Introduction

Inflammatory bowel diseases (IBD) refer to idiopathic, chronic, relapsing, inflammatory conditions arising from an underlying immunological deregulation triggered by currently unknown factors; Crohn’s disease (CD) and ulcerative colitis (UC) are the two major clinical forms of IBD.1, 2 Mounting evidence suggests that mesenchymal cells, such as intestinal fibroblasts and myofibroblasts (IMF), are actively involved in the inflammatory process of the intestinal mucosa.3 Stable alterations in their phenotype and function during persistent inflammation can regulate the transition to chronic gut inflammation by promoting tissue destruction and supporting the recruitment, retention and activation of immune cells by producing a diverse array of cytokines.4 However, mechanistic data, as how activated IMF maintain their inflammatory phenotype, are limited.

MicroRNAs (miRNA) are small non-coding RNAs that primarily bind to the 3′-untranslated region (UTR) of their specific target mRNAs, thus preventing protein synthesis.5 Intracellular signaling pathways induce or repress specific miRNA to switch gene expression programs of cells, strongly influencing signal transduction pathways and cellular phenotype.6, 7 Deregulated miRNA expression has been linked to a variety of disorders such as autoimmune diseases, cancers and chronic inflammatory pathologies, such as IBD.8, 9, 10, 11 MiR-155 is a typical multifunctional miRNA involved in hematopoiesis, inflammation and immune responses,12 whose expression is induced by inflammatory cytokines and toll-like receptor ligands.13 Therefore, it is not surprising that abnormal levels of miR-155 have been observed in the colonic mucosa of IBD patients.14, 15, 16, 17 However, the cellular sources of miR-155 in the colonic mucosa and its contribution to disease pathophysiology have not been investigated.

In the present study, we found that miR-155 is significantly upregulated in IMF isolated from UC patients but not from CD patients. Its expression in IMF is regulated by pro-inflammatory signals, such as TNF-α and lipopolysaccharide (LPS), but not by pro-fibrogenic mediators, such as TGF-β1. Moreover, we showed that the overexpression of miR-155 in control IMF recapitulates the inflammatory phenotype of UC-derived IMF, whereas miR-155 knockdown in UC-derived IMF corrects their pro-inflammatory phenotype. Furthermore, we found that miR-155 directly targets the suppressor of cytokine signaling 1 (SOCS1), whose expression is significantly downregulated in UC-derived IMF. Altogether, these results provide a better understanding in UC pathophysiology and identify a potential new molecular target to design anti-inflammatory therapies.

Materials and methods

Patients

We collected biopsies from patients with UC (n=8) or CD (n=8) undergoing colonoscopy for their ordinary follow-up. Biopsies were obtained from macroscopically normal colonic mucosa, and histological analysis of an immediately adjacent area confirmed the absence of active inflammatory damage. In Table 1, patients’ clinical details are summarized. Controls were biopsies collected from subjects undergoing colonoscopy for cancer screening (n=8), showing no endoscopic signs of disease (Table 1). All study procedures were complied with the principles of the amended Declaration of Helsinki and were approved by the Ethical Committee of the University Hospital of Padua. Each patient was provided with detailed information about the study aims and protocol and gave their written informed consent before enrollment. Biopsies were obtained from the sigmoid region, at 8–10 cm from the anus. In UC patients, disease activity was scored using the Schroeder (a.k.a. Mayo) endoscopic scoring system,18 whereas CD activity was determined using the Crohn’s disease activity index.19

