Tension-Free Weight-Bearing Model of Steroid-Induced Osteonecrosis of Femoral Head in Rats

The protocol shows building a simple and cost-effective rat weight-bearing training model for steroid-induced osteonecrosis of the femoral head using elastic therapeutic tape.

Unlike humans, rats are animals that walk on all fours, while humans are bipedal animals that stand, and the hips are subjected to tremendous pressure when walking and standing. In rat steroid-induced femoral head necrosis models, the biomechanical characteristics of the human hip under higher pressure often need to be simulated. Some scholars try to emulate the state of human hip pressure by making rats bear a certain weight, but fixing the weight-bearing object on the rat is tough. Rats can easily break free of immobilization, and sticking the weight on the rats with adhesive tape will cause the rats to suffocate or die from intestinal obstruction. Our research group used elastic therapeutic tape to perform tension-free immobilization of weight-bearing objects in rats so that the rats could breathe freely and not break away from the immobilization under weight-bearing conditions. Compared to the usual steroid-induced femoral head necrosis rat model, we found that this weight-bearing intervention can aggravate the progression of femoral head necrosis in rats.

The administration of glucocorticoids is the most common risk factor for non-traumatic osteonecrosis of the femoral head (ONFH)¹. Numerous pieces of evidence suggest that, in addition to glucocorticoids, the pressure load on the hip joint in patients is also associated with the occurrence of ONFH. Factors such as body weight and physical labor intensity are recognized as risk factors for ONFH². Multiple clinical studies have demonstrated a close relationship between hip joint loading conditions and the timing and incidence of joint replacement^3,4,5,6. Therefore, establishing a model that reflects the relationship between load bearing and steroid-induced osteonecrosis of the femoral head (SONFH) is important for a comprehensive investigation of this condition.

Large bipedal birds, such as ostriches and emus, serve as good models to simulate hip joint stress, resembling human leg loads^7,8. However, maintaining large bird species is challenging, and the associated research costs are high. Spontaneously hypertensive rat models^9,10 can exhibit higher rates of ONFH, but the compartment pressure load in the marrow generated by spontaneous hypertension differs significantly from mechanical pressure and is unsuitable for studying the impact of mechanical pressure on SONFH.

Small animal models are commonly used in research on SONFH. However, quadrupedal reptiles have lower hip joint stress, and their hip models cannot simulate the biomechanical environment of human hip joints during bipedal walking. Single limb immobilization models¹¹ and partial unloading models¹² are common, but both reduce limb loading. As humans are bipedal organisms with substantial lower limb loads while standing and walking, reducing the load in these models diminishes the connection between animal models and human diseases.

This study aims to establish a simple and cost-effective model for weight-bearing training to investigate the impact of weight-bearing training on steroid-induced osteonecrosis of the femoral head in rats. Currently, rat models have been used to study steroid-induced osteonecrosis of the femoral head^13,14, but there is still no model that can provide long-term, safe fixation with minimal disruption to movement, which is also relatively simple and inexpensive. In this study, a high-adhesive fixation material is applied, adopting tension-free fixation, which maintains the rats' mobility and reduces the suffering and even death caused by improper fixation.

The protocol adheres to the ethical guidelines set by the Institutional Animal Care and Use Committee (IACUC) at Beijing University of Chinese Medicine, Protocol number BUCM-4-2022062001-2109. The protocol uses Sprague Dawley (SD) rats (SCXK(Jing)2019-0008), aged 8-10 weeks and weighing 200 g-250 g.

1. Adaptation training

1. Albino rats generally have a visual range of less than 60 cm¹⁵. During the modeling process, place the rats in opaque cages and keep them as far away as possible from non-modeled rats (at least, other rats should be outside the maximum visual range of the modeled rats) to prevent them from experiencing traumatic stress.
2. Turn on the power and set the treadmill to 10 m/min with a 0° incline. Place the rats on the treadmill and ensure the movements are as gentle as possible to avoid stress. Do not use additional stimulation; the rats will naturally crawl forward because they are unfamiliar with the environment.
3. The training lasts for 15 min. After the training, clean up the feces and urine left by the rats and spray alcohol to eliminate the animal odor. Then, prepare the next rat for training.
   NOTE: If a rat refuses to walk on the treadmill for 5 minutes, exclude it from the study.
4. Continue adaptive training until the end of the 1^st week.

