close

Вход

Забыли?

вход по аккаунту

?

The intrinsic migratory capacity of memory T cells contributes to their accumulation in rheumatoid synovium.

код для вставкиСкачать
1434
THE INTRINSIC MIGRATORY CAPACITY OF
MEMORY T CELLS CONTRIBUTES TO THEIR
ACCUMULATION IN RHEUMATOID SYNOVIUM
JOHN J. CUSH, PETER PIETSCHMANN, NANCY OPPENHEIMER-MARKS, and PETER E. LIPSKY
Objective. Mechanisms controlling the infiltration
of T cells into rheumatoid synovium have not been fully
characterized. These studies were undertaken to investigate the relationship between T cell phenotype and
migratory capacity, so as to elucidate mechanisms that
might contribute to the accumulation of T cells at
inflammatory sites.
Methods. The characteristics of in vivo migrating
cells were studied by dual-immunofluorescence FACS
(fluorescence-activated cell sorter) analysis of rheumatoid synovial and peripheral blood T cells. Migratory
cells were also characterized using a recently developed
in vitro assay, wherein peripheral blood T lymphocytes
(PBTL) with the capacity to migrate through endothelial
cell monolayers were retrieved and assessed.
Results. Migratory CD4+ T cells from rheumatoid arthritis (RA) and normal individuals were characterized as being CD45RA-, CD29bright,CDllab"@",
L-selectin-, CD54+, and CD58+. Migrating RA
PBTL (compared with normal PBTL), however, were
significantly enriched in activated HLA-DR+ T cells.
RA synovial tissue lymphocytes exhibited a similar
phenotype, but with decreased surface density of CD4
and an increase in HLA-DR and VLA-1. RA synovial
From the Harold C. Simmons Arthritis Research Center,
University of Texas Southwestern Medical Center, Dallas, Texas.
Supported by NIH grants AR-09989, SCOR-AR-39169,and
a grant from the North Texas Chapter of the Arthritis Foundation.
John J. Cush, MD; Peter Pietschmann, MD; Nancy Oppenheimer-Marks, PhD; Peter E. Lipsky, MD.
Address reprint requests to Peter E. Lipsky, MD, Department of Internal Medicine, University of Texas Southwestern
Medical Center, 5323 Harry Hines Boulevard, Dallas, TX 75235.
Submitted for publication June 16, 1992; accepted in revised form August 20, 1992.
Arthritis and Rheumatism, Vol. 35, No. 12 (December 1992)
lymphocytes exhibited a 2-3-fold increase in migratory
capacity over normal and RA PBTL.
Conclusion. These studies demonstrate the inherent migratory proficiency of CD4+ T cells that express
a memory phenotype (CD29brighf,CDllabr'ght, and
CD58+). In addition, enhanced transendothelial migration was observed for CD4+ T cells that were CD54+
and L-selectin-. These studies demonstrate that the
migratory patterns of circulating lymphocytes may be
correlated with their surface phenotype and that the
intrinsic migratory capacity of memory T cells is one
component contributing to their accumulation in the
rheumatoid synovium.
Chronic inflammatory diseases, such as rheumatoid arthritis (RA), are histologically characterized
by the infiltration of mononuclear cells into sites of
tissue inflammation. RA is distinguished by the intensity of the cellular infiltration and the accumulation of
T lymphocytes in the sublining layer of the synovium,
often in a perivascular distribution (1,2). Despite an
increasing understanding of the immunopathogenesis
of RA, the mechanisms governing the migration and
accumulation of cells in these inflammatory lesions
remain obscure. A number of physiologic events could
contribute to the development and nature of the cellular infiltrate within the synovial tissue and fluid. These
include the intrinsic migratory capacity of various T
cell subsets, the activation status of circulating T cells,
as well as local activation events and/or retention of
specific subsets at inflammatory sites. The potential
role of each of these events in the development and
perpetuation of T cell infiltration in the rheumatoid
synovium has not been clearly delineated.
Lymphocytes eluted from the synovial mem-
T CELL MIGRATION INTO RA SYNOVIUM
1435
Table 1. Clinical characteristics of the 15 rheumatoid arthritis patients studied*
Patientlagelsex
RN44lF
DBISOIM
JSJl56lF
JCl73lF
GDl29lF
JMPI57IF
MFHl67lF
PGl38lF
NGl57lF
JPl531M
GJl76lF
TCl46lF
VGl56lF
GBl38lF
VAl6 1IM
Source
Tenosyn
Knee
MCP
MCP
MCP
Wrist
wrist
MCP
Tenosyn
SF
SF
PBTL
PBTL
PBTL
PBTL
Disease
duration
(years)
6
5
13
22
11
25
16
3
5
2
2
1
9
I
9
Ex traarticular
manifestations
+
0
0
0
0
0
+
0
0
0
0
0
+
+
0
Medication
NSAID, MTX
NSAID, PCM
NSAID, Gold
NSAID, AZA
NSAID, PCM
NSAID, PCM
NSAID, PCM
NSAID, Gold
NSAID
NSAID, Pred
Pred, MTX
NSAID, Pred
NSAID, MTX
NSAID
NSAID, Pred, Gold
* The ages shown represent the patient’s age at the time of tissue sampling. Peripheral blood T
lymphocytes (PBTL) were assessed in the last 4 patients because synovial fluid (SF) and tissues were
not available. M = male; F = female; Tenosyn = tenosynovium; NSAID = nonsteroidal antiinflammatory drug; MTX = methotrexate; PCM = penicillamine; MCP = metacarpophalangeal; AZA =
azathioprine; Pred = prednisone.
branes of RA patients have been shown to differ from
those found in the circulation. Thus, synovial tissue T
cells are enriched in CD4+ T cells bearing the “memory” phenotype (CD45RO+, CD29b‘igh‘) (3-5). In addition, synovial tissue T cells express a variety of
activation markers and an increased density of adhesion-related molecules (5-7). These findings are not
unique to rheumatoid synovial tissue cells. Other
examples of chronic inflammation involving the intestine, thyroid, skin, lung, or liver (8-11) have demonstrated a similar enrichment in activated, CD4+
“memory” T cells in perivascular tissues. These findings could have a number of explanations. First, it is
possible that resting memory T cells exhibit an inherent migratory capacity (8). Alternatively, a subset of
activated memory cells may possess a unique capacity
to enter inflammatory sites, as has recently been
suggested (12). Finally, activation of cells within the
inflammatory site could lead to uniform differentiation
to a memory phenotype. The experiments described
herein were undertaken to examine these possibilities.
