Animals
Mice with the Lrp5 conditional knock-in HBM alleles (p.G171V and p.A214V), were previously described [32, 33]. Animals used in this study were initially crossed to a ubiquitous Cre (CMV-Cre) transgenic mouse to excise the neomycin-resistant-cassette flanked by loxP sites and then backcrossed to C57BL/6J animals to breed out the Cre transgene. We used wildtype (+/+) and heterozygous Lrp5-A214V (A/+) or Lrp5-G171V (G/+), as previously described [36]. Female and male 3-month-old +/+ (wild-type, WT) and heterozygous Lrp5-A214V (A/+) or Lrp5-G171V (G/+) mice were treated daily with either 3 mg/kg LGK974 (Novartis) or with vehicle for 5 weeks by oral gavage at a dosing volume of 10 µL/g animal body weight. LGK974 is poorly soluble and is therefore administered as a suspension in 0.5% methylcellulose / 0.5% Tween 80 [23, 53]. Lrp5 HBM mice were maintained in accordance with institutional animal care and use guidelines, and experimental protocols were approved by the Institutional Animal Care and Use Committee of the Van Andel Institute Mice were housed in Thoren Maxi-Miser IVC caging systems with a 12-h/12-h light/dark cycle and fed a breeder rodent diet containing 23% protein and 24% fat with an energy content of 19.3 MJ/kg (5021, LabDiet St. Louis MO) with food and water provided ad libitum.
Female three-month-old wild-type C57BL/6J (Charles River Laboratories Germany) and Sost deficient (Sost KO) mice [34] were treated twice daily for six weeks with either vehicle (0.5% methylcellulose / 0.5% Tween 80 in water) or GNF-6231 suspension at 0.3, 1 or 3 mg/kg by oral gavage at a dosing volume of 10 µL/g animal body weight. Mice were housed at 22°C with a 12-h/12-h light/dark cycle and were fed a standard rodent diet containing 18.2% protein and 3.0% fat with an energy content of 15.8 MJ/kg (3890, Provimi Kliba SA) with food and water provided ad libitum. Procedures with Sost mice conformed to Swiss federal law for animal protection controlled by the Basel-Stadt Cantonal Veterinary Office, Switzerland.
For dynamic histomorphometry analysis of Lrp5 HBM models, animals were administered the fluorescent dye calcein (10 mg/kg split equally between i.p. and sub.q.; MilliporeSigma) 17 and 7 d before euthanasia. All animals were euthanized at 17-wk-old, following five weeks of treatment. Femurs were isolated and fixed in 10% neutral-buffered formalin (NBF) at room temperature for 48 h then changed to 70% ethanol before analysis and histological processing.
Table 1 details the number of specimens analyzed for each application.
DEXA
We performed whole body dual-energy X-ray absorptiometry (DEXA) to measure areal bone mineral density (aBMD; gm cm− 2) and bone mineral content (BMC; gm) for the postcranial skeleton. A PIXImus II bone densitometer (GE Lunar) was used for analysis. aBMD values were collected at 0, 7,14, 21, 28, and 35 days into treatment.
Microcomputed Tomography (µct)
The Lrp5 HBM models were analyzed using a SkyScan 1172 µCT system (Bruker MicroCT: Kontich, Belgium). Femora were scanned using an X-ray voltage of 60 kV, current of 167 µA, and 0.5 mm aluminum filter. Images were obtained with 2000 x 1200 pixel resolution and 7.98 µm pixel size. Femoral images were reconstructed using NRecon 1.7.4.6 (Bruker MicroCT). The mineralized tissue was oriented, and a volume of interest (VOI) was defined using DataViewer 1.5.6.3 (Bruker MicroCT). Regions of interest (ROI) were defined for cortical and trabecular bone using CTAn 1.18.8.0 (Bruker MicroCT). A trabecular ROI was drawn in the distal epiphysis for each femur, beginning 0.25 mm proximal to the growth plate and 2.5 mm in height. To define the position of each cortical ROI, the distal end of the region was set to be 45% of the femur length. The ROI was 0.8 mm in height toward the proximal end of the bone, within the midshaft. Trabecular 3D analysis was performed to quantify bone mineral density (BMD), bone volume/tissue volume (BV/TV), bone surface/bone volume, trabecular thickness (Tb.Th), trabecular separation, (Tb.Sp), and trabecular number (Tb.N). Cortical 2D analysis was performed to quantify tissue mineral density (TMD), tissue area, bone area, cortical area fraction (bone area/tissue area, CAF), cross-sectional thickness, and bone perimeter.
