with a 59-reporter dye FAM (6-carboxy-fluorescein) and the 39-quencher dye TAMRA (6-carboxy-N,N,N’,N9-tetramethyl-rhodamine). RT reactions were diluted 1 in 10 in distilled H₂O, and 5 μl of template was added to 6.5 μl of 2× master mix (Eurogentech, Seraing, Belgium) containing 1.2 μmol/l forward and reverse primers and 0.248 μmol/l of probe in a total volume of 12.5 μl. The PCR protocol was as follows: 50°C for 2 minutes and 95°C for 10 minutes, followed by 40 cycles of denaturation at 95°C for 15 seconds, and annealing/extension at 60°C for 1 minute. Thermocycling and real-time detection of PCR products were performed on asequence detection system (iCycler iQ; Bio-Rad Laboratories). Expression levels were normalized against 18S, β-actin, and glyceraldehyde 3-phosphate dehydrogenase (GAPDH), and values were calculated using the ∆∆C_t method.

Total soluble forms of collagen were measured in supernatants of tissue explants and myofibroblasts (Sircol Collagen Assay Kit; Biocolor Ltd, Belfast, UK), in accordance with the manufacturers’ instructions. The collagen content in each sample was calculated as an average of three readings.

Western blotting was performed using a modification of a previously described method [36]. Tissue samples or myofibroblasts were homogenized in ice-cold lysis buffer, and the amount of protein was determined by a protein assay (Bio-Rad Laboratories). Equal amounts of protein or 15 μl of cell- culture supernatants were loaded into each lane and run in 10% sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) gels under reducing conditions. After electrophoresis, protein was transferred onto nitrocellulose membranes (Bio-Rad Laboratories). Membranes were blocked with 5% non-fat dry milk, followed by incubation overnight at 4°C with the following antibodies: goat anti-human IL-17A (1 μg/ml), goat anti-human IL-17E (0.1 μg/ml) (both R&D Systems), mouse anti-human IL-17RB (1:500 dilution; LifeSpan Biosciences, Seattle, WA), mouse anti-human IL-17RC (1:750 dilution), rabbit anti-human MMP-3 (1:200 dilution), rabbit anti-human MMP-12 (1:1000 dilution) (all three from Abcam Ltd, Cambridge, UK), or mouse anti-human TIMP-1 (1 μg/ml; Oncogene Research, Nottingham, UK). Appropriate antibodies conjugated to horseradish peroxidase (Dako, High Wycombe, Buckinghamshire, UK) were used as secondary antibodies, and the reaction was developed with enhanced chemiluminescence (ECL Plus Kit;Amersham Biosciences, Little Chalfont, Buckinghamshire, UK). Blots were then stripped and analyzed for β-actin, as an internal loading control, using a rabbit anti-β-actin antibody (1:5000 dilution; Abcam Ltd). Bands were quantified by scanning densitometry (LKB Ultrascan XL Laser Densitometer; Kodak, Hemel Hempstead, UK).

Data were analyzed by the GraphPad Prism statistical PC program (GraphPad Software, San Diego, CA) using the paired t test and the Mann–Whitney U-test. P<0.05 was considered significant.