β2-Adrenergic receptor expression in subchondral bone of patients with varus knee osteoarthritis

Abstract An important causative factor in osteoarthritis (OA) is the abnormal mechanical stress-induced bone remodeling of the subchondral bone. β2-adrenergic receptor (Adrb2) plays a major role in mechanical stresses that induce bone remodeling. The medial tibial plateau (MTP) and lateral tibial plateau (LTP) of patients with varus Knee osteoarthritis (KO) bear different mechanical stresses. The present study aimed to investigate the expression of Adrb2 in medial tibial plateau subchondral bone (MTPSB) and lateral tibial plateau subchondral bone (LTPSB) in patients with varus KO. A total of 30 tibial plateau samples from patients undergoing total knee arthroplasty for varus KO and MTPSB and LTPSB were studied. Statistical analysis was performed using paired sample t-tests. Safranin O-Fast Green staining and Micro-computed tomography showed significant differences in the bone structure between MTPSB and LTPSB. Tartrate-resistant acid phosphatase (TRAP)-positive cell density in MTPSB was higher than that in LTPSB. Immunohistochemistry, reverse transcription-quantitative polymerase chain reaction, and Western blot analysis revealed that compared to LTPSB, the levels of Adrb2, tyrosine hydroxylase (TH), and osteocalcin increased significantly in MTPSB. Double-labeling immunofluorescence showed Adrb2 was present in the majority of TRAP-positive multinuclear cells of the MTPSB. The expression of Adrb2 and TH was significantly higher in MTPSB than in LTPSB, confirming the involvement of these molecules in the development of OA.


Introduction
Osteoarthritis (OA) is the most common degenerative joint disease, affecting about 250 million people worldwide. It mainly affects weight-bearing joints, such as knee and hip, and is characterized by progressive cartilage degeneration and subchondral bone changes [1]. Abnormal subchondral bone remodeling plays a major role in the pathogenesis of OA. Subchondral bone is the mechanical support for articular cartilage and undergoes bone resorption and remodeling in response to changes in mechanical stresses [2].
Mammalian bones exhibit a sympathetic nervous system (SNS) that is regulated by autonomic and sensory nerves in response to mechanical stresses [3,4]. During normal bone remodeling, sympathetic nerves effectuate via catecholamine, a critical sympathetic nerve neurotransmitter norepinephrine that activates the β2-adrenergic receptor (Adrb2), which in turn inhibits osteoblast proliferation and differentiation and promotes osteoclast precursor maturation and bone resorption activity. In this process, tyrosine hydroxylase (TH) acts as the rate-limiting enzyme for norepinephrine biosynthesis [5,6]. In the orthodontic tooth movement (OTM), mechanical stresses induced Adrb2 activation in alveolar bone remodeling [7]. Sympathetic nerve fibers are detected in the subchondral bone of OA and in a rat temporomandibular experimental OA model [8]. Adrb2 antagonists inhibit subchondral bone loss and osteoclast activity, while Adrb2 agonists aggravate these reactions, and Adrb2 plays a critical role in mechanical stress-induced bone remodeling [9].
When normal walking on a healthy knee, the medial compartment experiences 60-80% weight bearing, and the lesions in the medial compartment of the knee are common due to significant stress and knee adductive torque during weight-bearing activities [10,11]. In the early stage of knee osteoarthritis (KO), mechanical stresses cause changes in the subchondral bone, and the progression of OA increases the varus deformity of the knee [12,13]. This phenomenon increases the mechanical stresses through the medial compartment, further aggravating the degree of OA and varus deformity [14]. However, the expression of Adrb2 in the subchondral bone of human KO has not yet been reported.
The present study selected the same patients with KO varus deformity of the medial tibial plateau (MTP) and the lateral tibia plateau (LTP) and conducted a paired comparison of the subchondral bone structure change to assess the differences in Adrb2 expression. Also, Adrb2 with KO was investigated with respect to the underlying mechanism to provide a reference for the treatment of KO.

