Platelets and inflammation—insights from platelet non-coding RNA content and release in the Bruneck study and the PACMAN-AMI trial

Figures

• 

  Figure 1. Novel and conventional platelet function measurements in the Bruneck study. (A) Schematic describing the measurements performed in all 338 participants of the sixth quinquennial examination of the community-based prospective Bruneck study. (B) Circular heatmap displaying RNA levels and PF4 protein levels in PPP and eight different platelet releasates sampled from all 338 Bruneck participants. (C and D) Representative box and whisker plots for hepatocyte-derived miR-122 (C) and platelet-derived miR-126 (D) in all 338 Bruneck participants, using Kruskal–Wallis tests for statistical comparisons. The middle line represents the median, the upper and lower box borders represent the IQR and the whiskers represent 1.5 times the IQR of log2-RQ RNA levels.

• 

  Figure 2. Ex vivo RNA and protein release is agonist-specific and inhibited by aspirin in the Bruneck study. (A) Spearman correlation between aggregation responses and releasate RNA and PF4 protein levels as fold changes (FC) of platelet releasate levels to PPP levels in all 338 Bruneck participants. The size and colour intensity of dots reflect correlation strength. Blue reflects inverse and red reflects positive correlations. Significant (P < 0.05) correlations are highlighted by a black circle around the dot. (B) Circular heatmap displaying RNA levels and PF4 protein levels as FC of platelet releasate levels to PPP levels, comparing participants not on aspirin (‘−’, n = 183) to participants on aspirin (‘+’, n = 155). (C and D) Representative box and whisker plots for hepatocyte-derived miR-122 (C) and platelet-derived miR-126 (D), comparing participants not on aspirin (n = 183) to participants on aspirin (n = 155) using Mann–Whitney U tests in each agonist group. The middle line represents the median, the upper and lower box borders represent the IQR and the whiskers represent 1.5 times the IQR of log2-FC RNA levels. (E) Hierarchically clustered heatmap displaying ex vivo RNA and PF4…

• 

  Figure 3. Platelet aggregation and release measurements reveal platelet exhaustion in inflammation in the Bruneck study. (A) Correlation between clinical variables and aggregation responses in participants not on aspirin (left panel, n = 183) and participants on aspirin (right panel, n = 155). Size and colour intensity of dots reflect correlation strength. Blue reflects inverse and red reflects positive correlations. Significant (P < 0.05) correlations are highlighted by a black circle around the dot. Correlations between continuous variables are based on Spearman correlation. Correlations between a continuous and a binary variable are based on Point-biserial correlation. (B) Circular heatmaps displaying Spearman correlations between CRP and releasate RNA and PF4 protein levels in non-aspirin users (left panel, n = 183) and aspirin users (right panel, n = 155). Correlations with P ≥ 0.05 are uncoloured.

• 

  Figure 4. Intra-platelet RNA levels and their association with aggregation responses in the Bruneck study. (A) Hierarchical cluster analysis on a Spearman correlation heatmap of intra-platelet RNAs measured in all 338 Bruneck participants. Cluster one is labelled in blue and highlights RNAs that cluster with known leucocyte markers such as PTPRC and S100A8/A9 mRNAs. Cluster two is labelled in red and highlights RNAs that cluster with known platelet markers such as PF4 and PPBP mRNAs. (B) Spearman correlations between intra-platelet RNA levels and platelet aggregation responses in participants not on aspirin (upper panel, n = 183) and participants on aspirin (lower panel, n = 155). Size and colour intensity of dots reflect correlation strength. Blue reflects inverse and red reflects positive correlations. Significant (P < 0.05) correlations are highlighted by a black circle around the dot. Intra-platelet RNA labels are coloured according to their clustering in panel (A), i.e. leucocyte-derived RNA cluster (cluster one, blue font) or platelet RNA cluster (cluster two, red font). RNAs in panel (B) are ranked from left to right from the most negative to the most positive average correlation…

• 

  Figure 5. Platelet-derived miRNAs and YRNAs are predominantly carried by proteins, while circRNAs, lncRNAs and mRNA are carried by vesicles in samples from healthy donors. (A) The upper panel displays a schematic of the ultracentrifugation experiment conducted on plasma-free platelet releasate. The middle panel shows the relative distribution of protein markers using dot blots on fractions of small EVs, large EVs and EV-poor supernatant generated by ultracentrifugation. The lower panel displays the relative distribution of RNAs in the same fractions. Ultracentrifugation was performed on four sample pools, with each pool containing samples from six healthy donors. (B) The upper panel displays a schematic of the size-exclusion chromatography (SEC) experiment conducted on plasma-free platelet releasate (n = 9). The middle panel displays the SEC chromatogram and shows how the six SEC fractions (F1-F6) were collected. The lower panel displays the relative distribution of protein markers in vesicle-rich early fractions and protein-rich late fractions using dot blots. (C) Distribution of RNAs in the SEC fractions described in panel (B). (D) The upper panel displays a schematic of the degradation…

• 

  Figure 6. Platelet-derived miRNAs and YRNAs are readily released, while circRNAs, lncRNAs and mRNA are preferentially retained in samples from healthy donors. (A) Schematic describing the design of the experiments conducted on paired plasma-free platelet releasate and platelet pellet samples, to assess preferential release and preferential retainment of platelet RNAs (n = 24). (B) Cq levels of RNAs measured in plasma-free platelet releasate (y-axis) and paired, not activated platelet pellets (x-axis, n = 24). (C) Delta-Cq values calculated by subtracting the Cq value of plasma-free platelet releasate from the Cq value of the platelet pellet from the same donor. RNAs are coloured from low delta-Cq (preferentially released, red font) to high delta-Cq (preferentially retained, turquoise font). Bar graphs display means with 95% CI and individual values (n = 24). (D) Visual summary of compartmentalization and release tendency of platelet-derived RNAs.

• 

  Figure 7. Platelet function measurements in the PACMAN-AMI trial. (A) Platelet aggregation was measured with the VerifyNow aspirin assay (n = 126 paired samples) and the VerifyNow P2Y12 assay (n = 131 paired samples). Forest plots show log2 fold changes with 95% confidence intervals of platelet aggregation measured as aspirin reaction units (ARU) and P2Y12 reaction units (PRU). (B) NcRNA levels were measured by RT-qPCR in 265 patients of the PACMAN-AMI trial at 24 h (n = 85), 4 weeks (n = 265) and 52 weeks (n = 265) post-hospital admission and percutaneous coronary intervention due to AMI. MiR-208b-3p and miR-133a are myocyte-derived, miR-150 and RNY5 are linked to inflammation, all other ncRNAs are platelet-derived. Forest plots show log2 fold changes with 95% confidence intervals of ncRNAs between week 4 and 24 h (n = 85 paired samples, left panel) and between week 52 and week 4 (n = 265 paired samples, right panel). Wilcoxon signed-rank test was applied in panel (A) and (B) to determine statistical significance. P-values are colour-coded.