Activation of NLRX1 by NX-13 Alleviates Inflammatory Bowel Disease Through Immunometabolic Mechanisms in CD4+ T Cells

Home , NLRX1, NOD1

Activation of NLRX1 by NX-13 Alleviates Inflammatory Bowel Disease through Immunometabolic Mechanisms in CD4+ T Cells This information is current as of September 25, 2021. Andrew Leber, Raquel Hontecillas, Victoria Zoccoli-Rodriguez, Catherine Bienert, Jyoti Chauhan and Josep Bassaganya-Riera J Immunol published online 6 November 2019 http://www.jimmunol.org/content/early/2019/11/05/jimmun Downloaded from ol.1900364

Supplementary http://www.jimmunol.org/content/suppl/2019/11/05/jimmunol.190036 http://www.jimmunol.org/ Material 4.DCSupplemental

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists by guest on September 25, 2021 • Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2019 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published November 6, 2019, doi:10.4049/jimmunol.1900364 The Journal of Immunology

Activation of NLRX1 by NX-13 Alleviates Inﬂammatory Bowel Disease through Immunometabolic Mechanisms in CD4+ T Cells

Andrew Leber,*,† Raquel Hontecillas,*,† Victoria Zoccoli-Rodriguez,* Catherine Bienert,* Jyoti Chauhan,* and Josep Bassaganya-Riera*,†

Inﬂammatory bowel disease (IBD) is a complex autoimmune disease with dysfunction in pattern-recognition responses, including within the NLR family. Nucleotide-binding oligomerization domain, leucine rich repeat containing X1 (NLRX1) is a unique NLR with regulatory and anti-inﬂammatory functions resulting in protection from IBD in mouse models. NX-13 is an orally active, gut- restricted novel drug candidate that selectively targets and activates the NLRX1 pathway locally in the gut. In vitro and in vivo

efficacy of NLRX1 activation by NX-13 was examined. Oral treatment with NX-13 alleviates disease severity, colonic leukocytic Downloaded from infiltration, and cytokine markers of inflammation in three mouse models of IBD (dextran sulfate sodium, Mdr1a2/2, and CD45RBhi adoptive transfer). Treatment of naive CD4+ T cells with NX-13 in vitro decreases differentiation into Th1 and Th17 subsets with increased oxidative phosphorylation and decreased NF-kB activation and reactive oxygen species. With

stimulation by PMA/ionomycin, TNF-a,orH2O2, PBMCs from ulcerative colitis patients treated with NX-13 had decreased NF-kB activity, TNF-a+ and IFN-g+ CD4+ T cells and overall production of IL-6, MCP1, and IL-8. NX-13 activates NLRX1 to

mediate a resistance to both inﬂammatory signaling and oxidative stress in mouse models and human primary cells from http://www.jimmunol.org/ ulcerative colitis patients with effects on NF-kB activity and oxidative phosphorylation. NX-13 is a promising oral, gut- restricted NLRX1 agonist for treating IBD. The Journal of Immunology, 2019, 203: 000–000.

ver 3 million people in America and 5 million world- an unmet clinical need exists for novel targets capable of re- wide suffer from inﬂammatory bowel disease (IBD). storing immune tolerance and controlling the wide array of im- O This widespread and debilitating illness results in de- mune responses. creased quality of life and signiﬁcant healthcare related costs NLRs are a family of cytosolic pattern-recognition receptors that

(1). The underlying immune dysfunction in Crohn disease (CD) function in surveillance of metabolic stress, discrimination between by guest on September 25, 2021 (2, 3) and ulcerative colitis (UC) (4, 5) is with involvement of pathogenic bacteria versus beneficial symbionts, and recognition of both innate and adaptive immunity culminating in a perturbed inflammatory signaling (8–12). Genetic polymorphisms in several immune tolerance to the gut microbiome and other environ- NLRs have been described in IBD such as NLRC1 (NOD1) and mental factors. Despite this complexity, current therapeutics are NLRC2 (NOD2) (13–15). Although NOD1 and NOD2, in addi- unrefined steroids, immunosuppressants, and neutralizing Abs to tion to inflammasome-associated NLRs, are well-characterized to single cytokines such as TNF-a. Although current treatments for contribute to human disease, a smaller class of negative regulatory IBD have improved (6, 7), these therapies remain modestly NLRs remain relatively unknown. Indeed, genetic abnormalities successful with adverse side effects, including immunosuppres- in these well-characterized NLR genes are linked to IBD patho- sion, increased risk for infection, malignancies, and death. Thus, genesis (16–19). This smaller family of NLRs comprises NLRX1, NLRP12, and NLRC3 and has shown potent effects in the atten- *Landos Biopharma, Inc., Blacksburg, VA 24060; and †BioTherapeutics, Inc., uation of inflammation in nonclinical models of autoimmune and Blacksburg, VA 24060 infectious disease. NLRX1 regulates inflammation and IFN pro- ORCIDs: 0000-0001-9586-9953 (R.H.); 0000-0002-3837-3000 (J.C.); 0000-0001- duction in response to pathogens (20–23) and antagonizes NF-kB 8967-6012 (J.B.-R.). (20, 23, 24). However, many of these preclinical studies focus Received for publication March 29, 2019. Accepted for publication October 8, 2019. solely on the worsening of disease in loss-of-function designs. The This work was supported by National Institutes of Health Public Service Grant ability to modulate function by pharmacologic activation of the 1R43DK121561. receptor, the prevalence of function-disrupting mutations, and Conception and design of the study: A.L. and J.B.-R. Generation and interpretation of data: A.L., R.H., V.Z.-R., C.B., and J.B.-R. Drafting and revision of manuscript: A.L., native expression patterns in response to inflammation is unknown J.C., and J.B.-R. Approval of final version: A.L., R.H., V.Z.-R., J.C., and J.B.-R. in human disease. Address correspondence and reprint requests to Dr. Josep Bassaganya-Riera, Landos NLRX1 is a mitochondria-associated receptor involved in Biopharma, Inc., 1800 Kraft Drive, Suite 216, Blacksburg, VA 24060. E-mail ad- downregulating inflammation during bacterial and viral exposure, dress: [email protected] acute noninfectious colitis, colorectal cancer, and chronic pul- The online version of this article contains supplemental material. 2 2 monary disease (20, 22, 23). NLRX1 / mice with dextran sul- Abbreviations used in this article: CD, Crohn disease; DSS, dextran sulfate sodium; GI, gastrointestinal; IBD, inflammatory bowel disease; LDH, lactate dehydrogenase; fate sodium (DSS) colitis have significantly worse disease activity PMA/I, PMA (5 ng/ml) and ionomycin (500 ng/ml); qRT-PCR, quantitative RT-PCR; associated with more severe gut mucosal inflammation and fi- ROS, reactive oxygen species; Treg, regulatory T cell; UC, ulcerative colitis; WT, brosis when compared with their wild-type (WT) counterparts wild-type. (25). Colonic NLRX1 expression is significantly suppressed in Copyright Ó 2019 by The American Association of Immunologists, Inc. 0022-1767/19/$37.50 colons of mice with colitis (26). RNA-seq data from colons of WT

