Microbial short-chain fatty acids modulate CD8+ T cell responses and improve adoptive immunotherapy for cancer

The intestinal microbiota can influence the efficacy of cancer immunotherapies, including immune checkpoint inhibitors (ICIs) and adoptive cell therapies using tumor-specific CD8+ cytotoxic T lymphocytes (CTLs). Specific commensal bacteria (e.g., Akkermansia muciniphila and Bifidobacterium spp.) and defined microbial consortia have been shown to augment anti-tumor immunity and improve responses to PD-1/CTLA-4 blockade. While microbial organisms can modulate therapy outcomes, less is known about soluble microbial metabolites as tools to enhance cellular immunotherapies. Short-chain fatty acids (SCFAs) such as acetate, propionate, and butyrate are abundant microbial metabolites; butyrate is linked to Treg expansion and protection from autoimmunity and GVHD, yet SCFAs can also enhance effector T cell functions. Pentanoate (valerate), produced by low-abundance commensals like Megasphaera massiliensis, is not widely produced by dominant gut bacteria. The central question is whether ex vivo exposure of CTLs or CAR T cells to microbial SCFAs, particularly pentanoate and butyrate, can reprogram these cells metabolically and epigenetically to improve anti-tumor activity in vivo.

Prior studies show gut microbiota composition affects responses to ICIs and adoptive T cell therapies, with specific taxa (Akkermansia, Bifidobacterium) and defined bacterial consortia enhancing CD8+ T cell-mediated anti-tumor immunity. SCFAs (acetate, propionate, butyrate) modulate T cell biology: butyrate and propionate enhance Treg suppressive function and can promote CD4+ effector functions; SCFAs influence CD8+ effector and memory formation during viral infection. SCFAs act via epigenetic mechanisms (HDAC inhibition) and metabolic effects (influencing glycolysis/mTOR). Pentanoate has been less studied in cancer immunotherapy but has been reported to modulate CD4+ T cell function via HDAC inhibition and mTOR activation. These findings set the rationale to test SCFAs as ex vivo modulators to boost CTL and CAR T cell efficacy.

• Bacterial metabolite profiling: Multiple human gut commensals were cultured to stationary phase; cell-free supernatants were collected by sequential filtration. SCFA content (including pentanoate, butyrate, acetate, propionate) was quantified by chromatography/mass spectrometry. Megasphaera massiliensis and related species were assessed for pentanoate production.
• CTL generation and treatments: Murine CD8+ T cells were isolated from spleen/lymph nodes and activated under suboptimal CTL-polarizing conditions with anti-CD3/anti-CD28 plus IL-2. Cells were treated in vitro with SCFAs (acetate, propionate, butyrate, pentanoate; dose ranges optimized for efficacy/toxicity), with focus on butyrate and pentanoate. HDAC inhibitors were used as controls/comparators (trichostatin A, mocetinostat [class I], TMP-195 [class II]).
• HDAC activity assays: Fluorometric assays measured pan-HDAC and isoform-specific class I vs class II activity in the presence of SCFAs or bacterial supernatants and in CTL lysates.
• Metabolic assays: mTOR and S6 phosphorylation were analyzed by phospho-flow or immunoblot after SCFA treatment. Glycolytic activity was assessed by extracellular acidification rate (ECAR) using a Seahorse XFe96 flux analyzer. Metabolic inhibitors (rapamycin for mTOR; 2-deoxyglucose for glycolysis) were applied to test pathway dependence.
• GPCR dependence: CTLs deficient for SCFA receptors GPR41 (FFAR3) and GPR43 (FFAR2) were tested for IFN-γ induction and HDAC inhibition after pentanoate treatment to assess receptor involvement.
• In vivo CTL efficacy: Antigen-specific OT-1 CTLs (CD45.1+) were pretreated ex vivo with pentanoate or butyrate (or bacterial supernatant, or HDAC inhibitors) and adoptively transferred into tumor-bearing mice (e.g., B16-OVA melanoma; Panc02-OVA pancreatic model). Tumor volume and mass were measured; transferred CTLs in draining lymph nodes/tumors were quantified for frequency, number, and cytokine production (IFN-γ, TNF-α). Additional experiments included Rag1-deficient mice with co-transfer of Tregs to model suppressive microenvironments; CD25 expression, IL-2 production, STAT5 phosphorylation, and proliferation (CFSE) were assessed.
• CAR T cell studies (murine): ROR1-specific murine CD8+ CAR T cells were generated, treated with pentanoate or butyrate ex vivo, and tested for activation markers (CD25), cytokines (IFN-γ, TNF-α), and in vivo efficacy in Panc02/ROR1 pancreatic tumor model after adoptive transfer.
• CAR T cell studies (human): Human CD8+ T cells and ROR1-CAR T cells from healthy donors were pretreated (e.g., pentanoate 4 mM; butyrate 1 mM; mocetinostat 300 nM; TMP-195 2.5 μM). Readouts included mTOR/S6 phosphorylation, CD25 expression, cytokine secretion (ELISA), proliferation (CFSE), and cytotoxicity against ROR1+ K562 target cells at varying E:T ratios.
• Statistics: Two-tailed unpaired Student’s t-tests and linear mixed-effects models with Tukey correction for multiple comparisons were used; data typically presented as mean ± SEM.

