вход по аккаунту


Synovial interleukin-1 receptor antagonist and interleukin-1 balance in rheumatoid arthritis.

код для вставкиСкачать
Number 5 , May 1994, pp 644-652
0 1994, American College of Rheumatology
Objective. To quantify interleukin-1 receptor antagonist (IL-lra) and IL-1 production and gene expression by rheumatoid arthritis (RA) synovial tissue (ST)
Methods. IL-la, IL-lP, and IL-lra protein levels
were measured by enzyme-linked immunosorbent assay
in fresh and cultured ST cells, purified synovial macrophages, and fibroblast-like synoviocytes (FLS). The
relative expression of the secreted form of IL-lra (sILIra) and the alternatively spliced intracellular form
(icIL-lra) was determined by reverse transcription polymerase chain reaction (RT-PCR) techniques.
Results. IL-la, IL-lP, and IL-lra were present
in fresh and cultured ST cell samples of synovium from
RA and osteoarthritis patients. IL-1ra:IL-1 ratios
ranged from 1.2 to 3.6, which is below the l&lOO-fold
excess of IL-lra needed to inhibit IL-1 bioactivity.
Isolated CD14+ synovial macrophages secreted IL-lra,
but the amount was much less than that of alveolar or in
vitro-derived macrophages. Cultured FLS contained
intracellular IL-lra but secreted little IL-lra into the
culture supernatants. RT-PCR showed that icIL-lra
Supported in part by NIH grants AR-40525 (Dr. Firestein),
AR-39576 (Dr. Firestein), AR-39950 (Dr. Arend), and AR-40135 (Dr.
Arend), and a grant from H z e r Central Research (Dr. Firestein).
Gary S. Firestein, MD: University of California at San
Diego Medical Center, and Gensia, Inc., San Diego, California;
David L. Boyle, BA: University of California at San Diego Medical
Center; Carolyn Yu, BA: University of California at San Diego
Medical Center; Mary M. Paine, BA: University of California at San
Diego Medical Center; Teri D. Whisenand, BS: University of
Colorado Health Sciences Center, Denver; Nathan J. Zvaifler, MD:
University of California at San Diego Medical Center; William P.
Arend, MD: University of Colorado Health Sciences Center.
Address reprint requests to Gary S. Firestein, MD, Gensia,
Inc., 9360 Towne Centre Drive, San Diego, CA 92121.
Submitted for publication February 28, 1993; accepted in
revised form October 12, 1993.
mRNA was more abundant than sIL-lra mRNA in FLS
and unfractionated ST cells.
Conclusion. IL-lra production by RA ST cells is
deficient relative to total production of IL-1.
Cytokine networks play an essential role in the
pathogenesis of rheumatoid arthritis (RA) (1). Of the
factors identified in arthritic joints, interleukin-1 (IL-1)
appears to be especially important because it stimulates fibroblast-like synoviocytes (FLS) to proliferate
and secrete metalloproteinases, prostaglandins, and
cytokines (2,3). IL-1 also directly induces arthritis
when injected into joints (4) and exacerbates experimental antigen-induced arthritis (5). However, there
are multiple cytokines in the rheumatoid joint, and the
proinflammatory effects of IL-1 might be modified by
the presence of IL-1 receptor antagonists (IL-lra) and
other cytokines such as transforming growth factor /3
(TGFP) (6).
IL-lra is a specific receptor antagonist of IL-1
that was originally described in urine and in the
supernatants of human monocytes and macrophages
(7,8). This protein binds to types I and I1 IL-1 receptors without activating target cells and serves as a
competitive inhibitor of IL-1 (9-14). Two structural
variants of IL-lra have been described: sIL-lra, or
secretory IL-lra, which is synthesized with a leader
peptide and is a major product of monocytes and
macrophages; and icIL- Ira, or intracellular IL- Ira,
which lacks a leader peptide and remains intracellular
in keratinocytes and other epithelial cells (15,16). Both
forms of IL-lra may be produced simultaneously by
fibroblasts or alveolar macrophages (17,18).
IL-1 and IL-lra are present in the joints of
patients with RA (19-25). Rheumatoid synovial tissue
through a nylon mesh, and extensively washed. Previous
studies have shown that this population of rheumatoid
synovial tissue cells comprises approximately 20% macrophages, 30-50% T cells, and 20% fibroblast-like cells (3). OA
synovial tissue has a lower percentage of T cells (10-20%)
and higher percentage of macrophages (3040%) (3).
Aliquots of freshly isolated cell pellets were immediately snap frozen in liquid nitrogen. Unseparated cells were
also cultured at 2.5 x lo6 celldm1 in DMEM supplemented
with 10% fetal calf serum ([FCS], endotoxin content <0.006
ng/ml; Gibco), penicillin, streptomycin, and L-glutamine in a
humidified 5% C 0 2 atmosphere for 3 days. The supernatants
were harvested and snap frozen after separating the nonadherent cells by centrifugation at 400g. The adherent cells
were subsequently harvested from the plates using trypsitd
EDTA. These cells were then combined with the nonadherent population, aliquoted, and snap frozen.
CD 14-positive ST macrophages were separated from
the unfractionated cells using anti-CD14 antibody conjugated to magnetic beads as described by the manufacturer
(Biomag Beads; Collaborative Research, Bedford, MA).
