Fibrostenotic Phenotype of Myofibroblasts in Crohn’s Disease is Dependent on Tissue Stiffness and Reversed by LOX Inhibition

Fibrostenotic Phenotype of Myofibroblasts in Crohn’s Disease is Dependent on Tissue Stiffness... Abstract Background and Aims Crohn’s disease is a chronic inflammatory disorder of the intestine and often leads to fibrosis, characterized by excess extracellular matrix [ECM] deposition, increased tissue stiffness, and stricture formation. Here we evaluated the contribution of myofibroblast–ECM interactions to the development of intestinal fibrosis in Crohn’s disease. Methods Matched primary human myofibroblasts were isolated from stenotic, inflamed and normal-appearing small intestine within the same Crohn’s disease patient [n = 10]. Cells were analyzed by gene expression profiling, microscopy and functional assays, including matrix metalloproteinase [MMP] production and ECM contraction. Results We demonstrated that myofibroblasts isolated from stenotic intestine differed both in phenotype and function from those isolated from purely inflammatory or normal-appearing intestine of the same patient. Stenotic myofibroblasts displayed increased expression of genes associated with ECM modulation and collagen deposition. Upon culture in a fibrotic environment, normal myofibroblasts increased expression of MMPs to counteract the mechanical force exerted by the matrix. Interestingly, stenotic myofibroblasts showed a paradoxical response with decreased expression of MMP3. In addition, stenotic myofibroblasts expressed increased levels of the collagen crosslinking enzyme lysyl oxidase [LOX] and induced significantly more ECM contraction than both normal and inflamed myofibroblasts. Importantly, LOX inhibition completely restored MMP3 activity in stenotic myofibroblasts grown in a fibrotic environment, and prevented excessive ECM contraction. Conclusions Together these data indicate aberrancies in the myofibroblast–ECM interaction in Crohn’s disease, and identify LOX inhibition as a potential anti-fibrotic agent in this condition. Fibrosis, Crohn’s disease, extracellular matrix, lysyl oxidase 1. Introduction At diagnosis, most patients with Crohn’s disease present with a primarily inflammatory disease phenotype. Around 5% of patients have a stricture within 90 days of diagnosis,1 but over time this increases up to 30%.2 Although the therapeutic armamentarium for Crohn’s disease has improved significantly over the past decade, this has not led to a reduction in the volume of surgery performed for intestinal strictures,3 indicating a clear unmet need for these patients. Fibrosis is characterized by the formation of excess amounts of extracellular matrix [ECM] components such as collagen and fibronectin.4 In the intestine, the end stage of fibrosis is stricture formation with narrowing and obstruction of the lumen. Extracellular matrix production is a normal component of tissue repair, but can evolve into an irreversible, progressive fibrotic response if tissue injury is severe or if there are repeated cycles of inflammation—as is the case in Crohn’s disease. In line with this, intestinal inflammation is a necessary precondition for the initiation of fibrosis, but once the fibrotic process has started, progression may occur independently of inflammation.5 In normal wound healing, ECM deposition is regulated by production of its components and the activity of modulating enzymes, including matrix metalloproteinases [MMPs], which cleave ECM proteins. However, a dysregulation of MMP and/or ECM production results in fibrosis instead of normal wound healing.6,7 Accumulation of ECM components, of which collagen is most abundant, can lead to increased tissue stiffness, which in turn activates myofibroblasts, leading to further enhanced ECM production,8,9 thus resulting in a vicious circle. In a rat model for liver fibrosis, as well as in patients with hepatitis C, it has been shown that increased tissue stiffness precedes fibrosis,10,11 highlighting the importance of the ECM and tissue rigidity in the fibrotic process. Previous studies on the role of fibroblasts in Crohn’s disease have mainly compared the fibroblasts of patients with Crohn’s disease with those of control patients,12–14 or made use of the human colonic fibroblast cell line ccd18-co.5,15 One study compared stenotic and normal tissue–derived fibroblasts within the same patients, with a focus on transforming growth factor-β–driven mechanisms,6 and another focused on the role of microRNA in fibrosis.16 Unfortunately, these studies did not include fibroblasts from inflamed ileum. Because of the sequential order in Crohn’s disease, in which normal ileum evolves into inflamed ileum and eventually into stenotic ileum, we aimed to investigate the phenotype and function of myofibroblasts isolated from stenotic ileum compared with the phenotype and function of myofibroblasts from inflamed and normal-appearing ileum from the same patients. Furthermore, we used an unbiased approach and analyzed the full transcriptome of these cells. Our results show that human stenotic ileal myofibroblasts have a phenotype that is distinct from myofibroblasts isolated from normal-appearing or inflamed [but not stenotic] areas of the same patient. Specifically, stenotic myofibroblasts show a paradoxical response to increased tissue stiffness, with decreased expression of MMP3 activity upon increased stiffness. In addition, stenotic myofibroblasts express increased levels of the collagen crosslinking enzyme lysyl oxidase [LOX], further contributing to tissue rigidity. LOX inhibition completely reverses this behavior and prevents matrix contraction by stenotic myofibroblasts. These results suggest that LOX inhibition may be a possible anti-fibrotic therapy in Crohn’s disease. 2. Methods 2.1. Patient selection Myofibroblasts were isolated from the ileum of 10 Crohn’s disease patients undergoing resection of symptomatic ileal strictures in the Academic Medical Center, the Netherlands [patient characteristics in Table 1]. The local medical ethical committee [IRB] approved the study. All patients provided signed informed consent. Table 1. Baseline characteristics of patients included in the study. Gender [% female] 60% Age at diagnosis [years, mean +- SD] 22.2 [5.7] Age at surgery [years, median, IQR] 27.6 [22.7 – 33.3] Smoking [% yes] 0 Disease location [%]  Ileum 80%  Ileum + colon 20% Disease behaviour [%]  Stricturing 60%  Stricturing + penetrating 40% Medication at time of surgery [%]  None 20%  5-ASA 10%  Budesonide 10%  Antibiotics 10%  Azathioprine 20%  Anti-TNF 10%  Combination therapy azathioprine + anti-TNF 20% Gender [% female] 60% Age at diagnosis [years, mean +- SD] 22.2 [5.7] Age at surgery [years, median, IQR] 27.6 [22.7 – 33.3] Smoking [% yes] 0 Disease location [%]  Ileum 80%  Ileum + colon 20% Disease behaviour [%]  Stricturing 60%  Stricturing + penetrating 40% Medication at time of surgery [%]  None 20%  5-ASA 10%  Budesonide 10%  Antibiotics 10%  Azathioprine 20%  Anti-TNF 10%  Combination therapy azathioprine + anti-TNF 20% 5-ASA: 5 aminosalicylic acid; anti-TNF: anti-tumor necrosis factor alpha. View Large Table 1. Baseline characteristics of patients included in the study. Gender [% female] 60% Age at diagnosis [years, mean +- SD] 22.2 [5.7] Age at surgery [years, median, IQR] 27.6 [22.7 – 33.3] Smoking [% yes] 0 Disease location [%]  Ileum 80%  Ileum + colon 20% Disease behaviour [%]  Stricturing 60%  Stricturing + penetrating 40% Medication at time of surgery [%]  None 20%  5-ASA 10%  Budesonide 10%  Antibiotics 10%  Azathioprine 20%  Anti-TNF 10%  Combination therapy azathioprine + anti-TNF 20% Gender [% female] 60% Age at diagnosis [years, mean +- SD] 22.2 [5.7] Age at surgery [years, median, IQR] 27.6 [22.7 – 33.3] Smoking [% yes] 0 Disease location [%]  Ileum 80%  Ileum + colon 20% Disease behaviour [%]  Stricturing 60%  Stricturing + penetrating 40% Medication at time of surgery [%]  None 20%  5-ASA 10%  Budesonide 10%  Antibiotics 10%  Azathioprine 20%  Anti-TNF 10%  Combination therapy azathioprine + anti-TNF 20% 5-ASA: 5 aminosalicylic acid; anti-TNF: anti-tumor necrosis factor alpha. View Large 2.2. Myofibroblast isolation Myofibroblasts were isolated from the lamina propria of normal-, inflamed-only– and stenotic-appearing ileum of the same patient [method adapted from that of reference Owens et al.17]. Surgical specimens were cut open and fecal content was removed. The mucosa was separated from the submucosa mechanically and subsequently washed extensively in ice-cold PBS supplemented with 1% penicillin/streptomycin and 40 μg/ml G418 [PGA]. The mucosa was chopped up very finely, placed in RMPI-1640 [Invitrogen] supplemented with 1.5 mg/mL collagenase A [Roche, Germany], minced using the Gentlemacs Dissociator [Miltenyi Biotec, Leiden, the Netherlands], incubated at 37°C for 60 min, and further dissociated using another cycle of Gentlemacs. Cells were washed extensively with PGA before plating in RMPI-1640 with 10% FCS, 1% penicillin/streptomycin, 1% L-glutamin, G418 [40 μg/ml; Lonza, Leusden, the Netherlands] and amphotericin B [0.025 μg/ml; Gibco, Rockford, IL]. Importantly, myofibroblasts of normal-, inflamed-only– and stenotic-appearing ileum were paired from the same patient and at the same passage with the same cell density for individual experiments. Primary myofibroblasts were used at Passages 1–4. Ccd18-co colonic human fibroblast cell line was obtained from ATCC [Manassas, VA] cultured in DMEM [Lonza] supplemented with 10% FCS, 1% penicillin/streptomycin, 1% L-glutamin, G418 [40 μg/ml] and amphotericin B [0.025 μg/ml]. All cells tested negative for Mycoplasma. Cell proliferation of myofibroblasts from normal-, inflamed-only– and stenotic-appearing ileum was measured by plating 50 × 10E3 cells/well in a flat-bottomed 96-well plate. 3H-thymidine was added for 72 h and incorporation measured using a MicroBeta 2 Microplate Counter [Perkin Elmer, Waltham MA]. Additionally, proliferation was measured by impedance measurement using the xCelligence system as described before.18 For LOX inhibition experiments, cell cultures were treated with β-aminopropionitrile [βAPN, Sigma Aldrich], which blocks both LOX and LOXL219 at 500 μM concentration, for 3 days. For inhibiting focal adhesion kinase [FAK], cell cultures were treated with the FAK inhibitor Defactinib [10 μM, Selleckchem, Houston, TX] for 3 days. 2.3. RNA isolation, complementary DNA synthesis, quantitative reverse-transcription polymerase chain reaction and transcription analysis RNA was isolated using the RNAeasy mini kit [Qiagen, Valencia, CA]. Quantative RT–PCR was performed using SybrGreen [Roche] according to the manufacturer’s protocol on a BioRad iCycler. Glyceraldehyde-3-phosphate dehydrogenase [GAPDH] was used as the housekeeping gene [Supplementary Table 1 for primer sequences]. RNA was amplified using a TotalPrep RNA amplification kit for Illumina [Ambion] and labeled using a cRNA labeling kit for Illumina Arrays [Ambion], followed by hybridization with Illumina Human HT-12_v4 BeadChip slides. Expression profiles were deposited in the GEO repository [GSE90607]. Initial normalization was performed using Genome studio software [Illumina], gene set enrichment analyses were done using GSA software [Broad Institute of MIT and Harvard] and pathway analysis was done using Qiagen’s Ingenuity Pathway Analysis software [IPA, Qiagen Redwood City]. 2.4. Immunohistochemistry and immunofluorescence Paraffin-embedded tissue sections [4 μm] were deparaffinized, rehydrated and immersed in 0.3% H2O2 in methanol for 30 min. After antigen retrieval by sodium citrate for LOX staining and Tris-EDTA for CD3 staining, slides were blocked by PBS with 0.1% Triton X-100 and 1% bovine serum albumin [PBT] for 30 min, followed by incubation with primary antibody overnight at 4°C (rabbit polyclonal anti-human LOX [ab31238 Abcam, Cambridge, UK], rabbit monoclonal anti-human CD3 [clone SP7, Thermo Scientific] and goat polyclonal collagen I [Southern Biotech, Birmingham, AL, USA]) in PBT. Slides were incubated with Brightvision [Immunologic, Duiven, the Netherlands], counterstained with Haematoxylin and mounted. For immunofluorescence, cell cultures were fixed using 4% PFA, permeabilized using 0.1% Triton X, and stained using rabbit monoclonal anti-focal adhesion kinase [clone EP695Y, Abcam] followed by secondary donkey anti-rabbit 546 [Invitrogen, Carlsbad, CA] and phalloidin alexa fluor 488 [Thermo Scientific]. Samples were embedded in SlowFade Gold containing DAPI [Invitrogen] and imaged using a Leica DM6000B microscope. For flow cytometry, cells were isolated, fixed in 2% PFA, permeabilized in 0.1% Triton X/5 mM EDTA in PBS and stained using anti-desmin, anti-vimentin and anti-alpha-sma, followed by anti-rabbit-AF546 and anti-goat-AF647. Analysis was performed using a FACS Fortessa [BD Biosciences, Bedford, MA] and FlowJo software [FlowJo Inc. Ashland, Oregon]. 2.5. Protein analysis Matrix metalloproteinase activity was measured in the supernatant of myofibroblasts cultured in RPMI1640 without phenol red [Lonza], supplemented with 1% FCS, 1% penicillin/streptomycin, 1% L-glutamin, G418 [40 μg/ml] and amphotericin B [0.025 μg/ml]. Global MMP activity was assessed by adding 20 nmol/L OmniMMP substrate [Mca-Pro-Leu-Gly-Leu-Dpa-Ala-Arg-NH; Enzo Life Science, Lausen, Swiss] [in an appropriate assay buffer] to the supernatant. For global MMP activity, this ×10 buffer was 500 mM HEPES, 100 mM CaCl2, 0.5% Brij-35, pH 7.020; for more specific MMP2 activity, the ×10 buffer consisted of 500 mM MOPS, 100 mM CaCl2, 100 μM ZnCl2, 0.5% Brij-35, pH 7.0, and for even more specific MMP3 activity, the ×10 buffer consisted of 500 mM MES, 100 mM CaCl2, 0.5% Brij-35, pH 6.0, all diluted with cell culture supernatant. Fluorescence was detected by Novostar 700-0055 apparatus [BMG Labtech, Ortenberg, Germany] at a 320/405 wavelength by kinetic measurements every 100 s. Results are expressed as relative fluorescence increase compared with fluorescence intensity at the moment of stable background levels. LOXL2 and TGFb1 protein concentrations were measured using commercial human Duoset kits [R&D Systems, Minneapolis, MN]. 