Expression of The Factors Associated With Wnt/β- Catenin, BMP-2/Runx2/Osterix, OPG/RANKL and LGR4/RANKL Pathways in Postmenopausal Osteoporosis Fracture


 Objective Detect and analyse the correlation between factors of related pathways (Wnt/β-catenin, BMP-2/Runx2/Osterix, OPG/RANKL and LGR4/RANKL) and postmenopausal osteoporotic fracture (PMOPF). Methods The postmenopausal patients with tibial fracture were divided into control group (36 cases) and PMOPF group (36 cases). Using RNAiso Plus method to extract total RNA of bone tissue, RT-qPCR method was used to detect the expression of each factor. Detected the levels of serum of factors by ELISA method in control group. PMOPF group was divided into group A-F according to the blood collection time interval (at different time period), use ELISA method to detect each factor’s level. Compared the changes between control group and PMOPF group, and the subdivided groups of PMOPF group. Results (1) RT-qPCR detected the expressions of LRP5, β-catenin, Runx2, C-myc, Osterix, OPG and LGR4 in PMOPF group were lower than control group (P<0.05), but the expression of RANKL was increased (P<0.05). (2) ELISA detected the serum levels of LRP5, β-catenin, Runx2, C-myc, Osterix, OPG and LGR4 decreased significantly (P<0.05), and RANKL increased significantly (P<0.05). LRP5 and Runx2 appeared the lowest in Group B (2-3 days after fracture); β-catenin and C-myc appeared the lowest in Group C (4-7 days after fracture); RANKL appeared the highest in Group C; Osterix appeared the lowest in Group D (8-14 days after fracture); OPG and LGR4 appeared the lowest in Group E (15-28 days after fracture). Conclusion The related factors of Wnt/β-catenin, BMP-2/Runx2/Osterix, OPG/RANKL and LGR4/RANKL pathway are closely related to the occurrence of PMOPF. LRP5 and Runx2 decreased to the lowest level within 3 days after fracture, β-catenin and C-myc decreased to the lowest level within 7 days after fracture, the results showed that these changes in Wnt/β-catenin osteogenesis pathway were consistent; Osterix decreased to the lowest level within 14 days after fracture, OPG and LGR4 decreased to the lowest level within 28 days after fracture, which may be related to the difficulty of short-term healing of PMOPF; RANKL increased to the highest level within 7 days after fracture, which may be associated with the increase in bone formation after PMOPF. According to the changes and characteristics of these factors in above pathways, we can regulate or intervene the occurrence and progression of PMOPF.


Introduction
Osteoporosis is a metabolic disorder associated with systemic bone aging and degradation. It is characterized by decreased bone mass, structural degradation, increased brittleness and prone to fracture [1] . There are more than 70 million Osteoporosis patients in China at present, the morbidity rate among wemen over 50 years old in China is 20.7% [2] . The incidence of osteoporosis in postmenopausal women is 2-3 times that of non-menopausal women [3] . Postmenopausal osteoporosis (PMOP) is an osteoporosis in women after menopause due to estrogen de ciency, resulting in bone loss and bone structure changes; PMOPF is a serious consequence of PMOP, which can signi cantly increase the disability rate, mortality rate and bring great family and socio-economic burden [4] . Therefore, the study of the pathogenesis, affecting factors, and treatment of osteoporotic fracture (OPF) become the focus of scienti c research and clinical practice [5] . Osteogenesis and osteoclastogenesis have been hot spots in the prevention and treatment of osteoporosis. The classical Wnt/β-catenin, BMP-2/Runx2/Osterix, OPG/RANKL and LGR4/RANKL signaling pathways play important roles in regulating osteogenesis and osteoclastogenesis. LRP5, β-catenin, Runx2, C-myc, Runx2 and Osterix are key osteogenic factors, while OPG, RANKL, LGR4 and RANKL are key related osteoclast factors [6][7][8][9][10][11] . In this study, RT-qPCR was used to compare the expression of LRP5, β-catenin, Runx2, C-myc, Osterix, OPG, RANKL and LGR4 in bone tissues of PMOPF patients and postmenopausal patients with other types of fracture (control group). ELISA was used to compare the expression of above factors in serum of the two groups. The level of each factor at different time period of PMOPF was further detected by ELISA. Based on histological observation and molecular studies of fracture healing and healing stages, PMOPF group was divided into Group A (1 day after fracture, before operation), Group B (2-3 days after fracture), Group C (4-7 days after fracture), Group D (8-14 days after fracture), Group E (15-28 days after fracture), Group F (29-42 days after fracture). This study aimed to investigate the level of above factors in PMOPF patients in order to understand their expression in the occurrence and development at different time period of PMOPF.

