Mechanical force modulates macrophage proliferation via Piezo1‐AKT‐Cyclin D1 axis

Abstract

Orthodontic tooth movement (OTM) is induced by biomechanical stimuli and facilitated by periodontal tissue remodeling, where multiple immune cells participate in this progression. It has been demonstrated that macrophage is essential for mechanical force‐induced tissue remodeling. In this study, we first found that mechanical force significantly induced macrophage proliferation in human periodontal samples and murine OTM models. Yet, how macrophages perceive mechanical stimuli and thereby modulate their biological behaviors remain elusive. To illustrate the mechanisms of mechanical force‐induced macrophage proliferation, we subsequently identified Piezo1, a novel mechanosensory ion channel, to modulate macrophage response subjected to mechanical stimuli. Mechanical force upregulates Piezo1 expression in periodontal tissues and cultured bone‐marrow‐derived macrophages (BMDMs). Remarkably, suppressing Piezo1 with GsMTx4 retarded OTM through reduced macrophage proliferation. Moreover, knockdown of Piezo1 effectively inhibited mechanical force‐induced BMDMs proliferation. RNA sequencing was further performed to dissect the underlying mechanisms of Piezo1‐mediated mechanotransduction utilizing mechanical stretch system. We revealed that Piezo1‐activated AKT/GSK3β signaling was closely associated with macrophage proliferation upon mechanical stimuli. Importantly, Cyclin D1 (Ccnd1) was authenticated as a critical downstream factor of Piezo1 that facilitated proliferation by enhancing Rb phosphorylation. We generated genetically modified mice in which Ccnd1 could be deleted in macrophages in an inducible manner. Conditional ablation of Ccnd1 inhibited periodontal macrophage proliferation and therefore delayed OTM. Overall, our findings highlight that proliferation driven by mechanical force is a key process by which macrophages infiltrate in periodontal tissue during OTM, where Piezo1‐AKT‐Ccnd1 axis plays a pivotal role.

Abbreviations

BMDMs

bone‐marrow‐derived macrophages

Ccnd1

Cyclin D1

DAPI

4′,6‐diamidino‐2‐phenylindole

FC

fold change

FDR

false discovery rate

GEO

gene expression omnibus

GSEA

gene set enrichment analysis

KEGG

Kyoto encyclopedia of genes and genomes

Micro‐CT

micro‐computed tomography

MS

mechanical stretch

NC

negative control

OTM

orthodontic tooth movement

RNA‐seq

RNA‐sequencing

siRNA

small interfering RNA

α‐MEM

alpha‐minimum essential medium

1. INTRODUCTION

Malocclusion is a serious problem because of high incidence and their deleterious effects on both oral functions and appearance. ¹ Consequently, the demand for orthodontic treatment continues to increase every year. Orthodontic tooth movement (OTM) is the principle of orthodontic treatment promoted by mechanical force‐induced aseptic inflammatory response and periodontal tissue remodeling. During this process, multiple immune cells involved in the progression through modulating local inflammation, secretion of cytokines, and interaction with other cells. ² , ³ Due to the ability to rapidly adapt to the local microenvironment, macrophages play a vital role during bone modeling and remodeling by serving as progenitors of osteoclasts and effectors of mechanical force. ⁴ Previous studies provided accumulating evidence that macrophages modulated OTM in response to local inflammation microenvironment. ⁵ However, little is known about the response of macrophages toward mechanical stimuli during OTM.

Macrophages could sense the mechanical stimuli and respond through variety of processes such as cell division, cell proliferation, and differentiation. ⁶ , ⁷ A recent study constructed a single‐cell atlas of murine alveolar bone immune cells. It revealed a proliferative population of macrophage exclusively distributed in bone marrow tissue with biological functions enriched in cell cycle, DNA replication, and cell proliferation‐related pathways. ⁸ Most notably, macrophages manifest an increased proliferative phenotype when submit to mechanical strain compared to the steady state. ⁹ Hence, elucidating the potential mechanisms of how macrophages affect their proliferative behavior subjected to mechanical stimuli is increasing in importance.

The Piezo family was discovered as a novel class of mechanosensory ion channels with high affinity for calcium in 2011, ¹⁰ which was classified into two subtypes: Piezo1 and Piezo2. Piezo1 is mainly expressed by non‐sensory tissues and is critical to sense and transduce various mechanical stimuli. ¹¹ In contrast, Piezo2 is primarily expressed in sensory neurons, such as ganglia, enterochromaffin cells, and Merkel cell. ¹² , ¹³ Recent studies have identified that Piezo1 can function as mechanostat that directly senses mechanical stimuli to determine cell behaviors or disease progression. ³ For example, Piezo1 activity has been verified to affect macrophage polarization in response to IFN‐γ/LPS stimulation. Myeloid cells from Piezo1^ΔLysM mice showed decreased inflammatory meditator release. ¹⁴ In addition, Piezo1 can modulate fibroblasts proliferation ¹⁵ and accelerate oral squamous cell carcinoma growth via mechanical input. ¹⁶ Considering macrophages infiltrate during mechanical force‐induced bone remodeling and exhibit extremely high expression of Piezo1 compared to other reported mechanosensory ion channels, ¹⁷ Piezo1 may be a vital mediator in macrophage proliferative behavior in periodontal environment.

In this study, we established OTM model in vivo and used Flexcell tension system in vitro to explore macrophage behaviors variation upon mechanical stimuli. We demonstrated Piezo1 as a critical point to affect macrophage proliferative phenotype. RNA‐seq was further conducted to dissect the potential molecular mechanism. Mechanistically, Piezo1 may execute its regulatory impact on macrophage proliferation coordinating with Ccnd1. Finally, we generated conditional‐knockout mice, and identified that mice with Ccnd1 deficiency are resistant to macrophage proliferation induced by mechanical stimuli. Taken together, our study reveals a novel mechanism that modulates macrophage proliferation, which may provide precise therapeutic targets for clinical orthodontic treatment.