Table 1 Characteristics of patients
Full size table

Isolation, cultivation and characterization of human IMF

For IMF isolation, colonic biopsies were immediately placed in ice-cold Roswell Park Memorial Institute (Gibco, Life Technologies Inc., Grand Island, NY, USA). Mucus was removed by washing in phosphate-buffered saline containing 0.04% dithiotreitol (Applichem Gmbh, Darmstadt, Germany) for 10 min. The epithelial layer was removed by incubation with 10 mM EDTA for 30 min at 37 °C. Tissues were washed in Roswell Park Memorial Institute containing 1% penicillin/streptomycin and subjected to enzymatic digestion in 0.05% w/v collagenase type IV Clostridium histolyticum (Sigma-Aldrich, St Louis, MO, USA) and 4 mg ml−1 DNase (Calbiochem, San Diego, CA, USA) dissolved in Roswell Park Memorial Institute for 30 min at 37 °C. Tissues were washed three times in Roswell Park Memorial Institute and resuspended in Dulbecco’s modified Eagle's medium supplemented with 20% v/v heat inactivated fetal bovine serum (Gibco), 2 mM L-glutamine, 0.1 mM non-essential amino acids, 1 mM sodium pyruvate, 100 U ml−1 penicillin, 100 μg ml−1 streptomycin, 0.25 μg ml−1 amphotericin B (all from Gibco, and indicated as complete medium). Tissues were seeded in culture plates in a controlled atmosphere of 5% CO2 and 37 °C for 48 h without changing the medium to allow IMF to migrate from the tissue and adhere to plastic.20 Cells were characterized after four passages by immunocytochemistry using mouse monoclonal anti-α-SMA and anti-desmin antibodies (dilution 1:300) (all from Sigma-Aldrich) and used for experiments up to passage 8.

MiRNA isolation and quantification by quantitative RT-PCR

IMF (5 × 105) were seeded in six-well plates and after 48 h fresh medium containing 5% v/v, fetal bovine serum was added to the cells. Total miRNA was isolated after an additional 24 h using Qiagen miRNA isolation kit (Ambion, Austin, TX, USA) according to the manufacturer’s protocol. miRNA was quantified using a NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific Inc., Waltham, MA, USA). Expression level of mature miR-155 was determined with the TaqMan MicroRNA Reverse Transcription kit, and the TaqMan MicroRNA Assay for miR-155 and RNU6b small nuclear RNA (used as internal control) purchased from Applied Biosystems (Foster City, CA, USA). All experimental procedures were performed according to the manufacturer’s protocols. In total, 10 ng of miRNA were reverse transcribed in a final volume of 15 μl, and quantitative reverse transcription-PCR was carried out on an ABI PRISM 7000 Sequence Detection System (Applied Biosystems) in a final volume of 20 μl. All quantitative reverse transcription-PCR reactions were run in duplicate. The expression of miR-155 was calculated using the formula 2−ΔΔCT, where ΔCT=(CTmiRNA−CTmiRNAcontrol) and CT (that is, threshold cycle) indicates the fractional cycle number at which the amount of amplified target reaches a fixed threshold.21

Ectopic expression and knockdown of miR-155 in IMF

To examine the effect of miR-155 overexpression, pre-miRNA155 (PM12601) and pre-miRNA scramble control (AM 17150) (Invitrogen Corp.) were transfected into control IMF. Similarly, to examine the effect of miRNA155 knockdown, the anti-miR-155 inhibitor (AM12601) and anti-miR inhibitor negative control (AM 17010, all from Invitrogen) were transfected into UC-derived IMF. IMF (4 × 104 cells/9.6 cm2) were cultured for 24 h, washed, placed in serum-free Dulbecco's modified Eagle's medium without antibiotics and transfected. Transfection conditions and efficiencies were established in preliminary assays measuring the level of the target gene Hmga2 following IMF transfection with Let7c-miRNA inhibitor. IMF transfection using 6 μl Lipofectamine 2000 (Invitrogen Corp.) with 200 pmol of anti-mir let-7c miRNA inhibitor was the most efficient combination with a sevenfold increase in Hmga2 mRNA level (data not shown) after 24 h. These conditions were then used for all miRNA transfection experiments. IMF were harvested after 24 h and used for the specified experiments.