Sample collection
Specimens of the tibial plateau were obtained from patients with primary varus deformity KO who underwent total knee arthroplasty in the Department of Orthopedics, Ningxia Medical University General Hospital, Yinchuan, China, from 2019 to 2020. The cohort consisted of 30 cases, including 12 males and 18 females, aged 62-78 (mean age, 66.77 ± 3.48 years) years. The diagnosis of OA was based on the criteria of the American College of Rheumatology [15]. The patients with OA secondary to other diseases, such as trauma and connective tissue disease, were excluded and had no history of β-adrenergic receptor agonist and antagonist drug intake. The preoperative knee X-ray was taken. The measured varus angle was 12.0-25.5°(average: 17.6°). Most of the lesions occurred in the medial compartment ( Figure 1a). The severity of OA knee was assessed using Kellgren-Lawrence (K-L) scale (1-4) on weight-bearing radiographs [16]. Of 30 knees, 12 knees were K-L grade 3 and 18 knees were K-L grade 4.
Ethical approval and informed consent: This study was approved by the Ethics Committee of Ningxia Medical University General Hospital (No. 2020-985), and informed consent was obtained from each patient prior to inclusion in the study.

Sample processing
The tibial plateau material was obtained during surgery and placed on a clean table. Then, the bone and cartilage in the central bearing area of the MTP (n = 30) and LTP (n = 30) samples from each patient were selected and trimmed into blocks (2.0 cm 3 × 2.0 cm 3 × 1.0 cm 3 ). The coronal sections were used to divide the bone and cartilage into two parts randomly. The part of the samples that contained cartilage and subchondral bone was immobilized in 4% paraformaldehyde for 48 h. After micro-computed tomography (CT) analysis, the samples were decalcified with 10% ethylenediaminetetraacetic acid at room temperature for 3 weeks and embedded in paraffin. Safranin O-fast green staining, tartrate-resistant acid phosphatase (TRAP) staining, immunohistochemistry, and immunofluorescence were performed in 4-µm-thick discontinuous sections. The other part removes cartilage, and the subchondral bone was wrapped in foil and stored at −80°C to measure the mRNA and protein levels subsequently.

Safranin O-fast green staining
The processes were described previously [17]. The sections were stained with hematoxylin (ZLI-9610, ZSGB-Bio, Beijing, China) for 3-5 min and washed with tap water for 10 min. The specimens were dyed with 0.3% solid green (F8130, Solarbio, Beijing, China) for 5 min and quickly washed with the weak acid solution for 10-15 s to remove the residual solid green. The slices were added into 1% Safranin staining solution (S8020, Solarbio, Beijing, China), soaked for 5 min, then rinsed three times with distilled water for 5 min, and dehydrated with 95% ethanol and absolute ethanol. Xylene was transparent and sealed with optical resin. The samples were examined using a microscope. The Image-Pro Plus 6.0 image analysis software (Media Cybernetics, Bethesda, MD, USA) was employed to assess the structural parameters of the cartilage and subchondral bone deduced with morphometric methods. The parameters included total articular cartilage (TAC; the perpendicular distance between the cartilage surface and cement line), subchondral bone plate (SCP) thickness (the distance from the cement line to the interface between the SCP and trabecular bone), and trabecular bone area/total area (BA/TA), which are calculated as follows: (trabecular bone area in the region of measurement)/ (trabecular bone area + marrow cavity area) × 100.
Each section was evaluated using the Osteoarthritis Research Society International (OARSI) score [18]. The final score was based on the Safranin O/fast green staining results and the OARSI scoring items. The OARIS score was obtained by multiplying the histological grading of cartilage degeneration (in six grades) by the histological staging of cartilage degeneration (in five stages) following Safranin O/fast green staining.

TRAP staining
The paraffin slices were immersed in xylene for 5 min each to remove paraffin and then re-hydrated in an alcohol series and distilled water for 5 min. Tissue sections were pre-incubated at 37°C for 20 min in TRAP buffer (0.1 M acetate buffer pH 5.0 and 50 mM sodium tartrate). The enzyme reactivity was visualized by incubating the sections in TRAP buffer containing 0.1 mg/mL naphthol AS-MX (N-4875, Sigma, USA) and 0.3 mg/mL fast red violet LB salt (F-3381, Sigma, USA) at 37°C for 1 h. The stained sections were rinsed in phosphate-buffered saline (PBS) and counterstained with hematoxylin (ZLI-9610, ZSGB-Bio, Beijing, China). The slices were dehydrated by placing them in 70, 80, 90, and 100% alcohol and xylene for 1 min each, then sealed with neutral glue, and covered with a coverslip. TRAP-positive osteoclasts were counted within 400 µm of the cement line in the osteochondral junction and divided by the length of the subchondral bone to provide an osteoclast density expressed as TRAPpositive cells per mm 2 [19]. One dark purplish or reddish cell with ≥3 nuclei was counted as one osteoclast.

