ADAMTS4 or ADAMTS5: Which is the Key Enzyme in the Cartilage Degradation of Osteoarthritis and Kashin-Beck Disease?

Background: This study aims to investigate the altered expression of a disintegrin and metalloproteinase with thrombospondin motifs 4(cid:0)ADAMTS4(cid:0)and a disintegrin and metalloproteinase with thrombospondin motifs 5(cid:0)ADAMTS5(cid:0)in the human articular cartilage between osteoarthritis(OA) and Kashin-Beck disease(KBD) and compare their roles in the cartilage injury. Methods: Articular samples were collected from conrmed OA patients and KBD patients then divided into three groups, and the articular cartilages from the normal donors were used as controls. The morphology(cid:0)location and expression of ADAMTS4 and ADAMTS5 as well as aggrecan were detected by histochemical staining and immunohistochemical staining. Results: Compared to the control, the number of living cells in OA and KBD groups declined at three zones. Meanwhile, the results of toluidine blue staining showed that there was a loss of aggrecan in extracellular matrix of KBD and OA cartilages. The amounts of chondrocytes positively stained for ADAMTS4 were lower in the middle and deep zones of OA and KBD cartilages than those in the control samples. On the contrary, the immunostaining for ADAMTS5 signicantly increased in OA and KBD group compared with control group at all three zones. Notably, although there was no statistical signicance, the expression of ADAMST5 in KBD was slightly higher than the ones in OA in supercial, middle and deep zones. Conclusions: The role ADAMTS4 acted in aggrecan degradation was seemingly minor. It could be inferred that ADAMTS5 is the key enzyme in the cartilage destruction, particularly in KBD cartilage, which could explain the complex pathogenesis of KBD and provide a potential therapeutic target for KBD patients.


Background
Osteoarthritis(OA) is the most prevalent degenerative joint disease in the world. As the main cause of chronic pain and disability in the senior citizens, it is clinically characterized by stiffness, arthralgia, and limitation of movement (1). Kashin-Beck disease (KBD) is an endemic, chronic osteochondropathy, which can directly result in the joint deformity, dwar sm, depression and even disability (2), posing an increasing burden and challenge on society and individual. In China, this disease is widespread from northeast to the southwest, it can also be found in some regions of Eastern Siberia in Russia and in North Korea. Three hypotheses have been formulated to explain the occurrence of KBD: 1) an endemic de ciency of Selenium, 2) a cereal contamination by mycotoxins and 3) high humic acid levels in drinking water (3)(4)(5).
Despite extensive investigations, none of these has been veri ed by epidemiological and experimental evidences. Thus, KBD is still regarded to have a multi-factorial etiology and complex pathogenesis.
Through reviewing previous studies, a conclusion can be drawed that KBD shares its clinical symptoms and pathology with OA, particularly at earlier stages of the disease, although gene expression and pathogenesis of KBD have been indicated to differ from OA(6, 7). For example, painful joints, narrowed joint space, osteophytes, and movement restrictions of joints can be observed in the cartilage from both OA and KBD patients. Some pathological changes, such as predominant degradation of extracellular matrix (ECM) in affected articular cartilage, apoptosis and necrosis of chondrocytes overlap in OA and KBD. Given this, KBD should be undoubtedly viewed as a speci c type of OA.
The imbalance of the synthesis and degradation of ECM is one of the important causes in the cartilage destruction. The chondrocytes maintain the cartilage structure and the homeostasis of the cellular environment by synthetizing and catabolizing ECM macromolecules. The amount of cartilage matrix will decline, if the e ciency of degradation of ECM surpass their synthesis, and eventually leads to the destruction of cartilage(8). Aggrecan is the important component of ECM, imparting the compressive ability for cartilage (9). The depletion of aggrecan is a hallmark event happening in the advanced stage of OA, which is mediated by aggrecanases. Aggrecanases are the principal proteinases, present in the articular cartilage and growth plate cartilage, which belong to the a disintegrin and metalloproteinase with thrombospondin motifs (ADAMTS) family. As the indispensable members of ADAMTS, recent studies have indicated that ADAMTS4 and ADAMTS5 are responsible for the degradation of aggrecan in human OA models (10)(11)(12). It is plausible that ADAMTS4 and ADAMTS5 make major contributions to the loss of proteoglycans during the progression of KBD.
In this study, the presences of ADAMTS4, ADAMTS5 and aggrecan were tested by immunohistochemical stainings of cartilage from the patients of KBD and OA, as well as the normal group. The comparison between OA, KBD and the normal groups could not only clarify which one of ADAMTS4 or ADAMTS5 is the key aggrecanase in cartilage depletion, but also explore the roles of these aggrecanases in the pathogenesis of KBD, which will be helpful to illustrate the etiology of KBD and put forward new potential therapeutic targets for KBD.

