Assemble Flavone of Rhizoma Drynariae Promotes Differentiation of Osteoblasts and Growth of Bone Graft in Induced Membrane Partly by Activating Wnt/β-Catenin Signal Pathway

Background: Assemble Flavone of Rhizoma Drynariae (AFRD) is not only the extract of Rhizoma Drynariae, but also the effective component of Qianggu capsule, which is widely used in the treatment of fracture, bone defect, osteoporosis and other orthopedic diseases, and has achieved good results. Purpose of this trial was to explore effects of AFRD on bone graft mineralization and osteoblast differentiation in Masquelet induced membrane in rats. Methods: Twenty male Sprague-Dawley rats aged 10-12 weeks were randomLy divided into AFRD group (n=10) and control group (n=10). Critical-sized defects model of rats was established with 10 rats in each group. Polymethyl methacrylate (PMMA) was implanted into the defect of femur in rats. After the formation of the induced membrane, autogenous bone was implanted into the induced membrane. After 10 weeks of bone grafting, bone tissue in the bone graft area was examined by X-ray, Micro-CT and hematoxylin-eosin (HE) staining to evaluate the growth of the bone graft. Serums of the two groups of rats were extracted respectively, and these serums were used to culture osteoblasts in vitro. CKK8 method, Alkaline phosphatase (ALP) staining, Western blot and RT-PCR and other methods were used to evaluate the effects of AFRD on the proliferation of osteoblasts and the regulation of Wnt/β-catenin signal pathway. Results: In vivo experiment showed that the growth and mineralization effect of bone graft in AFRD group was better. In vitro experiment showed that osteoblasts proliferated faster, ALP activity was higher, number of mineralized nodules was more, and expression of proteins related to Wnt/β-catenin signal pathway and osteogenesis were more in AFRD group. Conclusions: AFRD can promote mineralization of bone graft and differentiation of osteoblasts during the bone graft growth period of induced membrane technique, which is related to the activation of Wnt/ β-catenin signal pathway.

technique is high, it also has the problem of long healing time. Studies have shown that the postoperative bone healing time of induced membrane technique is from 3 months to 94 months [7].
Bone formation is a series of complex physiological and pathological processes, including intramembranous ossi cation and endochondral ossi cation. Previous studies have shown that the bone formation function of osteoblasts and the bone resorption function of osteoclasts play a key role in the process of bone formation and remodeling [8,9]. Osteoblasts originate from bone mesenchymal stem cells (BMSCs) and differentiate and mature from BMSCs under the action of osteoblast differentiation factor. Osteoblasts are not only the main effector cells of mechanical stress in bone tissue, but also the main functional cells of bone formation. responsible for the synthesis, secretion and mineralization of bone matrix. Its differentiation and proliferation mainly determines the bone mass of bone formation.
Therefore, the regulation of osteoblasts has become an vital target to promote bone formation.
Wnt/β-catenin signaling pathway is an important regulatory mechanism involved in osteoblast differentiation [10][11][12], which is essential for bone development, bone mass maintenance and bone remodeling. When Wnt protein binds to speci c frizzled transmembrane receptors and the Low-density lipoprotein receptor related protein (LRP, Lrp5/6) co-receptor on cell surface, β-catenin is released in the cytoplasm and no protein degradation occurs. The accumulated β-catenin is transferred to the nucleus, where it binds to T-cell factor 4 (TCF 4) or lymphoid enhancer factor 1 (LEF 1) to activate downstream target genes such as c-myc, cyclinD, Runx2 transcription [13]. CyclinD 1, β-catenin and c-myc are the main functional molecules in Wnt/β-catenin. Dickkopfs (Dkks) bind and sequester the Lrp5/6 and Krm1/2 membrane complex to inhibit Wnt activity [14]. Dkk1 is a secretory Wnt inhibitor with good speci city and is active in many tissues [15]. Data from different animal models have con rmed that Dkk1 can inhibit Wnt signal pathway and thus inhibit bone formation. Anti-Dkk 1 neutralizing antibodies against the epitopes necessary for LRP 5 and LRP 6 binding to Dkk 1 increased bone mass in normal mice [16]. The experimental model of multiple myeloma showed that anti-Dkk 1 antibody treatment reversed the inhibitory effect of Dkk 1 on osteoblast formation and bone formation, thus reducing bone loss [17].
Traditional Chinese medicine (TCM) of tonifying kidney and strengthening bone has accumulated rich experience in promoting osteogenesis. Yellow Emperor's Inner Classic: Basic Questions believes that the kidney governs the bone. It is pointed out that the formation and development of bone are closely related to the kidney. Only when the kidney qi is su cient and the kidney essence is full, can the bone marrow be lled and the bones can be strong. Rhizoma Drynariae is a representative medicine of tonifying kidney and strengthening bone in orthopedics of TCM. It was rst recorded in the Supplement to 'The Grand Compendium of Materia Medica' in the Tang Dynasty. It is the dried rhizome of Drynaria fortunei (Kunze) J.S m, which can be picked all the year round and has been traditionally used in many Asian countries including China, Korea, and Japan for the treatment of diverse orthopedic diseases such as fracture, osteoporosis, bone defect, arthritis, etc., and has pharmacological activities to promote osteogenesis, anti-in ammation and anti-oxidative damage [18][19][20][21][22]. Animal experiments showed that Rhizoma Drynariae and its extracts could increase the number of bone trabeculae and bone mineral density (BMD), improve the morphology of bone tissue, promote new bone formation and increase biomechanical strength in bone defect or osteoporotic rats, and no systemic side effects such as infection were found [23,24]. In addition, in vitro experiments showed that Rhizoma Drynariae could accelerate the differentiation and mineralization of osteoblasts [25] and inhibit the bone resorption of osteoclasts [26].
Assemble Flavone of Rhizoma Drynariae (AFRD) is an effective ingredient extracted from the dried rhizome of Rhizoma Drynariae. The main component of Qianggu capsule used in this study is the AFRD, which can promote bone healing and relief the pain [27], but its speci c mechanism is not clear in terms of microstructure and cellular and molecular biology.
Idea of this trial was to explore effects of AFRD on bone graft mineralization and osteoblast differentiation in Masquelet induced membrane from the point of view of Wnt/β-catenin signal pathway.
In addition, it is also hoped to provide experimental basis for traditional Chinese medicine (TCM) in promoting bone formation and mineralization of induced membrane technique, shortening the treatment cycle of bone healing and improving the quality of osteogenesis.

