Chewing loads alter TMJ metabolism and Zn homeostasis via PIEZO1/ TRPV4 - PubMed (original) (raw)

Chewing loads alter TMJ metabolism and Zn homeostasis via PIEZO1/ TRPV4

Yongmei Wang et al. J Biol Eng. 2025.

Abstract

The biomechanical effects of bite forces on a temporomandibular joint within the craniofacial masticatory complex remain unclear. Herein, we will discuss the effect of recovering a hypofunctioning temporomandibular joint using normal bite forces. Specifically, we will correlate spatiotemporal: (1) biomolecular expressions to (2) physical characteristics of the inner and outer regions of the temporomandibular condyles. In this study, we divided twelve male Sprague-Dawley rats into three groups: hard-food (N = 3, hard pellets), soft-food (N = 3, soft-chow – hard-pellets grounded to powder), and recovery (N = 6, rats on soft-food for 9 weeks and switched abruptly to hard-food for an additional 8 weeks). Rats were euthanized at 17 weeks, and their condyles were imaged using a preclinical micro-computed tomography to estimate bone mineral density and volume fraction. The condyles were prepared for histology (proteoglycans using Safranin O counterstained with Fast Green, and matrix using Picrosirius Red). We performed immunolocalization of mechanosensory ion channels, PIEZO1 and TRPV4, hypoxia inducible factor-1α (HIF-1α), Indian hedgehog (Ihh), mitofusin 2 (MFN2), and cell senescence (p16). Analogous to in vivo conditions, the PIEZO1 and TRPV4 MS-ion channels of the ATDC5 cells were mechanically activated at 5 and 7 days in vitro. Corresponding agonists and antagonists were used for activation of the mechanosensory ion channels. Staining for zinc (Zn) alongside reverse transcription quantitative polymerase chain reaction (RT-qPCR) for quantification of PIEZO1 and TRPV4, superoxide dismutase 2 (SOD2), dynamin-related protein -1 (DRP-1), metal transcription factor-1 (MTF-1), HIF-1⍺, and metabolism (MFN2), Zn transporters1, 5, and 8 (ZnT1, ZnT5, ZIP8) were performed. The oxygen sensor HIF-1α was associated with regional metabolism, as indicated by MFN2, SOD2, and DRP expressions in proliferating (Ihh) and senescent (p16) cells across all groups. Increased Zn levels, along with MTF-1 and ZIP8 and ZnT1/5, were correlated with PIEZO1 in the proliferative zone, and TRPV4 in the inner hypertrophic and subchondral zones of the condyle. These temporal biochemical changes correlated with increased widths of the hypertrophic region and cartilage with a matrix structure loss but recovered bone mineral density and bone volume fraction in condyles of the recovery group. Results illustrate biochemical changes in the temporomandibular condyle in response to a change in bite force. A major finding with significant novelty is that the mechanically induced biochemical changes involve micronutrient Zn metabolism in relation to the activation of two mechanosensory ion channel proteins PIEZO1 and TRPV4. Hard food stimulation of cells that also are HIF-1α positive can alter Zn levels regulated by Zn transporters; a proposed mechanochemo cascade necessary for HIF-1α regulation and subsequently condylar function. In summary, though the mechanical properties are not yet known, chewing hard foods may help regain cartilage and bone thickness lost through TMJ disuse or chewing on soft chow only.

Supplementary Information: The online version contains supplementary material available at 10.1186/s13036-025-00564-2.

Keywords: Chewing loads; Mitochondria; PIEZO1; ROS; TRPV4; Zinc.

PubMed Disclaimer

Conflict of interest statement

Declarations. Ethics approval and consent to participate: The University of California San Francisco (UCSF) Institutional Animal Care and Use Committee approved the experimental procedures used in this study (approval no. IACUC protocol AN200023-00I) on February 11, 2025. Consent for publication: Not applicable. Competing interests: The authors declare no competing interests.

