вход по аккаунту


Release of cartilage macromolecules into the synovial fluid in patients with acute and prolonged phases of reactive arthritis.

код для вставкиСкачать
Objective. Extensive changes in articular cartilage metabolism occur during the acute phase of reactive
arthritis, as indicated by altered release of cartilage
macromolecules into synovial fluid (SF) demonstrated
immunochemically. Nevertheless, permanent cartilage
lesions are rare in this disease. To monitor specific
events during the evolution of reactive arthritis, we
investigated the content of cartilage macromolecules in
sequentially obtained SF samples from 22 patients.
Methods. Two groups of proteoglycan epitopes,
the glycosaminoglycan-rich region of aggrecan (referred
to as proteoglycan) and its hyaluronan-binding region
(HABr), as well as one matrix protein, cartilage oligoFrom the Departments of Rheumatology and Physiological
Chemistry, Lund University, Lund, Sweden, the Oslo City Department of Rheumatology, Norwegian Lutheran Hospital, and the
Department of Microbiology, Ulleva University Hospital, Oslo,
N orwdy.
Supported by the Medical Faculty of Lund, the Swedish
Medical Research Council, CIBA-Geigy Corporation, Alfred Osterlunds Stiftelse, Greta och Johan Kocks Stiftelser, Riksforbundet
mot Reumatism, Crafoords Stiftelse, Lasarettets i Lund fonder,
Konung Gustaf V:s 80-irsfond, Axel och Margareta Ax:son
Johnsons Stiftelse, and The German Rheumatism Research Centre,
Berlin, FRG. The clinical portion of the study was supported by a
research grant from Pfizer AS.
Tore Saxne, MD, PhD: Departments of Rheumatology and
Physiological Chemistry, Lund University; Anne Glennis, MD,
PhD: Oslo City Department of Rheumatology, Norwegian Lutheran
Hospital; Tore K. Kvien, MD, PhD: Oslo City Department of
Rheumatology, Norwegian Lutheran Hospital; Kjetil Melby, MD,
PhD: Department of Microbiology, Ullev&l University Hospital;
Dick Heinegird, MD, PhD: Department of Physiological Chemistry,
Lund University.
Address reprint requests to Tore Saxne, MD, PhD, Department of Rheumatology, Lund University Hospital, s-221 85, Lund,
Submitted for publication June 3, 1992; accepted in revised
form August 27, 1992.
Arthritis and Rheumatism, Vol. 36, No. 1 (January 1993)
meric matrix protein (COMP), were quantified by immunoassay.
Results. SF proteoglycan concentrations, which
were initially elevated, decreased significantly with prolonged arthritis, whereas COMP levels changed less
markedly and levels of HABr remained stable. There
was a positive correlation between SF and serum concentrations of COMP in samples obtained during the
early phase of the disease.
Conclusion. Cartilage involvement in reactive
arthritis is transient, in contrast to findings in rheumatoid arthritis. Reactive arthritis should therefore be a
suitable model for studies of repair processes in cartilage, which will facilitate understanding of the pathophysiology of cartilage involvement in arthritis.
Reactive arthritis is characterized by the acute
onset of arthritis in one or a few joints following an
infection, most often of the urinary or gastrointestinal
tract (1). Viable organisms are not found in the joint.
In the majority of cases the disease is self-limited, and
radiographically visible joint damage rarely develops.
Nevertheless, previous work has indicated that articular cartilage is involved in the disease process (2).
Major constituents of the organic matrix of
articular cartilage are type I1 collagen fibers and the
cartilage-specific, large aggregating proteoglycan, aggrecan (3). The function of cartilage depends on both
the highly polyanionic proteoglycan aggregates and
the collagen fibers. A number of other matrix constituents have essential roles in regulating the assembly of
the matrix and in participating in interactions necessary for the integrity of the tissue. One such noncol-
lagenous, cartilage-specific protein is cartilage oligomeric matrix protein (COMP) (4-7). This protein consists of 5 subunits with apparent M, of 100 kd. It is
somewhat more abundant in articular cartilage than in
other cartilages. The functions of COMP are presently
unclear (4).
