вход по аккаунту


Regulation of adhesion molecule expression by human synovial microvascular endothelial cells in vitro.

код для вставкиСкачать
Vol. 39, No. 3, March 1996, pp 467-477
8 1996, American College of Rheumatology
Objective. To examine the in vitro expression of
E-selectin, P-selectin, intercellular adhesion molecule 1
(ICAM-I), ICAM-2, vascular cell adhesion molecule 1
(VCAM-l), and platelet-endothelial cell adhesion molecule 1 (PECAM-1) by synovial microvascular endothelial cells (SMEC) in comparison with microvascular
neonatal foreskin endothelial cells (FSE) and macrovascular human umbilical vein endothelial cells (HUVE).
Methods. Cultured endothelial cells were treated
for 4 hours with medium alone or tumor necrosis factor
a (TNFa). The expression of endothelial adhesion molecules was evaluated by flow cytometry, cell enzymelinked immunosorbent assay, and Northern blot analysis.
Results. SMEC continuously expressed E-selectin
under basal culture conditions, whereas FSE and HUVE
did not. TNFa treatment of rheumatoid arthritis (RA)
SMEC resulted in sustained peak expression of Eselectin for up to 24 hours, which subsequently declined
but remained elevated even at 72 hours. In contrast,
peak E-selectin expression in FSE and HUVE occurred
between 4 hours and 16 hours after TNFa treatment
and then declined to near basal levels by 24-48 hours.
SMEC expressed significantly higher levels of ICAM-1
compared with HUVE under basal culture conditions.
There was no difference between SMEC, FSE, and
Supported by grants from the National Health and Medical
Research Council of Australia and the Wenkart Foundation. Mr.
To’s work was supported by a Postgraduate Biomedical Scholarship
from the National Health and Medical Research Council of Australia.
Shing S. T. To, MSc, Peter M. Newman, BSc(Hons),
Valentine J. Hyland, PhD, Bruce G. Robinson, MD, Leslie Schrieber, MD: University of Sydney at Royal North Shore Hospital, St.
Leonards, New South Wales, Australia.
Address reprint requests to Shing S. T. To, MSc, Department of Rheumatology, University of Sydney, Royal North Shore
Hospital, St. Leonards, NSW, 2065, Australia.
Submitted for publication June 26, 1995; accepted in revised form September 20, 1995.
HUVE in the expression of P-selectin, VCAM-1,
ICAM-2, or PECAM-1. Northern blot analysis demonstrated that the levels of E-selectin expression by T N F e
stimulated endothelial cells correlated with their respective messenger RNA levels.
Conclusion. Regulation of E-selectin and
ICAM-1 expression in RA synovial endothelium is different from that in neonatal foreskin and human umbilical vein endothelium. The augmented expression of
adhesion molecules in RA synovial endothelium may
facilitate the recruitment of leukocytes to this site.
A prominent feature of rheumatoid arthritis
(RA) is the chronic accumulation of leukocytes in the
inflamed synovium. This may be due to uncontrolled
local synovial production of proinflammatory cytokines, such as tumor necrosis factor a (TNFa), which
leads to the induction andor up-regulation of endothelial adhesion molecules (1,2). The latter play important
roles in the adhesion of circulating leukocytes and
their subsequent transendothelial migration into the
inflamed joint (3,4). There are at least 2 families of
adhesion molecules present on vascular endothelium:
1) the selectin family, which includes E-selectin
(endothelial leukocyte adhesion molecule 1) and Pselectin (GMP-140), and 2) the immunoglobulin supergene family, which includes intercellular adhesion
molecule 1 (ICAM-I), ICAM-2, vascular cell adhesion
molecule 1 (VCAM-I), and platelet-endothelial cell
adhesion molecule 1 (PECAM-1).
There is compelling evidence that endothelium
derived from different anatomic sites is heterogeneous
(5). The endothelium at the postcapillary venules in
lymphoid tissues is morphologically and functionally
distinct. It is formed into high endothelial venules,
which play a critical role in lymphocyte recirculation.
In pathologic conditions, for example in RA syno-
vium, high endothelial venules develop, which are not
found in normal synovium (6). Furthermore, RA synovial endothelium exhibits differences in adhesion molecule expression in situ compared with normal synovium (7). Therefore, to study synovial endothelial cell
properties in vitro, it is preferable to isolate purified
synovial microvascular endothelial cells (SMEC) from
the target tissue. However, the number of reports of in
vitro studies with these cells has been limited, due t o
the difficulties in isolation and culture (8-10). We have
successfully used Ulex-coated Dynabeads, a method
originally developed in our laboratory (1l), to isolate
SMEC from RA, osteoarthritic (OA), and normal
synovia, as well as microvascular endothelial cells
from neonatal foreskin (FSE) and from deciduum. Our
results indicate that E-selectin expression and ICAM-1
expression are differentially regulated in SMEC, particularly in SMEC derived from RA and OA patients,
compared with FSE and HUVE.