Figures

Fig. 1

Fig. 1

Experimental Design: Investigating bite forces and mechanobiology in TMJ and chondrocytes using both animal (in vivo) and alternate models (in vitro). (A) Twelve male Sprague-Dawley rats received hard-food (HF, N = 3) and soft-food (SF, N = 3), and a third group, the recovery (REC, N = 6) was given 9 weeks of SF and was immediately switched to 8 weeks of HF Theoretical projection of reactionary loading from the dentoalveolar joint to the temporomandibular joint in the rats from HF, SF, and Rec groups are shown. The condyles from the temporomandibular joints (TMJs) were prepared for CT imaging, histology, and immunohistochemistry. (B) The ATDC5 chondrocyte behavior at 5 days ((D5), N = 12) and 7 days ((D7), N = 12) following treatments with agonists of PIEZO1 (Yoda1, N = 6) and TRPV4 (GSK101790A, N = 6), and antagonists of PIEZO1 (GsMTx4, N = 6) and TRPV4 (GSK205, N = 6) for 2 h were mapped. Quantitative analyses of biometal zinc (Zn), and biomolecules PIEZO1, TRPV4 (MS-ion channels), SOD2, DRP-1, MTF-1, HIF-1α, and MFN2 (metabolism), ZnT1, ZnT5, and ZIP8 (Zn transporters) was performed.

Fig. 2

Fig. 2

Differential expressions of PIEZO1 and TRPV4 in TMJ cartilage and bone under different bite forces. (A) PIEZO1 was expressed in the fibrous outer layer (OL) of HF, and was not observed in SF and REC. On the contrary, TRPV4 was expressed in the IL (subchondral) and bone regions of the HF, SF, and REC. PIEZO1 and TRPV4 were expressed in the OL (proliferative zone) of SF and REC compared to HF. In contrast, PIEZO1 and TRPV4 were expressed in the IL (hypertrophic chondrocytes and subchondral bones) of HF and REC groups. SF animals had significantly lower proliferation than HF, determined by a decrease in expression of the proliferative marker Indian hedgehog (Ihh), increase in cell senescence marked by p16, and decrease in metabolism marked by MFN2. The REC group displayed similar proliferation to HF (Ihh), decreased cell senescence (p16 stats), and an increase in metabolism (MFN2). HIF-1⍺ and Zn were decreased in the SF group but restored in the REC group. Counter-stained by hematoxylin (DAB staining, a1) or DAPI (immunofluorescence, a2). Scale bar = 200 μm. (B) Spearman correlation illustrates that Zn is positively correlated with PIEZO1 and TRPV4, and MFN2 and negatively correlated with HIF-1α, Ihh, and p16. However, in SF, Zn is positively correlated with PIEZO1, MFN2 and p16, and continues to be negatively correlated with HIF-1α and Ihh. In REC, Zn is positively associated only with MFN2, negatively associated with p16 and the MS-ion channels, and has no association with HIF-1α and Ihh. Data are presented as median, IQR, and 95% CI for bar charts or mean (SD) otherwise. A significant increase (+ ve) or decrease (-ve) with a p < 0.05 are presented in Table 1. Scale bars: 200 μm

Fig. 3

Fig. 3

PIEZO1 and TRPV4 agonists and antagonists influenced Zn levels and temporal changes in gene expression in ZnTs, and cellular metabolism. (A) At D5 and D7, PIEZO1 and TRPV4 channels were activated (an increase relative to vehicle) or blunted (a decrease relative to vehicle) by PIEZO1 (Yoda1, GSK1016790A) and TRPV4 (GsMTx4, GSK205) agonists and antagonists respectively. Zn levels decreased between vehicle and cells treated with PIEZO1 agonist and TRPV4 antagonist at D5 and increased at D7. No significant differences were observed between vehicle, and PIEZO1 antagonist and TRPV4 agonist at D5 and D7. Scale bar − 20 μm. (B) The logFC of gene expressions with PIEZO1 agonist and antagonist at D5 and D7 were of heterogenous relationships, while for TRPV4 agonist and antagonist were consistent with an increase in D5 and a decrease in D7. (C) Hiearchical correlations indicated increased expressions of all transcripts with PIEZO1 agonist at D5 (box a) and to a lesser extent with antagonist (box b). However, at D7 the increased expressions of SOD2, and ZnT1 and ZnT5 remained steady (box c) with PIEZO1 agonist and decreased with antagonist (box d). With TRPV4 agonist, at D7, an increased expression in MTF1 with HIF-1α (box d), ZIP8, DRP, and MFN2 (box e), and to a lesser extent SOD2 were observed. It appears that with an increase in load (agonist) that there is a concomitant PIEZO1 and TRPV4 axes development from D5 through D7 with a decrease in HIF-1α in D7. The two axes include TRPV4-MTF1-MFN2-DRP-ZIP8 axis where intracellular Zn is controlled by ZIP8, and PIEZO1-SOD2-ZnT5 axis with an increase in Zn levels controlled by ZnT5. A decrease in load however encourages a flipped axes with the MSion channels in that TRPV4-SOD2-ZnT1/5, and PIEZO1-MFN2-DRP-ZIP8 were observed. Note: D5 and D7 – at Day 5 and Day 7; Log fold change (logFC) is the ratio of biomolecular expression levels in test condition (treated specimen) compared to control (vehicle). Scale bars: 20 μm

