Biochemical Pharmacology

RIPK1 inhibitor ameliorates colitis by directly maintaining intestinal barrier homeostasis and regulating following IECs-Immuno crosstalk
Huimin Lu, Heng Li, Chen Fan, Qing Qi, Yuxi Yan, Yanwei Wu, Chunlan Feng, Bing Wu, Yuanzhuo Gao, Jianping Zuo, Wei Tang
PII: S0006-2952(19)30450-2
Reference: BCP 113751

To appear in: Biochemical Pharmacology

Received Date: 11 October 2019
Accepted Date: 9 December 2019

Please cite this article as: H. Lu, H. Li, C. Fan, Q. Qi, Y. Yan, Y. Wu, C. Feng, B. Wu, Y. Gao, J. Zuo, W. Tang, RIPK1 inhibitor ameliorates colitis by directly maintaining intestinal barrier homeostasis and regulating following IECs-Immuno crosstalk, Biochemical Pharmacology (2019)

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

© 2019 Published by Elsevier Inc.

Title page: RIPK1 inhibitor ameliorates colitis by directly maintaining intestinal barrier homeostasis and regulating following IECs-IMMUNO crosstalk
Huimin Lu1,2, Heng Li1,2, Chen Fan1, Qing Qi1, Yuxi Yan1,2, Yanwei Wu1, Chunlan Feng1, Bing Wu1,2, Yuanzhuo Gao1,2, Jianping Zuo1,2*, Wei Tang1,2*
Affiliations:1 Laboratory of Immunopharmacology, State Key Laboratory of Drug Research,
Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Shanghai 201203, China

This work was funded by the “Personalized Medicines—Molecular Signature-based Drug Discovery and Development”, Strategic Priority Research Program of the Chinese Academy of Sciences (Grant No. XDA12020370) and the Science & Technology Commission of Shanghai Municipality, China (Grant No.18431907100).

1 RIPK1 inhibitor ameliorates colitis by directly maintaining intestinal barrier
2 homeostasis and regulating following IECs-IMMUNO crosstalk
3 Abstract
4 Background: The receptor-interacting protein kinase 1 (RIPK1) has emerged as a key upstream
5 regulator that controls the inflammatory response via its kinase-dependent and independent
6 functions, which makes it an attractive target for developing new drugs against
7 inflammation-related diseases. Growing evidences illustrate that RIPK1 is certainly associated
8 with pathogenesis of multiple tissue-damage diseases. However, what are intricate regulatory
9 codes of RIPK1 inhibitor in diseases is still obscure.
10 Methods: We used DSS-induced colitis model in vivo to study the therapeutic effects and the
11 mechanisms of RIPK1 inhibitor. We next characterized the barrier function and the interaction
12 between intestinal epithelial cells (IECs) and immunocytes both in vivo and in vitro. As a
13 candidate in clinical study, GSK2982772 is the most well-developed drug of RIPK1 inhibitors,
14 and we chose it as our study object.
15 Results: We demonstrated that RIPK1 inhibitor could ameliorate the intestinal barrier injury by
16 reducing tight junctions’ disruption and accompanying oxidative stress. Moreover, the release of
17 chemokines and adhesion molecules from damaged IECs was suppressed by RIPK1 inhibitor
18 treatment. And these protective effects were not only dependent on the suppression of necroptosis
19 but also on the compromised activity of NF-κB. Taken together, RIPK1 inhibitor exerts
20 suppressive function in intestinal inflammatory response possibly via protecting the intestinal
21 epithelial barrier and maintaining the homeostasis of immune microenvironments. Eventually, the
22 positive feedback immune response which triggered progressive epithelial cells injury could be
23 restrained.

24 Keywords: RIPK1 inhibitor; Colitis; Barrier homeostasis; IECs-immunocytes crosstalk;
25 necroptosis; NF-κB;
26

27 1. Introduction
28 The receptor-interacting protein kinase 1 (RIPK1), has emerged as an important upstream
29 kinase which could affect multiple cellular pathways associated with regulating inflammation.
30 While, there is controversy about the feature of RIPK1 in the pathogenesis of colitis. For one side,
31 current studies deem that RIPK1 has a pathogenic mechanism in colitis due to epithelial
32 necroptosis[1, 2], which process is dependent on the its kinase activity. The activated RIPK1 could
33 recruit RIPK3 and MLKL to form necrosome to contribute proinflammatory response[3]. For
34 another, in RIPK1 deficient mice, the essential protective effects of RIPK1 in intestinal
35 homeostasis reside in its kinase-independent scaffold function, which can inhibit RIPK3-mediated
36 necroptosis by RIP homotypic interaction motif (RHIM)[4-6]. However, without genetic
37 intervention, evidence about the role of RIPK1 in human inflammatory bowel disease (IBD)
38 remains limited[3]. The exact contribution of RIPK1 in colitis is an indispensable question for
39 drug development. It is worth mentioning that the inhibitors targeting RIPK1 are implementing in
40 pharmacodynamics studies of inflammatory diseases. Currently, GSK2982772 is in multiple Phase
41 2 clinical trials to treat inflammatory diseases [7]. However, the pharmacological mechanism of it
42 has not been reported and the effects of RIPK1 inhibitor in colitis have abundant room for further
43 progress in determining.
44 Ulcerative colitis (UC) is an inflammatory bowel disease characterized by mucosal barrier
45 damage and immune cells infiltration [8]. The intact barrier of intestinal could defense the
46 invasion of pathogens and antigens[9]. In pathologic states, a leaky epithelial barrier results in
47 excessive exposure to microbial antigens, recruitment of immune cells, release of inflammatory
48 mediators, and eventually leading to progressive intestinal mucosa injury [10]. The integrity of the
49 epithelial barrier largely depends on intestinal epithelial cells (IECs) and intercellular junctions
50 that include tight junctions (TJ) [9].
51 As an indispensable component of the mucosal barrier, IECs are of great importance on
52 maintaining barrier integrity and homeostasis. Some reports about the relationship between RIPK1
53 function and IECs homeostasis were limited to necroptosis, which lead to loss of integrity of
54 intestinal barrier[11]. Meanwhile, dying cells indirectly trigger inflammation by releasing
55 damage-associated molecular patterns (DAMPs)[3]. Besides necroptosis, there may be some
56 interesting mechanisms involve in the effects of RIPK1 in mediating barrier structural and

57 functional disorders. Myosin light chain kinase (MLCK) is a key effector of barrier dysfunction
58 and a potential therapeutic target. Intraperitoneal TNF administration can induce barrier loss and
59 enhance intestinal epithelial MLC phosphorylation[12]. Further work is required to establish
60 whether there is a pertinence between RIPK1 and MLC phosphorylation. Moreover, Reactive
61 oxygen species (ROS) is generally labeled as a proinflammatory factor, and its overproduction has
62 been strongly related to Crohn’s disease and pancolitis[13].The oxidative stress of IECs and the
63 action of RIPK1 in this process are deserved to study.
64 Finally, in consideration of the limited reports about the immunosuppressive function of
65 RIPK1 inhibitor on immunocytes, it is a novel point to focus on the pharmacological mechanisms
66 of RIPK1 inhibitor in IECs and following IECs-immunocytes crosstalk. Delicate and precise
67 interactions between epithelial cells and immune cells determine mucosal homeostasis. Even a
68 slight deviation might lead to epithelial barrier injury, translocation of luminal antigens and
69 immune responses[9]. Intercellular adhesion molecules (ICAMs) and chemokines generated by
70 epithelial cells could recruit immune cells and have become integral in matters of crosstalk
71 between epithelial cells and immunocytes. Elevated ICAMs are identified as trigger factors for
72 immunocytes adhesion, migration and local lymphocytes activation, and are responsible for
73 immunocytes trafficking in the focal inflammation. Epithelial ICAM-1 is related to inflammation
74 response by contributing to immune cells migration to peri-epithelial sites, and its blockade has
75 become a therapeutic target for colitis[14, 15]. Chemokines and their receptors as well as perform
76 indispensable functions in orchestrating tissue-specific leukocytes trafficking[16]. There is a
77 growing body of clinical data that the elevated levels of chemokines and their receptors were
78 observed in patients with IBD, and decreasing their production may help to ameliorate
79 diseases[17-19]. There have arisen several novel drugs target chemokines to treat colitis in clinical
80 trials and showed promising therapeutic results[20]. It is meaningful to explore the RIPK1-related
81 adhesion and chemotaxis behavior of leukocytes trafficking to epithelial cells, due to a noticeable
82 role of the crosstalk between IECs and immunocytes in colitis etiology.
83 In this study, we developed the murine colitis model to explore the possible mechanisms how
84 RIPK1 inhibitor protects barrier injury and interferes with the interaction between IECs and
85 immunocytes.

