Gut Microbiota-Derived Metabolites as Immune Modulators in Aging and Age-Related Chronic Inflammatory Diseases

Abstract

Age-associated dysbiosis, marked by shifts in the composition of gut microbiota and gut microbiota-derived metabolites (GMDMs), is increasingly implicated in driving systemic low-grade inflammation during aging. The disrupted GMDM pools, including altered levels of short-chain fatty acids (SCFAs), secondary bile acids (BAs) and tryptophan (Trp) metabolites, lead to mucosal barrier dysfunction, immunometabolic dysregulation, and modulation of innate and adaptive immune cells. In turn, this cascade of events drives tissue degeneration, chronic inflammation, and the onset of age-related diseases (ARDs). Here, we summarize the immunomodulatory role of major GMDMs and how aging may increase susceptibility to ARDs through changing GMDMs. We then explore the latest findings linking altered GMDM profiles to immune dysfunction across major gut-organ axes, including the liver, adipose tissue, muscle, and brain. Last, we highlight recent advances in harnessing GMDMs as geromedicine to improve aging parameters and discuss the potential of artificial intelligence (AI) in accelerating the bench-to-bedside translation of GMDM research. Together, this review positions GMDMs as actionable targets in a dysbiosis-driven network of immune aging, offering new possibilities for the development of healthspan-extending precision geromedicine.

1. Introduction

With over 1.4 billion people worldwide projected to be age 60 or older by 2030¹, the rapid expansion of aging population poses substantial challenges for healthcare and financial systems. Several core biological processes driving age-related physiological decline, named “the hallmarks of aging”, have been identified². Among these hallmarks, gut dysbiosis, characterized by shifts in the composition of gut microbiota and their derived metabolites, has gained prominence in recent decades for its broad involvement in various age-related diseases (ARDs), like neurodegenerative diseases, metabolic syndromes, and sarcopenia³.

A key mechanism by which age-associated dysbiosis accelerates ARDs is through chronic inflammation, another hallmark of aging⁴. Gut microbiota-derived metabolites (GMDMs), such as short-chain fatty acids (SCFAs), secondary bile acids (BAs), certain vitamins (vitamin K2 and Bs), and several amino acid metabolites (e.g., indole derivatives), are microbial products synthesized de novo or transformed from diet or host-derived molecules, with broad immunometabolic and neuroendocrine effects⁵. Dysbiosis alters GMDM profiles, disrupts the gut barrier, and drives systemic inflammation via endotoxemia, contributing to neuroinflammation, senescence, mitochondrial dysfunction, and DNA damage across multiple organs^6,7. Notably, in 2025, additional aging hallmarks have been proposed, including psychosocial isolation and extracellular matrix changes⁸, revealing unexplored pathways by which dysbiotic GMDM profiles can promote chronic inflammation.

The potent immunomodulatory effects of GMDMs raise a crucial question: can GMDMs be harnessed as therapies to control chronic inflammation and extend healthspan in aging populations? While probiotics, prebiotics, and postbiotics have been extensively studied for their anti-inflammatory potential, recent studies suggest that GMDMs can also shape drug absorption, metabolism, and toxicity in older individuals by modulating immune and endocrine circuits⁹. Despite increasing recognition of their therapeutic potential and pathogenic links to multiple hallmarks of aging, less is known about how dysbiotic GMDM profiles drive immune dysfunction and promote ARDs. In 2021, we reviewed the immune modulation by dysbiotic microbiota in obesity¹⁰; here, we extend this framework to aging, examining diverse mechanisms through which GMDMs contribute to immune dysfunction in aging across several key gut-organ axes, including the liver, adipose tissue, muscle, and brain, where bidirectional feedback loops between GMDMs and host immunity reinforce inflammatory aging, known as inflammaging, and ARD progression. We also discuss targeting GMDMs as therapeutics in gerosciences, using artificial intelligence (AI) and multi-omics, including metagenomics and metabolomics across aging populations, paving the way for the development and clinical translation of precision geromedicine and personalized interventions.

2. Age-associated microbiome remodeling and the interplay between dysbiosis and immune aging

An unresolved question related to GMDMs in aging is whether age-associated gut dysbiosis is a cause or consequence of immune aging and dysfunction. Accumulating evidence indicates that this relationship is best understood as bidirectional and self-reinforcing. Within this context, immune aging and microbial remodeling evolve in parallel, with progressive changes in GMDM availability acting as molecular intermediates that link microbial ecology to age-related immune phenotypes^11–13.

Aging is associated with heterogeneous changes in gut microbiome composition. In older adults, particularly those with frailty and co-morbidities, numerous studies have reported reduced microbial diversity and depletion of taxa linked to maintenance of epithelial integrity and immunoregulatory capacity, including Akkermansia muciniphila, Faecalibacterium prausnitzii, Roseburia hominis, Eubacterium rectale, and others^11,14,15. Importantly, this pattern is not universal: Centenarians and individuals undergoing “healthy aging” often retain a distinct and functionally adapted microbial community, highlighting that microbiome aging reflects physiological and immunological state rather than chronological age alone^15–17. Beyond compositional shifts, aging is marked by functional reprogramming of microbial metabolism, including reduced SCFA biosynthesis, altered BAs transformation, and changes in Trp-derived metabolites^18,19. These functional changes are particularly relevant to immune aging^11,20, as immune cells respond primarily to metabolite-mediated signaling cues, positioning GMDMs as critical effectors of host-microbe communication during aging.

Immune and gut barrier aging can actively contribute to dysbiosis. Age-related impairments in epithelial barrier repair, mucus production, and antimicrobial peptide secretion, in part from increased age-related apoptosis of intestinal lining cells, coupled to impaired IgA-mediated microbial targeting and chronic low-grade inflammation, including increased IL-1β and IFN-γ, reshape the intestinal niche in ways that favor pathobiont expansion and loss of beneficial commensals^21–24. These immune-driven changes are further reinforced by aging-associated environmental pressures, including dietary narrowing, polypharmacy, recurrent infections, and reduced physical activity, all of which exert selective pressure on microbiota composition^25,26. Concurrently, substantial experimental and clinical evidence supports dysbiosis as a contributor to immune aging. Age-associated gut dysbiosis is linked to diminished production of anti-inflammatory metabolites, increased epithelial permeability, and heightened innate immune activation, collectively promoting systemic inflammaging and impaired adaptive immunity^20,27. Notably, microbiota transfer studies demonstrate that transplantation of aged-associated microbiota into young or germ-free hosts is sufficient to alter GMDM and induce inflammaging-like immune features^28,29, providing causal support for a microbiome-driven component of immune aging.

Together, these findings support a model in which gut microbiome remodeling and immune aging are reciprocally linked. This bidirectional perspective provides an essential conceptual framework for interpreting age-associated changes in microbial metabolites and helps explain the variability observed across aging populations. Accordingly, the following section (Section 3) builds directly on this foundation by detailing how specific classes of GMDMs are altered during aging, and how these metabolite changes mechanistically regulate immune cell function, thereby shaping inflammaging and susceptibility to ARDs.

3. GMDMs and their changes in aging

Through its diverse metabolite repertoire, the gut microbiota exerts a profound influence on host immunity. These GMDMs modulate immune cell development, function, and plasticity, with emerging relevance to chronic inflammation, aging, and ARDs. Below, we delineate the immunomodulatory roles of key classes of GMDMs and their altered dynamics during chronic inflammation and aging.

3.1. Short-chain fatty acids (SCFAs)

SCFAs, primarily acetate, propionate, and butyrate, are microbial metabolites generated from the fermentation of dietary fiber³⁰. SCFAs-producing consortia can shape the host immunity while being shaped by it reciprocally. During aging, chronic low-grade inflammation, gut barrier dysfunction, and immunosenescence disturb this delicate balance, resulting in gut dysbiosis and reduced capacity for SCFAs biosynthesis, as seen in murine models and aging human subjects^31,32.

SCFAs exert broad immunomodulatory effects via multifaceted mechanisms, such as through G-protein-coupled receptors (GPR41, GPR43, GPR109A), histone deacetylase (HDAC) inhibition, and downstream metabolic reprogramming (Table 1). Mechanistically, SCFAs may modulate immune responses via HDAC inhibition^33,34 and GPR43-mediated signaling³⁵, which in general reduces inflammation by suppressing the NF-κB pathway. Interestingly, activation of GPR43 and GPR109A by SCFAs also contributes to the activation of the NLRP3 inflammasome through K⁺ efflux, which can have beneficial effects in promoting interleukin (IL)-18-mediated colonic epithelial homeostasis³⁶. SCFAs may also modulate immune responses through the mTOR-S6K and MAPK signaling pathways by HDAC inhibition, and can promote differentiation of effector (Th1, Th17) or regulatory T cells (Tregs)³⁷, depending on the immunological milieu³⁸ (Figure 2). These receptor- and pathway-specific effects provide mechanistic insight into how SCFAs actively shape immune function rather than acting as non-specific anti-inflammatory metabolites. In immune cells, SCFAs exert broadly anti-inflammatory effects by promoting Tregs, inhibiting IL-6, and tumor necrosis factor-alpha (TNF-α), while increasing IL-10 production^34,39 and regulatory B cells (Bregs) frequency across various disease models^40,41. SCFAs also modulate antigen-specific CD8⁺ T cell function via inhibiting dendritic cell (DC)-derived IL-12 production⁴². With aging, microbial dysbiosis reduces SCFA-producing taxa, leading to lower intestinal SCFAs, predisposing to ARDs and inflammaging (Table 1)^31,43. Importantly, SCFA supplementation in aged mice restores gut barrier function, enhances mucosal immunity, and decreases systemic inflammation⁴⁴. Clinical studies also link reduced fecal SCFAs with frailty and cognitive impairment in aging populations, revealing their role in preserving immunometabolic homeostasis and healthspan^45,46.

Table 1.

Immunomodulatory roles of major gut microbiota-derived metabolites (GMDMs) in aging and age-related diseases.

