RIP3 knockdown inhibits necroptosis of human intestinal epithelial cells via TLR4/MyD88/NF-κB signaling and ameliorates murine colitis

Normal human intestinal epithelial cells (HIEC-6, ATCC® CRL-3266) were cultivated in RPMI-1640 medium (Gibco, CA, USA) with 10% fetal bovine serum (16000-044, Gibco, CA, USA). Cells were seeded in a six-well plate and incubated in incubator at 37 °C with 5% carbon dioxide (CO2) and 95% air. When density reached about 70%, plasmids were transfected according to the instruction of Lipofectamine™ 3000 (Thermo Fisher Scientific, MA, USA). RNA was extracted after 48 h, and protein was after 72 h. The mass of plasmid transfected in each well is 800 ng.

Primers were designed according to the Sigma-Aldrich website (http://​rnaidesigner.​thermofisher.​com/​rnaiexpress/​) and the pSilencer2.1_U6 vector instructions. The sequences for shRNA were as follows: F: GATCCGCAAGTCTGGATAACGAATTCTTCAAGAGAGAATTCGTTATCCAGACTTGCTTTTTTGGAAA.

R:AGCTTTTCCAAAAAAGCAAGTCTGGATAACGAATTCTCTCTTGAAGAATTCGTTATCCAGACTTGCG. The pSilencer2.1-U6 neo vector was purchased from Invitrogen. The primers were annealed according to the instructions of Annealing Buffer for DNA Oligos (Shanghai Biyuntian Biotechnology Co., Ltd.). Procedures are as follows: 95 °C, 2 min; decrease by 0.1 °C every 8 s to 25 °C, ~ 90 min; 4 °C, ∞. pSilence2.1-U6 neo vector was digested with HindIII and BamHI; enzyme linked overnight at 16 °C; enzyme linked product was recombined, transformed and plated. After the colony is amplified, plasmids are extracted and sent to BGI for sequencing. The universal primer for sequencing is U6-promoter-human (GGACTATCATATGCTTACCG).

The experimental procedures were performed as previously reported [2]. The extraction of total protein was performed on ice throughout the entire process. The protein concentration was determined by Bradford method. Total proteins were separated on a 10% SDS-PAGE gel, then transferred onto PVDF membranes (Bio-Rad Laboratories Inc., CA, USA). After blocked with 5% non-fat milk for 1 h, membranes were incubated with antibodies RIP3 (ab62344, Abcam, Cambridge, UK), MLKL (ab184718, Abcam), MyD88 (ab219413, Abcam), TLR4 (ab13556, Abcam), p-IκB (CY6280, Abways Technology Inc., Shanghai, China), IκB (CY2327, Abways) and p65 (ab32536, Abcam). β-actin (AB0011, Abways) or histone H3 (ab1791, Abcam) was used as the loading control. Then, membranes were incubated with Peroxidase AffiniPure Goat Anti-Rabbit IgG (H + L) (111-035-003, 1:10,000, Jackson ImmunoResearch Inc., PA, USA) for 1 h and then HRP chemiluminescence detection reagent (Tiangen Biotech, Beijing, China). Tanon 5200 automatic chemiluminescence image analysis system (Tanon, Beijing, China) was applied for ECL imaging. Image J software (Rawak Software Inc., Stuttgart, Germany) performed semi-quantitative analysis.