2. MATERIALS AND METHODS

2.1. Human sample collection

Human periodontal ligament samples were obtained from five subjects who need to extract the maxillary first premolars for orthodontic treatment in Department of Orthodontics, the Affiliated Stomatological Hospital of Nanjing Medical University. First, we extracted the first premolars on one side (control group), and then extracted another first premolar on the other side 7 days after the force was applied (experimental group). The study was conducted following the Declaration of Helsinki and research protocol was approved by the Research Committee of the Affiliated Stomatological Hospital of Nanjing Medical University (PJ2021‐058‐01). All persons gave their informed consent prior to their inclusion in this study. Tissues were fixed with 4% paraformaldehyde for further processing.

2.2. Animals

Adult wild‐type mice (8–10 weeks) were used in this study. Ccnd1 ^f/f mice were crossed with Cx3cr1 ^CreERT2 mice from Gem Pharmatech Company to generate Ccnd1 ^f/f Cx3cr1 ^CreER2T mice (Stock Number T004322) and mice used for the experiment at the age of 8–10 weeks. All mice analyzed were maintained on the C57BL/6 genetic background and housed in a specified pathogen‐free facility in the Medical Experimental Animal Center of Nanjing Medical University, China. All animal protocols were approved by the Committee of Nanjing Medical University for Animal Resources (Approval ID: IACUC‐1601118‐2) and preformed according to the ARRIVE 2.0 guidelines for preclinical studies.

2.3. Orthodontic force application

The mechanical force was applied on mice (n = 5 or 6) based on previous studies. ¹⁸ Briefly, the left first molar was ligated to the incisors with nickel–titanium coil spring and delivered a force to induce tooth movement after anesthesia, no force was applied on the contralateral side as a control. All mice were fed with soft diet after operation.

2.4. Drug administration

For Piezo1 inhibitor treatment, GsMTx4 (Med Chem Express) was dissolved in PBS and thus administered 10 μl (5 μM) to the tooth movement sides of mice through intraperitoneal injection every 2 days according to the previous study. ¹⁵

To induce Cre‐mediated gene recombination, tamoxifen (T5648; Sigma) was dissolved in corn oil (C7030; Solarbio) to get the solution of 20 mg/ml. The transgenic mice were injected with tamoxifen (75 mg/kg) for continuous 5 days intraperitoneally.

2.5. Micro‐CT

Freshly dissected maxillae samples were fixed and scanned at a resolution of 15.6 μm using micro‐computed tomography (Micro‐CT) machine (vivaCT80, Switzerland). A single‐blinded rater renamed the micro‐CT data with randomly generated codes and files were imported into InVivoDental for reconstruction. The distance of tooth movement was measured with Mimics software as previously described. ¹⁹

2.6. Flow cytometry analysis

For tissue flow cytometry analysis, mice were euthanized and the alveolar bone tissues around maxillary first molar were washed with ice‐cold PBS with 1% fetal bovine serum. Then the alveolar bone was digested with collagenase type I (Sigma‐Aldrich) for 30 min at 37°C. For surface staining, antibodies were added to the cells in a volume of 100 μl in the dark for 30 min. For BrdU incorporation assays, BrdU (B23151; Thermo) was injected intraperitoneally at 1 mg in 100 μl PBS per mouse 16 h prior to sacrifice as previously described. ⁹ BrdU staining was conducted according to the instruction of BrdU staining kit (Invitrogen, 8811‐6600‐42). The data were collected on a flow cytometer (FACS Verse8, BD) and analyzed with Flow Jo (10.6.2). Antibodies used were Zombie Violet™ (#423113, BioLegend), CD45‐PE (#103105, BioLegend), CD11b‐APC (#4295321, eBioscience), F4/80‐BV510 (#123135, BioLegend), Ly6C‐APC‐Cy™7 (#128025, BioLegend), Ly6G‐PerCP/Cy5.5 (#560602, BD Biosciences), and BrdU‐FITC (#2271505, Invitrogen). Flow cytometry gating strategies: Macrophage: CD45⁺CD11b⁺F4/80⁺; proliferative macrophages: CD45⁺CD11b⁺F4/80⁺BrdU⁺; neutrophils: CD45⁺CD11b⁺F4/80⁻Ly6G⁺; circulating monocytes: CD45⁺CD11b⁺F4/80⁻Ly6G⁻Ly6C⁺; patrolling monocytes: CD45⁺CD11b⁺F4/80⁻Ly6G⁻Ly6C⁻ (Figure S1A–C).

To assess in vitro cell proliferation, after Piezo1 silence or GsMTx4 administration, we exchanged the culture medium and added 1 mg BrdU to each well and then initiated stretching as described below. Cells were harvested and performed surface staining. Intracellular staining for BrdU was carried out following the manufacturer's protocol (8811‐6600‐42, Invitrogen). For intracellular nuclear antigen Ki67 staining, cells were fixed and permeabilized with fixation/permeablization buffer (00‐5523‐00; Thermo) according to the manufacturer's protocol, Ki67 (ab16667; Abcam) and fluorescein CoraLite488‐conjugated Affinipure Goat Anti‐Rabbit secondary antibody (SA00013‐2; Proteintech) were used in this study.