GMDMs	Major Gut Microbiota Producers	Immune Targets and Mechanisms	Changes in Ageing	Key References
Short-chain fatty acids (SCFAs): primarily acetate, butyrate, propionate	Akkermansia, Bifidobacterium, Coprococcus, Eubacterium, Faecalibacterium, Roseburia	Activates GPR41/43/109A, inhibits HDACs; promotes Tregs and IL-10 induction; suppresses Th17 and pro-inflammatory cytokine production; modulates DC. ILC2 and CD8+ T cell function; enhances barrier integrity	↓ SCFA-producing taxa, ↓ SCFA production; impaired mucosal barrier, increased inflammaging	31,34,39,44,45,212,234
Tryptophan metabolites Kyn, kynurenic acid, 3-HAA, tryptamine, indole derivatives (lAld, ICA, IPA, ILA, IAA)	Escherichia coli, Clostridium spp., Ruminococcus gnavus, Bifidobacterium spp., Lactobacillus spp., Peptostreptococcus spp., Prevotella	Acts via AhR, PXR, and GPR35; regulates Th17/Treg balance; modulates GLP-1 secretion and macrophage-epithelial interactions; regulates oxidative stress	↑ IDO1/TDO activity; ↑ Kyn/Trp ratio; ↓ tryptophan and indoles; impaired epithelial integrity, increased inflammaging	47,48,56,66,215,313
Bile acids (BAs) Primary: CA, CDCA; Secondary: DCA, LCA, isoallo-LCA, 3-oxoLCA	Bacteroides, Actinobacteria, Akkermansia, Lactobacillus spp., Clostridium spp., Extibacter spp., Ruminococcus gnavus, Peptostreptococcus spp., Proteobacterium spp.	Acts via FXR, LXR, TGR5; Secondary BAs modulate macrophage, DC, monocyte, and T cell activity; inhibits NLRP3 inflammasome; regulates Th17/Treg balance	↑ primary BAs; ↓secondary BAs (DCA, LCA), ↑ neuroinflammation; impaired intestinal barrier integrity	70–73,89,91,346
Trimethylamine N-oxide (TMAO)	Desulfovibrio, Firmicutes, Actinobacteria (TMA-producing taxa)	Acts via PERK, TAAR5; induces oxidative stress, inflammation, endothelial activation; drives Ml macrophage polarization; impairs efferocytosis; enhances platelet activation	↑ with aging; promotes oxidative stress, neuroinflammation, and cognitive decline; gut dysbiosis favors TMA-producing taxa, further exacerbating inflammaging	103,107,116,120–122,449
Branched-chain and aromatic amino acid metabolites (BCAA, AAA) BCAA: Leu, Ile, Val, AAA: Phe, Tyr derivatives	Bacteroides, Firmicutes, Parabacteroides merdae, Clostridium spp.	Modulates mTORC1 signaling, SLC7A5 transporter activity, TFH/GC B cell function; regulate oxidative stress; BCFA (isovalerate) reduces inflammation	↑ plasma BCAA/AAA; ↓ Bacteroides; ↑ mTORC1 activation and inflammaging; ↑ metabolic risk	131–135,174
Vitamins B-group (B1, B2, B6, B9, B12), K2, D, E, A	Fibrobacter succinogenes, Lactiplantibacillus plantarum, Ruminococcaceae, Bacteroides	Regulates ILC2, DC, Treg development and T cell responses; modulates cytokine responses; antioxidant and anti-inflammatory functions	↓ microbial biosynthetic pathways (folate, B12) and absorption; vitamin deficiencies (riboflavin, thiamine, and folate) promote immunosenescence	138,144,150,151,450
Fatty acids & organic acids LCFAs (PA, SA, OA, LA), organic acids (succinate, α-KG)	Lactobacillus, Monoglobus, Lachnoclostridium, Bacteroidetes	Regulates ILC3s, macrophages, B cells; succinate → IL-1β via AMPK; α-KG promotes IL-10 and longevity	↑ PA, SA; ↓ α-KG; enhanced IL-β and ROS; metabolic reprogramming of immune cells	93–95,101,102
Polyphenol-derived metabolites Urolithin A, resveratrol, quercetin, 5-(3’,4’-dihydroxyphenyl)-γ-valerolactone	Eggerthellaceae, Eubacterium ramulus, Flavonifractor plautii, Eubacterium limosum	Inhibits NF-κB; reduces ROS and SASP; induces autophagy and mitophagy; modulates macrophage and T cell metabolism; restores immune homeostasis	↓ microbial diversity for polyphenol metabolism; ↓ UA and RSV bioavailability; ↑ oxidative stress	156,158,159,161,162

Abbreviations: SCFA, short-chain fatty acid; Treg, regulatory T cell; Th, T helper; DC, dendritic cell; IL, interleukin; TNF, tumor necrosis factor; AhR, aryl hydrocarbon receptor; PXR, pregnane X receptor; FXR, farnesoid X receptor; LXR, liver X receptor; TGR5, Takeda G protein-coupled receptor 5; Kyn, kynurenine; MASLD, metabolic dysfunction-associated steatotic liver disease; CRC, colorectal cancer; PD, Parkinson’s disease; mTORC1, mechanistic target of rapamycin complex 1; AMP-activated protein kinase (AMPK); TFH, T follicular helper; GC, germinal center; BCFA, branched-chain fatty acid; ILC, innate lymphoid cell; LCFAs, long-chain fatty acids; PA, palmitic acid; SA, stearic acid; OA, oleic acid; LA, linoleic acid; α-KG, alpha-ketoglutarate; NF-κB, nuclear factor kappa-light-chain-enhancer of activated B cells; ROS, reactive oxygen species; SASP, senescence-associated secretory phenotype; UA, urolithin A; RSV, resveratrol; ARDs, age-related diseases.

Figure 2. GMDM-mediated molecular pathways alter immune responses during aging-associated dysbiosis.

Aging-associated dysbiosis reshapes GMDM profiles, characterized by reduced SCFAs, beneficial secondary BAs, and indole derivatives, alongside increased TMA/TMAO, BCAAs, and dysregulated kynurenine metabolism. Reduced SCFA signaling, through altering HDAC inhibition and GPR activation, enhances NF-κB and MAPK signaling. Concurrent reductions in mitochondrial biogenesis and oxidative phosphorylation may also prevent a restorative/reparative phenotype in cells. Microbe-derived TMA acts directly on TAAR5 and is converted into TMAO by hepatic FMO3. TMAO promotes oxidative stress, NLRP3 inflammasome activation, and NF-κB signaling, which may occur through PERK-mediated endoplasmic reticulum stress. Altered BA pools, particularly microbe-derived secondary bile acids, and reduced TGR5 and FXR signaling, modulates inflammatory responses through NLRP3 inflammasome activation and NF-κB signaling. Increased IDO1 activity elevates the Kyn/Trp ratio, enhancing Kyn-AhR signaling. AhR signaling is associated with context-dependent tolerance or immunosuppressive effects, which are further influenced by GCN2 activation by Trp depletion and GPR35-mediated signaling. Disruption of amino acid metabolism during aging increases BCAAs, driving mTORC1 overactivation, insulin resistance via S6K, and reciprocal amplification of NF-κB signaling. 3-HAA = 3-hydroxyanthranilic acid, Akt = protein kinase B, AhR = aryl hydrocarbon receptor, BA = bile acid, BCAA = branched-chain amino acid, CAMKKb = calcium/calmodulin-dependent protein kinase kinase beta, cAMP = cyclic adenosine monophosphate, FGF15/19 = fibroblast growth factor 15/19, FMO3 = flavin-containing monooxygenase 3, FXR = farnesoid X receptor, GCN2 = general control nonderepressible 2, Gdf3 = growth differentiation factor 3, GPR = G protein-coupled receptor, HDAC = histone deacetylase, IDO1 = indoleamine 2,3-dioxygenase 1, IKB = Inhibitor of nuclear factor kappa-B, IR = insulin receptor, Kyn = kynurenine, MAPK = mitogen-activated protein kinase, mTOR = mechanistic target of rapamycin, NLRP3 = NOD-, LRR- and pyrin domain-containing protein 3, NF-κB = nuclear factor kappa-light-chain-enhancer of activated B cells, NRF2/SKN-1 = nuclear factor erythroid 2-related factor 2 / SKN-1 (C. elegans ortholog), PERK = protein kinase R (PKR)-like endoplasmic reticulum kinase, PGC-1α = peroxisome proliferator-activated receptor gamma coactivator 1-alpha, PI3K = phosphoinositide 3-kinase, PPARγ = peroxisome proliferator-activated receptor gamma, PKA = protein kinase A, PXR = pregnane X receptor, ROS = Reactive oxygen species, S6K = ribosomal protein S6 kinase, SCFA = short-chain fatty acid, SHP = small heterodimer partner, SIRT1 = sirtuin 1, TAAR5 = trace amine-associated receptor 5, TMA = trimethylamine, TMAO = trimethylamine N-oxide, TGR5 = G protein-coupled bile acid receptor 1, TSC1/2 = tuberous sclerosis complex 1/2.

3.2. Tryptophan (Trp) metabolites

The interaction between immune regulation and Trp metabolism is coordinated bidirectionally, where immune-derived signals regulate the Trp metabolism via distinct catabolic routes, while Trp metabolites modulate immune cell phenotypes and function. Age-associated inflammation often drives indoleamine 2,3-dioxygenase 1 (IDO1) and tryptophan 2,3-dioxygenase (TDO) to catabolize Trp into kynurenine (Kyn) and its metabolite, kynurenic acid, impacting immune responses and promoting immunosenescence⁴⁷.

Besides the classical Kyn pathway occurring mainly in the host liver, the gut microbiota can metabolize Trp into tryptamine, indoles and their derivatives like indole-3-aldehyde (IAld), indole-3-propionic acid (IPA), indole-3-lactic acid (ILA), indole-3-acetic acid (IAA), and Kyn⁴⁸. These GMDMs activate host receptors, most prominently the aryl hydrocarbon receptor (AhR), pregnane X receptor (PXR), and GPR35, to induce immunomodulation (Figure 1A)^48,49. Mechanistically, AhR activation induces context-dependent transcriptional programs to promote tolerance or immunosuppression. AhR activation suppresses NF-κB signaling⁵⁰ and NLRP3 inflammasome activity⁵¹, promotes IL-22 production in Th17 cells to support barrier integrity⁵², and induces FoxP3⁺ Tregs differentiation⁵³, collectively, suppressing inflammation. Persistent AhR activation may be an important contributor to tumor immunosuppression^52,54. This may be further exacerbated by gut dysbiosis and an age-associated increase in IDO1⁵⁵, which depletes Trp and promotes Kyn-AhR signaling. Clinically, higher Kyn/Trp ratios in elderly people correlate with frailty and mortality, establishing this metabolic axis as a biomarker of immunosenescence⁵⁶. Additionally, IDO1-driven Trp depletion activates general control nonderepressible 2 kinase (GCN2)^57,58, and suppresses mTOR^59,60 to further limit immune activation, which may promote tumor progression. Kyn may also promote immunosuppression through GPR35-mediated suppression of the ERK1/2, p38 MAPK, and Akt pathways, as well as accumulation of β-catenin⁶¹ (Figure 2). Importantly, gut dysbiosis is associated with altered Trp metabolism. Various studies have demonstrated that Trp metabolites, for instance, Kyn, tryptamine, IPA, and ILA, produced by bacteria like Clostridium sporogenes, Lactobacilli, and Bifidobacterium, etc., regulate the balance and function of Th17/Treg cells, modulate glucagon-like peptide (GLP-1) secretion from intestinal endocrine L cells, and direct epithelium-macrophage interactions in inflammatory diseases^62–65. Concurrently, indole derivatives produced by Lactobacillus spp. such as indole-3-carboxaldehyde (ICA), IAld, and IAA, activate AhR in goblet cells in an IL-10-dependent manner in geriatric mice⁶⁶, or in IL-22-producing ILC3s⁶⁷, preserving epithelial integrity and enhancing mucosal protection, processes that become dysregulated during chronic inflammation and aging. Furthermore, 3-hydroxyanthranilic acid (3-HAA) is reported to promote lifespan and resist oxidative stress by activating Nrf2/SKN-1⁶⁸, while picolinic acid suppresses CD4⁺ T cell proliferation and metabolic activity by inhibiting c-Myc activation⁶⁹, though both are compromised with aging. Collectively, Trp metabolism acts as an immune rheostat, with kynurenines suppressing immunity and indoles supporting barrier integrity, together offering therapeutic potential in aging and inflammation.

Figure 1. GMDMs sculpt immune tone at the intestinal interface across aging.

(A) A young, eubiotic microbiota yields SCFAs (via GPR41/43/109A, HDAC inhibition), indole derivatives (via AhR/PXR/GPR35), polyphenol metabolites, secondary bile acids (via FXR/LXR/TGR5), and vitamins, which reinforce mucin production and tight junctions, limit LPS access, and program the immune tone toward IL-10-rich states. These cues expand Tregs and maintain Treg/Th17 balance, bolster ILC3-mediated barrier integrity, increase Bregs, and support antigen-specific CD8⁺ T-cell fitness and macrophage anti-inflammatory functions, collectively reducing pro-inflammatory cytokines and systemic inflammation. (B) Age-associated dysbiosis erodes mucus, increases LPS translocation, and shifts metabolites toward BCAA/AAA derivatives and other pro-inflammatory metabolites (e.g., TMA to TMAO via hepatic FMO3, while PCS/PAGln/HPAA changes are context-dependent), driving ROS accumulation and pro-inflammatory cytokine production. Secondary BAs exhibit sex-dependent changes, while increased exposure to fatty acid derivatives drives the decrease in ILC3 functions and a skew favouring Th1 and M1-like macrophages, and increases the presence of autoimmune responses, contributing to barrier dysfunction and systemic inflammation. Additional immune remodeling includes heightened mTORC1 activity in CD4⁺ T cells, reduced TFH/GC B-cell programs, and broader senescence-skewed immunity, amplifying neuroinflammation and cognitive dysfunction and predisposing to inflammaging. Overall, youth-associated GMDMs maintain barrier function and mucosal immunity, whereas aging-shifted GMDMs promote pathobiont expansion, barrier failure, and systemic inflammatory tone. AAA = aromatic amino acids, AhR = aryl hydrocarbon receptor, BA = bile acids, BCAA = branched-chain amino acids, Bregs = regulatory B cells, DC = dendritic cells, FXR = farnesoid X r-coupled receptor 43, GPR109A = G protein-coupled receptor 109A, GC = germinal center, GLP-1 = glucagon-like peptide-1, HDAC = histone deacetylase, IL = interleukin, IFN-γ = interferon gamma, ILC3 = group 3 innate lymphoid cells, LPS = lipopolysaccharide, LXR = liver X receptor, mTORC1 = mechanistic target of rapamycin complex 1, Muc2 = mucin 2, PXR = pregnane X receptor, ROS = reactive oxygen species, SCFAs = short-chain fatty acids, TGR5 = Takeda G protein-coupled receptor 5, Th1 = T helper 1 cells, TFH = T follicular helper cells, TMA = trimethylamine, TMAO = trimethylamine N-oxide, TNF-α = tumor necrosis factor alpha.