Mucosal biopsies were collected from 20 patients with UC undergoing colonoscopy or surgery at the Suzhou Hospital Affiliated to Nanjing Medical University between March 2021 and July 2021. Informed consent was obtained from all the patients, and the hospital ethics committees approved this study. Clinical disease activity was evaluated using a modified Truelove–Witts severity index [11]. Clinical characteristics of included ulcerative colitis patients have been shown in Table 1. The experimental procedures were performed as previously reported [12]. After fixing the tissues with 4% paraformaldehyde at 4 °C for 24 h, the sections were embedded in paraffin, and then the IHC steps were as follows: dewaxing in xylene and ethanol; preheating sodium citrate (pH = 6.0) antigen retrieval solution at high temperature for 10 min, the tissue section is boiled in the solution for 30 min; 3% H2O2 for 10 min; 3% BSA for 1 h. Immunohistochemical staining of paraffin-embedded tissue was conducted. RIP3 antibody (ab62344, Abcam), MLKL antibody (ab184718, Abcam) and CD68 antibody (ab125212, Abcam) were diluted with PBS buffer at a ratio of 1:100. The slides were incubated with antibodies overnight at 4 °C and then DAB solution for 3 min; hematoxylin counterstain for 2 min; 1% hydrochloric acid ethanol differentiation for 2 s; dehydrate with ethanol and xylene. After the xylene was drained, neutral gum was added to mount the slide. Staining results were examined under microscope with 20 × objective. Eight fields of view were randomly selected under microscope. Image-Pro Plus 6.0 software (Rawak Software Inc., Stuttgart, Germany) was used for semi-quantitative analysis.

The experimental procedures were performed as previously reported [13]. Total RNA was extracted by Trizol (Invitrogen, CA, USA). cDNA was synthesized by Hifair II 1st Strand cDNA Synthesis SuperMix (Yeasen, Shanghai, China). Follow the instructions of Hieff UNICON® qPCR SYBR Green Master Mix (Yeasen, Shanghai, China) to configure the Real-Time PCR reaction system and run qPCR program: step1—95 °C, 30 s; step2—95 °C, 5 s; step3—60 °C, 20 s; step4—72 °C, 20 s; step1–3 repeat 40 cycles. GAPDH was used as a housekeeping gene. The sequences of specific primers are as follows. GAPDH-F: CCTTCCGTGTCCCCACT, GAPDH-R: GCCTGCTTCACCACCTTC; RIP3-F: TCCAGGGAGGTCAAGGC; RIP3-R: ACAAGGAGCCGTTCTCCA. The expression level of each group of genes was analyzed using 2^−ΔΔCT.

Human intestinal epithelial cells were treated with TNF-α (10 ng/ml, 12 h) (96-300-01A-10, PeproTech) and then shRIP3 plasmid or 2 μM RIP3 inhibitor GSK-872 (CAS No. 1346546-69-7, MedChemExpress) for 24 h. 20 μl of 5 mg/ml MTT was added to each well. After 4 h of treatment with MTT (Beyotime Biotech, Shanghai, China), the supernatant was discarded, and 100μL of dimethyl sulfoxide (DMSO) was added. Optical density value (OD value) was detected at wavelengths of 492 nm and 630 nm, and the cell survival rate was calculated according to the formula: cell survival rate (%) = [experimental group OD average value-background group OD average value]/[Blank control group OD average value-background group OD average value] × 100%

1 × 10⁶ human intestinal epithelial cells were cultured in a six-well plate and treated with TNF-α and then shRIP3 plasmid or RIP3 inhibitor GSK-872 for 24 h. The cells were incubated with 5 μl V-FITC and 5 μl PI (BD Biosciences Pharmingen, CA, USA) for 15 min in the dark and then resuspend in 400 μl 1 × binding buffer. The proportion of apoptotic cells was analyzed by flow cytometer (Becton, Dickinson and Company, NJ, USA). FlowJo 7.6 software was applied for data analysis.

The experimental procedures were performed as previously reported [5]. Slices were prepared from 4% paraformaldehyde-fixed and paraffin-embedded mice colon tissues.

The steps of dewaxing and antigen retrieval was the same as those of immunohistochemistry. The sections were blocked with 3% BSA-PBS at room temperature for 1 h and incubated with anti-mouse antibody against ZO-1 (ab221547, Abcam, dilute at 1:100) overnight at 4 °C, then secondary antibody goat anti-rabbit IgG (H + L) with FITC fluorescent label (Jackson ImmunoResearch Inc., PA, USA, dilute at 1:100) at room temperature for 1 h. Nuclei was stained with 1 μg/ml DAPI (Sigma-Aldrich, MO, USA) for 10 min. Neutral gum is added to mount the slide. Slices were examined under fluorescence microscope with a 20 × objective.

2 × 10⁵ human intestinal epithelial cells were cultured in a six-well plate. A fluorescent probe 10 μM DCFH-DA (37 °C, 20 min) was used to detect the level of intracellular ROS. Eight fields of view were randomly photographed under fluorescence microscope.

