Society: AGA

Background: Emerging evidence suggest that gut microbiota influence human health such as digestive functions and risk of sepsis. We hypothesize that using a novel oral microbiome formula (SIM01) can restore dysbiosis and reduce adverse clinical events among vulnerable individuals after COVID-19 vaccination.

Methods: In a single center, double-blind, randomized, parallel arm, placebo-controlled trial, we recruited COVID-19 vaccination-naïve subjects (elderlies aged ≥ 65 years and patients with type 2 diabetes). After receiving the first dose of COVID-19 vaccine (CoronaVac or BioNTech), eligible subjects were randomized in a 1:1 ratio to receive 3 months of SIM01 (containing 20 billion CFU of probiotics per day) or placebo (2 mg of vitamin C per day). Intervention (SIM01 or placebo) was initiated within 1 week of receiving COVID-19 vaccines to minimize confounding effects of COVID-19 infection. Follow-up assessments were arranged at 1 month and 3 months after second vaccine dose. Outcomes included (1) Adverse events (AEs), defined as new symptoms/diseases which exerted unfavourable impacts on subjects; and (2) Restoration of gut dysbiosis, defined as improvement in (i) gut microbiome composition and diversity; (ii) functional potential (i.e., MetaCyc pathway abundances); and (iii) proliferation of beneficial bacteria genus (i.e., bifidobacteria, eubacterium, roseburia and other short-chain fatty acids producers). Fecal microbiome analysis using shotgun metagenomic sequencing were performed at baseline and after 3 months. The study is registered in a clinical trial registry (NCT04884776).

Results: Between April 2021 and March 2022, of 497 subjects assessed for eligibility, we recruited 453 individuals (50.3% females; mean age 67.5 years). We randomly assigned 224 subjects to SIM01 and 229 to placebo. There was no difference in baseline characteristics between the two groups (Table 1). The incidence of AE (2.9% vs. 10.6%, p=0.002) and combined AE/serious AE (2.9% vs. 12.6%, p<0.001) were significantly lower in SIM01 group than placebo group at 1 month. Rates of AE (0% vs. 2.5%, p=0.053) and combined AE/serious AE (0% vs. 3.1%, p=0.025) remained significantly lower in SIM01 than placebo group at 3 months. Overall, 70.4%, 18.5%, and 11.1% of the AEs and serious AEs at 1 month included gastrointestinal disturbance, dermatological conditions and infections, respectively. Probiotics in SIM01 successfully colonized in the gut. SIM01, but not placebo, led to a significant increase in beneficial bacteria and butyrate producers in fecal samples, enterotype shift from bacteroides to bifidobacteria and a strengthened microbial ecology network at 3 months after second vaccine dose.

Conclusion: SIM01 restored gut dysbiosis, reduced adverse clinical outcomes and strengthened gut microbiome resilience in elderly people and diabetes patients after COVID-19 vaccination.

Background: Roseburia intestinalis is an anaerobic bacterium that has been shown to prevent intestinal inflammation by producing metabolites. We aimed to study the role of R. intestinalis in regulating colorectal tumorigenesis and immunotherapy through its metabolites.
Methods: The abundance of R. intestinalis was evaluated in stool samples of human CRC patients (n=444) and healthy subjects (n=575). The carcinogenic and immunomodulatory effects of R. intestinalis were studied in Apc^Min/+ and AOM-induced CRC mouse models. The metabolites produced by R. intestinalis were evaluated by untargeted metabolomic profiling and targeted gas chromatography-mass spectrometry (GC-MS). The bio-functions of R. intestinalis cultured medium and candidate metabolites were examined in CRC cell lines (HCT116, LoVo, SW480). The change of immune landscape was evaluated by multicolor flow cytometry and immunohistochemistry staining. The effects of R. intestinalis and its metabolites on anti-PD-1 efficacy were assessed in two syngeneic mouse models of CT26 (MSI-low) and MC38 (MSI-high).
Results: R. intestinalis was significantly depleted in stool samples of CRC patients compared to healthy controls (P<0.05). R. intestinalis administration significantly inhibited colorectal tumorigenesis by reducing tumor incidence, tumor number and tumor size in Apc^Min/+ mice, and further confirmed in AOM-induced CRC mice. R. intestinalis restored gut barrier functions as indicated by improved intestinal permeability and enhanced tight junction proteins ZO-1 and Claudin-3. R. intestinalis produced the metabolite butyrate which was identified as a functional component based on the metabolomic profiling assay and bio-functional validation in CRC cell lines. R. intestinalis significantly activated cytotoxic CD8⁺ T cells in colonic tumors of AOM-induced CRC mice. This was confirmed in both MC38 and CT26 orthotopic mouse models that R. intestinalis and butyrate suppressed tumor growth by inducing cytotoxic Granzyme B⁺, IFN-γ⁺ and TNF-α⁺ CD8⁺ T cells. Moreover, R. intestinalis and butyrate significantly improved anti-PD-1 efficacy in MSI-low CT26 orthotopic mouse model. Mechanistically, butyrate binds to TLR5 receptor of CD8⁺ T cells to induce its activity through activating NF-κB signaling. TLR5 inhibitor TH1020 abolished the effects of butyrate on activating CD8⁺ T cells and inducing NF-κB signalling.
Conclusions: R. intestinalis protects against CRC by producing butyrate in mice. R. intestinalis derived butyrate improved anti-PD-1 efficacy in CRC by inducing the functional CD8⁺ T cells through binding to the receptor of CD8⁺ T cells to activate the NF-κB signalling. R. intestinalis is a potential novel adjuvant therapy to augment anti-PD-1 efficacy against CRC.

