- Email: [email protected]
platelet-derived endothelial cell growth factor in synovial fluid and serum of rheumatoid arthritis
ELSEVIER
Clinica Chimica Acta 218 (1993) 1-4
@
High concentrations of immunoreactive gliostatin/platelet-derived endothelial cell growth factor in synovial fluid and serum of rheumatoid arthritis Kiyofumi Asai *a, Takayoshi Hirano a, Kohei Matsukawa a, Jun-ichi Kusada b, Masanori Takeuchi b, Takanobu Otsuka b, Nobuo Matsui b, Taiji Kato a aDepartments of Bioregulation Research, bDepartmentof Orthopedic Surgery, Nagoya City University Medical School, Mi=uho.ku, Nagoya 467, Japan (Received 8 Decembzr1992; revision received4 March 1993; accepted 8 March 1993)
Abstract
Since neovascularization plays an important role in the propagation of rheumatoid synovitis, we analyzed the concentration of gliostatin/platelet-derived endothelial cell growth factor (GLS/PD-ECGF), a potent angiogenic and chemotactic factor, in the synovial fluid and serum of rheumatoid arthritis (RA) patients. The immunoreactive GLS/PD-ECGF concentrations (mean value • S.D.) in synovial fluid, measured by a sandwich enzyme immunoassay, were significantly higher in RA patients than in osteoarthritis (OA) patients (233.02 ± 219.40 vs. 9.09 ± 14.86 ng/g, P < 0,001), and the serum concentrations were also higher in RA patients than in age-matched controls (8.77 ± 7.60 vs. 3.74 ,- 2.61 n~ml, P < 0.005). These results suggest that GLS/PD-ECGF may participate in the endothelial proliferation resulting in initiation of the extensive emigration of mononuclear cells and proliferation of the synovial tissues in rheumatoid arthritis, and that the immunoreactive GLS/PD-ECGF in serum as well as synovial fluids may be a useful diagnostic marker of RA.
Key words: Rheumatoid arthritis; Synovial fluids; Angiogenesis; Enzyme immunoassay; Gliostatin; Platelet-derived endothelial cell growth factor
* Corresponding author. 0009-8981/93/$06.00 © 1993 ElsevierScience Publishers B.V. All rights reserved. SSD10009-8981(93)05563-T
2
K. Asai et al. /Clin. Chim. Acta 218 (1993) 1-4
1. Introduction Giiostatin/platelet-derived endothelial cell growth factor (GLS/PD-ECGF) is a multifunctional protein factor, which induces angiogenesis including chemotactic migration and proliferation of endothelial cells [1], inhibits growth of gliai cells [2], and evokes neuronal survival and neuritogenic actions on cortical neurons [3]. Apart from the actions on gliai growth and neuronal survival, it is characteristic in its angiogenic and chemotactic actions in conjunction with the pathogenesis of angiogenic diseases. As a representative of such diseases, rheumatoid synovitis is characterized by ~eovascalarization of small vessels that induce the migration of mononuclear cells and proliferation of the synovial tissues. Therefore, we have, for the first time, measured the concentrations of the immunoreactive GLS/PD-ECGF in synovial fluids and serums of RA patients, and addressed an etiologic relation of the factor to RA. 2. Materials and methods Synovial fluids free of cells or platelets were obtained from the rheumatoid arthritis (RA) patients (n = ! 