Activation of nuclear factor κB at the onset of ossification of the spinal ligaments

J Orthop Sci (2000) 5:572–578

Activation of nuclear factor kB at the onset of ossification of the spinal ligaments Taiichi Kosaka1, Atsuhiro Imakiire1, Fumio Mizuno2, and Kengo Yamamoto1 1 2

Department of Orthopaedic Surgery, Tokyo Medical University, 6-7-1 Nishi-shinjuku, Shinjuku-ku, Tokyo 160-0023, Japan Department of Microbiology, Tokyo Medical University, Tokyo, Japan

Abstract We examined the correlation between the activation of nuclear factor kB (NFkB), stimulated by environmental factors involving cytokines and growth factors in ligament cells, and the onset of ossification of the spinal ligaments (OSL) or diffuse idiopathic skeletal hyperostosis (DISH). Aseptic samples were taken carefully from non-ossified sites during surgery (75 patients). We carried out preliminary hematoxylin and eosin and toluidine blue staining, using five portions of each specimen, and excluded samples containing chondrocytic, osteoblastic, or inflammatory cells (n 5 25). We used specimens from the remaining 50 patients (35 men and 15 women, ranging in age from 45–81 years); average age, 59.5 years (18 nuchal ligament specimens, and 32 yellow ligament specimens). OSL or DISH had occurred in 25 patients, 20 patients were in the non-OSL group (8 with cervical spondylotic myelopathy, and 12 with lumbar canal stenosis), and the remaining 5 samples were collected from patients with injury. For culture study, we used portions of the 14 largest samples from the above 50 patients. We extracted nuclear proteins and cytoplasmic proteins from non-ossified spinal ligaments in 50 patients and detected p65RelA/NFkB by Western blotting. Tumor necrosis factor-α (TNF α), interleukin 1β (IL-1β), platelet-derived growth factor BB (PDGF-BB) and transforming growth factor-β1 (TGF-β1) in cytoplasm were quantified by enzyme-linked immunosorbent assays (ELISA). Cultured cells from the 14 samples were then stimulated with 10, 100, 250, or 500 ng/ml of recombinant human (rh)PDGF-B or TGFβ1. A control experiment was performed without rhPDGF-BB or TGFβ1 stimulation. Alkaline phosphatase (ALP) activity was standardized by the DNA content of the cells. The number of NFkB-positive samples was significantly higher in patients with OSL or DISH than in non-OSL patients. This tendency was obvious in the case of OSL or DISH with non-insulin-dependent diabetes mellitus (NIDDM). In OSL and in DISH patients, significantly greater amounts of PDGF-BB and TGFβ1 were seen in ligament cells than in non-OSL patients (P , 0.05). There was a positive correlation

Offprint requests to: T. Kosaka Received: March 27, 2000 / Accepted: June 22, 2000

between the detection of p65RelA/NFkB band and the content of PDGF-BB and TGFβ1 in ligament cells (P , 0.05). ALP activity tended to be higher in cells in the OSL group not receiving any other treatment. Our results indicate the possibility that NFkB, stimulated by environmental factors involving PDGF-BB and TGFβ1 in ligament cells, influences the osteoblastic differentiation of undifferentiated mesenchymal cells. Key words Nuclear factor kB · Ossification of the spinal ligaments · Diffuse idiopathic skeletal hyperostosis · Noninsulin-dependent diabetes mellitus

Introduction Ossification of the spinal ligaments (OSL) is a disease in which compression of the spinal nerve results from the ossification and enlargement of the spinal ligaments. Because the disease is frequently associated with ossification of the iliotibial band and the ligaments of the pelvis and hip joint, ossification of the spinal ligaments has been thought of as a type of ankylosing spinal hyperostosis (ASH)5 or diffuse idiopathic skeletal hyperostosis (DISH).22 The etiology of OSL has not yet been clarified, although it is thought that the disease is related to complex environmental and genetic factors inducing osteogenic differentiation of undifferentiated mesenchymal cells. Because OSL is probably caused by many hereditary factors, genetic factors have attracted much attention.12 However, to facilitate early detection and treatment of OSL, it is also necessary to analyze the specific roles of environmental factors. OSL or DISH is frequently associated with non-insulin-dependent diabetes mellitus (NIDDM),3 the onset of which is related to inflammatory cytokines such as tumor necrosis factor-α (TNF-α) and interleukin-1β (IL-1β), and vascular growth factors such as platelet-derived growth factor-BB (PDGF-BB) and transforming growth

