MiRNA-140-5p Modulates Cartilage Mechanical Environment by Preserving Surface Stress Homeostasis

Abstract

Introduction

This study explores the interplay between miRNA-140-5p expression and mechanical stress in cartilage within joint biomechanics.

Methods

Tibial plateau specimens, CT data, and mechanical parameters were obtained from healthy and OA donors. Twenty-four mice, including 12 miRNA-140-5p knockdown (MUT-group) and 12 wild-type (WT group), underwent anterior-cruciate-ligament-transection (ACLT) or sham-operation (SHAM). Finite element analysis, 3D simulation, CT scans, immunohistochemical staining, and fluorescence in situ hybridization were conducted. Primary chondrocytes with or without miRNA-140-5p agomir were loaded mechanically and analyzed by RT-qPCR, Western blot, and phalloidin staining.

Results

The mechanical coupling unit comprised articular cartilage and subchondral bone, with cartilage apparent compressive modulus linked to the trabecular bone structure (P < 0.05). Healthy-joint samples and low-stress regions in animal samples exhibited high miRNA-140-5p expression (P < 0.05) and low RhoA expression. OA or high-stress regions showed the opposite trend (P < 0.05). MiRNA-140-5p knockdown increased joint loading in mice. In vitro, miRNA-140-5p overexpression reduced RhoA and cytoskeletal remodeling, maintaining chondrocyte mechano-responsiveness.

Conclusions

Our study reveals a link between mechanical stress and miRNA-140-5p, implying its role in maintaining joint mechanical homeostasis. These findings enhance understanding of biomechanical–molecular interplay, though further studies are needed to assess therapeutic potential in osteoarthritis.

Introduction

Articular cartilage is essential for preserving joint function by facilitating load distribution and enabling smooth articulation. ¹ Together with the underlying subchondral bone, it forms a mechanically coupled unit that plays a critical role in maintaining biomechanical homeostasis in joints.^2,3 Continuous or abnormal mechanical loading can disrupt this coupling system, compromising tissue integrity and mechanical responsiveness. ⁴

Cells within articular cartilage sense and respond to external mechanical cues through a process known as mechanotransduction. ⁵ This process converts physical forces—such as compression, tension, or shear—into intracellular biochemical signals.^6,7 At the core of this process lies Ras homolog family member A (RhoA), a pivotal protein in sensing changes in external stiffness and regulating cell cytoskeletal organization by modulating actin dynamics, ⁸ which significantly contribute to cell properties. The quantity and dynamic assembly of actin significantly affect cell stiffness, while traction on extracellular collagen further influences tissue elasticity through reassembly.

MicroRNA-140-5p (miRNA-140-5p) is a cartilage-enriched, evolutionarily conserved non-coding RNA that regulates gene expression post-transcriptionally. ⁹ It is known to play an important role in chondrocyte differentiation and extracellular matrix maintenance. ¹⁰ Research has shown that miRNA-140-5p can regulate the expression of RhoA protein, ¹¹ suggesting its involvement in cellular responses to mechanical stimulation. However, its role in modulating chondrocyte mechanics and cytoskeletal remodeling under load has not been fully elucidated.

In this study, we investigate the function of miRNA-140-5p in regulating chondrocyte mechanical adaptation under mechanical loading, with a particular focus on its interaction with RhoA and its impact on cytoskeletal remodeling. By examining both tissue-level and cellular responses, we aim to uncover how miRNA-140-5p contributes to maintaining the mechanical homeostasis of cartilage. While osteoarthritis is one potential context for such mechanical dysregulation, our findings are intended to inform broader mechanisms of cartilage adaptation and degeneration in response to mechanical stress.

Methods

Human Subjects

Tibial plateau specimens were obtained from 13 OA patients undergoing total knee replacement, and 5 control samples were derived from cadaveric donations (Supplementary Material 1 and Supplementary Table S1). Knee joint CT examinations were performed prior to surgery or joint collection, and the specimens underwent histological examination. Ethical approval was obtained from the Harbin Medical University ethics committee (Approval number: hrbmuecdc20210310), and a waiver of consent was approved by the institution.

