Osteoporosis of the vertebra and osteochondral remodeling of the endplate causes intervertebral disc degeneration in ovariectomized mice

Lower back pain and spinal compression nerve pain are the major symptoms caused by intervertebral disc degeneration (IVDD) in the clinic [1, 2]. Spinal instability or disability is common in serious IVDD which leads to enormous human suffering and significant socioeconomic losses [3]. Unfortunately, there are no currently effective methods to repair IVDD [4] and disc resection with interbody fusion are often the final choice. Therefore there is an urgent need to explore the key mechanism of IVDD and to develop drugs for its treatment.

It has long been recognized that musculoskeletal degeneration disorders such as osteoporosis, IVDD, and osteoarthritis are a difficult focus in locomotor disease research. An intimate relationship between cartilage and subchondral bone has been proven in recent years. Anatomically, the vertebrae and the intervertebral discs are combined in bundles to form the motion segments of the spine. From the mechanical and biological points of view they are closely linked and are considered as a functional unit [5‐7]. Although it is not fully clear whether it precedes or follows nucleus pulposus degeneration, the modic change in the endplate is an important feature of IVDD physiopathology [8]. Therefore, the health of the bone and its attached nonosseous tissues such as cartilage and disc are tightly associated. Research has shown that crosstalk between bone and cartilage is elevated in osteoarthritis where the coupling of bone and cartilage turnover is even aggravated [9‐11]. Moreover, remodeling of the subchondral bone microstructure due to osteoporosis could further exacerbate experimental osteoarthritis [12]. All of this suggests that the subchondral bone is an indispensable factor in the process of osteoarthritis. In terms of IVDD, however, there is a lack of research on the pathological mechanisms of subchondral bone in IVDD, and it is also unclear how osteoporosis affects the nonosseous tissues such as the intervertebral disc.

Because of the structural similarities between joint and intervertebral discs [13, 14], it is well known that the intervertebral disc is a nonvascular structure and the exchange of substances between the intervertebral disc and the vertebra depends on the cranial and caudal endplates [15‐17]. The endplate contains marrow contact channels which provide nutrients for the intervertebral disc and discharge metabolic waste through diffusion and liquid flow under cyclic loading [4, 18, 19]. Hence, changes in the vertebra microenvironment resulting from rapid bone turnover during vertebral osteoporosis and obstruction of marrow contact channels induced by calcification of the cartilaginous endplate may accelerate IVDD.

Recent studies have provided increasing evidence that osteoporosis is associated with the evolution of IVDD. The osteoporosis of vertebrae in postmenopausal women was correlated with IVDD [20, 21], and sex hormones can affect the severity of IVDD [21‐23]. Furthermore, in osteoprotegerin (OPG) knockout mice, ossification occurred in the cartilage endplate and resulted in IVDD [24]. IVDD often occurs with osteoporosis of the vertebrae, indicating that the development of osteoporosis and IVDD might be a coupling process which could explain why postmenopausal women have more lower back pain than men. Some studies have reported the prospect of delaying the course of disc degeneration through improving bone metabolism and vertebral osteoporosis. For example, it was found that alendronate could retard the progression of lumbar IVDD in ovariectomized rats by improving the bone quality [25, 26]. Calcitonin could also suppress intervertebral disk degeneration and preserve lumbar vertebral bone mineral density and bone strength [27]. However, it is still puzzling that some clinical and epidemiological studies have shown that osteoporosis is inversely related to spinal degenerative diseases and IVDD [19, 28, 29]. These studies support osteoporosis in increasing endplate permeability and delaying IVDD [19]. In addition, radiographic features of lumbar disc degeneration were associated with an increased bone mineral density (BMD) in the spine [28, 30]. Thus, the relationship between osteoporosis and IVDD is still controversial and confusing. Consequently, whether a positive correlation exists between osteoporosis and IVDD and how the vertebral body affects the intervertebral disc are still undefined and remain to be further clarified.