74) and osteoarthritis (OA) patients (n = 38). Sera were obtained from RA patients (n = 58), OA patients (n = 7) and their age-matched volunteers (n = 24). Synovial tissues were harvested during surgery on RA patients (n = 5) or OA patients (n = 3). All specimens were stored at -80°C until analysis. Frozen synovial tissues were thawed in 20 mmol/1 Tris-HC! buffer, pH 7.5, and homogenized by Polytron homogenizer on ice in the presence of 10 ~mol/I leupeptin, 10 ~mol/! pepstatin A, and I mmol/I phenyimethane sulfonyl fluoride. The homogenates were centrifuged at 25,000 x g for i h at 4°C, and the supernatant was applied to the enzyme immunoassay. Protein contents were determined by BCA Protein Assay Kit (Pierce). A sandwich enzyme immunoassay system for GLS/PDECGF was composed of two kinds of antibodies, the polyclonai antibodies for the primary antibody immobilized on the solid phase and the monoclonal antibody for the secondary antibody labeled with/~-galactosidase [41. There was a good linearity for GLS/PD-ECGF concentration of 0. I-10 rig/well (the detection limit: 30 pg/weli, the variation coefficients: 8.0% for the intra-assay, 5.7% for the inter-assay). This assay system provides no cross-reactivity with varied cytokines or growth factors [4], and no difference in the concentration either with serum and plasma or between heparinized and untreated synovial fluids. The immuno-reactive GLS/PD-ECGF values were expressed as nanogram per gram for synovial fluid, nanogram per milliliter for serum, and ~g/mg of extracted proteins for synovial tissue. All data are analyzed by non-parametric Mann-Whitney u-test. 3. Remits and discussion The immunoreactive GLS/PD-ECGF concentration (mean value-~-S.D.) in synovial fluid was significantly higher in RA patients than in OA patients (233.02 4. 219.40 vs. 9,09 ± 14.86 ng/g, P < 0.001); and the concentration in serum was also higher in RA patients than in either OA patients or age-matched controls (8.77 ~ 7.60 vs. 3.91 ± 2.48 or 3,74 ± 2.61 ng/ml, P < 0.005) as shown in Fig. I,
3
K. Asai et al,/ Clin. Chim. Acta 218 (1993) I - 4
A
B (1)
.-. 18[
(2)
(3)
~ 500 r
::~'~" 2.0
is
14 f
E 1.s
Q/
--"
•
RA OA Cont.
OA
A
Synovlal fluids (GLS/PI)-ECGFng/g)
Fig. I, (A) GLS/PD-ECGF concentrations in serum, synovial fluids or synovial tissues obtained from RA and OA patients. (I) Serum GLS/PD-ECGF levels. RA (n = 58), OA (n = 7) and age-matched controls (n = 24), (2) GLS/PD-ECGF levels in synoviai fluids, RA (n ffi 174) and OA (n = 38) (3) GLS/PD-ECGF contents in synovial tissue extracts. RA (n ffi 5) and OA ( n - 3 ) . Vertical bars indicate the S,D. (B) Correlation between serum and synovial fluids gliostatin/PD-ECGF levels in RA patients., ffi 20; r ffi0.856; P < 0.01. All data ware analyzed by non-parametric Mann-Whitney U-test.
Exceptionally, the serum from malignant RA patients showed a high concentration (20-40 ng/ml). There was a significant correlation between the individual immunoreactive GLS/PD-~CGF concentrations in synovial fluid and in serum ( n - 20, r -- 0.856, P < 0.01). The concentration in the extracts of synovial tissue was higher in RA patients than in OA patients (992 ± 54 vs. 107 .~ 93 ng/mg protein, P < 0.05). The histology of rheumatoid synovial tissues is characteristic in an aberrant proliferation of the connective tissue, an infiltration of chronic inflammatory ceils, and an extensive proliferation of small blood vessels or neovascularization. The extensive network of small vessels supports not only the infiltration of inflammatory cells to the synovium but the fibroblast proliferation essential for the growth of synovium developing into pannus. Therefore, making clear the mechanism of proliferation of blood vessels is important to realize the chronic inflammation such as the rheumatoid synovitis. Numerous factors, such as basic fibroblast growth factor (b-FGF, which is lately found to be identical to a heparin-binding ECGF) [5], platelet-derived growth factor (PDGF) [6] and interleukin-I (IL-I) [7], have been reported to support the process of neovascularization. Since GLS/PD-ECGF had been reported to be one of the angiogenic factors possibly causing an aberrant anglogenesis of RA [1], we have measured the concentration in synovial fluids and sera of rheumatoid arthritis patients. Taking into account the high concentration of immunoreactive OLS/PD-ECGF in synovial fluids in RA patients, we assume that GLS/PD-ECGF participates in the
