Expression and significance of protein-degrading enzymes in cartilage and subchondral bone in osteoarthritis

Background TGFβ1 plays an important role in the metabolism of articular cartilage and bone; however, the pathological mechanism and targets of TGFβ1 in cartilage degradation and uncoupling of subchondral bone remodeling remain unclear. Therefore, in this study, we investigated the relationship between TGFβ1 and major protein-degrading enzymes, and evaluated the role of high levels of active TGFβ1 in the thickening of subchondral bone and calcification of articular cartilage. Materials and methods The expression of TGFβ1 and protein-degrading enzymes in clinical samples of articular cartilage and subchondral bone obtained from the knee joint of patients with osteoarthritis was detected by immunohistochemistry. The expression levels of TGFβ1, MMP-3, MMP-13 and IL-1β in cartilage and subchondral bone tissues were detected by absolute real-time quantitative RT-PCR. The expression of TGFβ1, nestin and osterix in subchondral bone was detected by Western blot analysis and immunohistochemistry. The degree of subchondral bone thickening was determined by micro-computed tomography (CT) imaging. Results Expression of TGFβ1 and cartilage-degrading enzymes was higher in the cartilage-disrupted group than that in the intact group. Furthermore, expression of TGFβ1, nestin and osterix was significantly higher in the OA group than that in the control group. Micro-CT imaging showed that the OA group had abnormal hyperplasia of subchondral bone. Conclusions The data suggest that highly active TGFβ1 activates the expression of cartilage-degrading enzymes. Abnormally activated TGFβ1 may induce formation of the subchondral bone and expansion of the calcified cartilage area, eventually leading to degradation of the cartilage tissue. released the matrix-degrading enzymes [35] degradation of the extracellular matrix (ECM). analyzed the expression of TGFβ1 and the major cartilage-degrading enzymes MMP-3/13 and IL-1β in the OA knee tibia plateau. Our results indicate that the high levels of TGFβ1 expression promote the synthesis of matrix degrading-enzymes, leading to cartilage degradation. Matrix metalloproteinase MMP3 is a mesenchymal lytic enzyme [36] that degrades the proteoglycans and glycoproteins in the ECM. Studies have shown that MMP3 in OA cartilage, synovium, synovial fluid and peripheral blood [37] degradation, hypertrophy of chondrocytes vascularization calcification of articular