Experimental animals
Twenty healthy male Sprague-Dawley (SD) rats of 10-12 weeks old, weighing 250 g to 310 g (280.3 ± 21.4 g), were selected and provided by Guangdong Medical Laboratory Animal Center. License No.: SCXK (Yue) 2018-0002, Experimental Animal Certi cate No.44007200064529. All the selected experimental animals were raised in the SPF animal room of Guangzhou University of traditional Chinese Medicine (the laboratory temperature was 22 ~ 24℃, the humidity was 60 ~ 70%, and the light and dark cycle was 12 h/12 h), feeding feed was provided by the Experimental Animal Center of Guangzhou University of traditional Chinese Medicine. The experiment was carried out one week after feeding. According to the method of random number table, the experimental animals were randomly divided into two groups: Assemble Flavone of Rhizoma Drynariae group (AFRD n = 10), Conrrol group (Control n = 10).

Establishment of animal models
Before the experiment, the rats fasted for 12 hours, and 40,000 U of penicillin sodium was injected intramuscularly to prevent infection. Anesthesia was given intraperitoneally with 3% pentobarbital sodium (1.5 mL/kg). After the anesthesia takes effect, the right hindlimb is shaved to prepare the skin.
The rats were taken from the left recumbent position to expose the right hindlimb, sterilized and covered with aseptic hole towels.
The rst stage operation: The skin and fascia were cut longitudinally from the lateral greater trochanter to the lateral condyle of the femur, and the subcutaneous muscles were separated to expose the lateral side of the femur. Place a custom six-hole plate on the anterolateral side of the femur. After drilling, two cortical self-tapping screws were used to x the plate at the distal and proximal ends, and a wire saw was used to cut the bone at the center of the femoral shaft, the length of which was 4 mm. The bone defect area was lled with a prefabricated columnar cement block of 4 mm length. Finally, the incision was sutured layer by layer (Fig. 1a).
The second stage operation: Six weeks after the rst stage operation, two caudal vertebrae were taken from the middle or distal segment of the rat tailbone and cut into ne bone particles for bone grafting.
The subcutaneous muscles were separated to the induced membrane by longitudinal incision along the rst stage surgical incision. After the bone cement was removed, the prepared bone particles were lled into the bone defect area. Finally, the induced membrane, muscle fascia and skin were sutured with silk thread (Fig. 1b).
Within 3 days after rst or second stage, 40,000 u of penicillin sodium was injected intramuscularly every day to prevent infection and raised in a single cage for 14 days.

