From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
Tumor Angiogenesis Is Accompanied by a Decreased Inflammatory Response
of Tumor-Associated Endothelium
By Arjan W. Griffioen, Cora A. Damen, Geert H. Blijham, and Gerard Groenewegen
We previously showed that endothelial cells (EC) from the
vasculature of human solid tumors have a decreased expression of intercellular adhesion molecule-I (ICAM-1)and ICAM2 as compared with normal tissue EC. This effect is explained
by EC exposure t o angiogenic factors. It is known that upregulation of endothelial adhesion molecules (EAM) is a sign of
EC activation in inflammatory responses. We therefore
tested the effect of angiogenic factors on upregulation of
EAM on tumor EC and human umbilical vein EC (HUVEC) by
proinflammatory cytokines. Incubation of tumor-derived EC
in tumor necrosis factor a (TNFa) did result in expression
levels of only 20% of the level of similarly treated normal
tissue-derived EC. Pretreatment of HUVEC with 10 ng/mL
basic fibroblast growth factor (bFGF) for 3 days, before
TNFa- or interleukin-la (IL-la) stimulation, resulted in
ICAM-1 levels of only 30% t o 60% of cells without pretreatment. Also, the induction of vascular EC adhesion molecule1 (VCAM-1) and E-selectin by TNFa was significantly inhibited by prior exposure t o bFGF. Vascular endothelial growth
factor had similar but less prominent effects. The effect of
transforming growth factor-p and IL-8 was studied as well.
The functional relevance of the finding of a decreased EC
inflammatory response was confirmed by adhesion assays.
Our results show that tumor angiogenesis induces EC anergy. This may serve as a tumor-protecting mechanism by
impairing the development of an efficient leukocyte infiltrate
in tumors.
0 1996 b y The American Society of Hematology.
E
and RNA 1 e ~ e l .This
l ~ suggested an angiogenesis-accompanied escape mechanism from host defense by interference
with the mechanism of effector cell extravasation. In the
present study, we analyzed the effect of angiogenic factors
on the inflammatory response and activation of EC. We provide evidence that tumors have protective mechanisms, operative via angiogenic factors, against increased exposure to
leukocytes not only by downregulation of EAM expression,
but also by a decreased response to inflammatory cytokines.
NDOTHELIAL CELLS (EC) normally exist in the vasculature as quiescent cells. During inflammation, they
are activated to upregulate a number of genes including the
genes of adhesion molecules such as intercellular adhesion
molecule- 1 (ICAM- I), vascular cell adhesion molecule-I
(VCAM-l), and E-selectin. The regulatory mechanisms involved in the activation proces have not been fully elucidated. Cellular immunity against tumor cells requires adhesion molecules on vascular endothelium that mediate arrest
and extravasation of leukocytes into tumors.’” Treatment of
cancer patients with adoptive imm~notherapy~,~
is dependent
on the selective accumulation of leukocytes in the tumor
guided by endothelial adhesion molecules (EAM). Expression of EAM is controlled by cytokines such as tumor necrosis factor a (TNFa), interleukin-I (IL-l), and interferon y
( m y ) .These cytokines facilitate leukocyte adhesion to EC
and extravasation into extravascular tissues by inducing an
enhanced expression of ICAM- 1 , VCAM- 1, and E-selectin.l.6-8 Transforming growth factor P (TGFP), in contrast,
reduces adhesion of resting and activated lymphocytes to
EC, although without affecting expression of ICAM-1 and
VCAM- 1 .9,‘0
Angiogenesis, the formation of new microvasculature
from pre-existing vessels by capillary sprouting, is a process
occurring in tumors and wound healing.”,’’ This process,
regulated by angiogenic factors, expands the endothelial surface and thus the possible interaction between leukocytes
and vessel wall. In tumors, these interactions might be detrimental to the tumor itself. However, because cytotoxic effector cells form a major threat to the tumor, escape mechanisms avoiding infiltration of leukocytes into the tumor may
have evolved. Recent reports suggest that protective mechanisms, controlled by the tumor cells, are operative to prevent
infiltration of leukocytes.’3-’6Although altered expression
of certain tumor endothelium-associated antigens has been
described, ”-”insight in regulation of adhesion molecule expression and endothelial functions by angiogenic factors is
largely lacking. In a recent report, we described the decreased expression of ICAM-l and ICAM-2 on tumor-associated EC. We observed that exposure of normal resting
endothelial cells to basic fibroblast growth factor (bFGF)
decreases the expression of these EAh4 both at the protein
Blood, VOI 88,NO 2 (July 15). 1996: pp 667-673
MATERIALS AND METHODS
Monoclonal antibodies and cytokines. Antiadhesion molecule
antibodies M E M l l l anti-ICAM-UCD54, lG11B1 anti-VCAM-l/
CD106, ENAl anti-E-selectin/CD62E, and EN4 EC-marker were
from Sanbio (Uden, The Netherlands). The polyclonal antibody
against HLA class I was from Dako (Glostrup, Denmark). The cytokines TNFa, IL- la, IL-8, bFGF, vascular endothelial growth factor
(VEGF), and TGFP were all purchased from R&D Systems (Abington, UK). IFNy was obtained from Schering (Kenilworth, NJ).