In these studies, we utilized a system that
permits the isolation and identification of cells of
different transendothelial migratory capacities. Utilizing this approach, T cells exhibiting transendothelial
migration were analyzed to determine their phenotype
and to compare it with the phenotype of cells found in
the rheumatoid synovium. These studies demonstrate
that memory CD4+ T cells exhibit a greater migratory
capacity than do naive T cells. The expanded pool of
activated T cells in the circulation of RA patients
manifests an even greater capacity for transendothelial
migration. Finally, T cells recovered from the synovium of RA patients are enriched in CD4+ memory T
cells that exhibit an enhanced intrinsic capacity for
transendothelial migration. These observations are
consistent with the conclusion that intrinsic migratory
capacity plays an important role in the accumulation of
T cells in the rheumatoid synovium.
PATIENTS AND METHODS
Patients. Fifteen patients who fulfilled the American
College of Rheumatology (formerly, the American Rheumatism Association) 1987 revised criteria (13) for the diagnosis
of RA were studied. Table 1 details the characteristics of
these 15 patients. There were 12 women and 3 men, with a
mean age of 53 years and a mean disease duration of 9.1
years. All patients had active, rheumatoid factor-positive,
inflammatory polyarthritis at the time of evaluation. Thirteen
demonstrated significant morning stiffness (21 hour), and 4
patients had evidence of extraarticular disease (nodules or
vasculitis). All patients were receiving nonsteroidal antiinflammatory agents and/or low-dose prednisone (I 10 mgl
day), and 11 patients were receiving second-line antirheumatic therapy at the time of tissue specimen collection. In 9
patients, synovium or tenosynovium was retrieved during
tenosynovectomy or joint reconstruction surgery; synovial
fluid was obtained from 2 additional patients. Peripheral
blood samples were obtained from all individuals, and simultaneous, paired samples of peripheral blood were obtained at
1436
the time of surgical procedure or synovial fluid aspiration
from 11 of the RA patients.
Cell preparation. Peripheral blood mononuclear cells
(PBMC) were obtained from 14 normal, healthy volunteer donors between 20 and 49 years of age. PBMC were
prepared from heparinized venous blood samples by densitygradient centrifugation with sodium diatrizoate/Ficoll
gradients and were washed 3 times with Hanks’ balanced
salt solution (HBSS). PBMC were enriched for peripheral
blood T lymphocytes (PBTL) by passage over nylon-wool
columns as previously described (14). Human umbilical vein
endothelial cells (HUVEC) were obtained as previously
described (15).
HUVEC were cultured at 37°C in medium containing
RPMI 1640 supplemented with 15% fetal bovine serum
(FBS), 10% human serum (HS), 50 units/ml of heparin, 20
mM L-glutamine, 100 unitdm1 penicillin, 50 pgml gentamicin, 100 units/ml nystatin, and 24 &ml endothelial cell
growth supplement (ECGS; Collaborative Research, Bedford, MA) and were used at the third-fifth passage. HUVEC
were removed from the culture surface by trypsinization in
the presence of Puck’s EDTA and suspended in experimental culture medium consisting of RPMI with 10% HS, 24
pg/ml ascorbic acid, and 50 pgml ECGS.
Synovial tissue digestion and preparation of synovial
tissue lymphocytes (STL) and synovial fluid lymphocytes
(SFL). Sterile synovial tissue specimens (2-10 gm) were
obtained and enzymatically digested using a modification of
methods previously described (7). Briefly, after mincing,
synovial tissue was suspended in collagenase solution containing 100 ml of HBSS with 1% FBS, 5% HEPES buffer,
0.1% collagenase, and 0.001% deoxyribonuclease. Synovial
tissue was digested for 90 minutes at 37°C with constant
stirring. After washing twice with HBSS containing 10%
FBS, the synovial tissue cell suspension was rapidly passed
over a nylon-wool column to deplete adherent cells and
enrich for T cells. The number of STL recovered from the
digestion and purification of 9 patient samples ranged from
10 to 80 million cells.
Synovial fluid (20-60 ml) was aseptically withdrawn
(into heparinized tubes) from the anesthetized knees of 2
patients who had active synovitis with effusion. SFL were
prepared in the same manner as PBTL described above.
Transendotheli migration studies. The ability of
lymphocytes to bind to and migrate through endothelium
was assayed using a recently developed in vitro method,
wherein confluent HUVEC monolayers were established
over hydrated bovine collagen gels, as described elsewhere
(16). Briefly, collagen gels were established in 16-mm polystyrene macrowells (Costar, Cambridge, MA) using 0.75
d w e l l (4-mm thick) of a 60% bovine collagen solution in
phosphate buffered saline (PBS) (Vitrogen 100; Collagen
Corp., Palo Alto, CA). Confluent monolayers of endothelium were established by adding lo6 HUVEC/well and
incubating overnight at 37°C. After gently washing with
warm (37°C) culture medium, 3 x lo6 PBTL/ml in RPMI
with 0.6% bovine serum albumin (BSA) were added to each
macrowell and allowed to interact with the HUVEC for 4
hours at 37°C. The nonadherent, nonmigrating PBTL were
retrieved by gentle washing with warm RPMI and 0.6%
BSA; these were designated nonadherent (NAD) cells.
