Role of Fzd6 in Regulating the Osteogenic Differentiation of Adipose-derived Stem Cells in Osteoporosis

Objective: Although it has been demonstrated that adipose-derived stem cells (ASCs) from osteoporosis mice (OP-ASCs) exhibit impaired osteogenic differentiation potential, the molecular mechanism has not yet been elucidated. We found that Fzd6 was decreased in OP-ASCs compared with ASCs. This study investigates the effects and underlying mechanisms of Fzd6 in the osteogenic potential of OP-ASCs. Methods: Fzd6 expression in ASCs and OP-ASCs was measured by PCR gene chip. Fzd6 overexpression and silencing lentiviruses were used to evaluate the role of Fzd6 in the osteogenic differentiation of OP-ASCs. Real-time PCR (qPCR) and western blotting (WB) was performed to detect the expression of Fzd6 and bone-related molecules, including runt-related transcription factor 2 (Runx2) and osteopontin (Opn). Alizarin red staining and Alkaline phosphatase (ALP) staining was performed following osteogenic induction. Microscopic CT (Micro-CT), hematoxylin and eosin staining (H&E) staining, and Masson staining were used to assess the role of Fzd6 in osteogenic differentiation of osteoporosis (OP) mice in vivo. Results: Expression of Fzd6 was decreased signicantly in OP-ASCs. Fzd6 silencing down-regulated the osteogenic ability of OP-ASCs in vitro. Overexpression of Fzd6 rescued the impaired osteogenic capacity in OP-ASCs in vitro. We obtained similar results in vivo. Conclusions: Fzd6 plays an important role in regulating the osteogenic ability of OP-ASCs both in vivo and in vitro. Overexpression of Fzd6 associated with the Wnt signaling pathway promotes the osteogenic ability of OP-ASCs, which provides new insights for the prevention and treatment of OP.


Introduction
OP is a disease characterized by abnormal bone metabolism, trabecular bone structure disorders, and bone loss, which can lead to increased bone fragility [1,2]. As the homeostasis between osteoblasts and osteoclasts is destroyed, bone resorption by osteoclasts becomes faster than bone formation by osteoblasts, which leads to a decrease in bone mass and an increase in bone fragility, thus increasing the risk of fracture [3,4]. OP is more prevalent in women than men, especially in postmenopausal women, whose bone metabolism is accelerated because of estrogen de ciency [5]. OP seriously affects the quality of life of postmenopausal women, and nearly 50% of women may experience osteoporotic fractures after menopause [6]. At present, the number of people suffering from OP is increasing, and OP has become a major disease worldwide that seriously impacts people's health [7,8]. Many researchers are currently focusing on the issues of reducing bone damage and repairing bone defects in OP.
In recent years, bone tissue engineering has become a topic of interest for repairing bone defects and has explored the combination of seed cells, biological scaffold materials, and bioactive molecules to promote osteogenesis [9,10]. Bone marrow mesenchymal stem cells (BMSCs) are attractive for the treatment of osteoporotic bone defects, but in elderly patients with OP, BMSCs tend to differentiate into adipocytes rather than osteoblasts [11]. Furthermore, bone marrow aspiration to obtain BMSCs is an extremely painful process, and the number of cells obtained is usually low [12,13]. ASCs can also differentiate into osteoblasts and chondrocytes [14][15][16][17]. Because of their ability to multi-directionally differentiate, their extensive proliferative capacity, and low immunogenicity, ASCs have become an attractive source of seed cells for regenerative medicine and tissue engineering [18][19][20][21][22]. Therefore, researchers hope to transplant autologous ASCs from osteoporotic patients for the treatment of bone defects. However, the osteogenic differentiation potential of OP-ASCs is signi cantly impaired compared with ASCs [23]. Moreover, the mechanisms associated with the reduced osteogenic capacity of OP-ASCs are not clear.
The Wnt signaling pathway plays an important role in the regulation of bone differentiation and bone formation by ASCs [15,24]. Wnt proteins can activate downstream pathways by binding to the corresponding Frizzled (FZD) receptors, including Wnt/β-catenin, Wnt/planar cell polarity, and Wnt/Ca 2+ [25]. Due to its unique seven transmembrane structure, FZD is the link to the Wnt signaling pathway from the extracellular to the intracellular, and it is an important part of Wnt signaling pathway [26,27]. Among the members of the FZD family, Fzd6 is broadly expressed in various tissues [28]. Cvjetkovic N et al [29] found that Fzd6 down-regulates the activity of the TCF/LEF complex through the Wnt/Ca 2+ signaling pathway to inhibit the canonical Wnt signaling pathway. Abuna RPF et al [30] found that Fzd6 enhanced the high osteogenicity of nanotopography by activating the canonical Wnt/β-catenin signaling pathway. Olivares-Navarrete et al [31] demonstrated that the up-regulation of Fzd6 associated with the noncanonical Wnt signaling pathway in osteoblasts to improve osteogenic properties in vitro. However, the regulatory mechanism of Fzd6 in the differentiation of OP-ASCs into osteoblasts has not been reported.
In this study, gene chip data suggested that the expression of Fzd6 was decreased in OP-ASCs after the induction of osteogenesis. Based on the literature and our previous work, we hypothesized that Fzd6 may have a positive impact on the physiological functions of OP-ASCs through the Wnt signaling pathway.
This study aimed to explore the effects of Fzd6 on the osteogenic ability of OP-ASCs through the overexpression or silencing of Fzd6. Moreover, Fzd6 can be used as a speci c therapeutic target to treat osteoporotic bone defects. Ovariectomy (OVX) was performed when the weight of each mouse reached 20 g. The mice were weighed and anaesthetized by intraperitoneal injection of 1% sodium pentobarbital (30 mg/kg). After the removal of the bilateral ovaries, the abdominal muscle tissue and skin were sutured.

