Spontaneous colitis in IL‐10‐deficient mice was ameliorated via inhibiting glutaminase1

Abstract Immunity imbalance and barrier damage in the intestinal mucosa are the main pathogenic factors of Crohn's disease (CD). Bis‐2‐(5‐phenylacetamido‐1,2,4‐thiadiazol‐2‐yl) ethyl sulfide (BPTES) is a glutaminase 1 (Gls1) inhibitor with the dual functions of increasing glutamine levels and immune regulation. In this study, we focused on the role of BPTES in CD‐like enteritis and the possible mechanisms. We found that Gls1 expression was significantly increased in CD intestinal tissue compared with control tissue. Bis‐2‐(5‐phenylacetamido‐1,2,4‐thiadiazol‐2‐yl) ethyl sulfide treatment significantly ameliorated chronic colitis in the IL‐10−/−, as manifested by decreased disease activity index, body weight change, histological inflammatory degree and inflammatory cytokine expression. Bis‐2‐(5‐phenylacetamido‐1,2,4‐thiadiazol‐2‐yl) ethyl sulfide treatment exerted protective effects on CD that were associated with the maintenance of intestinal barrier integrity and the Th/Treg balance. Bis‐2‐(5‐phenylacetamido‐1,2,4‐thiadiazol‐2‐yl) ethyl sulfide treatment may act in part through TCR‐mediated mammalian target of rapamycin complex 1 (mTORC1) signalling activation. In conclusion, inhibition of Gls1 expression attenuated chronic colitis by maintaining intestinal barrier integrity and the Th/Treg balance, thereby ameliorating CD‐like colitis.


| INTRODUC TI ON
Crohn's disease (CD) is a chronic condition that causes inflammation of the lining of the digestive system with increasing incidence worldwide. All segments of the CD gastrointestinal tract can be affected, causing fibrosis, intestinal obstruction and fistula. 1 However, the exact aetiology is poorly understood, which limits diagnostic and treatment improvements. The interaction between the intestinal barrier and the immune system may play an important role in the pathogenesis of CD. 2 Defects in intestinal epithelial barrier function and immune disorders are the characteristic features of CD. 3,4 Increased permeability of the intestinal barrier can increase the absorption of bacteria, toxins, etc, which generally cannot pass through the normal intestinal mucosa, and stimulate a series of antigen-specific immune responses and inflammatory changes. Consequently, compensatory immune reactions are excessively triggered a process that is believed to finally result in chronic intestinal inflammation. 5 Among the variety of inflammatory cells in the intestine, mucosal CD4 + lymphocytes are believed to play central roles in both the induction and persistence of chronic inflammation by producing pro-inflammatory cytokines. 2,6 Crohn's disease is mainly characterized by an enhanced Th1 response 7,8 together with an increased Th17 immune response, which plays a critical role in the pathogenesis of CD. 9 Local immune balance of intestinal tissue is maintained by regulatory T cells (Tregs) in the intestine that inhibit the proliferation and response of other helper-T cells (Th cells). 10 Studies have indicated that CD is associated with impaired Treg responses and enhanced Th1/Th17 responses. [11][12][13] Due to the crucial role of the Th/Treg balance in inducing and sustaining intestinal damage in CD, regulating the balance of immune responses in CD patients may be a new therapeutic strategy.
Previous studies have shown that glutamine (Gln) therapy improves the outcome of experimental colitis by maintaining intestinal barrier integrity and decreasing intestinal permeability through the enhancement of tight junctions in experimental colitis. [14][15][16] Glutaminase (Gls) is the first enzyme in the glutaminolysis pathway, and it converts Gln to glutamate. 17 In mammals, there are two different genes encoding Gls, Gls1 (the kidney isoform) and Gls2 (the liver isoform); and Gls1 has greater enzymatic activity than Gls2. 18  Here, we detected increased Gls1 levels in CD patients and IL-10deficient (IL-10 −/− ) mice. The inhibition of Gls1 by BPTES significantly increased Gln expression in IL-10 −/− mice and showed a protective effect in experimental colitis. Furthermore, this study investigated the effects of a Gls1 inhibitor on Th/Treg cell homeostasis, intestinal mucosal inflammation and intestinal barrier integrity in IL-10 −/− mice.