2. Measuring the maximum weight-bearing capacity in rats

1. Prepare two test tubes, a hook, and a loop fastener approximately 80 cm long. Place steel balls into the test tubes, ensuring that the initial combined weight of the test tubes, the hook, and the loop fastener is 150 g. Prepare test tubes with different total weights in 10 g increments, with the maximum weight being 350 g.
2. A researcher uses both hands to fasten rats using gloves (Figure 1). Another researcher places the test tube on the back of the rat about 1 cm from the midline of the back. Since this step does not require movement of the rat, perform fixation by simply wrapping the hook and loop fastener around the rat without using additional measures to secure the tube.
3. Change the tube until the rat's maximum weight is reached. When the rat reaches maximum weight-bearing, it cannot stand when prodded or the experimenter leaves. Reduce the weight by 10 g to allow the rat to stand. This weight represents the rat's maximum weight-bearing capacity.

3. Preparation of weight load

1. Since secure fixation is required during training, secure the objects using 1-1.5 m elastic therapeutic tape based on rat size, which has strong adhesion. Use clay to adjust the weight of the test tubes to reduce the swinging of the load.
2. After weighing the weight of the test tubes and the fixation objects, add an appropriate amount of clay into the test tubes to adjust them to the target weight. Use a glass stirring rod to evenly spread the clay inside the test tubes. Clay concentrated at one end of the test tube may cause it to detach easily during training.
3. Adjust the total weight of the adjusted test tubes and clay at 50% of the rat's maximum weight-bearing capacity (Figure 1A). Exceeding this weight will make it difficult for the rat to complete the training. Do not adjust the weight of the load again until the end of the experiment.

4. Establishing steroid-induced osteonecrosis of the femoral head model

1. Use the methylprednisolone combined with lipopolysaccharide (LPS+MPS) method for SONFH model establishment^16,17. Perform intraperitoneal injection of LPS (20 µg/kg) using a standard needle at the midline of the rat's abdomen once every 24 h, for a total of 2 injections.
2. After the last LPS injection, inject MPS (40 mg/kg) into the unilateral gluteal muscle 24 h after regular feeding and continue injections every 24 h, alternating between the two gluteal areas, for a total of three injections.