Background: The recently enhanced capacity to test individuals for respiratory viruses, such as influenza virus (IAV), respiratory syncytial virus (RSV), and SARS/Cov-2, has led to the appreciation that severities of respiratory viral infections (RVI) are highly heterogenous ranging from asymptomatic infection to severe lung pathology and/or death.
Aim: We sought to investigate the hypothesis that intestinal microbiota composition is one factor that might influence severity of RVI.
Approach: Mice born with differing microbiomes were acquired from different vivaria or by administering mice the commensal bacteria segmented filamentous bacteria (SFB), which is naturally present in some vivaria. Nasal administration of H1N1 influenza virus (IAV) or respiratory syncytial virus (RSV) served as models of RVI.
Results: We observed that colonization of the intestine by SFB, naturally acquired or exogenously administered, lastingly protected mice against influenza virus (IAV) challenge, assessed by viral titers or clinical-type indices. Such protection was independent of type I and III interferons and adaptive immunity, IL-17/IL-22, and Stat1 signaling. Instead, it is associated with, and required, maintenance of alveolar macrophages (AMs), which, in mice lacking SFB, were quickly depleted as pandemic IAV replicated, leaving lung epithelial cells unguarded and thus highly prone to infection. Transplant of AMs from SFB+ to SFB- mice recapitulated the AM-emboldened phenotype and protected mice against IAV infection.
Conclusions: Gut microbiota composition can have dramatic impacts on alveolar macrophage phenotype and, consequently, be a major determinant RVI outcomes.

Background
Intestinal crypts form a pristine gut biogeographical niche, homing the intestinal stem cells and being the closest neighbors to underlying lamina propria. Initially believed to be sterile, the crypt lumen was recently shown to inhabit a conserved microbial population. However, the identity of the crypt-associated microbiota (CAM) remains elusive. The present study is among the first to illustrate the CAM composition in health and its deviation during ulcerative colitis (UC), and the effect of faecal microbiota transplantation along with an anti-inflammatory diet (FMT-AID) on the structure of this community.

Methods
Recto-sigmoidal biopsies from controls, and from patients with mild-moderate UC before and after FMT-AID (n=26), were subjected to methacarn-fixation, followed by paraffin-embedding, sectioning and laser-capture microdissection-assisted crypt isolation. DNA isolated from micro-dissected samples was subjected to 16S rRNA gene sequencing. The bacterial presence in colonic crypts was confirmed using fluorescence in-situ hybridization(FISH). Microbiome data analysis was carried out by using QIIME2 and R packages.

