Animals

All experiments received approval from the Animal Care Committee at Osaka University (Approval #310,018–017). Our objective was to create a mouse model with a variant equivalent to that found in a patient with type III OI. This patient exhibited multiple fractures and blue sclera at birth. At the age of 10, her height was 108.1 cm (− 2.16 SD), and her lumbar spine bone mineral density (BMD) was 0.313 g/cm2 (Z-score = − 5.99). The patient carried a heterozygous variant (p.Gly821Ser) in the COL1A1 gene, resulting in an alteration of the 643rd amino acid from glycine to serine in the helical region. Consequently, we introduced a G-to-A transversion at nucleotide 2428 in the Col1a1 gene, mimicking the same amino acid alteration in C57BL/6 J mice using the CRISPR-Cas9 system at The Institute of Experimental Animal Sciences, Faculty of Medicine, Osaka University, Japan. In addition to c.2428G > A, we introduced c.2427 T > C and c.2433A > G alterations, which do not alter the amino acid sequence and are considered silent variant, for genotyping. For the knock-in, we designed single guide RNA (sgRNA) targeting Col1A1 exon 36, where c.2428 is located, using CRISPOR (http://crispor.tefor.net/) and CRISPRdirect (https://crispr.dbcls.jp/). Additionally, we designed single-strand oligo DNA as a donor DNA template, incorporating the c.2428G > A variant along with the insert sequence, flanked by homology arms complementary to the target site. Microinjection of the sgRNA, Cas9 mRNA, and donor DNA into C57BL/6JJcl mouse zygotes was performed using a super electroporator (NEPA21®; NEPAGENE, Ichikawa, Japan). The electroporation reagent included 100 ng/µL of Cas9 endonuclease (Alt-R® S.p. HiFi Cas9 Nuclease 3NLS; Integrated DNA Technologies, Coralville, IA, USA), 200 ng/µL of sgRNA (Alt-R® CRISPR-Cas9 System; Integrated DNA Technologies), and 300 ng/µL of donor DNA (4 nmol Ultramer® DNA Oligos; Integrated DNA Technologies). Following overnight culture, 140 two-cell zygotes were injected into the gonaduct of five pseudo-pregnant ICR mice. We initially obtained 10 F0 mice including two male and two female knock-in mice. These heterozygous knock-in mice were then bred with wild-type C57BL/6 J mice to establish the colony. For this study, we analyzed mice from the F5 generation. The reproductive rate of this model mouse was comparable to that of wild-type C57BL/6 J mice, with litter sizes ranging from approximately six to ten pups.

For genotyping, target genes were amplified through polymerase chain reaction (PCR) using forward (5′-TCTAGGGAGACCGTGGTGAG-3′) and reverse (5′-CCAAGTCCAAGGCTATCCAA-3′) primers. Left and right femurs and whole vertebrae were harvested to analyze skeletal characteristics. After weaning at 3 weeks, all mice were housed in groups of one to three per cage, adhering to the regulations of the Animal Care Committee at Osaka University. Mice were kept under a 12 h light–dark cycle (light on from 08:00 to 20:00) with ad libitum access to food and water. The health and well-being of the animals were closely monitored by both the research team and staff of the Institute of Experimental Animal Sciences Faculty of Medicine, Osaka University. Monitoring included regular assessments of body weight (twice weekly), food and water intake, and general assessment of animal activity. We fed these mice with standard certified solid rodent diets (MF, Oriental Yeast Co., Ltd, Tokyo, Japan, Table S1). We collected blood samples via cardiac puncture at 8 weeks to measure serum levels of calcium, phosphate, alkaline phosphatase, and albumin. The laboratory analyses were conducted at Oriental Yeast Nagahama Lifescience laboratory (Shiga, Japan). After soft tissue stripping, the left femur and vertebrae were placed in 70% ethanol at 20 °C for microcomputed tomography (µCT) analysis, and the right femur was frozen in saline-soaked gauze at − 80 °C for three-point bending mechanical tests. These samples were kept from drying during the subsequent procedures without rehydration. To evaluate the growth longitudinally and analyze the young mouse bone phenotype, we analyzed 12-week-old mice, as reported by Jakobsen et al. and Kaupp et al.[19, 20]

µCT and Architectural Analysis

µCT scans of the L5 vertebrae and left femurs from 12-week-old Col1a1G643S/+ mice and their normal littermates were conducted using a Scan Xmate-L090H system (Comscantecno Co., Ltd., Yokohama, Japan) operating at 75 kV and 100 µA X-ray source, according to the manufacture’s instructions, at Kureha Special Laboratory (Iwaki, Japan). The resolution was set to 10.944 and 15.894 µm/pixel for femurs and vertebrae, respectively. For the distal femur, trabecular bone was scanned in a region starting from 0.5 mm proximal to the distal femoral growth plate with a 2,604 µm section of bone in a proximal direction. Data quantification was performed using the TRI/3D-BON bone structure analysis software (Ratoc System Engineering Co., Ltd. Tokyo, Japan).