5. Tension-free weight-bearing immobilization using elastic therapeutic tape and treadmill training

1. Mark the rat's back midline at 1-2 cm on both sides; this is the reference point for the weight load. Ensure the marks are appropriately positioned to prevent the rat from tipping or hindering movement during treadmill training.
2. With one investigator fixing the rat using gloves, place one hand under the rat's armpit to secure the forelimb and jaw while the other hand secures the hind limbs and tail. Avoid excessive force to prevent stress or suffocation (Figure 1B).
3. Adhere the elastic therapeutic tape to the test tube without applying tension to secure the load on the rat's back. In the process of winding, do not stretch the elastic therapeutic tape to achieve tension-free fixation.
   NOTE: The elastic therapeutic tape used in this model is designed for taping during intense physical activities, providing high adhesion and elasticity¹⁸. Due to its high elasticity, it greatly reduces the impact on the rat's movement while crawling.
4. Ask the second investigator to fix the test tube to the rat's back parallel to the rat's spine, with the fixation position based on the markings made, i.e., 1-2 cm away from the midline on both sides of the rat's back.
5. To minimize the suffering of the rat, reduce the impact on the rat's movement, and decrease mortality caused by fixation, perform fixation without applying tension. Stick the test tube with adjusted weight onto the elastic therapeutic tape, then wrap the test tube around the rat's body using a tension-free technique.
6. Ensure tension-free immobilization and maintain the original length of the elastic therapeutic tape without stretching in order to prevent any adverse effects on rat breathing and movement (Figure 1C).
7. After wrapping the test tube around the rat's body, place the rat in an open space or a single cage, monitoring its breathing and direction. If signs of rapid breathing or mouth opening appear, promptly release the rat to prevent suffocation.
   1. Ideal immobilization allows the rat to move freely and perform activities like drinking, lifting forelimbs, and feeding (Figure 1D). If immobilization is unsatisfactory, release temporarily and re-immobilize after 20 min to minimize stress on the rat due to repeated immobilization stimuli.
8. Turn on the power and set the treadmill to 1 m/min with a 0° incline. The treadmill training time is 30 min. Gently transfer the rat onto the treadmill and begin the timer.
9. Do not provide additional stimulation to the rat. If the rat cannot complete the 30 min training, place it in a cage for a 10 min rest without removing the fixation.
   NOTE: The elastic therapeutic tape used in this study has excellent stretchability and adhesiveness. When properly secured without tension, long-term fixation will not cause the test tube to detach or result in suffocation and death of the rat.
10. After the training, clean up the feces and urine left by the rats and spray alcohol to eliminate the animal odor.
11. Avoid the above steps being seen by other rats and handle the animals as gently as possible to prevent animals from vocalizing, thus causing traumatic stress to other rats.
12. Immediately release the rat from fixation after completing training (Figure 1E). During release, use fingers to press down on the rat's fur to protect it, minimizing fur loss during release (Figure 1F).

6. Animal grouping

1. To study the impact of weight-bearing training on SONFH, randomly divide 60 rats into four groups.
2. In the Control group, do not establish the common steroid-induced osteonecrosis of the femoral head (SONFH) model or perform weight-bearing training. In the Control+Load group, carry out weight-bearing training only, without establishing the steroid-induced osteonecrosis of the femoral head model. In the Model group, establish the steroid-induced osteonecrosis of the femoral head model only. In the Model+Load group, perform weight-bearing training after establishing the steroid-induced osteonecrosis of the femoral head model.

7. Euthanasia and specimen collection

1. Place the rats into a small animal euthanasia chamber and inject enough carbon dioxide to achieve a concentration higher than 5%. Once the rats lose consciousness, administer a dose of 0.3 mL/kg of body weight of a 20% potassium chloride solution, totaling 20 mL, via an intracardiac injection.
2. Monitor the rats' breathing and heartbeat to ensure they experience no pain or discomfort during euthanasia. Confirm death by checking for the absence of breathing and heartbeat. Spray alcohol to clean away any odor.
3. Post-euthanasia, dissect the rat's hip region and extract the femoral head while keeping the femur intact. Perform a microCT scan following pretreatment, followed by Hematoxylin-Eosin (HE) staining.