Patients. Synovial tissues were obtained from RA
and OA patients undergoing synovectomy or joint replacement surgery. Normal synovial tissues were obtained from
the wrist joints of patients undergoing carpal tunnel decompression for carpal tunnel syndrome who had no evidence of
inflammatory disease. Seven of the patients with RA met the
American College of Rheumatology (formerly, the American
Rheumatism Association) criteria (12), with at least 4 of 7
criteria. Two patients fulfilled only 3 criteria. RA specimens
were obtained from either the knee or the wrist except in 1
case, where the specimen was obtained from the hip. All OA
specimens were from the knee joints of patients who had
radiologically proven OA and no clinical evidence of previous inflammatory arthritis. Neonatal foreskin was obtained
at routine circumcision, and deciduum was obtained at
elective caesarean section.
Cell culture. Synovial microvascular endothelial cells
were isolated from surgical specimens using Ulex europaeus
agglutinin type I-coated Dynabeads (11). The partially purified endothelial cells were usually contaminated with fibroblasts, and sometimes with dendritic cells. Contaminating
cells were physically removed using a diathermy needle
under a phase-contrast microscope (Nikon TMS, Tokyo,
Japan) in a laminar flow cabinet. The endothelial cells were
grown in Biorich medium (ICN Biomedicals, Costa Mesa,
CA) supplemented with 30% normal pooled human serum,
15 mM NaHCO,, 50 &ml ampicillin, 50 &ml bovine lung
heparin (Sigma, St. Louis, MO) and 100 &ml bovine brain
extract (prepared according to the method of Maciag et a1
1131). At confluence, cells were passaged with 0.05%
(weightlvolume) trypsin and 0.02% (w/v) EDTA (Sigma).
HUVE cells were obtained by collagenase digestion
as described by Jaf€e et a1 (14). Cells were maintained in
growth medium consisting of M199 medium (ICN Biomedi-
cals) supplemented with 20% heat-inactivated fetal calf
serum (Commonwealth Serum Laboratories, Victoria, Australia), 15 mM NaHCO,, 50 &ml bovine lung heparin, and
100 &ml bovine brain extract.
Neonatal FSE was a &t from Dr. Chris Jackson
(Department of Rheumatology, Royal North Shore Hospital), and deciduum endothelial cells @EC) were kindly
provided by Dr. Eileen Gallery (Department of Renal Medicine, Royal North Shore Hospital). These cells were maintained in the same growth medium as was used for synovial
endothelial cells. Endothelial cells at passages 4-6 were used
for adhesion molecule expression analysis.
Monoclonal antibodies (MAb) and Bow cytometric
analysis. Confluent endothelial cells grown on 60-mm Petri
dishes were treated with either control medium (Biorich
medium + 10% normal pooled human serum) alone or
control medium containing recombinant human TNFa (a &
from Dr. Debbie Rathjen, Peptide Technology, Sydney,
Australia) at 10 ng/ml for 4 hours at 37°C. The cells were
washed with phosphate buffered saline (PBS) and detached
nonenzymatically with 5 mM EDTA in PBS. The resulting
single-cell suspensions were incubated with MAb against
ICAM-1 (1:4,000 ascites, clone 1.H4; from Dr. Andrew
Boyd, The Walter and Eliza Hall Institute of Medical Research, Melbourne, Australia), ICAM-2 (1 &ml, clone
CBR-IC212; from Dr. T. Springer, The Center for Blood
Research, Boston, MA), VCAM-1 (1 d m l , clone BBAS;
from British Biotechnology, Abington, UK), E-selectin (1
&ml, clone H18R; from Becton Dickinson, San Jose, CA),
P-selectin (1:1,OOO ascites, clone AK6; from Dr. M. Berndt,
Baker Medical Institute, Melbourne, Australia), PECAM-1
(1:2,000 ascites, clone WM-59; from Dr. E. Favaloro, Institute of Clinical Pathology and Medical Research, Sydney,
Australia), or isotype-matched immunoglobulin controls,
followed by fluorescein isothiocyanate-conjugated sheep
anti-mouse Ig F(ab’), (1:50 dilution; Silenus Laboratory,
Hawthorn, Victoria, Australia). The cells were then fixed
with 1% (w/v) paraformaldehyde and analyzed by flow
cytometry (Coulter Elite, Hialeah, FL).
Cell enzyme-linked immunosorbent assay (ELSA).
Endothelial cells were plated at 4 x lo4 celldwell on
gelatin-coated %well plates. After overnight incubation, the
cells were treated with control medium or TNFa at 10 ng/ml.
At the end of the incubation period, the cells were fixed with
0.025% (volumeholume) glutaraldehyde for 10 minutes at
room temperature, washed with PBS, and stored in 0.02%
gelatin/l mM azidePBS at 4°C until use. The cell ELISA
procedure involved washing the plates with PBS/O.OS% (v/v)
Tween 20 (Sigma) followed by incubation with MAb against
E-selectin (1 &ml) for 1 hour at 37°C. Binding of MAb was
detected using horseradish peroxidase-conjugated goat antimouse Ig (1:1,OOO dilution; Tago, Burlingame, CA) and
2,2’-azino-bis(3-ethylbenzthiazole)-6-sulfonicacid (Sigma).
Plates were read in a Titertek Multiscan plate reader (ICN
Biomedicals) at 414 nm and 492 nm.