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1900364 2 NX-13 ALLEVIATES IBD BY IMMUNOMETABOLISM

2 2 and NLRX1 / mice unveil that loss of NLRX1 during colitis fixed and permeabilized for staining of intracellular Ags (IFN-g, Tbet, IL- causes overzealous inflammatory chemokine and cytokine sig- 10, FOXP3, IL-17, RORgT, and IL-4). Data were acquired on a naling (.10-fold increase) in myeloid and epithelial cell response BD FACSCelesta II and analyzed using BD FACSDiva version 9.0 (BD Biosciences). elements such as Il-6, Tnf, Ifng, Il-21, and Mcp1. Previously, we demonstrated that the deficiency of NLRX1 worsens inflammatory Gene expression + responses in vitro in both CD4 T cells and epithelial cells of mice Colon or cells were homogenized within RLT buffer. RNA was isolated and shifts the metabolic preferences of cells. When the metabolic from homogenate by RNAeasy Mini Kit, according to manufacturer’s differences are prevented, the immune effects resulting from the instructions. From RNA, cDNA was produced by iScript reverse tran- loss of NLRX1 are abrogated (25) both in vitro and in vivo that are scriptase. qRT-PCR was conducted with SYBR Green Master Mix. Start- ing quantity was calculated according to standard curve and normalized to integrated with changes in inflammation-associated bacterial taxa b-actin, as previously described. within the colonic microenvironment (27). In line with these beneficial preclinical responses in IBD, we Histopathology have developed 1, 3, 5-tris (6-methylpyridin-2-yloxy) benzene Colons were fixed within 10% buffered formalin, embedded in paraffin, (NX-13), as an orally active, gut-restricted, NLRX1-activating sectioned, and stained with H&E for examination. Colons were scored (0– therapeutic with limited systemic exposure for the treatment of 4) for leukocytic infiltration, mucosal thickening, and epithelial erosion by IBD. NX-13 was generated by medicinal chemistry-based opti- blinded pathologist. mization of novel scaffolds designed to target the previously PBMC culture categorized binding pocket of NLRX1. In preliminary safety and PBMCs were isolated from fresh whole blood of male and female UC pharmacokinetic studies, NX-13 is gut-restricted with no effects patients. PBMCs were separated by LeucoSep tube (Greiner Bio-One), and Downloaded from on systemic hematology parameters and a “no observable ad- remaining RBCs were lysed by hypotonic lysis. Isolated cells were treated verse effect level” $1000 mg/kg (28). Following through in silico with NX-13 (0, 50, 100, and 500 nM). Cells were stimulated by TNF-a testing of affinity, safety, and pharmacokinetics, NX-13 was syn- (0.5 ng/ml) for 24 h, hydrogen peroxide (1 mM) for 24 h, or PMA/I for 6 h. Cytokine responses were measured by Luminex, and NF-kB activity was thesized to characterize its efficacy in the treatment of IBD. To our measured by TransAM NF-kB p65 Kit (Active Motif). knowledge, this manuscript provides the first evidence for NX-13 Statistical analysis as a promising immunometabolic therapeutic in three models of http://www.jimmunol.org/ IBD and primary cells from human UC patients. Data are presented as mean 6 SEM. Quantitative data were analyzed using ANOVA, followed by the Scheffe multiple-comparisons test. ANOVA was Materials and Methods performed using the general linear model procedure of SAS (SAS Institute, Cary, NC). Statistical significance was determined at p , 0.05. Mice WT and Rag22/2 mice were used on a C57BL/6 background. Mdr1a2/2 Results were used on an FVB background. All procedures and experiments were NX-13 modulates CD4+ T cell differentiation and metabolism approved by an institutional animal care and use committee. Experimental groupings were age-, gender-, and litter-matched with equal distribution to NX-13 is a symmetric triphenyl compound designed to activate groups in cohoused conditions. For DSS experiments, 8-wk-old WT mice NLRX1 in the GI tract. By Cellular Thermal Shift Assay, NX-13 by guest on September 25, 2021 were administered DSS in drinking water for 7 d. Mice were scored daily for disease activity. For adoptive transfer experiments, 7-wk-old Rag22/2 has in situ binding effects on the stability of NLRX1 at con- mice were injected with 4 3 105 naive CD4+ T cells (CD45RBhi) per centrations below 100 nM, with an estimated EC50 of 58 nM mouse by i.p. injection. Mice were scored weekly posttransfer for disease (Fig. 1). As NLRX1 has previously been associated with 2/2 2/2 activity for 8 wk. For Mdr1a experiments, Mdr1a mice were started immunometabolic effects in CD4+ T cells in IBD (25), the dif- on treatment at 4 wk of age and continued treatment through 10 wk of age. ferentiation of naive CD4+ T cells was assessed in the presence Mice were scored weekly. Disease activity was scored on a range from 0 to 4, encompassing parameters pertaining to rectal inflammation, stool con- of NX-13. Treatment with NX-13 significantly decreased the sistency, presence of blood in stool, and overall appearance/behavior. differentiation of Th1 (Fig. 2A) and Th17 (Supplemental Fig. 1) General assignment of score corresponded to 0, no symptoms; 1, soft/ at 0.05 mM, with a dose response and suppressive effect ob- loose/watery stool; 2, blood or moderate rectal inflammation; 3, blood served down to 0.005 mM. The effect on CD4+ T cell differen- and moderate or greater inflammation; and 4, severe gastrointestinal (GI) tiation was observed to be dependent on the presence of NLRX1, symptoms with reduced activity or weight loss. NX-13 treatments, or + vehicle control, were administered by orogastric gavage daily for the du- whereby CD4 T cell deficient in NLRX1 did not respond to ration of each experiment. NX-13 treatment (Supplemental Fig. 1). Reduction of Th1- phenotyped cells corresponded with a marked decrease in IFN- CD4+ T cell culture g production (Fig. 2G, 2H). In line with the previously described Naive CD4+ T cells were obtained from spleens of WT mice by magnetic metabolic functions of NLRX1 activation, NX-13 decreased sorting of a cellular suspension of splenocytes. Naive CD4+ T cells were LDH activity (Fig. 2B) and glucose uptake (Fig. 2C) in in vitro– cultured in IMDM supplemented with FBS within anti-CD3/anti-CD28 coated 96 well plates. Cells were differentiated into Th1 or Th17 cells differentiated Th1 cells paired with a decrease in proliferation by cytokine mixture: IL-12 (10 ng/ml) or anti–IL-4 (500 ng/ml); or (Fig. 2D, 2I). No effect on cell death or apoptosis was observed IL-6 (10 ng/ml), TGF-b (5 ng/ml), or anti–IL-4 (500 ng/ml), respec- (Supplemental Fig. 1). NX-13 also decreased NF-kB activity tively. During differentiation, cells were treated with NX-13. After (Fig. 2E) and intracellular ROS levels (Fig. 2F), two inflamma- 48 h, cells were stimulated with PMA (5 ng/ml) and ionomycin tory signaling elements inhibited by NLRX1 activation. Addi- (500 ng/ml) (PMA/I) for 6 h. After stimulation, cells were assayed for cellular phenotype by flow cytometry, gene expression by quantitative tionally, after the observation of effects on metabolism and ROS RT-PCR (qRT-PCR), proliferation by CFSE staining, or NF-kB, reac- levels, we further characterized the effects of NX-13 treatment tive oxygen species (ROS), lactate dehydrogenase (LDH) activity, and by qRT-PCR of in vitro–differentiated Th1 cells. Although NX- glucose uptake, according to manufacturer’s instructions. 13 is associated with increased oxidative phosphorylation Flow cytometry (Fig. 2J–M), it is also associated with increased expression of antioxidant enzymes, including Gpx1, Gstm1, and Txnrd1 (Fig. Lamina propria lymphocytes were obtained from colons by collagenase/ DNase digestion followed by purification by Percoll gradient. Cultured 2N–P). Meanwhile, Eno1, a marker of anaerobic glycolysis ac- cells were obtained directly from culture. Cells were washed and stained for tivity and a moonlighting enzyme associated with inflammatory extracellular Ags (CD45, CD3, CD4, CD8, CD25, and Gr1). Cells were then signaling, is decreased by NX-13 (Fig. 2Q). The Journal of Immunology 3