87 2. Materials and methods
88 2.1. Reagents and antibodies
89 GSK2982772, a RIPK1 inhibitor, was purchased from Med Chem Express (NJ, USA) and
90 dissolved in dimethyl sulfoxide (DMSO) as a stock solution. Dextran Sulfate Sodium (DSS,
91 molecular weight 36-50 kDa) was purchased from MP Biomedicals (Irvine. CA, USA).
92 (Hydroxypropyl)methyl cellulose (HPMC) was purchased from Sigma-Aldrich (St Louis, MO,
93 USA). Recombinant human TNF-α was obtained from Peprotech (London, UK). Fecal occult
94 blood test kit was obtained from Nanjing Jiancheng Bioengineering Institute (Nanjing, China).
95 RIPK3i (GSK872), MLKLi (NSA), NF-κBi (TPCA-1), SM164 and ZVAD were purchased from
96 Med Chem Express (NJ, USA). N-acetyl-cysteine was purchased from Sigma-Aldrich (St Louis,
97 MO, USA). Proteinase and phosphatase inhibitor and BCA assay kit were purchased from
98 Thermo Scientific (Pittsburgh, PA, USA). FITC-dextran was purchased from Sigma (St Louis,
99 MO, USA). Cell Titer-Glo Luminescent Cell Viability Assay was purchased from Promega
100 (Madison, USA). Reactive Oxygen Species Assay Kit was purchased from Beyotime (Haimen,
101 China). Calcein AM was purchased from Abcam (Cambridge, UK). In Situ Cell Death Detection
102 Kit (TUNEL) was purchased from Roche Diagnostics GmbH (Mannheim, Germany).
103 Anti-mCD16/CD32, PE-anti-F4/80, PE-anti-Ly6G PE-anti-CD25, PE-anti-CD278,
104 PE-anti-IL-17, PercP-Cy5.5-anti-CD11b, PercP-Cy5.5-anti-CD69, APC-anti-γδTCR,

105 APC-anti-Ly6c, APC-anti-IFNγ were purchased from BD Biosciences (Franklin Lakes, NJ, USA).

106 ELISA kits for mouse IL-17, IL-1β, IL-6, TNF-α were purchased from eBioscience (San Diego,
107 CA, USA). Human IL-8, CXCL10 ELISA kits were purchased from Biolegend (San Diego, CA,
108 USA). Anti-p-RIPK1, Anti-p-RIPK3, Anti-ICAM1, Anti-IκBα, Anti-p-P65, Anti-p-MLCK and
109 anti-Ecadherin were obtained from Cell Signaling (Danvers, MA, USA). Anti-mICAM1,
110 Anti-CD3, Anti-HSP90 and FITC-anti-CD11b were purchased from Abcam (Cambridge, UK).
111 Anti-Zo-1 and Anti-p65 were purchased from Proteintech (Chicago, IL, USA). Anti-Occludin was
112 purchased from Thermo Fisher Scientific (Waltham, MA, USA). Anti-HMGB1 was purchased
113 from Serotec (Oxford, UK).
114 2.2. Cell culture
115 The human colonic adenocarcinoma cell line HT-29, Caco-2 cells, the human acute

116 monocytic leukemia cell line THP-1 cells and T lymphocyte cell line Jurkat cells were purchased
117 from American Type Culture Collection (ATCC, Manassas, VA, USA). Cells were culture din
118 McCoy’s 5a Medium, DMEM and RPMI 1640 medium (Gibco, Grand Island, NY, USA),
119 respectively, containing 10% fetal bovine serum (HyClone, Logan, UT, USA), 100 U/ml penicillin
120 and 100 µg/ml streptomycin. Cells were cultured in a humidified incubator with 5% CO2 at 37°C.
121 2.3. Mice
122 Female C57BL/6J mice (6-8 weeks, 18-20 g) were obtained from Shanghai Lingchang
123 Biotechnology Co.Ltd. (certificate no. 2013-0018, shanghai, China), and were maintained at the
124 specific pathogen-free (SPF) animal facilities of Shanghai Institute of Materia Medica. All
125 experiments were performed on the basis of the guidelines of the Association Assessment and
126 Accreditation of Laboratory Animals Care International. And all of the procedures were carried
127 out strictly in accordance with the animal care and use protocol (2018-03-TW-06) approved by the
128 Institutional Animal Care and Use Committee (IACUC) at Shanghai Institute of Materia Medica.
129 2.4. Models of colitis
130 Acute colitis was achieved by feeding of 3% (w/v) DSS (MP Biomedicals, molecular mass
131 36,000-50,000 Da) in drinking water. Mice were received either regular drinking water (normal
132 control and normal+RIPK1i) or DSS drinking water (vehicle and RIPK1i) for 6 days followed by
133 3 days of regular drinking water. Mice were randomly divided into 4 groups with 10 mice per
134 group: normal control (HPMC), Normal+RIPK1i (GSK2982772, 20 mg/kg), vehicle control
135 (HPMC) and RIPK1i (GSK2982772, 20 mg/kg). Mice were orally administered once daily for 9
136 days and sacrificed at day 10.
137 Additionally, we performed an independent experiment and sacrificed mice at disease active
138 stage (Day6) for observing the treatment effect of RIPK1i in acute phase
139 2.5. Clinical assessment of colitis
140 For monitoring the severity of colitis, body weight, stool consistency and rectal bleeding
141 were assessed daily. Values assessed prior to DSS exposure served as baseline. Weight changes
142 were calculated in relation to the weight at baseline (100%). And the disease activity index (DAI)

143 was calculated based on the scoring system[21].

144 2.6. Histology

145 Intestinal tissues were fixed in 4% paraformaldehyde, embedded in paraffin. For
146 histopathological analysis, hematoxylin and eosin staining was performed according to standard
147 protocol. Histological evaluation of H&E-stained colonic sections was achieved by two
148 independent observers blinded to the experimental conditions and graded as previously
149 described[22]. For immunofluorescence colonic tissues were embedded in OCT compound, and
150 sectioned on a cryostat (6 µm thick). After fixed in paraformaldehyde, sections were blocked with
151 5% BSA for 60 min and then incubated with FITC-anti-CD11b (Abcam) at 4°C overnight. The
152 sections were counterstained with 4′, 6-diamidino-2-phenylindole (DAPI) (Abcam). Fluorescent
153 sections were visualized and images were captured using a Leica TCS SP8 STED confocal
154 microscope. Immunohistochemical staining was performed on formalin-fixed paraffin-embedded
155 tissues with rabbit anti-CD3(Abcam), rat anti-HMGB1 (Serotec) antibodies. These sections
156 microscopy was performed using a Leica DM 6B microscope.
157 2.7. Immunofluorescence cytochemistry
158 HT-29 cells on coverslips were fixed in fixing solution (Beyotime, China) for 15 min. After
159 treated blocking buffer (Beyotime) for 60 min, cells were incubated with rabbit anti-Ecadherin
160 (Cell signaling), Zo-1(Proteintech), Occludin (Thermo) respectively overnight at 4°C. The FITC
161 or PE-conjugated anti-rabbit secondary antibodies (Proteintech, Rosemont, USA) was added, after
162 washing with 1% PBS-Tween. Negative control reactions were included in each experiment and
163 carried out by replacing primary antibodies with PBS. The cells were counterstained with DAPI.
164 All images were captured using a Leica TCS SPS CFSMP microscope.
165 2.8. TdT-mediated dUTP Nick-End Labeling (TUNEL)
166 Cell death was analyzed with an in-situ cell death detection kit (TMR-red, Roche).
167 Fluorescence microscopy was performed using a Leica TCS SPS CFSMP microscope.
168 2.9. Assessment of myeloperoxidase (MPO) activity
169 For measured the neutrophil infiltration into inflamed colonic mucosa, MPO activity was
170 detected by O-dianisidine method as previously described[23]. The results were showed as activity
171 units per gram tissue.
172 2.10. Single cell preparation and Flow cytometry analysis
173 For single cell suspension preparation, spleens and mesenteric lymph nodes (MLNs) from