Effect of pro-inflammatory mediators on miR-155 in IMF

IMF isolated from normal or UC patients were seeded (5 × 105 per 9.6 cm2) and after 24 h the culture medium was changed. Then rhTNF-α, rhIL-1α, rhTGF-β1 (all at 10 ng ml−1) or bacterial LPS (1 μg ml−1, Sigma-Aldrich) was added to the culture medium. Cells were harvested after 24–72 h, and miR-155 quantified as described above.

SOCS1 silencing

Specific SOCS1 Silencer Select siRNA (s16468 and s16470) and Silencer Select-negative control siRNA no. 1 were purchased from Invitrogen and 20 μmol l−1 stock solutions were prepared in diethyl pyrocarbonate water. None of them has any significant homology to human genes (Invitrogen-held data). SiRNA were transfected when IMF reached 50% confluence. Lipofectamine 2000 and Silencer Select siRNA (80 nmol l−1) were used in a total transfection volume of 2 ml per well. After 8 h, 2 ml of culture medium were added to each well. Negative controls, including non transfected IMF and cells incubated with either Lipofectamine 2000 alone or negative control siRNA, were established in parallel. IMF were cultured for an additional 2 days before performing the experiments.

Western blot analysis

IMF treated as specified were washed with ice-cold phosphate-buffered saline and then homogenized in extraction buffer (100 mM KCl, 3 mM NaCl, 3.5 mM MgCl2 and 10 mM HEPES; pH 7.4) containing 1% v/v Triton X-100, 1% v/v Protease Inhibitor Cocktail Set I (Calbiochem) and phosphatase inhibitors (25 mM NaF and 10 mM Na3VO4). After 45 min on ice, cells were centrifuged (1500 × g for 5 min at 4 °C) to remove cellular debris. Then the supernatant designated as the total protein fraction was collected and the protein concentration was determined by the bicinchoninic acid assay. In total, 20 μg proteins for each condition were fractionated through sodium dodecyl sulphate-polyacrylamide gel electrophoresis gel and transferred to nitrocellulose membranes (Trans-Blot Transfer Medium, Bio-Rad Laboratories, Hercules, CA, USA). Membranes were blocked in phosphate-buffered saline containing 0.1% Tween-20 with 5% (w/v) non-fat dry milk or 5% bovine serum albumin for 1 h at room temperature and then incubated with rabbit anti-SOCS1 (1:1000, Cell Signaling Technology, Danvers, MA, USA) and mouse anti-β-actin antibody (1:5000, Sigma-Aldrich), respectively, overnight at 4 °C. After extensive washing, membranes were incubated for 1 h with proper secondary horseradish peroxidase -conjugated antibody (1:2000, all from Sigma-Aldrich). Immunocomplexes were visualized using enhanced chemiluminescence western blotting detection reagents (Pierce, Milan, Italy). Images were acquired using a VersaDoc imaging system (Bio-Rad Laboratories) and a densitometric quantification measurement of the band intensity was performed using ImageJ software.

Secretion of inflammatory cytokines

IMF isolated from normal subjects or UC patients were seeded in 24-well plates (4 × 104 cells per well) and, after 24 h, were placed in serum-free Dulbecco's modified Eagle’s medium without antibiotics. When indicated cells were transfected as described above with either pre-miRNA (PM12601) or anti-miRNA-155 (AM12601) or SOCS1 Silencer Select siRNA (s16468 and s16470) and then cultured for additional 48 h in culture medium. Medium was then refreshed in the presence or absence of LPS (1 μg ml−1). Culture medium was collected after 24 h and used to quantify interleukin (IL)-6 and IL-8 release by enzyme-linked immunosorbent assay (Immunotools).