CUSH ET AL
PBTL that remained tightly bound to endothelium were
removed from the HUVEC monolayer by vigorous washing
twice each with warm Puck’s EDTA and EGTA (0.5 mM)
and once with cold (5°C) Puck’s EGTA; these were designated bound cells. Finally, the migrated PBTL were recovered by mincing the gel and digesting it with the collagenase
solution (described above) for 45 minutes at 37°C with
constant stimng.
After retrieval, all 3 subsets were greater than 96%
viable, as determined by trypan blue exclusion. The NAD,
bound, and migrated populations were washed twice in
HBSS, quantified by light microscopy using a hemacytometer, and were analyzed for surface phenotype using a fluorescence-activated cell sorter (FACS). When samples from
individual donors were assayed on multiple occasions, a
strikingly similar distribution of T cells into NAD, bound,
and migrated populations was observed.
Monoclonal qntibodies (MAb). To assess the surface
phenotype of retrieved cells, MAb directly labeled with
fluorescein isothiocyanate (FITC) or phycoerythrin (PE)
were utilized at saturating concentrations. These MAb (recognizing distinct surface determinants) included: msIgG1RDUmsIgG2-FITC (control), T4-FITC/2H4-RDl (CD4/
CD45RA), T4-FITC/4B4-RDl (CD4/CD29), Mo~-FITC/
KC56-PE (monocyte/lymphocyte gating control), and T3FITC (CD3) were obtained from Coulter Immunology
(Hialeah, FL); Leu-&FITC/HLA-DR-PE (CD3/HLA-DR),
Leu-ZPEILeu-7-FITC (CD8/CD57), Leu-3-FITCILeu8-PE (CDWL-selectin), and interleukin-2 receptor (IL2R)-PE (CD25) were obtained from Becton Dickinson
(Thousand Oaks, CA); TS1/22 (CDlla) and TS2/9 (CD58)
were obtained from the American Type Culture Collection
(Rockville, MD) and were labeled with FITC as described
(17); VLA-1-FITC (CD49dCD29) was obtained from T Cell
Sciences (Cambridge, MA); and RR1/1 (CD54) was a gift
from Dr. R. Rothlein, Boehringer Ingelheim (Ridgefield, CT)
and was labeled with FITC (17).
Dual-immuno6uorescence studies. Lymphocytes
were stained using directly labeled MAb as previously
described (7). Briefly, lo5 to lo6 lymphocytes were washed
twice with 2% HS in PBS and 0.1% sodium azide, and
stained with saturating concentrations of the appropriate
MAb at 4°C for 30 minutes. Cells were then washed twice
and analyzed immediately for immunofluorescence using a
FACScan analyzer (Becton Dickinson). Lymphocytes were
selectively analyzed by gating according to forward and
orthogonal scatter. Accuracy of gating was affirmed by
staining with Mo2/KC56. Relative antigen density was ascertained by recording the mean fluorescence intensity
(MFI), which was reported in arbitrary units, for the sample
analyzed. Like samples were assessed after appropriate
compensation and with identical photomultiplier tube (PMT)
settings on the FACScan. In 4 separate experiments, the
effect of collagenase pretreatment on the staining profiles of
normal blood lymphocytes was tested and demonstrated not
to alter the percentage or fluorescence intensity of cells
stained by the above MAb (data not shown).
Statistical analysis. Unless indicated otherwise, all
data are reported as the mean ? SEM. Paired samples were
analyzed using Student’s t-test for paired samples, with
Bonferroni’s correction for multiple comparisons when ap-
T CELL MIGRATION INTO RA SYNOVIUM
1437
Table 2. FACS analysis of rheumatoid synovial tissue lymphocytes and PBTL from RA patients and
normal subiects*
Rheumatoid arthritis
Surface antigen
Normal
PBTL
(n = 14)
n
PBTL
Synovial
lymphocytes
CD3
CD4
CD4 MFI
CD29t
CD29 MFI
CD45RAt
CD45RA MFI
L-selectint
CD8
CD57t
CD3+, HLA-DR+
CD3+, CD25+
CDlla
CDlla MFI
CD49dCD29
75.9 f 2.0
54.4 f 1.5
279 f 32
70.7 f 4.7
171 f 24
48.1 f 5.5
531 f 59
51.7 f 11.9
31.7 f 10.1
23.5 f 2.8
2.4 f 1.0
1.2 f 0.8
92.4 f 2.9
135 f 11
15.4 f 5.9
8
9
9
9
9
9
9
4
9
9
8
6
6
6
7
77.9 f 3.6
55.2 f 4.9
294 f 70
75.6 f 7.4
197 f 17
46.7 f 3.9
451 f 92
63.0 f 13.0
30.6 f 4.0
34.3 f 2.5$
6.2 f 1.3$
4.9 f 2.2t
96.9 f 0.1
151 f 18
28.5 f 7.2$
60.1 f 2.5$
43.7 f 1.9$
149 f 37$
92.6 f 3.7$
282 f 42$
8.1 f 1.8$
216 f 37$
9.5 f 0.7$
24.1 f 4.5
20.7 f 1.9
32.6 f 7.3$
0.7 f 0.4
85.4 f 2.2
235 f 24$
50.1 f 6.4$
* Values are the mean f SEM % positive or mean fluorescence intensity (MFI; in relative units).
FACS = fluorescence-activated cell sorter; PBTL = peripheral blood T lymphocytes; RA =
rheumatoid arthritis.
t Values are expressed as a percentage of CD4+ or CD8+ cells, as indicated.
$ P 5 0.05 versus the other two groups.
propriate (18). Unpaired data were analyzed using population means and the 2-sample t-test.
