Novel antiobestic and antidiabetic probiotics

Abstract

To provide novel antiobestic and antidiabetic probiotics.SOLUTION: A composition includes a bacterium of the genus Alistipes for prophylaxis and / or treatment of type 2 diabetes or obesity. A composition includes a bacterium of the genus Dorea or a portion of genomic DNA of the bacterium for detection of type 2 diabetes or obesity.SELECTED DRAWING: None

Description

[Technical Field]

[0001] The present invention relates to novel anti-obesity and anti-diabetic probiotics. More specifically, the present invention relates to a composition comprising bacteria of the genus Aristipes for the prevention and / or treatment of type 2 diabetes or obesity, or a composition comprising bacteria of the genus Drea or a portion of the genomic DNA of said bacteria for the detection of type 2 diabetes or obesity. [Background technology]

[0002] Insulin resistance (IR) is a major pathophysiological feature of metabolic syndrome (MetS) and type 2 diabetes (T2D). Numerous studies have shown that IR is caused by long-term chronic mild inflammation and oxidative stress due to nutritional excess, and that IR itself is a risk factor for serious complications such as atherosclerosis (ASCVD). Meanwhile, the gut microbiota has been suggested to be involved in the pathology of obesity and IR in humans and mice (Non-Patent Literature 1). In particular, recent advances in next-generation sequencing technology have made shotgun metagenomic sequencing possible, leading to a better understanding of the relationship between gut microbiota, including unculturable strains, and populations (Non-Patent Literature 2).

[0003] It is believed that by identifying and controlling bacteria involved in intestinal glucose metabolism in individuals with type 2 diabetes and obesity, it may be possible to regulate the production of dietary metabolites and, as a result, suppress the absorption of nutrients by the host. Among the human gut bacteria associated with metabolic abnormalities such as obesity, the genera Drea and Aristipes, which are dominant in the human gut, have been reported to be associated with the human gut metabolic state (Non-Patent Literature 3). However, meta-analyses have shown that probiotics reported to date have limited effects on metabolic diseases such as type 2 diabetes and obesity (Non-Patent Literature 4). One reason for this is that the mechanism by which the metabolism of human gut bacteria affects host diseases has not been demonstrated. [Prior art documents] [Non-patent literature]

[0004] [Non-Patent Document 1] J. Qin et al., A metagenome-wide association study of gut microbiota in type 2 diabetes. Nature. 490, 55-60 (2012). [Non-Patent Document 2] LB Thingholm et al., Obese Individuals with and without Type 2 Diabetes Show Different Gut Microbial Functional Capacity and Composition. Cell Host Microbe. 26, 252-264.e10 (2019). [Non-Patent Document 3] BD Piening, et al., Integrative Personal Omics Profiles during Periods of Weight Gain and Loss. Cell Syst. 6, 157-170.e8 (2018). [Non-Patent Document 4] Tao YW et al., Effects of probiotics on type II diabetes mellitus: a meta-analysis. J Transl Med. 2020 Jan 17;18(1):30. [Overview of the project] [Problems that the invention aims to solve]

[0005] In recent years, the role of gut microbiota in the metabolism of major nutrients, namely amino acids, fats, and carbohydrates, has been highlighted, as these may influence the pathophysiology of diabetes, obesity, and cardio-metabolic diseases. For example, amino acids and their derivatives, particularly branched-chain amino acids (BCAAs), have been shown to exacerbate IR and visceral steatosis in humans, and several reports suggest this is partly due to gut microbiota. Similarly, the production of trimethylamine, a lipid metabolite that promotes ASCVD in mice and humans, is mediated by gut microbial enzymes. These studies demonstrate the importance of gut microbiota-derived metabolites in the pathophysiology of the host. Carbohydrate metabolism in symbionts has been suggested to contribute up to 10% of the host's total energy intake and is related to the pathogenesis of obesity and diabetes. However, the mechanistic involvement has remained unclear due to limited evidence from metagenomic or small-scale metabolome studies. Given that microbial metabolites are direct and important factors influencing human pathology, multi-platform metabolomics technologies should be applied to host-derived samples, such as fecal samples as well as plasma samples, which may complement metagenomic information. The objective of this invention is to elucidate novel host-microbe interactions involved in metabolic diseases and microbial carbohydrate metabolites through large-scale studies of human samples performing omics analysis including comprehensive metabolite analysis, and to provide novel anti-obesity and anti-diabetic probiotics. [Means for solving the problem]

[0006] To comprehensively investigate the role of the gut microbiota in intravascular respiration (IR), the inventors conducted a large-scale human sample study applying a multi-omics platform integrating metabolomics, metagenomics, and transcriptomics to a cohort of healthy individuals newly diagnosed as prediabetic. Specifically, they identified candidate gut bacteria involved in human metabolic diseases and the metabolites they produce. Plasma and fecal samples were collected from 306 individuals, including healthy individuals, obese individuals, and prediabetic individuals (those with prediabetes), and gut bacteria and metabolites were comprehensively analyzed. In conducting research using this analytical method, the inventors demonstrated through human sample analysis that intestinal glucose metabolism is involved in the underlying pathological conditions of type 2 diabetes and obesity, such as insulin resistance and metabolic syndrome, and identified the genera Drea, Alistipes, and Bacteroides as bacteria involved in this metabolism. Furthermore, we discovered that administering bacteria of the genus Aristipes, which were reduced in insulin-resistant subjects, to mice fed a high-fat diet improved insulin resistance, thus completing the present invention.

[0007] In other words, the present invention is as follows: [1] A composition comprising bacteria of the genus Aristipes and / or Bacteroides for the prevention and / or treatment of type 2 diabetes or obesity. [2] The composition according to [1], wherein the Alistipes bacterium is Alistipes indistinctus or Alistipes finegoldii. [3] The composition according to [1], wherein the Bacteroides bacterium is Bacteroides thetaiotaomicron. [4] The composition according to [1], [2], or [3], wherein the composition is orally administered. [5] The composition according to [3], wherein the composition is a food or a food supplement. [6] A composition comprising a bacterium of the genus Drea or a portion of the genomic DNA of said bacterium for the detection of type 2 diabetes or obesity. [7] The bacteria undergo the following steps: (i) A process of collecting samples including feces from healthy individuals and subjects, (ii) A step in which the gut microbiota and bacterial metabolites collected in step (i) are profiled and correlation analysis is performed. (iii) A step in which bacteria showing a significant difference as a result of the correlation analysis of step (ii) are determined to be bacteria that are effective against the host. A composition according to any one of [1] to [5], wherein the bacteria are identified by large-scale studies of human samples containing [the specified bacteria]. [8] The large-scale human trial further (iv) collects samples containing plasma and performs correlation analysis with plasma metabolites, determining that bacteria showing a significant difference are bacteria that are effective against the host. The composition according to [7], comprising: [9] The composition according to [7] or [8], wherein the subject is obese and / or has impaired glucose tolerance.

[10] The composition according to any one of [7] to [9], wherein the bacterial metabolite is a carbohydrate metabolite. [Effects of the Invention]

[0008] The present invention provides novel anti-obesity and anti-diabetic probiotics. In one embodiment, a composition comprising Aristipes bacteria for the prevention and / or treatment of type 2 diabetes or obesity is provided. Unlike the prior art, the present invention uses bacteria identified from large-scale human sample studies and bacteria that are metabolized in the human gut. Therefore, unlike past probiotics, it enables the provision of new treatment and prevention strategies that are more based on the mechanisms of metabolic disorder onset and exacerbation. [Brief explanation of the drawing]

