Animal Experiments

The experimental procedures involving animals were approved by the Ethics Committee in Gothenburg, Västra Götaland, or Umeå University, and all methods were carried out in accordance with relevant guidelines and regulations and in accordance with ARRIVE guidelines. Female mice were used to ensure consistency across the different models and because osteoporosis-related fractures occur more often in women than in men. The mice were housed in a standard animal housing facility with 12-h of darkness and 12-h of light photo period. The temperature was controlled (22 °C) and food and water were available ad libitum.

Mouse Model of Antigen-Induced Arthritis (AIA)

At day 0, seven-week-old wildtype (WT) C57BL/6N female littermate mice (n = 14) (Charles River, Germany) were primary immunized with 100 μl of 2 mg/ml of methylated bovine serum albumin (mBSA; Sigma-Aldrich, Stockholm, Sweden) dissolved in phosphate-buffered saline (PBS) and emulsified 1:1 in Freund’s complete adjuvant (Sigma-Aldrich) (Fig. 1B). A total volume of 100 μl was injected intradermally at the base of the tail (50 μl on each side) [14].

Fig. 1

TNF-α reduces Wnt16 mRNA expression in osteoblasts, but WNT16 treatment does not prevent periarticular bone loss in antigen-induced arthritis. A Wnt16 mRNA expression in primary mouse calvarial osteoblasts cultured for 24 h in control medium with or without TNF-α (50 ng/ml). Values are presented as mean ± SEM (n = 4). Data were analyzed using an unpaired Student’s t-test. B Schematic representation of the antigen-induced arthritis model. Upper panel: Control group after secondary immunization (left knee = arthritic; right knee = non-arthritic). Lower panel: Liposome-treated group (left knee = arthritic + liposomes with WNT16; right knee = arthritic + empty liposomes). (C-F) Peripheral quantitative computed tomography (pQCT) analysis of bone mineral density (BMD): total BMD in the metaphyseal regions of the distal femur C and proximal tibia D, and trabecular BMD in the same regions of the distal femur E and proximal tibia F. Data were collected from bones adjacent to non-arthritic and antigen-induced arthritic mouse knees treated with PBS, as well as from bones adjacent to arthritic knees treated with liposomes with WNT16 or empty liposomes. Samples were obtained from 9-week-old female wild-type mice. Values are mean ± SEM (n = 5–9 mice per group). Data assessed by paired Student’s t-test. lipo. = liposomes; inj. = injection; w = with; and w/o = without

Liposomes were prepared by dissolving 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC; 850345C, Avanti Polar Lipids, Inc., Alabaster, AL) in chloroform and drying the solution to a thin lipid film in a round-bottom flask using a gentle nitrogen stream, followed by vacuum for at least 2 h [12]. The lipid film was then hydrated with 25 μg WNT16 protein (7790-WN-025/cf; R&D Systems) in 400 μl PBS prewarmed to 32 °C. The final lipid concentration was 13.2 mg/ml. Control liposomes were prepared by hydration in PBS alone. The mixture was vortexed until the solution appeared cloudy and no visible lipids remained at the bottom of the flask. Liposomes were subsequently extruded by passing the suspension 51 times through a 100-nm polycarbonate membrane (Whatman, UK) at 32 °C. After cooling to room temperature, liposomes were stored at 4 °C until use.

At days 6, 9, and 12, nine of the mice received a 30 µl intra-articular injection of WNT16 liposomes (62.5 µg/ml WNT16 in PBS) into the left hindlimb knee joint, whereas the right knee joint received a 30 µl intra-articular injection of liposomes without WNT16 diluted in PBS [12]. A second local immunization was performed at day 7, where 20 mg/ml mBSA (dissolved in PBS) was mixed 1:1 with WNT16 liposomes (125 µg/ml WNT16 in PBS), and the mice received a 30 µl intra-articular injection into the left hindlimb knee joint [12, 14]. The right knee received a 30 µl intra-articular injection of 20 mg/ml mBSA (dissolved in PBS) and liposomes without WNT16 (dissolved in PBS) at a ratio of 1:1.

To control that the second immunization induced arthritis, five of the primary immunized mice were given a 30 µl intra-articular injection of 10 mg/ml mBSA in PBS in the left hindlimb joint at day 7, whereas the right joint received a 30 µl intra-articular injection of PBS. On days 6, 9, and 12, these mice received 30 µl intra-articular injections of PBS in both hindlimbs joints.

Fourteen days after the primary immunization, the mice were euthanized using Ketador (Richter Pharma) mixed with Dexdomitor (Orion Pharma), followed by exsanguination and cervical dislocation. The hindlimbs were cleaned from the skin, separately fixed in formalin and stored in ethanol until CT analyses.