Animals and Isolation of Primary Chondrocytes

C57BL/6J-background mice and miRNA-140-5p knockdown mice were purchased from the Nanjing Biomedical Research Institute of Nanjing University (Supplement Method 1), China. Ethical approval for animal experiments was obtained from the same institute’s ethics review panel. Genotyping was performed via PCR using genomic DNA isolated from the mouse tail (Supplemental Fig. S1A). The primer sequences are shown in Supplementary Table S2. After anesthesia with Avertin, 3-month-old male mice were randomly assigned to the WT-SHAM, WT-ACLT, MUT-SHAM, and MUT-ACLT groups. Following anesthesia, all mice (n = 5~6) underwent anterior cruciate ligament transection (ACLT) or SHAM surgery (SHAM) (Supplementary Method 2). ACLT was used to induce mechanical instability-associated OA in the right knee, and SHAM surgery was performed on independent mice. The mice were euthanized after 30 days postoperatively, and the tibial plateau of the knee joint was collected. The knee joint cartilage of wild-type mice was used for primary chondrocyte extraction (Supplementary Method 3).

Cell Culture and Treatment and Transfection

Primary cells were cultured in 6-well plates for 24 hours (Supplementary Fig. S1B). Compressive forces of 1, 1.925, and 2.425 g/cm² were applied for 3, 8, 24, and 48 hours, respectively. ¹² The cells underwent siRNA treatment (RIBOBIO, RN: R10034.8) and were transfected with chemically synthesized mature miRNA-140-5p and RhoA (Supplementary Table S3). Transfection efficiency was verified using the RT-qPCR method (Supplementary Fig. S2G. S2I). Cells were subjected to the designated forces, and subsequent analyses were conducted.

RT-qPCR and Western Blotting

RNA extraction was performed using the TRIzol reagent. Quantitative analysis of miRNA-140-5p (GCCAGTGGTTTTACCCTATGGTAG) was conducted using RT-qPCR. For Western blot analysis, cell lysis was performed with RAPI. After SDS-PAGE gel electrophoresis, the protein samples were transferred onto a PVDF membrane, followed by blocking. The membrane was then incubated with primary antibodies (RhoA, CY5640-ABWAYS; COLI, AB260043-Abcam; COLII, AB188570-Abcam) overnight at 4 °C, followed by secondary antibody incubation. After washing, the bands were visualized using an enhanced chemiluminescence reagent.

Histochemistry and Immunohistochemistry

Knee joint samples were collected, fixed, embedded in paraffin, and sectioned at 4 μm (Supplementary Method 4). Hematoxylin and eosin (H&E) and Safranin O–Fast Green staining (Supplementary Fig. S1C, S1E) were performed. OARSI histological scores were evaluated by H&E. Three distinct locations (i.e., lateral condyle, middle area, and medial condyle) were chosen, each 80 μm apart. Immunohistochemical staining for RhoA was performed according to standard protocols. Fluorescence-in-Situ-Hybridization (FISH) was performed using a detection reagent kit (BOSTER, MK10434) with probe sequences targeting hsa-miRNA-140-5p (GCCAGTGGTTTTACCCTATGGTAG). F-actin was labeled using Phalloidin Conjugates (YEASEN, 40735ES75) for cell staining. Morphological assessment was performed using ImageJ software.

Identification of Mouse Chondrocytes

For Type II collagen (COL-II) immunofluorescence, chondrocytes were seeded, fixed, and incubated with primary antibody and Rhodamine-Conjugated AffiniPure Goat Anti-Rabbit IgG. DAPI staining was performed to visualize cell nuclei. The stained cells exhibited red fluorescence in the cytoplasm and cell membrane under a fluorescence microscope. For toluidine blue staining, chondrocytes were seeded and fixed with paraformaldehyde and ethanol. The samples were stained with 1% toluidine blue solution at 60 °C for 3 hours. After rinsing, coverslips were air-dried, fixed with neutral gum, and observed under a microscope. Toluidine blue staining revealed dark blue nuclei, light blue cytoplasm, and extracellular matrix (Supplementary Fig. S2A, S2B, S2C).

Measurement of Mechanical Parameters

The static mechanical properties of cartilage tissue (Supplementary Fig. S2D, S2E) were evaluated using an electronic universal testing machine (INSTRON 68TM-5). A compression test was performed at a constant loading rate of 0.5 mm/min,^13,14 with cartilage samples compressed up to approximately 100% strain. The stress–strain (σ–ε) curve obtained during this loading was used to calculate the apparent compressive modulus (E, in MPa) by linear fitting of the linear region of the curve.