RA PBTL
RA Synovial
Lymphocytes
RESULTS
Phenotypic characterization of rheumatoid lymphocytes. FACS analysis revealed that the majority of
the cells retrieved from the rheumatoid synovium were
CD3-t T cells (Table 2). CD4+ T cells predominated
over CD8+ cells in the synovium and PBTL. The
CD4+ synovial T cells were enriched in CD29+
memory T cells and were virtually devoid of
CD45RA+ naive T cells (Figure 1). This contrasted
with the near-equal expression of both CD29-t and
CD45RA+ in normal and patient PBTL (Table 2 and
Figure 1). The CD4+ T cells retrieved from the
synovium were also significantly enriched in
CD29brigh'cells, compared with the levels in autologous and control PBTL (Figure 1). Thus, the mean
(+SEM) CD29 MFI for normal PBTL, RA PBTL, and
RA synovial lymphocytes were 171 + 24, 197 + 17,
and 282 2 42, respectively (P5 0.05). In 4 patients,
L-selectin expression on RA PBTL and STL was
assessed (Table 2 and Figure 2). A dramatic reduction
in the expression of this molecule by STL was observed (P< 0.001).
The percentages of CD8+ T cells detected in
normal PBTL and patient PBTL and STL were nearly
CD4
Figure 1. Histograms of dual-immunofluorescence studies of cells
from a representative patient with rheumatoid arthritis (RA), comparing the expression of CD29 and CD45RA by RA peripheral blood
T lymphocytes (PBTL) and RA synovial tissue lymphocytes. Numbers in the upper right are the percentages of CD4+ T cells that
express either CD29 (top panel) or CD45RA (bottom panel).
CUSH ET AL
1438
-
Nonadherent
-
Migrated
I
I
25.4
E
.
I
w
t
A
0
Q)
-
RA
RA Synovial
Lymphocytes
PBTL
83.8
I
@
I
22.7
I
CD4
Figure 2. Histograms of dual-immunofluorescence studies of Lselectin expression by CD4+ T cells from patients with RA.
Numbers in the upper right are the percentages of CD4+ T cells that
express L-selectin, among a population of nonadherent and one of
migrated CD4+ T cells retrieved from the collagen gel assay (top
panel), as well as synovial T cells and autologous PBTL (bottom
panel). See Figure 1 for definitions of abbreviations.
equal. Although, the number of CD8+ T cells expressing CD57 was not significantly different between normal PBTL and RA synovial tissue samples, RA PBTL
demonstrated a significant increase in CD8+, CD57+
cells compared with the levels in RA STL (Table 2).
Normal PBTL
When compared with controls, circulating RA
PBTL exhibited a small, but significant (P 5 0.05),
increase in HLA-DR, CD25, and CD49a/CD29
(VLA-1) expression. Moreover, lymphocytes eluted
from the rheumatoid synovium demonstrated a greater
degree of activation, as indicated by the far greater
percentage of synovial T cells expressing HLA-DR (P
< 0.01) (Table 2) and a decrease in the antigen density
of CD4 (P 5 0.05) when compared with normal and
RA PBTL (7). Despite the evidence of T cell activation, synovial tissue lymphocytes did not demonstrate
an increase in CD25 expression.
Although nearly 90% of the lymphocytes expressed leukocyte function-associated antigen-1 , or
LFA-I (CD1la/CD18), RA synovid tissue lymphocytes exhibited a significantly increased density of
CDlla (P = 0.05). Thus, nearly all T cells in the RA
synovium were CD1 labrightwhen compared with autologous and control PBTL (Figure 3). By comparison
with normal and RA PBTL, synovial T cells also
exhibited a significant increase in the number of cells
expressing very late-activation antigen-1 , or VLA- 1
(CD49dCD29).
Transendothelial migration by normal PBTL.
The ability of normal PBTL to interact with endothelium and undergo transendothelial migration was assessed next. After a 4-hour incubation with HUVEC
monolayers, 61.4 4.1% (range 4140%) of the recovered cells (mean k SEM) were nonadherent, 25.8
3.7% (range 1147%) were bound to EC, and 12.5 f
1.7% (range 420%) had migrated through EC monolayers and were recovered from the collagen gels (n =
8). Table 3 lists the mean results of 10 experiments
using normal donor PBTL. The percentage of CD3 + T
cells did not differ significantly among the 4 popula-
*
RA Synovial
Lymphocytes
RA PBTL
t
L
al
n
5
2
d
Fluorescence intensity
-
Figure 3. The expression of CDlla on normal PBTL and on RA PBTL and RA synovial tissue
lymphocytes from a representative patient. See Figure 1 for definitions of abbreviations.
