Wnt/β-Catenin Signaling Pathway Regulates Osteogenesis for Breast Cancer Bone Metastasis: Experiments in an In Vitro Nanoclay Scaffold Cancer Testbed

ABSTRACT: Breast cancer shows a high affinity toward bone, causing bone-related complications, leading to a poor clinical prognosis. The Wnt/β-catenin signaling pathway has been well- documented for the bone regenerative process; however, the regulation of the Wnt/β-catenin pathway in breast cancer bone metastasis is poorly explored. Here, we report that the Wnt/β- catenin signaling pathway has a significant effect on osteogenesis during breast cancer bone metastasis. In this study, we have created a 3D in vitro breast cancer bone metastatic microenviron- ment using nanoclay-based scaffolds along with osteogenically differentiated human mesenchymal stem cells (MSCs) and human breast cancer cells (MCF-7 and MDA-MB-231). The results showed upregulation in expressions of Wnt-related factors (Wnt- 5a, β-catenin, AXIN2, and LRP5) in sequential cultures of MSCs
with MCF-7 as compared to sequential cultures of MSCs with MDA-MB-231. Sequential cultures of MSCs with MCF-7 also showed higher β-catenin expression on the protein levels than sequential cultures of MSCs with MDA-MB-231. Stimulation of Wnt/β-catenin signaling in sequential cultures of MSCs with MCF-7 by ET-1 resulted in increased bone formation, whereas inactivation of Wnt/β-catenin signaling by DKK-1 displayed a significant decrease in bone formation, mimicking bone lesions in breast cancer patients. These data collectively demonstrate that Wnt/β-catenin signaling governs osteogenesis within the tumor- harboring bone microenvironment, leading to bone metastasis. The nanoclay scaffold provides a unique testbed approach for analysis of the pathways of cancer metastasis.

1. INTRODUCTION
Breast cancer shows a high affinity toward bone, causing bone- related complications, leading to poor clinical prognosis.1 Approximately 80% of breast cancer patients die within 5 years after the primary cancer has metastasized to the bones.2 The survival rate of the patients with breast cancer metastasized to bones is highly influenced by the tumor stage at the time of diagnosis. The interactions between the bone microenviron- ment and breast cancer cells have been shown to contribute toward the development of bone metastases.3 Breast cancer cells colonize within the bone marrow and impede bone remodeling processes either by activation of osteoclast differentiation or by the promotion of osteoblast activity.4,5 Unraveling the metastatic cascade at the cellular level would help develop new and effective therapeutic approaches to detect bone metastases at early stages to improve patient survival rates. To better understand the metastatic cascade within the bone, in vivo mouse models and two-dimensional (2D) in vitro models have been developed. In vivo mouse models offer a natural three-dimensional (3D) microenviron-ment,6 where human-derived cancer cells can be injected or human bone fragments can be implanted into immune- deficient animals to create a metastatic xenograft;7 however, these animal models are expensive, do not possess an immune system, and exhibit low efficiency in generating metastases.8 In contrast to in vivo models, 2D in vitro models offer simplicity and a low-cost platform for cancer research. There have been a few studies to recapitulate cancer-induced bone metastasis using 2D in vitro models. Breast cancer cells were co-cultured with osteoblasts to evaluate the role of osteoblasts in proliferation and migration of cancer cells;9,10 also, condi- tioned media have been extensively used to assess the paracrine effect of cancer-cell-secreted cytokines.11−13However, 2D in vitro models simplify the inherent 3Dmicroenvironment of cancer cells due to a lack of spatial cues.In contrast, 3D in vitro models represent a more physiologically relevant environment, better at recapitulating the in vivo interactions between cancer cells and the bone microenviron- ment, leading to improved predictions.14 There have been a lot of studies focusing on the development of 3D in vitro models for studying breast cancer bone metastases to date.