4
K. Asai et al./Clin. Chim. Acta 218 (1993) 1-4
proliferation of endothelial cells and neovascularization in rheumatoid synovial tissues. GLS/PD-ECGF in synovial fluid might be produced by synovial tissue inasmuch as synoviai tissues contain high concentration of the immunoreactive GLS/PD-ECGF, although it is hard now to specify which type of cells in rheumatoid synovial tissues produces this factor. High concentration of GLS/PD-ECGF in serum from RA patients may reflect that the active proliferation of endothelial cells occurs in systemic tissues as well as in synovial tissues. We have recently demonstrated that GLS/PD-ECGF possesses not only growth inhibitory actions on 81ial cells but also neurotrophic actions on neuronal cells [2,3]. None of the current cases of RA, however, was affected in the central or peripheral nervous system except in joint movements. The neurological relevance of this factor in RA remains uncertain. To this day, the clinical diagnosis of RA consists mainly of rheumatoid factor (RF), C.reactive protein (CRP), erythrocyte sedimentation rate (ESR) besides the American Rheumatism Association (ARA) criteria for rheumatoid arthritis [8]. 3"o further assess whether serum GLS/PD-ECGF level is useful as one of diagnostic indicator of RA or not, the correlation with such conventional indicators was examined. Serum level showed a statistically significant correlation with CRP (n = 102, r ffi 0.309, P < 0.01), ESR (n = 102, r = 0.236, P < 0.05), RAHA (n = 101, r = 0.258, P < 0.01) and Lansbury Score (n -- 54, r - 0.330, P < 0.05) but not with RF-IgG, RF-IgA or RF-IgM. The current work demonstrates the diagnostic potential of GLS/PD-ECGF level as one of clinical indicators of RA. 4. Acknowledgments This work was supported by the Grant-in-Aid for Scientific Research (B), Scientific Research on Priority Areas, Cancer Research and Encouragement of Youn 8 Scientists from the Ministry of Education, Science and Culture, Japan.
5. Referepces I Ishikawa F, Miyazono K, Hellman U et al, Identification of angiogenic activity and the cloning and
expression of platelet.derived endothelial ceil growth factor, Nature 1989;338:557-561. 2 Asai K, Hirano T, Kaneko S et al. A no~el glial growth inhibitory factor, gliostatin, derived from neurofibroma. J Neurochem 1992;59:307-317. 3 Asai K, Nakanishi K, lsobe ! et el. Neurotrophic action of 81iostatin on cortical neurons: Identification of gliostatin and platelet-derived endothelial cell growth factor (PD-ECGF). J Biol Chem 1992;267:20311-20316. 4 Hirano T, Asai K, Matsukawa K ¢t el. Establishment of enzyme immunoassay system for gliostatin/platelet.derived endothelial cell growth factor (PD-ECGF), Biochim Biophys Acta 1993; ! 176:229-304. $ Rifkin DB, Moscatelli D. Recent developments in the cell biology of basic fibroblast growth factor. $ Cell Biol 1989;109:!-6. 6 Reuterdahl C, Tingstrom A, Terracio L, Funa K, Heldin CH, Rubin K, Characterization of platelet. derived growth factor 0.receptor expressing cells in the vasculature of human rheumatoid synovium. Lab Invest 1991;3:321-329, 7 Cozzolino F, Torcia M, Aldinucci D et el. lnterleukin I is an autocrine regulator of human endothelial cell growth, Proc Natl Acad Sci USA 1990;87:6487-6491. 8 Arnett FC, Edworthy SM, Bloch DA et el. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis, Arthr Rheum 1988;31:315-324.