Materials And Methods
Samples and materials OA patients (n=9, age: 45-65 years old, sex ratio: M/F=4/5) and KBD patient(n=11, age: 50-63years, sex ratio: M/F=8/3) were primarily enrolled from those who underwent surgical treatment. All KBD patients were diagnosed and con rmed according to Chinese diagnosis criterion (13). The normal cartilage samples were collected from the patients who suffered from tra c accidents or polydactyly treatment surgically (n=12, age: 37-70years, sex ratio: M/F=5/7). All normal samples are from donors living in non-KBD endemic areas. The detailed information about subjects is shown in the Table 1.
This study was permitted by medical research permission granted by the Human Ethical Committee in Xi'an Jiaotong University. All the participants were informed of this studies, after which they provided a written informed consent.

Immunohistochemical staining
These cartilage tissues xed in the 4% paraformaldehyde were decalci ed, embedded in para n within 8h after surgery, then cut in coronary direction into 5μm thickness after dehydration, and prepared for the applications described below. For depara nization, 5μm sections were immersed in xylene and rehydrated with a series of gradually decreasing concentrations of ethanol(100%-80%), then incubated with 3% hydrogen peroxide H 2 O 2 for 15 min at room temperature. After rinsing with phosphate-buffered saline (PBS) for three times, the sections were incubated 2 mg/ml hyaluronidase for 20 min. After incubation in blocking buffer (the normal goat serum, from Zhong Shan Gold Bridge Rabbit SP reagent, Beijing Zhong Gold Bridge Biological Technology Co., 18112A11) for 20 min, the samples were incubated with the primary antibodies of ADAMTS4 at 1:200 dilutions and ADAMTS5 at 1:50 dilutions in 1% bovine serum albumin in PBS at 4 °C overnight. The primary antibody of negative control was replaced by PBS.
Then the sections were incubated with horseradish peroxidase-conjugated goat anti-rabbit IgG at 37°C for 20 min. After rinsing in PBS, hematoxylin counterstaining was performed. For visualization, a diaminobenzidine (DAB) kit (ZLI-9017, ZSGB-BIO, China) was applied. Finally, dehydration, clearing and sealing coverslips were performed.

Hematoxylin and eosin (H&E) and toluidine blue (TB) staining
For hematoxylin-eosin (H&E) and toluidine blue (TB) stainings the depara nization followed the procedure used for immunohistochemical samples. Then sections were stained with H&E or 0.1% (w/v) TB dye solution. After washing with running water, dehydrated by a series of gradually increasing concentrations of ethanol solution (80-100%) and xylene, then samples were sealed by coverslips.

Image capture and quanti cation
The stained section images were acquired by a Leica SCN400 slide scanner (Leica Microsystems GmbH, Wetzlar, Germany) under a 100× magni cation eld, then three representative regions were randomly selected from super cial, middle and deep zones were quanti ed by Image J software (NIH, USA). The positive staining rate was calculated as follows: The positive staining cell rate = Statistical analysis SPSS 22.0 software (SPSS Inc, USA) was applied for the data entry and statistical analysis. If the data from OA, KBD and control groups were in accordance with normal distribution by the test for normality and equal variance, the rank-based ANOVA test was used to examine the statistical discrepancy of positive staining cell rate among OA, KBD and control groups. If not, nonparametric rank sum test was used, and all statistical tests were bilateral tests. P 0.05 was regarded for statistically signi cant difference.

