Gut microbiota and tuberculosis

Abstract

Tuberculosis (TB), caused by Mycobacterium tuberculosis (MTB), remains a significant global health challenge. Recent advancements in gut microbiota (GM) research have shed light on the intricate relationship between GM and TB, suggesting that GM alterations may influence host susceptibility, disease progression, and response to antituberculosis drugs. This review systematically synthesizes and analyzes the current research progress on the relationship between GM and TB, focusing on six key aspects: (1) bidirectional effects between GM dynamics and TB progression; (2) the interaction between GM and anti-TB drugs; (3) GM and TB immune response; (4) GM as a potential target for diagnosis and treatment of TB; (5) multi-omics and artificial intelligence (AI) technologies in GM-TB research; (6) current challenges and future directions in GM-TB research. We highlight the bidirectional nature of the GM-TB interaction, where MTB infection can lead to GM dysbiosis, and changes can affect the host's immune response, contributing to TB onset and progression. Advanced molecular techniques, such as next-generation sequencing and metagenomics, along with AI, play pivotal roles in elucidating these complex interactions. Future research directions include investigating the relationship between GM and TB vaccine efficacy, exploring GM's potential in TB prevention, developing microbiome-based diagnostic and prognostic tools, and examining the role of GM in TB recurrence. By addressing these areas, we aim to provide a comprehensive perspective on the latest advancements in GM and TB research and offer insights for future studies and clinical applications. Ultimately, the development of novel microbiome-based strategies may offer new tools and insights for the effective control and management of TB, a disease that continues to pose a significant threat to public health.

Keywords: Mycobacterium tuberculosis; artificial intelligence; gut microbiota; microbiome‐based diagnostics; omics technologies; tuberculosis.

FIGURE 1

Relationship between gut microbiota (GM) dysbiosis and tuberculosis (TB) progression. The figure illustrates the association between GM dysbiosis and TB pathogenesis. Dysbiosis is characterized by a loss of biodiversity, particularly a depletion of key short‐chain fatty acid (SCFA)‐producing taxa such as Blautia spp., Roseburia spp., Ruminococcus spp., Bifidobacterium spp., and Eubacterium spp. This is accompanied by an overrepresentation of potentially pathogenic genera, including Bacteroides spp., Prevotella spp., Enterococcus spp., and Fusobacterium spp. These compositional shifts contribute to increased intestinal permeability, disruption of mucosal immune homeostasis, and impaired host immune responses against Mycobacterium tuberculosis, ultimately facilitating disease progression. ATD, antituberculosis drugs.

FIGURE 2

Comparative characterization of healthy versus pathogenic gut microbiota states. This figure highlights the contrasting features of a healthy versus dysbiotic gut microbiota. In healthy state, commensal bacteria such as Lactobacilli and Bifidobacteria reinforce intestinal barrier function through the production of bacteriocins and short‐chain fatty acids (SCFAs), particularly butyrate. These microbes preferentially colonize the mucosal layer, prevent pathogenic colonization, and modulate host immunity by promoting dendritic cell migration and the release of gut‐derived hormones. In pathogenic state, microbial diversity is reduced (1), with a concomitant overrepresentation of mucin‐degrading bacteria (2) and a loss of beneficial SCFA‐producing species (3). This altered microbial composition compromises intestinal epithelial integrity and enables translocation of pathogens such as Mycobacterium tuberculosis (MTB), which contribute to tissue damage via toxin‐mediated mechanisms (4). The reduction in SCFA‐producing bacteria and rise in pathogenic Enterobacteriaceae promote immune dysregulation through lipopolysaccharides (LPS)‐mediated activation of Toll‐like receptor 4 (TLR4) and G‐protein signaling. This cascade triggers a chronic inflammatory response characterized by elevated cytokine production, neutrophil degranulation, and impaired regulatory T cell function, ultimately disrupting immune homeostasis.

FIGURE 3

Bidirectional interactions between anti‐TB drugs (ATD) and gut microbiota (GM), and their impact on GM homeostasis. This figure illustrates: (1) The detrimental effects of ATD on GM compositions and functions, mediated through multiple mechanisms including the induction of intestinal inflammation, disruption of epithelial barrier integrity, metabolic perturbations, and direct bactericidal activity. (2) How GM dysbiosis reciprocally modulates the pharmacodynamics and toxicity of ATD, through mechanisms including altered drug metabolism, impaired immune modulation, reduced drug bioavailability, enhanced hepatotoxicity, and the facilitation of antimicrobial resistance. (3) Emerging therapeutic strategies aimed at restoring GM homeostasis, such as probiotics supplementation and administration of magnesium isoglycyrrhizinate, which may attenuate inflammation, promote barrier repair, and re‐establish microbial balance. Overall, this figure highlights the complex, bidirectional crosstalk between the GM and ATD, and suggests the potential of microbiota‐targeted interventions to optimize antituberculosis therapy. FMT, fecal microbiota transplantation. MgIG, magnesium isoglycyrrhizinate.

