вход по аккаунту


Regulation of the expression of intercellular adhesion molecule 1 in cultured human endothelial cells derived from rheumatoid synovium.

код для вставкиСкачать
Objective. To examine the regulation of intercellular adhesion molecule 1 (ICAM-1) in human synovial
microvascular endothelial cells (HSE) and human umbilical vein endothelial cells (HUVE) upon exposure to a
variety of agents.
Methods. Cultured endothelial cells were treated
with various cytokines alone and in combination. The
expression of ICAM-1 was evaluated at several levels,
including an investigation of messenger RNA (mRNA)
and surface protein expression.
Results. Treatment of HSE with interleukin-la
(IL-la) or tumor necrosis factor a (TNFa) resulted in
minimal increases in ICAM-1 expression, in contrast to
findings with HUVE. Incubation of HUVE or HSE with
IL-1 or TNF in combination with interferon-y (IFNy)
greatly potentiated the increase in ICAM-1 surface
expression. The synergistic effect of IFNy and TNF was
confirmed by several methods, including a cell-based
enzyme-linked immunosorbent assay, tluorescenceactivated cell sorting, immunofluorescence staining, and
determination of mRNA levels. IFN y also augmented
the actions of several other agonists on HSE, i.e., IL-4,
lipopolysaccharide, and TNFPllymphotoxin. Immunoprecipitation of TNFa IFN ystimulated, '251-labeled
HSE cells with anti-ICAM-1 revealed a single 90-kd
band, similar in size to ICAM-1 from HUVE treated in
From the Institute for Inflammation and Autoimmunity,
Miles Inc., West Haven, Connecticut.
Mary E. Gemtsen, PhD; Keith A. Kelley, BS; Gwenda
Ligon, BS; Carol A. Perry, BA; Chien-Ping Shen, MS; Andrew
Szczepanski, PhD; William W. Carley, PhD.
Address reprint requests to Mary E. Gemtsen, PhD, Institute for Inflammation and Autoimmunity, Miles Inc., 400 Morgan
Lane, West Haven, CT 06516.
Submitted for publication September 17, 1992; accepted in
revised form December 15, 1992.
Arthritis and Rheumatism, Vol. 36, No. 5 (May 1993)
an identical manner. Unexpectedly, IFNy alone was a
potent stimulus for HSE ICAM-1 mRNA synthesis, but
was relatively ineffective in HUVE.
Conclusion. These studies indicate that IFNy
plays a critical synergistic role in the regulation of
ICAM-1 expression in human synovial endothelial cells.
Inflammation of the rheumatoid joint involves
at least two components of cellular infiltration: an
exudative component characterized by population of
the synovial effusion with polymorphonuclear cells,
and a chronic inflammatory phase in the sublining
layer, associated with mononuclear cells. During
transvascular migration, both neutrophils and mononuclear cells must bind to the microvascular endothelium (predominantly the postcapillary venule), pass
through the vascular wall, and migrate to the appropriate microenvironment. The interaction of the integrin lymphocyte function-associated antigen 1
(CDl ldCD18) with its ligand, intercellular adhesion
molecule 1 (ICAM-l), is a major molecular pathway in
leukocyte-endothelial cell recognition (1-3). ICAM-1
surface expression occurs at low levels on many cell
types and can be induced by a variety of cytokines,
including interleukin-1 (IL- l), tumor necrosis factor a
(TNFa), and interferon-y (IFNy) (3-6).
Endothelial, fibroblastic, and epithelial cells
have shown differences in terms of which cytokines
are more capable of inducing ICAM-1 expression, and
thus, the locally generated mediators may serve to
regulate patterns of inflammatory cell localization (6).
The majority of studies characterizing the regulation of
ICAM- 1 expression in endothelial cells have been
performed using cultured human umbilical vein endothelial cells (HUVE). These cells, although providing a
useful and well-characterized model system, may not
exhibit the same properties of gene regulation as their
microvascular counterparts. Heterogeneity in morphology, as well as gene expression and regulation in
endothelium derived from various sized vessels and
different vascular beds, is a concept that is now widely
accepted. Therefore, in order to better assess the role
of endothelium in rheumatoid arthritis (RA), we characterized the regulation of ICAM- 1 expression in
human synovial microvascular endothelial cells
(HSE), derived from both normal subjects and RA
patients, and compared the findings with those obtained using cultured HUVE. The results suggest that
the cell surface expression of ICAM-1 and its messenger RNA (mRNA) levels are differentially regulated in
HSE versus HUVE.
Materials. Nu-Serum and Matrigel were obtained
from Collaborative Research (Bedford, MA). Fetal calf
serum (FCS) was purchased from Hyclone (Logan, UT). All
other cell culture materials were purchased from Gibco
(Grand Island, NY). Heparin, rhodamine-conjugated lectins,
nonimmune mouse IgG, urease-conjugated rabbit antimouse IgG, lipopolysaccharide (LPS) (Escherichia coli
0 1 11:B4), phenylmethylsulfonyl fluoride (PMSF), aprotinin,
Staphylococcus aureus, and protein A were from Sigma (St.
Louis, MO). Recombinant human TNFP and recombinant
human IL-4 were from R & D Systems (Minneapolis, MN).