Micro-Computed Tomography, 3D Simulation, and Analysis

Mouse knee joints were dissected, fixed overnight in 70% ethanol, and subjected to high-resolution micro-CT analysis (μCT, SCANCO; BRÜTTISELLEN, SWITZERLAND). Image reconstruction and analysis were performed using AVIZO software. The scanned image stack was normalized using the median filter module. The tibia’s subchondral bone model was constructed from the thresholded tibia data and presented in the axis ortho view. Morphometry module analysis of the region-of-interest (ROI) bone area included parameters such as average trabecula thickness and separation.

Finite Element Analysis (FEA)

COMSOL software was employed to construct a meshed model of tibias obtained through μCT data for FEA. A uniform continuous pressure was applied to the tibia’s articular surface to assess the von Mises stress distribution. The subchondral bone was treated as a linear elastic material with a Poisson’s ratio of 0.47 and Young’s modulus of 6 MPa. ¹⁵ The force was applied along the Z-axis, with detailed formulas and algorithms available in our previous study. ¹⁶

A force-controlled loading approach was used. Femoral cartilage was first brought into contact with tibial cartilage or the meniscus. Stress and strain distributions at an applied load of ~0.3 N ¹⁷ were selected for analysis. The distal tibia was fully fixed, and the superior surface of the femur was dynamically coupled to a reference point to ensure accurate load transfer. Ligaments and other soft tissues were not modeled, assuming they maintain joint stability. Cartilage contact was defined as frictionless hard contact, with perfect bonding assumed between the meniscus and tibial cartilage.

Ethical Statement

Ethical approval was obtained from the ethics committee (approval number: hrbmuecdc20210310), and a waiver of consent was approved by the institution.

Statistics

All assays were performed in triplicate. Data were presented as mean ± standard deviation Student t test, multifactorial ANOVA, and nonparametric tests were employed for group comparisons. Post-hoc tests were conducted to analyze individual differences in ANOVA. Simple linear regression is used to calculate the elastic modulus of articular cartilage in healthy individuals and osteoarthritis patients. Pearson correlation coefficients were used to analyze the correlation between apparent compressive modulus and the parameters of the subchondral bone trabeculae. Statistical significance was set at P < 0.05, and all tests were performed using SPSS statistics software (version 25.0).

Results

Subchondral Bone and Cartilage Structure as a Mechanical Coupling Unit

The physical properties of articular cartilage and its underlying bone are believed to play a crucial role in regulating the mechanical stress distribution across the joint surface. Given the well-documented cartilage degeneration and subchondral bone (SB) remodeling in osteoarthritis (OA) patients, tibial plateau samples from OA patients and healthy individuals were selected for comparative analysis. To assess the relative mechanical properties, vertical compression was applied at a constant loading rate and the resulting stress–strain curves were analyzed to estimate the apparent compressive modulus (E). Under compressive stress, healthy cartilage exhibited a strain of approximately 70%, compared to only 50% in OA cartilage (Fig. 1A, B), suggesting a higher compressive compliance in healthy cartilage. In contrast, OA cartilage exhibited reduced load-bearing performance, with lower maximum stress and strain values (Fig. 1C, D). The apparent compressive modulus was relatively higher in healthy cartilage, suggesting enhanced stiffness and resilience ( Fig. 1E ).

Figure 1.

Mechanical coupling unit was built by SB and AC. (A, B) The stress-strain curve of tibial plateau articular cartilage and the linear regression fitting line, where the slope of the fitting line represented the elastic modulus. (A) Ctr sample (n = 5). (B) OA sample (n = 13). (C-E) Comparisons of the physical properties of tibial plateau articular cartilage between the OA group (n = 13) and the Ctr (n = 5) group. Student t test was used, and value was presented as mean ± SD, with CI = 95%. (C) Max-strain. (D) Max-stress. (E) Elasticity modulus. (F) Two-dimensional reconstructed image of the tibial plateau subchondral bone. (G-I) Quantitative analysis of subchondral bone structure in CT reconstruction. Student t test was used and data was operated as mean ± SD, with CI = 95% and n = 5-9. (G) TbSp. (H) TbTh. (I) BVF. (J-L) Pearson correlation analysis of the mechanical properties of articular cartilage and structural abnormalities in the subchondral bone; n = 13. (J) Pearson correlation analysis between TbSp and E. (K) Pearson correlation analysis between TbTh and E. (L) Pearson correlation analysis between the BVF and E.