Recent studies have shown a significant role of the hydroxyapatite (HAP) in regulating breast cancer bone metastasis using porous poly lactide-co-glycolide (PLG) scaffolds.15 In another study, a 3D model was created by depositing bone-related proteins from differentiated mesenchymal stem cells on chitosan/HAP scaffolds to study metastatic abilities of breast cancer cells.16 Researchers have also developed 3D silk fibroin- based scaffolds to study interactions between cancer cells and osteoblasts.17 Although most studies are based on static experiments in Petri dishes, some works on the interactions between 3D bone tissue and cancer cells in real time using a bioreactor system18 are also reported. Other materials reported for the development of scaffolds include polyurethane foams.19 Although these studies have been useful in elucidating some individual characteristics of the bone environment, further studies need to be carried out to develop a true bone-mimetic environment. To this end, we have developed a 3D in vitro model of breast cancer bone metastasis using nanoclay-based scaffolds along with osteogenically differentiated human mesenchymal stem cells (MSCs) and human breast cancer cells (MCF-7 and MDA-MB-231). We have investigatedengineered nanoclays extensively20−22 in the context of asimulation-based design of polymer clay nanocomposites and also reported their application for bone tissue engineering.23 We have previously reported osteogenic induction of MSCs and MSC-mediated mineralization via vesicular delivery on nanoclay based scaffolds and composites without the use of osteogenic supplements.23,24 Nanoclay based scaffolds are highly porous with pore sizes of 100−300 μm and possess a compressive modulus of 2.495 MPa,23 which is higher than themodulus (∼1.1 kPa) of MSCs,25 however, it is lower than that of tissue culture polystyrene (TCPS) (∼5 GPa).26 The reasons mentioned above make the nanoclay-based scaffold an ideal candidate for recapitulating a bone-mimicking microenviron-ment.

Recently, we have reported late stage prostate and breast cancer pathogenesis to bone using a sequential culture of MSCs with human prostate and breast cancer cells on nanoclay-based scaffolds, respectively.27−31 We have also evaluated the feasibility of the developed 3D in vitro model of breast cancer bone metastasis for spectral biomarker discovery using FT-IR spectroscopy.32 Commercially available nanoclays have also been used for preparation of bone tissue engineering scaffolds.33−36 Overall, the testbed developed using the nanoclays enables many evaluations of cancer metastasis including changes in morphological and critical pathways at metastasis to bone.The Wnt/β-catenin pathway plays a crucial role in the bone regenerative process including fracture healing.37,38 Wnt signaling is mediated by activation of low-density lipoprotein (LDL) receptor-related protein 5/6 (LRP-5/6) and frizzled (FZD) by secreted Wnt ligands.39 In the absence of a Wnt signal, β-catenin is phosphorylated by casein kinase 1 (CK1) and the glycogen synthase kinase-3β (GSK-3β)/adenomatous polyposis coli (APC)/Axin complex, leading to ubiquitination and proteasome degradation of β-catenin.40−42 In the presence of a Wnt signal, the coreceptor LRP5/6 forms a complex withWnt-bound FZD, which in turn promotes nuclear trans- location of β-catenin and activates transcription of target genes, such as c-myc, cyclin D1, and RUNX2.43−45 Although the Wnt/β-catenin pathway is well studied for osteogenesis,46−56 little is known about regulation of the Wnt/β-catenin pathwayby breast-cancer-derived factors such as dickkopf-related protein 1 (DKK-1) and endothelin-1 (ET-1) during osteo- genesis in breast-cancer-induced bone metastasis. DKK-1, an inhibitor of canonical Wnt signaling, operates by sequestering LRP-5/6 from FZD and thus blocks Wnt/β-catenin pathway mediated osteogenesis.57−59 Increased DKK1 expressions have been associated with osteolytic bone metastases mediated by MDA-MB-231.60 Breast cancer cells have also been shown to control the secretion of DKK1 during bone metastasis via ET-1.ET-1 has been shown to downregulate secretion of DKK-1, causing increased osteoblastic activity and bone formation.61 Researchers have shown that breast cell lines T47D, MCF-7, and ZR75-1 secreted ET-1 and caused osteoblastic metastases, while osteolytic breast cancer cell line MDA-MB-231 did not secrete ET-1.62In the present study, we hypothesized that the Wnt/β- catenin pathway might be responsible for osteogenesis in breast cancer bone metastasis. Here, we aim at investigating the role of the Wnt/β-catenin pathway on osteogenesis in breast cancer bone metastasis by utilizing the 3D in vitro model.

