From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
Both Tumor Necrosis Factor Receptors, TNFR60 and TNFRSO, Are Involved
in Signaling Endothelial Tissue Factor Expression by Juxtacrine
Tumor Necrosis Factor a!
By Esther F. Schmid, Klaus Binder, Matthias Grell, Peter Scheurich, and Klaus Pfizenmaier
We have investigated the role of the two distinct tumor
necrosis factor (TNF) receptors (TNFRGO and TNFRIO) in endothelial cell activation employing an in vitro model of tumor necrosis factor a (TNF-a)-dependent tissue factor production of human umbilical vein endothelial cells (HUVECs).
In this model, tissue factor is produced either on addition of
exogeneous TNF-a, or by induction of endogenous TNF-a
via adhesion molecule-linked signal pathways. Under both
conditions,
tissue
factor
expression
could be partially
blocked by antagonistic antibodies against either TNFRGO or
TNFR80 and was fully inhibited by simultaneous application
of both antibodies. Selective inhibitors of either TNFRGO or
TNFRBO-induced signal pathways inhibited tissue factor expression, and selective triggering of either of the two TNF
receptors by agonistic antibodies induced this response in
HUVECs. Furthermore, a coculture system of HUVECs and
Chinese hamster ovary transfectants expressing a noncleavable, exclusively membrane-bound form of TNF-a resulted
in a potent activation ofHUVECs with synergistic action
of both TNF receptors. Together, these data underline the
importance of juxtacrine pathways in endothelial cell activation of procoagulant functions and show that membrane
TNF-a and both TNFR types play a critical role.
0 1995 by The American Societyof Hematology.
T
expression in primary HUVEC cultures already declines at
very early passages, ( 2 ) other studieswith distinct celltypes,
showing a genuine, TNFR60 independent signal capacity of
TNFR80 on appropriate activation,’6”8 and (3) mouse tumor
models suggesting involvement of TNFR80andthe host
endothelium in induction of tumor necrosis,”,’” we scrutinized the functional role of TNFR80 in endothelial cell activation and analyzedpotentialmechanisms
of cooperation
with TNFR60. To this end, we have used second passage
HUVECs in a model of in vitro tissue
factor induction by
endogenously induced or exogenously added,solubleor
membrane bound TNFa. By using receptor subtype specific
agonistic and antagonisticantibodies,aswell
as selective
inhibitors of TNFR60and TNFR80-inducedsignalpathways, we show that both TNF receptors inducesignal pathways in HUVECs thatsynergisticallyinteract
toproduce
tissue factor.
UMOR NECROSIS factor a (TNF-a) isapotentinflammatory cytokine thatissecreted
by several cell
types, in particular activated macrophages and T cells,’ but
also endothelial cells and othertissues.’.“ On monocytes and
endothelial cells, TNF-a exerts multiple biologic effects including the induction of tissue factor,”’ which has been implicated in a number of severe pathologic conditions,including systemiccoagulationoccuringduringseptic
shock.’”‘’
Recent studies in our laboratoryhave shown thatinvitro
antibodycross-linkingof
thecell adhesionmolecules Eselectin and intercellular adhesion molecule (ICAM) l on the
surface of human umbilicalcord
vein endothelialcells
(HUVECs)mimicsleukocyteadhesiontoendothelial
ce1ls.loaThis initiates an intracellular signal cascade leading
to the production of TNFa that acts in an autocrine or juxtacrine manner and induces tissue factor expression.”
