A novel mouse trauma model for osteonecrosis of the femoral head

Abstract

Osteonecrosis of the femoral head (ONFH) is a devastating condition characterized by the disruption of vascular supply, progressive bone degeneration, and eventual collapse of the femoral head. While corticosteroid- and alcohol-induced ONFH models are well established, a reliable trauma-induced ONFH model in mice has not been previously described. However, traumatic ONFH is commonly seen in clinical settings—such as following FN fracture, dislocation, or hip trauma—but remains poorly understood at the mechanistic level due to a lack of appropriate small-animal models. To address this gap, we established and validated a novel, reproducible trauma-induced ONFH model in C57BL/6 mice. Eight-week-old male C57BL/6 mice (n = 45) underwent surgical disruption in the left hip, while the right hip served as an internal control. Mice were sacrificed 1-12 wk post-surgery at 2-wk intervals. Gross morphological changes, trabecular bone loss, and histological features, such as marrow adiposity, empty osteocyte lacunae, and myeloid cell infiltration, were observed over time, indicating ONFH development. Histological analyses demonstrated decreased glycosaminoglycan content, suggesting cartilage degeneration, and increased empty osteocyte lacunae in the operated femoral heads. Immunofluorescence analyses demonstrated fragmentation of CD31- and endomucsin (EMCN)-positive vessels, vascular invasion into cartilage and growth plate, and dynamic changes in vascular architecture accompanying trabecular bone degeneration. The knowledge gained from this reproducible trauma-induced ONFH model, which mimics key features of human disease, provides a valuable platform for mechanistic studies of vascular dysfunction, cartilage degeneration, and bone remodeling, and may facilitate the development of early diagnostic markers and novel therapeutic strategies for osteonecrosis.

Graphical Abstract

Graphical Abstract.

Osteonecrosis of the femoral head is a debilitating disease caused by impaired blood supply to the bone. We developed a novel trauma-induced mouse model that recapitulates key pathological features of the disease, enabling investigation of its early vascular and skeletal changes. This model provides a valuable tool for understanding disease mechanisms and for developing future therapeutic approaches.

Introduction

Osteonecrosis of the femoral head (ONFH) is a vascular disease characterized by the inadequate delivery of oxygen, nutrients, cytokines, and other factors through the blood to the bone tissue. In the United States, between 10 000 and 20 000 new cases are diagnosed annually, and from 10% to 80% of these patients eventually experience FN fractures.^1–3 Based on a national epidemiological study reporting an annual incidence of 2.51 per 100 000 population, the number of newly diagnosed ONFH cases in Japan is estimated to be approximately 3000 per year.⁴

Osteonecrosis of the femoral head is broadly classified into 2 main categories: traumatic and non-traumatic. While steroid- and alcohol-induced ONFH are among the most frequent causes of the non-traumatic type, sickle cell disease, lupus, genetic predisposition, and human immunodeficiency virus infection have also been implicated.^2,5 Among non-traumatic causes, steroid-induced ONFH is the most common, followed by alcohol-related cases.^1,5 The condition primarily affects young individuals between the ages of 30 and 50, and accounts for approximately 30% of all orthopedic disorders worldwide.^5,6 Although ONFH may initially progress asymptomatically, a reported 80%-94% of asymptomatic cases will develop symptoms within 1-5 yr, frequently affecting both hips.^3,6–8

A primary clinical challenge is that ONFH is rarely diagnosed before the onset of pain, by which time the disease has typically advanced to mid or late stages, resulting in collapse of the femoral head and joint degeneration.^2,9 At that point, surgical intervention becomes unavoidable. While surgical procedures can alleviate pain, they often cannot reverse the damage to the necrotic bone. Therefore, early diagnosis and intervention are essential to halt disease progression.

