Peripheral kynurenine/tryptophan ratio is not a reliable marker of systemic indoleamine 2,3-dioxygenase: A lesson drawn from patients on hemodialysis

Abstract

Indoleamine 2,3-dioxygenase (IDO) has emerged as a pivotal enzyme for mediating immune tolerance. Because IDO metabolizes tryptophan into kynurenine, the plasma kynurenine/tryptophan (Kyn/Trp) ratio has been widely used as a marker of systemic IDO. Here, we evaluated the clinical value of using the plasma Kyn/Trp ratio to estimate cell-mediated immune responses to tuberculin skin testing and risk of new bacterial infection. We also compared the Kyn/Trp ratio to a novel IDO marker, the IDO median fluorescence index (MFI) of peripheral blood mononuclear cells, which was determined by flow cytometry. In 228 patients from two hemodialysis centers, the two IDO markers were higher in patients than in healthy controls but were not correlated with each other. In vitro experiments demonstrated that peripheral blood mononuclear cells could not metabolize tryptophan into kynurenine, indicating that the increased Kyn/Trp ratio was IDO-independent. Skin induration diameters of tuberculin skin testing were correlated with the IDO MFI (negatively), but not the Kyn/Trp ratio. Further, in a 24-month prospective cohort, the Kyn/Trp ratio was not correlated with clinical infection. Alternatively, patients with a higher IDO MFI had a lower accumulative infection-free survival rate. Using a Cox proportional hazard model, it was also revealed that a higher IDO MFI was significantly associated with new bacterial infection. Taken together, these results indicate that the Kyn/Trp ratio is not a reliable circulating IDO marker in hemodialysis patients. However, the IDO MFI reflects an immunocompromised state and thus might be a potential clinical marker of bacterial infection.

Keywords: 3-dioxygenase; hemodialysis; immune tolerance; indoleamine 2; infection; kynurenine to tryptophan ratio.

Figure 1. IDO expression in PBMCs

(A) Typical blueprints of CD303a(+) IDO^high subsets in the controls and hemodialysis patients. Data were obtained by flow cytometry. (B) Difference in the IDO MFI of PBMCs between the controls and hemodialysis patients (*P<0.05, controls vs. patients). MFI: median fluorescence index; PBMCs: peripheral blood mononuclear cells.

Figure 2. Relationships between peripheral blood Trp and its metabolites with IDO

(A) Trp levels were significantly lower in the hemodialysis patients than in the controls. (B) Kyn levels were significantly higher in the hemodialysis patients than in the controls. (C) The Kyn/Trp ratio was significantly higher in the hemodialysis patients than in the controls. (*P<0.05, controls vs. patients). (D) The Kyn/Trp ratio did not show a correlation with the IDO MFI in PBMCs (GGH: r_s0.059, P=0.426; WHH: r_s 0.106, P=0.485). MFI: median fluorescence index; PBMCs: peripheral blood mononuclear cells.

Figure 3. In vitro ability of PBMCs to metabolize Trp

(A) Cells were cultured with 100 μmol/L Trp for 4 h. After culturing PBMCs isolated from the controls and the hemodialysis patients for 4 h, the Trp and Kyn levels in the culture medium did not change significantly. In 293 cells transfected with IDO(+), the concentration of Trp decreased and the Kyn level increased, but addition of the IDO-specific inhibitor 1-MT inhibited these changes (*P<0.05, IDO(+) cells vs. IDO(+) cells + 1-MT). (B) Co-culture of PBMCs and Trp for 12 h. At 4 and 6 h, generation of Kyn or loss of Trp was not observed, whereas at 8 and 12 h, Kyn generation of up to 2.0 μmol/L and a 40% decrease in Trp were observed. The addition of 1-MT did not influence the levels of Trp or Kyn. MFI: median fluorescence index; PBMCs: peripheral blood mononuclear cells.

Figure 4. Relationship between IDO and tuberculin test results

(A) The IDO MFI was higher for positive TST results than for negative TST results (*P<0.05). (B) The Kyn/Trp ratio did not differ significantly between negative and positive TST results. (C) The diameter of the TST induration was negatively correlated with the IDO MFI (r_s -0.297, P=0.045). (D) The TST induration was not correlated with the Kyn/Trp ratio (r_s -0.248, P=0.096). MFI: median fluorescence index; PBMCs: peripheral blood mononuclear cells; TST: tuberculin skin test.

Figure 5. Infection-free survival curves based on the interquartile range of the IDO MFI or the Kyn/Trp ratio

(A) Comparison of Q4 of the IDO MFI with the Q1 revealed a lower cumulative infection-free survival rate (P<0.01). (B) Comparison of the Q4 and Q1 of the Kyn/Trp ratio revealed no difference in the cumulative infection-free survival rate (P=0.787). MFI: median fluorescence index.