Society: AGA

Background
Aconitate decarboxylase 1 (ACOD1; also known as IRG-1) is the ubiquitous source of the metabolite itaconate, which dampens inflammasome activation by preventing HIF1α production of IL-1β and prevents TNF production through NRF2 activation. We reasoned that expression of ACOD1 is an inhibitory pathway within monocytes to curtail their inflammatory function, which could be manipulated in IBD therapy.

Methods
We performed single-cell RNA sequencing of monocytes from colonoscopic biopsies of treatment naïve IBD patients, as well as complementary analysis of DSS treated C57BL/6J (WT) mice, using single-cell protein-mRNA (Primeflow) technologies for validation. To determine the role of ACOD1 in colitis, Acod1^-/- mice were administered 2% DSS in drinking water for 6 days (acute colitis) or 4 days followed by 14 day recovery period (inflammation resolution) and assessed the colon myeloid compartment by multi-parameter flow cytometry. This was complemented with competitive bone marrow chimeric mice to assess the intrinsic role of ACOD1 controlling monocyte survival and proinflammatory status. Transcriptional profiling (Nanostring) was used to determine the molecular pathways influenced by ACOD1.

Results
We found that ACOD1/Acod1 represented an evolutionarily conserved of feature of inflammation-associated monocytes, with expression present in mucosal but not blood monocytes. Acod1^-/- mice demonstrated increased susceptibility to acute intestinal injury with DSS compared to co-housed WT controls, characterised by increased weight loss and colon shortening with significant accumulation of colonic neutrophils and CD64⁺Ly6C⁺ MHC-II^+/–monocytes. Consistent with Acod1 limiting inflammasome activation, Acod1^-/- monocytes demonstrated significantly greater IL-1b producing capabilities than WT controls.

To assess Acod1 deficient and sufficient haemopoietic cells in in the same environment, we generated competitive bone marrow chimeric mice. Acod1 deficiency conferred a significant advantage to monocyte recruitment to the colon, characterised by a 10fold increase in Ly6C⁺ MHC-II^–colon monocytes in colitis.

Transcriptional profiling of WT and Acod1^-/- colonic monocytes revealed significantly greater expression of molecules required for monocyte extravasation, including Sell (CD62L), Icam1, Ccr2 and Itgal (CD11a). In addition, expression of genes indicative of macrophage maturation such as Itgax (CD11c), Cd163 and Cx3cr1 were lower in the context of Acod1 deficiency, suggesting that normal maturation of these cells may be perturbed by loss of Acod1.

Conclusion
ACOD1/Acod1 represents an evolutionarily conserved, monocyte specific immuno-metabolic pathway that acts to limit monocyte recruitment, fate and pro-inflammatory function within the inflamed intestine.

% change in starting weight of Acod1-/- or C57BL6/JCrl (WT) mice after 6 days of 2% DSS (A) or 4 days plus 14 days normal drinking water (B). (<b>C</b>) Flow cytometry plots of colon lamina propria CD64<sup>+</sup> cells isolated from (<b>B</b>) with relative proportions of Ly6C<sup>+</sup> MHC-II<sup>-</sup> monocytes (P1) Ly6C<sup>+</sup> MHC-II<sup>+</sup> monocytes (P2) and Ly6C<sup>-</sup> MHC-II<sup>+</sup> macrophages (P3/4) and the proportion of IL-1β expressing P1 cells observed after 2hours incubation with 1ug/ml Momensin (<b>D</b>).Congenic CD45.1/CD45.2<sup>+</sup> mice were lethally irradiated and reconstitiuted with WT CD45.1 or <i>Acod1</i><sup>-/-</sup> CD45.2 BM, before treatment with 6 days 2% of DSS or normal drinking water. (<b>E</b>) Chimerism of colon CD64<sup>+</sup> cells relative to blood monocytes from gating strategy in (<b>C</b>). (<b>F</b>) CD45.1 and CD45.2 P1 and P2 moncytes from DSS treated mice were purified by flow cytometry, RNA extracted and profiled by Nanostring.