Mouse Model of Inflammation-induced Bone Loss Caused by Toll-like Receptor 2 Activation

Inflammation-induced bone loss was initiated by subcutaneous injection of 75 μg synthetic diacylated lipopeptide, Pam2CSK4 (PAM2; InvivoGen, Toulouse, France) in 100 µl at the top of the skull [12, 17] of 6-week-old Obl-Wnt16 (WNT16 expression driven by the rat procollagen type I alpha1 promoter) female mice and WT female littermate controls, on C57BL/6N background [13]. All experiments were carried out on female mice born from crossing a male Obl-Wnt16 mouse with a female C57BL/6N mouse. PAM2 is an agonist of Toll-like receptor (TLR) 2. NaCl (9 mg/ml) was used as a vehicle. All injections were done under anesthesia with isoflurane (Baxter Medical AB, Kista, Sweden). Five days post-inoculation with PAM2, the mice were euthanized using Ketador (Richter Pharma) mixed with Dexdomitor (Orion Pharma), followed by exsanguination and cervical dislocation. The skull bones and femurs were dissected, separately fixed in formalin, and stored in ethanol until CT analyses. Trabecular-rich vertebral bodies were dissected, placed in RNAprotect Tissue Reagent (76 106, Qiagen), and stored at − 80 °C until total RNA extraction (Trizol Reagent, 15596018, Thermo Fisher Scientific) followed by the RNeasy Mini Kit (74116, Qiagen). To obtain cDNA, the RNA was reversed transcribed (4368814, Applied Biosystems) and real-time PCR analyses were performed using the StepOnePlus Real-Time PCR system (Thermo Fisher Scientific) using the predesigned Mm00446420_m1 assay for Wnt16. Relative gene expression was calculated using the 2−∆∆Ct method using the ribosomal subunit 18S as internal standard.

Mouse Model of S. Aureus-induced Arthritis

In the S. aureus arthritis experiments, food was placed at the bottom of the cage and water bottles with extra-long tips were provided to minimize discomfort for mice with inflamed paws during feeding. Pre-prepared batches of S. aureus strain LS-1, which produces Toxic Shock Syndrome Toxin 1 (TSST-1), were thawed, washed and diluted. Nine-week-old Obl-Wnt16 female mice [13] and WT female littermate controls on C57BL/6N background were inoculated via the tail vein with 0.2 ml of S. aureus at a concentration of 5 × 107 [] CFU/mouse or PBS as control.

Following inoculation, mice were weighed daily, and clinical assessments of S. aureus-induced arthritis were performed by observers blinded to the treatment group. The observers visually inspected each mouse’s wrists, ankles, fingers, and toes throughout the infection period [18]. Arthritis was defined as erythema and/or joint swelling. To assess arthritis severity, a clinical scoring system from 0 to 3 was used for each paw, as previously described [18, 19]. After eight days, the mice were euthanized using Ketador (Richter Pharma) mixed with Dexdomitor (Orion Pharma), followed by exsanguination and cervical dislocation. Lumbar vertebra 5 were dissected, fixed in formalin and stored in ethanol until CT analyses.

Assessment of Bone ParametersDetermination of Periarticular Bone Mineral Density (BMD)

To determine periarticular bone loss in the AIA mouse model, both the femur and tibia were analyzed using Stratec pQCT XCT Research M software version 5.4B (Norland) at a resolution of 70 μm, as described previously [20]. Total and trabecular BMD were determined using metaphyseal scans performed 0.4 mm from the growth plate, proximally in the distal femur and distally in the proximal tibia, respectively. The trabecular bone compartment was defined as the inner 45% of the bone area.

High-resolution Micro-computed Tomography (μCT)

Bone morphology was analyzed using high-resolution micro-computed tomography (μCT). A SkyScan 1172 system was used for the skull and a 1275 system for the femur and fifth lumbar vertebra (L5) (Bruker MicroCT, Aartselaar, Belgium). Scanning was conducted with an X-ray tube voltage of 50 kV for the skull and 40 kV for the femur and vertebra. The current was set to 200 μA. An aluminum filter of 0.5 mm was used for the skull, and 1 mm for the femur and vertebra. The scanning angular rotation was 180° with an angular increment of 0.70°, and an isotropic voxel size of 14 μm for the skull and 7 μm for the femur and vertebra. For femur analysis, the trabecular bone proximal to the distal growth plate was selected within a conforming volume of interest, excluding cortical bone, starting 504 μm from the growth plate and extending 210 μm proximally. For skull bones, the volume of interest was 6 mm wide, extending from 2 mm anterior to the posterior tip of the frontal bone to 1 mm posterior to the anterior tip of the interparietal bone, maintaining bilateral symmetry along the interfrontal and sagittal sutures. In the vertebra, the trabecular bone in the vertebral body caudal of the pedicles was selected for analysis within a conforming volume of interest (cortical bone excluded) commencing at a distance of 7 μm caudal of the lower end of the pedicles and extending a further longitudinal distance of 245 μm in the caudal direction. Datasets were reconstructed using NRecon (version 1.6.9.8, Bruker) and further analyzed with CTAn software (version 1.20.8.0, Bruker).

Calvarial Osteoblast Cultures and Gene Expression Analysis

Osteoblasts from CsA mice from our own inbred colony were isolated from 2–3-day-old mouse parietal bones by time sequential digestion with bacterial collagenase [21]. Cells from digestions 6–10 were used and plated at a density of 104 cells/cm2 and incubated with or without TNF-α (50 ng/ml) for 24 h. RNA was isolated and gene expression analysis performed using TaqMan Fast Advanced Master Mix as described previously [22]. Amplifications were performed with the StepOnePlus Real-Time PCR system. β-actin was used as housekeeping gene.

Statistical Analyses

All statistical analyses were performed using GraphPad Prism (version 10.3.1) and data are presented as mean ± standard error of the mean (SEM). Depending on the experimental design, statistical significance was assessed using either the paired or unpaired Student’s t-test, or the Mann–Whitney U test, as appropriate. To assess the effects of genotype, treatment and interaction, two-way ANOVA was used. Mouse survival rates were analyzed using the Kaplan–Meier method, with group comparisons assessed by the log-rank test. A P value of < 0.05 was considered statistically significant.