Mimosine and deferoxamine were purchased from Sigma-Aldrich Co. (St. Louis, MO, USA). A mouse osteoprotegerin enzyme-linked immunosorbent assay (ELISA) kit, mouse IL-6 ELISA kit, mouse VEGF ELISA kit and PGF_2α were obtained from R&D Systems, Inc. (Minneapolis, MN, USA). HIF-1α antibodies (ab16066) were obtained from Abcam (Cambridge, UK). Phospho-specific p38 MAP kinase antibodies (#4511), p38 MAP kinase antibodies (#9212), phospho-specific p44/p42 MAP kinase antibodies (#9101), p44/p42 MAP kinase antibodies (#9102), phospho-specific SAPK/JNK antibodies (#4671) and SAPK/JNK antibodies (#9252) were all obtained from Cell Signaling Technology, Inc. (Beverly, MA, USA). Actin antibodies were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA). An ECL western blot detection system was obtained from GE Healthcare (Buckinghamshire, UK). Other materials and chemicals were obtained from commercial sources. PGF_2α was dissolved in ethanol. Mimosine was disolved in phosphate-buffered saline (PBS) supplemented with 0.01% bovine serum albumin (BSA) containing 7.5% NaHCO₃. Deferoxamine was disolved in PBS supplemented with 0.01% BSA. The maximum concentration of ethanol was 0.1%, which did not affect the assay for osteoprotegerin release, IL-6 release, osteoprotegerin mRNA expression, or western blot analysis.

Cloned osteoblast-like MC3T3-E1 cells derived from newborn mouse calvaria (20) were maintained as previously described (21). Briefly, the cells were cultured in α-minimum essential medium (α-MEM) containing 10% fetal bovine serum (FBS) at 37°C in a humidified atmosphere of 5% CO₂/95% air. The cells were seeded into 35-mm diameter dishes (5×10⁴ cells/dish) or 90-mm diameter dishes (2×10⁵ cells/dish) in α-MEM containing 10% FBS. After 5 days, the medium was exchanged for α-MEM containing 0.3% FBS. The cells were then used for experiments after 48 h.

The cultured cells were pretreated with various doses of mimosine or deferoxamine for 60 min, and then stimulated with 10 µM PGF_2α or the vehicle (PBS supplemented with 0.01% BSA containing 0.1% ethanol) in 1 ml of α-MEM containing 0.3% FBS for the indicated periods of time. The conditioned medium was collected at the end of incubation, and the concentrations of osteoprotegerin or IL-6 were subsequently measured using an osteoprotegerin ELISA kit or IL-6 ELISA kit according to the manufacturer's instructions.

The cultured cells were treated with various doses of mimosine or deferoxamine in 1 ml of α-MEM containing 0.3% FBS for 48 h. The conditioned medium was collected at the end of incubation, and the VEGF concentrations were then measured using a VEGF ELISA kit according to the manufacturer's instructions.

Cultured cells were pretreated with 700 µM mimosine, 500 µM deferoxamine or vehicle for 60 min, and then stimulated with 10 µM PGF_2α or vehicle in α-MEM containing 0.3% FBS for 3 h. Total RNA was isolated and transcribed into cDNA using TRIzol reagent (Invitrogen Co., Carlsbad, CA, USA) and an Omniscript Reverse Transcriptase kit (Qiagen, Inc., Valencia, CA, USA), respectively. RT-qPCR was performed using a LightCycler system with capillaries and the FastStart DNA Master SYBR-Green I provided with the kit (Roche Diagnostics, Basel, Switzerland). Sense and antisense primers for mouse osteoprotegerin mRNA were purchased from Takara Bio, Inc. (Tokyo, Japan) (primer set ID: OPG; MA026526) (osteoprotegerin primer sequences (5′→3′): forward, CAATGGCTGGCTTGGTTTCATAG and reverse, CTGAACCAGACATGACAGCTGGA), whereas mouse VEGF mRNA or glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA primers were synthesized based on the study of Simpson et al (22) (VEGF primer sequences (5′→3′): forward, TTACTGCTGTACCTCCACC and reverse, ACAGGACGGCTTGAAGATG; and GAPDH primer sequences (5′→3′) forward, AACGACCCCTTCATTGAC and reverse, TCCACGACATACTCAGCAC. The amplified products were determined using a melting curve analysis and agarose gel electrophoresis. The mRNA levels of osteoprotegerin or VEGF were normalized to those of GAPDH mRNA, respectively.

The cultured cells were pretreated with various doses of mimosine or deferoxamine for 60 min, and then stimulated with 10 µM PGF_2α or vehicle in α-MEM containing 0.3% FBS for the indicated periods of time. The cells were washed twice with phosphate-buffered saline, and then lysed, homogenized and sonicated in a lysis buffer containing 62.5 mM Tris/HCl, pH 6.8, 2% sodium dodecyl sulfate (SDS), 50 mM dithiothreitol and 10% glycerol. SDS-polyacrylamide gel electrophoresis (PAGE) was performed according to the method of Laemmli (23) in 10% polyacrylamide gels. The protein was fractionated and transferred onto Immun-Blot PVDF membranes (Bio-Rad, Hercules, CA, USA). The membranes were blocked with 5% fat-free dry milk in Tris-buffered saline-Tween-20 (TBS-T; 20 mM Tris-HCl, pH 7.6, 137 mM NaCl, 0.1% Tween-20) for 1 h before incubation with primary antibodies. Western blot analysis was performed as described previously (24) using HIF-1α antibodies, actin antibodies, phospho-specific p38 MAP kinase antibodies, p38 MAP kinase antibodies, phospho-specific p44/p42 MAP kinase antibodies, p44/p42 MAP kinase antibodies, phospho-specific SAPK/JNK antibodies or SAPK/JNK antibodies as primary antibodies, and peroxidase-labeled antibodies raised in goat anti-rabbit IgG were used as secondary antibodies (074-1506, KPL, Inc., Gaithersburg, MD, USA). The primary and secondary antibodies were diluted at 1:1,000 with 5% fat-free dried milk in TBS-T. The peroxidase activity on the PVDF membranes was visualized on X-ray film by means of the ECL western blot detection system.