Mice are commonly used as an animal model for osteoporosis and intervertebral disc degeneration [31‐33]. Many studies have demonstrated that the ovariectomized mouse is a good model for postmenopausal osteoporosis [34]. Thus, the purpose of this study was to determine the roles of postmenopausal osteoporosis in IVDD and to further clarify its underlying mechanism by assessing the detailed pathological changes in L4–L5 spine motion segments including vertebrae, the endplate, and the intervertebral disc in ovariectomized mice.

IVDD and osteoporosis are the most common degenerative diseases in the spine, both of which are often accompanied with the other. However, the detailed relationship between them is not clear. In the present study, we evaluated the effects of estrogen deficiency on the bone mass and microarchitecture in vertebrae, the microarchitecture and porosity of the endplate, and the histopathology of the adjacent intervertebral disc to elucidate the possible relationship between osteoporosis and IVDD and to explore the superficial mechanism of IVDD associated with osteoporosis.

Our study showed that OVX could cause vertebrae osteopenia. Additionally, OVX promoted bone turnover and osteochondral remodeling at the junction of the vertebra and intervertebral disc, leading to an increased ossification and hypertrophy of the endplate, abnormal pores within the cartilaginous endplate, high porosity between the vertebra and intervertebral disc, and narrowing of the intervertebral disc space. Therefore, OVX exerted a detrimental effect on subchondral bone structure, particularly in the subchondral plate, which was closely related with disorders of the overlying cartilaginous endplate and played a crucial role in the development of IVDD.

At the same time, decreasing proteoglycans in the nucleus pulposus, increasing cracks within the annulus fibrosus, and osteochondral remodeling of the endplate could be also found in the intervertebral disc histomorphology of OVX mice. TRAP staining showed that the osteoclasts in the subchondral bone were significantly increased and particularly appeared in the endplate. The immunohistochemistry showed a corresponding increase in OSX expression, indicating that the ovariectomy induced a fast bone turnover which led to the structural remodeling of the endplate and changes in the porosity. It has been reported that there is a significant correlation between the effective permeability and marrow contact channel or porosity of the endplate [35, 36]. Thus, the increase in permeability induced by postmenopausal vertebral osteoporosis and endplate remodeling might be the most important factor contributing to the lesions of the soft tissue of the intervertebral disc in OVX mice. Additionally, the results of immunohistochemistry showed that OVX weakens the expressions of Col II and upregulates OPN and VEGF expressions in the endplate and annulus fibrosus, suggesting that abnormal ossification and angiogenesis are involved in the process of IVDD related to osteoporosis.

Estrogen as an endocrine hormone influences the metabolism of various tissues and organs in the body, such as the rich collagenous tissues of bone, cartilage, disc, artery, and skin, etc. [23]. It has been clearly shown that a rapid decline of estrogen levels is an important factor for osteoporosis in postmenopausal women. Insufficient estrogen caused by OVX is the common method to mimic postmenopausal status and can effectively induce osteoporosis. Many studies have confirmed that ovariectomy could cause osteoporosis in various animals, such as rats, mice, monkeys, and so forth [25, 32, 37]. In the current study, our results show that ovariotomy caused an increased body weight with abdominal fat accumulation but a decreased uterine weight with a severe atrophy in morphology. Furthermore, a mass of fat in the bone marrow of the tibia and thin trabeculae were observed in OVX mice, which supports that estrogen deficiency may induce adipogenic differentiation of bone marrow stem cells, but not osteogenesis. Moreover, the poorer results for BMDtv and the bone parameters including BV/TV, Tb.N, Tb.pf, CONN.D, and SMI were seen in the L5 vertebra of the OVX group when compared with the control group suggesting a deterioration in bone quality and quantity due to ovariectomy, consistent with previous studies [32, 37, 38]. Progressively, we investigated the effects of OVX on the intervertebral disc. As expected, OVX mice showed increases in the osteochondral remodeling and porosity of the endplate accompanied by decreases in the height and volume of the intervertebral disc, loss of proteoglycans and cells, and the formation of clefts, suggesting that ovariotomy effectively accelerates the deterioration of the endplate and induces IVDD. Considering that the intervertebral disc is avascular, material exchange with the vertebrae mainly relies on the endplate. Thus, it can be speculated that some effects of estrogen on the intervertebral disc are indirect. The initial occurrences of osteoporosis and cartilaginous endplate remodeling and subsequent disorders of intervertebral disc metabolism may be a reasonable explanation for the induction of IVDD by estrogen deficiency.