H&E staining
H&E staining manifested that the quantity of living cells of OA (super cial zone, P 0.001, middle zone, P=0.001, deep zones, P=0.003) and KBD(super cial zone, P 0.001, middle zone, P=0.004, deep zones, P 0.001) declined evidently among three zones compared to the control group. This suggested that the cartilage may undergo a focal chondrocyte death (necrosis), which is associated with proteoglycans depletion. In the super cial zones, the number of living cells of OA were less than KBD(P=0.036) (Fig.1.).

IHC staining
Theoretically, the expression aggrecanases should increase in OA and KBD cartilage. However, there were no difference ADAMTS4 stainings in the super cial zone among the three groups. In the middle zone, the positive staining rate of ADAMTS4 in OA cartilage reduced compared to control one(P=0.001). Similarly, this kind of decrease also happened in the deep zone in both OA and KBD cartilage (P=0.008, P<0.001). As for the comparison of expression of ADAMTS4 between KBD and OA cartilage, the positive staining rates in KBD group had a rise when compared to OA in both super cial and middle zones, wherein the statistically difference were observed in the middle zones(P=0.009). In the deep zone, condition changed: as the number of positive staining cells in KBD were lower than in OA(Fig3.).
The immunostaining of ADAMTS5 were much stronger in OA and KBD group than those in control group in the super cial zone (P 0.001, P 0.001). Meanwhile, the positive staining cells in the middle zones of OA and KBD also had an obviously increase compared to the healthy cartilage (P=0.003, P 0.001). In the deep zone, the distribution of positive staining cells in OA and KBD were more intensive than in the control (P 0.001, P 0.001). Totally, there was a tendency that the expression of ADAMTS5 in KBD was slightly higher than OA among all three zones in cartilage, although without statistical difference (Fig4.).

