Piezo1-mediated mechanosensation in bone marrow macrophages promotes vascular niche regeneration after irradiation injury

Abstract

Background: Irradiation disrupts the vascular niche where hematopoietic stem cells (HSCs) reside, causing delayed hematopoietic reconstruction. The subsequent recovery of sinusoidal vessels is key to vascular niche regeneration and a prerequisite for hematopoietic reconstruction. We hypothesize that resident bone marrow macrophages (BM-Mφs) are responsible for repairing the HSC niche upon irradiation injury. Methods: We examined the survival and activation of BM-Mφs in C57BL/6 mice upon total body irradiation. After BM-Mφ depletion via injected clodronate-containing liposomes and irradiation injury, hematopoietic reconstruction and sinusoidal vascular regeneration were assessed with immunofluorescence and flow cytometry. Then enzyme-linked immunosorbent assay (ELISA) and flow cytometry were performed to analyze the contribution of VEGF-A released by BM-Mφs to the vascular restructuring of the HSC niche. VEGF-A-mediated signal transduction was assessed with transcriptome sequencing, flow cytometry, and pharmacology (agonists and antagonists) to determine the molecular mechanisms of Piezo1-mediated responses to structural changes in the HSC niche. Results: The depletion of BM-Mφs aggravated the post-irradiation injury, delaying the recovery of sinusoidal endothelial cells and HSCs. A fraction of the BM-Mφ population persisted after irradiation, with residual BM-Mφ exhibiting an activated M2-like phenotype. The expression of VEGF-A, which is essential for sinusoidal regeneration, was upregulated in BM-Mφs post-irradiation, especially CD206⁺ BM-Mφs. The expression of mechanosensory ion channel Piezo1, a response to mechanical environmental changes induced by bone marrow ablation, was upregulated in BM-Mφs, especially CD206⁺ BM-Mφs. Piezo1 upregulation was mediated by the effects of irradiation, the activation of Piezo1 itself, and the M2-like polarization induced by the phagocytosis of apoptotic cells. Piezo1 activation was associated with increased expression of VEGF-A and increased accumulation of NFATC1, NFATC2, and HIF-1α. The Piezo1-mediated upregulation in VEGF-A was suppressed by inhibiting the calcineurin/NFAT/HIF-1α signaling pathway. Conclusion: These findings reveal that BM-Mφs play a critical role in promoting vascular niche regeneration by sensing and responding to structural changes after irradiation injury, offering a potential target for therapeutic efforts to enhance hematopoietic reconstruction.

Keywords: Piezo1; hematopoietic reconstitution; irradiation; macrophages; sinusoidal regeneration.

Figure 1

BM-Mφs persist and are activated upon irradiation injury. (A) Number of bone marrow nucleated cells (BMNCs) per femur after 5-Gy irradiation (IR) (n = 3-8 mice, Tukey test). (B) White blood cell (WBC), red blood cell (RBC), and platelet counts were measured in peripheral blood samples after 5-Gy irradiation (n = 3-8 mice). (C) Representative flow cytometry analysis of bone marrow-resident macrophages (BM-Mφs) (CD11b⁺F4/80⁺) without irradiation (Non) vs. 3 days after irradiation (IR 3d). (D) Number of BM-Mφs per femur and (E) percentage of BM-Mφs in the BMNC population after 5-Gy irradiation (n = 3-8 mice, Tukey test). (F) Representative flow cytometry analysis of BM-Mφs activation and persistence of M1-like Mφs (CD11b⁺F4/80⁺CD206^-CD11c⁺), M1/M2-like Mφs (CD11b⁺F4/80⁺CD206⁺CD11c⁺), and M2-like Mφs (CD11b⁺F4/80⁺CD206⁺CD11c^-). (G) Percentage of polarized BM-Mφs among all BM-Mφs (n = 5 mice, t-test). Graphs show the mean ± SD of pooled data from at least two independent experiments. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; ns, not significant.

Figure 2

Depletion of residual BM-Mφs impedes hematopoietic regeneration. (A) Schematic of experimental design for Mφs depletion and irradiation treatment. (B) Representative flow cytometry analysis of BM-Mφs (CD11b⁺F4/80⁺) in mice treated with PBS- or clodronate-containing liposomes (PBS-lip or Clo-lip). (C) Percentage of BM-Mφs among BMNCs (left panel) and number of BM-Mφs per femur (right panel) at 3 days after PBS-lip or Clo-lip injection without irradiation (n = 5 mice, t-test). (D) Percentage of BM-Mφs among BMNCs (left panel) and number of BM-Mφs per femur (right panel) at different time points after 5-Gy irradiation, in mice pretreated with Clo-lip or PBS-lip (n = 6-13 mice, t-test). (E) Percentages of polarized BM-Mφs among all BM-Mφs. (n = 5 mice). *P < 0.05, t-test; # P < 0.05, Tukey test, compared to non-irradiation (Non) group with PBS-lip injection. (F) Representative flow cytometry analysis of hematopoietic stem/progenitor cells. (G-J) Number of LSKs (Lin^-Sca-1⁺c-Kit⁺) (G), HSCs (LSK CD48^-CD150⁺) (H) HPCs (LSK CD48⁺CD150^+/-) (I), and MPPs (LSK CD48^-CD150^-) (J) per femur at 3 days after PBS-lip or Clo-lip injection without irradiation (n = 5-6 mice, t-test). (K-N) Number of LSKs (K), HSCs (L), HPCs (M), and MPPs (N) per femur after 5-Gy irradiation, for mice injected with Clo-lip vs. PBS-lip (n = 4-7 mice, t-test). (O-P) Kaplan-Meier survival curve (n = 10 mice, Log-rank test) (O) and body weight (P) for mice with Mφs depletion and 7.5-Gy irradiation treatment. Graphs show mean ± SD. *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 3