FIGURE 1. NX-13 binding to NLRX1 in situ. Spleno- cytes were cultured with NX-13 (1, 10, 50, 100, 200, and 500 nM) for 3 h, following which cells were exposure to acute thermal stress. Protein was isolated, and NLRX1 content was assessed by Western blot and compared at a ratio of soluble to total (A). Melting temperature (B) was calculated by ﬁve-parameter logistic ﬁt of ratio of soluble to insoluble/denatured NLRX1. EC50 estimated to be 58 nM in situ by protection from denaturation.

NX-13 reduces inflammation in a model of DSS colitis reduction of fecal calprotectin in the DSS model when com- Upon validation of NLRX1-mediated activity in vitro, NX-13 pared with current UC therapeutics (Supplemental Fig. 3). By was tested in a DSS model of colitis to characterize its thera- flow cytometry immunophenotyping, 1 mg/kg oral NX-13 peutic efficacy and underlying immunological changes induced significantly decreased Th1, Th2, and Th17 (Fig. 3F–H) lo- by NX-13 in the gut. Oral treatment with NX-13 reduced disease cally within the colon. A significant decrease in neutrophils was observed at 10 mg/kg and higher, with a noted suppressive activity index (Fig. 3A), a summarized score encompassing + Downloaded from diarrhea, rectal bleeding, rectal inflammation, and overall ac- trend at lower dosages (Fig. 3I). In contrast, CD25 regulatory tivity, with significant reduction retained throughout the 7-d T cells (Tregs) were significantly increased at the highest test DSS period, at dosages as low as 1 mg/kg. Similar trends were dosage, 20 mg/kg, and slightly increased at lower dosages (Fig. 3J). Similar trends in cellular populations were also ob- observed microscopically with reductions in leukocytic infil- + tration, epithelial erosion and mucosal thickening recorded at 1 served based on percentage of CD45 cells (Supplemental Fig. 2). When NX-13 was dosed to unchallenged mice, splenic

mg/kg and higher upon examination of colon (Fig. 3B–D). http://www.jimmunol.org/ Reduction in leukocytic infiltration was observed at both the immune cell populations were unchanged (Supplemental Fig. mucosal and submucosal levels, whereas NX-13 also protected 2), suggestive of local action. against ulceration and loss of crypt architecture (Fig. 3K, 3L). hi Treatment with NX-13 resulted in an increased colonic ex- NX-13 maintains immune homeostasis after CD45RB pression of Nlrx1 (Supplemental Fig. 2), with significant in- adoptive transfer creases observed at 10 and 20 mg/kg. Fecal calprotectin, a Based on the in vitro response to NX-13 in CD4+ T cells, its clinical biomarker for intestinal inflammation and treatment therapeutic efficacy was also assessed in the CD45RBhi adoptive response and remission, was decreased by NX-13 treatment on transfer model of IBD. Oral NX-13 treatment significantly re- day 7 (Fig. 3E). Notably, NX-13 treatment resulted in a greater duced signs of disease posttransfer with reduction of overall by guest on September 25, 2021

FIGURE 2. In vitro responses to NX-13. Naive CD4+ T cells were obtained from WT spleens and differentiated into Th1 cells in vitro in presence of NX-13 (0–0.5 mM) (A). After differentiation, LDH activity (B), glucose uptake (C), proliferation (D), NF-kB activity (E), and ROS (F) were measured. Representative plots of Th1 populations in untreated (G) and 0.5 mM NX-13 (H). Representative histogram (I) of untreated (orange) and 0.05 mM NX-13 (blue). Expression of mt-Co3 (J), mt-Nd3 (K), Odgh (L), Idh2 (M), Gpx1 (N), Gstm1 (O), Txnrd1 (P), and Eno1 (Q) were measured by qRT-PCR. Data presented as mean 6 SEM (n = 6). Asterisks mark signiﬁcance. *p # 0.05. 4 NX-13 ALLEVIATES IBD BY IMMUNOMETABOLISM Downloaded from http://www.jimmunol.org/