Endothelial cell injury ending in abnormal angiogenesis and inadequate repair may be the mechanism behind ONFH; all such handicaps can affect the vasoactive substances that disorder the vascular response.^1,5 Also, marrow fibrosis and osteocyte loss occurring in the subchondral bone (SCB) are signs of osteonecrosis.¹ In time, ischemic changes in bone, especially in the femoral head, decreases the trabecular bone number, causing femoral fractures and/or weakening of the bone structure.² Because vascular disruption is considered one of the earliest events in trauma-induced ONFH, we examined CD31- and endomucsin (EMCN)-positive bone endothelial populations, which have been implicated in angiogenesis–osteogenesis coupling and skeletal regeneration.^10,11

The pathophysiology of osteonecrosis is believed to result from a complex interplay of vascular dysfunction, disrupted bone cell physiology, genetic predisposition, and environmental risk factors.^3,5,12 Vascular impairment may stem from underlying coagulation abnormalities seen in conditions, such as sickle cell disease, hereditary thrombophilia, antiphospholipid syndrome, malignancies, or inflammatory bowel disease. In parallel, altered mesenchymal differentiation can compromise bone integrity by impairing osteogenesis and promoting adipogenesis, both of which contribute to reduced bone perfusion and structural deterioration.³ Under normal conditions, new bone formation with sufficient mechanical strength takes approximately 3 mo, whereas osteoclastic bone resorption can compromise trabecular architecture in as little as 3 wk. Together, these alterations drive the progression of ONFH. Following the onset of ischemia, histological evidence of BM necrosis and osteocyte death typically emerges within 24-72 h. This is followed by saponification of free fatty acids within the extracellular matrix and calcium deposition, which together initiate an inflammatory response. Over time, the acellular trabecular bone is replaced by mechanically inferior woven bone, predisposing the femoral head to structural collapse under normal physiological loading.^2,3,13

In the literature, many small animal models of steroid- or alcohol-induced ONFH (SONFH or AONFH) are used to study the mechanisms underlying ONFH and to evaluate therapeutic strategies; however, to our knowledge, no mouse model of trauma-induced ONFH has been reported.⁵ The reported success rate of ONFH induction in steroid- and alcohol-induced models ranges from 40% to 70%, potentially resulting in unnecessary animal use.^6,12,14,15 Additionally, these models rely on different induction methods and pharmacological agents, which may themselves influence the molecular mechanisms underlying ONFH development.⁵

Therefore, to minimize the confounding effects of the different pharmacological agents used to induce ONFH in steroid- and alcohol-induced models, as well as to reduce unnecessary animal use, a trauma-induced osteonecrosis model is needed.^6,12,14–16 Trauma-induced ONFH models have previously been developed in several larger animal species, including rabbits, dogs, and sheep, and have provided valuable insights into disease pathogenesis.^6,12,16,17 However, these models are relatively expensive, technically demanding, and offer limited opportunities for genetic manipulation. To best of our knowledge, no mouse model of trauma-induced ONFH has been reported. Therefore, the aim of this study was to establish a reproducible trauma-induced ONFH model in C57BL/6 mice for investigating the molecular mechanisms underlying disease development. Mice are cost-effective, breed rapidly, have skeletal development that shares many similarities with humans, mature quickly, and are easy to handle, making them well suited for this purpose. This model may facilitate mechanistic studies of trauma-induced ONFH and provide a valuable platform for evaluating future therapeutic strategies.

Materials and methods

Surgery

Seven-week-old C57BL/6 male mice were purchased from Japan SLC. Animals were housed in the animal experimentation facility at the University of Osaka and acclimated to the environment for 1 wk prior to surgery.

Briefly, for the surgery, 8-wk-old male C57BL/6 mice underwent anesthesia. The left hindlimbs were shaved and sterilized with ethanol prior to opening the animal. A small incision was made parallel to the third trochanter, followed by careful dissection until the third trochanter was reached. From the superior aspect of the third trochanter, the femoral capsule was accessed and incised. The femoral head was then dislocated from the acetabulum. A circumferential incision was made around the FN to disrupt the retinacular vessels. The remnants of the ligamentum teres were transected, and the femoral head was kept dislocated for at least 30 s to 1 min, before being repositioned into the acetabulum, to prevent any potential for revascularization. The incision was then closed.