Isolation and Culture of OP-ASCs
OP-ASCs were obtained from the subcutaneous fat at the inguinal sites of the op mice. Adipose tissue was washed with 10% penicillin-streptomycin PBS, 5% penicillin-streptomycin PBS, and penicillinstreptomycin-free PBS in turn. Then, adipose tissue were seeded in 25 cm 2 culture asks with modi ed eagle's medium and were cultured in an atmosphere of 5% CO 2 at 37 °C for 5 days. Adherent adipose tissue were cultured in a monolayer and non-adherent adipose tissue were removed. The medium was substituted for fresh medium every 3 days and cells were subcultured upon reaching 80% con uence.
Cells were passaged three times in the subsequent experiments.

PCR Gene Chip Screening of Wnt Signaling Pathway Gene Changes in ASCs and OP-ASCs
OP-ASCs and ASCs were cultured. Total RNA was extracted from the LiCl group (50 mg/mL LiCl), DKK-1 group (0.1 µg/mL DKK-1), and control group at day 1 and day 3. Genes related to the Wnt signaling pathway were then screened by Wnt signaling pathway gene chip combined with qPCR.

Expression of Fzd6 mRNA and Protein in OP-ASCs
The mRNA expression of Fzd6 was assessed by qPCR and protein expression of FZD6 was detected by WB.

Construction of Lentivirus-mediated Overexpression and Silencing of Fzd6
The construction, sequencing, puri cation and titer determination of lentivirus vectors were completed by Heyuan Biotechnology Co., Ltd. All lentivirus vectors contained green uorescent protein (GFP) and puromycin resistance genes. Lentivirus-mediated overexpression of Fzd6 carried 3-Flag, while the other groups did not carry 3-Flag. 3-Flag can be used to verify the successful transfection of lentivirus. The e ciency of successful transfection was veri ed by uorescence microscopy.

