Expression and role of Sphingosine 1-phosphate receptors in intervertebral disc degeneration

Abstract

BACKGROUND:

IVD degeneration is a widespread problem all over the world, which a variety of inflammatory cytokines have been implicated in, while Sphingosine 1-phosphate (S1P) is an important lipid mediator that may play a role in IVD degeneration.

OBJECTIVE:

To study the expression and role of S1PRs in the intervertebal disc (IVD) degeneration to enhance understanding of disc degeneration.

METHODS:

Degenerated and normal IVD were harvested from patients through surgery. Expression of S1P receptor subtypes was evaluated using real-time PCR, immunohistochemistry, and western blotting. The effect of S1PR on inflammation induced by interleukin-1 $\beta$ in nucleus pulposus (NP) cells was also assessed by real time PCR and western blotting.

RESULTS:

The nucleus pulposus mainly expressed the S1PR1/2/3, and the expression decreased in the severe degenerated nucleus pulposus cells. The ligand, S1P, inhibited the up-regulation of matrix metallopeptidase-3 (MMP-3) and ADAM metallopeptidase with thrombospondin type 1 motif 4 (ADAMTS4) induced by IL-1 $\beta$ .

CONCLUSIONS:

The results show that an the expression of S1PRs in degenerative discs is down-regulated as degeneration, and S1P can inhibit the inflammation response induced by IL-1 $\beta$ in NP cells, implicating that S1P/S1PR may contribute to IVD degeneration.

Keywords

Sphingosine 1-phosphate receptors nucleus pulposus cells intervertebral disc degeneration IL-1β

1. Introduction

Chronic low back pain (LBP) is a widespread problem all over the world, which affects about 50 to 80% of adults during their lifetime [1]. Numerous studies have shown that intervertebral (IVD) degeneration is one of the major underlying factors in chronic LBP [2, 3]. Various changes occur with the degeneration, such as fissures in the annulus fibrosus, and dehydration in the nucleus pulposus (NP). Alterations in the extracellular matrix and expression profile of the NP cells, including decreases in proteoglycan and type II collagen synthesis, increases in type I collagen synthesis and aggrecan fragment accumulation, is the major pathogenic characterization of IVD degeneration [4].

Furthermore, a variety of inflammatory cytokines have been implicated in IVD degeneration [5, 6, 7]. Previous studies have demonstrated that IL-1 $\beta$ and TNF- $\alpha$ are overexpressed in degenerated IVD, and led to matrix degradation in degenerated disc [8, 9]. Thus, inhibition of the effect of these cytokines may be a target for therapeutic intervention.

Sphingosine 1-phosphate (S1P) is an important lipid mediator, formed by phosphorylation of sphingosine and catalyzed by sphingosine kinase. It has shown to be implicated in many biological processes, including cell migration, differentiation, inflammation and angiogenesis [10, 11, 12]. S1P exerts its various functions by binding to specific G protein-coupled receptors, and 5 functionally different isoforms (termed S1PR 1-5) have been identified. Strikingly, previous research have shown that in human chondrocytes, S1P can inhibit the catabolic response induced via activation of the S1P receptor [13, 14], thus the activation of the receptor may be a therapeutic target for OA. In light of similar biological property of nucleus pulposus cells, S1P and its receptors reflect a certain application prospect in the treatment and control of intervertebral degeneration. To the best of our knowledge, the expression of S1P receptors or the effect of S1P in IVD degeneration have not been explored.

In the current study, we investigated the expression of S1P receptors in nucleus pulposus, and compared the expression in NP cells with different degrees of degeneration in order to verify the relation between the S1PRs and IVD degeneration. Finally, we preliminary explored the effects of S1P on IL-1 $\beta$ induced inflammation response in NP cells.

2. Materials and methods

2.1 Patients and samples

Thirty-four patients with lumbar degeneration (lumbar herniation or spondylolisthesis) and 5 patients with vertebral fracture were selected for this study. Study protocols were approved by the Ethics Committee of our institution, and the patients provided informed consent forms. Patients with an infection, tumor, immunological and endocrine disease were excluded from the current study.

The patients with degeneration were 51.17 $\pm$ 11.77 years old, ranging from 24 to 68. The group comprised of 20 males and 14 females. The patients with vertebral fracture were 19.60 $\pm$ 1.14 years old, ranging from 18 to 21, and had no documented clinical history of low back pain. As controls 2 males and 1 female were included.

