BACKGROUND: At diagnosis, up to one-third of patients with Crohn's disease (CD) have a complicated phenotype with stricturing (B2) or penetrating (B3) behavior or require early surgery. We evaluated protein biomarkers and antimicrobial antibodies in serum archived years before CD diagnosis to assess whether complicated diagnoses were associated with a specific serological signature.METHODS: Prediagnosis serum was obtained from 201 patients with CD and 201 healthy controls. Samples were evaluated with a comprehensive panel of 1129 proteomic markers (SomaLogic) and antimicrobial antibodies. CD diagnosis and complications were defined by the International Classification of Diseases-Ninth Revision and Current Procedural Terminology codes. Cox regression models were utilized to assess the association between markers and the subsequent risk of being diagnosed with complicated CD. In addition, biological pathway and network analyses were performed.RESULTS: Forty-seven CD subjects (24%) had a B2 (n = 36) or B3 (n = 9) phenotype or CD-related surgery (n = 2) at diagnosis. Subjects presenting with complicated CD at diagnosis had higher levels of antimicrobial antibodies six years before diagnosis as compared with those diagnosed with noncomplicated CD. Twenty-two protein biomarkers (reflecting inflammatory, fibrosis, and tissue protection markers) were found to be associated with complicated CD. Pathway analysis of the altered protein biomarkers identified higher activation of the innate immune system and complement or coagulation cascades up to six years before diagnosis in complicated CD.CONCLUSIONS: Proteins and antimicrobial antibodies associated with dysregulated innate immunity, excessive adaptive response to microbial antigens, and fibrosis precede and predict a complicated phenotype at the time of diagnosis in CD patients.

BACKGROUND & AIMS: Anti-granulocyte macrophage-colony stimulating factor autoantibodies (aGMAbs) are detected in patients with ileal Crohn’s disease (CD). Their induction and mode of action during or before disease are not well understood. We aimed to investigate the underlying mechanisms associated with aGMAb induction, from functional orientation to recognized epitopes, for their impact on intestinal immune homeostasis and use as a predictive biomarker for complicated CD. METHODS: We characterized using enzyme-linked immunosorbent assay naturally occurring aGMAbs in longitudinal serum samples from patients archived before the diagnosis of CD (n = 220) as well as from 400 healthy individuals (matched controls) as part of the US Defense Medical Surveillance System. We used biochemical, cellular, and transcriptional analysis to uncover a mechanism that governs the impaired immune balance in CD mucosa after diagnosis. RESULTS: Neutralizing aGMAbs were found to be specific for post-translational glycosylation on granulocyte macrophage-colony stimulating factor (GM-CSF), detectable years before diagnosis, and associated with complicated CD at presentation. Glycosylation of GM-CSF was altered in patients with CD, and aGMAb affected myeloid homeostasis and promoted group 1 innate lymphoid cells. Perturbations in immune homeostasis preceded the diagnosis in the serum of patients with CD presenting with aGMAb and were detectable in the noninflamed CD mucosa. CONCLUSIONS: Anti-GMAbs predict the diagnosis of complicated CD long before the diagnosis of disease, recognize uniquely glycosylated epitopes, and impair myeloid cell and innate lymphoid cell balance associated with altered intestinal immune homeostasis.

Mutations of the NOD2 gene are risk factors for Crohn's disease. Many aspects of how they contribute to the condition are unknown. The discovery of cell populations that are involved suggests new therapeutic options. Cell populations shed light on how NOD2 mutations contribute to disease.