Briefly, ST cells were incubated in phosphate buffered saline
supplemented with 5% human AB serum for 20 minutes to
block nonspecific binding. Anti-CDlkonjugated magnetic
beads were washed several times and added to the cells at a
ratio of approximately 5 beads per macrophage. The suspension was incubated on ice for 1 hour and inverted every 15
minutes. The CD14+ cells were separated using a magnet,
washed 3 times, and cultured for 2-3 days at 5 x lo5 cells/ml
in DMEM plus 10% FCS (which approximates the number of
Cytokines. Recombinant human tumor necrosis facmacrophages in cultures of unseparated ST cells). Supernator a (TNFa; specific activity 5 X lo7 units/mg, purity
tants and cell pellets were harvested and snap frozen.
>95%, lipopolysaccharide [LPS] <0.008 ng/mg) and recomFLS were obtained by culturing unfractionated
binant human IL-IP (specific activity 2 x lo7 unitshg,
synovial cells overnight at 1 x lo6 celldml. Nonadherent
purity >98%, LPS <0.048 ng/mg) were a gift from Amgen
cells were removed and adherent cells were cultivated in
Biologicals (Thousand Oaks, CA).
DMEM plus 10% FCS. At confluence, cells were
Complementary DNA (cDNA) probes and primers.
trypsinized, split at a 1:3 ratio, and recultured in medium.
The sIL-lra cDNA was obtained from Don Carter (The
FLS were used from passages 3 through 9 in these experiUpjohn Co., Kalamazoo, MI) (11). The icIL-lra cDNA was
during which time they were a homogeneous popuobtained from J. Stephen Haskill (Duke University,
lation of fibroblast-like synoviocytes (3). For IL-lra producDurham, NC) (16). The common 3’ primer for both forms of
IL-lra was S’-GTTCTACGTTCGGAAGTCTTAGACCC-3’. tion studies, FLS were grown to near confluence in 60-mm
petri dishes (approximately 2 x lo5 cells) and then treated
The upstream primer for icIL-lra was 5’-CAGGTACTGCwith
either 1 ng/ml of IL-lp, 100 ng/ml of TNFa, or medium
CCGGGTGCTACTTTAT-3’. The upstream primer for sILalone in a total volume of 3 ml for 48 hours. Supernatants
and cell pellets were harvested and frozen.
There was no cross-hybridization in the polymerase chain
Enzyme-linked immunosorbent assays (ELISAs). Spereaction (PCR) experiments (i.e., the icIL-lra primer did not
for IL-lra, IL-la, and IL-lp were used in these
amplify the sIL-lra cDNA and the sIL-lra primer did not
experiments. The IL-lra ELISA has recently been described
amplify the icIL-lra cDNA). The primers for the housekeepin detail (28) and used affinity-purified rabbit IgG antibodies
ing gene G3PDH were: 5’-CCCTTCATTGACCTCAACTAto IL-lra as the capture reagent and the biotinylated IgG
CATGGT-3’ and 5’-CATGGTGGTGAAGACGCCAGTGfraction of the antiserum as the secondary antibodies. This
GACT-3’. The amplified fragment for G3PDH is 211 baseELISA equally recognizes both structural variants of IL-lra
as well as the glycosylated and nonglycosylated forms of
Synovial tissue cells. ST cells were isolated by enzysIL-lra. The IL-la and IL-lp ELISA used monoclonal
matic dispersion of synovial tissues obtained from patients
antibodies as the capture reagents (compliments of Dr. Ann
with RA or OA undergoing joint surgery, as previously
Berger, Upjohn) and the biotinylated IgG fractions of spedescribed (26). The tissues were minced and incubated with
cific rabbit antisera as the secondary antibodies. The IL-la
1 mglml of collagenase (Worthington, Freehold, NJ) in
serum-free Dulbecco’s modified Eagle’s medium (DMEM;
ELISA recognizes equally well the 31-kd precursor and
Gibco, Grand Island, NY) for 2 hours at 37”C, filtered
17-kd mature forms of IL-la. However, as recently empha-
(ST) contains messenger RNA (mRNA) for IL-la and
IL-1p (26,27), and isolated tissue cells produce these
proteins in vitro. Both IL-1 and IL-lra proteins have
been localized to synovial lining cells and to perivascular macrophages in RA and osteoarthritis (OA)
synovium (22-24). Although we previously observed
biologically active IL-lra in supernatants of cultured
ST cells (22), the relative amounts of IL-1 and IL-lra
were not quantitatively determined.
Despite the apparent abundance of IL-lra in
rheumatoid synovium, we hypothesized that the concentration of IL-lra is too low to inhibit the activity of
IL-1. Hence, the present study was designed to investigate the absolute and relative amounts of IL-la,
IL-10, and 1L-lra proteins produced by cultured rheumatoid synovial cells using quantitative assays. In
addition, the relative amounts of mRNA for sIL-lra
and icIL-lra in synovial cells were also determined.
Our data indicate that synovial IL-Ira production is
insufficient to inhibit IL-1 biological activity. This is
due, in part, to alternative splicing of IL-lra mRNA to
the intracellular form and defective IL-lra production
b y synovial macrophages.
sized by other investigators (29), the IL-1p ELISA did not
fully detect the biologically inactive intracellular 3 1-kd precursor of IL-1p. Control experiments using Western blot
analyses indicated that only about 10% of the 31-kd form of
IL-1p was detected by our ELISA (Dr. Robert W. Janson,
University of Colorado Health Sciences Center, Denver,
CO). Furthermore, 95% of the IL-lp in synovial cell supernatants was the 17-kd mature form, whereas up to 50% of the
IL-lp in the cell lysates was the 31-kd precursor.
The sensitivities of the 3 ELISAs were 20 pg/ml.
Proteins were measured in cell supernatants and lysates. An
amount of lysis buffer equal to the volume of culture medium
was used for lysates, and the cells were frozen and thawed
for 3 cycles.
