вход по аккаунту


Serum Clq levels as a prognostic guide to articular erosions in patients with rheumatoid arthritis.

код для вставкиСкачать
Clq was measured serially by single radial
immunodiffusion in 54 rheumatoid arthritis (RA) patients over a period of more than 5 years, and values
were correlated with laboratory, radiographic, and
clinical findings. The number of joints with erosion
(NJE) was determined retrospectively from radiographs
of patients who had RA of >7 years duration. In
patients with clinically “burned out” RA, Clq levels
were not statistically different from those of healthy
adults. During the period of active disease, each patient’s Clq level remained very constant, irrespective of
erythrocyte sedimentation rate, C-reactive protein
(CRP) level, or whether the RA was active or in
remission. No sustained correlation was found between
the Clq level and the other 2 acute phase reactants, but
patients with Clq levels of at least 250 pg/ml showed a
positive CRP over a period of years, in contrast to those
Presented in part at the 47th Annual Meeting of the American Rheumatism Association, San Antonio, TX. June 1983.
From the Department of Orthopedic Surgery, Osaka University School of Medicine. Osaka. Japan and the Department of
Microbiology, Nara Medical College, Nara, Japan.
Takahiro Ochi, MD: Lecturer. Department of Orthopedics.
Osaka University School of Medicine; Kunio Yonemasu. MD:
Associate Professor, Department of Microbiology, Nara Medical
College; Rokuro Iwase, MD: Assistant Professor of Orthopedics,
Osaka University School of Medicine; Takako Sasaki, PhD: Assistant Professor of Microbiology, Nara Medical College; Kenichiro
Tsuyama, MD: Fellow in Orthopedics, Osaka University School of
Medicine; Keiro Ono, MD: Professor of Orthopedics, Osaka University School of Medicine.
Address reprint requests to Dr. Takahiro Ochi, Department
of Orthopedic Surgery, Osaka University School of Medicine.
1-1-50 Fukushima, Fukushima-ku, Osaka 553, Japan.
Submitted for publication October 3, 1983; accepted in
revised form March 12, 1984.
Arthritis and Rheumatism, Vol. 27, No. 8 (August 1984)
with Clq levels below 250 Crglml. Patients with an initial
Clq above 250 &ml had more erosive RA when
compared with those having Clq levels below 250 pg/
ml. These data suggest that active R A can be classified
into two subsets by Clq levels, one with persistent
inflammation and a high NJE and another without
persistent inflammation and with a low NJE.
Serum C Iq levels in healthy individuals remain
very constant between the ages of 3 days and 50 years.
The average serum C l q level of 346 healthy Japanese
adults (between 15 and 75 years of age) was found to
be 136.5 pg/ml (SD = 22.6) (1). However, in patients
with rheumatoid arthritis (RA), the average serum C l q
level was 21 1.9 pg/rnl (SD = 26.9), significantly higher
than that of the healthy group (2). Most of the RA
patients studied had active disease, but some had had
no signs of disease activity for at least 3 years, after
previously having long periods of active disease. Their
condition is termed “burned out” RA. Since Clq
levels in “burned out” RA, even with extensive
articular damage, wcre frequently within the normal
range, it was of interest to learn whether they remained constantly high during the entire clinical
course of RA and declined to the normal range only
after prolonged inactivity, or if these levels were
sometimes normal during active disease as well. Another question of interest was what other clinical,
laboratory, and radiologic data might be correlated
with a high serum C lq level, and whether the C l q level
could be of prognostic value in relation to the disease
course. To answer these questions, a longitudinal
study of 54 patients with active RA was undertaken.