Micro-CT evaluation
The LTP and MTP samples were observed under a micro-CT scanner (SkyScan 1076, Bruker, Kontich, Belgium) with the following scanning parameters: 18 μm isotropic voxel size, 55 kV voltage, 109 μA current, 200 ms integration time, and 4,000 projections. The microstructure of After standing, the mechanical shaft moves inward, and the stress concentrates on the inside of the knee joint; local magnification of the X-ray film showed that the medial joint space was significantly smaller than the lateral joint space, and the local osteosclerosis on the medial side of the joint was significantly greater than that on the lateral side. (b) Intraoperative gross pathological images: knee medial tibial plateau than the lateral cartilage defect is severe, and some bone is exposed, corresponding to the medial femoral condyle the same performance. The cartilage of the lateral femoral condyle of the knee joint is relatively complete with good gloss. (c) Safranin O/Fast Green staining. (d) OARSI score. Statistical analysis results: The difference between the MTP and LTP groups was statistically significant. Magnification ×50, scale bar 500 µm; magnification ×200, scale bar 100 µm; **** P < 0.0001. the subchondral plate and trabecular bone was visible in the two-dimensional (2D) reconstructed images that were converted into discrete binary objects by the global thresholding and binarization procedures combined with image filtering such as despeckling to remove noise. Subsequently, the data were processed for three-dimensional (3D) measurement in SkyScan CTAn software. The surface-rendered 3D models were created for visualization using the SkyScan CTVol software [20]. For subchondral trabecular bone, a cubic region of interest of 10 mm 3 × 10 mm 3 × 2 mm 3 was selected, and the parameters such as bone volume/total volume (BV/TV), bone mineral density (BMD), trabecular thickness/number/separation (Tb.Th, Tb.N, Tb.Sp), and structure model index (SMI) were calculated.

Immunohistochemistry staining
The paraffin sections were incubated at 65°C overnight, dewaxed by xylene, and rehydrated, followed by antigen retrieval with 0.1% trypsin for 20 min at 37°C. Subsequently, the sections were incubated with 3% hydrogen peroxide (H 2 O 2 ) for 10 min to remove endogenous peroxidase and blocked with Goat serum for 20 min at 37°C. Next, the slides were incubated with primary antibodies, including rabbit anti-Adrb2 (13096-1, 1:300, Proteintech, Wuhan, China), rabbit anti-TH (25859-1-AP, 1:500, Proteintech, Wuhan, China), and rabbit anti-osteocalcin (OCN; 23418-1-AP, 1:300, Proteintech, Wuhan, China) overnight at 4°C. The following day, the sections were treated with horseradish peroxidase-conjugated streptavidin detection system (PV-9001, ZSGB-Bio, Beijing, China) to detect the immunoreactivity. The sections were visualized using 3,3′-diaminobenzidine (ZLI-9018, ZSGB-Bio, Beijing, China), while hematoxylin (ZLI-9610, ZSGB-Bio, Beijing, China) was used for counterstaining. Routine dehydration and mounting were performed. After image capture, the number of positively stained cells was quantified using Image-Pro Plus 6.0 (Media Cybernetics, Inc.) as described previously [21,22]. Then, the histological score was calculated as the average optical density of positively stained cells, which equals the ratio of the overall optical density of the positive cells to the positive area. Positive cells are those with clear brownishyellow granules in the cytoplasm or nucleus.