Two distinct TNF receptors of 55 to 60 kD (TNFR60)
and 75 to 80 kD (TNFR80) have been described, and both
receptors were found to be expressed on HUVECS.”.’~ The
functional significance of TNFR60 was evident from
several
studies, whereas the role of TNFR8O in endothelial cellactivation
remained
example,
For
TNF-a-mediated
inductionofadhesionmolecules
has beenattributed
to
TNFR60signaling, with only a minor functional role of
TNFR80.I2 More recent studiesprovided some evidence for
a TNFR60-dependent, supportive function of TNFR80,’3.’4
compatible with the model of ligand passing.” Based on (1)
our own unpublished observations that TNFR80 membrane
From the Department of Pharmacological Research, Dr Karl Thomae GmbH, Biberach; and the Institute of Cell Biology and Immunology, University oj Stuttgart, Stuttgart, Germany.
Submitted March 22, 1995; accepted April 28, 1995.
Supported in part by Grant
No. 0310816from the German Federal
Ministly of Education and Research.
Address reprint requests to Klaus Pjizenmier, PhD, Institute of
Cell Biology and Immunology, University of Stuttgart, Allmandring
31, 70569 Stuttgart, Germany.
The publication costsof this article were defrayed in part by page
chargepayment. This article must therefore be hereby marked
“advertisement” in accordance with 18 U.S.C. section 1734 solely to
indicate this fact.
0 1995 by The American Society of Hematology.
0006-4971/95/8605-0016$3.00/0
1836
MATERIALS AND METHODS
Reagents. Recombinant human interferon y (Dr K. Thomae
GmbH, Biberach, Germany) and recombinant human TNF-a (Genentech, San Francisco, CA) were usedat a final concentration of
50 ng/mL. Pargyline and nordihydroguaiaretic acid (NDGA) were
purchased from Sigma Chemical CO (Deisenhofen, Germany) and
were used at final concentrations of 250 and 3 pg/mL, respectively.
D609 (Karniya Biomedical Company, Thousand Oaks, CA) was
diluted to a final concentration of 40 &mL. The generation of the
monoclonal antibody (MoAb) H398, a TNFR60 specific antagonist,2’
the rabbit antihuman TNFR80 serum M80:’ and the TNFR60-specific agonistic MoAb htr-l*’ have been described previously. The
purified IgG fraction from M80 serum was used for stimulation
experiments. For TNFR80 inhibition studies, Fab fragments, generated from the IgG fraction of M80 serum were used as selective
antagonists, as described.2’ MoAb to E-selectin (ELAM, IgG2,
mouse), aswell as a MoAb specific for ICAM-I (IgG1, mouse)
were provided by Bender (Vienna, Austria). All antibodies were
devoid of preservatives and were used at a final concentration of 50
pg/mL, except for cytofluorometric analysis.
Isolation and culture of endothelial cells. Endothelial cells from
HUVECs were isolated according to the method of Jaffe et alz4and
propagated in culture in medium M199 supplemented with CLEX
(5% wt/vol; Interchem, Miinchen, Germany), heat inactivated fetal
calfserum (5% wt/vol) andhuman serum (5% wtlvol) until they
reached confluence. The cells were detached from the culture dish
by collagenase treatment ( 1 mg/mL in phosphate-buffered saline
Blood, Vol 86,No 5 (September l), 1995:pp 1836-1841
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
'l837
TNF RECEPTOR SIGNALING
[PBS], 2 minutes, 37°C). washed in M199 plus supplementations
and then seeded on four-well plates (Nunclon, Nunc, Denmark)
coated with extracellular matrix derived from bovine corneae endothelial cells. For assays, cells of the second passage were grown to
confluence for 2 to 3 days in medium M199 plus supplements.
Eachwell (diameter = 16 mm) contained approximately 1 X IO*
HUVECs.
Coculture of HUVECs and membrane TNF expressing
transfectants. Stable C H 0 transfectants expressing a noncleavable, membrane-bound form of TNF-a (clone B2.20) have been constructed
as described by Perez et al." TNF-a membrane expression and the
complete lack of soluble, bioactive TNF of clone B2.20 was verified
by fluorescence-activated cell sorter (FACS) analysis and standard
TNF-a bioassay," respectively, and is described in detail elsewhere
(Grell et al, submitted). For the experiments described herein, B2.20
and control C H 0 cells, respectively, were added to the endothelial
cells at 2 X 10s cells per well.