Cells and cultures. Human umbilical vein derived endothelial
cells (HUVEC) were harvested from normal human umbilical cords
by perfusion with 0.125% trypsin as described previously.z’ Cells
were cultured in fibronectin (FN)-coated tissue culture flasks in
RPMI-1640 with 20% human serum (HS), supplemented with glutamine and antibiotics. Confluent HUVEC cultures were subculture
1.3. After reaching confluence in the second passage, EC were
seeded in flatbottomed tissue culture plates and used for adhesion
From the Department of Internal Medicine and Medical Oncology, Laboratov of Angiogenesis Research, University Hospital
Utrecht, Utrecht, The Netherlands.
Submitted January 17, 1996; accepted March 18, 1996.
Supported by the Dutch Research Foundation ‘De Drie Lichten”
and by the “Nijbakker Morra Stichting.”
Address reprint requests to Arjan W. Grifloen, PhD, Department
of Internal Medicine and Medical Oncology, Laboratory for Angiogenesis Research, F.02.126, University Hospital Utrecht, Heidelberglaan 100, 3584 CX Utrecht, The Netherlands.
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
“advertisement” in accordance with 18 U.S.C. section 1734 solely to
indicate this fact.
0 1996 by The American Society of Hematology.
OOO6-4971/96/8802-0002$3.OO/O
667
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
668
GRlFFlOEN ET AL
experiments (96-well plate, 5,000 cellslwell) and immunotluorescence (6-well plate, 2 X IO' cellslwell) when the cells reached 80%
to 90% confluence. Peripheral blood mononuclear cells (PBMC)
were prepared by density gradient centrifugation on Ficoll-Isopaque
(Pharmacia, Uppsala, Sweden) of heparinized peripheral blood from
healthy adult donors. Phytohemagglutinin (PHA) blasts were prepared by culturing 2 X lo6 PBMC/well (24-well plate) with 10 p&/
mL PHA. After 3 days, cells were used for adhesion assays. The
LFA-I -expressing (promyelocytic) cell line HL60 was also used in
the adhesion assays.
Immunojfuorescence. Immunofluorescence using indirect phycoerythrin (PE)-conjugated reagents required three separate incubations. A total of 1 X lo5 EC (harvested by incubation in 10 mmol/
L EDTA in phosphate-buffered saline [PBS]) were washed, resuspended in 20 pL appropriately diluted monoclonal antibody and
incubated for I hour on ice. Subsequently, cells were washed two
times in PBS/O.I% bovine serum albumin (BSA) and incubated for
another 30 minutes with hiotinylated goat-antimouse Ig (Dako).
After another two washings, cells were incubated with streptavidinphycoerythrin conjugate (Dako). Stained cells were analyzed on a
FACScan flow cytometer. Of each sample, forward scatter (FSC),
side scatter (SSC), green fluorescence (525/20 nm), and red fluorescence (640/20 nm) signals of 5,000 cells were recorded. Data analysis was performed using PClysys software (Becton Dickinson,
Mountain View, CA). Statistical significance of observed differences
was determined using the Student's t-test.
Flow cytometric analysis of tumor-derived EC. Tumor and normal tissues, obtained from surgical procedures as first treatment of
renal cell carcinoma, were processed immediately after resection
under sterile conditions. An ethical review committee approved informed consent procedure was used to obtain the relevant materials.
Normal and tumor tissue was minced and treated for 1 hour with I
mg/mL collagenase and 1 mg/mL dispase (both from Sigma Chemical Co, St Louis, MO) in PBS. Tissue was subsequently sieved and
single cells were washed and resuspended in RPMI-1640/20% HS.
Enzyme treatment and sieving was repeated and cells were collected
and cultured briefly in an FN-coated culture flask in RPMI-I640/
20%HS. Then, nonadhering cells and debris were decanted and adhering cells were stained with antiadhesion molecule antibodies and
fluorescein isothiocyanate (F1TC)-conjugated goat antimouse Ig
(Dako). Identification of EC was performed by subsequent indirect
staining with EC-specific antibody EN4-biotin and streptavidin-PE.