2.MATERIALS AND METHODS
Preparation of 3D Scaffolds of Polycaprolactone (PCL) and in Situ HAPclay. Preparation of PCL/in situ HAPclay scaffolds is described elsewhere.23 In brief, PCL/in situ HAPclay scaffolds wereprepared by adding 10% in situ HAPclay to polycaprolactone (PCL), using a freeze-extraction method. HAPclay was prepared by in situ biomineralization of hydroxyapatite (HAP) into intercalated nanoclay galleries resulting due to modification of Na-MMT clay (Clay Minerals Respiratory at the University of Missouri, Columbia) by 5- aminovaleric acid as described previously.63−65Cell Culture. Human MSCs (Lonza) were cultured inMSCGM Bulletkit medium (Lonza, PT3001). Human breast cancer cell lines MCF-7 and MDA-MB 231 (shortened as MM 231; ATCC) were cultured in Eagle’s Minimum Essential Medium (EMEM), 10% FBS, 0.01 mg/mL human recombinant insulin, and 1% P/S and 90% Dulbecco’s Modified Eagle medium/Nutrient Mixture F-12 DMEM- F-12(1:1), 10% FBS, and 1% P/S, respectively. Prior to cell culture, PCL/in situ HAPclay scaffolds (12 mm diameter and thickness of 3 mm) were sterilized under UV light for 45 min, immersed in 70% ethanol for 12 h, washed in phosphate buffered saline (PBS), and kept in a humidified 5% CO2 incubator at 37 °C immersed in culture medium for 24 h. For the sequential culture, MSCs were seeded at a density of 5 × 104 cells per scaffold (1.47 × 105 cells/cm3 of scaffold) and cultured for 23 days to obtain bone extracellular matrix (ECM) formation on scaffolds. Further, breast cancer cells MM 231/MCF-7 were seeded on the newly formed bone ECM in the 3D scaffolds at a density of 5 × 104 cells per scaffold (1.47 × 105 cells/cm3 of scaffold) and maintained in 1:1 MSCs and breast cancer cell medium (Scheme 1).Cellular Morphology. Cell-seeded scaffolds were fixed with 2.5% glutaraldehyde, dehydrated in a graded series solution (10%, 30%, 50%, 70%, and 100%), and dried in hexamethyldisilazane. Then, the samples were gold sputter coated and observed in SEM (JEOL JSM-6490LV) to evaluate cell morphology.

Gene Expression Studies. RNA was isolated from cell- seeded scaffolds using a Direct-zol RNA MiniPrep kit (Zymo Research). Then, cDNA was synthesized using 2 μgof RNA, random primers, and M-MLV reverse transcriptase (Promega) in a thermal cycler (Applied Biosystems). Real-time polymerase chain reaction (PCR) experiment was performed using a 7500 Fast Real-Time PCR system (Applied Biosystems). Forward primer, reverse primer, SYBR Green dye, and cDNA were added to make a final volume of 20 μL and run using a thermal profile with a holding stage (2 min at 50 °C, 10 min at 95 °C) and a cycling stage (40 cycles of 15 s at 95 °C, and 1 min at 60 °C). The mRNA expressions of alkaline phosphatase (ALP), runt related transcription factor 2 (RUNX2), osteocalcin (OCN), Wnt5a, β-catenin, FZD4, AXIN2, and LRP5 were quantified and normalized to housekeeping gene glyceraldehyde-3-phosphate- dehydrogenase (GAPDH). Target gene expressions were analyzed using the comparative Ct method (2−ΔΔCt).Table 1 lists the sequence of primers used.Immunofluorescence Staining. Cell-seeded scaffolds were washed in PBS and fixed in 4% paraformaldehyde (PFA) for 30 min, followed by permeabilizing with 0.2% TritonX-100 in PBS for 5 min. Further, the samples were blocked with 0.2% fish skin gelatin (FSG) for 45 min, followed by incubation with the primary antibody overnight at 4 °C. RUNX2 (Abcam), OCN (Santa Cruz), E-cadherin (Abcam), and β-catenin (Santa Cruz) antibodies were diltuted in a blocking buffer (0.2% FSG in PBS with 0.02% Tween20) at a dilutionof 1:100. Finally, Alexa Flour 488/647 conjugated secondary antibodies corresponding specifically to the origin of the used primary antibodies were added at 1:250 dilution and incubated for 45 min at25 °C. The nuclei were counterstained with DAPI. The stained samples were observed under confocal microscope (Zeiss Axio Observer Z1 LSM 700).

