Figure 1. FO promotes osteogenic activity in prednisolone (PDS)-pretreated MC3T3-E1 cells. Mouse preosteoblast MC3T3-E1 cells (1 × 10⁴ cells/mL) were pretreated with 10 μM PDS for two days prior to treatment with FO (0–200 μg/mL) for five days. Fresh media with FO and/or PDS were replenished every two days. At day 7, (A) bone mineralization was evaluated by alizarin red staining, and (B) ALP activity was evaluated using a TRACP & ALP Double-Staining Kit. (C) In a parallel experiment, total mRNA was extracted, and RT-PCR was performed to evaluate the gene expressions of ALP, RUNX2, and OSX. GAPDH was used as the internal control. (D) Total proteins were extracted, and Western blotting was performed to evaluate the expression of ALP, RUNX2 and OSX proteins. β-Actin was used as the internal control. All data are presented as means ± standard error of the mean (^### p < 0.001 vs. untreated MC3T3-E1 cells; * p < 0.05 and *** p < 0.001 vs. PDS-treated MC3T3-E1 cells). ALP: alkaline phosphatase; RUNX2: runt-related transcription factor 2; OSX: osterix; GAPDH: glyceraldehyde 3-phosphate dehydrogenase.

Figure 2. FO prevents PDS-induced bone resorption in zebrafish larvae. Zebrafish larvae (n = 20) at 5 days post fertilization (dpf) were pretreated with 20 μM PDS for two days prior to treatment with FO (0–100 μg/mL) for another two days. Fresh media with FO and/or PDS were replenished at 7 dpf. (A) At 9 dpf, zebrafish larvae were stained with 0.03% calcine and observed under fluorescence microscopy. Numbers show vertebrae (right panels). (B) Each vertebral number was manually counted and indicated. (C) Relative bone area and (D) bone intensity were calculated using imageJ software and expressed. All data are presented as means ± standard error of the mean (* p < 0.05 and *** p < 0.001). In a parallel experiment, total mRNA was extracted, and RT-PCR was performed to evaluate the gene expression of (E) osteoblast-related marker genes such as zRUNX2a, zRUNX2b, zOSX, and zALP and (F) osteoclast-related marker genes such as zCTSK, zNFATc-1, zRANK, and zACP5b. zβ-Actin was used as the internal control. All data are presented as means ± standard error of the mean (^### p < 0.001 vs. untreated zebrafish larvae; ** p < 0.01, and *** p < 0.001 vs. PDS-treated zebrafish larvae). z: zebrafish; RUNX2a/b: runt-related transcription factor 2a/b; OSX: osterix; ALP: alkaline phosphatase; CTSK: cathepsin K; NFATc-1: nuclear factor of activated T-cells; cytoplasmic 1; RANK: receptor activator of nuclear factor κB; and ACP5b: Acid phosphatase 5b.

Figure 3. Prednisolone (PDS)-induced anti-osteogenic activity was inhibited by pretreatment with FO in MC3T3-E1 cells. MC3T3-E1 cells (1 × 10⁴ cells/mL) were pretreated with FO (0–200 μg/mL) for 2 h prior to treatment with 10 μM PDS for seven days. Fresh media with FO and/or PDS were replenished every two days. At day 7, (A) bone mineralization and (B) ALP activity were evaluated using alizarin red staining and a TRACP & ALP Double-Staining Kit, respectively. (C) Total mRNA were extracted, and RT-PCR was performed to evaluate the gene expressions of ALP, RUNX2, and OSX. GAPDH was used as the internal control. (D) Total proteins were extracted, and Western blotting was performed to evaluate the expression of ALP, RUNX2, and OSX. β-Actin was used as the internal control. All data are presented as means ± standard error of the mean (^### p < 0.001 vs. untreated MC3T3-E1 cells; * p < 0.05, and *** p < 0.001 vs. PDS-treated MC3T3-E1 cells). ALP: alkaline phosphatase; RUNX2: runt-related transcription factor 2; and OSX: osterix.

