Piezo1-Mediated Mechanotransduction Contributes to Disturbed Flow-Induced Atherosclerotic Endothelial Inflammation

Abstract

Background: Disturbed flow generates oscillatory shear stress (OSS), which in turn leads to endothelial inflammation and atherosclerosis. Piezo1, a biomechanical force sensor, plays a crucial role in the cardiovascular system. However, the specific role of Piezo1 in atherosclerosis remains to be fully elucidated.

Methods and results: We detected the expression of Piezo1 in atherosclerotic mice and endothelial cells from regions with disturbed blood flow. The pharmacological inhibitor Piezo1 inhibitor (GsMTx4) was used to evaluate the impact of Piezo1 on plaque progression and endothelial inflammation. We examined Piezo1's direct response to OSS in vitro and its effects on endothelial inflammation. Furthermore, mechanistic studies were conducted to explore the potential molecular cascade through which Piezo1 mediates endothelial inflammation in response to OSS. Our findings revealed the upregulation of Piezo1 in apoE-/- (apolipoprotein E) atherosclerotic mice, which is associated with disturbed flow. Treatment with GsMTx4 not only delayed plaque progression but also mitigated endothelial inflammation in both chronic and disturbed flow-induced atherosclerosis. Piezo1 was shown to facilitate calcium ions (Ca²⁺) influx in response to OSS, thereby activating endothelial inflammation. This inflammatory response was attenuated in the absence of Piezo1. Additionally, we identified that under OSS, Piezo1 activates the Ca²⁺/CaM/CaMKII (calmodulin/calmodulin-dependent protein kinases Ⅱ) pathways, which subsequently stimulate downstream kinases FAK (focal adhesion kinase) and Src. This leads to the activation of the OSS-sensitive YAP (yes-associated protein), ultimately triggering endothelial inflammation.

Conclusions: Our study highlights the key role of Piezo1 in atherosclerotic endothelial inflammation, proposing the Piezo1-Ca²⁺/CaM/CaMKII-FAK/Src-YAP axis as a previously unknown endothelial mechanotransduction pathway. Piezo1 is expected to become a potential therapeutic target for atherosclerosis and cardiovascular diseases.

Keywords: Piezo1; atherosclerosis; endothelial cells; inflammation; shear stress.

Figure 1. Piezo1 is highly expressed in atherosclerotic plaque and disturbed flow‐exposed endothelial regions in apoE−/− mice.

A and B, Detection and quantification the protein expression level of Piezo1 in the aorta of apoE−/− mice from the CON and AS groups. C and D, Immunofluorescence images and analysis showed the expression and localization of Piezo1 in mice aortas from the CON and AS groups. Scale bar = 20 μm; n=3 mice per group. E, DF and LF patterns in atherosclerosis. F, Representative immunofluorescence images for Piezo1 in the outer and inner aortic arch. Scale bar = 20 μm. G, The anatomical diagram of partial carotid artery ligation. Ligation was performed at the left side, and the right side was the sham operation control. H and I, Immunofluorescence images and analysis for Piezo1 in the PL and non‐PL carotid arteries. Scale bar = 20 μm; n=3 mice per group. All data were statistically analyzed with unpaired 2‐tailed Student t test. ApoE indicates apolipoprotein E; AA, aortic arch; AS, atherosclerosis groups; CD31, platelet endothelial cell adhesion molecule 1; CON, control group; DF, disturbed flow; L‐CCA, left common carotid artery; L‐ECA, left external carotid artery; LF, laminar flow; L‐ICA, left internal carotid artery; N, 6 mice per group; PL, partial carotid artery ligation, and R‐CCA, right common carotid artery.

Figure 2. Inhibition of Piezo1 delays long‐term and disturbed flow‐induced atherosclerotic plaque progression and endothelial inflammation.

A, En face Oil Red O staining of the whole aorta. B, Analysis and quantification the percentage of plaque area to total area; n=6 mice per group. C and D, Typical images of HE staining of the aortic root and quantification of the plaque area; n=6 mice per group. Scale bar = 200 μm. E and F, All mice were raised on a Western diet for 4 weeks. HE staining of carotid arteries and quantification of plaque area; n=6 mice per group. Scale bar = 200 μm. G through I, Western blot analysis and quantification for VCAM‐1 and ICAM‐1 in mice aortas; n=6 mice per group. J and K, ApoE−/− mice were injected with GsMTx4 or an equivalent dose of saline and maintained on a high‐fat diet for 12 weeks. L and M, PL‐apoE−/− mice were injected with GsMTx4 or an equivalent dose of saline and maintained on a high‐fat diet for 4 weeks. J and K, Immunofluorescence staining was conducted to evaluate the VCAM‐1 expression; n=3 mice per group. Scale bar = 20 μm. B, D, H, I, and K, Unpaired 2‐tailed Student t test. F and M, One‐way ANOVA with Tukey post hoc tests. ApoE−/− indicates apolipoprotein E; AS, atherosclerosis (Western diet fed) + saline (0.1 mg/kg per day) injection; AS+GsMTx4, atherosclerosis (Western diet fed) + GsMTx4 (0.1 mg/kg per day) injection; CD31, platelet endothelial cell adhesion molecule 1; GsM, GsMTx4; GsMTx4, Piezo1 inhibitor; HE, hematoxylin–eosin; ICAM‐1, intercellular adhesion molecule 1; PL, partial carotid ligation; PL + GsMTx4, partial carotid artery ligation+GsMTx4 (0.1 mg/kg per day) injection; and VCAM‐1, vascular cell adhesion molecule 1.

