Abstract
Background: Most cases of prostate and breast cancer metastasis occur to the bone, and are responsible for the majority of cancer-related deaths. Osteocytes constitute over 90% of adult bone cells. They orchestrate bone remodelling through determining osteoclast activity and affecting osteoblasts. The osteocyte lacuno-canalicular network is also intimately associated with the blood vessel network in the bone matrix. However, the roles of osteocytes in cancer cell invasion and metastasis remain unknown. Materials and Methods: In this study, we investigated the effects of early osteocytes on the behaviour of breast and prostate cancer cells. The proliferation of cultured cells was assessed using the AlamarBlue assay. The electric cell-substrate impedance sensing (ECIS) system was used to measure spreading, attachment and migratory behaviour of cancer cells in response to conditioned medium (CM) from mouse osteocytes. Other cell assays, including in vitro wound healing and transwell migration/invasion assays, were also applied to evaluate the effect of osteocytes on cancer cells. Results: We found that CM from osteocytes from both monolayer and three-dimensional (3D) cultures, stimulated proliferation of DU145 and PC3 prostate cancer cells but not LNCaP cells compared to control medium. Osteocyte CM also stimulated proliferation of MDA-MB-231 and MCF-7 breast cancer cells. However, osteocyte CM promoted the migration and adhesion of PC3 and DU145 in prostate cancer cells but had the reverse effect on PZHPV7, a normal prostate epithelial cell line. In the breast cancer cells studied, osteocyte CM inhibited post-wound migration of MCF-7 and ZR-75.1 cells but not MDA-MB-231 cells. Moreover, osteocyte CM stimulated transwell chemotactic migration of MDA-MB-231 cells but not of MCF-7 and ZR-75.1 cells. Conclusion: Osteocytes play diverse roles in the proliferative and migratory potential of breast and prostate cancer cells that may be associated with cancer-specific bone metastasis and requires further investigation.
In bone, osteocytes are derived from osteoblasts and make up over 90% of bone cells. Osteoblasts become osteocytes after they are trapped in the matrix they secrete and are networked to each other via long cytoplasmic extensions that occupy tiny canals called canaliculi, used for exchange of nutrients and waste (1). It has been postulated that osteocytes act as master orchestrators in bone remodelling and secrete metastatic-mediated factors such as osteopontin (OPN) and receptor activator of nuclear factor kappa-B ligand (RANKL), that have been found up-regulated in breast and lung cancer with osteolytic-like metastasis (2, 3). Osteocytes play a key role in determining osteoclast formation and activity (4) secreting factors, such as RANKL, that signal to osteoclast precursors, and cytokines, such as interleukin-1β (IL1β), and can increase RANKL expression in osteocytes (5). Osteocytes are also able to affect osteoblasts (6). Furthermore, there is evidence that mechanical loading reduces the production of catabolic factors but enhances production of substances such as nitric oxide (NO) and cyclo-oxygenase 2 (COX2) in osteocytes (7, 8). NO is known to be messenger molecule in bone intercellular communication (9), while COX2 facilitates osteoblast proliferation, particularly under inflammatory conditions (10). Therefore, osteocytes are highly involved in the regulation of bone homeostasis through fine-tuning the dynamics of bone formation and resorption (11). Osteocyte apoptosis occurs due to aging, changes of hormone and soluble factors, pharmacological agents, and mechanical forces, and in turn, apoptotic osteocytes initiate targeted bone resorption by recruiting osteoclasts (12). However, the molecular regulation of osteocyte survival and subsequent bone remodelling are not completely characterized (13). Osteocytes also appear to be enriched in proteins that are resistant to hypoxia, which appears to be due to their embedded location and restricted oxygen supply (12). Oxygen tension may regulate the differentiation of osteoblasts into osteocytes, and osteocyte hypoxia may play a role in disuse-mediated bone resorption, which needs to be further investigated (12).
It is known that intensive cancer chemotherapy leads to significant bone loss, although the underlying mechanism remains unclear. A study in rats showed that the chemotherapy anti-metabolite methotrexate caused a 4.3-fold increase in the number of apoptotic osteocytes in tibial metaphysis, and this was accompanied by an increase in the number of tartrate-resistant acid phosphatase-positive bone resorbing osteoclasts and a loss of trabecular bone (14). Murine long bone osteocyte Y4 (MLO-Y4) cells are useful for studying the effects of mechanical stress on osteocyte function and for determining the means whereby osteocytes communicate with other bone cells, including osteoblasts and osteoclasts (15). In cultured MLO-Y4 cells, methotrexate treatment increases caspase-3-mediated apoptosis, which further induces osteoclast formation in primary bone marrow osteoclast precursors. Therefore, methotrexate-induced apoptosis of osteocytes increases osteoclast formation, which could contribute to loss of bone homeostasis in vivo (14). Osteocyte apoptosis also leads to increased secretion of osteocytic interleukin 6 (IL6) and soluble IL6 receptor, which enhances osteoclast precursor adhesion to endothelial cells via intercellular adhesion molecule 1 signaling pathway (16).