2.7. Histologic and Immunofluorescent staining

4% paraformaldehyde‐fixed maxillae were then decalcified in 14% EDTA (pH 7.2–7.4). The samples were embedded in paraffin for paraffin sections after dehydration. Paraffin sections were stained with hematoxylin and eosin for histomorphological analysis. For tissue immunofluorescent staining, the samples were embedded in Tissue‐Tek O.C.T and sectioned into 8 μm thickness. Then the sections were washed with PBS for 30 min and blocked for 30 min with 10% BSA at room temperature. Frozen sections were next stained with primary antibodies for anti‐CD68 (1:100, ab955; Abcam) anti‐Ki67 (1:200, ab16667; Abcam), anti‐Piezo1 (1:100, 15939‐1‐AP; Proteintech), anti‐Cyclin D1 (1:100, ab134175; Abcam), and anti‐F4/80 (1:100, ab16911; Abcam). Then, the sections were then incubated with fluorescein Cy3‐conjugated secondary antibody (Beyotime) and fluorescein CoraLite488‐conjugated secondary antibody (Proteintech). Nuclei were counterstained with 4′,6‐diamidino‐2‐phenylindole (DAPI). Images were captured with a fluorescent microscope (Carl Zeiss).

2.8. Macrophage culture

To culture BMDMs, BM cells were isolated from the femurs and tibias of 4‐ to 6‐week‐old mice, lysed red blood cells were cultured in alpha‐minimum essential medium (α‐MEM) and supplemented with 25 ng/ml recombinant M‐CSF (315‐02; Peprotech), 10% FBS (Gibco), and 1% double antibiotics (streptomycin and penicillin; Gibco), and then incubated at 37°C in a humidified atmosphere with 5% CO2 for 5 days. To inhibit Piezo1, 5 μM GsMTx4 was used. To inhibit AKT or GSK3β, 1 μM ARQ 092 (Med Chem Express) or 1 μM AR 014418 (Med Chem Express) was used.

2.9. Application of mechanical stretch

To apply mechanical stimulation to macrophages, BMDMs were seeded uniformly onto six‐well collagen I‐coated BioFlex culture plates with flexible silicon membrane bottoms (Flexcell International Corporation, USA). Using an FX‐5000 Flexcell Tension System (Flexcell International Corporation, USA), cyclic mechanical tension was applied in a sinusoidal pattern at frequency of 0.5 Hz with 5% strains as previously described. ⁹ Cells cultured in the same plates but were not subjected to stretching as controls.

2.10. Quantitative real‐time PCR

Total RNA was extracted from cells and mouse alveolar bones using Trizol reagent (Invitrogen), and converted to cDNA with HiScript II Q RT SuperMix (Vazyme). Next, qPCR was performed with ChamQ Universal SYBR qPCR Master Mix (Vazyme) in ABI QuantStudio7 (Applied Biosystems). The relative gene expression was quantified using the 2^−ΔΔCt method. All the primer sequences are listed in Table S1.

2.11. Calcium imaging

BMDMs were cultured in BioFlex culture plates, after Piezo1 silence or GsMTx4 administration, cells were loaded with Fluo‐4 AM (5 μM; Beyotime) and simultaneously subjected to mechanical stretch (5%, 0.5 Hz) for 1 h. The BMDMs were washed with PBS three times after mechanical stretch. Ca⁺ fluorescence images were captured with a fluorescent microscope (Leica).

2.12. Cell immunofluorescent staining

Cells cultured in BioFlex culture plates were fixed with 4% paraformaldehyde and permeabilized with Triton X‐100 (0.2%), and then incubated with the following specific antibodies overnight at 4°C: anti‐F4/80 (1:100, ab16911; Abcam) and anti‐Ki67 (1:200, ab16667; Abcam). Subsequently, the species‐matched secondary antibodies were used, and the nucleus was stained with DAPI.

2.13. RNA interference

For small interfering RNA (siRNA) experiment, we examined three different sequences and selected the most effective one or two to knock down the target gene. BMDMs were transiently transfected in six‐well plates with Piezo1 siRNA at a final concentration of 100 nM using Opti‐MEM (Gibco) supplemented with Lipofectamine 2000 Reagent (Invitrogen) according to the manufacturer's instructions. The sequences were as follows: Piezo1‐siRNA, 5′‐AGGAAGAAGCCAGAAGCUAATT‐3′ and 5′‐UUAGCUUCUGGCUCUUCCUTT‐3′, negative control (NC)‐siRNA, 5′‐UUCUCCGAACGUGUCACGUTT‐3′ and 5′‐ACGUGACACGUUCGGAGAATT‐3′.

2.14. RNA‐seq and bioinformatics analysis

Total RNA of macrophages from si‐NC and si‐Piezo1 groups were extracted using Trizol reagent following the manufacturer's procedure. After purification and fragmentation, the cleaved RNA fragments were reverse‐transcribed to create the cDNA by SuperScript™ II Reverse Transcriptase (Invitrogen, cat. 1896649; USA), which were next used to synthesize U‐labeled second‐stranded DNAs. Dual‐index adapters were ligated to the fragments, and size selection was performed with AMPureXP beads. The ligated products were enriched with 8 cycles of PCR to create the final cDNA library. The RNA libraries were sequenced on the illumina Novaseq™ 6000 (LC Bio Technology CO., Ltd., Hangzhou, China). The reads were filtered by Cutadapt (https://cutadapt.readthedocs.io/en/stable/, version: cutadapt‐1.9). The RNA‐seq data have been deposited in the Gene Expression Omnibus (GEO) datasets under accession code GSE190470. The FPKM values were calculated to estimate the mRNAs expression abundance. Genes differentially expression analysis was performed by DESeq2 with false discovery rate (FDR) < 0.05 and fold change (FC) > 2. Pathway Enrichment Analysis (KEGG) and Gene Set Enrichment Analysis (GSEA) were performed to further understand the biological functions of differentially expressed genes.

