Cancer Biology

A mediterranean-mimicking diet harnesses gut microbiota–derived 3-IAA to rejuvenate T cell

Xin Yu
Wenge Li
Hongfang Feng
Zhiyu Li
Hongmei Zheng
Shengrong Sun
Juanjuan Li author has email address
Bei Li author has email address
Qi Wu author has email address

Department of Breast Surgery, Fujian Provincial Hospital, Fuzhou University Affiliated Provincial Hospital, Fuzhou, China
Department of Breast and Thyroid Surgery, Renmin Hospital of Wuhan University, Wuhan, China
Department of Oncology, Fuzhou Pulmonary Hospital of Fujian, Fuzhou Pulmonary Hospital of Fujian, Fuzhou, China
Department of Breast Surgery, Hubei Cancer Hospital, Tongji Medical College, Huazhong University of Science and Technology, Hubei Provincial Clinical Research Center for Breast Cancer, Wuhan Clinical Research Center for Breast Cancer, Wuhan, China
Department of Pathology, Renmin Hospital of Wuhan University, Wuhan, China
Zhejiang Provincial Key Laboratory of Pancreatic Disease, Hangzhou, China
Department of Hepatobiliary and Pancreatic Surgery, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China

https://doi.org/10.7554/eLife.110468.1

Open access
Copyright information

Figures and data

MedDiet inhibits tumor progression.
(A) Composition of different dietary regimens. (B–F) Subcutaneous tumor models subjected to dietary interventions: (B) Schematic of the experimental design; (C–F) tumor growth curves, body-weight changes, blood glucose levels, and cumulative food intake in (C) male C57BL/6 mice bearing MC38 colon carcinoma, (D) female C57BL/6 mice bearing MC38 colon carcinoma, (E) male C57BL/6 mice bearing MCA205 sarcoma, and (F) female BALB/c mice bearing EMT6 breast carcinoma. (G) Prophylactic MedDiet intervention in male C57BL/6 mice challenged with MC38 cells: schematic of the experimental design; tumor growth curves, body-weight changes, blood glucose levels, and cumulative food intake. (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001)

The antitumor effect of the MedDiet depends on adaptive immunity.
(A) Nude mice bearing subcutaneous MC38 colon carcinoma subjected to dietary interventions: schematic of the experimental design; tumor growth curves. (B) Male C57BL/6 mice bearing subcutaneous MCA205 fibrosarcoma subjected to dietary interventions: schematic of the experimental design; tumor growth curves. (C, D) (C) Representative images of CD8 immunohistochemistry; (D) quantification of CD8-positive cells. (E) Quantification of CD8⁺ T-cell infiltration levels analyzed using CIBERSORT based on RNA-seq data. (F) Bubble plot displaying pathways upregulated in the MedDiet group compared to other groups, as determined by GSEA. (G) GSEA comparing IFNγ response pathway activation between the MedDiet and control groups. (Scale bar, 20 μm. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001)

Gut microbiota-dependent mechanisms reveal the pivotal role of B. thetaiotaomicron in MedDiet-mediated antitumor effects.
(A) Subcutaneous MC38 colon carcinoma model subjected to dietary interventions following antibiotic-mediated depletion of gut microbiota: schematic of the experimental design and tumor growth curves. (B–E) Metagenomic sequencing analysis of gut contents from diet-intervened mice: (B) PCoA; (C) bacterial taxonomic composition at the genus level; (D, E) LEfSe results—(D) LDA score histogram and (E) cladogram. (F) Subcutaneous MC38 colon carcinoma model subjected to bacterial gavage interventions: schematic of the experimental design and tumor growth curves. (G, H) (G) Representative images of CD8 immunohistochemistry; (H) quantification of CD8-positive cells. (I–N) Flow cytometric analysis of intratumoral CD8⁺ T cells: (I, J) assessment of intratumoral CD8⁺ T cells—(I) representative plots and (J) quantification of CD8⁺ T-cell infiltration; (K–N) assessment of intratumoral effector CD8⁺ T cells (IFNγ⁺ CD8⁺ T cells) and exhausted CD8⁺ T cells (PD1⁺ TIM3⁺ CD8⁺ T cells)—(K, M) representative plots, (L) quantification of effector CD8⁺ T cells, (N) quantification of exhausted CD8⁺ T cells. (Scale bar, 20 μm. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001)

Metabolomics reveal 3-IAA as a key effector molecule in the microbiota–immune axis.
(A, B) Untargeted metabolomic analysis of plasma samples from diet-intervened mice: (A) PCA; (B) KEGG enrichment analysis highlighting pathways upregulated in the MedDiet group; (C) heatmap of abundance of differential metabolites in the tryptophan metabolism pathway. (D) Quantification of 3-IAA concentrations in plasma samples from bacteria-gavaged mice using ELISA. (E) Subcutaneous MC38 colon carcinoma model subjected to B. thetaiotaomicron gavage and/or tryptophan-restricted diet: schematic of the experimental design; tumor growth curves. (F–J) Subcutaneous MC38 colon carcinoma model treated with oral 3-IAA: (F) schematic of the experimental design; (G) quantification of plasma 3-IAA concentrations using ELISA; (H) tumor growth curves; (I, J) (I) representative images of CD8 immunohistochemistry; (J) quantification of CD8-positive cells. (K) Subcutaneous MC38 colon carcinoma model subjected to CD8 in vivo neutralizing antibody and/or oral 3-IAA treatment: schematic of the experimental design; tumor growth curves. (L) Subcutaneous MC38 colon carcinoma models treated with 3-IAA, anti-PD1 antibody, or their combination: schematic of the treatment regimen; tumor growth curves over time; Kaplan–Meier survival curves. (Scale bar, 20 μm. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001)

