close

Вход

Забыли?

вход по аккаунту

?

Local production of B lymphocyte stimulator protein and APRIL in arthritic joints of patients with inflammatory arthritis.

код для вставкиСкачать
ARTHRITIS & RHEUMATISM
Vol. 48, No. 4, April 2003, pp 982–992
DOI 10.1002/art.10860
© 2003, American College of Rheumatology
Local Production of B Lymphocyte Stimulator Protein and
APRIL in Arthritic Joints of Patients With
Inflammatory Arthritis
Soon-Min Tan,1 Dong Xu,1 Viktor Roschke,2 James W. Perry,2 Daniel G. Arkfeld,1
Glenn R. Ehresmann,1 Thi-Sau Migone,2 David M. Hilbert,2 and William Stohl1
and total nucleated cell counts. Although SF and serum
BLyS protein levels correlated with each other, SF and
serum APRIL levels did not, suggesting that SF BLyS
protein levels are more dependent upon systemic factors
than are SF APRIL levels. Moreover, in 8 patients who
underwent sequential arthrocenteses, changes in SF
BLyS protein levels did not immutably parallel changes
in SF APRIL levels, indicating their differential regulation.
Conclusion. BLyS protein and APRIL are locally
produced in inflamed joints. Their respective SF levels
are differentially regulated, suggesting that they serve
different functions. Together, their local production
may foster survival and/or expansion of B cells that
produce pathogenic autoantibodies and/or promote local T cell activation and consequent joint destruction.
Objective. To determine whether synovial fluid
(SF) levels and cell-surface expression of B lymphocyte
stimulator (BLyS) protein and SF levels of APRIL are
elevated in patients with inflammatory arthritis (IA).
Methods. Same-day blood and SF samples from
89 patients with 103 knee effusions (81 knees with IA
and 22 with noninflammatory arthritis [NIA]) were
evaluated for BLyS protein and APRIL levels by
enzyme-linked immunosorbent assay. Blood and SF
mononuclear cells were double-stained for surface BLyS
protein and surface CD14 (monocyte marker) and were
analyzed by flow cytometry. Complete blood cell counts
and SF nucleated cell counts were performed by the
clinical hematology laboratory.
Results. BLyS protein levels were higher in SF
than in corresponding serum samples from both IA and
NIA patients. SF BLyS protein levels, but not surface
expression of BLyS protein, were disproportionately
elevated in IA patients. APRIL levels were higher in SF
than in corresponding serum samples from most IA
patients but not NIA patients. SF BLyS protein and
APRIL levels correlated with each other, and each
correlated with SF monocyte, lymphocyte, neutrophil,
B lymphocyte stimulator (BLyS; trademark of
Human Genome Sciences, Rockville, MD) protein is a
recently identified 285–amino acid member of the tumor
necrosis factor (TNF) ligand superfamily (1–6). Other
names for this protein are TALL-1, BAFF, THANK,
TNFSF20 (subsequently renamed TNFSF13B), and
zTNF4. It is expressed as a type II transmembrane
protein, which is cleaved from the cell surface to release
a biologically active soluble 17-kd protein (1–5). Mice
genetically deficient in BLyS protein harbor markedly
reduced numbers of mature B cells in secondary lymphoid organs and manifest reduced baseline serum
levels of Ig and Ig responses to T cell–dependent and T
cell–independent antigens (7,8). Conversely, in vivo administration of recombinant BLyS (rBLyS) protein to
mice induces B cell expansion and polyclonal hypergammaglobulinemia (1), which is, at least in part, consequent to inhibition of B cell apoptosis and enhanced B
cell survival (9–12). Constitutive overexpression of BLyS
Supported in part by grants from the NIH (AR-41006), the
Alliance for Lupus Research, and the Arthritis Foundation, Southern
California Chapter.
1
Soon-Min Tan, MD, Dong Xu, MD, Daniel G. Arkfeld, MD,
Glenn R. Ehresmann, MD, William Stohl, MD, PhD: Los Angeles
County ⫹ University of Southern California Medical Center, and
University of Southern California Keck School of Medicine, Los
Angeles, California; 2Viktor Roschke, PhD, James W. Perry, BA,
Thi-Sau Migone, PhD, David M. Hilbert, PhD: Human Genome
Sciences, Inc., Rockville, Maryland.
Address correspondence and reprint requests to William
Stohl, MD, PhD, Division of Rheumatology, University of Southern
California, 2011 Zonal Avenue, HMR 711, Los Angeles, CA 90033.
E-mail: stohl@usc.edu.
Submitted for publication August 5, 2002; accepted in revised
form December 13, 2002.
982
LOCAL PRODUCTION OF BLyS AND APRIL IN INFLAMMATORY ARTHRITIS
protein in blys-transgenic mice leads to elevated serum
titers of multiple autoantibodies, including anti–doublestranded DNA and rheumatoid factor autoantibodies
(6,13,14).
APRIL (also called TNFSF13A) is a 250–amino
acid member of the TNF ligand superfamily that shares
substantial homology with BLyS protein and binds to 2
of the 3 BLyS protein receptors (BCMA and TACI)
(15–19). APRIL is not expressed on the cell surface but
is processed intracellularly and secreted in its biologically active form (20). As is the case for rBLyS
protein, recombinant APRIL (rAPRIL) costimulates B
cells in vitro and in vivo (16,17), albeit with considerably
less potency. Constitutive overexpression of APRIL in
april-transgenic mice leads to enhanced T cell survival
and enhanced antigen-specific antibody responses (21).