174 mice were grinded and filtered through a 40 µm nylon mesh strainer. Colons were cut into small
175 pieces after removing intestinal contents and residual fat. And turned the tissue inside out by
176 cannulating the intestinal segments with curved forceps. Cleaned colon fragments were incubated
177 in extraction media (30 ml RPMI + 93 µl 5% (w/v) dithiothreitol (DTT) (Meilun Biotechnology,
178 Dalian, China)+ 60 µl 0.5 M EDTA (Sigma, USA))+ 500 µl fetal bovine serum (FBS)) for 15min
179 on constant temperature shaker (600 rpm) to dissociate epithelial cells, and then added the minced
180 small intestine to a centrifuge tube containing 25 ml of digestion media (25 ml RPMI + 12.5 mg
181 dispase + 37.5 mg collagenase II + 300 µl FBS) for incubating at 500 rpm for 30 min at 37 °C.
182 Filter digested tissue through a 100 µm cell strainer into a 50 ml tube. After centrifugation, filter
183 resuspended cells through a 40 µm cell strainer into a 50 ml tube. Finally, resuspend pellet in 1 ml
184 of RPMI containing 2% FBS.
185 The single-cells unpensions were blocked with anti-mCD16/CD32 (2.4G2) and then stained
186 with antibodies in dark. Flow cytometric analysis was performed on BD LSR Fortessa, and data
187 were analyzed using FlowJo 10 software (Treestar, Ashland, OR).
188 2.11. FITC–dextran intestinal permeability assay
189 Intestinal permeability was assessed by oral gavage of FITC–dextran 4KD (Sigma), a
190 macromolecule that cannot be metabolized and is used as a permeability probe. Mice were
191 administered 200 μl FITC–dextran (600mg/kg bodyweight) by oral gavage 4h before killing.
192 Whole blood was obtained by removing eyeball, and FITC–dextran levels of serum were
193 measured by fluorometry (Ex:488 nm, Em:525 nm). For observing permeable fluorescence signal
194 directly, mice were imaged in vivo by PE IVIS spectrum (Perkin Elemer).
195 In vitro, FITC-dextran permeability assays were performed as previously described[24]. In
196 brief, 5×104 Caco-2 cells were seeded onto 0.4μm pore size trans-well inserts (Corning, NY, USA)
197 and grown to confluence. After hTNFα (100ng/ml) treatment for 24 hours, 10 μg/mL
198 FITC-dextran 4KD solution was added to the inserts. After allowing 2h for diffusion, the medium
199 from the lower chamber was collected and analyzed on a fluorescence plate reader.
200 2.12. Transepithelial electrical resistance (TEER)
201 To measure the intestinal barrier function, 5×104 Caco-2 cells were seeded on the apical side
202 of 0.4μm pore size transwell polyester membrane filters,(Corning, NY, USA), and integrity of cell

203 monolayers was determined with trans epithelial electrical resistance (TEER) using an epithelial
204 Volt-Ohm Meter (Millicell ERS2)[25].
205 2.13. In vitro leukocytes adhesion
206 HT-29 cells plated on 24-well plates were treated with RIPK1i (50nM) for 30min and then
207 stimulated with TNFα (100ng/ml) for 24 h. THP-1 monocytes (2×105 cells/mL) were fluorescently
208 labeled with 2.5 μg/mL Calcein AM (Abcam) in RPMI-1640 medium for 30min. After washing
209 twice with PBS, Calcein AM-labeled THP-1 cells were resuspended in fresh RPMI-1640 medium
210 and then added at 105 cells/well onto a HT-29 monolayer and incubated at 37℃ for 30 min.
211 Non-adherent THP-1 cells were washed away by PBS. The representative image of adherent
212 THP-1 cells were captured under a fluorescence microscope (Olympus IX73)[26].
213 2.14. Chemotaxis assay
214 HT-29 cells were treated with RIPK1i (50nM) for 30min and then stimulated with TNFα
215 (100ng/ml) for 24 h. Subsequently, conditional medium (600μl) were collected and added into the
216 lower chamber of 24-well trans-well chambers with 8 µm pores (Corning, NY, USA). THP-1 and
217 Jurkat cells (1 × 106 cells/mL) were labeled with 2.5 μg/mL Calcein AM in RPMI-1640 medium
218 for 30min. After washing twice with PBS, THP-1 and Jurkat cells were resuspended in fresh
219 RPMI-1640 medium and then added at100μl /well onto upper chamber of trans-well for 2 or 4 h
220 incubated at 37℃ respectively. The number of cells in lower chamber was detected by cytometer
221 and fluorescence microscope (Olympus IX73)[27].
222 2.15. Quantitative real-time PCR
223 Total RNA was isolated from HT-29 cells and colonic tissues by using RNA simple total
224 RNA kit (Tiangen, Beijing, China). Then total RNA was reverse transcribed by an All-in-One
225 cDNA Synthesis SuperMix (Biotool, Houston, TX, USA). Quantitative real-time PCR was
226 performed with SYBR® Green Realtime PCR Master Mix (TOYOBO, Osaka, Japan) on a 7500
227 Fast Real-Time PCR System (Applied Biosystems, Foster city, CA, USA).
228 2.16. Cytokine analysis by ELISA
229 Colons from mice were homogenized with lysis buffer to extract total protein as described by
230 Janice J. Kim et al [23]. The concentration of total protein was determined by BCA protein assay
231 kit. Cytokines level of TNF-α, IL-1β, IL-6, IL-17 in colon homogenate and chemokines level of

232 CXCL10, IL-8 in HT-29 cell culture supernatant were quantified by ELISA kit (Biolegend).
233 2.17. Western blot analysis
234 The colon tissues were homogenized and HT-29 cells were lysed in sodium dodecyl sulfate
235 lysis buffer (Beyotime) containing proteinase and phosphatase inhibitor. The protein
236 quantification was determined by BCA Protein Assay Kit (Thermo Scientific) The equal amounts
237 of proteins were separated on SDS-polyacrylamide gel electrophoresis and transferred to a
238 nitrocellulose membrane (Amersham Pharmacia Biotech, Buckinghamshire, UK). After blocking,
239 the membranes were incubated with primary antibodies overnight at 4 ℃. After washing with TBS
240 with Tween-20, the secondary antibodies (1:20000, Bio-Rad, Richmond, CA, USA) were added,
241 and HRP-conjugated monoclonal mouse anti-GAPDH (1:10000, Kangcheng, Shanghai, China) as
242 control for normalization. Signals were detected with ECL system (Amersham Bioscience,
243 Buckinghamshire, UK) and exposed to classic autoradiography film or Amersham Imager 600
244 (GE).
245 2.18. Statistical analysis
246 Data were presented as mean ± SEM and all group data subjected to statistical analysis in the
247 present research have a minimum of n=3 individuals per group (the value of n is indicated in the
248 specific legend). The results of western blot, flow cytometry and morphology analysis were
249 presented as representative images. Statistical analyses were conducted using GraphPad Prism 5.0
250 software. All experiments were repeated at least three times, with similar results. Significant
251 differences between groups were determined using a one-way ANOVA with Dunnet’s multiple
252 comparisons test with no significant variance inhomogeneity (F achieved p<0.05) and p<0.05 was
253 considered to represent a significant difference.
254 3. Results
255 3.1. RIPK1 inhibitor ameliorated DSS-induced colitis and suppressed the proinflammatory
256 cytokines and DAMPs production
257 In order to confirm the pharmacological action of RIPK1 inhibitor in colitis, Dextran Sulfate
258 Sodium (DSS) induced acute colitis model was performed . To get more information
259 about the possible toxicity, the naïve mice were treated with the RIPK1 inhibitor. The evaluated
260 indicators, including body weight, DAI, colon length, serum ALT and ALP and pathological
261 sections, suggested that RIPK1 inhibitor treatment has no adverse events in indicated dose.

262 1B-E). Sustained body weight loss and disease activity index (DAI) rise were relieved to a certain
263 extent during RIPK1 inhibitor (RIPK1i, GSK2982772 20mg/kg) administration ). The
265 serum ALT and ALP, which indicate the liver function, showed improvement under RIPK1i
266 treatment . The severity of histology disruption was analyzed by H&E staining.
267 Representative microscopic images and histopathological scores of colonic sections reflected the
268 improvement effect of RIPK1 inhibitor . As expected, the phosphorylation level of
269 RIPK1 did increase in the vehicle group, which indicated the pathogenetic role of RIPK1
270 activation in colitis . Increased release of proinflammatory cytokines and DAMPs
271 contributed to the development of colitis. RIPK1 inhibition could suppress these cytokines levels
272 both in serum and colonic tissue . The production of high mobility group box1
273 (HMGB1) and heat shock protein (HSP90) also was suppressed by RIPK1 inhibitor treatment
275 3.2. RIPK1 inhibitor suppressed the immune response in colitis but had few direct effects on
276 immunocytes in vitro
277 The secondary immune organs, such as spleen and mesenteric lymph nodes, can be driven to
278 exert immune defense functions with the development of colitis. However, aberrant immunocytes
279 activation can deteriorate the progression of disease. We prepared a single cell suspension isolated
280 from spleen and mesenteric lymph nodes to test the marker of T cell activation, including CD25,
281 CD69 and CD278 gated from CD3 and CD4 positive cellular population. Compared to the vehicle
282 controls, RIPK1 inhibitor-treated mice showed an obvious reduction in the percentage of activated
283 T lymphocytes. (. It has been reported that γδT cells are involved in the exacerbation of
284 colitis[28]. In colitis model, we observed the increase of γδT cells in spleen can be controlled by
285 RIPK1 inhibitor administration. . Blockage the differentiation of Th17 cells could
286 ameliorate colitis[29]. In our study, RIPK1 inhibition intervened the differentiation of Th17 cells
287 in spleen . In conclusion, inhibition of RIPK1 delayed the immune response by
288 modulating the activation and differentiation of T cells in immune organs. Furthermore, we found
289 that RIPK1 inhibitor could reduce a high level of innate immunocytes both in mesenteric lymph
290 nodes and spleen, such as neutrophil (CD11b+ly6c+/ly6G+) and macrophage (CD11b+F4/80+)