Additional identification of EC was performed on basis of EC-specific FSC/SSC characteristics. Previous studies indicated that this
procedure identified the cells that stain after DiI-acetylated-LDL2'
incubation. ''
Adhesion assay. Cultures of untreated or cytokine-treated EC in
96-well plates were used for adhesion assays. Before the addition
of 4 nglmL TNFa, the EC were counted to correct for differences
in numbers of EC. HL-60- or PHA-activated peripheral blood T
lymphocytes were washed twice and subsequently added (in quadruplo, 5 x IO4 to 2 x IO' cells/well) to the EC cultures.'' After an
adhesion step of 1 hour at 37"C, unadhered cells were removed by
washing the wells gently with prewarmed culture medium using a
multichannel pipette. Adhered cells were enumerated using an inverted microscope. Adhesion is expressed as the number of adhered
leukocytes/high power field standardized for the number of EC.
RESULTS
Suppressed TNFa-induced EAM expression after HUVEC
exposure to bFGF. To study the effect of angiogenic factors on the inflammatory response of EC on stimulation with
proinflammatory cytokines, HUVEC were cultured for 3
days with 10 ng/mL bFGF. Subsequently, cells were acti-
r
1
500
60
400
50
-n
E
0
300
40
u
30
200
0
3
4
30
3
6
= 20
100
10
"
I
I
0
6
I
16
"
40
TNFa (hours)
Fig 1. Upregulation of EAM is inhibited by bFGF. HUVEC were
pretreated for 3 days with 10 nglmL bFGF and subsequently activated with 4 nglmL TNFa. The time course of TNFa-induced ICAM1 (0
and H; right vertical axis), E-selectin (V and V; left vertical axis),
and e; left vertical axis) upregulationwas measured
and VCAM-1 (0
in cells without pretreatment (solid symbols) and after bFGF incubation (open symbols). The mean fluorescence intensity is shown of
one experiment representative of three. The peak levels of TNFainduced adhesion molecules on bFGF-treated EC were only 3096 to
60% of levels obtained without bFGF.
vated with 4 ng/mL TNFa and expression of EAM ICAM1 , VCAM- I , and E-selectin was measured. We found that
upregulation of all three EAM on these cells was markedly
inhibited as compared with cells activated with TNFa without prior incubation with bFGF (Fig 1). Peak levels of
ICAM-1 on bFGF-pretreated cells, reached at 16 hours after
TNFa activation, were only 30% to 60% of levels obtained
without bFGF pretreatment. Induction of VCAM-1 and Eselectin, which reaches maximal levels at 6 to 8 hours after
activation with TNFcu, was also significantly reduced by
prior exposure to bFGF. It was excluded that this observed
reduced expression was dependent on the TNFa concentration because concentrations of up to 40 ng/mL TNFa showed
the same results (data not shown). The time-dependent kinetics of EAM expression by TNFa remained the same after
pretreatment of EC with bFGF (Fig 1).
Functional relevance for these findings was studied in
adhesion assays. Leukocyte adhesion was measured in EC
that were activated for 16 hours with 4 ng/mL TNFa. These
experiments were performed with untreated and bFGF-pretreated HUVEC. For the adhesion assays, two different leukocyte cell types expressing the major ICAM-1 ligand LFA1, PHA-activated peripheral blood T lymphocytes and the
promyelocytic HL-60 cell line, were selected. We found
that spontaneous adhesion of HL-60 cells (Fig 2) and PHAactivated peripheral blood T cells (data not shown) to EC is
inhibited by pretreatment of EC for 3 days with 10 ngl
mL bFGF. Although leukocyte adhesion is upregulated by
activation of EC with TNFa, it is shown in Fig 2 that the
TNFa-induced adhesion is efficiently suppressed by pretreatment with bFGF.
Inhibitory effect of bFGF on EAM upregulation by other
injammatory cytokines. We subsequently studied whether
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
THE INFLAMMATORY RESPONSE OF TUMOR ENDOTHELIUM
70
8o
.-0
c
1
angiogenic factors on the inflammatory response of EC is
summarized in Table 1. As shown, VEGF had largely similar
but less prominent effects on TNFa-induced EAM expression as bFGF, with the notable exception of the IL-lainduced expression of VCAM-1. TGFP also inhibited to
some extent the effects of TNFa-induced EAM expression
but not on the EAM induction by IL-la and IFNy. Whereas
ICAM-1 upregulation by IL-la is inhibited by VEGF, as is
the case for bFGF, the induction of VCAM- 1 and E-selectin
by IL-la remains unaltered by VEGF. IL-8 had no significant inhibitory effects in these studies. Because IFNy does
not upregulate all EAM, the effect of angiogenic factors on
LFNy stimulation is given only for ICAM-1. It appeared that
an inhibitory effect was detected only after bFGF and VEGF
and not after TGFP and IL-8. In these experiments, the effect
on upregulation of HLA class I by IFNy was also studied.
