A beneficial effect of a novel DKK1 monoclonal antibody and its sequential alendronate on bone and muscle properties of orchiectomized mice

Abstract

Background

Insights into basic bone biology and genetic disorders have led to advances in the treatment of osteoporosis. However, new drugs for the treatment of osteoporosis in men remain insufficient. We investigated the efficacy of a novel anti-Dkk1 monoclonal antibody (Dkk1-mAb) and its sequential alendronate treatment in a mouse model of osteoporosis induced by orchiectomy.

Methods

A specific Dkk1-mAb was generated using the hybridoma technology. A total of 40 male C57BL/6 mice aged 12 weeks underwent orchidectomy or sham surgery. At 8 weeks postoperatively, the orchiectomized mice were randomly divided into 4 groups (8 in each group) to receive treatment of Dkk1-mAb, alendronate, Dkk1-mAb followed by alendronate, or placebo. After 8 weeks of treatment, the mice were euthanized, and bone mineral density (BMD), skeletal mechanical properties, bone histology, bone turnover biomarkers, serum levels of Dkk1 and sclerostin, muscle properties, and activity of the Wnt signaling were evaluated.

Results

After 8 weeks of treatment, serum Dkk1 levels were lower in Dkk1-mAb and sequential groups than placebo group (P < 0.001, P < 0.05). L1-5 BMD was higher in Dkk1-mAb, alendronate, and sequential groups by 11.9 %, 22.8 %, and 23.1 % than placebo group (P < 0.05 or P < 0.001), with BMD at left femur increased by 12.8 % and 21.0 % in alendronate and sequential groups than placebo group (P < 0.05 or P < 0.001). The sequential group exhibited higher vertebral trabecular volume/total volume (39.9 %), bone surface/total volume (25.8 %), trabecular BMD (4.0 %), and trabecular thickness (13.1 %) than Dkk1-mAb group (P < 0.01 or P < 0.001), and higher cortical BMD, thickness than Dkk1-mAb (2.9 %, P < 0.01; 6.6 %, P < 0.05) and alendronate (2.2 %, 5.7 %, all P < 0.05) groups. All treatment groups showed greater bone strength, cross-sectional area of muscle, and grip strength than placebo group. The expression of Lef1 was significantly increased in bones of Dkk1-mAb and sequential groups. No pathological abnormalities in vital organs were found in all groups, suggesting that Dkk1-mAb had a good safety profile.

Conclusions

The novel Dkk1-mAb has therapeutic potential in osteoporosis, which can increase BMD and bone strength, improve bone microarchitecture and muscle performance of orchiectomy-induced osteoporotic mice through activating the WNT pathway, and its sequential treatment with alendronate achieves additive benefits.

The Translational Potential of this Article

This study found that the novel Dkk1-mAb can increase BMD and bone strength, improve bone microarchitecture and muscle performance of orchiectomy-induced osteoporosis mice, and sequential treatment with Dkk1-mAb and alendronate achieves additive benefits. Dkk1-mAb has translational potential for the treatment of osteoporosis.

Graphical abstract

1. Introduction

Although osteoporosis and its consequences, such as fragility fractures, lead to a significant and continuously growing public health concern in older men [1,2], new drugs for the treatment of osteoporosis in men remain insufficient. Bone health depends on the continuous renewal and homeostasis of bone tissue. Bone remodeling is a tightly regulated process involving a delicate balance between bone formation and resorption [3]. Current treatments for osteoporosis target either osteoclasts, by inhibiting bone resorption, or osteoblasts, by stimulating bone formation [4,5].

The Wnt signaling pathway plays a critical role in regulating bone metabolism. Inhibitors of this pathway, such as sclerostin (Sost) and Dickkopf-related protein 1 (Dkk1), are key regulators of bone formation and resorption, making them attractive targets for treatment of osteoporosis [6,7]. Sost, an antagonist of the Wnt signaling pathway, is primarily secreted by osteocytes and can inhibit bone anabolism. Several Sost inhibitors have been developed, of which romosozumab has been approved for treatment of osteoporosis [8,9]. Dkk1, another inhibitor of the Wnt pathway, is expressed by mature osteoblasts and osteocytes and plays a crucial role in regulating bone homeostasis [7]. A series of studies have shown that inhibiting Dkk1 is beneficial for multiple myeloma and glucocorticoid-induced osteoporosis [10,11]. However, the effects of anti-Dkk1 monoclonal antibodies (Dkk1-mAbs) on osteoporosis in man remain unclear. Recent studies have demonstrated that a variety of osteoanabolic drugs can effectively reduce the risk of fractures, and the sequential administration of an antiresorptive agent is recommended after discontinuing bone formation promoter. Nevertheless, the effectiveness of Dkk1-mAb sequential bone resorption inhibitor therapy is still uncertain.

Therefore, we utilized a biopharmaceutical research platform to generate a novel Dkk1-mAb via the hybridoma technology. We evaluated the effects of the Dkk1-mAb and its sequential treatment with alendronate on the bone and muscle of osteoporotic mice induced by orchiectomy (ORX).

2. Methods and materials

2.1. Generation of the Dkk1-mAb

A novel Dkk1-mAb was generated using the hybridoma technology through the biopharmaceutical platform (Taizhou Mabtech Pharmaceuticals Co., Ltd. China). To this end, the Dkk1 gene was cloned into a eukaryotic expression vector. Subsequently, CHO cells were transfected with this expression construct. After 10 d, high-expression stable cell lines were selected. The antigen was purified and then verified via SDS-PAGE. Three 6-week-old female Balb/c mice were immunized subcutaneously with 100 μg of the purified Dkk1 protein every two weeks for three rounds. Serum concentrations of anti-Dkk1 antibodies were assessed using enzyme-linked immunosorbent assay (ELISA). To produce Dkk1-mAbs, hybridomas were generated by fusing spleen cells of the immunized mice with NS-1 myeloma cells, and positive clones were identified through ELISA. The Dkk1-mAb was purified via Protein-A chromatography, and the purity of the antibody was confirmed via SDS-PAGE. The affinity, biological activity, and thermal stability of the Dkk1-mAb were assessed using the Biacore surface plasmon resonance technology, luciferase reporter assay, and differential scanning calorimetry, respectively. In the luciferase reporter assay and differential scanning calorimetry, DKN-01 served as the control antibody. DKN-01 is a Dkk1-targeted antibody generated using proprietary antibody construction and expression platform (Taizhou Mabtech Pharmaceuticals Co., Ltd. China).