Synovial cell culture. Synovial cell cultures were
prepared as described by Dayer et a1 (7). Fresh samples of
synovium were obtained at synovectomy from patients with
clinically defined, definite rheumatoid arthritis or osteoarthritis. The superficial layer of synovium was washed with
calcium- and magnesium-free phosphate buffered saline
(PBS). This layer was then dissected, minced, and incubated
for 90 minutes in 20 ml of Dulbecco's modified Eagle's
medium (DMEM) (Gibco, Grand Island, N Y ) without serum, 4 mdml of clostndiopeptidase A (125-200 units/mg,
Worthington Biochemicals, Freehold, NJ), 100 units
of penicillin, and 100 pg of streptomycin/ml (Gibco) at 37°C
in 5% COz. The suspension was mixed frequently during this
period. This was followed by addition of an equal volume of
0.25% trypsin for 30 minutes. The resulting suspension was
then centrifuged at room temperature for 10 minutes at 400g
and the pellet resuspended in 0.05% trypsin and 0.02%
EDTA, mixed for 10 minutes, and re-centrifuged. Cells were
then washed twice in DMEM-10% fetal calf serum (FCS).
Cells were counted and placed in plastic petri dishes
(Falcon Plastics, Oxnard, CA) at 1 x 206 cellslml in DMEM10% FCS. After overnight incubation the nonadherent cell
population was removed and the adherent population was
vigorously washed with PBS and then cultured in DMEM10% FCS. After several passages, cells were plated in 16-mm
diameter wells (Costar, Cambridge. MA) at I x lo' cells/
well. Cells were near confluency at this concentration, and
all wells were allowed to grow to confluency before use.
Therefore, cell counts at the end of all incubation conditions
were within 26% of the mean.
Mononuclear cell cultures. Heparinized peripheral
blood from healthy volunteers was layered onto FicollHypaque (Pharmacia Fine Chemicals, Piscataway, NJ) and
centrifuged at 400g for 40 minutes at room temperature, and
the mononuclear cell suspension was removed from the
interface. Plasma was saved and used to coat gels for
adherent cell preparation (8). Cells were washed 3 times with
Hanks' balanced salt solution (HBSS) and resuspended in
DMEM-10% FCS supplemented with penicillin (100 units/
ml), streptomycin (100 ndml). and L-glutamine (100 mM) at
a cellular concentration of 8 x lo6 cells/cc. Ten milliliters of
the resulting single cell suspension was layercd over previously prepared plasma-coated gels in 75-ml flasks (Corning)
using the method of Freundlich et a1 (8).
Adherence was allowed for 40 minutes at 37"C, after
which nonadherent cells were removed and flasks washed 4
times with prewarmed media. Adherent cells were then
incubated in DMEM-1% FCS plus 10 mM EDTA for 15
minutes at 37°C. After 15 minutes, cells no longer adhered
and were removed, washed at 4"C, and plated at desired
concentrations (i.e., 1-2 x 106/wellfor MCF production, and
4-6 x 106/well for binding). The resultant monolayer contained more than 98% monocytes, as determined by nonspecific estcrase staining, ingestion of latex particles, and
morphologic appearance. After overnight incubation in
DMEM-10% FCS, various test substances were added in
DMEM-10% FCS and incubated for another 48 hours.
Supernatants were harvested and stored at -20°C. Viability
of monocytes after cxposure to chloroquine was confirmed
by trypan blue exclusion, lactate dehydrogenase (LDH)
determination from supernatants, and by ingestion of latex
Bioassay for MCF activity. Assay of supernates for
PGE-stimulating activity was carried out by diluting thawed
supernatant at least I:20 in DMEM-10% FCS. The 400-p1
sterile, diluted supernatants were then incubated with ASC
for 48 hours, and cell-free supernates were removed and
assayed for PGE. PGE was assayed in undiluted monocyte
supernates. Chloroquine, 1.25 pg/ml (a 1:20 dilution of 25
pglml) was incubated directly on ASC in the presence of
control condition monocyte supernates and partially purified
Prostaglandin E, lysozyme, and lactate dehydrogenase. Prostaglandin E was assayed in culture media using a
monoclonal antibody specific for PGE (kindly provided by
Dr. Edward Lally, University of Pennsylvania School of
Dentistry) and using ammonium sulfate precipitation. Results are expressed as total PGE, although prior observations have shown that the major portion of PGE produced by
synovial cells in culture is PGEz (9).
Lysozyme activity was assayed in supernatants by
using suspensions of Micrococcus lysodeiktikus in 1% agar
(1). LDH was assayed in supernatants by measuring the
decrease in absorbance over time at 340 nm when 250 pI of
supernate was incubated with 150 pg/ml nicotinamide dehydrogenase and 150 ccg/ml of sodium pyruvate in PBS.
