Decoding the Heart Failure Peptidome

Figures

• 

  Figure 1.. Triangular strategy and regulated peptides in heart failure (HF). A, Peptides from BIOSTAT-CHF (Biology Study to Tailored Treatment in Chronic HF) and PREVEND (Prevention of Renal and Vascular End-Stage Disease) cohorts were identified based on regulation, predicted peptide variant (PPV) scores, and association with outcome. B, Regulated peptides between patients with and without HF. P values (y axis), and log10 ratio (x axis), family-wise error rate (FWER) <0.01 cutoff (blue line). C, Distribution of peptides: log10 regulation change (x axis), logit PPV scores14 (y axis). D, Angiotensinogen peptides; blue if HF-No HF ratio log10>0.176. E, Integrin alpha-7 peptides in patients with HF. Log10 ratio, FWER, and PPV in box. Proteolytic activation of ITA7 transduces the contractile force between the extracellular matrix and the cytoskeleton. F, ITA7 structure (alpha-fold: AF-Q13683-F1). Green arrow subtilisin-like endoprotease cleavage site for integrin activation. G, Identified peptides from isoform Q13683-9 align with the canonical isoform (uniprot ID: Q13683). ACE indicates angiotensin-converting enzyme; ACE2, dipeptidyl/mono carboxypeptidase ACE; ADM, adrenomedullin; AGT,…

• 

  Figure 2.. Prognostic peptide biomarkers for clinical outcome. A, Ratio difference and Cox regression analysis of clinically relevant peptides for outcome. Left, Forest plot of selected regulated peptides between BIOSTAT-CHF (Biology Study to Tailored Treatment in Chronic Heart Failure) and PREVEND (Prevention of Renal and Vascular End-Stage Disease; log10 ratio heart failure [HF]/no-HF, color-coded blue if log10>0.176, false discovery rate [FDR] <0.001). Right, Cognate hazard ratios (HRs) with 95% CI for peptides associated with cumulative incidence of all-cause death or HF hospital admission in the BIOSTAT-CHF cohort (FDR<0.05 colored blue). One unit change in Log10 peptide abundance is associated with the value of the x axis in HR. B, Global Cox regression volcano plot with Log2 HR at x axis and Log10 P value at y axis. Light blue emphasizes significant regulated peptides, and gray is nonsignificant from Figure 1B. The line signifies the FDR<0.05 cutoff for Cox regression outcome. ADM indicates adrenomedullin; AGT, angiotensinogen; BGLAP, osteocalcin; CCK, cholecystokinin; GIP, gastric inhibitory polypeptide; ITA7, integrin alpha-7; KNG1, kininogen-1; NPR-C, atrial natriuretic receptor…

• 

  Figure 3.. Characterization of patients with heart failure. A, Hierarchical clustering and correlation to outcome. Pearson correlation plot based on all pairwise peptide intensities. Medical history of heart failure New York Heart Association (MHHF-NYHA) score and all-class mortality and heart failure hospitalization at 2 years (MH_2Y). B, Kaplan-Meier survival curves on major clusters by cumulative incidence on all-cause death, including 95% CI. P value for cluster 3 vs the other 2 clusters together, adjusted for sex and stratified by batch. Censored subjects are indicated with a cross and reported events are shown in the table. C, Cluster 3 patients have more unique peptides (n=2203) not found in cluster 1 and 2 and significantly lower predicted peptide variant (PPV) score (t test, P<10-5). x axis logit PPV score. PPV calculations on peptides exclusively found within each cluster. D, Differential expression analysis between cluster 3 patients and cluster 1 and 2 combined (equally weighted, adjusted for albumin). Log10 (P value) on y axis, and Log10 ratio difference on x axis. Peptides from emphasized protein marked in dark blue.