Transcriptional Profiling of Human Endothelial Cells Unveils PIEZO1 and Mechanosensitive Gene Regulation by Prooxidant and Inflammatory Inputs

Figures

• 

  Figure 1. High shear stress regulates PIEZO1 and mechanotransduction-related gene (MRG) levels in HUVEC. (A) Hierarchical clustering of HUVEC samples exposed to static, 15 dyn/cm2, and 30 dyn/cm2 in vitro shear stress. (B) Correlation graph for 15 dyn/cm2 and 30 dyn/cm2 exposed HUVEC. Black dots represent similar gene regulation patterns between groups. Red dots represent different gene regulation between groups. (C) Transcription factor enrichment in DEG in HUVEC exposed to 15 dyn/cm2 and 30 dyn/cm2. (D) Fold change in PIEZO1 transcript levels in HUVEC exposed to static, 15 dyn/cm2, and 30 dyn/cm2 in vitro shear stress (E) Average expression heat map for MRG in HUVEC exposed to different shear stress levels. Values are expressed as mean ± S.E.M., * p < 0.05, one-way ANOVA, FDR post hoc test.

• 

  Figure 2. Effect of silencing shear stress-responsive genes on PIEZO1 and MRG transcript levels in HUVECs. (A) Average expression heatmap for mechanotransduction-related genes in HUVEC treated with siKLF4. (B) PIEZO1 transcript level fold change in HUVEC treated with siKLF4. (C) Average expression heatmap for mechanotransduction-related genes in HUVEC treated with siYAP. (D) PIEZO1 transcript level fold change in HUVEC treated with siYAP. Values are expressed as mean ± S.E.M., * p < 0.05, Mann–Whitney U test.

• 

  Figure 3. TNF-α increases PIEZO1 and mechanosensing-related transcript levels in diverse EC. (A) Fold change heatmap for mechanotransduction-related genes in HUVECs treated with TNF-α, 25 ng/mL (1 h), and 10 and 50 ng/mL (24 h). (B) PIEZO1 transcript level fold changes in response to TNF-α treatments at different times and concentrations. (C) Time-lapse curves representing selected transcript level changes in response to 480 min of TNF-α (10 ng/mL) treatment. (D) Dots graph representing the correlation between PIEZO1 and RELA transcript levels during time-lapse shown in (C). Values are represented as fold changes related to the basal from one sample from each data point. Values are expressed as mean ± S.E.M., * p < 0.05, one-way ANOVA, FDR post hoc test.

• 

  Figure 4. Effects of TNF-α stimulation on MRG and PIEZO1 transcript levels in endothelial cells from different vascular origins. (A) Hierarchical clustering of diverse endothelial cells exposed to different TNF-α treatments; (B) t-SNE projection showing clustering between diverse endothelial cell types and conditions. Endothelial cells exposed to TNF-α are represented in red, orange, and pink colors. Endothelial cells under basal conditions (control) are represented in blue and light blue colors. (C) Graph bars show seven enriched pathways in response to TNF-alpha treatment in diverse endothelial cells. (D,E) Fold change of PIEZO1 transcript levels in response to TNF-α treatment (50 ng/mL) for 24 h in HUVEC and HUAEC, respectively. Values are expressed as mean ± S.E.M., * p < 0.05 to control, Mann–Whitney U test.

• 

  Figure 5. Interaction of PIEZO1/RELA and the effects of NF-kB inhibitors on TNF-α-induced PIEZO1 expression. (A) Predicted binding sites for NF-κB in the PIEZO1 promoter region zone (TSS: transcription start site; 1000 bp). (B) ChIP-seq diagram representing p65 interaction with PIEZO1 promoter region in control and treated cells. The overlay shows graphical differences between conditions. (C) ChIP-seq diagram representing Pol II interaction with PIEZO1 promoter in control and treated cells. The overlay shows no qualitative differences between conditions. (D) Fold change in PIEZO1 transcript levels in response to TNF-α treatment (10 ng/mL) for 24 h alone or with NOD, a p65 inhibitor. (E) Fold change in PIEZO1 transcript levels in response to TNF-α treatment (50 ng/mL) for 24 h alone or with CORM-3, a CO-donor, and p65 inhibitor. (F) Fold change in PIEZO1 transcript levels in response to TNF-α treatment (25 ng/mL) for 1 h alone or with JQ1, a BRD4 inhibitor. Values are expressed as mean ± S.E.M., * p < 0.05 to control, # p < 0.05 to TNF-α treated, one-way ANOVA, FDR post hoc test.

• 

  Figure 6. Summary of proposed PIEZO1 regulation by pro-inflammatory conditions in in vitro endothelial cells. TNF-α receptor activation by TNF-α led to NF-κB complex p65-p50 subunit activation and translocation to the nucleus after IkB-alpha ubiquitination. Cytoplasmic levels and activity of p65 are reduced by NOD and CORM-3, respectively. NOD also reduces nuclear p65 levels. BRD4, a transcriptional NF-κB enhancer, is inhibited by JQ1, but no effect in the regulation of PIEZO1 transcription was found. This diagram is simplified and does not consider every TNF-α signaling pathway (created in Biorender.com).