Intervention measures
Qianggu capsule (each Qianggu capsule contains AFRD of 180 mg) was added to steamed water to make a certain concentration solution. The equivalent dose of AFRD was calculated according to the body surface area. The rats in AFRD group were given AFRD of 0.22 g/kg/d by intragastric administration, and the rats in control group were given the same dose of stroke-physiological saline solution (SPSS). In the course of the experiment, the weight was weighed every 2 weeks, and the dosage was adjusted in time according to the change of body weight. The rats were given medicine for ten weeks.

Collection and treatment of specimens
Two rats in each group had internal xation failure and failed to carry out follow-up trial. Finally, there were 8 rats in each group. Two hours after the last intragastric administration, the blood was taken from the abdominal aorta under anesthesia, and the serum was obtained by centrifugation, and the AFRD serum and SPSS serum could be isolated and stored in the refrigerator at -20℃ to prepare for the culture of osteoblasts. Four rats in each group were selected, and the bone tissue of the bone graft area was removed and xed in 4% paraformaldehyde xed liquid for 24 hours. Morphological and structural changes of bone tissue were observed by hematoxylin-eosin (HE) staining. The other four rats were examined by X-ray and Micro CT to evaluate the growth of bone graft.

Histological observation of bone tissue in bone graft area
The bone tissue in the induced membrane area was cut and treated with dehydration and para n embedding, then the tissue slicer was used for continuous slicing with a thickness of 5 µm. After baking at 68℃ in a constant temperature baking machine, the bone tissue sections were stained with hematoxylin-eosin (HE) staining solution. After sealing, the osteogenic process was observed and evaluated under light microscope.
2.7 X-ray and Micro-CT examination of bone in bone graft area X-ray mainly analyzed the growth of bone graft and the loosening or prolapse of steel plates and screws.
After X-ray examination, the right femur was removed, the muscle, fascia and other soft tissue around the femur were removed, and the femur was put into the Micro-CT sample tube for Micro-CT examination.
After the completion of the scan, the scanning results were analyzed by CT-An software, and the bone graft area was manually selected to establish a three-

Extraction and identi cation of osteoblasts
Five suckling rats of SD rats were killed, soaked in 75% alcohol for 2 minutes, the calvaria was taken under strict sterile conditions, the connective tissue attached to the bone surface was removed, PBS was washed repeatedly until the bone tissue was whitened, the bone tissue was cut to the size of 1 mm × 1 mm with scissors, the phosphate buffer (PBS) was rinsed to the bone tissue whitening, and the bone tissue fragments were placed in a centrifuge tube and digested with 0.25% trypsin. After 30 min, trypsin was discarded, 0.1% type II collagenase of 8 mL was added, digested for 1 hour, the Supernatant uid was collected and transferred to another centrifugal tube, and centrifuged by 1000r/min for 10 minutes.
0.1% type collagenase was added to the centrifuge tube containing bone tissue for 1 hour, and the Supernatant uid was collected and centrifuged to collect cells; added to the prepared cell culture medium. The cells were inoculated in the 25 cm 2 culture bottle at the concentration of 2 × 10 4 /mL and cultured in incubator (37℃, 5% CO 2 ). The adhesion and growth of cells were observed every day. After the cells were pasted to the bottom, they were digested and passaged with trypsin. The third generation osteoblasts were used in the experiment. After the cells adhered to the wall, the culture asks containing primary osteoblasts and passage osteoblasts were observed and photographed under inverted uorescence microscope. Osteoblasts were identi ed by morphological observation and alkaline phosphatase staining.

Culture of osteoblasts in vitro
The third generation osteoblasts were used in the experiment, and the osteoblasts of four groups were cultured in different pre-prepared culture media. Among them, in the process of osteoblast culture of Dkk1 group and AFRD + Dkk1 group, Recombinant Human Dickkopf-related protein 1 (Dkk1) was added to each culture medium at a concentration of 0.5 µg/ml. In the process of cell culture, the liquid was changed every 2 to 3 days, and the growth status of the cells was observed.