Using CT-based structural analysis, significant alterations were observed in the SB trabecular structure of OA group ( Fig. 1F ), accompanied by an increase in trabecular separation (TbSp) and a decrease in trabecular thickness (TbTh) (Fig. 1G, H). However, no significant difference in bone volume fraction (BVF) was observed between the two groups ( Fig. 1I ). It was found that the greater the TbSp in the subchondral bone, the smaller the apparent compressive modulus (E) of the overlying cartilage, showing a negative correlation ( Fig. 1J ). Conversely, increased TbTh and BVF in the subchondral bone were associated with improved apparent stiffness of the overlying cartilage, as inferred from structural correlation analysis (Fig. 1K, L).

MiRNA-140 Expression and Alterations in Subchondral Bone Structure

To examine the association between miRNA-140-5p expression and subchondral bone microarchitecture, FISH analysis was performed on tibial plateau cartilage from healthy donors and OA patients. The results showed that miRNA-140-5p expression was diminished in OA cartilage compared to controls (Fig. 2A, B). CT imaging displayed a rough tibia surface and chaotic subchondral bone structure in OA patients ( Fig. 1F ).

Figure 2.

SB alteration and miRNA-140 expression in three regions of tibial plateau. (A) Fluorescent in situ hybridization of miRNA140-5p (green) in tibial plateau articular cartilage (×200). Nuclei were labeled with DAPI (blue). (B, C) Quantitative analysis of normalized fluorescence intensity of miRNA-140-5p expression within human and mouse tibial plateau articular cartilage. (B) Human. Student t test was used, and value was presented as mean ± SD with n = 5-13. *P < 0.001. (C) Mouse. ANOVA was used, and post-hoc tests were conducted to analyze individual differences in ANOVA. Value was presented as mean ± SD, with n = 5-6. *P < 0.001. (D) Tibia platform pathological slice location schematic diagram. 1 represents the lateral condyle of the tibial plateau, 2 represents the central region of the tibial plateau, and 3 represents the medial condyle of the tibial plateau. (E, F) Micro-CT two-dimensional reconstruction image (the first column) and FEA analysis image (the second column) of the subchondral bone in the tibial plateau of the WT-group mouse in the sagittal plane. The three numbers correspond to the positions shown in Figure 2D. (G) Fluorescence in situ hybridization images of miRNA-140-5p (green) in different regions of the tibial plateau cartilage in WT–ACLTmouse knee joints (×400). Nuclei were labeled with DAPI (blue). The three numbers correspond to the positions shown in Figure 2D.

To reduce potential confounding factors and enable region-specific analysis under controlled conditions, we examined tibial plateaus in mice subjected to ACLT or sham surgery. The plateau was divided into exterior, middle, and medial regions ( Fig. 2D ), and structural characteristics of each region were independently evaluated. Substantial structural differences were observed in the exterior regions between ACLT and sham-operated mice, whereas no notable differences were detected in the middle region (Fig. 2E, F). FISH results indicated a substantial decrease in miRNA-140-5p expression in the exterior and medial condyle areas, while levels remained high in the middle region ( Fig. 2G ).

MiRNA-140-5p’s Role in Mechanical Force Distribution on Articular Cartilage

To explore the relationship between articular surface mechanical distribution and miRNA-140-5p expression, finite element analysis (FEA) was performed. The result demonstrated an inverse correlation between miRNA-140-5p expression and mechanical stress distribution in ACLT-mice. A three-dimensional model of stress distribution was simulated based on subchondral bone structure. ACLT-operated mice exhibited rough, porous subchondral bone with elevated von Mises stress, particularly in the exterior region and medial condyle of the plateau, compared to sham-operated mice (Fig. 2E, F). Correspondingly, miRNA-140-5p expression was reduced in these high-stress regions.