Background & Aims: Anti-GM-CSF autoantibodies (aGMAb) are detected in ileal Crohns Disease (CD) patients. Their induction and mode of action impacting homeostasis during, or prior to disease are not well understood. We aimed to investigate the underlying mechanisms leading to the induction of aGMAb, from functional orientation to recognized epitopes, for their impact on intestinal immune homeostasis and use as predictive biomarker for complicated CD. Methods: Using longitudinally collected sera from active component US personnel, we characterize naturally occurring aGMAb in a subset of CD patients years before disease onset. We employed biochemical, cellular, and transcriptional analysis to uncover a mechanism that governs the impaired immune balance in CD years prior to diagnosis. Results: Neutralizing aGMAb are specific to posttranslational glycosylations on GM-CSF, detectable years prior to diagnosis, and associated with complicated CD at presentation. Glycosylation and production of GM-CSF change in CD patients, altering myeloid homeostasis and destabilizing group 3 innate lymphoid cells. Perturbations in immune homeostasis precede the inflammation and are detectable in the non-inflamed CD mucosa of patients presenting with anti-GM-CSF autoantibodies. Conclusions: Anti-GM-CSF autoantibodies predict the diagnosis of complicated CD, have unique epitopes, and impair myeloid cell homeostasis across the ILC3-GM-CSF-myeloid cell axis, altering intestinal immune homeostasis long before the diagnosis of disease.

BACKGROUND & AIMS: Biomarkers are needed to identify patients at risk for development of inflammatory bowel diseases. We aimed to identify serum biomarkers of Crohn's disease and ulcerative colitis that can be detected and quantified before diagnosis. METHODS: We obtained serum samples from patients archived before a diagnosis of Crohn's disease (n = 200) or ulcerative colitis (n = 199), as well as from 200 healthy individuals (controls), collected from 1998 through 2013 as part of the US Defense Medical Surveillance System. We measured levels of antibodies against microbes (anti-Saccharomyces cerevisiae IgA or IgG, anti-Escherichia coli outer membrane porin C, anti-CBir1, anti-flagellin 2, anti-flagellin X, and perinuclear anti-neutrophil cytoplasmic antibodies) and 1129 proteins in each sample. We then used functional principal component analysis to derive the time-varying trajectory for each marker, which then was used in a multivariate model to predict disease status. Predictive performances at different prediagnosis timepoints were evaluated using area under the receiver operating characteristic curves (AUROCs). Biological pathways that were up-regulated in serum from patients with Crohn's disease were identified based on changes in protein abundance at different time periods preceding diagnosis. RESULTS: We identified a panel of 51 protein biomarkers that were predictive of Crohn's disease within 5 years with an AUROC of 0.76 and a diagnosis within 1 year with an AUROC of 0.87. Based on the proteins included in the panel, imminent development of CD was associated with changes in the complement cascade, lysosomes, innate immune response, and glycosaminoglycan metabolism. Serum antibodies and proteins identified patients who received a diagnosis of ulcerative colitis within 5 years with an AUROC of only 0.56 and within 1 year with an AUROC of 0.72. CONCLUSIONS: We identified a panel of serum antibodies and proteins that were predictive of patients who will receive a diagnosis of Crohn's disease within 5 years with high accuracy. By contrast we did not identify biomarkers associated with future diagnosis of ulcerative colitis.