MiR-155 target validation

Through TargetScan, a bioinformatics prediction program that can identify hypothetical mRNA-binding-specific miRNA, we identified SOCS1 as a potential target of miR-155. Genomic DNA extracted from peripheral blood mononuclear cells was used as template to amplify a fragment of the 3′-UTR of SOCS1, which contains the putative miR-155-binding sequence. The following primers were used: forward 5′-gagaGAGCTCAGATTTGACCGGCAGCGCCC-3′, reverse 3′-gagaTCTAGATGCACAGCAGAAAAATAAAGCCAGA-5′. The genomic fragment was cloned into a luciferase reporter construct (pmir-GLO, Promega, Madison, WI, USA). A control reporter plasmid (pmir-GLO-3′UTR Mut) with a deletion of seven nucleotides corresponding to miR-155 seed sequence (GCATTA) was constructed by site-directed mutagenesis using forward primer 5′-gagaccggcagcgcccgccgtgcacgcactgggatgccgtgttat-3′ and the same reverse primer was used in wild-type reporter. Then, each reporter plasmid was co-transfected into HEK 293A cells in 24-well plates together with Firefly luciferase plasmid pGL3 (Promega) and pre-miR-155 or pre-miR scramble as control. The luciferase activity in cell extracts was assayed 48 h after transfection using a Dual-Luciferase Reporter Assay System (Promega) and a 96-well plate reader. All experiments were repeated at least three times.

Statistical analysis

Statistical analysis was performed using the GraphPad Prism 3.03 (San Diego, CA, USA). Unpaired groups of values were compared according to the non-parametric Mann–Whitney test. Comparison of more than two groups of data was performed by Kruskal–Wallis test. Data were considered statistically significant for P<0.05.

Results

MiR-155 is upregulated in UC-derived IMF

The expression of miR-155 was determined in human primary IMF using quantitative reverse transcription-PCR. As shown in Figure 1a, miR-155 is significantly upregulated in UC-derived IMF as compared with cells isolated from CD and control colonic mucosa. These results suggest that IMF are cellular sources of miR-155 in the colonic mucosa of UC patients and may contribute to disease pathophysiology.

Figure 1
figure 1

Increased miR-155 expression in UC-derived IMF. (a) miR-155 levels were quantified by real-time qPCR, and data normalized against U6B RNA expression in IMF isolated from mucosa of control, CD and UC patients (n=8 each group). Data are presented as box–whisker plots (box, 25–75%; whisker, 10–90%; line, median); *P<0.01 vs control- and CD-derived IMF.

Full size image

TNF-α and bacterial LPS upregulate miR-155 expression in IMF

IMF phenotype results from the elaboration of a variety of stimuli arising from the extracellular milieu, thus we investigated the influence of different soluble mediators involved in the development, recurrence and exacerbation of the inflammatory process in UC on miR-155 expression. As shown in Figure 2, although TGF-β1 significantly influences IMF physiology,22, 23 it failed to modify miR-155 level in IMF isolated from control subjects (Figure 2a) and had only transient effects on UC-derived IMF (Figure 2b). However, an inflammatory cytokine such as TNF-α and bacterial LPS significantly modulated miR-155 level in IMF. Thus, bacterial LPS and TNF-α increased miR-155 level following 72 h exposure in control-derived IMF and further upregulated miR-155 level in UC-derived IMF. On the other hand, IL-1β had no significant effects on either control- and UC-derived IMF.

Figure 2
figure 2

TNF-α and bacterial LPS upregulate miR-155 expression in IMF. Control (a) and UC-derived (b) IMF were exposed for 24–72 h to either TNF-α, LPS, TGF-β1 or IL-1β. MiR-155 levels were quantified by real-time qPCR. Data were normalized against U6B RNA expression. Data are presented as mean±s.e.m. (n=3). **P<0.01 and *P<0.05 vs respective control.