To verify whether mechanical force directly reduces miRNA-140-5p expression, chondrocytes were subjected to sustained mechanical pressure, and miRNA-140-5p levels were quantified by RT-qPCR. The expression of miRNA-140-5p in chondrocytes subjected to mechanical loads of 1 and 1.925 g/cm² initially increased after 3 hours, then decreased below baseline over the following 5 hours, and remained stable for up to48 hours. In contrast, exposure to a higher mechanical load of 2.425 g/cm² resulted in a continuous decline in miRNA-140-5p expression, suggesting that excessive mechanical stress suppresses miRNA-140-5p levels ( Fig. 5E-I ).

Figure 5.

Regulation of miRNA-140 by mechanical stress and the regulatory role of miRNA-140-5p on RhoA under stress. (A) The immunohistochemistry image revealed the expression levels of RhoA in articular cartilage. (B) quantitative analysis of AOD of RhoA expression to total interested area in AC of human knee joint specimens. Student t test was used and value was presented as mean ± SD, (Ctr, n = 4, OA, n = 9). *P < 0.001. (C, D) The regulatory effect of miRNA-140-5p on RhoA expression under mechanical stress. (C) Western blot of RhoA protein in primary chondrocytes subjected to hydrostatic pressure for 0, 3, 48 hours with or without miRNA-140-5p agomir (represented by miR-140+ in the image). (D). Quantitative analysis of average gray values for RhoA Western blot. ANOVA was used, and post-hoc tests were conducted to analyze individual differences in ANOVA. Value was presented as mean ± SD, with n = 3. *P < 0.001. (E-I) The expression of miRNA-140-5p under different pressures for varying durations of time using RT-qPCR. (E) Line graph illustrating the expression of miRNA-140-5p under different pressures for varying durations of time using RT-qPCR (n = 3). (F-I) Histogram illustrating quantitative analysis of the miRNA-140-5p expression by RT-qPCR. Nonparametric tests were employed for group comparisons. Value was presented as mean ± SD, with n = 3. *P < 0.001.

To further investigate miRNA-140-5p’s role in mechanical loading of the tibial plateau articular cartilage, FEA was performed on miRNA-140-5p knocked-down mice (MUT). Compared to wild-type-SHAM (WT-SHAM) mice, MUT-SHAM mice showed uniformly increased mechanical load across the articular cartilage surface. In MUT-ACLT mice, mechanical load was further elevated in the lateral and middle condyles, leading to structural damage of the subchondral bone with numerous penetrating pores ( Fig. 3A ). Analysis of Tb Sp and Tb.Th revealed a sequential increase in Tb.Sp from WT-SHAM to MUT-ACLT groups, while Tb.Th was significantly lower in MUT mice compared to WT mice ( Fig. 3B ). These findings indicate that miRNA-140-5p knockdown results in increased mechanical loading and subsequent structural damage to the subchondral bone.

Figure 3.

SB structure reshaping and redistribution of mechanical loads on the upper cartilage. (A) The three-dimensional reconstruction image of four groups’ mouse tibia plateau in the horizontal plane using micro-CT (the first row). FEA of mechanical load distribution on the articular cartilage surface of four groups’ mouse tibia plateau: Horizontal plane mapping (the second row). 1 represents the lateral condyle of the tibial plateau, 2 represents the central region of the tibial plateau, and 3 represents the medial condyle of the tibial plateau. (B) Quantitative analysis of the subchondral bone structure of mouse tibia plateau beneath the articular cartilage through micro-CT reconstruction. TbSp. TbTh. ANOVA was used, and post-hoc tests were conducted to analyze individual differences in ANOVA. Data was operated as mean ± SD, with n = 5-6. *P < 0.001.