A total of 20 6-week-old male BALB/c mice were purchased from Jie Si Jie Laboratory, Shanghai, China. Five RIP3 knockout mice [15] (C57BL/6N-Ripk3em1) was purchased from Cyagen, Guangzhou, China. The animal research was approved by the Ethics Committee of Nanjing Medical University (approval no. SYXK(S) 2020-0022). Mice are kept in a sterile environment with constant temperature and humidity, and are allowed to eat and drink freely. Mice were randomly divided into five groups (n = 5 per group). The experimental colitis model was established by 3% (w/v) DSS (Sigma-Aldrich; Merck KGaA) in drinking water for 7 days. On the 8th day, mice were intraperitoneally administered with Necrostatin-1 (Nec-1, a specific cell necroptosis inhibitor) (MedChemExpress, CAS No. 4311-88-0, 5 mg/kg), SASP (Merck KGaA, cat. no. 599-79-1, 50 mg/kg) and 1% DMSO, respectively, for 14 consecutive days. Mice were euthanized with an intraperitoneal injection of 150 mg/kg sodium pentobarbital, and then the heartbeat of the mice was observed. The duration of animal experiments was 2 weeks from the first day of treatment with DSS to sac.

Mice of each group were examined daily for body mass, diarrhea and blood in the stool. Disease activity index (DAI) = (weight loss score + fecal trait score + fecal occult blood score)/3. Colonic mucosal damage index (CMDI) was scored according to the criteria [16].

Distal colon tissue was fixed in 4% paraformaldehyde for 24 h at 4 °C, and then embedded in paraffin, sectioned to a thickness of 8 µm. The sections were subjected to the following steps: xylene dewaxing; gradient ethanol hydration; hematoxylin staining for 5 min; water washing for 5 min; 1% hydrochloric acid ethanol differentiation for 5 s; weak alkaline aqueous solution for 3 s; eosin staining for 5 min; gradient ethanol dehydration; Mounting with neutral gum. Histopathological changes were observed under a light microscope (magnification, 20 ×) and scored according to the criteria described by Andújar et al. [17]. The colon histopathology index (CHPI) was determined as previously described [18].

The whole spleen is separated, cut into pieces, and filtered through a 200-mesh sieve. The suspend mouse spleen cells were washed and adjusted to 1 × 10⁶/ml and blocked with 2.5% BSA for 1 h and then incubated with FITC-rat anti-mouse CD4 (catalog no RM4-5; eBioscience; Thermo Fisher Scientific, Inc.) for 0.5 h in the dark and fixed with Transcription Factor Staining Buffer Set (catalog no. 00-5523-00; eBioscience; Thermo Fisher Scientific, Inc.) for 15 min. Cells were then incubated with PE-rat anti-mouse Foxp3 (catalog no. 72-5775-40; eBioscience; Thermo Fisher Scientific, Inc.) for 1 h at 4 °C in the dark and detected by CytoFLEX V0-B5-R3 Flow Cytometer (Life Sciences, Indianapolis, Indiana).

The data were analyzed by Graphpad Prism 8.0 (GraphPad Software Inc., CA, USA), and expressed as Mean ± standard deviation (SD). For in vitro experiments, we repeated three times, each containing at least three replicate samples. Results comparing multiple groups were analyzed by one-way ANOVA followed by Tukey’s post hoc test. Results comparing two groups were analyzed by student’s t test. Difference among groups was considered significant when P value < 0.05.

Necroptosis is a novel regulatory necrosis pathway that requires receptor-interacting protein kinase 3 (RIP3) and mixed-lineage kinase domain-like protein (MLKL). It has been shown that changes in the expression of RIP3 and MLKL involved in excessive death of epithelial cells by a manner of necroptosis, thereby triggering intestinal inflammation. MLKL was the target of RIP3, acts downstream of RIP1 and RIP3, and is recruited into the necrosome through interaction with RIP3 [19]. We first applied The Human Protein Atlas (https://​www.​proteinatlas.​org/​search/​RIPK3) to observe the RNA expression of RIP3 in more than 60 human tissues and organs. We compared the RNA expression of RIP3 in duodenum, small intestine, colon, and rectum (Fig. 1A). Interestingly, among more than 60 human tissues and organs, RIP3 has the highest RNA expression in small intestines.