Northern blot analysis. Total RNA was extracted
from untreated confluent cultures of endothelial cells using
the guanidium thiocyanate method as described by Chomczynski and Sacchi (15). Five micrograms of denatured total
RNA was electrophoresed through a 1.4% agarose gel containing 1.25M formaldehyde, then transferred to Genescreen
Figure 1. A, Phase-contrast microscopy of synovial endothelial cells isolated from a rheumatoid arthritis (RA)patient, showing cobblestone
morphology (original magnification x 215). B, Indirect immunofluorescence, showing expression of von Willebrand factor in the synovial
endothelial cells of an RA patient (original magnification x 950).
- control lg
-antj-E-selectin mAb
TN Fa-treated
control lg
- antLE-selectjn mAb
Log fluorescence
Figure 2. E-selectin expression by rheumatoid arthritis endothelial cells, as determined by flow cytometry (A and C) and indirect
immmofluorescence (B and D). A and B represent results obtained under basal culture conditions; C and D represent results obtained after
stimulation with tumor necrosis factor a (TNFa; 10 ng/ml for 4 hours). Arrows in B indicate E-selectin-positive cells. mAb = monoclonal
antibody. (Original magnification x 950 in B and D.)
nylon membrane (NEN Research Products, Boston, MA)
and ultraviolet cross linked (Stratalinker, Stratagene Cloning
Systems, La Jolla, CA). The membrane was further baked
for 2 hours at 80°C. Prehybridization was camed out overnight in a buffer containing 5 x SSPE (0.75M sodium
chloride, 50 mM sodium phosphate, 5 mM EDTA), 2% (w/v)
sodium dodecyl sulfate, 10% (wh) dextran sulfate, and 100
&ml sheared salmon sperm DNA, at 65°C. The membrane
was then hybridized overnight with E-selectin DNA probe
(1,841 basepairs; a gift from Dr. Eugene Butcher, Stanford
University, Palo Alto, CA) that had been labeled with
[a-”P]dCTP by the random primer method. Autaradiography was performed and quantitated using a Phosphor-Imager
(Molecular Dynamics, Sunnyvale, CA). The membrane was
sequentially probed with radiolabeled ICAM-1 DNA (1.5 kb;
a gift from Dr. Andrew Boyd) and mouse pactin DNA
(-600 bp; a gift from Dr. B. Spiegelman, Dana Farber
Cancer Institute, Boston, MA).
Statistical analysis. Data from flow cytometric analyses were expressed as mean fluorescence values. Before
statistical analyses were performed, these data were transformed to log,, to stabilize variances between sources of
50 A
f --I
C 40
0 ,
FndoMeUal cell sour=
Figure 4. E-selectin expression after stimulation with tumor necrosis factor a (TNFa; 10 ng/ml) for 4 hours. Open and closed symbols
represent basal and TNFwinduced E-selectin expression, respectively. Values in parentheses are the number of specimens examined. Bars represent the mean fluorescence for each source of
endothelial cells. * = P < 0.05 by ANCOVA, RA and normal SMEC
expression of E-selectin versus FSE and HUVE expression of
E-selectin. See Figure 3 for other definitions.
endothelial cells and types of adhesion molecules. Analysis
of covariance (ANCOVA) was used to compare leastsquares mean fluorescence values for different sources of
tissue and different types of adhesion molecules. Differences
in the background measure of isotype-matched Ig controls
were taken into account before the analyses were performed. P values less than 0.05 were considered significant.
Endothelialcell source
Figure 3. Basal E-selectin expression by endothelial cells from
various sources, as determined by flow cytometry. Closed symbols
represent mean fluorescence intensities when cells were stained
with monoclonal antibodies against E-selectin; open symbols represent mean background fluorescence obtained using isotype-matched
Ig controls. Values in parentheses are the number of specimens
examined. Bars represent the mean fluorescence for each source
of endothelial cells. * = P < 0.05 by analysis of covariance
(ANCOVA), synovial microvascular endothelial cell (SMEC) expression of E-selectin versus human umbilical vein endothelial cell
(HUVE) expression of E-selectin. t = P < 0.05 by ANCOVA,
rheumatoid arthritis (RA) SMEC expression of E-selectin versus
neonatal foreskin endothelial cell (FSE) expression of E-selectin.
OA = osteoarthritis.
Cell culture. Endothelial cells isolated from
synovial tissues exhibited typical cobblestone morphology (Figure 1A). More than 98% of the cells in the
monolayer were confirmed to be endothelial in origin,
based on their expression of von Willebrand factor
(Figure IB). Endothelial cells derived from other
sources were also positively identified.
E-selectin expression. Under basal culture conditions, a subpopulation of RA SMEC expressed low
levels of E-selectin (Figures 2A and B), whereas FSE
and HUVE did not. Incubation with TNFa (10 ng/ml)
for 4 hours resulted in E-selectin expression by all RA
SMEC, and at much higher levels compared with
unstimulated cells (Figures 2C and D). RA, OA, and
normal SMEC expressed low but significant levels of
E-selectin (P < 0.05 by ANCOVA) compared with
HUVE, undcr basal conditions (Figure 3). SMEC
47 1
1- -
a 4.05kb
J18s Figure 5. Northern blot analysis of E-selectin and intercellular adhesion molecule 1 (ICAM-1) messenger
RNA expression by endothelial cells grown under basal culture conditions. H = human umbilical vein
endothelial cells; F = neonatal foreskin endothelial cells; R, 0, and N = synovial endothelial cells from
rheumatoid arthritis patients, osteoarthritis patients, and normal subjects, respectively; D = decidual
endothelial cells; + = positive RNA control from tumor necrosis factor *treated human umbilical vein
endothelial cells.
derived from RA patients expressed higher levels of
E-selectin than those derived from OA patients and
normal subjects, but the dBerence did not reach
statistical significance, probably due to the limited
number of samples available for evaluation. Three
DEC cell lines were also examined, but no E-selectin
expression was detectable (results not shown).