Circulating levels of APRIL in patients with
rheumatic diseases have heretofore not been reported.
However, circulating levels of BLyS protein have been
measured in patients with a variety of rheumatic diseases, including systemic lupus erythematosus (SLE),
rheumatoid arthritis (RA), and Sjögren’s syndrome, and
have been found to be elevated in a substantial proportion of such patients (22–24). Although it is not yet
known what effects BLyS protein/APRIL antagonists
may have on human disease or on human in vivo
immune responses, the experience with such antagonists
in murine rheumatic disease models is telling. Mice
genetically prone to the development of SLE ([NZB ⫻
NZW]F1 and MRL-lpr/lpr mice) have elevated circulating levels of BLyS protein, and treatment of these
SLE-prone mice with a genetically engineered soluble
fusion protein between one of the BLyS protein receptors (TACI) and IgG Fc (TACI-Ig) ameliorates progression of disease and improves survival (6). Moreover, the
joint inflammation and joint destruction of collageninduced arthritis, a model of noninfectious inflammatory
arthritis (IA), are inhibited by treatment with TACI-Ig
before or after induction of disease (7,25).
Since TACI-Ig binds both BLyS protein and
APRIL (16–18), it is not clear whether its inhibition of
IA reflects inhibition of BLyS protein activity, inhibition
of APRIL activity, or a combination of both. In any case,
these observations raise the possibility that local expression and/or production of BLyS protein and/or APRIL
in inflamed joints may contribute to the development
and/or propagation of disease. We took the first step in
addressing this issue by assessing blood and synovial
fluid (SF) levels of BLyS protein and APRIL as well as
BLyS protein cell surface expression in patients with IA
and in patients with noninflammatory arthritis (NIA).
983
PATIENTS AND METHODS
Subjects. Patients who were admitted to the Los
Angeles County ⫹ University of Southern California Medical
Center (LAC⫹USC MC) and were seen in consultation by the
Rheumatology Service or patients who were receiving outpatient medical care at the Rheumatology Clinics of LAC⫹USC
MC, the Edward R. Roybal Comprehensive Health Center, the
USC Ambulatory Health Center, or the USC Center for
Arthritis and Joint Implant Surgery were recruited for this
study. Criteria for admission to the study were the clinically
indicated need for a therapeutic and/or diagnostic arthrocentesis of one or both knees and a willingness to participate in the
study. No exclusions were made on any basis other than an
inability to give informed consent.
Diagnoses were based on established clinical criteria
(26). Samples obtained from a total of 103 knee arthrocenteses
(89 patients) were evaluated: 22 SF samples from patients with
NIA (15 with osteoarthritis [OA] and 7 with traumatic arthritis), 49 from patients with RA, 14 from patients with crystalinduced arthritis (12 with gout and 2 with calcium pyrophosphate dihydrate crystal deposition disease), 18 from patients
with other noninfectious IA (5 with ankylosing spondylitis, 7
with reactive arthritis, 4 with SLE, 1 with polymyalgia rheumatica, and 1 with unspecified culture-negative IA). A patient
with features of more than one rheumatic disease was classified as having the disease which clinically predominated.
Infectious arthritis was excluded by appropriate microbiologic
studies.
Blood and SF studies. Venous blood and knee SF were
collected and sent to the appropriate clinical laboratories; a
complete blood cell count with differential cell count was
performed on the blood samples, and a nucleated cell count
with differential cell count and crystal examination was performed on the SF samples. In addition, nonheparinized and
heparinized venous blood and SF samples were used for
experimental studies. For serologic studies, sera obtained from
nonheparinized blood and SF samples were shipped to Human
Genome Sciences (HGS; Rockville, MD) for measurement of
BLyS protein and APRIL.
BLyS protein levels were determined by enzyme-linked
immunosorbent assay (ELISA) as previously described (23,27).
APRIL levels were determined by ELISA in a similar manner,
using mouse anti-human APRIL monoclonal antibody (mAb)
16F11 (HGS) as the capture antibody and biotinylated rabbit
anti-human APRIL polyclonal antibody (HGS) as the detection antibody. For determination of either BLyS protein or
APRIL, Immulon-II plates (Labsystems, Franklin, MA) were
coated overnight at 4°C with anti-BLyS protein or anti-APRIL
capture mAb (100 ␮l at 3 ␮g/ml), washed, and blocked for 2
hours at room temperature with blocking buffer (3% bovine
serum albumin [BSA] in phosphate buffered saline [PBS]).
Serial dilutions of serum or SF samples were prepared
in diluent (0.1% Tween 20 ⫹ 0.1% BSA in PBS) and added to
the antibody-coated plates for 2 hours at room temperature.
The plates were washed, incubated for 2 hours with the
biotinylated detection antibodies (100 ␮l at 0.25 ␮g/ml),
washed, incubated for 1 hour with streptavidin–peroxidase
conjugate (100 ␮l at 0.25 ␮g/ml in diluent; Vector, Burlingame,
CA), and developed with the tetramethylbenzidine substrate
kit (Sigma, St. Louis, MO).
984
TAN ET AL
Table 1.
Demographic and medication data of the 89 study patients, by diagnostic group
Parameter
Age, years
Mean
Range
Sex, no.
Female
Male
Race, no.
Arabic
Asian
Black
Hispanic
White
Treatment, no.