2.2. Establishing the male osteoporosis mouse model and its intervention

This study was approved by the Animal Welfare and Ethics Committee of Peking Union Medical College Hospital (PUMCH) (Approval No.: XHDW-2024-24).

Forty 10-week-old male C57BL/6 mice, weighing 24–28 g, were acclimated for 2 weeks in an AAALAC-certified, specific pathogen-free facility in PUMCH. After the acclimation, the mice underwent either a bilateral orchiectomy (ORX) (n = 32) or a sham operation (n = 8) under 2 % isoflurane anesthesia.

At 8 weeks postoperatively, the mice of ORX group were randomly assigned to the following four groups (8 in each group): Dkk1-mAb group, which received subcutaneous injections of 25 mg/kg Dkk1-mAb twice a week for 8 weeks [12]; alendronate group, which received subcutaneous injections of 1 mg/kg alendronate (Merck & Co., Inc., Rahway, NJ, USA) weekly for 8 weeks; sequential treatment group, which received subcutaneous injections of 25 mg/kg Dkk1-mAb twice a week for 4 weeks, followed by 1 mg/kg alendronate weekly for 4 weeks; and ORX and Sham groups, which received subcutaneous injections of 1 mL/kg phosphate-buffered saline weekly for 8 weeks.

Throughout the study, body length, tail length, and body weight were measured weekly. After 8 weeks of treatment, all mice were euthanized. To detect de novo bone formation, subcutaneous injections of 20 mg/kg calcein (Sigma–Aldrich, St. Louis, MO, USA) were administered on the 10th and third days before euthanasia. Blood, bone, muscle, adipose, and internal-organ samples were collected postmortem.

2.3. Measurement of serum levels of bone-turnover biomarkers

Blood samples were allowed to coagulate at room temperature, followed by centrifugation at 3000 rpm for 10 min at 4 °C to separate the serum. Serum levels of β-isomerized carboxy-telopeptide of type I collagen (β-CTX, a bone resorption marker), procollagen I N-terminal peptide (P1NP, a bone formation marker), Dkk1, and Sost were measured using ELISA. P1NP and β-CTX were detected using kits from Immunodiagnostic Systems Inc. (Boldon, UK), and Dkk1 and Sost were detected using kits from R&D Systems, Inc. (USA). All the tests were performed in the Key Laboratory of Endocrinology, Department of PUMCH, by following the instructions of the manufacturers.

2.4. Measurement of bone mineral density and body composition

Bone mineral density (BMD) and body composition of mice were measured by dual-energy X-ray absorptiometry (DXA, OsteoSys Co., Ltd., Korea) using small animal measurement software after 8 weeks of ORX or sham surgeries, and after 4 weeks and 8 weeks of treatment. BMD and bone mineral content of the total body, L1-L5 lumbar vertebra and femur, as well as fat mass and lean body mass were measured after intraperitoneal anesthetized by 1.25 % tribromoethanol (Avertin, Nanjing Aibei Biotechnology Co., Ltd., China).

2.5. Assessment of bone micro-architecture

Specimens of the left femur and L5 vertebra were fixed in 4 % paraformaldehyde. Subsequently, and bone microstructural parameters were evaluated using micro–computed-tomography (Venus Micro CT, PINGSENG Healthcare Kunshan Inc., China) under the following ex vivo scanning conditions: X-ray tube voltage of 90 kV, current of 90 μA, and a voxel size of 26 μm [13]. The region of interest (ROI) for the trabecular bone was defined as 1.5 mm below the growth plate, extending 1 mm proximally into the distal metaphysis. The central cancellous area of the L5 vertebral body was also analyzed for trabecular microarchitecture. For the cortical bone, the ROI was defined as the 1-mm section at the midpoint of the diaphysis of the left femur. Microstructural parameters included trabecular bone mineral density (Tb.BMD), bone volume per total volume (BV/TV), bone surface area per total volume (BS/TV), trabecular thickness (Tb.Th), trabecular number (Tb.N), trabecular separation (Tb.Sp), trabecular connectivity density (Conn.D), cortical bone mineral density (Ct.BMD), cortical thickness (Ct.Th), cortical bone area (Ct.Ar), cortical bone area to total area ratio (Ct.Ar/Tt.Ar), endosteal area, endosteal area ratio, and endosteal perimeter. Three-dimensional images were reconstructed using the Recon software (PINGSENG Healthcare Kunshan Inc., China).

2.6. Evaluation of bone mechanical properties

A three-point bending test of the right femur and a compression test of the L4 vertebra were performed to evaluate bone mechanical properties using a material testing system (Model 858 Mini Bionix II, MTS Systems, Minnesota, USA). Maximum load, stiffness, yield load, and breaking load were derived from the load–displacement curves of the femurs. The maximum load for the vertebra was obtained from the force-displacement measurement.

2.7. Evaluation of muscle function

Grip strengths of the forelimb and four-limb were measured at baseline, after 4 weeks and 8 weeks of treatment using a dynamometer (ZS-ZL, ZHONGSHI SCIENCE & TECHNOLOGY, China). Each mouse was tested three times, and the peak force was recorded. Coordination, strength, and balance of the mice were assessed using a rotary rod fatigue tester (ZL-200C, Anhui Yaokun Biotechnology, China). Mice were placed on the rod at 5 rpm, with the speed increasing to 40 rpm over 240 s. The time until the mouse fell from the rod was recorded, and the mean times from two trials were calculated for each mouse.

2.8. Histological analyses of bone and muscle

The right tibia was decalcified, embedded in paraffin, sectioned to a thickness of 4 μm using a microtome (Leica RM2016, Leica Microsystems), and stained with hematoxylin and eosin (HE) and TRACP (Servicebio, Cat# G1050). The left tibia and L3 vertebra were embedded in methyl methacrylate without decalcification. The embedded samples were sectioned into 10-μm-thick slices, de-plasticized, and stained with a Von Kossa stain kit (Servicebio, Cat# G1043). For unstained slices, the distance between the two calcein labels was measured to assess the bone mineral apposition rate (MAR). Ct.MAR measurements were specifically taken along the periosteal surface of the cortical bone. ImageJ was used for analysis following ASBMR standards [14].

The right quadriceps were dehydrated and then embedded in paraffin. Tissue sections at 4-μm thickness were prepared, deparaffinized with an eco-friendly solution, and stained with HE. Cross-sectional areas of muscle fibers were quantified using ImageJ under a light microscope, and the average area per sample was calculated. Immunohistochemical (IHC) staining for the key myogenic regulatory factors of MyoD1 and Myogenin were also performed in the muscles.

