Dulbecco's modified Eagle's medium (DMEM) with 4.5 g/l glucose, L-glutamine and sodium pyruvate and antibiotics (penicillin and streptomycin) were purchased from Corning Cellgro (Mediatech, Inc. Manassas, VA, USA). α-Minimum essential medium (α-MEM) was purchased from Sigma-Aldrich (St. Louis, MO, USA). Fetal bovine serum (FBS) was from HyClone Laboratories (Logan, UT, USA). Tumor necrosis factor-α (TNF-α) was from R&D Systems (Minneapolis, MN, USA). Sodium butyrate, roscovitine, sulforaphane, PD98059, staurosporine, Bay K 8644, wortmannin, 5,6-dichloro-1-β-D-ribofuranosylbenzimidazole (DRB), caspase-3 inhibitor, Alizarin red, lypopolysaccharide (LPS) and all other reagents were purchased from Sigma-Aldrich unless otherwise specified. Gemcytabine was obtained from Hospira, Inc. (Lake Forest, IL, USA). Antibodies for western blot analysis were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Gemcytabine and caspase-3 inhibitor were diluted in phosphate-buffered saline (PBS) and other reagents were dissolved in 100% ethanol to use in experiments.

Gene expression and survival data of 87 breast cancer patients were obtained through the Gene Expression Omnibus (GEO) database (GSE6532) for outcome analysis (23–25). These datasets contained gene expression data derived from the Affymetrix U133 Plus2 platform. For microarray analysis, expression and raw expression data (CEL files) were summarized and normalized using the Robust Multi-array Average algorithm and the Bioconductor package affy (http://www.bioconductor.org/packages/2.0/bioc/html/affy.html).

Human breast cancer MDA-MB-231 cells lack estrogen, progesterone and human epithelial growth factor type 2 (HER2) receptors, and are therefore considered as triple-negative (26). They express high levels of the epithelial growth factor receptor (EGFR) and activation of this receptor and its downstream signaling events enhance migration, proliferation, invasion and progression of the malignant phenotype of these cells (26). MDA-MB-231 cells were obtained from the American Type Culture Collection (ATCC; Rockville, MD, USA).

Stable regucalcin transfectants overexpressing full length or truncated regucalcin proteins in regucalcin in MDA-MB-231 cells were generated as follow. The cDNA encoding human regucalcin with full length (900 bp), deleted exon 4 (684 bp), and deleted exon 4 and 5 (552 bp) were cloned into the pBluescript vector (27,28). The complete regucalcin coding cDNA was cloned into the EcoRI site of the pCXN2 expression vector (27). The resultant plasmid was designated as regucalcin/pCXN2 (24). For transient transfection assay, MDA-MB-231 cells were grown on 24-well plates to ~70% confluence. Each of regucalcin (900 bp), deleted exon 4 (684 bp), and deleted exons 4 and 5 (552 bp) and pCXN2 vector alone were transfected into MDA-MB-231 cells using Lipofectamine reagent, according to the manufacturer's instructions (Promega, Madison, WI, USA) (27). After overnight incubation, gemciticin (G418) (600 μg/ml of medium; Sigma-Aldrich) was added to culture wells for selection and cells were cultured for 2 weeks. After that, cells were plated at limiting dilution to isolate stable transfectants. Multiple surviving clones were isolated, transferred to 35-mm dishes, and grown in medium without G418. The increase in regucalcin in transfectants was 15.5-fold of wild-type cells. In experiments, transfectants were cultured in DMEM containing 10% FBS and 1% penicillin and streptomycin for 1–7 days in a water-saturated atmosphere containing 5% CO₂ and 95% air at 37°C.