Fig. 4

Fig. 4

The physical properties and bone mineral density of REC trended toward healthy levels compared to SF. (A). A decrease in bone volume fraction (BVF) and condylar morphology (radius of curvature) and bone mineral density (BMD) were observed in SF contrasting an increase in REC. Physical properties of SF diverged from HF, but REC converged to HF (bone volume fraction (BVF)), and condylar morphology (radius of curvature)) and bone mineral density (BMD)). (B). Depth correlations between BMD and BVF within and across all groups are shown in (b1). Average BMD vs. BVF of REC converged with HF, but SF group diverged from HF. Standard deviation (SD) and average BMD and BVF also are plotted (b2). Histology sections shown in C are correlated with X-ray intensity maps of the same condyle/rat within each group to illustrate BMD variations. (C). Safranin-O/Fast Green stain (organ/red tissue above blue bone) illustrated a significantly decreased cartilage thickness in SF (P < 0.001) compared to no difference in REC relative to HF (not significant - ns) (c1). Porosity depth profiles indicated two significant peaks. Peak A closer to cartilage, and peak B in bone. Peaks A and B were of similar depths in HF and REC, but peak B was closer to SF condylar surface (c2). Scale bars: 400 µm

Fig. 5

Fig. 5

Working Hypothesis - Bite force restoration activates mechanosensitive pathways and Zn signaling associated with TMJ cartilage and bone recovery through MS-ion channel activities and HIF-1α related cellular responses. The recovery of a hypo-functioning TMJ by switching the bite forces from SF to HF (ΔL(+ve); grey block arrow) will induced deformation in cartilage, bone, and the hypertrophic cartilage-subchondral bone interface, and activate PIEZO1 and TRPV4 Ms-ion channels in tissue and zone-specific cells. Activation of MS-ion channels will induce mechanotransduction and alter Zn levels and flow of Zn through Zn transporters (ZnT1/T5) and regulate Ihh to stimulate differentiation of proliferating cells into chondrocytes and maintain normal chondrocyte density in the cartilage. These cell responses maintain normal/low oxidative (HIF-1α) and ROS production (SOD2), increase/maintain mitochondrial function (MFN2) and cell metabolism, cell fate (p16), and recover the subchondral bone volume (BV/TV) and mineral density (BMD).

References

    1. Botello-Smith WM, Jiang W, Zhang H, Ozkan AD, Lin YC, Pham CN, Lacroix JJ, Luo Y. A mechanism for the activation of the mechanosensitive piezo1 channel by the small molecule yoda1. Nat Commun. 2019;10(1):4503. -DOI -PMC -PubMed
    1. Bürkner PC. Bayesian item response modeling in r with Brms and Stan. J Stat Softw. 2021;100(5):1–54. -DOI
    1. Cardoneanu A, Macovei LA, Burlui AM, Mihai IR, Bratoiu I, Rezus II, Richter P, Tamba BI, Rezus E. 2022. Temporomandibular joint osteoarthritis: pathogenic mechanisms involving the cartilage and subchondral bone, and potential therapeutic strategies for joint regeneration. Int J Mol Sci. 24(1). -PMC -PubMed
    1. Cox CD, Bavi N, Martinac B. Biophysical principles of ion-channel-mediated mechanosensory transduction. Cell Rep. 2019;29(1):1–12. -DOI -PubMed
    1. Deng H, Qiao X, Xie T, Fu W, Li H, Zhao Y, Guo M, Feng Y, Chen L, Zhao Y et al. 2021. Slc-30a9 is required for zn(2+) homeostasis, zn(2+) mobilization, and mitochondrial health. Proc Natl Acad Sci U S A. 118(35). -PMC -PubMed

LinkOut - more resources