2.2 Histological grading of IVD tissues

To histologically assess the grade of degenerated IVD, samples obtained from the surgical operation were fixed with formalin and embedded with paraffin. Five microns sections of the tissue were acquired and processed into HE staining and Saf-O staining. The degree of degeneration was graded according to the previously published histological scoring system [15]. Two readers independently graded each of the lumbar IVDs. Consensus was reached when the initial reading of the two investigators differed.

2.3 Immunohistochemistry

Paraffin sections (5 $\mu$ m thick) were treated with xylene to remove paraffin and were rehydrated in graded alcohol baths followed by three rinses with phosphate-buffered saline (PBS). Slides were immunostained with the streptavidin-biotin peroxidase (SABC) technique. The primary antibodies of S1PR1 (Santa Cruz, sc-48356), S1PR2 (Santa Cruz, sc-25491) and S1PR3 (Santa Cruz, sc-30024) 1:50 dilution were used to stain.

Antigen retrieval was performed by heating the sections in 10 mmol/L citrate buffer (pH 6.0) up to 95 ${}^{\circ}$ C for 10 min and allowing them to cool down to room temperature for 20 min. Then, the sections were incubated in 1% H2O2 for 15 min to block the endogenous peroxidase activity. After pre-incubation with 5% normal goat serum for 30 min at room temperature, the sections were incubated overnight at 4 ${}^{\circ}$ C room temperature with rabbit polyclonal antibody of S1PR1/2/3. Then, the sections were incubated with the corresponding biotinylated secondary antibody, applied for 30 min at a dilution of 1:200, followed by a triple wash in PBS. Finally, the sections were incubated in ABC complex for 30 min at room temperature. Staining was visualized with DAB peroxides substrate solution for 3 min, followed by a brief rinse in distilled water. The slides were dehydrated in graded ethanol, cleared in xylene, and mounted with cover glass after counterstaining with hematoxylin solution for 3 min.

Figure 1.

HE and Saf-O stain of the intervertebral disc: normal NP shows abundant extracellular matrix and stained red evenly, small-size colonies in mild degenerated NP and less stained red, severe degenerated NP shows huge stained and stained red mainly in the cell colonies, bar 100 $\mu$ m.

Figure 2.

The expression of S1PR in NP was shown by immunohistochemistry. S1PR1, S1PR2 and S1PR3 are all positively expressed in NP tissues, and it is down-regulated with degeneration, bar 100 $\mu$ m.

2.4 Nucleus pulposus (NP) cell isolation and culture

The IVD tissues were harvested from three patients through surgery and transported to the laboratory within 30 min. NP tissue were carefully picked and digested in a 0.5% type II collagenase (Sigma-Aldrich, Saint Louis, MO, USA) solution in serum-free medium. The digested tissue/cell suspension was passed through a 100 $\mu$ m cell strainer to remove tissue debris, and cells were then collected by centrifugation at 1200 rpm for 5 minutes. The supernatant was removed, and cells were resuspended and cultured to confluence in a 25-cm ${}^{2}$ flask with Dulbecco’s Modified Eagle Medium: Nutrient Mixture F-12 (DMEM-F12), supplemented with 10% Fetal Bovine Serum (Gibco, Waltham, MA, USA), 1% penicillin-streptomycin (PS), at 37 ${}^{\circ}$ C in a humid atmosphere containing 5% CO ${}_{2}$ , with medium changed every 3 days. The cells were subcultured when confluence was 75–80%. The NP cells for subsequent experiment were used at passage 1 or 2.

2.5 Cell treatment

To explore the role of S1PRs in human NP cells, the cells were stimulated with ligand, Sphingosine 1-phosphate (S1P), and the effect was assessed by subsequent real-time PCR. Specific information is as follows: firstly, S1P (Sigma-Aldrich) was dissolved in methanol, evaporated, and then resuspended in 0.4% fatty acid-free bovine serum albumin. Recombinant human IL-1 $\beta$ (10 ng/ml; Sigma-Aldrich) was dissolved in the DMEM-F12 culture medium. The cells were treated with the IL-1 $\beta$ solution for 24 hours with or without 5 $\mu$ mol/ml S1P.