The fact that HUVE and SMEC were cultured
in different media might have resulted in differences in
E-selectin expression. However, when HUVE were
cultured in the same medium used for SMEC, no
Time (hr)
Figure 6. Kinetics of E-selectin expression by different sources of
endothelial cells after tumor necrosis factor a (TNFa) stimulation.
All endothelial cells (HUVE [O],
FSE [XI,RA [A], OA [+I, and
normal [O] synovium) were treated with TNFa (10 ng/ml) in Biorich
medium supplemented with 10% normal pooled human serum. Data
shown are the means of triplicate wells from 1 experiment. Similar
results were obtained in 2 other experiments using endothelial cells
from different individuals. See Figure 3 for other definitions.
E-selectin expression was observed (results not
shown). Basal E-selectin expression by SMEC could
have been due to the presence of contaminating cells
such as monocytes/macrophages, since the latter are
able to produce cytokines. However, treatment of
HUVE cells with RA SMEC-conditioned medium for
4 hours did not result in the induction of E-selectin.
Furthermore, flow cytometric analysis and immunoperoxidase staining using MAb (clone RPA-m1) did
not show any monocyte/macrophages in the RA
SMEC cultures (results not shown).
To determine if the observed differences between SMEC, FSE, and HUVE in basal E-selectin
expression could be maintained after cytokine activation, the endothelial cells were stimulated with TNFa,
and E-selectin expression was quantitated by flow
cytometry. A 4-hour incubation period and 10 nglml
TNFa concentration were chosen for maximal Eselectin expression. All endothelial cells expressed
increased levels of E-selectin following TNFa stimulation (Figure 4). RA and normal SMEC had a significant increase in E-selectin expression when compared
with FSE and HUVE.
To examine the basal steady-state E-selectin
messenger RNA (mRNA) levels in endothelial cells,
we performed Northern blot analysis on total RNA
extracted from HUVE, FSE, SMEC, and DEC (Figure 5). One of 5 HUVE and 2 of 6 FSE samples
showed low levels of E-selectin mRNA. In contrast, 3
of 5 RA SMEC expressed high E-selectin mRNA
levels compared with the HUVE and FSE samples.
Two of 4 OA SMEC samples expressed low levels of
E-selectin mRNA, while 1 of 4 normal SMEC expressed moderate levels. Unexpectedly, 2 of 4 DEC
also showed moderate E-selectin mRNA expression.
Levels of pactin in endothelial cells derived from
different specimens varied over a 3-fold range (data
3u 2.3
l 1.2
4 12 24 48 72
o 4 12 24 48
4 12 24 48 72
Time (hr)
Figure 7. Kinetics of E-selectin messenger RNA (mRNA) and protein expression by HUVE, FSE, and RA
SMEC after tumor necrosis factor a (TNFa) stimulation. + = positive RNA control from TNFa-treated
HUVE. Black bars = E-selectin protein levels as determined by flow cytometry; grey bars = ratio of
E-selectin mRNA levels to Pactin mRNA levels. See Figure 3 for other definitions.
not shown), and thus were not used as a standard with
which to compare E-selectin levels. We therefore used
18s ribosomal RNA as our standard for comparison.
The difference in E-selectin mRNA expression by
different endothelial cells was not due to uneven
loading of the RNA, since ethidium bromide staining
of the 18s ribosomal RNA showed that approximately
equal amounts of total RNA were loaded for gel
The maintenance of low levels of E-selectin by
SMEC under basal culture conditions (Figure 3) suggested that their regulation of E-selectin expression
may be different from that by HUVE or FSE. To
compare the kinetics of E-selectin expression, a variety of endothelial cells were incubated with TNFa (10
ng/ml) for up to 72 hours, and E-selectin protein
expression was quantitated by cell ELISA. All cell
lines tested responded to TNFa stimulation in a similar manner. Peak E-selectin expression occurred between 4 and 16 hours and then declined over time
(Figure 6). HUVE E-selectin expression declined rapidly and reached baseline levels after 48 hours of
incubation, whereas FSE and SMEC maintained elevated levels at this time. In the case of RA and OA
SMEC, peak E-selectin expression was sustained up
to 24 hours, and there was continued expression of
high levels of E-selectin even after 72 hours of incubation.
The kinetics of TNFwstimulated E-selectin expression in HUVE, FSE, and RA SMEC were further
examined by flow cytometry and Northern blot analysis (Figure 7). RA SMEC showed sustained high
levels of E-selectin mRNA for up to 24 hours, and
these remained elevated even after 72 hours. In contrast, HUVE and FSE E-selectin mRNA declined to
baseline levels by 48 hours. TNFa stimulation did not
alter the pactin mRNA levels in endothelial cells.