The amount of OCN (Invitrogen), DKK- 1(RayBiotech), and ET-1 (RayBiotech) released in the cell culture media was determined using the ELISA assay kits following the manufacturer’s instructions. To this end, we kept the cell-seeded scaffolds in a serum-free medium for 48 h before protein harvest.Cell-seeded scaffolds were fixed with 4% paraformaldehyde for 30 min and washed with PBS two or three times to remove residual fixative agent. Further 2% Alizarin Red S (ARS) staining solution of 50 μL was dropped on the washed scaffold and kept for 2 min and 30 s. After 2 min and 30 s, the scaffold was washed using PBS many times in order to remove the unbound stain and dried at room temperature for imaging. Photos were taken at 20× magnification. For quantification, stained samples were immersed into 700 μL of 10% acetic acid solution and incubated at room temperature for 5 min to solubilize the stain. The absorbance of the released Alizarin Red S stain was measured at 405 nm.Total protein was extracted from cell-seeded scaffolds using a mammalian cell extraction kit following the manufacturer’s instructions (Biovision), and protein concen- tration was estimated using Bradford’s reagent (Thermo Fisher). Further, 100 μg of total protein per sample was loaded per lane and resolved through SDS-polyacrylamide gel electrophoresis in 8−16% separating gel. The resolved proteins were transferred onto a 0.2 μm nitrocellulose membrane (Bio-Rad) and blocked for 1 h with 5% bovine serum albumin (BSA) in PBS with 0.05% Tween-20 (PBST) (Alfa Aesar). The membranes were washed with PBST and incubated with primary mouse monoclonal antibody against human β-catenin (Santa Cruz; 1:250 dilutions) at 4 °C overnight. The protein expression of β-catenin was studied relative to β-actin, which was usedas a loading control (mouse monoclonal against human β-actin, Abgent; 1:1000 dilutions). The blots were washed and incubated with horse radish peroxidase (HRP)-conjugated secondary goat antimouse IgG (Azure Biosystems, 1:10000 dilutions) for 1 h. After a brief washing step, the blots were visualized using an enhanced chemiluminescence method (Amersham ECL Prime Western Blotting Detection Reagent, GE Healthcare) through a gel documentation system (Alpha Innotech FluorChem FC2 Imaging System).All the experiments were carried out in triplicate (n = 3) unless otherwise mentioned, and the data are presented as mean ± standard derivation. The statistical significance (p value) among multiple comparisons was determined using one-way ANOVA followed by the appropriate post hoc test, while statistical significance (p value) between two groups was determined using the Student’s unpaired t test, using GraphPad Prism v7.04.

