IL-21 mediates microRNA-423-5p /claudin-5 signal pathway and intestinal barrier function in inflammatory bowel disease

Abnormally expression of IL-21 in ulcerative colitis patients

IL-21 stimulates fibroblasts to secrete extracellular matrix-degrading enzymes and epithelial cells to secrete T cell chemical inducer MIP-3α, which is an inflammatory factor. To explore the role of IL-21 in IBD, we examined the expression of IL-21 in colonic mucosa from UC patients and healthy volunteers. RT-PCR results performed the significantly increased expression level of IL-21 in UC patients, while there was almost no expression of IL-21 in the normal colon tissues (Figure 1A). Meanwhile, the protein level of IL-21 was also upregulated in colonic mucosa from UC patients (Figure 1B). Similar results were also performed in the IHC staining of colonic samples (Figure 1C).

For further research, we established a C57BL/6 mouse model of acute colitis with DSS and treated it with IL-21 neutralizing antibody. Compared with the DSS group, the bodyweight of mice treated with anti-IL-21 recovered (Figure 1D). After assessing the level of IL-21, we found that DSS induced the expression of IL-21, which was recovered by anti-IL-21 (Figure 1E). Further, we found that DSS induced the expression of inflammatory cytokines (IL1β, IL6, and TNFα). Compared with the DSS group, the expression of IL1β, IL6, and TNFα were decreased in anti-IL-21 treated mice (Figure 1F). We found that mice treated with IL-21 neutralizing antibody had a protective effect on experimental colitis compared with the DSS group (Figure 1G). As a biochemical test of acute enteritis, colonic MPO activity also showed that IL-21 neutralizing antibody could significantly reduce inflammation (Figure 1H). Further, we detected the intestinal barrier function of mice, and the results showed that IL-21 neutralizing antibody could decrease the level of FD40 in the serum of DSS mice (Figure 1I). Meanwhile, we detected the components of the NF-κB/MAPKs/JNK signaling pathway. DSS-induced the phosphorylation of NF-κB, ERK1/2, JNK, and P38, which were inhibited by anti-IL-21 (Figure 1J). These results suggested that fighting against IL-21 can relieve the symptoms of UC.

IL-21 regulates the intestinal epithelial barrier function by targeting claudin-5

The value of trans-epithelial resistance is an index to detect the tightness of colonic intercellular connection. The effect of IL-21 on intestinal epithelial barrier function was detected via TEER experiments. We found that IL-21 decreased the TEER, while IL-21 neutralizing antibodies markedly recovery the tight junction (Figure 2A).

The damage of the colonic mucosal barrier function is one of the pathogenesis of UC. Occludin, claudin, and zonulae occludente (ZO) in colonic epithelial cells play an essential role in intestinal tight junction function. We detected the level of claudin-1, claudin-4, claudin-5, claudin-8, claudin-11, occludin, and ZO-1 by employing the RT-PCR assay. We observed that claudin-1, claudin-5, claudin-8, occludin, and ZO-1 were decreased after IL-21 treatment in cells (Figure 2B). Except for claudin-1 and claudin-8, claudin-5 is the most significantly abnormal expression in Caco-2 cells after IL-21 treatment. We speculated that claudin-5 might be downstream of IL-21 in UC. Then we constructed the plasmid for upregulating the level of claudin-5 (claudin-5), vector plasmid (vector) was described as a negative control, we transfected claudin-5/vector into Caco-2 cells after treated with or without IL-21, we observed that overexpression of claudin-5 restored the injured-intestinal epithelial barrier function after IL-21 treatment (Figure 2C). We also evaluated the permeability of Caco-2 monolayer with fluorescein isothiocyanate-dexamethasone (FD4). Similar to the results of TEER, the overexpression of claudin-5 restored the change of FD 4 level caused by IL-21 (Figure 2D). Meanwhile, we assessed the protein level of claudin-5 in Caco-2 cells after IL-21 and IL-21 neutralizing antibodies treatment. Western blot assay results showed that IL-21 inhibited the level of claudin-5, and IL-21 neutralizing antibodies recovered the level of claudin-5 (Figure 2E, 2F). Further, we detected the expression of claudin-5 after IL-21 administrated by immunofluorescence experiment, similar to western blot assay, IL-21 blocked the expression of claudin-5 (Figure 2G). In summary, IL-21 regulated the function of the intestinal barrier by controlling the level of claudin-5.