T
50
40
u
30
20
10
0
669
1
@-
100
zn
Fig 2. Upregulationof leukocyte EC adhesion by TNFa is inhibited
by bFGF. HL-60 promyelocytic cell line was adhered for 1 hour on
resting or preactivated (10 ng/mL bFGF for 3 days) EC. Where indicated, EC were cultured for the last 16 hours with 4 nglmL TNFa.
The adhesion of quadruplicate cultures is expressed as described in
the Materials and Methods (mean 2 SEW. The negative control is
performedby paraformaldehydefixation before adhesion. The results
of one experiment representative of four are shown. "Significantly
different from values without bFGF ( P < .01).
I
-cn
Q)
0
Y-
O
,
*f
L
.
Q)
.-.'
.
I
n
E
3
c
0
1
the inflammatory response, ie, the upregulation of EAM,
induced in vitro by other cytokines such as IL-la and IFNy
is also influenced by pretreatment with bFGF. In Fig 3,
FACS analysis experiments with HUVEC are shown in
which the effect of pretreatment with 10 ng/mL bFGF is
measured on the upregulation ICAM-1 by IL-la and IFNy.
It was observed that bFGF also effectively inhibits the upregulation of endothelial ICAM-1 by both IL-la (Fig 3A) and
IFNy (Fig 3B).
Figure 4 summarizes the inhibitory effect of bFGF on the
EAM regulation by TNFa and IL-la. bFGF inhibits the
TNFa-induced upregulation of all three EAM tested. IL-la
has been shown to upregulate expression of ICAM-1,
VCAM-1, and E-selectin. Pretreatment of EC with bFGF
inhibited the upregulation of ICAM-1 and VCAM-1 by ILl a , but the expression of E-selectin was inducible to comparable levels in nontreated and bFGF pretreated EC (Fig 4).
Effect of other angiogenic factors on the in$a"atory
response of EC. In a next set of experiments, angiogenic
factors other than bFGF were tested for their influence on
the inflammatory response of EC. VEGF was chosen because
of its mitogenic effect on EC. TGFP and IL-8,which has
recently been described to elicit angiogenic activity, were
tested as cytokines without mitogenic potential. Flow cytometric measurement of ICAM-1, VCAM-1, and E-selectin
was performed on cultured HUVEC. The influence of these
100
I4
I
B
cn
I
I
Q)
0
YI
0
. .
L
Q)
n
I
E
..
3
c
0
1
I4
ICAM-1 fluorescence
Fig 3. Upregulation of CAM-1 by IL-la and IFNy is decreased by
bFGF. HUMC were cultured for 3 days in the presence of 10 ng/mL
bFGF. IL-la (100 UlmL) or IFNy (300 U/mL) were added for the last
16 hours of the culture. CAM-1 fluoresccHlce intensity was measured
in unstimulated cells (-. -1, IL-la-stimulated ce~lls(A), IFNystimulated (B) cells (- 4, and cells cultured with bFGF and IL-la or lFNy
(-1. (. * Conjugate controls.
-
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
GRlFFlOEN ET AL
670
X inhibition by bFGF
0
25
50
75
100
I
1
1
ICAM-1
TNFa
VCAM-1
E-adactin
ICAM-1
lLla
VCAM-1
E-rrlectln
Fig 4. Inhibition of inflammatory cytokine-mediated upregulation
of EAM by bFGF. HUVEC were cultured with 10 ng/mL bFGF. For the
last 16 hours of the culture, TNFa (n = 7) and IL-la (n = 4) were
added. Inhibition of the CAM-1, VCAM-1, and E-selectin upregulation
is measured and the percentage of inhibkion (mean f SEMI of mean
fluorescence intensity values is shown. The percentage of inhibition
of the upregulation of €AM is calculated as follows: 100 x (1 - [a bl/[c - bl), in which a is the fluorescence after TNFa incubation, b
is the constitutive expression, and c is the fluorescence after bFGF
end TNFa incubation.
We observed a similar significant inhibitory effect of 61%
(?14%, P < .01) and 19% (?7%, P < .01) after bFGF and
VEGF, respectively.
To identify the functional impact of these findings, adhesion assays were performed in which cultured EC are pretreated for 3 days with the indicated angiogenic factors. The
last 16 hours of this incubation was performed in the presence of 4 ng/mL TNFa. Whereas only bFGF and not VEGF,
IL-8, and TGFP was found to inhibit the adhesion of leukocytes to unstimulated EC, bFGF and VEGF both significantly
inhibited TNFa-induced adhesion of both leukocyte cell
types. The inhibitory effect of VEGF was less prominent
than the effect of bFGF (Fig 5). These results are compatible
with the regulatory effects of angiogenic factors on the EAM
expression under these conditions (Table l ) . ’ ? TGFP and
IL-8 exposure of HUVEC did not result in any significant
inhibition of adhesion.