*
T CELL MIGRATION INTO RA SYNOVIUM
1439
Table 3. FACS analysis of normal PBTL of differing migratory capacities*
Suface antigen
n
Initial
Nonadherent
Bound
Migrated
CD3
CD4
CD4 MFI
CD29t
CD29 MFI
CD45RAt
CD45RA MFI
L-selectinf
CD54t
CD58t
CD44
CD44 MFI
CD8
CD57t
CD3+ , HLA-DR+
CD3+ , CD25+
CD1la
CDl la MFI
CD49dCD29
5
9
9
9
9
9
9
8
7
4
3
3
7
7
5
3
8
8
5
73.5 f 6.6
49.5 f 3.4
813 f 50
65.4 f 7.7
248 f 30
53.9 f 4.0
1227 f 234
52.8 f 10.0
7.7 f 3.8
24.5 f 1.7
85.1 f 1.1
609 f 45
48.1 f 7.2
14.3 f 2.0
1.6 f 0.7
1.4 f 0.6
89.4 f 3.1
743 f 54
9.3 f 1.6
79.0 f 2.7
52.4 f 2.8
707 f 52
65.2 f 7.0
235 f 23
58.8 f 3.5
1003 f 100
64.1 f 8.5
6.4 f 3.3
13.9 f 6.6
90.6 f 2.4
492 f 46
37.9 f 3.0
9.9 f 2.6
1.5 f 0.4
1.7 f 0.7
87.5 f 2.6
601 f 69
14.3 f 1.6
75.4 f 3.2
47.2 f 2.3
762 f 36
83.7 f 3.57
324 f 23#
33.7 f 3.1#
845 f 75
54.5 f 8.9
16.4 f 6.2
50.0 f 12.67
94.7 f 0.7
837 f 120
47.3 f 2.2
19.1 f 1.1
4.3 f 1.1
1.3 f 0.4
93.9 f 0.7
883 2 727
24.3 2 8.6
68.6 f 2.9
37.6 f 2.7$
651 f 358
96.2 f 4.07
559 f 44$
18.1 f 2.6$
582 f 38$
39.5 f 7.87
24.4 f 6.9$
51.5 i 8 . u
95.9 f 0.6
993 f 129
49.3 f 5.6
25.2 f 5.6
7.0 f 1.7
1.2 f 0.4
94.4 f 0.6
941 f 697
7.1 f 2.9
* Values are the mean f SEM % positive or mean fluorescence intensity (MFI; in relative units).
FACS = fluorescence-activated cell sorter; PBTL = peripheral blood T lymphocytes.
t Values are expressed as a percentage of CD4+ or CD8+ cells, as indicated.
$ P 5 0.0083 versus the nonadherent and the bound groups.
8 P 5 0.0083 versus the bound group.
7 P 5 0.0083 versus the nonadherent group.
# P 5 0.0083 versus the nonadherent and the migrated groups.
tions, as defined by migratory capacity. However, the
number of CD4+ T cells was significantly decreased in
the migrated population (P< 0.008) compared with the
NAD and bound groups. Moreover, the CD4+ T cells
in the migrated population were distinctly different
from those in the other populations by virtue of the
significantly enhanced number of CD29b"ghtcells (indicated by the augmented CD29 MFI), as well as the
increased number of cells expressing CD54 and CD58
(P < 0.008).
Figure 4 shows that nearly all of the migrated
CD4+ T cells were CD29b"ph'when compared with the
bound, NAD, and initial subsets (P < 0.0001). In
addition, migrated CD4+ T cells exhibited a significant
reduction in the expression of the CD45RA isoform (P
< 0.001) when compared with the initial, NAD, and
bound cells (Table 3 and Figure 4). Migrated CD4+ T
cells also contained significantly fewer (P = 0.002)
L-selectin+ cells when compared with the nonadherent population (Table 3 and Figure 2). Thus, the
migrated CD4+ T cell population was markedly enriched in CD29b"gh' "memory" cells, but contained
few CD45RA "naive" T cells. It should be noted
however, that only a subset of memory cells demonstrated a migratory capacity in this assay, with many
other memory cells remaining in the nonadherent and
bound populations.
Although CD8+ T cells were found in the
migrated population, no significant enrichment of this
subset was observed (Table 3). A modest increase in
the numbers of CD8+ ,CD57+ and CD3+ ,DR+ T cells
(P = 0.03 and P = 0.02, respectively) was found in the
bound and migrated populations. However, neither
achieved the statistical significance required using the
Bonferroni method (18). The majority of T cells in
each population expressed CD44. Although the intensity of CD44 expression was increased in the bound
and migrated populations, this did not achieve statistical significance because of the small sample size. No
enrichment of cells expressing CD57, HLA-DR,
CD25, or VLA-1 was noted in either the bound or the
migrated populations (Table 3).
Whereas the percentage of CD1la+ T cells was
not significantly different between the populations of
cells with differing migratory capacities, the density of
CDlla was significantly increased in the bound and
migrated populations (Table 3) when compared with
that in the initial and NAD populations (P = 0.004).
The majority of migrating T cells were CDllabngh'
(Figure 5).
CUSH ET AL
1440
CD4
Figure 4. Histograms of dual-immunofluorescence studies of CD29 and CD45RA expression by
normal T cell populations of differing migratory capacities, after incubation with endothelial cells.
Numbers in the upper right are the percentages of CD4+ T cells that express either CD29 (top panel)
or CD45RA (bottom panel).
Migration of rheumatoid lymphocytes. The ability of normal or RA PBTL and RA synovial lymphocytes (STL from 3 patients and SFL from 2 patients) to
migrate through EC monolayers into the collagen gel
was also assayed. As shown in Figure 6, rheumatoid
synovial T cells demonstrated a significantly greater
migratory capacity compared with the control and
autologous circulating PBTL (P < 0.01). No obvious
differences between the migration of SFL and STL were
apparent in the limited number of samples examined.
To determine whether the presence of activated
T cells in PBTL might influence the migratory capacity
Non a d 17 e r en t
of this population, PBTL from normal subjects and 4
RA patients were analyzed. The phenotype of RA
PBTL recovered in the nonadherent, bound, and migrated populations was similar to that observed with
normal PBTL with regard to the expression of CD3,
CD4, CD29, and CD45RA (Table 4). Table 4, however, demonstrates that migrated RA PBTL were
significantly enriched in CD8+ and HLA-DR+ T cells.
By contrast, the percentages of HLA-DR-expressing
T cells in the nonadherent, bound, and migrated
populations in 10 experiments carried out with normal
PBTL were 4.9 ? 1.4, 10.3 ? 2.5, and 8.6 f 1.9,
Bound
Migrated
4
Fluorescence Intensity -------- >
Figure 5. CDlla expression by normal T cell populations of differing migratory capacities, after
incubation with endothelial cells.