2.15. Western blot

Total protein was extracted by protein extraction buffer (Beyotime). Nuclear proteins were prepared using the Nuclear and Cytoplasmic extraction reagents (Thermo Scientific™ NE‐PER™) according to the manufacturer's instructions. The protein concentration of each sample was measured using the BCA protein assay kit (Beyotime). Nuclear or total cell proteins were separated on 10% SDS‐PAGE and transferred by electroblotting to PVDF membranes (Millipore, IPVH00010). The membranes were blocked in 5% non‐fat milk and incubated with the following primary antibodies: anti‐p‐AKT (1:2000, #4060; CST), anti‐AKT (1:1000, #4685; CST), anti‐Piezo1 (1:500, 15 939‐1‐AP; Proteintech), anti‐p‐GSK3β (1:1000, #9323; CST), anti‐PI3K (1:1000, #4292), anti‐GSK3β (1:1000, #12456; CST), anti‐Cyclin D1 (1:5000, ab134175; Abcam), anti‐p‐Rb (1;1000, #8180; CST), anti‐Lamin B (1:1000, AF5161; Affinity), or anti‐β‐actin (1;1000, #4970; CST) overnight at 4°C, following by incubating with species‐matched secondary antibodies. Immunobands were obtained using ECL solution (Tanon).

2.16. Lentivirus construction and infection

The FLAG‐tagged mouse Ccnd1 (NM_001379248.1) was subcloned to the same vector (pLVX‐SPC1‐FLAG‐c‐Myc‐ZsGreen). The sequence of Ccnd1 overexpression was provided in Table S2. Plasmids were purchased from Viraltherapy Technologies Co. (Wuhan, China) and packaged into lentivirus particles using HEK 293 T cells.

2.17. Statistical analysis

All quantitative data were expressed as the mean ± SD based on at least three independent samples. The Student's t‐test was used for statistical comparisons of two groups and statistical comparisons of multiple groups were performed by one‐way ANOVA followed by post hoc Bonferroni correction. p < .05 was considered to indicate a statistically significant difference. GraphPad Prism 8.0 (Graph Pad Prism Software, Inc, San Diego, CA) was used for statistical analysis.

3. RESULTS

3.1. Mechanical force induces macrophage proliferation in periodontal tissues

To determine the role of mechanical force in modulating macrophage proliferation, we first detected the macrophage quantities and proliferation in human OTM periodontal samples. Immunofluorescence staining showed CD68⁺ macrophages, along with CD68⁺Ki67⁺ proliferating macrophages, were highly increased after force application (Figure 1A). Next, the OTM model was established on 2‐month‐old wild‐type mice to detect periodontal macrophages dynamics at days 0, 3, 7, and 14. The distance of mesial movement in maxillary first molar was measured by micro‐CT (Figure 1B). Flow cytometry measurements revealed that the CD45⁺CD11b⁺F4/80⁺ macrophages peaked at day 7 and slowly decreased to the control level at day 14 (Figure 1C). Further analysis found that Ly6C⁺ circulating monocytes showed no significant change, while the Ly6C⁻ patrolling monocytes elevated from day 0 to day 7 and decreased thereafter (Figure 1D). Conversely, there was a sequential augment in CD45⁺CD11b⁺F4/80⁺BrdU⁺ macrophages during OTM (Figure 1E). Moreover, immunofluorescence staining of the indicated sites was performed to assess the macrophage accumulation (Figure 1F,G), the F4/80⁺Ki67⁺ co‐localization showed the number of proliferation status of macrophages was increased after force application (Figure 1H). And the greatest number of F4/80⁺Ki67⁺ cells was observed at day 14 on the tension side of the periodontal ligament Figure 1I). It has been demonstrated that proliferation provides a general mechanism for the expansion of both peripherally derived and resident macrophages lineages during the resolution of inflammation, ²⁰ which was consistent with the findings above. Taken together, these results suggest that periodontal local macrophage proliferation dominates the macrophage accumulation during the later stage of OTM instead of monocyte recruitment.

FIGURE 1.

Mechanical force induces macrophage proliferation in periodontal tissues. (A) Representative immunofluorescence staining and quantification of CD68⁺ (red) Ki67⁺ (green) cells in normal or loaded periodontium of human. Values are mean ± SD. n = 5. **p < .01. White arrows indicated CD68⁺ Ki67⁺ cells. (B) Representative 3D micro‐computed tomography images of murine tooth movement after orthodontic force application for days 3, 7, and 14. Statistics of tooth movement distance are shown. n = 5. The white arrow indicates the direction of orthodontic force. (C–E) Flow cytometric analysis of macrophages (C), circulating monocytes, patrolling monocytes (D), and proliferative macrophages (E) in murine periodontal tissues after OTM treatment. Values are mean ± SD. n = 5. *p < .05. **p < .01. ***p < .001. ns, no significance. (F) Representative H&E staining of murine first molar after orthodontic force application for days 3, 7, and 14. Scale bars, 400 μm. (G, H) Representative immunofluorescence staining and quantification of F4/80⁺ (red) Ki67⁺ (green) cells on the control and OTM side 3, 7, or 14 d after OTM treatment. Values are mean ± SD. n = 5. **p < .01. ***p < .001. DE, dentin; ns, no significance; P, pulp; PDL, periodontal ligament. White arrows indicated F4/80⁺ Ki67⁺ cells. (I) Statistics of F4/80⁺ Ki67⁺ cells on compression and tension side acquired 14 d after OTM treatment. Values are mean ± SD. n = 5. **p < .01.