Recombinant human IL-la (lo5 units/pg) was from Genzyme (Boston, MA), and recombinant human TNFa (>2 x
10' unitshg), recombinant human IFNy (>2 x lo7 units/
mg), and collagenase B were from Boehringer Mannheim
(Indianapolis, IN). DiI-Ac-LDL (7) was purchased from
Biomedical Technologies (Staughton, MA). RNA extraction
kits were purchased from Pharmacia (Piscataway, NJ).
Nytran membranes were from Schleicher and Schuell
(Keene, NH). 32P-dCTP (3,000 Ci/mmole) and lZ5Iradionuclide (17 Ci/mg) were purchased from Amersham
(Arlington Heights, IL) and Dupont-New England Nuclear
(Boston, MA), respectively. Sulfated Bolton Hunter reagent
(s-SHPP) was from Pierce Chemical (Rockford, IL).
Cell culture. Synovial tissues were obtained from the
wrist, finger, and knee joints of 3 female patients with RA.
The patients ranged in age from 16 to 50 years. Normal
synovium was obtained from an amputated limb removed as
a result of accidental trauma. Immediately upon surgical
removal, and under aseptic conditions, 0.5-2 gm of tissue
was placed in growth medium (RPMI 1640-10% Nu-Serum10% FCS, supplemented with retinal-derived growth factor
[4 pl/ml] as described previously [8] and heparin [25 pglmll)
and transported to Miles Research Center (West Haven,
CT), over a 1-2-day period.
In the cell culture laboratory, the tissue was removed
from growth medium, minced finely with scissors, and
incubated for 30 minutes in 0.2% collagenase-O.l% bovine
serum albumin (BSA) in RPMI 1640 with 2 mM L-glutamine,
with agitation. After 30 minutes, the material was sheared 10
times through a 10-ml pipette, after which the tissue was
agitated an additional 30 minutes. The material was filtered
through sterile 100-pm nylon mesh, pelleted by centrifugation, resuspended in growth medium, and seeded to gelatincoated tissue culture dishes (35 mm in diameter; Falcon,
Oxnard, CA). Primary cultures attained confluence within
3-5 days and were subcultured twice with trypsin-EDTA
(1 :3 split). The subcultured cells were grown to confluence in
two T-75 flasks, incubated for 4 hours with 5 pg/ml DiI-AcLDL, and the upper 1% of fluorescent cells was selected
using a fluorescence-activated cell sorter (FACS) as described by Voyta et al (7). Cells used in this study were from
the third-to-eighth passage following sorting.
Human umbilical vein endothelial cells were purchased from Clonetics (San Diego, CA) and were grown in
EGM-UV medium (Clonetics, San Diego, CA) containing
10% FCS and bovine brain extract. Cells used in this study
were from the first-to-eighth passage.
Cell identification. Endothelial cells were identified
on the basis of a number of endothelial-specific cell markers:
Factor VIII-related antigen (9), angiotensin-converting enzyme (lo), uptake of DiI-Ac-LDL (7), tube formation in
Matrigel (1 l), and staining with rhodamine-conjugated Ulex
europaeus I (UEA) (12), using methods detailed previously
(8) or as detailed below. All cultures used in this study were
>99% positive for all of the above markers. All histologic
markers (i.e., Factor VIII-related antigen, angiotensinconverting enzyme, and UEA) were cross-evaluated and
confirmed with fresh-frozen sections of rheumatoid tissue
and were demonstrated to stain microvascular endothelial
cells in situ (results not shown).
To evaluate tube formation, confluent HSE were
removed from their growth media, grown for 48 hours in
RPMI 1640-10% FCS, trypsinized, and cells (5 x lo4) were
then plated in RPMI 1640-10% FCS on Matrigel (0.6ml/
35-mm dish). After various periods of time, the cells were
examined by phase microscopy and fixed in 3.7% formalin in
phosphate buffered saline (PBS) for 30 minutes. Actin was
stained by incubation overnight with 100 ng/ml rhodaminephalloidin, and photographs were taken using Hoffman modulation optics or rhodamine fluorescence illumination on a
Leitz Fluorovert FS microscope.
Confluent HSE were incubated for 4 hours with 5
pg/ml DiI-Ac-LDL in growth medium, washed 3 times with
PBS, and fixed in 3.7% formalin in PBS for 10 minutes.
Photographs were taken using rhodamine fluorescence illumination on a Leitz Fluorovert FS microscope.
Cytokine treatment. Endothelial cells were grown to
confluence on 96-well plates for enzyme-linked immunosorbent assay (ELISA), 6-well plates for FACS analysis, or
T-150 flasks for mRNA isolation in growth medium. When
confluence was reached, the media were replaced with
RPMI 1640-10% FCS and the cells incubated an additional
48 hours. Cells were then incubated with the indicated
concentration of cytokine in RPMI 1640-10% FCS for various time periods. Under these conditions, incubations with
cytokines had no significant effects on cell number, nor was
there any apparent toxicity as assessed by cell morphology
and by MTT assay (13). Toxicity, as assessed by the M'IT
assay, was expressed as % inhibition, calculated using the
1 - (control MTT optical density [OD] - test
Control MTT OD
x 100
Cell-based ELISA. Endothelial cells were incubated
with the indicated cytokines in RPMI 1640-10% FCS for
various periods of time. The media were removed and the
cells fixed for 10 minutes at 23°C with 3.7% formalin in PBS.