8. Hematoxylin-Eosin staining

1. After the modeling is completed, euthanize the rats. Remove the femurs for HE staining.
2. Soak the rat femurs in 4% paraformaldehyde for 24 h to preserve biological tissues as specimens. Soak the rat femur specimens in 5% formic acid for decalcification for 5 days.
3. Section the specimens into 5 µm slices for HE staining.
   NOTE: Empty lacunae refer to the situation in bone tissue¹⁹ where the lacunae, which should normally contain osteocytes, are devoid of these cells. It is an important indicator for assessing the health of bone tissue. In the pathological process of bone necrosis, insufficient blood supply may cause osteocyte death, leading to empty lacunae. In the diagnosis and evaluation of bone necrosis, the number and distribution of empty lacunae are important parameters for measuring the severity of the disease^20,21.
   1. Place rat femoral specimens into melted paraffin (56-60 °C), first soaking in low-melting-point paraffin, if necessary, then gradually transfer to higher melting point paraffin, each time for about 1 h, to ensure thorough infiltration of the tissue with paraffin.
   2. Prepare embedding molds and pour melted paraffin into the molds to the appropriate depth. Retrieve the tissue samples from the paraffin using forceps after soaking, remove excess paraffin, and place them in molds. Cool to room temperature to solidify the paraffin, forming paraffin blocks.
   3. Using a paraffin microtome, cut the embedded paraffin blocks into 5 µm thick sections.
4. Bake the rat femoral specimens with attached slices in a slide warmer for 1 h to enhance the adhesion between the slices and coverslip and reduce floating during subsequent staining processes.
5. Immerse the baked slices sequentially in pure xylene to remove paraffin, followed by a series of dehydration through absolute ethanol, gradient ethanol (100%, 95%, 80%, 70%), and water, for 2 min each, completing the deparaffinization and rehydration process.
6. Immerse the deparaffinized slices in a hematoxylin staining solution and stain for 10 min. Rinse with running water to remove unbound hematoxylin.
7. Immerse the hematoxylin-stained sections in eosin staining solution and stain for 2 min to color the cytoplasm and connective tissue. Rinsed with a 1% acetic acid solution to fix the staining effect.
8. Immerse the rat femoral bone specimen slices sequentially in 70%, 80%, 90%, 95%, and 100% ethanol, with each gradient for 2 min, gradually removing moisture.
9. Immerse the rat femoral bone specimen slices in pure xylene for transparency, followed by rinsing with running water to remove the xylene. Apply neutral mounting resin, covering the slides, gently tapping to remove air bubbles, and allowing the mounting medium to solidify to complete the mounting process.
10. Using the randomizr package in R, select 30 HE-stainedslides for each group and allocate 15 slides each to two researchers. Without informing them of the groupings, ask the researchers to calculate the empty lacunae rate of the slidesand collect the results for statistical analysis.

9. MicroCT analysis

1. After the modeling is completed, euthanize the rats. Remove the femurs for Hematoxylin-Eosin staining.
2. Soak the rat femurs in 4% paraformaldehyde for 24 h to preserve biological tissues as specimens. Place the rat femurs in the small animal-specific microCT device.
3. Set the microCT scanning conditions as follows: X-ray voltage: 80 kV, Current: 100 µA, Single exposure time: 50 ms, scanning resolution: 25 µm. Perform scans at intervals of 0.5°, focusing on the trabecular bone from beneath the cartilage to the upper part of the epiphysis as the region of interest.
4. Define the area above the rat femoral neck as the region of interest.
5. Perform coronal plane reconstruction of the scanning results using microCT device and collect bone morphometric measurements generated by the built-in software of the microCT device. Record the following observed indices: Total Volume (TV), Percent Bone Volume (BV/TV), Bone Surface/Volume Ratio (BS/BV), Structure Thickness (Tb.Th), Structure Separation (Tb.Sp), Number of Bone (Tb.N), Bone Density (BD).

10. Statistical analysis

1. Present the data as the mean ± standard deviation (SD). Conduct statistical analyses for micro-CT and quantitative histologic evaluations using the independent sample t-test. Consider a p-value < 0.05 as statistically significant.

Histopathology analysis
Hematoxylin and eosin staining revealed that in the Control group and Control+Load group, bone trabeculae were intact and arranged regularly. Blood vessel endothelial cells were present in bone dimples, and the cell morphology appeared plump. In contrast, the Model group and Model+Load group exhibited fractured and disordered bone trabeculae, along with a significantly higher number of empty lacunae. The Model+Load group had more empty lacunae compared to the Model group. In the Model group, some cells within the bone marrow exhibited lipid accumulation, while the bone marrow cavity in the Model+Load group appeared comparatively vacant (Figure 2A). The rates of empty lacunae in the Control group, Control+Load group, Model group, and Model+Load group were 6.0 ± 2.5, 6.4 ± 3.8, 57.6 ± 29.6, and 78.2 ± 15.5, respectively (Figure 2B).