Antibody staining and cytofluorometric analysis. HUVECs were
washed and detached from the extracellular matrix by collagenase
treatment as described above. All MoAbs (H398 and isotype controls) were used at a final concentration of 5 pg/mL. M80 serum
was used as a 1:300 final dilution. The antibodies were incubated
at 4°C for 45 minutes in PBS containing 0.2%bovine serum albumin
(BSA). The cells were washed with PBSiBSA and stained with a
fluoresceinisothiocyanate (FITC)-conjugated F(ab)2 goat antimouse
and goat antirabbit IgG, respectively, for 45 minutes at 4°C. The
cells were washed again and the pellet was resuspended in PBSI
BSA. Control antibodies of the corresponding isotype and preimmune serum were used to determine nonspecific binding. Cytofluorometric analysis was performed with an Epics CS (Coulter, Hialeah,
FL) at a wavelength of 488 nm and a Laser setting of 250 mW.
Antibody cross-linkingof E-selectin and ICAM-I and TNF-aproduction of HWECs. Anti-E-selectin (ELAM) and anti-ICAM-1
were incubated with HUVECs, preactivated by interferon (1FN)y
treatment (50 ng/mL, 6 hours, 37"C), and washed extensively before
further treatment with PBS as described." Antibody incubations
were performed at 4°C for 1 hour. For antibody cross-linking, F(ab),
fragments of a goat antimouse IgG were added at a final concentration of 500 pg/mL after washing the cultures with PBS. The F(ab)z
fragments were incubated for 1 hour at 4"C, as described above.
After washing with PBS, the cells were maintained in M199 (plus
supplements) at 37°C for 12 hours. At that time, TNF-a production of
E-selectin/ICAM-l -stimulated HUVECs was determined in culture
supernatants by a human TNF-a-specific enzyme-linked immunosorbent assay (ELISA) kit (Boehringer Mannheim, Germany) according to the manufacturer's instructions.
Blockade of TNFRs by antagonistic antibodies. After cross-linking of E-selectin and ICAM-I as described above, HUVECs were
washed with PBS and incubated at 4°C for 45 minutes with antibodies specific for TNFR60 (H398) andor TNFR80 (M80 Fab fragments). The cells were washed with PBS and further incubated in
0.5 mL medium M199 (plus supplements) at 37°C for 12 hours. For
blockade of membrane TNF-a-induced signal transduction, HUVECs were incubated, before addition of B2.20 cells to the cultures,
in accordance with the reagents above.
StimulationofTNFRs
withagonisticantibodies.
Endothelial
cells were washed with PBS and separately incubated with agonistic
antibodies against TNFR60 (50 pg/mL, htr-l) or TNFR80 (50 pg/
mL, M80 IgG fraction), at 4°C for 45 minutes. After washing with
PBS, 0.5 mL medium M199 (plus supplements) were added and the
cells were incubated at 37°C for 12 hours.
Inhibition of 7NF-induced signalpathways. For blockade of the
TNF signal pathways, the inhibitors pargyline (250 pg/mL), NDGA
(3
and D609 (40 pg/mL) were added to HWECs after
cross-linking of E-selectin and ICAM-I, being present during the
12-hour incubation period.
Tissue factor activity. The surface procoagulant activity of the
HUVEC monolayers was measured in a two-stage thrombin formation assay in human whole blood. A total of 0.6 mL of freshly drawn
citrated human blood was recalcified and immediately added to the
HUVECs. The four-well plate was shaken gently at 37°C. A total
of IO pL samples were taken at l-minute intervals, for 20 minutes.