Determination of the proliferation rate of osteocytes
Proliferation of cells in each group was detected by Cell Counting Kit-8 (CCK-8) method. Osteoblast suspension (100 µl/well) was inoculated in 96-well culture plate and cultured in different culture media.
10 µl CCK 8 solution was added to each well and incubated in the incubator for 4 hours. The absorbance (O.D value) at 450 nm was determined by enzyme labeling instrument.

Alkaline phosphatase (ALP) activity measurement
After 3 days, 6 days, 9 days and 12 days of cells culture, the activity of alkaline phosphatase (ALP) of osteoblasts was measured by Para-nitrophenyl phosphate (pNPP). The culture medium was removed, PBS was washed three times, 50 µl 0.5% Triton X-100 cell lysate was added and lysed at 4 °C for 1 hour.
The 20 µl lysate was taken on a 96-well plate and operated according to the operation table of the Alkaline Phosphatase Assay Kit. The OD value was determined by enzyme labeling instrument at 405 nm wavelength. According to the OD value of the sample, the activity value of ALP (U/L) was read on the ALP standard curve.

Alkaline phosphatase (ALP) staining
After 6 days of cells culture, an appropriate amount of osteoblasts were cultured on cover slides. The osteoblasts were washed with PBS and xed with cold acetone for 10 minutes after 70%-80% con uence. PBS washed the cells twice, then added alkaline phosphatase staining solution for 4-6 hours, rinsed with tap water several times, soaked in 2% cobalt nitrate for 3-5 minutes, then immersed in 1% ammonium sul de for 2 minutes. Finally, rinse with tap water and seal the lm after drying.

Mineralized nodule staining
Osteoblasts were cultured in 24-well plate for 5 days, then mineralized induction solution was added and cultured for two weeks. The dish was gently washed with PBS for 2 times, xed with 95% ethanol for 10 minutes, washed with distilled water for 3 times, added 1% alizarin red S solution, and dyed at 37 °C for 10 minutes. Finally, the dish is washed with distilled water. After drying, the mineralized nodules were observed under microscope and their number and area were analyzed.

Immuno uorescence detection
After the predetermined time of cells culture ( 3days, 6days, 9days ), culture medium was quickly absorbed and rinsed with cold PBS for 3 times. The cells were xed with 4% paraformaldehyde xed solution for 30 minutes and washed with PBS for 3 times and dried naturally. Add 0.25% Triton X100 and rest at 37℃ for 10 min and wash for 3 times, then air-dry naturally. Add sealed serum and seal at 37℃ for 30 minutes, pay attention to moisturizing. Remove the sealed serum, add β-catenin (diluted by 1 : 200) and spend the night at 4℃. Then, the residual liquid of primary antibody was sucked out and washed by PBS for 10 minutes for 3 times. Under the condition of avoiding light, the second antibody with FITC labeling was added (diluted by 1 : 200). After 1 hour, the cells were washed with PBS for 10 min. Finally, add DAPI for 10 minutes and seal the lm. The expression and localization of β-catenin were observed under uorescence microscope.

Protein extraction and western blotting analysis
After 6 days of osteoblasts culture, osteoblasts were fully washed with PBS, were fully lysed by the cell lysate, and the total protein of the sample was extracted, and the protein content was measured by BCA method. After boiling denaturation, 40 µg protein was taken for 15% sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE) electrophoresis and transferred to the membrane. After the end of the lm transfer, seal with 5% skimmed milk powder for 1 hour. Then the rst antibody was added and incubated overnight at 4 °C. After the incubation membrane was washed, the secondary antibody was added. ECL kit was used for photoluminescence development, and GAPDH (glyceraldehyde-3-phosphate dehy drogenase) was used as the internal reference protein.

RNA isolation and real-time PCR (RT-PCR)
After 6 days of osteoblasts culture, total RNA was extracted by Trizol method and reverse transcribed into cDNA by PrimeScript RT reagent Kit (Japan, RR037A). After the 25 µl reaction system was constructed, the relative mRNA expression of β-catenin, TCF, LEF-1, cyclinD, c-myc and Runx2 was detected by realtime PCR., The relative expression levels were determined according to the 2 −△△Ct manner. ACTB was used as the internal reference gene. The primer sequence was list in Table 1. Table 1 Prime sequences for RT-PCR.