Con A binding assays. Monocytes were isolated as
previously described and plated at 4-8 x lo6 cells per well of
a 24-well cluster tray. Cells were always used the day after
isolation. All binding studies were performed at 4°C after
appropriate incubations were carried out. At time 0,0.2 pCi
of 'H Con A (28.8 CilmM, New England Nuclear, Boston,
MA) was added to each well. Nonspecific binding was
calculated as binding in the resence of cold Con A at 1,OOO
times the concentration of PH Con A. After 90 minutes at
4"C, cells were washed twice with cold HBSS and then lysed
with 200 pI 2% triton X100.One-hundred-microliter aliquots
were counted in a Packard Scintillation Counter.
Chloroquine inhibition of MCF production. Results of chloroquine incubation with monocyte-macrophage preparations are shown in Table 1. Chloroquine
w a s present throughout t h e 48-hour incubation,
whether thcre was a preincubation period or not.
Control monocyte populations produced measurable
amounts of MCF activity. Addition of Con A increased MCF activity, while chloroquine inhibited this
PGEz-stimulating activity. Chloroquine also inhibited
monocyte PGE2 production, but in amounts insufficient t o affect subsequent target cell measurements.
Results could not be ascribed to cell death (Table 2) or
chloroquine effect on target cell PGE production.
Inhibition of MCF activity occurred at concentrations as low as 15 pghl and was dose-dcpendent
and maximal between 15 and 25 pg/ml. When CQ was
positive CRP, and 3 had a constantly negative CRP.
Among the 21 patients with the lowest C l q levels
(group HI), 7 had an intermittently positive CRP, while
14 always had a negative CRP.
In the same patients, the NJE was determined
from radiographs and correlated with the initial C l q
value (Figure 2). In all but I of the patients with the
highest initial Clq levels (group I), there were more
joints with erosion than in those in groups I1 and 111.
Patients in group I had a mean NJE of 26.8 2 2.0 SE
(Table 3). The mean NJE in groups I1 and 111 was 5.3
2 0.7 and 4.0 ? 0.4, respectively. These latter 2 values
did not differ significantly ( P > 0.05). In Table 3, the
NJE values for the 38 patients in groups I1 and 111 are
combined, with a mean NJE of 4.6 ? 0.4 SE. There is
a highly significant difference in NJE when this combined group is compared with group I ( P < 0.001).
Time ( ~ o n t h )
Figure 1. Serial measurements of C l q , C-reative protein (CRP).
and erythrocyte sedimentation rate (ESR) in 2 representative patients who had major fluctuations in the ESR and CRP (shown in
parentheses) over a period of I year. Shaded areas represent the
mean valuc ? SD of Clq for normal controls.
during the first 4 years we observed him, and then his
Clq level began to decline gradually.
In each of these patients, the Clq level was
very elevated and remained constant during the period
of active RA, irrespective of fluctuations in the ESR
and CRP and of whether their disease was remitting or
Correlation between initial serum Clq, serial
CRP levels, and NJE. Patients with active RA were
divided into 3 groups based on the C l q levels at their
first clinic visit. Group I was those with Clq levels 250
pg/ml or higher (16 patients); group I1 was those with
Clq levels between 200 and 249 pglml (17 patients);
group 111 was those with C l q levels lower than 200 pg/
ml (21 patients).
Clq levels and CRP were measured serially for
5 years. As shown in Table 2, 14 of 16 patients in group
1 had a positive CRP at every measurement and the
other 2 occasionally had a negative CRP during approximately 5 years of this study. Among the 17
patients with intermediate initial C l q levels (group II),
I had a constantly positive CRP, 13 an intermittently
High levels of Clq in patients with RA could
have some relationship to the inflammatory reaction
and tissue damage in this disease. There are at least
two possible ways that C I q could participate in the
pathogenesis of RA. One is through the classical
pathway of the complement (C) activation (8), and the
other is through stimulation of peripheral blood mononuclear cells (9,10), polymorphonuclear cells (1 I ) , and/
or endothelial cells ( I 2). There arc reports concerning
the production and secretion of C l q by fibroblasts
(13.14) and macrophages (IS), the interaction of human plasma fibronectin with C l q (16), and the cornpetitive inhibition of collagen-induced platelet aggregation by Clq (17,18). These reports suggest that Clq
has a relationship to connective tissue physiology.