17 human-derived Clostridium strains (VE202), belonging to a bacterial cluster under-represented in active IBD, induce colonic IL-10-producing FOXP3+ regulatory T cells and prevent colitis in murine IBD models (Nature 2013). The VE202 consortium is a promising live biotherapeutic product for clinical application. However, its 1) therapeutic benefit on established colitis and 2) effect on intestinal microbiota composition remain unclear. Models: We established colitis in ex-germ-free (exGF) Rag2−/− mice with naïve T cell transfer (Hu-nT) and Il10−/− (Hu-Il10−/−) mice colonized with healthy human fecal microbiota, using stool with low abundance of Clostridiales. In Hu-Il10−/−, we tested two other healthy donor stools (one with low Clostridiales and another with Fusobacteria, which is a potential pathobiont in IBD). For a selected bacteria-colonized model, we selected three aggressive Enterobacteriaceae strains cultured from the human donor stool used for the previous studies, based on colonic Th1 response and colitis in monoassociated exGF-Il10−/−. We then established an aggressive Klebsiella (K), Enterobacter (E) and E. coli (E) strain -colonized exGF Il10−/− colitis model (KEE-colitis). In addition, we established a human IBD-derived E.coli and Fusobacterium varium consortium-colonized Il10−/− (EF-colitis) model. In vitro, we anaerobically co-cultured human stools with VE202 and analyzed bacterial composition. Therapy: We treated mice by oral administration of VE202 or vehicle twice weekly for 2–4 weeks. Evaluation: Colitis severity was assessed by blinded histological score, fecal lipocalin-2, stool consistency score and colonic IFNγ +CD4+ T cells. Bacteria profiles in cecal contents were analyzed by metagenomics or qPCR. Hu-nT and Hu-Il10−/− (Fig.1A, B): VE202 significantly reversed histological colitis and other inflammatory endpoints compared to vehicle. Consistently VE202 attenuated colitis induced by three different donor stools. Bacteria profiling revealed that VE202 attenuated the colitis-associated bloom of Enterobacteriaceae and Fusobacteria. In KEE-colitis (Fig.1C) VE202 reversed colitis with reduction of all Enterobacteriaceae strains. In EF-colitis (Fig.1D) VE202 reversed colitis with reduction of E. coli and F. varium. Co-culture (Fig.1E): VE202 reduced abundance of Enterobacteriaceae in a dose dependent manner. Our findings suggest a novel IL-10-independent protective mechanism for human Clostridium VE202 strains, i.e. correction of dysbiosis with reduction of levels of Enterobacteriaceae and Fusobacteria. In addition, VE202 treatment is consistently effective for different human donor microbiota in vivo and in vitro. These results provide a rationale and target for therapeutic use of rationally selected resident protective bacterial cocktails in IBD patients.

Iron deficiency, a common comorbidity of gastrointestinal inflammatory disorders such as inflammatory bowel diseases (IBD), is often treated with oral iron supplementation. However, the safety of oral iron supplementation remains controversial because of its association with exacerbated disease activity in a subset of IBD patients. Because iron modulates bacterial growth and function, one possible mechanism by which iron may exacerbate inflammation in susceptible hosts is by modulating the intestinal microbiota. We, therefore, investigated the impact of dietary iron on the intestinal microbiota, utilizing the conventionalization of germ-free mice as a model of a microbial community in compositional flux to recapitulate the instability of the IBD-associated intestinal microbiota. Our findings demonstrate that altering intestinal iron availability during community assembly modulated the microbiota in non-inflamed wild type (WT) and colitis-susceptible interleukin-10-deficient (Il10(-/-)) mice. Depletion of luminal iron availability promoted luminal compositional changes associated with dysbiotic states irrespective of host genotype, including an expansion of Enterobacteriaceae such as Escherichia coli. Mechanistic in vitro growth competitions confirmed that high-affinity iron acquisition systems in E. coli enhance its abundance over other bacteria in iron-restricted conditions, thereby enabling pathobiont iron scavenging during dietary iron restriction. In contrast, distinct luminal community assembly was observed with dietary iron supplementation in WT versus Il10(-/-) mice, suggesting that the effects of increased iron on the microbiota differ with host inflammation status. Taken together, shifts in dietary iron intake during community assembly modulate the ecological structure of the intestinal microbiota and is dependent on host genotype and inflammation status.

The phosphoinositide 3-kinase catalytic subunit p110δ (PI3Kδ) gene maps to a human inflammatory bowel diseases (IBD) susceptibility locus, and genetic deletion of PI3Kδ signaling causes spontaneous colitis in mice. However, little is known regarding the role of PI3Kδ on IL-10-producing B cells that help regulate mucosal inflammation in IBD. We investigated the role of PI3Kδ signaling in B cell production of IL-10, following stimulation by resident bacteria and B cell regulatory function against colitis. In vitro, B cells from PI3KδD910A/D910A mice or wild-type B cells treated with PI3K specific inhibitors secreted significantly less IL-10 with greater IL-12p40 following bacterial stimulation. These B cells failed to suppress inflammatory cytokines by co-cultured microbiota-activated macrophages or CD4+ T cells. In vivo, co-transferred wild-type B cells ameliorated T cell-mediated colitis, while PI3KδD910A/D910A B cells did not confer protection from mucosal inflammation. These results indicate that PI3Kδ-signaling mediates regulatory B cell immune differentiation when stimulated with resident microbiota or their components, and is critical for induction and regulatory function of IL-10-producing B cells in intestinal homeostasis and inflammation.