3.2. Piezo1 activity‐mediated mechanotransduction promotes macrophage proliferation

Previous study has illustrated that bone‐marrow‐derived macrophages (BMDMs) could sense mechanical force and enhance relative proliferative potential. ⁹ To explore how macrophages perceived mechanical force and translated it into biological signals, we applied a mechanical stretch (MS) model in vitro to mimic the mechanical force during OTM. RT‐PCR revealed that the relative expression of Piezo1 was significantly upregulated after MS (Figure S2A). Meanwhile, Piezo1 mRNA was highly expressed in the alveolar bone compared to other mouse tissues (Figure 2A). The immunofluorescence staining also revealed an obvious rise of Piezo1 co‐localized with F4/80 cells in murine periodontal tissue after force application (Figure 2B). These results led us to speculate that Piezo1 may be a key gene affecting macrophage proliferation capacity.

FIGURE 2.

Piezo1 activity‐mediated mechanotransduction promotes macrophage proliferation. (A) qRT‐PCR analysis of Piezo1 mRNA in alveolar bone and other tissues from WT mice. n = 5. (B) Representative immunofluorescence co‐staining of F4/80 (red) and Piezo1 (green) on the control and OTM side 14 d after OTM treatment. Scale bars, 100 μm, 20 μm. White arrows indicated F4/80⁺ and Piezo1‐positive cells. (C) Immunofluorescence staining and quantification of macrophage proliferative capacity in the condition of GsMTx4 administration with or without in vitro mechanical loading. Scale bars, 50 μm. Values are mean ± SD. n = 3. *p < .05. **p < .01. ns, no significance. (D) Immunofluorescence staining and quantification of proliferative capacity in macrophages infected with si‐NC or si‐Piezo1 with or without in vitro mechanical loading. Scale bars, 50 μm. Values are mean ± SD. n = 3. *p < .05. **p < .01. ns, no significance. (E) Flow cytometric analysis of macrophage proliferation with BrdU in the condition of GsMTx4 administration with or without in vitro mechanical loading. Values are mean ± SD. n = 3. *p < .05. ns, no significance. (F) Representative immunofluorescence staining and quantification of F4/80⁺ (red) Ki67⁺ (green) cells on the OTM side 14 d after OTM treatment with or without GsMTx4 intraperitoneal injection. Scale bars, 100 μm. Values are mean ± SD. n = 6. **p < .01. White arrows indicated F4/80⁺ and Ki67‐positive cells. (G) Representative 3D micro‐computed tomography images of murine tooth movement after orthodontic force application for 14 d with or without GsMTx4 intraperitoneal injection. Statistics of tooth movement distance are shown. Values are mean ± SD. n = 6. **p < .01.

To investigate whether Piezo1 participated in mechanical force‐induced macrophage proliferation, the siRNA knockdown against Piezo1 (Figure S2B) and Piezo1 antagonist GsMTx4 were applied. Results indicated that Piezo1 activated by MS increased intracellular calcium concentrations loaded with fluo‐4. Nevertheless, Piezo1 blockage and knockdown abolished intracellular calcium concentrations upon MS in macrophages (Figure S2C). Immunofluorescence staining of stretched dishes documented increased proliferative capacity in macrophages after MS, while both administration of GsMTx4 and knock‐down of Piezo1 significantly blunted the MS‐induced macrophages proliferation (Figure 2C,D). Additionally, flow cytometry analysis further confirmed that block Piezo1 with GsMTx4 inhibited the number of proliferative macrophages upon MS (Figure 2E).

Subsequently, we evaluated the in vivo effect of GsMTx4 treatment during OTM in mice. We observed a remarkable decrease in F4/80⁺Ki67⁺ cells on the tension side of the periodontal tissue (Figure 2F). More importantly, micro‐CT analysis showed that mice treated with GsMTx4 significantly suppressed tooth movement compared with those treated with PBS (Figure 2G). Collectively, these findings support that the activation of Piezo1 could be related to MS‐mediated macrophage proliferation both in vitro and in vivo.

3.3. Piezo1 modulates macrophage proliferation via PI3K‐AKT pathway and Ccnd1

To probe the mechanisms by which Piezo1 promotes macrophage proliferation upon mechanical stimuli, RNA‐seq experiments were conducted on BMDMs transfected with negative control siRNA (si‐NC) and Piezo1 siRNA (si‐Piezo1), and then exposed to MS. A total of 17 upregulated and 30 downregulated differentially expressed genes (fold change >2 or <0.5, FDR <0.05) were identified in Piezo1‐knockdown BMDMs compared to control BMDMs (Figure 3A). KEGG analysis showed that the PI3K/AKT signaling pathway was inhibited in macrophages with Piezo1 knockdown compared to that in control group (Figure 3B). GSEA analysis also revealed a conspicuous downregulation of genes related to the PI3K/AKT signaling pathway as a result of Piezo1 knockdown (Figure 3C). PI3K/AKT signaling was shown to be activated upon mechanical strain and has been considered as essential prerequisite for cell growth and proliferation. ²¹ , ²² Western blot confirmed that mechanical stimuli positively regulated Phosphorylated AKT (Ser473) and GSK3β (Ser9) in macrophages. As we expected, knockdown or suppression of Piezo1 evidently reduced the protein levels of phosphorylated AKT and GSK3β, which was consistent with the results above. However, the protein expression level of PI3K did not change in each group, which revealed that Piezo1 mediated AKT activation potentially through a PI3K independent pathway (Figure 3D,E).

FIGURE 3.

