вход по аккаунту


Expression and function of CD80 and CD86 costimulator molecules on synovial dendritic cells in chronic arthritis.

код для вставкиСкачать
Vol. 39, No. 8, August 1996, pp 1287-1291
Q 1996, American College ol Rheumatology
Objective. To examine CDS6 expression on dendritic cells isolated from the synovial fluid (SFDC) of
patients with chronic arthritis, and to determine the
importance of both CDSO and CD86 molecules in
SFDC-T lymphocyte interactions.
Methods. CDS6 messenger RNA (mRNA) and
surface expression were analyzed in SFDC using reverse
transcriptase-polymerase chain reaction and flow cytometry, respectively. The costimulator activity of the
SFDC CDSO and CD86 molecules was determined by
allogeneic mixed lymphocyte reaction (MLR). CDSO and
CDS6 induction on SFDC during in vitro culture was
also examined.
Results. Fresh SFDC either lacked or showed very
weak surface expression of CDS6 molecules (as shown
previously for CDSO), yet contained CD86 mRNA. CDSO
antibodies minimally inhibited an allogeneic MLR,
whereas CDS6 antibodies and CTLA-4 Ig showed significant inhibition. Both CDSO and CD86 molecules were
inconsistently induced on SFDC following culture in
either media, interferon-y, o r granulocyte-macrophage
colony-stimulating factor.
Conclusion. SFDC may be defective antigenpresenting cells in vivo. The ability of CDSO and CD86
molecules to be induced and become functional on
SFDC in vitro implies the presence of a negative regulatory compound(s) in the synovial environment.
Dendritic cells (DC) are antigen-presenting cells
that are highly efficient at acquiring and presenting
Supported by a Canterhury Medical Research Foundation
grant, and a New Zealand Lotteries Board grant.
Kelly L. Summers, HBSc, John L. O’Donnell, FRACP,
FRCPA, Lisa A. Williams, HBSc, Derek N. J. Hart, FRACP, FRCPA,
DPhil: Christchurch Hospital, Christchurch, New Zealand.
Address reprint requests to John L. O’Donncll, FKACP,
FRCPA, P.O. Box 151, Private Bag, Christchurch Hospital, Christchurch, New Zealand.
Submitted for publication June 2, 1995; accepted in revised
form March 18, 1996.
antigens to T lymphocytes (1). For complete T lymphocyte activation, major histocompatibility complex
(MHC) class 11-rcstricted antigen presentation (signal
1) and a costimulatory signal (signal 2) are required
(2-4). In DC-T cell interactions, the CD80/CD86CD28/CTLA-4 costimulatory pathway appcars important. The likely involvement of DC in the pathogenesis
of chronic arthritis encouraged us to study the expression of these costimulator molecules on DC in this
We previously made the unexpected observation
that DC isolated from synovial fluid (SFDC) of patients
with chronic arthritis had the high surface density of
MHC class I1 molecules necessary for the induction of
signal 1, yet had low to undetectable levels of the CD80
costimulator molecule ( 5 ) . Given the apparent predorninant role of CD86 in DC costimulation (6), and the
availability of CD86 monoclonal antibodies (MAb), we
undertook the present study to examine SFDC for CD86
expression, and to determine the functional importance
of the CD80/CD86-CD28/CTLA-4 costimulatory pathway in chronic arthritis.
Patients and controls. After informed consent was
established, SF was collected by routine kncc joint aspiration
from 22 patients with chronic arthritis (9 rheumatoid arthritis
seronegative oligoarthritis, 6 psoriatic arthritis, and 1
monarthritis). The patients with KA wcrc classified according
to the 1987 revised criteria of the American College of
Rheumatology (7). Patients with seronegative oligoarthritis
had clinical involvement of at least 2, but fewer than 5, lower
limb joints for at least 12 months, as did patients with psoriatic
arthritis, who also had cutancous psoriasis. The patient with
monarthritis had experienced chronic knee joint swelling (despite synovectomy) for >8 years. Blood samples were donated
by healthy volunteers to serve as controls. All samples were
collected into EDTA blood tubes.