Statistical analysis
All the data were analyzed by Stata 12.0 software, and the metrological data such asβ-catenin, TCF and LEF protein content were expressed as means ± standard deviation (SD). After satisfying the normal distribution, the mean among the four groups were compared by one-way ANOVA, P < 0.05 was considered statistically signi cant.

AFRD accelerates the growth rate of bone graft
After 10 weeks of bone grafting, cartilage area and osteogenic area could be clearly seen in the bone graft area of the two groups, but osteogenic area of AFRD was larger than that of control group, and control group was dominated by cartilage growth at this stage (Fig. 2). From the X-ray results, new bone formation could be seen in the femoral defect area of both groups, and there was no loosening or displacement of screws and steel plates. In AFRD group, continuous callus lled the bone defect area, and cortical bone was basically formed, while in the control group, callus volume was small, only part of it passed through defect area, and cortical bone was not completely formed (Fig. 3). The above results show that AFRD can accelerate the growth rate of bone graft in induced membrane.

AFRD enhances mineralization effects of bone graft
From the results of Micro-CT cross-sectional scanning, bone mass and bone molding effect of AFRD group were signi cantly better than those of the control group (Fig. 4). Parameters of Micro-CT showed that the Bone Volume Fraction (BV/TV), Trabecular Number (Tb.N), Trabecular Thickness (Tb.Th), Connectivity Density (Conn.D.) in AFRD group were signi cantly higher than control group. The Bone Surface Fraction (BS/BV) and Trabecular Separation/Spacing (Tb.Sp) in AFRD group were smaller than control group, and the difference was statistically signi cant. However, there was no signi cant difference in Structural Model Index (SMI) between the two groups (P > 0.05) (Fig. 5). This is consistent with the results of X-ray and HE staining, which shows that AFRD can promote the formation of bone trabeculae and increase bone mass in the area of bone graft.

AFRD promotes the proliferation of osteoblasts
According to the results of CKK8 method, on the whole, the proliferation of osteoblasts in different groups had a certain rule: during 6 days of culture, cells proliferation increased gradually, the proliferation was the most exuberant on the 6th day, and entered the plateau in 6-9 days, and gradually decreased. The increment rate of cells in each group was almost the same on the rst day. On the 3rd day, 6th day and 9th day, there were signi cant differences in the increment rate of osteocytes. The sequence of osteocyte increment rate of each group was as follows: AFRD > control > AFRD + Dkk1 > Dkk1, the difference was statistically signi cant, indicating that the serum of AFRD could promote the proliferation of osteoblasts and was related to the activation of Wnt/ β-catenin signal pathway. ( Fig. 6)

AFRD increases the activity of ALP in osteoblasts
According to the results of Para-nitrophenyl phosphate (pNPP) detection, the activity of ALP of osteocytes in each group also showed a certain rule: within 9 days of cell culture, the ALP activity of osteoblasts showed an upward trend, with the fastest increase within 3-6 days, and the osteoblasts still increased within 6-9 days, but the range of proliferation was small. From 9 to 12 days, the activity of ALP showed a downward trend, which may be related to the slow proliferation of osteoblasts and the decrease of ALP secretion in the later stage. In our detection time, the activity value of osteoblasts was in the following order: AFRD > control > AFRD + Dkk1 > Dkk1, and the difference was statistically signi cant (Fig. 7). The results showed that the ALP activity of osteoblasts signi cantly increased under the intervention of AFRD, which was related to activation of Wnt/β-catenin signal pathway.

AFRD increases the number of ALP in osteoblasts
In order to further analyze the difference of ALP expression in each group, we performed ALP staining after osteoblasts were cultured for 6 days. The results showed that the ALP region showed grayish-brown ake deposition in the cytoplasm. There are some differences in the amount of ALP secreted by different groups of osteocytes. Among them, the expression region of ALP was the most in the AFRD group, followed by the control group, the AFRD + Dkk1 group was less than that in the control group, and the Dkk1 group was the least (Fig. 8). It is suggested that the AFRD can stimulate osteoblasts to secrete ALP, which is related to the activation of Wnt/β-catenin signal pathway.