FIGURE 3. Efficacy of NX-13 in DSS colitis. WT mice were challenged with DSS and treated daily with oral NX-13 (0–20 mg/kg). Disease activity was by guest on September 25, 2021 assessed daily (A). Colon histology was scored on day 7 of DSS challenge for leukocytic infiltration (B), epithelial erosion (C), and mucosal thickening (D). Fecal calprotectin was assessed in colonic contents on day 7 (E). Th1 [(F) CD4+ CD82 Tbet+ IFN-g+], neutrophils [(G) Gr1hi CD11b+], Treg [(H) CD4+ CD25+ FOXP3+ IL-10+], IL-4+ CD4+ (I), and Th17 [(J) CD4+ CD82 RORgT+ IL-17+] were quantified by flow cytometry on day 7. Representative photomicrographs of H&E-stained colon from vehicle (K) and 10 mg/kg NX-13 (L) groups; asterisks mark leukocytic infiltration, and arrows mark ulceration. Data presented as mean 6 SEM (n = 9). Asterisks mark significance. Scale bars, 100 mm. *p # 0.05. disease activity (Fig. 4A). Significant reductions in colonic his- colon, Th1 and neutrophil numbers were also significantly re- topathology parameters (Fig. 4B–D) were similarly observed with duced, whereas the total number of Tregs were unchanged NX-13 with notable decreases in leukocytic infiltration and mu- (Supplemental Fig. 2). At the transcriptional level, colonic ex- cosal thickening (Fig. 4E, 4F). In an environment of altered im- pression of Ifng and Tnf was reduced with oral NX-13 treatment mune tolerance, NX-13 decreased Th1 and Th17 cells, whereas (Fig. 5G, 5H). the number of CD25+ Tregs was unchanged (Fig. 4G–I). Reduced In vivo validation of NX-13–induced metabolic changes in neutrophils were also observed in the colonic lamina propria 2 2 Mdr1a / colitis of NX-13–treated mice (Fig. 4J). In regards to the percentage of CD45+ cells, similar trends were observed in the proportion of Upon demonstration of therapeutic efficacy, we examined local Th1 and neutrophil cells, whereas the proportion of Tregs was biomarkers of NX-13 activity. A panel of cytokines was used to increased (Supplemental Fig. 2). determine changes in inflammatory response, representative of the 2/2 effects of NX-13 (Fig. 6A). Notable cytokines prominent in IBD, Therapeutic efficacy of NX-13 in an Mdr1a model of colitis such as TNF, IL-6, and MCP1, were decreased by NX-13, whereas The knockout of Mdr1a was used as a spontaneous model of severe an increase in IL-10 was observed in NX-13–treated mice. Ad- colitis. Similar to other models of colitis, oral NX-13 treatment ditional cytokines with a lower degree of characterization in IBD, results in an early reduction of inflammation that is maintained such as CXCL1, MIP1a, IL-9, and IL-1a were also decreased, throughout the course of the experiment in Mdr1a2/2 (Fig. 5A). By indicating an ability of NX-13 to extend anti-inflammatory histopathology, NX-13 reduces the occurrence of crypt destruction properties beyond a single cytokine. Next, expression of meta- and structural abnormalities with reduced infiltration of neutro- bolic genes was measured in Mdr1a2/2 mice after treatment with phils and eosinophils (Fig. 5B, 5C). Cellular reductions in in- 10 or 20 mg/kg NX-13 to determine local metabolic biomarkers flammation were confirmed by flow cytometry by observation of of NX-13 activity. In a similar manner to that observed in vitro, reduced Th1 cells and neutrophils and increased proportion of NX-13resultedinanincreaseinmarkersofoxidativephos- IL-10–producing Tregs in the colonic lamina propria (Fig. 5D–F). phorylation, including the mitochondrially encoded Nd3 and Because of a decrease in total immune cell infiltration in the Co3, and rate-limiting enzyme, Ogdh (Fig. 6B–D). In contrast, The Journal of Immunology 5 Downloaded from http://www.jimmunol.org/ by guest on September 25, 2021

FIGURE 4. Efficacy of NX-13 in CD45RBhi adoptive transfer. Rag22/2 were adoptively transferred 4 3 105 naive CD4+ T cells and treated daily with oral NX-13 (10 mg/kg). Mice were scored twice weekly for disease activity until 8 wk posttransfer (A). Colon histology was scored at 8-wk posttransfer for leukocytic infiltration (B), mucosal thickening (C), and epithelial erosion (D). Representative photomicrographs of H&E-stained colon from vehicle (E) and 10 mg/kg NX-13 (F) groups; asterisks mark leukocytic infiltration, and bars mark mucosal thickening. Th1 [(G) CD4+ CD82 Tbet+ IFN-g+], Th17 [(H) CD4+ CD82 RORgT+ IL-17+], Treg [(I) CD4+ CD25+ FOXP3+ IL-10+], and neutrophils [(J) Gr1hi CD11b+] were quantified by flow cytometry in colon at 8 wk posttransfer. Data presented as mean 6 SEM (n = 10). Asterisks mark significance. Scale bars, 100 mm. *p # 0.05.

Eno1 was decreased by NX-13 (Fig. 6E). The change in oxidative production upon stimulation with PMA/I, TNF, or hydrogen phosphorylation was paired with increased expression of anti- peroxide (Fig. 7E, 7F). Importantly, NX-13 greatly reduced oxidant enzymes, Gpx1 and Gstm1 (Fig. 6F, 6G). In an activity- ROS production with oxidative stress caused by hydrogen based assay, colonic glutathione peroxidase was also observed peroxide. Additionally, NX-13 decreased inflammatory cyto- to be increased with NX-13 (Fig. 6H), with similar increases kines implicated in IBD, including IL-8, MCP1, and IL-6 upon observed in colonic glutathione levels (Fig. 6I). Similar to the stimulation with PMA/I or TNF (Fig. 7G–I). These results DSS model, these cytokines and metabolic changes correlated suggest a translatable potency of NX-13 between mice and to increased colonic expression of Nlrx1 at 10 and 20 mg/kg humans, with clear therapeutic benefit ex vivo for the reduction (Supplemental Fig. 2). of inflammation in primary cells from UC. NX-13 reduces inflammation and alters NLRX1 signaling in Discussion PBMCs from UC patients NX-13 is a promising immunometabolic oral, gut-restricted ther- Based on the efficacy of NX-13 in mouse models of IBD, the apeutic for CD and UC that is effective at reducing inflammation translational value and potency of response to NX-13 treatment was in DSS, adoptive transfer, and Mdr1a2/2 mouse models of IBD tested within PBMCs from moderate to severe UC patients. At plus in in vitro studies using primary murine and human concentrations of 0.01 mM and higher, NX-13 significantly cells. Through the activation of NLRX1, NX-13 increases ox- reduced TNF-a– and IL-4–producing cells after 24 h of culture idative phosphorylation in immune cells resulting in a decrease (Fig. 7A, 7B). Meanwhile, at 0.05 mM and above, IL-10–proin the inflammation-associated anaerobic glycolysis. However, ducing cells were increased, and IFN-g–producing cells were despite the increase in oxidative metabolism, NX-13 treat- decreased (Fig. 7C, 7D). NX-13 reduces both NF-kB and ROS ment decreases cellular ROS through activation and increased 6 NX-13 ALLEVIATES IBD BY IMMUNOMETABOLISM Downloaded from