Discussion
Apoptosis, necrosis, chondrocytes aging and loss of proteoglycans were universally acknowledged as the primary features of cartilage injure (17)(18)(19)(20)(21), As direct consequences of these, the density of the living cells in the cartilage decreases. The results of H&E staining veri ed that such a phenomenon really was obvious in the cartilage of OA and KBD.
Proteoglycans are complex macromolecules distributing in the ECM of articular cartilage, having a protective effect on joints. Dominant proteoglycans in cartilage are aggrecan (22). Increasing number of research has demonstrated that there is a close association between loss of aggrecan and cartilage diseases (23)(24)(25).It could be observed in the TB staining that the staining intensity of proteoglycans in KBD was quite low compared to OA group and control group, which showed that the depletion of aggrecan is more serious in cartilage of KBD.
It is generally accepted that two families of metalloproteases are the major contributors in the progression of degradation of ECM: matrix metalloproteinase (MMP) family and ADAMTS family(26). However, in comparison to the MMPs, aggrecanases seemingly have more association with the depletion of aggrecan, because it was the greatest in areas of cartilage adjacent to sites of cartilage erosion (25).
When referring to the roles of ADAMTS family played in the degradation of aggrecan, researchers often pay more attention to the ADAMTS4 and ADAMTS5, since other members of family related to degradation of aggrecan, such as ADAMTS1, ADAMTS8, ADAMTS9, has been shown to have a rather low aggrecan-degrading activity (27,28). ADAMTS4 and ADAMTS5 leave the aggrecan core protein at the aggrecanase-speci c Glu373-Ala374 bond in the interglobular domain (IGD) region (29,30), therefore, they have been regarded as the key aggrecanase. ADAMTS4 and ADAMTS5 have similar structure, containing a disintegrin domain, a thrombospondin domain, a cysteine-rich domain, and a spacer domain, but ADAMTS5 has an additional TS domain after the spacer domain. A study reported that in the IGD region the aggrecanolytic activity of ADAMTS5 are fourfold compared to ADAMTS4, in the CS-2 region of aggrecan 2.5 fold, and under physiological conditions the aggrecanase activity of ADAMTS5 was at least 1000-fold greater than that of ADAMTS4 (31,32).
Based on the research of activity and molecular basis of ADAMTS4 and ADAMTS5, advanced works from lots of laboratories have further illustrated the impact of aggrecanase on increase of aggrecan loss related to cartilage diseases in different ways (33)(34)(35)(36). Although there are some reviews comprehensively summarizing the regulation of ADAMTS4 and ADAMTS5 in OA(26, [37][38][39], it is still controversial that which of them is the main aggrecanaes in the destruction of cartilage. Some reports argue that ADAMTS4 is primarily expressed in an active form in osteoarthritic cartilage, while ADAMTS5 would be constitutively expressed in both normal and OA cartilage (40).
Others suggested that owing to the high a nity of its non-catalytic domains for glycosaminoglycan chains, ADAMTS5 is the most active aggrecanase to increase the loss of aggrecan (32). It has also been proposed that ADAMTS4 is the most inducible aggrecanase upon cytokine stimulation, whereas ADAMTS5 is the most abundant aggrecanase (41), which manifests that ADAMTS5 is constitutive. In the murine models of OA, ADAMTS5 de ciency could protect mice from OA, but deletion of ADAMTS4 didn't have any in uence on the normal growth and physiology of mice (42,43). As discussed above, ADAMTS5 appears to play the key role in the cartilage depletion, though these conclusions are limited to animal model, and the discrepancy between human and murine should not been ignored. Therefore, aims of this study not only investigated the main aggrecanases involving the cartilage degradation, but also explored whether ADAMST4 and ADAMTS5 play the equivalent roles in the progression of KBD.
In our study, the expression of ADAMTS4 is not in accordance with the points of some previous studies: the percentage of positive staining of the control was more intensive than OA and KBD in the middle and the deep zones. Generally speaking, ADAMTS4 should be overexpressed in the damaged cartilage.
However, the mRNA expression of ADAMTS4 was signi cantly repressed at the early stage of OA (44). Besides, in the super cial zone, the expressions of ADAMTS4 of OA and KBD had a similar proportion with the control. These results indicated that ADAMTS4-induced cartilage degradation maybe not happened at the early stage of cartilage diseases, and the role of ADAMTS4 is less signi cant, especially in the super cial zone.
Nevertheless, the expression of ADAMTS5 in OA and KBD were greatly increased compared to the control, particularly in the cartilages from KBD patients. In addition, there was gradual upregulation of ADAMTS5 in the OA group from super cial to deep zone, while the expression in KBD group down-regulated in the deep zone. It con rmed that ADAMTS5 was the crucial aggrecanase, which led to the cartilage damage. This adverse effect is evidence in the deep zone of OA cartilage. What's more, the expression of ADAMTS5 in KBD group was higher than the one in OA group. Based on the results of TB staining of the aggrecan positive cell rates among three groups, it could be inferred that the increased expression of ADAMTS5 is more likely to cause the decrease of aggrecan in OA and KBD.
However, mainly owing to the ethical issue, it was hard to collect enough cartilage from KBD patients. It couldn't manifest adequately the casual relationships between cartilage degradation and ADAMTS5. In order to investigate the role of ADAMTS5 in the cartilage disease, increasing amount of research is required, which could further explain the complex pathogenesis of KBD and provide a potential therapeutic target for KBD patients.
In conclusion, we found that the function of ADAMTS4 in degradation aggrecan is seemingly slight.
Hence, we prefer to favor of the opinion that ADAMTS5 is the key enzyme in the cartilage destruction, in particularly in KBD cartilage. Most of previous studies have focused on the roles of ADAMTS4 and ADAMTS5 in OA, while it is rare to investigate the importance of ADAMTS4 and ADAMTS5 in KBD. The test of the expression of ADAMTS4 and ADAMTS5 in OA and KBD cartilage further a rmed our standpoint that ADAMTS5 was more signi cant in the loss of aggrecan and cartilage damage.

Declarations
Funding This research was supported by the national natural science fund (No: 81620108026).
Con icts of interest/Competing interests The authors declare no con ict of interest. The funders had no role in the design of the study, in the collection, analyses, or interpretation of data, in the writing of the manuscript, or in the decision to publish the results.
Ethics approval and consent to participate The experimental protocol was established, according to the ethical guidelines of the Helsinki Declaration and was approved by the Human Ethics Committee of Xi'an Jiaotong University. Written informed consent was obtained from individual or guardian participants.
Consent for publication Not applicable.
Availability of data and material All data generated or analysed during this study are included in this