Results
FISH performed using pan-bacterial probes revealed the presence of sparse microbial clusters in colonic crypts, distinct from the overlying layer of Mucosa-associated microbiota (MAM), in both controls as well as in UC(Fig.1a and 1b). While MAM is dominated by members of phyla Firmicutes (45%) and Bacteroidetes (26%),CAM is comprised predominantly of aerobic members of Actinobacteria(54%) and Proteobacteria(38%), followed by minor proportions of Firmicutes(4%), Acidobacteria and Cyanobacteria (2% each). The significant members of CAM included aerobic genera -Cutibacterium(35%),Sphingobium(13%), Paracoccus(11%), Micrococcus(3%), Lawsonella(3%), Rothia(2%), Prauserella(2%), Kocuria, Corynebacterium, Acinetobacter and Brevundimonas (1% each)(Fig.1c and 1d). Analysis of CAM diversity in controls, and in patients with UC before and after FMT-AID, showed no significant alterations in the α- and β diversity matrices(Fig.1e and 1f).CAM demonstrated UC-associated dysbiosis of specific taxa which was restored after FMT-AID(Fig.2a and 2b). These FMT-restored CAM taxa correlated negatively with disease-associated parameters - Fecal calprotectin (FCP), Simple Clinical Colitis Activity Index (SCCAI) and Ulcerative Colitis Endoscopic Index of Severity (UCEIS)(Fig.2c and 2d). The positive effects of FMT-AID further refurbished the CAM-MAM interaction networks, which were obliterated in UC.(Fig.2e)

Conclusion
A gut bacterial community, enriched in aerobic bacteria, resides in the colonic crypts, and undergoes taxa-level alterations during UC and in response to FMT.

<b>Fig.1.</b> <b>(A and B)</b> Confocal images of Fluorescent in-situ hybridisation (FISH) of pan-bacterial probes performed on methacarn fixed tissue sections derived from the recto-sigmoid biopsies from non-UC controls and from patients with mild-moderate UC. Arrows indicate the crypt-associated microbiota (CAM), and ellipses indicate mucosa-associated microbiota (MAM) <b>(C and D)</b> Krona plots highlight the CAM and MAM compositions<b>(E)</b> Comparison of α diversity indices across the three groups – Controls, Pre-FMT (UC) and Post-FMT, assessed by Wilcoxon rank-sum test. <b>(F) </b>PCA plot depicting the clustering of the three sample groups of CAM on the basis of the Bray Curtis diversity indices. <b>(G)</b> Table depicting the ANOSIM values performed on the Aitchison Distances, for CAM and MAM samples. <b>(H) </b>Core bacterial features were identified along the three sample groups of CAM

Fig.1. (A and B) Confocal images of Fluorescent in-situ hybridisation (FISH) of pan-bacterial probes performed on methacarn fixed tissue sections derived from the recto-sigmoid biopsies from non-UC controls and from patients with mild-moderate UC. Arrows indicate the crypt-associated microbiota (CAM), and ellipses indicate mucosa-associated microbiota (MAM) (C and D) Krona plots highlight the CAM and MAM compositions(E) Comparison of α diversity indices across the three groups – Controls, Pre-FMT (UC) and Post-FMT, assessed by Wilcoxon rank-sum test. (F) PCA plot depicting the clustering of the three sample groups of CAM on the basis of the Bray Curtis diversity indices. (G) Table depicting the ANOSIM values performed on the Aitchison Distances, for CAM and MAM samples. (H) Core bacterial features were identified along the three sample groups of CAM

<b>Fig.2. (A and B)</b> Differentially abundant bacterial taxa between the Controls and UC and between UC Pre-FMT and Post-FMT were performed using R package DeSeq2<b>(C)</b> Biplot ordination depicting clustering of CAM abundance data and UC-associated clinical parameters (SCCAI, UCEIS and FCP levels)<b>(D) </b>Spearman correlation plots between the abundance of three CAM taxa found to be enhanced post-FMT (<i>Methylobacterium</i>, <i>Sphingobium</i> and <i>Rothia</i>) and the disease-associated clinical parameters – SCCAI, UCEIS and FCP levels.<b> (E) </b>Inter-niche interaction networks for five CAM members in post-FMT cohort constructed using Cytoscape-based CoNet application. The node colour depicts one of the four categories (as per the legend) in which each of the interactions falls. The node sizes vary as per the abundance of the taxa. Co-existence (or positive correlation) is represented by green edges, while mutual exclusion (or negative correlations) are denoted by red edges.

Fig.2. (A and B) Differentially abundant bacterial taxa between the Controls and UC and between UC Pre-FMT and Post-FMT were performed using R package DeSeq2(C) Biplot ordination depicting clustering of CAM abundance data and UC-associated clinical parameters (SCCAI, UCEIS and FCP levels)(D) Spearman correlation plots between the abundance of three CAM taxa found to be enhanced post-FMT (Methylobacterium, Sphingobium and Rothia) and the disease-associated clinical parameters – SCCAI, UCEIS and FCP levels. (E) Inter-niche interaction networks for five CAM members in post-FMT cohort constructed using Cytoscape-based CoNet application. The node colour depicts one of the four categories (as per the legend) in which each of the interactions falls. The node sizes vary as per the abundance of the taxa. Co-existence (or positive correlation) is represented by green edges, while mutual exclusion (or negative correlations) are denoted by red edges.