2.6 Real-time PCR

Total RNA was extracted from the NP cells by Trizol (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. The mRNA was analyzed with real-time polymerase chain reaction using a real-time polymerase chain reaction system (Geneline9640, Bioer, Hangzhou, China). The cDNA was then reverse transcribed (model R0037A, TaKaRa, Dalian, China) according to the manufacturer’s instructions. Real-time polymerase chain reaction reactions were done in triplicate in 96-well plates in a final volume of 20 $\mu$ L, including 10 $\mu$ L of Sybr Green, 2 $\mu$ L of cDNA, 1 $\mu$ L of each primer, and 6 $\mu$ L of sterile distilled water. All primers (shown in Table 2; obtained from Sangon, Shanghai, China) were designed on the basis of coding sequences. The cycle threshold values were obtained and data were normalized to GADPH expression using the 2 ${}^{-\Delta\Delta\text{Ct}}$ method.

2.7 Western blotting

The total protein was extracted using RIPA supplemented with 1% protease inhibitor cocktail (both from Sigma-Aldrich). Four independent experiments using chondrocytes derived from different patients were performed. The protein concentration was measured with BCA method (Bio-Rad). Proteins were separated by SPS-PAGE electrophoresis using a 10% polyacrylamide gel, and then transferred to a nitrocellulose membrane (Bio-Rad). After blocking in 5% skim milk-Tris buffered saline, the membranes were incubated with primary antibodies overnight. Thereafter, the membranes were rinsed in blocking solution and incubated for 1 hour with a secondary antibody conjugated to horseradish peroxidase. Bands were visualized using an acridan-based substrate detection system (ECL Plus, Amersham, UK) and protein levels were relatively quantified by densitometry using ImageJ.

2.8 Statistical analysis

Histological staining and western blot data are described qualitatively. The data of Real-PCR is shown as means $\pm$ standard deviation. In accordance with the data distribution and the homogeneity of variance, one-way ANOVA or Kruskal-Wallis H-test were selectively utilized. The $P$ value of less than 0.05 was considered statistically significant.

Table 1
The scores of histological grading of the IVD tissues harvested from patients through surgery

Code	Age	IVD	Diagnosis	Histological
		segment		grading
1	51	L4/5	LDH	9
2	67	L4/5	LDH	10
3	57	L5/S1	LDH and spondylolisthesis	10
4	24	L5/S1	LDH	7
5	61	L5/S1	LDH and spondylolisthesis	10
6	61	L4/5	LDH	10
7	66	L4/5	LDH and spondylolisthesis	12
8	49	L4/5	LDH	8
9	67	L4/5	LDH	10
10	60	L5/S1	LDH and spondylolisthesis	10
11	44	L4/5	LDH	8
12	46	L5/S1	LDH	7
13	49	L4/5	LDH	8
14	65	L5/S1	LDH and spondylolisthesis	9
15	68	L5/S1	LDH and spondylolisthesis	6
16	40	L4/5	LDH	7
17	42	L5/S1	LDH	6
18	20	L2/3	Vertebral fracture	2
19	47	L4/5	LDH	7
20	40	L5/S1	LDH	7
21	39	L5/S1	LDH	9
22	47	L4/5	LDH	9
23	21	L2/3	Vertebral fracture	2
24	57	L5/S1	LDH	10
25	51	L5/S1	LDH	9
26	47	L5/S1	LDH	9
27	41	L5/S1	LDH	7
28	18	L2/3	Vertebral fracture	2
29	34	L4/5	LDH	8
30	31	L4/5	LDH	7
31	64	L4/5	LDH	10
32	39	L4/5	LDH	7
33	67	L4/5	LDH	12
34	41	L4/5	LDH	6
35	61	L4/5	LDH	11
36	57	L4/5	LDH and spondylolisthesis	10
37	60	L4/5	LDH	10
38	19	L1/2	Vertebral fracture	3
39	20	L2/3	Vertebral fracture	2

Figure 3.

Expression of 3 types of S1PRs (S1PR1/2/3) in NP cells with different degrees of degeneration were measured by real-time PCR and western blotting. The typical grayscale of western blot from some samples was also shown (B). They all decrease in the severe degenerated cells. The relative expression to the severe degenerated cells is shown as mean $\pm$ SEM (A and C) (N, normal; MD, mild degenerated; SD, severe degenerated), ${}^{*}$ means $P<$ 0.05.

Figure 4.