AFRD promotes the maturation of osteoblasts
The formation of mineralized nodules is one of the important signs in the process of osteoblast maturation. According to the results of alizarin red S staining of osteoblasts, deep red mineralized nodules were observed in all groups of osteoblasts after 19 days of cell culture, and there were signi cant differences in the number of mineralized nodules in different groups. Among them, the number of mineralized nodules in the AFRD group was the most, followed by the control group, the number of mineralized nodules in the AFRD + Dkk1 group was less than that in the control group, and the number of mineralized nodules in the Dkk1 group was the least ( Fig. 9). The appeal results showed that AFRD can promoted the maturation of osteoblasts.

AFRD upregulates Wnt/β-catenin signal pathway in osteoblasts
According to the results of Western blotting, there were signi cant differences in the average protein expression of β-catenin, TCF, LEF, cyclin D, c-myc and Runx2 among different groups. The average expression of signal pathway-related proteins in the AFRD group was higher than that in the control group and the AFRD + Dkk1 group, indicating that AFRD could upregulate the activation of Wnt/β-catenin signal pathway and promote the expression of pathway-related proteins in osteoblasts. At the same time, the control group was higher than the Dkk1 group (Fig. 10), indicating that Dkk1 had a de nite blocking effect on Wnt/β-catenin signal pathway. In addition, RT-PCR results showed that there were signi cant differences in the relative mRNA expressions of β-catenin, TCF, LEF, cyclinD, c-myc and Runx2 among different groups. The relative expression of mRNA related to Wnt/β-catenin signaling pathway in osteocytes was in the following order: AFRD > control > AFRD + Dkk1 > Dkk1 (Fig. 11). This is generally consistent with the results of Western blot.

AFRD upregulates Wnt/β-catenin signal pathway with time difference
In order to dynamically observe the regulatory effect of AFRD on Wnt/β-catenin signal pathway, the expression status of β-catenin protein in different time periods was detected by immuno uorescence. The existence of β-catenin was observed in all groups of osteocytes after 3 days, 6 days and 9 days of cells culture, and β-catenin was positive in cell membrane, cytoplasm and nucleus, showing high intensity green uorescence, and the expression region increased gradually with the extension of culture time. Generally speaking, the uorescence signal intensity of β-catenin was more obvious at 9th day, and weaker at 3rd day and 6th day. From the comparison between the groups, the most positive areas of uorescence signal were found in the AFRD group, followed by the control group. The expression of AFRD + Dkk1 group and Dkk1 group was the least (Fig. 12). These results suggest that AFRD can continuously upregulate Wnt/β-catenin signal pathway.

AFRD promotes expression of osteogenesis-related proteins
There are many proteins related to bone formation in the process of osteoblast differentiation, such as collagen type I alpha 1 (COL1A1), bone morphogenetic protein 2 (BMP-2) and osteopontin (OPN). We found that AFRD signi cantly increased the expression of the three osteogenic marker genes. The relative expression of osteoblast-related proteins in osteoblasts was in the following order: AFRD > control > AFRD + Dkk1 > Dkk1 (Fig. 13). It was con rmed that the AFRD exerted its osteogenic effect partly by activating the Wnt/β-catenin signal pathway on osteoblasts.