Clinica Chimica Acta 218 (1993) 1-4
@
High concentrations of immunoreactive gliostatin/platelet-derived endothelial cell growth factor in synovial fluid and serum of rheumatoid arthritis Kiyofumi Asai *a, Takayoshi Hirano a, Kohei Matsukawa a, Jun-ichi Kusada b, Masanori Takeuchi b, Takanobu Otsuka b, Nobuo Matsui b, Taiji Kato a aDepartments of Bioregulation Research, bDepartmentof Orthopedic Surgery, Nagoya City University Medical School, Mi=uho.ku, Nagoya 467, Japan (Received 8 Decembzr1992; revision received4 March 1993; accepted 8 March 1993)
Abstract
Since neovascularization plays an important role in the propagation of rheumatoid synovitis, we analyzed the concentration of gliostatin/platelet-derived endothelial cell growth factor (GLS/PD-ECGF), a potent angiogenic and chemotactic factor, in the synovial fluid and serum of rheumatoid arthritis (RA) patients. The immunoreactive GLS/PD-ECGF concentrations (mean value • S.D.) in synovial fluid, measured by a sandwich enzyme immunoassay, were significantly higher in RA patients than in osteoarthritis (OA) patients (233.02 ± 219.40 vs. 9.09 ± 14.86 ng/g, P < 0,001), and the serum concentrations were also higher in RA patients than in age-matched controls (8.77 ± 7.60 vs. 3.74 ,- 2.61 n~ml, P < 0.005). These results suggest that GLS/PD-ECGF may participate in the endothelial proliferation resulting in initiation of the extensive emigration of mononuclear cells and proliferation of the synovial tissues in rheumatoid arthritis, and that the immunoreactive GLS/PD-ECGF in serum as well as synovial fluids may be a useful diagnostic marker of RA.
Key words: Rheumatoid arthritis; Synovial fluids; Angiogenesis; Enzyme immunoassay; Gliostatin; Platelet-derived endothelial cell growth factor
* Corresponding author. 0009-8981/93/$06.00 © 1993 ElsevierScience Publishers B.V. All rights reserved. SSD10009-8981(93)05563-T
2
K. Asai et al. /Clin. Chim. Acta 218 (1993) 1-4
1. Introduction Giiostatin/platelet-derived endothelial cell growth factor (GLS/PD-ECGF) is a multifunctional protein factor, which induces angiogenesis including chemotactic migration and proliferation of endothelial cells [1], inhibits growth of gliai cells [2], and evokes neuronal survival and neuritogenic actions on cortical neurons [3]. Apart from the actions on gliai growth and neuronal survival, it is characteristic in its angiogenic and chemotactic actions in conjunction with the pathogenesis of angiogenic diseases. As a representative of such diseases, rheumatoid synovitis is characterized by ~eovascalarization of small vessels that induce the migration of mononuclear cells and proliferation of the synovial tissues. Therefore, we have, for the first time, measured the concentrations of the immunoreactive GLS/PD-ECGF in synovial fluids and serums of RA patients, and addressed an etiologic relation of the factor to RA. 2. Materials and methods Synovial fluids free of cells or platelets were obtained from the rheumatoid arthritis (RA) patients (n = ! 74) and osteoarthritis (OA) patients (n = 38). Sera were obtained from RA patients (n = 58), OA patients (n = 7) and their age-matched volunteers (n = 24). Synovial tissues were harvested during surgery on RA patients (n = 5) or OA patients (n = 3). All specimens were stored at -80°C until analysis. Frozen synovial tissues were thawed in 20 mmol/1 Tris-HC! buffer, pH 7.5, and homogenized by Polytron homogenizer on ice in the presence of 10 ~mol/I leupeptin, 10 ~mol/! pepstatin A, and I mmol/I phenyimethane sulfonyl fluoride. The homogenates were centrifuged at 25,000 x g for i h at 4°C, and the