Determination of Lentivirus Multiplicity of Infection (MOI)
The lentiviruses evaluated included Fzd6(+), Fzd6(-), shRNA-1, shRNA-2, shRNA-3, and shRNA(-). The MOI value was set using the following gradient: 0, 20, 40, 60, 80, 100. The volume of virus solution needed in each group was calculated according to the following formula: virus volume = cell number × MOI/virus titer. After transfection for 72 h, the morphology and uorescence expression of the cells was observed and recorded by uorescence microscopy to determine the optimal MOI value for each lentivirus.
2.2.7 Acquisition of OP-ASCs Stably Transfected with Lentivirus-mediated Overexpression or Silencing of Fzd6 Lentivirus transfection was performed when the optimal MOI value was 80. After 72 h of transfection, stably transfected cells were selected with 0.6 µg/mL of puromycin.

Cell Viability Assays
Second-passage OP-ASCs (5 × 10 5 cells per 1 mL) and BCP were co-cultured in 96-well plates. Cell proliferation was measured by cell counting kit-8 (CCK-8, Solarbio, China) at 1 d, 3 d, 5 d, and 7 d. The absorbance of each sample was measured at 450 nm by a microplate reader (Spectra Thermo, Switzerland).

Proliferation and Adhesion of OP-ASCs on BCP was Detected by Scanning Electron Microscope (SEM)
BCP was placed in 24-well plates and 1 mL second-passage OP-ASCs cell suspension (5×10 5 cells per mL) was added to the surface of the BCP in each well. 24-well plates were cultured in 5% CO 2 at 37°C.
Samples were xed with paraformaldehyde on the 3rd and 7th day of culture. Alcohol gradient dehydration and critical point drying were then performed and cells were sprayed with gold for SEM observation.

Osteogenic Induction
OP-ASCs were seeded onto 6-well plates at a density of 5×10 4 cells/mL. The medium was changed to osteogenic induction medium (Cyagen, Guangzhou, China) when OP-ASCs reached a fusion of 80%.

Alkaline Phosphatase Staining (ALP) and Alizarin Red Staining
After osteogenic induction for 7 days, OP-ASCs transfected with Fzd6 lentivirus were rinsed with PBS and xed with 4% paraformaldehyde at 4°C for 30 minutes. ALP activity was detected using an Alkaline Phosphatase Assay Kit (Beyotime, Shanghai, China).
After 14 days of osteogenic induction, OP-ASCs were washed three times with PBS and xed with 4% paraformaldehyde at room temperature for 30 minutes. Cells were then rinsed three times with PBS and incubated in 0.1% Alizarin red staining for 15 minutes at 37°C.

Extraction of RNA and qPCR
Total cellular RNA was isolated using the Total RNA Extraction Kit (Tiangen, Shanghai, China) and subsequently reverse transcribed into cDNA using the RevertAid First Strand cDNA Synthesis Kit (Thermo, Waltham, USA). All primers in this study were synthesized by Shenggong Bioengineering Co., Ltd (Shanghai China). The speci c primer sequences were as follows: Gapdh, (forward) 5′-GGGTGAAGGTCGGTGTGAACG-3′ and (reverse) 5′-CTCGCTCCTGGAAGATGGTG-3′; Fzd6, (forward) 5′-AGACAACATTAGCGGCGTTT-3′ and (reverse) 5′-AGAGGAGAGACAGCCCAACA-3′; Runx2, (forward) 5′-CCGAACTGGTCCGCACCGAC-3′ and (reverse) 5′-CTTGAAGGCCACGGGCAGGG-3′; Opn, (forward) 5′-GGATTCTGTGGACTCGGATG-3′ and (reverse) 5′-CGACTGTAGGGACGATTGGA-3′. The qPCR reaction was in accordance with the instructions of QuantiNova SYBR Green PCR Kit. The reaction system was added to qPCR 96-well plates. The reaction liquid was collected at the bottom of the plate by horizontal centrifugation and the plate was then placed in a real-time PCR instrument. After the reaction was completed, the ampli cation curve and dissolution curve of each gene was analyzed by 7900 System SDS software and the relative expression of the target genes was calculated and statistically analyzed.