The cells were washed and then incubated with a 1:1,OOO
dilution of the first antibody (anti-ICAM-1 C78.5 [14]; kindly
supplied by Dr. J. Greve, Molecular Therapeutics Inc., West
Haven, CT) or nonimmune control IgG in PBS-0.1% BSA
for 2 hours at 23°C. The cells were washed in PBS-BSA, then
blocked for 1 hour in PBS-3% BSA. After washing, the cells
were incubated with urease-conjugated rabbit anti-mouse
IgG. Urease activity was then determined by monitoring urea
metabolism in the presence of bromcresol purple at pH 4.8. OD
at 590 nm was determined on a Molecular Devices kinetic
microplate reader (Menlo Park, CA). Data are expressed as the
percent of control (untreated) cell ICAM-1 expression and are
corrected for the background OD obtained using nonimmune
FACS analyses. Endothelial cells were incubated
with cytokines as described above. Cells were removed from
the culture dishes by washing once with Versene followed by
brief incubation in 1 ml trypsin. Cells were diluted in RPMI
1640-10% serum, pelleted by centrifugation, and washed
twice in PBS. Cells were incubated with a 15,000 dilution of
anti-ICAM-1 (C78.5 IgG) for 60 minutes in PBS-2% human
serum, washed, and incubated with a 1500 dilution of
fluorescein isothiocyanate-conjugated anti-mouse IgG. After washing, cells were analyzed in a FACScan (Becton
Dickinson, Mountain View, CA), using forward and orthogonal light scatter to select viable intact cells. Data for 10,000
cells were collected in list mode on a CONSORT 30 computer and subsequently analyzed using LYSYS software.
Fluorescence intensity was measured on a 4-decade log
amplifier. Mean channel fluorescence values were converted
to relative linear values for comparison of intensity differences by the method of Schmid et a1 (15), as previously
described (16).
Immunofluorescence studies. Endothelial cells were
grown on gelatin-coated 4-well Lab-Tek chamber slides (Nunc,
Naperville, IL), and treated with cytokines as described above.
Cells were fixed in 3.7% formalin in PBS, washed once with
RPMI 1640 and twice with PBS, and stained for ICAM-1 using
a 1 5 0 dilution of purified IgG ((38.5) as the first antibody and
a 1500 dilution of rhodamine-conjugated anti-mouse IgG as the
second antibody. Slides were examined on a Leitz fluorescence microscope, set for rhodamine excitation (546 pm) and
emission (570 pm), and photographs were taken. No staining of
the cells was observed when only the rhodamine-conjugated
anti-mouse or nonspecific mouse immunoglobulin was used.
Messenger RNA analyses. Endothelial cells were
treated with cytokine(s) for various periods of time, and total
RNA was prepared using guanidium isothiocyante and a
cesium chloride-trifluoroacetate gradient. Total RNA (10
pglwell) was then run on an RNA gel and transferred to a
Nytran nylon membrane. 32P-dCTP was used to label frag-
Figure 1. Morphologic analysis of human synovial microvascular
endothelial cells (HSE) from a patient with rheumatoid arthritis. a,
Low-power fluorescence micrograph of DiI-Ac-LDL seen in HSE
(bar = 40 pm). b, Phase micrograph of confluent culture of HSE
(third passage after sorting). The endothelial cells exhibit a somewhat flattened and elongated appearance, typical of many microvascular cells in culture (8,11,26) (bar = 10 pm). c, HSE plated on an
extracellular matrix (Matrigel; see Materials and Methods). Tubelike structures were rapidly formed. d, Same field as in c, viewed
under rhodamine fluorescent optics. Cellular actin was stained with
rhodamine phalloidin (bar = 52 pm).
ments of full-length ICAM-1 complementary DNA (1.9 kb;
kindly provided by Dr. Alan McClelland, Molecular Therapeutics Inc.), using a nick-translation system (GibcoBethesda Research, Grand Island, NY). Hybridization was
carried out at 42°C for 20 hours, in 50% formamide, 2 . 5 ~
Denhardt’s solution containing 100 pg/ml denatured salmon
sperm DNA, 0.1% sodium dodecyl sulfate (SDS), and 5x
SSC (1 x Denhardt’s solution = 0.02% polyvinylpyrrolidone,
0.02% Ficoll, and 0.02% BSA; 1 x SSC = 0.15M NaCl and
0.015M sodium citrate).
Immunoprecipitation. Cells were grown to confluence in T-75 flasks and transferred to growth factor-free
medium for 48 hours. Cells were incubated with o r without
cytokines for 20 hours, and were washed once with 12 ml
RPMI 1640 and then with 12 ml Dulbecco’s phosphate
buffered saline with 1 mM CaCl, and 1 mM MgCI, (PBS-Ca).