MAb and cytokines. MAb used in this study were BB-1
(CD80; Dr. J. Ledbetter, Seattle, WA), L307 (CD80; 5th
Leukocyte Differentiation Antigen Workshop [LDAW] and
Becton Ilickinson, Sydney, Australia), I3U-63 (CD86; Serotec,
Oxford, UK), IT-2 (CD86; Ilr. M. Azuma, Tokyo, Japan),
FUN-1 (CD86; LDAW), C‘I’LA-4 Ig fusion protein (I37 ligand;
Dr. P. I,insley, Seattle, WA), FMC-63 (CD19; Dr. H. Zola,
Adelaide, Australia), CMKF-3 1 (CD14; our laboratory), and
CMRF-7 (CD15; our laboratory). Those MAb obtained from
the American Type Culture Collection (Rockville, MD) included OKT3 (CD3), OKM1 (CDllb), and HNK-1 (CD57).
Negative controls included X63 (IgG1; Dr. H. Zola, Adelaidc,
Australia), CMRF-7 and CMRF-15 (IgM; our laboratory),
human Ig (our laboratory), and mouse Ig (our laboratory).
Cytokines included human interferon-y (IFN-y; Boehringer
Ingclheim, Auckland, NZ) and human granulocyte-macrophage
colony-stimulating factor (GM-CSF; Ixucomax, SandozPharma,
Auckland, NZ).
Isolation of fresh SFDC. SFDC were isolated as prcviously described ( 5 ) . Briefly, SF mononuclear cells (MC)
obtained over a Ficoll-Hypaque gradient (FI 1; density 1.077)
were depleted of T lymphocytcs by rosetting, and either the
isolation was continued or cells were kept overnight at 4°C.
The next day, cells were labeled with a “cocktail” of CD3,
CDllb, Cll14, CD15, CD19, and CD57 MAb, and depleted at
4°C by anti-mouse Ig panning. Residual labeled cells were
further depleted using either immunomagnetic MACS beads
(Miltenyi Biotcc GmbII, I3ergisch Gladbach, Germany) or a
Vantage fluoresccnce-activated cell sorter (Becton Dickinson).
The remaining unlabeled cells constituted the final S F l X
population. Purity ranged between 85% and 100% (mean
94.6%; n = 18), as defined by the absence of mature lineagespecific markers, homogenous size, and strong expression of
MHC class I1 antigens.
Preparationof normal peripheral blood (PB) T lymphocytes. PBMC were isolated over an FH gradient, and T
lymphocytes were isolated by rosetting with neuraminadasetreated sheep erythrocytes (5). This resulted in 90-98% pure
CD3-positive cells, as determined by flow cytometry.
Immunofluorescence labeling. Purified SFDC wcrc
incubated with primary MAb or recombinant proteins, and
binding was detected by cithcr fluorescein isothiocyanatc
(F1TC)-conjugated sheep anti-mouse IgG (Silenus Laboratories, Hawthorn, Australia) or FII‘C-conjugated goat
anti-human IgG (‘I’ago, Rurlingamc, CA), respectively, as
prcviously described ( 5 ) . lmmunofluorescencc analysis was
performed on an Epics flow cytometcr (Coulter Electronics,
Hialeah, FL). Since detectable levels of the CD80iCD86
molecules were so low, positive expression was defined as
either a change in mean channel fluorescence >2, or a
percentagc shift >5, above the value for the isotype-matched
ncgative control. The I Iodgkin’s disease-derived cell line,
L428, was used as a CD80iCD86 positive control (Dr. V. Diehl,
Cologne, Germany).