Subsequently, we collected clinical colon tissues from patients with mild, moderate, and severe UC according to the criterion of the modified Truelove & Witts severity classification [11]. Western blot results showed that the expression levels of RIP3 and MLKL significantly increased with the severity of UC increased. (Fig. 1B). Immunochemistry assay show that RIP3 was mainly distributed in the cytoplasm in the intestinal tissue of patients with mild and moderate UC, while RIP3 was mainly distributed in the cytoplasm and nucleus in the intestinal tissue of patients with severe UC (Fig. 1C). IHC semi-quantitative analysis showed that the expression level of RIP3 in the intestinal tissue of severe UC patients was higher than that in moderate UC patients (P < 0.01) (Fig. 1D). Compared with mild and moderate UC, the expression of MLKL in the intestinal tissue of patients with severe UC was higher (P < 0.05). (Fig. 1D).

TNF-α can induce either apoptosis or necroptosis in different cell lines [3]. RIP3 is involved in NF-κB-regulated programmed cell death, including apoptosis and necroptosis. RIP3 interacts with RIP1 with RIP homotypic interaction motif and then promotes apoptosis by activating caspases and/or inhibiting TNFR1-induced NF-κB activation [20, 21]. To investigate the effect of RIP3 knockdown on the survival of intestinal epithelia cells, we constructed the shRIP3 knockdown cell line in vitro. RT-qPCR verified that it could effectively reduce the mRNA level of RIP3 in normal human intestinal epithelial cells (Fig. 2A). Subsequently, MTT assay detected the effect of shRIP3 knockdown and RIP3 inhibitor GSK872 on cell viability after TNF-α (10 ng/ml, 12 h) induction. The results showed that TNF-α significantly inhibited cell proliferation (P < 0.001) (Fig. 2B). In contrast, RIP3 knockdown and GSK872 could significantly rescue the proliferation inhibitory effect of TNF-α on intestinal epithelial cells (P < 0.01) (Fig. 2B).

Then, we used Annexin V-FITC and PI double staining to detect the effects of shRIP3 knockdown and GSK872 on cell apoptosis. Flow cytometry results showed that TNF-α enhanced the proportion of apoptotic cells (P < 0.01) while RIP3 knockdown and GSK872 treatment significantly reversed the apoptosis-inducing effect of TNF-α on intestinal epithelial cells (P < 0.05) (Fig. 2C–D). Next, we observed PI staining results with a fluorescence microscope and found that TNF-α promoted red fluorescence intensity, and RIP3 knockdown and GSK872 inhibited red fluorescence intensity (Fig. 2E). Taken together, RIP3 inhibition could rescue TNF-α-induced apoptosis in normal human intestinal epithelial cells.

To explore the effect of RIP3 on activation of inflammatory cytokines, we performed ELISA to detect the effect of shRIP3 knockdown and GSK872 on the secretion of inflammatory factors IL-16, IL-17 and IFN-γ after TNF-α (10 ng/ml, 12 h) treatment. We observed that TNF-α significantly induced the secretion of IL-16, IL-17 and IFN-γ (P < 0.05) (Fig. 3A–C). RIP3 knockdown plasmid could inhibit the secretion of IL-6 and IL-17 (P < 0.05), and RIP3 inhibitor GSK872 inhibited the secretion of IL-6 and IFN-γ (P < 0.05) (Fig. 3A–C). It was reported that RIP3 could promote apoptosis by inhibiting NF-κB activation [21]. However, the influence of RIP3 on NF-κB activation remains controversial. Subsequently, we used western blot to detect the effects of RIP3 knockdown and GSK872 on the expression of key molecules in the TLR4/MyD88/NF-κB signaling pathway. Experimental results showed that TNF-α (10 ng/ml, 12 h) significantly promoted the expression of Myd88, TLR4, p-IκB, and p-65 in intestinal epithelial cells. RIP3 plasmid and GSK872 suppressed the expression of IL-16, IL-17 and IFN-γ induced by TNF-α (Fig. 3D). Under normal physiological conditions, cells control ROS levels by balancing the production and elimination of ROS. Under oxidative stress conditions, excessive ROS can destroy cellular proteins, lipids and DNA, leading to cellular carcinogenesis [22]. Next, we explored whether the pro-apoptotic effect of RIP3 was related to the regulation of intracellular ROS levels. After treating the cells with TNF-α (10 ng/ml, 12 h), RIP3 knockdown and GSK872, we incubated the cells with DAFH-DA probe. We observed under fluorescence microscope that TNF-α could obviously induce intracellular ROS accumulation compared with the blank control group. Compared with the TNF-α control group, RIP3 knockdown and GSK872 significantly inhibited ROS production (Fig. 3E).