Trabecular bone parameters, based on the scanning charts, included bone surface/bone volume (BS/BV, /mm) and bone volume/total volume (BV/TV, %) for the analysis of the volume. Trabecular thickness (Tb. Th, µm), trabecular number (Tb. N, /mm), trabecular separation (Tb. Sp, µm), and trabecular spacing (Tb. Spac, µm) were calculated using parallel plate models. BMD (g/cm2) data were obtained from the frontal section image of L5 vertebrae and femora, calculated using phantoms of defined density. The fractal dimension (FD) was measured to assess the complexity of unevenness. The trabecular bone pattern factor (TBPf, /mm) was calculated as the ratio of the delta bone surface to delta bone volume (δBD/δBV). As an index of osteoporosis severity, star volume was analyzed and presented as the marrow space star volume (V*m. space, mm3) and trabecular star volume (V*tr, mm3). The Node-Sturt model was used to assess the trabecular structure and node (Nd, the joint point of more than three trabeculae or trabeculae with different widths) and terminus (Tm, the end without a joint with other trabeculae), and the joint with cortical bone (Ct) was measured. From these parameters, the number (N) and mean length (E) of the bone between Nds (NdNd), Nd and Tm (NdTm), Ct and Nd (CtNd), Ct and Tm (CtTm), Cts (CtCt), and Tms (TmTm) were evaluated. The total strut length (TSL) was also calculated, and the ratio to the total length of each parameter was evaluated as follows: total length of NdNd/TSL (%), NdTm/TSL (%), CtNd/TSL (%), CtTm/TSL (%), CtCt/TSL (%), and TmTm/TSL (%). To analyze the cortical bone at the midshaft femur, scanning started at 50% of the total femur length from the distal end and continued for 1 mm proximally.

For the cortical bone, cortical volume (Ct. V, mm3), bone volume (Bv, mm3), medullary volume (Mv, mm3), average cortical thickness (Ct. Th, µm), cortical bone area (Ct. Ar, mm2), and total pore volume (Po. V, mm3) were measured. The total cross-sectional area inside the periosteal envelope (Tt. Ar, mm2) was calculated as the total of Ct. V, Mv, and Po. V. Cortical bone density (Cort Bone Density, %) and total bone density (Total Bone Density, %) were calculated using the following formulas: (Ct. V) / (Ct. V + Po. V) and (Bv + Ct. V) / (Tt. Ar), respectively.

Three-Point Bending Mechanical Test

Destructive three-point bending was conducted on the right femurs of 10 male and 10 female Col1a1G643S/+ mice and their littermates at 12 weeks of age using an MZ-500D system (Maruto Instrument Co., Ltd., Tokyo, Japan) at Kureha Special Laboratory. The posterior surface was subjected to tension at a vertical displacement rate of 2 mm/min, with a 6 mm distance between the lower supports. The maximum load (N) and displacement (mm) were measured from the load–displacement curve, and the stiffness (N/mm) was determined through linear regression of the initial region of the curve. The fracture energy (Nmm) was calculated by measuring the area under the load-deformation curve.

Bone Histomorphometry

Bone histomorphometry was performed on lumber vertebrae at 8-week-old mice, including 4 WT and 3 Col1a1G643S/+ mice. Tetracycline was administered subcutaneously, followed by calcein injection 5 and 2 days before vertebrae removal. The vertebrae were fixed with 70% ethanol immediately after the sacrifice. Sagittal sections of the lumber vertebrae (L3) were prepared, and fluorescence imaging was performed at the Ito Bone Histomorphometry Institute (Niigata, Japan). All bone sections were analyzed histomorphometrically using a light microscope (BX-53, Olympus, Tokyo, Japan) with an image analyzer (CSS-840: System Supply Co., Nagano, Japan). The analysis was conducted at 400 × magnification, with histomorphometric parameters measured in an area of 0.0625 mm2 (0.25 × 0.25 mm). Sixteen to twenty areas were analyzed per sample.