[0009] [Figure 1A]Overview of multi-omics analysis. Individuals who had not received a prior diagnosis of diabetes, diabetes drug treatment, or intestinal disease were included in this study (n = 306). The main clinical phenotypes were insulin resistance (IR) and metabolic syndrome (MetS). To evaluate the host-microbe relationship, clinical, plasma metabolome, transcriptome of peripheral blood mononuclear cells (PBMCs), and plasma cytokine data were collected as host factors, and 16S rRNA gene sequencing, metagenome, and fecal metabolome data were collected as microbial factors. The numerical values described in the figure indicate the elements detected in the omics data after quality filtering. [Figure 1B] Overview of multi-omics analysis. Workflow of this multi-omics analysis. To identify the microbial-metabolite relationships associated with metabolic phenotypes, first, the discriminative properties (signatures) of metabolites related to the phenotypes were analyzed by clustering the metabolites according to their respective correlations. 16S rRNA gene sequencing and metagenome datasets were used to determine the microbial signatures and analyze their associations with metabolites. To gain insights into the microbe-host relationship, the associations between fecal metabolites / microbes and host plasma metabolites, cytokines, and gene expression of peripheral blood mononuclear cells (PBMCs) were analyzed. Finally, bacterial culture experiments and animal experiments were conducted to verify the effects of candidate metabolites / microbes on metabolic phenotypes. [Figure 2A] Carbohydrate metabolites in feces clearly change in IR. Heatmap of hierarchical clustering showing the associations between clusters of fecal metabolites (CAGs) and clinical phenotypes and markers. The results of significant metabolite clusters in the comparison of insulin sensitivity (IS) vs insulin resistance (IR) and non-MetS vs MetS are described in the left column (yellow or sky blue). Also, the results of partial Spearman correlation corrected by age and sex are described in the heatmap. The category names of CAGs were determined based on the most abundant metabolites in the CAGs. [Figure 2B]Carbohydrate metabolites in feces clearly change with IR. Correlation between IR and the fecal concentration of glucose. The molar concentration (nmol / mg) of fecal metabolites for each participant was plotted in ascending order of the IR values. The color of the dots indicates IS (blue), intermediate (yellow), and IR (red). The Spearman coefficient (ρ) and p-value are described. The line and the gray zone indicate linear regression and the 95% confidence interval. [Figure 2C] Carbohydrate metabolites in feces clearly change with IR. Correlation between IR and the fecal concentration of fructose. The molar concentration (nmol / mg) of fecal metabolites for each participant was plotted in ascending order of the IR values. The color of the dots indicates IS (blue), intermediate (yellow), and IR (red). The Spearman coefficient (ρ) and p-value are described. The line and the gray zone indicate linear regression and the 95% confidence interval. [Figure 2D] Carbohydrate metabolites in feces clearly change with IR. Correlation between IR and the fecal concentration of galactose. The molar concentration (nmol / mg) of fecal metabolites for each participant was plotted in ascending order of the IR values. The color of the dots indicates IS (blue), intermediate (yellow), and IR (red). The Spearman coefficient (ρ) and p-value are described. The line and the gray zone indicate linear regression and the 95% confidence interval. [Figure 2E] Carbohydrate metabolites in feces clearly change with IR. Correlation between IR and the fecal concentration of xylose. The molar concentration (nmol / mg) of fecal metabolites for each participant was plotted in ascending order of the IR values. The color of the dots indicates IS (blue), intermediate (yellow), and IR (red). The Spearman coefficient (ρ) and p-value are described. The line and the gray zone indicate linear regression and the 95% confidence interval. [Figure 2F] Carbohydrate metabolites in feces clearly change with IR. Correlation between IR and the fecal concentration of maltose. The molar concentration (nmol / mg) of fecal metabolites for each participant was plotted in ascending order of the IR values. The color of the dots indicates IS (blue), intermediate (yellow), and IR (red). The Spearman coefficient (ρ) and p-value are described. The line and the gray zone indicate linear regression and the 95% confidence interval. [Figure 2G]Fecal carbohydrate metabolites change clearly with IR. Correlation between IR and fecal sucrose concentration. Fecal metabolite molar concentrations (nmol / mg) for each participant are plotted in descending order of IR value. The dot colors indicate IS (blue), intermediate (yellow), and IR (red). Spearman coefficient (ρ) and p-value are indicated. Lines and gray zones indicate linear regression and 95% confidence intervals. [Figure 3A] IR-related fecal metabolites are associated with altered gut microbiota and microbial genetic function. A hierarchical clustering heatmap shows the abundance of gut bacteria at the genus level among study participants. Gut bacteria were classified into four clusters based on their correlations, and study participants were further clustered into three groups, A to C, according to their microbial profiles. The proportion of individuals with insulin sensitivity (IS), intermediate, and insulin resistance (IR) is shown in a pie chart above the heatmap. [Figure 3B] IR-related fecal metabolites are associated with altered gut microbiota and microbial genetic function. HOMA-IR, triglyceride (TG), HDL cholesterol (HDL-C), and adiponectin levels between clusters A, B, and C based on microbial profiles. [Figure 3C] IR-related fecal metabolites are associated with altered gut microbiota and microbial genetic function. The study examined water-soluble and gut-bacteria-associated lipid metabolites and the microbial metabolite network of gut bacteria. Spearman correlations between genus-level gut bacterial abundance and metabolites were calculated, demonstrating significant positive correlations. [Figure 3D] IR-related fecal metabolites are associated with altered gut microbiota and microbial genetic function. A hierarchical clustering heatmap shows the associations between the KEGG pathway, fecal carbohydrates, and metabolic markers. KEGG ortholog genes significantly correlated with fecal carbohydrates associated with IR and metabolic syndrome were analyzed as enrichments in the KEGG pathway. The top 20 pathways positively or negatively associated with fecal carbohydrates are shown. The correlations between KEGG ortholog genes in these key pathways and HOMA-IR, TG, adiponectin, and BMI were analyzed, and enrichments were similarly analyzed and illustrated in the left column. [Figure 3E] IR-related fecal metabolites are associated with altered gut microbiota and microbial genetic function. This is linked to the gene levels related to the KEGG pathway concerning the phosphotransferase system, as well as the relative abundances of square-root transformed IS and IR-related genera. The line shows a generalized linear model using Poisson regression. [Figure 3F] IR-related fecal metabolites are associated with altered gut microbiota and microbial genetic function. This is linked to the gene levels associated with the KEGG pathway related to galactose metabolism, as well as the relative abundances of IS and IR-related genera after square root transformation. The line shows a generalized linear model using Poisson regression. [Figure 3G] IR-related fecal metabolites are associated with altered gut microbiota and microbial genetic function. This is linked to the gene levels related to the KEGG pathway in the citric acid cycle, as well as the relative abundances of square-root transformed IS and IR-related genera. The line shows a generalized linear model using Poisson regression. [Figure 4A] IS and IR-related microorganisms characteristically alter carbohydrate metabolites in culture supernatants. Principal component analysis was performed on metabolites from culture supernatants inoculated with Aristipes finegordii (AF), Aristipes indistinctus (AI), Dreah formisigenens (DF), Dreah longicatena (DL), and PBS (negative control). The resulting plots are shown. [Figure 4B] IS and IR-related microorganisms characteristically alter carbohydrate metabolites in the culture supernatant. The concentration of the disaccharide maltose in the culture supernatant is shown. Error bars indicate standard deviation (SD). Two-way ANOVA was used for testing. [Figure 4C] IS and IR-related microorganisms characteristically alter carbohydrate metabolites in the culture supernatant. The concentration of the disaccharide lactose in the culture supernatant is shown. Error bars indicate standard deviation. Two-way ANOVA was used for testing. [Figure 4D] IS and IR-related microorganisms characteristically alter carbohydrate metabolites in the culture supernatant. The concentration of the disaccharide trehalose in the culture supernatant is shown. Error bars indicate standard deviation. Two-way ANOVA was used for testing. [Figure 4E]IS and IR-related microorganisms characteristically alter carbohydrate metabolites in the culture supernatant. The concentration of the disaccharide cellobiose in the culture supernatant is shown. Error bars indicate standard deviation. Two-way ANOVA was used for testing. [Figure 4F] IS and IR-related microorganisms characteristically alter carbohydrate metabolites in the culture supernatant. The concentration of the monosaccharide glucose in the culture supernatant is shown. Error bars indicate standard deviation. Two-way ANOVA was used for testing. [Figure 4G] IS and IR-related microorganisms characteristically alter carbohydrate metabolites in the culture supernatant. The concentration of the monosaccharide mannose in the culture supernatant is shown. Error bars indicate standard deviation. Two-way ANOVA was used for testing. [Figure 4H] IS and IR-related microorganisms characteristically alter carbohydrate metabolites in the culture supernatant. The concentration of the monosaccharide sorbose in the culture supernatant is shown. Error bars indicate standard deviation. Two-way ANOVA was used for testing. [Figure 4I] IS and IR-related microorganisms characteristically alter carbohydrate metabolites in the culture supernatant. Concentration of the monosaccharide ribulose in the culture supernatant. Error bars indicate standard deviation. Tested by two-way ANOVA [(B) to (M)]. [Figure 4J] IS and IR-related microorganisms characteristically alter carbohydrate metabolites in the culture supernatant. The concentration of succinate, a citric acid cycle intermediate, in the culture supernatant is shown. Error bars indicate standard deviation. Two-way ANOVA was used for testing. [Figure 4K] IS and IR-related microorganisms characteristically alter carbohydrate metabolites in the culture supernatant. The concentration of malic acid, a citric acid cycle intermediate, in the culture supernatant is shown. Error bars indicate standard deviation. Two-way ANOVA was used for testing. [Figure 4L] IS and IR-related microorganisms characteristically alter carbohydrate metabolites in the culture supernatant. The concentration of fumarate, a citric acid cycle intermediate, in the culture supernatant is shown. Error bars indicate standard deviation (SD). Two-way ANOVA was used for testing. [Figure 4M] IS and IR-related microorganisms characteristically alter carbohydrate metabolites in the culture supernatant. The concentration of the citric acid cycle intermediate 2-ketoglutarate in the culture supernatant is shown. Error bars indicate standard deviation. Two-way ANOVA was used for testing. [Figure 5A]IS-related microorganisms exhibit clear metabolic effects in IR mouse models. Body weight change from baseline in high-fat diet (HFD) induced obesity and diabetes mice after administration of Aristipes indistinctus (AI) or PBS (bacterial solvent) as representative IS-related microorganisms. Error bars indicate SD. Two-way ANOVA. [Figure 5B] IS-related microorganisms show clear metabolic effects in IR mouse models. Blood glucose levels in an insulin loading test 3-4 weeks after the start of administration of bacteria (Aristipes indistinctus (AI), Aristipes finegordi (AF), Bacteroides citiotaomicron (BT), or Vehicle (bacterial solvent)). Insulin was administered intraperitoneally at a dose of 0.85 U / kg after 5 hours of fasting. Representative data from three independent experiments. Error bars indicate SD. * p < 0.05, ** p < 0.01, *** p < 0.001; 2-way repeated measures ANOVA. [Figure 5C] IS-related microorganisms show clear metabolic effects in the IR mouse model. Area under the curve (AUC) 3 weeks after the start of administration of bacteria (Aristipes indistinctus (AI), Aristipes finegordi (AF), Bacteroides citiotaomicron (BT), or Vehicle (bacterial solvent)). Insulin was administered intraperitoneally at a dose of 0.85 U / kg after 5 hours of fasting. Representative data from three independent experiments. Error bars indicate SD. *p < 0.05, ***p < 0.001; One-way ANOVA and Tukey's test. [Figure 5D] IS-related microorganisms show clear metabolic effects in IR mouse models. Fructose concentrations in the cecum of mice administered AI (sky blue) or PBS (gray). Error bars indicate standard deviation. * p < 0.05; Wilcoxon rank-sum test. [Figure 5E] IS-related microorganisms show clear metabolic effects in IR mouse models. Cecal mannose concentrations in mice administered AI (sky blue) or PBS (gray). Error bars indicate SD. * p < 0.05. [Figure 5F]IS-related microorganisms show clear metabolic effects in IR mouse models. Correlation between AUC in insulin loading tests and cecal fructose concentration in mice administered AI (sky blue) or PBS (gray). Representative data from two independent experiments. Spearman coefficient (ρ) and p-values ​​are indicated. Lines and gray zones indicate 95% confidence intervals for linear regression. [Figure 5G] IS-related microorganisms show clear metabolic effects in IR mouse models. Correlation between AUC in insulin loading tests and cecal mannose concentration in mice administered AI (sky blue) or PBS (gray). Representative data from two independent experiments. Spearman coefficient (ρ) and p-values ​​are indicated. Lines and gray zones indicate 95% confidence intervals for linear regression. [Figure 5H] IS-related microorganisms show clear metabolic effects in IR mouse models. Correlation between AUC in insulin loading tests and glucose cecal concentration in mice administered AI (sky blue) or PBS (gray). Representative data from two independent experiments. Spearman coefficient (ρ) and p-values ​​are indicated. Lines and gray zones indicate 95% confidence intervals for linear regression. [Figure 5I] IS-related microorganisms show clear metabolic effects in IR mouse models. Correlation between AUC in insulin loading tests and cecal concentration of tagatose in mice administered AI (sky blue) or PBS (gray). Spearman coefficient (ρ) and p-values ​​are shown. The lines and gray zones indicate the 95% confidence intervals for linear regression. [Figure 5J] IS-related microorganisms show clear metabolic effects in IR mouse models. Correlation between AUC in insulin loading tests and psicose cecal concentration in mice administered AI (sky blue) or PBS (gray). Spearman coefficient (ρ) and p-values ​​are shown. The lines and gray zones indicate the 95% confidence intervals for linear regression. [Figure 5K]IS-related microorganisms show clear metabolic effects in the IR mouse model. Postprandial blood glucose levels 4 weeks after the start of administration of Aristipes indistinctus (AI), Aristipes finegordi (AF), Bacteroides cetyoteomicron (BT), Bacteroides xylanisorbens (BX), Parabacteroides meldae (PM), Clostridium spirome (CS), Faecalibacterium prausnitzii (FP), or Vehicle (bacterial solvent). Error bars indicate SD. Representative data from two independent experiments. Data counts are 12 for Vehicle, 10 for AI and AF, and 5 for others. Kruskal-Wallis and Dunn tests were performed. *p < 0.05, **p < 0.01, ***p < 0.001. [Figure 6A] IS-related microorganisms also improve insulin resistance at the signaling pathway level in IR mouse models. Western blots of Akt protein phosphorylated at position 473 (S473) (p-Akt) and whole Akt protein in the liver (top) and epididymal fat (eWAT, bottom) of mice administered with Aristipes indictinctus (AI), Aristipes finegordii (AF), or Vehicle (bacterial solvent), 5 minutes after insulin injection. [Figure 6B] IS-related microorganisms improve insulin resistance at the signaling pathway level in IR mouse models. The graph shows densitometry results of bands detected by antibodies specific to phosphorylation at S473 (p-Akt) and bands detected by antibodies specific to the entire Akt protein in the liver of mice administered with Aristipes indistinctus (AI), Aristipes finegordi (AF), or Vehicle (bacterial solvent), 5 minutes after insulin injection. The vertical axis represents the relative value of the normalized p-Akt band densitometry measurement to the Akt band densitometry measurement. The number of data points is 4 for AI, and 5 each for Vehicle and AF. [Figure 6C]IS-related microorganisms improve insulin resistance at the signaling pathway level in IR mouse models. The graph shows densitometry results of bands detected by antibodies specific to phosphorylated Akt protein at S473 and bands detected by antibodies specific to all Akt protein in epididymal fat, 5 minutes after insulin injection into mice administered with Aristipes indistinctus (AI), Aristipes finegordii (AF), or Vehicle (bacterial solvent). The vertical axis represents the relative value of the normalized p-Akt band densitometry measurement to the Akt band densitometry measurement. The number of data points is 4 for AI, and 5 each for Vehicle and AF. [Figure 7A] IS-related microorganisms increase serum components with insulin resistance-improving properties in IR mouse models. HDL-cholesterol (HDL-C) levels in the serum of mice administered with Aristipes indistinctus (AI) or Vehicle (a solvent for the bacterial cells). Data set count: 5. [Figure 7B] IS-related microorganisms reduce serum components that worsen insulin resistance in IR mouse models. Triglyceride (TG) levels in the serum of mice administered with Aristipes indistinctus (AI) or Vehicle (a solvent for the bacterial cells). Data set count: 5. [Figure 7C] IS-related microorganisms increase serum components with insulin resistance-improving properties in IR mouse models. Adiponectin levels in the serum of mice administered with Aristipes indistinctus (AI) or Vehicle (a solvent for the bacterial cells). Data set count: 8. [Figure 8A] Fructose levels in the serum of mice administered with Aristipes indistinctus (AI) or Vehicle (bacterial solvent). The p-value was calculated using the Wilcoxon rank-sum test. [Figure 8B] Serum rhamnose levels in mice administered with Aristipes indistinctus (AI) or Vehicle (bacterial solvent). p-values ​​were calculated using Wilcoxon's rank-sum test. [Figure 8C] Serum glucose-6-phosphate levels in mice administered with Aristipes indistinctus (AI) or Vehicle (bacterial solvent). p-values ​​were calculated using the Wilcoxon rank-sum test. [Modes for carrying out the invention]