Bone metastasis is one of the most common complications in advanced malignancies of three cancer types: breast, prostate and lung. It is currently incurable and causes severe morbidities, including bone pain, hypercalcaemia, pathological fracture, spinal cord compression and subsequent paralysis. Over 90% of patients die within 5 years of diagnosis of bone metastasis from prostate cancer (17, 18). Physical bone remodelling and calcium homeostasis require a balance between osteoclast-mediated bone resorption and osteoblastic-mediated bone deposition. Dysregulation of this balance by aggressive tumour cells leads to either osteoblastic or osteolytic bone lesions. However, the precise molecular mechanisms underlying the predisposition of particular malignancies and subsequent colonisation and development of metastatic tumours in bone have not been elucidated. Previous studies on tumour bone metastasis were too limited to unveil the key process checkpoints of tumour bone metastasis. In particular, osteocytes have been ignored despite being the most abundant type of bone cells within the bone matrix, and playing a key role in determining osteoclast and osteoblast activity (2). Many questions need to be addressed. For instance, it is not known whether aggressive tumour cells divert the transformation of osteoblasts to osteocytes, induce apoptosis of osteocytes under hypoxic microenvironment, or disturb the physical mediating role of osteocytes in the balance of osteoblasts and osteoclasts (through IL6 and RANKL, or unknown signaling pathways). It is also known that hypoxia develops at the central part of a tumour, while osteocytes are enriched with hypoxic-resistant proteins. Furthermore, angiogenesis is considered a critical step of tumour bone metastasis. Osteocytes induce angiogenic properties of endothelial cells through signaling pathways of vascular endothelial growth factor (VEGF)- and mitogen-activated protein kinases (MAPKs) (19). It is, however, not known whether osteocytes play a part in tumour-associated angiogenesis.
In this preliminary study, we evaluated the effect of cells representing early-stage osteocytes (ML-OY4 cells) on the behaviour of breast and prostate cancer cells. We sought to investigate whether these cells influence proliferation, migration and invasion of breast and prostate cancer cells, and may have an implication in tumour bone metastasis.
Materials and Methods
Cell culture. All the human cell lines used for this study were obtained from the American Type Culture Collection (Rockville, MD, USA). The human breast cancer cell lines MDA-MB-231, MCF-7 and ZR-75.1, and prostate cancer cell lines PC3 and DU145 were routinely maintained in Dulbecco's modified Eagle's medium/Ham's F12 (Sigma–Aldrich, Irvine, UK) supplemented with 10% foetal bovine serum (FBS), and 1X penicillin and streptomycin. The PZ-HPV-7 human prostate epithelial cells were maintained with Keratinocyte Serum-Free Medium supplemented with a K-SFM kit containing 0.05 mg/ml bovine pituitary extract and 5 ng/ml human recombinant epidermal growth factor (Life Technologies, Paisley, UK). MLO-Y4 cells, mouse osteocytes at an early stage of differentiation (20), were a kind gift from Professor L. Bonewald, University of Missouri, Kansas City, USA and were cultured on collagen-coated surfaces (rat tail collagen type I, 0.15 mg/ml) in T75 flasks with α-modified essential medium (α-MEM) supplemented with 2.5% heat-inactivated FBS and 2.5% heat-inactivated newborn calf serum. All cells were maintained at 37°C, with 5% CO2 and 95% humidity.
Collection of conditioned medium. Following cultivation of MLO-Y4 cells for 48 h (monolayers 80% confluent), the conditioned medium (CM) was collected by centrifugation at 1000 × g for 10 minutes and filtered using 0.2 μm filters. CM was then aliquoted and stored at −80°C for 1 week before use.
In vitro cell-proliferation assay. The proliferation of the cultured cells was assessed using the AlamarBlue assay which detected redox reduction during cell growth. Briefly, cells were seeded into 96-well plates at a density of 3×103 cells/well and cultured overnight. After treatment with CM from osteocytes or fresh medium as a control for 48 h, 10 μl of the AlamarBlue reagent (Serotec Ltd., Oxford, UK) was added to each well containing 100 μl of fresh culture medium. Cells were then incubated for 3 h at 37°C. The fluorescence was measured with a fluorescence plate reader (Promega, Southampton, UK) with excitation at 544 nm and emission at 590 nm. The fluorescence value was proportional to the number of viable cells.