3-IAA alleviates the ISR to inhibit CD8⁺ T cell exhaustion.
(A–D) Flow cytometric analysis of CD8⁺ T cells under acute or chronic stimulation over 8 days with or without 3-IAA treatment: (A) schematic of T-cell activation, stimulation, and treatment; (B) representative plots on day 8; (C) quantification of effector CD8⁺ T cells; (D) quantification of exhausted CD8⁺ T cells. (E) Lactate dehydrogenase (LDH) release assay evaluating cytotoxicity of 3-IAA-treated OT1 mouse-derived T cells against MC38-OVA cells. (F) Bubble plot illustrating gene set enrichment analysis (GSEA) of pathways downregulated in chronically stimulated T cells following 3-IAA treatment. (G) Western blot analysis showing expression levels of p-eIF2α, total eIF2α, ATF4, and CHOP in acutely and chronically stimulated T cells with or without 3-IAA treatment. (H–M) Flow cytometric analysis of intratumoral CD8⁺ T cells following oral administration of 3-IAA, evaluating infiltration, effector function, exhaustion levels, and CHOP expression: (H) representative plots; (I) quantification of CD8⁺ T-cell infiltration; (J) quantification of effector CD8⁺ T cells; (K) quantification of exhausted CD8⁺ T cells; (L) histogram of CHOP expression levels; (M) quantification of CHOP expression. (N) Western blot analysis of p-eIF2α, total eIF2α, ATF4, and CHOP in chronically stimulated T cells treated with 3-IAA, GSK, or their combination. (O–S) Flow cytometric analysis of CD8⁺ T cells treated in vitro with 3-IAA, GSK, or their combination, assessing effector function, exhaustion levels, and CHOP expression: (O) representative plots; (P) quantification of effector CD8⁺ T cells; (Q) quantification of exhausted CD8⁺ T cells; (R) histogram of CHOP expression levels; (S) quantification of CHOP expression. (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001).

A MedDiet score correlates with enhanced antitumor immunity and predicts favorable outcomes in human cancers.
(A–D) Correlation analyses between MedDiet scores and (A) CD8⁺ T-cell infiltration levels, (B) cytotoxic scores, (C) central memory scores, and (D) effector memory scores.in patients from the TCGA pan-cancer cohort (E) Kaplan–Meier survival curves for patients with high versus low MedDiet score-expressing tumors. (F–N) Kaplan–Meier survival curves for patients with high versus low MedDiet score-expressing tumors following immunotherapy: (F) melanoma; (G) non-small cell lung cancer (NSCLC); (H) small cell lung cancer; (I) urothelial carcinoma; (J) hepatocellular carcinoma; (K) renal cell carcinoma; (L) pancreatic cancer; (M) head and neck cancer; and (N) mixed cancer cohorts.

Graphical abstract.
This schematic illustrates how a MedDiet suppresses tumor growth by activating a microbiota–metabolite–immune axis centered on gut microbiota-derived 3-IAA, thereby rejuvenating antitumor CD8⁺ T-cell immunity. A Mediterranean-mimicking diet selectively enriches B. thetaiotaomicron in the gut microbiota; administration of this bacterium alone recapitulates the diet’s tumor-suppressive effects. Both the MedDiet and B. thetaiotaomicron colonization significantly elevate systemic levels of the tryptophan-derived metabolite 3-IAA. Mechanistically, 3-IAA potentiates CD8⁺ T-cell cytotoxicity and reverses T-cell exhaustion by suppressing the integrated stress response, thereby sustaining robust antitumor immunity in vivo. Created with BioRender.com.

Antitumor effect of MedDiet is dependent on adaptive immunity.
Immune cell infiltration levels analyzed using the CIBERSORT algorithm based on RNA-seq data. (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001).

3-IAA alleviates the ISR to inhibit CD8+ T cell exhaustion.
(A) CCK-8 assay quantifying relative cell numbers of MC38 cells treated with 3-IAA. (B, C) CFSE staining assessing proliferation of CD8⁺ T cells treated with 3-IAA: (B) Proliferation index; (C) Representative plots.

Serum 3-IAA levels are associated with favorable clinical outcomes in pancreatic-cancer patients.
(A–C) Hamburg cohort: (A) serum 3-IAA concentrations in responders versus non-responders; (B) PFS and (C) OS Kaplan–Meier curves for patients with high versus low serum 3-IAA. (D) Munich cohort: PFS Kaplan–Meier curve for patients with high versus low serum 3-IAA.

Sign up for email alerts