The right limb did not undergo any surgical procedure and was used as the control. Animals were sacrificed on weeks 2, 4, 6, 8, 10, and 12, postoperatively.

All experimental procedures were conducted in accordance with the guidelines of the Osaka University Committee for Animal Experiments and were approved by the Osaka University Institutional Review Board.

Macroscopic observation of the femoral heads

At each time point, femoral heads were harvested, macroscopically observed and photographed. Color changes were evaluated by three blinded observers to discoloration in the operated femoral heads compared to the contralateral controls.

Assessment of bone loss with ex vivo micro-CT

Each femoral head was scanned using ex vivo micro-CT (ScanXmate-S090R; Comscan) to evaluate differences in bone morphometric parameters between the operated and contralateral control femoral heads. Specifically, the bone volume to tissue volume ratio (BV/TV), trabecular number (Tb.N), and trabecular separation (Tb.Sp) were assessed using TRI/3D-BON software (RATOC System Engineering). Scans were performed at 70 kV and 86 mA, with isotropic voxel size of approximately 13.2 μm. Coronal and sagittal images were acquired 1 d after euthanasia at each designated time point. The ROI for the femoral head included the entire epiphyseal trabecular compartment, ensuring consistent alignment and orientation across all samples.

Tissue histology

Bone tissues were fixed in 4% paraformaldehyde for at least 3 d and then dehydrated through a graded ethanol series. Following dehydration, the samples were decalcified in 10% EDTA (pH 7.4) for 15 d and embedded in paraffin. Serial 5-μm sections were prepared and stained with H&E, Masson’s Trichrome (MT), and Safranin O (SO) to evaluate general tissue morphology, bone histopathology, and cartilage integrity. Histological slides were digitized using an Olympus VS200 Research Slide Scanner (Olympus Corporation) at high resolution. Image analyses were performed using QuPath (version 0.4.3) and Fiji (ImageJ, NIH).

For histological quantification, 3 randomly selected ROIs were evaluated from each femoral head on H&E-, MT-, and SO-stained serial sections using QuPath (version 0.4.3). The percentage of empty osteocyte lacunae was determined for each ROI according to the following formula:

A total of 3, 5, 4, 4, 3, and 3 femoral heads per group were analyzed at 2, 4, 6, 8, 10, and 12 wk postoperatively, respectively.

Preparation of frozen bone cryosections

Frozen bone tissue sectioning was performed according to the protocol described by Kusumbe et al.,¹⁸ with slight modifications. Immediately after euthanasia, femurs were harvested and fixed in 4% paraformaldehyde at 4 °C for 4-6 h under constant agitation. Following fixation, bones were washed with PBS at 4 °C for another 4-6 h with continuous agitation. For decalcification, the samples were immersed in 0.5 M EDTA (pH 7.4) at 4 °C on a rotator for a maximum of 48 h. After decalcification, tissues were washed 6 times in PBS for 10 min each to remove residual EDTA. Subsequently, the samples were immersed in cryoprotectant solution and incubated overnight at 4 °C under constant agitation.

Following cryoprotection, tissues were embedded in super-cryo embedding medium (SCEM; Section-Lab), and immediately frozen. After at least 24 h, 10- to 12-μm-thick cryosections were prepared using a Leica CM 3050S cryostat (Leica Biosystems) as previously described.¹⁹

Immunohistochemistry

Immunohistochemistry was performed as previously described elsewhere.^11,20 The following primary antibodies were used: hamster anti-mouse CD31 monoclonal (1100, clone 2H8, Merck Millipore), and rat anti-mouse endomucin monoclonal (1200, clone V.7C7, eBioscience, Thermo Fisher Scientific). Primary antibodies detected using goat anti-hamster Alexa-Fluor-488, and goat anti-rat Alexa-Flour-647 secondary antibodies (Thermo Fisher). Nuclei were counterstained with DAPI, and sections were mounted using VECTASHIELD Antifade Mounting Medium (Vector Laboratories). Images were acquired using an Olympus FV3000 confocal microscope (Olympus Corporation).