IL-1 bioassay. IL- 1 biological activity in synovial
tissue culture supernatants was determined using a mouse
thymocyte costimulation assay. Thymocytes were obtained
from 4-6-week-old C3H/HeJ mice and cultured in 96-well
microtiter plates at a density of lo6 cells/well, in 100 pl of
RPMI 1640 supplemented with 5% FCS, L-glutamine, 5 x
lO-’M 2-mercaptoethanol, and 12.5 &ml polymyxin B
(Sigma, St. Louis, MO). The cells were stimulated with 1
pg/ml of phytohemagglutinin plus recombinant human IL-10
or a 1:8 dilution of 3-day ST cell supernatants. A supernatant
of LPS-stimulated human monocytes was also used as a
positive control.
Prior to addition to the thymocytes, the supernatants
were preincubated for 1 hour at 4°C with 1:100 dilutions of
neutralizing rabbit anti-human IL-la and IL-1p antisera
(from WPA) or a similar dilution of control rabbit serum. The
cells were cultured with the samples for 44 hours at 37°C and
then pulsed with 1 pCi/well of 3H-thymidine. Four hours
later, the cells were harvested, and thymidine incorporation
was determined in a liquid scintillation counter.
Reverse transcription PCR (RT-PCR). Poly(A)+
RNA was prepared from isolated ST cells or FLS using the
Microfast Track kit (Invitrogen, San Diego, CA), and the
cDNA was synthesized using Superscript reverse transcriptase (BRL, Bethesda, MD). The efficiency of reverse
transcription was consistent from experiment to experiment,
with 1&20% variation (data not shown). For PCR experiments, cDNA from lo5 cells was amplified in a 50-pl reaction
volume containing 30 pmoles of each primer and 200 pM
dNTP, 20 mM Tris (pH 8.3), 50 mM KCl, 3.5 mM MgCI,,
and 1 unit of AmpliTaq DNA polymerase (Perkin-Elmer
Cetus, Norwalk, CT). The number of cycles was adjusted so
that the intensity of the amplified sequence in the test sample
was in the linear portion of the standard curve (sIL-lra 30
cycles; icIL- Ira 35 cycles). Simultaneous reactions were
performed in separate tubes with known amounts of icIL-lra
or sIL-lra cDNA.
The amplified fragments (167 bases for icIL-lra and
109 bases for sIL-lra) were resolved on a 4% agarose gel
(BRL) stained with ethidium bromide. The gel was photographed under ultraviolet light with Polaroid 667 high-speed
black-and-white film. The photographs were scanned at 256
grey scale into a Macintosh I1 computer using The Complete
Half-Page Scanner (The Complete PC, San Jose, CA) and
image analysis was performed with NIH Image 1.43, using
the gel analysis macros. The macros were optimized for
evaluating 1-dimensional gel electrophoresis patterns from
Fresh cell
3 d cell
4.5 1
1 .-5 4
Fresh cell
3 d cell
Figure 1. Concentrations of interleukin-1 receptor antagonist (ILIra) and IL-1 in lysates of frozen unfractionated synovial cell pellets
(fresh or 3-day cultured) or supernatants of cultured synovial cells
obtained from patients with rheumatoid arthritis (RA; n = 11) or
osteoarthritis (OA; n = 12). Proteins were measured by enzymelinked immunosorbent assay. Differences between OA and RA
samples were not statistically significant. Fresh cell lysates = fresh,
enzymatically dispersed synovial tissue (ST)cells; 3 d cell lysates =
dispersed ST cells after a 3-day culture; 3 d supernat = supernatants
of dispersed ST cells after a 3-day culture.
grey-scale images and calibrated using the standard curve for
each experiment. First, the lanes of the standards in the
digitized image were identified and the number of plasmid
copies defined. The software then analyzed the image,
Table 1. IL-la, IL-lp, and IL-lra protein levels in ST cell lysates
and supernatants*
Fresh cell
3-day cell
f 0.18
* Values are the mean
0.48 f 0.24
1.68 f 0.74
2.00 2 0.59
f 0.18
SEM nd2.5 x lo6 unfractionated synovial
tissue (ST) cells (n = 23). Osteoarthritis and rheumatoid arthritis
data were pooled since the values for the 2 disease groups were not
significantly different. IL-I = interleukin-1 ; IL-Ira = IL-I receptor
interpolated the standards, and generated a standard curve.
The lanes with the unknowns were then identified, and the
number of copies calculated from the standard curve. Finally, the total number of copies was divided by the number
of cells to yield the number of copies per cell.
Statistical analysis. Student’s r-test was used to determine statistical significance in all cases, except for the
analysis of the correlation between IL-laand IL-lP production, which employed simple linear regression analysis (Statworks; Cricket Software, Philadelphia, PA).
The ratios of IL-la to IL-Ip levels were similar
in fresh cell lysates, 3-day cell lysates, and 3-day
culture supernatants, and there was a significant correlation between IL-la and IL-lp production (see
Figure 2 for the 3-day supernatant data). Of note, the
IL-1p protein levels in the cell lysates might be underestimated due to the inability of the monoclonal antibodies to detect the biologically inactive 3I-kd precursor in the ELISA.
IL-lra production by fibroblast-like synoviocytes
and fresh synovial tissue macrophages. Two populations of synovial cells were studied to assess their
contribution to IL-lra production in arthritis. Immunoreactive IL-lra was previously detected in CD14+
synovial tissue macrophages, using double-label immunohistochemistry (22). However, a large number of
CD14- synovial lining cells (possibly FLS) also contained IL- Ira. Therefore, cultured FLS were studied
to determine if they produced the receptor antagonist.