Introduction
Osteoarthritis (OA) is the most common chronic degenerative joint disease in orthopedics. immune response, elevated intraosseous pressure, enzyme degradation of cartilage, cytokines, genetic polymorphisms and other factors are related to OA; however, the specific pathological mechanism remains unclear [7,8] remodeling of OA subchondral bone is highly active in a process that involved both bone resorption and formation. Inhibiting subchondral bone remodeling effectively slowed down articular cartilage degeneration, indicating a close relationship between these two processes as well as an important role in the early stages of cartilage degeneration [7].
Previous studies suggest that structural changes in the subchondral bone are closely related to the pain experience by patients with OA, and the severity of OA can be expected [11]. However, the current relationship between the changes in subchondral bone and the pathogenesis of OA as well as the underlying pathological mechanisms are not clear.
Transforming growth factor (TGFβ1) is a pleiotropic cytokine that is closely related to the pathological changes of OA [12,13]. During the early stages of joint development, TGFβ1 is important for the maintenance of a normal chondrocyte phenotype [14]. The articular cartilage performs the biological functions of shock absorption and buffering mechanical stress. Chavez et al. [15] reported that TGFβ increases the expression of type II collagen (ColII) and the proteoglycan aggrecan (ACAN) by regulating Smad2/3. In contrast, there are also extensive reports indicating that TGFβ1 is an important factor in joint destruction.
Therefore, the current evidence for the role of TGFβ1 in the development of OA is contradictory [16]. In many animal models of OA, active TGFβ1 is expressed at high levels in epiphyseal and subchondral bone calcification [17]. Mesenchymal stem cells (MSCs), also known as pluripotent stem cells, are capable of self-renewal and differentiation into a variety of cells, allowing tissue growth and regeneration [18]. Nestin was identified in neural stem cells and is now used as a marker of bone marrow-derived MSCs [19]. MSCs first differentiate into osteogenic precursor cells and then into osteoblasts. The osteogenic precursor-specific transcription factor, osterix, which was first discovered in mice [20], is one of the most important transcription factors involved in osteoblast differentiation, and its loss leads to a direct loss of bone formation ability. Therefore, osterix expression is used to evaluate osteoblast differentiation. Clinically, TGFβ1 is highly expressed in the subchondral bone of OA patients, leading to increased formation of MSC clusters and calluses, and ultimately to the occurrence of OA. Moreover, TGFβ1 antibody treatment improves bone microstructure, reduces the severity of OA, and delays the progression of joint deterioration [21].Therefore, TGFβ1 is crucial for the metabolic homeostasis and structural integrity of articular cartilage, although the specific mechanism of action of TGFβ1 in OA is still controversial.
Subchondral bone is the main link between cartilage and bone and bone plays an important role in the stress transmission process. Therefore, a role for subchondral bone plays a role in OA is increasingly being proposed [22,23]. In our preliminary study, we found that active TGFβ1 was highly expressed in the subchondral bone of OA patients, and the reactivity of the patient's chondrocytes to TGFβ1 was decreased, which was consistent with the changes in TGFβ1 receptor expression in elderly OA cases. The level of active TGFβ1 is also elevated in a mouse anterior cruciate ligament rupture OA model (ACLT-OA) and a spontaneous OA guinea pig model. Cui et al. [24]showed that reduced TGFβ1 expression in the mouse ACLT-OA model impeded articular cartilage degeneration. The subchondral bone undergoes a process of plastic bone and bone remodeling. During osteolysis caused by osteoclasts, TGF-β1 induces proliferation of nestin-positive MSCs and transcription factor osterix-positive osteogenic precursor cells, which couples bone remodeling and angiogenesis [25]. These reports indicate that TGFβ1 plays different roles in the remodeling of subchondral bone and the pathological changes of articular cartilage, although the underlying mechanism requires further clarification.
Clinical OA joint samples are obtained long after the lesions have formed and after a very long period of wear. Animal models of OA offer the advantage that samples can be obtained at different stages during the development of the lesion. Such models provide the opportunity to investigate differences in the microstructure of the subchondral bone at different stages and different degrees of cartilage wear, the relationship between these differences and TGFβ1expression and the relationship with OA articular cartilage degeneration and subchondral bone remodeling.
In this study, control normal cartilage specimens and clinical OA specimens were collected to investigate the changes in cartilage and subchondral bone structure, the activity of TGFβ1 and its correlation with cartilage-degrading enzymes and OA lesions. This information will provide an important basis for the clinical application of TGFβ1-targeted therapy for OA.

Histological examination
The knee tibial plateau tissue sections were dewaxed in xylene and then hydrated in a series of graded ethanol solutions. The sections were placed in 0.4% pepsin and fixed at 37℃ for 30 min. After washing three times with PBS, sections were immersed in 3% hydrogen peroxide for 30 min, washed again and then incubated in 5% fetal bovine serum blocking solution for 30 min. The sections were then placed in hematoxylin solution for 30 s, acidified for 3-5 s, and rinsed with tap water until no color, tap water returned to blue for 10 min. Subsequently, sections were washed three times with PBS followed by 95% alcohol before dehydration in anhydrous ethanol, and then immersed in xylene to render the tissue transparent. Sections were sealed with optical resin. Immunohistochemical staining (percentage positive cells and staining intensity) was scored quantitatively in six fields randomly selected randomly under the microscope. The percentage of positive cells was scored as follows: 0, <5%; 1, 6%-25%; 2, 26%-50%; 3, 51%-75%; and 4, 76%-100%. Staining intensity was scored as follows: 0, no staining; 1, light brownish yellow; 2, brownish yellow; and 3, tan. The final score for each section was calculated by multiplying the scores for percentage of positive cells and staining intensity.