Allogeneic mixed lymphocyte reaction (MLR). Enriched S F I X were incubated with 25 Fgiml of mitomycin-C for
30 minutes at 3TC, and washed 3 times in phosphate buffered
saline. Treated cells (1-2.5 X 10‘) were further incubated with
titrated amounts of MAb for 30 minutes at 4°C prior to the
addition of 0.5-1 X 105 allogeneic normal PB T lymphocytes.
Cultures were maintained in 10% fetal calf serum-RPMI
culture media at 37°C and 5% CO, in 96-microwell roundbottomed plates. After 5 days of culture, T cell proliferative
activity was measured on a liquid scintillation analyzer following 16-18 hours incorporation of ’H-thymidine (5 Ci/mmole;
Amersham, Amersham, UK). Responses were reported as the
mean value of triplicate wells.
Induction of C1)SO and CD86 expression. SFDC wcrc
incubated in either media, 500 unitsiml IFN-y, or 500 units/ml
GM-CSF at 37°C in 5% CO, for variable amounts of time.
Cells wcrc labeled frcsh after culture with MAb directed
against the CD80 and CD86 molecules, as well as the CTLA-4
Ig fusion protein, and isotype-matched negative controls, and
analyzed by flow cytometry. Cells from the 1,428 line were
simultaneously labeled with each MAb to act as a positive
Reverse transcriptase-polymerase chain reaction (RTPCK). S1:IlC (1 X lo5) wcrc lysed (10 mM’I‘ris HCI [p€I 7.51,
200 mM NaCI, 1 mM ED’I’A, 0.5% sodium dodecyl sulfate)
and incubated at 42°C for 1 hour with 0.2 m g h l proteinasc K.
Samples were reverse transcribed and analyzed by PCK as
previously described (5).
Primers for CD86 wcrc 5’-GACCTGCTCATCTATACAC-3’ (forward, nuclcotides 593-61 1) and 5’-CTTCATCAGATC‘7TCAGG-3’ (reverse, nucleotides 1039-1OS7), and
primers for hypoxanthine phosphoribosyl transferase (HPRT)
were 5‘-GAACCAGGlTATGACCG-3’ (forward, nucleotides 139-157) and 5’-AGTCAAGGGCAI’ATCCTAC-3’
(reverse, nucleotides 663-681). Amplified DNA was hybridized to
internal oligonuclcotides: CD86 5’-GAGCITGAGGACCCTCAGCC-3’ (nucleotides 832-851) and HPRT 5’-CGTCITGCTCGAGATGTGA-3‘ (nucleotides 241-259).
Statistical analysis. Repeated-measured analysis of
variance was used to compare CD80 and CD86 induction on
SFDC and CDSO MAb, CD86 MAb, and Cn‘LA-4 Ig inhibition
of the MLR. When this indicated a significant difference
between the means, pairwise comparisons were madc using the
Tukey-Kramer multiple comparisons test.
Expression of CD80 and CD86 on Fresh synovial
DC. Fresh SFDC werc analyzed by flow eytometry for
surface expression of the costimulator molccules, CD80
(B7-1) and CD86 (B7-2), using 2 different anti-CD80
MAb (BB-1 and L307), 3 diffcrcnt anti-CD86 MAb
(BU-63, IT-2, and FUN-l), and the CTLA-4 Ig fusion
protein, which rccognizes both CD80 and CD86 molccules (Figure 1). There was no differcncc in staining
obscrved between the different MAb (data not shown).
Low levels of CD80 were detectcd in the SFDC
samples from none of 6 patients with RA, 1 of 5 with
seronegative oligoarthritis, 5 of 6 with psoriatic arthritis,
and 1 of 1 with monarthritis. CIX6 was cxpressed at vcry
low lcvels on thc SFDC samples from 1 of 6 patients with
RA, 1 of 5 with scronegativc oligoarthritis, 5 o f 6 with
psoriatic arthritis, and 1 of 1 with monoarthritis. Strong
staining of the L428-positive control cells was noted with
the CD80 and CD86 MAb (data not shown). To cnsurc
that the isolation procedure did not alter CD80 and
CD86 surface expression, B cells, T cells, and monocytes
from 2 samples were analyzed at each step and found to
be unaltered (data not shown).