Cells were incubated with PBS-Ca, pH 8.0, at 4°C for 20
minutes with 15 p10.2 mg s-SHPP/ml DMSO (final concentration 3 pglml), 5 pl 1251 (camer free; final concentration 0.5
mCi/ml), 15 pl chloramine T (5 mglrnl H20-0.5M Na2HP0,,
pH 7.5; final concentration 75 pdml), 15 pl hydroxyphenylacetic acid (10 mg/ml H20; final concentration 150 dml), and 15
pl sodium metabisulfite (12 mg/ml; final concentration 180
d m l ) . The reaction was stopped by addition of 10 ml RPMI
1640, and the cells washed with an additional 10 ml RPMI 1640.
The cells were lysed in 1 mi of lysis buffer (0.15M
400 .
500 .
0.01 0.1
600 .
700 .
100 .
0.01 0.1
10 1001000
[ I L - l a ] (Ulml)
Figure 2. Effects of interleukin-la (IL-la) (A), tumor necrosis factor a (TNFa) (B),and interferon y (IFNy) (C) on the
up-regulation of intercellular adhesion molecule 1 , as determined by enzyme-linked immunosorbent assay, in human
umbilical vein endothelial cells (HUVE) (A) and human synovial microvascular endothelial cells (HSE; from rheumatoid
synovium) (0).HSE and HUVE were from the fourth passage from primary culture. Endothelial cells were pretreated with
the indicated concentration of cytokine for 24 hours. Data are expressed as the percentage of unstimulated HSE or HUVE
control values (considered to be 100%) and are the mean of 4 experiments.
NaCl, 20 mMTris HCl, pH 7.5,1.0% NP40,l mMPMSF, 10
d m l aprotinin, 1 mM CaCl,, and 10 mM NaF). The cell
lysates were spun at 14,000 revolutions per minute for 15
minutes in an Eppendorf centrifuge, and the supernatant was
used for immunoprecipitations. Purified anti-ICAM-1 IgG
(C78.5) was added to 35 pl of S aureus-precleared superna-
Table l. Synergy between cytokines in their induction of ICAM-I
expression in HSE and HUVE*
Treatment, cell type
T N F a (10 ng/ml)
IFN y (25 ng/ml)
TNFa (10 ng/ml)
+ I F N y (25 ng/ml)
Increase in MCF
tants along with protein A-agarose, and the mixture was
incubated overnight at 4°C. The suspensions were pelleted
by centrifugation at 14,000 rpm for 30 seconds. The pellet
was washed in 0.4M NaCl in lysis buffer, mixed for 5
minutes, and pelleted again. The pellets were then washed
twice in lysis buffer without extra salt, and the beads
resuspended in 50 pl sample buffer (2% SDS, 60 mM Tris,
pH 6.8, 2% P-mercaptoethanol, and 2 mM EDTA), loaded
on 4 1 2 % gradient SDS-polyacrylamide gels, and run at 20
mA for 2 hours. After running, the gels were fixed in 50%
methanol:lO% acetic acid for 1 hour. The gels were dried,
and autoradiographs developed by exposure to XAR-5 film
(Eastman Kodak, Rochester, NY) at -85°C for 2-6 days
with amplifying screens.
* Human synovial microvascular endothelial cells (HSE) or human
umbilical vein endothelial cells (HUVE) were treated with tumor
necrosis factor a (TNFa), interferon-y (IFN-y), or both in combination, for 24 hours. Cells were prepared for flow cytometric analysis
as described in Materials and Methods. Mean channel fluorescence
(MCF) values were converted from logarithmic units to linear values
by the method of Schmid et al (IS). ICAM-I = intercellular adhesion
molecule 1.
Identification of endothelial cells. Endothelial
cells selectively take up and degrade the acetylated
form of low density lipoprotein, a property that distinguishes them from fibroblasts, smooth muscle cells,
and pericytes, which are common contaminants in
primary cultures (7,8). Voyta and coworkers (7) developed a fluorescent lipophilic probe, DiI-Ac-LDL,
which allows the identification, by fluorescence, of the
cells that incorporate the probe. In the present study,
DiI-Ac-LDL was used in conjunction with FACS to
isolate endothelial cells from the mixed cell population
of the synovium.
Figure l b illustrates the morphology of a con-
fluent population of sorted HSE. A low-power view of
a nearly confluent population of HSE that were incubated with DiI-Ac-LDL and photographed under rhodamine fluorescence is shown in Figure la. HSE
formed tubes within 4-8 hours when plated to Matrigel
in the absence of exogenous cytokines or growth
factors (see Materials and Methods). This is another
property characteristic of endothelial cells that is not
shared by fibroblasts, pericytes, or smooth muscle
cells (11) (Figures lc and d). The HSE also exhibited
>99% positive staining for the other endothelial markers, Factor VIII-related antigen and UEA (results not
shown). There were no apparent differences in purity,
morphology, or phenotypic expression of the endothelial markers among the different HSE preparations
derived from normal or rheumatoid synovium, and,
unless otherwise indicated, all data reported herein
were derived from HSE prepared from rheumatoid
Effects of cytokines on ICAM-1 expression in
HUVE and HSE. Incubation of HSE with IL-1 or
TNFa led to only small increases in ICAM-1 expression detectable by ELISA, compared with findings in
similarly treated HUVE (Figure 2). The cytokinemediated regulation of ICAM-1 expression in HUVE
Time (hrs)
Figure 4. Time course of the up-regulation of intercellular adhesion
molecule 1 in response to TNFa (10 ng/ml) (A),IFNy(25 ng/ml) (O),
and TNFa + IFNy (0)in HSE from rheumatoid synovium, as
determined by enzyme-linked immunosorbent assay. Values are the
mean of 4 experiments. See Figure 2 for definitions.