FIGURE 5. Efficacy of NX-13 in Mdr1a2/2 mice. Mdr1a2/2 mice were administered oral NX-13 (20 mg/kg) daily for 6 wk, during which disease activity was assessed weekly (A). Representative photomicrographs of vehicle (B) and NX-13–treated (C) mice after 6 wk of treatment; asterisks mark sites of leukocytic infiltration, and brackets mark mucosal thickening. Th1 [(D) CD4+ CD82 Tbet+ IFN-g+], Treg [(E) CD4+ CD25+ FOXP3+ IL-10+], and neutrophils [(F) Gr1hi CD11b+] were quantified by flow cytometry in colon after 6 wk of treatment. Expression of colonic Ifng (G) and Tnf (H) were http://www.jimmunol.org/ measured by qRT-PCR. Data presented as mean 6 SEM (n = 9). Scale bars, 100 mm. Asterisks mark significance. *p # 0.05. expression of antioxidant enzymes. Downstream of NLRX1, T cells (Th1 and Th17) drive glycolytic responses to produce these immunometabolic changes result in decreased NF-kB energy quickly (31). Immune cells unable to establish fatty acid activity and lower production of inflammatory cytokines such metabolism can show insensitivity to regulatory and suppressive as TNF-a and IFN-g. Both in vitro and in vivo, these signaling signals (32). Our data suggest that NLRX1 acts as a key regu- changes result in lower levels of Th1 cells. Oral NX-13 treat- lator of metabolic reprogramming within immune cells with ment reduces colonic lesions and overall disease severity at a greater oxidative phosphorylation by NX-13 treatment in line dosage as low as 1 mg/kg, indicating a high potential for further with the previously described augmentation of glycolytic flux by guest on September 25, 2021 development as an investigational new drug for CD and UC. and decreased substrate-switching capability displayed by The concept that immune and metabolic pathways have im- NLRX1-deficient cells (25). Indeed, NLRX1 directly influences portant codependencies has clear relevance to the understanding the maturation of mitochondrial RNA for important oxidative of disease and therapeutic development, particularly IBD with phosphorylation enzymes (33). Further, Eno1, which is de- diminished activity of the electron transport chain and lower ATP creased by NX-13, is connected to TNF-a,IL-1b,andIFN-g,is levels observed in the inflamed GI tract (29, 30). Effector CD4+ increased in colonic tissues, and may serve as an important Ag in

FIGURE 6. In vivo biomarker and metabolic effects of NX-13 in Mdr1a2/2 mice. Mdr1a2/2 mice were administered oral NX-13 (0, 10, or 20 mg/kg) daily for 6 wk. Protein levels (A) of cytokines and chemokines in colon after 6 wk were measured by Luminex; single asterisks mark significance relative to vehicle in the 20 mg/kg group, and double asterisks mark significance at both 10 and 20 mg/kg. Expression of mt-Nd3 (B), mt-Co3 (C), Odgh (D), Eno1 (E), Gpx1 (F), and Gstm1 (G) were measured by qRT-PCR. Glutathione peroxidase (H) and glutathione (I) were measured in colon after 6 wk of treatment. Data presented as mean 6 SEM (n = 8). Asterisks mark significance. *p # 0.05. The Journal of Immunology 7 Downloaded from http://www.jimmunol.org/

FIGURE 7. Efficacy of NX-13 in UC PBMCs. PBMCs were isolated from blood samples of UC donors and cultured ex vivo with NX-13 (0, 0.01, 0.05, 0.1, and 0.5 mM) for 24 h. Cells were stimulated with PMA/I, TNF (0.5 ng/ml), or hydrogen peroxide (1 mM), as indicated. TNF-a+ (A), IL-4+ (B), IL-10+ (C), and IFN-g+ (D) CD4+ T cells by flow cytometry. NF-kB activity (E) and ROS levels (F) after 24 h treatment. Secreted IL-8 (G), MCP1 (H), and IL-6 (I) protein levels by Luminex. Data presented as mean 6 SEM (n = 5). Asterisks mark significance. *p # 0.05. by guest on September 25, 2021 IBD with 50% of UC and CD patients presenting with serum the pathogenesis of IBD and a cause of the increased intestinal anti-ENO1 Abs (34, 35). The remaining question was whether oxidative stress observed in UC and CD (45). Additionally, GPX1 the increased oxidative phosphorylation would lead to an in- activity has been connected to a decreased ability of TNF to ac- crease in intracellular ROS and activation of inflammatory tivate expression of VCAM and ICAM molecules (46), which pathways. However, activation of NLRX1 by NX-13 leads to a would lead to further support of in vivo anti-inflammatory effects decrease in ROS that is paired with increased antioxidant en- locally in the GI tract, and serve as a mechanistic link for the zyme activity. Because of its previous connection to the regu- observed decreased leukocytic infiltration observed in the three lation of MAVS, NLRX1 could be acting as an important switch models of IBD. Although current data support a lack of effect on through c-Abl that would serve to activate GPX1 (36, 37). The systemic immune profiles (28), risk for infection remains a con- role in this regulation merits further exploration. With previous cern in the development of immune-targeted therapeutics. The connections to the prevention of oxidative stress (25, 38) and the immunometabolic effects induced by NX-13 may support specific described data in this study, including increases to Gpx1, Gstm1, antiviral responses through a depletion of lactate and modulation and Txnrd1, NLRX1 is a potent controller of the cellular oxi- of the overall cellular energy phenotype. Many of the viral re- dative state. Meanwhile, impaired function of these enzymes has sponses connected to NLRX1 have been through its interaction been linked to enhanced susceptibility to IBD and leukocytic with MAVS; however, MAVS is also highly responsive to meta- infiltration in colons of patients (39–41). bolic change with recent publications linking high lactate to im- The implication of ROS in the NLRX1 pathway extends beyond paired MAVS and Type I IFN production (47) as well as the the metabolic effects but the downstream signaling. Among other necessity of O-GlcNAcylation in MAVS signaling (48). Therefore, molecules, ROS are proposed to be a key mediator in the down- the metabolic change combined with NLRX1 activation provide a stream effects of NLRX1 on NF-kB and potentially other in- balance in antiviral signaling. However, additional studies on the flammatory pathways including the phosphorylation of ERK1 and effect of NX-13 on both the susceptibility and severity of bacterial ERK2 leading to upregulation of MAPK activity (42). Although and viral infections is needed. cytoplasmic NLRX1 has direct effects on NF-kB activity, the Via control of NF-kB, it is proposed that NLRX1 activation control of Gpx1, and other antioxidant enzymes may provide influences a wide array of cytokine responses. Based on the pre- further inhibitory support of the pathway as IKKa can be inhibited sented findings, oral NX-13 treatment provides anti-inflammatory by the decrease in hydrogen peroxide levels (43). Conversely, control over a number of cytokines within the colonic mucosa. NF-kB activity has also been linked to antioxidant enzyme tran- These cytokines cover a broad range of functions including potent scription, suggesting a potential coregulation and feedback in inflammatory mediators of activation and differentiation (TNF and this linkage in parallel to NF-kB–mediated upregulation of ROS IL-6), neutrophil and monocyte chemokines (MCP1 and CXCL1), producers (44). Dysfunction of this feedback may be important in as well as those with newly emergent functions in IBD (MIP1a, 8 NX-13 ALLEVIATES IBD BY IMMUNOMETABOLISM