MiR-423-5p binds with 3’UTR of claudin-5.

Through the prediction of the bioinformatics website, we found that there are binding sequences between miR-423-5p and claudin-5. By performing luciferase assay, we ensured that miR-423-5p could interact and bind with 3'UTR of claudin-5 (Figure 3A). Then we transfected the miR-423-5p mimic/miR-NC into three colorectal epithelial cells (Caco-2, NCM460, and SW480 cells). Then the protein level of claudin-5 was assessed in different cell lines. Figure 3B showed that forced expression of miR-423-5p could inhibit the level of claudin-5. Further, si-miR-423-5p/si-NC were transfected into cells for detecting the level of claudin-5. The assays performed, blocking the level of miR-423-5p would restore the level of claudin-5 (Figure 3C). Meanwhile, we found that the level of miR-423-5p was abnormally elevated in UC patients (Figure 3D). Further, we detected the level of claudin-5 in Caco-2 cells by immunofluorescence experiment, similar to western blot assay. MiR-423-5p blocked the expression of claudin-5 (Figure 3E).

MiR-423-5p regulates the intestinal epithelial barrier function by targeting claudin-5

We found that miR-423-5p mimic transfection reduced the TEER of Caco-2 cells. However, si-miR-423-5p significantly enhanced the tight junction of Caco-2 cells (Figure 4A). Then we detected the level of claudin-1, claudin-4, claudin-5, claudin-8, claudin-11, occludin, and ZO-1. We observed that the mRNA level of claudin-5 was also decreased after miR-423-5p mimic transfection in cells (Figure 4B). We transfected claudin-5 into Caco-2 cells after transfecting with miR-423-5p mimic/miR-NC, we observed that overexpression of claudin-5 restored the injured-intestinal epithelial barrier function after forced expression of miR-423-5p (Figure 4C). Similar to the results of TEER, the overexpression of claudin-5 restored the change of FD4 level caused by overexpression miR-423-5p (Figure 4D). Taken together, miR-423-5p would control the intestinal epithelial barrier function by targeting claudin-5.

MiR-423-5p antagomir administrated performs protection in DSS mice.

Next, we established a C57BL/6 mouse model of acute colitis with DSS and treated it with miR-423-5p antagomir. Compared with the DSS group, the bodyweight of mice treated with miR-423-5p antagomir recovered (Figure 5A). Further, we found that DSS induced the expression of IL1β, IL6, and TNFα. Compared with the DSS group, the expression of IL1β, IL6 and TNFα were decreased in miR-423-5p antagomir injected mice (Figure 5B)

We also found that mice treated with miR-423-5p antagomir had a protective effect on experimental colitis compared with the DSS group (Figure 5C). Colonic MPO activity also showed that miR-423-5p antagomir could significantly reduce inflammation (Figure 5D). Further, miR-423-5p antagomir down-regulated the level of FD40 in the serum of DSS mice (Figure 5E). H&E staining also performed that miR-423-5p antagomir alleviated the inflammation level induced by DSS (Figure 5F). Meanwhile, we determined the level of claudin-5. MiR-423-5p antagomir restored the down-regulated of claudin-5 in DSS mice (Figure 5G). Consistent with the results of western blot, the immunofluorescence experiment was performed on rectal tissue sections, and it was found that the level of claudin-5 increased in DSS mice injected with miR-423-5p antagomir (Figure 5H). Meanwhile, we detected the components of the NF-κB/MAPKs/JNK signaling pathway. DSS-induced the phosphorylation of NF-κB, ERK1/2, JNK, and P38, which were inhibited by miR-423-5p antagomir (Figure 5I).

Relationship between IL-21 and MiR-423-5p in UC

Based on the above experimental results, we confirmed that miR-423-5p participated in UC progression. Next, we need to confirm the relationship between IL-21 and miR-423-5p. Colorectal epithelial cells were treated with IL-21 and IL-21 neutralizing antibody, the level of miR-423-5p was assessed by RT-PCR. We observed that IL-21 induced expression of miR-423-5p, and IL-21 neutralizing antibody blocked the expression of miR-423-5p (Figure 6A). Then the Caco-2 cells were transfection with si-miR-423-5p/si-NC after IL-21 treatment. By TEER detecting, we observed that si-miR-423-5p restored the injured-intestinal epithelial barrier function after IL-21 treatment (Figure 6B). Down-regulated of miR-423-5p recovered the change of FD4 level caused by IL-21 (Figure 6C). In vivo, compared with DSS mice, the level of miR-423-5p was down-regulated in DSS mice after IL-21 neutralizing antibody injection (Figure 6D). Taken together, miR-423-5p may be downstream of IL-21 and regulated by IL-21

Human colon samples

Human IBD colonic tissue was taken from IBD patients and healthy volunteers in Shanxi Provincial People's Hospital. All participants signed the informed consent form. The research scheme was approved by the Ethics Committee of Shanxi Provincial People's Hospital, and the experimental method was in line with the Helsinki Declaration.