Full size image

MiR-155 promotes a pro-inflammatory phenotype in IMF

In line with previous reports,24, 25, 26 UC-derived IMF show at baseline an activated phenotype as demonstrated by the increased constitutive release of the pro-inflammatory cytokines IL-6 and IL-8 (Figures 3a and b). Indeed, overexpression of miR-155, obtained by transfecting IMF with pre-miR-155, augmented IL-6 in control IMF to a level comparable to LPS (Figures 3c and d). Furthermore, normalizing miR-155 level in UC-derived IMF by transfecting an anti-miR-155 significantly reduced IL-6 and IL-8 release (Figures 3e and f).

Figure 3
figure 3

MiR-155 modulates the inflammatory phenotype of IMF. (a) and (b) Control- and UC-derived IMF were incubated in medium alone or LPS (1 μg ml-1) for 24 h (n=3 for each line with triplicate determinations); IL-6 and IL-8 levels in the supernatants were determined by ELISA. Data are expressed as mean±s.e.m. ** indicates P<0.01 and *P<0.05 vs respective control, ° indicates P<0.01 vs control IMF. (c) and (d) Control-derived IMF were transfected with pre-miR-155 to obtain miR-155 overexpression, and after 48 h cells and culture media were collected; IL-6 and IL-8 release was determined by ELISA. Data are expressed as mean±s.e.m. (n=3 with duplicate determinations). ** indicates P<0.01 and *P<0.05 vs pre-miRNA scramble. (e) and (f) UC-derived IMF were transfected with anti-miR-155 to obtain miR-155 knockdown and after 48 h cells and culture media were collected. IL-6 and IL-8 release was determined by ELISA (n=3 with duplicate determinations). **P<0.01 and *P<0.05 vs respective control.

Full size image

MiR-155 and SOCS1 expression are inversely correlated in UC-derived IMF

To select potential target genes of miR-155, we used TargetScan prediction program (www.targetscan.org) to identify genes that harbored miR-155 target sites in the 3′-UTR of their mRNA. Because miR-155 influenced the inflammatory phenotype of IMF, we focused on putative candidates known to be associated with control of inflammatory processes. We identified SOCS1 as a potential miR-155 target in IMF (Figure 4a). Interestingly, recent studies revealed that SOCS1 deficiency exacerbates intestinal inflammation.27, 28 Moreover, we found that SOCS1 protein level was significantly lower in UC-derived IMF than in control-derived IMF (Figure 4b). These findings suggest an inverse relationship between miR-155 and SOCS1 expression because increased miR-155 expression is associated with the downregulation of SOCS1 in UC-derived IMF.

Figure 4
figure 4

SOCS1 expression is reduced in UC-derived IMF. (a) Schematic representation of putative miR-155-binding site in the 3′-untranslated region (3′-UTR) of SOCS1. (b) Protein levels of SOCS1 were determined by western blot analysis control- and UC-derived IMF. Densitometric quantification performed using β-actin as internal control (n=3). *P<0.05 vs control IMF.

Full size image

MiR-155 inhibits SOCS1 expression by directly targeting its 3′-UTR

To demonstrate a direct interaction between miR-155 and SOCS1, we performed luciferase reporter assays in HEK 293A cells co-transfected with reporter plasmids carrying a fragment of the 3′-UTR region of SOCS1 gene and pre-miR-155 to overexpress this miRNA. As shown in Figure 5a, miR-155 mimic significantly reduced the luciferase activity of the reporter fused with the 3′-UTR of human SOCS1 but not of the construct with a mutated 3′-UTR of SOCS1 in which the seven nucleotides of the miR-155 seed-matched sequence were deleted (3′-UTR Mut). Furthermore, transfection of miR-155 mimic in control IMF significantly reduced SOCS1 basal expression as shown by western blot analysis (Figure 5b), whereas anti-miR-155 transfection normalized SOCS1 level in UC-derived IMF (Figure 5c). These results indicate that SOCS1 is a direct target of miR-155 in IMF.