FIGURE 4

Multimodal regulation of host immunity by the gut microbiota (GM). The GM modulates the host immune system through multiple pathways. First, microbial components are recognized by pattern recognition receptors (e.g., TLRs) expressed on innate immune cells, activating downstream signaling cascades such as the phosphoinositide 3‐kinase (PI3K) pathway. This leads to the inactivation of glycogen synthase kinase 3β (GSK‐3β) and subsequent activation of cyclic adenosine monophosphate (cAMP) response element‐binding protein (CREB)‐dependent transcription of anti‐inflammatory genes, promoting the release of cytokines and other immunomodulatory factors. Second, microbial metabolites, such as SCFAs, bind to G protein‐coupled receptors (GPCRs) to drive the expansion of mucosal regulatory T cells (Tregs), thereby suppressing pro‐inflammatory responses. SCFAs also promote macrophage polarization through the gut–lung axis and enhance the memory function of CD8 + T cells, while stimulating B cell‐mediated secretion of immunoglobulin A (IgA), thereby strengthening mucosal and systemic immunity. In addition, specific bacteria taxa (e.g., Bacteroides, Clostridium, and Prevotella) promote the differentiation of CD4⁺ T cells into T helper 17 (Th17) and Tregs, contributing to the regulation of adaptive immunity. Moreover, bacterial polysaccharides can regulate T helper cell fate through activating signal transducer and activator of transcription 4 (STAT4) signaling in the presence of IL‐12 to promote Th1 differentiation, or STAT6 signaling under IL‐4 to promote Th2 responses, thus maintaining Th1/Th2 homeostasis.

FIGURE 5

Gut microbiota (GM) as a diagnostic and therapeutic target in tuberculosis (TB). Emerging evidence highlights the GM as a key modulator of TB pathogenesis and a promising target for diagnosis and treatment. At the molecular level, the GM influences TB development through systemic circulation, with alterations in microbial composition, fungal–bacterial interactions, gut‐immune crosstalk, and metabolite profiles offering diagnostic potential. Elucidating these mechanisms is critical for advancing TB therapeutics. Microbiota‐targeted interventions, including probiotics, prebiotics, synbiotics, and FMT can restore GM homeostasis, reduce treatment‐associated complications, and improve gut and immune health.

FIGURE 6

Overview of the next‐generation sequencing (NGS) workflow and its applications. (A) Schematic representation of the core stages in the NGS workflow: DNA fragmentation and library preparation, high‐throughput sequencing, and bioinformatics analysis and data interpretation. (B) Illustration of the diverse applications of NGS across genomics and biomedical research.

FIGURE 7

Applications of NGS in gut microbiota‐tuberculosis (TB) research. (A) Microbial composition analysis: Relative abundances of intestinal bacterial phyla in TB patients versus healthy controls (HC) were compared using the Wilcoxon rank‐sum test. Differentially abundant taxa were identified by linear discriminant analysis effect size (LEfSe) with LDA score > 4 and p < 0.05. Adapted with permission from [24], Copyright 2024, BMC. (B) Metabolic pathway analysis: Heatmap shows differential metabolic pathway abundance between TB and HC groups based on NGS‐derived functional profiling. Pathways were ranked according to consensus functional classification; statistical significance was determined using the Wilcoxon rank‐sum test with false discovery rate (FDR) < 0.1. Abundances were standardized as row Z‐scores. Adapted with permission from [27], Copyright 2019, Frontiers. (C) Longitudinal microbiota dynamics during anti‐TB therapy: Temporal changes in the average relative abundance of bacterial families were tracked using stool sample sequencing data across multiple treatment time points and experimental groups. Bacterial families are grouped by phylum and class in the accompanying color key. Time points are indicated along the x‐axis, grouped by treatment conditions. Adapted with permission from [29], Copyright 2017, BMC.