[0010] 1. Bacteria of the genus Aristipes The present invention provides a composition comprising bacteria of the genus Aristipes (hereinafter also referred to as "the Aristipes bacterial composition of the present invention") for the prevention and / or treatment of type 2 diabetes or obesity.

[0011] The genus Alistipes is a relatively new genus of bacteria, primarily isolated from medical clinical samples, and was previously classified under the genus Bacteroides. While isolated from the human gut microbiome, various species of this genus have been isolated from patients with appendicitis and abdominal and rectal abscesses. There are also research reports suggesting that Alistipes bacteria may have protective effects against several diseases, including hepatic fibrosis, colitis, and cardiovascular disease. For example, the following species exist, but are not limited to: Alistipes finegoldii, Alistipes indistinctus, Alistipes onderdonkii, Alistipes putredinis, Alistipes senegalensis, and Alistipes shahii. Preferably, Aristipes faingoldii or Aristipes indistinctus, and more preferably, Aristipes indistinctus.

[0012] Type 2 diabetes is a disease in which blood glucose (blood sugar) levels are higher than normal. If high blood sugar levels are left untreated, it gradually damages blood vessels and nerves throughout the body, leading to various complications. Type 2 diabetes is thought to be caused by factors such as genetic predisposition, high-calorie diet, high-fat diet, and lack of exercise. As a result, insulin secretion and insulin sensitivity (IS) decrease, leading to insulin deficiency. Foods containing sugar are broken down into glucose by saliva and digestive enzymes and absorbed into the bloodstream from the small intestine. When blood glucose levels increase after a meal, insulin is secreted from the pancreas, and glucose is sent to muscles and other tissues for use as energy. However, if "insulin deficiency" occurs, the body cannot properly process the glucose in the blood, and high blood sugar levels persist. There are two causes of "insulin deficiency": one is that the function of the pancreas weakens and insulin secretion decreases (insulin secretion deficiency), and the other is that tissues such as the liver and muscles become less sensitive to the action of insulin, making it less effective even when some insulin is secreted (insulin resistance).

[0013] Prediabetes refers to a state in which fasting or postprandial blood glucose levels fall between normal and abnormal (values ​​that would be considered diabetes). If left untreated, the risk of developing diabetes increases. It is also called pre-diabetes or borderline diabetes. Prediabetes occurs when the body secretes too little insulin or when insulin function is impaired, leading to an increase in blood sugar levels (hyperglycemia). For example, it is diagnosed when fasting blood glucose levels are between 110 and 125 mg / dl, or when blood glucose levels two hours after an oral glucose tolerance test are between 140 and 199 mg / dl.

[0014] Obesity is defined not only as having a high body weight, but also as a condition in which excessive body fat is accumulated. Obesity is a cause of numerous diseases, including lifestyle-related diseases such as diabetes, dyslipidemia, hypertension, and cardiovascular disease, making obesity prevention and management crucial for overall health. The degree of obesity is determined using the internationally standardized BMI (Body Mass Index) = [Weight (kg)] ÷ [Height (m)]. 2The term ] is used. The standard BMI for both men and women is 22.0, and obesity is defined as "a condition in which there is an excessive accumulation of fat in adipose tissue, resulting in a body mass index (BMI) of 25 or higher."

[0015] "Diet-induced type 2 diabetes or obesity" is defined herein as insulin resistance and weight gain resulting from excessive dietary intake of fat (particularly saturated fat) and, depending on the context, carbohydrates. For a given subject, excessive dietary intake, particularly fat and, depending on the context, carbohydrates, means the consumption of more food (diet) than is necessary to meet the subject's physiological needs and maintain energy balance, particularly fat and, depending on the context, carbohydrates. The effectiveness of a treatment for reducing (or preventing) diet-induced insulin resistance or weight gain in a subject can be evaluated by comparing the insulin resistance and weight gain observed in the treated subject with the insulin resistance or weight gain observed in the same subject who was given the same diet and had the same level of physical activity but did not receive the treatment.

[0016] In this specification, “improving insulin resistance” means restoring or reducing the level of insulin resistance induced by a given food in a subject compared to the level of insulin resistance induced by the food in a subject not administered with the composition containing Aristipes bacteria. Tests for evaluating insulin resistance in subjects are known in the art (see, for example, Ferrannini E, Mari A. How to measure insulin sensitivity. J Hypertens. 1998 Jul;16(7):895-906.). The level of insulin resistance in a subject can be easily estimated using any insulin resistance index known in the art (e.g., homeostatic model assessment of insulin resistance; HOMA-IR).

[0017] The present invention also encompasses compositions comprising bacteria of the genus Aristipes for use in the treatment, prevention, or mitigation of medical conditions resulting from food-induced insulin resistance or weight gain. Examples of medical conditions resulting from food-induced insulin resistance or weight gain include overweight, obesity, and related diseases (e.g., type 2 diabetes, impaired glucose tolerance (also known as prediabetes), non-alcoholic fatty liver disease (NAFLD), hypertension, and atherosclerosis (ASCVD)). One aspect of the present invention is to use a composition comprising Aristipes bacteria as a composition to reduce food-induced weight gain, improve food-induced insulin resistance, and, depending on the situation, alleviate inflammation in a subject, or to use the composition of the present invention to treat a subject. Another aspect of the present invention is to use a composition comprising Aristipes bacteria as a diagnostic marker to predict whether a subject has type 2 diabetes or obesity by detecting the Aristipes bacteria or their metabolites in the intestinal product, e.g., feces. Detection of Aristipes bacteria or their metabolites in the intestinal product may qualitatively mean whether they are detectable or not. Alternatively, detection of Aristipes bacteria or their metabolites in the intestinal product may quantitatively include determining the relative or absolute value of the abundance, detection frequency, or detection amount or concentration compared to other healthy individuals, obese individuals, and / or individuals with impaired glucose tolerance. The present invention also includes diagnosing conditions resulting from food-induced insulin resistance or food-induced weight gain by detecting Aristipes bacteria or their metabolites in intestinal products. Yet another aspect of the present invention refers to using Aristipes bacteria or their metabolites in the intestines qualitatively or quantitatively, such as by detecting them in intestinal products, for example, feces, and then administering them to supplement when the levels of Aristipes bacteria or their metabolites are lower than in healthy individuals.