1 1

1

1 1

1 1 1

1

OPLL, Ossification of the posterior longitudinal ligaments; DISH, diffuse idiopathic skeletal hyperostosis; OYL, ossification of the yellow ligament; CSM, cervical spondylotic myelopathy; LCS, lumbar canal stenosis; INJ, injury; Com (complication); NIDDM, non-insulin-dependent diabetes mellitus

49 M 76 F 72 M 65 M 59 M 60 M 59 F 62 M 56 M 46 M 50 M 53 M 45 F 52 M LCS LCS LCS LCS LCS LCS LCS LCS LCS INJ INJ INJ INJ INJ 37 38 39 40 41 42 43 44 45 46 47 48 49 50 50 M 74 M 54 M 52 M 56 F 57 M 59 F 64 M 53 M 51 F 55 M 70 F 75 M 63 M 67 F 54 M 59 M 67 F 63 M 72 F 64 M 72 M 74 M 64 M 66 M 78 F 81 M 46 M 51 M 76 F 78 M 58 F 56 M 51 F 47 M 63 F 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

OPLL 1 DISH OPLL 1 DISH OPLL 1 DISH OPLL OYL 1 DISH OPLL OPLL OPLL OPLL OPLL OYL OYL OYL OYL OYL DISH DISH DISH

Com Age (years)/ Sex Diagnosis Patient no.

Table 1. Patients’ background

A microtome was used to prepare 50-µm slices. Phosphate-buffered saline (PBS) containing type 1 collagenase (1.0%) was added to each specimen, and the mixture was stirred at 37°C for 1 h. Then, the specimen was centrifuged at 700 g for 20 min, in accordance with a modification of the method of Dignam and colleagues (Dignam et al.;4 Bethea et al.2). The precipitate was dissolved in buffer A (containing 10 mM hydroxyethylpiperazine ethanesulfonic acid [HEPES] [pH 7.9 at 4°C], 1.5 mM MgCl2, 10 mM KCl, and 0.5 mM dithiothreitol [DTT]), the amount of which was adjusted to five times that of the specimen. After being placed on ice for 10 min, the extracts were treated with 1.0% Nonidet P-40. The specimen was homogenized three to four times, using a microhomogenizer, and was then centrifuged at 12 000 g for 10 min. The supernatant and the precipitate were collected as, respectively, cytoplasmic protein and crude nuclear fractions. The collected nuclear fraction was mixed with buffer C (containing 20 mM HEPES [pH7.9], 25% [v/v] glycerol,

Patient no.

Extraction of nuclear proteins and cytoplasmic proteins from non-ossifying spinal ligaments

OYL DISH DISH OYL 1 DISH DISH DISH DISH CSM CSM CSM CSM CSM CSM CSM CSM LCS LCS LCS

Diagnosis

Age (years)/ Sex

Com

Aseptic samples were taken carefully from non-ossified sites during surgery (75 patients) and rapidly frozen in liquid nitrogen. We carried out preliminary hematoxylin and eosine and toluidine blue staining using five portions of each sample, and excluded those samples containing chondrocytic, osteoblastic, or inflammatory cells (n 5 25). We used samples from the remaining 50 patients (35 men and 15 women, aged 45 to 81 years); average age, 59.5 years (18 nuchal ligament samples and 32 yellow ligament samples). OSL or DISH had occurred in 25 patients; 20 patients were in the non-OSL group (cervical spondylotic myelopathy, n 5 8, and lumbar canal stenosis, n 5 12); and the remaining 5 samples were collected from patients who were injured (Table 1). For culture study, we used portions of the 14 largest samples from the above 50 patients (Table 2, 3).