BCP Combined with OP-ASCs was Transplanted into a Mouse Model of Critical-sized Calvarial Defects
OP-ASCs transfected with Fzd6 overexpression lentivirus, silencing lentivirus, or untransfected lentivirus were prepared and cultured in osteogenic induction medium. 1 mL cell suspension (5×10 5 cells per 1 mL) was added to the surface of the BCP in 24-well plates. 24-well plates were cultured in a CO 2 incubator for 48 h. The osteoporosis mice were treated with prone xation, skin preparation, and disinfection at the top of the skull. The skin was cut at the midline of the skull, the skull was exposed, and the periosteum was bluntly separated. Then, full-thickness defects of 4 mm were created symmetrically. The BCP combined with OP-ASCs was transplanted into the skull defect area and the operation area was then sutured. After 4 and 8 weeks, mice were euthanized and the skull specimens were obtained.

Micro-CT, H&E staining, and Masson staining
Micro-CT scans were performed on mice with skull defects at 4 and 8 weeks. Then, tissues samples of mice skull defect were stained with H&E and Masson staining.

Statistical Analysis
All experiments were repeated a minimum of three times independently, and representative data are provided as mean ± the standard deviation (SD). Differences between two groups were analyzed by student's independent t-test, while differences between multiple groups were analyzed by one-way analysis of variance (ANOVA). Statistical analysis was completed using SPSS 20.0 software. Differences were accepted as statistically signi cant if P < 0.05.

Primary Culture and Subculture of OP-ASCs
The OP-ASCs of generation P0 were passaged to generations P1 and P2 (Fig. 1A). The cell morphologies remained stable and uniform.

Osteogenic Differentiation Potential of OP-ASCs
Second-passage OP-ASCs were cultured in 6-well plates. After 14 days of induced osteogenesis, many mineralized nodules were observed (Figure. 1B). The cells were tightly arranged and the cell morphology was polygonal or oval.

The Results of Gene Chip
Gene chip data showed that many genes related to the Wnt signaling pathway changed after osteoinduction in ASCs and OP-ASCs. Interestingly, Fzd6 expression was signi cantly decreased (Fig. 2).

Gene and Protein Expression of Fzd6 in OP-ASCs
Fzd6 gene expression was decreased in OP-ASCs (Fig. 3A). The protein expression of FZD6 was also signi cantly decreased in OP-ASCs (Fig. 3B).

WB Veri ed the Successful Transfection of Fzd6 Overexpression or Silencing Lentivirus
While the overexpression group contained 3-FLAG tag protein, no 3-FLAG tag protein was found in the other groups, which could support that Fzd6 lentivirus was successfully transferred into OP-ASCs and integrated 3-Flag into the OP-ASC genome. The protein expression of FZD6 in the overexpression group was signi cantly higher than in the control group, while in the [shRNA(+)-2] and [shRNA(+)-3] groups, the expression of FZD6 was signi cantly lower. The expression of FZD6 was no statistically signi cant difference in [Fzd6(-)] and [shRNA(-)] compared with the control group (Fig. 5B).

CCK-8 Assay for the Detection of the Proliferation of OP-ASCs and Cytotoxicity of BCP
OP-ASCs and BCP were co-cultured for 1 d, 3 d, 5 d, and 7 d and the proliferation of OP-ASCs and cytotoxicity of BCP were detected by CCK-8 assay. The results (Fig. 8A) showed that the number of cells gradually increased as the length of co-culture time increased. These results indicated that OP-ASCs could grow and proliferate in an environment with BCP and that BCP had no cytotoxic effect on OP-ASCs.

