Piezo1 Upregulation in Monocyte-Derived Macrophages Impairs Post-Myocardial Infarction Cardiac Repair via Defective Efferocytosis and Enhanced Ferroptosis

Abstract

The regulation of macrophage function, particularly that of monocyte-derived macrophages (MoMs), by mechanical forces during myocardial infarction (MI) remains poorly understood. Consistently upregulated Piezo1 expression in cardiac macrophages and MoMs post-MI is found. Elevated Piezo1 expression in MoMs directly contributes to increased Piezo1 levels in cardiac macrophages. Myeloid cell-specific Piezo1-deficient mice (Piezo1^Lyz2) exhibit significant improvements in ventricular function/remodeling after MI, accompanied by decreased apoptotic cardiomyocytes and decreased inflammation, increased numbers of macrophages, and increased border zone efferocytosis. In vitro, Piezo1 activation by Yoda1 increased oxygen-glucose deprivation (OGD)-induced ferroptosis and impaired MoM efferocytosis. Conversely, Piezo1 deficiency in MoMs decreases ferroptosis and increases efferocytosis. SLC7A11 is shown to mediate Piezo1-induced defective efferocytosis in MoMs. Piezo1 activation aggravated OGD-induced macrophage ferroptosis via Ca²⁺ influx followed by SLC15A3 upregulation. Piezo1 upregulated SLC7A11 in macrophages via a Ca²⁺/ATF4-dependent pathway. MoM-specific SLC7A11 knockdown significantly increases efferocytosis, reduces cardiomyocyte apoptosis and inflammation, and ameliorates post-MI left ventricular remodeling and function. In conclusion, early Piezo1 activation in MoMs is identified during MI, which governs the fate and function of recruited macrophages. These data establish an ischemic heart-bone marrow functional network and provide a novel therapeutic strategy in which MoM Piezo1 is targeted for post-MI heart repair.

Keywords: Piezo1; SLC15A3; SLC7A11; macrophages; myocardial infarction.

Figure 1

Piezo1 expression in cardiac macrophages increased after MI. A–D) scRNA‐seq analysis of the levels of four mechanosensitive molecules in macrophages from the heart tissues of sham and MI/R‐operated mice (n = 22903 cells for sham, n = 5042 cells for MI/R 3D). E) Representative images of immunostaining for cardiac troponin T (cTNT), Piezo1, F4/80 and nuclei (DAPI) in murine hearts subjected to the sham or MI procedure. The images of the MI groups were selected from the infarct area. F) Quantification of Piezo1⁺F4/80⁺ cells in (E) (n = 6 mice per group). G) Schematic diagram of the experimental procedure performed on Piezo1^GFP mice. H) Representative images of F4/80 (red), GFP and nuclei (DAPI, blue) in Piezo1^GFP murine hearts. I) Quantification of GFP⁺F4/80⁺ cells in (H) (n = 6 mice per group). The data in (A–D) were analyzed using an unpaired 2‐tailed Mann‒Whitney U test. The data in F were analyzed using one‐way ANOVA, followed by the Bonferroni post hoc correction. The data in (I) were analyzed using unpaired Student's t test.

Figure 2

Upregulation of Piezo1 in monocyte‐derived macrophages contributed to increased Piezo1 expression in cardiac macrophages following MI. A) Representative fluorescence lifetime images of freshly isolated bone marrow cells and cultured BMDMs from sham and MI‐operated mice. B,C) The mean lifetime of Flipper‐TR (n = 6 mice per group, 4 measurements per animal). D) Quantitative PCR analysis of Piezo1 and Piezo2 expression in BMDMs (n = 5 mice per group). E) Representative Western blots and quantification of Piezo1 protein levels in BMDMs (n = 5 mice per group). F) Expression of Piezo1 mRNA in blood mononuclear cells (n = 8 mice per group). G) Piezo1 mRNA expression in whole blood from patients with acute MI and control patients (n = 10 patients per group). H) Schematic diagram showing the experimental design. I–N) Representative fluorescence‐activated cell sorting (FACS) plot of the GFP⁺ and F4/80⁺ subsets and statistics of their frequencies in BMDMs I–K) and blood mononuclear cells L–N) from Piezo1^GFP mice (n = 6 mice per group). O) Schematic diagram showing the strategy for the bone marrow transplantation (BMT) experiment. P,Q) Representative images and statistics of GFP⁺F4/80⁺ cells in heart tissues from mice that underwent the Sham or MI procedure after BMT (n = 5 mice per group). The data in (F,G,J,K,M and N) were analyzed using unpaired Student's t test. Other variables were analyzed using one‐way ANOVA, followed by the Bonferroni post hoc correction.