Fluorescence image analyses were performed using Fiji (ImageJ, NIH). The percentages of CD31- and endomucin-positive areas were quantified within manually defined ROIs. A total of 4, 5, 2, 3, and 4 femoral heads for CD31 staining and 4, 6, 3, 6, and 5 femoral heads for EMCN staining were analyzed and 2, 4, 6, 8, and 10 wk post-operation, respectively.

Statistical analysis

All statistical analyses were performed using GraphPad Prism 9 (GraphPad Software). Paired t-tests were used to compare the measured variables between the operated and contralateral control femoral heads. A p-value <.05 was considered statistically significant. Statistical significance was denoted as p < .05 (^*), p < .01 (^**), p < .001 (^***), and p < .0001 (^****).

Results

Surgery

To establish a reproducible trauma-induced ONFH model, the blood supply to the femoral head was surgically disrupted by transecting the retinacular vessels and ligamentum teres. Femoral heads were harvested at 2, 4, 6, 8, 10, and 12 wk postoperatively to evaluate the temporal progression of pathological changes using μCT, histological, and immunofluorescence analyses.

Assessment of macroscopic and structural changes

Macroscopic examination of the operated femoral heads revealed their progressive discoloration between 6 and 12 wk postoperatively compared with the contralateral controls. By 8 and 10 wk, a distinct yellow-brown hue became apparent, indicating possible lipid accumulation and subchondral ischemia in the operated femoral heads. At wk, the operated femoral heads exhibited a dark reddish-brown discoloration, suggesting hemorrhagic necrosis and bone resorption (Figure 1A). μCT images (Figure 1B) revealed a progressive decrease in trabecular bone starting from week 4 through week 12 postoperatively. Quantitative analysis demonstrated significant decreases in BV/TV and Tb.N, together with a significant increase in Tb.Sp in the operated femoral heads compared with the contralateral controls (Figure 1C).

Figure 1.

Gross morphology and micro-CT analysis of the femoral head following trauma-induced osteonecrosis of the femoral head (ONFH). (A) Representative images showing morphological differences in the femoral heads of control and operated femoral heads at 6, 8, 10, and 12 wk post-surgery. Black dashed boxes indicate the region of the femoral head. (B) Representative 3D micro-CT (μCT) reconstruction of control and operated femoral heads at 4, 6, 8, 10, and 12 wk after surgery. Progressive trabecular bone loss is observed in the operated femoral heads. (C) Quantitative μCT analysis of bone volume fraction (BV/TV), trabecular number (Tb.N), and trabecular separation (Tb.Sp). Data are presented as mean ± SD with individual biological replicates were significantly decreased in the surgery group, beginning at week 6, while trabecular separation (Tb.Sp) was markedly increased. (Week 2: n = 3; week 4: n = 5; weeks 6, 8, 12: n = 9; week 10: n = 10; ^*p < .05, ^**p < .01, ^***p < .001, ^****p < .0001; ns., not significant).

Histological evaluation of bone and cartilage degeneration

H&E staining showed increased BM adipocyte accumulation from 2 wk postoperatively, followed by reduced cartilage thickness at 4 wk and irregularity of the articular cartilage surface at 6 wk in the operated femoral heads (Figure 2A). Reduced trabecular bone was also observed from 4 wk postoperatively, consistent with the μCT findings (Figures 1B and 2A). Increased adipocyte and myeloid cell accumulation within the BM was observed in the operated femoral heads compared with the contralateral controls (Figure 2B and C).

Figure 2.

H&E staining of the femoral head during the early stage of trauma induced osteonecrosis of the femoral head (ONFH). (A) Representative H&E-stained sections comparing control and operated femoral heads at 2, 4, and 6 wk after surgery. The black arrow indicates thinning of the articular cartilage in the operated femoral heads at 4 wk, whereas the double-headed black arrow indicates the articular cartilage thickness in the corresponding control femoral head. Black dashed boxes indicate the regions shown at higher magnification in panels (B) and (C). Scale bars = 200 μm. (B) Higher magnification image of the boxed region from the operated femoral head at 2 wk after surgery, showing adipocyte accumulation and increased eosinophilic staining within the BM. Scale bar = 50 μm. (C) Higher-magnification image of the boxed region from the operated femoral head at 6 wk after surgery, showing increased cellular infiltration within the BM. Scale bar = 50 μm.