2.6. Stiffness-controlled extracellular matrix gels Tissue stiffness experiments were performed as described before.21 In brief, round glass coverslips were treated with HCl, 3-aminopropyltriethoxysilane and glutaraldehyde each time, followed by extensive washing with PBS. Collagen-coated polyacrylamide gels of 2.8 and 30 kPa were generated using varying ratios of 40% acrylamide and 2% bisacrylamide [Bio-rad], supplemented with 0.6% ammonium persulfate and 0.4% TEMED. Acrylamide gels were polymerized on treated coverslips by inverting the coverslips onto the acrylamide/bisacrylamide solution. After polymerization, coverslips were rinsed extensively with PBS. The acrylamide gels were crosslinked with 2.5 mg/mL sulfo-SANPAH [Thermo Scientific, Rockford, IL] at a wavelength of 365 nm in a Stratagene UV crosslinker oven [Stratagene, La Jolla, CA] delivering 7500 J energy in total, followed by UV sterilization. The coverslip-gels were placed in a 12-well tissue culture plate and coated with 2 mg/mL rat tail collagen I [BD Biosciences]. After 4 h incubation, cells were seeded at a density of 40 000 cells/well. For all experiments, paired cells were cultured for 3 days. 2.7. Extracellular matrix contraction assay To assess force-mediated matrix remodeling,22 150 000 myofibroblasts were embedded in 50 μL matrigel/collagen I mix, yielding a final collagen concentration [BD Bioscience] of 4.6 mg/mL and a final Matrigel concentration [VWR, Radnor, PA] of 2.2 mg/mL, and seeded on a 35-mm glass-bottom dish [MatTek, Ashland, MA]. After 30 min polymerization, cells were maintained in full growth medium with or without LOX inhibitor. Images were obtained daily for 7 days, and the diameter of the gel was measured using ImageJ software. Data are shown as the decrease of the circumference of the gel relative to Day 0. 2.8. Statistical analysis Data were analyzed using Graphpad Prism 7.0 [Graphpad Software Inc., La Jolla, CA]. The Kruskal–Wallis test with post-hoc analysis, or the Mann–Whitney-U test, were performed to compare data. Extracellular matrix contraction assays were analyzed as area under the curve, with a paired t-test afterwards. A P-value of <0.05 was considered statistically significant. 3. Results 3.1. Stenotic myofibroblasts differed in ECM organization and collagen production In order to investigate the phenotype and function of myofibroblasts from different regions of Crohn’s disease intestinal tissue, cells were isolated from patients undergoing surgery for stricturing disease of the ileum. The ileum of all these patients was considerably narrowed and the bowel wall was thickened. Macroscopically non-inflamed as well as non-stenotic [henceforth referred to as ‘normal’], inflamed-only and stenotic regions were identified within each individual patient [Figure 1a], and cells were isolated from all three areas separately. The nature of the regions used for isolation was confirmed by immunohistochemistry [Figure 1b]. Indeed, areas designated as ‘inflamed-only’ displayed significantly more CD3+ cells than either ‘normal’ or ‘stenotic’ regions. Areas designated ‘stenotic’ showed increased presence of ECM compared with normal and inflamed-only areas. Numbers of CD3+ cells in stenotic regions were intermediate between ‘normal’ and ‘inflamed-only’, in line with an expected mixed fibrotic/inflammatory phenotype. Of note, some infiltration of CD3+ cells was observed in the ‘normal’ areas, representing the gradient that occurs in the inflammatory response seen in Crohn’s disease patients. No apparent differences could be observed in cellular morphology, nor in expression of [myo]fibroblast markers alpha smooth muscle actin, desmin and vimentin [Supplementary figure 1]. Myofibroblasts isolated from inflamed-only areas proliferated much faster than myofibroblasts from normal and stenotic mucosa, as quantified both by thymidine incorporation assay and xCelligence impedance measurements [Figure 1c]. This effect remained stable over multiple passages [data not shown], excluding the confounding effect of inflammatory cytokines present directly after myofibroblast isolation. Interestingly, myofibroblasts obtained from stenotic regions displayed proliferation similar to normal myofibroblasts. Figure 1. View largeDownload slide Stenotic myofibroblasts differ from inflamed-only and normal myofibroblasts in extracellular matrix [ECM] organization and collagen production. [A] Macroscopic and low-power image of location selection. Resection specimen of Crohn’s ileum with narrowing at site of stenosis. Arrows indicate macroscopically normal, inflamed-only and stenotic regions. Low-power image of H&E staining; all magnifications: ×40. [B] Microscopic analysis of macroscopically normal, inflamed-only and stenotic mucosa by immunohistochemistry for CD3 and collagen I; bottom row depicts zoom of collagen I staining. [C] Proliferation of myofibroblasts as measured by xCelligence [upper graph, n = 5 individual patients, representative data shown] and 3H thymidine incorporation assay [lower graph, summarized data of eight experiments in individual patients, bars represent mean, error bars s.e.m.]. [D–E] mRNA was isolated from normal, inflamed-only and stenotic myofibroblasts obtained from seven individual patients, and gene expression was analyzed by Gene Set Enrichment Analysis. Vertical lines below x-axis display relative distribution of expression per gene included in the geneset. [F] Expression patterns of individual genes present in the Reactome geneset ‘ECM organization’. Paired samples of each condition were obtained from seven individual patients. Each line represents an individual sample. Expression is depicted as Z-score within the individual gene. Figure 1. View largeDownload slide Stenotic myofibroblasts differ from inflamed-only and normal myofibroblasts in extracellular matrix [ECM] organization and collagen production. [A] Macroscopic and low-power image of location selection. Resection specimen of Crohn’s ileum with narrowing at site of stenosis. Arrows indicate macroscopically normal, inflamed-only and stenotic regions. Low-power image of H&E staining; all magnifications: ×40. [B] Microscopic analysis of macroscopically normal, inflamed-only and stenotic mucosa by immunohistochemistry for CD3 and collagen I; bottom row depicts zoom of collagen I staining. [C] Proliferation of myofibroblasts as measured by xCelligence [upper graph, n = 5 individual patients, representative data shown] and 3H thymidine incorporation assay [lower graph, summarized data of eight experiments in individual patients, bars represent mean, error bars s.e.m.]. [D–E] mRNA was isolated from normal, inflamed-only and stenotic myofibroblasts obtained from seven individual patients, and gene expression was analyzed by Gene Set Enrichment Analysis. Vertical lines below x-axis display relative distribution of expression per gene included in the geneset. [F] Expression patterns of individual genes present in the Reactome geneset ‘ECM organization’. Paired samples of each condition were obtained from seven individual patients. Each line represents an individual sample. Expression is depicted as Z-score within the individual gene. To further evaluate similarities and differences between normal and stenotic myofibroblasts, we performed transcriptional analysis on myofibroblast populations isolated from eight individual patients. When comparing normal and stenotic myofibroblasts, pathway analysis using Ingenuity software revealed alterations in pathways involved in cell–cell and cell–matrix interactions, including fibrotic responses and MMPs [Table 2]. Gene set enrichment analysis [GSEA] showed significant enrichment of genes associated with ECM organization and collagen production in stenotic myofibroblasts compared with in those from normal tissue regions [Figure 1d]. As stenotic areas did display varying degrees of inflammation, stenotic myofibroblasts were also compared with those obtained from the inflammation-only areas. Again, genes associated with ECM organization and collagen production were increased in the stenotic myofibroblasts, suggesting the results are not merely the consequence of residual inflammation in the stenotic areas [Figure 1e and f]. Collectively, we showed that myofibroblasts isolated from stenotic, inflamed-only and normal-appearing regions of the ileum within the same patient have different phenotypes and that stenotic myofibroblasts are distinct from both inflamed-only and normal myofibroblasts in terms of ECM organization and collagen production. Table 2. Top 10 canonical Ingenuity pathways differentially activated between normal and stenotic myofibroblasts. Ingenuity canonical pathways -log[p-value] Hepatic fibrosis / Hepatic stellate cell activation 7.99E+00 Retinoate biosynthesis I 3.56E+00 Bile acid biosynthesis, neutral pathway 3.28E+00 Altered T cell and B cell signaling in rheumatoid arthritis 3.25E+00 Inhibition of matrix metalloproteases 2.99E+00 Agrin interactions at neuromuscular junction 2.97E+00 Agranulocyte adhesion and diapedesis 2.94E+00 Atherosclerosis signaling 2.89E+00 IL-8 signaling 2.76E+00 Granulocyte adhesion and diapedesis 2.71E+00 Ingenuity canonical pathways -log[p-value] Hepatic fibrosis / Hepatic stellate cell activation 7.99E+00 Retinoate biosynthesis I 3.56E+00 Bile acid biosynthesis, neutral pathway 3.28E+00 Altered T cell and B cell signaling in rheumatoid arthritis 3.25E+00 Inhibition of matrix metalloproteases 2.99E+00 Agrin interactions at neuromuscular junction 2.97E+00 Agranulocyte adhesion and diapedesis 2.94E+00 Atherosclerosis signaling 2.89E+00 IL-8 signaling 2.76E+00 Granulocyte adhesion and diapedesis 2.71E+00 IL: Interleukin. View Large Table 2. Top 10 canonical Ingenuity pathways differentially activated between normal and stenotic myofibroblasts. Ingenuity canonical pathways -log[p-value] Hepatic fibrosis / Hepatic stellate cell activation 7.99E+00 Retinoate biosynthesis I 3.56E+00 Bile acid biosynthesis, neutral pathway 3.28E+00 Altered T cell and B cell signaling in rheumatoid arthritis 3.25E+00 Inhibition of matrix metalloproteases 2.99E+00 Agrin interactions at neuromuscular junction 2.97E+00 Agranulocyte adhesion and diapedesis 2.94E+00 Atherosclerosis signaling 2.89E+00 IL-8 signaling 2.76E+00 Granulocyte adhesion and diapedesis 2.71E+00 Ingenuity canonical pathways -log[p-value] Hepatic fibrosis / Hepatic stellate cell activation 7.99E+00 Retinoate biosynthesis I 3.56E+00 Bile acid biosynthesis, neutral pathway 3.28E+00 Altered T cell and B cell signaling in rheumatoid arthritis 3.25E+00 Inhibition of matrix metalloproteases 2.99E+00 Agrin interactions at neuromuscular junction 2.97E+00 Agranulocyte adhesion and diapedesis 2.94E+00 Atherosclerosis signaling 2.89E+00 IL-8 signaling 2.76E+00 Granulocyte adhesion and diapedesis 2.71E+00 IL: Interleukin. View Large 3.2. Paradoxical response to tissue stiffness of myofibroblasts in MMP3 activity Transcriptional analysis suggested that several MMP transcripts were significantly more highly expressed in myofibroblasts from stenotic ileum compared with in myofibroblasts from both inflamed-only and normal ileum [Figure 1f]. This is counterintuitive, as MMPs are endopeptidases involved in the degradation of ECM proteins, which cause stenosis, so it would be reasonable to expect reduced MMP activity in stenotic lesions in patients with Crohn’s disease. Indeed, previously published studies have shown that MMP protein expression and activity levels decreased in mucosa overlying stenotic ileum in vivo.6 We corroborated the transcriptional data using enzyme activity assays, which not only measure MMP protein levels but also take into account the effects of their natural inhibitors [TIMPs]. In accordance with the transcriptional data, MMP activity was increased in stenotic myofibroblasts compared with in both normal and inflamed-only myofibroblasts when cultured in the absence of ECM [Figure 2a]. This was seen when analyzing global MMP activity as well as more specific MMP2 and MMP3 activity. Figure 2. View largeDownload slide Matrix metalloproteinase [MMP] activity in myofibroblasts isolated from normal, inflamed-only and stenotic ileal regions. [A] Isolated primary myofibroblasts were cultured in the absence of ECM for 3 days, and MMP activity was measured in the supernatant using a quenched fluorescent probe. Representative graphs of experiments in three separate patients are shown. [B–C] Primary myofibroblasts [B] or the fibroblast cell line ccd18-co [C] were cultured in matrix with a normal [2.8 kPa, blue] or high [30 kPa, red] compliance for 3 days. MMP3 activity was measured in the supernatant. Representative data of four experiments are shown. Figure 2. View largeDownload slide Matrix metalloproteinase [MMP] activity in myofibroblasts isolated from normal, inflamed-only and stenotic ileal regions. [A] Isolated primary myofibroblasts were cultured in the absence of ECM for 3 days, and MMP activity was measured in the supernatant using a quenched fluorescent probe. Representative graphs of experiments in three separate patients are shown. [B–C] Primary myofibroblasts [B] or the fibroblast cell line ccd18-co [C] were cultured in matrix with a normal [2.8 kPa, blue] or high [30 kPa, red] compliance for 3 days. MMP3 activity was measured in the supernatant. Representative data of four experiments are shown. This remarkable discrepancy might originate from a different physical environment in stenotic lesions compared with that in normal ileum, with high tissue stiffness in fibrostenosis. It has previously been shown that normal ileum has a compliance of ~2.8 kPa, which increases up to 30 kPa in the stenotic region of a Crohn’s ileum.23 The effect of stiffness can be mimicked in vitro by embedding cells in gels containing varying amounts of ECM components.21 As expected, myofibroblasts isolated from normal-appearing mucosa increased MMP3 activity when cultured in stiff 30 kPa conditions, indicating that an increase in tissue stiffness activates a compensatory activation of MMPs in normal myofibroblasts. In sharp contrast, the MMP3 activity of stenotic myofibroblasts was low in a matrix that mimicked their native environment [30 kPa] and paradoxically increased when environmental pressure was decreased to 2.8 kPa [Figure 2b]. Interestingly, we found that the human colonic fibroblast cell line ccd18-co had the same MMP3 activity pattern, with low activity in a stiff environment and a paradoxical increase in MMP3 activity in soft matrix, indicatingthat this cell line may display a more stenotic phenotype [Figure 2c]. Collectively our data show that normal myofibroblasts increase their MMP3 activity in stiff matrix, presumably in order to degrade ECM, and counteract the increased compliance. In contrast, stenotic myofibroblasts display an aberrant response to a stiff tissue environment, with reduced MMP3 activity. 