Some studies have reported the therapeutic effects of RIP3 knockout and Nec-1 on dextran sulfate sodium (DSS)-induced colitis in mice [15, 23‐25]. However, the effect of RIP3 deficiency on inflammation is still controversial [15, 25]. To investigate the effect of RIP3 on intestinal pathology and histopathology in DSS-induced colitis mouse model in vivo, we gave BALB/c wild-type and RIP3 knockout mice 3% DSS solution, freely drinking for 7 days to establish the experimental colitis model. On the 8th day, Nec-1 (5 mg/kg), sulfasalazine (50 mg/kg) and placebo (1% DMSO) were injected intraperitoneally for 14 consecutive days. No mice died during the modeling process. After 24 h of modeling, the model animals began to show loose stools, poor mental state, rough fur, reduced food intake, and weight loss. Compared with the WT + DSS control group, the general state of the mice in the WT + DSS + SASP and WT + DSS + Nec-1 group gradually improved, and the DAI score gradually decreased. Compared with wild-type mice, RIP3 knockout mice gradually recovered from DSS-induced UC after stopping drinking DSS water, (Fig. 4A).

Next, the gross morphology of the colon and CMDI score showed that the surface of the colon of the normal control group was smooth and the texture was clear. In the WT + DSS group, the surface of the intestinal cavity was uneven, and inflammatory congestion and edema were more obvious, accompanied with the enhanced adhesion of intestinal wall to the mesentery. Compared with the WT + DSS group, the intestinal cavity surface congestion and edema in the WT + DSS + SASP group were significantly relieved. The length of the colon was similar to the normal control group, and the CMDI score was significantly reduced (P < 0.05). Local congestion and edema were seen in the colon of mice in the WT + DSS + Nec-1 group, mainly in the distal colon, and the CMDI score was significantly decreased (P < 0.05) (Fig. 4B–D). The colon of RIP3 KO mice still showed obvious congestion and edema, and the CMDI score was slightly lowered, but it was not statistically significant (Fig. 4B–D). The results of H&E staining indicated that there was no obvious inflammatory cells infiltration in normal control group, and a complete intestinal mucosal structure was visible. The structure of the colon glands in the WT + DSS group was severely damaged, and the glands were arranged disorderly. A large number of lymphocytes and neutrophils infiltrated. Compared with the WT + DSS group, the histopathological changes of the colon in the WT + DSS + SASP group were significantly reduced; the colonic mucosa of the WT + DSS + Nec-1 group was relatively intact with a small amount of inflammatory cell infiltration (Fig. 4E).

The balance of intestinal epithelial cell death and proliferation is the basis for maintaining the intestinal mucosal barrier function. Excessive cell death and shedding may cause intestinal barrier dysfunction [26]. The intestinal mucosal barrier is composed of an epithelial cell layer and a mucous layer, which can effectively prevent the invasion of bacteria, toxins and inflammatory mediators in the intestinal lumen. Tight junction (TJ) between intestinal mucosal epithelial cells can prevent harmful substances from penetrating into the submucosa, and the main tight junction proteins include ZO-1 protein [27]. The results of immunofluorescence staining showed that the colon of the model group showed decreased fluorescence intensity of ZO-1compared with the normal control group. Compared with the model control group, the fluorescence intensity of ZO-1 of the WT + DSS + SASP group and the WT + DSS + Nec-1 group increased significantly. However, there was no significant change in the RIP3 KO + DSS group (Additional file 1: Figure S1A).