Subjects
This study was approved by Ethics Committee of People's Hospital of Sanshui and all participants signed the informed consent. The patients completely consented to this experimental study, signed informed consent and other related matters. Total 72 postmenopausal patients with closed tibial fracture were recruited at the hospital of People's Hospital of Sanshui from January 2018 to November 2020. The Bone Mineral Density (BMD) of lumbar was measured by dual-energy X-ray absorptiometry. BMD, height, and body weight were recorded. Bone turnover markers including β-CTX, PINP, N-MID-OT, 25(OH) D and estradiol were measured. Patients with secondary osteoporosis, osteoarthritis, and pathological fracture due to non-osteoporosis were excluded. They were divided into two groups: PMOPF group (36 cases), and control group (36 cases) described as postmenopausal cases (Non-osteoporotic). Operation of Internal xation to fractures in all patients were performed within one day of injury. Take the specimens (the curetted bone mass is equal or more than 200 mg) from bone tissue of the fracture end during operation. Bone tissue samples were collected and stored in liquid nitrogen rapidly. Fasting peripheral venous blood was collected within 1 day after injury (before operation) in all cases, After centrifugation, the serum was stored in a deep cryogenic refrigerator. Blood was further continuedly collected in group B-F of PMOPF group as described above (according to the blood collection time interval, at different time period), save the serum as above. Blood collected within 1 day in PMOPF group was belonged to group A. 100 mg bone tissues were grinded in liquid nitrogen. Total RNA was extracted from bone tissue by using RNAiso Plus (Bio-Rad, USA), and cDNA was synthesized by using PrimeScript RT Master Mix (Bio-Rad, USA). Real-time was performed using primers synthesized by Thermo Fisher Scienti c (the sequences of the primers were shown in Table 1), and SYBR Premix kit (Bio-Rad, USA). Conditions of PCR were as follow: denaturation at 94℃ for 5 minutes; 30 cycles of denaturation at 94℃ for 30 seconds, annealing at 58℃ for 30 seconds and extension at 72℃ for 40 seconds; extension at 72℃ for 10 minutes. GAPDH was selected as the internal reference. Data were expressed using the comparative Ct (2 −ΔΔCT ) method and normalized against GAPDH.

Data analysis
All experiments were performed at least 3 times. All data were expressed as mean ± standard deviation. The data were analyzed by nonparametric test (Mann-Whitney) using SPSS (version 16.0; Chicago, IL, USA). A value of P < 0.05 was considered statistically signi cant. Graphpad Prisma 5 software was used for statistical analysis.

Results
The comparison of general data In this study, 72 cases of fracture were included, and all of them were treated by surgery operation within 1 day. As shown in Table 2, there was signi cant difference in Lumbar BMD (T value) between two groups, but there were no signi cant differences in age, height and weight between two groups (P > 0.05).
In Table 3, the level of β-CTX and PINP in control group was signi cantly lower than that of PMOPF group (P < 0.05), the other parameters including N-MID-OT, 25-(OH)D and estradiol were not statistically signi cant between two groups (P > 0.05). The mRNA expression of factors in control group and PMOPF group in bone tissues RT-qPCR showed that LRP5, β-catenin, Runx2, C-myc, Osterix, OPG and LGR4 mRNA levels in bone tissues were lower in PMOPF group than in control group (P < 0.05), while RANKL mRNA level was higher in PMOPF group than in control group (P < 0.05) (Fig. 1).  Discussion PMOP is due to the rapid decline in estrogen levels in women after menopause and osteoclast resulting in a signi cant increase in bone resorption, while osteoblasts did not increase synchronously, resulting in bone resorption greater than bone formation, it's a metabolic disease [12] . The classical Wnt/β-catenin, BMP-2/Runx2/Osterix, OPG/RANKL and LGR4/RANKL/RANK signaling pathways play important roles in regulating osteogenesis and osteoclastogenesis [6][7][8][9][10][11] . Based on histological observation and molecular studies of fracture healing, the early stage of fracture healing is divided into early in ammatory response stage (within 1 days after fracture), non-speci c anabolic stage (within 3 days after fracture), non-speci c catabolism stage (3 days to 1 week after fracture) and more speci c anabolic stage of bone tissue (1 week after fracture); while the typical fracture healing stage was divided into three stages: hematoma organization stage (2-3 weeks after fracture), original callus formation stage (4-6 weeks after fracture), callus reconstruction molding stage (more than 1 years after fracture) [13][14] . Therefore, PMOPF group was further divided into group A (within 1 day after fracture), group B (2-3 days), group C (4-7 days), Group D (8-14 days), Group E (15-28 days) and Group F (29-42 days).