19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

Patient no.

Materials and methods

1 1 1

Diagnosis

Age (years)/ Sex

Com

factor-β1 (TGF-β1).6,7,17,18,20 These factors may be included among environmental factors of OSL. The nuclear factor (NF)kB/Rel family exists at sites downstream to those where many factors exist, and is activated to control the expression of various genes involved in cell division and growth.13,15,16,19,24 It has been shown that NFkB regulates the differentiation of multipotential cells.1,9,11 Accordingly, we examined the correlation of NFkB activation, stimulated by environmental factors involving these cytokines and growth factors in ligament cells, and the onset of OSL.

1

573 1

T. Kosaka et al.: Activation of NFkB and onset of OSL

574

T. Kosaka et al.: Activation of NFkB and onset of OSL

Table 2. Samples in which NFkB band was detected Group

n

OSL or DISH OSL and DISH Non · OSL INJ OSL or DISH 1 Com OSL or DISH 2 Com Non · OSL 1 Com Non · OSL 2 Com

25 5 20 5 13 12 6 14

NFkB-Positive cases 15 2 3 0 10 5 2 1

**

*

*** *

*P , 0.05; **P , 0.01; ***P , 0.001 OSL, ossification of the spinal ligaments; OSL or DISH 5 OPLL or OYL or DISH; OSL and DISH 5 OPLL and DISH/OYL and DISH; Non · OSL 5 CSM or LCS; Com 5 (complication) NIDDM

0.42 M NaCl, 1.5 mM MgCl2, 0.2 mM ethylenediaminetetraacetic acid [EDTA], 0.5 mM phenylmethylsulfonyl fluoride [PMSF], and 0.5 mM DTT. Nuclear extracts were briefly sonicated on ice, then centrifuged at 12 000 g for 10 min. Sodium dodecylsulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and Western blotting The amounts of cytoplasmic proteins in buffer A and nuclear proteins in buffer C were determined by colorimetric microassay (Bio-Rad, Richmond, CA, USA). The protein concentration of each sample was adjusted to 1 µg/µl. Then, 10 µl of electrophoresis sample buffer and 20 µl of specimen were mixed and heated at 100°C for 5 min, and a 15-µl aliquot was subjected to SDSPAGE in a 7.5% acrylamide gel. The electrophoresed proteins were transferred to a cellulose acetate membrane (0.5 mA, 16 h). Anti-P65RelA homodimers (rabbit IgG) antibodies19,21 and anti-actin monoclonal antibodies (Sigma, St Louis, MO, USA) were used as primary antibodies. Anti-rabbit IgG antibodies (Amersham, Buckinghamshire, England) and anti-mouse IgGhorseradish peroxidase antibodies (Amersham) were used as secondary antibodies. The protein bands were detected by chemiluminescence, using an enhanced chemiluminescence (ECL) system (Konica, Tokyo, Japan). Kaleidoscope standards (Bio-Rad) and biotinylated SDS-PAGE standards (Bio-Rad) were used as molecular weight markers (Bio-Rad). Rheumatic synovium and human osteosarcoma cells (HOS), stimulated with TNFα were used as positive controls. Statistical significance was assessed using hypothesis testing of proportion. Enzyme-linked immunosorbent assay (ELISA) IL-1β, TNFα, PDGF-BB, and TGFβ1 in cytoplasm were quantified with an ELISA kit (R and D Systems, Minneapolis, MN, USA), as in previous studies.

