Osteocytes

Chondrocytes (1×10⁴ cells/well) were seeded into 96-well plates. After 2 days of incubation, α-MEM (with 1% FBS) containing 10⁻⁶M to 10⁻¹⁰M Ecd was added 1 day before test. The 3-(4, 5-dimethylthiazolyl-2)-2, 5-diphenyltetrazolium bromide (MTT; Sigma Co., St. Louis, MO, USA) assay for cell viability was performed on the 1^st, 3^rd, 7^th, 10^th day of culture. During the experiment, the treatment (including medium and medication) was changed every 3 days and fresh Ecd was added at each media change. The level of mitochondrial activity of the bone cells after Ecd treatments were determined by colorimetric assay, which detects the conversion of MTT to insoluble formazan. The plates were read on the ELISA reader (Spectra max 340, molecular Devices; CA, USA) at a wavelength of 570 nm.

After dissection, the condyles at the proximal end of the tibiae were sawn off with a diamond saw and the distal end was cut off with scissors. The tibiae were immediately fixed for 24 hours at 4°C in 4% paraformaldehyde in phosphate buffered saline (PBS). After fixation the tibiae were decalcified in 10% EDTA and 0.5% paraformaldehyde in PBS at 4°C for 4 weeks. Proper decalcification was verified by x-ray photography. The tibiae were then dehydrated and embedded in paraffin and cut into 5 μm thick sections. Sections were rehydrated and endogenous peroxidase was quenched with 3% H₂O₂ in 40% methanol/PBS. Antigen retrieval was performed through incubation with 0.5% trypsin + 0.1% CaCl2 for 15 minutes at 37°C. After the blocking of the non-specific binding sites with blocking solution (Invitrogen, Waltham, MA, USA) for 1 hour, the sections were incubated overnight at 4°C with 1/1000 anti-intact FGF23 antibody (Santa Cruz Biotechnology Inc, Dallas, TX, USA). The sections were incubated with 1/100 biotinylated second antibody (Dako, Agilent Technologies, Inc., Santa Clara, CA, USA). Further enhancement was performed with the Tyramide Signal Amplification kit (Invitrogen, Waltham, MA, USA) and color development was performed with AEC (Zymed technologies, Waltham, MA, USA). The sections were counterstained with Mayer’s hematoxylin. Quantitative evaluation of the specific staining for FGF23 was performed with NIS Elements digital imaging software (Nikon, Tokyo, Japan). Digital images were made from the medial cortex of two sections of each tibia at 10x magnification. The area of positively-stained osteocytes, the area of positively-stained capillairy bloodvessels and the total bone area was measured. The percentage of positively-stained bone area and the percentage of positively stained vessel area were calculated. Data from the two sections of each tibia were averaged.

CX₃CR1-EGFP/TRAP-tdTomato mice were used in this analysis. The surface tool of Imaris software (Bitplane) was used to perform automatic cell-surface segmentation of each area of SHG⁺ bones, tdTomato⁺ mature osteoclasts, and EGFP⁺ osteoclast precursors. The 3D surface objects of bones were manually adjusted to the 3D surface objects of mature osteoclasts tightly adhering to the bone surface. EGFP⁺ surface objects ≤ 50 µm³ in volume were not included in the analysis, because such groupings were unlikely to represent cells. The surface tool was then used to detect colocalized SHG and EGFP voxels, defined as the adhesion areas between bones and osteoclast precursors (Supplementary Fig. S2A). EGFP⁺ osteoclast precursor areas and adhesion areas were binarized using Otsu’s thresholding method and automatically extracted from the edited images by NIS-Elements software (Nikon). The ratio of the sum of osteoclast precursor areas to the sum of adhesion areas was calculated.