Reactivity of isolated tumor EC. In the in vivo situation,
tumor EC are exposed to high levels of angiogenic factors.
Therefore, we were prompted to study the reactivity of tumor-associated EC to activation by inflammatory cytokines.
Freshly isolated EC from normal and tumor renal tissue were
analyzed flow cytometrically for constitutive and TNFa-induced expression of ICAM-1 (Fig 6). We found that the
expression of ICAM- 1 on EC from normal renal tissue could
be enhanced by incubation in TNFa. EC obtained from renal
cell carcinoma tissue exhibit low levels of ICAM-I expression. Stimulation of these cells with TNFa did result in some
elevation of ICAM-1 levels, but significant levels (resembling expression of resting cells from normal tissue) were
not achieved (Fig 6). The suppressed upregulation of ICAM1 after TNFa was also observed with concentrations of
TNFa as high as 0.4 pg/mL .
DISCUSSION
We showed in a recent report that tumor-associated EC
have a suppressed expression of ICAM- 1 and ICAM-2 compared with EC isolated from healthy tissue. We also showed
that the presence and the local production of angiogenic
factors is responsible for this phenomenon.” In this report,
it is shown that angiogenic factors suppress the cytokineinduced upregulation of EAM. This downregulation is due
neither to different time-kinetics of the response (Fig 1) nor
to different dose requirements, as could be shown by doseresponse curves (data not shown). Thus, not only the basal
expression of EAM is decreased in the tumor vasculature, but
also the possibility of inflammation-mediatedupregulation is
hampered.
The present results suggest that a general mechanism for
regulation of EC inflammatory responses in relation to proliferation is operative. The mitogenic cytokines used (bFGF
and VEGF) were found to be the most effective inhibitors
of EAM upregulation. Treatment of EC with angiogenic factors not only modifies EAM expression, but also effectively
inhibits HLA-ABC upregulation by 1FN-y. However, differ-
Table 1. Percentage of Inhibition of Cytokine-Induced Upregulation of EAM by Angiogenic Factors
EAM
Fold Upregulationt
TNFa (n = 7)
ICAM-1
VCAM-1
E-selectin
8.4
9.6
11.9
I L - l a (n = 4)
CAM-1
VCAM-1
E-selectin
lFNy (n = 5)
ICAM-1
Cytokine*
~ ~ _ _ _ _ _ _ _
~
bFGF*
VEGFt
L8*
TGFBS
33 f 80
59 i 311
55 t 511
15 f 40
24 I 511
33 2 311
454
7 2 4
11 I 4
14 I 41
31 I l o ?
31 t 411
4.1
3.7
2.0
49 -t 130
81 2 711
6 2 6
27 i 511
7 1 5
8 2 7
6 1 8
8 2 3
38 2 25
-6 5 5
14 I 1 0
25 t 17
2.5
63 f 181
23 t 61
121
~~
* TNFa was administered at 4 ng/mL,
2 t 24
I L - l a at 40 ng/mL, and IFNy at 100 U/mL.
t Mean upregulation of adhesion molecule expression by TNFa, IL-la, or IFNy in n experiments.
Mean percentage of inhibition (fSEM) of inflammatory cytokine-induced upregulation of EAM expression by the angiogenic factors bFGF.
VEGF, IL-8, and TGFP. Inhibition is calculated as indicated in Fig 4.
1 P < .01.
*
1 P < .os.
11 P C ,001.
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
THE INFLAMMATORY RESPONSEOFTUMOR
671
ENDOTHELIUM
Fig 5. Upregulationofleukocyteendothelial c e l l
adhesion by TNFm is inhibited by bFGF and VEGF
but not by TGFp and IL-8.HL-60 promyelocyticcell
line (A) and PHA-activated peripheral blood T cells
IBI were adhered for 1 houronrestingorpreactivated EC (10 nglmL bFGF, 100 ng/mL VEGF, 10 n g l
mL TGFP, or 30 ng/mL 11-8). Bars represent mean
adhesion in quadruplicate
cultures I S E M ) . (m)Experiments in which, before adhesion, EC were cultured with 4 nglmL TNFn during the last16 hours.
The resultsof one representative experimentoffour
are shown. Bars identified
with
asterisks are significantly
different
from
values without angiogenic factors ( * P < .01; * * P < .001).
120
-
IO0
-
80
-
700
A
600
$
60
500
-
40
-
20
-
m
9
H
g.
2 300
0
0
=I
200
0
30
100
0
0 -
0
W g L - l g
LLLW
ucoL
-
>
E
c
0
entia1 effects were observed after treatment of EC with the
mitogenic factors bFGF and VEGF; eg, E-selectin expression induced by TNFa but not by IL-la can be inhibited by
pretreatment of EC with bFGF. This finding is in line with
those of earlier studies indicating that induction of E-selectin
by IL-la is not affected by bFGF." These findings probably
reflect differences in signalling via IL-la and TNFa receptors leading to differential responses to angiogenic stimuli.