3.3. Enhanced ECM contraction in stenotic myofibroblasts in vitro In vivo, intestinal fibrosis is characterized by excess ECM production, of which collagen is one of the major components.4 Increased collagen deposition leads to tissue scarring and subsequent contraction of the tissue, and may result in stenosis.24 Indeed stenotic myofibroblasts displayed increased expression of COL1A1, COL5A1 and COL11A1 compared with cells isolated from the normal but not the inflamed-only regions [Figure 3a]. The functional capacity of stenotic myofibroblasts to induce ECM contraction in vitro, using the collagen/matrigel contraction assay, is a validated method for measuring cellular contractility within an ECM.22 Cells are cultured in a gel matrix that contracts upon collagen deposition [similar to the contraction seen during fibrosis in vivo]. In this assay, stenotic myofibroblasts caused substantially more ECM contraction than myofibroblasts isolated from either inflamed-only or normal-appearing mucosa from the same patient [Figure 3b], consistent with tissue contraction of stenotic tissue in vivo. The ccd18-co cell line caused contraction of the ECM as well, again showing a striking resemblance to the phenotypic characteristics of stenotic myofibroblasts [Figure 3c]. Figure 3. View largeDownload slide Enhanced extracellular matrix [ECM] contraction by stenotic myofibroblasts. [A] Expression of COL1A1, COL5A1 and COLL11A1 in myofibroblasts isolated from stenotic, inflamed-only or normal regions as measured by qPCR and normalized to GAPDH. Dots represent individual patients. [B] Myofibroblasts were embedded in a matrigel/collagen I mix and cultured for 7 days. Contraction was measured as percentage decrease in matrix surface. Representative graphs and images of three independent experiments. Error bars represent median and standard error. Asterisk indicates p ≤ 0.05. Figure 3. View largeDownload slide Enhanced extracellular matrix [ECM] contraction by stenotic myofibroblasts. [A] Expression of COL1A1, COL5A1 and COLL11A1 in myofibroblasts isolated from stenotic, inflamed-only or normal regions as measured by qPCR and normalized to GAPDH. Dots represent individual patients. [B] Myofibroblasts were embedded in a matrigel/collagen I mix and cultured for 7 days. Contraction was measured as percentage decrease in matrix surface. Representative graphs and images of three independent experiments. Error bars represent median and standard error. Asterisk indicates p ≤ 0.05. 3.4. LOX inhibition restored MMP3 activity and decreased ECM contraction in stenotic myofibroblasts Collagen is produced and secreted as a soluble protein. After secretion, crosslinking of individual fibrils results in collagen deposition and consequently increased tissue stiffness.25 Key enzymes in this process are the LOX family members LOX, lysyl oxidase-like-1 [LOXL1] and lysyl oxidase-like-2 [LOXL2]. These enzymes are primarily responsible for the covalent crosslinking of collagen and elastin in the ECM.26 In addition, LOX can interact with fibronectin, another ECM component. In turn, fibronectin can increase LOX activity, leading to a positive feedback loop with further collagen crosslinking and tissue stiffness.27,28 Enhanced LOX activity may thus be associated with increased formation of stenotic lesions. Interestingly, on the transcriptional level, LOX and LOXL2 expression were increased in myofibroblasts isolated from stenotic ileum, with a similar trend for LOXL1 [Figure 4a]. Likewise, secretion of LOXL2 protein was significantly higher in stenotic myofibroblasts than in normal myofibroblasts [Figure 4b]. Finally, immunohistochemistry confirmed increased expression of LOX in stenotic regions in vivo [Figure 4c]. Figure 4. View largeDownload slide Enhanced lysyl oxidase [LOX] expression in stenotic myofibroblasts. [A] Expression of LOX, LOX-like-1 [LOXL1] and LOX-like 2 [LOXL2] mRNA in primary myofibroblasts normalized to GAPDH. Dots represent individual patients. [B] Isolated myofibroblasts were cultured and LOXL2 in the supernatant measured by ELISA. n = 8 patients per condition. [C] Resection specimens from normal, inflamed-only and stenotic regions of the ileum were stained for LOX. Arrows indicate positive staining. Magnification: ×400. Single asterisk indicates p ≤ 0.05; two asterisks indicate p ≤ 0.01; three asterisks indicate p ≤ 0.001. Figure 4. View largeDownload slide Enhanced lysyl oxidase [LOX] expression in stenotic myofibroblasts. [A] Expression of LOX, LOX-like-1 [LOXL1] and LOX-like 2 [LOXL2] mRNA in primary myofibroblasts normalized to GAPDH. Dots represent individual patients. [B] Isolated myofibroblasts were cultured and LOXL2 in the supernatant measured by ELISA. n = 8 patients per condition. [C] Resection specimens from normal, inflamed-only and stenotic regions of the ileum were stained for LOX. Arrows indicate positive staining. Magnification: ×400. Single asterisk indicates p ≤ 0.05; two asterisks indicate p ≤ 0.01; three asterisks indicate p ≤ 0.001. As we observed that stenotic myofibroblasts showed an aberrant suppression of MMP activity in response to a stiff microenvironment, altering the microenvironment might change the phenotype of myofibroblasts. Therefore, we hypothesized that blocking collagen crosslinking by inhibition of the LOX enzyme family might restore MMP activity in stenotic myofibroblasts, which could result in resolution of the accumulated ECM. Focal adhesion kinase [FAK] is a downstream signaling transducer of LOX which is important for cell–matrix interaction,29 and inhibiting LOX reduced expression of FAK, as seen by immunofluorescence staining [Figure 5a]. In line with our hypothesis, inhibiting LOX in ccd18-co fibroblasts [which show responses to environmental cues comparable with those of stenotic myofibroblasts] strongly increased the activity of MMP3 when cultured under stenotic conditions [30 kPa]. In fact, MMP3 activity under these conditions was comparable with that seen in myofibroblasts cultured under normal 2.8 kPa conditions [Figure 5b]. Similarly, in primary myofibroblasts isolated from stenotic mucosa, MMP3 activity increased when these myofibroblasts were cultured in a stiff matrix in the presence of the LOX inhibitor [Figure 5c]. Interestingly, direct inhibition of FAK using the small molecule Defactinib did not alter MMP3 activity [Figure 5d]. Figure 5. View largeDownload slide LOX inhibition increases matrix metalloproteinase 3 [MMP3] activity and decreases extracellular matrix [ECM] contraction induced by stenotic myofibroblasts. [A] Ccd18-co cells were cultured in the presence or absence of LOX inhibitor [βAPN, 500 μM] for 3 days. Cells were stained for actin [phalloidin, green] and FAK [red]. [B–C] Ccd18-co fibroblasts [B] or primary stenotic myofibroblasts [C] were cultured in matrix with a normal [2.8 kPa, blue] or high [30 kPa, red] compliance in the presence of LOX inhibitor for 3 days. MMP3 activity was measured in the supernatant. Representative graphs shown, n = 5 independent experiments for ccd18-co cells; n = 3 independent experiments for primary cells. [D] Ccd18-co fibroblasts were cultured in matrix with a normal [2.8 kPa, blue] or high [30 kPa, red] compliance in the presence of FAK inhibitor Defactinib [10 μM] for 3 days. MMP3 activity was measured in the supernatant. Representative graph shown, n = 3 independent experiments. [E–F] Ccd18-co [E] or primary stenotic [F] myofibroblasts were embedded in a matrigel/collagen I mix and cultured for 7 days in the presence of LOX inhibitor. Contraction was measured as percentage decrease in matrix surface. Left graph depicts representative experiment, right graph shows data of four [E] or two [F] individual experiments as area under the curve. Asterisk indicates p ≤ 0.05. Figure 5. View largeDownload slide LOX inhibition increases matrix metalloproteinase 3 [MMP3] activity and decreases extracellular matrix [ECM] contraction induced by stenotic myofibroblasts. [A] Ccd18-co cells were cultured in the presence or absence of LOX inhibitor [βAPN, 500 μM] for 3 days. Cells were stained for actin [phalloidin, green] and FAK [red]. [B–C] Ccd18-co fibroblasts [B] or primary stenotic myofibroblasts [C] were cultured in matrix with a normal [2.8 kPa, blue] or high [30 kPa, red] compliance in the presence of LOX inhibitor for 3 days. MMP3 activity was measured in the supernatant. Representative graphs shown, n = 5 independent experiments for ccd18-co cells; n = 3 independent experiments for primary cells. [D] Ccd18-co fibroblasts were cultured in matrix with a normal [2.8 kPa, blue] or high [30 kPa, red] compliance in the presence of FAK inhibitor Defactinib [10 μM] for 3 days. MMP3 activity was measured in the supernatant. Representative graph shown, n = 3 independent experiments. [E–F] Ccd18-co [E] or primary stenotic [F] myofibroblasts were embedded in a matrigel/collagen I mix and cultured for 7 days in the presence of LOX inhibitor. Contraction was measured as percentage decrease in matrix surface. Left graph depicts representative experiment, right graph shows data of four [E] or two [F] individual experiments as area under the curve. Asterisk indicates p ≤ 0.05. To test the hypothesis that LOX inhibition also functionally decreases ECM contraction, we cultured ccd18-co fibroblasts in the contraction assay. Upon LOX inhibition, ECM contraction decreased significantly in the gels with LOX inhibition compared with that in the control gels [Figure 5e]. As we showed earlier, myofibroblasts from stenotic ileum exhibited more ECM contraction than normal myofibroblasts [Figure 3a], but this phenotype was reversed upon culture in the presence of a LOX inhibitor [Figure 5f]. Importantly, the inhibitor had very limited effect on the ECM contraction levels of myofibroblasts from normal and inflamed-only mucosa [data not shown], showing that LOX activity specifically decreases tissue contractility in stenotic myofibroblasts. 4. Discussion These data show that myofibroblasts isolated from stenotic intestinal areas of patients with Crohn’s disease displayed a pathological phenotype when compared with myofibroblasts isolated from normal-appearing and inflamed-only areas from the same patient. Intriguingly, the pathologic phenotype depended at least partially on the nature of the surrounding matrix, suggesting modulation of the ECM may alter myofibroblast behavior and thus serve as a potential therapeutic target for stenotic complications of Crohn’s disease. For this study, we compared paired samples from within the same patient, and designated these as ‘normal’, ‘inflammation-only’ or ‘stenotic’, based on macroscopic appearance. Although subsequent histological analysis confirmed increased levels of CD3+ cells in ‘inflamed-only’ and increased ECM deposition in ‘stenotic’ areas, it has to be acknowledged that Crohn’s disease patients display gradients in the level of inflammation and stenosis. As a result, areas designated in this study as ‘normal’ still contain a certain level of inflammatory cells. In addition, ‘stenotic’ areas are not only stenotic, but also display considerable levels of inflammation, as is to be expected in these patients. However, despite this gradual rather than black-and-white scale between the different phenotypes, clear phenotypic and functional differences were observed between the three types of myofibroblasts, supporting our selection method in the context of Crohn’s disease. Caution has to be exerted when extending these findings to healthy ‘normal’ tissue [which is likely to contain less inflammatory cells] or non–IBD-related cases of stenosis. Earlier studies have aimed at identifying alterations in fibroblasts obtained from strictures in Crohn’s disease. For example, increased levels of vascular endothelial growth factor [VEGF] as well as N-Cadherin were described on the protein level.30,31 Indeed, we observed increased levels of CDH2 transcript, the gene encoding for N-Cadherin, while we did not observe alterations in VEGF expression. Furthermore, transcriptional profiling was performed earlier, comparing fibroblasts from strictured and non-strictured regions.32 Although this study did not include macroscopically normal regions from Crohn’s disease patients, a number of transcripts upregulated in this study were confirmed in our experiments. Additionally, previous studies reported increased transcription of the matrix-associated cytokine TGFβ, an important cytokine involved in matrix remodeling in stenotic intestinal regions.14 Interestingly, although total protein levels of TGFb1 did not differ between the various locations [Supplementary figure 2], upstream analysis of the transcriptome indicated a modestly enhanced TGFb1 activity in inflamed regions and significantly increased activity in stenotic regions [data not shown]. The ECM is a dynamic tissue component undergoing constant remodeling, mediated among others by MMPs. Defective MMP3 secretion due to genetic ablation results in enhanced sensitivity to colitis in a murine model,33 and Crohn’s disease patients carrying a single-nucleotide polymorphism in the gene encoding for MMP3 have a 20% higher risk of developing stenotic complications.34 Additionally, expression of MMP3 was decreased in mucosa overlying strictured intestine in Crohn’s disease, while expression of the MMP inhibitor TIMP1 was increased.6,35 Our data show that this defective MMP3 activity is not an intrinsic defect of stenotic myofibroblasts per se, but occurs specifically in the context of a stenotic ECM. The assays used in this study do not differentiate fully between the various MMP family members, but rather are based on conditions favoring one family member over the others. Consequently, we cannot exclude the possibility that the increased activity seen in the MMP3 assays is partly caused by the other stromelysins MMP10 and MMP11. Others have shown earlier that transcription of MMP12 was decreased in myofibroblasts isolated from stenotic regions, independent of the presence of ECM.6 This may indicate differential regulatory mechanisms between various MMP family members. A recent study described various alterations in DNA methylation patterns in myofibroblasts isolated from stenotic intestine,36 suggesting intrinsic and potentially long-lasting alterations. This is in line with our findings that some of the aberrancies remain throughout several passages in culture. It has also been shown previously that myofibroblasts isolated from stenotic regions of the small intestine display increased proliferative capacity,37 whereas we describe increased proliferative capacity in inflamed-only, but not in stenotic myofibroblasts. This difference may be the result of alternatively selected regions, as the former study specifically included regions simultaneously stenosed and inflamed. In our study, the stenotic region did show inflammatory activity, but this was considerably less than in the regions designated as inflamed. Interestingly, therapeutic application of ECM modulation has been known for decades in a more intuitive form. In dermatology, scar massage is anecdotally effective for the treatment of postsurgical scars.38 Similarly, intestinal fibrosis is currently treated through balloon endoscopy or surgical strictureplasty.39 During strictureplasty, the narrowed ileum is incised, thereby relieving local tissue pressure. Despite the fact that local myofibroblasts are left in situ, the recurrence rate at the site of