Depletion of residual BM-Mφs impedes sinusoidal endothelial cell (SEC) regeneration. (A) In situ immunofluorescence showing bone marrow sinusoids (green, CD105) after 5-Gy irradiation. Nuclei were stained with DAPI (blue). (B) Immunofluorescence images of bone marrow sinusoids (green, CD105) at 3 days after 5-Gy irradiation in PBS controls and mice with Mφs depletion by Clo-lip. Scale bar, 50 μm. (C-H) Flow cytometry analysis to determine relative percentage of SECs (CD45^-Ter119^-CD31⁺CD105⁺) among BMNCs (C), number of SECs per femur (D), relative percentage of CD31⁺ ECs (CD45^-Ter119^-CD31⁺CD105^-) among BMNCs (E), number of CD31⁺ ECs per femur (F), relative percentage of CD105⁺ stromal cells (CD45^-Ter119^-CD31^-CD105⁺) among BMNCs (G) and number of CD105⁺ stromal cells per femur (H) at indicated times after 5-Gy irradiation and pretreatment with PBS- or Clo-lip injection (n = 3-6 mice). Data are shown as mean ± SD. *P < 0.05; **P < 0.01; ***P < 0.001 (t-test).

Figure 4

Residual BM-Mφs upregulate VEGF-A after irradiation. (A) Quantification of bone marrow VEGF-A levels by ELISA at indicated times after 5-Gy irradiation and treatment with PBS-lip or Mφs depletion by Clo-lip (n = 4-5 mice). *P < 0.05; **P < 0.01, t-test; # P < 0.05, compared to non-irradiation with PBS-lip injection, Tukey test; & P < 0.05, compared to non-irradiation with Clo-lip injection, Tukey test. (B) Representative flow cytometry analysis of VEGF-A expression in BM-Mφs from mice after 5-Gy irradiation or no irradiation. (C) Histogram showing mean fluorescence intensity (MFI) of VEGF-A in BM-Mφs (n = 5 mice, Tukey test). (D) Representative flow cytometry analysis of VEGF-A expression in CD206⁺ and CD206^- BM-Mφs from mice after irradiation or non-irradiation, with or without Mφs depletion by Clo-lip. (E) Histogram showing MFI of VEGF-A in CD206⁺ and CD206^- BM-Mφs (n = 5 mice). **P < 0.01; ***P < 0.001, ***P < 0.0001, t-test; # P < 0.05, compared to non-irradiation with PBS-lip injection, Tukey test; & P < 0.05, compared to non-irradiation with Clo-lip injection, Tukey test. Data are from one experiment and representative of three independent experiments. (F) RT-PCR analysis of VEGF-A mRNA levels in BMDMs 24 h after irradiation or non-irradiation in vitro (n = 3 independent experiments, t-test). (G) Quantification by ELISA of VEGF-A expression in BMDMs 24 h after irradiation or non-irradiation (n = 3 independent experiments, t-test). Data are shown as mean ± SD; ns, not significant.

Figure 5

Residual BM-Mφs respond to mechanical stretching. BM-Mφs were sorted from mice 3 days after 5-Gy irradiation (IR_3d) or non-irradiation (C) for RNA sequencing analysis. (A) Heat map showing differentially expressed genes (|Log₂(FC)| > 1, false discovery rate (FDR) < 0.05) in BM-Mφs from irradiation-treated mice, compared to controls. (B) Bubble diagram showing enriched gene ontology (GO) biological process terms for differentially expressed genes. (C) In situ immunofluorescence showing the morphology of BM-Mφs (green, F4/80) in frozen femur sections from irradiated mice. Nuclei were stained with DAPI (blue). Scale bar, 20 μm. (D) Nuclear area of F4/80⁺Mφs in the bone marrow. Values represent mean ± SD, n = 306, 311, 200, 260 cells, respectively, from three separate experiments. ****P < 0.0001 (Kruskal-Wallis test). (E) Frequency distribution of nuclear area. (F) Cell area of F4/80⁺Mφs in the bone marrow. Values represent mean ± SD, n = 155, 157, 131, 184 cells, respectively, from three separate experiments. ***P < 0.001; ****P < 0.0001 (Kruskal-Wallis test). (G) Frequency distribution of cell area.