| Patient specimen preparation
The study was approved by the local ethics committee, and informed consent was obtained from the patients for the use of surgical specimens in this study. Intestinal specimens were collected from patients with CD (n = 13) who underwent intestinal resection, and uninjured bowel was collected from colon cancer patients (control, n = 17

| Mice
Both wild-type (WT) mice and IL-10 −/− mice (C57BL/6J background) were purchased from The Jackson Laboratory and were housed in a specific pathogen-free (SPF) environment. The IL-10 −/− mice consistently developed colitis at 15 weeks of age when maintained in the SPF environment as previously reported. 24 The animal experiments were conducted in compliance with the guidelines issued by the China Council for Animal Care and Utilization Committee of Bengbu Medical College (Bengbu, China).

| BPTES administration protocol and enteritis symptom assessment
The IL-10 −/− mice (15 weeks old, male) were divided into the control (IL-10 −/− , n = 10) and BPTES-treated groups (IL-10 −/− +BPTES, n = 10); all mice showed spontaneous enteritis. The WT mice (15 weeks old, male, n = 10) were used as negative controls. A total of 60 μg/mice BPTES (Gls1 inhibitor; Selleck Chemicals, USA) or dimethyl sulfoxide (DMSO) in PBS was intraperitoneally administered twice a week for 4 weeks as reported previously. 20 The IL-10 −/− mice were scored weekly using the inflammatory bowel disease activity index (DAI) and a numerical system, as reported previously. 24 In short, DAI was calculated by scoring 1 point for the appearance of each of the following features: ruffled fur; occult faecal blood; rectal prolapse <1 mm; and soft stool. An additional point was given for diarrhoea or severe rectal prolapse >1 mm.
Therefore, DAI was obtained on a 6-point (0-5) scale. Mouse body weight was monitored daily.

| Histological examinations
Histologic evaluations were performed on disease activity indexstained sections of the intestines fixed in 10% formalin solution, as previously described. 25 In brief, intestinal inflammation was scored on a scale of 0-4 according to inflammatory cell infiltration in the intestinal lamina propria and changes in the intestinal mucosal architecture. All histological scoring was conducted by two independent pathologists who were blinded to treatment group.

| Immunohistochemical analysis
Intestinal Gls1 levels were determined by immunohistochemical analysis as previously described. 26 Briefly, intestinal tissue sections were deparaffinized, rehydrated, subjected to antigen retrieval, blocked with normal goat serum for 30 minutes and incubated at 4°C overnight with rabbit polyclonal antibody against Gls1 (1:100; Abcam). Then, the samples were incubated for 60 minutes at RT with a 1:300 dilution of biotinylated goat anti-rabbit immunoglobin G (IgG) antibody (Beyotime, Haimen, China) in PBS, followed by development with DAB (Beyotime). Ten fields from each tissue section were randomly selected to quantify Gls1 expression.

| Measurement of Gln
The Gln concentration in the intestinal homogenate was determined using the Gln/glutamate determination kit (GLN-1; Sigma-Aldrich, USA) according to the manufacturer's instructions. Absorbance was read at 340 nm using a Tecan Infinite M200 plate reader (Tecan, Austria).

| Enzyme-linked immunosorbent assay
Interleukin-17A, IFN-γ and TNF-α expression levels in the intestine were determined by ELISA. Briefly, intestinal tissue was homogenized in 1 mL of normal saline with protease inhibitors (Sigma-Aldrich). Then, the homogenates were centrifuged at 1000 g at 4°C for 30 minutes, and the supernatant was stored at −80°C until analysis. Interleukin-17A, IFN-γ and TNF-α levels (pg/mg) were measured by ELISA (eBioscience, San Diego, CA).