Electric cell-substrate impedance sensing (ECIS) assay. An ECIS Zθ instrument and 96W1E arrays (Applied Biophysics, Inc., Troy, NY, USA) were applied for the measurement of spreading, attachment and migratory behaviour of cancer cells. Briefly, the 96W1E array plates were stabilized using normal culture medium in a tissue culture incubator for 2 h. Cancer cells were seeded at a density of 4×104 cells/well and cultured for 24 h. Each treatment group was set up with six repetitions. Cells were then treated with control medium or CM from osteocytes. Cell behaviour was monitored using the ECIS system with multiple frequencies. Post-wound migration of cells was measured after electrical wounding at 2,600 μA for 20 s.
In vitro scratch-wound assay. Cancer cells were seeded into a 48-well plate at a density of 1×105 cells per well and allowed to form a monolayer which was then scratched with a pipette tip to create a linear wound approximately 200 μm wide. Migration of the cells into the wounding gap was monitored by serial time-lapse imaging using an EVOS® FL imaging system (Life technologies, Carlsbad, CA, USA) with a ×10 objective. The percentage of wound gap closure was measured using Image J software (National Institutes of Health, Bethesda, MD, USA) and a customized macro.
In vitro Matrigel invasion and migration assays. An in vitro Matrigel invasion assay was used to assess the invasiveness of breast cancer cells. Briefly, transwell inserts (8-μm pores) for 24-well plates were pre-coated with 50 μl/insert of 1 mg/ml Matrigel (BD Bioscience, Oxford, UK), for 1 h at 37°C. Subsequently 2×104 cells were seeded into the upper chamber of each insert in 100 μl basal medium and 650 μl control medium or CM from osteocytes was added to each well (lower chamber) under the inserts. After incubation for 24 h, cells that had penetrated the Matrigel-coated membrane and adhered to other side of the inserts were dissociated with cell dissociation solution (MerkMillipore, Watford, UK) containing 4 μg/ml Calcein AM (eBiosciences, Hatfield, UK) for 1 h at 37°C. The solution containing invaded cells was transferred to a 96-well black-well plate at a volume of 100 μl/well. Invaded cells labelled with Calcein AM were then quantified using a fluorescence plate reader (Promega) with excitation at 490 nm and emission at 520 nm. The migration assay was performed similarly to the invasion assay described above, but in the absence of Matrigel.
Statistical analysis. The statistical comparisons of the data were performed using SPSS version 20 for Windows (SPSS, Chicago, IL, USA). The significance of differences in the ECIS data was analysed using repeated-measures (RM) analysis of variance (ANOVA) and one-way ANOVA for other multiple group data. Two-group comparisons were analysed using Student's t-test when data were normally distributed (via Shapiro-Wilk W test) or Mann–Whitney U-test when not normally distributed. Differences were considered statistically significant when p-values were less than 0.05.
Results
CM from monolayer-cultured MLO-Y4 cells stimulated proliferation of some prostate and breast cancer cell lines. We used the AlamarBlue assay to determine the proliferation of cancer cells in response to MLO-Y4 CM. As shown in Figure 1, after culture for 48 h, 30% (v/v) MLO-Y4 CM from monolayers stimulated the proliferation of DU145 (p<0.05, n=6) and PC3 (p<0.05, n=6) prostate cancer cells, but did not have an effect on proliferation of LNCAP prostate cancer cells. Moreover, 30% MLO-Y4 CM from monolayers stimulated the proliferation of MDA-MB-231 (p<0.05, n=6) and MCF-7 (p<0.05, n=6) breast cancer cells, but did not stimulate proliferation of AGS gastric cancer cells (p<0.05, n=6). MLO-Y4 CM at 20% did not affect the three types of cancer cells considered.
CM from 3D-cultured MLO-Y4 cells stimulated proliferation of some prostate and breast cancer cell lines. Similar experiments were repeated using CM from 3D-cultured MLO-Y4 cells. As shown in Figure 2, after culture for 48 h, 30% CM from 3D-cultured MLO-Y4 cells also stimulated the proliferation of DU145 (p<0.05) and PC3 (p<0.05) prostate cancer cells. CM from 3D-cultured MLO-Y4 cells did not have an effect on proliferation of LNCAP prostate cancer cells. For breast cancer cells, 30% CM from 3D-cultured MLO-Y4 cells stimulated the proliferation of MDA-MB-231 (p<0.05) and MCF-7 (p<0.05) breast cancer cells, whereas the same amount of CM did not stimulate proliferation of AGS gastric cancer cells (p>0.05). CM from 3D-cultured MLO-Y4 cells at 20% did not affect the three types of cancer cells considered.