and HSE was further evaluated by FACS analysis
(Table l), and the increment in ICAM-1 expression in
response to TNFa was again shown to be much
smaller in HSE than the response of HUVE. For example, a 24-hour incubation of HUVE with 10 ng/ml TNFa
resulted in a 12.7-fold increase in ICAM-1, compared
with only a 3.5-fold increase in similarly treated HSE
(Table 1). Incubation with TNFa combined with IFNy
led to a much greater increment in both HUVE and HSE
ICAM-1 expression (Table l), and IFNy in a concentra-
Table 2. Effects of T N F a and IFNy on ICAM-1 expression in
normal and rheumatoid synovial endothelial cell preparations*
ICAM-1 expression (OD units)
lFNg Concentration (nglrnl)
Figure 3. Dose-dependence of the effects of IFNy (IFNg), in the
absence (A) or presence ( 0 )of 10 ng/ml TNFa, on the up-regulation
of intercellular adhesion molecule I in HSE from rheumatoid
synovium. Values are the mean of 4 experiments. See Figure 2 for
other definitions.
(10 ngiml)
(25 ng/ml)
TNFa +
0.24 ? 0.01 0.31
0.37 2 0.01 0.40
0.37 ? 0.01 0.46
0.01 0.35 f 0.01 0.73 & 0.04t
0.01 0.50 f 0.07t 0.82 5 0.02t
0.03 0.45 ? 0.01 0.86 C 0.02t
* Cells were incubated with the indicated cytokine or cytokine
combination for 16 hours, fixed, and ICAM-1 expression was
determined by enzyme-linked immunosorbent assay as described in
Materials and Methods. Values are the mean f SEM optical density
(OD) units, after subtraction of background. NSE = human synovial
endothelial cells derived from normal synovium; HSE 1 and HSE 2
are from 2 rheumatoid arthritis patients (other than the patient
whose results are shown in Figure 1). See Table 1 for other
t P < 0.05 versus unstimulated control (basal ICAM- I expression),
by one-way analysis of variance followed by the Bonferroni modified t-test.
Table 3. Synergistic effect of IFNy on cytokine-induced ICAM-1
expression in HSE*
% of control ICAM-1 expression
IL-1 (10 unitshl)
TNFa (10 ng/ml)
TNFP (12 ng/ml)
1L-4 (10 ng/ml)
LPS (1 pglrnl)
PMA (100 nM)
IL-I (10 unitslml)
(10 ng/ml)
+IFNy (25 ng/ml)
168 t 2 t
138 2 t
138 t 2"
115 f 6"
107 t 3
160 t 4't
138 t 2 t
+ TNFo
156 f 15$
227 f 7$
431 2 16$
227 f 15$
227 f 1%
278 f 8$
114 f 9
430 5 16$
* HSE were incubated with the indicated cytokines for 16 hours,
and ICAM-1 expression was determined by enzyme-linked immunosorbent assay. Values are the mean t SEM. IL-1 = interleukin-1;
LPS = lipopolysaccharide; PMA = phorbol myristate acetate. See
Table 1 for other definitions.
t P < 0.05 versus untreated control.
f P < 0.05 versus untreated control and versus cells treated with the
same cytokine in the absence of I F N y
tion as low as 1 ndml was able to significantly augment
the actions of TNFa (Figure 3).
A representative example of the time course of
ICAM-1 expression in response to TNFa (10 ng/ml),
IFNy (25 ng/ml), or both cytokines is presented in
Figure 4. Increased ICAM-1 expression was observed
as early as 4 hours after addition of the cytokines, and
maximal levels were attained between 8 hours and 16
hours. Elevated ICAM- 1 expression was maintained
as long as 48 hours after the addition of cytokine. At
every time point, the combination of TNFa + IFNy
was significantly more effective than either cytokine
alone. In addition, the increased ICAM- 1 expression
with TNFa or IFNy reached maximal levels by 8
hours, whereas the expression of ICAM-1 in response
to the combination of cytokines continued to increase
between 8 and 16 hours. The synergistic effect of the
cytokines was observed in all HSE isolates prepared
from 2 additional RA patient specimens and 1 normal
control specimen (Table 2).
The synergistic actions of IFNy on HSE cells
were not specific to TNFa. As shown by ELISA
(Table 3), coincubation with IFNy together with IL-1,
TNFP, IL-4, or LPS also augmented ICAM-1 expression. In contrast, the stimulatory actions of the protein
kinase C activator phorbol myristate acetate were reduced by coincubation with IFNy. Treatment with IL-1
(1 unitlml) in combination with TNFa (10 ng/ml) did not
result in the synergism seen with IFNy (Table 3).