By quantifying cartilage components released
into synovial fluid (SF), with the use of an immunoassay, we have found that extensive cartilage involvement occurs early in reactive arthritis, despite the
favorable prognosis of the disease. SF content of the
glycosaminoglycan-rich region of the core protein of
the proteoglycan (for simplicity here referred to as
proteoglycan) and of COMP is increased in acute
stages of the disease, which suggests rapid initial loss
of matrix components (6). Results of a cross-sectional
study indicate that the release of these components
decreases during recovery, which suggests that the
release reflects acute changes in cartilage metabolism
and that compensatory mechanisms, which promote
synthesis of matrix components, occur in this disease
(6). We have also found that release of the hyaluronanbinding region (HABr) of the proteoglycan is not
increased either in the acute stage of reactive arthritis
or in early rheumatoid arthritis (RA). However, in
later stages of RA, increased SF concentrations of
HABr are seen, which suggests that release of this
portion of the proteoglycan reflects advanced cartilage
damage (8).
In the present study we measured concentrations of proteoglycan, HABr, and COMP, in sequentially obtained SF samples from patients with reactive
arthritis, to elucidate whether the increased levels of
proteoglycan and COMP found during the acute stage
of the disease reflect cartilage involvement, which
subsequently subsides. We further wanted to determine whether initial levels of these markers have any
prognostic implications with regard to total duration of
arthritis, and whether the pattern of release is related
to the type of precipitating agent or to the inflammatory process as assessed by various traditional markers. Serum levels of COMP were measured to determine whether they correlate with SF COMP levels,
which would open the possibility of using blood samples for monitoring the disease process.
Patient characteristics and SF collection. Subjects in
the present study were identified from a study on the
epidemiology of reactive arthritis in Oslo, from March 1988
to March 1990. The latter study included 186 consecutive
patients between 18 and 60 years old, who presented with
possible reactive arthritis (9). A subgroup of 22 of these
patients was selected for the present investigation. These
patients had gonarthritis with synovial effusion, which required 2 or more joint fluid aspirations at intervals of at least
10 days. All had either Chlamydia-induced arthritis proven
by urethralkervical culture (Chlamydia trachomatis, n = 9),
serologically or culture-proven enteroarthritis (Salmonella
species, n = 4; Yersinia enterocolitica, n = l), or possible
reactive arthritis, defined as arthritis triggered by an unknown agent (negative findings on culture and serologic
tests) with clinical remission within 24 weeks (n = 8).
Findings on radiographs of the affected knee joints, taken at
entry into the study, were normal in all cases. N o patient
received intraarticular or oral glucocorticoid treatment during the study period.
There were 8 women and 14 men in the patient
group. The median age was 26 years (range 19-50 years), and
the median duration of arthritis at the first joint aspiration
was 1 week (range 1-19 weeks).
Synovial fluid was collected in heparinized tubes, centrifuged, and then stored at -70°C until examination. Blood
was collected in tubes without additives, allowed to clot,
centrifuged, and stored similarly to the SF until analysis.
Immunoassays. SF concentrations of proteoglycans
were quantified by enzyme-linked immunosorbent assay
(ELISA) (lo), using goat polyclonal antibodies against purified human proteoglycan monomer, which preferentially
recognize the chondroitin sulfate-rich region of the core
protein of the large, aggregating proteoglycan of cartilage
(8). SF concentrations of HABr were measured in an ELISA
using antibodies that selectively react with epitopes in the
HABr, prepared from antiproteoglycanantibodies by affinity
chromatography (8). SF and serum concentrations of COMP
were determined with an ELISA using rabbit polyclonal
antibodies against the bovine protein. Microtiter plates were
coated with purified human COMP, and a standard of
dilutions of human COMP was included in each plate (6).
Analysis of inflammation markers. Serum and SF
white blood cells (WBC) were counted with an STKS
counter (Coulter, Hialeah, FL). The erythrocyte sedimentation rate (ESR) was determined by the Westergren method.
C-reactive protein (CRP) was quantified by a turbidometric
method (TDX turbidometer; Abbott, Irving, TX). Morning
stiffness was reported by the patient as a mean of the
duration (minutes) of morning stiffness on each day during
the week prior to the examination. The degree of pain was
indicated on a 1Wmm visual analog scale. The joint index
was a modified Lansbury index (1 1,12). An “active” joint
was defined as a swollen joint andor a joint with pain or
tenderness in combination with limited motion.
Statistical analysis. Correlations were calculated using Spearman’s correlation coefficient. Comparisons between groups were calculated using the Mann-Whitney U
test (2-tailed), and comparisons between SF and serum
concentrations of COMP were made using the Wilcoxon
matched pairs test (2-tailed). P values less than 0.05 were
considered significant.