Expression of MMP-3 and ADAMTS4 in NP cells treated with IL-1 $\beta$ in presence or absence of S1P was measured by real-time PCR and western blotting. S1P significantly inhibited the up-regulation of MMP-3 and ADAMTS4 induced by IL-1 $\beta$ . The effect of S1P on NP cell may be mediated by the NF- $\kappa$ B p65. The relative gene expression to the control is shown as mean $\pm$ SEM (A), calculated by the 2 ${}^{-\Delta\Delta\text{Ct}}$ method. ${}^{*}$ means $P<$ 0.05; The grayscale of western blotting is also shown (B).

3. Results

3.1 Evaluation of the degeneration of IVD samples

In order to evaluate the grade of degeneration in NP cells, histological grading was proceeded by HE and Saf-O staining. The scores of histological grading of the IVD tissues harvested from patients through surgery is shown in Table 1. The representative HE and SAF-O staining of NP tissue with different levels of degeneration is shown in Fig. 1. Overall, with the progress of the degeneration, the NP cells get together into colonies and the extracellular matrix loose.

Table 2
Primer sequences for real-time PCR

Gene	Primer sequences
S1PR1	Forward TATCAGCGCGGACAAGGAGAACAG
	Reverse ATAGGCAGGCCACCCAGGATGAG
S1PR2	Forward CATCGTGCTAGGCGTCTTTA
	Reverse GGCATAGTCCAGAAGGAGGA
S1PR3	Forward CTGCCTGCACAATCTCCCTGACTG
	Reverse GGCCCGCCGCATCTCCT
MMP-3	Forward CTCGTTGCTGCTCATGAAAT
	Reverse GAACCGAGTCAGGTCTGTGA
ADAMTS4	Forward TTTGTGGAGACACTGGTGGT
	Reverse ACCACCAAGCTGACAGGATT
GADPH	Forward TGGTATCGTGGAAGGACTCA
	Reverse CCAGTAGAGGCAGGGATGAT

According to the grading system [15], which generates a score between 0 and 12, we defined that a grade of 0 to 4 represents histologically normal degeneration (N), grades 5 to 8 indicate mild degeneration (MD), and grades 9 to 12 severe degeneration (SD). Then all the samples were divided into three groups, the N group included 5 samples, the MD group included 12 samples, and the SD group included 22 samples.

3.2 Expression of S1PRs in NP tissues

To investigate the distribution of S1P receptors in IVD tissue, we performed immunohistochemical staining. S1PR1/2/3 was detected on the IVD sections (Fig. 2). Both cytoplasmic and cytomembrane staining of S1PR1/2/3 immunoreactivity were observed in the nucleus pulposus cells. Then we correlated the immune-expression with degree of degeneration, it seems down-regulated during disease progression.

3.3 Expression of S1PRs in NP cells with different levels of degeneration

To confirm the trend shown in immunohistochem, we used real-time PCR to compare the gene expression of S1PR1/2/3 in human NP cells with different levels of degeneration (Fig. 3). The results confirmed that human degenerative NP cells demonstrated a significant decrease in S1PR1/2 expression at mRNA level in severe degenerated NP cells, S1PR3 was also decreased but not significantly. Western blotting shows that the decrease tendency is the same in the protein level, which the expression of S1PR1/2/3 decreases in severe degenerated NP cells.

3.4 The effect of the ligand (S1P) in NP cells

Previous studies have confirmed that in human chondrocytes S1P can reduce the induction of catabolic genes in the presence of IL-1 $\beta$ via activation of the S1P receptor 2(S1PR2). Thus we simulated the same pattern (10 ng/ml IL-1 $\beta$ with or without 5 $\mu$ mol/ml S1P) on the NP cell isolated from mild degenerated IVD, and observed that S1P counteracted the IL-1 $\beta$ induced up-regulation of ADAMTS-4 and MMP-3 after treatment for 24 hours by real-time PCR (Fig. 4). The suppressive effect of S1P on IL-1 $\beta$ is also verified by western blotting in the protein level. Furthermore, we tested the signal pathway, which showed that S1P inhibits the phosphorylation of NF- $\kappa$ B P65 caused by IL-1 $\beta$ .

4. Discussion

The results of this study demonstrate for the first time that S1PR1/2/3 are expressed in the human nucleus pulposus, and it is down-regulated with the intervertebral disc degeneration. The activation of the receptor (stimulated by the ligand, S1P) inhibits the up-regulation of MMMP-3 and ADAMTS4 induced by IL-1 $\beta$ in NP cells.