OA and RA FLS constitutively produced IL-Ira, but
the protein was not secreted (Table 3). When FLS
were stimulated for 2 days with IL-lp or TNFa,
cell-associated IL- Ira levels increased significantly but
there was only a slight increase in IL-Ira secretion.
Hence, FLS, like dermal fibroblasts and keratinocytes
IL-1 and IL-lra production by synovial tissue
cells. Twenty-three synovial tissues were processed
(11 RA and 12 OA); the results are summarized in
Figure 1. Fresh cell lysates, 3-day cultured cell lysates, and 3-day supernatants from RA patients contained significant amounts of immunoreactive IL- 1
(IL-la plus IL-1p) and IL-lra. There were no correlations between cytokine levels and the patients’ medications or disease activity levels. The amounts were
relatively low in the freshly isolated cell lysates and
increased during the 3-day culture period (P < 0.05).
The concentrations of IL-1 and IL-Ira in OA samples
were similar to those in RA samples.
Overall (Le., RA plus OA), the ratios of IL-Ira
to total IL-1 were about 1.2 in fresh cell lysates, 1.8 in
3-day cell lysates, and 3.6 in culture supernatants
(Table 1). Figure 1 also shows that culture supernatants and cell lysates contained similar amounts of
IL-Ira, indicating that about half of the receptor
antagonist was not secreted by the cells. Despite the
presence of substantial amounts of IL-lra and other
IL-1 antagonists (such as TGFP and soluble IL-1
receptors), IL- 1 biological activity was still present in
ST supernatants (see Table 2).
Table 2. Biologically active IL-1 in culture supernatants of ST
ST 1
+ anti-IL-1
ST 2
+ anti-IL-l
ST 3
+ anti-IL-1
ST 4
+ anti-IL-1
ST 5
+ anti-IL-1
+ anti-IL-1
* Three-day
culture supernatants of ST cells were assayed in an
IL-I bioassay (see Materials and Methods). Biological activity
(3H-thymidine incorporation and ng/ml net biological IL-I activity,
respectively) and immunoreactive IL-I (ng/ml) are shown for each
supernatant. Normal rabbit serum (NRS) and polyclonal rabbit
anti-IL-la and anti-IL-lp antisera were used at 1:lOO dilutions. See
Table 1 for other abbreviations.
(16,17), appeared to produce a form of IL-Ira that
remained intracellular.
IL-lra production was also studied in CD14t
synovial tissue macrophages. In contrast to FLS,
IL-lra was present in the supernatants of cultured
synovial macrophages but was barely detectable in the
3-day cell lysates (Table 3). Of interest, the amount of
IL-lra produced by ST macrophages was small (about
0.30 ng/105 cells) compared with that of macrophages
isolated from other sites (which constitutively secrete
10-100 times more [30,31]). To determine if low IL-lra
secretion by synovial macrophages was due to the
isolation method, monocyte-derived macrophages
(which produce very large amounts of IL-Ira [301)
were treated with anti-CDlkonjugated magnetic
beads. The amount of IL-lra secreted was similar to
that observed with untreated monocyte-derived
macrophages (see Table 3).
IL-lra structural variants. RNA transcripts encoding intracellular and secreted forms of IL-lra
mRNA have been described (11,13,16). Intracellular
IL-lra mRNA lacks a signal peptide, which leads to
intracellular accumulation of the protein. To study the
relative levels of sIL-lra and icIL-lra mRNA in ST
cells, specific primers for each form were synthesized
and RT-PCR was performed. Both icIL-lra and sILIra mRNA were detected in synovial tissue cells.
0 4
0 8
1 0
1 2
1 4
(ng12.5~106 cells)
Figure 2. Correlation between IL-la and IL-lp production in supernatants of unfractionated cultured synovial cells obtained from
patients with RA (n = 11) or OA (n = 8). Data are pooled for the 19
available samples (8 of these samples had no detectable IL-1). r =
0.964, P < 0.001. See Figure 1 for abbreviations.
Table 3. IL-lra protein levels in synovial cell supernatants and
FLS (n = 4)
TNFa (100 nglml)
IL-lp (1 ng/ml)
CD14+ ST cells (n = 7)
Monocyte-derived macrophages
0.08 f 0.08
0.03 f 0.03
0.30 2 0.18
* 0.02t
* Fibroblast-like synoviocytes (FLS) were cultured for 2 days in
medium alone or in the presence of tumor necrosis factor a (TNFa)
or interleukin-lp (IL-lp). CD14+ synovial tissue (ST) cells were
cultured for 3 days in medium alone. Three-week-old monocytederived macrophages were prepared as previously described (30),
and were treated either with medium or with antLCD14 antibody
conjugated to magnetic beads prior to 3 days of culture. Values are
the mean 2 SEM ng/ml of IL-1 receptor antagonist (IL-lra) per lo5
CD14+ cells or per confluent plate of FLS. ND = not determined.
t P < 0.04 versus supernatants.
Figure 3 shows a representative experiment using
FLS. Using computer-assisted image analysis, the
number of copies of mRNA per cell was estimated.
The method selected could potentially underestimate
the number of copies if reverse transcription was
inefficient but, because icIL-Ira and sIL-lra mRNA
are identical except for a small upstream region, the
icIL-1ra:sIL-lra ratio should be accurately described.
Table 4 shows the results of RT-PCR experiments with RA synovial cells in which the number of
mRNA copies of icIL-lra and sIL-lra was estimated.