Background: Microbiota dysbiosis is known to contribute to the aetiology of irritable bowel syndrome (IBS). We conducted a randomized trial to assess the efficacy of two infusions of faecal microbiota transplantation (FMT) in patients with IBS and to determine gut microbial changes associated with clinical improvement.
Methods: This randomised, double-blind, placebo-controlled study randomised 56 patients with IBS fulfilling ROME III criteria at a ratio of 1:1 to FMT or placebo. Eligible patients received frozen FMT from healthy donors or placebo (saline) via the duodenal route at baseline and 1 month. Primary outcome was a reduction in the IBS symptoms at 3 months after FMT (response). Clinical response was defined as a decrease of 50 or more points in the total IBS symptom score. Secondary outcomes were overall symptom improvement, improvement in bloating and faecal microbiota composition assessed by shotgun metagenomic sequencing at 3 months after FMT. Open-label FMT (two infusions 1 month apart) was provided to patients who received placebo.
Results: Clinical response occurred in 52.2% of patients who received FMT and 37.5% who had placebo at 3 months (p=0.31). More patients who had FMT had general symptom relief (81.5% vs 53.6%, p=0.027) and bloating relief (75% vs 30%; p=0.0028) at 3 months compared with patients who had placebo. In the open-label extension, 57.9% and 82.0% of patients who received open-label FMT achieved clinical response and relief in bloating symptoms, respectively. Faecal microbiome of patients who received FMT showed a reduction in pathogenic bacteria such as Ruminococcus gnavus and enrichment of beneficial bacteria such as Lawsonibacter asaccharolyticus at 3 months compared with baseline, whilst no change in gut microbiome was observed in the placebo arm. Functional microbiota analyses showed that hydrogen sulphide-producing pathway, which was associated with the bloating in IBS patients, significantly decreased in patients who had FMT (p< 0.05) accompanied by a reduction in the total relative abundance of related bacteria, including Escherichia coli and several Klebsiella. There was no serious adverse event related to FMT.
Conclusions: Though FMT did not lead to overall IBS symptom improvement, more patients had general symptom and bloating improvement after FMT in association with reduction in hydrogen sulphide-producing bacteria in the gut. Further studies involving subtypes of IBS (bloating predominant) and well-defined donors are needed to personalized FMT treatment based on bacterial signatures for better clinical outcomes.

<b>Figure 1: Self-reported general symptom and bloating relief rate between groups at week 12 </b><br /> (A) Self-reported general symptom relief rate between groups at week 12. (B) Self-reported bloating relief rate between groups at week 12. FMT: faecal microbiota transplantation

Figure 1: Self-reported general symptom and bloating relief rate between groups at week 12
(A) Self-reported general symptom relief rate between groups at week 12. (B) Self-reported bloating relief rate between groups at week 12. FMT: faecal microbiota transplantation