IL-9, and IL-1a). Recently, CXCL1 has been identified as a dif- 11. Agostini, L., F. Martinon, K. Burns, M. F. McDermott, P. N. Hawkins, and J. Tschopp. 2004. NALP3 forms an IL-1beta-processing inflammasome with ferentiator between responders and nonresponders clinically (49) increased activity in Muckle-Wells autoinflammatory disorder. Immunity 20: and IL-9 positively correlated to clinical scores, including mu- 319–325. cosal healing and endoscopic appearance in UC (50), whereas 12. Lara-Tejero, M., F. S. Sutterwala, Y. Ogura, E. P. Grant, J. Bertin, A. J. Coyle, R. A. Flavell, and J. E. Galań. 2006. Role of the caspase-1 inflammasome in changes in MIP1a and IL-1a indicate a potential involvement of Salmonella typhimurium pathogenesis. J. Exp. Med. 203: 1407–1412. myeloid cells in the response to NX-13 treatment (51). 13. Hugot, J. P., M. Chamaillard, H. Zouali, S. Lesage, J. P. Ce´zard, J. Belaiche, As regulators of many of these molecular changes, this manu- S. Almer, C. Tysk, C. A. O’Morain, M. Gassull, et al. 2001. Association of + NOD2 leucine-rich repeat variants with susceptibility to Crohn’s disease. Nature script largely focuses on the effect of NX-13 in CD4 T cells 411: 599–603. because of their role in the pathogenesis and chronicity of IBD 14. Lesage, S., H. Zouali, J. P. Ce´zard, J. F. Colombel, J. Belaiche, S. Almer, C. Tysk, C. O’Morain, M. Gassull, V. Binder, et al; GETAID Group. 2002. and the previous implication of NLRX1 in this cell type. NX-13 CARD15/NOD2 mutational analysis and genotype-phenotype correlation in 612 + shows a direct ability to decrease effector CD4 T cells, including patients with inflammatory bowel disease. Am. J. Hum. Genet. 70: 845–857. Th1 and Th17 cells. Recent evidence on the dual regulation of 15. Van Limbergen, J., K. Geddes, P. Henderson, R. K. Russell, H. E. Drummond, J. Satsangi, A. M. Griffiths, D. J. Philpott, and D. C. Wilson. 2015. Paneth cell these cell types (52), along with the development of anti–IL-23 marker CD24 in NOD2 knockout organoids and in inflammatory bowel disease Abs (53), which serve to suppress both Th1 and Th17 differenti- (IBD). Gut 64: 353–354. ation, support that the potential downmodulation of these cell 16. Feldmann, J., A. M. Prieur, P. Quartier, P. Berquin, S. Certain, E. Cortis, D. Teillac-Hamel, A. Fischer, and G. de Saint Basile. 2002. Chronic infantile types independently can reduce GI inflammation in IBD. Based on neurological cutaneous and articular syndrome is caused by mutations in CIAS1, prior knowledge of NLRX1 and therapeutic efficacy across mouse a gene highly expressed in polymorphonuclear cells and chondrocytes. Am. J. Hum. Genet. 71: 198–203. models of IBD, NX-13 is likely to also induce direct effects on 17. Hoffman, H. M., J. L. Mueller, D. H. Broide, A. A. Wanderer, and epithelial and myeloid cells, although currently unexplored. Given R. D. Kolodner. 2001. Mutation of a new gene encoding a putative pyrin-like Downloaded from the shared immunometabolic pathways in each of these cell types, protein causes familial cold autoinflammatory syndrome and Muckle-Wells syndrome. Nat. Genet. 29: 301–305. these direct effects would be supportive of the overall anti- 18. Villani, A. C., M. Lemire, G. Fortin, E. Louis, M. S. Silverberg, C. Collette, inflammatory nature of NX-13 treatment. N. Baba, C. Libioulle, J. Belaiche, A. Bitton, et al. 2009. Common variants in Combined with the described therapeutic efficacy in this man- the NLRP3 region contribute to Crohn’s disease susceptibility. Nat. Genet. 41: 71–76. uscript, NX-13 is safe up to oral dosages of 1000 mg/kg in pre- 19. Henckaerts, L., M. Pierik, M. Joossens, M. Ferrante, P. Rutgeerts, and liminary safety studies in rats (28). This benign safety profile is S. Vermeire. 2007. Mutations in pattern recognition receptor genes modulate http://www.jimmunol.org/ mediated in part by gut-restricted pharmacokinetics, with local seroreactivity to microbial antigens in patients with inflammatory bowel disease. Gut 56: 1536–1542. colonic concentrations over 1000-fold higher than those in plasma 20. Allen, I. C., C. B. Moore, M. Schneider, Y. Lei, B. K. Davis, M. A. Scull, (28). NX-13 is effective in vivo at oral dosages as low as 1 mg/kg, D. Gris, K. E. Roney, A. G. Zimmermann, J. B. Bowzard, et al. 2011. NLRX1 protein attenuates inflammatory responses to infection by interfering with the with in vitro responses below 100 nM. With a wide therapeutic RIG-I-MAVS and TRAF6-NF-kB signaling pathways. Immunity 34: 854–865. window permitted by this large safety margin, NX-13 is a prime 21. Lei, Y., H. Wen, Y. Yu, D. J. Taxman, L. Zhang, D. G. Widman, K. V. Swanson, candidate for advancement into investigational new drug–enabling K. W. Wen, B. Damania, C. B. Moore, et al. 2012. The mitochondrial proteins NLRX1 and TUFM form a complex that regulates type I interferon and auto- studies and clinical testing in UC and CD patients. By immuno- phagy. Immunity 36: 933–946. metabolic mechanisms targeting enhanced oxidative phosphory- 22. Moore, C. B., D. T. Bergstralh, J. A. Duncan, Y. Lei, T. E. Morrison, lation and controlled ROS levels, NLRX1 activation is an A. G. Zimmermann, M. A. Accavitti-Loper, V. J. Madden, L. Sun, Z. Ye, et al. by guest on September 25, 2021 2008. NLRX1 is a regulator of mitochondrial antiviral immunity. Nature 451: innovative first-in-class strategy for the treatment of autoimmune 573–577. disease. Further, the categorization of NLRX1 as a regulatory 23. Philipson, C. W., J. Bassaganya-Riera, M. Viladomiu, B. Kronsteiner, V. Abedi, NLR offers the potential to intrinsically restore immune tolerance S. Hoops, P. Michalak, L. Kang, S. E. Girardin, and R. Hontecillas. 2015. Modeling the regulatory mechanisms by which NLRX1 modulates innate im- to the gut. mune responses to Helicobacter pylori infection. PLoS One 10: e0137839. 24. Coutermarsh-Ott, S., A. Simmons, V. Capria, T. LeRoith, J. E. Wilson, B. Heid, C. W. Philipson, Q. Qin, R. Hontecillas-Magarzo, J. Bassaganya-Riera, et al. Disclosures 2016. NLRX1 suppresses tumorigenesis and attenuates histiocytic sarcoma All authors are employees of Landos Biopharma. J.B.-R. is a shareholder of through the negative regulation of NF-kB signaling. Oncotarget 7: 33096– Landos Biopharma. 33110. 25. Leber, A., R. Hontecillas, N. Tubau-Juni, V. Zoccoli-Rodriguez, M. Hulver, R. McMillan, K. Eden, I. C. Allen, and J. Bassaganya-Riera. 2017. NLRX1 regulates effector and metabolic functions of CD4+ T cells. J. Immunol. 198: References 2260–2268. 1. Kappelman, M. D., S. L. Rifas-Shiman, C. Q. Porter, D. A. Ollendorf, 26. Lu, P., R. Hontecillas, V. Abedi, S. Kale, A. Leber, C. Heltzel, M. Langowski, R. S. Sandler, J. A. Galanko, and J. A. Finkelstein. 2008. Direct health care costs V. Godfrey, C. Philipson, N. Tubau-Juni, et al. 2015. Modeling-enabled char- of Crohn’s disease and ulcerative colitis in US children and adults. Gastroen- acterization of novel NLRX1 ligands. PLoS One 10: e0145420. terology 135: 1907–1913. 27. Leber, A., R. Hontecillas, N. Tubau-Juni, V. Zoccoli-Rodriguez, V. Abedi, and 2. Baumgart, D. C., and W. J. Sandborn. 2012. Crohn’s disease. Lancet 380: 1590– J. Bassaganya-Riera. 2018. NLRX1 modulates immunometabolic mechanisms 1605. controlling the host-gut microbiota interactions during inflammatory bowel 3. Torres, J., S. Mehandru, J. F. Colombel, and L. Peyrin-Biroulet. 2017. Crohn’s disease. Front. Immunol. 9: 363. disease. Lancet 389: 1741–1755. 28. Leber, A., R. Hontecillas, V. Zoccoli-Rodriguez, M. Ehrich, J. Chauhan, and 4. Orda´s, I., L. Eckmann, M. Talamini, D. C. Baumgart, and W. J. Sandborn. 2012. J. Bassaganya-Riera. 2019. Exploratory studies with NX-13: oral toxicity and Ulcerative colitis. Lancet 380: 1606–1619. pharmacokinetics in rodents of an orally active, gut-restricted first-in-class 5. Ungaro, R., S. Mehandru, P. B. Allen, L. Peyrin-Biroulet, and J. F. Colombel. therapeutic for IBD that targets NLRX1. Drug Chem. Toxicol. DOI: 10.1080/ 2017. Ulcerative colitis. Lancet 389: 1756–1770. 01480545.2019.1679828. 6. Camilleri, M. 2003. GI clinical research 2002-2003: the year in review. Clin. 29. Novak, E. A., and K. P. Mollen. 2015. Mitochondrial dysfunction in inflam- Gastroenterol. Hepatol. 1: 415–420. matory bowel disease. Front. Cell Dev. Biol. 3: 62. 7. Kozuch, P. L., and S. B. Hanauer. 2008. Treatment of inflammatory bowel dis- 30. Santhanam, S., S. Rajamanickam, A. Motamarry, B. S. Ramakrishna, ease: a review of medical therapy. World J. Gastroenterol. 14: 354–377. J. G. Amirtharaj, A. Ramachandran, A. Pulimood, and A. Venkatraman. 2012. 8. Ting, J. P., R. C. Lovering, E. S. Alnemri, J. Bertin, J. M. Boss, B. K. Davis, Mitochondrial electron transport chain complex dysfunction in the colonic R. A. Flavell, S. E. Girardin, A. Godzik, J. A. Harton, et al. 2008. The NLR gene mucosa in ulcerative colitis. Inflamm. Bowel Dis. 18: 2158–2168. family: a standard nomenclature. Immunity 28: 285–287. 31. Fox, C. J., P. S. Hammerman, and C. B. Thompson. 2005. Fuel feeds function: 9. Franchi, L., C. McDonald, T. D. Kanneganti, A. Amer, and G. Nuń˜ez. 2006. energy metabolism and the T-cell response. Nat. Rev. Immunol. 5: 844–852. Nucleotide-binding oligomerization domain-like receptors: intracellular pattern 32. Michalek, R. D., V. A. Gerriets, S. R. Jacobs, A. N. Macintyre, N. J. MacIver, recognition molecules for pathogen detection and host defense. J. Immunol. 177: E. F. Mason, S. A. Sullivan, A. G. Nichols, and J. C. Rathmell. 2011. Cutting 3507–3513. edge: distinct glycolytic and lipid oxidative metabolic programs are essential for 10. Harton, J. A., M. W. Linhoff, J. Zhang, and J. P. Ting. 2002. Cutting edge: CAT- effector and regulatory CD4+ T cell subsets. J. Immunol. 186: 3299–3303. ERPILLER: a large family of mammalian genes containing CARD, pyrin, 33. Singh, K., L. Sripada, A. Lipatova, M. Roy, P. Prajapati, D. Gohel, K. Bhatelia, nucleotide-binding, and leucine-rich repeat domains. J. Immunol. 169: 4088–4093. P. M. Chumakov, and R. Singh. 2018. NLRX1 resides in mitochondrial RNA The Journal of Immunology 9

