Heterotopic ossification of tendons and ligaments (HOTL) is a chronic and disabling disease that causes significant damage to patients’ health and economic burden to society. Currently, there are several known causes of HOTL, including genetic factors, tissue injury, abnormal mechanical stress, age, and metabolic diseases (Biz et al., 2015; Zhang et al., 2020). As a controllable factor, mechanical stress is a topic of interest in research. In vitro experimental results demonstrated that primary cells derived from patients with heterotopic ossification exhibit significantly elevated expression of osteogenic marker genes under dynamic culture conditions compared to static cultures, suggesting that tensile stimulation may enhance the osteogenic differentiation of ligaments (Katsumi et al., 2016; Kang et al., 2019; Huber et al., 2020). Moreover, clinical reports suggest that cervical laminoplasty combined with instrumented fusion is more effective in suppressing the progression of the ossification of the posterior longitudinal ligament (OPLL) than stand-alone laminoplasty (Katsumi et al., 2016; Lee et al., 2017; Kang et al., 2019). In vivo experiments in animal models have shown that repeated excessive traction stimulation of the rat tail or thoracic vertebra can induce heterotopic ossification of local ligaments (Tsukamoto et al., 2006; Zhao et al., 2021). Given that internal spinal fixation can effectively delay the progression of OPLL, it is essential to construct a model that limits animal physical activity to study specific physiological mechanisms.

Additionally, there is an ongoing debate regarding the role of physical activity in the development of heterotopic ossification. Some scholars assert that passive physical activity can effectively delay traumatic heterotopic ossification and increase joint mobility (Vasileiadis et al., 2021; Vasileiadis et al., 2023). Conversely, other researchers suggest that movement of the spine or Achilles tendon may actually enlarge the ossification range and severely impact the patient’s quality of life (Kang et al., 2019). It is important to note that mechanical stimulation may have additional effects on secondary chronic spinal cord injury (SCI) caused by spinal stenosis beyond affecting the size of the ossification range in OPLL. However, there is no consensus on the role of proper running exercise in OPLL-related secondary chronic spinal cord injury. Takahiro argued that running exercise is an important rehabilitation measure for spinal cord function recovery, as proper exercise can not only promote spinal cord recovery but also improve motor function (Shibata et al., 2021). However, secondary SCI caused by OPLL differs fundamentally from traumatic SCI, and exercise cannot relieve spinal cord compression secondary to spinal stenosis. In contrast, excessive bending of the spine may promote repeated friction stimulation between the spinal cord and the ossification site, and a large number of inflammatory factors may further damage spinal cord tissue.

Our previous study demonstrated the significance of yes-associated protein (YAP), a transcriptional coactivator that plays a crucial role in promoting osteogenic differentiation of cells under tensile stress (Zhu et al., 2023). However, its in vivo relevance has yet to be validated. Therefore, we generated a mouse model with Enpp1 gene deletion that develops spontaneous heterotopic ossification in tendons and ligaments. Our objective was to assess abnormal YAP activation by identifying the ossification focus, which is essential for our subsequent pathological mechanism research.

In this study, we utilized this gene-deficient mouse model to develop an external fixation brace that restricts the spine’s flexion and extension and evaluated its effect on delaying ossification progression. Similarly, a sloped treadmill was utilized to induce spine and Achilles tendon activity, allowing us to observe the ossification range and gait changes (Figure 1). Through this experimental design, we aimed to explore the effect of mechanical stimulation on heterotopic ossification in vivo.

Experimental animals are biological models obtained through carefully designed experiments that offer a high degree of uniformity. They allow researchers to study the pathophysiological mechanisms of diseases and potential treatments more accurately. Previous research has shown that mechanical stimulation has a significant impact on cell osteoblastic differentiation at the cellular level, and this process is associated with the nuclear translocation of the mechanical signal response factor YAP (Zhu et al., 2023). In this study, we aimed to establish an appropriate animal model that can serve as a basis for further in vivo investigation of this mechanism.

To our knowledge, few animal models that address the mechanical factors that promote the progression of heterotopic ossification have been reported. Tsukamoto’s team designed a device that could apply circular stretching to the rat’s caudal ligament (Tsukamoto et al., 2006). The rat’s pelvis was secured to a distal fixator with a steel pin inserted into the 8th caudal vertebra, while its body was restrained on the edge of a moving bed that moved back and forth to exert twice its body weight on the anterior tail segment (2-5 caudal vertebrae). Histological staining revealed that when this traction was sustained for 2 weeks, heterotopic ossification occurred near the intervertebral space in the stressed area of the coccygeal ligament. A significant number of mature chondrocyte-like cells appeared near the ossification area, and the duration of stretching time had a positive correlation with these pathological changes.