Background and Aim. The mucosal barrier established by intestinal epithelial cells prevents inappropriate immune responses to luminal content. Impairment of the intestinal barrier contributes to a wide range of diseases, including inflammatory bowel disease (IBD). We have shown that 2’-fucosyllactose (2’-FL) prevents colitis in mice. However, the mechanism remains unclear. The aim of this study was to define if the protective effect of 2’-FL on the intestinal barrier is through regulation of gut microbial metabolism.
Methods. Gut microbial metabolic pathways in non-IBD individuals (n=21) and IBD patients (n=82) were studied by HUMAnN3 and MaAsLin2 analysis of whole genome shotgun (WGS) sequencing of fecal samples. C57BL/6 mice at 6-weeks-old received 2’-FL in drinking water (1 mg/mL) for 7 (n=5) or 28 days (n=7), or water as control (n=5). Mouse fecal samples were analyzed by WGS sequencing and HUMAnN3. Bifidobacterium infantis (ATCC 15702) was cultured in RCM broth with or without 2’-FL (1%) supplementation (n=5/group); supernatants were subjected to untargeted metabolomics. Caco2 cells were pre-treated with pantothenate (20-40 ug/mL) before H₂O₂ to disrupt tight junctions.TER was measured to determine barrier function. Mouse colon epithelial cell line (YAMC) were pre-treated with pantothenate (1ug/mL) before scratch assay and remaining wound areas 6h after wounding were recorded.
Results. PCoA showed significant separation among control, 2’-FL-7D and 28D groups (P=0.003). Three genera (Bifidobacterium, Lactobacillus, and Enterorhabdus) were significantly upregulated and one (Anaerotruncus) downregulated in the 2’-FL-28D group compared to the control group (P<0.05). HUMAnN3 analysis identified 34 microbial metabolic pathways in 2’-FL 7- and/or 28-day groups that were not found in the control group, mainly involved in metabolite biosynthesis. In Bifidobacterium, 2’-FL upregulated 29 total metabolic pathways: 14 involved in amino acid biosynthesis and 8 in nucleotide biosynthesis. Notably, 3 bacterial metabolic pathways increased by 2’-FL were identified as downregulated in IBD patients (FDR<0.05), including purine salvaging and formaldehyde detoxification. Metabolomics of B. infantis culture supernatant found that 2’-FL significantly upregulated the abundance of 15 secreted metabolites compared to the control (P<0.05), including pantothenol (68-fold). Pantothenol promoted wound healing of YAMC (P=0.0347) and mitigated barrier disruption by H₂O₂ insult in Caco2 cells (P<0.05).
Conclusions. These results identify specific 2’-FL-modulated metabolite production by B. infantis which can protect the intestinal barrier and demonstrate the novel relevance of 2’-FL for mitigating the disturbance of microbial metabolic pathways in IBD patients. Thus 2’-FL may exert protective effects on intestinal integrity during IBD via modulation of microbial metabolism.

Gut bacteria- derived metabolites have important roles in maintaining intestinal homeostasis and modulating the progression of intestinal inflammatory diseases. Dysbiotic fecal samples from IBD patients exhibit higher concentrations of primary bile acids (BA), enrichment of taurine- conjugated BAs and lower levels of secondary BAs. Our ongoing studies of isolated bacterial strains from healthy donors identified strains that preferentially stimulate IL-10 but low IFNγ (high IL-10/IFNγ ratio- HI) fail to induce colitis in selectively colonized susceptible gnotobiotic mice, while low ratio (LO) strains induce moderate to severe colitis. We now explore non-immunologic mechanisms of colitis induced by the LO bacterial consortia. Hypothesis: colonic luminal accumulation of toxic BAs potentiates aggressive colitis in the LO bacterial-colonized mice.
Methods: We colonized germ-free Il10^-/- mice with selected HI, LO or mixed bacteria for 10 weeks and harvested cecal contents, feces and intestinal tissues. LO bacteria-colonized Il10^-/- mice were treated with the ileal BA transport inhibitor GSK223672 (iAsbt) for 2 weeks. Cecal and fecal BA profiles were measured by Vanquish UHPLC, colonic inflammation was blindly scored histologically, and host inflammatory genes, BA receptor and transporter gene expression were analyzed by qPCR. Human cell line, Caco-2 cells, exposed to taurocholic acid (TCA) in vitro was used to quantify mitochondrial dysfunction by specific probes, MitoSox and Mitotracker.
Results: Il10^-/- mice colonized with LO strains exhibited much higher luminal total BA pools than HI ratio mice that was dominated by taurine-conjugated BAs, primarily TCA, and de-conjugated cholic acid (CA), a primary BA (Fig. 1). The ratio of free BA to taurine-conjugated BAs were higher in the LO than the HI bacterial- colonized mice, while the reverse was true for the secondary BA, deoxycholic acid (DCA). Blocking ileal BA absorption with iAsbt increased the cecal BA pool and concentrations of taurine-conjugated and primary BAs, but did not increase the secondary BAs DCA and LCA, leading to even higher CA/DCA and CDCA/LCA ratios. iAsbt treatment of LO strains colonized mice enhanced colitis measured by histology scores and cecal inflammatory cytokines. iAsbt treated mice had increased colonic expression of the pro-inflammatory genes and membrane receptor for conjugated BAs, sphingosine-1-P-receptor-2 (S1PR2). Exposure of Caco-2 to high dose (500uM) TCA, similar to levels found in the cecal lumen of LO strains colonized mice in vivo, induced mitochondrial dysfunction (Fig.2).
Conclusion: Luminal toxic BA accumulation after colonization by LO ratio bacteria adversely affects mucosal homeostasis and potentiates experimental colitis, laying the foundation for diverse novel therapeutic approaches to prevent and reverse intestinal inflammation.

Jeremy Herzog

Appearances