3.3. Bile acids (BAs)

BAs, cholesterol-derived substances, are synthesized in the liver as primary BAs such as cholic acid (CA) and chenodeoxycholic acid (CDCA), and their conjugated forms. A variety of secondary BAs, like deoxycholic acid (DCA), lithocholic acid (LCA), and their derivatives, are produced by gut microbiota, such as Lactobacillus spp, Clostridium clusters, Eggerthella lenta, Ruminococcus gnavus, and others⁷⁰. An aging immune system disrupts gut barrier function and favors dysbiosis, enhancing the growth of pathobionts (e.g., some Firmicutes, Muribaculaceae, etc) with altered bile salt hydrolase activity and production of differential deconjugated and secondary BAs, such as 3-oxoLCA, 3-oxoalloLCA, isoallo-LCA, DCA, and LCA (Table 1)^71,72.

Accordingly, BAs and their derivatives regulate innate and adaptive immunity via nuclear receptor including farnesoid X receptor (FXR), and via G-protein coupled receptor such as Takeda G protein receptor 5 (TGR5/GPBAR1)⁷³ (Figure 2). FXR activation suppresses NF-κB signaling^74,75 by stabilizing nuclear receptor corepressor 1 complexes⁷⁶. Additionally, FXR activation induces small heterodimer partner (SHP), which suppresses the expression of AP-1 inflammatory target genes⁷⁷ and limits RelA/p65-mediated Ccl2 promoter activity⁷⁸. Independent of SHP, FXR activation also directly restrains NLRP3 inflammasome priming by physical interaction with NLRP3 or caspase-1⁷⁹. While these pathways highlight anti-inflammatory BA-FXR signalling, specific BAs (CDCA, TCDCA, DCA) can act as danger-associated molecular patterns that activate NLRP3 via prolonged Ca²⁺ influx and TLR3-related pathways, independent of FXR, underscoring their context-dependent effects^79,80. TGR5 signaling reduces oxidative stress⁸¹ and activates the cAMP-PKA axis^82,83. PKA-dependent phosphorylation suppresses NLRP3⁸² and NF-κB activation^81,83, and promotes IL-10 production via CREB recruitment to the IL-10 promoter⁸⁴. TGR5 activation also constrains chemokine expression by driving Akt-mTORC1-mediated translation of the C/EBPβ isoform liver inhibitory protein⁸⁵. With aging, microbiota-driven changes in BA composition weaken these FXR/TGR5-dependent tolerogenic circuits and favour pro-inflammatory BA species, thereby amplifying immune dysfunction⁸⁶. Several studies demonstrate that secondary BAs like DCA, 3-oxoLCA, isoLCA, and isoallo-LCA have anti-inflammatory properties by regulating monocytes, macrophages, DCs, and T cell functions in various disease models⁸⁷. Notably, aging-related dysbiosis, including elevated primary BAs and reduced DCA and LCA, correlates with impaired barrier function, inflammation, and ARD risks⁸⁸. In murine senescent brains, tauro-β-muricholic acid (TβMCA) accumulation (associated with gut dysbiosis) induces microglial activation, neuroinflammation, and behavioral impairment⁸⁹. In contrast, calorie restriction-mediated upregulation of LCA metabolite confers anti-aging benefits by activating the AMPK pathway in Caenorhabditis elegans and Drosophila melanogaster⁹⁰, whereas taurocholic acid and some conjugated primary BAs have emerged as regulators of neuroinflammation in older adults and aging mice⁹¹. Taken together, these findings suggest that BAs are a class of host-microbial co-metabolites that reciprocally interact with the immune system, rendering BAs an appealing therapeutic target for chronic inflammation and ARDs.

3.4. Fatty acids and organic acids

Aside from SCFAs and BAs, a rich array of host- and microbiota-derived long-chain fatty acids (LCFAs) and organic acids modulate immune responses and ARDs (Figure 1B). One study reported that supplementing saturated LCFAs, such as palmitic acid (PA) and stearic acid (SA), protects against alcohol-induced inflammation by maintaining intestinal eubiosis and reducing oxidative stress⁹². Contrarily, another study demonstrated that a fat-rich diet containing PA diminishes beneficial microbes (specifically Lactobacillus, Monoglobus, and Lachnoclostridium), impairs mucosal immunity by inhibiting ILC3 populations and decreasing IL-22 production, compromises gut barrier integrity, and exacerbates systemic inflammation in a high-fat diet (HFD)-fed murine model⁹³, highlighting the context-dependent effects of PA. Similarly, in elderly individuals, prolonged exposure to PA, which increases with age and obesity, leads to lipid accumulation and metabolic reprogramming in B cells, resulting in enhanced T-bet expression and autoreactive IgG secretion, worsening B cell immunosenescence and autoimmunity⁹⁴.

Additionally, some unsaturated microbe-derived LCFA, such as elaidate and linoleic acid (LA), are shown to have pro-inflammatory function^95,96. In contrast, other unsaturated LCFAs, especially omega-3 polyunsaturated and monounsaturated oleic acid (OA), exert anti-inflammatory effects. An imbalanced omega-6/omega-3 ratio [(derivative of LA and linolenic acid (LLA) from the microbiota⁹⁷ during aging skews Th1 and M1 responses via arachidonic acid-derived eicosanoids, fueling inflammaging, whereas omega-3 fatty acids promote pro-resolving mediators and restore immune homeostasis⁹⁸. Among organic acids, succinate, which is a well-known byproduct of the tricarboxylic acid (TCA) cycle and a metabolite produced by gut microbiota⁹⁹, promotes IL-1β production via AMPK activation in pro-inflammatory macrophages, which in turn drives obesity-related inflammation¹⁰⁰. This may contribute to inflammaging since the phenomenon of persistent, low-grade IL-1β inflammation is associated with aging. Conversely, α-ketoglutarate (KG), an organic acid and TCA cycle intermediary¹⁰¹, appears to mitigate the effects of immunosenescence by elevating IL-10 in T cells and lowering systemic inflammation, thus extending lifespan¹⁰². These studies illustrate the critical role of particular organic and fatty acids in age-associated immune response and demonstrate the possibilities for therapeutics to target immune dysregulation in the elderly.

3.5. Trimethylamine N-oxide (TMAO)

TMAO has emerged as a pivotal mediator of immunometabolic dysregulation in various diseases¹⁰³. Its biosynthesis initiates in the gut, where distinct microbes in the Desulfovibrio, Firmicutes, etc., taxa catabolize dietary choline, carnitine, betaine, and ergothioneine to generate trimethylamine (TMA)¹⁰⁴. Age-related dysbiosis promotes the overgrowth of TMA-producing taxa^43,105, resulting in elevated circulating levels of TMA, which is then rapidly oxidised to TMAO in the liver by the flavin-containing monooxygenase 3 (FMO3) enzyme¹⁰⁴. This results in elevated levels of TMAO, which has pro-inflammatory and gut dysbiotic properties^43,105,106, thus establishing a feedback cycle where dysbiosis fosters elevated TMAO levels consistently observed in ARDs, implicating age-dependent shifts in host-microbiota co-metabolism¹⁰⁷.

Once in circulation, TMAO can dysregulate both innate and adaptive immune responses^108–110. Elevated TMAO levels contribute to various disease progression via diverse mechanisms, including inflammation, oxidative stress, fibrosis and gut microbiome dysbiosis (Figure 1B)¹¹¹. At the molecular level, TMAO modulates immune responses by stimulating ROS production¹¹², activating the NLRP3 inflammasome^106,113,114, and NF-κB signaling¹¹⁵. Specifically, TMAO may drive vascular inflammation through ROS production that activates NF-κB signalling through PKC-mediated phosphorylation in endothelial cells¹¹⁶ and Nox4/PRMT5-mediated epigenetic modification in vascular smooth muscle cells¹¹⁷, which together can increase VCAM-1 expression to promote leukocyte recruitment and endothelial dysfunction. Furthermore, TMAO activates endoplasmic reticulum stress kinase PERK-mediated unfolded protein response^118,119 and increases apoptosis¹¹⁹ (Figure 2), linking TMAO accumulation to tissue injury and age-related immune dysfunction. With aging, increasing levels of TMAO induce cognitive dysfunction by enhancing oxidative stress and inhibiting the mTOR signalling pathway in a murine model¹²⁰. Consistently, age-associated increase in TMAO in both humans and mice correlates with neuroinflammation, astrocyte activation, and cognitive decline, and chronic treatment with TMAO in young mice recapitulates aging-like memory deficits and glial activation¹²¹. A recent study has revealed that HFD feeding to aging rats promotes gut dysbiosis, and an increase in TMAO levels, which aggravates sarcopenic obesity development via a ROS-AKT/mTOR signaling axis¹²². Together, these findings demonstrate that age-associated increase in TMAO biotransformation has the potential to exacerbate immune dysfunction and accelerate the development of ARDs. Thus, TMAO plays a central role as a microbial product and as a crucial product linking gut dysbiosis to the pathophysiology of aging. While a large body of research has focused on the role of TMAO, TMA was recently found to alter circadian rhythms through TAAR5¹²³, and further research into the direct role of TMA during aging is warranted.

3.6. Branched-chain and aromatic amino acid metabolites

While diet is the main source, branched-chain amino acids (BCAAs: mainly leucine, isoleucine, valine), and aromatic amino acids [AAAs: phenylalanine (Phe) and tyrosine (Tyr), aside from well-studied Trp] are both synthesized and catabolized by the gut microbiota (Bacteroidetes, Firmicutes, etc.) via fermentation to generate microbial metabolites^124–126. Importantly, gut dysbiosis, particularly the depletion of Bacteroides spp., is associated with higher concentrations of AAA and BCAA in circulation, contributing to chronic inflammation, a hallmark of aging¹²⁷.

Microbial BCAA and AAA metabolites use metabolic reprogramming and nutrition-sensing pathways to alter immune cell activity in the gut and abroad (Figure 1B). Mechanistically, BCAAs may modulate immune responses through persistent activation of mTORC1¹²⁸, which is further potentiated by IKKβ¹²⁹, linking increased BCAAs to chronic inflammation and insulin resistance^128,130 (Figure 2). Chronic elevation of leucine in aging maintains aberrant mTORC1 activity in CD4⁺ T cells via the SLC7A5 transporter, which hinders the maintenance of T follicular helper (TFH) cells and germinal center (GC) B cells, and reduces vaccine responsiveness¹³¹. Isoleucine plays a complex role in both chronic inflammation and aging. Interestingly, restricting isoleucine intake has been shown to improve metabolic health, reduce frailty, and extend lifespan, suggesting a potential role in healthy aging in mice¹³². Similarly, Parabacteroides merdae generated BCFAs (BCAA metabolites), including isovalerate, 2-methylbutyrate, and isobutyrate, can delay aging in mice by reducing oxidative stress and inflammation, improving muscle capacity, reversing brain acetylcholine levels, and regulating blood glucose levels¹³³.