• SCFA production by commensals: Low-abundance Megasphaera massiliensis strains were strong producers of pentanoate and butyrate, unlike many abundant gut species. Megasphaera elsdenii produced pentanoate at lower levels.
• Effector function boost in CTLs: Supernatants from M. massiliensis and direct treatment with pentanoate or butyrate increased frequencies of IFN-γ+ TNF-α+ CD8+ T cells and elevated TNF-α secretion versus controls; acetate had minimal effect. Dose-finding established effective, low-toxicity concentrations; pentanoate and butyrate were prioritized.
• Epigenetic mechanism: Pentanoate and butyrate inhibited class I HDAC activity (but not class II) in CTLs and in enzyme assays. Pan-HDAC inhibitor TSA and class I inhibitor mocetinostat phenocopied increases in IFN-γ/TNF-α, whereas class II inhibitor TMP-195 did not. Valproate (a branched SCFA/HDAC inhibitor) similarly enhanced IFN-γ and induced T-bet/Eomes.
• Metabolic mechanism: Pentanoate increased phosphorylation of mTOR and downstream S6 protein and elevated glycolysis (ECAR). Rapamycin or 2-DG reduced IFN-γ induction, supporting mTOR/glycolysis involvement. Mocetinostat/TMP-195 did not activate mTOR, indicating SCFAs exert both HDAC-dependent and metabolic (HDAC-independent) effects.
• GPCR independence: Enhancement of CTL effector function and HDAC inhibition by pentanoate occurred in GPR41/FFAR3 and GPR43/FFAR2 double-deficient CTLs, indicating effects are independent of these SCFA GPCRs.
• In vivo CTL efficacy: Ex vivo pretreatment of OT-1 CTLs with pentanoate or butyrate reduced tumor volume and mass in B16-OVA melanoma; higher frequencies and absolute numbers of transferred IFN-γ+ TNF-α+ CTLs were detected in draining lymph nodes. M. massiliensis supernatant and class I HDAC inhibitor (mocetinostat) improved, while class II inhibitor (TMP-195) did not improve anti-tumor responses. In Panc02-OVA pancreatic model, pentanoate-pretreated CTLs showed improved engraftment/proliferation, elevated CD25, stronger IL-2/STAT5 signaling, and increased IL-2 production; 2-DG blocked pentanoate-induced CD25 upregulation.
• CAR T cell enhancement (murine): ROR1-CAR T cells pretreated with pentanoate or butyrate upregulated CD25 and produced more IFN-γ/TNF-α. In Panc02/ROR1 tumors, pentanoate-pretreated CAR T cells reduced tumor volume/weight and showed higher frequencies and numbers of IFN-γ+ TNF-α+ CAR T cells in tumors. Doses used included ~0.75 mM butyrate and 2.5 mM pentanoate in cited experiments.
• CAR T cell enhancement (human): In human CD8+ T cells and ROR1-CAR T cells, pentanoate and butyrate increased mTOR and S6 phosphorylation, CD25 expression, and secretion of IFN-γ/TNF-α. Pentanoate-pretreated human CAR T cells exhibited enhanced proliferation and cytolytic activity against ROR1+ targets compared with untreated CAR T cells.
• Representative quantitative example (from figure panel): Percent IFN-γ/TNF-α+ cells increased versus untreated (2.4% ± 0.1) with butyrate (9.3% ± 0.2) and pentanoate (8.3% ± 0.3); mocetinostat showed intermediate increase (~4.1% ± 0.1), TMP-195 minimal (~3.8% ± 0.1).

The study addresses whether microbial metabolites can be harnessed to improve adoptive cellular immunotherapies. Pentanoate and butyrate reprogram CTLs and CAR T cells through dual mechanisms: inhibition of class I HDACs leading to epigenetic activation of CTL-associated genes, and metabolic activation via mTOR/glycolytic pathways, independent of GPR41/43 signaling. This reprogramming increases expression of effector molecules (CD25, IFN-γ, TNF-α), augments IL-2-STAT5 signaling, and enhances proliferation and persistence in vivo, translating into superior tumor control in murine melanoma and pancreatic cancer models. The approach extends to human ROR1-CAR T cells, where SCFA pretreatment improved activation, cytokine production, and cytotoxicity in vitro, suggesting translational potential. These findings integrate with literature showing microbiota modulation of cancer immunotherapy, highlighting a metabolite-based strategy that is compatible with ex vivo T cell manufacturing and potentially with in vivo administration. The results are significant for enhancing the potency of adoptive T cell therapies and may inform GMP-compliant process optimizations.

Pentanoate and butyrate, two microbiota-derived SCFAs, enhance CTL and CAR T cell anti-tumor activity via class I HDAC inhibition and activation of mTOR-driven metabolism, increasing effector cytokine production and activation markers and improving tumor control in vivo. The effects are independent of SCFA GPCRs GPR41/43. Pentanoate also augments human ROR1-CAR T cell function, indicating translational promise. These microbial metabolites can potentially be integrated into ex vivo manufacturing of therapeutic T cells or considered for adjunctive administration post-infusion. Future work should define optimal dosing, timing, delivery routes, and safety, and evaluate efficacy across tumor models and CAR specificities, including clinical studies.

• Preclinical models: Findings are based on murine models and in vitro human assays; clinical efficacy and safety remain untested.
• Dosing and delivery: Optimal concentrations, exposure durations, and administration routes for SCFAs in clinical settings are not established and require further investigation.
• Specificity and breadth: Results are shown for OT-1 CTLs and ROR1-CAR T cells; generalizability to other antigens, CAR designs, and solid tumor microenvironments needs validation.
• Mechanistic scope: While class I HDAC inhibition and mTOR activation are implicated, the precise molecular targets, epigenomic landscapes, and long-term effects on memory/exhaustion were not comprehensively mapped.
• GPCR expression context: Lack of involvement of GPR41/43 was shown in spleen/LN-derived CTLs; effects in tissue-resident or gut-associated T cells with higher receptor expression were not examined.