To identify factors that regulate gut microbiota density and the impact of varied microbiota density on health, we assayed this fundamental ecosystem property in fecal samples across mammals, human disease, and therapeutic interventions. Physiologic features of the host (carrying capacity) and the fitness of the gut microbiota shape microbiota density. Therapeutic manipulation of microbiota density in mice altered host metabolic and immune homeostasis. In humans, gut microbiota density was reduced in Crohn's disease, ulcerative colitis, and ileal pouch-anal anastomosis. The gut microbiota in recurrent Clostridium difficile infection had lower density and reduced fitness that were restored by fecal microbiota transplantation. Understanding the interplay between microbiota and disease in terms of microbiota density, host carrying capacity, and microbiota fitness provide new insights into microbiome structure and microbiome targeted therapeutics. Editorial note:This article has been through an editorial process in which the authors decide how to respond to the issues raised during peer review. The Reviewing Editor's assessment is that all the issues have been addressed (see decision letter).

Environmental triggers is part of five focus areas of the Challenges in IBD research document, which also includes preclinical human IBD mechanisms, novel technologies, precision medicine and pragmatic clinical research. The Challenges in IBD research document provides a comprehensive overview of current gaps in inflammatory bowel diseases (IBD) research and delivers actionable approaches to address them. It is the result of a multidisciplinary input from scientists, clinicians, patients, and funders, and represents a valuable resource for patient centric research prioritization. In particular, the environmental triggers section is focused on the main research gaps in elucidating causality of environmental factors in IBD. Research gaps were identified in: 1) epidemiology of exposures; 2) identification of signatures of biological response to exposures; and 3) mechanisms of how environmental exposures drive IBD. To address these gaps, the implementation of longitudinal prospective studies to determine disease evolution and identify sub-clinical changes in response to exposures is proposed. This can help define critical windows of vulnerability and risk prediction. In addition, systems biology analysis and in silico modeling were proposed as approaches to integrate the IBD exposome for the identification of biological signatures of response to exposures, and to develop prediction models of the effects of environmental factors in driving disease activity and response to therapy. This research could lead to identification of biomarkers of exposures and new modalities for therapeutic intervention. Finally, hypothesis-driven mechanistic studies to understand gene-environment interactions and to validate causality of priority factors should be performed to determine how environment influences clinical outcomes.

In the version of this article initially published, a word ("neutraling") in sentence 2 of paragraph 5 is incorrect. The correct phrase is "...neutralizing properties...". The error has been corrected in the HTML and PDF version of the article.