Therefore, pactin mRNA levels were used as a standard to quantitate the kinetics of E-selectin mRNA
expression. The kinetics of cell surface E-selectin,
measured by immunofluorescence, corresponded to its
steady-state mRNA levels, expressed as ratios of
E-selectin mRNA to pactin mRNA. There was a
strong linear relationship (r = O.%) between the protein and mRNA levels of E-selectin in TNFastimulated endothelial cells (data not shown)
ICAM-1 expression. HUVE and FSE expressed
low levels of ICAM-I under basal culture conditions.
All SMEC had significantly higher levels of ICAM-1
than did HUVE (Figure 8). High ICAM-1 levels were
not restricted to SMEC, since DEC also consistently
expressed significantly higher levels compared with
705 1
Figure 8. Intercellular adhesion molecule 1 (ICAM-1) expression
under basal culture conditions, as determined by flow cytometry.
Closed symbols represent mean fluorescence intensities when cells
were stained with monoclonal antibodies against ICAM-1; open
symbols represent mean background fluorescence obtained using
isotype-matched Ig controls. Values in parentheses are the number
of specimens examined. Bars represent the mean fluorescence for
each source of endothelial cells. * = P < 0.0005 by ANCOVA,
expression of ICAM-1 by RA, OA, and normal SMEC and by DEC
versus expression of ICAM-I by HUVE. t = P < 0.005 by
ANCOVA, expression of ICAM-1 by RA and OA SMEC and by
DEC versus expression of ICAM-1 by FSE. See Figure 3 for other
would maintain higher ICAM-1 expression compared
with FSE and HUVE at the initial phase of cytokine
activation. Figure 9 shows that all endothelial cells
expressed increased levels of ICAM-1 following T N F a
stimulation. SMEC ICAM-1 expression was further
up-regulated by T N F a treatment. RA and normal
SMEC had significant increases in ICAM-1 expression
when compared with FSE and HUVE.
Expression of other adhesion molecules. The
expression of P-selectin, VCAM- 1, ICAM-2, and
PECAM-1 by various sources of endothelial cells is
summarized in Table 1. Compared with isotypematched Ig control, there was no basal expression of
P-selectin on any of the endothelial cells tested. T N F a
stimulation did not result in the translocation of Pselectin expression to the endothelial cell surface, and
was therefore not evaluated. VCAM-1 was not expressed under basal conditions but was detectable
after 4 hours of stimulation with T N F a (10 ng/ml).
However, there was no significant difference in
VCAM- 1 expression between the various endothelial
cell sources at this time. The kinetics of T N F e
stimulated VCAM-I expression were similar in all
FSE and HUVE. RA and OA SMEC and DEC, but
not normal SMEC, expressed significantly higher
ICAM-1 levels when compared with FSE (P < 0.005
The basal steady-state ICAM-1 mRNA levels in
endothelial cells were also evaluated (Figure 5 ) . There
was little or no detectable ICAM-I mRNA expression
in 4 of the 5 HUVE samples examined; 1 sample
expressed high levels. Four of 6 FSE samples showed
low-to-moderate levels of ICAM-1 mRNA. All SMEC
and DEC showed varying amounts of ICAM-1 mRNA.
Three of 5 RA SMEC samples, 3 of 4 OA samples, 3 of
4 normal SMEC samples, and all DEC samples expressed moderate-to-high levels of ICAM-1 mRNA.
To examine whether the high level of basal
ICAM-1 expression by SMEC could be augmented by
cytokine stimulation, the endothelial cells were treated
with 10 ng/ml T N F a for 4 hours, and ICAM-1 expression was quantitated by flow cytometry. Although
ICAM-1 up-regulation peaked at -24 hours, a 4-hour
stimulation period was chosen to determine if SMEC
i 807
t I
EadPtbellal cell s
Figure 9. Intercellular adhesion molecule 1 (ICAM-I) expression
after stimulation with tumor necrosis factor a (TNFa; 10 np/ml) for
4 hours. Open and closed symbols represent basal and TNFw
stimulated ICAM- 1 expression, respectively. Values in parentheses
are the number of specimens examined. Bars represent the mean
fluorescence for each source of endothelial cells. * = P < 0.05 by
ANCOVA, expression of ICAM-1 by RA and normal SMEC versus
expression of ICAM-1 by FSE and HUVE. See Figure 3 for other
11.2 (4)
4.0 (3)
f 0.34
26.2 2 7.2
15.9 (3) 36.6
8.5 (3)
0.08 (4)
f 0.12
11.2 32.2
(3) 35.6
8.0 (3)
f 2.86
f 2.66
0.09 (3)
f 0.57
0.07 (2)
f 0.50
2.15 (2)
3.9 (2)
f 3.05
19.2 f 3.7
(2) 2.67
0.10 (2)
f 0.22
1.05 0.39
39.9 2 6.75 (2) 38.7 2 6.10
22.0 f 3.7 (2)
' Endothelial cells were either untreated or were treated with tumor necrosis factor a (TNFa; 10 n g / d for 4 hours). Values shown are the mean f SEM mean fluorescence
intensities, as determined by flow cytometry. Values in parentheses are the number of different cell lines examined under both conditions. VCAM-1 = vascular cell adhesion
molecule 1; ICAM-1 = intercellular adhesion molecule 1; PECAM-1 = platelet-endothelial cell adhesion molecule 1; HUVE = human umbilical vein endothelial cells; FSE
= foreskin endothelial cells; RA = rheumatoid arthritis; OA = osteoarthritis; ND = not done.