Cell cultures

Caco-2 cells were seeded in DMEM (Biological Industries, Israel) supplemented with 10% FBS (Biological Industries). SW480 cells were cultured in 1640 (Biological Industries, Israel) medium supplemented with 10% FBS. NCM460 cells were cultured in M3: BaseF medium (Biological Industries, Israel) supplemented with 10% FBS and 1% penicillin and streptomycin sulfate (Gibco, USA).

Cell transfection

For the siRNA/mimic/plasmid transfection, 2 × 105 cells per well were plated in a 6-well plate. After adhering for 24 hours, si-miR-423-5p/si-NC, miR-NC/miR-423-5p, or claudin-5/vector (RiboBio, China) were added to the transfection medium with Lipofectamine 2000 (Thermo Fisher Scientific, USA) for 6 hours at 37°C in a CO2 incubator. After transfection, the cells were supplemented with a normal culture medium and cultured at 37°C/5% CO2 for up to 48 hours before harvest.

Western blot

Total proteins were extracted from cells and tissues. The cells and tissues were lysed with 20 μL of RIPA Buffer (Roche Molecular Biochemicals, Switzerland), and protein concentrations were detected using the BCA kit (Beyotime, China), according to the manufacturer’s guidelines. Protein samples (20~40 μg) were separated on an SDS-PAGE gel and transferred to nitrocellulose membranes. After blocking the membranes with 5% skim milk in phosphate-buffered saline (PBS) for 1 h at room temperature, the membranes were probed with anti-IL-21 (Abcam, UK), anti-claudin-5 (GeneTex, USA), anti-p-NF-κB (Santa Cruz Biotechnology, USA), anti-NF-κB (Santa Cruz Biotechnology, USA), anti-p-ERK1/2 (Proteintech, USA), anti-ERK1/2 (Proteintech, USA), anti-p-JNK (Santa Cruz Biotechnology, USA), anti-JNK (Proteintech, USA), anti-p-P38 (Santa Cruz Biotechnology, USA), anti-P38 (GeneTex, USA) and GAPDH (Zhongshanjinqiao, China) antibodies at 4°C overnight. The nitrocellulose membranes were washed with PBST (PBS containing 0.5% Tween 20) three times for 7 min each, which was followed by the incubation with a fluorescence-labeled secondary antibody (IRDye700/800 mouse and rabbit antibodies) (Santa Cruz Biotechnology, USA). Protein levels were determined using an Odyssey infrared scanning system (LI-COR Biosciences, USA), and the band intensities were quantified using ImageJ software.

Quantitative real-time PCR

Total RNA was isolated from cells and tissues using Trizol reagent (Invitrogen, USA) according to the manufacturer’s protocols. Total RNA (1 μg) was used for synthesizing the first-strand cDNA using the cDNA Reverse Transcription Kit (Applied Biosystems, USA). qRT-PCR was performed with the SYBR Green Mix kit (Applied Biosystems, USA) according to the manufacturer’s instructions. The relative RNA levels were calculated using the Δ ΔCt method. GAPDH levels served as an internal control. miR-423-5p expression was assessed using the qRT-PCR miRNA Detection Kit (Applied Biosystems, USA), with U6 levels as an internal control. Gene expression was calculated using the 2-ΔΔCt method. The primer sequences were shown as follow:

IL-21

Forward: 5'-TGTCGCTAGCTCCTGG AGACTCAGT TCTG-3';

Reverse: 5'-CCGGGATATCCTAGGAGAGATGCTG A TG-3';

claudin-1

Forward: 5'-ATCTCGAGATGGCCAACGCGGGGCTGCAG-3';

Reverse: 5'-GTGGATCCTCACACGTAGTCTTTCCCGCTG-3';

claudin-4

Forward: 5'-ACTGGTACCAGCATGGTTGGGAAGA GAGATG-3';