T CELL MIGRATION INTO RA SYNOVIUM
T
U
t
W
20
U
L
10
NONADHERENT
BOUND
MIGRATED
Figure 6. Migration of normal PBTL and RA PBTL and RA synovial lymphocytes. Values are the mean and SEM of cells from 8
normal donors and 5 RA patients. See Figure 1 for definitions of
abbreviations.
respectively (mean & SEM). Therefore, the percentage of HLA-DR-expressing T cells in the migrated
populations of normal and RA PBTL was significantly
different (P = 0.04). These findings suggest that HLADR-expressing, activated T cells in RA patients may
have an enhanced capacity to enter the synovium.
DISCUSSION
These studies were undertaken to elucidate
whether an association exists between the phenotype
and the migratory behavior of T lymphocytes, and if
so, whether such an association may influence the
phenotype of cells accumulating at sites of tissue
inflammation. We analyzed T cells recovered from
inflammatory synovial tissues and compared their phenotype and migratory behavior with those of circulat-
1441
ing PBTL from the same RA patients and from normal
control subjects. RA synovial lymphocytes exhibited a
different phenotype than circulating PBTL, with the
former demonstrating a significant increase in CD4+
memory T cells that were CD29brigh', CD45RA-,
CD1 labfigh', and L-selectin-. Similar findings have
been reported by other researchers analyzing synovial
fluid and tissue lymphocytes and frozen synovial tissue specimens (3-8). Moreover, increased numbers of
cells bearing a memory phenotype have also been
detected in samples taken from inflammatory sites in
other disorders, including bronchoalveolar lavage
fluid, delayed-type hypersensitivity (DTH) lesions,
and cerebrospinal fluid (8-1 1,19,20). Although several
studies have demonstrated the accumulation of memory T cells at sites of tissue inflammation, it is unclear
whether memory T cells preferentially migrate, are
uniquely sequestered after other subsets have exited
the synovium, or have differentiated locally.
Using an vitro model of transendothelial migration, we were able to analyze, compare, and contrast 3
subsets of normal donor PBTL with distinctly different
migratory capacities: the nonadherent, bound, and
migrated PBTL. After normal PBTL were separated
according to their migratory behavior, no significant
alterations were noted in the number of T cells in the
various populations expressing CD3, CD8, CD57,
CD44, CDlla, CD25, HLA-DR, and VLA-1. Although CD4+ cells tended to be less migratory than
CD8+ cells in this 4-hour assay, both populations
were able to migrate through the endothelial cell
monolayers. Several studies have noted the predominance of either CD4 (in DTH lesions, RA synovium,
and Reiter's syndrome synovial fluid) or CD8 (in RA
synovial fluids and rejected organ graft tissues) T cells
at inflammatory sites. These findings suggest that
following the initial entry of cells into the inflammatory
Table 4. FACS analysis of RA PBTL of differing migratory capacities*
Surface antigen
n
Nonadherent
Bound
Migrated
CD3
CD4
CD29t
CD45RAt
CD8
CD3 + , HLA-DR+
4
4
4
5
3
4
84.0 f 4.3
68.6 f 3.0
65.5 f 11.9
69.4 f 8.6
23.3 2 6.1
6.5 f 3.2
76.9 f 2.4
64.6 f 6.5
76.9 2 9.7
49.8 2 16.4
32.4 2 7.4
13.0 f 5.3
67.0 f 5.8
56.8 f 6.8
95.7 2 3.0$
14.7 f 1.28
44.1 2 1.8$
20.2 2 7.0$
* Values are the mean 2 SEM % positive. FACS = fluorescence-activated cell sorter; RA =
rheumatoid arthritis; PBTL = peripheral blood T lymphocytes.
t Values are expressed as a percentage of CD4+ T cells.
$ P I0.017 versus the nonadherent group.
6 P I0.017 versus the nonadherent and the bound groups.
1442
tissue, local events may influence the final mix of cells
that accumulate, but also emphasize that the intrinsic
migratory capacity of cells is one important determinant of this process.
One of the striking findings from these studies
was the discordant representation of the T cell differentiation antigens identifying “naive” and “memory”
T cells among the migrated population. Naive CD4+ T
cells, expressing CD45RA, exhibited a significant deficit in their ability to bind EC and were poorly capable
of transendothelial migration. By contrast, memory
CD4+ T cells that were CD29b“gh‘were significantly
increased among bound and migrated T cells. The
enhanced transendothelial migration by memory T
cells has been noted in other in vitro and in vivo
studies (12,19). Related studies have indicated that
memory T cells (CD29+ or CD45RO+) are significantly more adherent to resting or activated (i.e.,
IL-l-stimulated) endothelium than are naive T cells
(CD45RA+ or CD29-) (15,21). Moreover, Matsuoka
et a1 have noted CD45RA+ T cells to be poorly
adherent, while T cells expressing CD29, LFA-1, and
CD2 were more adherent to IL-l-stimulated synovial
fibroblasts (22). The current study extends these findings by demonstrating that migrated CD4+ T cells
were predominantly CD29bngh‘memory cells and virtually devoid of CD45RA+ naive cells. Moreover, a
distinct subpopulation binds to EC but fails to migrate
in this assay, suggesting that different subsets of
memory cells may manifest adhesive or migratory
properties. The memory CD4+ T cell population
among migrated cells was further characterized by the
increased number of cells expressing other surface
markers that characterize memory T cells, including
LFA-3, an increased density of CDl l a (Table 3), and
CD45RO (data not shown) (see ref 16).