Figure 5
figure 5

MiR-155 directly targets SOCS1 in IMF. (a) Luciferase activity in HEK 293A cells co-transfected with each reporter plasmid (pmir-GLO-Socs1 3′-UTR or pmir-GLO-Socs1 3′-UTR mut) and pre-miR-155 or pre-miR-scramble. *P<0.05 vs pre-miRNA scramble. (b) Western blotting analysis of SOCS1 expression in control-derived IMF following pre-miRNA-155 transfection. Densitometric quantification performed using β-actin as internal control (n=3). *P<0.05 vs pre-miRNA scramble. (c) Western blot analysis of SOCS1 expression in UC-derived IMF following transfection of anti-miRNA-155 or scramble oligonucleotide as control (n=3). *P<0.05 vs anti-miRNA scramble control.

Full size image

SOCS1 influences IMF inflammatory phenotype

As SOCS1 is a known feedback inhibitor of inflammation, we asked whether SOCS1 could regulate the inflammatory phenotype of IMF. We initially determined the ability of specific SOCS1 siRNAs to knockdown SOCS1 in control IMF. As shown in Figures 6a and b, transfection with specific SOCS1 siRNAs reduced SOCS1 levels by ~70% after 48 h, whereas negative control siRNA was ineffective. As shown in Figures 6c and d, SOCS1 knockdown in control-derived IMF caused a significant increase in basal and LPS stimulated IL-6 and IL-8 release, supporting its regulatory role in the development of an inflammatory phenotype of IMF.

Figure 6
figure 6

SOCS1 regulates inflammatory cytokines in IMF (a) Validation of SOCS1 silencing by western blot analysis of SOCS1 expression in control-derived IMF transfected with SOCS1 siRNA no. 1 and siRNA no. 2 or scramble control. (b) Densitometric quantification performed using β-actin as internal control (n=3). *P<0.05 vs scramble control. (c) IL-6 and (d) IL-8 release determined by ELISA in control-derived IMF supernatants following SOCS1 knockdown (n=3). **P<0.01 vs control, °°P<0.01 and °P<0.05 vs LPS alone.

Full size image

Discussion

In IBD, the deregulated activity of several immune and non immune cell populations generates, amplifies and maintains tissue damage.29 However, the specific gene expression profile and the signaling pathways altered in cellular populations relevant to IBD pathophysiology are scarcely known. In this study, we report that miR-155 has an important role in the aberrant phenotype of IMF in UC. In addition, we demonstrate that SOCS1 is targeted by miR-155 and that restoring miR-155 levels in UC-IMF recovers their phenotype. These findings are significant to the IBD field for several reasons. First, they provide an miRNA expression signature of the inflammatory phenotype of IMF in UC. Second, they establish a direct link between miR-155 and IMF pro-inflammatory behavior in UC. Third, they identify IMF as a cellular population, contributing to the inadequate SOCS1 expression during colitis. Finally, our data ascertain SOCS1 as a direct target of miR-155 in primary human IMF.

MiRNAs are increasingly recognized as an important element in the development and function of the innate and adaptive immune system, and changes in miRNA expression profile are described in many immune-related diseases.30 Elucidating the role of miRNA in IBD represents a new frontier to advance our knowledge of disease mechanisms, and to develop new diagnostic and therapeutic tools as miRNA controls basic biological functions such as proliferation, apoptosis and cell differentiation.31 Previous studies have reported that the expression of several miRNA is altered in the intestinal mucosa during inflammatory disorders, including UC.14, 15, 16, 17, 32 However, our study identified IMF as a cellular source of miR-155 deregulation in the colonic mucosa of UC patients, in addition to B and T lymphocytes and macrophages following exposure to pathogens.33, 34 Indeed, the key role of this miRNA deregulation in colonic mucosa of UC patients has been recently underscored in the study by Singh et al.35 showing that miR-155 deficit protects mice from experimental colitis.