The samples were diluted 1:400 in reaction buffer (TRIS-HC1 50
mmoVL, NaCl 100 mmoVL, EDTA 20 mmoVL, BSA 0.5 g L , pH
7.9) containing 200 pmoVL chromogenic substrate S2238 (Kabi
Vitrum, Molndal, Sweden). The reaction mixture was kept on ice
for the duration of the experiment, then cellular components were
removed by centrifugation. The chromogenic reaction was started
by rapidly warming to 37°C. Absorption at 405 nm was measured
at two time points (5 minutes and 10 minutes after the start of the
warming) and optical density (OD) was determined. The amount of
functionally active thrombin generated was calculated from a standard curve derived from purified human thrombin (kindly provided
by Professor H.C. Hemker, Maastricht, Netherlands). Next, a dose
response curve of recombinant human tissue factor (kindly provided
by Professor W. Konigsberg, Yale University, New Haven, CT) was
prepared. We determined the time delay (in minutes) until the rate
of thrombin formation exceeded 30 nmoVL as a parameter to convert
the onset of thrombin formation to tissue factor concentration. The
same method was used to determine the amount of tissue factor on
the surface of endothelial cells.
RESULTS
Both TNF receptors are involvedin endothelial cell tissue
factor production. To investigate the role of both TNF re-
ceptors in TNF-a signaling of tissue factor production, membrane expression of TNFR60 and TNFR80 on H W E C s
from second passage in vitro culture was verified by FACS
analysis. A typical result is shown in Fig 1, indicating that
both TNFRs were expressed at approximately equal levels
in these cells, with 63% and 68% of HUVECs staining at
above background levels with TNFR60 and TNFRSO specific
antibodies, respectively.
In the first set of experiments, tissue factor production
was initiated by simultaneous cross-linking of E-selectin and
ICAM-1, previously shown to stimulate endogenous TNFa,'Oawhich subsequently induces tissue factor expression in
endothelial cells. In the experiments depicted in Fig 2A, a
tissue factor production of 110 f 15 pg/well was noted,
whereas, in the absence of cross-linking adhesion moleculespecific antibodies, no tissue factor production was observed. When antagonistic antibodies against TNFR60 and
TNFR80 were separately added to preactivated HUVECs
directly after cross-linking of E-selectin and ICAM-1, the
expression of functional tissue factor activity decreased to
50 f 21 pg/well and 60 f 17 pg/well, respectively (Fig 2A).
Simultaneous blocking of both TNFRs reduced the amount
of functionally active tissue factor to 30 t 16 pg/well (Fig
2A). Independently, involvement of TNFR60 and TNFR80
in tissue factor induction was confirmed by stimulation with
recombinant TNF-a (50 ng/mL), instead of endogenously
produced TNF-a.Again, tissue factor activity (145 2 18 pg/
well) could be inhibited, both by antibodies against TNFR60
(50 2 1 1 pg/well) and by antibodies against TNFR80 (85
2 16 pglwell) (Fig 2B). Simultaneous blockade of both
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
SCHMID ET AL
1838
2
1
R
I
FITC
FITC
A
--
ELAM I CAM1 crosslink
TNF& antagonist
TNF$, antagonist
B
160
4
TNF
T N F h antagonist
T N F k antagonist
+
-
+
+
+
+
-
+
+
+
+
+
+
+
-
-
+
T
--
+
-
+
+
Fig 2. Inhibition of endogenous and exogenous TNF-cr-induced
tissuefactor expression on HUVECs byantagonisticantibodies
against TNFR6O and TNFR8O. Treatment of HUVECs for induction of
tissue factor expression and inhibition by receptor-specific antagonistic antibodies was done as described in Materials and Methods.
(A) E-selectinIICAM-1-cross-linking-induced tissuefactorproducexogenous TNF-cr. Data
tion. (B)Tissue factor production induced by
represent mean t SD of three independent experiments.
1OOE
Fig 1. Analysis of TNFR expression on second
passage HUVECs. Cells were incubated with (1)
MoAb H398, specific for TNFRGO and (2) rabbit antihuman TNFRIO serum (1:300). Isotype-matchedcontrols(filledhistograms)were
used t o determine
background staining (C) and for calculation of percent positivecells.