The striking finding was the abundance of icIL-lra
mRNA compared with sIL-lra mRNA. In unfractionated fresh ST cells, the ratio of icIL-lra to sIL-lra
mRNA was 29:l. The ratio for unfractionated 3-day
cultured cells was similar (data not shown). This high
ratio was surprising, considering that a significant
amount of IL-lra protein was detected in ST culture
supernatants. The relative abundance of icIL-lra
mRNA compared with sIL-Ira in fresh CD14+ ST
macrophages was lower (16: I), but still high compared
with the low amount of the protein in cell lysates. In
contrast to ST macrophages, FLS showed a better
correlation between protein and mRNA results. This
cell type produced a cell-associated form of IL-lra
(see Table 3) and exhibited a marked predominance of
icIL-lra mRNA (70:l).
IL-1 plays a central role in the pathogenesis of
rheumatoid arthritis (1,2). Small amounts of IL-lP
IL-lra icll-ira
I I0
copy number
20.0001 lG3PDH IMarkers
copy number
Figure 3. Example of RT-PCR for estimation of icIL-lra and sIL-lra mRNA levels in cultured FLS derived from ST sample STM222. G3PDH
(glyceraldehyde-3-phosphatedehydrogenase) is included as a control for the integrity of the mRNA. Abbreviated standard curves for purified
icIL-lra and sIL-Ira templates are shown (most curves include 4-6 concentrations of control cDNA to ensure that the amplification is in the
linear range). Occasionally, faint primer-dimer bands were observed (as shown in the STM222 icIL-Ira lane). The molecular weight markers
are a mixture of DNA molecular weight markers 111 and V (Boehringer Mannheim, Indianapolis, IN). The photograph of the gel was digitized
and individual bands quantified by computer-assisted image analysis as described in Materials and Methods. See Table 4 for abbreviations.
protein have been detected in RA synovial effusions
(19-21), and mRNA for both IL-la and IL-1p are
found in synovial tissue macrophages (26,27). Although the precise role of IL-1 in arthritis is not
completely known, it probably activates fibroblast-like
synoviocytes in the membrane to proliferate and secrete a variety of proinflammatory mediators, including cytokines, prostaglandins, and metalloproteinases
(1-3). However, the presence of natural inhibitors to
IL-1, such as TGFp (6), in RA articular samples raises
some questions regarding the consequences of IL- 1
production in vivo. In addition, a naturally occurring
receptor antagonist of 1L-1 (IL-lra) was identified in
the synovium of RA and OA patients (22-24).
IL-Ira competitively inhibits the binding of
IL-la and IL-Ip to both type I and I1 IL-1 receptors
(IL-1R) (14,16). IL-lra is a pure receptor antagonist
since it does not induce signal transduction or internalization of the ligand-receptor complex (32). Intracellular and secreted forms of IL-lra that result from
alternative splicing of RNA have been identified
(13,16). Although IL-lra binds to the IL-1R with high
Table 4. Quantification of icIL-lra and sIL-lra mRNA by
Total ST cells (n
FLS (n = 5 )
4,664 +- 561
3,568 2 407
158 2 21
51 +- 17
* Values are the mean
? SEM copies of messenger RNA (mRNA)
= intracellular interleukin-1 receptor antago-
per lo4 cells. icIL-lra
nist; sIL-lra = secretory IL-lra; RT-PCR = reverse transcription
polymerase chain reaction; ST = synovial tissue; FLS = fibroblastlike synoviocytes.
affinity, 10-100-fold excess amounts of IL-lra are
needed to inhibit 50% of the IL-1 biological response
in cells that express the type I IL-1R (10,33-35). This
is because target cells exhibit full biological response
when only a small percentage of the IL-1 receptors on
each cell are occupied.
Immunohistochemical studies of RA synovium
have demonstrated large amounts of IL-lra protein
and mRNA in the intimal lining and in sublining
mononuclear cells (22-24). We have also shown that
immunoreactive and biologically active IL- Ira is secreted by cultured ST cells (22). However, net IL-I
agonist activity was detected in most ST supernatants
using a biological assay. These functional assays,
however, can be confounded by the myriad of other
cytokines and inhibitors that are abundant in biological
samples, possibly including soluble type I1 IL- 1R.
In the present study, we determined the relative
levels of IL-1 and IL-lra protein production by synovial tissue cells from RA and OA patients, using
sensitive and specific immunoassays. Our data show
that the ratio of IL-lra to IL-1 in unfractionated ST
cells is relatively low in fresh cell lysates, 3-day
cultured cell lysates, and 3-day cell supernatants. The
highest average ratio was observed in culture supernatants. However, even at this level, one would not
anticipate inhibition of 1L-1-mediated activation of
chondrocytes or synoviocytes. This was borne out by
studies identifying net IL-1 biological activity in ST
culture supernatants. Since the 3 I-kd precursor form
of IL-la but not IL-1p is biologically active (36),
intracellular IL-1a probably comprises most of the
active cytokine in cell lysates. The IL-1ra:IL-1 ratio
could, therefore, be somewhat higher (about 2-fold) if
one discounted the values for IL-1p in the lysates,
although it would still be well below the level required
to inhibit IL-1 activity. The failure of the ELISA to
fully detect the 31-kd precursor form of IL-1p does not
change this conclusion. Both IL-la and IL-1p precursor and mature forms contribute to the total IL-1 in the
supernatants, but secreted IL- l p is predominantly
processed to the 17-kd form.
The relatively low ratio of immunoreactive ILIra to IL-1 likely reflects a combination of high IL-1
production (37) and defective IL-lra production by
monocyte-derived cells (30). It should be emphasized
that the rheumatoid synovium may contain other molecules that inhibit IL-l activities, such as TGFp or
soluble type I1 IL-1R. Thus, the functional inhibition
of IL-1 by these proteins might be additive or synergistic with the limited amounts of IL-lra produced.