To evaluate the cell surface distribution of
ICAM-1, the cytokine-treated cells were examined by
fluorescence microscopy. Confirming the findings of
the ELISA and FACS studies, TNFa alone increased
HUVE ICAM-1 expression, but had only a minimal
effect in the HSE. Furthermore, coincubation with
IFN y and TNFa again resulted in a marked synergistic
effect on the expression of ICAM-I in both HSE and
HUVE. The distribution of ICAM-1 in the HSE stimulated with TNFa + IFNy was qualitatively similar to
that of HUVE incubated under identical conditions
(Figure 5).
Immunoprecipitation of ICAM-1. ICAM- 1 was
immunoprecipitated with antiserum (C78.5) directed
against the first domain of this immunoglobulin generelated protein. As shown in Figure 6, this antiserum
precipitated a single molecular weight species (M,90
Figure 5. Immunoreactive intercellular adhesion molecule 1
(ICAM-1) expression in HSE (a-d) and HUVE (e-h). Cells were
incubated for 24 hours in the absence of cytokine (a and e) or in the
presence of TNFa (10 ng/ml; b and f), IFNy (25 ng/ml; c and g), or
TNF + IFNy (d and h), fixed, and stained for ICAM-1 as described
in Materials and Methods. (Bar = 48 p n . ) See Figure 2 for other
kd), which appeared to be identical in both HSE and
HUVE stimulated with TNFa + IFNy.
Effects of cytokines on ICAM-1 mRNA in HUVE
and HSE. To determine the effects of cytokines on
ICAM-1 mRNA, we measured the steady-state levels
of mRNA by Northern blot analysis (Figure 7). The
results shown in Figures 7 and 8 are from studies using
somewhat low concentrations of TNFa (1 ng/ml) and
IFNy (5 ng/ml); however, similar observations were
obtained using higher concentrations (TNFa 10 ng/ml,
IFNy 25 ng/ml) (Gerritsen M et al: unpublished observations). In HUVE, at 24 hours, TNFa alone or in
combination with IFN y induced the accumulation of
ICAM-1 mRNA, while IFNy alone increased mRNA
levels only slightly. In contrast, in HSE, TNFa alone
induced a relatively small amount of ICAM-1 mRNA,
while IFNy alone increased ICAM-1 mRNA to a
greater extent and, together with TNFa, resulted in
greatly enhanced ICAM-1 mRNA levels. The time
courses of the effects of TNFa and IFNy in combination on ICAM- 1 mRNA in HUVE and HSE are shown
in Figure 8. The time courses were similar in the two
cell types, although the HUVE demonstrated a more
pronounced peak in steady-state mRNA levels at 8
hours, compared with HSE. In both cell types,
ICAM-1 mRNA was observed as early as 2 hours,
attained maximal levels between 4 and 8 hours, and
then declined by 24 and 48 hours.
Intercellular adhesion molecule 1 (ICAM-1) messenger
RNA (mRNA) regulation in HSE and HUVE. Cell monolayers were
stimulated with TNFa (TNFa; 1 ng/ml) or IFNy (gIFN; 5 ng/ml) or
both for 20 hours (see Materials and Methods), and total mRNA was
extracted and analyzed for ICAM-1 levels by Northern blot analysis. Ten micrograms of total RNA was applied to each lane. The 28s
ribosomal RNA subunit intensity on ethidium bromide staining is
shown across the bottom. See Figure 2 for other definitions.
Figure 7.
Figure 6. Intercellular adhesion molecule 1 immunoprecipitation
from HUVE and HSE. Confluent cell monolayers were treated with
TNFa (10 ndml) + IFNy (25 ndml) for 24 hours, surface-labeled
with 1251, solubilized, immunoprecipitated, electrophoresed, and
autoradiographed as described in Materials and Methods. Control =
HSE cells stimulated and immunoprecipitated with nonimmune
antiserum. See Figure 2 for definitions.
The major conclusion of the present study is
that IFNy appears to play a critical role in the regulation of ICAM- 1 expression in synovial endothelial
cells. Incubation with cytokines such as TNFa or IL-1
provides a relatively ineffective stimulus for ICAM- 1
expression in these cells, in contrast to both published
data on HUVE as well as the observed responses with
HUVE in the present study. In addition, IFN y alone
was shown to be as effective, or nearly as effective, as
T N F when
its actions on HSE were tested, whereas
in HUVE, IFNy is much less potent than TNFa.
These observations from the initial samples tested
were reproduced in cells derived from 2 additional RA
Figure 8. Time course of intercellular adhesion molecule 1 (ICAM-1) messenger RNA (mRNA) expression
in HSE and HUVE. Cell monolayers were incubated with TNFa (1 ng/ml) and IFNy (5 ng/ml) for the
indicated time periods. Total mRNA was extracted, and 10 pg was applied to each lane and analyzed for
ICAM-I levels by Northern blot analysis. The 28s ribsomal RNA subunit intensity on ethidium bromide
staining is shown across the bottom. See Figure 2 for other definitions.
patients and 1 normal subject. The cytokine synergism
requirements for ICAM- 1 expression were confirmed
at several levels, including population distribution by
FACS, immunofluorescence staining, and mRNA
The size of the ICAM-1 in HSE appears to be
identical to that in HUVE, and the size of the ICAM-1
mRNA also appears to be the same for both cell types.