Duration of arthritis (weeks)
Duration of arthritis (weeks)
Figure 1. Concentration of proteoglycan (glycosaminoglycan-rich
region) by duration of arthritis at the time of joint aspiration, in
synovial fluid (SF) from the knees of individual patients with
reactive arthritis. Note the logarithmic scale of the horizontal axis.
SF proteoglycan content. The proteoglycan concentration in SF was elevated initially, but decreased
with time (r, = -0.3402, P < 0.02) (Figure 1). The
differences between acute and later stages of disease
were even more pronounced when the data were
corrected for the amount of SF aspirated (r, =
-0.4139, P < 0.01) (data not shown).
SF HABr content. Neither the concentration nor
the total content of proteoglycan HABr in SF changed
significantly during the observation period (r, =
-0.0633 and rs = -0.1700, respectively) (Figure 2).
These concentrations were in a similar range to those
observed in RA patients without extensive joint destruction (8).
Synovial fluid and serum COMP content. There
was no significant change in either the concentrations
or the total content of SF COMP during the observation period (r, = -0.0689 and rs = -0.1266, respectively) (Figure 3). Although the COMP level appeared
to decrease during the early phases of reactive arthritis, statistical analysis showed that the change was not
significant. As expected, the S F concentration of
COMP was higher than the serum concentration in all
Figure 2. Concentration of the hyaluronan-binding region of proteoglycan (HABr) by duration of arthritis at the time of joint
aspiration, in synovial fluid (SF) from the knees of individual
patients with reactive arthritis. Note the logarithmic scale of the
horizontal axis.
cases (P < 0.001). There was a significant positive
correlation between serum concentrations of COMP
and the concentrations of this protein in the affected
joint fluid (r, = 0.5632, P < 0.01) (Figure 4). Overall,
however, the concentrations of COMP in the serum
Duration of arthrith (weeks)
Figure 3. Concentration of cartilage oligomeric matrix protein
(COMP) by duration of arthritis at the time of joint aspiration, in
synovial fluid (SF) from the knees of individual patients with
reactive arthritis. Note the logarithmic scale of the horizontal axis.
8 1 0
S-COMP (UQ/ml)
Figure 4. Synovial fluid (SF) and serum ( S ) concentrations of
cartilage oligomeric matrix protein (COMP) at the time of the first
knee joint aspiration, in individual patients with reactive arthritis. A
positive correlation is seen (r, = 0.5632, P < 0.01).
from reactive arthritis patients were not increased
compared with levels in healthy blood donors (6).
Ratio between SF proteoglycan Concentrations
and SF COMP concentrations. Analysis of multiple SF
components introduces advantages in that ratios can be
calculated. Such ratios are less sensitive to altered volume and turnover of SF. Calculation of the SF proteog1ycan:COMP ratio over time revealed a significant
correlation (r, = -0.4287, P < 0.01), providing further
verification of the major alterations in proteoglycan
release occurring in reactive arthritis (Figure 5).
Relationship between initial SF proteoglycan or
COMP levels and total duration of arthritis or type of
precipitating agent. The median total duration of arthritis was 21 weeks (range 4-108 weeks). We found no
significant correlation between the initial concentration of proteoglycans or COMP, or the initial ratio
between the concentrations of these markers, and total
duration of arthritis in the 16 patients whose duration
of synovitis was <2 weeks at the initial sampling (data
not shown). Initial levels of the cartilage markers in
this subgroup did not differ between patients whose
disease was precipitated by Chlamydia and those in
whom no causative agent could be identified (n = 6 in
each group). Four of these 16 patients had disease
precipitated by a serologically verified gastrointestinal
Duration of arthritis (weeks)
Figure 5. Ratio of the synovial fluid (SF) concentration of proteoglycan (glycosaminoglycan-rich region) to the SF concentration of
cartilage oligomeric matrix protein (COMP),by duration of arthritis
at the time of knee joint aspiration, in individual patients with
reactive arthritis. Note the logarithmic scale of the horizontal axis.