Another indication of the diverse response of EC to angio-
genic factors is the fact that EAM upregulation by TNFa
but not by IL-la can be significantly suppressed by TGFD
(Table 1). It has been described previously that TGFO inhibits leukocyte adhesion to endothelial cells."."' These results
could not be explained by downregulation of VCAM-I and
ICAM-I. Because we observed an inhibited membrane expression of adhesion molecules induced by TGFD (Table l),
a minor reduction of adhesion might have been expected.
The differential regulation of the signals provided by differ-
10 4
10 4
60
.g 400
W
m
T T
c
C
-2
90
B
lo'
normal
.
.
.
.
.
'
10'
1 i o zi o
10'
io4
1
IO
10'
10'
io4
10 4
tumor
1
10
10'
IO'
coni. control
1
l
1
IO
10'
ICAM-1
103
1
L
tumor
1
IO
10'
ioJ
10'
ICAM-1
Fig 6. Suppressed upregulation of ICAM-1 b y TNFa in renal cell carcinoma-derived €C. FACS analysis (1 representative experiment of 3)of
EN4-PE (vertical axis) and ICAM1-FITC (horizontal axis) of normal renaltissue-derived EC (upper panels) and renal cellcarcinoma-derived EC
flower panels), of untreated EC (B and E), or of EC after TNFa activation (C and F). Conjugate controls for FITC-fluorescence are presented in
(A) and (D). The mean ICAM-1 (FITC) fluorescence intensity of all ENdpositive cells is givenin the upper left corner.
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
GRlFFlOEN ET AL
672
ent inflammatory cytokines by TGFP is currently under investigation.
Verification of the biologic significance of the observed
regulation of EAM expression was performed in adhesion
assays. It should be mentioned that the adhesion experiments
performed in this study do not allow discrimination between
the different adhesion molecules. Therefore, we should consider the adhesion measurements as the net result of the
regulated EAM expression. The adhesion molecules studied
are critical in rolling of leukocytes along the vascular lining
cells, firm adhesion to the EC, and extravasation into surrounding tissues. In a previous study, we have already shown
that the adhesion measured in these kind of experiments is
mainly mediated by ICAM-1 and its ligands.’’
Activation experiments with tumor-derived EC were performed by flow cytometric analysis of freshly isolated EC
from renal cancer tissue. This tumor type was selected for
these studies for reasons of high vascularization of the tumor
and the availability of normal renal control tissue from the
same donor. Induced artefacts due to the isolation procedure
can be discounted because control tissue is treated similarly.
It has been described previously that angiogenic factors can
be measured in the urine of cancer patients.23 However, our
results are measured despite the presence of these circulating
factors.
However, due to the highly specialized functions of renal
EC, it is likely that these EC do not fully represent the
phenotype of EC in normal tissue. In a number of experiments, we also used umbilical vein- and dermis-derived EC.
With respect to the expression of ICAM- 1, VCAM- 1, and
E-selectin, these cells resembled EC from control renal tissue
(Griffioen et al, unpublished observation). A drawback of
the use of freshly isolated EC from cancer tissue might be
that they grow in cultures together with other adherent cell
types. Identification of EC should therefore be performed,
for instance, by double staining with EN4-PE. However,
because EC in tumors are stable from a genetic point of
view, this mixed-culture model of EC with cells that are also
present in tumor microenvironment guarantees the integrity
of tumor EC phenotype.
In the present report, a reduced upregulation of ICAM-I
on TNFa-exposed, tumor-derived EC was noted. Combining
these findings with those of the in vitro HUVEC experiments, we conclude that there is an inhibited cytokine-mediated upregulation of EAM on endothelium induced by angiogenic factors. For VCAM-1 and E-selectin, this conclusion
is valid because both molecules are absent on unstimulated
EC. However, ICAM-1 is constitutively expressed on normal
EC; also, on tumor-associated EC, there is (a very low)
expression of ICAM-1. Activation of EC by TNFa induces
a measurable level of upregulation on both normal and tumor
EC. Although the magnitude of the response might therefore
be comparable (measured as percentage increase) in both
cell types, we show that the expression on tumor EC is still
decades lower than the expression of normal EC, even in
the presence of TNFa concentrations of up to 0.4 pg/mL.