2.9. Assessment of the safety of Dkk1-mAb

The upper half of the heart, including the aortic root, was saline-perfused, embedded in Optimal Cutting Temperature compound (Sakura Finetek, Tokyo, Japan), frozen in liquid nitrogen, and cryosectioned at a thickness of 10 μm. Atherosclerotic plaques were assessed by the surface area of Oil-Red-O–positive lesions on the aortic arches, and the ratio of plaque area to the total aortic root cross-sectional area was calculated [15].

Important visceral organs (heart, liver, spleen, lung, and kidney) and perirenal adipose tissue were dehydrated, embedded in paraffin, sectioned into 4.0-μm-thick slices, and stained with HE. These samples were observed and imaged using a light microscope. Additionally, serum levels of interleukin-6 (IL-6) and tumor necrosis factor α (TNF-α) were measured using ELISA (Thermo Fisher Scientific, USA).

2.10. Assessment of the expression of key genes in WNT pathway of bone and muscle

Expression levels of important members of the Wnt pathway were evaluated in the radius bones and quadriceps femoris using reverse transcription–quantitative polymerase chain reaction (RT-qPCR). Total RNA was extracted from the radius bones and quadriceps using TRIzol reagent and reverse-transcribed into cDNA by using the PrimeScript RT Reagent Kit (Takara, Kusatsu, Japan). The expression levels of Runx2, Lef1, Ctnnb1, Axin2, Dkk1, and Sost were measured using Viia 7 Real-Time PCR system (Life Technologies, USA) and TB Green Premix Ex Taq II (Tli RNase H Plus, Takara). Relative mRNA levels were calculated using the 2^−ΔΔCT method and normalized to Gapdh mRNA levels. The sequences of the primers used in RT-qPCR were shown in Supplementary Table 1.

2.11. Statistical analysis

The Shapiro–Wilk or Kolmogorov–Smirnov test was employed to assess the normality of data distribution. Normally distributed continuous variables were expressed as mean ± SD or SE, while non-normally distributed variables were expressed as the median (interquartile range, IQR). For comparisons, normally distributed data were analyzed using an independent sample t-test (two groups) or one-way ANOVA (multiple groups). When ANOVA was significant, post-hoc pairwise comparisons were performed using Tukey’s test, which adjusts for multiple comparisons by controlling the family-wise error rate (FWER). Non-normally distributed data were analyzed using the Mann–Whitney U test (two groups) or the Kruskal–Wallis test (multiple groups). Upon a significant Kruskal–Wallis test, Dunn’s post-hoc test with FWER correction was applied for pairwise comparisons. Longitudinal analysis was performed using a linear mixed-effects model to evaluate the impact of the therapeutic interventions over time (pre-versus post-treatment). Statistical analyses were performed using SPSS 25.0 (IBM Corp., Armonk, NY, USA) and GraphPad Prism 10.0 (GraphPad Software, San Diego, CA, USA). For all post-hoc tests (Tukey’s and Dunn’s), the P-values reported were those adjusted for multiple comparisons. A P-value <0.05 was considered to indicate statistical significance.

3. Results

3.1. Generation of a novel Dkk1-mAb with good performance

A new Dkk1-mAb was successfully generated using the hybridoma technology (Fig. 1A–C). The KD (M) of the antibody was measured at 2.036 × 10⁻¹¹ using the Biacore Surface Plasmon Resonance technology, indicating good affinity. In the luciferase reporter gene assay, the antibodies effectively relieved Dkk1 inhibition of Wnt3a, activating Tcf/Lef transcription factors and increasing downstream gene expression, thereby demonstrating their role in activating the Wnt/β-catenin signaling pathway (Fig. 1D). Thermal analysis revealed that the Dkk1-mAb had a higher Tm value than control antibody of DKN01, indicating nice thermal stability (Fig. 1E).

Fig. 1.

SDS-PAGE analysis, bioactivity assay and thermal analysis of DKK1-mAb A. DKK1 antigen bands in SDS-PAGE The band corresponds to the expected molecular weight of the DKK1 antigen, confirming successful expression and purification of DKK1 antigen in CHO cells B. DKK1-mAb bands in SDS-PAGE The presence of intact antibody structures in non-reducing conditions indicates proper assembly of the DKK1-mAb C. Heavy and light chains of DKK1-mAb in SDS-PAGE Reducing conditions result in the separation of the heavy and light chains, validating the antibody’s structural integrity D. Bioactivity Assay of DKK1-mAb RGA DKK1-mAb could induce downstream luciferase gene expression. The detected luciferase signal intensity was positively correlated with the antibody concentration. The humanized antibody exhibited a significantly greater upregulation of gene expression than the control antibody DKN01. E. Thermal Analysis of DKK1-mAb DKK1-mAb exhibited a higher Tm value than the control antibody DKN01, indicating superior thermal stability.

3.2. Generation of a mouse model with male osteoporosis

After 8 weeks of surgery, BMD at the total body, femurs, and L1–5 vertebrae was significantly lower in the ORX group than sham group (all P < 0.001) (Table 1). Likewise, lean mass, lean mass percentage, and body weight were lower, with increased fat percentage, in ORX group (all P < 0.001) (Table 1). Additionally, four-limb grip strength was lower in ORX group than the sham group (P < 0.01) (Table 1), indicating that the ORX mice exhibited characteristics of osteoporosis and sarcopenia.

Table 1.

Comparison of bone mineral density, body composition, grip strength between ORX and Sham group after 8 weeks of operation.

	ORX (n = 32)	Sham (n = 8)	P value
General BMD (mg/cm2)	78.7 ± 3.0	85.1 ± 4.1	<0.001
BMD at L1-5 (mg/cm2)	34.8 ± 3.9	45.5 ± 4.6	<0.001
BMD at left femur (mg/cm2)	70.3 ± 5.2	88.5 ± 9.6	<0.001
BMD at right femur (mg/cm2)	68.1 ± 5.2	89.9 ± 10.2	<0.001
Body composition
BMC (g)	0.558 ± 0.044	0.565 ± 0.039	0.684
Fat (g)	5.636 ± 0.946	5.229 ± 0.958	0.285
Fat (%)	20.504 ± 1.777	16.337 ± 2.648	<0.001
Lean (g)	21.696 ± 1.395	26.721 ± 1.110	<0.001
Lean (%)	77.900 ± 1.725	82.225 ± 2.627	<0.001
Total Weight (g)	28.278 ± 2.405	32.625 ± 1.597	<0.001
Forelimbs grip strength (N)	0.79 ± 0.15	0.89 ± 0.14	0.051
Four limbs grip strength (N)	1.64 ± 0.17	1.84 ± 0.22	0.008
Time to fall in Rotarod test (s)	27.1 ± 23.9	31.6 ± 20.2	0.396

Bold values indicate that there was a signification difference between the two groups.