How Clq participates in the damage to connective
tissue is still unclear, but it is possible that Clq plays a
role in the destruction of connective tissue not only
through the classical pathway (8) of C in the presence
of immunoglobulin, but also through some receptormediated mechanism independent of immunoglobulin.
Table 2. Correlation between the initial serum Clq level and serial
C-reactive protein (CRP) levels
Figure 2. Correlation between the Clq level and the number of
joints with erosion. (Groups I. 11. and 111 contain patients with Clq
levels equal to or more than 250 pg/ml, between 200 and 249 pdml.
and less than 200 pg/ml, respectively.) Solid horizontal lines represent mean values; broken horizontal lines represent 1 SD.
Thus, Clq may play a more important role in the
pathogenesis of RA than was previously suspected.
This study has shown that Clq levels in
“burned out” RA remain in the normal range and that
the high levels found during the active disease period
return to the normal range at the “burned out” stage.
These findings suggest that serum C Iq has a relation to
active inflammation, although the mechanism is unknown at present.
Other studies have shown that ESR and CRP
seem to correlate with disease activity (19,20). Serial
comparison of Clq levels and the ESR and CRP in
individual RA patients in this study showed that the
ESR and CRP fluctuated with exacerbation or remisTable 3. Correlation between the initial serum Clq level and the
number of joints with erosion
No. of joints with erosion
Serum Clq level
2250 ccg/ml (16)
<250 d m l (38)
P < 0.001 between the 2 groups.
sion, whereas Clq levels remained stable (Figure 1).
Thus, Clq does not seem to be a typical acute inflammatory protein. Retrospective examination of Clq
levels before “burn out” of RA, however, has shown
that the Clq level reflects very well the degree of
continuation of active inflammation over a period of 5
years. The extent of later erosive disease appears to be
predictable from the Clq level at the first clinic visit
irrespective of whether the CRP is high, with 250 pg/
ml of Clq as the boundary level (Table 2 and Figure 2).
It is generally believed that prognosis in a
disease is affected by the treatment. However, since
our data indicate that the prognosis for a patient with
RA can be determined by the Clq level in the early
stage of disease, one might consider that RA can be
classified into two subsets with different natural histories. One subset, RA with a Clq level 2250 pg/ml, is
disease with continuance of severe inflammation; another, with a Clq level <250 &ml, is disease without
continued inflammation.
The normal level of Clq found in “burned out”
RA might be relevant to our finding that the abnormally high level of Clq in the early stages of active RA
sometimes begins to decline toward normal after 7
years of disease (Ochi T: unpublished data), possibly
reflecting the beginning of the “burned out” stage.
This does not contradict the fact that each patient
whose Clq level was higher than 250 pg/ml at the first
clinic visit had extensive erosive joint destruction after
7 years of disease, whereas only a few joints were
similarly involved in patients whose Clq level was
lower than 250 CLg/ml (Figure 2). It could further
indicate the possible existence of two subsets of RA
with different natural histories.
A conspicuously high titer of CRP (21) and/or
the continuance of a significantly high level of CRP
(22) have been reported as clues to the prediction of
prognosis in RA. Our data suggest that the Clq level in
active RA may better reflect the continuance of significant inflammation and that measurement of Clq early
in the disease can separate RA into two subsets with
different prognoses. Thus, the early measurement of
Clq is a very simple, rapid, and useful tool to predict
the extent ofjoint erosions in RA. A prospective study
is now being undertaken to confirm these findings.
The authors wish to thank Dr. Chester W. Fink,
Professor of Pediatrics, University of Texas Health Science
Center at Dallas, for great help in preparing the manuscript,
and Miss Y. lnagaki for typing the manuscript.