MicroCT analysis
The microCT results clearly display the microscopic structure of bone trabeculae, reflecting the integrity of tissue structure and changes in bone mass. In this study, the trabeculae in the Control group and Control+Load group were dense, well-arranged, and clearly visible. In the Model group, trabeculae were sparse and exhibited disordered and irregular arrangements. Above the epiphyseal line in the Model+Load group, trabeculae morphology was incomplete. The femoral heads in all four groups appeared smooth without significant collapse (Figure 3A).

Quantification of skeletal structure, morphology, and dimensions was done. TV represents the total volume of the region of interest, which can indirectly reflect whether there has been a significant change in bone morphology¹⁹. TV showed no significant differences among the four groups, indicating that both the construction of the glucocorticoid-induced damage model and the intervention of mechanical pressure load may not lead to collapse in the femoral heads of rats. BV/TV can reflect bone mass²². The Control+Load group had significantly higher bone volume than the Control group, and both the Control group and Control+Load group had much higher bone volume than the Model group and Model+Load group. Compared to the Model group, the BV/TV in the Model+Load group was the lowest. Concurrently, BD results corroborated with BV/TV²³, indicating that mechanical pressure load intervention may promote bone formation in the femoral heads without hormonal intervention while inhibiting bone formation in the femoral heads subjected to glucocorticoid intervention.

BS/BV can indirectly reflect the disorder of trabeculae²⁴. The disorder of trabeculae increased sequentially in the Control group, Control+Load group, Model group, and Model+Load group, suggesting that mechanical pressure load intervention may exacerbate trabecular disorder in normal bone tissue and bone tissue damaged by glucocorticoids. Tb.Th represents the trabecular thickness¹⁹, and Tb.N represents the number of trabeculae²⁰. Both of these indicators are positively correlated with bone formation. In this study, Tb.N and Tb.Th in the Control group were significantly lower than in the Control+Load group. Tb.N and Tb.Th in the Model group and Model+Load group were significantly lower than in the Control group and Control+Load group, with a significant reduction in the Model+Load group compared to the Model group. This indicates that mechanical pressure load may increase the number and thickness of trabeculae, while mechanical pressure load may have the opposite effect on bones after glucocorticoid intervention.

Tb.Sp represents the gap between trabeculae and is negatively correlated with bone mass²². In this study, Tb.Sp exhibited an opposite trend to Tb.N and Tb.Th, corroborating the results of Tb.N and Tb.Th (Figure 3B). The microCT results suggest that mechanical pressure load may have opposing effects on the femoral heads with or without glucocorticoid intervention. For femoral heads without glucocorticoid intervention, mechanical pressure load may promote bone formation. However, when bones are subjected to glucocorticoid damage, mechanical pressure load intervention may exacerbate steroid-induced osteonecrosis of the femoral head.

Figure 1: Tension-free weight-bearing immobilization. (A) Measure the weight of the load. (B) Researchers secure the rat. (C) Fixation without tension. (D) Fixation completed. (E) Two-person collaboration to release fixation. (F) Protect the rat's fur during fixation contact. Please click here to view a larger version of this figure.

Figure 2: Hematoxylin and eosin staining. (A) Hematoxylin and eosin staining. (B) Percentage of empty lacunae (n =15). Statistical analyses were done using the independent sample t-test. Error bars show mean ± standard deviation. Please click here to view a larger version of this figure.

Figure 3: MicroCT analysis. (A) Two-dimensional reconstructed image from microCT. (B) Bone morphometric results (n = 15). Statistical analyses were done using the independent sample t-test. Error bars show mean ± standard deviation. Please click here to view a larger version of this figure.

Currently, various animals, such as rabbits²⁵, rats²⁶, mice²⁷, pigs²⁸, broilers²⁹, ostriches⁸, and emus³⁰ can be used to establish models of femoral head necrosis. Among them, rats, mice, and rabbits are the most commonly used species. The rat, as a model for femoral head necrosis, offers numerous advantages. Rats are easy to feed and breed, grow rapidly, and share physiological and metabolic characteristics similar to humans. Their femoral heads are of moderate size, making them suitable for radiographic and pathological studies²⁶. However, as smaller quadrupedal animals, rats bear lower loads on their hind limbs during locomotion, making it challenging to investigate the relationship between mechanical pressure loading and the development of femoral head necrosis using conventional rat models.