| Intestinal permeability in vivo
After being fasted for 4 hours, the mice were administered FITCdextran (4 kDa; Sigma) by gavage at a dose of 600 mg/kg. Then, the mice received isoflurane anaesthesia through inhalation and were killed by spinal dislocation. Blood was collected through cardiac puncture, and serum was isolated using centrifugation. Serum FITC levels were evaluated using fluorometry. 27

| Bacterial translocation
Sterile isolation of mouse liver and spleen was performed for bacteriological cultures. The tissue samples were weighed, and 0.1 g of each sample was homogenized with 0.9 mL of sterile saline. The homogenates were diluted and cultured (100 μL) on MacConkey's agar (Sigma-Aldrich) at 37°C for 24 hours. Bacterial growth on the plates was expressed as colony forming units/g of tissue, and the presence of more than 10 2 colonies/g of tissue indicated a positive result. 28

| Flow cytometry
T-cell responses were analysed by flow cytometry as described previously. 29 For the Treg analysis, antibodies specific for CD4, CD25 and Foxp3 (eBioscience) were used to analyse the proportion of

| Western blot analysis
Total protein extracts were obtained from intestinal mucosa tissue, and the expression levels of target proteins were analysed by western blot analysis. In short, after SDS-PAGE, the proteins were transferred to a PVDF membrane, which was immunoblotted with antibodies against Gls-1, claudin-1, occludin, ZO-1, p-p70 S6K, p70 S6K, p-4E-BP1, 4E-BP1 or β-actin. Densitometric analysis of protein band intensity was performed with Imagej (National Institutes of Health, USA).

| Total RNA extraction and real-time quantitative PCR
Freshly intestinal mucosa tissues were lysed by Trizol reagent (Invitrogen) and cDNA was generated from 1 μg of isolated RNA using the PrimeScript RT reagent kit with gDNA Eraser (Takara).

| Statistical analysis
Statistical analyses were performed with GraphPad Software (San Diego, CA). Means and SDs were calculated. Binary and categorical data were compared by chi-squared tests for contingency tables.
The parametric Student's t test was used to assess the significance of differences between the IL-10 −/− and IL-10 −/− +BPTES groups, and differences were considered significant at P < 0.05.

| Increased Gls1 expression in the intestines of CD patients and IL-10 −/− mice
The presence of Gls1 can aggravate inflammatory reactions in autoimmune diseases. We found significantly increased Gls1 expression in intestinal mucosa tissue of CD patients compared with control patients ( Figure 1A,B). We also analysed the correlation of CD patients' clinical parameters with Gls1 expression, but we did not find a definite correlation. Glutaminase 1 expression was also higher in the inflamed areas of CD patients' intestinal tissue than in the uninflamed areas ( Figure 1C,D). In the animal experiments shown in Figure 1C Figure 2E). Our data also demonstrated that BPTES could significantly inhibit Gls1 expression and increase Gln levels in intestinal tissues ( Figure 2F-H), consistent with previous studies. 20,23 F I G U R E 1 Gls 1 is highly expressed in the intestines of CD patients and Il-10 -/mice. Immunohistochemical staining with an antibody that recognizes Gls1 was performed on the intestines of control and CD patients (A). The quantitative analysis presented in (B) shows Gls1 expression in the intestines of CD (n = 13) and control patients (n = 17). (C,D) Western blot analysis of Gls1 in the intestinal mucosa in the intestines of CD patients (inflamed and uninflamed areas) and control patients. (E,F) The expression of Gls1 in the intestines of Il-10 -/mice and WT mice (n = 8 in each group). CD, Crohn's disease; Gls1, glutaminase 1; IOD, integrated optical density; WT, wild-type. The data are presented as the relative IOD ± SD. ***P < 0.001 Consequently, BPTES treatment was found to ameliorate the signs of experimental colitis in IL-10 −/− mice.

| Treatment with BPTES improves intestinal barrier function in IL-10 −/− mice
Intestinal barrier dysfunction is one of the suggested causes of CD pathogenesis, and tight junction proteins (including claudin-1, occludin and ZO-1) contributing to barrier function; decreased levels of those proteins lead to altered tight junction structure. Thus, immunofluorescence was used to evaluate the intestinal mucosa of BPTES-treated mice and control mice. The results are shown in Figure 3A

| Gls1 regulates different T-cell subset responses in IL-10 −/− mice involved in activating mTORC1
In T cells, it has been reported that Gln transporter-deficient T cells have a decreased Th1/Th17 response and reduced TCR-mediated mammalian target of rapamycin complex 1 (mTORC1) activity. 32 In addition, inhibition of Gls1 can reduce the phosphorylation of p70 S6K, which is downstream of mTORC1. 20 We have suggested that Gls1 up-regulates Th1/Th17 and down-regulates the Treg response involved in regulating mTORC1 activity. As expected, BPTES treatment down-regulated p70 S6K ( Figure 5A,B) and 4E-BP1 ( Figure 5C) phosphorylation in the intestines of IL-10 −/− mice.