Crohn’s disease (CD) is a progressive and destructive disease. At diagnosis, up to 1/3 of patients have a complicated phenotype defined as stricturing (B2), internal penetrating (B3) CD, or surgery. We evaluated anti-microbial antibodies and protein markers in multiple samples long before diagnosis to assess whether complicated vs. non-complicated CD at diagnosis was associated with pre-diagnostic biomarkers. Pre-diagnosis serum samples (~2, 4, and 6 years prior to diagnosis) were obtained from 200 patients with CD and 200 healthy controls (HC). Samples were tested for a panel of anti-microbial antibodies and 1129 protein markers (SomaLogic panel). A complicated CD phenotype at diagnosis was defined based on ICD-9 or CPT codes (B2, B3, or surgery). The association between each marker and complication was assessed via Cox regression. Significant markers passing a false-discovery rate of 20% were selected for different time before diagnosis (−2Y, −4Y, and -6Y). In addition, for proteomic markers, biological pathways enriched in the set of predictive markers were identified via Fisher exact test. Forty-seven subjects (24%) had a B2 (n = 36) or B3 (n = 9) phenotype or surgery (n = 2) at diagnosis. Figure 1 shows anti-microbial antibodies as well as protein markers selected at different years before diagnosis. At 6 years prior to diagnosis, the difference of anti-microbial antibody titers was already significant between complicated and non-complicated CD. Thirty protein markers, involved in the activation of immune system and/or inflammation, were associated with complicated CD. Most protein markers such as CRP, C9, and C5 were up-regulated, while few markers including SERPINA4 and c-Kit were down-regulated in patients developing complications. Pathway analysis identified higher activation of the innate immune system and complement/coagulation cascades for both 4 and 6 years before diagnosis in complicated CD vs. uncomplicated CD. The difference of those activated pathways was prominent at 6 and 4 years before diagnosis and disappeared as approaching the time of diagnosis. Most of pre-diagnostic biomarkers were increased in both complicated and non-complicated CD vs. healthy controls. Complicated CD at diagnosis is associated with higher serum levels of anti-microbial antibodies and a different profile of proteins, vs. non-complicated CD, years before diagnosis. Findings suggest that innate immune activation with involvement of complement pathways occur early in the natural history of complicated CD many years before diagnosis. Abstract DOP035 – Figure 1. Heatmap showing the association between Serologic (a) as well as SomaLogic markers (b) with complication in Crohn's disease based on Cox regression. (a) For all Serologic markers, the heatmap shows the p-value of association in the −log10 scale multiplied by the sign of the coefficient in the Cox regression. OmpC and pANCA were not found to be significantly associated to complications (p > 0.05). (b) Selected proteomic markers passing a 20% false discovery rate, which were selected in at least two of the three time points. As shown, some of the markers, including KIT, AFM, GSN, SERPINA4, and LRPP8, were found to be down-regulated in patients developing complication.

Background & aims Originally believed to be primarily a disorder of T-cell signaling, evidence shows that macrophage-lineage cells also contribute to the pathogenesis of Crohn’s disease (CD). Colony stimulating factor-1 (CSF-1) is a key regulator of the macrophage lineage, but its role in CD has not been well established. We examined transcriptional data from CD mucosa for evidence of CSF-1 pathway activation and tested JNJ-40346527 (PRV-6527), a small molecule inhibitor of CSF-1 receptor kinase (CSF-1R), for its ability to inhibit disease indices in murine colitis. Methods A CSF-1 pathway gene set was created from microarray data of human whole blood cultured ex vivo with CSF-1 and compared to a TNFα-induced gene set generated from epithelial-lineage cells. Gene set variation analysis was performed using existing Crohn’s mucosa microarray data comparing patients who either responded or failed to respond to anti-TNFα therapy. Commencing day 14 or day 21, mice with T-cell transfer colitis were treated with vehicle or JNJ-40346527 until study termination (day 42). Endpoints included colon weight/length ratios and histopathology scores, and macrophage and T cells were assessed by immunohistochemistry. Mucosal gene expression was investigated using RNAseq. Results Both the CSF-1 and the TNFα gene sets were enriched in the colonic mucosal transcriptomes of Crohn’s disease and in mouse colitis, and expression of both gene sets was highest in patients who did not respond to anti-TNFα therapy. In these patients neither set was reduced by therapy. In the mouse model, JNJ-40346527 inhibited the increase in colon weight/length ratio by ∼50%, reduced histological disease scores by ∼60%, and reduced F4/80+ mononuclear cell and CD3+ lymphocyte numbers. RNAseq analysis confirmed the CSF-1 gene set was sharply reduced in treated mice, as were gene sets enriched in “M1” inflammatory and “M0” resident macrophages and in activated T cells. Conclusions CSF-1 biology is activated in Crohn’s disease and in murine T cell transfer colitis. Inhibition of CSF-1R by JNJ-40346527 was associated with attenuated clinical disease scores and reduced inflammatory gene expression in mice. These data provide rationale for testing JNJ-40346527 (PRV-6527) in human inflammatory bowel disease.