Similarly, AAA metabolites such as phenylacetylglutamine (PAGln) and p-cresol sulfate (PCS), derived from Phe and Tyr catabolism via specific gut microbiota species (Clostridium sp. AF50–3, Bacteroides sp. CF01–10NS, and Bacteroides vulgatus), have been linked to age-related neurodegenerative disorders in elderly individuals¹³⁴. A recent study has revealed that Tyr-derived catecholamines (dopamine, norepinephrine) are responsible for the regulation of neural-immune circuits essential for cognition and stress resilience. However, their dysregulation with age contributes to neuroinflammation and cognitive decline¹³⁵. Notably, 4-hydroxylphenylacetic acid (4-HPAA), a microbial tyrosine catabolite enriched in lean individuals, modulates intestinal immunity by enhancing colonic B cell responses, suppressing ILC subsets, and attenuating chronic mucosal inflammation independent of T cells and DCs¹³⁶. Though not yet examined in aged cohorts, its impact on immune-metabolic homeostasis positions 4-HPAA as a potential therapeutic candidate in age-associated metabolic inflammation. Collectively, these findings establish BCAA and AAA metabolites as immunomodulating effectors whose dysregulation integrates amino acid metabolism with inflammaging.

3.7. Vitamins

Although vitamins are often considered general nutritional supplements, several vitamins relevant to aging and immune regulation are either directly synthesized by the gut microbiota or critically dependent on microbial metabolic activity for their bioavailability. Importantly, age-associated dysbiosis alters microbial diversity and vitamin biosynthetic capacity, rendering changes in vitamin availability a functional indicator of microbial ecosystem integrity^137–139. In this context, microbiota-dependent vitamins, particularly vitamin B and K groups^137,140, act as immune-metabolic intermediates that can reflect dysbiosis-driven immune dysfunction during aging^141–143. A metagenomic investigation of ~8,000 human gut microbiomes indicate a significant age-related reduction in microbial diversity and biosynthetic pathways for folate (vitamin B9) and cobalamin (vitamin B12)¹⁴⁴. Several other studies have found that dysbiosis, resulting in vitamin deficiencies (such as riboflavin, thiamine, and folate), promotes immunosenescence and increases the risk of chronic diseases^138,145,146. Reciprocally, changes in vitamin levels impact the growth of specific bacteria, altering the composition and function of microbiota-mediated immune responses. For instance, Vitamin A (all-trans-retinol), which regulates microbial composition¹⁴⁷, delays skin aging through TGF-β signaling¹⁴⁸, probably by promoting fibroblast proliferation¹⁴⁹. Notably, feeding mice with a vitamin B1-deficient diet compromises the number of intestinal IL-4⁺ ILC2s¹⁵⁰, indicating a potential mechanism driving barrier dysfunction in aging. Additionally, vitamin D supplementation promotes tolerogenic DC phenotypes and boosts Treg development while simultaneously attenuating Th1/Th17 effector responses, changes that may reduce inflammaging^151,152. Collectively, these findings support the view that microbiota-dependent vitamins should be considered functional GMDMs whose dysregulation integrates microbial dysbiosis, immune aging, and barrier dysfunction, rather than merely exogenous nutritional supplements.

3.8. Polyphenol-derived metabolites

Polyphenol-derived metabolites, generated by microbial catabolism of dietary polyphenols, are major immunomodulators in aging and chronic diseases, and have significantly expanded their functional roles beyond traditional antioxidant activity (Figure 1A). The gut microorganisms (e.g., Eubacterium ramulus, Flavonifractor plautii, Eubacterium limosum, etc.) metabolize dietary polyphenols through a series of enzymes, such as flavone reductase, phloretin hydrolase, enoate reductase, and chalcone isomerase, to break down the dietary polyphenols into smaller phenolic and aromatic acids readily absorbed by the host and have immunoregulatory activity^153–155. A recent study has shown that resveratrol can also be metabolized by gut microbiota¹⁵⁶, potentially contributing to its ability to rescue immunosenescence by limiting pro-inflammatory cytokines and inhibiting NF-κB activation in aged mice, restoring both innate and adaptive immunity¹⁵⁷. Accordingly, supplementation of urolithin A (UA), an ellagic acid metabolite produced by the gut microbiota, in sleep-deprived mice significantly reduces gut dysbiosis and inflammation and improves intestinal permeability and motor function¹⁵⁸. In humans, UA offers protection against age-related inflammation and mitochondrial dysfunction by reducing the production of pro-inflammatory cytokines, improving mitochondrial gene expression, autophagy, and fatty acid oxidation pathways. These processes facilitate immune cell oxidative metabolism linked to reduced inflammation; therefore, UA emerges as an efficient immunometabolic modulator to promote healthy aging¹⁵⁹. Similarly, quercetin, a senolytic flavonol metabolized by the gut-associated microbe Eubacterium ramulus¹⁶⁰, is reported to remove senescent immune cells, inhibiting senescence-associated secretory phenotype (SASP)-induced inflammation and restoring immune homeostasis in aged murine models¹⁶¹. Most recently, microbial metabolite 5-(3′,4′-dihydroxyphenyl)-γ-valerolactone, a GMDM derived from proanthocyanidins, can inhibit neuroinflammation and oxidative stress in cortical microglia by downregulating NLRP3 expression, and modulating the NOX2/Nrf2 signaling pathway¹⁶², processes engaged in inflammaging. Collectively, these metabolites represent a mechanistically diverse group against chronic disease and immunosenescence.

Table 1 summarizes the multifaceted roles of GMDMs, their major microbial sources, age-associated alterations, immunomodulatory mechanisms, and common molecular pathways implicated in aging.

4. GMDMs and multi-organ immune axes in aging

An altered GMDM profile and systemic spreading of pro-inflammatory microbial products can impact various metabolite-immune cell axes in distant organs, contributing to systemic, chronic low-grade inflammation, cellular senescence, and the pathogenesis of a spectrum of ARDs¹⁶³. In return, the inflammatory changes in distant organs can also influence the gut microbiota and GMDM composition¹⁶⁴, implicating a pathogenic feedback loop (Figure 3). This section will explore these bidirectional interactions across several key organs, elucidating how age-altered GMDM profiles contribute to immune dysregulation and accelerate adverse aging outcomes.

Figure 3. Gut-multi-organ GMDM axes in aging and age-related inflammatory diseases.

Core GMDM classes, SCFAs, tryptophan-indole derivatives, bile acids, pro-inflammatory uremic/pro-atherogenic metabolites (TMA/TMAO, IS, PCS), and additional moieties (BCFAs, organic acids, vitamins, BCAA/AAA metabolites) reprogram immunometabolic circuits across liver, adipose, muscle, brain, lung, heart, pancreas, kidney, retina, and joints, thereby shaping organ-specific ARDs. Each organ shows cell-type-specific effects, yet dysbiotic GMDM profiles drive convergent pathology (barrier loss, chronic inflammation, fibrosis) and cross-organ propagation (e.g., renal-muscular and renal–cardiovascular axes). Collectively, this metabolite network synchronizes systemic aging trajectories and the progression of diverse ARDs. AAA = aromatic amino acids; AD = Alzheimer’s disease; AMD = age-related macular degeneration; ARDs = age-related diseases; BCAA = branched-chain amino acids; BCFAs = branched-chain fatty acids; CKD = chronic kidney disease; COPD = chronic obstructive pulmonary disease; CVD = cardiovascular disease; GMDMs = gut microbiota-derived metabolites; IR = insulin resistance; IS = indoxyl sulfate; MASLD = metabolic dysfunction-associated steatotic liver disease; PCS = p-cresyl sulfate; PD = Parkinson’s disease; PDAC = pancreatic ductal adenocarcinoma; SCFAs = short-chain fatty acids; T2D = type 2 diabetes; TMA = trimethylamine; TMAO = trimethylamine N-oxide.

4.1. GMDMs and gut-liver axis

The liver harbours a diverse network of immune cells, including Kupffer cells, DCs, ILCs, B cells, NKT and T cells, which continuously monitor gut-derived pathogens and antigens¹⁶⁵. Among the altered GMDMs, BAs, SCFAs, BCAAs, and tryptophan derivatives play important roles in shaping the liver immune landscape during aging⁸⁶, compounding with other age-related factors, such as hepatocyte senescence and metabolic stress¹⁶⁶. While the total BA level is generally reduced during aging due to hepatocyte senescence and reduced CYP7A1 expression¹⁶⁷, age-associated dysbiosis further disturbs BA balances (especially secondary BA). For instance, aging can increase liver DCA abundance in male mice, as well as decrease liver and serum LCA in female mice⁸⁸. DCA and LCA are potent activators of the NLRP3 inflammasome in Kupffer cells, contributing to liver fibrosis in mice¹⁶⁸, potentially rendering male mice more susceptible to age-related liver diseases and metabolic dysfunction (e.g., insulin resistance). However, DCA and LCA can also suppress splenic and liver CD8⁺ T cell effector functions in a TGR5- and FXR-independent manner^169,170, indicating the complex immunomodulatory features of BAs in the liver immune niche.

The age-associated decrease in SCFAs may also impact the liver by promoting macrophage activation and reducing Treg differentiation, critical for controlling age-related metabolic liver diseases (see section 5.1). Furthermore, age-associated changes in tryptophan metabolism lead to reduced production of indole and indole derivatives¹⁷¹, which can alleviate diet-induced liver inflammation by suppressing pro-inflammatory macrophage activation in mice¹⁷². Notably, not all changes in microbiota and GMDMs during aging are associated with increased inflammaging. P-cresol and its derivative PCS, which are linked to oxidative stress and inflammation during aging, are shown to protect bile duct epithelium damaged by LPS and reduce bile duct inflammation by driving M2 polarization of Kupffer cells¹⁷³. In addition, older individuals were found to host an increased population of P. merdae, which is capable of catabolizing BCAAs, activators of mTOR that contribute to immune aging and chronic inflammation¹⁷⁴. Indeed, supplementing P. merdae in aged mice significantly lowered BCAAs and reversed hepatocyte focal necrosis caused by D-galactose-induced senescence¹³³, implicating the multifaceted role of GMDMs in regulating liver aging.

4.2. GMDMs and gut-adipose axis

Adipose tissue (AT), especially visceral adipose tissue (VAT), serves as both an endocrine organ and a metabolically active immune organ populated by T cells, B cells, adipose tissue macrophages (ATMs), DCs, and ILCs. In the aging gut-adipose axis, dysbiosis facilitates endotoxin translocation and GMDM imbalance, representing two interwoven inflammatory pathways. Together, they predispose a shift in the local adipose tissue immune landscape, decreased thermogenic potential, and an accumulation of senescent immune cells that drive chronic low-grade inflammation, insulin resistance, and other age-related metabolic syndromes¹⁷⁵.

SCFAs and tryptophan metabolites are particularly critical regulators of the gut-adipose axis^176,177. Propionate can dampen LPS-stimulated human ATM activation in vitro¹⁷⁸, and butyrate modulates ATM polarization to M2-like populations in obese mice¹⁷⁹. As a promoter of Treg differentiation, the reduction of these GMDMs during age-related dysbiosis may underlie the reduced functionality of naive Tregs during aging¹⁸⁰. Butyrate supplementation restores VAT ST2⁺ Tregs in DIO mice via PPARγ signaling¹⁸¹, suggesting relevance in aging, where dysfunctional Tregs accumulate in VAT, driving insulin resistance, fibrosis, and worsened aging pathology¹⁸⁰. Given that the accumulation of dysfunctional immune cells, especially ILC2 and Tregs, in aging VAT also incurs decreased thermogenesis and cold tolerance¹⁸², it is tempting to investigate if GMDMs can potentially restore adipose thermogenesis by reviving local immune cell functions.