Masson’s trichrome staining demonstrated an increased number of empty osteocyte lacunae in the operated femoral heads at 2, 4, and 6 wk. Quantitative analysis further demonstrated a significantly higher empty osteocyte lacunae ratio in the operated femoral heads than in the contralateral controls at 2, 4, 6, 8, and 10 wk postoperatively, whereas no significant difference was observed at 12 wk (Figure 3A and B).

Figure 3.

Histological evaluation of osteocyte loss and cartilage changes following trauma-induced osteonecrosis of the femoral head (ONFH). (A) Representative Masson’s trichrome (MT)-stained sections of control and operated femoral heads as 2, 4, and 6 wk after surgery. Black boxes indicate the regions shown at higher magnification. Yellow arrows indicate representative empty osteocyte lacunae. The yellow dashed boxes indicate the regions shown at higher magnification in the rightmost column. Red arrows indicate representative empty osteocyte lacunae. Scale bars = 100 μm. (B) Quantification of the ratio of empty osteocyte lacunae to total osteocyte lacunae. Data presented as mean ± SD with individual biological replicates. (Week 2: n = 3; week 4: n = 5; weeks 6 and 8: n = 4; week 10 and 12: n = 3; ^*p < .05, ^**p < .01, ^***p < .001, ^****p < .0001; ns., not significant). (C) Representative Safranin O-stained sections of control and operated femoral heads at 4, 8, and 12 wk after surgery. Reduced Safranin O staining in the operated femoral heads indicates decreased glycosaminoglycan (GAG) content. Scale bars = 50 μm.

Safranin O staining demonstrated reduced staining intensity in the articular cartilage of the operated femoral heads compared with the contralateral controls at 4, 8, and 12 wk postoperatively, suggesting decreased glycosaminoglycan (GAG) content (Figure 3C).

Immunohistochemical analysis of aberrant vascularization and subchondral invasion

Immunofluorescence staining was performed to evaluate changes in vascular architecture within the femoral heads (Figure 4A, merged images). Quantitative analysis demonstrated a significant increase in the EMCN-positive area fraction in the operated femoral heads at 2 and 6 wk postoperatively compared with the contralateral controls (Figure 4B). Although no significant differences were detected in the CD31-positive area fraction, both CD31- and EMCN-positive vessels in the operated femoral heads appeared fragmented and disorganized compared with those in the contralateral controls. Moreover, vascular invasion extending from the growth plate toward the SCB was observed at earlier time points in the operated femoral heads than in the contralateral controls and was accompanied by disruption of the normal vascular architecture (Figure 4A, merged images).

Figure 4.

Immunofluorescence analysis of CD31 and EMCN expression in the femoral head following trauma-induced osteonecrosis of the femoral head (ONFH). (A) Representative immunofluorescence images of control and operated femoral heads at 2, 4, 6, 8, and 10 wk after surgery showing CD31 (red), EMCN (green), merged CD31/EMCN (yellow), and merged CD31/EMCN/DAPI (yellow/cyan). Comparisons were performed between the control and operated groups at the corresponding time point. Representative growth plates are outlined by white lines in the 2- and 4-wk panels. White arrows indicate vascular extension from the growth plate toward the subchondral bone (SCB). The boxed regions are shown as enlarged views in the rightmost panels. Yellow arrows indicate representative areas of fragmented and disorganized vascular morphology. SCB, subchondral bone. Scale bars = 100 μm. (B) Quantification of the CD31- and EMCN-positive area fractions. For CD31 quantification, week 2: n = 4; week 4: n = 5; week 6: n = 2; week 8: n = 3; week 10: n = 4. For EMCN quantification, week 2: n = 4; week 4: n = 6; week 6: n = 3; week 8: n = 6; week 10: n = 5. ^*p < .05.

