Sclerostin antibody promotes bone formation through the Wnt/β-catenin signaling pathway in femoral trochlear after patellar instability

ABSTRACT Purpose The molecular mechanism of patellar instability (PI) remains unknown. The purpose of this study was to explore the function of SOST/sclerostin in PI and examine the effect of sclerostin antibody (Scl-Ab). Materials and Methods We randomly divided 60 male 3-week-old C57Bl/6 mice into four groups: sham, PI, Scl-Ab intraperitoneal injection (Scl-Ab IP), Scl-Ab intraarticular injection (Scl-Ab IA). PI was established in the latter three groups. The Scl-Ab IP/IA groups were administered with an intraperitoneal/intraarticular Scl-Ab injection (100 mg/kg, 20 µl), respectively, at 5-day intervals. Distal femurs were collected 30 days after the surgery. The SOST/sclerostin, β-catenin, ALP, OPG and RANKL expression in distal femur were determined. Trochlear morphology and structural parameters of the trabecular and cortical bone compartments were determined by micro-CT. Further sub-regional analysis was performed. HE staining and Masson’s trichrome staining were performed to evaluate cartilage changes. Results PI increased the expression of SOST/sclerostin and RANKL, and decreased β-catenin, ALP and OPG levels, while Scl-Ab IP reversed these changes. Scl-Ab IP brought trochlear morphology closer to normality. Additionally, Scl-Ab IP significantly improved most of the bone parameters. Importantly, both PI and Scl-Ab IP acted mainly on trabecular bone. Histological analysis showed that Scl-Ab IP protected cartilage from degeneration. However, Scl-Ab IA did not protect against bone loss or cartilage degradation. Conclusions SOST/sclerostin plays an important role in PI and systemic Scl-Ab use promotes bone formation through the Wnt/β-catenin signaling pathway in the femoral trochlear after PI.


Introduction
Patellar instability (PI) is a frequent and debilitating musculoskeletal disorder among the young active population, particularly in females. Trochlear dysplasia is considered as a predisposing factor for PI, whereas our previous studies found that PI could also lead to trochlear dysplasia during the growth and development period 1,2 . Insufficient patellar stress is the main reason behind this. Mechanical stress plays a crucial role in promoting osteogenesis and maintaining bone homeostasis. Following the first episode of PI, the common process of mechanotransduction from the patellar to femoral trochlear is disrupted and results in abnormal bone formation 3 . Furthermore, the bone loss leads to morphological abnormality of the femoral trochlear and further exacerbates PI, thus creating a vicious circle. Currently, no nonsurgical intervention is available to restore trochlear development after PI. Break the loop at an early stage may be a potential treatment.
The SOST gene, which is expressed exclusively in osteocytes, plays an essential role in bone homeostasis 4 . Sclerostin, a glycoprotein encoded by the SOST gene, is secreted in a paracrine manner and subsequently acts on osteoblast lineage cells 5 . It inhibits canonical Wnt signaling by binding to the receptors of low-densitylipoprotein receptor-related protein 5 and 6 6 . Sclerostin not only mediates osteoblast differentiation and apoptosis, but also promotes osteoclastogenesis by regulating the secretion of nuclear factor kappa-B ligand (RANKL) and osteoprotegerin (OPG) 7,8 . Furthermore, sclerostin is a key negative regulator of mechanotransduction. The decrease in SOST/sclerostin during mechanical loading is accompanied by a significant increase in bone formation 9 . This enhancement of bone anabolism can be reversed by sclerostin upregulation 10 .
Neutralizing antibodies to sclerostin (Scl-Ab) are considered a promising anabolic treatment for bone loss. Recent animal studies and phase I and II clinical trials have found that Scl-Ab therapy could increase bone formation and mass in osteopenia, osteoporosis and other metabolic bone diseases 11,12 . Additionally, Scl-Ab therapy decreased bone resorption in postmenopausal osteoporotic women 13 . However, it remains unclear whether SOST/sclerostin is engaged in PI and whether Scl-Ab therapy is effective in promoting trochlear development after PI.
Currently, the molecular mechanism after PI remains unknown. The purpose of our study was to investigate the function of SOST/sclerostin in PI and investigate whether Scl-Ab therapy can promote bone formation to compensate for its loss in the femoral trochlear caused by PI. We also compared the effects of systemic or local Scl-Ab application. To the best of our knowledge, this study is the first to reveal that SOST/sclerostin plays an important role in trochlear development and could serve as a therapeutic target to prevent and treat PI-induced bone loss.