MiRNA-140-5p Influences RhoA Expression Under Mechanical Stress

To investigate miRNA-140-5p’s regulatory role in the mechanical coupling unit, immunohistochemistry was used to assess RhoA expression in various regions of articular cartilage in different experimental mouse groups. In ACLT-treated mice, the average optical density (AOD) of RhoA staining was significantly higher in the exterior and medial condyle regions compared to the middle region, while no significant changes were observed in sham-operated mice (Fig. 4A, C). In MUT mice, RhoA-positive cells increased across the articular cartilage regardless of the surgical procedure. RhoA expression exhibited a pattern similar to that in WT group, with the highest expression level observed in the lateral and middle condyles (Fig. 4A, C). This suggests that miRNA-140-5p may act as a protective factor by inhibiting RhoA’s role in regulating knee joint mechanical homeostasis. Further analysis in human samples showed elevated RhoA AOD in the OA group compared with controls. To verify the protective role of miRNA-140-5p, overexpression of miRNA-140-5p was performed under pressure conditions. Western blotting showed that miRNA-140-5p overexpression via miRNA-140-5p agomir decreased RhoA expression in chondrocytes exposed to mechanical stress for 48 hours compared to those without miRNA-140-5p agomir (Fig. 5C, D). These results suggest that miRNA-140-5p can inhibit RhoA expression under mechanical stress.

Figure 4.

RhoA was regulated by miRNA-140 under mechanical stress. (A) Immunohistochemical localization of RhoA expression in the lateral condyle, mid-region, and medial condyle of mouse tibial plateau articular cartilage. 1 represents the lateral condyle of the tibial plateau, 2 represents the central region of the tibial plateau, and 3 represents the medial condyle of the tibial plateau. Immunohistochemistry images in the first row and the third row: (×200). Immunohistochemistry images in the second row and the fourth row: (×400). (B, C) Quantitative analysis of AOD of RhoA expression to total interested area within mouse tibial plateau articular cartilage. The numbers on the x-axis represent the same meaning as depicted in Figure 4A. ANOVA was used, and post-hoc tests were conducted to analyze individual differences in ANOVA. Value was presented as mean ± SD, *P < 0.001. (B) Comparison of RhoA expression in the articular cartilage of four groups of mice (n = 9). (C) Comparison of RhoA expression in different locations of the articular cartilage of each group of mice (n = 5-6).

MiRNA-140-5p Mitigates Cytoskeletal Remodeling Under Mechanical Stress

In our previous investigations, miRNA-140-5p expression increased modestly after 3 hours of mechanical force. However, at 1.925 g/cm² for 24 and 48 hours, miRNA-140-5p levels declined sharply ( Fig. 5G ). Interestingly, RhoA exhibited an opposite trend after 48 hours of loading (Fig. 5C, D), suggesting a possible regulatory relationship with miRNA-140-5p. RhoA regulates the dynamic assembly of actin, particularly in response to mechanical force, influencing cellular stiffness and OA development.^18,19 Given RhoA’s role in actin regulation, we examined F-actin organization under mechanical stress to assess potential downstream cytoskeletal effects ( Fig. 6A ). Prior to mechanical stimulation, chondrocytes displayed a rounded or polygonal morphology with loose intercellular organization ( Fig. 6B ). After 48 hours of hydrostatic pressure, F-actin polymerization was markedly enhanced, leading to dense and organized dendritic structures ( Fig. 6A ). These findings suggest that mechanical loading promotes cytoskeletal remodeling and contributes to changes in chondrocyte morphology. However, chondrocytes overexpressing miRNA-140-5p under identical conditions retained a more rounded morphology compared to those exposed to mechanical loading alone for 48 hours (P < 0.05, Fig. 6B ). This suggests that miRNA-140-5p may counteract the cytoskeletal reorganization and associated morphological changes induced by mechanical stress, which are commonly associated with increased cellular stiffness.

Figure 6.

MiRNA-140-5p maintain the elasticity of chondrocytes and cartilage tissue. (A) Immunofluorescence staining image of F-actin with or without transfection under mechanical pressure with 1.925 g/cm² (×200). Stress+ means chondrocytes cultured under 1.925 g/cm²; stress- represents cells cultured without mechanical load; MiRNA-140+ represents cells cultured with miRNA-140 agomir transfection. (B). Quantitative analysis the influence of miRNA-140-5p overexpression on chondrocyte cytoskeleton circularity and roundness under pressure. ANOVA was used, and post-hoc tests were conducted to analyze individual differences in ANOVA. Value was presented as mean ± SD, with n = 3, *P < 0.001. (C-F) MiRNA-140-5p regulation of COL-I and COL-II expression under mechanical stress. (C) Western blot analysis of COL-I and COL-II protein abundance. (D, E) Quantitative analysis of average gray values for COL-I and COL-II Western blot. ANOVA was used, and post-hoc tests were conducted to analyze individual differences in ANOVA. Value was presented as mean ± SD, with n = 3. (F) Quantitative analysis of the ratio of the average gray values for COL-I and COL-II Western blot. ANOVA was used, and post-hoc tests were conducted to analyze individual differences in ANOVA. Value was presented as mean ± SD, with n = 3. *P < 0.001.