ORX: orchiectomy, BMD: bone mineral density, BMC: bone mineral content.

3.3. Effects of different treatments on bone turnover markers

After 8 weeks of treatment, serum β-CTX levels were lower by 48.5 %, 58.3 %, and 54.5 % in the Dkk1-mAb, sequential treatment, and alendronate groups, respectively, than in the ORX group (all P < 0.001). Serum P1NP levels were lower by 70.1 % and 59.9 % in the sequential treatment and alendronate groups, respectively, than in the ORX group (all P < 0.001), while it was higher by 49.4 % in Dkk1-mAb group than ORX group (P < 0.001) (Fig. 2A and B).

Fig. 2.

Effects of different treatments on serum bone turnover markers and expression levels of DKK1 and SOST in bone tissue A. Serum levels of CTX measured by ELISA in different groups (n = 7–8 per group) B. Serum levels of P1NP measured by ELISA in different groups (n = 7–8 per group) C. Serum levels of DKK1 measured by ELISA in different groups (n = 7–8 per group) D. Serum levels of SOST measured by ELISA in different groups (n = 7–8 per group) E. Relative expression levels of DKK1 to GAPDH measured by qPCR in radius bones (n = 5 per group) F. Relative expression levels of SOST to GAPDH measured by qPCR in radius bones (n = 5 per group). DKK1-mAb: DKK1 monoclonal antibody, ALN: alendronate, SHAM: sham, CTX: the C-terminal cross-linking telopeptide of type I collagen, P1NP: the N-terminal propeptide of type 1 procollagen, SOST: sclerostin ∗: P < 0.05, ∗∗: P < 0.01, ∗∗∗: P < 0.001 between the two groups.

Serum Dkk1 level in the Dkk1-mAb group was close to zero, significantly lower than that of the other groups. The sequential treatment group also exhibited lower Dkk1 levels by 42.9 % and 45.0 % than ORX and alendronate groups, respectively (all P < 0.05) (Fig. 2C). Additionally, serum Sost level was higher in Dkk1-mAb group than ORX, alendronate, and sham groups by 27.8 % (P < 0.01), 79.3 % (P < 0.001), and 24.4 % (P < 0.05), respectively. In contrast, the serum Sost level in alendronate group was lower than that in the other groups, with reductions of 28.7 % compared to ORX group (P < 0.01), 44.1 % compared to Dkk1-mAb group (P < 0.001), 40.8 % compared to sequential treatment group (P < 0.001), and 30.6 % compared to the sham group (P < 0.001) (Fig. 2D).

In bone tissue, the expression of Dkk1 showed no significant difference among the various groups (Fig. 2E). The Sost mRNA level in bone tissue of Dkk1-mAb group was numerically higher than that in ORX and Sham groups, although it did not reach statistical significance. The Sost expression in alendronate group showed a decreasing trend, which were significantly lower than that in Dkk1-mAb group (P < 0.05) (Fig. 2F).

3.4. Effects of different treatments on BMD

After 8 weeks of treatment, the Dkk1-mAb, alendronate, and sequential treatment significantly increased BMD (P < 0.05 or P < 0.001 vs. baseline) (Table 2). L1-5 BMD in Dkk1-mAb group was higher by 11.9 % than that in ORX group (P < 0.05). The alendronate group exhibited higher BMD at the left femur (12.8 %, P < 0.001) and L1-5 (22.8 %, P < 0.001) than ORX group (Fig. 3A–D). The sequential treatment group had higher BMD at L1-5 (10.0 %, P < 0.05) than Dkk1-mAb group and exhibited higher BMD at the total body (9.4 %), bilateral femurs (21.0 % and 27.5 %), and L1-5 (23.1 %) than ORX group (all P < 0.001). Additionally, the sequential treatment group had higher BMD in total body than alendronate and sham groups (5.1 % and 6.8 %, all P < 0.05) (Fig. 3A–D).

Table 2.

Comparison of bone mineral density, grip strength at baseline and after treatment.

	Groups	Before treatment (n = 8)	After treatment (n = 7–8)	P value
General BMD (mg/cm2)	ORX	78.8 ± 3.6	82.2 ± 2.3	0.18
	DKK1-mAb	78.6 ± 2.7	85.6 ± 2.1	<0.001
	DKK1-mAb-ALN	78.6 ± 3.2	89.7 ± 2.9	<0.001
	ALN	78.8 ± 3.3	85.4 ± 2.3	<0.001
	SHAM	85.1 ± 4.1	84.0 ± 3.9	0.96
BMD at L1-5 (mg/cm2)	ORX	32.9 ± 3.0	39.0 ± 3.1	<0.01
	DKK1-mAb	35.0 ± 4.7	43.6 ± 2.9	<0.001
	DKK1-mAb-ALN	35.4 ± 4.8	48.0 ± 3.0	<0.001
	ALN	35.9 ± 3.2	47.9 ± 2.4	<0.001
	SHAM	45.5 ± 4.6	47.6 ± 3.2	0.74
BMD at left femur (mg/cm2)	ORX	68.4 ± 6.2	74.1 ± 4.2	0.29
	DKK1-mAb	72.1 ± 5.6	81.8 ± 5.3	<0.05
	DKK1-mAb-ALN	71.7 ± 5.6	89.7 ± 9.1	<0.001
	ALN	69.9 ± 3.0	83.6 ± 4.4	<0.001
	SHAM	88.5 ± 9.6	93.0 ± 6.1	0.56
BMD at right femur (mg/cm2)	ORX	68.0 ± 6.0	72.4 ± 5.2	0.71
	DKK1-mAb	70.0 ± 5.6	81.1 ± 5.9	<0.05
	DKK1-mAb-ALN	68.9 ± 4.8	92.3 ± 13.2	<0.001
	ALN	65.8 ± 4.8	83.3 ± 6.0	<0.001
	SHAM	89.9 ± 10.2	91.5 ± 5.8	0.99
Forelimbs grip strength	ORX	0.82 ± 0.12	0.56 ± 0.06	<0.001
	DKK1-mAb	0.75 ± 0.08	0.74 ± 0.04	0.99
	DKK1-mAb-ALN	0.85 ± 0.21	0.79 ± 0.04	0.84
	ALN	0.73 ± 0.16	0.75 ± 0.08	0.99
	SHAM	0.89 ± 0.14	0.80 ± 0.06	0.42
Four limbs grip strength	ORX	1.78 ± 0.16	1.61 ± 0.06	0.10
	DKK1-mAb	1.61 ± 0.18	1.99 ± 0.15	<0.001
	DKK1-mAb-ALN	1.61 ± 0.17	1.94 ± 0.08	<0.001
	ALN	1.58 ± 0.11	1.94 ± 0.15	<0.001
	SHAM	1.84 ± 0.22	2.15 ± 0.10	<0.001
Time to fall in Rotarod test (s)	ORX	29.8 ± 24.5	56.4 ± 29.1	0.35
	DKK1-mAb	21.9 ± 12.5	46.6 ± 42.6	0.43
	DKK1-mAb-ALN	21.6 ± 26.3	82.0 ± 44.3	<0.01
	ALN	28.4 ± 27.6	61.9 ± 32.3	0.14
	SHAM	31.6 ± 20.2	45.2 ± 27.2	0.90