Piezo1 modulates macrophage proliferation via activation of PI3K‐AKT pathway and Ccnd1. (A) Heatmap of DEGs of BMDMs infected with si‐NC or si‐Piezo1 after in vitro mechanical loading. Blue and orange colors represent low and high expression values, respectively. (B) Representative downregulated KEGG pathway analysis influenced by Piezo1 knockdown. (C) Gene set enrichment analysis (GSEA) was used to confirm the distribution of genes in PI3K/AKT pathway gene set of si‐NC and si‐Piezo1 groups. NES, normalized enrichment score. (D, E) Western blot results indicating the effect of GsMTx4 treatment (E) or Piezo1 knockdown (F) on the protein expression level of p‐AKT/AKT, p‐GSK3β/GSK3β, and PI3K after in vitro mechanical loading. (F) Heatmap of genes involved in macrophage proliferation was performed based on the result of RNA‐seq. Gray and Orange colors represent low and high expression values, respectively. (G) qRT‐PCR analysis of Ccnd1 mRNA in macrophages infected with si‐NC or si‐Piezo1 with or without in vitro mechanical loading. Values are mean ± SD. n = 3. *p < .05. **p < .01. ns, no significance. (H, I) Western blot results indicating the effect of GsMTx4 treatment (H) or Piezo1 knockdown (I) on the nuclear or total protein expression level of Ccnd1 with or without in vitro mechanical loading. (J) Western blot of p‐AKT/AKT, p‐GSK3β/GSK3β, Ccnd1 (total), and Ccnd1 (nuclear) protein level treated with ARQ 092 (an inhibitor of AKT) or (an inhibitor of AR 014418), with or without in vitro mechanical loading. (K) Flow cytometric analysis of Ki67 expression in Piezo1 suppression, Ccnd1 overexpressing macrophages with or without mechanical stretch. Values are mean ± SD. n = 3. *p < .05. **p < .01. ***p < .001.

We further explored the proliferation‐related genes based on the RNA‐seq results and qRT‐PCR was performed to detect mRNA expression level of the top five genes with significant difference in the RNA‐seq data with or without mechanical loading. Analysis showed that the relative mRNA expression of Ccnd1 was significantly upregulated after mechanical loading, while others were downregulated or showed no significant difference (Figure S3). Thus, we identified Ccnd1, an indispensable protein for regulating cell cycle, serving as a potential downstream target factor of Piezo1‐activated AKT/GSK3β pathway (Figure 3F). These results were also confirmed by qRT‐PCR (Figure 3G). It has been previously proved that phosphorylation of AKT induced GSK3β deactivation through phosphorylation of GSK3β, contributing to the stabilization and nuclear translocation of Ccnd1. ²³ , ²⁴ Concomitantly, Western blot was performed to substantiate that both total and nuclear Ccnd1 protein increased when subjected to mechanical stimuli. Importantly, these increases were blocked after administration of GsMTx4 or Piezo1 knockdown (Figure 3H,I). The immunofluorescence staining also confirmed the same results as the Western blot results (Figure S4A,B). Small molecule inhibitors targeting AKT (ARQ 092) or GSK3β (AR 014418) were further used to confirm the AKT/GSK3β pathway is involved in Piezo1‐mediated stabilization and nuclear localization of Ccnd1. As shown in Figure 3J, AKT and GSK3β signaling was suppressed by ARQ 092, and phosphorylated GSK3β was activated by AR 014418 in BMDMs after force application. As expected, both total and nuclear‐localized Ccnd1 decreased after treatment with ARQ 092 and increased after treatment with AR 014418 (Figure 3J).

Furthermore, macrophages were infected with negative control lentivirus or Ccnd1 overexpressing lentivirus. The qRT‐PCR and Western blot data demonstrated that the expression level of Ccnd1 significantly altered in the overexpressing lentivirus‐transfected macrophages (Figure 5A,B). Flow cytometry measurements revealed that overexpression of Ccnd1 obviously stimulated macrophage proliferation, which was significantly amplified under MS. Importantly, Ccnd1 overexpression restored the macrophage proliferative capacity suppressed by the Piezo1 antagonist GsMTx4 in the context of mechanical stretch (Figure 3K). We also examined the number of F4/80⁺Ccnd1⁺ cells in stretched tissues and normal tissues, showing a prominent upregulation of Ccnd1 during OTM (Figure 4A). Together, these results indicated that Ccnd1 may serve as a critical target of Piezo1 in the regulation of macrophage proliferation through PI3K/AKT/GSK3β pathway.

FIGURE 5.

Macrophage Ccnd1 deficiency represses mechanical‐force‐induced bone remodeling. (A) Experimental design: Ccnd1 ^f/f or Ccnd1 ^f/f Cx3cr1 ^CreERT2 mice were injected with tamoxifen (75 mg/kg body weight intraperitoneally) for 5 consecutive days. Orthodontic force was applied to mice at the 6th day after the first tamoxifen dose. After 14 days of OTM, the maxillary was harvested. (B) Representative 3D micro‐computed tomography images of control and OTM side after orthodontic force application for 14 days in Ccnd1 ^f/f or Ccnd1 ^f/f Cx3cr1 ^CreERT2 mice. Statistics of tooth movement distance are shown. Values are mean ± SD. n = 8. **p < .01. (C) Representative H&E staining of murine first molar after orthodontic force application for 14 d in Ccnd1 ^f/f or Ccnd1 ^f/f Cx3cr1 ^CreERT2 mice. Quantification of BV/TV is shown which represents bone volume fraction. Scale bars, 400 μm, 100 μm. Values are mean ± SD. n = 5. **p < .01. ***p < .001. ns, no significance. (D) Gating and quantification of periodontal tissue macrophage radio and proliferative level after OTM treatment for 14 d in Ccnd1 ^f/f or Ccnd1 ^f/f Cx3cr1 ^CreERT2 mice. Values are mean ± SD. n = 6 or 7. **p < .01. (E) Representative gating and quantification of periodontal tissue monocytes variation after OTM treatment for 14 d in Ccnd1 ^f/f or Ccnd1 ^f/f Cx3cr1 ^CreERT2 mice. Values are mean ± SD. n = 6 or 7. ***p < .001. ns, no significance. (F) Representative immunofluorescence staining and quantification of F4/80⁺ (red) Ki67⁺ (green) cells on the OTM side 14 d after OTM treatment in Ccnd1 ^f/f or Ccnd1 ^f/f Cx3cr1 ^CreERT2 mice. Scale bars, 100 μm. Values are mean ± SD. n = 5. **p < .01. White arrows indicated F4/80⁺ and Ki67‐positive cells. (G) Graphic abstract of this study.