[0018] The terms "treatment" or "treating" are used interchangeably herein, and these terms can represent a method for obtaining a beneficial result or a desired result. This includes, but is not limited to, therapeutic benefits and / or prophylactic benefits. Therapeutic benefits can mean the eradication or improvement of the primary disorder being treated. Similarly, therapeutic benefits can be achieved by eradicating or improving one or more physiological symptoms related to the primary disorder so that improvement is seen in the subject, even though the subject may still be at risk of suffering from the primary disorder. The prophylactic effect may include delaying, preventing or eliminating the occurrence of a disease or condition, delaying or eliminating the onset of symptoms of a disease or condition, delaying, stopping or reversing the progression of a disease or condition, or any combination thereof. To obtain a prophylactic effect, a subject at risk of developing a particular disease, or a subject reporting one or more physiological symptoms of a disease, can be treated even if the disease has not been diagnosed.

[0019] The composition of the present invention can be in any form suitable for administration, particularly oral administration. This form includes, for example, solids, semi-solids, liquids and powders. In the composition of the present invention, preferably, the strain is a live bacterium or a lyophilized preparation.

[0020] When the bacterium of the genus Allistipes is a live bacterium, the composition typically contains 10 5 ~10 13 colony forming units (cfu) per gram of dry weight, preferably at least 10 6 cfu, more preferably at least 10 7 cfu, even more preferably at least 10 8 cfu, most preferably at least 10 9 cfu. In the case of a liquid composition, generally 10 4 ~10 12 colony forming units (cfu), preferably at least 10 5 cfu, more preferably at least 10 6 cfu, even more preferably at least 10 7cfu, most preferably at least 10 9 This corresponds to cfu / mL. When the Aristipes bacteria are not viable, the composition is typically 10 per gram of dry weight. 5 ~10 13 It may contain bacterial DNA of the genus Aristipes, which corresponds to the viable bacteria of the colony-forming unit (CFU).

[0021] The compositions of the present invention may be used alone or in combination with other intestinal bacteria, such as lactic acid bacteria.

[0022] The compositions of the present invention include compositions that are distributed and used as pharmaceuticals, foods, dietary supplements, and functional foods. A "dietary supplement" is a product made from compounds commonly used in food products, in any form such as tablets, powders, capsules, liquid preparations, or other forms not normally associated with food, that has beneficial effects on health. A "functional food" is a food that has beneficial effects on health. In particular, dietary supplements and functional foods may have physiological, protective, or therapeutic effects on diseases, such as chronic diseases.

[0023] The term “administration” means either “oral administration,” i.e., the subject orally ingesting a bacterial strain or a composition containing said strain according to the present invention, or “direct administration,” i.e., the direct administration of a bacterial strain or a composition containing said strain in situ, particularly by colonoscopy or via suppositories to the rectum. Oral administration is preferred. The compositions of the present invention may be in the form of gelatin capsules, capsules, tablets, powders, granules, oral solutions, or suppositories. Administration may be a single dose or multiple doses. In the case of multiple doses, doses may be administered once or two to five times a day, daily, every two, three, four, five, or six days, or every one, two, three, four, five, six, seven, or eight weeks. Determining the therapeutically effective dose will be well understood by those skilled in the art, in particular, considering the detailed disclosures provided herein.

[0024] 2. Bacteroides bacteria The present invention provides a composition comprising Bacteroides bacteria for the prevention and / or treatment of type 2 diabetes or obesity (hereinafter also referred to as "the Bacteroides bacterial composition of the present invention").

[0025] Bacteroides bacteria are characterized by the presence of sphingolipids in their cell membranes and are one of the dominant bacteria that make up the gut microbiota, present in large quantities from the oral cavity. Bacteroides bacteria have immunomodulatory effects on the intestinal immune system. For example, the following species exist, but are not limited to: Bacteroides thetaiotaomicron, Bacteroides xylanisolvens, Bacteroides ovatus, and Bacteroides caccae. Preferably, it is Bacteroides thetaiotaomicron.

[0026] The present invention also encompasses compositions comprising Bacteroides bacteria for use in the treatment, prevention, or mitigation of conditions resulting from food-induced insulin resistance or weight gain. Examples of conditions resulting from food-induced insulin resistance or weight gain include overweight, obesity, and related diseases (e.g., type 2 diabetes, impaired glucose tolerance (also known as prediabetes), non-alcoholic fatty liver disease (NAFLD), hypertension, and atherosclerosis (ASCVD)). The compositions comprising Bacteroides bacteria of the present invention may be used in combination with compositions comprising Aristipes bacteria of the present invention. One aspect of the present invention is to use a composition comprising Bacteroides bacteria as a composition to reduce food-induced weight gain, improve food-induced insulin resistance, and, depending on the situation, alleviate inflammation in a subject, or to use the composition of the present invention to treat a subject. The composition comprising Bacteroides bacteria of the present invention may be used in combination with the composition comprising Aristipes bacteria of the present invention. Another aspect of the present invention is to use a composition comprising Bacteroides bacteria as a diagnostic marker to predict whether a subject has type 2 diabetes or obesity by detecting Bacteroides bacteria or their metabolites in an intestinal product, e.g., feces. Detecting Bacteroides bacteria or their metabolites in an intestinal product may involve qualitatively determining whether they are detectable or not. Alternatively, detecting Bacteroides bacteria or their metabolites in an intestinal product may involve quantitatively determining the relative or absolute value of the abundance, detection frequency, or detection amount or concentration compared to other healthy individuals, obese individuals, and / or individuals with impaired glucose tolerance. The present invention also includes diagnosing conditions resulting from food-induced insulin resistance or food-induced weight gain by detecting Bacteroides bacteria or their metabolites in intestinal products. Yet another aspect of the present invention refers to using Bacteroides bacteria or their metabolites in the intestines qualitatively or quantitatively, such as by detecting them in intestinal products, for example, feces, and then administering them to supplement when the levels of Bacteroides bacteria or their metabolites are lower than in healthy individuals.

[0027] 3. Dorea bacteria The present invention provides a composition comprising a bacterium of the genus Drea or a portion of the genomic DNA of said bacterium for the prevention of type 2 diabetes or obesity.

[0028] Dorea bacteria are a type of intestinal bacteria and are generally classified as opportunistic bacteria. Opportunistic bacteria are bacteria that are neither beneficial nor harmful intestinal bacteria, but side with whichever group is stronger. Examples include non-pathogenic Escherichia coli, Clistridium, Streptococcus, Dorea, and Bacteroides. Dorea bacteria include, but are not limited to, the following species: Dorea formicigenrans and Dorea longicatena. Preferably, Dorea formicigenrans or Dorea longicatena.

[0029] In this specification, “a composition comprising a Dreia bacterium or a portion of the genomic DNA of said bacterium for the detection of type 2 diabetes or obesity” means the use of a Dreia bacterium or a portion of the genomic DNA of said bacterium as a diagnostic marker to predict whether a subject has type 2 diabetes or obesity by detecting it in an intestinal product, such as feces. Detecting a Dreia bacterium or a portion of the genomic DNA of said bacterium in an intestinal product includes qualitatively determining whether it is detectable or not. Alternatively, detecting a Dreia bacterium or a portion of the genomic DNA of said bacterium in an intestinal product includes quantitatively determining the relative or absolute value of its abundance compared to other healthy individuals, obese individuals, and / or individuals with impaired glucose tolerance. In the composition, “a portion of the genomic DNA of said bacterium” includes a single-stranded deoxyribooligonucleotide or modified oligonucleotide that can specifically hybridize with any DNA strand of the Dreia bacterium genomic DNA, which can specifically detect the Dreia bacterium genome, or a primer pair that can amplify a portion of the Dreia bacterium genomic DNA by PCR or other methods. An alternative aspect of the present invention is a risk detection method for type 2 diabetes or obesity, which involves determining the abundance of Dreya bacteria in a sample of a target intestinal organism, and determining that the target has a high risk of developing type 2 diabetes or obesity if the abundance is greater than that in a sample of a healthy person's intestinal organism. In the above method, the "step of determining the abundance of Dreya bacteria in the intestinal organism" means determining the abundance of Dreya bacteria by constructing a metagenomic shotgun library of the intestinal organism microbiome and sequencing analysis, and / or 16S rRNA gene amplicon sequencing analysis. As shown in the following examples, the abundance of Dreya bacteria has been shown to increase with IR. Therefore, if the abundance of Dreya bacteria is greater than that in a sample of a healthy person's intestinal organism, it is determined that the target has a high risk of developing type 2 diabetes or obesity.A further aspect of the present invention includes using Dreya bacteria or their metabolites in the intestines to qualitatively or quantitatively control Dreya bacteria or their metabolites in the intestines by detecting them in intestinal products, such as feces, and then administering them to supplement when the amount of Dreya bacteria or their metabolites is deficient compared to healthy individuals.

[0030] The present invention also provides a composition containing bacteria identified by a large-scale human study, comprising the steps of (i) collecting samples including feces from healthy individuals and subjects, (ii) profiling the intestinal microbiota and bacterial metabolites of the samples collected in step (i) and performing correlation analysis, and (iii) determining the bacteria that showed a significant difference as a result of the correlation analysis in step (ii) as bacteria that exhibit effects on the host.

[0031] In this specification, “subject,” “individual,” or “person” are interchangeable. “Subject” may be a living organism containing expressed genetic material. The living organism may be a plant, an animal, or a microorganism, including, for example, bacteria, bacterial plasmids, viruses, fungi, and protozoa. The subject may be in vivo collected or cultured tissues, cells, and their offspring from a living organism. Preferably, the subject is a mammal; more preferably, a human. The subject may be diagnosed with or suspected of being at high risk of a certain disease. In some cases, the subject may not necessarily be diagnosed with or suspected of being at high risk of that disease.

[0032] A biological sample may be any sample type from any microbiome of the body in question. Some examples of microbiomes that can be used in this disclosure include the skin microbiome, gastrointestinal microbiome, nasal microbiome, and oral microbiome. Preferably, it is the gastrointestinal microbiome. Depending on the application, the biological sample may be whole blood, serum, plasma, mucus, saliva, cheek swab, urine, feces, cells, tissues, body fluids, or a combination thereof. Preferably, it is feces or plasma. More preferably, it is feces. In this specification, "feces" includes not only feces after excretion but also the contents of the gastrointestinal tract.