Discussion

In this study, we successfully established a trauma-induced mouse model of ONFH that, to the best of our knowledge, is the first of its kind reported in the literature. This model recapitulates key pathological features of human ONFH. Specifically, the progressive morphological changes, bone loss, and cartilage degeneration observed in our model are consistent with previous reports from steroid-induced ONFH (SONFH) models,^14,15,21 other trauma-induced ONFH models,²² and the clinical progression of the disease,⁸ making this model a promising platform for mechanistic studies.

Decreased BMD, reflected by a reduced bone volume fraction (BV/TV) is a hallmark of osteonecrosis.²² In our study, μCT images revealed progressive trabecular bone loss beginning at 4 wk postoperatively, whereas significant morphometric changes, including decreased BV/TV and Tb.N and increased Tb.Sp, were detected from 6 wk onward. These findings are consistent with previous reports from experimental ONFH models and clinical studies. However, it should also be noted that the μCT analyses were performed ex vivo using independent cohorts at each time point. Therefore, temporal comparisons represent cross-sectional observations rather than longitudinal changes within the same animals, and apparent differences between consecutive time points should not be interpreted as progressive recovery or regeneration of bone mass.

In our study, H&E staining confirmed early pathological changes in the BM microenvironment. Adipocyte accumulation was observed in the operated femoral heads as early as 2 wk postoperatively, indicating early alterations in the BM microenvironment. Additionally, increased myeloid cell infiltration was observed within BM, suggesting the establishment of an inflammatory microenvironment that may contribute to ONFH progression. These observations are in agreement with earlier studies describing inflammation-associated mechanisms in ONFH.^15,22

Another defining feature of osteonecrosis is the presence of empty osteocyte lacunae, indicative of osteocyte death. In the present study, quantitative analysis demonstrated a significant increase in the empty osteocyte lacunae ratio in the operated femoral heads during the early stages after surgery, consistent with progressive osteocyte death following vascular disruption. The lack of a significant difference in the empty osteocyte lacunae ratio at 12 wk may be partly attributable to age-related skeletal changes. As the mice reached approximately 16 wk of age at this time point, age-associated alterations in osteocyte viability may have increased the baseline frequency of empty lacunae in the control group, thereby reducing the difference between the operated and control femoral heads.

The intensity of SO staining reflects GAG content, a critical component of the cartilage extracellular matrix. Decreased GAG content and structural deterioration has been associated with the progression of osteoarthritis (OA) in both humans and rodents.^12,14 Moreover, secondary OA is a known complication of ONFH.¹³ In our study, SO staining demonstrated reduced GAG content in the operated femoral heads at 4, 8, and 12 wk postoperatively. This sustained reduction in GAG, together with cartilage thinning and increased empty osteocyte lacunae, supports the presence of a chronic degenerative process. These findings further suggest that both cartilage degeneration and SCB deterioration are evident from 4 wk postoperatively, indicating that cartilage degeneration may represent an early pathological feature of trauma-induced ONFH.

A critical and novel finding of our study was the marked and rapid increase in vessel density as early as 2 wk postoperatively, accompanied by fragmentation of CD31-EMCN-expressing vessels in the necrotic femoral heads. These vessels may represent H-type or R-type vessels and play pivotal roles in osteogenesis–angiogenesis coupling and adult bone regeneration, respectively, and are essential for maintaining skeletal homeostasis.^5,11,23 One possible explanation for the increased vessel density at 2 wk postoperatively may be that disruption of the blood supply to the femoral head may induce local hypoxia, which could stimulate VEGF production by bone and cartilage cells, thereby promoting early angiogenic response and early invasion through growth plate to SCB. However, despite the increased density, the fragmented morphology of these vessels suggests that the newly formed vasculature fails to establish a stable vascular network. Therefore, this may be directly related to the impaired bone regeneration observed in ONFH. In addition, abnormal vascular structures, appearing barbed and fragmented, were observed invading the cartilage and growth plate regions—indicative of pathological neovascularization.