Amounts of cytokines and growth hormones in patients were determined and compared with the individual standard curves. Values for results are expressed as means 6 SEM. Values for cytokines and growth factors are expressed as ng/mg total protein. Statistical significance was assessed using analysis of variance (ANOVA), followed by the Tukey-Kramer multiple comparision test. A P value of less than 0.05 was considered to indicate significant differences. Isolation and culture of spinal ligament cells For cell culture, the collected ligaments were washed several times with phosphate-buffered saline (PBS), minced into pieces of about 1-mm2, and then placed in 60-mm culture dishes containing Dulbecco’s Modified Eagle’s Medium (DMEM) purchased from GIBCO Laboratories (Grand Island, NY, USA) supplemented with 10% heat-inactivated fetal calf serum (FCS) (CSL, Melbourne, Victoria, Australia) and 0.2 mM l-ascorbic acid 2-phosphate (Wako Pure Chemical Industries, Tokyo, Japan). The explants were incubated in a controlled 95% oxygen/5.0% CO2 atmosphere at 37°C. Cells migrating from the explants were harvested with 0.1% trypsin, replated in 100-mm culture dishes, and maintained in DMEM supplemented with 10% FCS and 0.2 mM l-ascorbic acid 2-phosphate. For the experiments, cells after the fifth passage were seeded in 12-well plates at a cell density of 5 3 104 cells/ well. Cell lines obtained from OSL groups and nonOSL groups were named OSL cells (OL) and nonOSL cells (NL), respectively. In the present study, seven OL lines and seven NL lines were used for the experiments. Stimulation of ligament cells When the ligament cells became confluent, the DMEM medium was replaced with DMEM containing 1.0% FCS and 0.2 mM l-ascorbic acid 2-phosphate. The cells were then stimulated with 10, 100, 250, or 500 ng/ml of recombinant human (rh)PDGF-BB or TGFβ1, and incubated for a further 96 h. A control experiment was performed without rhPDGF-BB or TGFβ1 stimulation. Determination of alkaline phosphatase activity and DNA content After 96 h of incubation with or without rhPDGF-B or TGFβ1, the alkaline phosphatase (ALP) activity of cultured cells was assayed. The cells were washed twice with cold PBS and lysed with 200 mM sodium carbonate, 100 mM glycine, 1 mM MgCl2, and 0.1% Triton X100. They were sonicated and then centrifuged at 9000 g for 10 min. The ALP activity of these supernatants

T. Kosaka et al.: Activation of NFkB and onset of OSL

was determined with an assay kit (Wako). Briefly, the substrate (7.5 mM p-nitrophenyl phosphate) was incubated with the cell extracts described above in 50 mM Tris-HCl (pH 9.6), 1 mM MgCl2 for 15 min at 37°C, and the amount of product (p-nitrophenol) was determined by absorbance at 405 nm. The DNA content of each cell layer was assayed using the method of Labarca and Paigen.14 The ALP activity was standardized by the DNA content of the cells, and was expressed as n moles of pnitrophenyl phosphate hydrolyzed/min per µg DNA. Data values are expressed as means 6 SEM. Statistical analysis was performed by Student’s t-test between cells without rhPDGF-BB or TGFβ1 (control) and cells with rhPDGF-BB or TGFβ1 stimulation at various concentrations. Significant differences were indicated by a P value of less than 0.01.

Results Western blotting In order to detect NFkB in nuclear protein extracts, molecular weight markers were used for comparison, and the 65-kDa band was considered to represent the p65RelA/NFkB (Fig. 1). Western blotting demonstrated that the cytoskeletal protein, actin, was detectable only in cytoplasmic protein fractions and not in nuclear protein fractions, confirming the absence of cytoplasmic proteins in nuclear protein extracts (Fig. 2). The number of samples in which the NFkB band was detected was significantly higher in patients with OSL or DISH than in non-OSL patients (P , 0.01) (Table 2) and the significant difference was even greater in those in whom such ossification was associated with NIDDM (P , 0.001).