3.RESULTS
Osteogenic Differentiation of MSCs on Nanoclay- Based Scaffolds Is Mediated by Wnt/β-Catenin Path- way. To evaluate the induction of osteogenic differentiation in MSCs on nanoclay-based scaffolds, we examined theexpressions of bone-specific genes (ALP, RUNX2, and OCN). ALP is an early stage osteogenic marker,66 while RUNX2 is a transcription factor that governs the early stages of osteoblastic differentiation.67,68 RUNX2 has been shown to promote the differentiation of mesenchymal cells into immature osteoblasts; however, the levels of RUNX2 reduce when osteoblasts undergo maturation.69,70 ALP expression has also been shown to downregulate during maturation of osteoblasts and formation of an extracellular matrix (ECM).71 OCN is a late stage bone marker, and the expression of OCN increases during osteoblast maturation.72 In accordance with these previous observations, we found increased gene expressions of ALP and RUNX2 at 8 days, indicating induction of osteogenic differentiation of MSCs; however, the expressions of ALP and RUNX2 were down- regulated by ∼2-fold along with a ∼2-fold increase in OCNexpression at 23 days, suggesting maturation of tissue- engineered bone (Figure 1a).We further confirmed gene expression results with immunostaining of RUNX2, OCN, and a quantified release of OCN in the culture medium using ELISA. We noticed intense staining of RUNX2 at 8 days, while it was reduced at 23 days along with intense staining of OCN, as shown in Figure 1c. We also observed a significant increase in the release of OCN (∼20 ng/mL) at 23 days as opposed to 8 days (∼3.38 ng/mL; Figure 1b). Next, we evaluated calcium deposition on scaffolds using ARS staining. ARS showed dispersed andlimited calcium deposition at 8 days, while enhanced calcium deposition was observed at 23 days, which was further confirmed by quantification of the released ARS (Figure 1d). The Wnt/β-catenin signaling pathway has been shown to play a critical role in various stages of osteogenesis.37 In order to evaluate if the Wnt/β-catenin signaling pathway is activated during osteogenesis on the nanoclay-based scaffolds, weanalyzed the expressions of Wnt/β-catenin pathway-specific genes (Wnt5a, β-catenin, AXIN2, FZD4, and LRP5). Wnt5a is a member of the Wnt family that plays a critical role in bone osteogenesis, while AXIN2 is a direct target gene of Wnt ligand binding and activation of the Wnt pathway.73,74 The Wnt/β- catenin pathway suggests Wnt-FZD-LRP complex formation in the presence of the Wnt ligand.

Further, the Wnt-FZD-LRP complex inhibits cytoplasmic degradation of β-catenin while promoting nuclear translocation of β-catenin to initiate transcription of bone-specific genes.44 It has previously been shown that β-catenin regulates early stages of osteogenic differentiation; however, the expression of β-catenin reduces during maturation of bone.75,76 In line with these observations, we noticed upregulated expressions of Wnt-related factors (Wnt5a, β-catenin, AXIN2, FZD4, and LRP5) at 8 days, whilethe expressions of all genes evaluated went down by ∼1.25-fold at 23 days (Figure 2a), which was further substantiated byimmunostaining (nuclear translocation; Figure 2b) andWestern blot of β-catenin (Figure 2c), indicating activation of the Wnt/β-catenin signaling pathway on nanoclay scaffolds during osteogenesis.Regulate Wnt/β-Catenin Pathway, Leading to Bone Metastasis. Breast cancer bone metastasis most often leads to either osteolytic lesions by activation of osteoclast differ- entiation or osteoblastic lesions by promotion of osteoblastic activity, leading to a weakened bone matrix with poor mechanical stability.77In the present study, we chose two breast cancer cell lines MM 231 and MCF-7. We observed very distinct behavior of two breast cancer cells on bone scaffolds, as shown in Figure 3. Sequential culture of MSCs with MCF-7 (MSCs + MCF-7 SC) gave rise to tumoroids with distinguishable cellular boundaries (Figure 3b), while sequential culture of MSCs with MM 231 (MSCs + MM 231 SC) showed the formation of disorganized aggregates of cells (Figure 3a).Further, to evaluate the effect of breast cancer cells on tissue- engineered bone, we analyzed the expression of late-stage bonemarker OCN. We noticed a ∼1.2-fold increase in OCN expression with MSCs + MCF-7 SC, while MSCs + MM 231 SC exhibited a ∼3-fold downregulation of OCN, as compared to MSCs, which was substantiated by immunostaining and quantification of OCN release in culture medium (Figure 4a− c).