CDSO and CD86 messenger RNA (mRNA) in
fresh SFDC. Using RT-PCR and Southern blot analysis,
CD86 mRNA was detected consistently in 6 of 6 fresh
SFDC samples (Figure 2). In comparison, we have
previously found CD80 mRNA to be either present in
low amounts or absent in fresh SFDC (5).
Functional importance of CD8O and CDS6 molecules. Thc variable presence of CD80 mRNA and
rcadily detectable CD86 mRNA in fresh SFDC suggested that these molecules had the potential to be
up-regulated on the cell surface. To determine if these
costimulator molecules were induced and functional in
SFDC-T lymphocyte interactions, we cocultured allogeneic normal PB T lymphocytes and SFDC in a primary
MLR in the presence of blocking CD80 MAb (BB-1;
n = 4), CD86 MAb (RU-63; n = 3), CD80/CD86 ligand
blocking CTLA-4 Ig fusion protein (n = 6), or their
isotype-matched controls (mouse Ig and human Ig). At
1.25 &ml and 2.5 pg/ml, CD86 MAb and CTLA-4 Ig
significantly inhibited the primary allogeneic MLR (P <
Figure 2. Presence of CD86 messenger RNA (mRNA) in fresh
synovial fluid dendritic cells (SFDC). Using reverse transcriptasepolymerase chain reaction, freshly isolated SFDC (lanes 1 and 2) were
analyzed for the presence of CD86 mKNA (94% and 95% pure,
respectively). Controls included LA28 complementary DNA (cDNA)
with ( +) or without (-) reverse transcriptase and no cDNA template
(C). Panels show the CD86 (upper) and hypoxanthine phosphoribosyl
transfcrase (HPRT) (lower) amplified polymerase chain reaction
products probed with specific g2P-ATP end-labeled internal oliE;onucleotides. SFDC from 2 patients with psoriatic arthritis are represented; however, SFDC from 2 patients with rheumatoid arthritis, 1
with seronegative oligoarthritis, and 1 with monarthritis were also
analyzed. In each case, CD86 mRNA was detected.
Figure 1. Flow cytometric analysis of CD80 (monoclonal antibody
[MAb] L307) and CD86 (MAb BU-63) costimulator molecule cxpres-
sion on 98% pure synovial fluid dendritic cells from a patient with
rheumatoid arthritis, labeled fresh and after 39 hours culture in media
(10% fetal calf serum-RPMI), granulocyte-macrophagc coionystimulating factor (GM-CSF; 500 units/ml), or interferon-y (IFN-7;
500 units/ml). This sample showed the greatest degree of induction.
Solid lines represent CD80, CD86, and CTLA-4 Ig staining, and dotted
lines represent the isotype-matched ncgative controls. Values shown
are the mean channel fluorescence/percentage of positive cells (minus
control). A minimum of 1 X lo4 cells wcre counted for each profile.
0.05 and I’ < 0.01, respectively), while CD80 MAb
provided no significant inhibition ( P = 0.88) (Figure 3).
This suggested that CD86 is more important for costimulation in an SFDC-stimulated primary immune
Induction of the CDSO and CD86 antigens on
SFDC. The inhibition of a primary allogeneic immune
reaction using CTLA-4 Ig implied that CD28/CTLA-4
ligands, primarily CD86 antigen, must be induced on
SFDC during the course of the MLR, perhaps by T
lymphocyte-derived stimuli. We therefore tested the
effect of culturing fresh SFDC in mcdia alone, IFN-y, or
GM-CSF on both CD80 and CD86 surface expression.
These cytokincs were used because wc have previously
demonstrated their ability to up-regulate, albeit weakly,
the CD80 antigen on normal D C (8).