| D ISCUSS I ON
In this study, we detected significantly increased Gls1 expression on the CD intestinal mucosa and identified the protective effect of Gls1 in a spontaneous mouse model of chronic colitis. We demonstrated that Gls1 inhibition by BPTES ameliorated colitis in IL-10 −/− mice, as shown by histopathology and by reductions in inflammation scores and the DAI, which was associated with significantly decreased Th1 and Th17 responses but increased Treg responses. Furthermore, we found that the regulatory effects of Gls1 on the T-cell subset response may involve mTORC1 activation.
Previous studies have shown that the glutaminolysis pathway plays an important protective role in experimental colitis models and that targeting Gls1 (the first enzyme in the glutaminolysis pathway) is an effective therapy for autoimmune diseases. 16,20,22,23,33 In this study, Gls1 was highly expressed in the intestines of CD patients and IL-10 −/− mice. We suspect that high Gls1 levels in the intestine of CD patients may be related to the disease extent. Indeed, we found that inhibiting Gls1 expression in IL-10 −/− mice could ameliorate experimental colitis, which was associated with reduced mean DAI, body weight loss, histological inflammation scores and pro-inflammatory Our study has some limitations. First, our data indicated that Gls1 expression was significantly higher in the intestine of CD patients than in normal tissue, but we did not investigate the possible mechanisms leading to this elevation. Second, the inhibition of Gls1 showed that the protective effect against colitis may occur through other pathways in addition to regulating the balance of Th/Treg responses. Finally, Gls1 regulates T-cell responses possibly through mechanisms other than mTORC1 activation.
In summary, our study demonstrated that Gls1 expression is increased in CD patients' intestinal mucosa. Inhibition of Gls1 by BPTES maintains the Th/Treg balance, controls inflammatory reactions and maintains intestinal barrier integrity, which ameliorates spontaneous colitis in IL-10 −/− mice. We suggest that the therapeutic effect of Gls1 on CD-like colitis is mediated at least partially through regulation of the T-cell subset response involved in the mTORC1 signalling pathway.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data used to support the findings of this study are included in the article. F I G U R E 4 BPTES treatment maintains the Th/Treg balance in IL-10 −/− mice. Splenocytes and mesenteric lymph node cells from each group were prepared at the fourth week after drug administration. IFN-γ-producing CD4 + T cells and IL-17A-producing CD4 + T cells were quantified by intracellular cytokine staining as described in the Section 22. (A,C) Representative flow cytometric images of IFN-γ-producing CD4 + T cells. Summary of the percentage (B,D) of IFN-γ-producing CD4 + T cells. (E,G) Representative flow cytometric images of IL-17Aproducing CD4 + T cells. Summary of the percentage (F,H) of IL-17A-producing CD4 + T cells. (I,J) Representative flow cytometric images of CD4 + CD25 + Foxp3 + T cells in different organs. Summary of the percentage (K,L) of CD4 + CD25 + Foxp3 + T cells. The data are expressed as the mean ± SD (n = 8 per group). BPTES, bis-2-(5-phenylacetamido-1,2, 4-thiadiazol-2-yl) ethyl sulphide; IFN, interferon; IL, interleukin; NS, no significance; *P < 0.05, **P < 0.01, ***P < 0.001 F I G U R E 5 Gls1 regulate different T-cell subset responses in IL-10 −/− mice involve inactivating mTORC1. Total protein extracts were prepared from intestinal tissue, and the protein levels of p-p70 S6K, p70 S6K, p-4E-BP1 and 4E-BP1 were analysed by western blot; β-actin was used as an internal reference A,. Densitometric assays of all protein band intensities (B,C). The data are expressed as the mean ± SD (n = 8 per group). Gls1, glutaminase 1; NS, no significance; *P < 0.05, **P < 0.01