575

ELISA OSL patients showed significantly higher TGFβ1 and PDGF-BB content in ligament tissue than nonOSL patients (P , 0.05) (Fig. 3a,b). In particular, there was a positive correlation between NFkB activation and the content of PDGF-BB and TGFβ1 (P , 0.05). However, no significant difference between OSL groups and non-OSL groups was observed for either IL1β or TNFα. ALP activity ALP activity tended to be higher in cells from patients in the OSL group without treatment than in the cells of patients other. Two cell lines from the OSL and DISH group showed high activity (3.49 n mole/min per µg DNA in OL1, 4.84 n mole/min per µg DNA in OL2), and four cell lines from the OSL or DISH group showed moderate activity (1.21 n mole/min per µg DNA in OL6, 0.98 in OL7, 1.13 in OL11, and 0.84 in OL13). All the NL cells showed marginal activity, of about 0.27–0.49 n

Fig. 2. Western blotting, demonstrating the cytoskeletal protein, actin, in cytoplasmic protein fractions (C) and its absence in nuclear protein fractions (N)

Fig. 1. The 65-kDa band was considered to represent the p65RelA/nuclear factor (NF) kB of aseptic samples from non-ossified sites. The positions of molecular size standards are indicated (in kDa) on the left. NS, Non-specific bands; RA, Synovium of rheumatoid arthritis; HOS, human osteosarcoma cells; OPLL 1 DISH, ossification of the posterior longitudinal ligaments 1 diffuse idiopathic skeletal hyperostosis; CSM, cervical spondylotic myelopathy; INJ, injury

576

T. Kosaka et al.: Activation of NFkB and onset of OSL

mole/min per µg DNA, and one cell line — OL12 — had low activity (0.47 n mole/min per µg DNA) (Table 3). There was no significant difference in the cell lines responding to rhPDGF-BB or TGFβ1.

Discussion

a

b Fig. 3a,b. Content of a transforming growth factor (TGF) β1 and b platelet-derived growth factor (PDGF)-BB in ligament OSL or DISH, j OSL and DISH, u tissue. *P , 0.05. Non-OSL, u Injured cases, u NFkB positive cases, u NFkB negative cases

It is thought that the onset of OSL is influenced by various environmental factors, but the exact nature of the intracellular signaling pathway remains to be determined. Activated NFkB enhances the expression of genes such as Sonic Hedgehog, Twist, and BMP-4, and it is also related to the initial differentiation of embryonic cells,10 and to the differentiation of osteoclasts.9 Thus, NFkB regulates the differentiation of multi-potential cells. In the present study, we took note of the fact that NFkB, together with inhibitor kB in the cytoplasm, forms a compound as part of the intracellular signaling pathway. This compound is dissociated by various stimuli, and NFkB alone is transferred into the nucleus to interact with its binding sites, thereby resulting in enhanced transcription.19 The amount of intranuclear NFkB is extremely low and, therefore, in-vivo assay is very difficult. In view of the possibility that the transfer of NFkB to the nucleus may be induced in tissue culture, we extracted nuclear proteins directly from intraoperatively collected samples and attempted to detect p65RelA/NFkB by Western blotting. The highly sensitive ECL technique which was adopted to detect actin, a cytoskeletal protein, did not reveal any in the nuclear protein extracts. This indicates a strong probability that the nuclear protein extracts used for NFkB detection contained no cytoplasm-derived proteins. Examination of the protein samples by SDS-PAGE revealed that NFkB had been transferred into the nucleus. Signifi-

Table 3. Culture study of specimens Cell line OL-1 OL-2 OL-6 OL-7 OL-11 OL-12 OL-13 NL-30 NL-31 NL-32 NL-33 NL-48 NL-49 NL-50

Patient no.