CD4+ T cells that expressed L-selectin were
more likely to remain in the nonadherent population,
and were significantly reduced in the migrated PBTL
when compared with the bound and nonadherent
subsets. Therefore, the lack of L-selectin expression
on T cells was associated with enhanced migratory
behavior. To an even greater extent, RA synovial
lymphocytes demonstrated a significant loss of Lselectin expression when compared with autologous
PBTL. This finding is consistent with those in other
reports showing that L-selectin expression is reduced
in rheumatoid synovium and is unlikely to play a
significant role in cellular entry into synovium and
gut-associated lymphoid tissue (23). L-selectin is
known to be the human equivalent of the murine
CUSH ET AL
homing receptor (Mel-14) and has primarily been
shown to be important in the homing of PBTL to
lymph nodes and their binding to high endothelial
venules (HEV) (23-25). A crucial role for L-selectin
expression in the trafficking of lymphocytes to lymph
node HEV and a less critical role for L-selectin
expression in the migration of lymphocytes to nonlymphoid tissues, such as the rheumatoid synovium, has
been suggested (23-25). The current data demonstrate
that cells exhibiting a migratory capacity express
considerably less L-selectin than do those that fail to
migrate, In addition, lymphocytes that have accumulated in the rheumatoid synovium minimally express
L-selectin. These findings are most consistent with the
conclusion that the cells entering the synovium lack
L-selectin expression and that down-modulation of
this molecule is not merely a consequence of local
activation.
The results from our in vitro assays indicate
that the expression of CD54 (intercellular adhesion
molecule type 1, or ICAM-I), CD58 (LFA-3), and
CD44 by CD4+ T cells is enhanced in cells that exhibit
migratory behavior. Although these surface determinants were not examined in the rheumatoid synovial
samples in this study, previous results indicate that
ICAM-1 and LFA-3 are expressed by a variety of cell
types within the rheumatoid joint, including synoviocytes of the macrophage lineage, fibroblasts, EC, and
infiltrating lymphocytes (26). CD44 is also expressed
by numerous cell types and has also been shown to be
involved in the binding of mononuclear cells to HEV,
the binding of activated T cells to activated HUVEC,
and possibly, in the trafficking of leukocytes to inflamed tissues (27). It has also been suggested that
CD44 is upregulated in inflamed synovium (27). The
current data suggest that CD54, CD58, and to a lesser
extent CD44, expression further defines a memory T
cell population with migratory potential, although the
role of each in promoting cellular migration to the
inflamed synovium will require further investigation.
The conversion from a mixed population of
naive and memory T cells in the circulation to a
predominantly memory T cell population among migrated T cells suggests a relationship between T lymphocyte maturatioddifferentiation and migratory potential. Moreover, these findings indicate that T cells
that have progressed to a particular stage of differentiation, manifested by a memory phenotype, are most
likely to be the cells exhibiting enhanced migratory
behavior. Activation-related events may also influence
the transendothelial migration by circulating T cells.
T CELL MIGRATION INTO RA SYNOVIUM
However, when normal PBTL were examined, bound
and migrated T cells did not exhibit a significant
increase in HLA-DR or CD25 expression, nor was the
density of CD4 down-modulated, suggesting that the
presence of activated T cells could not account for
adhesion to EC or for transendothelial migration.
These findings contrast with the results observed with
RA PBTL, wherein there was an enrichment of activated T cells in the migrated population (Table 4).
Previous reports have documented that activation
increases the capacity of T cells to migrate (12,15). It
should be noted, however, that the vast majority of
bound and migrated T cells from RA PBTL were not
activated. Thus, even in RA patients, most of the
migrating cells are nonactivated memory T cells.
Characteristically, activated T cells are abundant among rheumatoid synovial lymphocytes (3-8).
Not surprisingly, the analysis of synovial tissue lymphocytes from our patients demonstrated a number of
activation-related changes, including a 5-6-fold increase in HLA-DR expression, a significant reduction
in CD4 density, a marked reduction in L-selectin
expression, and a 2-5-fold increase in VLA-1, a
marker of persistent T cell activation (7). Moreover,
RA synovial lymphocytes exhibited a significantly
greater capacity for transendothelial migration compared with normal PBTL (Figure 6). This may reflect
the intrinsic migratory capacity of memory T cells and
the enhancement of migration from local activation,
and supports the conclusion that migratory capacity
resulting from differentiation or activation status contributes to the accumulation of cells in the rheumatoid
synovium.
These results do not permit a determination of
the comparative roles of transendothelial migration
and local activation in the accumulation of T cells in
rheumatoid synovium. Although it is possible that
memory cells enter the synovium and are activated in
situ, it is also possible that memory T cells enter the
synovium, but only the activated subset is retained
because they encounter specific stimuli or have an
enhanced capacity to bind matrix or cellular elements
in the tissue. Regardless of the specific details of
cellular accumulation, it is clear from the current data
that migratory capacity may be a primary determinant
of cellular accumulation at inflammatory sites.
ACKNOWLEDGMENTS
The authors would like to acknowledge the skillful
technical assistance of Amie Schall-Napier, Daphene Gibson, and Ellis Lightfoot.