Prednisone
Methotrexate
Azathioprine
Sulfasalazine
Hydroxychloroquine
Leflunomide
Etanercept
Infliximab
Osteoarthritis/
traumatic arthritis
(n ⫽ 22)
Rheumatoid
arthritis
(n ⫽ 39)
Other
noninfectious
inflammatory
arthritis
(n ⫽ 14)
58.6
31.9–79.8
47.5
19.9–76.1
42.5
20.8–81.4
Crystal-induced
arthritis
(n ⫽ 14)
52.9
35.9–72.2
12
10
33
6
5
9
2
12
0
0
0
16
6
1
2
0
32
5
0
1
1
11
1
0
1
2
10
1
2
2
0
0
2
0
0
0
18
18
4
5
11
4
2
2
3
2
1
5
4
0
0
0
1
1
0
0
1
0
0
0
Optical density at 450 nm was measured with a SpectraMax 3000 plate reader (Molecular Devices, Sunnyvale, CA).
Standards of rBLyS protein (HGS) and rAPRIL (R&D Systems, Minneapolis, MN) at 0.005–100 ng/ml were processed
along with test samples, and the concentrations of BLyS
protein or APRIL in the experimental samples were calculated
from the standard curves. To ascertain the validity of the BLyS
protein and APRIL ELISAs for both serum and SF, test sera
and SF were intentionally spiked with known amounts of either
rBLyS protein or rAPRIL. In all cases, detection of the
respective protein was as predicted, with no evidence of ELISA
inhibitors and no difference in the degree of recovery between
serum and SF (HGS: unpublished observations).
For cell-based studies, mononuclear cells were isolated
from heparinized blood and SF by Ficoll density-gradient
centrifugation and were double stained with fluorescein
isothiocyanate–conjugated anti-CD14 mAb (PharMingen, San
Diego, CA) plus biotinylated anti-BLyS mAb 9B6 followed by
phycoerythrin-conjugated streptavidin (Dako, Carpinteria,
CA). The specificity of mAb 9B6 for BLyS protein has been
established by the binding of mAb 9B6 to plate-bound rBLyS
protein (but not to rAPRIL, BSA, ovalbumin, or human IgG)
and by the ability of mAb 9B6 (but not control IgG1) to
immunoprecipitate a 36-kd band (detected by Western blotting
with anti-BLyS protein polyclonal antibodies) from monocytes
and BLyS-transfected 293T cells but not from nontransfected
293T cells or from 293T cells transfected with other
membrane-bound TNF family members (HGS: unpublished
observations). The cell surface staining observed with mAb
9B6 is not observed with control IgG1 mAb (27).
Stained cells were analyzed by flow cytometry, with at
least 10,000 events analyzed for each sample. Cell debris, as
determined by forward- and side-scatter characteristics, was
electronically excluded from the analysis.
Patient profiles. Each patient’s sex, race, age, and
medications at the time of the arthrocentesis and phlebotomy
were recorded. The demographic and medication data of the
study patients are listed in Table 1.
Statistical analysis. All analyses were performed using
SigmaStat software (SPSS, Chicago, IL). Neither serum nor SF
BLyS protein or APRIL levels were normally distributed, so
they were each log-transformed to achieve normality. Parametric testing between 2 matched or unmatched groups was
performed by the paired or unpaired t-test, respectively. Parametric testing among 3 or more groups was performed by
one-way analysis of variance (ANOVA). When log transformation failed to generate normally distributed data or the
equal variance test was not satisfied, nonparametric testing was
performed by the Mann-Whitney rank sum test between 2
groups and by Kruskal-Wallis one-way ANOVA on ranks
among 3 or more groups. Correlations were determined by
Pearson’s product-moment correlation for interval data and by
Spearman’s rank order correlation for ordinal data or for
interval data which did not follow a normal distribution.
Nominal data were analyzed by chi-square analysis-ofcontingency tables.
RESULTS
Disproportionate elevation of SF BLyS protein
and APRIL levels in IA patients. SF obtained from 103
clinically swollen knees of 89 patients and the corre-
LOCAL PRODUCTION OF BLyS AND APRIL IN INFLAMMATORY ARTHRITIS
985
Figure 1. Serum and synovial fluid (SF) levels of B lymphocyte stimulator (BLyS) protein and
APRIL in arthritis patients. Top left, BLyS protein levels from concurrently obtained serum and SF
from individual patients with osteoarthritis/traumatic arthritis (OA/Tr; 22 serum, 22 SF samples),
rheumatoid arthritis (RA; 45 serum, 49 SF samples), other noninfectious inflammatory arthritis
(Inflam; 17 serum, 18 SF samples), or crystal-induced arthritis (Crys; 14 serum, 14 SF samples) are
depicted as circles. The lines inside the boxes indicate the medians; the outer borders of the boxes
indicate the 25th and 75th percentiles; the bars extending from the boxes indicate the 10th and 90th
percentiles. Top right, The results are plotted as the ratio of SF to serum BLyS protein levels.
Bottom left, APRIL levels from concurrently obtained serum and SF from individual patients with
OA/traumatic arthritis (19 serum, 22 SF samples), RA (42 serum, 49 SF samples), other
noninfectious inflammatory arthritis (16 serum, 18 SF samples), or crystal-induced arthritis (13
serum, 13 SF samples). Bottom right, The results are plotted as the ratio of SF to serum APRIL
levels.
sponding serum samples were analyzed for BLyS protein
levels. Although the range of serum BLyS protein levels
was broad, the collective values were very similar among
all the patient cohorts (P ⫽ 0.644) (Figure 1, top). In
each patient cohort, BLyS protein levels in SF from
clinically affected knees were significantly higher than
the levels in concurrently obtained serum samples (P ⱕ
0.001 for each cohort). Nevertheless, BLyS protein levels
in SF from each of the IA cohorts were significantly
higher than the levels in SF from the NIA patients (P ⬍
0.001). SF and serum BLyS protein levels in an additional patient with gonococcal arthritis (13 and 3.3
ng/ml, respectively) and in an additional patient with
tuberculous arthritis (14 and 5.8 ng/ml, respectively)
were similar to those in the IA patients.