In this study, bone tissue and serum samples from fracture patients were used to explore the correlation between the expression of above factors and PMOPF. We analyzed 72 cases, there were no statistical differences in age, height and weight between the two groups. And there were also no statistical differences in N-MID-OT, 25-(OH)D and estradiol, which was consistent with the changes of bone turnover markers in the two groups. There were statistical differences in β-CTX and PINP between the two groups, which can be illustrated that the bone conversion of PMOPF group was higher than control group.
In the typical Wnt/β-cateinin signaling pathway, Wnt combined with LRP5/6 and FZD to construct complex, recruit Dvl and degradative complex, and inhibit the phosphorylation of β-catenin by GSK-3β. The non-phosphorylated β-catenin will gradually accumulate and enter into nucleus to activate downstream Runx2, C-myc and other factors, resulting in osteogenic differentiation [6,15] LRP5 exists on the surface of multiple cell membranes [16] . Some studies have found that the function and proliferation of osteoblast are blocked after LRP5 deletion, which affects bone formation [17] .
Glinka's [18] studies suggest that LGR5 can regulate embryonic patterns and the proliferation of stem cell through Wnt/β-catenin mediating agonist R-cavernous signaling. This experiment found that the expression of LRP5 in PMOPF group was signi cantly decreased, it was related to the limitation of osteogenesis, there may also be many factors such as sclerosis protein binding to LRP5/6, which can inhibited Wnt/β-catenin pathway and caused resistance to bone formation [19] . β-catenin is the most critical factor in Wnt/β-catenin pathway, by activating it's downstream factors, it can affect osteoblast and related expression. Hill TP [20] found that the knockdown of gene β-catenin in mice can increased cartilage production, but osteoblast was decreased, which revealed that β-catenin is necessary for early MSCs to differentiate. The expression of β-catenin decreased in PMOPF group, indicating that the classical Wnt pathway is in an inhibitory state, which could be one of the reasons for osteoporosis, this state re ects the insu cient osteogenesis after PMOPF. Runx2 is an earliest and highly speci c marker factor in osteogenesis, it is a key factor that led to bone fragility. It is a necessary gene in bone formation and bone development. Runx2 can up-regulates transcription of various mineralization-related protein genes in osteoblast [21,22] . Studies have con rmed that the activation of Wnt/β-catenin pathway can directly regulate Runx2, strengthen osteogenic differentiation and accelerate fracture healing [23] . The expression of Runx2 decreased signi cantly in PMOPF group, indicating that the low expression of Runx2 may be an important factor affecting PMOPF. C-myc is an important downstream factor of Wnt/β-catenin pathway, Wnt-3a can activate β-catenin signal, thus increase the expression of C-myc [24] , C-myc further promote cell cycle from G1 to S, accelerate osteoblast differentiation and proliferation [25] . The expression of C-myc decreased in PMOPF group, which suggest that C-myc as a target factor at the downstream of gene Wnt, can further veri ed the insu cient osteogenesis at osteoporosis transcription level.
The expression of LRP5, β-catenin, Runx2 and C-myc in PMOPF group was consistently decreased, which indicated that the osteogenesis of the four factors in the Wnt/β-catenin signaling pathway had highly coincident function, and further veri ed their close interaction of positive correlation and mutual promotion in the pathway. Studies have shown that LRP5 regulate osteoblast development and bone formation by activating Runx2 expression [23] . LRP5 and Runx2 both decreased to the lowest level within 3 days, and then increased, which highly indicated that their changes were consistent at the osteogenic stage. Chen Yan's [26] studies showed that β-catenin have different functions at different time of fracture repair. β-catenin can inhibit its downstream target C-myc by inhibiting glycolysis and glutamine [27] . The concentration of β-catenin in cytoplasm determines whether the expression of C-myc in nucleus can be activated [25] . β-catenin and C-myc all decreased to the lowest within 7 days after fracture and then increased, and the synchronism of the two factors was consistent with the above related studies. The above four factors in the same pathway increased so little (no statistical signi cance), which may be related to the di culty healing of PMOPF in the short term.
In BMP-2/Runx2/Osterix pathway, BMP-2 modulates the transcription of related osteogenic genes by activating Smads signal, thus induce the expression of Runx2 [28] . Smads transmit the signal TGF-β to nucleu and regulate the transcription of target factors, and then induce the expression of Runx2, Runx2 can further regulate Osterix [29] . Runx2 is the core factor affecting osteogenesis in Wnt/βcatenin and BMP-2/Runx2/Osterix pathways, so the expression of Runx2 in our study is obvious. In this study, the expression of Runx2 in PMOPF group was decreased, which indicated that Runx2 may affect the bone formation of PMOP by down-regulating the expression of osteoblasts. Osterix is an important osteogenic factor in the downstream of Runx2 [30] . The level of Osterix is dependent on the level of Runx2, Osterix only expressed in osteoblastic cells [31] . We found the expression of Osterix in PMOPF group was decreased, which can indicated that Osterix was an important downstream osteogenic factor and it may be one of the factors resulting in PMOP.