[0033] The microbiome refers to the trillions of microorganisms (also called bacterial communities) that inhabit various parts of a healthy individual's body. The gut microbiota alone is reported to contain approximately 40 trillion bacteria. Some examples of microbiome sites include the skin, digestive tract, oral cavity, conjunctiva, and vagina. To better understand the roles of these microbiomes and how they influence physiology and disease, we can analyze what microorganisms make up the microbiome, how they interact, and how they affect an individual's health and clinical responses.

[0034] One current standard in the field of phylogenetic classification of bacterial species is DNA sequencing of 16S ribosomal RNA (rRNA) genes. For example, 16S rRNA is generally present in all bacteria and contains nine variable regions that can be used to distinguish phylogenetic classifications, making it a viable target for classification.

[0035] Bacterial metabolites are substances produced by bacteria through metabolic processes (also called bacterial fermentation products), and examples include carbohydrate metabolites, lipid metabolites, amino acid metabolites, polyamines, choline metabolites, and vitamins. Carbohydrate metabolites include sugar metabolites such as monosaccharides and disaccharides, specifically monosaccharides such as glucose, fructose, galactose, and xylose, or disaccharides such as maltose and sucrose, but are not limited to these. Lipid metabolites include short-chain fatty acids, medium-chain fatty acids, and long-chain fatty acids, with short-chain fatty acids being preferred, and the main components being acetic acid, propionic acid, and butyric acid, but are not limited to these.

[0036] "Profiling and correlationally analyzing the gut microbiota and bacterial metabolites" refers to comprehensively integrating the composition and function of the gut microbiota by combining, for example, 16S rRNA gene sequencing data of the gut microbiota with metabolome analysis data of gut bacterial metabolites. Metabolite profiles obtained by mass spectrometry and gut microbiota profiles obtained by next-generation sequencing may be analyzed independently at each hierarchical level, as well as integrated analyses of both sets of data over time. This allows for the clarification of the characteristics of the target gut environment, comprehensive analysis of the correlation between individual bacterial species and metabolites, and the acquisition of clues for their control. Furthermore, bacterial composition and gene analysis can be performed using metagenomics, bacterial gene expression analysis using transcriptomics, and analysis of gut metabolites using metabolomics, allowing for a comprehensive analysis of the overall picture as multi-omics correlation analysis. In addition to the methods shown in the following examples, known analytical methods (e.g., Ohno et al., Experimental Medicine 29:2937-2942, 2011) can be used.

[0037] In this specification, a large-scale human study refers to a study involving, for example, 100 to 500 subjects. Preferably, the number of subjects is 200 or more, more preferably 300 or more. The upper limit of the subjects is preferably 200, 300, 400, or more. The upper limit of the subjects may also be 1000, 2000, 10000, 20000, or more. Specifically, as shown in the examples, examples of studies include, but are not limited to, the following: To investigate how intestinal bacteria affect metabolic abnormalities in humans, plasma and fecal samples are collected from 306 individuals, including healthy individuals, obese individuals, and individuals with impaired glucose tolerance (pre-diabetic patients), and intestinal bacteria and metabolites are analyzed. To exclude the effects of diabetes medications on intestinal bacteria, only untreated individuals with impaired glucose tolerance may be included.

[0038] In the present invention, detection of bacteria of the genera Aristipes, Bacteroides, and / or Dreya from fecal samples can be performed using detection methods such as metagenomic shotgun library construction and sequencing analysis of the fecal microbiome, or detection methods such as 16S rRNA gene amplicon sequencing analysis, as described in detail below. Alternatively, detection methods using PCR or other amplification techniques with primers or probes targeting specific genome sequences common to Aristipes, Bacteroides, or Dreya, or genome sequences specific to individual species of Aristipes, Bacteroides, or Dreya, can be used for detection of bacteria of the genera Aristipes, Bacteroides, and / or Dreya from fecal samples. The detection methods using primers or probes also include methods in which the primers or probes are immobilized on a solid support such as a DNA chip and detected by fluorescence, luminescence, color development, surface plasmon resonance, or other means.

[0039] The present invention will be described in more detail below with reference to examples, but these are merely illustrative and do not limit the scope of the present invention in any way. [Examples]

[0040] (material and method) Research participants and data collection From 2014 to 2016, participants were recruited for this study during health checkups at the University of Tokyo Hospital. Applicants were Japanese men and women aged 20 to 75. Exclusion criteria were as follows: a confirmed diagnosis of diabetes, daily use of medications for diabetes and / or gastrointestinal disorders, administration or use of antibiotics within two weeks prior to sample collection, and a weight loss of 3 kg within three months prior to sample collection. After explaining the study in detail, written consent was obtained from participants before the health checkups were conducted. To standardize the clinical characteristics of participants, we planned to recruit approximately 100 normal, 100 obese, and 100 pre-diabetic individuals based on clinical data, and stopped recruitment when the number of participants nearly reached the target. Sample size was determined based on previous metagenomic studies that demonstrate the microbial signature characteristics of diabetic patients. As a result, 112, 100, and 101 individuals were enrolled in the pre-targeted groups. Of these, two withdrew their consent to participate in the study. Additionally, five did not provide fecal samples. In total, 306 individuals who underwent physical examination, clinical tests, fecal sampling for fecal 16S rRNA gene sequencing and metabolome analysis of water-soluble metabolites, and plasma sampling for plasma metabolite analysis were included in this study. Due to the limited sample size, fecal metagenomic data were available for 290 individuals, and plasma cytokine and insulin data were available for 282 individuals. All clinical data and plasma samples were collected at the hospital during health checkups. Fecal samples were collected on the day of the hospital visit or at home and immediately sent to the hospital frozen within 1-2 days before or after the health checkup. Blood samples were immediately centrifuged to collect plasma. Plasma and fecal samples were stored at -80°C until sample preparation and analysis. This study was approved by the Ethics Review Boards of RIKEN and the University of Tokyo and was conducted in accordance with the guidelines of each institution.

[0041] Phenotypic results The results of the phenotypes used in this study are as follows. First, insulin resistance (IR), defined as HOMA-IR ≥ 2.5 in the Japanese population, was used. Insulin sensitivity (IS) was defined as HOMA-IR ≤ 1.6. HOMA-IR was calculated as follows: Fasting plasma insulin (μU / mL) × Fasting plasma glucose (mg / dL) / 405. Due to limited plasma insulin data, this phenotype was available to 282 individuals. Next, metabolic syndrome (MetS) was used. The diagnosis of MetS was based on the Japanese criteria (Y. Matsuzawa, Metabolic Syndrome - Definition and Diagnostic Criteria in Japan. J. Atheroscler. Thromb. 12, 301 (2005).). A waist circumference of ≥ 85 cm for men and ≥ 90 cm for women is a prerequisite for diagnosis. In addition, two of the following three clinical abnormalities are required for diagnosis. Dyslipidemia, defined as triglycerides ≥ 150 mg / dL and / or HDL cholesterol < 40 mg / dL; hypertension, defined as systolic blood pressure ≥ 130 mmHg and / or diastolic blood pressure ≥ 85 mmHg; and fasting hyperglycemia, defined as fasting blood glucose ≥ 110 mg / dL.

[0042] Measurement of plasma cytokines Plasma cytokines were measured using the Human Adipokine Magnetic Bead Panel 2 (Millipore, HADK2MAG-61K) and the Human Obesity Premixed Magnetic Luminex Performance Assay Kit (R&D, FCSTM08) according to the manufacturer's instructions. Data below the detection limit were considered zero, and data above the detection limit were considered the highest value of cytokine analyzed.

[0043] Preparation of fecal samples Aliquots (5g) of feces were blended with 30mL of methanol, filtered through a 100μm mesh filter, and vigorously vortexed to remove food residue. The filtrate was centrifuged at 15,000×g at 4°C for 10 minutes, and the supernatant (methanol extract) was used for metabolome analysis. DNA from the fecal microbiome was collected from the pellet.

[0044] Extraction and measurement of water-soluble metabolites from fecal and plasma samples Extraction of water-soluble metabolites was performed with modifications to the previous description by Nishiumi S. et al. (S. Nishiumi et al., Metabolomics. 6, 518-528 (2010)). To a 10 μL aliquot of plasma, 150 μL of methanol, 125 μL of Milli-Q water, 15 μL of internal standard solution (1 mmol / L 2-isopropylmalic acid), and 60 μL of CHCl3 were added. For the fecal sample, a 25 μL aliquot of methanol extract was added, along with 125 μL of methanol, 150 μL of Milli-Q water containing the internal standard (100 μmol / L 2-isopropylmalic acid), and 60 μL of CHCl3. The solutions were shaken at 1,200 rpm at 37°C for 30 minutes. After centrifugation at 16,000 × g at room temperature for 5 minutes, 250 μL of supernatant was transferred to a new tube and 200 μL of Milli-Q water was added. After mixing, the solution was centrifugated at 16,000 × g at room temperature for 5 minutes, and 250 μL of supernatant was transferred to a new tube. The sample was evaporated to dryness at 40°C for 20 minutes using a vacuum evaporator and then lyophilized using a lyophilizer. The dried extract was first methoxylated with 40 μL of 20 mg / mL methoxyamine hydrochloride (Sigma-Aldrich) dissolved in pyridine and shaken at 1,200 rpm for 90 minutes at 30°C. Next, the solution was silylated with 20 μL of N-methyl-N-trimethylsilyl-trifluoroacetamide (MSTFA, GL Science) at 37°C for 30 minutes with shaking at 1,200 rpm. After derivatization, the sample was centrifugated at 16,000 × g at room temperature for 5 minutes, and the supernatant was transferred to a glass vial. Analysis was performed using a gas chromatography-tandem mass spectrometry (GC / MS / MS) platform with a Shimadzu GCMS-TQ8030 triple quadrupole mass spectrometer (Shimadzu) equipped with a capillary column (BPX5, SGE Analytical Science). The GC oven program was as follows: held at 60°C for 2 minutes, increased to 330°C (15°C / min), and finally held at 330°C for 3.45 minutes. The GC was operated in constant linear velocity mode set to 39 cm / sec. The detector and injector temperatures were 200°C and 250°C, respectively. The injection volume was set to 1 μL with a split ratio of 1:30.The extraction and measurement of SCFAs were modified from the previously reported method ((Y. Sato et al., J. Dev. Orig. Health Dis.10, 659-666 (2019))). An internal standard (2 mM [1,2-]) was added to a plasma aliquot (90 μL). 13 C2] Acetate, 2mM[ 2 10 μL of Milli-Q water containing [H7]butyrate and 2 mM crotonate was added to the fecal sample. For the fecal sample, 25 μL of methanol extract was aliquoted to 10 μL of Milli-Q water containing an internal standard, then concentrated by centrifugation at 40°C and reconstituted with 100 μL of Milli-Q water. 50 μL of hydrochloric acid (HCl) and 200 μL of diethyl ether were added to this solution and mixed well. After centrifugation at 3,000 × g for 10 minutes, 80 μL of the organic layer was transferred to a glass vial and the sample was derivatized by adding 16 μL of N-tert-butyldimethylsilyl-N-trifluoroacetamide (MTBSTFA, Sigma-Aldrich). The vial was incubated at 80°C for 20 minutes and allowed to stand for 48 hours before injection. Analysis was performed using a Shimadzu GCMS-TQ8030 triple quadrupole mass spectrometer equipped with a capillary column (BPX5). The GC oven program was as follows: 60°C was maintained for 3 minutes, then increased to 130°C (8°C / min), then to 330°C (3°C / min), and finally maintained at 330°C for 3 minutes. The detector and injector temperatures were 230°C and 250°C, respectively. The GC was operated in constant linear velocity mode set to 40 cm / sec. The injection volume was set to 1 μL with a split ratio of 1:30. The data were processed using LabSolutions Insight (Shimadzu) to calculate the concentration.