The differences in responsiveness to TNF were unlikely to be due to differences in passage number since
(a) higher-passage HUVE (passages 8-10) still demonstrated 6-10-fold increases in ICAM-1 expression in
response to TNFa, and (b) preliminary experiments
with low-passage HSE (second passage after sorting)
also indicated no significant up-regulation of ICAM-1
in response to 1&100 ng/ml of TNFa (Gerritsen M et
a1: unpublished observations).
One explanation for the relatively poor response to IL-1 or TNFa in HSE could be a comparatively low receptor density for these cytokines, compared with HUVE. The actions of IFNy could
conceivably be mediated by an increase in cytokine
receptors. IFNy has been shown to regulate the expression of TNFa receptors in certain tumor cells
(17,18). However, Johnson and Pober (19) showed that
IFNy had no effect on TNF receptor number in human
endothelial cells. Coincubation with IFNy augmented
the responses of HSE to a number of cytokines,
suggesting that IFN y may affect a shared mechanism
common to all these stimuli. For example, IFNy could
augment a rate-limiting step in the transcriptional
activation of the ICAM-1 gene. The promoter region of
ICAM-1 contains AP-1 as well as NFKP consensus
sequences (20,21), and it is possible that IFNy alters
either the concentration and/or the activity of one or
more transactivating factors which bind the cis-acting
regulatory elements, leading to augmented gene expression.
The synergistic action between IFNy and
TNFa on the ability of various cell types to express
ICAM and to bind inflammatory cells, including polymorphonuclear cells, T cells, and monocytes, is well
documented (5,22-24). In contrast to HUVE, however, HSE did not show a large increment in ICAM-1
expression in response to either IL-1 or T N F a alone,
but appeared to require the presence of IFNy. The
pathophysiologic significance of the actions of IFN y
on HSE is unclear since rheumatoid synovial fluid
contains only low levels of I F N y compared with other
cytokines (25,26). However, even very low levels of
IFNy (1 ng/ml) were capable of augmenting the response to T N F a , and thus the levels in the synovial
fluid may be sufficient. Additionally, since IFNy also
augmented the expression of ICAM-1 in response to
other inflammatory cytokines (IL-4, TNFP, IL-l), the
ability of this interferon to act synergistically with
other local mediators may be of particular importance
in maintaining a prolonged, proadhesive state in synovial endothelium.
These observations clearly indicate functional
differences in the cytokine responsiveness of endothelial cells derived from different vascular beds. Differential regulation of adhesion molecule expression may
provide insight into yet another component of the
complex mechanism whereby specific populations of
leukocytes can target select vascular beds in response
to different pathogenic stimuli. These observations
also serve to further emphasize that, in studying a
specific vasculopathologic condition, the endothelial
cells evaluated should be derived from the organ of
interest to the given condition; extrapolations from
studies with large-vessel endothelial cells such as the
HUVE may be invalid or overgeneralized.
In conclusion, endothelial cells derived from
human rheumatoid synovium were demonstrated to
express ICAM-1 in response to inflammatory cytokines. The ICAM-I expressed by the HSE cells appears to be identical to that expressed by HUVE,
based on immunoprecipitation and Northern blot analyses. However, the regulation of ICAM-1 expression
in HSE differs from that in HUVE. TNFa, a potent
inducer of ICAM-1 expression in HUVE (1,6,31) as
well as in other cell types ( 6 ) , was a relatively ineffective stimulus in HSE. Optimal ICAM-1 expression in
HSE appeared to require the combination of IFN y and
TNFa. This observation is similar to earlier reports on
human synovial fibroblasts (22) and keratinocytes (24),
in which it was found that cytokine synergism was
required for optimal ICAM- 1 expression. The ability
of endothelial cells to differentially regulate common
adhesion molecules may play an important role in
directing specific immune responses to select vascular
60 1
The authors express their appreciation to John Flannagan and Kristen Huwiler for their technical assistance.
The authors are grateful to Drs. G. Rayan (Oklahoma City,
OK) and A. Freemont (Manchester, UK) for their invaluable
assistance in tissue procurement.
1. Dustin ML, Springer TA: Lymphocyte function associ-
ated antigen-1 (LFA-1) interaction with intercellular
adhesion molecule-1 (ICAM-1) is one of at least three
mechanisms for lymphocyte adhesion to cultured endothelial cells. J Cell Biol 107:321-331, 1988
2. Zimmerman GA, McIntyre TM: Neutrophil adherence
to human endothelium in vitro occurs by CDwl8 (Mol
Mac-l/LFA-1/GP 150,95) glycoprotein dependent and
independent mechanisms. J Clin Invest 81531-537,1988
3. Pohlman TH, Stanness KA, Beatty PG, Ochs HD,
Harlan JM: An endothelial cell surface factor(s) induced
in vitro by lipopolysaccharide, interleukin-1 and tumor
necrosis factor increases neutrophil adherence by a
CDw 18-dependent mechanism. J Immunol 136:45484553, 1986
4. Bevilacqua MP, Pober JS, Wheeler ME, Cotran RS,
Gimbrone MA Jr: Interleukin 1 acts on cultured human
vascular endothelium to increase the adhesion of polymorphonuclear leukocytes, monocytes and related cell
lines. J Clin Invest 76:2003-2011, 1985
5 . Doukas J, Pober JS: IFN-7 enhances endothelial activation induced by tumor necrosis factor but not IL-1. J