Zhao et al. (2021) developed a device capable of applying stretch stress to the ligamentum flavum in the thoracolumbar segment of rats. They immobilized the rats’ limbs in parallel splints, with their chests and abdomens pressed against elastic boards. When the device was turned on, an eccentric wheel drove the elastic board to periodically flex and extend the thoracolumbar vertebrae, thereby forcing the ligamentum flavum to bear cyclic tensile stress. Four weeks of cyclic stimulation induced heterotopic ossification of the ligamentum flavum at the stress site of the thoracolumbar region. Further in vitro verification indicated that this process might be associated with the high expression of integrin β3 induced by stretch stimulation.

Histological staining and micro-CT imaging results from previous animal models have confirmed the ability of mechanical factors to induce heterotopic tissue ossification in animals. However, non-traumatic heterotopic ossification in clinical cases often involves multiple pathogenic factors, making it difficult to simulate the complete pathological process of the disease with a single mechanical stimulus factor. Therefore, there is a need to establish more reasonable animal models. Previous clinical sequencing results have shown that the expression of the ENPP1 gene in the peripheral blood of OPLL patients was significantly lower than that of normal individuals (Saito et al., 2011; Mehta et al., 2012; Ferreira et al., 2021; Kato et al., 2022), and naturally mutated tiptoe walking (ttw) mice have been shown to have mutations in the gene site of Enpp1 (Mackenzie et al., 2012). Building on this, our research group constructed EIIa ^cre -Enpp1 ^flox/flox gene-defective mice using CRISPR‒Cas9 technology. These mice exhibited clinical symptoms similar to those of OPLL patients, including spontaneous ectopic bone formation, gradual spinal cord compression, and the onset of chronic SCI symptoms such as unsteady gait and difficulty walking. With this gene-defective mouse model, we can study the mechanism of mechanical stimulation in the progression of OPLL and propose corresponding intervention measures according to these mechanical factors.

Micro-CT imaging revealed that OPLL originated behind the intervertebral space and appeared relatively late behind the vertebral body. This difference in position may be due to variations in the stress placed on the ligaments. The PLL located in the intervertebral space is subjected to tensile stress due to the spine’s flexion and extension, while the PLL close to the vertebral body is subjected to less tensile stress due to its close connection to the vertebral body. Furthermore, immunofluorescence staining indicated a significant increase in YAP expression at the site of ligamentous heterotopic ossification, which is consistent with previous in vitro studies (Zhu et al., 2023). This finding suggests that mechanical stimulation plays a vital role in the pathological process.

Some scholars have also explored targeting mechanical factors as a new approach to delaying ossification. Previous studies have mainly focused on therapeutic measures for the gene mutation itself. For example, Hiratsuka et al. supplemented levamisole and exogenous inorganic pyrophosphate (PPi) in ttw mice to maintain a high level of serum PPi, thus delaying the progression of heterotopic ossification (Hiratsuka et al., 2018). Additionally, some studies have shown that supplementing ttw mice with exogenous Enpp1 recombinant enzyme (INZ-701) can maintain blood PPi at physiological levels for 3 consecutive days and that continuous medication for 8 weeks can effectively prevent heterotopic ossification. Furthermore, H2 receptor antagonists (Maeda et al., 2015), nuclear retinoic acid receptor γ agonists (Hiratsuka et al., 2018) and neurotrophic factors (Uchida et al., 2012) have also been reported to delay heterotopic ossification or alleviate secondary chronic SCI in ttw mice. Although these therapeutic measures may have clinical efficacy, the ethics, safety, and side effects associated with drug use make them difficult to use in the clinic in the short term. Therefore, safer and more feasible therapeutic approaches need to be considered.

To delay the progression of heterotopic ossification of the spinal ligaments, we wanted to limit the movement of the spine. To achieve this, we designed an external spinal fixation brace, which is the first of its kind to be used in mice. The brace limits spinal flexion and extension, thereby reducing ligament stretching stress. The weight of the external fixation brace was only 1–1.5 g, which is approximately 1/20 of the body weight of mice, and had no adverse effect on physiological activities. Additionally, the external fixation brace was replaced weekly to accommodate the changing body size of the mice. When wearing the brace, the double forelimbs were left fully exposed to avoid interfering with eating and drinking.