Discussion
Although Masqulet technique has achieved a high success rate in clinical practice, the composition and characteristics of induced membrane and mechanism of its bone healing are not clear. According to the diamond concept of bone healing [28,29], including osteoblasts, bone conduction scaffolds, blood vessels, osteogenic factors, and the stable mechanical environment, osteoblasts are essential for bone growth and mineralization. It is well known that bone formation of osteoblasts and bone resorption of osteoclasts play a key role in the process of bone formation and remodeling. In the process of bone formation, osteoblasts go through four stages: osteoblast proliferation, extracellular matrix maturation, extracellular matrix mineralization and osteoblast apoptosis. Therefore, promoting the proliferation, differentiation and mineralization of osteoblasts and increasing the secretion of osteogenesis-related proteins have become one of the trail ideas to accelerate the speed of bone healing in induced membrane.
Under guidance of the theory of tonifying kidney and strengthening bone in TCM, the use of kidneytonifying and bone-strengthening herbs to promote bone formation has become a unique method for the treatment of fracture and osteoporosis [30], but the speci c mechanism is not clear. Rhizoma Drynariae is a representative medicine of tonifying kidney and strengthening bone in orthopedics of TCM. AFRD is one of the effective components extracted from the dried rhizome of Rhizoma Drynariae, and it is also the effective component of Qianggu capsule, a Chinese patent medicine. An in vivo trail indicated that AFRD could increase BMD, mechanical strength, Bone Volume (BV), Bone Volume Fraction (BV/TV), Trabecular Number (Tb.N), Trabecular Thickness (Tb.Th) and decreased Trabecular Separation (Tb.Sp) in osteoporotic rats [31]. Yao et al found that after taking AFRD, chickens with Tibial dyschondroplasia (TD) recovered their walking ability earlier, repair and arrangement of chondrocytes were more regular, the vascular invasion of cartilage area was earlier, and the expression levels of BMP-2 and Runx2 were higher [32]. Both BMP-2 and Runx2 are important regulatory genes for bone formation and differentiation [33,34]. Most of these trails focus on the effect of AFRD on improving fracture and osteoporosis, but there are few reports on promoting the growth of bone graft after secondary operation of msquelet technique. Our results showed that bone remodeling ability of the bone graft area in AFRD group was better than that in control group in terms of histology and imaging results.
Studies have shown that bone formation is related to the activation of osteogenesis-related signal pathways in BMSCs and osteoblasts, including Wnt/β-catenin signal pathway, MAPK signal pathway, Smad signal pathway and so on [35]. Among these signal pathways, the most important signal pathway is Wnt/β-catenin signal pathway [36]. With the activation of Wnt/β-catenin signal pathway on osteoblasts, osteoblasts enter the mitotic phase. At this stage, the differentiation and proliferation of osteoblasts is accelerated, the synthesis of ALP is increased, and calci cation is initiated, thus promoting bone formation. The content of ALP in cells represents the degree and state of cell differentiation and is an early speci c marker of extracellular matrix maturation [37]. Studies found that local use of β-catenin enhancer can promote the proliferation and differentiation of osteoblasts, and then promote new bone formation [38]. In a experiment of transgenic mice, it was found that the expression level of β-catenin was directly related to bone formation, and the loss of β-catenin expression directly led to the decrease of osteoblast differentiation and the disturbance of bone formation [36]. In the Wnt/β-catenin signal pathway, Dkk1 is one of the important antagonists, which can speci cally inhibit the classical Wnt signal pathway [39]. In vitro experiments have con rmed that AFRD can induce the proliferation and differentiation of BMSCs and osteoblasts and inhibit the early apoptosis of osteoblasts [40,41]. In order to further con rm the speci c mechanism of AFRD on osteoblasts, we cultured osteoblasts in vitro. We found that in terms of osteoblast increment rate, ALP activity and the number of mineralized nodules, the AFRD group was more obvious than that of the control group and the AFRD + DKK1 group, indicating that the AFRD could signi cantly promote the proliferation of osteoblasts. This is related to the up-regulation of Wnt/β-catenin signal pathway by the AFRD. At the same time, Dkk1 can speci cally inhibit the classical Wnt signal pathway and reduce the rate of cells proliferation. Speci cally antagonizing the effect of Dkk1 in vivo to upregulate Wnt/β-catenin signal pathway may be one of the therapeutic targets to promote bone formation. In addition, the osteoblast increment rate and the expression of pathway protein in the AFRD + Dkk1 group were still higher than those in the Dkk1 group, indicating that even under the inhibition of Dkk1, AFRD can still play a small role in the Wnt/β-catenin signal pathway.
In addition, we also explore status of osteogenic-related proteins, including COL1A1, BMP-2 and OPN, induced by AFRD. COL1A1 is responsible for the synthesis of type 1 collagen, thus ensuring that bones and cartilage are resistant to tension, shear and compression [42]. Abnormal collagen production can lead to bone-related diseases, such as Paget disease and osteoporosis [43]. BMP-2 is highly involved in inducing mesenchymal cells to differentiate into osteoblasts and promoting osteoblasts to produce bone matrix [44]. OPN can stimulate osteoblast adhesion, proliferation and calci cation, and mediate the changes of bone metabolism caused by mechanical stress [45]. Our experimental results showed that the expression of three proteins increased signi cantly in AFRD group, which further con rmed the role of AFRD in promoting osteogenesis.

Conclusions
AFRD can promote the growth rate and mineralization of bone graft in the induced membrane, which is related to the fact that AFRD can promote osteoblast differentiation, mineralization and expression of osteogenesis-related proteins partly by activating Wnt/β-catenin signal pathway. However, whether AFRD also upregulates other signal pathways, and whether there is a synergistic effect between these signal pathways is the direction of our follow-up trails.