The rat has a spindle-shaped body with smooth fur, making it challenging to secure weights using conventional fixation methods, which can significantly impact rat mobility. In the early stages of our research, we attempted to use methods such as adhesive tape, nylon straps, and traction ropes to immobilize the rats, but none yielded satisfactory results. Through continuous experimentation, we successfully employed elastic therapeutic tape and tension-free techniques to allow rats to carry a certain load during weight-bearing walking training with minimal restriction. This model, which is simple and cost-effective, proved to be successful. It meets experimental requirements, enabling exploration of the effects of mechanical pressure loading on steroid-induced femoral head necrosis.

A few troubleshooting considerations for the rat model are discussed here. Skin lesions: Repeated application of elastic therapeutic tape may result in fur abscission and skin lesions on the rat's body. Do not perform any special treatment if the fur has only abscission and does not ulcerate. Elastic therapeutic tape can be applied to bare skin for 4 days without adverse effects. In case of skin breakdown, disinfect the rats with iodophor and continue to use the patch after the lesion has healed. Elastic therapeutic tape is safe, and allergies have not been observed in our research. Tension-free taping does not pull the skin and does not cause blisters; the elastic therapeutic tape can be applied directly to the bare skin. Death by asphyxiation: If the fixation is too tight, the rat may die from suffocation. After completion, observe the rat's breathing. For example, if the rat breathes with its mouth open or struggles vigorously, it may indicate poor fixation. Loosen the tape promptly; otherwise, the rat may die within 5 to 10 min. Gastric Content Reflux: If dark gastric contents are observed at the rat's mouth post-fixation, it indicates a serious problem with the fixation. This situation should be promptly corrected, and the fixation technique should be reviewed. At this point, the rat is in severe pain and should be immediately euthanized.

So far, the predominant models concerning mechanical pressure have been those focused on unloading¹², primarily used to simulate disuse osteoporosis or osteoporosis under low gravity conditions. Our model differs from unloading models in terms of intervention; we applied a higher load to rats, which is a less common intervention method. However, through this intervention, we provide a research model for studying the early stages of femoral head necrosis and how weight-bearing training should be conducted in the early stages of steroid-induced osteonecrosis of the femoral head. This study found that femoral heads unaffected by corticosteroids do not lose bone mass due to weight-bearing training. In steroid-influenced femoral heads, large weight-bearing training may promote the development of SONFH.

Spontaneously hypertensive rats also exhibit high-pressure conditions in the femoral head, and similar to mechanical pressure intervention, hypertensive rats are more prone to femoral head necrosis^31,32. Although both models draw similar conclusions, besides presenting symptoms of femoral head necrosis, spontaneously hypertensive rats also experience endothelial cell damage and marrow cavity adipogenesis⁹. However, there are significant differences in the mechanisms of femoral head necrosis caused by vascular and hip joint mechanical pressure. Spontaneous hypertension models have a longer modeling time and a lower success rate.

Furthermore, this study employs a tension-free fixation technique. Besides minimizing the impact on animal movement, this technique reduces traumatic stress and, when operated correctly, does not cause suffocation or death in rats. These two steps are crucial in this study.

This technique is not only applicable to the study of femoral head necrosis but can also serve as a reference model for research related to sports medicine, weight-bearing training, and endurance training. Existing weight-bearing fixation models cannot secure weight-bearing objects for extended periods. Therefore, this model presents a viable option for studies requiring prolonged weight-bearing in rats.

This study has some limitations. Due to cost constraints and other reasons, we did not measure changes in the femoral head at different stages, and the number of baseline measurements was relatively small. The steroid dosage in this study was also referenced from other models, and the impact of extremely high doses of steroids on this model was not discussed.

The author declares no conflicts of interest, affiliations, or collaborations that could potentially influence the objectivity or outcomes of this research.

This research is an independent study and did not receive any funding.