The ability to predict response to therapy in inflammatory bowel disease (IBD) is a significant unmet need. We previously described PROgECT, a Phase 2a open-label study of patients with moderate-to-severe ulcerative colitis (UC), which prospectively validated the ability of a molecular profile score (MPS) consisting of a colonic 13-gene expression panel to predict response to TNF-antagonist therapy. Although the MPS had low specificity in predicting responders to therapy, we evaluated whether the MPS could be a useful tool in accurately identifying a subset of non-responder patients to therapy. We evaluated the sensitivity and specificity of the MPS in identifying non-responders to therapy in four independent TNF-antagonist trials (ACT1, PURSUIT-SC, PROgECT, PURSUIT-J) and an anti-IL12/23 trial (UNITI). We also characterised the gene expression and microbiome profiles of predicted non-responders by the MPS using microarray and 16S sequencing in the PROgECT cohort. We report that the MPS can accurately predict non-responders, as defined by lack of mucosal improvement, to TNF-antagonist therapy in UC in four independent clinical trials, with a high negative predictive value (NPV) of 0.78 in ACT1, 0.79 in PURSUIT-SC, 0.89 in PROgECT, and 0.73 in PURSUIT-J. In addition, the MPS could predict non-responders, as defined by lack of endoscopic response, to anti-IL12/23 therapy in Crohn’s disease (CD) with an NPV of 0.85. The predicted non-responders by MPS did not differ compared with predicted responders in baseline disease severity as measured by Mayo Score, or baseline inflammatory markers including CRP, faecal calprotectin, or faecal lactoferrin levels. Transcriptomics and microbiome analysis revealed insights into potential ways to treat this predicted non-responder population, as predicted non-responders had 268 differentially expressed genes enriched in inflammatory pathways and also demonstrated significant microbial dysbiosis. The MPS consistently predicts non-responders to therapy in IBD regardless of ethnicity or whether the therapy targeted TNF or IL12/23 pathways. Clinical parameters and inflammatory markers by themselves lack the granularity to identify this subset of non-responder patients. The MPS is the first prospectively validated predictive biomarker that can accurately identify a discrete subset of non-responder patients to two different anti-inflammatory therapies and may be valuable in identifying subsets of patients in need of treatment with alternative therapies or for patient stratification in clinical trials.

Endoscopy, histology, and biomarker measures of inflammation have limitations of sensitivity, specificity, reproducibility, and range in evaluating inflammatory bowel disease (IBD). We explored whole transcriptome gene expression to define molecular scores of gut inflammation. These scores are applicable to both Crohn’s disease (CD) and ulcerative colitis (UC), enabling more granular, continuous measures across multiple states and location of disease. We present a molecular characterisation of IBD based on the transcription profiles of 719 endoscopically defined inflamed (Inf) and 1776 non-inflamed (NInf) intestinal biopsies from 498 CD, 419 UC patients in the Mount Sinai Crohn’s and Colitis Registry (MSCCR) during endoscopy. Genes differentially expressed between Inf and NInf biopsies were used to generate a biopsy-level molecular inflammation score (MIS) via gene set variation analysis.1 MIS was strongly associated with histological biopsy scores for CD (GHAS2) and UC (Nancy Index3) and independent of inflammatory status (Inf B = 3.1, NInf B = 2.73; p > 0.05) (Figure 1A), MIS of Inf biopsies was higher than NInf within the same histological score, indicating that MIS describes a broader range of inflammation signal than histologic assessment. MIS was also associated with endoscopically defined severity (SES-CD and Mayo-endo for UC); capturing the gradient from mild, moderate, to severe disease (Figure 1B). Association of MIS with clinical disease severity was significant for Inf biopsies for continuous measures (HBI for CD B = 0.65, p < 0.01; SCCAI for UC B = 1.94, p < 0.01) and could also differentiate between HBI and SCCAI defined active and inactive subsets (UC d = 11.5, p < 0.01; CD d = 5, p < 0.01). This was not the case for NInf biopsies (Figure 1C), indicating that the clinical scores track with inflammation but not with homeostatic features of the gut. Abstract OP017 Association of MIS with histological (A), endoscopic (B) and clinical (C) IBD severity. Means (black dots) and CI (blue) We generated a transcriptionally based intestinal inflammation score in IBD patients, which provides an objective quantification of disease state in IBD-relevant tissues. MIS scores are associated with features captured by histological, endoscopic, and clinical evaluations, but do so with a greater dynamic range, and as a common metric for CD and UC. Further work will explore whether MIS may improve patients subsetting, identify sub-clinical disease, predict flares or therapeutic response. Furthermore, MIS can be used to regress the inflammation component, revealing novel non-inflammatory mechanisms. References 1. Hänzelmann S, Castelo R, Guinney J. GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinf. 2013;14:7. 2. D’Haens GR, Geboes K, Peeters M, et al. Early lesions of recurrent Crohn’s disease caused by infusion of intestinal contents in excluded ileum. Gastroenterology. 1998;114:262–7. 3. Marchal-Bressenot A, Salleron J, Boulagnon-Rombi C, et al. Development and validation of the Nancy histological index for UC. Gut, 2017;66:43–9.

