Biomarkers of inflammation, immunosuppression and stress with active disease are revealed by metabolomic profiling of tuberculosis patients

Abstract

Although tuberculosis (TB) causes more deaths than any other pathogen, most infected individuals harbor the pathogen without signs of disease. We explored the metabolome of >400 small molecules in serum of uninfected individuals, latently infected healthy individuals and patients with active TB. We identified changes in amino acid, lipid and nucleotide metabolism pathways, providing evidence for anti-inflammatory metabolomic changes in TB. Metabolic profiles indicate increased activity of indoleamine 2,3 dioxygenase 1 (IDO1), decreased phospholipase activity, increased abundance of adenosine metabolism products, as well as indicators of fibrotic lesions in active disease as compared to latent infection. Consistent with our predictions, we experimentally demonstrate TB-induced IDO1 activity. Furthermore, we demonstrate a link between metabolic profiles and cytokine signaling. Finally, we show that 20 metabolites are sufficient for robust discrimination of TB patients from healthy individuals. Our results provide specific insights into the biology of TB and pave the way for the rational development of metabolic biomarkers for TB.

Figure 1. Examples of metabolite patterns in tuberculosis patients (TB^active), healthy uninfected (TST^–) and latently infected (TST⁺) individuals.

(A) Changes in relative abundance of three exemplary small metabolites: fibrinopeptide A, inosine and kynurenine. Red line indicates sample mean, blue line indicates sample median. Stars indicate significant differences between profiles (result from t-test corrected for multiple testing; *, p<0.05; **, p<0.01; ***, p<0.001) (B) Correlation between abundance of inosine and abundance of hypoxanthine and choline. Colors and symbols denote study groups: grey squares, TST^–; green circles, TST⁺; red triangles, TB^active. Spearman correlation coefficients (rho) and corresponding p-values are given.

Figure 2. Heat map showing fold changes of small metabolic compounds in the three study groups, TB patients, healthy uninfected and latently infected individuals.

Fold changes are relative to the average abundance in the TST^– group. Red indicates relative abundance higher than average in the TST^– group; blue indicates relative abundance lower than average in the TST^– group. Horizontal axis: samples belonging to the three study groups; vertical axis: top 50 compounds selected by variable importance in RF analysis, including compounds that could not be identified, but were strong predictors of sample status. Color bars above the heat map denote study groups: grey, TST^–; green, TST⁺; red, TB^active. See also Figure S2.

Figure 3. Network showing functional relationships between the small metabolic compounds in TB patients, healthy uninfected and latently infected individuals.

Nodes correspond to metabolites; edges correspond to statistically significant correlation between residual small metabolite profiles corrected for study classes. Colors correspond to differences between the TST⁺ and classes (see Figure S3 for additional comparisons). Color intensity indicates significance of difference with darker colors corresponding to more significant differences. Metabolites with adjusted p value >0.05 are not colored. Line thickness corresponds to the absolute Spearman correlation coefficients corrected for groups (see Methods). See also Figure S3.

Figure 4. Abundance of cytokines and their correlation with selected metabolites in TB patients, healthy uninfected and latently infected individuals.

(A) Strip charts showing abundances of eight cytokines that differed significantly between the study groups. Significance thresholds for a two-tailed t-test corrected for multiple testing: *, p<0.05; **, p<0.01; ***, p<0.001. Blue line indicates group median, red line indicates group mean. (B) Correlations between metabolic compounds and cytokines. Grey squares, healthy controls (TST^–); green circles, latently infected individuals (TST⁺); brown triangles, active TB patients (TB^active). Spearman correlation coefficient (rho) and p-values are given.

Figure 5. Demonstration of IDO1 expression and kynurenine production in response to M. tuberculosis infection and regulation of M. tuberculosis-specific T-cell responses by kynurenines.

(A) Immunohistochemistry staining of formalin-fixed, paraffin-embedded tissue of a murine pulmonary TB lesion stained with anti-IDO polyclonal antibody; staining representative of lesions from five animals. Bar is equal to 200 nm. Human monocyte-derived dendritic cells (DCs) (B) and macrophages (C) were infected with M. tuberculosis H37Rv or stimulated with irradiated and heat-killed M. tuberculosis H37Rv for 24 h and indoleamine 2,3 dioxygenase 1 (IDO1) gene expression was measured by qPCR. Mean and standard deviation (SD) of fold-change IDO1 gene expression of one donor representative of four. Line indicates minimal significant fold change threshold equal to 1.5. DCs (D) and macrophages (E) were infected with M. tuberculosis H37Rv and cell culture supernatants were collected at indicated times for measurement of kynurenines by HPLC; kynurenine levels from uninfected controls were subtracted. Means and SD of four donors are depicted (C and D). Star indicates significance (p<0.05) in Friedman test.(F) Human monocyte-derived DCs were pulsed with purified protein derivative (PPD) and mannosylated lipoarabinomannan (ManLAM) and co-cultured for 4 days with autologous CFSE-labeled purified T cells (DC:T cell ratio 1∶20) in the presence or absence of 1-methyl-DL-tryptophan (1-MT-DL) and 3-OH-kynurenine (Kyn). Cell proliferation was assessed by CFSE dilution using flowcytometry. Figure representative of three independent experiments (ANOVA F = 4.1 for CD⁺CD4⁺ cells and F = 3.3 for CD3⁺CD8⁺ cells, p<0.05).