Animal models
The entire animal study was approved by the animal ethics committee of our hospital. A total of 60 male 3-week-old C57Bl/6 mice were purchased from Vitalriver (Beijing, China). Mice were acclimatized for one week and then assigned randomly to four groups: the sham group (n = 15), the PI group (n = 15), the PI + Scl-Ab (Abclonal, A8213, Wuhan, China) intraperitoneal injection group (Scl-Ab IP) (n = 15) and the PI + Scl-Ab intraarticular injection group (Scl-Ab IA) (n = 15). Mice were anesthetized using sevoflurane. To establish the PI model, the right knees of the mice were subjected to medial joint capsule incision. A 3-µm incision was made on the medial retinaculum to induce lateral PI. The patellar was pushed outwards repeatedly to spontaneously induce a lateral dislocation at terminal knee extension and recovery with flexion. The sham group received only a skin and subcutaneous tissue incision. Immediately after the operation, the Scl-Ab IP group was administered with an intraperitoneal Scl-Ab injection (100 mg/kg, 20 µl) and intraarticular injection of equivalent normal saline; while the Scl-Ab IA group was administered with an intraarticular Scl-Ab injection (100 mg/ kg, 20 µl) and intraperitoneal injection of equivalent normal saline. Both the sham and PI groups received 20 µl normal saline intraperitoneal and intraarticular injections. The dose selection was based on that of a previous study 14 . The treatments were given at 5-day intervals, with a total of six administrations. Each animal engaged in a daily 60-min treadmill exercise postoperatively to avoid disuse atrophy. The distal femurs were collected 30 days after the surgery.

Real-time quantitative polymerase chain reaction (RT-qPCR)
Following sample collection, the attached muscle, connective tissue, cartilage and bone marrow were removed, and only the bone tissue remained. Total RNA was extracted with a total RNA extraction kit (Promega, WI, USA) in accordance with the manufacturer's instructions. Subsequently, reverse transcription was performed using a reverse transcription kit (Promega).
RT-qPCR was conducted in the Bio-Rad CFX96 system (Bio-Rad, CA, USA). Primers were synthesized by Sangon Biotech (Shanghai, China) and the sequences are listed in Table 1. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an internal control and the expression change of the target gene was calculated by the 2−∆∆Ct method.

Micro-ct
Images of right distal femurs were collected via PIXImus CT (Siemens, Germany) at a voxel size of 10 μm acquired at 0.5° steps over a total rotation of 360° at 60kV and 400 μA. Images were used for threedimensional reconstruction with the Inveon Research Workplace II software (Germany).
All micro-CT analyses focused on the distal femur. A region of interest (ROI) of 1.2 mm was selected, starting from 0.5 mm distal to the growth plate. The upper and lower slices were shown for a better understanding of the selected ROI ( Figure 1A). Sub-regional analysis was performed in three ROIs. Total ROI assessed trabecular and cortical bone compartments separately and together. Within this region, we further analyzed the medial and lateral compartments of the femoral condyle ( Figure 1B) and subchondral bone ( Figure 1C). Sagittal slices were cut to display the medial and lateral regions.
Bone parameters were assessed in accordance with previous studies 15,16 . Within both the trabecular and cortical ROIs, the bone volume (BV), tissue volume (TV), BV/TV and tissue mineral density (TMD) were assessed. Within trabecular ROIs, trabecular BV, TV, BV/TV and TMD and microarchitecture parameters of trabecular thickness (Tb.Th), number (Tb.N) and separation (Tb.Sp) were assessed. Within cortical ROIs, cortical BV, thickness (Ct.Th) and TMD were assessed.

Measurements of sulcus angle and trochlear depth
The sulcus angle and trochlear depth were determined from the micro-CT images ( Figure 1D,E). The sulcus angle is the angle constituted between the lowest point of the femoral trochlea and the highest point of medial and lateral condyles. The trochlear depth is defined as the distance from the connecting line of the medial and lateral femoral condyles to the deepest part of the trochlear groove.

Hematoxylin-eosin (HE) staining
Following micro-CT scanning, distal femurs were decalcified for 2 weeks with EDTA at 4°C. Following decalcification, the samples were embedded in paraffin and cut to a slice thickness of 4 μm. The slices were subjected to HE staining and observed under an optical microscope.

Masson's trichrome staining
The slices were stained using the Masson Trichrome Staining Kit in accordance with the manufacturer's instructions (Solarbio, China).

Statistical analysis
All of the experiments were repeated three times. Statistical analysis was performed using SPSS 21.0 (SPSS Inc., IL, USA). All results were presented as mean ± standard deviation (SD). One-way analysis of variance (ANOVA) and Tukey's multiple comparisons test were used to compare the differences between groups. P<0.05 was considered to be statistically significant.

SOST/Sclerostin expression increased in mouse PI model
Initially, the SOST/sclerostin expression levels in the distal femur after PI were determined and were found to be significantly increased after PI (Figure 2A-C). These data suggested that SOST/sclerostin may play an important role in PI.