Immunofluorescence
The procedure for the first day of the immunofluorescence protocol was the same as that used for immunohistochemistry (described above). The following day, the sections were incubated with goat anti-rabbit IgG-TRITC-conjugated secondary antibody (SA00007-2, 1:500, Proteintech, Wuhan, China) for 1 h at 37°C in the dark. To identify the cell morphology of the nuclei, the sections were counterstained with 4′6diamidino-2-phenylindole (DAPI; Solarbio, Beijing, China) for 3 min followed by PBS washes.
For double-labeling immunofluorescence, primary antibodies of Adrb2 were used as described earlier. To detect osteoclast cell in the tibial plateau subchondral bone, anti-TRAP antibody (D-3) (1:300, sc-376875, Santa Cruz, Texas, USA) was used as a primary antibody. The mixture of rabbit anti-Adrb2 antibody and mouse anti-TRAP antibody was incubated overnight at 4°C, washed in PBS, and incubated with goat anti-rabbit IgG-TRITC-conjugated secondary antibody (SA00007-2, 1:500, Proteintech, Wuhan, China) and goat anti-mouse IgG-fluorescein isothiocyanate (FITC)-conjugated secondary antibody (SA00003-1, 1:500, Proteintech, Wuhan, China) for 1 h at 37°C in the dark. The sections were then incubated with DAPI and then stimulated under a fluorescent microscope.
All sections were mounted in a fluorescence microscope (DM4B, Leica Microsystems GmbH) for examination. For immunofluorescence analysis, in a place with high light intensity, under a 200× magnification, the optical density value was analyzed under the same area and was equal to the ratio of the overall optical density to the area.
The immunoreactive bands were visualized by electrochemiluminescence on an ALS4000 gel image analysis system (GE Healthcare Life Sciences, Logan, UT, USA). Image J software (NIH, Bethesda, MD, USA) was used to quantify the protein band signal intensity and expressed as arbitrary units (a.u). The optical density was normalized to that of GAPDH to calculate the levels of the target protein.

Reverse transcription-quantitative PCR
The medial and lateral samples of subchondral bone were taken from the tibial plateau of each patient to evaluate the mRNA levels of the target genes. Briefly, the subchondral bone samples were thawed, weighed, and homogenized in liquid nitrogen. Subsequently, total RNA was extracted from the tissues using the RNA pure tissue kit (G3640, Wuhan Servicebio Technology, Wuhan, China), according to the manufacturer's instructions. Total RNA amount and purity were determined by UV spectrophotometry. The primers for Adrb2, TH, and OCN were designed and synthesized by Servicebio Biotech ( Table 1). Total RNA was reverse transcribed into cDNA using the RT First Strand cDNA Synthesis kit (G3331, Wuhan Servicebio Technology, Wuhan, China), according to the manufacturer's instructions. A 25 µL quantitative polymerase chain reaction (qPCR) kit consisted of cDNA (0.5 µL), SYBR Green qPCR Master Mix (G3320, Wuhan Servicebio Technology, Wuhan, China) (12.5 µL), forward and reverse primers (0.5 µL each), and diethylpyrocarbonate-treated water (11 µL). The amplification was carried out at 50°C (20 min), 95°C (10 min), and 40 cycles at 95°C (15 s) and 60°C (60 s). Then, the mRNA level was calculated using the cycle threshold method (2 −ΔΔCt ) and normalized against that of GAPDH that was used as an internal control.

Statistical analysis
Statistical analysis was performed using SPSS version 20.0 software (IBM, Chicago, IL, USA). Graphs were drawn using GraphPad Prism version 8.0 software (GraphPad Software Inc., San Diego, USA). The data were presented as mean ± SD. The clinical, micro-CT, histology, and immunohistochemistry data were tested for normality using the Shapiro-Wilk Test. Paired sample t-test was used to compare the measurements. P < 0.05 was considered statistically significant.

Macroscopic observations
The degeneration of MTP cartilage was more severe than that of LTP cartilage in all specimens. The surface of the MTP cartilage was rough, dull, and grayish-yellow, with extensive softening foci, huge fissures, exposed ivory-like subchondral bone covered by pannus tissue, and multiple osteophytes in the center and edge of the plateau. The surface of the LTP cartilage was flat and locally shiny, with scattered superficial ulcers and softening foci, rare fissure formation, no cartilage loss in the whole layer, no pannus tissue formation, and marginal osteophytes in some specimens (Figure 1b).

Histological assessment
Cartilage damage was investigated by performing Safranin O/fast green staining. The results revealed no notable damage and mild damage in the LTP and severe damage in the MTP. In the MTP, some of the subchondral bone was no longer covered by cartilage, and the subchondral bone was completely exposed. Compared to the MTP, the subchondral trabecular bone of the LTP had thinner and smaller, the chondrocytes of the LTP were relatively normal in size and shape, and the cartilage matrix was stained uniformly. In the MTP sample, the ivory-like subchondral bone, the chondrocytes were abnormal in morphology, swollen, and reduced in number. The cartilage matrix was not stained. Some samples displayed cartilaginous deposits (triangle) among trabecular bones (Figure 1c). Cartilage damage was evaluated using the OARSI score ( Table 2). The OARSI scores in the MTP were higher than that in the LTP. The analysis showed that there were statistically significant differences in OARSI score among groups (P < 0.0001) (Figure 1c). The TAC thickness of the MTP subchondral bone was significantly lower than that of the LTP, while BA/TA and SCP thickness were significantly higher than those of the LTP subchondral bone (P < 0.0001) ( Table 2).