Femora from Sost knockout mice were analyzed by high-resolution ex vivo mCT using a vivaCT 40 instrument (Scanco Medical AG) at an isotropic nominal resolution of 6 µm. A Gaussian filter (σ = 0.7, support of one voxel) was used in all analyses to suppress noise and a segmentation threshold of 275 was applied.
Peripheral Quantitative Computed Tomography (Pqct)
Treatment efficacy was assessed longitudinally by in vivo pQCT under isoflurane inhalation using an adapted Stratec-Norland XCT‐2000 instrument fitted with an Oxford (Oxford, UK) 50‐mmX‐ray tube (GTA6505M/LA) and a 0.5‐mm‐diameter collimator (voxel size: 0.07 mm x 0.07 mm x 0.4 mm).
Static And Dynamic Histomorphometry
Detailed methods for static and dynamic histomorphometry tissue preparation and analysis are found in our published protocol for skeletal tissue phenotyping [54].
Briefly, fixed fluorochrome-labeled femurs were dehydrated in graded ethanol and cleared using xylene. Samples were infiltrated and embedded in plastic using 85–100% MMA and 15% dibutyl phthalate until polymerization. Femurs were sectioned coronally, deplasticized, Goldner's Trichrome stained, and coverslipped for static histomorphometry.
Additional femurs were fixed and decalcified in 10% EDTA for 14 d, embedded in paraffin, and sectioned sagittally. Sections were stained with hematoxylin and eosin (H&E) or tartrate-resistant acid phosphatase (TRAP kit 387A; Millipore Sigma, St. Louis, MO, USA). Indices measured for Goldner's Trichrome stained slides included bone volume/tissue volume (BV/TV), adipocyte number/tissue volume (Ad.V/TV), osteoid volume/bone volume (OV/BV), osteoid surface/bone surface (OS/BS), and the number of osteoblasts/bone surface (N.Ob/BS). Parameters measured for TRAP-stained slides included osteoclast surface/bone surface and the number of osteoclasts/bone surface. All histomorphometric analysis was performed using BIOQUANT OSTEO software (v19.2.60; BIOQUANT Image Analysis Corporation, Nashville, USA).
For dynamic histomorphometry, femoral midshafts were cross-sectioned and coverslipped. Endocortical mineral apposition rate (MAR), endocortical bone formation rate (BFR), periosteal MAR, and periosteal BFR were measured for each sample.
Statistical Analyses
All analyses except DEXA were performed using the same methods. All percentage data, including BV/TV (µCT), CAF, BV/TV (histomorphometry), OV/BV, OS/BS, and Oc. S/BS were analyzed via beta regression. All other outcomes were analyzed via robust linear regression with natural log (y + 0.01) transformation to improve the models' fit. All analyses and outcomes were Benjamini-Hochberg false discovery rate (FDR) adjusted to account for multiple testing. To determine if the two groups were similar, an equivalence test was performed such that any non-significant difference with a 95% confidence interval entirely within the range of 0.9–1.1 (fold-change [FC]) is considered equivalent. For any differences we could confidently say are smaller than 10%, we can conclude that there is sufficient evidence to support a hypothesis that there is no difference between the groups.
DEXA data were analyzed using linear mixed-effects models with random slope and intercept for each animal. FDR-adjusted linear contrasts were used to test specific hypotheses. Analyses were stratified by sex. For all models, a 3-way interaction between days, genotype, and treatment was initially fit. All two-way interactions were also tested and dropped if p-values exceeded 0.15 and saved for the interaction between days and treatment. This allowed for testing of a treatment effect in all comparisons.