Scl-Ab IP promoted bone growth and development in young mice
To investigate Scl-Ab efficacy after PI induction, Scl-Ab was injected into the mice's abdominal or articular cavity at 5-day intervals starting immediately after surgery. Distal femur samples were collected and analyzed 30 days after the surgery ( Figure 3A). The mice in the Scl-Ab IP group were significantly larger than in the sham and PI groups (Supplementary Table S1). The Scl-Ab IP mice also exhibited a longer femur length and higher normalized femur weight (Table 2). By contrast, Scl-Ab IA did not show similar effects. These data suggested that systemic Scl-Ab application promoted bone growth and development in young mice.

Scl-Ab IP reversed the effects of PI on the Wnt/βcatenin pathway
To assess the effects of Scl-Ab, we analyzed the gene expression profiles of distal femurs from all the groups. Following PI induction, SOST and RANKL mRNA expression were increased, whereas the β-catenin, ALP and OPG mRNA levels were decreased (Figure 3 B-3). The protein levels were consistent with the mRNA level results (Figure 3 G-3). Additionally, we found that the effects of PI were reversed by Scl-Ab IP but not Scl-Ab IA. These data suggested that the bone loss in PI was associated with Wnt/β-catenin signaling pathway suppression, which can be prevented by systemic Scl-Ab treatment.

Scl-Ab IP partially reversed trochlear development after PI
To determine the impact of Scl-Ab IP on trochlear morphology, we measured sulcus angle and trochlear depth. The representative figures of each group are shown in Figure 4A. PI exhibited a larger sulcus angle and greater trochlear depth. Scl-Ab IP partially prevented these changes, while Scl-IA displayed the same pattern as the PI group ( Figure 4B,C). These data suggested that although systemic Scl-Ab application did not completely reverse the PI-induced changes, it brought trochlear morphology closer to normality.

Scl-Ab IP prevented PI-induced bone loss
A distal femur region was selected for detailed micro-CT ( Figure 5). In the PI group, total (combined trabecular and cortical) BV, BV/TV and TMD were decreased compared with the sham group. Scl-Ab IP not only reversed these reductions, but also increased these three parameters, whereas Scl-Ab IA did not display similar efficacy. Subanalyses confirmed that the effect of PI and Scl-Ab IP was greater in the trabecular region than in the cortical region. Trabecular parameters, including BV (−27%, +92%, +39%), BV/TV (−18%, +78%, +47%) and TMD (−7.8%, +24%, +14%), and cortical parameters, including BV (−14%, +13%, non-significant) and TMD (−4.3%, +8.3%, +3.6%), were all significantly greater with PI vs sham, Scl-Ab IP vs PI and Scl-Ab IP vs sham, respectively. These data suggested that both PI and systemic Scl-Ab application strongly impact anabolic responses and greater responses were found in the trabecular compartment. The medial and lateral femoral condyle regions ( Figure 6A) and subchondral bone region ( Figure 6B) were also assessed by micro-CT. The changes in all parameters were similar in these two regions. In the femoral condyle, the BV (−15% vs − 9%) and TV (−12% vs − 7%) changes observed in the medial compartment were more prominent than those in the lateral compartment with PI vs sham. The changes were similar across other parameters and groups. In subchondral bone, changing trends of bone parameters in the medial regions were similar to those in the lateral regions.

Scl-Ab IP reduced PI-induced cartilage degeneration
Histological assessments were performed to evaluate cartilage changes. The results of HE staining ( Figure 7A) and Masson's staining ( Figure 7B) showed that PI promoted cartilage degradation, while Scl-Ab IP effectively relieved this degradation. Scl-Ab IA failed to show any therapeutic efficacy. These data suggested that systemic Scl-Ab application protected cartilage from degeneration during PI pathogenesis.