Micro-CT evaluation
The micro-CT 2D images revealed that the MTP subchondral bone had increased bone mass and reduced porosity compared to the LTP subchondral bone. The micro-CT 3D images show significant thickening of the SCP in the MTP compared to the LTP (Figure 2a). The measurement of bone structural parameters showed that BV/TV, Tb.N,  and Tb.Th of the medial subchondral bone were significantly increased compared to those of the LTP subchondral bone, while SMI and Tb.SP were significantly decreased (P < 0.0001; Figure 2b and c).

TRAP staining
TRAP-positive multinucleated osteoclasts are observed at the bone surface of the subchondral bone ( Figure 3). In the MTP, subchondral bone TRAP-positive cells were widely distributed (Figure 3a), mainly distributed in the subchondral bone resorption pit region (Figure 3b). In the lateral tibia platform, subchondral bone TRAP-positive cells were rarely distributed (Figure 3a), mainly distributed in the junction between cartilage and subchondral bone (Figure 3b). The density of osteoclasts in the subchondral bone in the MTP was significantly higher than in the LTP (P < 0.0001; Figure 3c).

Adrb2, OCN, and TH levels detected by immunohistochemistry
Adrb2 is expressed in both osteoblasts and osteoclasts, but mainly in the cell membrane. The MTP subchondral bone Adrb2-positive cells are multinucleated clusters, primarily distributed in the trabecular bone edge and the subchondral bone resorption pits. The LTP subchondral bone Adrb2-positive cells constitute the distribution channel in bone cartilage and subchondral bone edge (Figure 4a). The MTP subchondral bone Adrb2-positive cell density was significantly higher than that of the LTP subchondral bone (P < 0.0001; Figure 4d). The THpositive cells were also partially expressed in cartilage, but mainly distributed in the subchondral bone resorption pits and bone marrow cavities (Figure 4b). The MTP subchondral bone density was significantly higher than TH-positive cells in the LTP subchondral bone (P < 0.001; Figure 4d). OCN is mainly expressed in the cell membrane and intercellular space of osteoblasts, with a scattered gritty distribution (Figure 4c). The LTP subchondral bone OCN-positive cell density was significantly lower than that of the MTP subchondral bone (P < 0.0001; Figure 4d).

Adrb2, OCN, and TH levels detected by immunofluorescence
Immunofluorescence staining was performed using TRITC, which is excited at 550 nm and emits red light, and then images of the samples were captured using a 40X fluorescence microscope. The results of Adrb2 (Figure 5a), TH  (Figure 5d). The differences in Adrb2, TH, and OCN expression were statistically significant between the MTP subchondral bone and LTP subchondral bone by analyzing the mean optical density (P < 0.0001, P < 0.001, and P < 0.0001, respectively).

Evaluation of Adrb2 and TRAP coexpression by double-labeling immunofluorescence
To further observe Adrb2 and TRAP co-expression levels in osteoclasts from the subchondral bone, double-labeling immunofluorescence staining was used. In the doublelabeling immunofluorescence staining, TRAP expression was performed using FITC, which is excited at 500 nm and emits green light, while Adrb2 expression was the same as before, and then images of the samples were captured using a 40X fluorescence microscope. Adrb2 and TRAP were detected in both MTP and LTP subchondral bones. Double-labeling immunofluorescence showed that Adrb2 was present in the majority of TRAP-positive multinuclear cells of the MTP subchondral bone (Figure 6a). By analyzing the mean optical density, the difference in Adrb2 and TRAP co-expression was statistically significant between MTP and LTP subchondral bone (P < 0.0001; Figure 6b).

Adrb2, OCN, and TH protein levels in subchondral bone determined by Western blot analysis
Western blot results showed the protein levels of Adrb2, OCN, and TH in subchondral bone of MTP and LTP by the gray scale of the bands (Figure 7a). The densities of the    bands were numerically quantified and compared. Compared to the values obtained from the LTP subchondral bone samples, the protein levels of Adrb2, OCN, and TH of the MTP subchondral bone samples increased (P < 0.0001; Figure 7b).