Altered expression of molecules involved in the necroptotic pathway, causing excessive epithelial cell death, has been shown to trigger intestinal inflammation [7, 28]. Regulatory T cells (Treg) are important for the homeostasis of the immune system. The functional defects of Treg cells or their low proportion in all T cells may lead to inflammatory diseases. Numerous studies have shown that the proportion of Treg cells in peripheral blood mononuclear cells in patients with UC has decreased. Treg has plasticity, such as loss of Foxp3 expression and increased levels of pro-inflammatory cytokine IL-17 and IFN-γ [29, 30]. ELISA results showed that compared with the normal control group, the level of cytokine IL-17 in the colon tissue of the model group was significantly increased, while the level of IL-10 was reduced (P < 0.01). Compared with the model control group, IL-17 and IFN-γ levels in the WT + DSS + SASP group were decreased (P < 0.01), while IL-10 levels increased (P < 0.05). The level of IL-17 in the WT + DSS + Nec-1 group increased slightly (P < 0.05), while the level of IL-10 increased significantly (P < 0.01) (Fig. 5A–C). Consistently, the level of IL-17 in the RIP3 KO + DSS group decreased slightly (P < 0.01). Further, flow cytometry results showed that compared with the blank control group, the ratio of CD4⁺Foxp3⁺ T cells to CD4⁺ T cells in the spleen of the model control group decreased (P < 0.01). Compared with the model control group, the proportion of CD4⁺Foxp3⁺ T cells in the spleen of mice in the WT + DSS + SASP group and WT + DSS + Nec-1 group increased significantly, and the difference was statistically significant (P < 0.05). The proportion of Treg cells in the RIP3 KO + DSS group was almost unchanged (Fig. 5D-E).

To understand the role of RIP3 in the regulation of inflammatory signaling pathways and ROS generation in vivo, we used IHC to detect the proportion of CD68⁺ cells in the mouse colon. CD68⁺ monocyte/macrophage infiltration is the hallmark for colonic inflammation. Semi-quantitative analysis showed that the proportion of CD68-positive cells in the colon of colitis mice was significantly increased (P < 0.05). Compared with the model control group, the proportion of CD68⁺ cells in the colon of the WT + DSS + SASP group decreased (P < 0.05), which the proportion of CD68⁺ cells in the WT + DSS + Nec-1 group also decreased, but it was not statistically significant. Consistent with WT + DSS + SASP group, CD68⁺ cells in the RIP3 KO + DSS group was slightly increased (P < 0.05) (Fig. 6A–B).

Subsequently, western blot detected the effect of RIP3 knockout or knockdown on key molecules of TLR4/NF-κB signaling pathway in vivo. The results showed that DSS induces the expression of TLR4, p65 and p-IκB in mouse colon tissue. Compared with the model control group, the expression of the TLR4, p65 and p-IκB in the colon of the WT + DSS + SASP group and WT + DSS + Nec-1 group was significantly down-regulated. The levels of p65 and p-IκB in the RIP3 KO + DSS group were significantly down-regulated (Fig. 6C). In order to explore whether RIP3 knockout or knockdown in vivo exerted an anti-apoptotic effect on cells in colon by regulating ROS, we collected the colon tissue of mice 14 days after the administration and tested the levels of SOD, MDA and MPO. The results showed that the SOD level in the colon tissue of the colitis mice was significantly reduced compared with the normal control group (P < 0.05), while the MDA and MPO levels were significantly increased (P < 0.05). Compared with the model control group, the SOD level in the WT + DSS + SASP group increased significantly (P < 0.05), while the levels of MDA and MPO decreased (P < 0.05). The SOD level in the WT + DSS + Nec-1 group increased significantly (P < 0.01), while the MDA level decreased significantly (P < 0.05). Besides, the MPO level in the RIP3 KO + DSS group decreased significantly (P < 0.01), suggesting that RIP3 inhibitor Nec-1 can alleviate the oxidative damage of the mouse colon induced by DSS (Fig. 6D–F).