291 To determine whether RIPK1 inhibitor has direct immunosuppressive effects on
292 immunocytes, we prepared spleen lymphocytes to investigate the suppressive function of RIPK1
293 inhibitor on the proliferation status, which stimulated by Concanavalin (ConA) and LPS. The
294 results suggested that RIPK1 inhibitor had the comparable value of CC50 and IC50. RIPK1 kinase
295 inhibition may have negligible effects on lymphocytes activation . Furthermore,
296 we performed relative experiments on bone marrow-derived macrophage (BMDM) and bone
297 marrow-derived dendritic cells (BMDC). The results revealed that RIPK1 and necroptosis
298 pathway inhibition had no effects on LPS-induced cytokines elevation both in BMDM and BMDC
299 . Taken together, RIPK1 inhibitor may have inapparent effects on immunocytes
300 function.
301 3.3. RIPK1 inhibitor restrained the immunocytes infiltration by maintaining intestinal barrier
302 homeostasis and chemotaxis process in colitis
303 According to the above data, we can infer that RIPK1 inhibitor may play a protective role in
304 the early stages of disease progression. Before the systemic immune response, there will be many
305 immunocytes infiltration in the lamina propria of the colon. In immunohistochemical results, the
306 infiltration of CD3 positive cells increased significantly in the vehicle group, and RIPK1 inhibitor
307 could relieve this symptom . As shown in 4B, the decreased Treg
308 (CD25+Foxp3+) and increased Th17 (IL-17+) in the lamina propria could be restored by this
309 inhibitor. Myeloperoxidase (MPO), which is most abundantly expressed in neutrophil
310 granulocytes, showed an enhanced activity in colon tissue of colitis mice. This indicated that
311 RIPK1i treatment may reduce the massive neutrophil infiltration in colon.
312 Correspondingly, CD11b expression was detected in situ, which can reflect the infiltration of
313 myeloid immune cells. The result of myeloid immune cells infiltration was consistent with the
314 above . To confirm this event, we prepared a single cell suspension isolated from
315 colon lamina propria to determinate the percentage of infiltrated cells by flow cytometry. A large
316 number of accumulating neutrophils (CD11b+Ly6G+) and macrophages (CD11b+F4/80+) was
317 lowered in lamina propria after RIPK1 inhibitor treatment compared with the vehicle
318 In view of the epithelium barrier as a defense system against pathogens, RIPK1 inhibitor may
319 have a positive role in barrier function and then reduce the infiltration of immunocytes in colon.

320 We next used the TUNEL assay to visually detect intestinal epithelium injury in situ. Inhibition of
321 RIPK1 obviously lessened epithelium injury in colitis . The expression of tight
322 junction proteins, including E-cadherin, Zo-1 and Occludin was restored in RIPK1i-treatment
323 group , indicating the protective effects of RIPK1 inhibition in maintaining barrier
324 integrity. Intestinal permeability was assessed by oral gavage of FITC–dextran. Both in serum
325 detection and in vivo imaging, we all found that the absorption of fluorescence was lower in
326 RIPK1i-treatment group compared with the vehicle , which demonstrated that RIPK1
327 inhibitor preserved the intact intestinal barrier.
328 IECs-immunocytes interactions, as well as inflammatory mediators release during this
329 process, are critical pathogenic factors in colitis. Therefore, we paid attention to the role of RIPK1
330 inhibitor in this pathological process. In colitis model, results suggested that the expression of
331 chemokines and its receptors in colon was RIPK1 dependent. This result has also been observed in
332 the level of ICAM1 . This finding, while preliminary, suggests that RIPK1
333 inhibitor protects the structure and function of the intestinal epithelial barrier and reduces the
334 secretion of chemokines and adhesion molecules from IECs, then lowering the infiltration of
335 immunocytes.
336 3.4. RIPK1 inhibitor maintained IECs homeostasis by alleviating cell death, MLCK-related
337 tight junctions’ injury and accompanying oxidative stress
338 Firstly, epithelial cells viability is essential for the integrity of intestinal barrier. The results
339 suggested that TSZ (TNF+SM164+ZVAD) treatment did induce evident HT-29 cell death, and
340 RIPK1 inhibitor restored this injury . As shown in e 5B, cell death did not occur
341 obviously under TNF stimulation alone. If RIPK1 inhibitor has a protective mechanism in IECs
342 besides of cell death? To answer this question, we explored whether RIPK1 inhibitor can protect
343 the injury of tight junctions in IECs, which as an indispensable factor in maintaining epithelium
344 integrity. The results suggested that RIPK1 inhibition ameliorated TNFα-triggered destruction of
345 E-cadherin, Zo-1 and Occludin . To further verify the effect of RIPK1 inhibitor on
346 functional profile of IECs, we detected the permeability of colonic epithelial cell, Caco-2. The
347 FITC-Dextran flux through monolayers was elevated with TNFα incubation and RIPK1 inhibition
348 alleviated this high permeability . The epithelial monolayers have a certain value of

349 trans epithelial electrical resistance (TEER). When barrier structural disturbances, the drop of
350 TEER value could be tested. RIPK1 inhibitor treatment rescued this compromise of TEER
351 triggered by TNFα . Some researches showed that MLCK activation resulted in
352 dysregulation of tight junctions, barrier loss and induction of colitis[12, 30]. We found that RIPK1
353 was related to MLC phosphorylation in HT-29 cells . RIPK1 inhibitor may play a
354 protective effect in IECs by suppressing the activation of MLCK. Moreover, RIPK1 inhibition
355 could reduce the ROS release of IECs, and it is RIPK3 dependent
356 3.5. RIPK1 inhibitor could influence the interaction between IECs and immunocytes by
357 suppressing the production of chemokines and adhesion molecules of epithelium
358 Based on the RIPK1 inhibition could reduce the immunocytes infiltration in intestinal lamina
359 propria, we would like to explore IECs-immunocytes crosstalk in vitro. The results showed that
360 RIPK1 inhibitor had evident effects on down-regulation of chemokines and ICAM1 genes
361 expression of HT-29 cells . To test the protein expression level, we detected the IL-8
362 and CXCL10 secretion by ELISA. As shown in , RIPK1 inhibition suppressed the high
363 level of chemokines in response to TNFα. The expression of ICAM-1 also was inhibited
364 significantly at different time points during this process . Functionally, enhanced
365 adhesive ability of HT-29 cells was observed under the stimulation of TNFα and RIPK1 inhibition
366 intervened this event . In accordance with the decreased chemokines level after RIPK1
367 inhibition, the chemotaxis behavior of THP-1 and Jurkat cells to conditional media from HT-29
368 cells was restrained by RIPK1 inhibitor treatment . We thought that RIPK1
369 inhibition suppressed the chemotaxis and adhesion processes of immunocytes towards IECs.
370 RIPK1 may participate in the interaction between IECs and immunocytes, which possibly
371 becomes one of the underlying mechanisms of RIPK1 mediating barrier injury-related
372 inflammation.
373 3.6. The suppressive effect of RIPK1 inhibitor on cytokines of IECs was related to necroptosis

374 and NF-κB pathway

375 We would be interested to know what mechanisms involved in the suppressive function of
376 RIPK1 inhibitor on the production of chemokines and ICAM1 from IECs. Firstly, we found that
377 RIPK1 inhibitor could reduce the release of cytokines both under TNF and TSZ stimulation,

378 which represented necroptosis and non-cell death conditions, respectively . For
379 necroptosis process, RIPK1 inhibitor prevented the activation of necrosome (p-RIPK3 and
380 p-MLKL). And then the production of pro-inflammatory factors was inhibited . For
381 non-cell death condition, did RIPK3 and MLKL still act effect? As shown in 7B, only
382 RIPK1 and NF-κB inhibition decreased the chemokines release. It can therefore be assumed that
383 RIPK1 inhibitor plays a suppressive role via NF-κB pathway in the absent of cell death. We next
384 detected the relative protein expression in this process. The result suggested that the degradation
385 of IκBα was blocked by RIPK1 inhibitor and it did not involve in the activation of necrosome
386 . We compared the P65 nuclear translocation between TNF and TSZ treatment in IECs.
387 An early transient intense nuclear import of p65 was observed at 15 and 30 min after TNF
388 stimulation, and a delayed weak translocation of it occurred at 12h to 24h . Meanwhile,
389 cells began to experience necroptosis process at 3h, and only a brief weak nuclear accumulation
390 was found in cells stimulated by TSZ . RIPK1 inhibitor could reduce the P65 nuclear
391 translocation mainly occurring in TNF induced IECs injury . This observation
392 may support the conclusion that the suppressive effects of RIPK1 inhibitor on chemokines and
393 adhesion molecules are necroptosis pathway and NF-κB pathway dependent in cell death and
394 non-cell death conditions, respectively.
395 3.7. RIPK1 inhibitor also ameliorated DSS-induced colitis in acute phase
396 In this study, we observed the therapeutic effect of RIPK1 inhibitor on colitis in the recovery
397 phase. To test it in the acute phase, we designed another independent experimental scheme
398 8A). The results suggested that RIPK1 inhibition ameliorated the progression of the disease: the
399 colon shortening and pathological score were restored. The immunocytes infiltration was reduced
400 in lamina propria and immune organ activation also was suppressed ( 8B-F). Therefore, the
401 RIPK1 inhibitor has a protective effect both on acute and recovery phase of colitis via the
402 aforesaid mechanism.
403 3.8. RIPK1 inhibitor alleviated the vicious circle which triggered IECs-immunocytes crosstalk
404 by targeting necroptosis and NF-κB pathway
405 We concluded that RIPK1 inhibitor could suppress inflammation in colitis by maintaining the
406 homeostasis of intestinal barrier, which mainly involves in keeping barrier intact and reducing

