Pulmonary primary oxysterol and bile acid synthesis as a predictor of outcomes in pulmonary arterial hypertension

Figures

• 

  Fig. 1.. Overall study design.

• 

  Fig. 2.. Metabolites associated with mortality in PAH discovery and validation cohorts. (A) Volcano plot of metabolites associated with mortality in discovery and validation cohorts. Data are derived from multivariable-adjusted Cox regression models. Red dots indicate deleterious metabolites significant in both discovery and validation cohorts; pink dots indicate deleterious metabolites significant only in the discovery cohort; dark blue dots indicate protective metabolites in both discovery and validation cohorts; light blue dots indicate protective metabolites significant only in discovery cohort; grey dots indicate all metabolites measured in discovery and validation cohorts. (B) Chemical networking of significant metabolites in both discovery and validation cohorts. Tandem mass spectra for the chromatographic features were clustered against tandem mass spectra for commercial standards with the resulting correlations displayed as a nested molecular network. Here, red nodes indicate tandem mass spectra for commercial standards, blue nodes indicate tandem mass spectra from glucuronidated compounds and grey nodes indicate tandem mass spectra from the 356 unknown compounds. (C) Manhattan…

• 

  Fig. 3.. Primary bile acid metabolites associate with PAH outcomes and physiology. (A) Heatmap of primary bile acid and oxysterol metabolites in idiopathic PAH patients (N=45) and control plasma (N=44). Each square indicates relative abundance of metabolite value in PAH versus control. Data originated from multivariable logistic regression analysis. Asterisks (*) indicate significance with P<0.05. (B) Forest plots of the top 20 glucuronide metabolites and their association to 6-minute walk distance (meters), WHO functional class, and pulmonary vascular resistance (Wood units) in three independent cohorts of PAH (Cohort 1, N=1913; Cohort 3, N=102; Cohort 4, N=122). Data are derived from multivariable regression models adjusting for age, sex, and body mass index. All data are presented as odds ratio of clinical/hemodynamic metric per 1-SD increase in metabolite concentration.

• 

  Fig. 4.. Elevation of primary oxysterol and bile acid metabolites in preclinical models of PAH. (A) Heatmap of oxysterol and primary bile acid metabolites in plasma samples of vehicle control, SU5146/hypoxic (SU/Hyp), or monocrotaline (MCT) PAH rat models. Each square indicates relative abundance of metabolite value. Data originated from student t-test. *P<0.05. (B to G) Significantly elevated oxysterols and primary bile acid metabolites in plasma samples of SU/Hyp or MCT rats compared to vehicle controls. (H to J) Significantly elevated oxysterols and primary bile acid metabolites in lung tissue of SU/Hyp or MCT rats compared to vehicle controls. *P<0.05; **P<0.01; ***P<0.001; ns, not significant with P>0.05. (K and L) Significantly elevated oxysterols and primary bile acid metabolites in pulmonary artery endothelial cells. PAEC were transfected with siRNA targeting BMPR2 (siBMPR2) and compared with scrambled siRNA (siNC, n=3). Data are presented as mean±SEM. Samples were tested in triplicate. Data were analyzed with an unpaired t test. *P<0.05

• 

  Fig. 5.. A poly-metabolite risk score predicts PAH disease severity and mortality. (A) Receiver-operating characteristic curves display the performance of models predicting mortality: the clinical model alone is represented by the black curve, the 13 PPOBAS poly-metabolite model by the blue curve, and 51 PPOBAS poly-metabolite model by the red curve. (B) Kaplan-Meier analyses display survival over time in patients with PAH within the discovery (Cohort 1) and validation (Cohort 2) cohorts, classified according to the PAH poly-metabolite risk score (PAH-PMRS). (C) Difference in overall survival between PAH-PMRS groups normalized to low-risk group. T indicates tertile with T1 representing lower tertile; T2, middle tertile; and T3, top tertile. P values originated from log rank test. ***P<0.001. (D) Reclassification of PAH clinical risk score (based on the European Society of Cardiology/European Respiratory Society [ESC/ERS] PAH risk calculator) using the PAH poly-metabolite risk score (PAH-PMRS). (E) Reclassification of PAH clinical risk score (European Society of Cardiology/European Respiratory Society [ESC/ERS]) in patients who died (N=520) in discovery (Cohort 1) and validation (Cohort 2)…