Many risk factors have been found to be involved in the IVDD process, including age, sex, injury, obesity, genetic predisposition, immune, nutrition, inflammation, and mechanical factors [7, 24, 39]. Increasing evidence indicates that IVDD is associated with the disruption of an intact spinal structure, such as adjacent structures including the vertebra and endplate [14, 16, 17, 22]. In recent years, the important role of subchondral bone in the development of osteoarthritis has been increasingly recognized [40]; in fact, it is the same with IVDD. Changes in vertebral strain energy were correlated with increasing Schmorl’s nodes in multilevel lumbar disk degeneration [41]. The changes in vertebral-endplate subchondral bone signal detected by magnetic resonance imaging (MRI) may serve as an ‘active discopathy’ judgment [8]. Therefore, as the bridge of communication between the vertebra and the intervertebral disc, the integrity of the endplate may be the key factor affecting the intervertebral disc [42, 43], and the deterioration of subchondral bone may be the trigger of IVDD. Interestingly, in some studies, ossification of the endplate with a reduction in porosity and permeability accelerated degeneration of the intervertebral disc during the osteoporosis process [17, 38]. It has been shown by a dynamic contrast-enhanced MRI study that ovariectomy induces a decrease in the second wash-in phase, indicating that the diffusion between the vertebra and the disc was impaired as a result of ovariectomy [44]. On the contrary, other studies have concluded that osteoporosis increases the porosity and permeability of the endplate, leading to a delay in disc degeneration [19]. Indeed, these conclusions are derived from the theory that nutrient acquisition of the intervertebral disc relies on endplate permeability. However, our study has found that endplate osteochondral remodeling causes a high endplate porosity/permeability leading to disc degeneration, and this may be associated with antigen exposure, immune inflammation, and loss of nucleus pulposus osmotic pressure in the intervertebral disc due to the increased endplate porosity. TRAP staining showed that a large number of activated osteoclasts appeared in the osteochondral interface of OVX mice, which indicated the osteochondral remodeling of the endplate. Here, we speculate that the rapid bone turnover caused by OVX should be responsible for the endplate remodeling and the increased porosity which could reflect the high permeability between vertebrae and the intervertebral disc and which is closely associated with disc degeneration. Several studies could support these etiological hypotheses. Rodriguez et al. [36] found that porosity and permeability of the endplate were increased with age and disc degeneration. Osteoprotegerin (OPG) knockdown mice had an increase in neovascularization and expression of inflammatory cytokines in the intervertebral disc, indicating that osteoporosis can induce inflammation and consequently become the cause of disc degeneration [24]. The modic change of the endplate is closely related to back pain [1, 16]. Endplate damage could also lead to decompression of the nucleus pulposus [6]. Furthermore, severe disc degeneration is more common in patients with endplate modic changes, which suggests that modic changes could result in the occurrence and development of IVDD [45]. Consistently, we found that remodeling of the endplate led to an increased surface porosity and permeability, which could lead to degeneration of the intervertebral disc.