supernatant was applied to the enzyme immunoassay. Protein contents were determined by BCA Protein Assay Kit (Pierce). A sandwich enzyme immunoassay system for GLS/PDECGF was composed of two kinds of antibodies, the polyclonai antibodies for the primary antibody immobilized on the solid phase and the monoclonal antibody for the secondary antibody labeled with/~-galactosidase [41. There was a good linearity for GLS/PD-ECGF concentration of 0. I-10 rig/well (the detection limit: 30 pg/weli, the variation coefficients: 8.0% for the intra-assay, 5.7% for the inter-assay). This assay system provides no cross-reactivity with varied cytokines or growth factors [4], and no difference in the concentration either with serum and plasma or between heparinized and untreated synovial fluids. The immuno-reactive GLS/PD-ECGF values were expressed as nanogram per gram for synovial fluid, nanogram per milliliter for serum, and ~g/mg of extracted proteins for synovial tissue. All data are analyzed by non-parametric Mann-Whitney u-test. 3. Remits and discussion The immunoreactive GLS/PD-ECGF concentration (mean value-~-S.D.) in synovial fluid was significantly higher in RA patients than in OA patients (233.02 4. 219.40 vs. 9,09 ± 14.86 ng/g, P < 0.001); and the concentration in serum was also higher in RA patients than in either OA patients or age-matched controls (8.77 ~ 7.60 vs. 3.91 ± 2.48 or 3,74 ± 2.61 ng/ml, P < 0.005) as shown in Fig. I,
3
K. Asai et al,/ Clin. Chim. Acta 218 (1993) I - 4
A
B (1)
.-. 18[
(2)
(3)
~ 500 r
::~'~" 2.0
is
14 f
E 1.s
Q/
--"
•
RA OA Cont.
OA
A
Synovlal fluids (GLS/PI)-ECGFng/g)
Fig. I, (A) GLS/PD-ECGF concentrations in serum, synovial fluids or synovial tissues obtained from RA and OA patients. (I) Serum GLS/PD-ECGF levels. RA (n = 58), OA (n = 7) and age-matched controls (n = 24), (2) GLS/PD-ECGF levels in synoviai fluids, RA (n ffi 174) and OA (n = 38) (3) GLS/PD-ECGF contents in synovial tissue extracts. RA (n ffi 5) and OA ( n - 3 ) . Vertical bars indicate the S,D. (B) Correlation between serum and synovial fluids gliostatin/PD-ECGF levels in RA patients., ffi 20; r ffi0.856; P < 0.01. All data ware analyzed by non-parametric Mann-Whitney U-test.
Exceptionally, the serum from malignant RA patients showed a high concentration (20-40 ng/ml). There was a significant correlation between the individual immunoreactive GLS/PD-~CGF concentrations in synovial fluid and in serum ( n - 20, r -- 0.856, P < 0.01). The concentration in the extracts of synovial tissue was higher in RA patients than in OA patients (992 ± 54 vs. 107 .~ 93 ng/mg protein, P < 0.05). The histology of rheumatoid synovial tissues is characteristic in an aberrant proliferation of the connective tissue, an infiltration of chronic inflammatory ceils, and an extensive proliferation of small blood vessels or neovascularization. The extensive network of small vessels supports not only the infiltration of inflammatory cells to the synovium but the fibroblast proliferation essential for the growth of synovium developing into pannus. Therefore, making clear the mechanism of proliferation of blood vessels is important to realize the chronic inflammation such as the rheumatoid synovitis. Numerous factors, such as basic fibroblast growth factor (b-FGF, which is lately found to be identical to a heparin-binding ECGF) [5], platelet-derived growth factor (PDGF) [6] and interleukin-I (IL-I) [7], have been reported to support the process of neovascularization. Since GLS/PD-ECGF had been reported to be one of the angiogenic factors possibly causing an aberrant anglogenesis of RA [1], we have measured the concentration in synovial fluids and sera of rheumatoid arthritis patients. Taking into account the high concentration of immunoreactive OLS/PD-ECGF in synovial fluids in RA patients, we assume that GLS/PD-ECGF participates in the