The CD86 molecule was variably induced on the
SFDC of 2 of 2 patients with RA, 1 of 1 with psoriatic
arthritis, and 2 of 3 with seronegative oligoarthritis,
following culture in either media alone, IFN-y, or GM-
U 30000
1 .o
T cells onty
concentration (Irglnl)
Figure 3. Rcpresentativc mixed lymphocytc reaction (MLR) inhibitiop study using synovial fluid dendritic cclls (SPIX) (95% purity)
fiom a paticnt with psoriatic arthritis. The stirnulatory ability of SFDC
could be blockcd significantly by 0 8 6 monoclonal antibody (MAb)
an'g CTLA-4 Ig at 1.25 pghl and 2.5 pg/ml (P < 0.05 [n = 41 and
0.01 [n = 61, rcspectively). but not by CD80 MAb (P = 0.88 [ n -31) in an allogrncic M I X . Fresh SFDC ( I X 10') and allogcncic
peiipheral blood T IymphcKytcs ( 1 X I d ) were coculturcd with CD80
MAb (BB-I). CD86 MAb (BU-63), CTLA-4 Ig, and the isotype
cogtrols, mousc Ig (mIg) and human Ig (hulg), at the concentrations
indicated. The control T lymphocyte rcsponse (no stimulators) was
a@ includcd.
CSF (Figure 1). Thc CD80 moleculc was also variably
induced pn 4 of thcse 5 samples (2 RA, 1 seronegative
oligoarthritis, and 1 psoriatic arthritis). One seronegativk oligoarthritis SFDC sample induced neither CD80
no1 CD86 exprcssion following culture under any of
thebe conditions. Both CD80 and CD86 induction rcquired at lcast 36 hours culture. Due to the variability of
induction, these increases failed to rcach statistical
siknificancc ( P = 0.33).
We have now dcmonstrated that thc CD86 (B70/
U7-2) costimulator molecule is either abscnt or expressed weakly OD the surface of freshly isolated SFDC
obrained from joints with chronic arthritis, similar to our
previous findings with the CD80 (B7/UB-l/B7-1) cos<mulator molcculc (5). The weakly cxprcssing SFDC
w&e predominantly confined to patients with psoriatic
arthritis. In contrast, thc majority of SFDC from patients
wiih RA and seroncgative oligoarthritis lacked CD80
a i d CD86 expression. A more extensive study would be
nepessary to determinc whether these differences were
statistically significant.
Our findings for CD80 expression are supportcd
by an immunohistologic study of normal and rheumatoid
synovium (9), although Thomas et al have rcported
CD80 expression on rheumatoid SFDC (10). Liu et a1
have recently reported the presence of CD86+ cells
within thc rheumatoid synovium; however, it is noteworthy that such cells were not dctected in 7 of the 18
samples examined (1 1). Such differences may reflect
patient variation, diffcring cell populations studied, differcnt treatment regimens, or aberrant activation as a
result of thc isolation technique.
Both the CD80 and CD86 molecules are present
on normal PBDC that arc isolated by conventional
methods involving overnight culturc, and can be induced
on fresh PBDC following in vitro activation (4,8). In
addition, tonsil DC express the CD80 antigen (8), and
cultured epidermal Langcrhans cells express both CD80
and CD86 (12). Immunohistochemical studies have further dcmonstrated the presence of CD80 antigens on
cclls located in inflammatory or reactive environments,
such as tonsils and malignant Reed-Sternberg cells in
Hodgkin's diseasc (9). Thus, it was surprising to find that
DC isolated from joints with chronic arthritis either
lacked or wcakly expresscd the CD80 ( 5 ) and CD86
costimulator molecules. Our rcsults imply that SFDC
are not in a fully activated state, despite their origin in a
site rich in proinflammatory cytokines. These cytokines
might have been anticipated to up-rcgulate these molecules, as shown for the induction of the CD16 antigen on
normal PB monocytes by rheumatoid SF (13).