Despite this, however, supernatants of most synovial
tissue cells exhibited IL-1 biological activity.
An unanticipated observation was that OA and
RA cells produced similar amounts of IL-lra. This is
particularly interesting since frozen sections of RA ST
appear to contain more immunoreactive IL-lra than
do OA ST (22). The likely reason for this discrepancy
is that equal numbers of OA and RA cells were used in
the in vitro experiments described herein. This corrects for the lesser number of cells normally found in
intact OA tissue sections. A similar phenomenon has
been observed for IL-1; namely, that significantly less
IL-1 protein was found in OA ST than in RA ST even
though a similar percentage of synovial cells in each
group contained IL-1p mRNA (25,26).
We also characterized some of the synovial
cells that express the IL-Ira gene. Both fresh CD14+
synovial macrophages and cultured FLS produced
IL-ha, albeit with distinctive patterns. IL-lra was
detected primarily in the supernatants of the synovial
macrophages, while FLS produced an intracellular
molecule. This differs somewhat from a recent study
that reported a much greater percentage of IL-lra in
the supernatants of cytokine-stimulated synovial fibroblasts (38). Our data suggest that FLS produce primarily icIL- Ira, while macrophages may produce sIL- Ira.
Even so, synovial macrophages secreted much less
IL-lra than did macrophages from other sites. For
instance, normal blood monocyte-derived macrophages and alveolar macrophages produce about 1&
100 times more IL-Ira (30,31). This difference might
result from a specific defect in CD14+ ST macrophages, since they are capable of secreting abundant
amounts of other cytokines, including granulocyte-
macrophage colony-stimulating factor, TNFa, and
IL-1 (26,39). Low IL-lra production could also reflect
synovial macrophage population immaturity because
the IL- 1ra:IL-1 ratio increases during macrophage
differentiation (40). Synovial macrophages express a
variety of differentiation surface markers, which suggests the presence of both mature and immature cells
in the inflamed synovium (41).
The relative levels of icIL-lra and sIL-lra
mRNA in synovial tissue cells were also determined.
RT-PCR was performed using a common 3' primer and
specific 5' primers that distinguish between the two
forms of the IL-lra mRNA. While the method of
RT-PCR used (like all RT-PCR methods) has some
theoretical drawbacks (such as the inability to correct
for inefficient reverse transcription), it minimizes other
artifacts that can interfere with other methods using
internal RNA standards, such as competition for enzyme or nonlinear amplification due to higher template
concentrations (42). Since reverse transcription efficiency should be similar for sIL-lra and icIL-Ira
mRNA, the calculated icIL-1ra:sIL-lra ratio using this
technique should reflect the actual levels.
A surprising finding was the predominance of
icIL-lra mRNA in the unfractionated ST cells studied
even though about half of the IL-lra was found in the
supernatants of the cultured cells. This suggests that
regulation of IL-lra production may occur, in part, at
the translational level and that the relatively high
amounts of icIL-lra mRNA may not result in commensurate levels of protein production. Alternatively, cell
injury or apoptosis could lead to the release of icIL-lra
into the culture medium. This mechanism appears to
play an important role in IL-I release (43). Since our
IL-lra ELISA does not discriminate between the two
structural variants, we cannot ascertain whether some
protein in the supernatants of unfractionated ST or
isolated macrophages is icIL-lra. The very high icIL1ra:sIL-lra mRNA ratio in FLS correlated with the
lack of IL-Ira secretion by this cell type.
In conclusion, we have examined the balance
between IL-1 and IL-lra production by synovial cells
and determined that the levels of IL-lra are low
compared with those needed to inhibit IL-1 action.
Since many of the proinflammatory effects of IL-1 in
RA result from stimulation of cells bearing type I
IL-1R (e.g., FLS and chondrocytes), it is doubtful that
locally produced IL-lra alone significantly inhibits this
aspect of synovitis. However, high levels of IL-Ira are
present in the synovial fluids of some patients with RA
(25), and a high IL-1ra:IL-1 ratio in the synovial fluid
of patients with Lyme arthritis was shown to correlate
with more rapid resolution of acute joint disease (44).
Thus, IL-lra in synovial fluid, possibly derived from
neutrophils ( 4 9 , might penetrate ST and compensate
for inadequate local production of IL-Ira. Furthermore, a significant percentage of ST IL-lra remains
intracellular and, hence, does not contribute to IL-1
antagonism in the extracellular milieu. This phenomenon appears to be due, at least in part, to the predominant production of icIL-lra by FLS. The results of
recent studies indicate that sIL-lra and icIL-lra are
equally potent at inhibiting IL-1 stimulation of thymocytes (15) and endothelial cells (46). However, the
excess icIL-1 mRNA in synovial tissue suggests that
this unique molecule might have additional effects in
joint disease beyond inhibition of IL-1 binding to cell
surface receptors.