4
K. Asai et al./Clin. Chim. Acta 218 (1993) 1-4
proliferation of endothelial cells and neovascularization in rheumatoid synovial tissues. GLS/PD-ECGF in synovial fluid might be produced by synovial tissue inasmuch as synoviai tissues contain high concentration of the immunoreactive GLS/PD-ECGF, although it is hard now to specify which type of cells in rheumatoid synovial tissues produces this factor. High concentration of GLS/PD-ECGF in serum from RA patients may reflect that the active proliferation of endothelial cells occurs in systemic tissues as well as in synovial tissues. We have recently demonstrated that GLS/PD-ECGF possesses not only growth inhibitory actions on 81ial cells but also neurotrophic actions on neuronal cells [2,3]. None of the current cases of RA, however, was affected in the central or peripheral nervous system except in joint movements. The neurological relevance of this factor in RA remains uncertain. To this day, the clinical diagnosis of RA consists mainly of rheumatoid factor (RF), C.reactive protein (CRP), erythrocyte sedimentation rate (ESR) besides the American Rheumatism Association (ARA) criteria for rheumatoid arthritis [8]. 3"o further assess whether serum GLS/PD-ECGF level is useful as one of diagnostic indicator of RA or not, the correlation with such conventional indicators was examined. Serum level showed a statistically significant correlation with CRP (n = 102, r ffi 0.309, P < 0.01), ESR (n = 102, r = 0.236, P < 0.05), RAHA (n = 101, r = 0.258, P < 0.01) and Lansbury Score (n -- 54, r - 0.330, P < 0.05) but not with RF-IgG, RF-IgA or RF-IgM. The current work demonstrates the diagnostic potential of GLS/PD-ECGF level as one of clinical indicators of RA. 4. Acknowledgments This work was supported by the Grant-in-Aid for Scientific Research (B), Scientific Research on Priority Areas, Cancer Research and Encouragement of Youn 8 Scientists from the Ministry of Education, Science and Culture, Japan.
5. Referepces I Ishikawa F, Miyazono K, Hellman U et al, Identification of angiogenic activity and the cloning and
expression of platelet.derived endothelial ceil growth factor, Nature 1989;338:557-561. 2 Asai K, Hirano T, Kaneko S et al. A no~el glial growth inhibitory factor, gliostatin, derived from neurofibroma. J Neurochem 1992;59:307-317. 3 Asai K, Nakanishi K, lsobe ! et el. Neurotrophic action of 81iostatin on cortical neurons: Identification of gliostatin and platelet-derived endothelial cell growth factor (PD-ECGF). J Biol Chem 1992;267:20311-20316. 4 Hirano T, Asai K, Matsukawa K ¢t el. Establishment of enzyme immunoassay system for gliostatin/platelet.derived endothelial cell growth factor (PD-ECGF), Biochim Biophys Acta 1993; ! 176:229-304. $ Rifkin DB, Moscatelli D. Recent developments in the cell biology of basic fibroblast growth factor. $ Cell Biol 1989;109:!-6. 6 Reuterdahl C, Tingstrom A, Terracio L, Funa K, Heldin CH, Rubin K, Characterization of platelet. derived growth factor 0.receptor expressing cells in the vasculature of human rheumatoid synovium. Lab Invest 1991;3:321-329, 7 Cozzolino F, Torcia M, Aldinucci D et el. lnterleukin I is an autocrine regulator of human endothelial cell growth, Proc Natl Acad Sci USA 1990;87:6487-6491. 8 Arnett FC, Edworthy SM, Bloch DA et el. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis, Arthr Rheum 1988;31:315-324.