1443
REFERENCES
1. Cush JJ, Lipsky PE: Cellular basis for rheumatoid
inflammation. Clin Orthop 2659-22, 1991
2. Harris ED Jr: Rheumatoid arthritis: pathophysiology
and implications for therapy. N Engl J Med 322:12771289, 1990
3. Nakao H , Eguchi K , Kawakami A, Migita K , Otsubo T,
Ueki Y, Shimomura C, Tezuka H, Mastsunaga M,
Maeda K, Nagataki S: Phenotypic characterization of
lymphocytes infiltrating synovial tissue from patients
with rheumatoid arthritis: analysis of lymphocytes isolated from minced synovial tissue by dual immunofluorescent staining. J Rheumatol 17:142-148, 1989
4. Hanly JG, Pledger D, Parkhill W, Roberts M, Gross M:
Phenotypic characteristics of dissociated mononuclear
cells from rheumatoid synovial membrane. J Rheumatol
17:1274-1279, 1990
5. Potocnik AJ, Kinne R, Menninger H , Zacher J, Emmrich F, Kroczek RA: Expression of activation antigens
on T cells in rheumatoid arthritis patients. Scand J
Immunol 1:213-224, 1990
6. Poulter LW, Duke 0, Panayi GS, Hobbs S, Raftery MJ,
Janossy G: Activated T lymphocytes of the synovial
membrane in rheumatoid arthritis and other arthropathies. Scand J Immunol 22:683-690, 1985
7. Cush JJ, Lipsky PE: Phenotypic analysis of synovial
tissue and peripheral blood lymphocytes isolated from
patients with rheumatoid arthritis. Arthritis Rheum 3 1:
1230-1238, 1988
8. Ziff M: Role of the endothelium in chronic inflammatory
synovitis. Arthritis Rheum 34: 1345-1352, 1991
9. Bos JD, Hagenaars C, Das PK, Krieg SR, Voorn WJ:
Predominance of “memory” T cells (CD4+, CDw29+)
over “naive” T cells (CD4+, CD45R+) in both normal
and diseased human skin. Arch Dermatol Res 281:2430,
1989
10. Dominique S, Bouchonnet F, Smiejan JM, Hance AJ:
Expression of surface antigens distinguishing “naive”
and previously activated lymphocytes in bronchoalveolar lavage fluid. Thorax 45:391-396, 1990
11. Volpes R, van den Oord JJ, Desmet VJ: Memory T cells
represent the predominant lymphocyte subset in acute
and chronic liver disease. Hepatology 13:82-29,
1990
12. Masuyama JI, Berman JS, Cruikshank WW, Morimoto
C, Center DM: Evidence for recent as well as long term
activation of T cells migrating through endothelial cell
monolayers in vitro. J Immunol 148:1367-1374, 1992
13. Arnett FC, Edworthy SM, Bloch DA, McShane DJ,
Fries JF, Cooper NS, Healey LA, Kaplan SR, Liang
MH, Luthra HS, Medsger TA Jr, Mitchell DM, Neustadt DH, Pinals RS, Schaller JG, Sharp JT, Wilder RL,
Hunder GG: The American Rheumatism Association
1987 revised criteria for the classification of rheumatoid
arthritis. Arthritis Rheum 31:315-324, 1988
CUSH ET AL
1444
14. Wernick RM, Lipsky PE, Marban-Arcos E, Maliakkal
JJ, Edelbaum D, Ziff M: IgG and IgM rheumatoid factor
synthesis in rheumatoid synovial membrane cell cultures. Arthritis Rheum 28:742-752, 1985
15. Oppenheimer-Marks N, Davis LS, Lipsky PE: Human T
lymphocyte adhesion to endothelial cells and transendothelial migration: alteration of receptor use relates
to the activation status of both the T cell and the
endothelial cell. J Immunol 145:140-148, 1990
16. Pietschmann P, Cush JJ, Lipsky PE, OppenheimerMarks N: Identification of subsets of human T cells
capable of enhanced transendothelial migration. J Immunol 149:1170-1 178, 1992
17. Goding JW: Immunofluorescence, Monoclonal Antibodies: h-inciples and Practice. London, Academic Press,
1986
18. Ottenbacher KJ: Statistical conclusion validity: multiple
inferences in rehabilitation research. Am J Phys Med
Rehab 70:317-322, 1991
19. Pitzalis C, Kingsley GH, Covelli M, Meliconi R, Markey
A, Panayi GS: Selective migration of the human-helperinducer memory T cell subset: confirmation by in vivo
cellular kinetic studies. Eur J Immunol21:369-376, 1991
20. Chofflon M, Weiner HL, Morimoto C, Hafler DA:
Decrease of suppressor inducer (CD4+2H4+) T cells in
multiple sclerosis cerebrospinal fluid. Ann Neurol 25:
494499, 1989
21. Damle NK, Doyle LV: Ability of human T lymphocytes
to adhere to vascular endothelial cells and to augment
22.
23.
24.
25.
26.
27.
endothelial permeability to macromolecules is linked to
their state of post-thymic maturation. J Immunol 144:
1233-1240, 1990
Matsuoka N, Euchi K, Kawakami A, Ida H , Nakashima
M, Sakai M, Terada K, Inoue S, Kawabe Y,Kurata A,
Fukuda T, Aoyagi T , Maeda K, Nagataki S: Phenotypic
characteristics of T cells interacted with synovial cells. J
Rheumatol 18:1137-1 142, 1991
Manolios N, Geczy C, Schrieber L: Lymphocyte migration in health and inflammatory rheumatic disease.
Semin Arthritis Rheum 20:339-352, 1991
Camerini D, James SP, Stamenkovic I, Seed B: Leu-8/
TQl is the human equivalent of the Mel-14 lymph node
homing receptor. Nature 342:7&82, 1989
Kishimoto TK, Jutila MA, Butcher EC: Identification of
a human peripheral lymph node homing receptor: a
rapidly down-regulated adhesion molecule. Proc Natl
Acad Sci U S A 87:2244-2248, 1990
Hale LP, Martin ME, McCollum DE, Nunley JA,
Springer TA, Singer KH, Haynes BF: Immunohistologic
analysis of the distribution of cell adhesion molecules
within the inflammatory synovial microenvironment.
Arthritis Rheum 32:22-30, 1989
Haynes BF, Hale LP, Denning SM, L e PT, Singer KH:
The role of leukocyte adhesion molecules in cellular
interactions: implications for the pathogenesis of inflammatory synovitis. Springer Semin Immunopathol 11:
163-185, 1989
Документ
Категория
Без категории
Просмотров
11
Размер файла
981 Кб
Теги
contributes, synovium, capacity, memory, intrinsic, migratoria, rheumatoid, cells, accumulation
1/--страниц
Пожаловаться на содержимое документа