The advice and support of Dr. Bruce Littman a n d
the technical assistance of Jon U y e s a k a are gratefully
1. Firestein GS, Zvaifler NJ: How important are T cells in chronic
rheumatoid synovitis? Arthritis Rheum 33:768-773, 1990
2. Arend WP, Dayer J-M: Cytokine and cytokine inhibitors or
antagonists in rheumatoid arthritis. Arthritis Rheum 33:305-315,
3. Alvaro-Gracia JM, Zvaifler NJ, Firestein GS: Cytokines in
chronic inflammatory arthritis. V. Mutual antagonism between
interferon-gamma and tumor necrosis factor-alpha on HLA-DR
expression, proliferation, collagenase production, and granulocyte macrophage colony-stimulating factor production by rheumatoid arthritis synoviocytes. J Clin Invest 86: 1790-1798, 1990
4. Chandrasekhar S, Harvey AK, Hrubey PS, Bendele AM: Arthritis induced by interleukin-1 is dependent on the site and
frequency of intraarticular injection. Clin Immunol Immunopathol 55382-394, 1990
5 . Henderson B, Pettipher ER: Arthritogenic actions of recombinant IL-1 and tumour necrosis factor alpha in the rabbit evidence for synergistic interactions between cytokines in vivo.
Clin Exp Immunol 75:306-310, 1989
6. Wahl SM, Allen JB, Wong HL, Dougherty SF, Ellingsworth
LR: Antagonistic and agonistic effects of transforming growth
factor-beta and IL-I in rheumatoid synovium. J Immunol 145:
2514-2519, 1990
7. Arend WP, J o s h JG, Massoni RJ: Effects of immune complexes on production by human monocytes of interleukin 1 or an
interleukin I inhibitor. J Immunol 134:3868-3875, 1985
8. Balavoine J-F, deRochemontiex B, Williamson K, Seckinger P,
Cruchaud A, Dayer J-M: Prostaglandin E2 and collagenase
production by fibroblasts and synovial cells is regulated by
urine-derived human interleukin 1 and inhibitor(s). J Clin Invest
78:1120-1124, 1986
9. Dripps DJ, Verderber E, Ng RK, Thompson RC, Eisenberg SP:
Interleukin-l receptor antagonist binds to the type. I1 interleukin-1
65 1
receptor on B cells and neutrophils. J Biol Chem 266:2031120315, 1991
Arend WP, Welgus HG, Thompson RC, Eisenberg SP: Biological properties of recombinant human monocyte-derived interleukin 1 receptor antagonist. J Clin Invest 85:1694-1697, 1990
Carter DB, Deibel MR Jr, Dunn CJ, Tomich CS, Laborde AL,
Slightom JL, Berger AE, Bienkowski MJ, Sun FF, McEwan
RN: Purification, cloning, expression and biological characterization of an interleukin-1 receptor antagonist protein. Nature
344:633-638, 1990
Arend WP: Interleukin 1 receptor antagonist: a new member of
the interleukin 1 family. J Clin Invest 88: 1445-1451, 1991
Eisenberg SP, Evans RJ, Arend WP, Verderber E, Brewer MT,
Hannum CH, Thompson RC: Primary structure and functional
expression from complementary DNA of a human interleukin- 1
receptor antagonist. Nature 343:341-346, 1990
Hannum CH, Wilcox CJ, Arend WP, J o s h FG, Dripps DJ,
Heimdal PL, Armes LG, Sommer A, Eisenberg SP, Thompson
RC: Interleukin-1 receptor antagonist activity of a human interleukin-l inhibitor. Nature 343:336340, 1990
Bigler CF, N o m s DA, Weston WL, Arend WP: Interleukin-1
receptor antagonist production by human keratinocytes. J Invest Dermatol 9k38-44, 1992
Haskill S, Martin G, Van Le L, M o m s J, Peace A, Biegler CF,
JaEe GJ, Hammerberg C, Sporn SA, Fong S, Arend WP, Ralph
P cDNA cloning of an intracellular form of the human interleukin 1 receptor antagonist associated with epithelium. Proc Natl
Acad Sci U S A 88:3681-3685, 1991
Chan LS, Hammerberg C, Kang K, Sabb P, Tavakkol A,
Cooper KD: Human dermal fibroblast interleukin-1 receptor
antagonist (IL-Ira) and interleukin-1 beta (IL-1 beta) mRNA
and protein are co-stimulated by phorbol ester: implication for a
homeostatic mechanism. J Invest Dermatol 99:3 15-322, 1992
Quay J, Arend WP, Becker S: IL-I and IL-1 receptor antagonist
production in human alveolar macrophages infected with respiratory syncytial virus. J Infect Dis (in press)
Miossec P, Dinarello CA, Ziff M: Interleukin-1 lymphocyte
chemotactic activity in rheumatoid arthritis synovial fluid. Arthritis Rheum 29:461470, 1986
Symons JA, McDowell TL, di Giovine FS, Wood NC, Capper
SJ, Duff GW: Interleukin 1 in rheumatoid arthritis: potentiation
of immune responses within the joint. Lymphokine Res 8:365372, 1989
Rooney M, Symons JA, Duff GW: Interleukin 1 beta in synovial
fluid is related to local disease activity in rheumatoid arthritis.