Adrb2, OCN, and TH mRNA levels in subchondral bone
Compared to the expression obtained from the LTP subchondral bone samples, the mRNA levels of Adrb2, OCN, and TH of the MTP subchondral bone samples increased (P < 0.0001, P < 0.001, and P < 0.0001, respectively; Figure 7c).

Discussion
In this study, we investigated changes in subchondral bone remodeling, microstructure, and their association with cartilage degradation in KO varus deformity patients. We found that MTP had abnormal bone remodeling and microstructural deterioration and cartilage severe damage. In addition, the level of OCN in the MTP subchondral bone was higher than that in the LTP, which was consistent with the positive correlation between increased subchondral osteogenesis and the progression of OA reported previously [23]. The results of micro-CT also showed significant differences in the bone structure between medial and lateral subchondral bone, which was consistent with the results of the study by Finnilä et al. [24]. These findings indicated that the severe the OA, the obvious the degree of subchondral bone hyperplasia and osteosclerosis.
In this study, we selected the tibial plateau with varus deformity of KO and found that the mechanical stresses of the MTP were higher than that of the LTP, and the lower limb mechanical stresses and mechanical axis of the lower limb were correlated with the local BMD [25]. The adduction moment of the knee is a major determinant of medial to lateral load distribution and also the cause of the biomechanical abnormality of KO in the medial compartment [11,26]. Intriguingly, lower limb dynamic load and tibial KO disease severity have been reported, and adduction torque increase might affect the medial tibial platform structure of OA and cause varus deformity [27]. Varus alignment increases the risk of progression of medial ventricular OA in KO [28]. The knee adductive moment reflects the dynamic load in the medial joint compartment and predicts the progression of radiographic OA [27,29]. Therefore, for patients with varus deformity of KO, medial OA is severe than lateral OA because the mechanical loading on the MTP is greater than that on LTP.
The SNS distributed in the periosteum, bone trabecula, bone marrow, and subchondral bone strongly affects bone remodeling [5,30]. In this study, we detected TH through testing the subchondral bone in MTP and LTP and illustrated that the sympathetic nerve is involved in the subchondral bone of bone remodeling. The previous animal model studies have shown that SNS suppresses bone formation by activating Adrb2 on osteoblasts, directly or indirectly accelerates osteoclast generation, and promotes bone loss [7,9]. The present study demonstrated Adrb2 in the human KO subchondral bone, and as far as we know, it was also the first report. To avoid differences in the results due to individual factors (such as hormone levels, weight, age, and occupation), we took the same knee joint on the medial and lateral sides of the tibial plateau for a paired comparison study. Interestingly, the expression of Adrb2 in the MTP subchondral bone with high bone mass, and obviously, osteosclerosis was higher than that in LTP and was mainly distributed in the subchondral bone absorption socket.
It has been reported that Adrb2 was expressed in both osteoblasts and osteoclasts [31]. To determine whether Adrb2 expression originated in osteoclasts, immunofluorescence and TRAP staining were performed. Immunofluorescence showed that Adrb2 was mainly expressed in multinucleated osteoclasts. TRAP staining showed that the number of osteoclasts in MTP subchondral bone was significantly higher than that of LTP, and they were mainly distributed in the subchondral bone absorption pit, which was similar to the distribution of Adrb2. Our study revealed that TRAP and Adrb2 were co-expressed in multinucleated cells by immunofluorescence double-label detection, and the expression level of MTP was higher than that of LTP. In this study, we did not use in vitro cell cultures for the detection of Adrb2 because Adrb2 serves as hormone receptors, and the properties of these cells change when they are removed from their environment. Thus, we inferred that Adrb2 on MTP with high mechanical stress is mainly expressed on osteoclasts. A comparative study on the medial sclerotic area and the lateral nonsclerotic area of the tibial plateau in patients with KO found that the number and the activity of osteoclasts in the sclerotic area were significantly higher than that in the nonsclerotic area [32]. The increased activity of osteoclasts in the sclerotic area of the bone indicated that the activity of bone formation and resorption were increased in human KO, but no coupling was observed.