Tissue of origin

Diagnosis

ALP activity n mole/min per µg DNA

1 2 6 7 11 12 13 30 31 32 33 48 49 50

NU NU NU NU YL YL YL NU NU NU NU YL YL YL

OSL and DISH OSL and DISH OSL or DISH OSL or DISH OSL or DISH OSL or DISH OSL or DISH Non · OSL Non · OSL Non · OSL Non · OSL INJ INJ INJ

3.49 4.84 1.21 0.98 1.13 0.47 0.84 0.27 0.198 0.14 0.25 0.34 0.29 0.49

ALP, Alkaline phosphatase; OL, cell lines obtained from OSL patients; NL, non-OSL; NU, nuchal ligament; YL, yellow ligament

T. Kosaka et al.: Activation of NFkB and onset of OSL

cantly more samples in which p65RelA/NFkB band was detected were observed in patients with OSL or DISH than in non-OSL patients. This tendency was particularly obvious in OSL or DISH with NIDDM. Patients with OSL or DISH showed significantly greater amounts of PDGF-BB and TGFβ1 in ligament cells than non-OSL patients. In particular, there was a positive correlation between the appearance of NFkB and the content of PDGF-BB and TGFβ1 in ligament cells. This would seem to indicate that PDGF-BB and TGFβ1 are factors capable of altering the kinetics of NFkB in patients with OSL or DISH, or both. In the culture study, we observed that ALP activity tended to be higher in cells from OSL or DISH groups than in cells from the other patients. ALP, which is an ectoenzyme expressed on the cell surface with phosphatidyl inositol anchorage, plays an important role in the calcification process during bone formation. High ALP activity has been viewed as a good indicator for the maturation stages of osteoblastic change.8,25,26 This suggests that spinal ligament cells in patients with OSL or DISH have characteristics of osteoprogenitor cells. However, there was no significant difference in the response of the cell lines with rhPDGF-BB or TGFβ1. Factors which are related to the onset of OSL can be divided into initiators and promoters. Accordingly, PDGF-BB and TGFβ1 are thought to operate mainly as initiators. Thus, we conclude that the activation of NFkB, stimulated by environmental factors involving PDGF-BB and TGFβ1 in ligament cells, influences the osteoblastic differentiation of undifferentiated mesenchymal cells which are genetically related to OSL and NIDDM (Fig. 4). Our results indicate the possibility that NFkB activation plays an important role in the

Fig. 4. The role of NFkB activation in undifferentiated mesenchyme of ligament tissue at the onset of ossification of the spinal ligments (OSL). NIDDM, Non-insulin-dependent diabetes mellitus

577

onset of OSL, but we need to clarify at what period it is activated, and how. This knowledge could lead to new methods of the disease prevention and treatment. Acknowledgments. The authors are indebted to Prof. J. Patrick Barron of the International Medical Communications Center of Tokyo Medical University for his review of this manuscript. This study was supported by a Research Grant for Specific Disease of the Ministry of Health and Welfare of Japan.