This study has several limitations. First, although our trauma-induced ONFH model successfully replicates key pathological features of human ONFH, it does not fully reflect the complexity and multifactorial nature of the disease seen in clinical settings. Second, the model was validated primarily through histological and morphometric analyses; molecular characterization of key signaling pathways involved in angiogenesis, hypoxia, and cartilage degradation (eg, vascular endothelial growth factor, hypoxia-inducible factor-1α, and matrix metalloproteinases) was not performed and remains to be addressed in future studies. Third, functional assessments, such as locomotor behavior or pain sensitivity, were not included in the current evaluation. Lastly, all analyses were conducted in a single mouse strain and sex, which may limit the generalizability of the findings. Another limitation of this study is that the contralateral femoral head was used as the internal control instead of including an independent sham-operated group. Although this paired design minimizes inter-animal variability and reduces the number of animals required, potential effects of altered mechanical loading or systemic responses on the contralateral femoral head cannot be completely excluded.^24,25 Future studies including a sham-operated control group will help to further validate the findings. Finally, compared with some previously reported ONFH models, our model exhibited relatively mild pathological changes, without pronounced femoral head collapse or severe cartilage degeneration during the observation period. This may reflect differences in disease etiology, as the present model reproduces trauma-induced ischemia rather than steroid- or alcohol-induced ONFH, as well as the intrinsic regenerative capacity of mice. While this may limit its suitability for studying end-stage structural collapse, it provides a valuable platform for investigating the early pathological events and vascular changes that precede irreversible joint destruction. In addition, the relatively small sample size may have limited the statistical power to detect subtle differences in CD31- and EMCN-positive vessel area fractions at certain time points. Therefore, these findings should be interpreted with caution and confirmed in future studies using larger cohorts of animals.

Future studies using this model will aim to identify key molecular regulators involved in vascular degeneration, adipose tissue remodeling, chondrocyte viability, and bone remodeling in ONFH. Integration of transcriptomic or proteomic approaches may further elucidate the early signaling events following trauma-induced vascular disruption. In addition, pharmacological interventions targeting angiogenic or anti-inflammatory pathways could be tested in this model to assess their therapeutic potential. Expanding this model to include different mouse strains, sexes, and long-term follow-up may also help to uncover disease modifiers and improve clinical relevance.

In conclusion, to the best of our knowledge, this model is the first mouse model to demonstrate such progressive and multifactorial pathological features of trauma-induced ONFH. Its reproducibility, accessibility, and pathological fidelity make it a valuable tool for investigating the molecular mechanisms of ONFH and for the development of potential therapeutic strategies.

Author contributions

Zeynep Bal (Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Visualization, Writing—original draft, Writing—review & editing), Yukika Tanabe (Writing—review & editing), Fitriana Nur Rahmawati (Writing—review & editing), Man Wang (Writing—review & editing), Faya Nuralda Sitompul (Writing—review & editing), Ru Su (Writing—review & editing), and Nobuyuki Takakura (Conceptualization, Funding acquisition, Methodology, Project administration, Resources, Supervision, Writing—review & editing)

Funding

This work was supported by the Japan Agency for Medical Research and Development (AMED) under Grant number (24ama221533h0001, 24ck0106727h0003, 25ama221533h0002) and the Japan Society for the Promotion of Science (JSPS) Grants-in-Aid for Scientific Research(S) (20H05698, 25H00444), Grants-in-Aid for Early Career Scientists (23K14502), Ministry of Education, Culture, Sports, Science and Technology (MEXT) (JPMXP1323015484), and OU Master Plan Implementation Project promoted under The University of Osaka.

Conflicts of interest

Z.B., Y.T., F.N.R., M.W., F.N.S., R.S., N.T.: None declared.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Ethics approval statement

Ethics approval was obtained from the Animal Resource Center for Infectious Diseases, RIMD and IFReC, The University of Osaka under the project titled “Elucidation of the angiogenesis process through in vivo imaging analysis” on June 13, 2023, with approval number R05-02-0.

Patient consent statement

Not applicable.