Western blot analysis
The hypoglossal bone tissue of the tibial plateau was ground under liquid nitrogen and then incubated overnight at 4℃ in high-efficiency lysate RIPA buffer and the PMSF protease inhibitor. The protein concentration was determined by ultraviolet spectrophotometry (A280). 1/4 of a loading buffer was added to the EP tube for 10 minutes to be used at -80℃. The protein sample was added to the loading buffer, electrophoresed on an SDS polyacrylamide gel. For TGFβ1 and osterix detection, proteins were transferred to the PVDF membrane using a semi-dry method, whereas a wet-transfer method was used for detection of nestin. Membranes were then blocked at room temperature for 2 h, washed three times with TBST and incubated overnight with the primary detection antibodies (TGFβ1, nestin, osterix antibodies, 1:1,000; β-actin antibody, 1:20,000). After washing three times with TBST, membranes with the secondary antibody for 2 h at room temperature. Protein immunoreactivity was visualized using XXXX. Image-Pro Plus software was used to analyze the grayscale values of the protein bands.

Real-time PCR analysis
Total RNA was extracted from the tibial plateau using Qiagen kit and TRIzol reagent (Invitrogen) and reverse transcribed into cDNA using SuperScript First-Strand Synthesis System (Invitrogen). Using cDNA as a template, SYBR GreenMaster Mix (Qiagen) was used for RT-PCR amplification of target genes (TGFβ1, MMP-3, MMP-13 and IL-1β) and internal control (β-actin) using the following PCR parameters: pre-denaturation at 95℃ for 2 min; denaturation at 95℃ for 30 s; extension at 60℃ for 30 s (40 cycles). The primer sequences are shown in Table 1. Relative gene expression was analyzed using the 2-ΔΔCt method.

Micro-CT joint imaging
The isolated tibia was imaged using the GEHC MicroView. The resolution was adjusted to 9 mm for scanning, and the 3-dimensional (3D) image was then reconstructed. After 3D reconstruction, the region of interest (ROI) of the subchondral bone in the human tibial plateau was selected. The ROI included all trabecular bone surrounded by a growth plate and a distal tibia (1-4 mm). For each ROI, we measured bone mineral density (BMD), bone volume fraction (BV/TV), trabecular thickness (Tb.Th), trabecular bone number (Tb.N), trabecular bone spacing (Tb.Sp), trabecular surface area/skeleton (BS/BV).

Statistical analysis
All statistical analyses were performed using SPSS 20.0 (IBM Corp, Armonk, NY, USA).
Differences between groups was evaluated by one-way ANOVA and chi-squared tests. Pvalues < 0.05 were considered to indicate statistical significance.