M DA-M B-231 wild-type cells (1×10⁵/ml/well) and MDA-MB-231 cells (1×10⁵/ml/well) transfected with regucalcin cDNAs of either full length, deleted exon 4 or deleted exons 4 and 5 were cultured using a 24-well plate in DMEM containing 10% FBS and 1% penicillin and streptomycin for 1, 2, 3 or 7 days in a water-saturated atmosphere containing 5% CO₂ and 95% air at 37°C (28,29). In separate experiments, MDA-MB-231 wild-type cells (1×10⁵/ml/well) or full length regucalcin transfectants were cultured in DMEM containing 10% FBS and 1% penicillin and streptomycin in the presence of sodium butyrate (10 and 100 μM), roscovitine (10 and 100 nM), sulphoraphan (1 and 10 nM), dibucain (0.1 or 1 μM), Bay K 8644 (1 or 10 μM), PD98059 (1 or 10 μM), wortmannin (0.1 or 1 μM), DRB (0.1 or 1 μM), or gemcitabine (50 or 100 nM) for 3 days. After culture, the cells were detached with trypsin from each culture dishes and counted.

MDA-MB-231 wild-type cells (1×10⁵/ml/well) and MDA-MB-231 cells (1×10⁵/ml/well) transfected with either full length, deleted exon 4 or deleted exons 4 and 5 regucalcin cDNAs were cultured using a 24-well plate in DMEM containing 10% FBS and 1% penicillin and streptomycin for 5 days. Subconfluent cells were cultured for additional 3 days in the presence or absence of LPS (0.1 or 1 μg/ml), TNF-α (0.1 or 1 ng/ml) (30). In separate experiments, wild-type MDA-MB-231 cells (1×10⁵/ml/well) or transfectants were cultured for 5 days to confluent, and then for an additional 24 h in the presence or absence of LPS (1 ng/ml) or Bay K 8644 (10 μM) with or without caspase-3 inhibitor (10 μM) (29). After culture, cells were detached with trypsin from each culture dish.

After trypsinization of each culture dish using 0.2% trypsin plus 0.02% EDTA in Ca²⁺/Mg²⁺-free PBS for 2 min at 37°C, the detached cells from the dish were collected by centrifugation (28–30). Cells were resuspended on PBS solution and stained with eosin. Cell numbers were quantified by counting under a microscope using a hemocytometer plate. For each dish, we took the average of two counts. Cell number is shown as number of cells per well.

MDA-MB-231 cells, which were transfected with control vector or regucalcin cDNAs with full length, deleted exon 4 and deleted exons 4 and 5 were plated in 35-mm dishes at a density of 1×10⁶ cells/well in 2 ml of medium, and they were cultured in DMEM containing 10% FBS and 1% penicillin and streptomycin for 3 days. Cells were washed twice with ice cold PBS and removed from the dish with a cell scraper. Recovered cells were disrupted by sonication in 1.0 ml of ice cold PBS containing protease and phosphatase inhibitors. The homogenate was centrifuged for 5 min at 1,500 x g to obtain cell debris, and then the supernatant including cytoplasm, nucleus and other cell fractions were collected. The concentration of protein was determined using Bradford dye reagent (Bio-Rad Laboratories, Inc., Hercules, CA, USA) using bovine serum albumin as a standard. Samples 30 μg of the supernatant protein per lane were separated by SDS-PAGE and transferred to nylon membranes for western blotting using specific antibodies against regucalcin (including 33, 25 and 20 kDa) (21) and other proteins (Santa Cruz Biotechnology). Loading controls consisted of β-actin for cytosolic proteins. A minimum of 3 blots from independent experiments were scanned on an Epson Perfection 1660 Photo scanner, and bands quantitated using ImageJ. Data from independent experiments were normalized as a percentage of control before averaging.

Female mice (CD1-Elite, wild-type, 2 months old), which were purchased from Charles River, were housed in a non-specific pathogen-free facility, and all procedures and protocols were approved through the Institutional Animal Care and Use Committee of Emory University. The femur and tibia were removed immediately after sacrifice (31). Bone marrow cells were isolated under sterile conditions from the femurs and tibias.

The preosteoblastic cell line MC3T3-E1, clone 14 (MC3T3), was purchased from the American Type Culture Collection (ATCC; Manassas, VA, USA) and cultured as previously described (31).