0.12 (5)
0.19 f 0.06 (3)
Table 1. P-selectin, VCAM-1, ICAM-2, and PECAM-1 expression by endothelial cells'
endothelial cells; peak expression occurred between
16 and 24 hours and remained elevated up to 72 hours
(data not shown).
Unlike ICAM-1, ICAM-2 is constitutively expressed at high levels by HUVE and FSE. We investigated whether expression of ICAM-2 by SMEC was
higher than that by HUVE and FSE, as observed with
ICAM-1. However, all endothelial cells examined
showed comparable levels of ICAM-2, and TNFa
treatment did not alter its expression. All endothelial
cells tested constitutively expressed very high levels
of PECAM-1. There was no significant difference
between the levels of PECAM-1 in macrovascular
HUVE and microvascular FSE and SMEC. Moreover, TNFa treatment did not increase PECAM-1
In the present study, we examined the expression and regulation of adhesion molecules by human
synovial microvascular endothelial cells. SMEC, particularly those derived from patients with RA, expressed low but significant levels of E-selectin compared with HUVE and FSE. The expression of
E-selectin was limited to a subpopulation of SMEC.
This contrasts with the findings of a study by Abbot et
a1 (8), in which E-selectin expression was observed
universally in RA SMEC. The discrepancy may be due
to the method of detection used. We used flow cytometry, whereas Abbot et a1 used a cell ELISA method.
The latter measures the presence of E-selectin in the
cells grown in microtiter plates, but does not discriminate between E-selectin-positive and -negative subpopulations. In contrast, flow cytometry enables determination of the pattern of E-selectin expression in
subpopulations of cells.
It is unclear why some SMEC maintain the
expression of E-selectin for up to 4 weeks in tissue
culture without exogenous cytokine stimulation. In
contrast, HUVE, which do not express E-selectin
under basal conditions, do express E-selectin after
TNFa stimulation, but only transiently. Koch et a1 (16)
observed that not all endothelial cells in RA synovium
expressed E-selectin in situ. It is likely that the isolated SMEC, even from the same synovium, are
heterogenous in origin, resulting in only a subpopulation of cells that are E-selectin positive.
To investigate the regulation of E-selectin by
SMEC, we performed kinetic studies on different
endothelial cells. RA and OA SMEC showed pro-
longed E-selectin expression after TNFa stimulation,
compared with FSE or HUVE (Figure 6). The levels of
E-selectin expression by RA and OA SMEC remained
elevated even after 72 hours, whereas FSE and HUVE
E-selectin levels declined to basal levels by 48 hours.
Endothelial cell E-selectin expression may be regulated by the transcription factor nuclear factor kappa B
(NF-KB) and an inhibitor of NF-KB (I&) (17). The
prolonged E-selectin expression by RA SMEC after
TNFa stimulation may be due to an imbalance of
NF-KB and IKB,resulting in attenuated transcription
of E-selectin.
E-selectin expression is observed in situ in RA,
OA, and normal synovial tissues (16,18). It is possible
that SMEC “memorize” their in vivo phenotype and
retain their pattern of adhesion molecule expression in
vitro. Alternatively, there may be an autocrine mechanism operating in SMEC, whereby E-selectin expression is sustained in vitro.
There were considerable variations in basal
E-selectin protein expression as well as mRNA levels
even within each source of SMEC. This is not unexpected since in situ adhesion molecule expression
studies have shown that not all RA and OA synovial
tissues are positive for E-selectin (16). However, it
is unclear why 1 of 5 HUVE, 2 of 6 FSE, and 2 of 4
DEC samples expressed low-to-moderate levels of
E-selectin mRNA and yet E-selectin protein expression was not detectable in any of these cells. It is
possible that E-selectin protein synthesis may be regulated at a posttranscriptional level, as described for
the translation of fenitin mRNA (19).
ICAM-1 is constitutively expressed on endothelial cells as well as other cell types, such as fibroblasts.
Szekanecz et a1 (20) have shown that significantly
more RA and OA synovial endothelial cells express
ICAM-1 compared with normal synovium. In this
study, we demonstrated that SMEC, as well as DEC,
consistently expressed high levels of ICAM-1 under
basal culture conditions. Our findings differ from those
of Gerritsen et a1 (9), who showed no difference
between SMEC and HUVE in their basal ICAM-1
expression. Those authors also showed that treatment
of SMEC with TNFa for 24 hours resulted in minimal
increases in ICAM-1 expression compared with
HUVE. In contrast, our study has shown that 4 hours
of TNFa stimulation further increased ICAM-I expression by SMEC (Figure 9). In experiments in which
endothelial cells were treated with TNFa for 24 hours
instead of 4 hours, comparable levels of ICAM-1
expression between SMEC and HUVE were observed
(data not shown). The differences between our results
and those of Gerritsen and colleagues may be due to
differences in the cell isolation technique used. Gerritsen et al isolated SMEC by fluorescence-activated cell
sorting following uptake of acetylated low density
lipoprotein labeled with a fluorescent probe, DiI. It is
possible that this isolation procedure affects SMEC
expression of adhesion molecules. The technique that
we used is unlikely to account for the difference in
SMEC adhesion molecule expression since FSE were
isolated using the same technique but did not show
high levels of basal ICAM-1 expression.
