Linking gut dysbiosis with metabolic perturbations in diabesity

Abstract

Diabetes mellitus (DM) is a pandemic metabolic disorder affecting ~537 million adults globally, with ~783 million projected by the year 2045. Type 1 DM (T1DM) and type 2 DM (T2DM) are the two major forms of DM and have been represented by distinct pathophysiological phenotypes. Diabesity, the convergence of T2DM and obesity, accounts for ~90-95% of DM cases and represents a major healthcare burden. Both metabolic dysfunction and gut microbiome dysbiosis have been observed in diabesity, however, their combined role in the causation of the same still remains to be understood. This thesis investigates the temporal relationship between gut microbiome dysbiosis and serum metabolic perturbations in diabesity through an integrated approach combining NMR-based metabolomics and 16S rRNA metagenomic analyses in murine models. In addition to this, a comparative account of the metabolic profile between the T1DM and diabesity models has also been presented. Male C57BL/6J mice (n = 6-8/group) were used for the induction of: (1) T1DM using streptozotocin (STZ, 55 mg/kg IP, 0-5 injections), that were monitored at 15-day and 60-day time-points, and (2) Diabesity using high fructose/high fat diet (HF/HFD) intervention (60% energy from fat + 10% w/v fructose) for 4, 8, 12, and 16 weeks and chow diet (ND; 60% energy from fat) which served as controls. The establishment of hyperglycemia and insulin resistance were established by performing the oral glucose tolerance tests (OGTT) and the insulin tolerance tests (ITT). Following this, 1H NMR spectroscopy (600 MHz, Bruker AVANCE III) was used to identify and profile the abundant and aqueous metabolites from serum and adipose tissues. Fecal samples were used to perform 16S rRNA sequencing (HiMedia MBS206B). Metabolomic and metagenomic data were integrated via MetOrigin 2.0 and analyzed using Partial Least Squares Discriminant Analysis (PLS-DA), Analysis of Variance (ANOVA), Metabolite Set Enrichment Analysis (MSEA), and Metabolomics Pathway Analysis (MetPA) (p  0.05, FDR  0.05, impact  0.2). The HF/HFD-induced diabesity model demonstrated a temporal metabolic progression characterized by early disruption of starch and sucrose metabolism at 4 weeks, involving 14 overlapping metabolic pathways; dominance of taurine and hypotaurine metabolism at 8–12 weeks; and pronounced alterations in pyruvate and glycerophospholipid metabolism at 16 weeks, indicative of gut–liver axis impairment. Using the information from the significantly perturbed metabolites from the serum, a Diabesity Metabolic Factor (DMF), comprising of lactate, o-phosphocholine, choline, and AMP, was developed that helped discriminated the early and the advanced disease stages. Likewise, the perturbed metabolites from the adipose tissue were used to develop Adipose-Derived Metabolic Indicator (ADMI), consisting of GPC, pyroglutamate, valine, choline, and glutamate, which helped identify the adipose metabolic remodeling that happens during the establishment of diabesity in the system. Interestingly, dysbiosis observed in the HF/HFD-induced diabesity model was characterized by depletion of short chain fatty acid (SCFA)-producing taxa (Faecalibacterium, Roseburia) and increased relative abundance of potentially pathogenic families Yersiniaceae and Micrococcaceae. Network analysis revealed progressive uncoupling of host-microbe metabolic coordination. The STZ-induced T1DM model exhibited a distinct metabolic profile, with the T1DM-Diagnostic Molecular Fingerprint (DMF) group showing alterations in key serum metabolites, namely, leucine, lysine, mannose, choline, and lactate. Interestingly, mannose emerged as an early marker of β-cell stress and branched-chain amino acids (BCAA) indicated a catabolic state. The adipose tissue- Adipocyte Molecular Fingerprint (AMF) profiles (represented by leucine, lysine, valine, choline, and o-phosphocholine), which were distinct from diabesity-driven patterns, highlighted dysregulation of lipid and amino acid metabolism. This research establishes that gut microbe dysbiosis drives systemic metabolic dysfunction in diabesity through stage-specific, coordinated shifts in microbiome-metabolome interactions. Early HF/HFD-induced disruptions favor carbohydrate processing pathway dysfunction, transitioning to gut-liver axis impairment and finally global metabolic collapse. Integration of metagenomics and metabolomics via MetOrigin 2.0 provided mechanistic insights beyond correlation, distinguishing co-metabolized pathways. The identified metabolic signatures (DMF, ADMI) can serve as biomarkers for disease staging and therapeutic monitoring. Divergent metabolic underpinnings between T1DM, characterized by BCAA elevation and glutathione dysfunction, and diabesity, driven by lipotoxicity and dysbiotic pathways, despite shared hyperglycemia, underscore disease-etiology-specific trajectories. These findings support stage-specific treatment in diabesity, such as early targeting of starch and sucrose metabolism, taurine or bile acid–based therapies at intermediate stages, and multi-omics-guided interventions later in disease. With future validation in human studies and tissue-specific and multi-omics profiling, these approaches could enable personalized and focused diabesity management.