We observed that during normal physiological activities in mice, spinal flexion and extension were used to maintain body balance, especially during standing and running. X-ray images showed that when mice were in a standing position, the cervicothoracic segments and the lower thoracic segments were in a state of hyperflexion. However, when mice wore the external fixation brace, their spinal mobility was significantly limited. While this animal model did not show a significant effect in delaying ossification in the cervicothoracic segment, it did show such an effect in the lower thoracic segment. There may be several reasons for this phenomenon. First, the external fixation brace was made by cutting and curling flat material, and several layers were wrapped around the abdomen of the mice to increase its mechanical strength, while there was only one layer around the cervical spine. Second, in addition to flexion and extension, the neck can also rotate, but the external fixation brace could not effectively limit the rotation of the neck. Conversely, the restrictions on the limbs made it difficult for the mice to make any abdominal movements.

To further evaluate the impact of changes in the ossification range on pathological signs in mice, we utilized the BBB score for SCI to observe the mobility of the hind limbs of mice. The results indicated that the external fixation group was able to effectively protect against SCI from 16 to 24 weeks of age, as evidenced by a significantly higher BBB score than that of the control group. However, by 28 weeks of age, there was no significant difference in BBB scores between the two groups. We speculated that the deletion of the Enpp1 gene might be a contributing factor to the mice’s strong ability to spontaneously develop ligamentous heterotopic ossification. While limiting spinal movement could delay the process of heterotopic ossification, it had no effect on the treatment or reversal of ossification.

In this part of the experiment, we also aimed to evaluate the impact of physical hyperactivity, specifically running, on the progression of ossification. The effect of physical activity on patients with heterotopic ossification remains a controversial topic in the medical community. While some scholars have suggested that passive limb exercise can effectively delay traumatic heterotopic ossification (Vasileiadis et al., 2021; Wang et al., 2021; Vasileiadis et al., 2023), others have concluded that repeated stress stimulation of the spine can accelerate ossification progression and negatively impact patients’ quality of life (Lee et al., 2017; Kang et al., 2019; Lee et al., 2021). Additionally, the benefits of appropriate physical exercise for the recovery of neurological function in OPLL patients have not been uniformly established. Some experts recommended running exercise as an essential rehabilitation measure for the recovery of spinal cord function after chronic SCI, while others argued that mechanical stimulation from running may not only increase the size of the ossification range but also cause additional damage to the spinal cord tissue (Wang et al., 2015; Yu et al., 2019; Shibata et al., 2021). Repeated flexion and extension of the spine may lead to a smaller spinal canal volume, increasing the compression of the spinal cord (Azuma et al., 2010; Ito et al., 2015). Moreover, the uneven surface of the protruding site of ossified tissue can cause direct physical damage to the spinal cord, leading to secondary inflammatory reactions (Fournely et al., 2020; Jannesar et al., 2021).

This study aimed to investigate whether excessive physical activity could accelerate the progression of ossification by introducing treadmill training. The intensity of the training was gradually increased by adjusting the speed, slope, and running time. Micro-CT results revealed a significant expansion of ossification in specific segments of the mice undergoing treadmill training compared to the control group. Additionally, the BBB score of the treadmill group at 12 weeks of age was significantly lower than that of the control group. These findings are consistent with previous reports indicating that repeated flexion and extension of the spinal canal may aggravate spinal cord compression when ossification compression is not relieved (Lee et al., 2017; Kang et al., 2019; Lee et al., 2021). Further experimental verification is needed to determine whether the aggravation of SCI is due to direct mechanical injury or secondary inflammatory reactions.

This study has several limitations that need to be acknowledged. First, while radiographs showed that the external fixation brace effectively limited spinal movement and micro-CT demonstrated delayed progression of lower thoracic OPLL, the brace was found to be less effective than internal fixation due to soft tissue cushioning. Second, the accuracy of HE staining of axial sections was limited by frozen section technology, making it difficult to position the staining precisely behind the intervertebral space. Therefore, the results of HE staining can only be used to supplement the findings of micro-CT. Third, owing to the diminutive size of mice, our current approach was confined to employing external fixation solely for the purpose of restricting spinal movement (from cervical to lumbar regions) rather than applying any alternative interventions aimed at curbing activity in other body areas, such as the limbs. Finally, the genetically defective mice used in this study develop ossification of ligament tissue throughout the whole body, rather than just in the PLL or Achilles tendon, which may affect experimental observations to some extent.

In this study, we designed an external fixation device to limit the flexion and extension of the spine and reduce the ossification progress of PLL in mice with Enpp1 gene deficiency. Additionally, we used treadmill training to promote heterotopic ossification of the PLL and Achilles tendon. Our in vivo experiment confirmed that mechanical stimulation is an important factor in the progression of OPLL. These rational animal models provided an experimental basis for further study of the specific pathological mechanism. Furthermore, we confirmed that YAP, a transcriptional coactivator protein, plays a crucial role in accelerating the process of heterotopic ossification of ligaments by mechanical stimulation. However, more in vivo studies are necessary to confirm these results.