TNFRs led to a further reduction in tissue factor activity (30
5 IS pg/well) (Fig 2B).
Signal transduction inhibitors block E-selectin/lCAM-I
induced tissue factor production at the level ?f TNFR signaling. The above results indicated an involvement of both
TNF receptors in induction of tissue factor gene expression.
It was now of interest to determine the role of each of the
TNFR in this response, in particular whether TNFR80 has
an activate, TNFR60 independent signaling function or is
only involved as a ligand passing receptor. To this end, we
used inhibitors previously shown to selectively interfere with
TNF-a-induced intracellular signal pathways.'".2h Among
several inhibitors shown to affect TNF-a signaling, two were
of special interest because they allowed to discriminate between TNFR60- and TNFR80-induced signal pathways in a
model of TNF-a-mediated tumor cell apoptosis: NDGA,
known as an inhibitor of lipoxygenases and mono-oxygenases, wasfoundto
interfere with TNFR80 signaling,
whereas Pargyline, a mono-amino-oxidase inhibitor, was
shown to selectively affect TNFR60 induced cell death.'"
HUVECs were activated by E-selectin/ICAM- 1 crosslinking, and cells were cultured in the absence or presence
of inhibitors for I2 hours before tissue factor expression was
determined. In parallel to tissue factor activity, TNF levels in
the culture supernatants were measured by ELISA to ensure
production of endogenous TNF-a (236 pg/mL) by E-selectin/ICAM-I cross-linking (Fig 3). The presence of D609,
a phosphatidylcholine-specific phospholipase C reagent,
known as an effective inhibitor of TNF responsiveness in
several."' but not all,'"cell types, resulted in a strong inhibition of tissue factor expression (Fig 3). Likewise, the separate
addition of the selective inhibitors of TNFR60 and TNFR80
signal pathways, pargyline and NDGA, in each case led to
a reduction of tissue factor expression from 180 5 16pg/
well to 40 5 I 1 and 38 2 12 pglwell, respectively (Fig
3). Equal TNF-a levels in all E-selectin/ICAM- 1 -activated
HUVEC cultures verified that the inhibitors used were, indeed, selective for TNF-a-induced signal pathways and did
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
TNF RECEPTOR SIGNALING
,-,
y 120
2
J
-
.B
1839
I
I
T
T
80
60
40
20
0
ELAM I CAM1 crosslink
40
-
+
+
+
+
4 255
0 236236
TNF production (pglml) 228
Fig 3. E-selectin/lCAM-l-induced, TNF-dependent stimulation of
tissue factor synthesis involves
both TNFRGO and TNFRIO signal
pathways. E-selectin/lCAM-l-cross-linking-induced tissuefactor
production in the absence or presence of variousTNF-a signal transduction inhibitors. HUVECs were treated on antibody cross-linking
with D609 (40 pg/mL), pargyline (250 pg/mL), or NDGA (3 pg/mL),
and TNF-a, as well as tissue factor synthesis, was determined after
a 12-hour treatment as described in Materials and Methods. Any
direct influence of these compoundson TNF-a-independent tissue
factor synthesis was
excluded in control experiments ofLPS-induced
tissue factor production of HUVECs, which was found t o be unaffected by all threesubstances (data not shown).
not interfere with adhesion molecule-induced signals leading to endogenous TNF-a production in HUVECs (Fig 3).
These data suggest that both TNFR are capable of signal
transduction in endothelial cells and that the pathways activated by TNFR60 and TNFR80, respectively, are distinct.