Results
Visual observation of specimen morphology By visually observing the specimen, it was found that the articular cartilage on the lateral side of the tibial plateau was lightly worn, and the cartilage surface was relatively intact.
No cartilage sclerosis, bare subchondral bone or osteophyte formation were found. In addition, the subchondral bone thickness was increased and the cartilage surface was uneven with obvious cracks. In the cartilage exfoliation and cartilage intact groups, almost completely disappeared than the hyaline cartilage. The subchondral bone was exposed and deformed obviously. The calcified cartilage thickening was obviously exposed, and the bone cell and bone structure in the subchondral bone area decreased although the accumulation of lipid droplets was increased. Accumulating evidence shows that transforming growth factor (TGFβ1) is closely related to the pathological changes of OA [28,29] . In normal joints, TGFβ1 is involved in the regulation of cartilage growth and metabolism of subchondral bone, and plays a key role in maintaining joint homeostasis. However, the effects and functions of TGFβ1 are changed in OA, in which the cytokine is involved in the induction of chondrocyte phenotype differentiation, chondrocyte proliferation, and destruction of the extracellular matrix synthesis [30] . Subchondral bone callus formation and abnormal bone reconstruction are symptoms of the development of arthritis [31] and important mechanisms responsible for the progression of OA [32] . Studies have shown abnormal elevation of TGFβ1 in OA human specimens and various animal models [33,34] , which is consistent with the findings of this study. Furthermore, it has been reported that the stimulatory factor TGFβ1 released from the chondrocytes of damaged joints stimulates the synthesis of protein matrix-degrading enzymes [35] , leading to degradation of the extracellular matrix (ECM). In this study, we analyzed the expression of TGFβ1 and the major cartilage-degrading enzymes MMP-3/13 and IL-1β in the OA knee tibia plateau. Our results indicate that the high levels of TGFβ1 expression promote the synthesis of matrix degrading-enzymes, leading to cartilage degradation. Matrix metalloproteinase MMP3 is a mesenchymal lytic enzyme [36] that degrades the proteoglycans and glycoproteins in the ECM. Studies have shown that MMP3 is highly expressed in OA cartilage, synovium, synovial fluid and peripheral blood [37] .
Similarly, we found that MMP3 is highly expressed in cartilage and subchondral bone.
Matrix metalloproteinase MMP13, also known as collagenase [38] , is the major type II collagen-degrading enzyme and is prominently expressed in the cartilage matrix [39] .
Studies have confirmed high expression [40] and enhanced activity [41,42] of MMP13 in OA cartilage tissue samples. Furthermore, MMP13 is reported to be highly expressed in early OA and at low levels in late OA. [43] , although the pathogenesis of OA is associated with increased MMP13 expression [44] . These reports are consistent with the results of our study. Although studies of MMP13 expression in the subchondral bone are rare, our results show high MMP13 expression in the subchondral bone.
Interleukin (IL)-1β is an inflammatory mediator that acts as a catabolic agent in arthritis.
Inflammation plays a crucial role in the development of OA, and IL-1β, which is highly expressed in OA patients [45] , plays an important pro-inflammatory role during inflammatory reactions. It also increases the secretion of major proteolytic enzymes in cartilage, including MMPs and A disintegrin and metalloproteinase with thrombospondin motifs (ADAMT-5) [46] . In addition, IL-1β induces NO production, and the enhances cyclooxygenase-2 (COX-2) activity, which ultimately leads to ECM degradation.
PTHrP was first discovered as a cancer-derived hormone, but it has since been shown to be an important factor in the development and maturation of cartilage and bone as well as an important regulator of endochondral bone formation and bone remodeling [47] . PTHrP is highly expressed in articular cartilage, and is induced by the lipoprotein receptor-related protein 6 (LRP6) 6 to inhibit the sclerostin SOST, which inhibits bone formation [48] . In this study, we show that the percentage of PTHrP-positive cells and expression intensity in cartilage and subchondral bone are increased, further indicating that PTHrP promotes bone formation. PTHrP is produced by chondrocytes to promote the proliferation of adjacent chondrocytes, and can delay its differentiation into hypertrophic chondrocytes.
When PTHrP cannot stimulate these cells, hypertrophic chondrocytes secrete IHH, which promotes the proliferation and differentiation of hypertrophic chondrocytes and accelerates the differentiation of adjacent chondrocytes into osteoblasts [49] . Accumulating evidence suggests the importance of the Hh signaling pathway that regulates the cartilage osteogenesis, especially the formation of IHH and SHH, in the formation of vertebrate bones [50] . IHH is a very important factor in the development of cartilage damage and OA.
Previous studies have shown that IHH is highly expressed in articular cartilage in OA mouse models and human OA patients and that high expression of IHH is related to the severity of OA [51] . Our results show high expression in both articular cartilage and subchondral bone, indicating that IHH upregulation is associated with the pathogenesis of OA.
SHH is a very important factor for embryonic and organ development, and plays an important role in bone formation in bone remodeling, regulating the osteogenic differentiation of bone marrow mesenchymal stem cells. SHH has been shown to be present and highly expressed in various cancers such as bladder cancer [52] , lung cancer [53] , and thyroid cancer [54] ; however, there are few studies in osteoarthritis. Our results showed that the expression of SHH protein in the cartilage and subchondral bone of the cartilage wear and cartilage exfoliation groups was significantly increased compared with that in the intact cartilage group (P < 0.05). Therefore, SHH is implicated as a potential influencing factor in the development of OA.
β-Catenin belongs to the intracellular glycoprotein family and is a key molecule in the classical Wnt pathway. It is widely distributed in the cell membrane, cytoplasm and nucleus and exerts various functions. Matrix metalloproteinases MMPs are highly expressed in articular cartilage of rats and rabbits and are activated by the Wnt/β-catenin signaling pathway [55,56] . Furthermore, β-catenin induces high expression of MMP-9, MMP-13 and BMP-2 genes in chondrocytes, which in turn, leads to OA production in mouse knee joints [57] . In accordance with previous reports, we show that high levels of β-catenin expression in the subchondral bone in OA patients. Typically, TGF-β/Smad signals through two main pathways; ALK5/Smad2/Smad3 and ALK1/SMAD1/SMAD5/SMAD. The TGFβ/ALK5/Smad2/Smad3 signal transduction pathway inhibits further differentiation of chondrocytes to maintain the normal phenotype [14] . The TGFβ/ALK1/SMAD1/SMAD5/SMAD8 signal transduction pathway induces the synthesis of MMP13, leading to cartilage matrix degradation, hypertrophy of chondrocytes and vascularization and calcification of articular cartilage [58] . Davidson [59] reported that the ALK1/ALK5 ratio was elevated in damaged articular cartilage in OA mice. Peter M [60] demonstrated that ALK 1 predominates and activates SMAD1/5/8 in OA in vitro, and the level of pSmad1 is increased in chondrocytes.
Similarly, in the present study, semi-quantitative immunohistochemical staining revealed a high expression of phosphorylated Smad1 in the cartilage and subchondral bone of the cartilage-destroyed group compared with that the intact cartilage group.
Bone remodeling is accomplished through tightly regulated coordination of bone formation and resorption mediated by osteoblasts that deposit calcified bone matrix and osteoclasts that resorb bone [61,62] . These processes do at specific anatomical sites and follow a welldefined sequence of events known as the bone remodeling cycle [63] . Studies have shown that active TGFβ1 acts as a pro-migratory factor involved in tissue damage and remodeling to induce MSC migration to normal and pathological tissues for repair and remodeling [64] . It has also been reported that in the process of osteolysis mediated by osteoclasts, one or multiple factors are released from the bone matrix, including transforming growth factor TGFβ1, which in the active form, functions as a coupling factor to induce migration of bone marrow MSCs to bone for the formation of new bone; however, excessive TGFβ1 activity leads to abnormal remodeling of subchondral bone [65] . In the pre-experiment, TRAP staining in the ACLT mouse model revealed an obvious increase in osteoclasts, and immunohistochemistry showed that increased levels pSmad2/3 (a downstream factor of TGFβ1), demonstrating that osteoclasts caused increased bone Some limitations of this study should be noted. First, relatively few tibial plateau specimens representing each cycle of OA in patients were available because knee replacement is performed late OA. Furthermore, a longitudinal study of cartilage and subchondral bone is required. Therefore, our findings may not reflect the subchondral bone pathology during disease progression. However, the advancement of imaging technology, such as micro-CT, may provide opportunities for non-invasive evaluation of patients that will facilitate longitudinal studies to identify new targets for OA diagnosis and treatment.