Bold values indicate that there was a signification difference between the two groups.

BMD: bone mineral density, ORX: orchiectomy, DKK1-mAb: DKK1 monoclonal antibody, ALN: alendronate, SHAM: sham.

Fig. 3.

Effects of different treatments on bone mineral density assessed by DXA A. Total-body BMD in different groups (n = 7–8 per group) B. BMD at left femur in different groups (n = 7–8 per group) C. BMD at right femur in different groups (n = 7–8 per group) D. BMD at L1-5 vertebrae in different groups (n = 7–8 per group). DKK1-mAb: DKK1 monoclonal antibody, ALN: alendronate, SHAM: sham, BMD: bone mineral density ∗: P < 0.05, ∗∗: P < 0.01, ∗∗∗: P < 0.001 between the two groups.

3.5. Effects of different treatments on bone microarchitecture

After 8 weeks of treatment, the Dkk1-mAb group exhibited higher femoral Tb.BMD (2.9 %, P < 0.01), Tb.N (29.6 %, P < 0.01), and lower Tb.Sp (25.8 %, P < 0.001) than ORX group. Both the sequential treatment and alendronate groups also led to higher Tb.BMD and Tb.N, with lower Tb.Sp than ORX group (Fig. 4A and B). The Dkk1-mAb group had higher femoral Ct.Th (7.2 %, P < 0.05), Ct.Ar (17.6 %, P < 0.01), and Ct.Ar/Tt.Ar (10.3 %, P < 0.05) than ORX group. Furthermore, the sequential treatment and alendronate groups showed improved femoral cortical bone microarchitecture parameters compared with those of ORX group. The sequential treatment group had a lower endosteal area ratio than ORX group (8.5 %, P < 0.05). Additionally, the sequential treatment group exhibited higher Ct.BMD, Ct.Th, and Ct.Ar/Tt.Ar than Dkk1-mAb group (all P < 0.05) (Fig. 4C and D).

Fig. 4.

Effects of different treatments on bone microstructure assessed by micro-CT A. Image of the trabecular microstructure in the left femur B. Microstructural properties of the trabecular bone in the left femur (n = 7–8 per group) C. Image of the cortical microstructure in the left femur D. Microstructural properties of the cortical bone in the left femur (n = 7–8 per group) E. Image of the trabecular microstructure in the 5th lumbar vertebra F. Microstructural properties of the trabecular bone in the 5th lumbar vertebra (n = 7–8 per group). DKK1-mAb: DKK1 monoclonal antibody, ALN: alendronate, SHAM: sham, BV/TV: bone volume per total volume, BS/TV: bone surface area per total volume, Tb.BMD: trabecular bone mineral density, Tb.Th: trabecular thickness, Tb.Sp: trabecular separation, Tb.N: trabecular number, Conn.D: trabecular connectivity density, Ct.BMD: cortical bone mineral density, Ct.Th: cortical thickness, Ct.ar: Cortical bone area, Ct.ar/Tt.ar: Cortical bone area to total area ratio ∗: P < 0.05, ∗∗: P < 0.01, ∗∗∗: P < 0.001 between the two group.

In the L5 vertebrae, the sequential treatment group showed higher BV/TV (57.4 %, P < 0.001), BS/TV (39.1 %, P < 0.001), Tb.BMD (4.6 %, P < 0.001), Tb.Th (15.0 %, P < 0.01), Tb.N (16.8 %, P < 0.05), Conn.D (60.2 %, P < 0.05), and lower Tb.Sp (23.3 %, P < 0.01) than ORX group (Fig. 4C and D). The alendronate group also had improved bone microstructural parameters in the L5 vertebra compared with those of ORX group. The effects were greater in the sequential treatment and alendronate groups than Dkk1-mAb group (Fig. 4E and F).

3.6. Effects of different treatments on bone mechanical properties

Three-point bending test of the femur indicated that Dkk1-mAb group had higher stiffness (26.5 %, P < 0.01) and yield load (64.5 %, P < 0.05) than ORX group. The alendronate group exhibited higher stiffness (21.7 %, P < 0.05), yield load (73.3 %, P < 0.01), and breaking load (12.8 %, P < 0.05) than ORX group (Fig. 5A and B). The sequential treatment group had higher maximum load (19.1 %, P < 0.05), stiffness (33.9 %, P < 0.001), yield load (63.3 %, P < 0.05), and breaking load (19.0 %, P < 0.01) than ORX group (Fig. 5A and B). No significant differences were found in mechanical parameters among the three treatment groups (Fig. 5A and B).

Fig. 5.

Effects of different treatments on bone biomechanical properties A. Typical load–displacement curve from the femur three-point bending test B. Mechanical parameters of the right femur by three-point bending test (n = 7–8 per group) C. Mechanical parameters of L4 in the indentation test (n = 7–8 per group). DKK1-mAb: DKK1 monoclonal antibody, ALN: alendronate, SHAM: sham ∗: P < 0.05, ∗∗: P < 0.01, ∗∗∗: P < 0.001 between the two groups.