FIGURE 4.

Ccnd1 is requisite for mechanical‐strain‐mediated macrophage proliferation. (A) Representative immunofluorescence co‐staining and quantification of F4/80 (red) and Ccnd1 (green) on the control and OTM side 14 d after OTM treatment. Scale bars, 100 μm, 20 μm. Values are mean ± SD. n = 5. **p < .01. White arrows indicated F4/80⁺ and Ccnd1⁺cells. (B) Immunofluorescence staining and quantification of Ki67⁺ cells in macrophages from Ccnd1 ^f/f and Ccnd1 ^cko with or without in vitro mechanical loading. Scale bars, 50 μm. Values are mean ± SD. n = 3. *p < .05. ns, no significance. (C, D) Flow cytometric analysis of macrophage proliferation with BrdU from Ccnd1 ^f/f and Ccnd1 ^cko mice with or without in vitro mechanical loading. Values are mean ± SD. n = 3. *p < .05. ***p < .001. (E) Western blot results of Ccnd1 and p‐Rb protein expression in macrophages from Ccnd1 ^f/f and Ccnd1 ^cko with or without in vitro mechanical loading.

3.4. Ccnd1 is requisite for mechanical stretch mediated macrophage proliferation

As a key regulator of cell cycle, Ccnd1 has been well described to involve in multiple cell proliferation. However, the detailed roles of Ccnd1 in modulating macrophage proliferation are still unknown. To determine this progress, we intercrossed Ccnd1 ^f/f mice with Cx3cr1 ^CreERT2 transgenic mice to create tamoxifen‐inducible Ccnd1 knockout (Ccnd1 ^f/f Cx3cr1 ^CreERT2) mice, in which Ccnd1 expression was inducibly reduced upon tamoxifen treatment in macrophages, allowing temporal control of Ccnd1 expression (Figure S5C,D). BMDMs were cultured from both Ccnd1 ^f/f and Ccnd1 ^f/f Cx3cr1 ^CreERT2 mice with M‐CSF and exposed them to MS. As expected, MS stimulated macrophages from Ccnd1 ^f/f mice exhibited higher proliferative capacity, whereas this effect was abrogated by Ccnd1 knockout (Figure 4B). In addition to the staining of Ki67 and F4/80, flow cytometry analysis demonstrated the positive role of Ccnd1 in regulating the proliferation of macrophages (Figure 4C,D). Western blot analysis also proved that Ccnd1 knockout could reduce downstream Rb phosphorylation level in macrophages under mechanical stretch (Figure 4E).

3.5. Macrophage Ccnd1 deficiency represses macrophage expansion during OTM

To evaluate whether Ccnd1 in macrophages is required for OTM in vivo, the orthodontic devices were applied to control Ccnd1 ^f/f mice and Ccnd1 ^f/f Cx3cr1 ^CreERT2 mice after tamoxifen injection (Figure 5A). Micro‐CT scanning and three‐dimensional (3D) reconstruction revealed an observable suppression of tooth movement distance (Figure 5B), and H&E stained sections indicated an increase in BV/TV (%) in Ccnd1 ^f/f Cx3cr1 ^CreERT2 mice compared to Ccnd1 ^f/f mice (Figure 5C).

The role of Ccnd1 in regulating periodontal tissues macrophage biological behavior was further dissected by flow cytometry and immunofluorescence staining. First, we observed a notable reduction of both macrophages and proliferative macrophages during OTM in Ccnd1 ^f/f Cx3cr1 ^CreERT2 mice compared with Ccnd1 ^f/f mice, which was determined by flow cytometry analysis (Figure 5D). Further measurements revealed that Ccnd1 deficiency led to a slight rise on Ly6C⁺ circulating monocytes. On the contrary, we found no difference in recruitment of Ly6C⁻ patrolling monocytes into periodontal tissues between Ccnd1 ^f/f and Ccnd1 ^f/f Cx3cr1 ^CreERT2 mice (Figure 5E). In addition, cellular proliferation with Ki67 staining documented decreased F4/80⁺Ki67⁺ cells in periodontal tissue in Ccnd1 ^f/f Cx3cr1 ^CreERT2 mice compared to their control littermates (Figure 5F). Taking in aggregate, these data support that Ccnd1 acts as a critical intracellular mechanical sensor responding to mechanical force and modulating macrophages' biological behavior.

4. DISCUSSION

Macrophages infiltration is critical to the inflammatory bone remodeling induced by orthodontic force. As osteoclast precursors, macrophages directly regulate osteoclast differentiation and facilitate bone resorption. ²⁵ The prevailing dogma is that circulating monocytes are recruited to the site of inflammation. ²⁶ Yet, recent findings have revealed that the proliferation of macrophage populations provides a general way for macrophage expansion at particular stages during inflammation. ²⁰ Although certain studies have indicated that mechanical force contributes to macrophage proliferation in several diseases, ⁹ its role in OTM is unclear. In this study, we demonstrated that macrophage proliferation controls the macrophage accumulation at the later stage of tooth movement. Natheless, the suppression of Piezo1, a novel mechanosensitive ion channel, could inhibit macrophage proliferation and bone remodeling upon mechanical stimuli. More importantly, we verified Piezo1 may regulate macrophage proliferation through AKT/GSK3β pathway induced nuclear translocation of Ccnd1 (Figure 5G).