Even though we do not know whether there is S1P in IVD tissue, the presence in synovial fluid, blood plasma and cerebrospinal fluid has been confirmed [16, 17]. With the development of IVD degeneration, blood vessel infiltrated the nucleus pulposus tissue [18], the plasma component may affect the metabolism of the NP cells. Furthermore, immune cells, such as macrophages, neutrophils and T cells, can be recruited to degenerated IVD [19], thus as an important lipid mediator, S1P probably participates in the inflammatory response.

NP cells showed a different pattern of gene expression with degeneration [20, 21] and these variations may lead to distinct cells function, and interact with the degeneration. Here we compared the expression of S1PR1/2/3 in NP cells with different levels of degeneration, and found that the expression of S1PR1/2/3 all decrease with the development of degeneration. The change of S1PR2 is in agreement with a previous study [13], while S1PR1 and S1PR3 is not. The disagreement may be due to the difference between NP cells and chondrocytes, or the different experimental method. This result may be helpful to assess the risk of recurrence after surgery.

In addition to the change of the expression of S1PRs, we confirmed the anti-inflammatory effect of S1P. This is also in line with previous studies in chondrocytes [13, 14]. Together with the decrease of S1PRs, this result may explain the accumulation of matrix degrading enzymes, such as MMP-3 and ADAMTS4 in IVD degeneration. Here we test the effect in NP cells from IVD with mild degeneration, which express S1PRs representatively. As the angiogenesis and tissue defect along with IVD degeneration, the level of S1P in IVD may increase; while the S1PRs is down-regulated, the effect of S1P/S1PR in severely degenerated IVD is doubtful, which needs more research.

Other effects of S1P in chondrocyte have also been reported [11, 22]. Notably, Masuko’s study drew the conclusion that S1P attenuates proteoglycan aggrecan expression via production of prostaglandin E2 from human articular chondrocytes and may thus promote cartilage degradation. Furthermore, S1P is also involved in spinal nociceptive processing [17]. Taken this into account, the down-regulation of S1PRs may be advantageous for the nucleus pulposus cells to adapt to the deterioration of internal environment caused by degeneration.

The biological action of S1P is largely ascribed to specific S1PRs that may evoke distinct biological responses. The key receptor of the anti-inflammatory effect in NP cells has not been studied. This has been controversial even in chondrocytes. Moon’s study shows that S1PR1 is involved in the chondrocyte anti-inflammatory effect of S1P [14]; while Stradner’s study drew the conclusion that S1P reduces the induction of catabolic genes in the presence of IL-1 $\beta$ via S1PR2. The latter conducted not only a pharmaceutical experiment but also siRNA of S1PRS, which is more dependable. Furthermore, the exact signaling pathway is also doubtful. We verified that NF- $\kappa$ B p65 may participate that effect in NP cells, but further study is needed.

In the current study, we investigated the role of S1P and its receptors in intervertebral degeneration. We drew the conclusion that human nucleus pulposus cells mainly express S1PR1/2/3, and the expression of decreased when severe degeneration appears. S1P can inhibit the inflammation response induced by IL-1 $\beta$ via the receptors, while further research is needed to study the key receptor in the effect and the signaling pathway related. These results add new aspects to our understanding of intervertebral degeneration and may have therapeutic implications in light of the novel class of S1P agonist drugs currently being developed.

Footnotes

Acknowledgments

We would like to thank Professor Yue Zhou (Department of Orthopedics, Xinqiao Hospital, The Third Military Medical University, China) for support of the study.

Conflict of interest

None of the authors have any conflicts of interest related to this manuscript.

References

McBeth

and Jones

. 2007. Epidemiology of chronic musculoskeletal pain. Best Practice & Research. Clinical Rheumatology 21, 403-25.

Goode

Carey

and Jordan

. 2013. Low back pain and lumbar spine osteoarthritis: how are they related? Current Rheumatology Reports 15, 305.

Livshits

Popham

Malkin

Sambrook

Macgregor

Spector

and Williams

. 2011. Lumbar disc degeneration and genetic factors are the main risk factors for low back pain in women: the UK Twin Spine Study. Annals of the Rheumatic Diseases 70, 1740-5.