MiRNA-140-5p Regulates Collagen Composition Related to Cartilage Mechanical Properties Under Mechanical Pressure

The mechanical properties of cartilage tissue are closely related to its collagen composition. ²⁰ To explore the relationship between miRNA-140-5p and cartilage matrix composition under mechanical loading, miRNA-140-5p were overexpressed and subsequent changes in collagen ratios were evaluated. Mechanical pressure induced an increase in Type I collagen (COL-I) secretion by chondrocytes, while COL-II secretion decreased significantly only after 48 hours of loading (Fig. 6C, D, E). Overexpression of miRNA-140-5p suppressed the secretion of COL-I and restored COL-II secretion (Fig. 6C, D, E). The average protein expression ratios of COL-II/COL-I in the KB, 3-h, 48-h, 140agomir + 3-h, and 140agomir + 48-h groups were 2.03, 1.46, 0.87, 1.67, and 1.60, respectively ( Fig. 6F ).

Discussion

This study primarily investigates the interaction between miRNA-140-5p and mechanical forces in articular cartilage. MiRNA-140-5p emerges as a pivotal player in the delicate balance governing chondrocyte functionality, development, and overall homeostasis. However, its role under abnormal mechanical stress conditions remains unclear.

Articular cartilage and subchondral bone together constitute a complex and integrated functional unit essential for joint health and biomechanics. The articular cartilage and subchondral bone, intimately connected through osteochondral junctions, are often studied as a cohesive biomechanical unit. ²¹ The shared vascular system and studies utilizing sodium fluorescein highlight the biological crosstalk between the articular cartilage and subchondral bone.^{22 -24} Notably, the subchondral bone, characterized by its greater stiffness and strength, serves as the primary load-bearing structure for force transmission and shock absorption. Given that OA patients exhibit altered joint structures, including subchondral bone sclerosis and changes in cartilage composition, we selected OA samples to model mechanically unbalanced joints. Considering the large variation in loading rates reported in literature and since our focus was on comparing physicochemical properties rather than replicating exact physiological loading, we selected a loading rate between static and slow dynamic conditions.^13,14 Our findings suggest that cartilage underlying bone with a high proportion of bone content and thick, dense trabecular bone often accompanies more elastic articular cartilage. This high resilience enables articular cartilage to disperse mechanical load impacts over a larger range.^{25 -27} Subsequently, using mouse knee joint instability model from different groups, we simulated alterations in surface load and examined corresponding changes in subchondral bone structure and load distribution. Analysis of miRNA-140-5p expression in different regions of the mouse knee joint in ACLT and SHAM-group revealed heterogeneous distribution both within individual joints and between experimental groups. In miRNA-140-5p knockdown mice, a significant increase in articular surface mechanical load was observed, accompanied by elevated RhoA expression. The results suggest that a high mechanical load environment may suppress miRNA-140-5p expression, whereas miRNA-140-5p, in turn, helps attenuate mechanical load transmission or bearing. This observation is consistent with previous studies suggesting that miRNA-140-5p contributes to the maintenance of cartilage homeostasis. ¹⁰ However, its potential involvement in OA progression warrants further investigation.