a previous strictureplasty is only 3–10%.39,40 Although requiring further study, it is tempting to speculate that, also in this case, the release of local tissue pressures affects myofibroblast behavior. The effects of mechanical forces on a cellular level have been appreciated in the context of medical therapy only recently. For example, increased ECM deposition and subsequently stiffness within the tumor niche enhances tumor cell survival and promotes malignant progression.41,42 In this case, the enhanced ECM deposition is mediated by collagen crosslinking via members of the LOX family.43 Inhibiting LOX activity in a mouse model for breast carcinoma reduced collagen deposits as well as the number of focal adhesions and subsequently limited tumor progression.42 In line with this, we observed less intracellular FAK staining after inhibition of LOX. However, direct inhibition of FAK did not alter MMP3 activity, suggesting that tissue stiffness is not [exclusively] translated by integrin-mediated FAK signaling in intestinal stenotic myofibroblasts. Previous studies have indicated the presence of FAK-independent integrin signaling specifically in fibroblasts.44 Indeed, LOX inhibition as a mode of affecting ECM stiffness has been shown to limit liver fibrosis in vitro as well as in animal models,45,46 although it had no benefits in patients with primary sclerosing cholangitis as well as non-alcoholic steatohepatitis.47,48 In summary, we showed that myofibroblasts isolated from stenotic ileum were phenotypically and functionally different from those isolated from normal- and inflamed-only–appearing regions in the same patient, mainly with respect to ECM organization and collagen production. Myofibroblasts from stenotic ileum aberrantly suppressed expression of MMP3 in the stiff tissue environment that corresponds to an intestinal stenosis in vivo. Altering the physical properties of the microenvironment by LOX inhibition resulted in a release of MMP3 activity by stenotic myofibroblasts as well as decreased matrix contraction. These data indicate that modulation of the ECM through inhibition of LOX may be an effective anti-fibrotic therapy in Crohn’s disease. Funding This research was partially funded by a VICI grant number 91814605 from the Netherlands Organization for Scientific Research [to GvdB]. Conflict of Interest GvdB is currently an employee of GSK. Author Contributions All authors have made substantial contributions to the following: conception and design of the study [JdB, MW, GD, GvdB], acquisition of data [JdB, MW, JS, SM, VM], providing patient material [CB, WB, AtV], analysis and interpretation of data [JdB, MW, JS, GvdB], statistical analysis [JdB, MW], drafting the article [JdB, MW], critical revision of the manuscript for important intellectual content [JdB, MW, GD, GvdB] and final approval of the version to be submitted [all authors]. All authors had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Data deposit Expression profiles were deposited in the GEO repository [GSE90607]. Supplementary Data Supplementary data are available at ECCO-JCC online. Abbreviations: Abbreviations: ECM extracellular matrix FAK focal adhesion kinase LOX lysyl oxidase LOXL2 lysyl oxidase–like 2 MMP matrix metalloproteinase TIMP tissue inhibitor of metalloproteinase VEGF vascular endothelial growth factor. References 1. Thia KT , Sandborn WJ , Harmsen WS , Zinsmeister AR , Loftus EV Jr . Risk factors associated with progression to intestinal complications of Crohn’s disease in a population-based cohort . Gastroenterology 2010 ; 139 : 1147 – 55 . Google Scholar CrossRef Search ADS PubMed 2. Cosnes J , Gower-Rousseau C , Seksik P , Cortot A . Epidemiology and natural history of inflammatory bowel diseases . Gastroenterology 2011 ; 140 : 1785 – 94 . Google Scholar CrossRef Search ADS PubMed 3. Cosnes J , Nion-Larmurier I , Beaugerie L , Afchain P , Tiret E , Gendre JP . Impact of the increasing use of immunosuppressants in Crohn’s disease on the need for intestinal surgery . Gut 2005 ; 54 : 237 – 41 . Google Scholar CrossRef Search ADS PubMed 4. Wynn TA , Ramalingam TR . Mechanisms of fibrosis: therapeutic translation for fibrotic disease . Nat Med 2012 ; 18 : 1028 – 40 . Google Scholar CrossRef Search ADS PubMed 5. Johnson LA , Luke A , Sauder K , Moons DS , Horowitz JC , Higgins PD . Intestinal fibrosis is reduced by early elimination of inflammation in a mouse model of IBD: impact of a “Top-Down” approach to intestinal fibrosis in mice . Inflamm Bowel Dis 2012 ; 18 : 460 – 71 . Google Scholar CrossRef Search ADS PubMed 6. Di Sabatino A , Jackson CL , Pickard KM et al. Transforming growth factor beta signalling and matrix metalloproteinases in the mucosa overlying Crohn’s disease strictures . Gut 2009 ; 58 : 777 – 89 . Google Scholar CrossRef Search ADS PubMed 7. Shimshoni E , Yablecovitch D , Baram L , Dotan I , Sagi I . ECM remodelling in IBD: innocent bystander or partner in crime? The emerging role of extracellular molecular events in sustaining intestinal inflammation . Gut 2015 ; 64 : 367 – 72 . Google Scholar CrossRef Search ADS PubMed 8. Wells RG . The role of matrix stiffness in regulating cell behavior . Hepatology 2008 ; 47 : 1394 – 400 . Google Scholar CrossRef Search ADS PubMed 9. Rustad KC , Wong VW , Gurtner GC . The role of focal adhesion complexes in fibroblast mechanotransduction during scar formation . Differentiation 2013 ; 86 : 87 – 91 . Google Scholar CrossRef Search ADS PubMed 10. Georges PC , Hui JJ , Gombos Z et al. Increased stiffness of the rat liver precedes matrix deposition: implications for fibrosis . Am J Physiol Gastrointest Liver Physiol 2007 ; 293 : G1147 – 54 . Google Scholar CrossRef Search ADS PubMed 11. Yin M , Talwalkar JA , Glaser KJ et al. Assessment of hepatic fibrosis with magnetic resonance elastography . Clin Gastroenterol Hepatol 2007 ; 5 : 1207 – 1213.e2 . Google Scholar CrossRef Search ADS PubMed 12. Grillo AR , Scarpa M , D’Incà R et al. TAK1 is a key modulator of the profibrogenic phenotype of human ileal myofibroblasts in Crohn’s disease . Am J Physiol Gastrointest Liver Physiol 2015 ; 309 : G443 – 54 . Google Scholar CrossRef Search ADS PubMed 13. Di Sabatino A , Pender SL , Jackson CL et al. Functional modulation of Crohn’s disease myofibroblasts by anti-tumor necrosis factor antibodies . Gastroenterology 2007 ; 133 : 137 – 49 . Google Scholar CrossRef Search ADS PubMed 14. Burke JP , Ferrante M , Dejaegher K et al. Transcriptomic analysis of intestinal fibrosis-associated gene expression in response to medical therapy in Crohn’s disease . Inflamm Bowel Dis 2008 ; 14 : 1197 – 204 . Google Scholar CrossRef Search ADS PubMed 15. Otte JM , Rosenberg IM , Podolsky DK . Intestinal myofibroblasts in innate immune responses of the intestine . Gastroenterology 2003 ; 124 : 1866 – 78 . Google Scholar CrossRef Search ADS PubMed 16. Nijhuis A , Biancheri P , Lewis A et al. In Crohn’s disease fibrosis-reduced expression of the miR-29 family enhances collagen expression in intestinal fibroblasts . Clin Sci (Lond) 2014 ; 127 : 341 – 50 . Google Scholar CrossRef Search ADS PubMed 17. Owens BM , Steevels TA , Dudek M et al. CD90+ stromal cells are non-professional innate immune effectors of the human colonic mucosa . Front Immunol 2013 ; 4 : 307 . Google Scholar CrossRef Search ADS PubMed 18. Ke N , Wang X , Xu X , Abassi YA . The xCELLigence system for real-time and label-free monitoring of cell viability . Methods Mol Biol 2011 ; 740 : 33 – 43 . Google Scholar CrossRef Search ADS PubMed 19. Barry-Hamilton V , Spangler R , Marshall D et al. Allosteric inhibition of lysyl oxidase–like-2 impedes the development of a pathologic microenvironment . Nat Med 2010 ; 16 : 1009 – 17 . Google Scholar CrossRef Search ADS PubMed 20. Asbun J , Manso AM , Villarreal FJ . Profibrotic influence of high glucose concentration on cardiac fibroblast functions: effects of losartan and vitamin E . Am J Physiol Heart Circ Physiol 2005 ; 288 : H227 – 34 . Google Scholar CrossRef Search ADS PubMed 21. Fischer RS , Myers KA , Gardel ML , Waterman CM . Stiffness-controlled three-dimensional extracellular matrices for high-resolution imaging of cell behavior . Nat Protoc 2012 ; 7 : 2056 – 66 . Google Scholar CrossRef Search ADS PubMed 22. Calvo F , Ege N , Grande-Garcia A et al. Mechanotransduction and YAP-dependent matrix remodelling is required for the generation and maintenance of cancer-associated fibroblasts . Nat Cell Biol 2013 ; 15 : 637 – 46 . Google Scholar CrossRef Search ADS PubMed 23. Johnson LA , Rodansky ES , Sauder KL et al. Matrix stiffness corresponding to strictured bowel induces a fibrogenic response in human colonic fibroblasts . Inflamm Bowel Dis 2013 ; 19 : 891 – 903 . Google Scholar CrossRef Search ADS PubMed 24. Gurtner GC , Werner S , Barrandon Y , Longaker MT . Wound repair and regeneration . Nature 2008 ; 453 : 314 – 21 . Google Scholar CrossRef Search ADS PubMed 25. Wynn TA . Cellular and molecular mechanisms of fibrosis . J Pathol 2008 ; 214 : 199 – 210 . Google Scholar CrossRef Search ADS PubMed 26. Kagan HM , Li W . Lysyl oxidase: properties, specificity, and biological roles inside and outside of the cell . J Cell Biochem 2003 ; 88 : 660 – 72 . Google Scholar CrossRef Search ADS PubMed 27. Barker HE , Cox TR , Erler JT . The rationale for targeting the LOX family in cancer . Nat Rev Cancer 2012 ; 12 : 540 – 52 . Google Scholar CrossRef Search ADS PubMed 28. Fogelgren B , Polgár N , Szauter KM et al. Cellular fibronectin binds to lysyl oxidase with high affinity and is critical for its proteolytic activation . J Biol Chem 2005 ; 280 : 24690 – 7 . Google Scholar CrossRef Search ADS PubMed 29. Wong VW , Rustad KC , Akaishi S et al. Focal adhesion kinase links mechanical force to skin fibrosis via inflammatory signaling . Nat Med 2011 ; 18 : 148 – 52 . Google Scholar CrossRef Search ADS PubMed 30. Burke JP , Cunningham MF , Sweeney C , Docherty NG , O’Connell PR . N-cadherin is overexpressed in Crohn’s stricture fibroblasts and promotes intestinal fibroblast migration . Inflamm Bowel Dis 2011 ; 17 : 1665 – 73 . Google Scholar CrossRef Search ADS PubMed 31. Beddy D , Watson RW , Fitzpatrick JM , O’Connell PR . Increased vascular endothelial growth factor production in fibroblasts isolated from strictures in patients with Crohn’s disease . Br J Surg 2004 ; 91 : 72 – 7 . Google Scholar CrossRef Search ADS PubMed 32. Lang M , Schlechtweg M , Kellermeier S et al. Gene expression profiles of mucosal fibroblasts from strictured and nonstrictured areas of patients with Crohn’s disease . Inflamm Bowel Dis 2009 ; 15 : 212 – 23 . Google Scholar CrossRef Search ADS PubMed 33. Li CK , Pender SL , Pickard KM et al. Impaired immunity to intestinal bacterial infection in stromelysin-1 (matrix metalloproteinase-3)-deficient mice . J Immunol 2004 ; 173 : 5171 – 9 . Google Scholar CrossRef Search ADS PubMed 34. Meijer MJ , Mieremet-Ooms MA , van Hogezand RA , Lamers CB , Hommes DW , Verspaget HW . Role of matrix metalloproteinase, tissue inhibitor of metalloproteinase and tumor necrosis factor-alpha single nucleotide gene polymorphisms in inflammatory bowel disease . World J Gastroenterol 2007 ; 13 : 2960 – 6 . Google Scholar CrossRef Search ADS PubMed 35. McKaig BC , McWilliams D , Watson SA , Mahida YR . Expression and regulation of tissue inhibitor of metalloproteinase-1 and matrix metalloproteinases by intestinal myofibroblasts in inflammatory bowel disease . Am J Pathol 2003 ; 162 : 1355 – 60 . Google Scholar CrossRef Search ADS PubMed 36. Sadler T , Bhasin JM , Xu Y et al. Genome-wide analysis of DNA methylation and gene expression defines molecular characteristics of Crohn’s disease–associated fibrosis . Clin Epigenetics 2016 ; 8 : 30 . Google Scholar CrossRef Search ADS PubMed 37. Lawrance IC , Maxwell L , Doe W . Altered response of intestinal mucosal fibroblasts to profibrogenic cytokines in inflammatory bowel disease . Inflamm Bowel Dis 2001 ; 7 : 226 – 36 . Google Scholar CrossRef Search ADS PubMed 38. Shin TM , Bordeaux JS . The role of massage in scar management: a literature review . Dermatol Surg 2012 ; 38 : 414 – 23 . Google Scholar CrossRef Search ADS PubMed 39. Rieder F , Zimmermann EM , Remzi FH , Sandborn WJ . Crohn’s disease complicated by strictures: a systematic review . Gut 2013 ; 62 : 1072 – 84 . Google Scholar CrossRef Search ADS PubMed 40. Yamamoto T , Fazio VW , Tekkis PP . Safety and efficacy of strictureplasty for Crohn’s disease: a systematic review and meta-analysis . Dis Colon Rectum 2007 ; 50 : 1968 – 86 . Google Scholar CrossRef Search ADS PubMed 41. Lo CM , Wang HB , Dembo M , Wang YL . Cell movement is guided by the rigidity of the substrate . Biophys J 2000 ; 79 : 144 – 52 . Google Scholar CrossRef Search ADS PubMed 42. Levental KR , Yu H , Kass L et al. Matrix crosslinking forces tumor progression by enhancing integrin signaling . Cell 2009 ; 139 : 891 – 906 . Google Scholar CrossRef Search ADS PubMed 43. Barker HE , Chang J , Cox TR et al. LOXL2-mediated matrix remodeling in metastasis and mammary gland involution . Cancer Res 2011 ; 71 : 1561 – 72 . Google Scholar CrossRef Search ADS PubMed 44. Lin TH , Aplin AE , Shen Y et al. Integrin-mediated activation of MAP kinase is independent of FAK: evidence for dual integrin signaling pathways in fibroblasts . J Cell Biol 1997 ; 136 : 1385 – 95 . Google Scholar CrossRef Search ADS PubMed 45. Liu SB , Ikenaga N , Peng ZW et al. Lysyl oxidase activity contributes to collagen stabilization during liver fibrosis progression and limits spontaneous fibrosis reversal in mice . FASEB J 2016 ; 30 : 1599 – 609 . Google Scholar CrossRef Search ADS PubMed 46. Meissner EG , McLaughlin M , Matthews L et al. Simtuzumab treatment of advanced liver fibrosis in HIV and HCV-infected adults: results of a 6-month open-label safety trial . Liver Int 2016 ; 36 : 1783 – 92 . Google Scholar CrossRef Search ADS PubMed 47. Muir AG , Goodman Z , Levy C et al. Efficacy and safety of simtuzumab for the treatment of primary sclerosing cholangitis: results of a phase 2b, dose-ranging, randomized, placebo-controlled trial . J Hepatol 2017 ; 66 : S73 . Google Scholar CrossRef Search ADS 48. Sanyal AA , Abdelmalek MF , Diehl AM et al. Efficacy and safety of simtuzumab for the treatment of nonalcoholic steatohepatitis with bridging fibrosis or cirrhosis: results of two phase 2b, dose-ranging, randomized, placebo-controlled trials . J Hepatol 2017 ; 66 : S54 . Google Scholar CrossRef Search ADS Copyright © 2018 European Crohn’s and Colitis Organisation (ECCO). Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Crohn's and Colitis Oxford University Press

Fibrostenotic Phenotype of Myofibroblasts in Crohn’s Disease is Dependent on Tissue Stiffness and Reversed by LOX Inhibition

Journal of Crohn's and Colitis , Volume Advance Article (7) – Apr 16, 2018

Loading next page...