This dissimilarity between IA and NIA patients
was further highlighted by analysis of the SF-to-serum
BLyS protein ratios. A ratio of 3 was arbitrarily defined
as a marked increased in SF BLyS protein level relative
to that in serum. In OA/traumatic arthritis patients, only
1 of 22 SF samples demonstrated a BLyS protein ratio
ⱖ3, whereas 18 of the 49 RA, 6 of the 18 other
noninfectious IA, and 5 of the 14 crystal-induced arthritis SF samples demonstrated BLyS protein ratios ⱖ3
(P ⫽ 0.041 for RA versus each of the other 3 groups, and
P ⫽ 0.009 for the entire IA group versus the NIA group).
APRIL levels were measured in 93 SF samples
and their corresponding sera. As was the case with BLyS
protein, serum APRIL levels were very similar among all
the patient cohorts (P ⫽ 0.405) (Figure 1, bottom). In
contrast to the case with BLyS protein, SF APRIL levels
in clinically affected knees from NIA patients were lower
than those in the corresponding sera (P ⬍ 0.001), with
an SF-to-serum APRIL ratio of ⬍1 in 17 of 19 cases.
Nevertheless, the SF-to-serum ratios in the majority (52
of 74) of IA samples were ⬎1 (P ⬍ 0.001), and SF
986
TAN ET AL
Figure 2. Surface expression of BLyS protein by blood and SF mononuclear cells in arthritis patients. Top, Mononuclear cells
from concurrently obtained blood and SF from 2 OA, 2 RA, 2 systemic lupus erythematosus (SLE), and 2 gout patients were
double-stained for surface CD14 and BLyS protein. Results are presented as contour plots. The numbers in the upper right
quadrants of each tracing indicate the percentages of CD14⫹ cells that stained positive for BLyS protein. Bottom, Percentages
of CD14⫹ cells that stained positive for BLyS protein from blood and SF from individual patients with OA/traumatic arthritis
(12 blood, 11 SF samples), RA (33 blood, 36 SF samples), other noninfectious inflammatory arthritis (16 blood, 17 SF samples),
or crystal-induced arthritis (9 blood, 9 SF samples) are depicted as circles. The lines inside the boxes indicate the medians; the
outer borders of the boxes indicate the 25th and 75th percentiles; the bars extending from the boxes indicate the 10th and 90th
percentiles. See Figure 1 for other definitions.
APRIL levels in clinically affected knees from IA patients
were collectively higher than those from NIA patients (P ⬍
0.001 for RA versus each of the other 3 groups, and P ⬍
0.001 for the entire IA group versus the NIA group).
Expression of surface BLyS protein by blood and
SF mononuclear cells. To assess whether SF BLyS
protein expression is also disproportionately elevated in
IA patients, blood and SF mononuclear cells from IA
and NIA patients were stained for surface BLyS protein.
Staining profiles from representative individuals within
each patient cohort are presented in Figure 2 (top). As
anticipated, the great majority of BLyS protein–positive
cells in both blood and SF was limited to the CD14⫹ cell
population (monocytes) (1–3,5,27), although in some
LOCAL PRODUCTION OF BLyS AND APRIL IN INFLAMMATORY ARTHRITIS
987
Figure 3. Correlations among levels of B lymphocyte stimulator (BLyS) protein or APRIL in synovial fluid (SF) or blood and counts of nucleated
cells. Each circle indicates an individual SF or blood sample. The lines are the calculated regression lines. Top, SF monocyte, lymphocyte, and
neutrophil counts were determined along with SF BLyS protein levels in 85 samples. SF nucleated cell counts were determined along with SF BLyS
protein levels in 102 samples. White blood cell (WBC) counts in blood were determined along with serum BLyS protein levels in 97 samples. Middle,
SF monocyte, lymphocyte, and neutrophil counts were determined along with SF APRIL levels in 83 samples. SF nucleated cell counts were
determined along with SF APRIL levels in 100 samples. WBC counts in blood were determined along with serum APRIL levels in 93 samples.
Bottom, Serum BLyS protein and APRIL levels were both determined in 90 blood samples. SF BLyS protein and APRIL levels were both
determined in 101 SF samples. Serum and SF BLyS protein levels were determined in 103 matched samples. Serum and SF APRIL levels were
determined in 93 matched samples.
samples, there were small CD14– cell populations that
also stained positive for BLyS protein.
Overall, there was considerable variability in
BLyS protein staining patterns, even among patients
within the same cohort. As illustrated in Figure 2
(bottom), there was no consistent increase in surface
BLyS protein expression by monocytes in blood or SF
from IA patients relative to that in blood or SF from
NIA patients. In fact, the percentages of BLyS protein–
positive monocytes in RA SF were actually lower than
the percentages in SF from the other patient cohorts
(including OA/traumatic arthritis, P ⬍ 0.001) despite
BLyS protein levels being significantly higher in RA SF
than in OA/traumatic arthritis SF (Figure 1).