References 1. Beck CW, Sutherland DJ, Woodland HR. Involvement of NF-kB associated proteins in FGF-mediated mesoderm induction. Int J Devel Biol 1998;42:67–77. 2. Bethea JR, Ohmori Y, Hamilton TA. A tandem GC box motif is necessary for lipopolysaccharide-induced transcription of the type II TNF receptor gene. J Immunol 1997;158:5815–23. 3. Denko CW, Boja B, Moskowitz RW. Growth promoting peptides in osteoarthritis and diffuse idiopathic skeletal hyperostosis — insulin, insulin-like growth factor-I, growth hormone. J Rheumatol 1994;21:1725–30. 4. Dignam JD, Lebovitz RM, Roeder RG. Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalism nuclei. Nucleic Acids Res 1983;11:1475–89. 5. Forestier J, Rotes-Querol J. Senile ankylosing hyperostosis of the spine. Ann Rheum Dis 1950;9:321–30. 6. Freter RR, Alberta JA, Hwang GY, et al. Platelet-derived growth factor induction of the immediate-early gene MCP-1 is mediated by NF-kB and a 90-kDa phosphoprotein coactivator. J Biol Chem 1996;271:17417–24. 7. Inaba T, Ishibashi S, Gotoda T, et al. Enhanced expression of platelet-derived growth factor-β receptor by high glucose. Involvement of platelet-derived growth factor in diabetic angiopathy. Diabetes 1996;45:507–12. 8. Ishida Y, Kawai S. Characterization of cultured cells derived from ossification of the posterior longitudinal ligament of the spine. Bone 1993;14:85–91 (in Japanese). 9. Jimi E. Role of NF-kB/Rel transcription factors for the differentiation and function of osteoclasts. The Bone 1998;12:91–8 (in Japanese). 10. Kanegae Y. Role of Rel/NFkB transcription factors during the outgrowth of the vertebrate limb. Nature 1998;392:611–14. 11. Kitajima I. Role of activation of NFkB and diseases. Clin Chem 1998;27:78–86 (in Japanese). 12. Koga H, Taketomi E, Sakou T, et al. Genetic analysis of OPLL Clin Orthop Surg 1998;33385–91 (in Japanese). 13. Kwon G, Corbett JA, Hauser S, et al. Evidence for involvement of the proteasome complex (26S) and NFkB in IL-1-β induced nitric oxide and prostaglandin production by rat islets and RINm5F cells. Diabetes 1998;47:583–91. 14. Labarca C, Paigen K. A simple, rapid, and sensitive DNA assay procedure. Anal Biochem 1980;102:344–52. 15. Li JM, Shen X, Hu PP, et al. Transforming growth factor β stimulates the human immunodeficiency virus 1 enhancer and requires NF-kB activity. Mol Cell Biol 1998;18:110–21. 16. Marumo T, Schini-Kerth VB, Fisslthaler B, et al. Platelet-derived growth factor-stimulated superoxide anion production modulates activation of transcription factor NF-kB and expression of monocyte chemoattractant protein 1 in human aortic smooth muscle cells. Circulation 1997;96:2361–7. 17. Nakamura T, Ebihara I, Fukui M, et al. Effect of a specific endothelin receptor A antagonist on mRNA levels for extracellu-

578

18.

19. 20.

21.

22.

T. Kosaka et al.: Activation of NFkB and onset of OSL lar matrix components and growth factors in diabetic glomeruli. Diabetes 1995;44:895–9. Nilsson J, Jovinge S, Niemann A, et al. Relation between plasma tumor necrosis factor-α and insulin sensitivity in elderly men with non-insulin-dependent diabetes mellitus. Arterioscler Thromb 1998;18:1199–202. Okamoto H. Redox control mechanism and NF-kB activation. Cyto-protection Biol 1997;15:1–6 (in Japanese). Pfeiffer A, Drewes C, Middelberg-Bisping K, et al. Elevated plasma levels of transforming growth factor-β1 in NIDDM. Diabetes Care 1996;19:1113–17. Renard J. The relationship of human serum and NF-kB in the cells of HIV patients. Annual report of the special study group on HIV. The Ministry of Health and Welfare of Japan; 1995. p. 31–3 (in Japanese). Resnick D, Shaul SR, Robins JM. Diffuse idiopathic skeletal hyperostosis (DISH). Forestier’s disease with extraspinal manifestations. Radiology 1975;115:513–24.

23. Sakurai H, Shigemori N, Hasegawa K, et al. TGF-β-activated kinase 1 stimulates NF-kB activation by an NF-kB-inducing kinase-independent mechanism. Biochem Biophys Res Commun 1998;243:545–9. 24. Sen CK, Khanna S, Reznick AZ, et al. Glutathione regulation of tumor necrosis factor-α-induced NF-kB activation in skeletal muscle-derived L6 cells. Biochem Biophys Res Commun 1997; 237:645–9. 25. Terakado A, Tagawa M, Goto S, et al. Elevation of alkaline phosphatase activity induced by parathyroid hormone in osteoblast-like cells from the spinal hyperostotic mouse TWY (twy/twy). Calcif Tissue Int 1995;56:135–9. 26. Yamasaki M, Kon T, Goto K, et al. Effect of cell growth factors on the cultured spinal ligament cells; a possible pathogenesis of OPLL. Clin Orthop Surg 1998;33:567–72 (in Japanese).