IMF are plastic cells sensing and reacting to a variety of stress to preserve intestinal mucosa homeostasis.3 However, in response to persistent inflammatory insults like in IBD, IMF acquire an activated phenotype, proliferate, cause an unnecessary remodeling of the extracellular matrix and produce an excess of soluble mediators, such as inflammatory cytokines, TGF-β1 and Wnts ligands, which profoundly influence neighboring epithelial, mesenchymal and immune cells. Compelling evidence indicates that miR-155 regulates several cellular functions involved in inflammatory processes, such as survival, growth and chemosensitivity. Therefore, deregulated miR-155 expression can profoundly impact cellular phenotype and functions.12, 13 In line with a previous report for the monocyte–macrophage lineage,33 pro-inflammatory mediators such as TNF-α and LPS positively regulated miR-155 expression in IMF. Indeed, in macrophages TNF-α upregulates miR-155 that contributes to increased miR-155 production in an autocrine loop.36 All together these data support the view that the persistently increased mucosal levels of TNF-α in UC patient37 can drive miR-155 overexpression in IMF, influencing their phenotype. TGF-β1 failed to directly regulate miR-155 in IMF in contrast to the direct upregulation observed in lymphoma cell lines38 and mammary gland adenocarcinoma cells.39 Thus, taking into account the growth-inhibitory effects of TGF-β1, a miR-155-mediated mechanism to escape TGF-β’s growth-inhibitory effects may be present in transformed cells but not in IMF isolated from non-cancerous tissues. Intriguingly, in several immune cells of IBD patients, the resistance to TGF-β1 signaling is due to enhanced activity of the inhibitory Smad740 as opposite to somatic mutations in SMADs or TGF-β receptors reported in malignant cells.41

On the basis of the fact that miR-155 induced inflammatory cytokine production in IMF, validation of the target genes was focused on putative candidates known to be associated with control of inflammatory processes, taking into account that miR-155 targets are estimated to be over 900.42 However, only a minority of in silico predicted targets have been found to be responsive to miRNA control upon experimental validation.42 Recent studies support a positive relationship between miR-155 upregulation and molecular events associated to inflammatory processes such as activation of nuclear factor kB.43 Furthermore, Min et al.16 recently identified FOXO3 as target of miR-155 in HT-29 cells where it contributes to the upregulation of inflammatory cytokines such as IL-8. Here, we identified SOCS1 as a direct target of miR-155, confirming recent functional studies in T-regulatory, Th17 and effector CD8 T lymphocytes, breast cancer cells, microglia and macrophages.44, 45, 46, 47 SOCS1 is a potent molecular switch that, tuning key signaling pathways such as the JAK kinases and activated cytokine-receptor complexes, regulates the development of a variety of cell populations, inflammatory processes and immune responses.48 Indeed, SOCS1-deficient mice, who are hypersensitive to TLR ligands, show a deregulated cytokine production that disturbs the activation of immune cells and triggers the development of systemic autoimmunity.49 Interestingly, SOCS1 expression has been described in the small and large intestines of humans and rodents50 and SOCS1 knockdown in mice deficient for T-cell receptor α markedly enhanced colitis severity, suggesting that this protein has a critical role in maintaining mucosal homeostasis.28 Previous studies have suggested that SOCS1 is inactivated in human IBD contributing to deregulated mucosa inflammation, although the cellular populations involved were not identified.27 Here, we identify IMF as a reduced source of mucosal SOCS1 and demonstrate that inadequate SOCS1 levels contribute to the inflammatory phenotype of these cells.

In summary, to our knowledge this study represents the first comprehensive demonstration that the expression of a specific miRNA has a key role in supporting the inflammatory phenotype of IMF in UC patients, a relevant pathophysiological process in IBD. Taken together, our data provide novel insight into the peculiar role of miR-155 in intestinal mucosa during chronic inflammatory disorders and identify one of its specific target genes in IMF. Thus, inhibiting miR-155 may represent an interesting and innovative therapeutic target to prevent the deregulated cytokine release from IMF that contributes to fueling the inflammation.