FIGURE 8

An overview of the host–microbe interactions in TB progression. The bidirectional influence of the lung–GM axis on host immunity and MTB survival. On the left (green), a balanced microbiome supports effective immune surveillance through enhanced immune signaling, a metabolically favorable state (characterized by elevated glycolysis and reduced oxidative phosphorylation), maintenance of stable immune epigenetic landscapes, and tightly regulated cytokines production, promoting MTB clearance. In contrast, the right side (pink) depicts a dysbiotic microbiome that contributes to impaired host defense through disrupted immune signaling (e.g., reduced cell–cell communication), metabolic dysregulation, and skewed pro‐/anti‐inflammatory responses, thus supporting MTB persistence and immune evasion. GM, gut microbiota; TB, tuberculosis.

FIGURE 9

Overview of the general metabolomics workflow. The metabolomics pipeline comprises several key steps, beginning with sample collection and preparation, followed by analytical profiling using high‐resolution platforms such as chromatography, mass spectrometry (MS), nuclear magnetic resonance (NMR) spectroscopy. Data acquisition includes pre‐processing, clean‐up, and statistical analysis. The final step involves the identification and interpretation of metabolic signatures to reveal underlying biological processes and disease mechanisms. Some of the illustrative elements were created using BioRender (https://BioRender.com).

FIGURE 10

Applications of artificial intelligence (AI)/machine learning (ML) in microbiome research: from data collection to implementation. The diagram outlines the workflow for AI/ML integration in microbiome studies. The process begins with sample collection and processing, followed by multi‐omics profiling approaches (e.g., DNA, RNA, metabolite, and protein‐based analyses) to generate high‐dimensional datasets (e.g., taxonomic composition, alpha and beta diversity, functional profiling). Unsupervised learning methods, such as clustering and dimensionality reduction, are used to explore microbial community structure and diversity. Supervised learning algorithms (e.g., classification and regression) enable prediction of clinically relevant outcomes, such as disease diagnosis and treatment response. Deep learning (DL) methods, using architectures such as neural networks and Long Short‐Term Memory (LSTM) networks, further enhance the ability to model complex, nonlinear relationships within microbiome data.

FIGURE 11

Integrative multi‐omics framework for investigating gut microbiota‐tuberculosis (GM–TB) interactions. This schematic illustrates a comprehensive strategy for multi‐omics integration in GM‐TB research, combining metabolomics, proteomics, transcriptomics, and genomics approaches. The framework incorporates key study parameters, including sample size, phenotypic data, disease models, and clinical or experimental disease characteristics, to facilitate a systems‐level understanding of TB pathogenesis. Central themes include genomic instability and mutations, immune modulation, microbial dysbiosis, and the mechanistic interplay between the lung and gut microbiota.

FIGURE 12

Challenges and Limitations in gut microbiota‐tuberculosis (GM‐TB) Studies. Current GM‐TB research faces several key challenges and limitations. Observational study designs limit the ability to infer causality, in contrast to the more rigorous evidence provided by randomized controlled trials (RCTs). Small sample sizes weaken statistical power and restrict the generalizability of findings. Population heterogeneity, including genetic background, environmental exposures, and lifestyle factors, is often underrepresented. Technical limitation in microbiome analysis, such as analytical complexity, inconsistence in computational methods, and the lack of standardization, further complicate data interpretation. Translational gaps between preclinical models and human studies remain a significant barrier. Lastly, the integration of multi‐omics data, including metabolomics and proteomics, poses analytical challenges due to differences in data structure, scale, and dimensionality.

FIGURE 13

Future directions and research priorities in gut microbiota‐tuberculosis (GM‐TB) studies. Future research directions in GM‐TB studies encompass elucidating the role of GM in modulating vaccine efficacy and shaping host immune responses, along with exploring the therapeutic potential of probiotics, prebiotics, and FMT in TB prevention. Efforts are also directed towards developing microbiome‐based diagnostic tools by identifying and validating microbial biomarkers and creating rapid, noninvasive diagnostic platforms. Treatment optimization is a growing focus, emphasizing the need to consider GM dynamics to support personalized therapeutic approaches, including tailored antimicrobial regimens. Understanding the contribution of the GM to TB recurrence and latent tuberculosis infection (LTBI) is also critical, with research aimed to identifying predictive markers and recurrence risk factors. Advanced multi‐omics data integration is being facilitated through the development of sophisticated algorithms and bioinformatics tools. Lastly, standardization of microbiome analysis workflows, spanning sample processing and computational analysis, is crucial to ensure reproducibility, comparability, and translational relevance across studies.