Figure 3

Myeloid deletion of Piezo1 ameliorated cardiac dysfunction and cardiomyocyte apoptosis after MI. A) Representative long‐axis M‐mode echocardiographic images 4 weeks after MI. B–D) The LVEF, LVEDD, and LVESD were evaluated by long‐axis M‐mode echocardiography (n = 7, 8, 24, and 25 at 1 D; n = 7, 8, 20, and 23 at 2 W; and n = 7, 8, 19, and 22 at 4 W). Orange p, compared with the Piezo1^fl/fl+sham group; green p, compared with the Piezo1^fl/fl+MI group. E,F) Representative images of Masson's trichrome staining and quantification of the fibrotic area of heart tissue 4 weeks after MI (n = 6). G–I) Expression of Col1a, ANP and BNP mRNAs in heart tissues 4 weeks after MI (n = 6–13 mice per group). J–M) Representative Western blots and quantification of Piezo1, cleaved caspase‐3 and caspase‐3 protein levels from Piezo1^fl/fl and Piezo1^Lyz2 mice at 7 days after MI (n = 6 mice per group). N–O) Representative images and quantification of TUNEL‐positive cardiomyocytes (red, indicated by the white arrows) in the border area at 1 day, 3 days, and 7 days after MI (n = 6 mice per group). The data in (B–D) were analyzed using a Mixed‐effects model by Bonferroni's multiple comparisons test. The data in O were analyzed by unpaired Student's t test. Other data were analyzed via two‐way ANOVA, followed by the Bonferroni post hoc correction.

Figure 4

Myeloid deletion of Piezo1 increased the number and potentiated the efferocytotic activity of cardiac macrophages. A–I) Flow cytometry analysis of F4/80⁺CD11b⁺ cells in the bone marrow A–C) and blood D–F) and cardiomyocyte‐depleted cardiac cells G–I) from mice in the Piezo1^fl/fl+MI and Piezo1^Lyz2+MI groups on Day 7 (n = 6 mice per group). The beads were used to calculate the number of cells. J) Immunofluorescence staining showing the colocalization of F4/80⁺ macrophages (green) with cTNT⁺ cardiomyocytes (red) in the peri‐infarct area 1 week after MI. K) Analysis of the internalization of cardiomyocyte‐derived proteins in macrophages. The macrophages that stained positive for cardiomyocyte cTNT were scored as having internalized cardiomyocyte‐derived proteins (n = 6 mice per group). L–O) Expression of the TNF‐α, IL‐1β, IL‐6 and IL‐4 mRNAs in heart tissues 1 week after MI (n = 8 mice per group). P) Fluorescence images revealing the uptake of PKH67‐labeled apoptotic Jurkat cells by BMDMs from Piezo1^fl/fl+MI and Piezo1^Lyz2+MI mice. Q) Confocal microscopy images and quantification of BMDMs (F4/80, red) that engulfed PKH67‐labeled apoptotic Jurkat cells (green) (n = 6 mice per group). R) The uptake of PKH67‐labeled apoptotic Jurkat cells by BMDMs was analyzed via flow cytometry (n = 6 mice per group). The data in (L–O) were analyzed by two‐way ANOVA, followed by the Bonferroni post hoc correction. Other variables were analyzed via unpaired Student's t test.

Figure 5

Piezo1 induced defective efferocytosis in MoMs via SLC7A11 upregulation. A) Cluster analysis of the differentially expressed genes (|log₂Fold Change| > 1, p < 0.05) identified by RNA‐seq of BMDMs treated with DMSO or Yoda1 (5 µm) for 12 h) B. GO analysis revealed the enriched biological processes (BPs). C) The 17 genes upregulated according to the GO_BP analysis, including genes with read counts >1000, were analyzed by quantitative PCR in BMDMs from Piezo1^fl/fl+MI and Piezo1^Lyz2+MI mice (n = 6 mice per group). D. Quantitative PCR analysis of the 17 genes downregulated according to the GO_BP analysis, including genes with read counts >1000 (n = 6 mice per group). E) Representative Western blots and quantification showing SLC7A11 and SLC15A3 protein levels in BMDMs treated with DMSO or Yoda1 (5 µm) for 12 h (n = 4). F,G) Representative Western blots and quantification of SLC7A11, SLC15A3 and Piezo1 protein levels in BMDMs from Piezo1^Lyz2+MI and Piezo1^fl/fl+MI mice (n = 6 mice per group). H) Confocal microscopy images and quantification of BMDMs (F4/80, red) that engulfed PKH67‐labeled apoptotic Jurkat cells (green) (n = 4). The data in (C, D, E and F) were analyzed by unpaired Student's t test. Other data were analyzed via one‐way ANOVA, followed by the Bonferroni post hoc correction.