Figure 3. Piezo1 promotes endothelial inflammatory activation in response to oscillatory shear stress.

A, Typical immunofluorescence staining image showing Piezo1 in HUVECs exposed to OSS (0.5 ± 4 dyn/cm²) or ST for 1 hour. Scale bar = 10 μm. B and C, HUVECs treated with Yoda1 (5 μm), GsMTx4 (5 μm), ST, and LSS (12 ± 4 dyn/cm² or 0.5 ± 4 dyn/cm²) for 8 hours. Intracellular Ca²⁺ imaging and quantitative analysis. Scale bar = 50 μm. D and E, Imaging and quantification the Ca²⁺ fluorescence in sh‐NC and sh‐Piezo1 HUVECs after ST or OSS treatment for 8 hours. Scale bar = 50 μm. F and G, HUVECs from sh‐NC and sh‐Piezo1 groups were treated with ST or OSS for 8 hours, detection and quantification for VCAM‐1 and ICAM‐1. C, One‐way ANOVA with Tukey post hoc tests. E and G, Two‐way ANOVA with Bonferroni multiple comparison post hoc test. Pairwise comparisons were performed among the 4 groups. All experiments were independently repeated 3 times. Ca2+: calcium ion; GsM, GsMTx4; GsMTx4, Piezo1 inhibitor; HUVECs indicates human umbilical vein endothelial cells; ICAM‐1, intercellular adhesion molecule 1; LSS, laminar oscillatory shear stress; OSS, oscillatory shear stress; sh‐NC, lentivirus infection negative control; sh‐Piezo1, lentivirus target‐Piezo1 silencing; ST, static state; Yoda1: Piezo1 agonist; and VCAM‐1, vascular cell adhesion molecule 1.

Figure 4. Piezo1 promotes endothelial inflammation through YAP activation.

A and B, HUVECs were subjected to OSS for 0, 1, 2, 4, 6, 8, and 16 hours, respectively. Immunoblot analysis and quantification p‐YAP (S127) and t‐YAP. C and D, HUVECs with sh‐NC and sh‐Piezo1 were treated with ST or OSS for 8 hours. Immunofluorescence staining images displayed and analyzed for the nuclear accumulation of YAP. Scale bar = 20 μm. E through G, HUVECs (sh‐NC and sh‐Piezo1) exposed to ST or OSS for 8 hours. Immunoblot analysis and quantification for p‐YAP (S127)/t‐YAP. H and I, Immunofluorescence staining analysis and quantification for YAP in apoE−/− mice carotid arteries; n=3 mice per group. Scale bar = 20 μm. J through L, Western blot analysis demonstrated that silencing Piezo1 reversed the upregulation of VCAM‐1 and ICAM‐1 induced by YAP overexpression. B and I, One‐way ANOVA with Tukey post hoc tests. D, F, G, K, and L, Two‐way ANOVA with Bonferroni multiple comparison post hoc test. Pairwise comparisons among the 4 groups. All experiments were independently repeated 3 times. ApoE−/− indicates apolipoprotein E; HUVECs, human umbilical vein endothelial cells; CD31, platelet endothelial cell adhesion molecule 1; GsM, GsMTx4; GsMTx4, Piezo1 inhibitorPLA:partial left carotid artery ligation; ICAM‐1, intercellular adhesion molecule 1; nuc., nuclear; OE‐YAP, YAP overexpression; OSS, oscillatory shear stress; p‐YAP (S127), phosphorylation of YAP at Ser127; sh‐NC, lentivirus infection negative control; sh‐Piezo1, lentivirus target‐Piezo1 silencing; ST, static state; t‐YAP, total YAP; VCAM‐1, vascular cell adhesion molecule 1; and YAP, yes‐associated protein.

Figure 5. Piezo1 activates YAP through kinase FAK/Src.