Conclusions
The findings of the present study show that abnormal changes in cartilage and subchondral bone are associated with high expression of TGF-β1. Furthermore, high levels  The authors claim that none of the material in the paper has been published or is under consideration for publication elsewhere.

Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests" in this section. [23] HENROTIN Y, PESESSE L, SANCHEZ C. Subchondral bone and osteoarthritis: biological and cellular aspects [J].

Inclusion criteria
The inclusion criteria were based on the diagnostic criteria for knee osteoarthritis developed by the International Osteoarthritis Research Association in 2014.
(1) The patient has a feeling of pain in the knee joint and has felt pain for most of the past month.
(2) In the medical examinations, X-ray imaging is characterized by narrowing and asymmetry of the joint space, sclerosis of the articular cartilage, and the presence of osteophytes. A cystic change can occur under the hardened articular cartilage.
(3) Patients must undergo two joint fluid examinations to rule out inflammatory arthritis.
The laboratory examination of the joint fluid must meet the OA standard: clear joint fluid, normal viscosity, and yellowish color. Inflammatory indicators such as erythrocyte sedimentation rate (ESR) and C-reactive protein are in the normal range.
(4) The activity of the knee joint can be basically maintained, and there is no manifestation of joint rigidity.
(6) Audible of palpable bone rubbing when the joint is active.

Exclusion criteria
Patients with OA who have a history of knee trauma, other forms of arthritis, metabolic bone disease, bone tumors, or drugs that affect bone remodeling were excluded. Not enough subchondral bone was included. The OA tibial plateau (the total thickness of the subchondral bone plate and subchondral trabecular bone <5 mm) was excluded from the study Figures Figure 1 General view of the tibial plateau. OA1 and OA2 are the lateral and medial sides of the tibial plateau, respectively.     Indicates statistical significance (P < 0.05).