Inflammatory bowel disease (IBD) is associated with immune dysregulation triggered by environmental factors, microbial dysbiosis, and genetical susceptibility. Regulatory T cells (Tregs) are critical in controlling intestinal immune homeostasis and Treg deficiencies trigger intestinal inflammation.1Brunkow M.E. et al.Nat Genet. 2001; 27: 68-73Crossref PubMed Scopus (2023) Google Scholar, 2Bennett C.L. et al.Nat Genet. 2001; 27: 20-21Crossref PubMed Scopus (2639) Google Scholar Interleukin (IL)-2 is a key cytokine controlling differentiation, survival, and function of Tregs.3Klatzmann D. et al.Nat Rev Immunol. 2015; 15: 283-294Crossref PubMed Scopus (399) Google Scholar In contrast to conventional T cells (Tcon), Tregs exhibit higher sensitivity to IL-2 due to constitutive expression of CD25, the high-affinity subunit of the IL-2 receptor.3Klatzmann D. et al.Nat Rev Immunol. 2015; 15: 283-294Crossref PubMed Scopus (399) Google Scholar Low-dose (LD) IL-2 has been reported to selectively expand Tregs and used as a therapeutic strategy in chronic graft-versus-host disease, hepatitis C virus–induced vasculitis, systemic lupus erythematosus, and Wiskott-Aldrich syndrome.4He J. et al.Nat Med. 2016; 22: 991-993Crossref PubMed Scopus (365) Google Scholar, 5Koreth J. et al.N Engl J Med. 2011; 365: 2055-2066Crossref PubMed Scopus (826) Google Scholar, 6Saadoun D. et al.N Engl J Med. 2011; 365: 2067-2077Crossref PubMed Scopus (588) Google Scholar, 7Jyonouchi S. et al.Clin Immunol. 2017; 179: 47-53Crossref PubMed Scopus (24) Google Scholar Thus, we sought to investigate LD IL-2 as an IBD therapeutic using humanized mice. Similar to other disease settings,4He J. et al.Nat Med. 2016; 22: 991-993Crossref PubMed Scopus (365) Google Scholar, 5Koreth J. et al.N Engl J Med. 2011; 365: 2055-2066Crossref PubMed Scopus (826) Google Scholar, 6Saadoun D. et al.N Engl J Med. 2011; 365: 2067-2077Crossref PubMed Scopus (588) Google Scholar, 7Jyonouchi S. et al.Clin Immunol. 2017; 179: 47-53Crossref PubMed Scopus (24) Google Scholar LD IL-2 specifically activated peripheral blood and colonic lamina propria Tregs from patients with IBD in vitro (Supplementary Figure 1). To study the efficacy of LD IL-2 in a preclinical setting, we treated NSG mice reconstituted with healthy donor peripheral blood mononuclear cells with phosphate-buffered saline (PBS) or LD IL-2 (1.0 × 104 IU/day [10K]; 5.0 × 104 IU/day [50K]) (Figure 1A). On day 5 following immune reconstitution, colitis was induced by rectal application of 2,4-dinitrobenzene sulfonic acid. Mice receiving 10K IL-2 had reduced weight loss and reduced histology scores compared with mice treated with PBS or 50K IL-2 (Figure 1B and C). Phospho-flow analysis of STAT5 confirmed that 10K IL-2 specifically activated Tregs, whereas STAT5 phosphorylation was also detected in Tcons from mice receiving 50K IL-2 (Figure 1D). Expansion of Tregs was observed in blood, spleen, and colon of mice treated with either dose of IL-2 (Figure 1E). However, reduced body weight