infection (Salmonella infection in 3 and Yersinia infection in 1). Levels of the cartilage markers in these
patients did not differ from those found in the other
Relationship between levels of the cartilage
markers and clinical and biochemical markers of inflammation. Laboratory and clinical data indicating the
degree of inflammation at the time of the first joint
aspiration are listed in Table 1- The SF concentrations
of proteoglycan, HABr, or COMP did not correlate
Table 1. Indicators of inflammation in the 22 patients with reactive
SF WBC, 109Aiter
Serum WBC, 109iliter
Serum CRP, mgAiter
ESR, mmihour
No. of active joints
Joint indext
Morning stiffness, minutes
Joint pain$
10.5 (2.6-25.4)
8.5 (5.2-13.8)
36 (0-278)
34 (6-130)
75 (25-150)
10 (0-240)
52 (2-85)
* Values are the median (range). SF = synovial fluid; WBC = white
blood cells; CRP = C-reactive protein; ESR = erythrocyte sedimentation rate.
t Modified Lansbury index (1 1.12).
$ Visual analog scale, where 0 = no pain, 100 = maximum pain.
with either ESR or with serum CRP. Furthermore, the
SF WBC count did not relate to SF concentrations of
any of the cartilage markers. Serum concentrations of
COMP did not correlate with ESR, with serum CRP,
or with the joint index or number of active joints.
Levels of the cartilage markers also did not correlate
with duration of morning stiffness or with degree of
joint pain as measured on a visual analog scale.
This study represents the first attempt to monitor cartilage involvement in reactive arthritis by longitudinally quantifying cartilage constituents released
into synovial fluid. We found that cartilage involvement, as shown by increased release of proteoglycan
fragments and COMP, occurs early in the disease,
which corroborates the findings in a cross-sectional
study (6). We also found that these concentrations
decrease during recovery, but this is statistically significant only for proteoglycan, which thus should be a
marker for early, apparently reversible cartilage damage that presumably does not seriously involve key
structures. In contrast, concentrations of the HABr
did not change during the study period, which suggests
that SF levels of this portion of the proteoglycan
reflect the normal continuous remodeling of cartilage
in these patients, and that increased release of HABr is
characteristic of more extensive cartilage damage (8).
The pattern of release of COMP and proteoglycan in reactive arthritis resembles that in RA, albeit
with a different time scale. The reasons for this pattern
appear to differ between the two diseases, however. In
reactive arthritis, the decreasing concentrations of
cartilage components in SF probably reflect subsiding
disease activity, with restoration of cartilage integrity.
This notion is supported by the fact that the SF
concentration of HABr does not increase during the
course of the disease, whereas the progressive tissue
destruction in RA corresponds to increasing SF HABr
levels (8). In addition, SF concentrations of proteoglycan and COMP during the recovery phase of reactive
arthritis are similar to those in healthy volunteers
(Saxne T, HeinegHrd D: unpublished observations).
Thus, the restored pattern of cartilage components in
the SF in reactive arthritis appears to follow from
normalization of the metabolism of matrix macromolecules.
The SF concentrations of proteoglycans and
COMP in the early stages of reactive arthritis correspond to those previously found not only in this
condition but also in other acute arthritides, such as
crystal arthropathies and transient synovitis of the hip
in children (2,13,14). This indicates that cartilage damage occurs in acute arthritis, regardless of etiopathology, and that the damage is, with few exceptions,
reversible. The pathophysiologic mechanisms in these
diseases thus seem different from those in RA, where
cartilage involvement also can be seen early, but
where the repair mechanisms are in many cases insufficient to prevent progressive joint destruction.
We found a positive correlation between SF
and serum levels of COMP, which supports our findings in a previous study (6). This indicates that a
significant proportion of the circulating antigens were
released from the affected joint. However, the serum
levels were not elevated compared with those in
healthy blood donors, which limits their diagnostic
value. In patients with monarthritis, however, repeated serum measurements of COMP might be useful
for monitoring the local disease process. We did not
measure serum concentrations of proteoglycans in this
study, but we have previously found these concentrations to be close to or below the limit of detection, and
not correlating with SF levels (Saxne T et al: unpublished observations).
The SF concentrations of proteoglycans or
COMP at the initial joint aspiration did not relate to
total duration of arthritis. The prognostic value of
measuring these markers thus seems limited in reactive arthritis. Since none of the patients had radiographically visible joint changes during the observation period, it was not possible to assess the relation
between initial release of cartilage constituents and
permanent joint damage. We could not find any difference in SF levels of the cartilage markers between
patients whose disease had been caused by a verified
Chlamydia infection, a gastrointestinal infection, or
those in whom no causative agent was identified.