Figure 6

Piezo1 increased SLC7A11 in a Ca²⁺/ATF4‐dependent manner and increased SLC15A3 in a Ca²⁺‐dependent manner in macrophages. A) Heatmap of the genes of interest (the same as the 17 genes that were upregulated and the 17 genes that were downregulated by Yoda1 shown in Figure 5) identified by a new RNA‐seq analysis of Piezo1^fl/fl and Piezo1^Lyz2 BMDMs, which were treated with DMSO or Yoda1 (5 µm) for 12 h. B) Levels of Atf4, Atf3, Nrf2 and P53 mRNAs in BMDMs after treatment with DMSO or Yoda1 (n = 6). C,D) Representative Western blots and quantification of the protein levels of ATF4, ATF3, NRF2 and P53 in BMDMs (n = 3). E) Representative images of immunofluorescence staining showing the ATF4 level and location in BMDMs treated with DMSO or Yoda1. F) Protein expression of ATF4 in the cytoplasm and nucleus of BMDMs treated with DMSO or Yoda1. G–I) Representative Western blots and quantification of ATF4 and SLC7A11 protein levels in BMDMs after control or Atf4 siRNA transfection and DMSO or Yoda1 treatment (n = 3). J) Flow cytometry analysis of intracellular Ca²⁺ levels by Fluo‐4 staining in BMDMs from Piezo1^fl/fl and Piezo1^Lyz2 mice treated with DMSO or Yoda1 for 30 min (n = 5 mice per group). K) Fluo‐4 fluorescence signals in BMDMs pretreated with 5 µM BAPTA‐AM for 24 h followed by DMSO or 5 µm Yoda1 for 30 min were analyzed via flow cytometry (n = 3). L–O) Representative Western blots and quantification of the protein levels of ATF4, SLC7A11 and SLC15A3 in BMDMs (n = 3). The data in (B and D) were analyzed by unpaired Student's t test. Other data were analyzed by one‐way ANOVA, followed by the Bonferroni post hoc correction.

Figure 7

MoM‐specific SLC7A11 knockdown restored macrophage efferocytosis and ameliorated cardiac dysfunction after MI. A) Schematic diagram of the experimental design for MoM‐specific SLC7A11‐knockdown mice. B) Relative SLC7A11 mRNA levels in BMDMs from AAV2‐F4/80‐SLC7A11‐treated and AAV2‐F4/80 empty vector‐treated mice (n = 8 mice per group). C) Immunofluorescence staining showing the colocalization of F4/80⁺ macrophages (green) with cTNT⁺ cardiomyocytes (red) in the peri‐infarct area 1 week after MI (n = 6 mice per group). D) Representative images of TUNEL‐positive cardiomyocytes (red, white arrows) in the peri‐infarct area 1 week after MI (n = 6 mice per group). E–H) Expression of the TNF‐α, IL‐1β, IL‐6 and IL‐4 mRNAs in heart tissues 1 week after MI (n = 5–7 mice per group). I) Representative long‐axis M‐mode echocardiographic images 4 weeks after MI. J–L) The LVEF, LVEDD, and LVESD were evaluated via long‐axis M‐mode echocardiography (n = 7, 13, 12, and 12 at 1 D; n = 7, 9, 8, and 9 at 2 W; and n = 7, 9, 8, and 9 at 4 W). Blue p, compared with the sham group; green p, compared with the MI+MoM‐SLC7A11^WT group. M,N) Representative images of Masson's trichrome staining and quantification of the fibrotic area of heart tissue 4 weeks after MI (n = 6 mice per group). O–Q) The levels of the ANP, BNP and Col1a mRNAs in heart tissues 4 weeks after MI (n = 5–6 mice per group). R) Schematic illustration of the Piezo1‐mediated regulation of macrophage ferroptosis and efferocytosis in post‐MI cardiac remodeling. MI results in increased mechanical stress in the bone marrow, which increases membrane tension and Piezo1 expression in bone marrow monocyte‐derived macrophages (MoMs). After being mobilized to the blood and recruited into the ischemic heart, macrophages with high Piezo1 expression exhibit increased Ca²⁺ influx, which increases SLC15A3 expression and nuclear ATF4 expression. ATF4 transcriptionally increases SLC7A11 expression. SLC15A3 exacerbates macrophage ferroptosis, whereas SLC7A11 inhibits macrophage efferocytosis. Dysfunctional MoMs in the heart delay the clearance of dead cells and the resolution of inflammation, leading to the aggravation of post‐MI cardiac remodeling and heart failure. The data in B, C, D and N were analyzed by unpaired Student's t test. The data in (J–L) were analyzed using a Mixed‐effects model by Bonferroni's multiple comparisons test. Other variables were analyzed via one‐way ANOVA, followed by the Bonferroni post hoc correction.