Vanhee et a1 (21) have recently shown that
continuous stimulation of HUVE with TNFa for >6
days results in persistent ICAM-1 expression even
when the stimulus is no longer present. SMEC and
DEC may have been chronically stimulated in vivo,
resulting in modification of ICAM-1 gene regulation.
The modified regulation of ICAM-1 gene expression
may be retained in the propagated SMEC and DEC
since late passages of these cells still have high levels
of ICAM-1 expression.
There is evidence that protein kinase C (PKC)
is involved in the up-regulation of ICAM-1 expression
in endothelial cells (22-24). Wertheimer et a1 (25) have
suggested that PKC acts primarily at a posttranscriptional level by stabilizing ICAM-1 mRNA. SMEC and
DEC may have up-regulated PKC activity, resulting in
basal induction and a longer half-life of ICAM-1
Bacterial lipopolysaccharide (LPS) induces
E-selectin and ICAM-1 expression on endothelial cells
(26). However, it is unlikely that the growth medium
for SMEC was contaminated with LPS since FSE
were maintained in the same growth medium but did
not express E-selectin or high levels of ICAM-1.
Furthermore, HUVE cells maintained in the same
growth medium did not show induction of these adhesion molecules. Autocrine regulation of ICAM-1 expression, by soluble factors such as interleukin-1a,has
been reported in transitional cell carcinoma cell lines
(27). It is possible that SMEC ICAM-1 expression is
similarly regulated in an autocrine manner. However,
treatment of HUVE cells with SMEC-conditioned
medium for up to 24 hours did not result in augmented
ICAM-1 expression (data not shown).
A recently described endothelial adhesion molecule, vascular adhesion protein l (VAP-l), has been
shown to be present on the endothelium in inflamed
synovium (28). Preliminary experiments were performed using MAb (clone 1B2; a gift from Dr. M.
Salmi, National Public Health Institute, Turku, Finland) to detect VAP-1 on RA SMEC by cell ELISA
and immunoperoxidase staining, However, no detectable levels of VAP-1 were found on FL4 SMEC under
basal or cytokine-activated conditions (results not
shown). This suggests that the expression of VAP-1 by
SMEC requires the presence of unknown in vivo
E-selectin is important in tethering polymorphonuclear leukocytes and memory T lymphocytes to
the endothelium (29,30). In contrast, ICAM-1 plays a
role in both basal leukocyte adhesion and transendothelial migration (31). The augmented expression of
E-selectin and ICAM-1 by SMEC implies a role for
these adhesion molecules in leukocyte extravasation
in the synovium. However, their presence on SMEC
from noninflamed OA or normal synovium, where
little or no leukocyte accumulation occurs, suggests
that a complex interplay of adhesion molecules may
operate in vivo.
The mechanism by which SMEC expression of
E-selectin and ICAM-1 is regulated remains to be
determined. Nevertheless, the augmented expression
of adhesion molecules may play an important role in
directing specific populations of leukocytes to adhere
to and transmigrate into the inflamed synovium.
We thank Dr. M. Cross (Mater Misericordiae Hospital, Crows Nest, NSW, Australia), Dr. M. Tonkin, and
orthopedic surgeons at the Centre for Bone and Joint Diseases (Ryde, NSW, Australia) for providing synovial specimens. We also thank Dr. M. King for performing flow
cytometric analysis, Ms Car0 Badcock for performing statistical analyses, and Dr. T. Longhurst for helpful discussion.
1 . Maini RN, Brennan FM, Williams R, Chu CQ, Cope AP,
Gibbons D, Elliott M, Feldmann M: TNF-alpha in rheumatoid
arthritis and prospects of anti-TNF therapy. Clin Exp Rheumato1 ll:S173-S175, 1993
Farahat MN, Yanni G, Poston R, Panayi GS: Cytokine expression in synovial membranes of patients with rheumatoid arthritis
and osteoarthritis. Ann Rheum Dis 52:870-875, 1993
Cronstein BN, Weissmann G: Current comment: the adhesion
molecules of inflammation. Arthritis Rheum 36: 147-157, 1993
Ziff M: Current comment: role of the endothelium in chronic
inflammatory synovitis. Arthritis Rheum 34: 1345-1352, 1991
Page C, Rose M, Yacoub M, Pigott R: Antigenic heterogeneity
of vascular endothelium. Am J Pathol 141:673-683, 1992
Ziff M: Role of endothelium in the pathogenesis of rheumatoid
synovitis. Int J Tissue React 15:135-137, 1993
El-Gabalawy H, Gallatin M, Vazeux R, Peterman G, Wilkins J:
Expression of ICAM-R (ICAM-3), a novel counter-receptor for
LFA-1, in rheumatoid and nonrheumatoid synovium: comparison with other adhesion molecules. Arthritis Rheum 37:846
854, 1994
8. Abbot SE, Kaul A, Stevens CR, Blake DR: Isolation and
culture of synovial microvascular endothelial cells: characterization and assessment of adhesion molecule expression. Arthritis Rheum 35:401-406, 1992
9. Gemtsen ME, Kelley KA, Ligon G, Perry CA, Shen C-P,
Szczepanski A, Carley WW: Regulation of the expression of
intercellular adhesion molecule 1 in cultured human endothelial
cells derived from rheumatoid synovium. Arthritis Rheum 36:
593-602, 1993
10. Szczepanski A, Moatter T, Carley WW, Gerritsen ME: Induction of cyclooxygenase I1 in human synovial microvessel endothelial cells by interleukin-I: inhibition by glucocorticoids.