Reverse: 5'-ACGCTCGAGCAGTTTCTCTGGATTCC TGAG-3;

claudin-5

Forward: 5-GATTGGCTCTCGAGCGATGACCGGCG -3;

Reverse: 5-CGTGCCCAAGCTTCTCAGACGTAGTT C-3;

claudin-8

Forward: 5-TCTGCAGTAGGA- CATAGAAACCCC TAA-3;

Reverse: 5-CGTTTAGG GGTTTCTATGTCCTACTG C-3;

claudin-11

Forward: 5-CCGGTGTGGCTAAGTACA-3;

Reverse: 5-ACTGAACCTGACCGTACACACACAGG GAACCAGATG-3;

miR-423-5p

Forward:5-ATGGTTCGTGGGTGAGGGGCAGAGAGCGAGAGCAGGGTCCGAGGTATTCG-3;

Reverse: 5-GTGCAGGGTCCGAGGT-3;

Dual Luciferase Reporter Assay

20 mmol/L miR-423-5p mimic/si- miR-423-5p or negative control (si-NC) and claudin-5 were co-transfected into HEK293T cells. Luciferase activity was detected with Luciferase Reporter Assay Kit (Biovision, China) on a luminometer (Berthold, Germany) 48 hr after the transfection.

Detection of barrier permeability

For the establishment of a monolayer model of Caco-2 cells. The determination of transepithelial resistance of cells, also known as transepithelial electrical resistance (TEER), is currently recognized as a simple and authoritative method for evaluation. Its value can reflect the permeability of the cell monolayer. TEER is formed by the flow of ions through the paracellular space. Because of its simple operation and repeatability, it has become a common index for detecting cell bypass permeability. The transmembrane resistance of cells was measured by MillicelI-ERS transmembrane resistance meter. The transmembrane resistance of Caco-2 cells = (sample hole measured value-blank hole measured value) 2x Millicell membrane area = resistance value .cm2.

Colitis model

Mice were purchased from Beijing Charles river and fed in the experimental Animal Center of Shanxi Provincial People's Hospital with the standard condition. Day 1 to Day 7: C57BL/6 in the model group was fed 3% Dextran Sodium Sulfate (DSS) aqueous solution, while the mice in the untreated group drank normal water. Day 8 to Day 14: replace fresh DSS-free drinking water. On day 14, mice were intraperitoneally injected with 3% pentobarbital sodium and were killed by excessive anesthesia with a dose of 90 mL/kg, and the organ and tissue were removed for follow-up study. The research protocol of this study was approved by the Animal Care and Use Committee of Shanxi Provincial People's Hospital.

Mice were treated with IL-21-neutralizing antibodies (0.5 mg/kg) by intraperitoneal injection at day 8-10 after DSS treated, at day 14, mice were intraperitoneally injected with 3% pentobarbital sodium and were killed by excessive anesthesia with a dose of 90 mL/kg, and the organ and tissue were removed for follow-up study.

For inhibition of miR-423-5p, mice in the antagomir-423-5p (50 nmol, RiBoBio, China) group were dosed by tail vein injection on the 7^th day.

Measurement of myeloperoxidase activity

Myeloperoxidase (MPO) activity was measured in the distal colonic tissue obtained from control and mice with colitis. The inflamed distal colon (5 cm) was removed. MPO is an enzyme found primarily in neutrophils; measurement of MPO has been widely used as a marker for intestinal inflammation [25]. MPO activity was measured according to the protocol described by Krawisz et al. [26]. Briefly, after the samples were weighed, tissue samples were homogenized in a buffer (0.5% hexadecyltrimethylammonium bromide in 50 mM potassium phosphate buffer, pH 6.0) for 1 min. The samples were frozen in liquid nitrogen, thawed three times, and centrifuged at 20,000 × g for 20 min at 4°C using a microcentrifuge. Aliquots of supernatants (20 ml) were mixed with 980 ml of O-dianisidine. Absorbance was recorded at 450 nm every 1 min over a period of 10 min by ELISA. MPO activity was expressed as units/g of tissue. An enzyme unit was defined as the conversion of 1 mol of H2O2 per min at 25°C.

Statistical analysis

Data were calculated as means ± SEM and analyzed by GraphPad 7.0. T-test and two-way ANOVA were used to analyze the data P < 0.05 was considered as a statistically significant difference.