% change in starting weight of Acod1-/- or C57BL6/JCrl (WT) mice after 6 days of 2% DSS (A) or 4 days plus 14 days normal drinking water (B). (C) Flow cytometry plots of colon lamina propria CD64⁺ cells isolated from (B) with relative proportions of Ly6C⁺ MHC-II^- monocytes (P1) Ly6C⁺ MHC-II⁺ monocytes (P2) and Ly6C^- MHC-II⁺ macrophages (P3/4) and the proportion of IL-1β expressing P1 cells observed after 2hours incubation with 1ug/ml Momensin (D).Congenic CD45.1/CD45.2⁺ mice were lethally irradiated and reconstitiuted with WT CD45.1 or Acod1^-/- CD45.2 BM, before treatment with 6 days 2% of DSS or normal drinking water. (E) Chimerism of colon CD64⁺ cells relative to blood monocytes from gating strategy in (C). (F) CD45.1 and CD45.2 P1 and P2 moncytes from DSS treated mice were purified by flow cytometry, RNA extracted and profiled by Nanostring.

Background: Better biomarkers for prediction of ulcerative colitis (UC) development and prognostication are needed. Anti-integrin αvβ6 autoantibodies (anti-αvβ6) have recently been described in adult and pediatric UC patients. Here, we tested for the presence of anti-αvβ6 autoantibodies in the pre-clinical phase of UC and studied their association with disease-related outcomes after diagnosis.

Methods:
Anti-αvβ6 autoantibodies were measured in 4 longitudinal serum samples collected from 82 subjects who later developed UC and 82 matched controls (HC) from a Department of Defense pre-clinical cohort (PREDICTS). The longitudinal samples were collected at the following timepoints: Sample A at the time of UC diagnosis, Sample B at a median of 2 years before Sample A, Sample C at a median of 4 years before Sample A and Sample D at a median of 10 years before Sample A. In a distinct, external validation cohort (GEM), we tested 12 pre-UC subjects and 49 matched controls. Further, anti-αvβ6 were measured in 2 incident UC cohorts (COMPASS n=55 and OSCCAR n=104) and associations between anti-αvβ6 and UC-related outcomes were defined using Cox proportional-hazards model.

Results:
Anti-αvβ6 autoantibodies were significantly higher among individuals who developed UC compared to controls up to 10 years before diagnosis in PREDICTS (Figure 1A), and this was validated in the GEM cohort in which samples were collected at a median of 4.2 years before diagnosis (Figure 1B). During the pre-clinical phase (Sample D to A), the proportion of anti-αvβ6 seropositive subjects increased from 12.2% (Sample D) to 20.7% (Sample C) to 30.5% (Sample B) to 52.4% (Sample A) of subjects who developed UC (Chi-square test for trend p<0.0001), compared with a mean of 2.7% in HC group across the 4 timepoints in the PREDICTS cohort (Figure 1C). Anti-αvβ6 predicted UC development with an AUC of at least 0.8 up to 10 years before diagnosis (Figure 1D). Finally, high levels of anti-αvβ6 were associated with a composite of adverse UC-outcomes including hospitalization, disease extension, colectomy, systemic steroid use and/or escalation to biologic therapy in both cohorts of recently diagnosed UC patients (Figure 1E).

Conclusion:
Anti-integrin αvβ6 autoantibodies precede the clinical diagnosis of UC by up to 10 years and are associated with adverse UC-related outcomes. The presence of anti-αvβ6 autoantibodies before UC diagnosis suggests that a pre-clinical phase, potentially amenable to therapeutic intervention may indeed predate UC diagnosis.