Arthritis Rheum 37:495-503, 1994
1 1 . Jackson CJ, Garbett PK, Nissen B, Schrieber L: Binding of
human endothelium to Ulex europaeus I-coated Dynabeads:
application to the isolation of microvascular endothelium. J Cell
Sci %:257-262, 1990
12. 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
13. Maciag T, Cerundolo J, Ilsley S, Kelley PR, Forand R: An
endothelial cell growth factor from bovine hypothalamus: identification and partial characterization. Proc Natl Acad Sci U S A
7656744678, 1979
14. Jaffe EA, Nachman RL, Becker CG, Minick CR: Culture of
human endothelial cells derived from umbilical veins: identification by morphologic and immunologic criteria. J Clin Invest
52:2745-2756, 1973
IS. Chomczynski P, Sacchi N: Single-step method of RNA isolation
by acid guanidinium thiocyanate-phenol-chloroformextraction.
Anal Biochem 162:156-159, 1987
16. Koch AE, Burrows JC, Haines GK, Carlos TM, Harlan JM,
Leibovich SJ: Immunolocalization of endothelial and leukocyte
adhesion molecules in human rheumatoid and osteoarthritic
synovial tissues. Lab Invest 64:313-320, 1 9 9 1
17. Read MA, Whitley MZ, Williams AJ, Collins T: NF-kappa B
and I kappa B alpha: an inducible regulatory system in endothelial activation. J Exp Med 179503-512, 1994
18. Fairburn K, Kunaver M, Wilkinson LS, Cambridge G, Haskard
D, Edwards JC: Intercellular adhesion molecules in normal
synovium. Br J Rheumatol 32:302-306, 1993
19. Harford JB, Klausner RD: Coordinate post-transcriptional reg-
ulation of fenitin and transfemn receptor expression: the role of
regulated RNA-protein interaction. Enzyme 44:28-41, 1990
20. Szekanecz Z, Haines GK, Lin TR, Harlow LA, Goerdt S,
Rayan G, Koch AE: Differential distribution of intercellular
adhesion molecules (ICAM-I, ICAM-2, and ICAM-3) and the
MS-I antigen in normal and diseased human synovia: their
possible pathogenetic and clinical significance in rheumatoid
arthritis. Arthritis Rheum 37:221-231, 1994
21. Vanhee D, Delneste Y, Lassalle P, Gosset P, Joseph M, Tonne1
AB: Modulation of endothelial cell adhesion molecule expression in a situation of chronic inflammatory stimulation. Cell
Immunol 155:446456, 1994
22. Renkonen R, Mennander A, Ustinov J, Mattila P: Activation of
protein kinase C is crucial in the regulation of ICAM-I expression on endothelial cells by interferon-gamma. Int Immunol
2:719-724, 1990
23. Myers CL, Wertheimer SJ, Schembri King J, Parks T, Wallace
RW: Induction of ICAM-1 by TNF-alpha, IL-I beta, and LPS in
human endothelial cells after downregulation of PKC. Am J
Physiol263:C767-C772, 1992
24. Sung CP, Arleth AJ, Nambi P: Evidence for involvement of
protein kinase C in expression of intracellular adhesion molecule-l (ICAM-I) by human vascular endothelial cells. Pharmacology 48:143-146, 1994
25. Wertheimer SJ, Myers CL, Wallace RW, Parks TP: Intercellular adhesion molecule-1 gene expression in human endothelial
cells: differential regulation by tumor necrosis factor-alpha and
phorbol myristate acetate. J Biol Chem 267:12030-12035, 1992
26. Leeuwenberg JF, Smeets EF, Neefies JJ, Shaffer MA, Cinek T,
Jeunhomme TM, Ahern TJ, Buurman WA: E-selectin and
intercellular adhesion molecule-I are released by activated
human endothelial cells in vitro. Immunology 77543-549, 1992
27. Alexandroff AB, Jackson AM, Esuvaranathan K, Prescott S,
James K: Autocrine regulation of ICAM-I expression on bladder cancer cell lines: evidence for the role of IL-I alpha.
Immunol Lett 40: 117-124, 1994
28. Salmi M, Jalkanen S: A 90-kilodalton endothelial cell molecule
mediating lymphocyte binding in humans. Science 257: 14071409,1992
29. Abbassi 0, Kishimoto TK, Mcintire LV, Anderson DC, Smith
CW: E-Selectin supports neutrophil rolling in vitro under conditions of flow. J Clin Invest 92:2719-2730, 1993
30. Shimizu Y, Shaw S, Graber N, Gopal TV, Horgan KJ, van
Seventer GA, Newman W: Activation-independent binding of
human memory T cells to adhesion molecule ELAM-I [see
comments]. Nature 349:799-802, 1991
31. Oppenheimer-Marks N, Davis LS, Bogue DT, Ramberg J,
Lipsky PE: Differential utilization of ICAM-I and VCAM-I
during the adhesion and transendothelid migration of human T
lymphocytes. J Immunol 147:2913-2921, 1991
Без категории
Размер файла
959 Кб
expressions, endothelial, microvascular, molecules, regulation, synovial, human, adhesion, vitro, cells
Пожаловаться на содержимое документа