407 immunocytes infiltration in tissue sites. For one side, RIPK1 inhibitor plays a protective role in
408 IECs via cell death inhibition. For another, it could ameliorate tight junctions’ injury and oxidative
409 stress. Subsequently, damaged IECs could release chemokines and adhesion molecules to recruit
410 immunocytes. RIPK1 inhibitor reduces the production of these molecules from IECs by
411 weakening NF-κB activation. Meanwhile, the DAMPs from IECs initiates immune responses.
412 Subsequently, cytokines released from activated immune cells result in progressive injury of IECs.

417 4. Discussion
418 Several researches directed at discovering the proinflammatory function of RIPK1 in
419 tissue-injury diseases which closely related to barrier dysfunction, for instance intestinal
420 inflammation, skin inflammation and renal injury diseases [5, 31, 32]. In this study, an animal
421 model of colitis, which is featured with intestinal mucosal barrier injury, was built to evaluate the
422 effect of RIPK1 inhibitor on intestinal mucosal barrier injury and find the possible mechanisms in
423 crosstalk between epithelium and immune microenvironments. To consider the adverse events of
424 RIPK1 inhibitor and the optimum dose in terms of body weight, we performed a dose dependent
425 study and added a group that normal mice treatment with this inhibitor. The results indicated that
426 the inhibitor had no obvious adverse effects in normal mice , and we chose 20mg/kg
427 as the optimum dose for further study (data not shown).
428 The results suggested that RIPK1 inhibitor could ameliorate DSS-induced colitis and
429 suppress the proinflammatory cytokines production ( 1). It is worth mentioning that the
430 systematic immune response was weakened under RIPK1i treatment ( 2). Nevertheless,
431 very little was found on the question that whether RIPK1 inhibitor has direct suppressive effects
432 on immunocytes. To determine it, we simply selected some classic immunocytes, such as spleen
433 lymphocytes, BMDM and BMDC, to observing whether RIPK1 inhibitor interfered the immune
434 response. As shown in 3, under antigen stimulation, the immune response was not
435 suppressed by RIPK1 inhibitor. It can therefore be assumed that there were no obvious effects of
436 RIPK1 inhibitor on immunocytes. We would like to focus on the IECs and following events about
437 IECs-immunocytes crosstalk.
438 Since the immunocytes infiltration in the lamina propria was reduced by RIPK1i inhibitor
439 administration (4A-E), it may play a protective role in the early stages of the disease
440 development. For mucosal barrier integrity, IECs are indispensable in maintaining barrier
441 homeostasis. Damaged IECs contribute to barrier disruption, allowing pathogens and
442 environmental microbes to invade the tissue. The results verified the protective role of RIPK1
443 inhibitor in TSZ induced necroptosis of IECs (5A). On account of the barrier integrity not
444 only depends on IECs but also on the capacity to keep the barrier sealed, it will be necessary to
445 determine whether RIPK1 regulates barrier-associated cell behaviors and characteristics such as

446 adhesion[3]. Tight junctions are important components ensuring the integrity and function of the
447 gut epithelial barrier by tightly sealing the intercellular junctions[33]. Our study adds to the
448 evidence that RIPK1 inhibitor could ameliorate the tight junctions’ injury (5C). Intestinal
449 permeability is an important feature of barrier function[34]. And TEER values indicate changes in
450 the cellular monolayer integrity. RIPK1 inhibition displayed positive effects on these indicators
451 ( 5D and E), which was supported by in vivo study in DSS-induced colitis model. RIPK1
452 inhibitor treatment alleviated barrier destruction during colitis (4F-H). Hence, it could
453 conceivably be assumed that the RIPK1 inhibition ensures intestinal barrier homeostasis by
454 reducing epithelial monolayers disruption and maintaining epithelial permeability. As for the
455 mechanism of RIPK1 inhibitor has a protective role in epithelial barrier injury, we prefer to
456 explore MLCK recruitment and oxidative stress in the following study. Noteworthily, we found
457 that RIPK1 inhibitor could regulate the level of total and cleaved caspase 8 both in colitis and
458 necrotic IECs (data not shown). According to the positive role of caspase 8 in maintaining barrier
459 homeostasis, it is valuable to focus on the mutual regulation of RIPK1 and caspase 8. It may bring
460 enlightenment to the study of cell death-associated diseases. [35].
461 Immunocytes adhesion and migration are critical steps when trafficking in intestine under
462 inflammation status[36]. We consider that RIPK1 plays a key role in the crosstalk between IECs
463 and immunocytes, which may be the pathogenic mechanism of colitis. ICAM-1, a key intercellular
464 adhesion molecule, was reported that it contributed to lymphocytes adhesion and migration across
465 endothelial cells monolayers[37]. There was a large increase of ICAM-1 during TNF stimulation
466 in HT-29 cells, and RIPK1 inhibitor had steady inhibitory effects on the expression level of it
467 ( 6A and C). Accordingly, the adhesion ability of IECs and its sticking ability to immune
468 cells were also restrained by RIPK1 inhibitor ( 6D). Chemokines released by IECs are an
469 initial pathological process for immune cells migration to the focus of inflammation, and it could
470 amplify the destructive immune-inflammatory response in colitis[38]. In the same way, RIPK1
471 inhibition decreased the chemokines production from damaged cells and the chemotaxis process
472 of immunocytes to epithelial cells was also intervened ( 6E and F). In vivo, the level of
473 ICAM1 and chemokines was increased in colitis model, consistently, the infiltration of
474 immunocytes in intestinal lamina propria was enhanced. According with in vitro data, RIPK1

475 inhibitor treatment ameliorated these pathological changes ( 4I and J). The results
476 demonstrate that RIPK1 has significant effects on epithelial cells to release adhesion molecules
477 and chemokines. So that the following adhesion and migration behaviors of immunocytes and the
478 cascade of epithelial cells disruption events towards immuno-inflammatory responses can be
479 regulated. Our study provides well-reason basis that RIPK1 takes part in the interaction between
480 IECs and immunocytes by promoting adhesion and migration, which should be attached
481 importance to the pathogenesis of injury-associated inflammatory diseases. This mechanism might
482 be a noteworthy point in RIPK1-mediated inflammatory response.
483 To investigate the mechanism of RIPK1-mediated the release of chemokines and adhesion
484 molecules from IECs, we tested which pathway is associated with this effect. Firstly, we found
485 that both under TSZ and TNF stimulation, the cytokines released from IECs could be suppressed
486 by RIPK1 inhibitor ( 7A). This suggests that RIPK1 inhibitor plays a role in
487 anti-inflammation response both in the presence and absence of cell death. In the cell death
488 process, RIPK1 inhibitor restrained the activation of necrosome, which is an important factor of
489 the inflammatory mediators’ release ( 7C). Under non-cell death status, RIPK1 inhibitor
490 acted to decrease these cytokines mainly via NF-κB inhibition ( 7D and E). This observation
491 may support the hypothesis that the ubiquitination and phosphorylation of RIPK1 have mutual
492 effects and this inhibitor plays an indispensable role during these processes. Further research
493 should be undertaken to investigate the involvement of deep molecular mechanisms.
494 Taken together, our findings demonstrate pharmacological mechanisms of RIPK1 inhibitor
495 for treatment colitis. We consider that RIPK1 inhibitor could maintain the homeostasis of barrier
496 and regulate the crosstalk between IECs and lymphocytes, which are not limited to its role in
497 protecting cell death. As a potential anti-inflammation target, RIPK1 has tricky effects in human
498 disease. Therefore, the pharmacological mechanisms of RIPK1 inhibitor in inflammatory diseases
499 urgently need to be clarified. We hope our study could be conducive to a clear understanding of
500 this target and provide new ideas for the treatment of colitis.