Proliferation and Adhesion of OP-ASCs on the BCP Surface was Detected by SEM
The proliferation and adhesion of OP-ASCs on the surface of BCP was detected by SEM. The results showed that OP-ASCs grew adherently on the surface and pores of BCP. On day 3 (Fig. 8B) and day 7 ( Fig. 8C) of co-culture, the number of cells increased and the extracellular matrix was gradually connected into pieces.
3.10 Micro-CT was Used to Detect the Osteogenic Capacity of OP-ASCs in Vivo Four weeks after the transplantation of OP-ASCs and BCP in the skull defect area, three-dimensional reconstruction of Micro-CT scan showed that the green part of BCP was the new bone matrix. The amount of green bone matrix observed was signi cantly higher in [Fzd6(+)] and was signi cantly lower in [shRNA(+)] (Fig. 9A). Through sagittal, coronal, and BV/TV analysis (Fig. 9C), we found that the osteogenic ability of [Fzd6(+)] was signi cantly increased, while the osteogenic ability of [shRNA(+)] was signi cantly decreased.
Eight weeks after transplantation of OP-ASCs, three-dimensional reconstruction showed that the amount of green bone matrix observed was signi cantly higher than at four weeks, and the amount of green bone matrix observed was signi cantly higher in [Fzd6(+)] (Fig. 9B). The results showed that the osteogenic ability of OP-ASCs was signi cantly increased in [Fzd6(+)] (Fig. 9C).

H&E and Masson Staining were Used to Detect Osteogenic Ability of OP-ASCs in Vivo
Four weeks after transplantation of OP-ASCs and BCP in the skull defect area, the osteogenic ability of OP-ASCs was detected by H&E and Masson staining. H&E staining showed that the new bone matrix was red, and the staining degree of [Fzd6(+)] was deeper than that of [shRNA(+)]. Additionally, Masson staining showed that the new bone matrix was blue, and the staining degree was deeper in [Fzd6(+)] (Fig. 10A). Eight weeks after the implantation of OP-ASCs, similar results were obtained (Fig. 10B).