Figure 4. Pretreatment with FO restores vertebral formation in prednisolone (PDS)-post-treated zebrafish larvae. Zebrafish larvae (n = 20) at 5 days post fertilization (dpf) were treated with FO (0–100 μg/mL) for two days, and 20 μM PDS was post-treated for another two days. (A) At 9 dpf, zebrafish larvae were stained with 0.03% calcine and observed under fluorescence microscopy. Numbers show vertebrae (right panels) (B) Vertebral number was manually counted and indicated. (C) Relative bone area and (D) bone intensity were calculated using imageJ software and expressed. All data are presented as means ± standard error of the mean (* p < 0.05, ** p < 0.01, and *** p < 0.001). Total mRNA was extracted, and RT-PCR was performed to evaluate the gene expression of (E) osteogenic genes, including zRUNX2a, zRUNX2b, zOSX, and ALP and (F) osteoclastogenic genes such as zCTSK, zNFATc-1, zRANK, and zACP5b. β-Actin was used as the internal control. All data are presented as means ± standard error of the mean (^# p < 0.05 and ^### p < 0.001 vs. untreated zebrafish larvae; ** p < 0.01, and *** p < 0.001 vs. PDS-treated zebrafish larvae). z: zebrafish; RUNX2a/b: runt-related transcription factor 2a/b; OSX: osterix; ALP: alkaline phosphatase; CTSK: cathepsin K; NFATc-1: nuclear factor of activated T-cells; cytoplasmic 1; RANK: receptor activator of nuclear factor κB; and ACP5b: acid phosphatase 5b.

Figure 4. Pretreatment with FO restores vertebral formation in prednisolone (PDS)-post-treated zebrafish larvae. Zebrafish larvae (n = 20) at 5 days post fertilization (dpf) were treated with FO (0–100 μg/mL) for two days, and 20 μM PDS was post-treated for another two days. (A) At 9 dpf, zebrafish larvae were stained with 0.03% calcine and observed under fluorescence microscopy. Numbers show vertebrae (right panels) (B) Vertebral number was manually counted and indicated. (C) Relative bone area and (D) bone intensity were calculated using imageJ software and expressed. All data are presented as means ± standard error of the mean (* p < 0.05, ** p < 0.01, and *** p < 0.001). Total mRNA was extracted, and RT-PCR was performed to evaluate the gene expression of (E) osteogenic genes, including zRUNX2a, zRUNX2b, zOSX, and ALP and (F) osteoclastogenic genes such as zCTSK, zNFATc-1, zRANK, and zACP5b. β-Actin was used as the internal control. All data are presented as means ± standard error of the mean (^# p < 0.05 and ^### p < 0.001 vs. untreated zebrafish larvae; ** p < 0.01, and *** p < 0.001 vs. PDS-treated zebrafish larvae). z: zebrafish; RUNX2a/b: runt-related transcription factor 2a/b; OSX: osterix; ALP: alkaline phosphatase; CTSK: cathepsin K; NFATc-1: nuclear factor of activated T-cells; cytoplasmic 1; RANK: receptor activator of nuclear factor κB; and ACP5b: acid phosphatase 5b.

Figure 5. FO is not related to estrogenic activity in ovariectomy (OVX)-induced mice and estrogen response element (ERE)-luciferase transfected MCF-7 cells. (A) OVX-induced mice were orally fed with 100 mg/kg and 200 mg/kg FO and intraperitoneally (IP) injected with 10 μg/kg 17β-estradiol (E2) for 4 weeks. After termination, uterine weight was determined by calculating uterine weight in body weight. (B–D) MCF-7 breast cancer cells (1 × 10⁴ cells/mL) transfected with human estrogen response element (ERE)-luciferase reporter plasmid for 24 h and (B) concentration–response curve of transcriptional activation for FO (2 × 10⁻⁴–2 × 10² μg/mL), corticosterone (10⁻¹¹–10⁻⁵ M), genistein (10⁻¹⁰–10⁻⁴ M), and E2 (10⁻¹⁴–10⁻⁸ M) were measured using luciferase activity. (C,D) In a parallel experiment, FO (200 μg/mL) and E2 (10⁻⁹ M) were treated in the presence or absence of 10 μM ICI 182,780 (ICI), and (C) relative luciferase activity and (D) relative cell proliferation were calculated. All data are presented as means ± standard error of the mean (and ^### p < 0.001 and ^## p < 0.01 vs. untreated MCF-7 cells; *** p < 0.001 and ** p < 0.01 vs E2-treated MCF-7 cells).