Figure 6

Piezo1 activation mediates the upregulation of VEGF-A in BM-Mφs. (A) Expression analysis of known mammalian mechanosensory ion channels in BM-Mφs after 5-Gy irradiation or non-irradiation (n = 3 mice, Tukey test). FPKM: Fragments per kilobase of transcript per million mapped fragments. (B) Representative flow cytometry analysis of Piezo1 expression in BM-Mφs after 5-Gy irradiation or non-irradiation. (C) Histogram showing means fluorescence intensity (MFI) of Piezo1 in BM-Mφs (n = 5 mice, Tukey test). (D) Representative flow cytometry analysis of Piezo1 expression in CD206⁺ and CD206^- BM-Mφs from mice after irradiation or non-irradiation with Mφs depletion by Clo-lip or not. (E) Histogram showing MFI of Piezo1 in CD206⁺ and CD206^- BM-Mφs (n = 5 mice). *P < 0.05, **P < 0.01, ***P < 0.001, ***P < 0.0001, t-test; # P < 0.05, compared to non-irradiated group with PBS-lip injection, Tukey test; & P < 0.05, compared to non-irradiated group with Clo-lip injection, Tukey test. (F) RT-PCR analysis of VEGF-A mRNA levels in BMDMs 6 h after Yoda1 treatment in vitro (n = 3 independent experiments, Tukey test). (G) ELISA analysis of VEGF-A expression in BMDMs 24 h after Yoda1 treatment in vitro (n = 3 independent experiments, Tukey test). (H-I), RT-PCR analysis of Piezo1 mRNA levels in BMDMs 24 h after irradiation (H) or Yoda1 (I) treatment. (J) RT-PCR (left panel) and immunofluorescence analysis (right panel) of CD206 expression in BMDMs 24 h after phagocytosis of irradiation-induced apoptotic bone marrow cells (n = 3 independent experiments, t-test). Scale bar, 50 μm. (K) RT-PCR analysis of Piezo1 expression in BMDMs 24 h after phagocytosis of irradiation-induced apoptotic bone marrow cells (n = 3 independent experiments, t-test). Data are shown as mean ± SD. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; ns, not significant.

Figure 7

Activation of calcineurin/NFAT/HIF-1α signaling induces a Piezo1-mediated increase in VEGF-A expression in BMDMs. (A) RT-PCR analysis of HIF-1α mRNA levels in BMDMs 6 h after Yoda1 treatment (n = 3 independent experiments, Tukey test). (B) Western blot analysis of HIF-1α accumulation in BMDMs 24 h after Yoda1 treatment. Blots are representative of three independent experiments. (C) RT-PCR analysis of VEGF-A mRNA levels in BMDMs pretreated with echinomycin (Ecn) for 30 min before 6-h Yoda1 (5 μM) treatment (n = 3 independent experiments, Tukey test). (D) ELISA analysis of VEGF-A expression in BMDMs pretreated with Ecn for 30 min before 24-h Yoda1 (5 μM) treatment (n = 3 independent experiments, Tukey test). (E) Western blot analysis of NFATC1 and NFATC3 in BMDMs treated with Yoda1 for 24 h. (F) Western blot analysis of HIF-1α in BMDMs pretreated with CsA or FK506 for 30 min before 24-h Yoda1 (5 μM) treatment. Blots are representative of three independent experiments. (G) RT-PCR analysis of VEGF-A mRNA levels in BMDMs pretreated with CsA or FK506 for 30 min before 6-h Yoda1 (5 μM) treatment (n = 3 independent experiments, Tukey test). (H) ELISA analysis of VEGF-A expression in BMDMs pretreated with CsA or FK506 for 30 min before 24-h Yoda1 (5 μM) treatment (n = 3 independent experiments, Tukey test). (I) Western blot analysis of NFATC1 and NFATC3 in BMDMs pretreated with BAPTA-AM or GsMTx4 for 30 min before 24-h Yoda1 (5 μM) treatment. Blots are representative of three independent experiments. (J) RT-PCR analysis of VEGF-A mRNA levels in BMDMs pretreated with BAPTA-AM or GsMTx4 for 30 min before 6-h Yoda1 (5 μM) treatment (n = 3 independent experiments, Tukey test). (K) ELISA analysis of VEGF-A expression in BMDMs pretreated with BAPTA-AM or GsMTx4 for 30 min before 24-h Yoda1 (5 μM) treatment (n = 3 independent experiments, Tukey test). Data are shown as mean ± SD. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. (L) Schematic model showing that Piezo1-mediated mechanosensation in BM-Mφs promotes vascular niche regeneration after irradiation injury.