Yuan et al. [42] successfully developed a rat model of IVDD using the injection of alcohol within the endplate to block the blood vessels. IVDD and osteoporosis can also result in endplate cartilage injury [46], supporting that the intervertebral disc interacts with the vertebra. A finite element analysis showed us that both decreased trabecular core density and IVDD have been suggested to play roles in vertebral fractures. IVDD caused a shift of the load from the nucleus pulposus to the anulus fibrosus, resulting in bone adaptation which was presented as a dramatically reduced density of the trabecular core and an increased density in the vertebral walls [47]. Furthermore, the positive correlation between the thickness of the subchondral bone and the proteoglycan content of the adjacent disc have been found in human cadaveric material, particularly in the region of the nucleus pulposus [48]. Bone responds to a greater hydrostatic pressure exerted by discs with higher proteoglycan content than that by discs with less proteoglycan present, which indicates that vertebral osteoporosis with resultant endplate-bone remodeling could affect the flow of solutes to and from the intervertebral disc by failing to maintain the normal hydrostatic pressure of the nucleus pulposus, resulting in the loss of proteoglycans [48]. Therefore, only by guaranteeing a certain appropriate permeability with the proper microstructure of the subchondral bone, being neither too high nor too low, could the endplate maintain the specific hydrostatic pressure and balance the microenvironment of the nucleus pulposus to ensure the stability of the intervertebral disc. The endplate might play the role of a biological semipermeable membrane.

Col II is the most abundant collagen in cartilaginous tissues and is often referred to as the major collagen. Therefore, its content is crucial for proper disc function, particularly in the cartilaginous endplate and nucleus pulposus [49]. Our study showed that OVX could effectively weaken Col II expression, especially in the zone of ossification of the endplate, which could be related to OVX-induced osteochondral remodeling. Furthermore, the levels of OPN and VEGF were markedly elevated in the endplate and annulus fibrosus of OVX mice, consistent with the results of IVDD caused by spinal instability [31]. It is noteworthy that colocalized expression of OPN and VEGF were visible in degenerative discs. The molecular pathological changes in the intervertebral disc indicate that ossification and angiogenesis of the intervertebral disc are not to supply more nutrient for disc repair but to accelerate the intervertebral disc fibrosis and ossification.

Osteoclastogenesis is a prerequisite for osteochondral remodeling, and the osteoclast resorption process is required to degrade subchondral bone and cartilage [50]. Our study found a significant increase in osteoclasts in the subchondral bone and cartilaginous endplate of the OVX group, consistent with the previous findings. Interestingly, alendronate and calcitonin can inhibit osteoclast activity and osteochondral remodeling to extenuate IVDD [26, 27]. Although parathyroid hormone 1–34 has substantial anabolic effects on bone mass and trabecular microarchitecture, nonsignificant effects have not yet been found on disc degeneration [51]. This leads us to suggest that preventing the remodeling of the osteochondral structure caused by the initial osteoclastogenesis after the menopause could be an important aspect in the fight against the occurrence of intervertebral disc degeneration. Furthermore, some reports have confirmed that suppressing osteoclastogenesis and aberrant angiogenesis could blunt IVDD and osteoarthritis [52, 53], which provides further promise for the treatment of IVDD in the future. It may be beneficial to renew the intervertebral disc by balancing bone metabolism and regulating permeability of the endplate.

In summary, this study uncovered the mystery of the deteriorative effects of OVX on IVDD, clearly showing that OVX can induce and accelerate the progression of disc degeneration. The underlying mechanisms could be related to the destruction of the structural integrity and the function of the vertebra and endplate induced by OVX, both of which are essential structures for maintaining disc function. OVX modulates bone turnover and results in osteochondral interface remodeling which could also influences the expressions of Col II, OSX, OPN, and VEGF to stimulate the disc degeneration. It is conceivable that osteoporosis of vertebrae and endplate remodeling resulting from estrogen depletion may affect the bone marrow microenvironment and endplate permeability which subsequently could alter the metabolism and biomechanics of the intervertebral disc. Simultaneously, the reduction in estrogen caused by OVX could also directly affect the metabolism of the intervertebral disc because of the presence of estrogen receptors in the nucleus pulposus. These changes may be a reasonable explanation for disc degeneration related to osteoporosis. However, our observations are preliminary and need to be further confirmed through clinical trials or additional animal experiments. Based on the above results, intervention for osteoporosis and improvement in the vertebral body and endplate structure may be an effective way to retard IVDD; the mechanisms of IVDD related to osteoporosis need to be further studied.