To determine the effects of breast cancer cells on osteoblastogenesis and mineralization of bone marrow or preosteoblastic MC3T3, we used mineralization medium (MM) containing ascorbic acid (100 ng/ml) and 4 mM β-glycerophosphate in DMEM with 10% FBS and 1% penicillin and streptomycin. Bone marrow cells (1×10⁶ cells/1 ml/well) or preosteoblastic MC3T3 (2×10⁵ cells/1 ml/well) were cultured for 3 days at 37°C in a humidified 5% CO₂ atmosphere, and then the cells were co-cultured with addition of breast cancer MDA-MB-231 cells (1×10⁴ cells/1 ml/well) of wild-type or transfectant using 12-well plates in α-MEM in the presence or absence of MM containing ascorbic acid (100 ng/ml) and 4 mM β-glycerophosphate for 18 days (31). The medium was changed every 3 days. After culture, cells were washed with PBS and stained with Alizarin red stain. For quantitation, 10% cetylpyridinium chloride solution was added to each well to elute the dye and absorbance was measured at 570 nm on a microtiter plate reader (31).

To determine the effects of breast cancer cells on bone marrow osteoclastogenesis, bone marrow cells (2×10⁵ cells/1 ml/well) were cultured in DMEM containing 10% FBS and 1% penicillin and streptomycin using 24-well plates (1.0 ml/well) (31). Bone marrow cells were co-cultured in the presence of wild-type (1×10⁴ cells/1 ml/well) or transfectant (1×10⁴ cells/1 ml/well) with full length of regucalcin cDNA for 3 days and then 0.5 ml of the old medium was replaced with fresh medium, and cultures were maintained for an additional 4 days. In other experiments, bone marrow cells (2×10⁵ cells/1 ml/well) were cultured for 3 days in medium and then fresh medium added. Cells were co-cultured with MDA-MB-231 cells [wild-type (1×10⁴ cells/1 ml/well) or transfectants (1×10⁴ cells/1 ml/well)] for additional 4 days (31). After culture for 7 days, the cells adherent to the 24-well plates were stained for tartrate-resistant acid phosphatase (TRACP), a marker enzyme of osteoclasts (32). Briefly, the cells were washed with phosphate-buffered saline solution and fixed with 10% neutralized formalin-phosphate (pH 7.2) for 10 min. After the culture dishes were dried, TRACP staining was applied (32). The fixed cells were incubated for 90 min at room temperature in acetate buffer (pH 5.0) containing naphthol AS-MX phosphate (Sigma) as a stain for the reaction product, in the presence of 10 mM sodium tartrate. TRACP-positive multinucleated cells (MNCs) containing three or more nuclei were counted as osteoclast-like cells. MNCs scored were the mean ± SDM of six cultures.

Survival curves were constructed by Kaplan-Meier analysis and were compared with the log-rank test as performed with IBM SPSS Statistics 18 software (IBM, Chicago, IL, USA; http://www.ibm.com). In the experiments with MDA-MB-231 cells, statistical significance was determined using GraphPad InStat version 3 for Windows XP (GraphPad Software, Inc., La Jolla, CA, USA). Multiple comparisons were performed by one-way analysis of variance (ANOVA) with Tukey-Kramer multiple comparisons post-test for parametric data as indicated. P<0.05 was considered statistically significant.

To understand the involvement of regucalcin in human patients with breast cancer, we compared the clinical outcome between 44 patients with higher regucalcin expression and 43 patients with lower regucalcin expression. There was a significant difference of regucalcin expression between the two groups (Fig. 1A). The reduction of regucalcin expression was associated with poor prognosis in patients with breast cancer. Breast cancer patients with the higher regucalcin gene expression were found to have prolonged relapse-free survival (Fig. 1B). These findings support the view that the suppression of regucalcin gene expression partly contributes to the development of carcinogenesis in human breast cancer cells and leads to a worse clinical outcome.