granules and regulates mitochondrial RNA processing and bioenergetic adap- 44. Morgan, M. J., and Z. G. Liu. 2011. Crosstalk of reactive oxygen species and tation. Biochim. Biophys. Acta Mol. Cell Res. 1865: 1260–1276. NF-kB signaling. Cell Res. 21: 103–115. 34. Bae, S., H. Kim, N. Lee, C. Won, H. R. Kim, Y. I. Hwang, Y. W. Song, 45. Tian, T., Z. Wang, and J. Zhang. 2017. Pathomechanisms of oxidative stress in J. S. Kang, and W. J. Lee. 2012. a-Enolase expressed on the surfaces of inflammatory bowel disease and potential antioxidant therapies. Oxid. Med. Cell. monocytes and macrophages induces robust synovial inflammation in rheuma- Longev. 2017: 4535194. toid arthritis. J. Immunol. 189: 365–372. 46. Lubos, E., N. J. Kelly, S. R. Oldebeken, J. A. Leopold, Y. Y. Zhang, J. Loscalzo, 35. Ji, H., J. Wang, J. Guo, Y. Li, S. Lian, W. Guo, H. Yang, F. Kong, L. Zhen, and D. E. Handy. 2011. Glutathione peroxidase-1 deficiency augments proin- L. Guo, and Y. Liu. 2016. Progress in the biological function of alpha-enolase. flammatory cytokine-induced redox signaling and human endothelial cell acti- Anim. Nutr. 2: 12–17. vation. J. Biol. Chem. 286: 35407–35417. 36. Cao, C., Y. Leng, W. Huang, X. Liu, and D. Kufe. 2003. Glutathione peroxidase 47. Zhang, W., G. Wang, Z. G. Xu, H. Tu, F. Hu, J. Dai, Y. Chang, Y. Chen, Y. Lu, 1 is regulated by the c-Abl and Arg tyrosine kinases. J. Biol. Chem. 278: 39609– H. Zeng, et al. 2019. Lactate is a natural suppressor of RLR signaling by tar- 39614. geting MAVS. Cell 178: 176–189.e15. 37. Jacobs, J. L., and C. B. Coyne. 2013. Mechanisms of MAVS regulation at the 48. Li, T., X. Li, K. S. Attri, C. Liu, L. Li, L. E. Herring, J. M. Asara, Y. L. Lei, mitochondrial membrane. J. Mol. Biol. 425: 5009–5019. P. K. Singh, C. Gao, and H. Wen. 2018. O-GlcNAc transferase links glucose 38. Stokman, G., L. Kors, P. J. Bakker, E. Rampanelli, N. Claessen, G. J. D. Teske, metabolism to MAVS-mediated antiviral innate immunity. Cell Host Microbe 24: L. Butter, H. van Andel, M. A. van den Bergh Weerman, P. W. B. Larsen, et al. 791–803.e6. 2017. NLRX1 dampens oxidative stress and apoptosis in tissue injury via control 49. Zwicker, S., R. Lira-Junior, C. Hoö¨g, S. Almer, and E. A. Bostro¨m. 2017. of mitochondrial activity. J. Exp. Med. 214: 2405–2420. Systemic chemokine levels with “Gut-Specific” vedolizumab in patients with 39. Karban, A., N. Krivoy, H. Elkin, L. Adler, Y. Chowers, R. Eliakim, and E. Efrati. inflammatory bowel disease-A pilot study. Int. J. Mol. Sci. 18: 1827. 2011. Non-Jewish Israeli IBD patients have significantly higher glutathione S- 50. Matusiewicz, M., K. Neubauer, I. Bednarz-Misa, S. Gorska, and M. Krzystek- transferase GSTT1-null frequency. Dig. Dis. Sci. 56: 2081–2087. Korpacka. 2017. Systemic interleukin-9 in inflammatory bowel disease: asso- 40. Mrowicki, J., M. Mrowicka, I. Majsterek, M. Mik, A. Dziki, and Ł. Dziki. 2016. ciation with mucosal healing in ulcerative colitis. World J. Gastroenterol. 23: Evaluation of effect CAT -262C/T, SOD + 35A/C, GPx1 Pro197Leu polymorphisms 4039–4046. in patients with IBD in the polish population. Pol. Przegl. Chir. 88: 321–327. 51. Arimura, K., H. Takagi, T. Uto, T. Fukaya, T. Nakamura, N. Choijookhuu, 41. Zhang, C., L. Guo, and Y. Qin. 2016. [Interaction of MIF gene -173G/C poly- Y. Hishikawa, Y. Yamashita, and K. Sato. 2017. Crucial role of plasmacytoid morphism and GPX1 gene 594C/T polymorphism with high-fat diet in ulcerative dendritic cells in the development of acute colitis through the regulation of in- Downloaded from colitis]. Zhonghua Yi Xue Yi Chuan Xue Za Zhi 33: 85–90. testinal inflammation. Mucosal Immunol. 10: 957–970. 42. Sharma, A., D. Yuen, O. Huet, R. Pickering, N. Stefanovic, P. Bernatchez, and 52. Harbour, S. N., C. L. Maynard, C. L. Zindl, T. R. Schoeb, and C. T. Weaver. J. B. de Haan. 2016. Lack of glutathione peroxidase-1 facilitates a pro- 2015. Th17 cells give rise to Th1 cells that are required for the pathogenesis of inflammatory and activated vascular endothelium. [Published erratum appears colitis. Proc. Natl. Acad. Sci. USA 112: 7061–7066. in 2017 Vascul. Pharmacol. 99: 83.] Vascul. Pharmacol. 79: 32–42. 53. Feagan, B. G., W. J. Sandborn, G. D’Haens, J. Pane´s, A. Kaser, M. Ferrante, 43. Li, Q., S. Sanlioglu, S. Li, T. Ritchie, L. Oberley, and J. F. Engelhardt. 2001. E. Louis, D. Franchimont, O. Dewit, U. Seidler, et al. 2017. Induction therapy GPx-1 gene delivery modulates NFkappaB activation following diverse envi- with the selective interleukin-23 inhibitor risankizumab in patients with