Although the patient groups were small, this suggests
that similar mechanisms are operating regardless of
the triggering agent, which corroborates the abovediscussed hypothesis that the changes reflect the acute,
reversible nature of the disease. However, it must be
cautioned that the group of patients in whom no causative agent was identified could include a few with
undiagnosed Chlamydia or gastrointestinal infection.
In accordance with previous findings (12), concentrations of the cartilage markers did not correlate
with standard biochemical or clinical measures of
inflammation. This emphasizes that quantification of
these tissue-specific macromolecules reflects processes
taking place locally in the joint, whereas the ESR and
CRP reflect degrees of generalized inflammation.
By combining analysis of tissue-specific markers and analysis of markers of inflammation, it will be
possible to monitor both local tissue destruction and
generalized inflammation. The finding of different patterns of release of cartilage markers in reactive arthritis a n d RA makes reactive arthritis a suitable model for
further studies to elucidate the basic pathophysiologic
mechanisms of tissue destruction and repair in arthritis. The ultimate goal of such studies is to develop
methods for preventing tissue destruction.
We are grateful to Mette Lindell and Kirsten Mossin
for their skillful technical assistance. We thank our colleagues, Drs. 0. Andrup, B. Karstensen, and J. E. Thoen,
who participated in the clinical portion of the study.
1 . Ahvonen P, Sievers K, Aho K: Arthritis associated with
Yersinia enterocolitica infections. Acta Rheumatol
Scand 15:232-255, 1969
2. Saxne T, Heineglrd D, Wollheim FA: Cartilage proteoglycans in synovial fluid and serum in patients with
inflammatory joint disease: relation to systemic treatment. Arthritis Rheum 30:972-979, 1987
3. HeinegHrd D, Paulsson M: Cartilage: structural and
contractile proteins. D. Extracellular matrix. Methods
Enzymol 145:336363, 1987
4. Hedbom E, Antonsson P, Hjerpe A, Paulsson M, RosaPimentel E, Sommarin Y, Wendel M, Oldberg A, Heineglrd D: Cartilage matrix proteins: an acidic oligomeric
protein (COMP) detected only in cartilage. J Biol Chem
267:6132-6136, 1992
5 . Morgelin M, HeinegHrd D, Engel J, Paulsson M: Elec-
tron microscopy of native cartilage oligomeric matrix
protein purified from the Swarm rat chondrosarcoma
reveals a five-armed structure. J Biol Chem 267:61376141, 1992
6. Saxne T , HeinegPrd D: Cartilage oligomeric matrix
protein: a novel marker of cartilage turnover detectable
in synovial fluid and blood. Br J Rheumatol 31583-591,
7. Fife RS, Brandt KD: Identification of a high-molecularweight (>400 OOO) protein in hyaline cartilage. Biochim
Biophys Acta 802506514, 1984
8. Saxne T, Heineglrd D: Synovial fluid analysis of two
groups of proteoglycan epitopes distinguishes early and
late cartilage lesions. Arthritis Rheum 35:385-390, 1992
9. Kvien TK, GlennHs A, Melby K, Granfors K , Andrup
0, Karstensen B, Thoen JE: Reactive arthritis: incidence, triggering agents and clinical presentation. In
10. Saxne T, HeinegHrd D, Wollheim FA: Therapeutic effects on cartilage metabolism in arthritis as measured by
release of proteoglycan structures into the synovial
fluid. Ann Rheum Dis 45:491-497, 1986
1 1 . Lansbury J: Report of a three-year study on the systemic and articular indexes in rheumatoid arthritis:
theoretic and clinical considerations. Arthritis Rheum
1505-522, 1958
12. Kvien TK, Hoyerall HM, Sandstad B: Assessment
methods of disease activity in juvenile rheumatoid arthritis: evaluated in a prednisolone/placebo double-blind
study. J Rheumatol9:696702, 1982
13. Ratcliffe A, Doherty M, Maini RN, Hardingham TE:
Increased concentrations of proteoglycan components
in the synovial fluids of patients with acute but not
chronic joint disease. Ann Rheum Dis 47:82&832, 1988
14. Lohmander LS, Wingstrand H , Heinegird D: Transient
synovitis of the hip in the child: increased levels of
proteoglycan fragments in joint fluid. J Orthop Res
6:420424, 1988
Без категории
Размер файла
531 Кб
patients, release, macromolecules, arthritis, prolonged, synovial, cartilage, acute, reactive, phase, fluid
Пожаловаться на содержимое документа