501 Correspondence
502 Wei Tang and Jianping Zuo, Shanghai Institute of Materia Medica, Chinese Academy of
503 Sciences, Shanghai 201203, China. E-mail address: [email protected] and [email protected]

504 Conflict of interest

505 The authors declare that they have no competing interests.

506 Acknowledgements
507 This work was funded by the "Personalized Medicines—Molecular Signature-based Drug
508 Discovery and Development", Strategic Priority Research Program of the Chinese Academy of

520 [1] C. Gunther, H. Neumann, M.F. Neurath, C. Becker, Apoptosis, necrosis and necroptosis:

521 cell death regulation in the intestinal epithelium, Gut 62(7) (2013) 1062-71.

522 [2] A. Negroni, E. Colantoni, M. Pierdomenico, F. Palone, M. Costanzo, S. Oliva, A. Tiberti, S.

523 Cucchiara, L. Stronati, RIP3 AND pMLKL promote necroptosis-induced inflammation and alter

524 membrane permeability in intestinal epithelial cells, Digestive and liver disease : official

525 journal of the Italian Society of Gastroenterology and the Italian Association for the Study of

526 the Liver 49(11) (2017) 1201-1210.

527 [3] M. Pasparakis, P. Vandenabeele, Necroptosis and its role in inflammation, Nature

528 517(7534) (2015) 311-20.

529 [4] K. Newton, K.E. Wickliffe, A. Maltzman, D.L. Dugger, A. Strasser, V.C. Pham, J.R. Lill, M.

530 Roose-Girma, S. Warming, M. Solon, H. Ngu, J.D. Webster, V.M. Dixit, RIPK1 inhibits

531 ZBP1-driven necroptosis during development, Nature 540(7631) (2016) 129-133.

532 [5] M. Dannappel, K. Vlantis, S. Kumari, A. Polykratis, C. Kim, L. Wachsmuth, C. Eftychi, J.

533 Lin, T. Corona, N. Hermance, M. Zelic, P. Kirsch, M. Basic, A. Bleich, M. Kelliher, M.

534 Pasparakis, RIPK1 maintains epithelial homeostasis by inhibiting apoptosis and necroptosis,

535 Nature 513(7516) (2014) 90-4.

536 [6] N. Takahashi, L. Vereecke, M.J. Bertrand, L. Duprez, S.B. Berger, T. Divert, A. Goncalves,

537 M. Sze, B. Gilbert, S. Kourula, V. Goossens, S. Lefebvre, C. Gunther, C. Becker, J. Bertin,

538 P.J. Gough, W. Declercq, G. van Loo, P. Vandenabeele, RIPK1 ensures intestinal

539 homeostasis by protecting the epithelium against apoptosis, Nature 513(7516) (2014) 95-9.

540 [7] P.A. Harris, S.B. Berger, J.U. Jeong, R. Nagilla, D. Bandyopadhyay, N. Campobasso, C.A.

541 Capriotti, J.A. Cox, L. Dare, X. Dong, P.M. Eidam, J.N. Finger, S.J. Hoffman, J. Kang, V.

542 Kasparcova, B.W. King, R. Lehr, Y. Lan, L.K. Leister, J.D. Lich, T.T. MacDonald, N.A. Miller,

543 M.T. Ouellette, C.S. Pao, A. Rahman, M.A. Reilly, A.R. Rendina, E.J. Rivera, M.C. Schaeffer,

544 C.A. Sehon, R.R. Singhaus, H.H. Sun, B.A. Swift, R.D. Totoritis, A. Vossenkamper, P. Ward,

545 D.D. Wisnoski, D. Zhang, R.W. Marquis, P.J. Gough, J. Bertin, Discovery of a First-in-Class

546 Receptor Interacting Protein 1 (RIP1) Kinase Specific Clinical Candidate (GSK2982772) for

547 the Treatment of Inflammatory Diseases, Journal of medicinal chemistry 60(4) (2017)

548 1247-1261.

549 [8] L.C. Tong, Y. Wang, Z.B. Wang, W.Y. Liu, S. Sun, L. Li, D.F. Su, L.C. Zhang, Propionate

550 Ameliorates Dextran Sodium Sulfate-Induced Colitis by Improving Intestinal Barrier Function

551 and Reducing Inflammation and Oxidative Stress, Frontiers in pharmacology 7 (2016) 253.

552 [9] A.C. Luissint, C.A. Parkos, A. Nusrat, Inflammation and the Intestinal Barrier:

553 Leukocyte-Epithelial Cell Interactions, Cell Junction Remodeling, and Mucosal Repair,

554 Gastroenterology 151(4) (2016) 616-32.

555 [10] F. Sanchez de Medina, I. Romero-Calvo, C. Mascaraque, O. Martinez-Augustin,

556 Intestinal inflammation and mucosal barrier function, Inflammatory bowel diseases 20(12)

557 (2014) 2394-404.

558 [11] A. Negroni, S. Cucchiara, L. Stronati, Apoptosis, Necrosis, and Necroptosis in the Gut

559 and Intestinal Homeostasis, Mediators of inflammation 2015 (2015) 250762.

560 [12] W.V. Graham, W. He, A.M. Marchiando, J. Zha, G. Singh, H.S. Li, A. Biswas, M. Ong,

561 Z.H. Jiang, W. Choi, H. Zuccola, Y. Wang, J. Griffith, J. Wu, H.J. Rosenberg, Y. Wang, S.B.

562 Snapper, D. Ostrov, S.C. Meredith, L.W. Miller, J.R. Turner, Intracellular MLCK1 diversion

563 reverses barrier loss to restore mucosal homeostasis, Nature medicine 25(4) (2019) 690-700.

564 [13] G. Aviello, U.G. Knaus, ROS in gastrointestinal inflammation: Rescue Or Sabotage?,

565 British journal of pharmacology 174(12) (2017) 1704-1718.

566 [14] R. Lyck, G. Enzmann, The physiological roles of ICAM-1 and ICAM-2 in neutrophil

567 migration into tissues, Current opinion in hematology 22(1) (2015) 53-9.

568 [15] A. Kaser, H. Tilg, Novel therapeutic targets in the treatment of IBD, Expert opinion on

569 therapeutic targets 12(5) (2008) 553-63.

570 [16] H. Wakita, T. Yanagawa, Y. Kuboi, T. Imai, E6130, a Novel CX3C Chemokine Receptor 1

571 (CX3CR1) Modulator, Attenuates Mucosal Inflammation and Reduces CX3CR1(+) Leukocyte

572 Trafficking in Mice with Colitis, Molecular pharmacology 92(5) (2017) 502-509.

573 [17] M. Eberhardson, P. Marits, M. Jones, P. Jones, P. Karlen, M. Karlsson, G. Cotton, K.

574 Woznica, B. Maltman, H. Glise, O. Winqvist, Treatment of inflammatory bowel disease by

575 chemokine receptor-targeted leukapheresis, Clinical immunology (Orlando, Fla.) 149(1) (2013)

576 73-82.

577 [18] I. Marafini, I. Monteleone, V. Dinallo, D. Di Fusco, V. De Simone, F. Laudisi, M.C. Fantini,

578 A. Di Sabatino, F. Pallone, G. Monteleone, CCL20 Is Negatively Regulated by TGF-beta1 in

579 Intestinal Epithelial Cells and Reduced in Crohn's Disease Patients With a Successful

580 Response to Mongersen, a Smad7 Antisense Oligonucleotide, Journal of Crohn's & colitis

581 11(5) (2017) 603-609.

582 [19] A. Egesten, M. Eliasson, A.I. Olin, J.S. Erjefalt, A. Bjartell, P. Sangfelt, M. Carlson, The

583 proinflammatory CXC-chemokines GRO-alpha/CXCL1 and MIG/CXCL9 are concomitantly

584 expressed in ulcerative colitis and decrease during treatment with topical corticosteroids,

585 International journal of colorectal disease 22(12) (2007) 1421-7.

586 [20] L. Mayer, W.J. Sandborn, Y. Stepanov, K. Geboes, R. Hardi, M. Yellin, X. Tao, L.A. Xu, L.

587 Salter-Cid, S. Gujrathi, R. Aranda, A.Y. Luo, Anti-IP-10 antibody (BMS-936557) for ulcerative

588 colitis: a phase II randomised study, Gut 63(3) (2014) 442-50.

589 [21] Y.X. Yan, M.J. Shao, Q. Qi, Y.S. Xu, X.Q. Yang, F.H. Zhu, S.J. He, P.L. He, C.L. Feng,

590 Y.W. Wu, H. Li, W. Tang, J.P. Zuo, Artemisinin analogue SM934 ameliorates DSS-induced

591 mouse ulcerative colitis via suppressing neutrophils and macrophages, Acta pharmacologica

592 Sinica 39(10) (2018) 1633-1644.

593 [22] L. Santucci, S. Fiorucci, N. Rubinstein, A. Mencarelli, B. Palazzetti, B. Federici, G.A.

594 Rabinovich, A. Morelli, Galectin-1 suppresses experimental colitis in mice, Gastroenterology

595 124(5) (2003) 1381-94.