Fresh SFDC contain detectable levels of CD86
mRNA and, as prcviously shown, cither no or minimal
amounts of CD80 mRNA (5). Whilc it is possible that
the CD86 mRNA detected may have originated from a
contaminating cell population, it is morc likely, in view
of the importance of CD86 in the MLR, that SFDC do
contain significant levels of CD86 mRNA. Thus, the
CD86 antigcn has the potcntiai to be transcribed and
expressed on the cell surfacc undcr optimal conditions.
To determine if CD80KD86 molccules were
critical to SFDC function, we coculturcd fresh SFDC
from patients with chronic arthritis with normal allogencic PB T lymphocytes, and analyzed the effcct of
blocking thc CD80 and CD86 antigens on T lymphocyte
proliferation. There wa. no significant inhibition with
CD80 blocking MAb; howcver, there was significant
inhibition with both CD86 blocking MAb and CTLA-4
Ig. Therefore, dcspite an initial lack of CD80 and CD86
surfacc expression, fresh SFDC are still able to induce
and utilize a CTLA-4 Ig costimulatory ligand in vitro.
Our data suggest that CD86 is the predominant costimulator molecule used by SFDC in the allogcneic MLR.
The cytokines GM-CSF and IFN-y havc bccn
shown to regulate expression of the CD80 and CD86
molcculcs on various cell types. We demonstrated that a
subpopulation of isolated SFDC cultured in media, with
and without thcse cytokines, could express both the
CD80 and CD86 molecules, although there was great
variability between SFDC samples. Due to this variability, neither the changes in mcan percentage of positive
cells nor the changes in mean fluorcscencc intensity for
the group reached statistical significance. As other actiit may bc
vation molecules are found on fresh SFDC (3,
that a selective regulatory mechanism, acting to suppress
CD80/CD86 molecules, is present within the chronic
arthritic joint. Currently, interleukin-10 (IL-10) is thc
only cytokine known to inhibit CD80/CD86 expression,
as demonstrated on PB macrophages (14). Expression of
IL-10 has been found on synovial lining cells, primarily
macrophages, located in rheumatoid tissue; however,
there are conflicting reports as to its presence (15) or
absence (16) in rheumatoid joint fluid.
Absence of CD8O and CD86 costimulator molecules on an antigen-presenting cell during its intcraction
with a T cell results in T cell anergy or hyporesponsive.
ness (2-4). Our results suggest that SFDC are unable to
direct antigen-specific SF T lymphocytes to proliferate
and differentiatc into effector cells, thus providing a
potential explanation for the depressed ccllular immunc
fesponse in RA. Further studies are required to determine if DC within the synovial membrane also lack these
costimulator molecules, and whether the duration of
disease and/or different therapeutic regimens have an
effect on SFDC function.
Clearly, there is a need to know more about the
regulation of CD80 and CD86 molecules on DC locatcd
in the chronic arthritic joint. These molcculcs may be
extraordinarily relevant to understanding the pathogenesis, effects of therapy, and development of future
immunomodulating trcatment of chronic inflammatory
W c wish to thank Drs. Christopher Frampton and
Elizabeth Wells for statistical advice, Drs. J. Highton and Peter
Moller for joint fluid samples, and Ms Christinc Martin for
secretarial support.