Correlations between local inflammatory cell response and BLyS protein or APRIL levels in SF. The
lack of positive association between the percentages of
BLyS protein–positive cells and the actual BLyS protein
988
TAN ET AL
Table 2. BLyS protein and APRIL levels in patients who underwent concurrent bilateral knee
arthrocenteses*
BLyS protein, ng/ml
APRIL, ng/ml
Synovial fluid
Synovial fluid
Patient
Diagnosis
Serum
Left knee
Right knee
Serum
Left knee
Right knee
A
B
C
D
E
RA
RA
RA
RA
ReA
1.4
3.2
6.3
3.4
6.6
5.8
4.2
18
13
18
6.6
5.5
20
11
18
ND
35
20
17
23
100
29
19
34
260
98
40
22
43
78
* RA ⫽ rheumatoid arthritis; ND ⫽ not determined; ReA ⫽ reactive arthritis.
levels in SF was an unexpected finding. However, the
absolute number of BLyS protein–positive cells (monocytes) in the affected joints of IA patients was higher
than that in the affected joints of NIA patients. Indeed,
there was a significant positive correlation between
SF monocyte counts and SF BLyS protein levels (Figure 3, top). Similar positive correlations were also
noted between SF lymphocyte or neutrophil counts and
SF BLyS protein levels. Importantly, the SF total nucleated cell counts correlated with SF BLyS levels the
strongest, suggesting that the intensity of the local
inflammatory response in the joint, rather than just the
monocyte count alone, is a major determinant of SF
BLyS protein levels.
Qualitatively similar, and quantitatively more
striking, correlations were also observed between SF
APRIL levels and SF cell counts (Figure 3, middle). No
correlations were noted between serum BLyS protein or
APRIL levels and white blood cell counts in blood,
which is consistent with circulating BLyS protein and
APRIL arising predominantly from extravascular sites
(i.e., production in secondary lymphoid tissues and
release into the circulation).
Lack of immutable concordance between BLyS
protein and APRIL levels in SF. SF and serum BLyS
protein levels correlated with each other, but SF and
serum APRIL levels did not (Figure 3, bottom). This
suggested that SF APRIL levels may be less dependent
Figure 4. Surface expression of B lymphocyte stimulator (BLyS) protein by blood and synovial
fluid (SF) mononuclear cells in arthritis patients who underwent bilateral knee arthrocentesis on
the same day. Mononuclear cells from concurrently obtained blood and SF from both knees of a
patient with rheumatoid arthritis (RA; patient D in Table 2) and a patient with reactive arthritis
(ReA; patient E in Table 2) were double-stained for surface CD14 and BLyS protein. Results are
presented as contour plots. The numbers in the upper right quadrants of each tracing indicate the
percentages of CD14⫹ cells that stained positive for BLyS protein.
LOCAL PRODUCTION OF BLyS AND APRIL IN INFLAMMATORY ARTHRITIS
upon global systemic factors than are SF BLyS protein
levels. If this were the case, then BLyS protein levels in
SF collected at a single time point from anatomically
distinct sites might not necessarily be paralleled by
corresponding APRIL levels. Moreover, changes in serum and/or SF BLyS protein levels over time might not
be paralleled by corresponding changes in APRIL levels.
Five patients underwent therapeutic arthrocenteses of both knees on the same day (Table 2). In 2 of
these patients, staining of blood and SF mononuclear
cells was also performed (Figure 4). In each patient, the
BLyS protein staining patterns for SF mononuclear cells
from each knee were remarkably similar but clearly
distinct from the staining pattern of corresponding blood
mononuclear cells. Although serum and SF BLyS protein levels varied considerably among the 5 patients, SF
BLyS protein levels for each patient were very similar in
each knee. Serum and SF APRIL levels also varied
considerably among these patients, with APRIL levels in
SF from each knee being similar in 4 of the patients.
However, in the fifth patient (patient E), there was a
marked disparity in SF APRIL levels in the two knees.
Eight patients underwent arthrocentesis on more
than one occasion (Table 3). In 5 of them (patients F, H,
I, J, and K), changes (or lack of changes) in SF BLyS
protein and APRIL levels occurred in parallel. In addition, the expression of BLyS protein on SF mononuclear
cells obtained from either arthrocentesis in patient F was
very similar (Figure 5, right). However, in 3 patients,
there was a clear discordance between interval changes
in SF BLyS protein levels and SF APRIL levels. In
patient E, BLyS protein levels in serum and SF at the
time of the second arthrocentesis were substantially
lower than those at the time of the first, despite there
being only a 2-day interval between the two procedures.
Of interest, BLyS protein expression on SF mononuclear
cells from this patient was relatively unchanged (Figure
5, left), pointing to an uncoupling of BLyS protein
expression from BLyS protein levels. In contrast to the
decrease in SF BLyS protein levels, serum APRIL levels
remained unaffected, and SF APRIL levels in 1 joint
modestly rose. In patients G and L, SF and serum BLyS
protein levels fell over time. In contrast, serum and SF
APRIL levels rose by the third and second arthrocenteses, respectively.
DISCUSSION
In patients with inflammatory arthritis, BLyS
protein and APRIL levels in SF from clinically affected
knees were significantly higher than the levels in corre-
989
sponding sera (Figure 1). Although BLyS protein levels
(but not APRIL levels) in patients with noninflammatory arthritis were also greater in SF from clinically
affected knees than in sera, the levels in SF from IA
patients were much higher than those in SF from NIA
patients, despite all patients having similar serum BLyS
protein levels.