The dysregulation of Trp metabolites also contributes to metabolic disturbances during aging⁵⁵. Age-related dysbiosis lowers indole availability, diverting Trp metabolism toward the Kyn pathway via age-associated increase in IDO1 activity⁵⁵. The resulting Trp starvation can activate GCN2 to promote FoxP3⁺ Tregs¹⁸³, while Kyn can further promote Tregs differentiation through AhR and PD-1 upregulation¹⁸⁴, potentially shaping the altered Treg dynamics in aging adipose tissue, though their integrated impact remains to be directly demonstrated.

Notably, multiple GMDMs, including SCFAs, Trp metabolites, and secondary BAs, can promote B-cell antibody class switching^185,186, Breg differentiation^40,187, and memory responses¹⁸⁸. The increasing inflammatory tone in aging AT is associated with increased presence of self-antigens and accumulation of autoreactive B cells^189,190, which exacerbate metabolic aging via pro-inflammatory responses and autoantibody (AutoAb) secretion²². In addition, B cells can also promote T cell immunosenescence, linking their expansion in VAT to systemic inflammaging¹⁹¹. How age-associated GMDM shifts impact B cells within AT and contribute to age-related immune dysfunction represents an intriguing frontier to explore.

4.3. GMDMs and gut-muscle axis

The gut-muscle axis is increasingly recognized as a regulator of muscle and metabolic health¹⁹². A young GMDM profile coordinates the immune responses required for muscle regeneration¹⁹³, whereas an imbalanced GMDM profile in aging can impair the necessary early pro-inflammatory phase required for debris clearance and the subsequent anti-inflammatory responses that enable satellite cell activation. In turn, defective muscle regeneration can exacerbate dysbiosis and systemic immune dysfunction (Figure 3).

SCFAs can regulate muscle regeneration through direct and indirect mechanisms. By preserving gut barrier integrity, they help create an anti-inflammatory, pro-resolving niche in the muscle that supports muscle stem cell function^194,195. With aging, mitochondrial dysfunction impacts both muscle cells and immune cells, reducing muscle regeneration and contributing to immunosenescence^196,197. SCFAs counter this by enhancing mitochondrial biogenesis in both immune and muscle cells and improving immune cell oxidative phosphorylation under inflammatory stress^198–200. Consistently, butyrate treatment improves muscle mass and reduces oxidative stress in aged mice²⁰¹.

Other GMDMs also preserve muscle integrity. Trp metabolites (e.g. indole-3-carbinol (I3C), IAA and IPA) reduce muscle inflammation and improve myogenesis²⁰², while promoting ILC3 and Tregs differentiation via AhR or PXR^48,203, and reinforcing gut integrity, thereby limiting systemic LPS exposure. Similarly, secondary BAs, such as isoalloLCA and 3-oxoLCA, can promote Tregs and macrophage polarization²⁰⁴, favoring a pro-resolving milieu conducive to satellite cell activation and muscle regeneration. In addition, pipecolic acid, a lysine-derived GMDM, and succinate, a Kreb’s cycle intermediate, preserve muscle mass and strength in murine disuse atrophy²⁰⁵, potentially via enhanced ATP output and mitochondrial complex activity²⁰⁶, implicating these GMDMs as candidates to mitigate sarcopenic decline in aging.

GMDMs may restore not only the abundance but also the functional quality of aged muscle immune cells. A hallmark of aging is declining AMPK activity across tissues with impaired mitochondrial function and persistent inflammation⁴. Mice lacking AMPKα1 in myeloid cells show defective phagocytosis of necrotic muscle fibers, persistent inflammation, and reduced myofiber sizes post-injury²⁰⁷. Butyrate activates macrophage AMPKα1, enhancing oxidative metabolism, efferocytosis^208,209, and the transition to restorative phenotypes, potentially via the CaMKKβ-AMPK axis^210,211. Similarly, SCFAs can activate PPARγ in skeletal muscle macrophages, enhancing macrophage oxidative metabolism and the expression of growth differentiation factor 3, potentially promoting myogenic cell fusion and tissue repair during late-stage regeneration^210,212. Thus, GMDMs may improve muscle repair in aging by metabolically reprogramming muscle-resident macrophages toward a pro-resolving, regenerative state.

4.4. GMDMs and gut-brain axis

From the enteric nervous system (ENS) to the central nervous system (CNS), GMDMs shape brain health and sustain cognitive functions through a complex immune and neuroendocrine network. Aging destabilizes this network and drives neuroinflammation, while age-related psychosocial stress further impairs gut microbiota and barrier integrity, accelerating cognitive decline¹⁸.

GMDMs can cross the blood-brain and cerebrospinal fluid (CSF) barriers to act directly on the CNS. SCFAs, particularly butyrate, traverse the blood-brain barrier (BBB) via endothelial monocarboxylate transporters, reaching concentrations an order of magnitude higher than in blood²¹³, where they modulate microglial maturation, neurogenesis, and NF-κB-driven inflammation, while strengthening BBB integrity²¹⁴. Other GMDMs affect CNS via distinct axes. BAs (UDCA and TUDCA) suppress neuroinflammation directly through microglial and neuronal FXR/TGR5 activation, and indirectly via gut-derived GLP-1 and FGF15/19-β-Klotho signaling in the hypothalamus²¹⁵. Indole derivatives (IAA and IPA) reduce Aβ deposition, diminish tau hyperphosphorylation, and improve cognitive function, potentially by inhibiting microglial activation via AhR and suppressing NLRP3-driven inflammation^216,217. In contrast, TMAO drives microglial activation and amplifies neuroinflammation, accelerating Aβ and tau pathology²¹⁸. Thus, age-related GMDM alterations can be direct neurochemical drivers of neuronal dysfunction and barrier breakdown, linking metabolic aging in the gut to neuroinflammatory decline in the brain.

Beyond sculpting neuronal and glial activities, GMDMs engage neuroendocrine pathways, such as the vagus nerve, a bidirectional communication “highway” between gut and brain²¹⁹. SCFAs can activate FFAR3 on vagal afferents, while BAs and indoles stimulate enteroendocrine cell signaling to excite vagal neurons, regulating feeding and emotional behaviors^220,221. Notably, SCFAs also trigger serotonin release from enterochromaffin cells that synapse with vagal terminals, activating 5-HT receptors and glutamatergic pathways controlling appetite, mood, and stress responses that are often dysregulated in aging²²². SCFA-induced GLP-1 and peptide YY further regulate satiety via the vagal-brainstem circuit²²³, while attenuating microglial activation, preserving dopaminergic neurons, restoring neuronal insulin signaling, and reducing α-synuclein burden^224–226. Through this integrated neuroimmune signaling, the microbiota maintains the gut-brain homeostasis, whose erosion with aging amplifies neuroinflammation. How GMDM alterations contribute to specific neurodegenerative diseases during inflammaging will be detailed in section 5.3.

4.5. GMDMs and gut-lung axis

While lungs harbor their own microbiota, dysbiotic GMDM profiles can also drive a pulmonary inflammaging niche, predisposing to prolonged lung infections, and age-related pulmonary diseases (see section 5.5)²²⁷. A key study showed that SCFAs markedly reduced pulmonary inflammaging, oxidative stress, and metabolic alteration, and ameliorated acute lung injury in old mice⁴⁴. In addition, SCFAs promote Ly6c⁻ patrolling monocytes, enhance CD8⁺ T cell effector function, and restrain excessive neutrophils during influenza^228,229, implicating their potential in protecting aged individuals from unresolved lung inflammation after viral infections, driven by excessive neutrophil recruitment and neutrophil extracellular traps^230–232. SCFAs also inhibit dysfunctional ILC2s that accumulate in aging lungs^233,234, which promote extracellular matrix (ECM) remodeling that exacerbates age-related fibrosis and the potentially associated autoimmune responses^235,236, the IgE-driven allergic inflammation²³⁷, and inflammatory responses that fuel age-related pulmonary diseases.

Beyond SCFAs, other GMDMs also shape the gut-lung axis with relevance to aged immunity, though direct studies in aging are limited. BCFAs, notably isovaleric acid, enhance upstream mucosal containment by increasing gut IgA²³⁸ and fortify the airway barrier by upregulating ZO-1 and occludin, thereby mitigating influenza injury²³⁹, a critical vulnerability in older adults. Indole derivatives similarly temper lung inflammation: I3C reduces monocyte/neutrophil accumulation after LPS exposure, and IAA pretreatment lowers neutrophil influx, edema, and tissue damage in intratracheal LPS challenge^240,241, processes that are typically exaggerated with age. In contrast, TMAO promotes IL-6 production by bone marrow-derived macrophages, a cue that can drive neutrophil recruitment and foster pulmonary inflammaging²⁴². Together, these metabolites map to barrier integrity, myeloid restraint, and mucosal immunoglobulin support that can be leveraged to rebalance inflammation in the aging lung.

4.6. GMDMs and gut-heart axis

Shaped by microbial metabolites and immuno/neurohormonal signalling, the gut-heart axis regulates cardiovascular health, which deteriorates with age²⁴³. GMDMs, especially SCFAs and TMAO, are central modulators of cardiac immune cells, with macrophages as key effectors driving chronic inflammation and cardiac dysfunction during aging^244,245.

SCFAs, especially butyrate, ameliorate cardiac fibrosis, a cardiac ARD, by promoting M2-like macrophage polarization and restoring mitochondrial function in rats²⁴⁶. Butyrate also reduces atherosclerotic lesions and systemic inflammation in high-fat-fed ApoE⁻/⁻ mice by modulating macrophage polarization via GPR signaling, HDAC3 inhibition, and miRNA regulation²⁴⁷. Notably, acetate, butyrate, and propionate can all inhibit inducible nitric oxide synthase in LPS-treated macrophages, suppressing the level of nitric oxide, which is mainly sourced from M1-like macrophages accumulating in aging hearts and known to promote heart failure and myocardial infarction when excessive^176,248,249. Though direct evidence for the effect of SCFAs on cardiac macrophages in aging has yet to be demonstrated, these findings suggest that aging-related decline in SCFA levels may impair macrophage function and mitochondrial health, exacerbating chronic cardiovascular inflammation and tissue remodeling.

In contrast, TMAO, elevated nearly twofold in aged individuals¹⁰⁷, amplifies vascular inflammation by activating NF-κB and VCAM-1 in human endothelial cells and accelerates atherosclerosis¹¹⁶, while driving M1-like macrophage polarization²⁵⁰, foam cell formation²⁵¹, and macrophage mitochondrial dysfunction¹⁰⁶. Interestingly, hydrogen sulfide (H₂S), an endogenous gasotransmitter, significantly inhibits the activation of NF-κB signaling and expression of pro-inflammatory cytokines in M1-like macrophages by upregulating Sirtuin-1, a NAD⁺-dependent deacetylase²⁵². With age-related declines in Sirt1 and H₂S activity^253,254, macrophages may become more sensitive to TMAO, exacerbating immune aging and contributing to chronic vascular inflammation and cardiovascular dysfunction. Further investigation of GMDMs regulating Sirt1 and H₂S levels may help advance therapies targeting cardiac aging.

4.7. GMDMs and gut-pancreas axis

The gut-pancreas axis regulates pancreatic health in aging by modulating its immune environment. Mice colonized with aged human gut microbiota exhibited exacerbated acute pancreatitis, likely due to reduced pancreatic and ileal antimicrobial peptides, and decreased expression of intestinal tight junction protein²⁵⁵. This environment allows microbial endotoxin to polarize macrophages toward the M1-like phenotype in mice with acute pancreatitis²⁵⁶. Following injury, in young pancreas, M2-like macrophages are usually transiently upregulated by Th2/ILC2 cell-derived IL-4 and IL-13, to effectively resolve inflammation and promote tissue regeneration²⁵⁷. However, during aging, a reduction in SCFAs, such as butyrate, may dampen suppression on Th2/ILC2 cells, and M2-like macrophages, especially the CD206⁺ M2a subtype, which fuels pancreatic stellate cell (PSC) activation and ECM deposition to cause fibrosis²⁵⁷. Compounding the reduction in SCFAs, decreased levels of isoalloLCA (Treg inducer), as well as 3-oxoLCA and isoLCA (Th17/Th22 inhibitors), dysregulate the Treg/Th17 balance²⁵⁸. IL-17A and IL-22, despite maintaining intestinal barrier integrity, can stimulate PSCs to induce collagen deposition and pancreatic fibrosis^259,260. Notably, an aged stromal environment is vastly immunosuppressive and can accelerate the progression of pancreatic ductal adenocarcinoma (PDAC)²⁶¹, while microbial indoles can activate tumor-associated macrophages via AhR, worsening PDAC progression in mice²⁶². Thus, the therapeutic potential of GMDMs in pancreatic ARDs may depend on T cell-macrophage-stromal interactions, particularly the stage-specific and paradoxical roles of Tregs and M2-like macrophages.