Discussion
OP is a major disease worldwide, and the prevention and treatment of osteoporotic bone defects continues to be an area of focus in bone tissue engineering research [32]. At present, composite scaffolds combined with autologous stem cells could effectively repair osteoporotic bone defects [33,34]. ASCs are ideal substitutes for BMSCs because of their larger storage capacity, easier acquisition, richer source, and lower immunogenicity [20,[35][36][37]. ASCs can grow on plastic petri dishes and proliferate actively, and can differentiate into a variety of cells, including osteoblasts, adipocytes, cartilage cells, muscle cells, and ligament cells [38][39][40]. Thus, ASCs may be one of the most promising cell types for bone regeneration in bone tissue engineering [41]. Cho et al found that the therapeutic effect of ASCs was similar to that of estrogen therapy and could prevent bone loss in OVX mice [42]. Mirsaidi et al treated elderly osteoporosis mice with autologous ASCs and found that osteogenesis related molecules were signi cantly increased and the quality of bone trabecula was signi cantly improved [43]. These studies suggest that autologous ASCs transplantation is an effective cell therapy for osteoporosis fractures and bone defects. Therefore, transplantation of OP-ASCs into an area of bone defect may be a potential treatment for osteoporosis patients. However, a previous study found that the osteogenic differentiation potential of OP-ASCs is signi cantly impaired compared with ASCs [23]. Thus, it is worthwhile to improve the osteogenic ability of OP-ASCs and enhance the bone formation in the bone defect area after the transplantation of OP-ASCs.
Nevertheless, there are currently few studies exploring the mechanisms underlying the decreased osteogenic ability of OP-ASCs and strategies to improve the osteogenic ability of these cells.
In this study, we established a mouse model of OP and obtained OP-ASCs. Consistent with previous studies, Alizarin red staining showed that many mineralized nodules were present in these cells [35]. This result demonstrates that OP-ASCs can proliferate, self-renew, and differentiate to bone, and therefore can be used as seed cells for the treatment of osteoporotic bone defects. Next, we screened out differentially expressed genes related to the Wnt signaling pathway in OP-ASCs and ASCs through PCR gene chip.
After analyzing the chip data results, we found that Fzd6 decreased signi cantly in OP-ASCs. Additionally, we found that the gene and protein level of Fzd6 was also reduced in OP-ASCs. Therefore, we aimed to investigate the osteogenic effect of Fzd6 on OP-ASCs and explore changes in the osteogenic capacity of OP-ASCs through the up-regulation and down-regulation of Fzd6.
The Wnt signaling pathway plays an important regulatory role in the process of osteogenic differentiation in ASCs and BMSCs [15,44]. Additionally, the WNT/FZD signaling pathway plays a vital role in the repair, regeneration, and plasticity of adult bone tissue [45][46][47]. Boland GM et al [48] found that Fzd6 was upregulated in the process of osteogenic differentiation in BMSCs in vitro. This result suggested that an increase in Fzd6 expression may activate the Wnt signaling pathway and promote the expression of downstream target genes and thereby promote osteogenesis. Vlacic-Zischke J et al [49] found that on the surface of hydrophilic titanium implants, the osteogenic ability of osteoblasts was signi cantly enhanced and the expression of Fzd6 in the Wnt signaling pathway was signi cantly increased. This result indicated that the enhanced osteogenic ability of osteoblasts may be the result of up-regulation of Fzd6 in the Wnt signaling pathway. However, there are currently few studies that involve the direct regulation of changes in Fzd6 expression and subsequent observation of the effect of these changes on osteogenesis. In the present study, we investigated the effect of the overexpression or silencing of Fzd6 on the osteogenic potential of OP-ASCs. We found that silencing Fzd6 in OP-ASCs decreased the expression of osteogenic-related regulation factor Runx2 and Opn during the process of bone formation. After overexpression of Fzd6, the expression of Runx2 and Opn increased over time. Furthermore, ALP staining and Alizarin red staining showed that silencing Fzd6 reduced the early and late osteogenic differentiation capacities of OP-ASCs. Conversely, overexpression of Fzd6 partially rescued the early and late osteogenic differentiation capacities of OP-ASCs. Correlation analysis determined that Fzd6 could up-regulate the osteogenic differentiation capacity of OP-ASCs in vitro. Therefore, we assert that Fzd6 regulates the osteogenic capacity of OP-ASCs.
It is worth noting that the osteogenic effects of Fzd6 in the regulation of OP-ASCs need to be further con rmed in vivo. Ye X et al [50] wrapped autologous ASCs with calcium alginate gel and implanted them into the distal femur of OVX rabbits and found that the femoral bone mineral density of OVX rabbits in the composite ASCs group increased signi cantly and promoted the formation of new bone. BCP is a new type of biological scaffold material that provides su cient space for cell attachment, growth, and proliferation [51,52]. Rath SN et al found that BMSCs can proliferate on BCP and promote bone formation following osteogenic induction [53]. In the present study, BCP combined with OP-ASCs was transplanted into a mouse critical-sized calvarial defect model for in vivo osteogenesis studies. Micro-CT reconstructed images showed that the formation of new bone increased with time in each group of composite BCP, and the [Fzd6(+)] group had the most new bone during the observed time period. These results indicate that overexpression of Fzd6 can partially rescue the decline in osteogenic ability of OP-ASCs. Furthermore, histological observations of the corresponding tissue samples were also consistent with the results of Micro-CT ,that is, the amount of new bone formation of [Fzd6(+)] is more obvious than the amount of new bone formation of [shRNA(+)]. These ndings suggest that Fzd6 can rescue the reduced osteogenic ability of OP-ASCs, and is a potential therapeutic target for the prevention and treatment of osteoporosis.

Conclusion
This study shows that the overexpression of Fzd6 can promote the osteogenic potential of OP-ASCs and rescue the decline in osteogenesis of OP-ASCs. Silencing Fzd6 effectively reduces the osteogenesis of OP-ASCs. Furthermore, given the role of Fzd6 as a key factor related to osteogenic regulation by the Wnt signaling pathway, Fzd6 could serve as an important target for the treatment of bone defects in OP. Competing InterestsThe authors declare no con ict of interests.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.