1. Williams LA, Egner W, Hart DNJ: Isolation and function of
human dcndritic cclls. Int J <)to1 153:41-103, 1994
2. Janeway CA, Bottomly K: Signals and signs for IymphocytE
responses. Cell 76:275-285, 1994
3. Guinan EC, Gribbm JG. Boussiotis VA, Freeman GJ, Nadlei LM:
Pivotal role of the B7:CD28 pathway in transplantation toleiancc
and tumor immunity. Blood 84:3261-3282, 1994
4. Kuchroo VK, Das MP, Brown JA, Ranger, AM, Zamuii SS, Sobtl
RA, Wcincr HI., Nabaui N, Glimber LII: B7-1 and B7-2 costin&
latory molcculcs activate differentially the Thl/Th2 develbpmen!bl
pathways: dpplication to autoimmune disease therapy. Cell 80:
707-718, 1995’
5. Summers KL, Danicl PB, O’Dqnncll JL, Hart DNJ: Dendritic J l l s
in synovial fluid 6f chronic inflammatory arthritis lack Ch30
surface expression. Clin Exp Immunol 100:81-89, 1995
6. McLcllan AD, Starling GC, Williams LA, IIock BD, Hart DNJ:
Activation of human pcriphcral blood dcndritic cel!s induces the
CD86 costimulatory molecule. Eur J Immunol25:2064-2068, 19b5
7. Arnctt FC, Edwurthy SM, Bloch DA, McShahe DJ, Fries JF,
Cooper NS, Healey LA, Kaplan SK, Liang MH, Luthra I,&,
Medsger TA Jr, Mitchell DM, Neustadt D k , Pinals RS, Schalkr
JG, Sharp JT, Wilder RL, Hunder GG: The American Rheum&
tism Association 1987 revised criteria for thc classificatiod of
rheumatoid arthritis. Arthritis Rhcum 31:315-324, 1988
8. Hart DNJ. Starling GC, Calder VL, Fernando NS: B7/BB-l *,a
lcucocyte differentiation antigcn on human dcndritic cclls induced
by activation. Immunology 79:616-620, 1993
9. Munro JM, Freedman AS, Aster JC, Gribbcn JG, Lcc N c ,
Khynhart KK, Banchereau J, Nadlcr LM: I n vivo expression of tbe
B7 costimulatory molecule by subsets of antigen-presenting cklk
and thc malignant cells of Hodgkin’s disease. Blood 83:793-7c)R.
10. Thomas R, Davies LS, Lipsky PE: Rhcumatciid synovium IS
enrichcd in mature antigen-presenting dendritic cells. J Imml~nbl
152:2613-2623, 1994
11. Liu MF, Kohsaka H. Sakurai H, Azuma M, Qkumura K, Saitb J,
Miyasaka N: Tlic prcscnce of costimulatory molecules CD86 and
CD28 in rheumatoid arthritis synovium. Arthritis Rheum 3Y:l Id114, 1996
12. I.udewig B, Graf D. Geldcrblom HR, Heck Y,Kroczek RA,Pi$
G: Spontaneous apoptosis of dendritic cclls is cficicntly inhihjlbd
by TRAP (CD40-ligand) and TNF-a, but strongly enhanced by
intcrlcukin-10. Eur J Immunol 25:1943-1950, 1995
13. Wahl SM, Allen JB. Welch GR, Wong I I L Transforming groFth
factor+ in synovial fluids modulates FcyRIIl (CD16) cxprcssiph
on mononuclear phagocytcs. J Immunol 148:485-490, 1992
14. Willcms F, Marchant A, I>elvilleJP, Gerard C, Delvaux A, Veld
dc Boer M, Goldman M: Intcrlcukin-10 inhibits B7 and intcroO , ular adhesion molecule-1 cxprcssion on human monocytcs. Eur
J Irnmunol 24:1007-1009, 1994
15. Cush JJ, Splawski JB, Thomas R. McFarlin JE, Schulze-Kodpb,
Davis IS,Fujita K, Lipsky PE: Elevated interleukin-10 lcvels ,in
patients with rheumatoid arthritis. Arthritis Khcum 38:96-104,
16. Katsikis PD. Chu C, Brennan FM, Maini RN, Feldmann M
lmmunoregulatory role of interleukin 10 in rhcumatoid arthritis. J
Exp Med 179:1517-1527. 1994
Без категории
Размер файла
529 Кб
expressions, cd80, dendriticum, molecules, arthritis, synovial, costimulator, function, cd86, chronic, cells
Пожаловаться на содержимое документа