Effect of CM from monolayer-cultured MLO-Y4 cells on post-wound migration of prostate cancer cells. As indicated by the ECIS system (Figure 3), 30% CM from monolayer MLO-Y4 cells stimulated post-wound migration of PC3 (p<0.01) prostate cancer cells (Figure 3). Osteocyte CM also stimulated the post-wound migration of DU145 cells, but inhibited the post-wound migration of PZHPV7, a non-tumorigenic human prostatic epithelial cell line.
Effect of CM from monolayer-cultured MLO-Y4 cells on post-wound migration of breast cancer cells. We also evaluated the effect of 30% CM from MLO-Y4 cells on post-wound migration of breast cancer cells. As shown in Figure 4, treatment of ZR-75.1 and MCF-7 breast cancer cells with osteocyte CM led to a reduction of post-wound migration of those cells (p<0.01). However, osteocyte CM did not have any effect on post-wound migration of MDA-MB-231 cells (Figure 4).
Conditioned medium from MLOY4 cells appeared to suppress wound-healing capacity of the aggressive MDA-MB-231 but not MCF7 breast cancer cells. Using the scratch-wound assay, we found that osteocyte CM reduced the wound-healing capacity of MDA-MB-231 cells. However, it appeared that osteocyte CM did not have any effect on the wound-healing capacity of MCF-7 cells (Figure 5).
Proliferation of selected human cancer cell lines in response to conditioned medium (CM) from monolayer-cultured MLO-Y4 cells. Proliferation was determined usin the AlamarBlue assay after cultivation for 48 h. Normal culture medium for human cancer cells and control medium for osteocytes were used for background reading without cells. Data represent mean±SD from one experiment with six replicates.*p<0.05.
MLOY4-conditioned medium appeared to stimulate a chemotactic migratory response of MDA-MB-231 cells but not their invasion into Matrigel. Using the transwell invasion and migration assays, we investigated the chemoattractant effect of osteocyte CM on breast cancer cells. As shown in Figure 6, the transwell migratory capacity of MDA-MBV-231 cells was increased in response to osteocyte CM. Osteocyte CM did not have any effect on the transwell migration of MCF-7 and ZR-75.1 cells. In addition, osteocyte CM did not have any effect on transwell invasion of MDA-MB-231, MCF-7 and ZR-75.1 cells (Figure 6).
Discussion
In this study, we investigated the possible interactions between cells at early stages of the pathway of differentiation from osteoblasts to osteocytes and cancer cells, based on previous studies on abundance, distribution and functions of osteocytes in bone. Our data indicated that CM from osteocytes at an early stage of differentiation stimulates the proliferation of some prostate cancer cell lines, namely DU145 and PC3, and breast cancer cell lines, namely MDA-MB-231 and MCF-7. The stimulatory effect was observed both with CM collected from MLO-Y4 cells maintained in monolayer and CM from cells in 3D culture. It is known that osteocytes secrete various growth factors and cytokines, including macrophage colony-stimulating factor, RANKL, matrix metalloproteinase-2 (MMP2), tumor necrosis factor-α, IL6, OPG, sclerostin and α2-Heremans-Schmid glycoprotein/fetuin-A (21-24). Of these soluble factors, IL6 is a potent, pleiotropic, inflammatory cytokine mediator. Interestingly, CM from apoptotic osteocytes contains elevated levels of IL6 and IL6R compared to CM from non-apoptotic osteocytes (16). Through activation of MAPK, signal transducer and activator of transcription–3 and phosphatidylinositol 3-kinase/ protein kinase B (AKT) signaling pathways, IL6 is able to promote cancer cell proliferation and survival, and abrogate apoptosis depending on tumour type (25). However, we found that CM from MLO-Y4 cells did not affect the proliferation of LNCAP prostate cancer cells and AGS gastric cancer cells, perhaps indicating that the stimulation of proliferation by osteocyte CM is cancer cell-dependent.
Proliferation of selected human cancer cell lines in response to conditioned medium (CM) from 3D-cultured ML-OY4 cells. Proliferation was determined using the AlamarBlue assay after cultivation for 48 h. Normal culture medium for human cancer cells and control medium for osteocytes were used for background reading without cells. Data represent mean±SD from one experiment with 6 replicates. *p<0.05.