Immunol 145:1727-1733, 1990
6. Springer TA: Adhesion receptors of the immune system.
Nature 346:425434, 1990
7. Voyta JC, Via DP, Butterlield CE, Zetter BR: Identification and isolation of endothelial cells based on their
increased uptake of acetylated low density lipoprotein. J
Cell Biol 99:2034-2040, 1984
8. Gerritsen ME, Carley WW, Milici AJ: Microvascular
endothelial cells: isolation, identification and cultivation, Advances in Cell Culture. Vol. 6. Edited by K
Maramorosch, GH Sato. San Diego, Academic Press,
9. 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
10. Del Vecchio PJ, Smith JR: Expression of angiotensin
converting enzyme activity in cultured pulmonary artery
endothelial cells. J Cell Physiol 108:337-345, 1981
11. Chung-Welch N, Patton WF, Yen-Patton A, Hechtman
H, Shepro D: Phenotypic comparison between mesothelial and microvascular endothelial cell lineages using
conventional endothelial cell markers, cytoskeletal pro-
tein markers and in vitro assays of angiogenic potential.
Differentiation 42:44-53, 1989
12. Holthofer H, Virtanen I, Karineimi AL, Hormia M,
Linder E, Miettinen A: Ulex europaeus I lectin as a
marker for vascular endothelium in human tissues. Lab
Invest 47:60-66, 1982
13. Denizot F, Lang R: Rapid colorimetric assay for cell
growth and survival: modifications to the tetrazolium
dye procedure giving improved sensitivity and reliability. J Immunol Methods 89:271-277, 1986
14. McClelland A, DeBear J, Yost SC, Meyer AM, Marlor
CW, Greve JM: Identification of monoclonal antibody
epitopes and critical residues for rhinovirus binding in
domain 1 of ICAM-1. Proc Natl Acad Sci U S A 88:
7993-7997, 1991
15. Schmid I, Schmid P, Giorgi JV: Conversion of logarithmic channel numbers into relative linear fluorescence
intensity. Cytometry 9533-538, 1988
16. Rothman BL, Blue ML, Kelley KA, Wunderlich D,
Mierz DV, Aune TM: Human T cell activation by OKT3
is inhibited by a monoclonal antibody to CD44. J Immuno1 147:2493-2499, 1991
17. Aggarwal BB, Seesalu TT,Hass PE: Characterization of
receptors for human tumor necrosis factor and their
regulation by yinterferon. Nature 318:665-667, 1985
18. Tsujimoto M, Yip YK, Vilcek J: Interferon-gamma
enhances expression of cellular receptors for tumor
necrosis factor. J Immunol 136:2441-2444, 1986
19. Johnson DR, Pober JS: Tumor necrosis factor and
immune interferon synergistically increase transcription
of HLA class I heavy and light chain genes in vascular
endothelium. Proc Natl Acad Sci U S A 8735183-5187,
Degits K, Lian-Jie L, Caughman SW: Cloning and
characterization of the 5' transcriptional regulatory region of the human intercellular adhesion molecule 1
gene. J Biol Chem 266: 1402414030, 1991
Voraberger G, Schafer R, Stratowa C: Cloning of the
human gene for intercellular adhesion molecule 1 and
analysis of its 5' regulatory region: induction by cytokines and phorbol ester. J Immunol 147:2777-2786, 1991
Krzesicki RF, Fleming WE, Winterrowd GE, Hatfield
CA, Sanders ME, Chin JE: T lymphocyte adhesion to
human synovial fibroblasts: role of cytokines and the
interaction between intercellular adhesion molecule 1
and CDlldCD18. Arthritis Rheum 34:1245-1253, 1991
Wankowicz Z, Megyeri P, Issekutz A: Synergy between
tumor necrosis factor-a and interleukin 1 in the induction of polymorphonuclear leukocyte migration during
inflammation. J Leukoc Biol 43:349-356, 1988
Barker JNWN, Sarma V, Mitra RS, Dixit VM, Nickoloff
BJ: Marked synergism between tumor necrosis factor
and interferon y in regulation of keratinocyte-derived
adhesion molecules and chemotactic factors. J Clin
Invest 85:605-608, 1990
Egelund T , Lund H: Immunoregulatory lymphokines in
rheumatoid joints. Scand J Immunol 25:101-106, 1987
Firestein GS, Zvaifler NJ: Peripheral blood and synovial
fluid monocyte activation in inflammatory arthritis. 11.
Low levels of synovial fluid and synovial tissue interferon suggest that ?interferon is not the primary macrophage activating factor. Arthritis Rheum 30:864-87 1 ,
Без категории
Размер файла
1 052 Кб
expressions, endothelial, synovium, molecules, culture, regulation, derived, intercellular, human, adhesion, rheumatoid, cells
Пожаловаться на содержимое документа