The cDNA-encoding human regucalcin with full length (33 kDa), deleted exon 4 (25 kDa), and deleted exons 4 and 5 (20 kDa) was cloned into the expression vector pCXN2. Human breast cancer MDA-MB-231 cells were transiently transfected with the pCXN2 vector or regucalcin/pCXN2 construct by lipofection. To generate the transfectants stably overexpressing regucalcin in MDA-MB-231 cells, pCXN2 vector- or regucalcin/pCXN2-transfected MDA-MB-231 cells were cultured in neomycin-containing medium. Multiple neomycin-resistant clones were selected, and the regucalcin content of these clones was analyzed by immunoblotting with an anti-regucalcin antibody. The regucalcin content in cells transfected with regucalcin cDNA vector of full length (33 kDa) was increased 15.5-fold as compared with that of the parental wild-type MDA-MB-231 cells (Fig. 2A). However, the proteins of 25 and 20 kDa were not expressed in MDA-MB-231 cells transfected with deleted exon 4 (25 kDa) and deleted exons 4 and 5 (20 kDa) (Fig. 2A).

To determine the effects of the over-expression of endogenous regucalcin on the proliferation of MDA-MB-231 cells in vitro, the cancer cells were cultured for 1, 2, 3 and 7 days. Numbers of wild-type cells were increased over time in culture (Fig. 2B–E). This increase was suppressed in MDA-MB-231 cells transfectanted with regucalcin cDNA of full length (34 kDa) for 1 (Fig. 2B), 2 (Fig. 2C), 3 (Fig. 2D) and 7 (Fig. 2E) days. The proliferations of MDA-MB-231 cells transfected with exon 4-deleted regucalcin cDNA or the exons 4 and 5-deleted regucalcin cDNA were not significantly suppressed with culture for 7 days as compared with that of the transfectants with the regucalcin cDNA of full length (Fig. 2B–E). Overexpression of regucalcin with full length was found to specifically exhibit suppressive effects on the proliferation of MDA-MB-231 cells in vitro.

Proliferation in MDA-MB-231 cells was determined in the presence of various inhibitors that induce cell cycle arrest in vitro (Fig. 3). Wild-type cells were cultured for 3 days in the presence of butyrate (10 and 100 μM), roscovitine (10 and 100 nM) or sulforaphane (1 and 10 nM) (28,33,34). Cell proliferation was suppressed in the presence of these inhibitors (Fig. 3A). Such effects were not revealed in the transfectants (Fig. 3B). Endogenous regucalcin was suggested to induce G1 and G2/M phase cell cycle arrest in MDA-MB-231 cells.

Next, to determine the mechanistic characterization for suppressive effects of regucalcin on cell proliferation, we examined whether suppressive effects of overexpression of regucalcin on the proliferation of MDA-MB-231 cells are modulated by various signaling factors that suppress the proliferation. Proliferation in MDA-MB-231 cells (wild-type) was suppressed in the presence of dibucaine (0.1 or 1 μM), an inhibitor of calcium/calmodulin-dependent protein kinases (28), or Bay K 8644 (0.1 or 1 μM), an agonist of calcium entry into cells (35) (Fig. 4A). Such effects were not seen in transfectants (Fig. 4B). Likewise, the proliferation of MDA-MB-231 cells (wild-type) was suppressed by culture with wortmannin (0.1 or 1 μM), an inhibitor of phosphatidylinositol 3-kinase (PI3K) (36), and PD98059 (1 or 10 μM), an extracellular signal-regulated kinase (ERK) inhibitor (37) (Fig. 4C). Suppressive effects of these inhibitors on cell proliferation were not revealed in transfectants (Fig. 4D). DRB is an inhibitor of transcriptional activity with RNA polymerase II inhibition (38). Gemcitabine is a strong antitumor agent that induces nuclear DNA damage (39). Proliferation of MDA-MB-231 cells was suppressed by culture with DRB (0.1 or 1 μM) or gemcitabine (50 or 100 nM) (Fig. 4E). However, the suppressive effects of DRB, but not gemcitabine, were not potentiated in transfectants (Fig. 4F).

To determine the effects of the overexpression of regucalcin on cell death in MDA-MB-231 cells, the cells were cultured for 5 days to reach subconfluency. Subconfluent cells were cultured for an additional 24 h. Number of wild-type cells was decreased in the presence of LPS (0.1 or 1 μg/ml) or TNF-α (0.1 or 1 ng/ml), which is known to induce apoptotic cell death (30) (Fig. 5A). Such effects were not exhibited in transfectants overexpressing regucalcin full length (Fig. 5B). In addition, stimulatory effects of LPS (0.1 or 1 μg/ml) or TNF-α (0.1 or 1 ng/ml) on apoptotic cell death were exhibited in MDA-MB-231 cells transfected with the regucalcin cDNA deleted with the exon 4 or with the exons 4 and 5 (Fig. 5C and D). Thus, overexpression of regucalcin with full length was found to specifically protect cell death induced by LPS or TNF-α in MDA-MB-231 cells.