[0045] Lipidomics of fecal and plasma samples Lipidomics (lipid metabolite analysis) was performed according to a previously reported study (H. Tsugawa et al., Nat. Biotechnol., in press, doi:10.1038 / s41587-020-0531-2). LC-MS grade methanol, isopropanol, chloroform, and acetonitrile were purchased from Wako (Tokyo, Japan). Ammonium acetate and EDTA were purchased from Wako and Dojindo (Tokyo, Japan), respectively. Milli-Q water was purchased from Millipore (Merck, Massachusetts, US). EquiSPLASH was purchased from Avanti Polar Lipids (Alabama, US). Palmitic acid-d3 and stearic acid-d3 were purchased from Olbracht Serdary Research Laboratories (Toronto, ON, Canada).

[0046] For plasma lipid extraction, an aliquot of 20 μL of human plasma sample was added to 200 μL of methanol containing 5 μL of EquiSPLASH (Avanti Polar Lipids, Inc.), 10 μM palmitic acid-d3, and 10 μM stearic acid, and vortexed for 10 seconds. Next, 100 μL of chloroform was added and vortexed for 10 seconds. After incubation at room temperature for 2 hours, the solvent tube was centrifuged at 2000 × g for 10 minutes at 20°C. 200 μL of the supernatant was transferred to an LC-MS vial (Agilent Technologies). For fecal lipid extraction, 50 μL of methanol extract was added to 145 μL of methanol containing 5 μL of EquiSPLASH, 10 μM palmitic acid-d3, and 10 μM stearic acid-d3 in a 2 mL glass tube, and vortexed for 10 seconds. Next, 100 μL of chloroform was added and vortexed for 10 seconds. After incubation at room temperature for 1 hour, 20 μL of water was added and vortexed for 10 seconds. After incubation at room temperature for 10 minutes, the solvent was centrifuged at 2000 × g at 4°C for 10 minutes, and the supernatant was transferred to an LC-MS vial. All samples were divided into four and five batches, and after randomization for plasma and fecal analysis, 70–80 and 55–60 samples were obtained from each batch, respectively. For each batch, a series of samples were prepared and subsequent LC-MS / MS measurements were performed. Quality control (QC) samples were prepared by mixing an equal volume of plasma from the first batch subjects. Procedure blanks were prepared using an equal volume of water instead of the biological samples. Blank samples were analyzed at the beginning and end of each analytical batch, and QC samples were injected for every 10 study samples.

[0047] The LC system consisted of a Waters Acquity UPLC system. Lipids were separated using an Acquity UPLC Peptide BEH C18 column (50 × 2.1 mm; 1.7 μm) (Waters, Milford, MA, USA). The column was maintained at a flow rate of 0.3 mL / min and 45°C. The mobile phase consisted of (A) 1:1:3 (v / v / v) acetonitrile:methanol:water containing ammonium formate (5 mM) and 10 nM EDTA, and (B) 100% isopropanol containing ammonium formate (5 mM) and 10 nM EDTA. Sample volumes of 0.5 to 3 μL were used for injection, depending on the biological sample. Separation was performed using the following gradients: 0 min 0%(B); 1 min 0%(B); 5 min 40%(B); 7.5 min 64%(B); 12 min 64%(B); 12.5 min 82.5%(B); 19 min 85%(B); 20 min 95%(B); 20.1 min 0%(B); and 25 min 0%(B). The sample temperature was maintained at 4°C.

[0048] Lipid mass spectrometry detection was performed using a TripleTOF 6600 quadrupole / time-of-flight mass spectrometer (SCIEX, Framingham, MA, USA). All analyses were performed in high-resolution MS1 ​​mode (full width at half maximum (FWHM) ~35,000) and high-sensitivity MS2 mode (~20,000 FWHM). Data-dependent MS / MS acquisition (DDA) was used. The parameters were: mass range of MS1 and MS2, m / z 70~1250; MS1 storage time, 250 ms; MS2 storage time, 100 ms; collision energy, +40 / -42 eV; collision energy spread, 15 eV; cycle time, 1300 ms; curtain gas, 30; ion source gas 1, 40(+) / 50(-); ion source gas 2, 80(+) / 50(-); temperature, 250℃(+) / 300℃(-); floating ion spray voltage, +5.5 / -4.5 kV; declustering potential, 80 V. Other DDA parameters were: dependent product ion scan number, 16; intensity threshold, 100 cps; precursor ion exclusion time, 0 sec; mass tolerance, 20 ppm; ignor peak, m / z within 200; and dynamic background subtraction, True. Mass calibration was performed automatically using the APCI positive / negative calibration solution via the Calibration Delivery System (CDS).

[0049] MS-DIAL version 4.16 was used with the following parameters. The set parameters were as follows: (Data acquisition) RT start, 1.0 min; retention end, 18 min; mass range start for MS1 and MS2, 0 Da; mass range end for MS1 and MS2, 2000 Da; MS1 tolerance, 0.01 Da; MS2 tolerance, 0.025 Da; (Peak detection) Minimum peak height, 3000 amplitude; mass slice width, 0.1 Da; smoothing method, linear weighted moving average; smoothing level, 3 scans; minimum peak width, 5 scans; exclusion mass list, none; (Identification) Retention time tolerance, 1.5 min; accurate mass tolerance for MS1, 0.01 Da; accurate mass tolerance for MS2, 0.05 Da; cutoff for discrimination score, 70%; all lipid subclasses were used as search space; (Alignment) Retention time tolerance, 0.15 min; MS1 tolerance, 0.015 Da. Default values ​​were used for the other parameters.

[0050] Metabolite clustering Clustering analysis was performed on 110 water-soluble metabolites and 2654 lipid metabolites selected from fecal metabolites that passed quality control and were detected in multiple samples. These metabolites were clustered based on their co-abundance using the R package "WGCNA". The following parameters were used for the analysis: For water-soluble metabolites, soft threshold β=12, minimum cluster size=3, deep split=4, cut height=0.9999, and PAM clustering=F. For lipid metabolites, soft threshold β=14, minimum cluster size=20, deep split=4, cut height=0.999, and PAM clustering=F. Since WGCNA could not cluster all metabolites, the remaining metabolites that did not meet the criteria were subsequently clustered based on biweight midcorrelation. The following parameters were used for secondary clustering: For water-soluble metabolites, minimum cluster size=3, deep split=4, cut height=0.9999, and PAM clustering=F. For lipid metabolites, the following settings were used: minimum cluster size = 6, deep split = 4, cut height = 0.999, and PAM clustering = F. Clusters with biweight intermediate correlations greater than 0.8 were merged. The primary principal component (PC1) of each cluster was calculated using the WGCNA "moduleEigengenes" command and used as a representative value for the cluster for subsequent analysis. The representative metabolite class included in each cluster was noted as the cluster annotation.

[0051] DNA extraction from fecal samples The detailed method described separately (T. Kato et al., DNA Res. 21, 469-480 (2014)) was slightly modified. Before DNA extraction, the fecal pellet was washed once with PBS and suspended in 10 mM Tris-HCl / 20 mM EDTA buffer (pH 8.0). Subsequently, lysozyme (Sigma), achromopeptidase (Wako), and proteinase K (Merck) were added to the cell lysis sample. DNA was recovered by phenol / chloroform extraction. To purify the extracted DNA, RNA was digested with RNase (Nippon Gene). Next, the DNA was precipitated in a solution containing polyethylene glycol 6000 solution (Hampton Research). DNA concentration was quantified using Quant-iT PicoGreen.

[0052] Shotgun metagenomic sequencing Metagenome shotgun libraries (insertion size 500 bp) were prepared using the TruSeq Nano DNA Kit (Illumina) and sequenced on the Illumina NovaSeq platform. After quality filtering, reads mapped to the human genome (HG19) and phiX bacteriophage genomes were removed. For each individual, NovaSeq reads that passed the filter were assembled using MEGAHIT (v1.2.4). Protein-coding genes (≧100 bp) were predicted as contigs (≧500 bp) and singletons (≧300 bp) using Prodigal (v2.6.3). Finally, 6,458,217 non-redundant genes were identified from 290 samples by clustering the predicted genes using CD-HIT with 95% nucleotide identity and a 90% length coverage cut-off. Functional assignment of non-redundant genes was performed against the Kyoto Encyclopedia of Genes and Genomes (KEGG) database (release 2019-10-07) using DIAMOND (e-value ≤ 1.0e-5) to obtain KEGG orthology genes (KO). Best-hit genes that correlated with eukaryotic genes were excluded from further analysis.

[0053] Quantification of annotated genes in the metagenomic. Using Bowtie2, 1 million metagenomic reads per individual were mapped to a reference gene catalog with a 95% identity cutoff. The reference gene catalog used was a combined version of non-redundant genes from this study and non-redundant genes from IGC (J. Li et al., Nat. Biotechnol. 32, 834-841 (2014)) and JPGM (S. Nishijima et al., DNA Res. 23, 125-133 (2016)). The number of reads that mapped equally to multiple genes was normalized by the proportion of reads that mapped uniquely to the gene, as was done for the mapping analysis to the reference genome. The proportion of knockouts (KOs) was calculated from the number of reads that mapped to them. For the KEGG pathway enrichment analysis, all upstream pathways linked to a KO were assigned +1 (-1) for significantly and positively (negatively) associated KOs, and these points were summarized as a ratio to the number of KOs in the pathway.