4PBA Treatment

4PBA was administered following the protocol described by Duran et al. [9]. Briefly, 50 mg/day of 4PBA (OrphanPacific, Inc., Tokyo, Japan) was added to the drinking water and changed once per week over a 9-week treatment period (days 28 to 84). Before the treatment, water consumption was as follows: male and female Col1a1G643S/+ mice drank 4 and 3 mL of water per day, respectively. Therefore, 1.25 g and 1.9 g of 4PBA were dissolved in 100 mL of drinking water for male and female mice, respectively, ensuring an approximate intake of 50 mg per day per mouse. The body length of mice was measured from nose to hindquarters at each time point. No adverse events associated with the treatment were observed.

SDS-PAGE and Immunofluorescence of Collagen From Fibroblast

Dermal fibroblasts were isolated from WT and Col1a1G643S/+ mice and cultured in Dulbecco Modified Eagle’s Medium (D-MEM) high glucose with L-glutamine, 10% fetal bovine serum (FBS), 25 U/mL penicillin, and 25 μg/mL streptomycin at 37 °C in a humidified incubator with 5% CO2. For collagen synthesis, fibroblasts were seeded at a predetermined density and cultured in D-MEM high glucose with L-glutamine, 2% FBS, 200 μM L-ascorbic acid phosphate magnesium salt n-hydrate (Wako, Tokyo, Japan), 25 U/mL penicillin, and 25 μg/mL streptomycin.

To analyze collagen, fibroblasts were cultured in collagen synthesis medium for 3 days. Collagen was extracted from the culture medium using pepsin digestion (0.1 mg/mL in 0.1 N HCl; Sigma, St Louis, MO) followed by salt precipitation with 1 M NaCl. The purified collagen was denatured at 95 °C for 3 min, subjected to electrophoresis on 10% polyacrylamide gel, and stained using Coomassie Brilliant Blue R-250 (Bio-Rad, Hercules, CA). The gel was subsequently destained in a solution containing 40% methanol and 10% acetic acid. Images of stained collagen were captured using ChemiDoc XRS Plus imaging system (Bio-Rad, Hercules, CA).

For Immunofluorescence, fibroblasts were cultured in the collagen synthesis medium with or without 5 mM 4-PBA (Sigma, St Louis, MO), for 4 days. After fixing with 4% paraformaldehyde and incubated with primary antibodies against type I collagen (Abcam, Cambridge, UK #ab34710) and protein disulfide isomerase (PDI: Enzo, New York, NY #ADI-SPA-891). This was followed by incubation with secondary antibodies: Alexa Fluor 488-conjugated anti-rabbit IgG (Jackson ImmunoResearch, West Grove, PA) and Alexa Fluor 647-conjugated anti-mouse IgG (Jackson ImmunoResearch, West Grove, PA). Muclei were stained with Hoechst 33,342 solution (Dojindo, Kumamoto, Japan). Fluorescence images were acquired from 25 fields per well using an automated IN Cell Analyzer 6000 microscope (GE Healthcare, Bucks, UK). Quantification of type I collagen accumulation in the ER was performed by calculating the ratio of the area of type I collagen co-localized with PDI to the total PDI-positive area, using IN Cell Investigator software (GE Healthcare, Bucks, UK).

ER Stress Markers

Fibroblasts derived from WT and Col1a1G643S/+ mice were cultured in the collagen synthesis medium for 7 days, with or without 5 mM 4-PBA. After the culture period, total RNA was extracted using RNeasy Mini Kit (Qiagen, Hilden, Germany). cDNA was synthesized from the extracted RNA using ReverTra Ace qPCR RT Master Mix (Toyobo, Osaka, Japan) according to the manufacturer’s instructions. Real-time quantitative PCR (qPCR) was performed with specific primer sets using THUNDERBIRD SYBR qPCR Mix (Toyobo, Osaka, Japan) and QuantStudio 7 Flex Real-time PCR System (Applied Biosystems, Framingham, MA). Gene expression levels were quantified based on the 2−ΔΔCt method and normalized to Gapdh expression.

Statistical Analysis

All statistical analyses were conducted using JMP® Pro software version 17.0.0 (SAS Institute Inc., Cary, NC, USA). Comparisons of mean values were performed using Student’s t test or 2-way ANOVA with Tukey–Kramer post hoc analysis. In the 4PBA treatment analysis, we separated males and females and performed 2-way ANOVA among four groups: wild type with placebo, wild type with 4PBA, Col1a1G643S/+ with placebo, and Col1a1G643S/+ with 4PBA. Differences with p-values < 0.05 were considered statistically significant.