Densitometric analysis was performed using scanner and image analysis software (ImageJ version 1.48; National Institutes of Health, Bethesda, MD, USA). The phosphorylated protein levels were calculated as follows: the background-subtracted signal intensity of each phosphorylation signal was normalized to the respective total protein signal and plotted as the fold increase in comparison to control cells which had not been stimulated by PGF_2α nor treated with mimosine or deferoxamine.

The data were analyzed by ANOVA followed by the Bonferroni method for multiple comparisons between pairs, and a p-value <0.05 was considered to indicate a statistically significant difference. All data are presented as the means ± SEM of triplicate determinations from three independent cell preparations.

Mimosine, which is an inhibitor of the prolyl hydroxylase domain proteins responsible for degrading HIF-1α, is known to act a normoxic inducer of HIF-1α (25). In the present study, we first examined the effect of mimosine on HIF-1α protein levels in osteoblast-like MC3T3-E1 cells. We noted that mimosine markedly increased the HIF-1α protein levels (Fig. 1A).

It has previously been noted that deferoxamine, an iron chelator, exerts its angiogenic effects through stimulation of the HIF-1α pathway (26), and deferoxamine is known to be another inducer of HIF-1α (26). We also found that HIF-1α protein expression levels were markedly upregulated by deferoxamine (Fig. 1B).

We have recently shown that PGF_2α stimulates osteoprotegerin synthesis in osteoblast-like MC3T3-E1 cells (10). Thus, in the present study we examined the effect of mimosine and deferoxamine on PGF_2α-induced osteoprotegerin release in MC3T3-E1 cells. Mimosine was noted to significantly reduce osteoprotegerin release induced by PGF_2α up to 48 h (Fig. 2). The suppressive effect of mimosine on osteoprotegerin release is clearly dose-dependent in the range between 300 and 700 µM (Fig. 3A); mimosine at 500 µM caused a 65% decrease in PGF_2α-induced OPG release. In addition, we also noted that deferoxamine significantly decreased the release of osteoprotegerin induced by PGF_2α (Fig. 2), and the inhibitory effect was dose-dependent in the range between 100 and 500 µM (Fig. 3B). The maximum inhibitory effect of deferoxamine in relation to OPG release was observed at 500 µM, which caused approximately an 80% decrease in PGF_2α-induced OPG release.

We have previously reported that PGF_2α stimulates IL-6 synthesis in osteoblast-like MC3T3-E1 cells (13,14). Thus, in the present study we examined the effect of mimosine or deferoxamine on the PGF_2α-induced release of IL-6. It was clear that mimosine up to 700 µM failed to markedly affect the IL-6 release induced by 10 µM PGF_2α (Fig. 4A). In addition, we noted that deferoxamine up to 500 µM did not markedly effect PGF_2α (10 µM)-induced IL-6 release (Fig. 4B).

To investigate whether the inhibitory effect of mimosine or deferoxamine on PGF_2α-induced osteoprotegerin release is mediated by transcriptional events in osteoblast-like MC3T3-E1 cells, we examined the effect of mimosine and deferoxamine on PGF_2α-induced osteoprotegerin mRNA expression. Mimosine (700 µM) and deferoxamine (500 µM), which alone hardly affected the osteoprotegerin mRNA level, significantly attenuated the increase in the mRNA expression level of osteoprotegerin induced by 10 µM of PGF_2α (Fig. 5).

It has been noted previously that HIF increases oxygen-regulated gene expression, including VEGF, and promotes angiogenesis and osteogenesis (27). Therefore, we examined whether mimosine or deferoxamine upregulates VEGF release in osteoblast-like MC3T3-E1 cells. The release of VEGF was significantly upregulated by mimosine in a dose-dependent manner in the range between 300 and 700 µM (Fig. 6A). Additionally, we noted that deferoxamine dose-dependently increased VEGF release in the range between 100 and 500 µM (Fig. 6B).

We further investigated the effects of mimosine or deferoxamine on VEGF mRNA expression in MC3T3-E1 cells. Both mimosine and deferoxamine significantly upregulated VEGF mRNA expression levels (Fig. 7).

In a previous study, we have demonstrated that PGF_2α-induced osteoprotegerin synthesis is mediated through activation of p38 MAP kinase, p44/p42 MAP kinase and SAPK/JNK in osteoblast-like MC3T3-E1 cells (10). Therefore, in the present study we examined whether mimosine and deferoxamine affected the PGF_2α-induced phosphorylation of p38 MAP kinase, p44/p42 MAP kinase or SAPK/JNK in these cells. We noted that the phosphorylation of p38 MAP kinase, p44/p42 MAP kinase and SAPK/JNK induced by PGF_2α was not markedly affected by mimosine, up to 700 µM (Fig. 8). It also became evident that deferoxamine exerted little effect on the PGF_2α-induced phosphorylation of p38 MAP kinase, p44/p42 MAP kinase and SAPK/JNK, up to 500 µM (Fig. 9).