C. A. Cason1, K. Dolan5, G. Sharma3, M. Tao3, A. Longchamp3, R. Kulkarni2, L. Xiong2, E. B. Chang5, M. M. McDermott4, C. K. Ozaki3, K. J. Ho2 1Northwestern University,General Surgery,Chicago, IL, USA 2Northwestern University,Vascular Surgery,Chicago, IL, USA 3Brigham And Women’s Hospital,Vascular Surgery,Boston, MA, USA 4Northwestern University,Medicine,Chicago, IL, USA 5University Of Chicago,Medicine,Chicago, IL, USA
Introduction: Altered gut microbial composition and gut microbe-dependent metabolism of dietary components has been associated with atherogenesis. To expand our understanding of the role of gut microbe-dependent metabolites in atherosclerosis, we selected a panel of 5 metabolites that are known to be differentially produced in germ-free versus conventional mice. To test the hypothesis that specific gut microbe-derived metabolite signatures are associated with severe human atherosclerosis, we determined circulating plasma levels of these metabolites in cohorts of patients with surgical peripheral arterial disease (PAD) and non-PAD controls.
Methods: After informed consent, plasma and medical information were collected from patients undergoing carotid endarterectomy (N=48) and lower extremity major amputation (N=12) or revascularization (femoral endarterectomy or bypass surgery; N=39). Control non-PAD plasma specimens (N=22) were obtained from consented age- and gender-matched patients with normal lower extremity ankle-brachial indices. Patients with serum creatinine ≥ 2 mg/dL were excluded. Metabolites and internal standards were extracted in methanol, separated by reverse-phase HPLC, and measured by tandem mass spectrometry on a triple quadrupole instrument operating in multiple reaction monitoring mode. Normalized peak areas (NPA) were calculated using internal standards and converted to concentration using standard curves. Median plasma metabolite concentrations were compared between groups using Wilcoxon signed rank tests.
Results: There were statistically significant differences in 2 of the 5 gut microbe-dependent metabolites between the patients and controls (see Table). In PAD patients, tryptophan was significantly lower (P < 0.001), indole-3-propionic acid was significantly lower (P<.0001), and there was a trend toward a lower kynurenine (P=.15) than in controls. The kynurenine/tryptophan ratio was higher (P = .0005) in PAD patients than in controls.
Conclusion: Specific gut microbe-derived metabolite signatures associate with severe human atherosclerosis, including differences in tryptophan, kynurenine, and indole-3-propionic acid levels in severe atherosclerosis patients. Beyond potential roles as powerful quantitative biomarkers for atherosclerotic disease burden, these findings suggest mechanistic links between defined microbial metabolic pathways and atherogenesis. ?