Considering that, in the resting normal situation, inflammatory infiltrate formation is absent and that this level of EAM
expression can not be reached after exposure to TNFa, there
is, in our view, an inhibited inflammatory response of tumor
EC or, similarly, there exists EC anergy. This anergy will
have impact on the effect of clinical immunotherapy. It is
suggested here that the antitumor effect of IFNa in patients
is based, at least partly, on counteracting this anergy, because
IFNa is reported to decrease bFGF production.’4
For an explanation of our findings it can be suggested that
triggering of the bFGF receptor (bFGF-R) leads to either a
decrease of TNF receptor sensitivity or downregulation of
the expression TNF receptor on the cell. The latter possibility
seems not likely because inflammatory responses to other
cytokines (IL-la and IFNy) are also decreased and other
angiogenic factors can have effects similar to bFGF. In addition, in preliminary experiments using anti-TNFp55R and
anti-TNFp75RZ5 antibodies we found expression of these
receptors on HUVEC, but no regulation of expression by
bFGF was found. To explain the present results by a decreased signalling efficiency of different cytokine receptors
by ligation of the bFGF-receptor seems feasible. The description of the bFGF-R as a transmembrane tyrosine kinase (reviewed in Mason”) may reflect the involvement of the receptor in cell activation processes. The finding that stimulation
via the VEGF-receptor (VEGF-R) results in similar physiologic effects and the characterization of the VEGF-R as a
tyrosine kinase-associated re~eptor’~.’~
suggests that part of
the effect may rely on phosphorylation of protein substrates
on tyrosine. Heterologous receptor desensitisation, a mechanism described in antigen-induced clonal anergy of B lymphocytes,” might then be involved in the similar effects of
different angiogenic factors and proinflammatory cytokines.
The present report describes the suppressed response of
tumor EC to activation by inflammatory cytokines compared
with EC from normal tissue. In experiments with HUVEC
it is shown that angiogenic factors are capable of inducing
this state of anergy, both at the protein level (expression)
and the functional level (adhesion). It can be concluded from
the current study that tumors that are dependent on angiogenesis by their production of angiogenic factors downregulate
the inflammatory response of the vascular endothelium. This
may serve as a tumor-protecting mechanism by impairing
the development of effective leukocyte infiltrates. Future
work will concentrate on the possib es to improve results
of antitumor immunotherapies by counteracting this EC anergy.
ACKNOWLEDGMENT
Dr N. van Adrichem (Amersfoort, The Netherlands) and Dr T.
Boon (Utrecht, The Netherlands) are greatly acknowledged for providing surgically removed tissue.
REFERENCES
1. Shimizu Y, Newman W, Tanaka Y, Shaw S: Lymphocyte interactions with endothelial cells. Immunol Today 13: 106, 1992
2. Zimmerman GA, Prescott SM, McIntyre TM: Endothelial cell
interactions with granulocytes: Tethering and signaling molecules.
Immunol Today 13:93, 1992
3. Staunton DE, Marlin SD, Stratowa C, Dustin ML, Springer
TA: Primary structure of ICAM- 1 demonstrates interaction between
members of the immunoglobulin and integrin supergene families.
Cell 52:925, 1988
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
THE INFLAMMATORY RESPONSE OF TUMOR ENDOTHELIUM
4. Rosenberg SA, Spiess P, Lafreniere R: A new approach to the
adoptive immunotherapy of cancer with tumor infiltrating lymphocytes. Science 233:1318, 1986
5. Griffith KD, Read ET, Carrasquillo JA, Carter CS, Yang JC,
Fisher B, Aebersold P, Packard BS, Yu MY, Rosenberg SA: In vivo
distribution of adoptively transferred indium-1 1 I-labelled tumor infiltrating lymphocytes and peripheral blood lymphocytes in patients
with metastatic melanoma. J Natl Cancer Inst 81:1709, 1989
6. Swerlick RA, Lee KH, Li LJ,Sepp NT, Caughman SW, Lawley TJ: Regulation of vascular cell adhesion molecule 1 on human
dermal microvascular endothelial cells. J Immunol 149:698, 1992
7. Ruco LP, Pomponi D, Pigott R, Stoppacciaro A, Monardo F,
Uccini S, Boraschi D, Tagliabue A, Santoni A, Dejana E: Cytokine
production (IL-I alpha, IL-I beta, and TNF alpha) and endothelial
cell activation (ELAM-I and HLA-DR) in reactive lymphadenitis,
Hodgkin’s disease, and in non-Hodgkin’s lymphomas. An immunocytochemical study. Am J Pathol 137:1163, 1990
8. Pober JS, Gimbrone MAJ, Lapierre LA, Mendrick DL, Fiers
W, Rothlein R, Springer TA: Overlapping pattems of activation of
human endothelial cells by interleukin 1, tumor necrosis factor, and
immune interferon. J Immunol 137:1893, 1986
9. Gamble JR, Vadas MA: Endothelial cell adhesiveness for human T lymphocytes is inhibited by transforming growth factor-beta
1. J Immunol 146:1149, 1991
10. Gamble JR, Vadas MA: Endothelial adhesiveness for blood
neutrophils is inhibited by transforming growth factor-beta. Science
242:97, 1988
1 1. Folkman J: What is the evidence that tumors are angiogenesis
dependent? J Natl Cancer Inst 82:4, 1990
12. Folkman J, Shing Y: Angiogenesis. J Biol Chem 267:10931,
1992
13. Griffioen AW, Damen CA, Martinotti S, Blijham GH,
Groenewegen G: Endothelial ICAM-1 expression is suppressed in
human malignancies; role of angiogenic factors. Cancer Res
56:1111, 1996
14. Wu NZ, Klitzman B, Dodge R, Dewhirst MW: Diminished
leukocyte-endothelium interaction in tumor microvessels. Cancer
Res 52:4265, 1992
15. Piali L, Fichtel A, Terpe HJ, Imhof BA, Gisler RH:Endothelial vascular cell adhesion molecule 1 expression is suppressed by
melanoma and carcinoma. J Exp Med 18I:8ll, 1995
16. Kitayama J, Nagawa H, Yasuhara H, Tsuno N, Kimura W,
Shibata Y, Muto T: Suppressive effect of basic fibroblast growth
factor on transendothelial emigration of CD4(+) T-lymphocyte.