 
/lp/ou_press/fibrostenotic-phenotype-of-myofibroblasts-in-crohn-s-disease-is-Urrzv58y5t
Publisher
Oxford University Press
Copyright
Copyright © 2018 European Crohn’s and Colitis Organisation (ECCO). Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com
ISSN
1873-9946
eISSN
1876-4479
D.O.I.
10.1093/ecco-jcc/jjy036
Publisher site
See Article on Publisher Site

Abstract

Abstract Background and Aims Crohn’s disease is a chronic inflammatory disorder of the intestine and often leads to fibrosis, characterized by excess extracellular matrix [ECM] deposition, increased tissue stiffness, and stricture formation. Here we evaluated the contribution of myofibroblast–ECM interactions to the development of intestinal fibrosis in Crohn’s disease. Methods Matched primary human myofibroblasts were isolated from stenotic, inflamed and normal-appearing small intestine within the same Crohn’s disease patient [n = 10]. Cells were analyzed by gene expression profiling, microscopy and functional assays, including matrix metalloproteinase [MMP] production and ECM contraction. Results We demonstrated that myofibroblasts isolated from stenotic intestine differed both in phenotype and function from those isolated from purely inflammatory or normal-appearing intestine of the same patient. Stenotic myofibroblasts displayed increased expression of genes associated with ECM modulation and collagen deposition. Upon culture in a fibrotic environment, normal myofibroblasts increased expression of MMPs to counteract the mechanical force exerted by the matrix. Interestingly, stenotic myofibroblasts showed a paradoxical response with decreased expression of MMP3. In addition, stenotic myofibroblasts expressed increased levels of the collagen crosslinking enzyme lysyl oxidase [LOX] and induced significantly more ECM contraction than both normal and inflamed myofibroblasts. Importantly, LOX inhibition completely restored MMP3 activity in stenotic myofibroblasts grown in a fibrotic environment, and prevented excessive ECM contraction. Conclusions Together these data indicate aberrancies in the myofibroblast–ECM interaction in Crohn’s disease, and identify LOX inhibition as a potential anti-fibrotic agent in this condition. Fibrosis, Crohn’s disease, extracellular matrix, lysyl oxidase 1. Introduction At diagnosis, most patients with Crohn’s disease present with a primarily inflammatory disease phenotype. Around 5% of patients have a stricture within 90 days of diagnosis,1 but over time this increases up to 30%.2 Although the therapeutic armamentarium for Crohn’s disease has improved significantly over the past decade, this has not led to a reduction in the volume of surgery performed for intestinal strictures,3 indicating a clear unmet need for these patients. Fibrosis is characterized by the formation of excess amounts of extracellular matrix [ECM] components such as collagen and fibronectin.4 In the intestine, the end stage of fibrosis is stricture formation with narrowing and obstruction of the lumen. Extracellular matrix production is a normal component of tissue repair, but can evolve into an irreversible, progressive fibrotic response if tissue injury is severe or if there are repeated cycles of inflammation—as is the case in Crohn’s disease. In line with this, intestinal inflammation is a necessary precondition for the initiation of fibrosis, but once the fibrotic process has started, progression may occur independently of inflammation.5 In normal wound healing, ECM deposition is regulated by production of its components and the activity of modulating enzymes, including matrix metalloproteinases [MMPs], which cleave ECM proteins. However, a dysregulation of MMP and/or ECM production results in fibrosis instead of normal wound healing.6,7 Accumulation of ECM components, of which collagen is most abundant, can lead to increased tissue stiffness, which in turn activates myofibroblasts, leading to further enhanced ECM production,8,9 thus resulting in a vicious circle. In a rat model for liver fibrosis, as well as in patients with hepatitis C, it has been shown that increased tissue stiffness precedes fibrosis,10,11 highlighting the importance of the ECM and tissue rigidity in the fibrotic process. Previous studies on the role of fibroblasts in Crohn’s disease have mainly compared the fibroblasts of patients with Crohn’s disease with those of control patients,12–14 or made use of the human colonic fibroblast cell line ccd18-co.5,15 One study compared stenotic and normal tissue–derived fibroblasts within the same patients, with a focus on transforming growth factor-β–driven mechanisms,6 and another focused on the role of microRNA in fibrosis.16 Unfortunately, these studies did not include fibroblasts from inflamed ileum. Because of the sequential order in Crohn’s disease, in which normal ileum evolves into inflamed ileum and eventually into stenotic ileum, we aimed to investigate the phenotype and function of myofibroblasts isolated from stenotic ileum compared with the phenotype and function of myofibroblasts from inflamed and normal-appearing ileum from the same patients. Furthermore, we used an unbiased approach and analyzed the full transcriptome of these cells. Our results show that human stenotic ileal myofibroblasts have a phenotype that is distinct from myofibroblasts isolated from normal-appearing or inflamed [but not stenotic] areas of the same patient. Specifically, stenotic myofibroblasts show a paradoxical response to increased tissue stiffness, with decreased expression of MMP3 activity upon increased stiffness. In addition, stenotic myofibroblasts express increased levels of the collagen crosslinking enzyme lysyl oxidase [LOX], further contributing to tissue rigidity. LOX inhibition completely reverses this behavior and prevents matrix contraction by stenotic myofibroblasts. These results suggest that LOX inhibition may be a possible anti-fibrotic therapy in Crohn’s disease. 2. Methods 2.1. Patient selection Myofibroblasts were isolated from the ileum of 10 Crohn’s disease patients undergoing resection of symptomatic ileal strictures in the Academic Medical Center, the Netherlands [patient characteristics in Table 1]. The local medical ethical committee [IRB] approved the study. All patients provided signed informed consent. Table 1. Baseline characteristics of patients included in the study. Gender [% female] 60% Age at diagnosis [years, mean +- SD] 22.2 [5.7] Age at surgery [years, median, IQR] 27.6 [22.7 – 33.3] Smoking [% yes] 0 Disease location [%]  Ileum 80%  Ileum + colon 20% Disease behaviour [%]  Stricturing 60%  Stricturing + penetrating 40% Medication at time of surgery [%]  None 20%  5-ASA 10%  Budesonide 10%  Antibiotics 10%  Azathioprine 20%  Anti-TNF 10%  Combination therapy azathioprine + anti-TNF 20% Gender [% female] 60% Age at diagnosis [years, mean +- SD] 22.2 [5.7] Age at surgery [years, median, IQR] 27.6 [22.7 – 33.3] Smoking [% yes] 0 Disease location [%]  Ileum 80%  Ileum + colon 20% Disease behaviour [%]  Stricturing 60%  Stricturing + penetrating 40% Medication at time of surgery [%]  None 20%  5-ASA 10%  Budesonide 10%  Antibiotics 10%  Azathioprine 20%  Anti-TNF 10%  Combination therapy azathioprine + anti-TNF 20% 5-ASA: 5 aminosalicylic acid; anti-TNF: anti-tumor necrosis factor alpha. View Large Table 1. Baseline characteristics of patients included in the study. Gender [% female] 60% Age at diagnosis [years, mean +- SD] 22.2 [5.7] Age at surgery [years, median, IQR] 27.6 [22.7 – 33.3] Smoking [% yes] 0 Disease location [%]  Ileum 80%  Ileum + colon 20% Disease behaviour [%]  Stricturing 60%  Stricturing + penetrating 40% Medication at time of surgery [%]  None 20%  5-ASA 10%  Budesonide 10%  Antibiotics 10%  Azathioprine 20%  Anti-TNF 10%  Combination therapy azathioprine + anti-TNF 20% Gender [% female] 60% Age at diagnosis [years, mean +- SD] 22.2 [5.7] Age at surgery [years, median, IQR] 27.6 [22.7 – 33.3] Smoking [% yes] 0 Disease location [%]  Ileum 80%  Ileum + colon 20% Disease behaviour [%]  Stricturing 60%  Stricturing + penetrating 40% Medication at time of surgery [%]  None 20%  5-ASA 10%  Budesonide 10%  Antibiotics 10%  Azathioprine 20%  Anti-TNF 10%  Combination therapy azathioprine + anti-TNF 20% 5-ASA: 5 aminosalicylic acid; anti-TNF: anti-tumor necrosis factor alpha. View Large 2.2. Myofibroblast isolation Myofibroblasts were isolated from the lamina propria of normal-, inflamed-only– and stenotic-appearing ileum of the same patient [method adapted from that of reference Owens et al.17]. Surgical specimens were cut open and fecal content was removed. The mucosa was separated from the submucosa mechanically and subsequently washed extensively in ice-cold PBS supplemented with 1% penicillin/streptomycin and 40 μg/ml G418 [PGA]. The mucosa was chopped up very finely, placed in RMPI-1640 [Invitrogen] supplemented with 1.5 mg/mL collagenase A [Roche, Germany], minced using the Gentlemacs Dissociator [Miltenyi Biotec, Leiden, the Netherlands], incubated at 37°C for 60 min, and further dissociated using another cycle of Gentlemacs. Cells were washed extensively with PGA before plating in RMPI-1640 with 10% FCS, 1% penicillin/streptomycin, 1% L-glutamin, G418 [40 μg/ml; Lonza, Leusden, the Netherlands] and amphotericin B [0.025 μg/ml; Gibco, Rockford, IL]. Importantly, myofibroblasts of normal-, inflamed-only– and stenotic-appearing ileum were paired from the same patient and at the same passage with the same cell density for individual experiments. Primary myofibroblasts were used at Passages 1–4. Ccd18-co colonic human fibroblast cell line was obtained from ATCC [Manassas, VA] cultured in DMEM [Lonza] supplemented with 10% FCS, 1% penicillin/streptomycin, 1% L-glutamin, G418 [40 μg/ml] and amphotericin B [0.025 μg/ml]. All cells tested negative for Mycoplasma. Cell proliferation of myofibroblasts from normal-, inflamed-only– and stenotic-appearing ileum was measured by plating 50 × 10E3 cells/well in a flat-bottomed 96-well plate. 3H-thymidine was added for 72 h and incorporation measured using a MicroBeta 2 Microplate Counter [Perkin Elmer, Waltham MA]. Additionally, proliferation was measured by impedance measurement using the xCelligence system as described before.18 For LOX inhibition experiments, cell cultures were treated with β-aminopropionitrile [βAPN, Sigma Aldrich], which blocks both LOX and LOXL219 at 500 μM concentration, for 3 days. For inhibiting focal adhesion kinase [FAK], cell cultures were treated with the FAK inhibitor Defactinib [10 μM, Selleckchem, Houston, TX] for 3 days. 2.3. RNA isolation, complementary DNA synthesis, quantitative reverse-transcription polymerase chain reaction and transcription analysis RNA was isolated using the RNAeasy mini kit [Qiagen, Valencia, CA]. Quantative RT–PCR was performed using SybrGreen [Roche] according to the manufacturer’s protocol on a BioRad iCycler. Glyceraldehyde-3-phosphate dehydrogenase [GAPDH] was used as the housekeeping gene [Supplementary Table 1 for primer sequences]. RNA was amplified using a TotalPrep RNA amplification kit for Illumina [Ambion] and labeled using a cRNA labeling kit for Illumina Arrays [Ambion], followed by hybridization with Illumina Human HT-12_v4 BeadChip slides. Expression profiles were deposited in the GEO repository [GSE90607]. Initial normalization was performed using Genome studio software [Illumina], gene set enrichment analyses were done using GSA software [Broad Institute of MIT and Harvard] and pathway analysis was done using Qiagen’s Ingenuity Pathway Analysis software [IPA, Qiagen Redwood City]. 2.4. Immunohistochemistry and immunofluorescence Paraffin-embedded tissue sections [4 μm] were deparaffinized, rehydrated and immersed in 0.3% H2O2 in methanol for 30 min. After antigen retrieval by sodium citrate for LOX staining and Tris-EDTA for CD3 staining, slides were blocked by PBS with 0.1% Triton X-100 and 1% bovine serum albumin [PBT] for 30 min, followed by incubation with primary antibody overnight at 4°C (rabbit polyclonal anti-human LOX [ab31238 Abcam, Cambridge, UK], rabbit monoclonal anti-human CD3 [clone SP7, Thermo Scientific] and goat polyclonal collagen I [Southern Biotech, Birmingham, AL, USA]) in PBT. Slides were incubated with Brightvision [Immunologic, Duiven, the Netherlands], counterstained with Haematoxylin and mounted. For immunofluorescence, cell cultures were fixed using 4% PFA, permeabilized using 0.1% Triton X, and stained using rabbit monoclonal anti-focal adhesion kinase [clone EP695Y, Abcam] followed by secondary donkey anti-rabbit 546 [Invitrogen, Carlsbad, CA] and phalloidin alexa fluor 488 [Thermo Scientific]. Samples were embedded in SlowFade Gold containing DAPI [Invitrogen] and imaged using a Leica DM6000B microscope. For flow cytometry, cells were isolated, fixed in 2% PFA, permeabilized in 0.1% Triton X/5 mM EDTA in PBS and stained using anti-desmin, anti-vimentin and anti-alpha-sma, followed by anti-rabbit-AF546 and anti-goat-AF647. Analysis was performed using a FACS Fortessa [BD Biosciences, Bedford, MA] and FlowJo software [FlowJo Inc. Ashland, Oregon]. 2.5. Protein analysis Matrix metalloproteinase activity was measured in the supernatant of myofibroblasts cultured in RPMI1640 without phenol red [Lonza], supplemented with 1% FCS, 1% penicillin/streptomycin, 1% L-glutamin, G418 [40 μg/ml] and amphotericin B [0.025 μg/ml]. Global MMP activity was assessed by adding 20 nmol/L OmniMMP substrate [Mca-Pro-Leu-Gly-Leu-Dpa-Ala-Arg-NH; Enzo Life Science, Lausen, Swiss] [in an appropriate assay buffer] to the supernatant. For global MMP activity, this ×10 buffer was 500 mM HEPES, 100 mM CaCl2, 0.5% Brij-35, pH 7.020; for more specific MMP2 activity, the ×10 buffer consisted of 500 mM MOPS, 100 mM CaCl2, 100 μM ZnCl2, 0.5% Brij-35, pH 7.0, and for even more specific MMP3 activity, the ×10 buffer consisted of 500 mM MES, 100 mM CaCl2, 0.5% Brij-35, pH 6.0, all diluted with cell culture supernatant. Fluorescence was detected by Novostar 700-0055 apparatus [BMG Labtech, Ortenberg, Germany] at a 320/405 wavelength by kinetic measurements every 100 s. Results are expressed as relative fluorescence increase compared with fluorescence intensity at the moment of stable background levels. LOXL2 and TGFb1 protein concentrations were measured using commercial human Duoset kits [R&D Systems, Minneapolis, MN]. 