Kaback's [32] study showed that after the fracture model in mice was established, cartilage and tissue started to form after 7 days and sustained 10 days, meanwhile the mRNA level of Sox9 increased; Osterix was mainly expressed in the osteoblast near the site of fracture about 14 days later, meanwhile the expression of sox9 decreased, during this time cartilage became hard bone at the injured place. Numerous studies and clinical manifestations con rmed that BMP had unique osteogenic effect, and ber junction was completed at at 2 weeks' time after fracture [8,9] . The level of Osterix in BMP-2/Runx2/Osterix pathway appeared the lowest in group D (8-14 days), which was consistent with above study.
OPG/RANKL/RANK pathway is a very important signaling pathway in osteoclast differentiation, including: RANKL, RANK located on the cell membrane and pseudoreceptor OPG. There is a high a nity between OPG and RANKL, and OPG can competitively inhibits the interaction between RANKL and RANK, further inhibits the differentiation of osteoclast, cause bone resorption and induce apoptosis [10] . The decrease of OPG expression in PMOPF indicated that its function of competitively inhibiting osteoclast is growing weakern and promotes the further occurrence of osteoporosis. RANKL is the only factor to stimulate the differentiation and maturation of osteoclast up to now, and it can prevent apoptosis [33] . This study found that the expression of RANKL increased in PMOPF group, which was consistent with the thoery that RANKL can cause osteoporosis by promoting differentiation of osteoclast. RANKL has strongest expression in bone tissue [34] , so our study found that the expression increased more obviously in bone tissue than in serum. The level of OPG decreased to the lowest within 28 days after fracture, which may be related to the di culty of short-term healing of PMOPF. RANKL increased to the highest within 7 days after fracture, and then a small increase or decrease occurred, re ecting their trend in osteoclast, which may be associated with the increase in bone formation after PMOPF.
LGR4/RANKL/RANK pathway plays an important role in osteoclastogenesis.
LGR4 is a receptor of RANKL.
LGR4 competes with RANK to combine with RANKL, and inhibit typical RANKL signals during osteoclast differentiation [10] . Binding to LGR4, RANKL activates Gαq and GSK-3β signaling pathway and modulates osteoclast differentiation and bone resorption [11] . We found that the expression of LGR4 was decreased in PMOPF group, which suggested that the fuction of inhibiting osteoclast of LGR4 became weakened, meanwhile RANKL increased in PMOPF group, which was consistent with the theory that the combination of LGR4 competes with RANK and combines with RANKL, then inhibits the osteoclast of RANKL/RANK signaling pathway. LGR4 decreased to the lowest level within 28 days after fracture, re ecting the slow and lasting trend of LGR4, which may be related to the di culty of short-term healing of PMOPF. Negative regulation to osteoclast of LGR4 can be used to treat osteoporosis and other disease [35] . The whole genome sequencing of human also found that LGR4 has a great function to osteoporosis [36] . Denosumab as an speci cal antibody through targeting RANK, can cause osteoclast inactivation by block the combination of RANKL and RANK, but its side effects include calcium homeostasis imbalance and so on [37] . The related studies in mice showed that compare with OPG to RANKL, LGR4-ECD protein has lower a nity with RANKL, and has little negative physiological effect on mice. The above studies show that the advantages of LGR4 and the side effects of antagonistic RANKL have great value in the treatment of osteoporosis [11] . Through the above analysis, There may exist some related relation and mechanism according with the variation tendency of RANKL and LGR4 (RANKL appeared highest during 4-7 days, LGR4 appeared lowest during 15-28 days) in our study. The peak time of RANKL appeared shorter than LGR4, Does it related to the therapeutic effect? We can further study it.
Although this study has limitations that we did not investigate the mechanisms underlying the abnormal expression of factors in bone tissue and serum of PMOPF patients, our results provide strong evidence that the factors related to Wnt/β-catenin, BMP-2/Runx2/Osterix, OPG/RANKL and LGR4/RANKL pathways affected osteogenesis and osteoclastogenesis, and each factor changed at a different subdivided time period in PMOPF group. According to the changes and characteristics of these factors in these pathways, we can regulate or intervene the occurrence and progression of PMOPF.  Line diagram of serum levels of each factor in group A-F (n=36)