In vertebral mechanical tests, the Dkk1-mAb, sequential treatment, and alendronate groups showed higher maximum load than ORX group (104.3 %, P < 0.05; 145.1 %, P < 0.001; 152.6 %, P < 0.001; respectively), without significant differences among the three treatment groups (Fig. 5C).

3.7. Effects of different treatments on skeletal histology

Under the microscope, the three treatment groups exhibited significantly thicker and denser trabeculae than ORX group (Fig. 6A). TRACP staining revealed that Dkk1-mAb, alendronate, and sequential treatment groups had significantly fewer osteoclasts than ORX group (Fig. 6B). Additionally, Von Kossa staining revealed that the three treatment groups had significantly higher bone mineralization, as well as increased trabecular area and thickness, than ORX group (Fig. 6C and D).

Fig. 6.

Staining of HE, TRACP, Von Kossa and bone mineral apposition rate of bone A. HE staining of the left tibia, scale bar = 100 um B. TRACP staining of the left tibia, scale bar = 100 um C. Von Kossa staining of the left tibia, scale bar = 1500 um D. Von Kossa staining of the L3 vertebra, scale bar = 1500 um E. Image of cortical bone mineral apposition rate at left tibia of the periosteal surface, scale bar = 60 um F. Image of trabecular bone mineral apposition rate at L3 vertebra, scale bar = 40 um G. Comparison of cortical bone mineral apposition rates in left tibia among different groups (n = 7–8 per group) H. Comparison of trabecular bone mineral apposition rates at L3 vertebrae among different groups (n = 7–8 per group). DKK1-mAb: DKK1 monoclonal antibody, ALN: alendronate, SHAM: sham, HE: hematoxylin and eosin, TRAP staining: Tartrate-Resistant Acid Phosphatase staining: Identifies osteoclast activity in bone tissue, with TRAP-positive cells (osteoclasts) appearing red, indicating regions of active bone resorption, Von Kossa staining: Highlights areas of mineralization in bone tissue by staining calcium deposits black, Ct.MAR: cortical bone mineral apposition rate, Tb.MAR: trabecular bone mineral apposition rate ∗: P < 0.05, ∗∗: P < 0.01, ∗∗∗: P < 0.001 between the two groups.

The Dkk1-mAb group had higher cortical and trabecular MAR than ORX, sequential treatment, and alendronate groups (P < 0.05, P < 0.01, and P < 0.001, respectively) (Fig. 6E–H). The sequential treatment group showed higher Ct.MAR than alendronate group (P < 0.05) (Fig. 6E and G). No significant difference in Tb.MAR was observed among the sequential treatment, alendronate, and ORX groups (Fig. 6F and H).

3.8. Effects of different treatments on muscle

After 8 weeks of treatment, grip strength in all four limbs was increased in the three treatment groups (all P < 0.001 vs. baseline) (Table 2), which was higher than those in ORX group (all P < 0.001) (Fig. 7A and B). Additionally, the sequential treatment group showed improvement in fall time during the rotarod test (P < 0.01 vs. baseline) (Table 2). Histological analysis revealed that Dkk1-mAb, sequential treatment, and alendronate groups all had larger muscle cross-sectional areas than ORX group (28.1 %, P < 0.001; 27.3 %, P < 0.001; 16.9 %, P < 0.05; respectively) (Fig. 7D and E). No significant differences were observed in fat and lean mass content among the treatment groups (Fig. 7F and G). There was a significant increase of MyoD1 protein expression in Dkk1-mAb group than in ORX group and alendronate group (all P < 0.05) (Fig. 7H and J). Myogenin protein expression had a higher trend in Dkk1-mAb group than that in ORX group, although this did not reach statistical significance (Fig. 7I and K).

Fig. 7.

Effects of various treatments on muscle of ORX mice A. Grip strength for forelimbs (n = 7–8 per group) B. Grip strength for four limbs (n = 7–8 per group) C. Rotarod test times (n = 7–8 per group) D. HE staining of the quadriceps femoris, scale bar = 60 um E. Cross sectional area of the quadriceps femoris (n = 7–8 per group) F. The percentage of fat mass (n = 7–8 per group) G. The percentage of lean mass (n = 7–8 per group) H. Image of immunohistochemistry staining for MyoD1 of the quadriceps femoris, scale bar = 100 μm I. Image of immunohistochemistry staining for Myogenin of the quadriceps femoris, scale bar = 100 μm J. Quantitative analysis of MyoD1 protein expression by immunohistochemistry (n = 5 per group) K. Quantitative analysis of Myogenin protein expression by immunohistochemistry (n = 5 per group). HE: hematoxylin and eosin, DKK1-mAb: DKK1 monoclonal antibody, ALN: alendronate, SHAM: sham ∗: P < 0.05, ∗∗: P < 0.01, ∗∗∗: P < 0.001 between the two groups.

3.9. Expression levels of key genes of the wnt pathway

The expression level of Lef1 in bone was higher in Dkk1-mAb and sequential treatment groups than in ORX (P < 0.05 and P < 0.01, respectively) and sham (P < 0.05 and P < 0.01, respectively) groups. Likewise, the expression level of Runx2 in bone was higher in sequential treatment group than alendronate group (P < 0.05) (Fig. 8A–C). Additionally, Axin2 mRNA expression in skeletal muscle of Dkk1-mAb group was significantly higher (P < 0.05) than that of ORX group, and Ctnnb1 expression was somewhat higher in Dkk1-mAb and sequential treatment groups than in ORX group, but the differences were not significant. The expression levels of Lef1 and Runx2 in muscle were similar across all groups (Fig. 8D–G).

Fig. 8.

Downstream gene expression of WNT pathway in radius bones and quadriceps by qPCR A. Relative expression levels of LEF1 to GAPDH in radius bones (n = 5 per group) B. Relative expression levels of CTNNB1 to GAPDH in radius bones (n = 5 per group) C. Relative expression levels of RUNX2 to GAPDH in radius bones (n = 5 per group) D. Relative expression levels of LEF1 to GAPDH in quadriceps (n = 5 per group) E. Relative expression levels of CTNNB1 to GAPDH in quadriceps (n = 5 per group) F. Relative expression levels of RUNX2 to GAPDH in quadriceps (n = 5 per group) G. Relative expression levels of Axin2 to GAPDH in quadriceps (n = 5 per group). DKK1-mAb: DKK1 monoclonal antibody, ALN: alendronate, SHAM: sham ∗: P < 0.05, ∗∗: P < 0.01, ∗∗∗: P < 0.001 between the two groups.