The difference in SF BLyS protein and APRIL
levels between IA and NIA patients almost certainly
related in large measure to the intensity of the local
inflammatory responses in the affected joints of the
respective patients (Figure 3). With regard to BLyS
protein, not only did SF BLyS protein levels correlate
significantly with SF monocyte counts (the putative
BLyS protein–producing cells), but they also correlated
significantly with SF lymphocyte counts and SF neutrophil counts. Although other interpretations are possible,
the simplest explanation is that increased numbers of
BLyS protein–producing cells, in the presence of increased numbers of activated lymphocytes and neutrophils, led to increased levels of BLyS protein.
The immediate stimulus to local production of
BLyS protein in inflammatory SF is unknown, but
proinflammatory cytokines such as interferon-␥ (IFN␥)
and IFN␣ are almost certainly contributory (27,28). We
did not measure BLyS protein levels in SF from clinically
unaffected joints of IA or NIA patients, but we predict
that there would be low BLyS protein levels in SF from
joints with low-to-absent degrees of inflammation.
Given the strong correlation between serum and SF
BLyS protein levels, we propose that SF BLyS protein
levels are determined by local factors (e.g., degree of
inflammation, cytokine milieu) that accentuate a systemic proclivity to BLyS protein production.
Of note, there was no consistent pattern of BLyS
protein expression by SF mononuclear cells on a per-cell
or a percentage basis, even within a given patient cohort
(Figure 2, top). The great majority of BLyS protein–
positive cells were CD14⫹, although in some patients,
CD14– cells staining positively for BLyS protein were
detected. Whether these cells represent CD14⫹ monocytes that have down-regulated their surface CD14 in
vivo, other myeloid lineage cells that are inherently
CD14–, or CD14– nonmyeloid lineage cells that can
express surface BLyS protein or to which BLyS protein
passively adheres in vivo remains to be established.
Also of note, staining of SF mononuclear cells
from patient E obtained from the 2 arthrocenteses
(Figure 5) failed to reveal a difference in surface BLyS
protein expression despite the substantial difference in
BLyS protein levels (Table 3). This suggests a divergence
990
TAN ET AL
Table 3.
BLyS protein and APRIL levels in patients with repeated arthrocenteses
Patient,
diagnosis,*
arthrocentesis
E, ReA
1
2
2
F, RA
1
2
G, RA
1
2
3
H, AS
1
2
I, ReA
1
2
J, RA
1
2
K, RA
1
2
L, RA
1
2
BLyS protein, ng/ml
Interval,
days
APRIL, ng/ml
Knee
Serum
Synovial
fluid
Serum
Synovial
fluid
Ipsilateral
Contralateral
26
6.6
6.6
29
18
18
22
23
23
210
78
260†
0
63
Ipsilateral
2.4
2.8
6.4
7.6
26
26
31
37
0
91
290
Ipsilateral
Ipsilateral
5.6
4.1
2.6
22
7.0
7.3
12
2.8
27
76
25
64†
0
14
Ipsilateral
3.8
2.3
7.2
9.1
24
21
16
14
0
35
Contralateral
1.7
1.5
5.1
4.1
36
39
6.9
7.0
0
78
Contralateral
2.6
6.3
12
18
18
13
23
39
0
82
Contralateral
1.9
3.2
6.2
8.7
9.3
21
28
30
0
147
Contralateral
4.2
1.0
11
3.2
13
30
9.3
37†
0
2
2
* ReA ⫽ reactive arthritis; RA ⫽ rheumatoid arthritis; AS ⫽ ankylosing spondylitis.
† Unambiguous discordance between the interval changes in synovial fluid levels of BLyS protein and
APRIL.
Figure 5. Surface expression of B lymphocyte stimulator (BLyS) protein by blood and synovial fluid (SF) mononuclear cells in
arthritis patients who underwent repeat arthrocentesis of the same knee. At the time of the first arthrocentesis (day 0, time 1),
mononuclear cells were isolated from blood and SF from the clinically affected knee of a patient with reactive arthritis (ReA; patient
E in Table 3) and a patient with rheumatoid arthritis (RA; patient F in Table 3) and were double-stained for surface CD14 and BLyS
protein. Repeat arthrocentesis of the same knee and phlebotomy were performed on day 2 (ReA patient) and day 63 (RA patient)
(time 2). Results are presented as contour plots. The numbers in the upper right quadrants of each tracing indicate the percentages
of CD14⫹ cells that stained positive for BLyS protein.
LOCAL PRODUCTION OF BLyS AND APRIL IN INFLAMMATORY ARTHRITIS
between BLyS protein expression and BLyS protein
production. Since soluble BLyS protein arises from the
cleavage of surface membrane BLyS protein (3), regulation of such cleavage may be critical to the ultimate
levels of soluble BLyS protein. At present, very little is
known about the regulation of the release of soluble
BLyS protein from the membrane-bound form. Indeed,
it may be that accelerated cleavage of surface BLyS
protein contributes to reduced percentages of BLyS
protein–positive monocytes in SF from RA patients
(Figure 2), despite elevated BLyS protein levels in these
same SF (Figure 1). Development of an assay that
identifies individual BLyS protein–producing cells (e.g.,
enzyme-linked immunospot assay) would permit formal
assessment of the relationship between BLyS protein–
expressing cells and BLyS protein–producing cells.
As was the case for SF BLyS protein levels, SF
APRIL levels also correlated significantly with SF
monocyte, lymphocyte, neutrophil, and total nucleated
cell counts. The identity of the factors which promote
APRIL production and the cells actually producing
increased amounts of APRIL remain to be established.