[0054] 16S rRNA gene amplicon sequencing analysis Fecal sample DNA was amplified by PCR using barcoded primers, targeting the variable region V1-V2 of the 16S rRNA gene. PCR amplicons were purified using AMPure XP magnetic beads (Beckman Coulter, Inc.) and quantified using the Quant-iT PicoGreen dsDNA assay kit (Life Technologies Japan, Ltd.). Equal volumes of PCR amplicons were mixed and sequenced using MiSeq (Illumina). Sequencing reads were sorted by sample according to barcode using bcl2fastq. Reads lacking both forward and reverse primer sequences were then excluded. Finally, reads with an average quality value < 25 were filtered to exclude chimeric sequences. The 16S database was created based on a public database (Ribosomal Database Project (RDP) v. 10.27, CORE (microbiome.osu.edu / ), NCBI FTP site (ftp: / / ftp.ncbi.nih.gov / genbank / , December 2011)). Operational Taxonomy Unit (OTU) clustering and UniFrac analysis were performed by randomly selecting 3000 reads after the above filtering. Subsequently, all selected reads were clustered using OTU clustering with a 97% homology threshold using UCLUST (www.drive5.com / ). Representative OTU sequences were classified by performing homology searches using the GLSEARCH program and the above 16S database.

[0055] Culture experiment Aristipes feingordii (JCM16770), Aristipes indistinctus (JCM16068), Bacteroides ceetiotaomicron, Bacteroides xylanisorbens, Parabacteroides merdae, Clostridium spiroforme, Faecalibacterium prausnitzii, Drea formisigenerans (JCM31256), and Drea longicatena (JCM11232) were obtained from the RIKEN BioResource Research Center. All bacteria were cultured in YCFA medium (JCM medium No. 1130) containing short-chain fatty acids. To measure metabolites in the cell-free supernatant, bacteria grown in YCFA medium were inoculated into experimental medium (YCFA medium) and cultured for 24 hours. The samples were centrifuged, and the cell-free supernatant was collected and analyzed. As described above, GC-MS was performed to measure water-soluble metabolites.

[0056] Animal experiments Six-week-old male C57BL6 / N mice were purchased from CLEA Japan. The mice were given Quick Fat (CLEA Japan) starting three weeks before bacterial administration and continued throughout the three weeks of bacterial administration. Aristipes indistinctus (JCM16068) was cultured overnight in YCFA medium and then diluted in PBS to a concentration of 2.5 × 10⁻⁶. 8 The concentration was adjusted to CFU / mL. Bacteria and PBS (a negative control) were orally administered at a dose of 200 μL per mouse. Bacteria and PBS were administered 2-3 times per week for 3 weeks. Body weight was measured 2-3 times per week prior to forced oral administration. An insulin loading test was performed 3 weeks after the start of bacterial administration. Mice were fasted for 5 hours prior to insulin injection, followed by intraperitoneal administration of 0.85 U / kg of insulin (Sigma). Blood glucose was collected from the tail vein and continuously measured using a Glucocard (Arkray). At necropsy, mice were anesthetized with isoflurane (MSD), and intestinal contents were collected from the cecum. All experimental procedures were approved and followed by the Animal Experiment Committees of RIKEN and Yokohama City University.

[0057] statistical analysis For general statistical comparisons, the Wilcoxon rank-sum test was used to compare two groups, and the Kruskal-Wallis test followed by Dunn's post-hoc analysis was used to compare three or more groups. Spearman's rank correlation was used to assess correlations. Two-way ANOVA was used to compare time-series data, such as insulin loading tests. P-values ​​less than 0.05 were considered significant. For multiple testing correction, p-values ​​were corrected using the Benjamini-Hochberg procedure with the R command "p.adjust".

[0058] To construct a genus-level microbial co-existence network, genus-level microorganisms observed in more than 50% of participants were selected, and correlations were calculated using the R package CCREPE (Compositionality Corrected by REnormalization and PErmutation) with initial settings. Interactions with q values ​​less than 0.2 were filtered for further analysis. Microorganisms positively related to each other were determined as members of independent co-existence microbial groups. To characterize the microbial profiles of research participants, individuals were clustered based on the abundance of 20 genera in four co-existence groups using the ward.D function of the R package pheatmap. Three distinct clusters of participants were determined, and IR ratios were compared among these three groups, A to C, using the Cochran-Armitage test. Other clinical parameters, such as HOMA-IR, were also compared using the Kruskal-Wallis test followed by Dunn's post-hoc analysis. To visualize the microbial metabolite network, the correlation between IS-associated co-existence groups (Bacteroidales + Clostridiales cluster) and IR-associated groups (Lachnospiraceae, Coriobacteriaceae + Bifidobacteriaceae, and other clusters) and their IS and IR-associated water-soluble and lipid metabolites was analyzed separately. Spearman correlations with q values ​​less than 0.05 were filtered for network analysis. The network was visualized using Cytoscape ver. 3.7.0.

[0059] To assess the association between host inflammatory characteristics and fecal metabolites, we first calculated cytokine scores, the primary principal component of PCA based on 10 cytokines measured in this study. The cytokine scores were then tested for partial Spearman correlations among 15 fecal carbohydrate metabolites associated with IR, adjusted for age, sex, BMI, and FBG, and corrected for multiple testing.

[0060] To construct and visualize correlation-based networks of omics data, metabolites, transcripts, and microorganisms associated with IS or IR were selected. Data were standardized, and partial Spearman rank correlations adjusted for age, sex, BMI, and FBG were calculated between all given factors. Correlations with a q value less than 0.2 were filtered for visualization. Networks, such as microbial metabolite networks, were visualized using Cytoscape ver. 3.7.0.

[0061] Example 1: Study setup and overview of multi-omics data. This study analyzed 306 men and women (70.9% male) aged 20 to 75 years (median 61 years) recruited through the annual health checkup at the University of Tokyo Hospital (Figure 1A). The cohort was designed to include approximately 100 individuals with normal blood glucose levels, 100 with obesity, and 100 with prediabetes. Individuals with a confirmed diagnosis of diabetes were excluded from the study to avoid the long-term effects of hyperglycemia. Similarly, individuals taking diabetes medications and those routinely taking intestinal disease medications were excluded as drug users who have been reported to contaminate the gut microbiome. Details of the exclusion criteria are described in the (Materials and Methods) section of the Examples section of this specification. As a result, this study included relatively healthy individuals. Median (IQR) body mass index (BMI), fasting blood glucose (FBG), and HbA1c were 24.9 (22.2–27.1) kg / m², respectively. 2The levels were 96(91-105) mg / dL and 5.8(5.5-6.1)%. The clinical phenotype primarily analyzed in this study was IR, defined as HOMA-IR ≥ 2.5. Furthermore, the association with MetS was analyzed to support the observations. Intestinal microbial information was obtained by 16S rRNA gene amplicon sequencing and metagenomic shotgun sequencing. Data for 482 intestinal microbiota species were obtained. There were 26 core genera with relative abundances exceeding 80%, and 4 co-existing groups. Non-targeted metabolomics using two mass spectrometry-based analytical platforms generated information on 195 and 100 fecal and plasma water-soluble metabolites, and 2654 and 635 fecal and plasma lipid metabolites, respectively. The following analyses included those detected in 75% or more of the samples (see the "(Materials and Methods)" section of the Examples Specified). Host transcriptome information was obtained using the Cap Analysis of Gene Expression (CAGE) method (M Kanamori-Katayama et al., Genome Res. 21, 1150-1159 (2011)). Transcription levels for 14,614 genes were analyzed. A schematic flowchart of the multi-omics analysis is shown in Figure 1B. First, we aimed to define the microbial identification characteristics of IR and MetS using fecal metabolomics and metagenomic data. Next, we attempted to identify potential interactions between host and microbial metabolites that influence the pathophysiology of IR and MetS. Fecal metabolome and metagenomic data were summarized into clusters (CAGs) based on their correlations, and fecal data were summarized into functional modules, namely KEGG orthologs and pathways, respectively. There were 18 water-soluble CAGs, 105 lipid CAGs, and 6,711 KEGG orthologs. Phenotypes and microbial-related metabolites were further analyzed for correlation with host cytokines, metabolome, and transcriptome changes. Finally, several experiments were conducted to investigate the metabolic effects of candidate microorganisms and their metabolites.

[0062] Example 2: Carbohydrate metabolites in feces change with IR. Next, we investigated whether fecal metabolomics provides mechanistic clues to the pathogenesis of metabolic diseases. Due to differences in mass spectrometry techniques and the following data analysis methods, water-soluble and lipid metabolites were analyzed separately (see Materials and Methods for details).

[0063] Carbohydrate metabolites, mainly monosaccharides (e.g., hydrophilic CAG 5, 12, and 15), were most strongly associated with IR and MetS (Figure 2A). CAG containing short-chain fatty acids, a major polysaccharide fermentation product (hydrophilic CAG 8), was also increased, particularly in IR. KEGG pathway analysis of CAG 5, 12, and 15 metabolites, which showed significant associations with both IR and MetS, revealed that the metabolites are indeed involved in carbohydrate metabolism. Specifically, monosaccharides such as glucose, fructose, galactose, and xylose were found to be significantly increased in IR (Figures 2B-E). In contrast, disaccharides such as maltose and sucrose showed weak or no association with IR or MetS (Figures 2F and G), suggesting that these sugars are broken down into monosaccharides.

[0064] Example 3: Interaction between microorganisms and metabolites in IR Next, we investigated the involvement of the gut microbiota in the production of carbohydrate fermentation products and specific lipid metabolites that were increased in IR and MetS. First, we assessed the changes in the gut microbiota associated with these phenotypes. Using 16S rRNA gene sequencing data, we profiled the genus-level microbiome of the study participants based on their co-existence (correlation). As a result, we identified four bacterial groups characterized by Bacteroideales + Clostridiales, Lachnospiraceae, Coriobacteriaceae + Bifidobacteriaceae, and others. Furthermore, based on the microbial profile, we identified three clusters A through C of the participants (Figure 3A). Individuals in cluster A possessed more microorganisms from the Bacteroideales + Clostridiales group, while individuals in cluster C possessed more bacterial species from the Lachnospiraceae group. In particular, the ratio of IR (Figure 3A, p = 0.025) to HOMA-IR (Figure 3B) was significantly higher in cluster C, which possessed more Lachnospiraceae groups. Other metabolic parameters associated with IR and MetS, such as triglycerides (TG), HDL-C, and adiponectin, differed between clusters A and C (Figure 3B). Specifically, the abundance of the Lachnospiraceae species Dorea formicigenerans increased in IR, while the Rikenellaceae genus Alistipes showed a negative association. Adiponectin, which improves insulin sensitivity, showed a correlation with these genera in the opposite direction. These findings characterize the IS and IR-associated microorganisms in this cohort.