3.10. Safety of the Dkk1-mAb

No histological abnormalities were obvious in visceral organs, including the heart, liver, spleen, lungs, and kidneys, in Dkk1-mAb and sequential treatment groups (Fig. 9A and Fig. S2). No significant pathological changes were detected in heart and aortic arch across all groups, and no increase in vascular atheromatous plaques was found (Fig. 9A–C). Additionally, serum levels of IL-6 and TNF-α were within normal ranges and showed no significant differences among the groups (Fig. 9D and E).

Fig. 9.

Cardiovascular safety of the treatment A. HE staining of the heart, scale bar = 500 um B. Oil Red O staining of the aortic roots, scale bar = 200 um C. Quantification of positive staining per cryo-section of aortic roots stained with Oil Red O (n = 7–8 per group) D. Serum levels of TNF-α by ELISA (n = 4–8 per group) E. Serum levels of IL-6 by ELISA (n = 6–8 per group). HE: hematoxylin and eosin, TNF-α: tumor necrosis factor-α, IL-6: interleukin 6, DKK1-mAb: DKK1 monoclonal antibody, ALN: alendronate, SHAM: sham ∗: P < 0.05, ∗∗: P < 0.01, ∗∗∗: P < 0.001 between the two groups.

4. Discussion

The emergence of osteoporosis as a significant public health concern has driven substantial advances in research of the mechanisms regulating bone turnover and disease pathophysiology. The Wnt pathway plays a crucial role in bone anabolism by regulating osteoblast activity. Dkk1, an endogenous negative regulator of Wnt signaling, has emerged as a key target of osteoporosis treatment. In this study, a novel Dkk1-mAb with high binding affinity, biological activity, and thermal stability was generated. This Dkk1-mAb can effectively reduce serum Dkk1 levels, upregulated key genes of the Wnt pathway, promoted bone formation, and mitigated bone loss. Additionally, the Dkk1-mAb significantly increased BMD at lumbar spine and femur and improved the microstructure and mechanical properties of bone in ORX mice. Moreover, the Dkk1-mAb enhanced muscle mass, cross-sectional area, and function, providing a promising treatment for addressing sarcopenia in osteoporosis. Interestingly, sequential Dkk1-mAb and alendronate treatment showed additive benefits than either monotherapy of Dkk1-mAb or alendronate alone. This provides preliminary evidence for the use of Dkk1-mAb in the treatment of male osteoporosis.

Previous studies have shown that Dkk1 is an extracellular inhibitor of Wnt pathway, playing a crucial role in bone development and homeostasis [6]. Certain rare coding variants in Dkk1 locus are partial loss-of-function mutations, which can lead to a high bone mass phenotype [16]. Dkk1 knockout mice exhibit increased BMD, surpassing that of heterozygous mice, with up to a 3-fold increase in trabecular bone volume in distal femur and up to a 2-fold increase in vertebrae than wild-type controls [17]. These findings indicate that a decrease in Dkk1 expression will significantly increase bone mass, suggesting that antagonizing Dkk1 may have the potential to promote osteogenesis. Recently, Wnt-pathway activators such as the Sost-neutralizing antibody (romosozumab) have been shown to significantly promote bone formation, increase BMD, and reduce the risk of fractures, and have been approved for treating osteoporosis patients with high fracture risk [[18], [19], [20]]. Both Sost and Dkk1 are natural inhibitors of the Wnt pathway, and inhibiting Dkk1 may be a promising therapeutic approach for osteoporosis.

The Wnt pathway is activated when a Wnt ligand binds to the Frizzled and LRP5/6 receptors on osteoblasts. This binding inhibits glycogen synthase kinase 3 (GSK3) and stabilizes β-catenin. The stabilized β-catenin then translocates to the nucleus, where it interacts with T-cell factor and lymphoid enhancer factor (Tcf/Lef) to promote the expression of multiple important genes in osteoblasts [21,22]. Dkk1 acts as a natural antagonist of the Wnt pathway by binding to LRP5/6 receptors, thereby reducing β-catenin levels and inhibiting osteoblast activity and bone formation [23]. In the present study, we developed a novel antibody with high affinity and biological activity for Dkk1, which not only significantly reduced serum Dkk1 levels, but also notably increased the expression of Lef1, thereby promoting bone formation and indirectly inhibiting bone resorption. This novel Dkk1-mAb effectively increased BMD, improved bone microstructure, and enhanced bone strength in ORX mice, supporting its therapeutic potential for male osteoporosis.

However, Dkk1-mAbs have failed to generate significant osteoanabolic effects on BMD at the lumbar spine and femur of ovariectomized rats [24]. Previous studies have detected a compensatory upregulation of Sost in the bone following Dkk1-mAb therapy or bone-specific deletion of Dkk1 [25], which can presumably weaken the therapeutic effect of Dkk1-mAb. However, a Dkk1-neutralizing antibody consistently and robustly improved BMD and bone strength in Sost knockout mice [25]. Additionally, combination therapy with Sost-mAb and Dkk1-mAb resulted in more than double the increase in bone mass than monotherapy [26,27]. The present study showed that the Dkk1-mAb and alendronate have opposite effects on serum Sost level, and Dkk1-mAb significantly increases Sost mRNA expression in bone tissue, whereas alendronate downregulates its expression. We hypothesize that the activation of Wnt/β-catenin signaling following Dkk1 inhibition may trigger an intrinsic compensatory feedback mechanism in osteocytes, of which Sost expression is upregulated to maintain bone metabolic homeostasis, and it may represent an inherent limiting factor for its therapeutic window. Sequential treatment of Dkk1-mAb and alendronate may combine the potent anabolic capacity of the former with anti-resorptive effect of the latter, which achieved better improvement in BMD, bone microstructure, and bone strength than monotherapy. These findings suggested that we need to know the mutual feedback regulation between Dkk1 and Sost in osteoblasts to develop appropriate sequential therapeutic strategies for osteoporosis. In the future, concurrent inhibition of Dkk1 and Sost may overcome this feedback and may yield more powerful anabolic therapies for osteoporosis.