2.6. Stiffness-controlled extracellular matrix gels Tissue stiffness experiments were performed as described before.21 In brief, round glass coverslips were treated with HCl, 3-aminopropyltriethoxysilane and glutaraldehyde each time, followed by extensive washing with PBS. Collagen-coated polyacrylamide gels of 2.8 and 30 kPa were generated using varying ratios of 40% acrylamide and 2% bisacrylamide [Bio-rad], supplemented with 0.6% ammonium persulfate and 0.4% TEMED. Acrylamide gels were polymerized on treated coverslips by inverting the coverslips onto the acrylamide/bisacrylamide solution. After polymerization, coverslips were rinsed extensively with PBS. The acrylamide gels were crosslinked with 2.5 mg/mL sulfo-SANPAH [Thermo Scientific, Rockford, IL] at a wavelength of 365 nm in a Stratagene UV crosslinker oven [Stratagene, La Jolla, CA] delivering 7500 J energy in total, followed by UV sterilization. The coverslip-gels were placed in a 12-well tissue culture plate and coated with 2 mg/mL rat tail collagen I [BD Biosciences]. After 4 h incubation, cells were seeded at a density of 40 000 cells/well. For all experiments, paired cells were cultured for 3 days. 2.7. Extracellular matrix contraction assay To assess force-mediated matrix remodeling,22 150 000 myofibroblasts were embedded in 50 μL matrigel/collagen I mix, yielding a final collagen concentration [BD Bioscience] of 4.6 mg/mL and a final Matrigel concentration [VWR, Radnor, PA] of 2.2 mg/mL, and seeded on a 35-mm glass-bottom dish [MatTek, Ashland, MA]. After 30 min polymerization, cells were maintained in full growth medium with or without LOX inhibitor. Images were obtained daily for 7 days, and the diameter of the gel was measured using ImageJ software. Data are shown as the decrease of the circumference of the gel relative to Day 0. 2.8. Statistical analysis Data were analyzed using Graphpad Prism 7.0 [Graphpad Software Inc., La Jolla, CA]. The Kruskal–Wallis test with post-hoc analysis, or the Mann–Whitney-U test, were performed to compare data. Extracellular matrix contraction assays were analyzed as area under the curve, with a paired t-test afterwards. A P-value of <0.05 was considered statistically significant. 3. Results 3.1. Stenotic myofibroblasts differed in ECM organization and collagen production In order to investigate the phenotype and function of myofibroblasts from different regions of Crohn’s disease intestinal tissue, cells were isolated from patients undergoing surgery for stricturing disease of the ileum. The ileum of all these patients was considerably narrowed and the bowel wall was thickened. Macroscopically non-inflamed as well as non-stenotic [henceforth referred to as ‘normal’], inflamed-only and stenotic regions were identified within each individual patient [Figure 1a], and cells were isolated from all three areas separately. The nature of the regions used for isolation was confirmed by immunohistochemistry [Figure 1b]. Indeed, areas designated as ‘inflamed-only’ displayed significantly more CD3+ cells than either ‘normal’ or ‘stenotic’ regions. Areas designated ‘stenotic’ showed increased presence of ECM compared with normal and inflamed-only areas. Numbers of CD3+ cells in stenotic regions were intermediate between ‘normal’ and ‘inflamed-only’, in line with an expected mixed fibrotic/inflammatory phenotype. Of note, some infiltration of CD3+ cells was observed in the ‘normal’ areas, representing the gradient that occurs in the inflammatory response seen in Crohn’s disease patients. No apparent differences could be observed in cellular morphology, nor in expression of [myo]fibroblast markers alpha smooth muscle actin, desmin and vimentin [Supplementary figure 1]. Myofibroblasts isolated from inflamed-only areas proliferated much faster than myofibroblasts from normal and stenotic mucosa, as quantified both by thymidine incorporation assay and xCelligence impedance measurements [Figure 1c]. This effect remained stable over multiple passages [data not shown], excluding the confounding effect of inflammatory cytokines present directly after myofibroblast isolation. Interestingly, myofibroblasts obtained from stenotic regions displayed proliferation similar to normal myofibroblasts. Figure 1. View largeDownload slide Stenotic myofibroblasts differ from inflamed-only and normal myofibroblasts in extracellular matrix [ECM] organization and collagen production. [A] Macroscopic and low-power image of location selection. Resection specimen of Crohn’s ileum with narrowing at site of stenosis. Arrows indicate macroscopically normal, inflamed-only and stenotic regions. Low-power image of H&E staining; all magnifications: ×40. [B] Microscopic analysis of macroscopically normal, inflamed-only and stenotic mucosa by immunohistochemistry for CD3 and collagen I; bottom row depicts zoom of collagen I staining. [C] Proliferation of myofibroblasts as measured by xCelligence [upper graph, n = 5 individual patients, representative data shown] and 3H thymidine incorporation assay [lower graph, summarized data of eight experiments in individual patients, bars represent mean, error bars s.e.m.]. [D–E] mRNA was isolated from normal, inflamed-only and stenotic myofibroblasts obtained from seven individual patients, and gene expression was analyzed by Gene Set Enrichment Analysis. Vertical lines below x-axis display relative distribution of expression per gene included in the geneset. [F] Expression patterns of individual genes present in the Reactome geneset ‘ECM organization’. Paired samples of each condition were obtained from seven individual patients. Each line represents an individual sample. Expression is depicted as Z-score within the individual gene. Figure 1. View largeDownload slide Stenotic myofibroblasts differ from inflamed-only and normal myofibroblasts in extracellular matrix [ECM] organization and collagen production. [A] Macroscopic and low-power image of location selection. Resection specimen of Crohn’s ileum with narrowing at site of stenosis. Arrows indicate macroscopically normal, inflamed-only and stenotic regions. Low-power image of H&E staining; all magnifications: ×40. [B] Microscopic analysis of macroscopically normal, inflamed-only and stenotic mucosa by immunohistochemistry for CD3 and collagen I; bottom row depicts zoom of collagen I staining. [C] Proliferation of myofibroblasts as measured by xCelligence [upper graph, n = 5 individual patients, representative data shown] and 3H thymidine incorporation assay [lower graph, summarized data of eight experiments in individual patients, bars represent mean, error bars s.e.m.]. [D–E] mRNA was isolated from normal, inflamed-only and stenotic myofibroblasts obtained from seven individual patients, and gene expression was analyzed by Gene Set Enrichment Analysis. Vertical lines below x-axis display relative distribution of expression per gene included in the geneset. [F] Expression patterns of individual genes present in the Reactome geneset ‘ECM organization’. Paired samples of each condition were obtained from seven individual patients. Each line represents an individual sample. Expression is depicted as Z-score within the individual gene. To further evaluate similarities and differences between normal and stenotic myofibroblasts, we performed transcriptional analysis on myofibroblast populations isolated from eight individual patients. When comparing normal and stenotic myofibroblasts, pathway analysis using Ingenuity software revealed alterations in pathways involved in cell–cell and cell–matrix interactions, including fibrotic responses and MMPs [Table 2]. Gene set enrichment analysis [GSEA] showed significant enrichment of genes associated with ECM organization and collagen production in stenotic myofibroblasts compared with in those from normal tissue regions [Figure 1d]. As stenotic areas did display varying degrees of inflammation, stenotic myofibroblasts were also compared with those obtained from the inflammation-only areas. Again, genes associated with ECM organization and collagen production were increased in the stenotic myofibroblasts, suggesting the results are not merely the consequence of residual inflammation in the stenotic areas [Figure 1e and f]. Collectively, we showed that myofibroblasts isolated from stenotic, inflamed-only and normal-appearing regions of the ileum within the same patient have different phenotypes and that stenotic myofibroblasts are distinct from both inflamed-only and normal myofibroblasts in terms of ECM organization and collagen production. Table 2. Top 10 canonical Ingenuity pathways differentially activated between normal and stenotic myofibroblasts. Ingenuity canonical pathways -log[p-value] Hepatic fibrosis / Hepatic stellate cell activation 7.99E+00 Retinoate biosynthesis I 3.56E+00 Bile acid biosynthesis, neutral pathway 3.28E+00 Altered T cell and B cell signaling in rheumatoid arthritis 3.25E+00 Inhibition of matrix metalloproteases 2.99E+00 Agrin interactions at neuromuscular junction 2.97E+00 Agranulocyte adhesion and diapedesis 2.94E+00 Atherosclerosis signaling 2.89E+00 IL-8 signaling 2.76E+00 Granulocyte adhesion and diapedesis 2.71E+00 Ingenuity canonical pathways -log[p-value] Hepatic fibrosis / Hepatic stellate cell activation 7.99E+00 Retinoate biosynthesis I 3.56E+00 Bile acid biosynthesis, neutral pathway 3.28E+00 Altered T cell and B cell signaling in rheumatoid arthritis 3.25E+00 Inhibition of matrix metalloproteases 2.99E+00 Agrin interactions at neuromuscular junction 2.97E+00 Agranulocyte adhesion and diapedesis 2.94E+00 Atherosclerosis signaling 2.89E+00 IL-8 signaling 2.76E+00 Granulocyte adhesion and diapedesis 2.71E+00 IL: Interleukin. View Large Table 2. Top 10 canonical Ingenuity pathways differentially activated between normal and stenotic myofibroblasts. Ingenuity canonical pathways -log[p-value] Hepatic fibrosis / Hepatic stellate cell activation 7.99E+00 Retinoate biosynthesis I 3.56E+00 Bile acid biosynthesis, neutral pathway 3.28E+00 Altered T cell and B cell signaling in rheumatoid arthritis 3.25E+00 Inhibition of matrix metalloproteases 2.99E+00 Agrin interactions at neuromuscular junction 2.97E+00 Agranulocyte adhesion and diapedesis 2.94E+00 Atherosclerosis signaling 2.89E+00 IL-8 signaling 2.76E+00 Granulocyte adhesion and diapedesis 2.71E+00 Ingenuity canonical pathways -log[p-value] Hepatic fibrosis / Hepatic stellate cell activation 7.99E+00 Retinoate biosynthesis I 3.56E+00 Bile acid biosynthesis, neutral pathway 3.28E+00 Altered T cell and B cell signaling in rheumatoid arthritis 3.25E+00 Inhibition of matrix metalloproteases 2.99E+00 Agrin interactions at neuromuscular junction 2.97E+00 Agranulocyte adhesion and diapedesis 2.94E+00 Atherosclerosis signaling 2.89E+00 IL-8 signaling 2.76E+00 Granulocyte adhesion and diapedesis 2.71E+00 IL: Interleukin. View Large 3.2. Paradoxical response to tissue stiffness of myofibroblasts in MMP3 activity Transcriptional analysis suggested that several MMP transcripts were significantly more highly expressed in myofibroblasts from stenotic ileum compared with in myofibroblasts from both inflamed-only and normal ileum [Figure 1f]. This is counterintuitive, as MMPs are endopeptidases involved in the degradation of ECM proteins, which cause stenosis, so it would be reasonable to expect reduced MMP activity in stenotic lesions in patients with Crohn’s disease. Indeed, previously published studies have shown that MMP protein expression and activity levels decreased in mucosa overlying stenotic ileum in vivo.6 We corroborated the transcriptional data using enzyme activity assays, which not only measure MMP protein levels but also take into account the effects of their natural inhibitors [TIMPs]. In accordance with the transcriptional data, MMP activity was increased in stenotic myofibroblasts compared with in both normal and inflamed-only myofibroblasts when cultured in the absence of ECM [Figure 2a]. This was seen when analyzing global MMP activity as well as more specific MMP2 and MMP3 activity. Figure 2. View largeDownload slide Matrix metalloproteinase [MMP] activity in myofibroblasts isolated from normal, inflamed-only and stenotic ileal regions. [A] Isolated primary myofibroblasts were cultured in the absence of ECM for 3 days, and MMP activity was measured in the supernatant using a quenched fluorescent probe. Representative graphs of experiments in three separate patients are shown. [B–C] Primary myofibroblasts [B] or the fibroblast cell line ccd18-co [C] were cultured in matrix with a normal [2.8 kPa, blue] or high [30 kPa, red] compliance for 3 days. MMP3 activity was measured in the supernatant. Representative data of four experiments are shown. Figure 2. View largeDownload slide Matrix metalloproteinase [MMP] activity in myofibroblasts isolated from normal, inflamed-only and stenotic ileal regions. [A] Isolated primary myofibroblasts were cultured in the absence of ECM for 3 days, and MMP activity was measured in the supernatant using a quenched fluorescent probe. Representative graphs of experiments in three separate patients are shown. [B–C] Primary myofibroblasts [B] or the fibroblast cell line ccd18-co [C] were cultured in matrix with a normal [2.8 kPa, blue] or high [30 kPa, red] compliance for 3 days. MMP3 activity was measured in the supernatant. Representative data of four experiments are shown. This remarkable discrepancy might originate from a different physical environment in stenotic lesions compared with that in normal ileum, with high tissue stiffness in fibrostenosis. It has previously been shown that normal ileum has a compliance of ~2.8 kPa, which increases up to 30 kPa in the stenotic region of a Crohn’s ileum.23 The effect of stiffness can be mimicked in vitro by embedding cells in gels containing varying amounts of ECM components.21 As expected, myofibroblasts isolated from normal-appearing mucosa increased MMP3 activity when cultured in stiff 30 kPa conditions, indicating that an increase in tissue stiffness activates a compensatory activation of MMPs in normal myofibroblasts. In sharp contrast, the MMP3 activity of stenotic myofibroblasts was low in a matrix that mimicked their native environment [30 kPa] and paradoxically increased when environmental pressure was decreased to 2.8 kPa [Figure 2b]. Interestingly, we found that the human colonic fibroblast cell line ccd18-co had the same MMP3 activity pattern, with low activity in a stiff environment and a paradoxical increase in MMP3 activity in soft matrix, indicatingthat this cell line may display a more stenotic phenotype [Figure 2c]. Collectively our data show that normal myofibroblasts increase their MMP3 activity in stiff matrix, presumably in order to degrade ECM, and counteract the increased compliance. In contrast, stenotic myofibroblasts display an aberrant response to a stiff tissue environment, with reduced MMP3 activity. 