Rheumatol Int 10:217-219, 1990
Firestein GS, Berger AE, Tracey DE, Chosay JG, Chapman
DL, Paine MM, Yu C, Zvaifler NJ: IL-1 receptor antagonist
protein production and gene expression in rheumatoid arthritis
and osteoarthritis synovium. J Immunol 149:1054-1062, 1992
Koch AE, Kunkel SL, Chensue SW, Haines GK, Strieter RM:
Expression of interleukin-1 and interleukin-1 receptor antagonist by human rheumatoid synovial tissue macrophages. Clin
Immunol Immunopathol65:23-27, 1992
Deleuran BW, Chu CQ, Field M, Brennan FM, Katsikis P,
Feldmann M, Maini RN: Localization of interleukin-1 alpha,
type 1 interleukin-] receptor and interleukin-1 receptor antagonist in the synovial membrane and cartilage/pannus junction in
rheumatoid arthritis. Br J Rheumatol 31:801-809, 1992
Malyak M, Swaney RE, Arend WP: Levels of synovial fluid
interleukin-1 receptor antagonist in rheumatoid arthritis and
other arthropathies: potential contribution from synovial fluid
neutrophils. Arthritis Rheum 36:781-789, 1993
Firestein GS, Alvaro-Gracia JM, Maki R: Quantitative analysis
of cytokine gene expression in rheumatoid arthritis. J Immunol
14413347-3353, 1990
Buchan G, Barrett K, Turner M, Chantry D, Maini RN,
Feldmann M: Interleukin-1 and tumour necrosis factor mRNA
expression in rheumatoid arthritis: prolonged production of IL-1
alpha. Clin Exp Immunol 733449-455, 1988
Malyak M, J o s h FG, Verderber EL, Eisenberg SP, Arend WP:
IL-lra ELISA: reduction and alkylation of synovial fluid eliminates interference by IgM rheumatoid factors. J Immunol
Methods 140:281-288, 1991
Herzyk DJ, Berger AE, Allen JN, Wewers MD: Sandwich
ELISA formats designed to detect 17 kDa IL-1 beta significantly
underestimate 35 kDa 1L-1 beta. J Immunol Methods 148:2432.54, 1992
Tsai V, Firestein GS, Arend W, Zvaifler NJ: Cytokine-induced
differentiation of cultured nonadherent macrophages. Cell Immunol 144:203-216, 1992
Moore SA, Stneter RM, Rolfe MW, Standiford TJ, Burdick
MD, Kunkel SL: Expression and regulation of human alveolar
macrophage-derived interleukin-1 receptor antagonist. Am J
Respir Cell Mol Biol 6569-575, 1992
Dripps DJ, Brandhuber BJ, Thompson RC, Eisenberg SP:
Interleukin-1 (IL-I) receptor antagonist binds to the 80-kDa IL-I
receptor but does not initiate IL-1 signal transduction. J Biol
Chem 266:10331-10336, 1991
Arend WP, Coll BP: Interaction of recombinant monocytederived interleukin 1 receptor antagonist with rheumatoid synovial cells. Cytokine 3:407413, 1991
Smith RJ, Chin JE, Sam LM, Justen JM: Biologic effects of an
interleukin-1 receptor antagonist protein on interleukin-lstimulated cartilage erosion and chondrocyte responsiveness.
Arthritis Rheum 34:78-83, 1991
Seckinger P, Klein-Nulend J, Alander C, Thompson RC, Dayer
JM, Raisz LG: Natural and recombinant human IL-I receptor
antagonists block the effects of IL-1 on bone resorption and
prostaglandin production. J Immunol 145:41814184, 1990
Mosley B, Urdal DL, Prickett KS, Larsen A, Cosman D,
Conlon PJ, Gillis S, Dower SK: The interleukin-1 receptor binds
the human interleukin-1 alpha precursor but not the interleukin-1
beta precursor. J Biol Chem 262:2941-2944, 1987
37. Fuji I, Shingu M, Nobunaga M: Monocyte activation in early
onset rheumatoid arthritis. Ann Rheum Dis 49:497-503, 1990
38. Krzesicki RF, Hatfield CA, Bienkowski MJ, McGuire JC,
Winterrowd GE, Chapman DL, Berger AE, McEwan RN,
Carter DB, Chosay JG, Tracey DE, Chin JE: Regulation of
expression of IL-1 receptor antagonist protein in human synovial and dermal fibroblasts. J Immunol 150:4008-4018, 1993
39. Alvaro-Gracia JM, Zvaifler NJ, Brown CB, Kaushansky K,
Firestein GS: Cytokines in chronic inflammatory arthritis. VI.
Analysis of the synovial cells involved in granulocytemacrophage colony-stimulating factor production and gene expression in rheumatoid arthritis and its regulation by IL-I and
tumor necrosis factor-alpha. J Immunol 146:336.5-3371, 1991
40. Janson RW, Hance KR, Arend WP: Production of IL-I receptor
antagonist by human in vitro-derived macrophages: effects of
lipopolysaccharide and granulocyte-macrophage colonystimulating factor. J Immunol 147:4218-4223, 1991
41. Hogg N, Palmer DG, Revell PA: Mononuclear phagocytes of
normal and rheumatoid synovial membrane identified by monoclonal antibodies. Immunology 56:67368 1, 1985
42. O’Garra A, Vieira P: Polymerase chain reaction for detection of
cytokine gene expression. Curr Opin Immunol4:211-215, 1992
43. Hogquist KA, Nett MA, Unanue ER, Chaplin DD: Interleukin
1 is processed and released during apoptosis. Proc Natl Acad
Sci U S A 8k848.5-8489, 1991
44. Miller LC, Lynch EA, Isa S , Logan JW, Dinarello CD, Steere
AC: Balance of synovial fluid 1L-Ip and IL-I receptor antagonist and recovery from Lyme arthritis. Lancet 341:146148,
45. Malyak M, Smith MF Jr, Abel AA, Arend WP: Peripheral blood
neutrophil production of interleukin-lp and interleukin-1 receptor antagonist. J Clin Immunol 14:20-30, 1994
46. Bertini R, Sironi M, Martin-Padura I, Colotta F, Rambaldi S,
Bernasconi S, Ghezzi P, Haskill SJ, Liu D, Mantovani A:
Inhibitory effect of recombinant intracellular interleukin I receptor antagonist on endothelial cell activation. Cytokine 4:4447, 1992
Без категории
Размер файла
941 Кб
balances, arthritis, antagonisms, interleukin, synovial, receptov, rheumatoid
Пожаловаться на содержимое документа