ronmental injuries through a speciﬁc subunit of the IKK complex. Antioxid. moderate-to-severe Crohn’s disease: a randomised, double-blind, placebo- http://www.jimmunol.org/ Redox Signal. 3: 415–432. controlled phase 2 study. Lancet 389: 1699–1709. by guest on September 25, 2021

Figure S1. In vitro CD4+ T cell responses to NX-13. Effect of NX-13 on cell viability after 24 h of treatment by Annexin V assay (A). Differentiation of naïve CD4+ T cells into Th17 cells (B). NX-13 reduces differentiation of Th17 cells, measured by flow cytometry. Differentiation of naïve CD4+ T cells into Th1 (C) and Th17 (D) cells in control and Nlrx1-deficient cells. Data presented as mean ± SEM (n = 5). Asterisks mark significance (P ≤ 0.05).

Figure S2. Colonic expression of NLRX1. mRNA expression of NLRX1 in DSS (A) and Mdr1a-/- (B) models of colitis after treatment with NX-13 by qRT-PCR. Expression measured at day 7 of DSS challenge and at 10 weeks of age in Mdr1a-/-, respectively. Splenic immune cells by flow cytometry after 1 week of treatment with vehicle or NX-13 (20 mg/kg) by oral gavage (C). Data presented as mean ± SEM (n = 5). Percentage of CD45+ cells after DSS challenge (D, n = 9) after 7 days NX-13 and adoptive transfer (E, n = 10) after 8 weeks NX-13. Total cell number of lamina propria immune cells in Mdr1a-/- colitis (F, n = 9) after six weeks of NX-13. Th1 (CD4+ CD8- Tbet+ IFNγ+), neutrophils (Gr1hi CD11b+), Treg (CD4+ CD25+ FOXP3+ IL10+), and Th17 (CD4+ CD8- RORγT+ IL17+) cells were quantified by flow cytometry. Asterisks mark significance (P ≤ 0.05).

Figure S3. Comparative efficacy of NX-13 to current UC therapeutics. Mean change in day 7 fecal calprotectin concentration relative to vehicle treated mice after DSS challenge (A). Lamina propria Th1 cells by flow cytometry (B) and colonic leukocytic infiltration score by H&E (C) after four weeks of treatment in Mdr1a-/- mice. NX-13 was evaluated at a once daily oral dose of 10 mg/kg. Mesalamine was administered at 25 mg/kg once daily by oral gavage. Anti-TNF was administered at 2 mg/kg twice weekly by intravenous injection. Tofacitinib was administered at 0.75 mg doses twice daily by oral gavage. Prednisone (0.625 mg/kg), budesonide (0.125 mg/kg), and azathioprine (2.5 mg/kg) were delivered by oral gavage once daily. Data presented as mean ± SEM (n = 5). Asterisks mark significance (*, P ≤ 0.05; **, P ≤ 0.01).