Animal Procedures

All experiments were approved by the Recombinant DNA Experiment Safety Committee and Animal Research Ethics Committee at Tokyo Dental College (DNA2101 and No. 300706). Mild-type HPP often exhibits autosomal dominant inheritance with a pathogenic mutation in only one allele, and heterozygous Akp2+/− mice are commonly used as a model for mild-type HPP [17]. Immediately after birth, genotyping was performed using forward, 5′-AGTCCGTGGGCATTGTGACTA-3′, and reverse, 5′-TGCTGCTCCACTCACGTCGAT-3′ primers according to the method previously reported, and Akp2+/+ and Akp2+/− mice were distinguished [18]. First, we investigated the distribution of the BP preparations within the femurs of Akp2+/− mice. We used Zol-FAM (BioVinc LLC, CA, USA), a BP in which zoledronic acid (Zol) is labeled with the fluorescent substance FAM (FAM-Zol). FAM-Zol was diluted with physiological saline (PBS) and subcutaneously administered to 4-week-old Akp2+/+ or Akp2+/− mice at 200 μg/kg (volume: 100 μL), and femurs were collected 24 h later [19].

Next, we investigated the effects of long-term administration of zoledronic acid hydrate on the femurs of Akp2+/− mice. Zol (Novartis, Basel, Switzerland) 1 mg/kg (volume: 200 µL) diluted in PBS was administered subcutaneously to 4-week-old Akp2+/+ or Akp2+/− mice [20, 21]. The mice were administered the drug once a week until they were 8 weeks old (five times total), and sampling was performed at 9 weeks of age. Akp2+/+ or Akp2+/− mice administered with the same amount of PBS were used as controls.

Distribution of BPs

The sampled femurs were cut into upper (containing the femoral head) and lower (containing the growth plate) sections. The samples were fixed using 4% paraformaldehyde (PFA) at 4 °C for 1 day. Sucrose substitution was performed, and the samples were encapsulated in Super Cryoembedding Medium (Leica Microsystems, Wetzlar, Germany) without decalcification. According to the Kawamoto method, the specimen was sliced to a thickness of 10 μm and attached to a Cryofilm Type IIC (9) (Leica Microsystems, Wetzlar, Germany). Hoechst 33342 solution (DOJINDO LABORATORIES, Kumamoto, Japan) was diluted 1/1000 for nuclear staining and mounted with SCMM-G1 (Leica Microsystems, Wetzlar, Germany), a glycerin-based mounting medium. The thin sections were mounted on MAS-coated glass slides (Matsunami Glass, Tokyo, Japan) and observed under a confocal microscope LSM880 (Carl Zeiss AG, Jena, Germany) (n = 3).

Histological Analysis

Harvested femurs were fixed in 0.1 M phosphorus acid buffer (pH 7.4) containing 4% paraformaldehyde (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan). They were decalcified in 10% ethylenediaminetetraacetic acid (Muto Junyaku, Tokyo, Japan) for 3 weeks, dehydrated in alcohol, cut into the upper part containing the femoral head and the lower part containing the growth plate, and embedded in paraffin. The prepared paraffin blocks were sectioned into serial sections of 4 μm thickness using a rotary microtome SM2000R (Leica Microsystems, Wetzlar, Germany). Thin sections were stained with hematoxylin–eosin (HE) staining (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) to confirm tissue morphology. The sections were stained with Alcian blue (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) to detect the chondrocyte matrix. Additionally, a TRAP/ALP Staining Kit (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) was used to identify the osteoclasts. Photographs were captured using an Axio Imager M2 upright microscope (Carl Zeiss AG; n = 3).

Micro-computed Tomography Imaging

Harvested femurs were stored in 4% PFA at 4 °C overnight and fixed, and micro-computed tomography (micro-CT) imaging was performed. Two bone samples were excluded from the study due to accidental breakage during sampling. The femoral head was imaged using a micro-CT50 (JEOL/SCANCO, Tokyo, Japan). Imaging conditions were as follows: tube voltage, 70 kV; tube current, 57 µA; voxel size, 14.8 × 14.8 × 14.8 µm. The growth plate area was imaged using a CosmoScan FX (Summit Pharmaceuticals International Corporation, Tokyo, Japan). Imaging conditions were as follows: tube voltage, 90 kV; tube current, 88 µA; voxel size, 10 × 10 × 10 µm. The captured data were imaged using software TRI/3D-SRF (Ratoc System Engineering, Tokyo, Japan; n = 6).