4.8. GMDMs and gut-kidney axis

Aging can accelerate various chronic kidney diseases (CKDs), such as nephrosclerosis, ischemic nephropathy, tubulointerstitial fibrosis, and diabetic nephropathy²⁶³. Dysbiotic microbiota and altered GMDM profiles drive these processes while being reciprocally shaped by CKDs.

Aging enriches proteolytic bacteria, increasing uremic toxins, including TMAO, indoxyl sulfate (IS), and PCS^264,265, which not only disrupts colonic tight junctions²⁶⁶ but also damage podocytes, leading to increased glomerular permeability and proteinuria, markers of CKD progression²⁶⁷. At pathogenic levels, IS and PCS activate kidney resident macrophages via AhR, enhance ROS production and oxidative stress, and boost M1-like macrophages producing MCP-1, IL-6, and TNF-α²⁶⁸. Notably, Klotho, a regulator of oxidative stress and senescence, is found to alleviate IS-induced renal inflammation by promoting M2-like macrophage differentiation²⁶⁹, and its reduction in aging is associated with an increased incidence of kidney failure^270,271. However, similar to the aging pancreas, in chronic kidney injury, sustained M2-like macrophage infiltration may drive fibrosis via IL-10, TGF-β, and myofibroblast activation. TMAO, IS, and PCS are also known to directly induce renal fibrosis by stimulating epithelial-to-mesenchymal transition, increasing renal cell carcinoma risk²⁷².

Uremic toxins also foster multi-organ immune dysfunction. For instance, beyond amplifying oxidative stress in type 2 diabetes (T2D) patients with diabetic nephropathy^273,274, IS and PCS promote mitochondrial damage, oxidative stress, and expression of atrophy-related genes (e.g., myostatin, Atrogin-1) in myoblasts²⁷⁵, accelerating sarcopenia. Accumulated uremic toxins also induce widespread endothelial dysfunction and heighten oxidative stress within the vascular wall, promoting retention of low-density lipoproteins and driving atherosclerotic plaque formation²⁷⁶.

Overall, GMDMs orchestrate complex, bidirectional immunometabolic interactions across organs, with context-specific and concentration-dependent effects. Disrupted GMDM networks can transform protective immune cells into drivers of fibrosis and inflammation. Decoding these nuanced interactions offers translational potential to restore immune homeostasis and mitigate ARDs through precision microbiome-based interventions.

5. GMDMs and age-associated chronic inflammatory diseases

Considering the central role of GMDMs in immunomodulation, age-related gut dysbiosis and the associated shift in GMDMs may create a permissive environment for the development of multiple ARDs (Figure 3).

5.1. Metabolic syndrome

Metabolic syndrome (MetS), including obesity, dyslipidemia, hypertension, and dysglycaemia, is highly prevalent in older individuals and contributes to late-life multimorbidity, such as T2D, cardiovascular disease (CVD), and metabolic dysfunction-associated steatotic liver disease (MASLD)^277,278. The chronically altered GMDMs pool in aging fosters visceral adiposity, insulin resistance, and steatosis, elevating risks for MetS. In young mice, SCFAs protect against diet-induced metabolic dysfunction and systemic inflammation by strengthening the gut barrier, modulating immune cells like ILCs and Tregs, and suppressing TNFα levels across adipose and hepatic tissues^279–281. However, age-associated loss of SCFA-producing taxa and reduced SCFA availability may contribute to increased gut permeability, metabolic endotoxemia, and the pro-inflammatory remodelling of adipose and hepatic immune cells, changes that promote insulin resistance and MASLD²⁸². Age-associated dysbiosis also reshapes the BA pool toward more hydrophobic, pro-inflammatory species. Secondary BAs such as DCA drive M1-like macrophage polarization and IL-6/TNFα production via TLR-NF-κB signalling in the colon, while reduced Bacteroides uniformis in aging lowers 3-succinylated cholic acid (3-sucCA), a BA that alleviates hepatic steatosis in mice by expanding A. muciniphila^283–285. In addition, declines in microbiota-derived indoles and other AhR agonists also predispose older hosts to hepatic and adipose inflammation-driven MetS. Loss of indole derivatives, such as IAA, lifts the restraint on macrophage activation and cytokine-mediated lipogenesis in hepatocytes²⁸⁶, while weakened AhR-dependent IL-22 production compromises gut barrier repair, facilitating metainflammation²⁸⁷. Notably, indigo, an AhR agonist with structural similarity to indoles, protects against insulin resistance and hepatic steatosis in obese mice by promoting IL-22, suggesting therapeutic potential in age-related MetS²⁸⁸. In contrast, elevated circulating TMAO and related metabolites, such as uremic toxins, lead to endothelial dysfunction, NLRP3 inflammasome activation, and pro-inflammatory monocyte responses that contribute to cardiovascular dysfunction, insulin resistance, and other MetS risks in aged populations^289–292. Together, these age-linked GMDM shifts create a permissive, barrier-compromised milieu that lowers the threshold for MetS in late life.

5.2. Cardiovascular diseases

Aging amplifies the cardiovascular impact of dysbiosis by layering metabolite-driven inflammation onto an aging vasculature undergoing endothelial senescence, arterial stiffening, and impaired repair. In general, older adults with higher plasma TMAO and imidazole propionate, which increase with age, show increased risks of developing atherosclerosis and related mortality^293,294, while SCFA and various indoles that are lost with age are often protective^295,296. TMAO promotes vascular inflammaging by activating NF-κB-dependent endothelial leukocyte recruitment, upregulating macrophage CD36 and SR-A1 to drive foam cell formation, and inducing endothelial cell senescence, oxidative stress and atherosclerotic plaque development^115,297. Meanwhile, elevated imidazole propionate drives atherosclerosis via imidazoline-1 receptor-induced mTOR signaling in myeloid cells²⁹⁸. The age-related reduction in microbial indole derivatives, such as IPA and IAA, further expedites vascular aging. IPA is among the most downregulated GMDMs in patients with coronary artery disease, and lower IPA associates with greater plaque burden²⁹⁶. In ApoE^−/− mice, IPA attenuates vascular inflammation and atherosclerosis, in part by enhancing macrophage cholesterol efflux, whereas TMAO impairs it^296,299. Atherosclerotic progression in aging coronary arteries can culminate in myocardial infarction (MI), where altered GMDM profiles markedly dictate the disease outcomes. Age-elevated TMAO and phenylacetylglutamine increase acute thrombotic risk by heightening platelet activation and thrombus formation^300,301. TMAO further destabilizes aged plaques by impairing M2-like macrophage polarization and efferocytosis, preventing the resolution of the inflammatory responses³⁰². In parallel, the age-related decline in SCFAs hinders post-MI cardiac repair by compromising the monocyte-induced healing processes necessary for survival³⁰³, further lowering the threshold for both atherothrombosis and myocardial repair failure in late life.

5.3. Neurodegenerative diseases

The age-related blood-brain and gut barrier dysfunction create a permissive milieu where dysbiotic shifts in GMDMs can amplify neuroinflammation and fuel the pathogenesis of several age-related neurodegenerative diseases, including Alzheimer’s disease (AD) and Parkinson’s disease (PD)³⁰⁴.

5.3.1. Alzheimer’s disease

In AD, elevated TMAO and secondary BAs, together with age-related loss of microbiota-derived indoles and SCFAs, converge on astrocyte and microglial activation and accelerate neurodegeneration. TMAO can cross the blood-brain barrier to drive hippocampal neuronal senescence and synaptic dysfunction^120,121, whereas pharmacologic TMAO lowering ameliorates cognitive impairment and neuroinflammation in AD models³⁰⁵. Meanwhile a concurrent shift from primary to secondary BAs removes neuroprotective factors, such as ursodeoxycholic acid (UDCA) and TUDCA, which normally suppress microglial NLRP3/IL-1β signalling, glial activation, amyloid pathology and associated autoantibody responses^22,306,307. This BA dysregulation is particularly pathogenic in aging, where impaired glymphatic and meningeal clearance promotes maladaptive antigen presentation and brain-reactive T-cell activation³⁰⁸. Simultaneously, age-related dysbiosis redirects Trp metabolism away from protective indoles, which typically enhance Aβ clearance via AhR signaling, towards a potentially neurotoxic kyn pathway, generating metabolites that trigger microglial-driven oxidative stress and motor decline³⁰⁹. Notably, prebiotic inulin boosts SCFA and Trp metabolites and attenuates AD markers and frailty in mice³¹⁰, while peripheral inhibition of KMO, IDO or TDO in AD fly models, together with supplementation of indoles (IPA, IAA), ameliorates neurodegeneration and cognitive impairment^{217,311–313}. This evidence suggests that restoring age-depleted GMDMs may raise the threshold for AD pathology.

5.3.2. Parkinson’s disease

PD arises predominantly in later life, during which gut dysbiosis can facilitate α-synuclein-driven neurodegeneration. Older individuals with PD show a consistent reduction in butyrate-producing bacteria, indicating a shift toward a pro-inflammatory gut niche³¹⁴. By improving gut barrier integrity in PD mice, butyrate supplementation or prebiotic high-fiber diets ameliorate dopaminergic neuron loss and motor impairment, thus dampening striatal microglial activation and favoring neuroprotective microglial phenotypes^315,316. However, SCFAs’ effects on microglia are context dependent. While acetate supports microglial maturation and metabolic fitness under homeostatic or neurodegenerative conditions, mixed SCFA supplementation in germ-free α-synuclein-overexpressing mice is sufficient to activate microglia and worsen motor dysfunction^317,318. These findings suggest that in aging PD, selective depletion of SCFA producers and altered SCFA balance may tip the gut-brain axis from immune homeostasis toward inflammaging. In parallel, elevated TMAO biosynthesis in PD and its capacity to activate glia and induce pro-inflammatory cytokines point to a metabolite amplifier of age-primed neuroinflammation^319–321. Finally, altered BA and Trp metabolism in PD, including reduced Trp flux into AhR ligands, may further weaken neuroprotective signaling pathways in the aged brain^322–324.

5.3.3. Psychosocial isolation as a risk factor

Psychosocial isolation, another hallmark of aging⁸, further reshapes the gut-brain axis along with GMDMs. Loneliness in healthy old adults correlates with hypothalamus-pituitary-adrenal (HPA)-axis dysregulation³²⁵, blunted cortisol rhythms, and heightened NF-κB activity³²⁶. This stress-induced HPA dysregulation can alter neuroimmune responses in the CNS and ENS, predisposing to dysbiosis^327,328 and neurodegenerative diseases^329,330. Chronic isolation in mice induces dysbiosis, depletes SCFA and Trp metabolites, and promotes neuroinflammation³³¹, whereas repletion of these metabolites mitigates neuronal declines by restoring HPA axis homeostasis^332,333. Together, these findings highlight how GMDMs intricately link psychosocial isolation as a risk factor for neurodegenerative disease.