loss upon treatment with 10K IL-2 associated with Treg expansion in the absence of significant Tcon activation suggest that a therapeutic range of LD IL-2 is critical (Figure 1C–E). To evaluate the efficacy of LD IL-2 in a fully reconstituted humanized murine system, we developed NSG mice that lack murine MHCII but express human HLA-DQ8 (NSGIIDQ8 mice). Sixteen-week-old mice reconstituted with human healthy donor CD34+ hematopoietic stem cells at birth were sensitized with 2,4,6-trinitrobenzenesulfonic acid (TNBS) and treated with 10K IL-2 daily followed by induction of colitis with TNBS rectal challenge (Figure 2A). In contrast to mice treated with PBS, mice receiving LD IL-2 exhibited significant improvement in histological disease activity with a trend in reduced weight loss (Figure 2B and C). LD IL-2 was associated with significant expansion of human Tregs in the blood and spleen but not in the mesenteric lymph nodes or colon (Figure 2D and E). Correspondingly, no difference in the frequency of FOXP3-expressing T cells was detected using RNAscope analysis of paraffin-embedded colonic lamina propria sections (Supplementary Figure 2). The expansion of peripheral Tregs and improvement in disease activity in LD IL-2-treated mice was not attributed to alterations in the human immune reconstitution between groups (Figure 2F) or frequency of effector T or natural killer cells (Supplementary Figure 3). Tregs have been reported to suppress pathogenic effector T cell function and autoimmunity through both contact-dependent and contact-independent mechanisms.8Sakaguchi S. et al.Nat Rev Immunol. 2010; 10: 490-500Crossref PubMed Scopus (1794) Google Scholar CyTOF analysis of splenic CD25+ cells showed an increased frequency of FOXP3+ T cells in mice receiving LD IL-2 with majority of expanded cells falling within the CD45RO+FOXP3+ Treg cluster (Figure 2G). The majority of FOXP3+ Tregs that expanded after LD IL-2-treatment exhibited increased expression of molecules associated with Treg activation or function (HLA-DR, CD45RO, and CTLA4) or chemokines important for trafficking and migration to sites of inflammation (CCR4 and CCR6) (Figure 2H). Our data suggest that expansion and activation of memory Tregs might be critical clinical determinants. Taken together, our study demonstrates that LD IL-2 expands Tregs and ameliorates experimental colitis in humanized mice. While these data support a rationale for LD IL-2 in IBD therapy, the safety of long-term drug administration needs further investigation to ensure continued selective activation of Tregs over Tcons. Based on these promising results, we have initiated a phase 1b/2a clinical trial investigating the safety and therapeutic efficacy of LD IL-2 in patients with moderate to severe ulcerative colitis (NCT02200445). The authors thank the Harvard Digestive Disease Center for their core support services. Download .pdf (2.13 MB) Help with pdf files Supplemental Methods Download .pdf (.12 MB) Help with pdf files Supplemental Figures 1–4