3.3. Enhanced ECM contraction in stenotic myofibroblasts in vitro In vivo, intestinal fibrosis is characterized by excess ECM production, of which collagen is one of the major components.4 Increased collagen deposition leads to tissue scarring and subsequent contraction of the tissue, and may result in stenosis.24 Indeed stenotic myofibroblasts displayed increased expression of COL1A1, COL5A1 and COL11A1 compared with cells isolated from the normal but not the inflamed-only regions [Figure 3a]. The functional capacity of stenotic myofibroblasts to induce ECM contraction in vitro, using the collagen/matrigel contraction assay, is a validated method for measuring cellular contractility within an ECM.22 Cells are cultured in a gel matrix that contracts upon collagen deposition [similar to the contraction seen during fibrosis in vivo]. In this assay, stenotic myofibroblasts caused substantially more ECM contraction than myofibroblasts isolated from either inflamed-only or normal-appearing mucosa from the same patient [Figure 3b], consistent with tissue contraction of stenotic tissue in vivo. The ccd18-co cell line caused contraction of the ECM as well, again showing a striking resemblance to the phenotypic characteristics of stenotic myofibroblasts [Figure 3c]. Figure 3. View largeDownload slide Enhanced extracellular matrix [ECM] contraction by stenotic myofibroblasts. [A] Expression of COL1A1, COL5A1 and COLL11A1 in myofibroblasts isolated from stenotic, inflamed-only or normal regions as measured by qPCR and normalized to GAPDH. Dots represent individual patients. [B] Myofibroblasts were embedded in a matrigel/collagen I mix and cultured for 7 days. Contraction was measured as percentage decrease in matrix surface. Representative graphs and images of three independent experiments. Error bars represent median and standard error. Asterisk indicates p ≤ 0.05. Figure 3. View largeDownload slide Enhanced extracellular matrix [ECM] contraction by stenotic myofibroblasts. [A] Expression of COL1A1, COL5A1 and COLL11A1 in myofibroblasts isolated from stenotic, inflamed-only or normal regions as measured by qPCR and normalized to GAPDH. Dots represent individual patients. [B] Myofibroblasts were embedded in a matrigel/collagen I mix and cultured for 7 days. Contraction was measured as percentage decrease in matrix surface. Representative graphs and images of three independent experiments. Error bars represent median and standard error. Asterisk indicates p ≤ 0.05. 3.4. LOX inhibition restored MMP3 activity and decreased ECM contraction in stenotic myofibroblasts Collagen is produced and secreted as a soluble protein. After secretion, crosslinking of individual fibrils results in collagen deposition and consequently increased tissue stiffness.25 Key enzymes in this process are the LOX family members LOX, lysyl oxidase-like-1 [LOXL1] and lysyl oxidase-like-2 [LOXL2]. These enzymes are primarily responsible for the covalent crosslinking of collagen and elastin in the ECM.26 In addition, LOX can interact with fibronectin, another ECM component. In turn, fibronectin can increase LOX activity, leading to a positive feedback loop with further collagen crosslinking and tissue stiffness.27,28 Enhanced LOX activity may thus be associated with increased formation of stenotic lesions. Interestingly, on the transcriptional level, LOX and LOXL2 expression were increased in myofibroblasts isolated from stenotic ileum, with a similar trend for LOXL1 [Figure 4a]. Likewise, secretion of LOXL2 protein was significantly higher in stenotic myofibroblasts than in normal myofibroblasts [Figure 4b]. Finally, immunohistochemistry confirmed increased expression of LOX in stenotic regions in vivo [Figure 4c]. Figure 4. View largeDownload slide Enhanced lysyl oxidase [LOX] expression in stenotic myofibroblasts. [A] Expression of LOX, LOX-like-1 [LOXL1] and LOX-like 2 [LOXL2] mRNA in primary myofibroblasts normalized to GAPDH. Dots represent individual patients. [B] Isolated myofibroblasts were cultured and LOXL2 in the supernatant measured by ELISA. n = 8 patients per condition. [C] Resection specimens from normal, inflamed-only and stenotic regions of the ileum were stained for LOX. Arrows indicate positive staining. Magnification: ×400. Single asterisk indicates p ≤ 0.05; two asterisks indicate p ≤ 0.01; three asterisks indicate p ≤ 0.001. Figure 4. View largeDownload slide Enhanced lysyl oxidase [LOX] expression in stenotic myofibroblasts. [A] Expression of LOX, LOX-like-1 [LOXL1] and LOX-like 2 [LOXL2] mRNA in primary myofibroblasts normalized to GAPDH. Dots represent individual patients. [B] Isolated myofibroblasts were cultured and LOXL2 in the supernatant measured by ELISA. n = 8 patients per condition. [C] Resection specimens from normal, inflamed-only and stenotic regions of the ileum were stained for LOX. Arrows indicate positive staining. Magnification: ×400. Single asterisk indicates p ≤ 0.05; two asterisks indicate p ≤ 0.01; three asterisks indicate p ≤ 0.001. As we observed that stenotic myofibroblasts showed an aberrant suppression of MMP activity in response to a stiff microenvironment, altering the microenvironment might change the phenotype of myofibroblasts. Therefore, we hypothesized that blocking collagen crosslinking by inhibition of the LOX enzyme family might restore MMP activity in stenotic myofibroblasts, which could result in resolution of the accumulated ECM. Focal adhesion kinase [FAK] is a downstream signaling transducer of LOX which is important for cell–matrix interaction,29 and inhibiting LOX reduced expression of FAK, as seen by immunofluorescence staining [Figure 5a]. In line with our hypothesis, inhibiting LOX in ccd18-co fibroblasts [which show responses to environmental cues comparable with those of stenotic myofibroblasts] strongly increased the activity of MMP3 when cultured under stenotic conditions [30 kPa]. In fact, MMP3 activity under these conditions was comparable with that seen in myofibroblasts cultured under normal 2.8 kPa conditions [Figure 5b]. Similarly, in primary myofibroblasts isolated from stenotic mucosa, MMP3 activity increased when these myofibroblasts were cultured in a stiff matrix in the presence of the LOX inhibitor [Figure 5c]. Interestingly, direct inhibition of FAK using the small molecule Defactinib did not alter MMP3 activity [Figure 5d]. Figure 5. View largeDownload slide LOX inhibition increases matrix metalloproteinase 3 [MMP3] activity and decreases extracellular matrix [ECM] contraction induced by stenotic myofibroblasts. [A] Ccd18-co cells were cultured in the presence or absence of LOX inhibitor [βAPN, 500 μM] for 3 days. Cells were stained for actin [phalloidin, green] and FAK [red]. [B–C] Ccd18-co fibroblasts [B] or primary stenotic myofibroblasts [C] were cultured in matrix with a normal [2.8 kPa, blue] or high [30 kPa, red] compliance in the presence of LOX inhibitor for 3 days. MMP3 activity was measured in the supernatant. Representative graphs shown, n = 5 independent experiments for ccd18-co cells; n = 3 independent experiments for primary cells. [D] Ccd18-co fibroblasts were cultured in matrix with a normal [2.8 kPa, blue] or high [30 kPa, red] compliance in the presence of FAK inhibitor Defactinib [10 μM] for 3 days. MMP3 activity was measured in the supernatant. Representative graph shown, n = 3 independent experiments. [E–F] Ccd18-co [E] or primary stenotic [F] myofibroblasts were embedded in a matrigel/collagen I mix and cultured for 7 days in the presence of LOX inhibitor. Contraction was measured as percentage decrease in matrix surface. Left graph depicts representative experiment, right graph shows data of four [E] or two [F] individual experiments as area under the curve. Asterisk indicates p ≤ 0.05. Figure 5. View largeDownload slide LOX inhibition increases matrix metalloproteinase 3 [MMP3] activity and decreases extracellular matrix [ECM] contraction induced by stenotic myofibroblasts. [A] Ccd18-co cells were cultured in the presence or absence of LOX inhibitor [βAPN, 500 μM] for 3 days. Cells were stained for actin [phalloidin, green] and FAK [red]. [B–C] Ccd18-co fibroblasts [B] or primary stenotic myofibroblasts [C] were cultured in matrix with a normal [2.8 kPa, blue] or high [30 kPa, red] compliance in the presence of LOX inhibitor for 3 days. MMP3 activity was measured in the supernatant. Representative graphs shown, n = 5 independent experiments for ccd18-co cells; n = 3 independent experiments for primary cells. [D] Ccd18-co fibroblasts were cultured in matrix with a normal [2.8 kPa, blue] or high [30 kPa, red] compliance in the presence of FAK inhibitor Defactinib [10 μM] for 3 days. MMP3 activity was measured in the supernatant. Representative graph shown, n = 3 independent experiments. [E–F] Ccd18-co [E] or primary stenotic [F] myofibroblasts were embedded in a matrigel/collagen I mix and cultured for 7 days in the presence of LOX inhibitor. Contraction was measured as percentage decrease in matrix surface. Left graph depicts representative experiment, right graph shows data of four [E] or two [F] individual experiments as area under the curve. Asterisk indicates p ≤ 0.05. To test the hypothesis that LOX inhibition also functionally decreases ECM contraction, we cultured ccd18-co fibroblasts in the contraction assay. Upon LOX inhibition, ECM contraction decreased significantly in the gels with LOX inhibition compared with that in the control gels [Figure 5e]. As we showed earlier, myofibroblasts from stenotic ileum exhibited more ECM contraction than normal myofibroblasts [Figure 3a], but this phenotype was reversed upon culture in the presence of a LOX inhibitor [Figure 5f]. Importantly, the inhibitor had very limited effect on the ECM contraction levels of myofibroblasts from normal and inflamed-only mucosa [data not shown], showing that LOX activity specifically decreases tissue contractility in stenotic myofibroblasts. 4. Discussion These data show that myofibroblasts isolated from stenotic intestinal areas of patients with Crohn’s disease displayed a pathological phenotype when compared with myofibroblasts isolated from normal-appearing and inflamed-only areas from the same patient. Intriguingly, the pathologic phenotype depended at least partially on the nature of the surrounding matrix, suggesting modulation of the ECM may alter myofibroblast behavior and thus serve as a potential therapeutic target for stenotic complications of Crohn’s disease. For this study, we compared paired samples from within the same patient, and designated these as ‘normal’, ‘inflammation-only’ or ‘stenotic’, based on macroscopic appearance. Although subsequent histological analysis confirmed increased levels of CD3+ cells in ‘inflamed-only’ and increased ECM deposition in ‘stenotic’ areas, it has to be acknowledged that Crohn’s disease patients display gradients in the level of inflammation and stenosis. As a result, areas designated in this study as ‘normal’ still contain a certain level of inflammatory cells. In addition, ‘stenotic’ areas are not only stenotic, but also display considerable levels of inflammation, as is to be expected in these patients. However, despite this gradual rather than black-and-white scale between the different phenotypes, clear phenotypic and functional differences were observed between the three types of myofibroblasts, supporting our selection method in the context of Crohn’s disease. Caution has to be exerted when extending these findings to healthy ‘normal’ tissue [which is likely to contain less inflammatory cells] or non–IBD-related cases of stenosis. Earlier studies have aimed at identifying alterations in fibroblasts obtained from strictures in Crohn’s disease. For example, increased levels of vascular endothelial growth factor [VEGF] as well as N-Cadherin were described on the protein level.30,31 Indeed, we observed increased levels of CDH2 transcript, the gene encoding for N-Cadherin, while we did not observe alterations in VEGF expression. Furthermore, transcriptional profiling was performed earlier, comparing fibroblasts from strictured and non-strictured regions.32 Although this study did not include macroscopically normal regions from Crohn’s disease patients, a number of transcripts upregulated in this study were confirmed in our experiments. Additionally, previous studies reported increased transcription of the matrix-associated cytokine TGFβ, an important cytokine involved in matrix remodeling in stenotic intestinal regions.14 Interestingly, although total protein levels of TGFb1 did not differ between the various locations [Supplementary figure 2], upstream analysis of the transcriptome indicated a modestly enhanced TGFb1 activity in inflamed regions and significantly increased activity in stenotic regions [data not shown]. The ECM is a dynamic tissue component undergoing constant remodeling, mediated among others by MMPs. Defective MMP3 secretion due to genetic ablation results in enhanced sensitivity to colitis in a murine model,33 and Crohn’s disease patients carrying a single-nucleotide polymorphism in the gene encoding for MMP3 have a 20% higher risk of developing stenotic complications.34 Additionally, expression of MMP3 was decreased in mucosa overlying strictured intestine in Crohn’s disease, while expression of the MMP inhibitor TIMP1 was increased.6,35 Our data show that this defective MMP3 activity is not an intrinsic defect of stenotic myofibroblasts per se, but occurs specifically in the context of a stenotic ECM. The assays used in this study do not differentiate fully between the various MMP family members, but rather are based on conditions favoring one family member over the others. Consequently, we cannot exclude the possibility that the increased activity seen in the MMP3 assays is partly caused by the other stromelysins MMP10 and MMP11. Others have shown earlier that transcription of MMP12 was decreased in myofibroblasts isolated from stenotic regions, independent of the presence of ECM.6 This may indicate differential regulatory mechanisms between various MMP family members. A recent study described various alterations in DNA methylation patterns in myofibroblasts isolated from stenotic intestine,36 suggesting intrinsic and potentially long-lasting alterations. This is in line with our findings that some of the aberrancies remain throughout several passages in culture. It has also been shown previously that myofibroblasts isolated from stenotic regions of the small intestine display increased proliferative capacity,37 whereas we describe increased proliferative capacity in inflamed-only, but not in stenotic myofibroblasts. This difference may be the result of alternatively selected regions, as the former study specifically included regions simultaneously stenosed and inflamed. In our study, the stenotic region did show inflammatory activity, but this was considerably less than in the regions designated as inflamed. Interestingly, therapeutic application of ECM modulation has been known for decades in a more intuitive form. In dermatology, scar massage is anecdotally effective for the treatment of postsurgical scars.38 Similarly, intestinal fibrosis is currently treated through balloon endoscopy or surgical strictureplasty.39 During strictureplasty, the narrowed ileum is