5.4. Colorectal cancer

Colorectal cancer (CRC) incidence rises with age, and age-related shifts in GMDMs help create a colonic niche that is more permissive to tumor initiation and progression. In older adults, depletion of butyrate-producing taxa such as Holdemanella biformis, Roseburia intestinalis, and Faecalibaculum rodentium^334–337, reduces luminal SCFAs, weakening HDAC-dependent restraints on colonocyte proliferation, c-Myc expression and apoptosis, and blunting SCFA-driven anti-tumor CD8⁺ T cell and NKG2D responses^{336,338–341}. Although strain-specific SCFA production by Fusobacterium nucleatum can instead fuel Th17- and IL-23-mediated tumour-promoting inflammation via FFAR2/GPR43³⁴², the net effect of age-associated SCFA loss still lowers the threshold for malignant transformation. While gut Trp metabolites and indole-AhR ligands are inversely associated with CRC risk, the role of Trp metabolites in age-related CRC can be dual immunomodulatory³³⁵. I3C and Lactobacillus reuteri derived ILA can suppress Th17 responses, support barrier function and limit tumorigenesis via AhR^343,344, whereas trans-2-indoleacrylic acid, enriched in CRC, promotes tumor growth by inhibiting ferroptosis through the AhR-ALDH1A3FSP1 axis³⁴⁵, underscoring context-dependent Trp effects in aged colons. Finally, aging-associated accumulation of secondary BAs such as DCA, together with diminished FXR signalling, skews gut macrophages toward M1-like states, enhances Il17a/Il23 expression, and impairs CD8⁺ T cell-mediated immune surveillance, thereby fostering an inflammatory, pro-tumorigenic microenvironment in aged colons^169,346.

5.5. Chronic obstructive pulmonary disease

Aging heightens susceptibility to chronic obstructive pulmonary disease (COPD) by predisposing the lung and immune system to dysbiosis-driven injury. Older individuals with COPD exhibit distinct fecal microbiome and metabolome profiles, including reduced amino acid biosynthesis and lower SCFA levels, partly linked to decreased fibre intake^347,348. Loss of butyrate in this context removes an important brake on age-primed airway inflammation, where butyrate normally alleviates COPD-like pathology by promoting anti-inflammatory immune regulation, limiting ILC2 proliferation via HDAC inhibition and dampening type 2 and Th9 responses^{234,349–351}. In addition to GMDM origin, Trp metabolites, such as IAA, can also be produced by the airway microbiome. Here, it can protect against COPD through AhR-dependent IL-22 crosstalk with lung macrophages, which suppresses epithelial apoptosis³⁵². As age-linked erosion of SCFA and Trp metabolite pools likely lowers the threshold for chronic airway obstruction synchronously with local microbiota in late life, it will be important for future studies to dissect how gut- versus lung-derived Trp metabolites differentially shape pulmonary disease in aging.

6. Modulating GMDMs as a therapeutic strategy

With the potential to regulate inflammation, metabolic balance, and healthspan, GMDMs are promising immunomodulators for ARDs. Treatments with dietary, microbial, behavioral, and biotechnological interventions are rapidly moving from theory to trials. Below, we outline key approaches, clinical progress, and persisting challenges in the translation of GMDMs (Figure 4).

Figure 4. Therapeutic strategies to harness GMDMs for healthy aging.

Individual phenotypes and multi-omics profiles, integrated with behavioral records and health history, are analyzed by AI to stratify patients and guide personalized interventions. Tailored biotics (pre/pro/postbiotics), bacterial therapies (FMT, GEMs), exercise, and circadian modulation converge on the microbiome to tune key GMDMs (SCFAs, BAs, Trp-indoles), strengthening barrier function, enhancing metabolic resilience, cognition, and muscle fitness, and reducing systemic inflammation. These improved outcomes feedback as new phenotypes for continuous optimization, establishing a technology-enabled, closed-loop “ecosystem” for healthy aging.

6.1. Prebiotics, probiotics, and postbiotics

Collectively termed “biotics,” these interventions enhance levels of bioactive metabolites with immunoregulatory and protective functions, thus restoring microbial balance, promoting gut health, and contributing to healthspan (Figure 4)³⁵³. Prebiotics, such as non-digestible components including inulin, FOS, GOS, fructans, and lactulose, selectively nourish beneficial taxa, particularly SCFA producers like Faecalibacterium prausnitzii and Bifidobacterium spp., while lowering pathobionts³⁵⁴. By selectively promoting the growth of these microbes, prebiotics can enhance the saccharolytic populations and SCFA output, which can strengthen mucosal barrier integrity, dampen inflammation, and thereby enhance metabolic resilience during aging.

Probiotics, typically Bifidobacterium and Lactobacillus strains delivered via fermented foods or supplements, support microbial balance, regulate mucosal immunity, and reduce inflammation^355,356. Yet, results from clinical trials in older individuals remain inconsistent, with generally modest immune benefits that are influenced by probiotic strain used, population heterogeneity, and variable timing of age-related microbiota decline^357,358. Of note, while viable probiotics are likely to offer the most potential benefits, because of metabolite production, even pasteurized probiotics can elicit some benefits in trials of insulin resistance³⁵⁹. While the mechanisms are not known, it could be linked to boosting oral tolerance to gut commensals¹⁰.

Postbiotics, the metabolites and structural components derived from probiotics, offer a more controllable means of disease intervention. Although human evidence remains limited, postbiotics have been shown to reduce inflammation and oxidative stress after stroke, despite modest neurological improvement³⁶⁰. A tyndallized probiotic-postbiotic combination was shown to lower microbial TMA and TMAO levels, mitigating TMAO-associated metabolic toxicity³⁶¹. Key GMDMs such as Lactobacillus-derived lipoteichoic acid reduce permeability and colitis-associated inflammation³⁶², while urolithin B suppresses HMGB1-TLR4-NF-κB signaling to alleviate age-related intestinal inflammation³⁶³. Similarly, sodium butyrate lowers Aβ burden and improves cognition in 5xFAD mice³⁶⁴.

While direct SCFA supplementation trials in generally healthy older adults remain uncommon, interventional studies in older-skewing populations show feasibility and signal. For example, resistant starch and inulin-type prebiotics can measurably increase SCFA production in older adults, accompanied by improvements in insulin sensitivity in some trials^365,366, while a randomized, double-blind trial in PD tested oral butyric/propionic acids and reported alleviated PD symptoms and improved intestinal immunity³⁶⁷. Together, these data demonstrate that SCFAs or pre/probiotics promoting SCFAs are promising candidates for restoring immunometabolic homeostasis in later life, while highlighting the need for more direct, adequately powered supplementation trials in broadly healthy aging cohorts.

6.2. Bacterial therapies

Manipulating bacterial communities or functions, such as via fecal microbiota transplantation (FMT) or genetically engineered microorganisms (GEMs), may recalibrate GMDM pools and extend healthspan. FMT from young donors to aged recipients restores SCFAs, BAs, and Trp-derived metabolites, improving gut barrier function, systemic metabolism, and cognition^368,369. In aged mice, FMT from young mice enriches tauro-conjugated BAs, acetate, and polyamines, while boosting vitamin B7/B9 biosynthesis, thereby rejuvenating the gut-brain and gut-retina axes³⁷⁰. FMT also mitigates inflammatory signaling, activates FoxO pathways, and promotes hematopoietic stem cell rejuvenation via Trp metabolites³⁷¹. Additionally, young murine microbiota reduces frailty and psychological decline, with lipid-derived metabolites, such as taurodeoxycholate and azelate, downregulating inflammation and supporting neuronal bioenergetics³⁷². In humans, early FMT studies in age-related neurodegenerative cohorts suggest that FMT can restore metabolite-immune homeostasis and improve healthspan. In a pilot capsule-FMT trial in cognitive impairment (ages 54–80), serum metabolomics showed remodeled BA compositions with decreased bilirubin, alongside a lower LPS-binding protein signal, implicating reduced endotoxin translocation³⁷³. Consistently, a randomized, placebo-controlled repeat-dose capsule-FMT trial in PD increased Firmicutes and enriched SCFA-producing taxa (e.g., Roseburia/Ruminococcaceae) while reducing Proteobacteria, indicating a shift away from pro-inflammatory, barrier-disruptive microbiota³⁷⁴. While FMT has yet to be tested in clinical trials for extending lifespan, by refreshing GMDM pools across multiple tissues, it may achieve metabolite-driven easing of inflammaging.

Complementing populational-level modification by FMT, GEMs allow precision engineering of GMDM landscapes. Strains of E. coli Nissle 1917 (EcN) and Bacteroides have been engineered to synthesize butyrate, IAA, or AhR agonists, thereby enhancing SCFA availability, remodeling BA metabolism, and protecting against intestinal and hepatic inflammation^375–377. EcN engineered to produce ILA alleviates colitis in mice³⁷⁸, and ILA itself acts as an anti-aging metabolite that mitigates intestinal senescence by reducing oxidative stress³⁷⁹. GEMs can also blunt harmful metabolite outputs, as shown by BA-hydrolase mutants that alter host lipid, circadian, and immune pathways³⁸⁰. Beyond metabolites, EcN-based platforms deliver therapeutic peptides including GLP-1, IGF1, and FGF19, alleviating diabetes, steatosis, and neurodegeneration while stabilizing microbial composition^381,382. Meanwhile, GEMs can be repurposed as diagnostic tools, which express biosensors detecting SCFAs³⁸³, indoles³⁸⁴, and kynurenines³⁸⁵, potentiating real-time monitoring of host-microbiome interactions that reflect the progression of “biome-aging”, a newly coined concept describing age-associated changes in the gut microbiome³⁵³.

6.3. Exercise

Though well-established as an anti-aging approach, exercise is also a potent modulator of the aging gut-metabolic axis, enriching health-associated, SCFA-producing taxa, such as A. muciniphila and Faecalibacterium, and thereby counteracting age-related dysbiosis and low-grade inflammation^386–388. In models of age-related cardiovascular disease, exercise-induced elevations in SCFAs attenuate atherosclerotic plaque development and vascular inflammation, effectively raising the threshold for cardiovascular events³⁸⁹. Beyond SCFAs, physical activity reshapes Trp metabolism. Voluntary exercise optimizes the indole pathway and may increase Trp availability for cerebral serotonin synthesis, buffering against neurodegenerative decline and psychosocial stress³⁹⁰. Together, through the gut-muscle axis, restoration of tolerogenic GMDMs and a more balanced immune profile preserves muscle function and immune resilience in late life. This reciprocal circuit generates a positive feedback loop, where exercise promotes protective GMDMs that maintain muscle capacity for continued activity, providing a non-pharmacological strategy to limit sarcopenia and inflammaging during aging.

6.4. Circadian rhythm modulation

Disruption of circadian rhythms during aging, reflected as fragmented sleep, blunted hormonal cycles, and dysregulated metabolism, exacerbates gut dysbiosis and alters diurnal GMDM oscillations^391,392. A recent study revealed that loss of circadian proteins REV-ERBα/β promotes gut ILC3-to-ILC1 conversion, increasing IFNγ at the expense of IL-22³⁹³, potentially fostering gut inflammation and dysbiosis. In turn, dysbiosis may reciprocally disrupt circadian clocks, exacerbating inflammaging³⁹⁴. Modulating either arm of this gut-mediated feedback loop could offer therapeutic value. Strengthening circadian clocks through time-restricted feeding (TRF) may alleviate metabolic disorders and restore the diurnal GMDM oscillations, a mechanism potentially contributing to extended lifespan in Drosophila³⁹⁵. TRF may also benefit younger individuals whose modern lifestyles, such as irregular mealtimes and calorie-dense diets, promote circadian misalignment and potentially accelerate aging³⁹⁶.

Conversely, alleviating dysbiosis with GMDM supplementation may restore and strengthen circadian rhythms. Butyrate and acetate are shown to resynchronize host circadian rhythms in peripheral tissues by modulating PER2 and BMAL1 rhythms, while an oat fiber diet is reported to increase SCFA concentrations and reverse the dysbiosis and circadian rhythm disruption in HFD-fed mice^397,398. Similarly, oral supplementation of microbe-derived unconjugated BAs enhances Clock and Arntl expression across murine ileum, colon, and liver, collectively re-aligning host circadian rhythms³⁹⁹. Trp/Phe metabolites and methionine further modulate circadian rhythms, while extending host sleep duration via melatonin synthesis and curbing oxidative stress, respectively^400,401. Thus, GMDMs may act as metabolic Zeitgebers that reinstate circadian rhythms, reduce dysbiosis, and eventually improve metabolic health during aging.