Background: The presence of inflammatory regulatory T cells (Tregs) in inflammatory bowel disease (IBD) patients is one of the hallmarks of resistance to therapies, such as anti-TNF biologics. Yet, the mechanisms leading to the derivation of inflammatory Tregs and their contribution to intestinal tissue damage have remained elusive. Our central hypothesis is that the inflamed gut microenvironment, which is rich in nutrients and cytokines, induces inflammatory Treg development through metabolic alterations that result in Treg defect and refractory IBD. Methods: Human Tregs derived from healthy PBMC donors were characterized by microarray/RT-qPCR, immunoblotting, fluorescent-activated cell sorting (FACS), mass cytometry by time of flight (CyTOF), metabolomics and electron microscopic (EM) and confocal imaging. IL21-induced transcriptome was validated in single-cell RNA sequencing (scRNA-seq) dataset from Crohn’s disease intestinal lesions of anti-TNF responders vs. non-responders. The in vivo relevance of wildtype (WT) and IL21 receptor-deleted (Il21r^-/-) Tregs were examined in two models of CD4⁺CD45Rb^high T cell-induced colitis in Rag1^-/- mice. Results: Through in vitro cytokine challenge of human Tregs and subsequent protein analysis by FACS and CyTOF, we identified IL21 as the unique cue capable of inducing Tregs co-expressing TNF, IFNγ and IL17. Through integration of transcriptomic and metabolomic data, we found significant upregulation of mRNA and metabolites associated with metabolic processes such as glycolysis (conversion of glucose-derived pyruvate to lactate), and amino acid, nucleotide and lipid metabolism. Of note, IL21-induced metabolic transcripts were enriched in Treg cluster within scRNA-seq dataset derived from intestinal lesions of anti-TNF non-responders, implying that IL21-induced metabolism in Tregs is a feature of IBD. Addressing how IL21 induces these metabolic alterations, confocal imaging, EM, and real-time bioenergetic analysis by Seahorse analyzer revealed disruption to mitochondria-ER interaction, coinciding with glycogen synthase kinase 3β (GSK3β) activation and glycolytic switch rather than pyruvate-mediated mitochondrial oxidative phosphorylation (OXPHOS). Congruently, pharmacologic inhibition of GSK3β or supplementation with membrane-permeable methyl pyruvate rescued IL21-induced metabolic imbalance and inflammatory phenotype. Finally, in contrast to wildtype (WT) Tregs, adoptive transfer of murine Il21r^-/- Tregs efficiently prevented and rescued CD4⁺CD45Rb^high T cell-induced colitis, as evidenced by the reduction in Mouse Colon Histology Index and serum inflammatory markers (TNF, IFNγ and IL6). Conclusions: IL21-induced metabolic dysfunction can incapacitate Treg immunosuppressive function. Desensitizing Tregs to IL21 through various therapeutic approaches can improve Tregs’ capacity in alleviating IBD.

Background
The serum metabolome contains a plethora of endogenously produced and environmentally absorbed biomarkers that could serve as Crohn’s disease(CD) markers. However, most metabolomic studies have used post-diagnosis samples, which may be a response to CD disease activity as opposed to a predictor of disease evolution. Moreover, the association between pre-diagnosis CD risk metabolites and other biomarkers remains unknown. We aimed to assess serum metabolite alterations in the pre-clinical phase of CD and assess the correlation of these alterations with other known risk factors
Methods
In a cohort of prospectively followed healthy first-degree relatives(FDRs) enrolled in the Genetics Environment Microbial Project(CCC-GEM), subjects who later developed CD(n=78) were matched up to 1:4 with control FDRs that remained healthy(n=312). At enrollment, we measured untargeted serum metabolites using the combination of four ultrahigh performance liquid chromatography-tandem mass spectroscopy; subclinical inflammation using fecal calprotectin(FCP) assay; gut barrier function using urinary fractional excretion of lactulose to mannitol ratio(LMR) assay. We used conditional logistic regression to identify serum metabolites associated with CD onset and used marginal correlation analysis to assess the association of specific CD-associated metabolites with CD risk factors described above. Finally, random forests were performed to assess individual pre-diagnostic risk biomarkers' predictive potential for CD and their area under the curve(AUC) based on 5-fold cross validation were compared using the Delong test. Associations with false discovery rate-adjusted P values(q)<0.05 were considered significant
Results
In total, 63 of 1001 pre-diagnostic serum metabolites were associated with CD onset(1×10^-4<q<0.05), including enrichments for sphingomyelin and aspartate, and depletions for isoleucine, leucine and valine. Among the identified serum metabolites, 20 were correlated with increased FCP(2×10^-5<q<0.05) and two with elevated LMR(0.002<q<0.02). A metabolome-based model to predict CD yielded an AUC of 0.89(95% CI 0.85-0.92), which was significantly superior to the LMR-based model with an AUC of 0.48(95% CI 0.40-0.56, p<0.001) and the FCP-based model having an AUC of 0.70(95% CI 0.59-0.81, p<0.001)
Conclusion
Our study revealed alterations in serum metabolites in pre-diagnostic CD, some of these were related to gut barrier function and subclinical inflammation. The model based on serum metabolites performed better than LMR and FCP in predicting CD, suggesting metabolomic pathways other than intestinal permeability and inflammation that contribute to CD development. Identified serum metabolites may provide potential insights into disease pathogenesis, and the prediction model's good performance may offer a novel opportunity for CD prevention and early diagnosis

Charles Bernstein

University of Manitoba

Appearances