incised, thereby relieving local tissue pressure. Despite the fact that local myofibroblasts are left in situ, the recurrence rate at the site of a previous strictureplasty is only 3–10%.39,40 Although requiring further study, it is tempting to speculate that, also in this case, the release of local tissue pressures affects myofibroblast behavior. The effects of mechanical forces on a cellular level have been appreciated in the context of medical therapy only recently. For example, increased ECM deposition and subsequently stiffness within the tumor niche enhances tumor cell survival and promotes malignant progression.41,42 In this case, the enhanced ECM deposition is mediated by collagen crosslinking via members of the LOX family.43 Inhibiting LOX activity in a mouse model for breast carcinoma reduced collagen deposits as well as the number of focal adhesions and subsequently limited tumor progression.42 In line with this, we observed less intracellular FAK staining after inhibition of LOX. However, direct inhibition of FAK did not alter MMP3 activity, suggesting that tissue stiffness is not [exclusively] translated by integrin-mediated FAK signaling in intestinal stenotic myofibroblasts. Previous studies have indicated the presence of FAK-independent integrin signaling specifically in fibroblasts.44 Indeed, LOX inhibition as a mode of affecting ECM stiffness has been shown to limit liver fibrosis in vitro as well as in animal models,45,46 although it had no benefits in patients with primary sclerosing cholangitis as well as non-alcoholic steatohepatitis.47,48 In summary, we showed that myofibroblasts isolated from stenotic ileum were phenotypically and functionally different from those isolated from normal- and inflamed-only–appearing regions in the same patient, mainly with respect to ECM organization and collagen production. Myofibroblasts from stenotic ileum aberrantly suppressed expression of MMP3 in the stiff tissue environment that corresponds to an intestinal stenosis in vivo. Altering the physical properties of the microenvironment by LOX inhibition resulted in a release of MMP3 activity by stenotic myofibroblasts as well as decreased matrix contraction. These data indicate that modulation of the ECM through inhibition of LOX may be an effective anti-fibrotic therapy in Crohn’s disease. Funding This research was partially funded by a VICI grant number 91814605 from the Netherlands Organization for Scientific Research [to GvdB]. Conflict of Interest GvdB is currently an employee of GSK. Author Contributions All authors have made substantial contributions to the following: conception and design of the study [JdB, MW, GD, GvdB], acquisition of data [JdB, MW, JS, SM, VM], providing patient material [CB, WB, AtV], analysis and interpretation of data [JdB, MW, JS, GvdB], statistical analysis [JdB, MW], drafting the article [JdB, MW], critical revision of the manuscript for important intellectual content [JdB, MW, GD, GvdB] and final approval of the version to be submitted [all authors]. All authors had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Data deposit Expression profiles were deposited in the GEO repository [GSE90607]. Supplementary Data Supplementary data are available at ECCO-JCC online. Abbreviations: Abbreviations: ECM extracellular matrix FAK focal adhesion kinase LOX lysyl oxidase LOXL2 lysyl oxidase–like 2 MMP matrix metalloproteinase TIMP tissue inhibitor of metalloproteinase VEGF vascular endothelial growth factor. References 1. Thia KT , Sandborn WJ , Harmsen WS , Zinsmeister AR , Loftus EV Jr . Risk factors associated with progression to intestinal complications of Crohn’s disease in a population-based cohort . Gastroenterology 2010 ; 139 : 1147 – 55 . Google Scholar CrossRef Search ADS PubMed 2. Cosnes J , Gower-Rousseau C , Seksik P , Cortot A . Epidemiology and natural history of inflammatory bowel diseases . Gastroenterology 2011 ; 140 : 1785 – 94 . Google Scholar CrossRef Search ADS PubMed 3. Cosnes J , Nion-Larmurier I , Beaugerie L , Afchain P , Tiret E , Gendre JP . Impact of the increasing use of immunosuppressants in Crohn’s disease on the need for intestinal surgery . Gut 2005 ; 54 : 237 – 41 . Google Scholar CrossRef Search ADS PubMed 4. Wynn TA , Ramalingam TR . Mechanisms of fibrosis: therapeutic translation for fibrotic disease . Nat Med 2012 ; 18 : 1028 – 40 . Google Scholar CrossRef Search ADS PubMed 5. Johnson LA , Luke A , Sauder K , Moons DS , Horowitz JC , Higgins PD . Intestinal fibrosis is reduced by early elimination of inflammation in a mouse model of IBD: impact of a “Top-Down” approach to intestinal fibrosis in mice . Inflamm Bowel Dis 2012 ; 18 : 460 – 71 . Google Scholar CrossRef Search ADS PubMed 6. Di Sabatino A , Jackson CL , Pickard KM et al. Transforming growth factor beta signalling and matrix metalloproteinases in the mucosa overlying Crohn’s disease strictures . Gut 2009 ; 58 : 777 – 89 . Google Scholar CrossRef Search ADS PubMed 7. Shimshoni E , Yablecovitch D , Baram L , Dotan I , Sagi I . ECM remodelling in IBD: innocent bystander or partner in crime? The emerging role of extracellular molecular events in sustaining intestinal inflammation . Gut 2015 ; 64 : 367 – 72 . Google Scholar CrossRef Search ADS PubMed 8. Wells RG . The role of matrix stiffness in regulating cell behavior . Hepatology 2008 ; 47 : 1394 – 400 . Google Scholar CrossRef Search ADS PubMed 9. Rustad KC , Wong VW , Gurtner GC . The role of focal adhesion complexes in fibroblast mechanotransduction during scar formation . Differentiation 2013 ; 86 : 87 – 91 . Google Scholar CrossRef Search ADS PubMed 10. Georges PC , Hui JJ , Gombos Z et al. Increased stiffness of the rat liver precedes matrix deposition: implications for fibrosis . Am J Physiol Gastrointest Liver Physiol 2007 ; 293 : G1147 – 54 . Google Scholar CrossRef Search ADS PubMed 11. Yin M , Talwalkar JA , Glaser KJ et al. Assessment of hepatic fibrosis with magnetic resonance elastography . Clin Gastroenterol Hepatol 2007 ; 5 : 1207 – 1213.e2 . Google Scholar CrossRef Search ADS PubMed 12. Grillo AR , Scarpa M , D’Incà R et al. TAK1 is a key modulator of the profibrogenic phenotype of human ileal myofibroblasts in Crohn’s disease . Am J Physiol Gastrointest Liver Physiol 2015 ; 309 : G443 – 54 . Google Scholar CrossRef Search ADS PubMed 13. Di Sabatino A , Pender SL , Jackson CL et al. Functional modulation of Crohn’s disease myofibroblasts by anti-tumor necrosis factor antibodies . Gastroenterology 2007 ; 133 : 137 – 49 . Google Scholar CrossRef Search ADS PubMed 14. Burke JP , Ferrante M , Dejaegher K et al. Transcriptomic analysis of intestinal fibrosis-associated gene expression in response to medical therapy in Crohn’s disease . Inflamm Bowel Dis 2008 ; 14 : 1197 – 204 . Google Scholar CrossRef Search ADS PubMed 15. Otte JM , Rosenberg IM , Podolsky DK . Intestinal myofibroblasts in innate immune responses of the intestine . Gastroenterology 2003 ; 124 : 1866 – 78 . Google Scholar CrossRef Search ADS PubMed 16. Nijhuis A , Biancheri P , Lewis A et al. In Crohn’s disease fibrosis-reduced expression of the miR-29 family enhances collagen expression in intestinal fibroblasts . Clin Sci (Lond) 2014 ; 127 : 341 – 50 . Google Scholar CrossRef Search ADS PubMed 17. Owens BM , Steevels TA , Dudek M et al. CD90+ stromal cells are non-professional innate immune effectors of the human colonic mucosa . Front Immunol 2013 ; 4 : 307 . Google Scholar CrossRef Search ADS PubMed 18. Ke N , Wang X , Xu X , Abassi YA . The xCELLigence system for real-time and label-free monitoring of cell viability . Methods Mol Biol 2011 ; 740 : 33 – 43 . Google Scholar CrossRef Search ADS PubMed 19. Barry-Hamilton V , Spangler R , Marshall D et al. Allosteric inhibition of lysyl oxidase–like-2 impedes the development of a pathologic microenvironment . Nat Med 2010 ; 16 : 1009 – 17 . Google Scholar CrossRef Search ADS PubMed 20. Asbun J , Manso AM , Villarreal FJ . Profibrotic influence of high glucose concentration on cardiac fibroblast functions: effects of losartan and vitamin E . Am J Physiol Heart Circ Physiol 2005 ; 288 : H227 – 34 . Google Scholar CrossRef Search ADS PubMed 21. Fischer RS , Myers KA , Gardel ML , Waterman CM . Stiffness-controlled three-dimensional extracellular matrices for high-resolution imaging of cell behavior . Nat Protoc 2012 ; 7 : 2056 – 66 . Google Scholar CrossRef Search ADS PubMed 22. Calvo F , Ege N , Grande-Garcia A et al. Mechanotransduction and YAP-dependent matrix remodelling is required for the generation and maintenance of cancer-associated fibroblasts . Nat Cell Biol 2013 ; 15 : 637 – 46 . Google Scholar CrossRef Search ADS PubMed 23. Johnson LA , Rodansky ES , Sauder KL et al. Matrix stiffness corresponding to strictured bowel induces a fibrogenic response in human colonic fibroblasts . Inflamm Bowel Dis 2013 ; 19 : 891 – 903 . Google Scholar CrossRef Search ADS PubMed 24. Gurtner GC , Werner S , Barrandon Y , Longaker MT . Wound repair and regeneration . Nature 2008 ; 453 : 314 – 21 . Google Scholar CrossRef Search ADS PubMed 25. Wynn TA . Cellular and molecular mechanisms of fibrosis . J Pathol 2008 ; 214 : 199 – 210 . Google Scholar CrossRef Search ADS PubMed 26. Kagan HM , Li W . Lysyl oxidase: properties, specificity, and biological roles inside and outside of the cell . J Cell Biochem 2003 ; 88 : 660 – 72 . Google Scholar CrossRef Search ADS PubMed 27. Barker HE , Cox TR , Erler JT . The rationale for targeting the LOX family in cancer . Nat Rev Cancer 2012 ; 12 : 540 – 52 . Google Scholar CrossRef Search ADS PubMed 28. Fogelgren B , Polgár N , Szauter KM et al. Cellular fibronectin binds to lysyl oxidase with high affinity and is critical for its proteolytic activation . J Biol Chem 2005 ; 280 : 24690 – 7 . Google Scholar CrossRef Search ADS PubMed 29. Wong VW , Rustad KC , Akaishi S et al. Focal adhesion kinase links mechanical force to skin fibrosis via inflammatory signaling . Nat Med 2011 ; 18 : 148 – 52 . Google Scholar CrossRef Search ADS PubMed 30. Burke JP , Cunningham MF , Sweeney C , Docherty NG , O’Connell PR . N-cadherin is overexpressed in Crohn’s stricture fibroblasts and promotes intestinal fibroblast migration . Inflamm Bowel Dis 2011 ; 17 : 1665 – 73 . Google Scholar CrossRef Search ADS PubMed 31. Beddy D , Watson RW , Fitzpatrick JM , O’Connell PR . Increased vascular endothelial growth factor production in fibroblasts isolated from strictures in patients with Crohn’s disease . Br J Surg 2004 ; 91 : 72 – 7 . Google Scholar CrossRef Search ADS PubMed 32. Lang M , Schlechtweg M , Kellermeier S et al. Gene expression profiles of mucosal fibroblasts from strictured and nonstrictured areas of patients with Crohn’s disease . Inflamm Bowel Dis 2009 ; 15 : 212 – 23 . Google Scholar CrossRef Search ADS PubMed 33. Li CK , Pender SL , Pickard KM et al. Impaired immunity to intestinal bacterial infection in stromelysin-1 (matrix metalloproteinase-3)-deficient mice . J Immunol 2004 ; 173 : 5171 – 9 . Google Scholar CrossRef Search ADS PubMed 34. Meijer MJ , Mieremet-Ooms MA , van Hogezand RA , Lamers CB , Hommes DW , Verspaget HW . Role of matrix metalloproteinase, tissue inhibitor of metalloproteinase and tumor necrosis factor-alpha single nucleotide gene polymorphisms in inflammatory bowel disease . World J Gastroenterol 2007 ; 13 : 2960 – 6 . Google Scholar CrossRef Search ADS PubMed 35. McKaig BC , McWilliams D , Watson SA , Mahida YR . Expression and regulation of tissue inhibitor of metalloproteinase-1 and matrix metalloproteinases by intestinal myofibroblasts in inflammatory bowel disease . Am J Pathol 2003 ; 162 : 1355 – 60 . Google Scholar CrossRef Search ADS PubMed 36. Sadler T , Bhasin JM , Xu Y et al. Genome-wide analysis of DNA methylation and gene expression defines molecular characteristics of Crohn’s disease–associated fibrosis . Clin Epigenetics 2016 ; 8 : 30 . Google Scholar CrossRef Search ADS PubMed 37. Lawrance IC , Maxwell L , Doe W . Altered response of intestinal mucosal fibroblasts to profibrogenic cytokines in inflammatory bowel disease . Inflamm Bowel Dis 2001 ; 7 : 226 – 36 . Google Scholar CrossRef Search ADS PubMed 38. Shin TM , Bordeaux JS . The role of massage in scar management: a literature review . Dermatol Surg 2012 ; 38 : 414 – 23 . Google Scholar CrossRef Search ADS PubMed 39. Rieder F , Zimmermann EM , Remzi FH , Sandborn WJ . Crohn’s disease complicated by strictures: a systematic review . Gut 2013 ; 62 : 1072 – 84 . Google Scholar CrossRef Search ADS PubMed 40. Yamamoto T , Fazio VW , Tekkis PP . Safety and efficacy of strictureplasty for Crohn’s disease: a systematic review and meta-analysis . Dis Colon Rectum 2007 ; 50 : 1968 – 86 . Google Scholar CrossRef Search ADS PubMed 41. Lo CM , Wang HB , Dembo M , Wang YL . Cell movement is guided by the rigidity of the substrate . Biophys J 2000 ; 79 : 144 – 52 . Google Scholar CrossRef Search ADS PubMed 42. Levental KR , Yu H , Kass L et al. Matrix crosslinking forces tumor progression by enhancing integrin signaling . Cell 2009 ; 139 : 891 – 906 . Google Scholar CrossRef Search ADS PubMed 43. Barker HE , Chang J , Cox TR et al. LOXL2-mediated matrix remodeling in metastasis and mammary gland involution . Cancer Res 2011 ; 71 : 1561 – 72 . Google Scholar CrossRef Search ADS PubMed 44. Lin TH , Aplin AE , Shen Y et al. Integrin-mediated activation of MAP kinase is independent of FAK: evidence for dual integrin signaling pathways in fibroblasts . J Cell Biol 1997 ; 136 : 1385 – 95 . Google Scholar CrossRef Search ADS PubMed 45. Liu SB , Ikenaga N , Peng ZW et al. Lysyl oxidase activity contributes to collagen stabilization during liver fibrosis progression and limits spontaneous fibrosis reversal in mice . FASEB J 2016 ; 30 : 1599 – 609 . Google Scholar CrossRef Search ADS PubMed 46. Meissner EG , McLaughlin M , Matthews L et al. Simtuzumab treatment of advanced liver fibrosis in HIV and HCV-infected adults: results of a 6-month open-label safety trial . Liver Int 2016 ; 36 : 1783 – 92 . Google Scholar CrossRef Search ADS PubMed 47. Muir AG , Goodman Z , Levy C et al. Efficacy and safety of simtuzumab for the treatment of primary sclerosing cholangitis: results of a phase 2b, dose-ranging, randomized, placebo-controlled trial . J Hepatol 2017 ; 66 : S73 . Google Scholar CrossRef Search ADS 48. Sanyal AA , Abdelmalek MF , Diehl AM et al. Efficacy and safety of simtuzumab for the treatment of nonalcoholic steatohepatitis with bridging fibrosis or cirrhosis: results of two phase 2b, dose-ranging, randomized, placebo-controlled trials . J Hepatol 2017 ; 66 : S54 . Google Scholar CrossRef Search ADS Copyright © 2018 European Crohn’s and Colitis Organisation (ECCO). Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Journal of Crohn's and ColitisOxford University Press

Published: Apr 16, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off