596 [23] J.J. Kim, M.S. Shajib, M.M. Manocha, W.I. Khan, Investigating intestinal inflammation in

597 DSS-induced model of IBD, Journal of visualized experiments : JoVE (60) (2012).

598 [24] T. Murakami, E.A. Felinski, D.A. Antonetti, Occludin phosphorylation and ubiquitination

599 regulate tight junction trafficking and vascular endothelial growth factor-induced permeability,

600 The Journal of biological chemistry 284(31) (2009) 21036-46.

601 [25] F. Heller, P. Florian, C. Bojarski, J. Richter, M. Christ, B. Hillenbrand, J. Mankertz, A.H. 602 Gitter, N. Burgel, M. Fromm, M. Zeitz, I. Fuss, W. Strober, J.D. Schulzke, Interleukin-13 is the 603 key effector Th2 cytokine in ulcerative colitis that affects epithelial tight junctions, apoptosis, 604 and cell restitution, Gastroenterology 129(2) (2005) 550-64.
605 [26] R.J. Roth Flach, A. Skoura, A. Matevossian, L.V. Danai, W. Zheng, C. Cortes, S.K.

606 Bhattacharya, M. Aouadi, N. Hagan, J.C. Yawe, P. Vangala, L.G. Menendez, M.P. Cooper,

607 T.P. Fitzgibbons, L. Buckbinder, M.P. Czech, Endothelial protein kinase MAP4K4 promotes

608 vascular inflammation and atherosclerosis, Nature communications 6 (2015) 8995.

609 [27] J. Shao, I. Stout, O.L. Volger, P.J. Hendriksen, H. van Loveren, A.A. Peijnenburg, 610 Inhibition of CXCL12-mediated chemotaxis of Jurkat cells by direct immunotoxicants, 611 Archives of toxicology 90(7) (2016) 1685-94.
612 [28] M. Nanno, Y. Kanari, T. Naito, N. Inoue, T. Hisamatsu, H. Chinen, K. Sugimoto, Y. 613 Shimomura, H. Yamagishi, T. Shiohara, S. Ueha, K. Matsushima, M. Suematsu, A. Mizoguchi, 614 T. Hibi, A.K. Bhan, H. Ishikawa, Exacerbating role of gammadelta T cells in chronic colitis of 615 T-cell receptor alpha mutant mice, Gastroenterology 134(2) (2008) 481-90.
616 [29] K. Cheung, G. Lu, R. Sharma, A. Vincek, R. Zhang, A.N. Plotnikov, F. Zhang, Q. Zhang, 617 Y. Ju, Y. Hu, L. Zhao, X. Han, J. Meslamani, F. Xu, A. Jaganathan, T. Shen, H. Zhu, E. 618 Rusinova, L. Zeng, J. Zhou, J. Yang, L. Peng, M. Ohlmeyer, M.J. Walsh, D.Y. Zhang, H. 619 Xiong, M.M. Zhou, BET N-terminal bromodomain inhibition selectively blocks Th17 cell 620 differentiation and ameliorates colitis in mice, Proceedings of the National Academy of 621 Sciences of the United States of America 114(11) (2017) 2952-2957.
622 [30] L. Su, S.C. Nalle, L. Shen, E.S. Turner, G. Singh, L.A. Breskin, E.A. Khramtsova, G. 623 Khramtsova, P.Y. Tsai, Y.X. Fu, C. Abraham, J.R. Turner, TNFR2 activates MLCK-dependent 624 tight junction dysregulation to cause apoptosis-mediated barrier loss and experimental colitis, 625 Gastroenterology 145(2) (2013) 407-15.
626 [31] P.S. Welz, A. Wullaert, K. Vlantis, V. Kondylis, V. Fernandez-Majada, M. Ermolaeva, P. 627 Kirsch, A. Sterner-Kock, G. van Loo, M. Pasparakis, FADD prevents RIP3-mediated epithelial 628 cell necrosis and chronic intestinal inflammation, Nature 477(7364) (2011) 330-4.
629 [32] Y. Xu, H. Ma, J. Shao, J. Wu, L. Zhou, Z. Zhang, Y. Wang, Z. Huang, J. Ren, S. Liu, X.

630 Chen, J. Han, A Role for Tubular Necroptosis in Cisplatin-Induced AKI, Journal of the

631 American Society of Nephrology : JASN 26(11) (2015) 2647-58.

632 [33] T. Oshima, H. Miwa, Gastrointestinal mucosal barrier function and diseases, Journal of

633 gastroenterology 51(8) (2016) 768-78.

634 [34] S.C. Bischoff, G. Barbara, W. Buurman, T. Ockhuizen, J.D. Schulzke, M. Serino, H. Tilg, 635 A. Watson, J.M. Wells, Intestinal permeability--a new target for disease prevention and 636 therapy, BMC gastroenterology 14 (2014) 189.
637 [35] K. Newton, K.E. Wickliffe, D.L. Dugger, A. Maltzman, M. Roose-Girma, M. Dohse, L. 638 Komuves, J.D. Webster, V.M. Dixit, Cleavage of RIPK1 by caspase-8 is crucial for limiting 639 apoptosis and necroptosis, Nature 574(7778) (2019) 428-431.
640 [36] M. Strazza, I. Azoulay-Alfaguter, M. Peled, A.V. Smrcka, E.Y. Skolnik, S. Srivastava, A. 641 Mor, PLCepsilon1 regulates SDF-1alpha-induced lymphocyte adhesion and migration to sites 642 of inflammation, Proceedings of the National Academy of Sciences of the United States of 643 America 114(10) (2017) 2693-2698.
644 [37] A. Teijeira, M.C. Hunter, E. Russo, S.T. Proulx, T. Frei, G.F. Debes, M. Coles, I. Melero, 645 M. Detmar, A. Rouzaut, C. Halin, T Cell Migration from Inflamed Skin to Draining Lymph 646 Nodes Requires Intralymphatic Crawling Supported by ICAM-1/LFA-1 Interactions, Cell 647 reports 18(4) (2017) 857-865.
648 [38] E. Franze, I. Marafini, V. De Simone, I. Monteleone, F. Caprioli, A. Colantoni, A. Ortenzi, 649 F. Crescenzi, R. Izzo, G. Sica, P. Sileri, P. Rossi, F. Pallone, G. Monteleone, Interleukin-34 650 Induces Cc-chemokine Ligand 20 in Gut Epithelial Cells, Journal of Crohn's & colitis 10(1)
651 (2016) 87-94.

653 legends
654 1. RIPK1 inhibitor ameliorated DSS-induced colitis and suppressed the
655 proinflammatory cytokines and DAMPs production.
656 (A) Experimental design of the DSS-induced colitis. 3% DSS was dissolved in sterile water for 6 657 days except normal control and normal+RIPK1i groups, then the mice were administered with 658 regular water for 3 days. Normal control (HPMC), Normal+RIPK1i (GSK2982772, 20 mg/kg), 659 Vehicle control (HPMC) and RIPK1i (GSK2982772, 20 mg/kg) were given by intragastric 660 administration from day 1 to 9, respectively. During the treatment, weight loss and disease activity 661 index (DAI) were monitored from day 0 to day 9. Ten mice of each group were sacrificed at day 662 10 and the colons, spleens and mesenteric lymph nodes were isolated for subsequent analysis.
663 (B) Body weight (left) and disease activity index (DAI) (right) of 3% DSS treated mice were
664 evaluated (n=10).
665 (C) The typical colon appearance (left) and the statistical analysis length of the colon (right)
666 (n=10).
667 (D) The level of serum ALP and ALT was tested (n=10).
668 (E) Representative microscopic pictures of H&E-stained colons are shown on the left (scale bars, 669 100 μm) and histopathological scores of colonic sections of each group are shown on the right, 670 (n=10).
671 (F) Western blot analysis of representative colon tissues of normal control, vehicle and RIPK1i
672 (20mg/kg) to test the phosphorylation level of RIPK1 (n=4).
673 (G) Pro-inflammatory cytokines of mice serum and colonic tissue were measured by quantitative
674 real-time PCR or ELISA (n=4).
675 (H) The expression of HMGB-1 and HSP90 in colonic tissue was determined through
676 immunohistochemistry (scale bars, 100 μm).
677 Data are presented as means ± SEM. * P < 0.05, ** P < 0.01.
678 2. RIPK1 inhibitor suppressed the immune response in colitis.
679 (A) The CD4+T cells activation marker CD25, CD69, CD278 were analyzed by flow cytometry in
680 spleen.
681 (B) The percentage of γδT cells in spleen was determined by flow cytometry.
682 (C) The subset of T cells, Th17 differentiation ratio of spleen. This cellular population was gated