Further cellular experiments confirmed that mechanical force reduces miRNA-140-5p expression. We found that miRNA-140-5p expression exhibited dynamic changes under mechanical stress, with a modest increase at the early stage (3 hours) followed by a significant decline under prolonged high mechanical pressure (24 and 48 hours). This suggests that miRNA-140-5p may be involved in the adaptive regulation of chondrocytes in response to mechanical stimuli. Overexpression of miRNA-140-5p significantly reduced RhoA expression and ameliorates mechanical pressure-induced changes in chondrocyte morphology. Changes in cell morphology reflect alterations in cellular physicochemical properties, particularly those related to cytoskeletal remodeling. ²⁸ RhoA, a key regulatory protein of the cellular cytoskeleton, is particularly sensitive to mechanical stimuli. ²⁹ It promotes stress fiber formation through its effectors, Rho-associated coiled-coil kinase (ROCK) and Mammalian diaphanous-related formin-1 (Mdia-1)^30,31 and phosphorylates and activates LIM Domain Kinase-1 (LIMK-1) and LIM Domain Kinase-2 (LIMK-2) to inhibit actin-depolymerizing factor (ADF) and actin filament-mediated actin depolymerization, ³² thereby stabilizing the stress fibers. It modifies cell stiffness in response to the external mechanical cues by regulating the arrangement and quantity of stress fibers. In animal studies, we observed a significant increase in the expression of RhoA with increasing stress on the joint surface. Moreover, in the absence of the inhibitory effect exerted by miRNA-140-5p, its expression was further upregulated. These results collectively suggest that mechanical loading alters the biophysical properties of chondrocytes, and that miRNA-140-5p plays an essential role in maintaining these properties under stress. Although we observed low expression levels of RhoA in healthy mouse knee joints, our in vitro experiments revealed that RhoA knockout resulted in complete chondrocyte death after 48 hours of loading (Supplementary Fig. S2F).

The elasticity parameter (E) of cartilage tissue is primarily determined by the collagen content. ²⁰ Among these, COL-II plays a crucial role in the elasticity and toughness of cartilage tissue, effectively dispersing and absorbing pressure to protect the joint surface from damage. The stiffness of cartilage tissue is determined by COL-II and with a minor contribution from COL-I. However, the presence of COL-I increases the stiffness of cartilage tissue. ³³ We examined the effect of mechanical pressure on collagen secretion by chondrocytes and observed an increase in COL-I alongside a decrease in COL-II, resulting in a reduced COL-II/COL-I ratio, indicative of altered cartilage property and matrix composition. Although overexpression of miRNA-140-5p did not fully restore collagen ratios, it significantly ameliorated these alterations.

However, due to the scarcityof the human samples, and the limited availability of transgenic mice, as well as their high cost, and the fact that many mice could not tolerate anesthesia and surgery, the number of effective human and animals’ samples was small. Additionally, the limited number of valid sections obtainable from serial slicing further constrained the sample size. Consequently, although the findings on mechanical coupling units were statistically significant, their generalizability may still be limited by the small sample size. Second, the limited sample size also hindered full control for potential confounding factors such as age, sex, and body weight, precluding the construction of complex models for further analysis. In our dataset, the healthy group included one 50-year-old donor and the OA group included one 59-year-old patient, while all other samples were from individuals over 60 years of age. Fisher’s exact tests for age, sex, and body weight all yielded P = 1, suggesting no statistically significant differences between groups; however, due to the small sample size, potential confounding cannot be entirely ruled out. To ensure robustness, we conducted both Pearson and Spearman correlation analyses, and the results were consistent. Third, although correlations between miR-140-5p expression and mechanical properties were observed, the study did not investigate the mechanotransduction signaling pathways involving miR-140-5p, limiting a deeper mechanistic understanding of its role in joint tissue remodeling. Fourth, given the inherent complexity and abstraction of mechanical properties and force transmission in human joints, we adopted a simplified finite element model. This simplification was necessary to ensure computational feasibility and to focus on the relative distribution of stress and strain within cartilage. We acknowledge that these assumptions may underestimate local shear forces and overestimate joint displacement; however, the relative mechanical differences between regions, which serve as the basis for correlation analysis with miR-140-5p expression, remain valid. Therefore, although the simplified model does not fully recapitulate the multifaceted mechanical environment of human cartilage, it provides a reasonable framework for exploring the mechanoregulation of miRNA expression under compressive loading.

In summary, we propose that the expression pattern of miRNA-140-5p exhibits a dynamic adaptive relationship with the mechanical loading experienced by articular cartilage. Mechanical stimulation induces a stage-specific alteration in miRNA-140-5p expression, which in turn modulates the morphology of chondrocytes and the composition of the extracellular matrix, ultimately influencing their physical and mechanical properties. These results highlight its potential role of miRNA-140-5p in cartilage mechanoregulation and provide new insights into cartilage biology under mechanical stress.