A through F, HUVECs of sh‐NC and sh‐Piezo1 were exposed to OSS or Yoda1 (5 μm) for 8 hours. Western blot (A through C) and immunofluorescence staining (D through F) analyzed and quantified the relative protein expression of p‐FAK (Y397) and p‐Src (Y419). Scale bar = 50 μm. G, Immunofluorescence staining examination of the spatial localization of FAK and Src in Piezo1‐silenced (sh‐Piezo1) and Piezo1‐normal (sh‐NC) HUVECs. Scale bars are shown in (G); n=4 independently repeated experiments. H, Coimmunoprecipitation analysis for the interaction of FAK and Src in Piezo1‐silenced (sh‐Piezo1) and Piezo1‐normal (sh‐NC) HUVECs. I through K, HUVECs were preincubated with Yoda1 (5 μm) followed by coincubation with PP2 (5 μm) or PF‐573228 (5 μm) for 24 hours. Western blot analyzed and quantified the relative expression of VCAM‐1 and p‐YAP (S127). All data were statistically analyzed with 1‐way ANOVA with Tukey post hoc tests. All experiments were independently repeated 3 times unless otherwise noted. FAK indicates focal adhesion kinase; p‐YAP (S127), phosphorylated YAP (Serine 127); p‐YAP, phosphorylated YAP; PF‐573228, FAK inhibitor; PP2, Src inhibitor; Src, proto‐oncogene tyrosine‐protein kinase; SRCt‐YAP, total YAP; Yoda1, Piezo1 agonist; HUVECs, human umbilical vein endothelial cells; IgG, immunoglobulin G; IP, immunoprecipitation; OSS, oscillatory shear stress; p‐FAK (Y397), phosphorylated FAK at Tyr397; p‐Src (Y419), phosphorylated Src at Tyr419; sh‐NC, lentivirus infection negative control; sh‐Piezo1, lentivirus target‐Piezo1 silencing; ST, static state; t‐FAK, total FAK; t‐Src, total Src; VCAM‐1, vascular cell adhesion molecule 1; and YAP, yes‐associated protein.

Figure 6. Piezo1‐mediated YAP activation and endothelial inflammation is dependent on Ca²⁺/CaM/CaMKII‐FAK/Src axis.

A and B, HUVECs were preincubated with Yoda1 (5 μm) for 8 hours and followed by coincubation with STO‐609 (1, 5, 10 μM) or KN‐93 (1, 5, 10 μM) for 24 hours. Western blot analyzed and quantified the relative protein level of p‐YAP (S127). C through E, HUVECs were preincubated with Yoda1 (5 μm) for 8 hours and followed by coincubation, respectively, with BAPTA (10 μm), trifluoperazine (30 μm), and KN‐93 (10 μm) for 24 hours. Western blot analyzed and quantified the relative protein level of VCAM‐1 and p‐YAP (S127). F through H, HUVECs were incubated with BAPTA (10 μm), trifluoperazine (30 μm), or KN‐93 (10 μm) for 24 hours, respectively. Western blot analyzed and quantified the relative p‐FAK (Y397) and p‐Src (Tyr419). I and J, HUVECs treated with or without KN‐93 (10 μm) for 24 hours. I, Coimmunoprecipitation analysis for the interaction of FAK and Src. J, Immunofluorescence staining examination for the spatial localization of FAK and Src. Scale bars are shown in (J); n=4 independently repeated experiments. B, Two‐way ANOVA with Bonferroni multiple comparison post hoc test. D, E, G, and H, One‐way ANOVA with Tukey post hoc tests. All experiments were independently repeated 3 times unless otherwise noted. FAK indicates focal adhesion kinase; HUVECs, human umbilical vein endothelial cells; IP, immunoprecipitation; VCAM‐1, vascular cell adhesion molecule 1; and YAP, yes‐associated protein. Ca2+/CaM/CaMKs, calcium ions/calmodulin/calmodulin‐dependent protein kinases; STO‐609, CaMKK inhibitor; KN‐93, CaMKII inhibitor; p‐YAP (S127), phosphorylated YAP (Serine 127); Yoda1, Piezo1 agonist;BAPTA, Ca2+ chelator; p‐FAK Y397, phosphorylated FAK at Tyrosine 397; p‐Src(Y419), phosphorylated Src at Tyrosine 419; Src, proto‐oncogene tyrosine‐protein kinase SRC; t‐YAP, total YAP; t‐FAK, total FAK; p‐Src(Y419), phosphorylated Src at Tyrosine 419

Figure 7. Schematic illustration of Piezo1 promotes atherosclerotic endothelial inflammation in response to OSS.

The atherogenic DF/OSS activates endothelial Piezo1. The activation of Piezo1 allows Ca²⁺ influx, which subsequently activates the Ca²⁺/CaM/CaMKII signaling pathway and downstream kinases FAK/Src. This process contributes to YAP S127 dephosphorylation and ultimately triggers endothelial inflammation, promoting the progression of atherosclerosis. DF indicates disturbed flow; FAK, focal adhesion kinase; ICAM‐1, intercellular adhesion molecule 1; OSS, oscillatory shear stress; VCAM‐1, vascular cell adhesion molecule 1; and YAP, yes‐associated protein; Ca2+, calcium ions; Ca2+/CaM/CaMKs , calcium ions/calmodulin/calmodulin‐dependent protein kinases; Src, proto‐oncogene tyrosine‐protein kinase SRC; P, phosphorylation; YAP S127, YAP Serine12.