In addition, we assessed calcium deposition on sequential culture constructs using ARS and observed intense calcium deposition on MSCs + MCF-7 SC, whereas MSCs + MM 231 SC showed abrogated calcium deposition, compared to MSCs (Figure 4d).Next, we assessed the expressions of Wnt-related factors (Wnt5a, β-catenin, AXIN2, and LRP5) to identify whether the Wnt/β-catenin signaling pathway is involved in osteogenesis during bone metastasis. We noticed a ∼2.2-fold increase in the expressions of Wnt-related factors in MSCs + MCF-7 SC while MSCs + MM 231 SC showed a ∼1-fold downregulation in the expressions of Wnt-related factors, compared to MSCs, indicating upregulation and downregulation of β-catenin inthe sequential culture of breast cancer cells MCF-7 and MM231, respectively (Figure 5a). To further validate our results, we performed Western blot experiments of β-catenin and found results in good agreement with gene expression (Figure 5c).To evaluate whether breast cancer cells can modulate expression of β-catenin in MSCs, we cultured MSCs for another 10 days in (1:1) MSCGM and breast cancer-derived conditioned medium (CM) after bone formation at 23 days.Interestingly, we observed reactivation of β-catenin expression when MSCs were treated with MCF-7 CM, and treating MSCs with MM 231 CM diminished β-catenin expression as shown by immunostaining (Figure 5b).Breast cancer has been shown to contribute to the formation of osteolytic and osteoblastic bone metastases under the influence of DKK-1 and ET-1, respectively.60,62 DKK-1 is an inhibitor of Wnt signaling and operates by sequestering LRP- 5/6 from the receptor FZD and thus blocks Wnt/β-cateninsignaling mediated osteogenesis,57−59 whereas ET-1 has been shown to promote Wnt/β-catenin signaling, leading to osteogenesis.78 To determine whether DKK-1 and ET-1 are involved in the regulation of the Wnt/β-catenin signaling pathway, we evaluated the expressions of these factors. We found upregulated expression of ET-1 in MSCs + MCF-7 SC, whereas MSCs + MM 231 SC exhibited higher levels of DKK-1, which was substantiated by quantifying the release of these factors in culture media using an ELISA assay (Figure 6a,b). Taken together, breast-cancer derived factors DKK-1 and ET-1 seem to regulate the Wnt/β-catenin pathway, leading to bone metastasis.

4.DISCUSSION
It is well-known that metastatic breast cancer cells colonize within the bone marrow and disrupt bone remodeling, either by activation of osteoclast differentiation or by the promotion of osteoblast activity. Given the multifaceted role of the Wnt/ β-catenin pathway in bone regeneration, we hypothesized that regulation of the Wnt/β-catenin pathway might influence thetypes of bone metastasis. Therefore, the primary goal of the study is to understand the influence of the Wnt/β-catenin pathway in osteogenesis during breast cancer mediated bone metastases in vitro. Due to the scarcity of availability of human breast cancer metastasized bone samples and in vivo mouse models for spontaneous breast cancer bone metastasis, we developed a 3D in vitro model using nanoclay-based scaffolds along with osteogenically differentiated MSCs and human breast cancer cells. We have previously reported uniform bone- like ECM formation on nanoclay scaffolds at 23 days when seeded with MSCs.23 In a recent study, we reported that migration of breast cancer cells was significantly enhanced in the presence of bone-mimetic scaffolds.30 Also, growing cancer cells on bone ECM led to the formation of in vivo like tumoroids as shown in Figure 3b. Collectively, these observations demonstrate successful recapitulation of the metastatic condition in vitro on nanoclay scaffolds.To this end, we cultured MSCs for 23 days on nanoclay- based scaffolds to obtain bone-like ECM formation by deposition of an inorganic matrix from osteogenically differ-entiated MSCs (Figure 1d); after this step, we seeded human breast cancer cells on the obtained bone-like tissue model to mimic the in vivo metastatic spreading condition (Scheme 1). In the present study, we confirmed activation of the Wnt/β- catenin pathway during bone formation on nanoclay-based scaffolds (Figure 2).