Anderson

and Tannoury

. 2005. Molecular pathogenic factors in symptomatic disc degeneration. The Spine Journal: Official Journal of the North American Spine Society 5, 260s-6s.

Le Maitre

Pockert

Buttle

Freemont

and Hoyland

. 2007. Matrix synthesis and degradation in human intervertebral disc degeneration. Biochemical Society Transactions 35, 652-5.

Hartman

Yurube

Jacobs

Sowa

and Kang

. 2013. Expression and regulation of metalloproteinases and their inhibitors in intervertebral disc aging and degeneration. The Spine Journal: Official Journal of the North American Spine Society 13, 331-41.

Phillips

Cullen

Chiverton

Michael

Cole

Breakwell

Haddock

Bunning

Cross

and Le Maitre

. 2015. Potential roles of cytokines and chemokines in human intervertebral disc degeneration: interleukin-1 is a master regulator of catabolic processes. Osteoarthritis and Cartilage/OARS, Osteoarthritis Research Society 23, 1165-77.

Seguin

Pilliar

Roughley

and Kandel

. 2005. Tumor necrosis factor-alpha modulates matrix production and catabolism in nucleus pulposus tissue. Spine (Phila Pa 1976) 30, 1940-8.

Le Maitre

Hoyland

and Freemont

. 2007. Catabolic cytokine expression in degenerate and herniated human intervertebral discs: IL-1beta and TNFalpha expression profile. Arthritis Research & Therapy 9, R77.

10.

Yang

Chang

Liu

Han

Zhu

Yang

and Li

. 2012. Bone marrow-derived mesenchymal stem cells differentiate to hepatic myofibroblasts by transforming growth factor-beta1 via sphingosine kinase/sphingosine 1-phosphate (S1P)/S1P receptor axis. The American Journal of Pathology 181, 85-97.

11.

Masuko

Murata

Beppu

Nakamura

Kato

and Yudoh

. 2012. Sphingosine-1-phosphate modulates expression of vascular endothelial growth factor in human articular chondrocytes: a possible new role in arthritis. International Journal of Rheumatic Diseases 15, 366-73.

12.

Ogawa

Takahashi

Hirose

Morimoto

Ise

Murakami

Yasue

Kuriyama

Hongo

Kobayashi

and Ikeda

. 2007. A novel sphingosine-1-phosphate receptor agonist KRP-203 attenuates rat autoimmune myocarditis. Biochemical and Biophysical Research Communications 361, 621-8.

13.

Stradner

Gruber

Angerer

Huber

Setznagl

Kremser

Moazedi-Furst

Windhager

and Graninger

. 2013. Sphingosine 1-phosphate counteracts the effects of interleukin-1beta in human chondrocytes. Arthritis and Rheumatism 65, 2113-22.

14.

Moon

Jeong

Lee

Seol

and Park

. 2012. Sphingosine-1-phosphate inhibits interleukin-1beta-induced inflammation in human articular chondrocytes. International Journal of Molecular Medicine 30, 1451-8.

15.

Rutges

Duit

Kummer

Bekkers

Oner

Castelein

Dhert

and Creemers

. 2013. A validated new histological classification for intervertebral disc degeneration. Osteoarthritis and Cartilage/OARS, Osteoarthritis Research Society 21, 2039-47.

16.

Cyster

and Schwab

. 2012. Sphingosine-1-phosphate and lymphocyte egress from lymphoid organs. Annual Review of Immunology 30, 69-94.

17.

Coste

Brenneis

Linke

Pierre

Maeurer

Becker

Schmidt

Gao

Geisslinger

and Scholich

. 2008. Sphingosine 1-phosphate modulates spinal nociceptive processing. The Journal of Biological Chemistry 283, 32442-51.

18.

David

Ciurea

Iencean

and Mohan

. 2010. Angiogenesis in the degeneration of the lumbar intervertebral disc. Journal of Medicine and Life 3, 154-61.

19.

Risbud

and Shapiro

. 2014. Role of cytokines in intervertebral disc degeneration: pain and disc content. Nature Reviews. Rheumatology 10, 44-56.

20.

Iida

Akeda

Kasai

Masuda

Morimoto

Sakakibara

Sato

and Uchida

. 2009. Expression of proteinase-activated receptor-2 in the intervertebral disc. Spine (Phila Pa 1976) 34, 470-8.

21.

Tsai

Lai

Liao

Niu

Chen

Lee