Our ECIS data demonstrated that osteocyte CM promoted post-wound migration of PC3 and DU145 prostate cancer cells, but suppressed migration of PZHPV7 prostate epithelial cells. This can be partly explained by the secretion of MMP2 from osteocytes. MMPs are a family of calcium-dependent zinc-containing endopeptidases comprising of 23 human MMPs and 23 mouse MMPs (26). They cleave extracellular matrix proteins and bind to other molecules which are involved in cell adhesion and cell–cell interaction. MMPs have been found to promote cancer progression by regulating cancer-cell growth, migration, invasion, metastasis and angiogenesis (27). In particular, MMP2 interacts with the vitronectin receptor integrin αvβ3 and degrades type IV collagen, which consequently facilitates tumour cell migration and invasion (28). Although there is evidence that PZHPV7 cells have molecular features distinct from prostate cancer cells, the mechanisms by which osteocyte CM has an opposite effect on post-wound migration of PZHPV7 need to be further investigated.
Effect of MLO-Y4-conditioned medium (CM) on post-wound migration of prostate cancer cell lines indicated by the electric cell-substrate impedance sensing system. Data represent means±SD from one experiment with six replicates. *p<0.05, ***p<0.001.
Effect of MLO-Y4-conditioned medium (CM) on post-wound migration of breast cancer cells indicated by the electric cell-substrate impedance sensing system. Data represent means±SD from one experiment with six replicates. ***p<0.001.
In contrast to the effect on prostate cancer cells, osteocyte CM suppressed post-wound migration of MCF-7 and ZR-75.1 cells but not aggressive MDA-MB-231cells. However, the scratch assay showed that osteocyte CM suppressed in vitro wound-healing capacity of the MDA-MB-231 but not of the MCF7 breast cancer cells. The transwell invasion and migration assays showed that more MDA-MB-231 cells migrated towards osteocyte CM. On the other hand, osteocyte CM did not have any effect on the transwell invasion capacity of MDA-MB-231, MCF-7 and ZR-75.1 cells. Interestingly, MCF-7 and ZR-75.1 cells are positive for estrogen receptor, while MDA-MB-231 cells are negative for this receptor and for progesterone receptors and human epidermal growth factor receptor 2. However, the exact molecular mechanisms underlying the effect of osteocyte CM on migration and wound-healing potential of breast cancer cells are not clear.
Overall, to the best of our knowledge, this is the first report describing the diverse effects of CM from MLO-Y4 cells on migration of human prostate and breast cancer cells. It is known that breast cancer metastasis tends to be more osteolytic, while prostate cancer metastasis can be more osteoblastic. Therefore, it would be interesting to further investigate how osteocytes interact with different human cancer cell types. Furthermore, it might be worth investigating whether human cancer cells disturb the mediating role of human osteocytes in the balance of osteoblasts and osteoclasts under hypoxic tumour conditions. All these questions might be better addressed if mature and functional human osteocytes were available for cultivation.
Scratch wound-healing assay of breast cancer MDA-231 and MCF-7 cells. Cells were treated with osteocyte-conditioned medium (CM) or control medium for 48 h. The cell monolayer was then scratched with a pipette tip to create a line gap. The 48-well plates containing scratched cells were loaded on to an EVOS system and monitored by time lapse photography. The gap closure was then automatically measured using a home-created macro in ImageJ. A: Closure of the wound gaps by MDA-MB-231 cells. B: Representative images of gap closure by MDA-MB-231 cells. C: Closure of the wound gaps by MCF-7 cells. D: Representative images of gap closure by MCF-7 cells. The data shown are means±SD (n=3). *p<0.05 vs. control medium at the same time point.
Transwell invasion (A) and migratory (B) capacities of cancer cells with response to the conditioned medium (CM) from osteocytes. Transwell inserts (8-μm pores) for 24-well plates were pre-coated with 50 μl/insert of 1 mg/ml Matrigel for invasion. Transwell migration assay without Matrigel was performed in a parallel manner. As chemoattractants, 30% CM from osteocytes or control medium was loaded to the lower chamber of each insert. After incubation for 24 h, cells adhering to other side of the inserts were dissociated with cell dissociation solution containing 4 μg/ml Calcein AM. Invaded cells labelled with Calcein AM were then quantified using a fluorescence plate reader (excitation 490 nm/emission at 520 nm). Data are means±SD (n=3). Significance of the invasion and migration data among two groups was tested using Student's t-test. *p<0.05 vs. control medium.
Acknowledgements
The Authors thank the Cardiff China Medical Research Collaborative and Cancer Research Wales for funding. We also thank Carole Elford for technical assistance.
Footnotes
Conflicts of Interest
The Authors declare no conflicts of interest.
- Received January 4, 2016.
- Revision received February 17, 2016.
- Accepted February 18, 2016.
- Copyright© 2016 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