1. Yonemasu K, Kitajima H, Tanabe S, Ochi T, Shinkai H:
Effect of age on Clq and C3 levels in human serum and
their presence in colostrum. Immunology 35:523-530,
Ochi T, Yonemasu K, Ono K: lmmunochemical quantitation of complement components of Clq and C3 in sera
and synovial fluids of patients with bone and joint
diseases. Ann Rheum Dis 39:235-240, 1980
Ropes MW, Bennett GA, Cobb S. Jacox R, Jessar RA:
1958 revision of diagnostic criteria for rheumatoid arthritis. Bull Rheum Dis 9:175-176, 1958
Mancini G, Carbonara AO, Hercmans JF: Immunochemical quantitation of antigens by single radial
immunodfiusion. Immunochemistry 2:235-254, 1965
Anderson HC, McCarty M: Determination of C-reactive
protein in the blood as a measure of the activity of the
disease process in acute rheumatic fever. Am J Med
8:445-455, 1950
The Cooperating Clinics Committee of the American
Rheumatism Association: A seven-day variability study
of 499 patients with peripheral rheumatoid arthritis.
Arthritis Rheum 8:302-334, 1965
Steinbrocker 0, Traeger CH, Batterman RC: Therapeutic criteria in rheumatoid arthritis. JAMA 140:659-662.
Whaley K, El-Ghobarey A: Complement, Immunological Aspects of Rheumatology. Edited by WC Dick.
Lancaster, England, MTP Press. 1981. pp 93-121
Tenner AJ, Cooper NR: Analysis of receptor-mediated
Clq binding to human peripheral blood mononuclear
cells. J Immunol 125:1658-1664, 1980
Tenner AJ, Cooper NR: Identification of types of cells in
human peripheral blood that bind Clq. J Immunol
126:1174-1179, 1981
Tenner AJ, Cooper NR: Stimulation of a human polymorphonuclear leukocyte oxidative response by the C lq
subunit of the first complement component. J lmmunol
I28:2547-2552, I982
12. Andrews BS, Shadforth M. Cunningham P,Davis JS IV:
Demonstration of a Clq receptor on the surface of
human endothclial cells. J Immunol 127: 1075-1080, 1981
13. Al-Adonani MS, McGee JO: Clq production and secretion by fibroblasts. Nature 263: 145-146, 1976
14. Reid KBM, Solomon E: Biosynthesis of the first component of complement by human fibroblasts. Biochem J
167:647-660. 1977
IS. Muller W, Hanauske-Abel H, Loos M: Biosynthesis of
the first component of complement by human and guinea
pig peritoneal macrophages: evidence for an independent production of the CI subunits. J lmmunol
121:1578-1584, 1978
16. Pearlstein E, Sovillo J, Gigli I: The interaction of human
plasma fibronectin with a subunit of the first component
of complement, Clq. J Immunol 128:2036-2039, 1982
17. Suba EA. Csako G: C l q (Cl) receptor on human platelets: inhibition of collagen-induced platelet aggregation
by Clq (CI) molecules. J Immunol 117:304-309, 1976
18. Wautier JL, Souchon H, Reid KBM, Peltier AP, Caen
JP: Studies on the mode of reaction of the first component of complement with platelets: interaction between
the collagen-like portion of Clq and platelets. Immunochemistry 14:763-766, 1977
19. McConkey B, Crockson RA, Crockson AP. Wilkinson
AR: The effect of some anti-inflammatory drugs on the
acute-phase proteins in rheumatoid arthritis. Q J Med
42:785-791, 1973
20. Maddock J, Rees P, Hooly F. Aylward M: The influence
of alcofenac treatment on acute-phase proteins, plasma
tryptophan, and erythrocyte sedimentation rate in patients with rheumatoid arthritis. Curr Med Res Opin
3:286-297, 1975
21. Nusinow S, Arnold WJ: Prognostic value of C-reactive
protein levels in rheumatoid arthritis (abstract). Arthritis
Rheum (suppl) 25:S24, 1982
22. Amos RS, Constable TJ, Crockson RA, Crockson AP,
McConkey B: Rheumatoid arthritis: relation of serum Creactive protein and erythrocyte sedimentation rates to
radiographic changes. Br Med J I:195-197, 1977
Без категории
Размер файла
467 Кб
prognostic, clq, level, patients, guide, arthritis, serum, erosion, articular, rheumatoid
Пожаловаться на содержимое документа