Cancer Res 54:4729, 1994
17. Schlingemann RO, Rietveld FJ, Kwaspen F, van de Kerkhof
673
PC,de Waal RM, Ruiter DJ: Differential expression of markers for
endothelial cells, pericytes, and basal lamina in the microvasculature
of tumors and granulation tissue. Am J Pathol 138:1335, 1991
18. Renkonen R, Paavonen T, Nortamo P, Gahmberg CG: Expression of endothelial adhesion molecules in vivo. Increased endothelial ICAM-2 expression in lymphoid malignancies. Am J Pathol
140:763, 1992
19. Kuzu I, Bicknell R, Fletcher CD, Gatter KC: Expression of
adhesion molecules on the endothelium of normal tissue vessels and
vascular tumors. Lab Invest 69:322, 1993
20. Doussis Anagnostopoulou I, Kaklamanis L, Cordell J, Jones
M, Turley H, Pulford K, Simmons D, Mason D, Gatter K: ICAM3 expression on endothelium in lymphoid malignancy. Am J Pathol
143:1040, 1993
21. Groenewegen G, De Leij M, Jeunhomme GMAA, Buurman
WA: Supematants of human leukocytes contain mediators, different
from interferon-gamma, which induces expression of MHC class I1
antigens. J Exp Med 164:131, 1986
22. Netland PA, Zetter B, Via DP, Voyta JC: In situ labeling of
vascular endothelium with fluorescent acetylated low density lipoprotein. Histochemical J 17: 1309, 1985
23. Nguyen M, Watanabe H, Budson AE, Richie JP, Hayes DF,
Folkman J: Elevated levels of an angiogenic peptide, basic fibroblast
growth factor, in the urine of patients with a wide spectrum of
cancers. J Natl Cancer lnst 86:356, 1994
24. Singh RK, Gutman M, Bucana CD, Sanchez R, Llanza N,
Fidler IJ: Interferons alpha and beta down-regulate the expression
of basic fibroblast growth factor in human carcinomas. Proc Natl
Acad Sci USA 92:4562, 1995
25. Brockhaus M, Schoenfeld HJ, Schlaeger El, Hunziker W,
Lesslauer W, Loetscher H: Identification of two types of tumor
necrosis factor receptors on human cell lines by monoclonal antibodies. Proc Natl Acad Sci USA 87:3127, 1990
26. Mason IJ: The ins and outs of fibroblast growth factors. Cell
78:547, 1994
27. Millauer B, Wizigmann-Voos S, Schnurch H, Martinez R,
Moller NPH, Risau W, Ullrich A: High affinity VEGF binding and
developmental expression suggest Flk-1 as a major regulator of vasculogenesis and angiogenesis. Cell 722335, 1993
28. Seetharam L, Gotoh N, Maru Y, Neufeld G, Yamaguchi S,
Shibuya M: A unique signal transduction from FLT tyrosine kinase,
a receptor for vascular endothelial growth factor VEGF. Oncogene
10:135, 1995
29. Rijkers GT, Griffioen AW, Zegers BJM, Cambier JC: Ligation of membrane immunoglobulin leads to inactivation of the signaltransducing ability of membrane immunoglobulin, CD19, CD21, and
B-cell gp95. Proc Natl Acad Sci USA 87:8766, 1990
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1996 88: 667-673
Tumor angiogenesis is accompanied by a decreased inflammatory
response of tumor-associated endothelium
AW Griffioen, CA Damen, GH Blijham and G Groenewegen
Updated information and services can be found at:
http://www.bloodjournal.org/content/88/2/667.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.