To determine whether the preventive effects of regucalcin on cell death are involved in caspase-3, MDA-MB-231 wild-type cells and transfectants (with full length of regucalcin) were cultured for 5 days to subconfluency, and then the cells were additionally cultured in the presence of LPS (1 μg/ml) or Bay K 8644 (1 μM) with or without caspase-3 inhibitors (10 μM) for 24 h (Fig. 5E and F). Stimulatory effects of LPS or Bay K 8644 on cell death were completely prevented in the presence of caspase-3 inhibitor (Fig. 5E). LPS- or Bay K 8644-induced cell death was not seen in transfectants in the presence or absence of caspase-3 inhibitor (Fig. 5F). Thus, overexpression of regucalcin prevents cell death due to decreasing the activity of caspase-3 that activates nuclear DNA fragmentation, which induces apoptosis of cells.

It was examined whether overexpression of regucalcin regulates protein levels related to cell signalings in MDA-MB-231 cells in vitro using westren blot analysis (Fig. 6). Protein levels of Akt, phospho-Akt, MAPK, phospho-MAPK, SAPK/JNK, and phospho-SAPK/JNK were decreased by overexpression of regucalcin (Fig. 6A). These results suggested that over-expression of regucalcin suppresses signaling pathways that are related to activation of EGFR in MDA-MB-231 cells. Moreover, overexpression of regucalcin increased protein level of p53, a tumor suppressor protein, and it decreased K-ras, c-fos and c-jun, an oncogene, in MDA-MB-231 cells (Fig. 6B). Notably, overexpression of regucalcin was found to decrease protein levels of β-catenin, a transcription factor related to Wnt signaling, and p65 related to NF-κB signaling (Fig. 6B). In addition, overexpression of regucalcin decreased protein levels of caspase-3 and cleaved caspase-3 (Fig. 6C).

To determine an involvement of regucalcin in the bone metastasis of MDA-MB-231 cells, we examined changes in mineralizations in bone marrow osteoblasts or of the osteoblastic cell line MC3T3 co-cultured with MDA-MB-231 cells in vitro (Fig. 7). Bone marrow cells were cultured in the presence or absence of mineralization medium (MM) (Fig. 7A). After 3 days, bone marrow cells were co-cultured with addition of MDA-MB-231 cells (wild-type) or transfectants for 18 days that revealed mineralization. Mineralization in bone marrow cells was suppressed by co-culture with MDA-MB-231 cells. This suppression was prevented in the presence of transfectants (Fig. 7A). Next, preosteoblastic MC3T3 cells were cultured for 3 days, and then the cells were co-cultured with addition of MDA-MB-231 cells (wild-type) or transfectants in medium containing MM for additional 18 days in vitro (Fig. 7B). Co-culture with MDA-MB-231 cells suppressed mineralization in preosteoblastic MC3T3 cells. This suppresstion was not exhibited in the case of transfectants (Fig. 7B).

Moreover, we examined the effects of overexpression of regucalcin on osteoclastogenesis in vitro (Fig. 8). Mouse bone marrow cells were co-cultured in the presence or absence of MDA-MB-231 cells (wild-type) or transfectants overexpressed with regucalcin of full length for 7 days (Fig. 8A). Osteoclastogenesis in bone marrow cells was markedly enhanced with MDA-MB-231 cells (wild-type). However, such an effect was not seen in the case of transfectants (Fig. 8A). Next, bone marrow cells were cultured for 3 days, and then MDA-MB-231 cells (wild-type) or transfectants were seeded on bone marrow cells, and those cells were cultured for additional 4 days (Fig. 8B). Overexpression of regucalcin markedly suppressed osteoclastogenesis enhanced by co-culture with MDA-MB-231 cells (Fig. 8B).