It is likely that temporal differences in the local production of various cytokines that may differentially affect
APRIL and BLyS protein production occur, and such
differences may substantially contribute to the discordance between SF APRIL and BLyS protein levels.
Given the lack of correlation between serum and SF
APRIL levels, we propose that SF APRIL levels are
largely locally regulated and highly independent of systemic APRIL production.
This presumed divergence between the regulation of SF BLyS protein levels and the regulation of SF
APRIL levels is highlighted by the findings in patients
from whom more than one SF sample was obtained. In
the 5 IA patients who underwent arthrocentesis of both
knees on the same day, SF BLyS protein levels were
remarkably similar in each joint (Table 2). In the 2
patients whose SF mononuclear cells from both knees
were stained, BLyS protein expression was also highly
similar in each (Figure 4). In contrast, SF APRIL levels
in both knees of one of these patients were markedly
disparate (Table 2). Moreover, in patients studied on
multiple occasions (Table 3), those whose SF showed
considerable sequential reduction in BLyS protein levels
were patient E, who was started on an around-the-clock
regimen of an antiinflammatory agent, patient G, who
received intraarticular corticosteroids, and patient L,
who was started on disease-modifying agents and a TNF
antagonist. No consistent parallel reductions in SF
APRIL levels were observed in these patients.
991
The apparent differences in their regulation notwithstanding, increased SF levels of BLyS protein and
APRIL are each likely to be a consequence of the local
inflammatory response rather than its proximate cause.
Nevertheless, rather than being an epiphenomenon of
uncertain biologic importance, there is considerable,
albeit inferential, evidence in animal models that points
to a vital role for B cell survival factors, such as BLyS
protein and/or APRIL, in noninfectious, non–crystalinduced arthritis. First, when CD4⫹ T cells isolated
from individual follicles of human RA synovium are
injected into SCID mice previously transplanted with
major histocompatibility complex–matched RA synovium, production of proinflammatory cytokines (as measured by messenger RNA levels) in the transplanted
synovium is dramatically up-regulated. However, if the
target synovium is inherently deficient in B cells or if the
recipient mice are treated with anti-CD20 mAb to
render the synovium deficient in B cells, then adoptively
transferred CD4⫹ T cells do not up-regulate the proinflammatory cytokines (29). These results imply that
factors, such as BLyS protein and/or APRIL, could
facilitate T cell–driven inflammation in RA by enhancing B cell survival in the synovium. In the absence of a
chronic T cell–driven inflammatory process, elevated
BLyS protein and/or APRIL levels may have little
biologic impact. Accordingly, elevated BLyS protein and
APRIL levels in crystal-induced arthritis may have no
(or little) pathogenetic ramifications due to no (or little)
chronic T cell–driven inflammation in this condition.
Second, the inflammation and joint destruction
associated with murine collagen-induced arthritis (a
disease highly dependent upon T cells) are inhibited,
even after their induction, by the BLyS protein/APRIL
antagonist, TACI-Ig (7,25). Since the mice treated with
TACI-Ig underwent a significant reduction in B cells, the
observations collectively suggest that B cells are vital to
disease pathogenesis. By extension, B cell survival factors, including BLyS protein and/or APRIL, could facilitate the development and/or exacerbation of IA in
humans. In situ studies of the spatial relationship between BLyS protein–expressing cells and B cell clusters
in synovial tissue are currently in progress. In any case,
BLyS protein/APRIL antagonists may have salutary
effects on IA and find an important niche in the
management of patients with IA.
ACKNOWLEDGMENTS
The authors thank Hal Soucier for performing the flow
cytometry, the attending physicians and house officers of the
992
TAN ET AL
Division of Rheumatology for their assistance in identifying
and recruiting patients for this study, and all the subjects for
their participation.
REFERENCES
1. Moore PA, Belvedere O, Orr A, Piwei K, LaFleur DW, Feng P, et
al. BLyS: member of the tumor necrosis factor family and B
lymphocyte stimulator. Science 1999;285:260–3.
2. Shu H-B, Hu W-H, Johnson H. TALL-1 is a novel member of the
TNF family that is down-regulated by mitogens. J Leukocyte Biol
1999;65:680–3.
3. Schneider P, MacKay F, Steiner V, Hofmann K, Bodmer JL,
Holler N, et al. BAFF, a novel ligand of the tumor necrosis factor
family, stimulates B cell growth. J Exp Med 1999;189:1747–56.
4. Mukhopadhyay A, Ni J, Zhai Y, Yu G-L, Aggarwal BB. Identification and characterization of a novel cytokine, THANK, a TNF
homologue that activates apoptosis, nuclear factor-␬B, and c-Jun
NH2-terminal kinase. J Biol Chem 1999;274:15978–81.
5. Tribouley C, Wallroth M, Chan V, Paliard X, Fang E, Lamson G,
et al. Characterization of a new member of the TNF family
expressed on antigen presenting cells. Biol Chem 1999;380:
1443–47.
6. Gross JA, Johnston J, Mudri S, Enselman R, Dillon SR, Madden
K, et al. TACI and BCMA are receptors for a TNF homologue
implicated in B-cell autoimmune disease. Nature 2000;404:995–9.
7. Gross JA, Dillon SR, Mudri S, Johnston J, Littau A, Roque R, et
al. TACI-Ig neutralizes molecules critical for B cell development
and autoimmune disease: impaired B cell maturation in mice
lacking BLyS. Immunity 2001;15:289–302.