It is worth noting that studies have shown that romosozumab can significantly increase BMD and reduce the risk of fractures [28,29], but it may potentially increase the risk of cardiovascular events, including myocardial infarction, stroke, and cardiovascular death [[30], [31], [32]]. Sost deficiency can activate the Wnt/β-catenin signaling pathway and regulate downstream genes essential for cardiovascular integrity [33]. Therefore, the potential cardiovascular risks should be alert when activating the Wnt signaling pathway for bone anabolic therapy. In this study, we measured the serum levels of inflammatory cytokines and assessed the area of atherosclerotic plaques in the aortic arch. The results showed that Dkk1-mAb did not increase serum levels of IL-6 and TNF-α, indicating that Dkk1-mAb did not trigger excessive systemic inflammation response. Histological examination and Oil-Red-O staining did not reveal significant adverse effects of Dkk1-mAb on the heart, aortic arch, or other internal organs, suggesting that the novel Dkk1-mAb had good safety.

In addition, there is close crosstalk between bone and muscle, and osteoporosis is often accompanied by sarcopenia [[34], [35], [36], [37]]. In this study, we observed characteristics of both osteoporosis and sarcopenia in ORX mice. We found that Dkk1-mAb, alendronate, and their sequential treatments all increased strength and cross-sectional area of muscle in ORX mice. Additionally, the sequential treatment enhanced motor performance in the rotarod test. Previous studies have highlighted the importance of the Wnt signaling in regulating muscle function, as it promotes muscle growth and differentiation [[38], [39], [40]]. Our study also shows that Dkk1-mAb may activate the Wnt/β-catenin pathway in skeletal muscle tissue. The expression of Wnt-related gene Ctnnb1 in muscle shows an increasing trend in Dkk1-mAb and sequential treatment groups, with an increasing expression of its direct transcriptional target, Axin2 in Dkk1-mAb group. Then, activation of Wnt/β-catenin pathway triggered a myogenic response, leading to a significant increase in MyoD expression and an elevating trend of Myogenin expression in Dkk1-mAb group, which may be the mechanism of improvements in muscle mass and strength after treatment of Dkk1-mAb. The critical role of β-catenin (encoded by Ctnnb1) in muscle homeostasis has been demonstrated by human genetics. Pathogenic heterozygous loss-of-function variants in Ctnnb1 were associated with a neurodevelopmental disorder, which is characterized by symptoms of truncal hypotonia, muscle weakness, hypertonia, dystonia, and microcephaly [41]. Our results suggest that Dkk1-mAbs followed by alendronate is beneficial for improving muscle volume and strength in ORX mice, but the exact mechanism is worth exploring further.

This study is the first to demonstrate that a Dkk1-mAb and its sequential alendronate treatment can effectively improve BMD, bone microarchitecture, bone strength, muscle strength, and muscle cross-sectional area in ORX mice, showing promising therapeutic potential for osteoporosis and sarcopenia. However, this study has several limitations. Firstly, we did not conduct experimental studies on the effects of Dkk1-mAb at different doses, but only chose a single dose based on previous studies [12]. In the future, we will further explore the efficacy of different doses of Dkk1-mAb, such as 10, 25, and 50 mg/kg. Secondly, this study did not dynamically monitor the production of Dkk1-mAb anti-drug antibodies, making it difficult to evaluate whether it will weaken the efficacy of Dkk1-mAb. Thirdly, we did not examine the differences in efficacy between the Dkk1-mAb and alendronate with varying sequential time regimens. Lastly, the mechanisms by which the Dkk1-mAb affects muscle mass and function were not thoroughly explored. Further studies are needed to enhance our understanding of the therapeutic potential of Dkk1-mAbs.

5. Conclusion

The novel Dkk1-mAb can increase bone mineral density, bone microarchitecture and bone biomechanical properties, and can enhance strength and cross-sectional area of muscle through activating the WNT pathway. Dkk1-mAb sequential administration with alendronate will achieve additive therapeutic effects. Dkk1-mAb has therapeutic potential for osteoporosis combined with sarcopenia induced by androgen deprivation, which is worth further clinical research.

Declaration of Generative AI in scientific writing

The authors did not use any generative AI tools in the preparation of this manuscript.

Ethical approval

All animal research procedures were approved by the Animal Welfare & Ethics Committee of PUMCH (No. for the application: XHDW-2024-24). All sections of the present study adhere to the “ARRIVE guidelines” for reporting in vivo experiments in animal research.

Authorship

Yanye Wang: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization, Writing – original draft, Writing – review & editing Lei Sun: Data curation, Formal analysis, Validation, Visualization, Writing – review & editing Lingyang Meng: Data curation, Formal analysis, Validation, Visualization, Writing – review & editing Bingna Zhou: Data curation, Formal analysis, Validation, Visualization, Writing – review & editing Dapeng Zhang: Data curation, Formal analysis, Resources, Validation, Visualization, Writing – review & editing Weizhu Qian: Data curation, Formal analysis, Resources, Validation, Visualization, Writing – review & editing Lijia Cui: Writing – review & editing Qian Zhang: Writing – review & editing Ou Wang: Writing – review & editing Yan Jiang: Writing – review & editing Weibo Xia: Writing – review & editing Xiaoping Xing: Writing – review & editing Mei Li: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

Conflict of interest

The authors have no conflicts of interest to disclose in relation to this article.

Abbreviations

Dkk1

Dickkopf-related protein 1

Dkk1-mAb

anti-Dkk1 monoclonal antibody

Sost

sclerostin

CTX

C-terminal cross-linking telopeptide of type I collagen

P1NP

N-terminal propeptide of type 1 procollagen

ELISA

enzyme-linked immunosorbent assay

DXA

dual-energy X-ray absorptiometry

BMD

bone mineral density

ROI

region of interest

BV/TV

bone volume per total volume

BS/TV

bone surface area per total volume

Tb.BMD

trabecular bone mineral density

Tb.Th

trabecular thickness

Tb.Sp

trabecular separation

Tb.N

trabecular number

Conn.D

trabecular connectivity density

Ct.BMD

cortical bone mineral density

Ct.Th

cortical thickness

Ct.ar

cortical bone area

Ct.ar/Tt.ar

Cortical bone area to total area ratio

HE

hematoxylin and eosin

TRAP staining

Tartrate-resistant acid phosphatase staining

Ct.MAR

cortical bone mineral apposition rate

Tb.MAR

trabecular bone mineral apposition rate

TNF-α

tumor necrosis factor-α

IL-6

interleukin 6

ORX

orchiectomy

RT-qPCR

reverse transcription–quantitative polymerase chain reaction

Lef1

lymphoid enhancer-binding factor 1

Ctnnb1

catenin beta 1

Runx2

runt-related transcription factor 2

Gapdh

glyceraldehyde-3-phosphate dehydrogenase

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Data availability statement

All data generated and/or analyzed during this study are not publicly available but are available from the corresponding author on reasonable request.