6.5. Dietary interventions

Nutritional interventions offer a feasible way to modulate health outcomes from daily life to long-term aging by reprogramming the metabolic states of the microbiota and the immune system. In 30-year cohort analyses, adherence to the Alternative Healthy Eating Index (AHEI), alternate Mediterranean Diet (aMED), Dietary Approaches to Stop Hypertension (DASH), and the Mediterranean-DASH Intervention for Neurodegenerative Delay (MIND) diet tracks with higher odds of healthy aging⁴⁰². These patterns are enriched in fermentable fibers, legumes, nuts, and phytochemicals, which push microbial metabolism toward SCFAs, polyphenol- and tryptophan-derived metabolites, and related products that strengthen barrier function, reduce antigen leak, and bias antigen-presenting cells toward tolerogenic programs, and expanding Tregs while dampening tonic myeloid activation that sustains inflammaging^403,404. Indeed, a polyphenol-rich diet in chronically inflamed adults aged ≥60 years reduces inflammatory indices with modest gains in microbial diversity⁴⁰⁵. Beyond observational links, the multicountry NU-AGE randomized trial in older adults supports causal directionality, showing that a Mediterranean-style intervention remodels gut ecology toward higher diversity and lower pathobiont burden alongside improved frailty- and inflammation-related phenotypes⁴⁰⁶. The Mediterranean diet is also enriched in vitamins, including K- and B-group vitamins (B1, B2, niacin, B6, folates)^407,408, and emerging human data indicate that colon-targeted vitamin delivery can measurably reshape microbial diversity and increase SCFA production, emphasizing that vitamin effects are context-dependent and microbiome-mediated rather than simple replacement⁴⁰⁹. In contrast, ultra-processed food (UPF)-heavy patterns promote dysbiosis with depletion of SCFA-producing taxa and enrichment of TMA-generating pathways, while high salt/saturated fat loads and food-additive exposure further drive salt-linked Th17 skewing⁴¹⁰ and emulsifier-associated mucus thinning that compromises epithelial barrier integrity⁴¹¹. While dietary patterns broadly shape aging outcomes, emerging precision nutrition aims to personalize these metabolite trajectories to baseline microbiome state.

6.6. Precision and personalized nutrition

As a precision medicine approach, personalized nutrition (PN) aims to tailor dietary strategies to improve healthy aging at the individual level⁴¹². PN programs based on individual postprandial responses, microbiome composition, and health history improve triglycerides, waist circumference, and microbiome diversity, outperforming the generalized advice⁴¹³. In older adults, digital PN integrating individualized foods and mobile tracking significantly reduces adiposity, steatosis indices, and blood pressure, while enhancing vitality and emotional well-being, critical for healthy aging⁴¹⁴. Moreover, the healthspan-extending efficacy of PN can be enhanced when coupled with metabolomic profiles that identify aging biomarkers, including altered amino acid, ketone body, lipoprotein, GlycA, and fecal SCFA/LCFA/BCAA compositions, permitting targeted supplementation of specific GMDMs based on individual “metabotypes”^415,416. Meanwhile, PN can target host nutrigenetics that modulate GMDM pathways. Polymorphisms in multiple genes, including NR1H4, IDO1, AHR, FFAR2, and FFAR3, are associated with age-related metabolic disorders or neurodegenerative diseases in humans^417–420, underscoring the potential of genotype-guided targeting of relevant GMDMs in PN. With emerging wearable/mobile technologies, future PN can benefit from real-time intake tracking and AI-driven analytics to capture not only what but also when and how to eat, with socio-culturally adapted designs to improve individual aging outcomes^421,422.

6.7. Artificial intelligence in precision geromedicine

While preclinical studies highlight the therapeutic promise of GMDMs, the translation to human aging is hampered by variable “exposomes”, including infections, lifestyles, and other factors, which shape individual aging trajectories. This “non-universality” of inflammaging, as shown by distinct inflammatory profiles in industrialized versus non-industrialized populations⁴²³ can be further confounded by sex, underscoring the need for precision geromedicine integrating microbiota, immune states, and large-scale behavioral and multi-omics data. This is where AI with its capacity for pattern recognition and predictive modeling becomes indispensable (Figure 4)⁴²⁴. Supervised machine learning (ML), from classical classifiers to graph neural networks, has already identified microbial biomarkers and predicted outcomes from fecal metagenomes⁴²⁵, while advanced models, like the Weighted Signed Graph Convolutional Neural Network (WSGMB), may capture microbial co-occurrence patterns during aging⁴²⁶. Meanwhile, explainable AI approaches, such as Shapley Additive Explanations (SHAP), can interpret if specific biomarkers were most influential in a particular patient, further augmenting the predictive power of ML into actionable and personalized treatments⁴²⁷.

Intriguingly, the recently developed MEBOCOST framework, which models metabolite-mediated cell-cell communication (mCCC) from transcriptomics and metabolomics data, may provide a transformative approach to decode how GMDMs interact with immune cells⁴²⁸. Its integration with temporal AI models, such as LSTM-based metabolomics⁴²⁹, Dynamo^430,431, and Dynamic Graph Neural Networks^432–434, could extend its ability to capture temporal dynamics of GMDM-immune cell interactions in aging, revealing time windows for maximal GMDM efficacy. This time-resolved mapping could also serve as an immunometabolic aging clock, predicting biological age and disease risks. With organ-specific multi-omics, it can further resolve local aging trajectories, and with sex-specific data, it may offer additional insights for developing sex-specific interventions, eventually accelerating bench-to-bedside translation beyond one-size-fits-all paradigms.

Collectively, modulation of diet, microbes, and lifestyles can harness GMDMs as multifaceted therapeutics against ARDs while expanding healthspan. AI-powered precision medicine offers particular promise for reprogramming inflammation, metabolism, and aging. However, realizing their full geroprotective potential will depend on continuous human validation with sustained integration of mechanistic and clinical research.

7. Conclusion

Far from simple byproducts of microbial metabolism, GMDMs are dynamic orchestrators weaving a network of bidirectional feedback loops between inflammaging and dysbiosis during “biome-aging”. However, biome-aging does not unfold the same way for everyone. While general patterns, such as depletion of anti-inflammatory metabolites (e.g., butyrate, IPA) and accumulation of pro-inflammatory metabolites (e.g., TMAO), may be conserved across populations, “exposomes” including infection, physical inactivity, polypharmacy, psychosocial isolation, and malnutrition³⁵³, can all contribute to individual aging phenotypes, or “ageotypes”, while cumulatively lowering the inflammatory threshold across organ systems.

The emergence of new facets of aging pathophysiology and unexplored gut-organ axes is enriching the pathways that GMDMs drive chronic inflammation and ARDs. Recent research shows that dysbiosis can worsen rheumatoid arthritis by driving AutoAb production^435,436. These studies not only unveil an understudied gut-joint inflammatory axis but also allude to the GMDM interplay with AutoAbs in aging, a potentially novel feature of aging research^22,437. Similarly, a recent study utilizing advanced ML tools reveals that dysbiosis drives age-related macular degeneration progression via inflammatory and metabolic pathways, indicating an emerging gut-retina axis where dysbiosis may fuel age-related ocular diseases⁴³⁸.

Several obstacles prevent the translation of feasible microbiome-targeting therapies (see Box 1). These challenges underscore a broader paradigm shift in geroscience, which should no longer be approached solely through observational studies but through integrated systemic profiling of host-microbiota interactions. Comprehensive multi-omics datasets, encompassing metagenomics, metabolomics, transcriptomics, and immune phenotyping, shall be collected from longitudinal cohorts with behavioral, nutritional, and exposomic data that reflect individual-specific ageotypes, marking the entrance into an era in which GMDM signatures may serve as actionable biomarkers of immune aging associated with tissue-specific dysfunctions. Future efforts that integrate AI-driven, spatially and temporally resolved, organ-specific immunometabolomics atlases will help translate GMDM therapies into clinical precision geromedicine. This approach aligns with the latest framework advocating organ-specific inflammaging clocks and longitudinal individualized aging assessments to improve the temporal reliability⁴³⁹, ultimately enabling proactive reprogramming of aging immunity to preserve organ functions and extend health span.

Box 1: Five important challenges in the field of GMDMs research and immune aging.

First, while interactions between individual GMDMs and specific immune cells have been extensively studied, the integrated effects of age-associated shifts in multiple GMDMs within the same immune niche are rarely investigated due to high complexity. For example, little research is done on the combined effects of reduced SCFAs and increased secondary BAs on immune cells during aging, though this is very likely the reality in actual aging processes. Future research should aim to investigate how simultaneous changes in multiple GMDM species can collectively reshape the immune landscape in aged tissue microenvironments.

Second, though sexual dimorphisms in GMDM-immune interactions are increasingly investigated, and many geroprotective compounds are found to be more effective in males⁴⁴⁰, current research remains fragmented, largely due to a disconnection between population-level observations and mechanistic insights. Broad, multi-cohort studies like de la Cuesta-Zuluaga et al. (2019), which focuses on how differences in hormonal milieu during puberty can contribute to age- and sex-dependent changes in gut microbiota diversity⁴⁴¹, are rarely further explored with cell-specific mechanistic validation. Meanwhile, mechanistic studies like Gao et al. (2021), focusing on how androgen-driven alterations to gut microbiota contribute to glucose intolerance⁴⁴², are rarely followed up in larger-scale populations. Future research should bridge this gap by coupling longitudinal, sex-stratified observations with targeted validation in animal/ex vivo models to identify translationally meaningful mechanisms.

Third, current research is limited within analyzing discrete GMDM changes between young and old models, while the temporal dynamic of continuous shifts in GMDM profiles are unstudied. Just like the temporal dynamics shown in DNA methylation in aged mouse colons⁴⁴³ and the development of autoantibodies that drive chronic inflammation²², the nonlinear waves of aging can be attributed to varying patterns of GMDM shifts, especially when associated with immune and endocrine fluctuations. Continuous monitorial studies combining multi-omics tools, including metabolomics, will help reveal potential windows during aging when GMDM interventions yield maximal benefit.

Fourth, fundamental differences in the GMDM reservoirs exist between humans and animal models, especially mice. The different primary BA pools in humans (cholic and chenodeoxycholic acids) and mice (muricholic acids) leads to distinct secondary BA pools⁴⁴⁴, while microbes mediating Trp metabolism⁴⁴⁵ and SCFA production⁴⁴⁶, as well as enzymes mediating TMAO synthesis⁴⁴⁷, also exhibit species-specific difference, limiting the extrapolation of mice data to human studies. Tools like humanized gnotobiotic mouse models, organoid-microbiome co-culture systems, and comparative metabolomics may help isolate conserved pathways shared among model organisms and humans, ensuring meaningful translation.

Fifth, much remains unexplored for the link between GMDM changes and psychosocial isolation, the 14th hallmark of aging. Though initial links have been established by a study revealing that psychosocial isolation in mice alters gut microbiota composition, reduces beneficial SCFAs, and promotes systemic inflammation and cognitive decline⁴⁴⁸, there is a lack of studies mechanistically resolving how psychosocial stressors reshape GMDM landscapes, leading to immunometabolic dysfunction or organ-specific aging. Integrating behavioral, neuroendocrine, metabolite, and immune profiling particularly within the gut-brain axis in both preclinical models and human cohorts will help elucidate how psychosocial stress drives molecular aging phenotypes through microbial alterations.

Highlights:

1. Gut-microbiota derived metabolites (GMDMs) orchestrate gut-immune circuits across multiple organs.

2. Age-related dysbiosis alters GMDM pools, driving systemic inflammation and age-related diseases (ARDs).

3. GMDMs can act as biomarkers of organ-specific aging states.

4. Changes in GMDMs may define personalized aging trajectories.

5. Artificial intelligence (AI) powered multi-omic analysis can identify therapeutic strategies for precision geromedicine.