Bone Morphology Analysis

Based on the micro-CT images taken above, trabecular bone (TB) and cortical bone (CB) morphometry were performed using a 2-mm-thick area 0.5 mm away from the growth plate as the region of interest. Bone mineral density (BMD), bone mineral content (BMC), bone volume/tissue volume (BV/TV), trabecular thickness (Tb.Th), trabecular number (Tb.N), trabecular separation (Tb.Sp), trabecular spacing (Tb.Spac), marrow space star volume (V* m.space), average cortical thickness (Ct.Th), and cortical area fraction (Ct.Ar), which are used as typical bone morphometry parameters, were automatically calculated according to the TRI/3D-SRF software (Ratoc System Engineering) manual (n = 6).

Measurement of ALP Activity, P1NP, CTX-1, and α-SMA

Blood collected from the inferior vena cava at 9 weeks of age was left at room temperature for 30 min and then centrifuged at 4 °C and 12,000 rpm for 15 min to obtain the serum. The collected bone marrow of the femur was flushed with PBS, and the remainder was placed in a CellCover solution (Anacyte Laboratories, Hamburg, Germany) and stored at 4 °C overnight. Thereafter, the cells were homogenized using EzRIPA Lysis Buffer (ATTO Corporation, Tokyo, Japan) to obtain a protein extract. The obtained serum and protein extract were used to measure ALP activity; procollagen 1 intact N-terminal propeptide (P1NP), a bone formation marker; type I collagen cross-linked C-terminal telopeptide (CTX-1), a bone resorption marker; and alpha-smooth muscle actin (α-SMA), a myofibroblast marker. As previously reported, ALP activity was determined by measuring the wavelength at 405 nm, taking advantage of its ability to convert the substrate p-nitrophenol phosphate to the strong yellow-soluble product p-nitrophenol under alkaline conditions. The amount of enzyme required to catalyze the production of 1 μmol p-nitrophenol per min was calculated as 1 U. Measurements were carried out using the MBS3805180 PINP ELISA Kit (MyBioSource Inc., San Diego, CA, USA) for P1NP, the AC-06FI CTX-1 ELISA kit (RatLapsTM, Immunodiagnostic System Ltd, Boldon, United Kingdom) for CTX-1, and the NBP2-66429 α-SMA ELISA kit (Novus Biologicals, Colorado, USA) for α-SMA according to the attached protocols. Protein concentrations of the serum and protein extracts were measured using a DC Protein Assay Kit (Bio-Rad, Hercules, CA, USA), and each measurement result was corrected to a value per milligram (n = 5).

RNA Extraction and Real-Time PCR

The collected femurs were pretreated in the same manner as for the protein extracts described prior. RNA was extracted using a RNeasy Mini Kit (QIAGEN, Hilden, Germany), according to the manufacturer’s instructions. The extracted RNA was immediately synthesized into cDNA using SuperScript® Reverse Transcriptase (Thermo Fisher Scientific, Massachusetts, USA) with random hexamers. Relative quantification by the ΔΔCT method was then performed using QuantStudio 7Pro (Applied Biosystems, Massachusetts, USA). Runx2, sp7, col1a1, Bglap, and col2a1 were measured on TB Green® Premix Ex Taq™ II (Clontech Laboratories, TaKaRa Bio Company, Ohtsu, Shiga, Japan) using the primers shown in Supplementary Table 1, and the remaining Alpl, BMP2, cartilage oligomeric matrix protein (COMP), col10a1, CCAAT/Enhancer Binding Protein-β (C/EBPβ), and Smad ubiquitination regulatory factors-2 (Smurf2) were measured using a custom TaqMan® array plate from Thermo Fisher Scientific. Both systems used 18S rRNA as the endogenous control. Gene names and Assay IDs are listed in Supplementary Table 1. The obtained data were analyzed using ExpressionSuite ™ Software v1.3 (Applied Biosystems; n = 5).

Spontaneous Locomotor Activity

After the last Zol administration, 8-week-old mice underwent spontaneous locomotor activity measurements using a wheel cage RWC-15 (MELQUEST, Toyama, Japan). The amount of movement of the mice was measured for the activity time, total distance, average speed, and acceleration for 3 days, and the average values were calculated from the results on the second and third days, excluding the first day as a training period. The average speed was calculated by averaging the values, excluding those where the distance traveled per minute was zero, and the maximum value was calculated for acceleration. The average speed was detected using the momentum data recording and analysis system CIF-4S (MELQUEST) and analyzed using the Actmaster software (MELQUEST) (n = 8).

Statistical Analysis

All data are presented as the mean ± SD. Statistical analyses were performed using the Prism 10 software (GraphPad Software, La Jolla, CA, USA). Significant differences between two groups were analyzed using an unpaired two-tailed Student's t test. Significant differences among the four groups were determined using two-way ANOVA and Šídák's multiple comparison tests. Statistical differences were considered significant at P < 0.05, and statistical analyses and graphing were performed using Prism 10 (GraphPad Software, La Jolla, CA, USA).