Activation of the Wnt/β-catenin pathway in MSCs could be attributed to the released silicate ions from nanoclay. The nanoclays are 2:1 phyllosilicates with one octahedral alumina sheet sandwiched between two tetrahedral silica sheets.79 Silicate-containing bioglass and bioceramics have also been reported to stimulate osteoblastic differ-entiation via the Wnt/β-catenin pathway.47,50,54,80 Breast cancer has been shown to form both osteolytic and osteoblastic bone metastases under the influence of DKK-1 and ET-1, respectively.60,62 In the present study, we observed that MSCs+ MCF-7 SC resulted in increased bone formation as shown by increased expression of OCN and intensive calcium deposition (Figure 4c,d). It is noteworthy to mention that we also found enhanced expression of Wnt-related factors (Wnt-5a, β- catenin, AXIN2, and LRP5) in the MSCs + MCF-7 SC, suggesting involvement of the Wnt/β-catenin pathway in regulating osteogenesis during bone metastasis. The activationof the Wnt/β-catenin pathway has been well-documented in the literature for osteoblast differentiation in vitro and bone formation in vivo.38,81,82 It is hence not surprising that the Wnt/β-catenin pathway may play an important role in increased bone formation during bone metastasis. We further noticed that MSCs + MCF-7 SC expresses higher levels of ET- 1 (Figure 6a,b), and previous studies have demonstrated that up-regulation of ET-1 resulted in osteoblastic metastases.62 In a recent study, ET-1 has been shown to promote osteogenic differentiation of periodontal ligament stem cells via the Wnt/ β-catenin pathway.78 In another study, ET-1 has been shown to activate β-catenin signaling through the endothelin-A receptor (ETAR)/β-arrestin complex in ovarian cancer metastasis.83 Hence, it seems likely that up-regulation of ET-1 may contribute toward activation of the Wnt/β-catenin pathway in MSCs + MCF-7 SC, leading to increased bone formation.Next, the inactivation of the Wnt/β-catenin pathway by breast-cancer-derived DKK-1 resulted in inhibition of bone formation in MSCs + MM 231 SC, as indicated by abrogated calcium deposition and reduced OCN expression (Figure 4c,d).

Our results are in good agreement with previous studies on the inhibitory effect of DKK-1 on osteoblast differentiation in vitro and bone formation in vivo.37,60,84−86 Hence, inhibited bone formation in our model is mediated through down- regulation of osteoblast differentiation by DKK-1 induced inactivation of the Wnt/β-catenin pathway. In summary, our 3D in vitro model allows investigation of the interactionsbetween breast cancer cells and bone and facilitates quantification of bone formation with different breast cancer cells. Most importantly, the model exhibited both inhibited and excessive bone formation with different cell lines, mimicking bone lesions observed in many breast cancer patients. Therefore, this model is suitable for studying cellular signaling mechanisms underlying the change during bone metastasis. However, it should be noted that our model does not take into consideration the early phases of bone metastases, such as local invasion, intravasation, dissemination via circulation, and extravasation. Thus, our data suggested that the Wnt/β-catenin pathway regulates osteogenesis at the secondary bone site. Future studies are planned to evaluate whether this pathway also governs in the early phases of breast cancer bone metastasis.

5.CONCLUSION
The present study suggests that osteogenesis in a 3D in vitro model of breast cancer bone metastasis is mediated by the Wnt/β-catenin signaling pathway. MCF-7 cells secrete ET-1, which promotes osteoblastic differentiation via Wnt/β-catenin signaling and induces bone-forming activity, leading to increased bone formation. Inactivation of Wnt/β-catenin signaling by MM 231 secreted DKK-1 leads to inhibited bone formation through down-regulation of osteoblast differ- entiation. Further, the 3D in vitro model exhibited mixed bone lesions with different cell types, mimicking bone lesions observed in breast cancer patients. However, it should be noted that this model as presented does not describe the very early stages of breast cancer bone metastasis. More studies should be carried out to understand BC-2059 the underlying molecular mechanism of Wnt-dependent pathways in tumor-harboring bone microenvironment before developing any therapeutic strategies for bone metastasis. The in vitro breast cancer model thus presents a viable testbed for studying metastasis.