[0065] Next, we analyzed the interaction between gut microbiota and IR-related fecal metabolites. A Spearman correlation-based network suggested that fecal carbohydrate metabolites were positively correlated with bacterial species belonging to the Lachnospiraceae family (such as *Drea*, *Coprococcus*, and *Blautia*) or the Coriobacteriaceae + Bifidobacteriaceae family (such as *Collinsella*), which are positively correlated with IR (Figure 3C). Conversely, they were negatively correlated with several genera of the Bacteroideales + Clostridiales order (such as *Aristipes*, *Bacteroides*, and *Parabacteroides*), which are negatively correlated with IR. These findings suggest that IR-related metabolites are positively or negatively associated with gut microbiota.

[0066] The IR and IS-related bacteria suggested in this study, including genera such as Dorea, Corinthella, Aristipes, and Bacteroides, have also been shown to be part of the gut microbiome of obese and lean individuals in Chinese cohorts (Liu, R., Hong, J., Xu, X. et al. Gut microbiome and serum metabolome alterations in obesity and after weight-loss intervention. Nat Med 23, 859-868 (2017).). Furthermore, previous studies have suggested that several Lachnospiraceae species are involved in carbohydrate fermentation (HJ Flint, KP Scott, SH Duncan, P. Louis, E. Forano, Microbial degradation of complex carbohydrates in the gut. Gut Microbes. 3 (2012), M. Vacca et al., The Controversial Role of Human Gut Lachnospiraceae. Microorganisms. 8, 573 (2020).).In contrast, according to LA David et al. (Diet rapidly and reproducibly alters the human gut microbiome. Nature. 505, 559-563 (2014)), the genus Aristipes is resistant to bile acids, increases with an animal-based diet rather than a polysaccharide-rich diet, and was abundant in IS subjects (BD Piening, W. Zhou, TL Mclaughlin, GM Weinstock, MP Snyder, Integrative Personal Omics Profiles during Periods of Weight Gain and Loss. Cell Syst. 6, 157-170.e8 (2018)) and lean mice (VK Ridaura et al., Gut microbiota from twins discordant for obesity modulate metabolism in mice. Science. 341 (2013), doi:10.1126 / science.1241214). These previous studies suggest that microbial functions related to major nutrients may be involved in the production of IS and IR-related metabolites.

[0067] To address this hypothesis, we investigated the function of gut microbiota using shotgun metagenomics. KEGG orthologues positively associated with fecal carbohydrate metabolites were indeed abundant in carbohydrate metabolism-related pathways such as the phosphotransferase system (PTS), starch and sucrose metabolism, and galactose metabolism (Figure 3D). Furthermore, these pathways were positively associated with HOMA-IR, TG, and BMI, and negatively associated with adiponectin. In contrast, KEGG pathways involving amino acids and energy metabolism (e.g., glycolysis / gluconeogenesis and the citric acid cycle) were negatively associated with fecal carbohydrate metabolites. Consistently, IR-associated bacteria such as Drea and Blautia were positively associated with KEGG pathways related to carbohydrate metabolism, such as PTS and galactose metabolism, while IS-associated bacteria such as Aristipes were positively associated with the citric acid cycle (Figures 3E-G), suggesting an overall change in fecal carbohydrate metabolism in IR. These consistent correlations between metagenomic and metabolome analyses suggest that increased fecal monosaccharides are partly attributable to abnormalities in microbial function in intravascular respiration (IR).

[0068] Example 4: Role of IS and IR-related microorganisms in experimental models These findings from human multi-omics analysis revealed a link between gut microbial fermentation products and the host pathophysiology of IR. To address the causal relationship between gut microbiota, carbohydrate utilization, and metabolic diseases, we first analyzed carbohydrate metabolism in bacterial cultures of several human fecal microorganisms representative of IS and IR. PCA plots show that the overall metabolome of the cell-free supernatant is distinctly different between the Aristipes group associated with IS, namely Aristipes finegordi (AF, JCM16770) and Aristipes indistinctus (AI, JCM16068), and the Dreah group associated with IR, namely Dreah formisigenerance (DF, JCM31256) and Dreah longicatena (DL, JCM11232) (Figure 4A). Disaccharides (i.e., maltose, lactose, trehalose, and cellobiose) were substantially low in the two Dreia groups (Figure 4B-E). Glucose was consumed by all bacteria, but other monosaccharides such as mannose were relatively low in the two Aristipes groups (Figure 4F-I). In particular, citric acid cycle intermediates such as succinic acid, malic acid, 2-ketoglutaric acid, and fumaric acid were abundantly produced by Aristipes strains but not in Dreia strains (Figure 4J-M), which is consistent with metagenomic analysis results (Figure 3G). These findings indicate that Aristipes and Dreia, representing IS and IR-associated microorganisms, metabolize carbohydrates in different ways.

[0069] Finally, the therapeutic effects of IS-related microorganisms on IR in an obese mouse model were investigated. C57BL / 6N mice fed a high-fat diet (HFD) were force-administered either AI, Aristipes finegordi (AF), Bacteroides citiotaomicron (BT), Bacteroides xylanisorbens (BX), or PBS 6-9 times over 3 weeks. Weight gain associated with high-fat dieting after bacterial administration was slightly improved with AI administration (Figure 5A). More significantly, insulin resistance in HFD-fed mice was improved with AI, AF, and BT, as measured by insulin loading tests (Figures 5B and C). Cecal concentrations of fructose and mannose were significantly reduced with AI (Figures 5D and E). Furthermore, cecal concentrations of fructose, mannose, glucose, tagatose, and psicose were positively correlated with AUC in insulin loading tests (Figures 5F-J). Postprandial blood glucose levels four weeks after the start of administration of Aristipes indistinctus (AI), Aristipes finegordi (AF), Bacteroides cetyoteomicron (BT), Bacteroides xylanisorbens (BX), Parabacteroides meldae (PM), Clostridium spirome (CS), Faecalibacterium prausnitzii (FP), or Vehicle (bacterial solvent) were significantly reduced with AI, AF, and BT (Figure 5K).

[0070] Furthermore, Western blots (Figure 6A) of Akt protein phosphorylated at serine position 473 (S473) and total Akt protein in the liver (top) and epididymal fat (eWAT, bottom) 5 minutes after insulin injection into mice administered with Aristipes indistinctus (AI), Aristipes finegordii (AF), or Vehicle (bacterial solvent) of the same Akt protein, and relative values ​​of normalized p-Akt band densitometry measurements to Akt band densitometry measurements in the liver (Figure 6B) and epididymal fat (Figure 6C) based on densitometry analysis of the Western blots, showed that AI and AF enhanced Akt protein phosphorylation more than Vehicle. This result demonstrates at the signaling level that AI and AF improve insulin resistance.

[0071] Furthermore, mice administered with Aristipes indistinctus (AI) showed elevated serum HDL-cholesterol (HDL-C) levels (Figure 7A), decreased triglyceride (TG) levels (Figure 7B), and elevated adiponectin levels (Figure 7C) compared to mice administered with Vehicle (bacterial solvent). These results biochemically demonstrate that AI increases serum components that improve insulin resistance and reduces serum components that worsen insulin resistance.

[0072] Furthermore, mice administered with Aristipes indistinctus (AI) showed significantly lower serum levels of fructose (Figure 8A), rhamnose (Figure 8B), and glucose-6-phosphate (Figure 8C) compared to mice administered with Vehicle (a solvent for the bacterial cells). This result biochemically demonstrates that AI reduces serum sugar concentrations that contribute to the worsening of insulin resistance. In particular, fructose concentrations were also reduced in the cecum (Figure 5D), confirming that AI reduces sugars that contribute to fat accumulation.

[0073] Overall, these findings are consistent with human cohort data showing that carbohydrate metabolism in the gut microbiota by specific microorganisms is associated with IR and metabolic abnormalities. [Industrial applicability]

[0074] The novel anti-obesity and anti-diabetic probiotics of the present invention are extremely useful because, unlike previous probiotics, they can provide new therapeutic and preventive methods that are more closely based on the mechanisms of metabolic disorder onset and exacerbation, as they are bacteria identified from large-scale human studies and are consistent with human intestinal metabolism. The probiotics of the present invention can be used as live bacteria or lyophilized preparations and are useful in that they can be used not only as preventive and / or therapeutic agents but also as foods or dietary supplements.

Claims

1. A composition for preventing and / or treating type 2 diabetes or obesity, comprising bacteria of the genus Alistipes and / or bacteria of the genus Bacteroides.

2. The composition according to claim 1, wherein the bacteria of the genus Alistipes are Alistipes indistinctus or Alistipes finegoldii.

3. The composition according to claim 1, wherein the bacteria of the genus Bacteroides are Bacteroides thetaiotaomicron.

4. The composition according to claim 1, 2 or 3, wherein the composition is orally administrable.

5. The composition according to claim 4, wherein the composition is a food or a dietary supplement.

6. The bacteria are obtained by the following steps: (i) (a) profiling the gut microbiota and bacterial metabolites for samples containing feces of healthy subjects and subjects, or (b) profiling the gut microbiota and bacterial metabolites for samples containing feces of healthy subjects and subjects, and profiling plasma metabolites for samples containing plasma of the healthy subjects and the subjects, (ii) performing a correlation analysis between the gut microbiota and the bacterial metabolites, or performing a correlation analysis between the gut microbiota and the bacterial metabolites and a correlation analysis between the gut microbiota and the plasma metabolites, and (iii) determining, based on the results of the correlation analysis in step (ii), the bacteria showing a significant difference as bacteria having an effect on the host and are identified by a large-scale test of human samples, wherein the subjects are obese subjects and / or subjects with impaired glucose tolerance. The composition according to any one of claims 1 to 3.

7. The composition according to claim 6, wherein the bacterial metabolite is a carbohydrate metabolite.

8. A composition for detecting type 2 diabetes or obesity, comprising bacteria of the genus Dorea or a part of the genomic DNA of the bacteria.

Images