Microbiota-dependent metabolite and cardiovascular disease marker trimethylamine-N-oxide (TMAO) is associated with monocyte activation but not platelet function in untreated HIV infection

Judith M. Haissman, Anna K. Haugaard, Sisse R. Ostrowski, Rolf K. Berge, Johannes R. Hov, Marius Trøseid, Susanne D. Nielsen

HIV infection is associated with increased risk of cardiovascular disease beyond that explained by traditional risk factors. Altered gut microbiota, microbial translocation, and immune activation have been proposed as potential triggers. The microbiota-dependent metabolite trimethylamine-N-oxide (TMAO) predicts myocardial infarction (MI) in the general population and has recently been shown to induce platelet hyperreactivity. In the present study, we investigated if TMAO was associated with platelet function, microbial translocation, and immune activation in both untreated and combination anti-retroviral therapy (cART) HIV infection. TMAO and the pre-cursors betaine, choline, and carnitine were quantified by mass-spectrometry in plasma samples from a previously established cross-sectional cohort of 50 untreated and 50 cART treated HIV-infected individuals. Whole-blood impedance aggregometry, C-reactive protein, sCD14, and lipopolysaccharide were assessed as measures of platelet function, inflammation, monocyte activation, and microbial translocation, respectively. TMAO was not associated with platelet aggregation response after stimulation with four different agonists, or with overall hypo- or hyperreactivity in untreated or treated HIV-infected individuals. In contrast, sCD14 a marker of both monocyte activation and microbial translocation was independently associated with TMAO in untreated HIV-infection (R = 0.381, P = 0.008). Lower levels of carnitine [32.2 (28.4-36.8) vs. 38.2 (33.6-42.0), P = 0.001] and betaine [33.1 (27.3-43.4) vs.37.4 (31.5-48.7, P = 0.02], but similar TMAO levels [3.8 (2.3-6.1), vs. 2.9 μM (1.9-4.8) P = 0.15] were found in cART treated compared to untreated HIV-infected individuals, resulting in higher ratios of TMAO/carnitine [0.12 (0.07-0.20) vs. 0.08 (0.05-0.11), P = 0.02] and TMAO/betaine [0.11 (0.07-0.17) vs. 0.08 (0.05-0.13), P 0.02]. In contrast to recent studies in HIV-uninfected populations, the present study found no evidence of TMAO-induced platelet hyperreactivity in HIV infected individuals. Microbial translocation and monocyte activation may affect TMAO levels in untreated individuals. Furthermore, the elevated ratios of TMAO/betaine and TMAO/carnitine in cART-treated individuals could possibly suggest a role of cART in TMAO metabolism.