Discussion
To the best of our knowledge, this study is the first to show SOST/sclerostin expression level as being elevated in PI. Systemic Scl-Ab application promoted bone formation to compensate for its loss caused by PI via the Wnt/β-catenin signaling pathway. Although systemic Scl-Ab application did not fully restore normal trochlear morphology, it significantly increased the BV and bone mass and reduced cartilage degeneration. Additionally, our study is the first, to our knowledge, to observe that bone loss in the distal femur was more prominent in the trabecular compartment. Interestingly, the promotion of bone formation on Scl-Ab treatment was also predominantly in the trabecular compartment. We further found that both the medial and lateral femoral condyles and subchondral bone underwent significant bone loss, with the changes in the medial femoral condyle being more prominent than those in the lateral femoral condyle. Overall, our findings indicated that SOST/sclerostin plays an important role in PI, and systemic Scl-Ab use could be a potential therapeutic approach (Figure 8).
In our study, the increased SOST/sclerostin in PI could lead to significant bone loss through Wnt/βcatenin signaling pathway inhibition. Inversely, systematic Scl-Ab administration restored bone formation in the femoral trochlear, suggesting that the Wnt/βcatenin signaling pathway plays an important role in PI. Although sclerostin was significantly decreased in the systematic Scl-Ab treatment group compared with the sham group, β-catenin expression did not differ between these two groups. A possible explanation is that there is a "self-regulation" mechanism during longterm Scl-Ab treatment 17 . Other Wnt inhibitors, such as Dickkopf 1, are elevated in the sclerostin-absent condition to compensate for sclerostin loss 18,19 .
Significant bone loss in the femoral trochlear was observed in our PI mouse model. As expected, the observed changes in bone parameters were similar in the femoral condyle and subchondral bone. Further analysis revealed that greater changes were observed in the medial femoral condyle, which was consistent with a previous study 20 . The different changes in bone mass between the medial and lateral compartments may be due to the different load distribution in the distal femur after PI. Unexpectedly, there was no difference between the medial and lateral subchondral bone in the extent of bone loss. In the patellofemoral joint, mechanical stresses are transmitted directly onto the articular cartilage and subchondral bone, and approximately 30% of loads can be attenuated by subchondral bone 21,22 . Therefore, the impact of mechanical alteration on subchondral bone was less pronounced. Furthermore, our results were the first, to our knowledge, to reveal that trabecular bone was most significantly altered in PI, but not cortical bone. There was evidence that trabecular bone is more sensitive than cortical bone to mechanical alteration, and mechanical stimulation is suggested to be a major determinant for maintaining trabecular bone structures 23,24 . Taken together, all these changes follow the general principles of ''bone functional adaptation" to altered stress 20,25 .
Scl-Ab therapy is considered a promising method to treat different conditions of bone loss 26 . Scl-Ab increased cortical BV and improved bone strength in an osteoporosis model 27 . Additionally, Scl-Ab enhanced implant fixation by increasing trabecular BV and improving the architecture 28,29 . In our study, Scl-Ab treatment for a 4-week period produced significant increases in bone mass. Furthermore, both our results and previous studies showed that Scl-Ab had a greater effect within the trabecular bone 14 .
PI was frequently accompanied by articular cartilage injury and degeneration 30 . The cartilage degeneration was exacerbated with time and further led to patellofemoral arthritis. Previous studies have shown that the Wnt/β-catenin signaling pathway plays an important role in cartilage development and homeostasis 31,32 . Sclerostin was found to participate in the occurrence and development of posttraumatic osteoarthritis, and could thus serve as a potential therapeutic target 33,34 . Our histological analysis showed that Scl-Ab also protected cartilage from degeneration. Taken together, these data indicated that Scl-Ab treatment displayed good efficacy in PI.
It should be noted that systematic Scl-Ab application provided only modest improvement in trochlear morphology. A possible explanation was that the vital mechanical stimulation for maintaining trochlear development was not restored. Our previous studies have found that recovery mechanics could promote trochlear development but failed to compensate for bone loss 2,3 . Because Scl-Ab can significantly increase bone formation after PI, combined surgical reduction and Scl-Ab treatment may solve these two problems simultaneously.
Our results showed that 100 mg/kg intraarticular Scl-Ab injection did not appear to impact bone formation or trochlear development after PI. It was reported that the treatment effect on bone formation was in a dose-dependent manner 11,35 . In our study, only a single dose was chosen for local application, and this may explain why it displayed no effect. A concentration gradient should be set in further studies to determine the optimum dose or verify that the local administration is inapplicable. Furthermore, after IA injection, the drug residence time within the joint cavity is influenced by the transport of the solute out of the joint 36 . The significant difference in efficacy between IP and IA injections may be attributed to the rapid elimination and clearance of drugs from the joint cavity.
Our study has several limitations. The Scl-Ab effects are time-dependent; the increased bone formation would return to the control level within 3-12 months after the first dose 17,37 . Because our study was only 4 weeks long, longer-term effects should be assessed in further studies. Moreover, because the estrogen receptor is involved in mechanotransduction 38,39 , Scl-Ab efficacy may differ by sex 40 . However, hormone levels in females are much more complex than in males, which may disturb the Scl-Ab effects. To minimize confounding variables, only male mice were used in our study. Additional studies are necessary to examine whether Scl-Ab is also applicable in females with PI.
In summary, our study was the first, to our knowledge, to find SOST/sclerostin elevation in PI. We further observed that the abnormal trochlear development after PI predominantly involved the medial trochlear region but did not affect the lateral trochlear. Although trochlear morphology did not return to normal, systemic Scl-Ab application significantly enhanced osteogenesis and reduced PI-induced bone loss and cartilage degeneration. These effects were achieved by activating the Wnt/β-catenin pathway. However, no efficacy was displayed in the local injection group. Therefore, our results revealed the potential mechanism of bone loss in the femoral trochlear after PI and that systemic Scl-Ab application exhibited a translational possibility of trochlear development when bone mass was low.