8. Schiemann B, Gommerman JL, Vora K, Cachero TG, ShulgaMorskaya S, Dobles M, et al. An essential role for BAFF in the
normal development of B cells through a BCMA-independent
pathway. Science 2001;293:2111–4.
9. Thompson JS, Schneider P, Kalled SL, Wang L, Lefebre EA,
Cachero TG, et al. BAFF binds to the tumor necrosis factor
receptor-like molecule B cell maturation antigen and is important
for maintaining the peripheral B cell population. J Exp Med
2000;192:129–35.
10. Do RKG, Hatada E, Lee H, Tourigny MR, Hilbert D, Chen-Kiang
S. Attenuation of apoptosis underlies B lymphocyte stimulator
enhancement of humoral immune response. J Exp Med 2000;192:
953–64.
11. Batten M, Groom J, Cachero TG, Quian F, Schneider P, Tschopp
J, et al. BAFF mediates survival of peripheral immature B
lymphocytes. J Exp Med 2000;192:1453–66.
12. Hsu BL, Harless SM, Lindsley RC, Hilbert DM, Cancro MP.
Cutting edge: BLyS enables survival of transitional and mature B
cells through distinct mediators. J Immunol 2002;168:5993–6.
13. Mackay F, Woodcock SA, Lawton P, Ambrose C, Baetscher M,
Schneider P, et al. Mice transgenic for BAFF develop lymphocytic
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
disorders along with autoimmune manifestations. J Exp Med
1999;190:1697–710.
Khare SD, Sarosi I, Xia X-Z, McCabe S, Miner K, Solovyev I, et
al. Severe B cell hyperplasia and autoimmune disease in TALL-1
transgenic mice. Proc Natl Acad Sci U S A 2000;97:3370–5.
Hahne M, Kataoka T, Schröter M, Hofmann K, Irmler M, Bodmer
JL, et al. APRIL, a new ligand of the tumor necrosis factor family,
stimulates tumor cell growth. J Exp Med 1998;188:1185–90.
Marsters SA, Yan M, Pitti RM, Haas PE, Dixit VM, Ashkenazi A.
Interaction of the TNF homologues BLyS and APRIL with the
receptor homologues BCMA and TACI. Curr Biol 2000;10:785–8.
Yu G, Boone T, Delaney J, Hawkins N, Kelley M, Ramakrishnan
M, et al. APRIL and TALL-1 and receptors BCMA and TACI:
system for regulating humoral immunity. Nat Immunol 2000;1:
252–6.
Wu Y, Bressette D, Carrell JA, Kaufman T, Feng P, Taylor K, et
al. Tumor necrosis factor (TNF) receptor superfamily member
TACI is a high affinity receptor for TNF family members APRIL
and BLyS. J Biol Chem 2000;275:35478–85.
Rennert P, Schneider P, Cachero TG, Thompson J, Trabach L,
Hertig S, et al. A soluble form of B cell maturation antigen, a
receptor for the tumor necrosis factor family member APRIL,
inhibits tumor cell growth. J Exp Med 2000;192:1677–84.
López-Fraga M, Fernández R, Albar JP, Hahne M. Biologically
active APRIL is secreted following intracellular processing in the
Golgi apparatus by furin convertase. EMBO Rep 2001;2:945–51.
Stein JV, López-Fraga M, Elustondo FA, Carvalho-Pinto CE,
Rodriguez D, Gomez-Caro R, et al. APRIL modulates B and T
cell immunity. J Clin Invest 2002;109:1587–98.
Zhang J, Roschke V, Baker KP, Wang Z, Alarcón GS, Fessler BJ,
et al. Cutting edge: a role for B lymphocyte stimulator in systemic
lupus erythematosus. J Immunol 2001;166:6–10.
Cheema GS, Roschke V, Hilbert DM, Stohl W. Elevated serum B
lymphocyte stimulator levels in patients with systemic immunebased rheumatic diseases. Arthritis Rheum 2001;44:1313–9.
Groom J, Kalled SL, Cutler AH, Olson C, Woodcock SA, Schneider P, et al. Association of BAFF/BLyS overexpression and altered
B cell differentiation with Sjögren’s syndrome. J Clin Invest
2002;109:59–68.
Wang H, Marsters SA, Baker T, Chan B, Lee WP, Fu L, et al.
TACI-ligand interactions are required for T cell activation and
collagen-induced arthritis in mice. Nat Immunol 2001;2:632–7.
Klippel JH, Crofford LJ, Stone JH, Weyand CM, editors. Primer
on the rheumatic diseases. 12th ed. Atlanta: Arthritis Foundation;
2001.
Nardelli B, Belvedere O, Roschke V, Moore PA, Olsen HS,
Migone TS, et al. Synthesis and release of B-lymphocyte stimulator
from myeloid cells. Blood 2001;97:198–204.
Litinskiy MB, Nardelli B, Hilbert DM, He B, Schaffer A, Casali P,
et al. DCs induce CD40-independent immunoglobulin class
switching through BLyS and APRIL. Nat Immunol 2002;3:822–9.
Takemura S, Klimiuk PA, Braun A, Goronzy JJ, Weyand CM. T
cell activation in rheumatoid synovium is B cell dependent.
J Immunol 2001;167:4710–8.
Документ
Категория
Без категории
Просмотров
1
Размер файла
185 Кб
Теги
production, local, april, patients, joint, protein, arthritis, inflammatory, stimulate, lymphocytes
1/--страниц
Пожаловаться на содержимое документа