683 from CD3+ and CD4+.
684 (D) The percentage of macrophage (CD11b+ F4/80+) and neutrophil (CD11b+ly6G+/ly6C+) in
685 spleen and mesenteric lymph nodes.
686 RIPK1 inhibitor has little direct suppressive effects on immunocytes in vitro.
687 (A) Cell counting kit (CCK8) analysis of lymphocytes viability isolated from spleen.
688 (B) Lymphocytes were stimulated with ConA or LPS in the presence or absence RIPK1 inhibitor
689 (GSK2982772). Analysis of cell specific proliferation under antigen treatment by H3-TdR
690 incorporation assay.
691 BMDM and BMDC were pre-incubated with RIPK1i (GSK2982772, 100nM) for 30min and then
692 stimulated with LPS (10μg/ml) for 24h.
693 (C) qPCR analysis of the pro-inflammatory factors in BMDM under LPS stimulation.
694 (D) qPCR analysis of the pro-inflammatory factors in BMDC under LPS stimulation.
695 4. RIPK1 inhibitor restrained the immunocytes infiltration by maintaining intestinal
696 barrier homeostasis and chemotaxis process in colitis
697 (A) The infiltration of CD3 positive lymphocytes in colon was reflected by immunohistochemistry
698 (scale bars, 100 μm).
699 (B) Flow cytometry analysis the percentage of Treg (CD25+Foxp3+) and Th17 (IL-17+) in
700 lamina propria infiltration. These cellular population were gated from CD3+ and CD4+.
701 (C) MPO activity in the colonic homogenates (n=5).
702 (D) The infiltration of CD11b positive monocytes in colon was determined through
703 immunofluorescence (scale bars, 100 μm).
704 (E) Flow cytometry analysis the percentage of macrophage (CD11b+F4/80+) and neutrophil
705 (CD11b+ly6G+) in lamina propria infiltration.
706 (F) Identification of intestinal epithelial cell injury by TUNEL assay (red) (scale bars, 100 μm). 707 (G) Western blot analysis of representative colon tissues of normal control, vehicle and RIPK1i 708 (20mg/kg) to test the expression level of tight junctions (n=4).
709 (H) Intestinal permeability assay in normal control, vehicle and RIPK1i treatment (20mg/kg) 710 using FITC-labelled dextran, fluorescence intensity in serum was test by microplate reader (n=3). 711 In vivo imaging system was used to reflect the intestinal absorption of fluorescence.

712 (I) Western blot analysis of representative colon tissues to test the expression level of adhesion
713 molecule ICAM1 (n=4).
714 (J) Chemokines and receptors of colonic tissue were measured by qPCR (n=4).
715 Data are presented as means ± SEM. * P < 0.05, ** P < 0.01.
716 5. RIPK1 inhibitor maintained IECs homeostasis by alleviating MLC
717 phosphorylation and accompanying oxidative stress.
718 HT-29 cells or Caco-2 cells (C-D) were pre-incubated with RIPK1i (50nM) or SM164 (50nM) and 719 Z-VAD (25μM) for 30 mins and then culture for different times in the presence or absence of 720 hTNF (100ng/ml).
721 (A) The cell viability was determined by CellTiterGlo after indicated times of TSZ treatment. And 722 TUNEL assay was used to reflect the death of HT-29 cells, damaged of intestinal epithelial cells 723 were stained with red (scale bars, 100 μm).
724 (B) The cell viability was determined by CellTiterGlo after indicated times of TNF treatment.
725 (C) Confocal laser-scanning microscope images of tight junctions E-cadherin, ZO-1 and occludin
726 were measured by immunofluorescence staining (scale bars, 100 μm).
727 (D-E) After TNF administration for 24h, permeability of 4-KDa FITC-dextran (D) and (E)
728 transepithelial electrical resistance (TEER) of Caco-2 cells were determined.
729 (F) After HT-29 cells were stimulated with TNF, cells were harvested respectively at different
730 times for western blot analysis to test the expression of p-MLC.
731 (G) Reactive oxygen species (ROS) production of HT-29 cells cultured with above system was 732 detected by flowcytometry. N-acetyl-cysteine (NAC) is a ROS scavenger, which was used as a 733 positive control.
734 Data are presented as means ± SEM. * P < 0.05, ** P < 0.01.
735 6. RIPK1 inhibitor could influence the interaction between IECs and immunocytes
736 by suppressing the production of chemokines and adhesion molecules of epithelium. 737 HT-29 cells were pre-incubated with RIPK1i (50nM) for 30 mins and then culture for 24 h in 738 presence or absence of hTNF (100ng/ml) (A-F).
739 (A) qPCR analysis of chemokines on RNA extracted from HT-29 cells.
740 (B) The secretion level of IL-8 and CXCL10 was measured via ELISA.

741 (C) Western blot analysis of adhesion molecule (ICAM-1) in HT-29 cells at different time points
742 after hTNF (50 or 100ng/ml) administration.
743 (D) Representative images of THP-1 cells adhesion to the HT-29 monolayer which treated with
744 the above culture system (scale bars, 100 μm).
745 (E) Representative images of THP-1 cells and Jurkat cells chemotaxis to the conditional 746 supernatant which collected from HT-29 cells treated with the above culture system (scale bars, 747 100 μm).
748 (F) The number of THP-1 and Jurkat cells that were chemotactic to lower chamber was counted
749 by cytometry.
750 Data are presented as means ± SEM of three independent experiments. * P < 0.05, ** P < 0.01. 751 7. The suppressive effect of RIPK1 inhibitor on cytokines of IECs was related to 752 necroptosis and NF-κB pathway.
753 HT-29 cells were pre-incubated with RIPK1i (50nM) or SM164 (50nM) and Z-VAD (25μM) for
754 30 mins and then culture for different times in presence or absence of hTNF (100ng/ml).
755 (A) qPCR analysis of cytokines on RNA extracted from HT-29 cells at different times after TNF
756 or TSZ treatment.
757 (B) Under necroptosis and NF-κB pathway inhibitors administration (RIPK3i: GSK872 1μM; 758 MLKLi: NSA 1μM; NF-κBi: TPCA-1 1μM), the genes expression of chemokines and adhesion 759 molecule was determined by qPCR with TNF treatment for 24h.
760 (C) Western blot analysis of the protein expression of NF-κB and necroptosis pathway in HT-29
761 cells at different time points after TSZ administration.
762 (D) Western blot analysis of the protein expression of NF-κB and necroptosis pathway in HT-29
763 cells at different time points after TNF administration.
764 (E) Immunostaining of p65 in HT-29 cells treated with TNF in the presence or absence RIPK1i
765 for the indicated time periods (scale bars, 25 μm).
766 (F) Immunostaining of p65 in HT-29 cells treated with TSZ in the presence or absence RIPK1i
767 for the indicated time periods (scale bars, 25 μm).
768 8. RIPK1 inhibitor also ameliorated DSS-induced colitis in acute phase.
769 (A) Experimental design of the DSS-induced colitis. DSS was dissolved in sterile water for 6 days

770 except normal control. Vehicle control (HPMC), RIPK1i (20 mg/kg), were given by intragastric 771 administration from day 1 to 6, respectively. Eight mice of each group were sacrificed at day 6 and 772 the colons, spleens and mesenteric lymph nodes were isolated for subsequent analysis.
773 (B) Body weight (above) and disease activity index (DAI) (below) of 3% DSS treated mice were
774 evaluated (n=6).
775 (C) The typical colon appearance (left) and the statistical analysis length of colon (left) (n=6). 776 (D) Representative microscopic pictures of H&E-stained colons are shown on the left (scale bars, 777 100 μm) and histopathological scores of colonic sections of each group are shown on the right 778 (n=6).
779 (E) The infiltration of γδT cells and monocytes (CD11b+) in lamina propria and its subset 780 percentage of macrophage (CD11b+F4/80+) and neutrophil (CD11b+ly6C+) was analysis by flow 781 cytometry.
782 (F) The CD4+T cells activation markers CD25, CD69, CD278 in spleen and the percentage of 783 macrophage (CD11b+F4/80+) and neutrophil (CD11b+ly6C+) in spleen and mesenteric lymph 784 nodes were analyzed by flow cytometry.
785 (G) Data are presented as means ± SEM. * P < 0.05, ** P < 0.01.
786 9. RIPK1 inhibitor alleviated the vicious circle which trigger IECs-immunocytes
787 crosstalk by targeting necroptosis and NF-κB pathway
788 RIPK1 inhibition ameliorates colitis at the early stage during disease progress via protecting
789 epithelial barrier injury and regulating IECs-immunocytes crosstalk.
790 Author Contributions
791 Huimin Lu performed the in vitro and in vivo experiments, interpreted the data, and wrote the 792 manuscript. Heng Li and Yuxi Yan performed the in vitro experiments and reviewed the 793 manuscript. GSK2982772 Chen Fan, Qing Qi, Yanwei Wu, Chunlan Feng, Bing Wu and Yuanzhuo Gao 794 performed the in vivo experiments and provided advice on experimental design. Jianping Zuo 795 provided advice on the experiments and reviewed the manuscript. Wei Tang conceived and 796 supervised the project, designed the experiments, and wrote the manuscript. All authors reviewed 797 the manuscript.