Associations of microbial and indoleamine-2,3-dioxygenase-derived tryptophan metabolites with immune activation in healthy adults

Abstract

Background: Tryptophan (Trp) metabolites from intestinal bacteria (indole, indole acetic acid [IAA] and indole propionic acid [IPA]), and the Trp metabolite kynurenine (Kyn) from the indoleamine 2,3-dioxygenase (IDO) pathway, are aryl hydrocarbon receptor (AhR) agonists and thus, can regulate immune activity via the AhR pathway. We hypothesized that plasma concentrations of these metabolites would be associated with markers of immune activation in a cohort of healthy adults in a manner consistent with AhR-mediated immune-regulation. We also hypothesized that the plasma Kyn/Trp ratio, a marker of IDO activity, would be associated with immune markers reflecting IDO activation in innate immune cells. Finally, we hypothesized that some intestinal bacteria would be associated with plasma indole, IPA and IAA, and that these bacteria themselves would be associated with immune markers.

Methods: A novel set of 88 immune markers, and plasma Trp metabolites, were measured in 362 healthy adults. Bacterial taxa from stool were identified by 16S rRNA gene analysis. Multiple linear regression analysis was used to identify significant associations with immune markers.

Results: The sum of indole and IAA was positively associated with natural killer T-cells levels. Kyn and Kyn/Trp were positively associated with neopterin and IP-10, markers of type 1 immunity, and TNF-α and C-reactive protein (CRP), markers of the acute phase response, and the regulatory cytokine IL-10. Three bacteria negatively associated with Trp metabolites were associated with markers of immune activation: the family Lachnospiraceae with higher lymphocyte counts but lower level of activated CD4 T-cells, the genus Dorea with higher production of IFN-γ by T-cells in PBMC cultures, and the genus Ruminococcus with higher production IL-6 in PBMC cultures stimulated with bacterial lipopolysaccharide (LPS).

Conclusions: In this cohort of healthy adults bacterial Trp metabolites were not strongly associated with immune markers. Conversely, the Kyn/Trp ratio was strongly associated with markers of systemic inflammation and the acute phase response, consistent with IDO activation in innate immune cells. Finally, commensal bacteria associated with lower plasma (and perhaps intestinal) levels of bacterial Trp metabolites were associated with greater immune activation, possibly reflecting decreased regulatory immune activity related to lower intestinal levels of bacterial indole metabolites.

Keywords: immunity; indole; indole acetic acid (IAA); indole producing bacteria; indole propionic acid (IPA); inflammation; kynurenine; tryptophan.

Figure 1

Distribution of plasma Trp and Trp metabolites across females (F) and males (M), three age categories (Cat.), A1-A3: [18–34 y (n=129), 35–49 y (n=119), and 50–66 y (n=114)] and three BMI categories, B1-B3: [18–24.99 kg/m² (n=139), 25.00–29.99 kg/m² (n=131), and 30.00–44 kg/m² (n=92)]. Data are presented as medians, and 5th and 95th percentiles. Significant differences between groups within the sex, age and BMI categories are marked by different letters. (A) Trp, (B) Kynurenine, (C) Kynurenine : Tryptophan (Kyn/Trp) ratio, (D) Indole, (E) Indole Acetic Acid, (F) Indole Propionic Acid.

Figure 2

Association of plasma indole, IAA and IPA with plasma markers of immune activity in a linear regression model adjusted for age, sex, and BMI categories and CMV infection status. The association of IPA with immune markers was also adjusted for YMCA step scores. Heatmaps (A–F) show coefficients of association with adjusted beta values for the six immune groupings (A) Effector/Memory T-cells and activation level (n = 24), (B) Other lymphocytes including Th cells, NK, NK T-cells and B cells (n=11), (C) PBMC cytokines (n=8), (D) Complete Blood Count (n = 15), (E) Innate cell activation (n=11), and (F) Plasma (n = 19). ‘Ϯ’ indicates significant with unadjusted p values <0.05. No significant associations were seen using with Benjamini-Hochberg adjusted p-values <0.05.

Figure 3

Association of plasma Kyn and Kyn/Trp with plasma markers of immune activity in a linear regression model adjusted for age, sex and BMI categories and CMV infection status. Heatmaps (A–F) show coefficients of association with adjusted beta values for the six immune groupings (A) Effector/Memory T-cells and activation level (n=24), (B) Other lymphocytes including Th cells, NK, NK T-cells and B cells (n=11), (C) PBMC cytokines (n=8), (D) Complete Blood Count (n=15), (E) Innate cell activation (n=11), and (F) Plasma (n=19). ‘Ϯ’ indicates significant association with unadjusted p values<0.05. ‘*’ indicates significant association with Benjamini-Hochberg adjusted p-values<0.05. (A) PD-1, Programmed Death-1; Mem, Memory; Eff, Effector; (B) NK, Natural killer cells, NKT, Natural killer T-cells; (D) Neut, Neutrophils; Lym, Lymphocytes; Plt, Platelets; SII, (Platelet×Neutrophil)/Lymphocyte; (E) Mono, Monocytes; Alt, Alternate; Inter, Intermediate; Eos, Eosinophils; (F) CRP, C-Reactive Protein; SAA, Serum Amyloid A; ICAM-1/VCAM-1, Intracellular/Vascular cell adhesion molecule; TNF-α, Tumor Necrosis Factor-α; sCD14, Soluble CD14; MMP, Matrix metalloproteinase; MPO, Myeloperoxidase; MDC, Macrophage-derived chemokines.

Figure 4

The standardized effect sizes, R², and p values for significant association between the immune biomarkers and (A) Kyn, (B) Kyn/Trp ratio. Dotted lines represent the range of the small (0.1-0.3), medium (0.3-0.5) and large (0.5-1) effects.

Figure 5

Relative abundance of microbial taxa across the three terciles of Trp metabolites. Variability of the percent relative abundance of microbial taxa across the three groups of Trp metabolites is presented using a Tukey's boxplot. The boxplot elements are defined as following: center line, median; box limits, upper and lower quartiles; whiskers, 1.5 × interquartile range. Circles indicate outliers. As described in Results, there was a significant negative association between (A) IAA and family Lachnospiraceae, and between (B) IPA and family Erysipelotrichaceae and (C) IPA and genera Ruminococcus and (D) Dorea. Significant differences between the groups of Trp metabolites are marked by different letters.