Orthodontic treatment complies with Wolff's law—the remodeling of bone and surrounding tissues is an adaptive biological response to this external force. ²⁷ However, how these physical interventions exert molecular reactions in cells and tissues in this process, especially around teeth, remains unclear. While much has recently been investigated the mechanobiology of stem cells, fibroblasts, and others, ²⁸ , ²⁹ the knowledge of macrophages mechanobiology is still insufficient. Macrophages are sensitive to mechanical stimuli, and affect their biological behavior by converting mechanical signals into cellular messages depending on various mechanosensors. ³⁰ Mechanosensitive ion channels mediated Ca²⁺ influx has been identified as a key second messenger in macrophage polarization. ¹⁴ , ³¹ The reduction of alveolar surface tension in vivo could regulate the shape of cells and increase the phagocytic activity of alveolar macrophages. ³² As dynamic and erratic mechanosensitive organelles for macrophage migration, invasion and degradation of surrounding matrix, podosomes were sensitive to substrate stiffness. ³³ Here, we confirmed Piezo1 was highly expressed in macrophages and alveolar bone and acted as a key sensor in response to mechanical stimuli.

Piezo1 is a calcium ion channel that could sense different stimuli in various cells and tissues. ³⁴ Piezo1 activation in osteoblasts is associated with regulating bone homeostasis via osteoblast–osteoclast crosstalk. ³ Stimulation of Piezo1 by mechanical signals promotes bone anabolism in osteocytes. ³⁵ More importantly, Piezo1 signaling is essential for macrophage mechanosensation, mutations in Piezo1 abrogate macrophage inflammatory‐related transcriptional reprograming. ¹⁷ In our study, we substantiated that Piezo1 mediated the mechanical signal transduction of macrophages both in vivo and in vitro. GsMTx4 and siRNA were used to silence Piezo1 and thus macrophages proliferation was effectively confined under mechanical stretch. Nevertheless, GsMTx4 is not a specific inhibitor of macrophage Piezo1. In the future, we may generate macrophage‐specific Piezo1 knockout mice by crossing Piezo1^f/f mice with Cx3cr1^CreERT2 mice in OTM model which could make our hypothesis more persuasive.

Various intracellular signaling pathways have been explored for macrophage proliferation including multiple ligands and receptors, second messengers, and transcription factors. ³⁶ , ³⁷ Our RNA‐seq based on KEGG pathway enrichment and GSEA reveal that Piezo1 associated with PI3K/AKT signaling pathway are vital in macrophage proliferation during bone remodeling. AKT is generally activated by PI3K. ³⁸ However, multiple researches showed that Ca2⁺ influx mediated by ion channels is also known to activate AKT in a PI3K‐independent pathway. ³⁹ Therefore, we further detected the expression level of PI3K by Western blot and results revealed that Piezo1‐mediated AKT activation was independent from PI3K activity, as administration of GsMTx4 or Piezo1 knockdown did not change the expression of PI3K in BMDMs. We further identified Ccnd1, also known as cyclin D1, as a potential downstream effector of Piezo1 activated AKT/GSK3β signaling pathway. As a member of the cyclin family, Ccnd1 is an indispensable factor in regulating cell cycle to maintain the normal growth and development of various cells. Besides, Ccnd1 exerts its function by binding to cyclin‐dependent kinase4/cyclin‐dependent kinase6 (CDK4/CDK6) and promoting the phosphorylation of Rb. ⁴⁰ Both Phosphorylated AKT and Ccnd1 were augmented after mechanical stretch stimuli in cultured macrophages. Moreover, we found that the expression level of Ccnd1 was elevated during orthodontics bone remodeling microenvironment and deletion of Ccnd1 in macrophages caused a proliferative phenotype variation that was similar to suppression of Piezo1.

The importance of macrophages to perceive and respond to their surrounding mechanical microenvironment is recognized increasingly. Understanding the mechanisms how macrophages interpret mechanosensitive information is critical for modulating OTM. Our findings reveal a unique role of Piezo1‐AKT‐Ccnd1 axis in regulating macrophage proliferation during OTM, which might be a novel target for manipulating the OTM rate.

AUTHOR CONTRIBUTIONS

Hao Xu contributed to the conception, design, data acquisition, analysis, and interpretation, drafted and critically revised the manuscript; Jiani Guan contributed to the data acquisition and analysis, interpretation, drafted and critically revised the manuscript; Zhichun Jin contributed to the conception, data acquisition, and critically revised the manuscript; Cheng Yin, Shengnan Wu, and Wen Sun contributed to the data acquisition and interpretation, critically revised the manuscript; Hanwen Zhang and Bin Yan contributed to the conception, design, drafted and critically revised the manuscript. All authors gave final approval and agreed to be accountable for all aspects of the work.

DISCLOSURES

The authors declared no potential conflicts of interest.

Supporting information

Table S1

Figure S1

Xu H, Guan J, Jin Z, et al. Mechanical force modulates macrophage proliferation via Piezo1‐AKT‐Cyclin D1 axis. The FASEB Journal. 2022;36:e22423. doi: 10.1096/fj.202200314R

DATA AVAILABILITY STATEMENT

The RNA‐seq data have been deposited to Gene Expression Omnibus database under accession code GSE190470. All the data that support the findings of this study are available from the corresponding author upon reasonable request.