Sympathetic mediator catechins induce differentiation of osteoclast precursors and stimulate osteoclast activity by binding to Adrb2 [33]. Since Adrb2 is a hormone receptor, the subchondral bone of the MTP and LTP of the knee of the same patient was used to conduct a paired experiment, suggesting that there were differences in topical catecholamines in the subchondral bone between MTP and LTP. This suggests that mechanical stress is the main cause of upregulation of Adrb2 in the subchondral bone, inducing osteoclast generation and participating in subchondral bone remodeling. Wnt signal promotes the proliferation and differentiation of osteoblasts [34]. Mature osteoclasts can reduce the Wnt signaling pathway inhibitor osteosclerosis protein expression and through the secretion of Wnt10b and BMP6 promoting osteoclast areas of osteoblast differentiation and bone formation [35]. Our research team has previously reported that the severity of KO is negatively correlated with the expression of sclerostin in the subchondral bone [36]. Recent studies have shown that osteoclasts downregulate the expression of sclerostin in the trabecular bone [37], which can explain the fact that the sclerotic MTP has high osteogenic activity and high osteoclast density in this study.
Pain is the main cause of disability in KO patients [38], and subchondral bone is the main cause of KO pain [39,40]. In overweight and obese KO patients, knee pressure and pain can be reduced by weight loss [41]. Similarly, high tibial osteotomy improved the mechanical axis of the lower limbs and relieved the load of the medial compartment of the knee, thus alleviating pain [42,43]. These studies suggest a positive correlation between intraarticular mechanical stresses and pain [44]. In studies on symptomatic KO, it was found that the increased nerve growth factor and osteoclast density in osteochondral channels seem to be key factors leading to KO bone pain [45]. In this study, we selected patients who underwent total knee arthroplasty due to medial knee pain. Therefore, we speculated that the MTP with high mechanical stress might elevate the number of osteoclasts through the high expression of β2-adrenergic receptor in the subchondral bone, leading to joint pain. In another study, β-adrenergic receptor inhibitors reduced the need for opioids in patients with joint pain and OA [46], which indirectly supports our view. However, a further study of the role of mechanical stress through the activation of subchondral bone osteoblasts by Adrb2 to regulate osteoblasts in OA, animal model experiments is still needed.
The current research related to the pathogenesis of KO of the most common animal models includes destabilization of medial meniscus and anterior cruciate ligament transection, etc., which are posttraumatic OA models and cannot objectively reflect abnormal subchondral bone remodeling of OA caused by mechanical stresses. The alveolar bone has a strong remodeling ability after mechanical stimulation, such as bite force, which is considered an ideal model for studying the bone mechanical response. In the OTM, mechanically modulating Adrb2 induces and promotes osteoclast generation and regulates alveolar bone remodeling [7]. In a model of OA of the temporomandibular joint, increased osteoclast activity by intraperitoneal injection of Adrb2 agonist caused bone loss following increased mechanical stresses [9]. Adrb2 is a G protein-coupled receptor [47], a signal converter that converts extracellular signals into intracellular signals; plays a key role in bone development, remodeling, and diseases; and is the primary drug target of human diseases [48]. It provides a new target for treating OA and alleviating symptoms.
In summary, we compared the MTP subchondral bone with the LTP subchondral bone in patients with varus knee deformity using a number of experimental methods. The bone structure of the MTP subchondral bone was significantly different from that of the LTP subchondral bone, showing increased bone mass and obvious sclerosis. However, the density of osteoclasts and OCN in the MTP subchondral bone were higher than that in the LTP subchondral bone, indicating that there was no coupling between bone formation and resorption. Furthermore, it is reported for the first time that the expression of Adrb2 in the MTP subchondral bone with high mechanical stresses is higher than that in the LTP subchondral bone, indicating that the signal transduction of Adrb2 plays an important role in mechanical stress-induced subchondral bone remodeling. However, the mechanism of Adrb2 in the whole pathogenesis of OA needs to be further studied, which may provide a new target for the treatment of OA and the reduction of symptoms. Acknowledgments: The authors would like to thank all study participants who were enrolled in this study.

List of abbreviations
Funding information: The present study was supported by the National Natural Science Foundation of China (grant no. 8186090221).
Author contributions: XC.Y., XG.L., and HH.G. carried out the studies, participated in collecting data, and drafted the manuscript. L.M., L.J., X.Z., and J.W. performed the statistical analysis and participated in its design. LL.Y., ZQ.M., and QH.J. participated in acquisition, analysis, or interpretation of data and draft the manuscript. All authors read and approved the final manuscript.

Conflict of interest:
The authors state no conflict of interest.
Data availability statement: The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.