TNFR80 independently induces tissue factor production
and syergizes with TNFR60 at the level of signal rransducan
active signaling,
tion. Direct experimental proof of
rather than a supportive. ligand passing functionI5 of
TNFR80 in HUVEC tissue factor production was obtained
in two independent approaches. First, HUVECs were treated
with agonistic antibodies htr-l and M80, specific for
TNFR60 and TNFRSO. respectively. With each of these agonistic antibodies, a selective activation of HUVECs could
be shown and costimulation of both TNFRs by agonistic
antibodies resulted in a further increase in tissue factor production (98 ? 12 pg/well, Fig 4). Thus, from antibody mediated activation of each of the TNFRs, their signal capability
in principle can be deduced. In quantitative terms, however,
receptor activation by antibodies depends on inherent features of these individual reagents and cannot be directly
compared with each other or to the natural ligand.
Therefore, in an approach to show an active signal capacity of each of the two TNFRs induced by natural ligand
under conditions that rule out a function of TNFR80 as a
ligand passing receptor, we used the transmembrane form
of TNF as a stimulus for tissue factor production of HUVECs. Stable C H 0 transfectants (B2.20) have been established expressing a mutated form ofTNF-a, in which normal
proteolytic processing to the soluble 17-kD form is abrogated
by internal deletion of 12 amino acids (nos. 1 through 12 of
the mature proteids). Similar constructs have previously
20
0
T N F b agonist
TNFR, agonist
-
-
+
-
+
+
+
Fig 4. Tissue factor expression on HUVECs induced by TNFR6Oand TNFRIO-specificagonistic antibodies. HUVECswere treated with
the respective agonistic antibodies as described in Materials and
Methods for45 minutes, washed with PBS, and cultured for
12 hours
before tissue factor activity was
determined. Data represent the
mean 2 SD of three independentexperiments.
been shown to possess bioactivity.” B2.20 cells express high
levels of TNF in the cell membrane, as shown by immunofluorescence microscopy, but do not release bioactive TNF
into the culture supernatant (Grell et al, submitted). Stimulation of HUVECs with B2.20 cells induced large amounts of
tissue factor, which rose from 2 ? 3 pg/well to 290 2 15
pg/well (Fig 3 , whereas coincubation of HUVECs with untransfected C H 0 cells did not induce tissue factor expression
(5 ? 5 pg/well, Fig 5). The membrane TNF-a-induced tissue factor activity could be strongly inhibited by antibodies
against TNFR60 (40 ? 7 pg/well) and TNFR80 (70 ? 12
pg/well). Simultaneous blockade of both TNFRs inhibited
tissue factor synthesis close to background levels (20 ? 15
320
82.20 C
-1
-
O C U ~ ~ U ~
TNFR, antagonist
T N F b antagonist
+
-
+
+
-
-+
+
+
+
+
Fig 5. The transmembrane form of TNF synergistically activates
both TNFRs. Tissue factor production induced by coculture of HUVECs with membrane-TNF-a-expressing CH0 transfectants (62.20)
and inhibition by TNFR-specific antagonistic antibodies. As a negative control, untransfected parental C H 0 cells were cocultured with
HUVECs lleft column). Before these experiments, control measurements were performed
t o determine tissue factor activity of parental
CH0 and the62.20 transfectant cell line. No significant tissue factor
activity could be detected in either cell line. Data shown represent
mean k SD of three independent experiments.
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
1840
SCHMfO ET A1
pg/well) (Fig 5). As expected, no secreted TNF was detectable in the supernatants of HUVECB2.20 cocultures with a
highly sensitive bioassay (KYM - 1 cytotoxicityz2;data not
shown). Together, membrane TNF-a and TNFR80 agonistinduced tissue factor production provide compelling evidence for a direct signaling function of TNFR80 and a synergistic cooperation at the level of signal transduction of both
TNF receptors in endothelial cell activation.
DISCUSSION
The relative importance of TNFR60 and TNFR80 in mediating the biologic effects of TNF-a, like the induction of
apoptosis and gene regulatory effects controlling cellular
differentiation, inflammatory and procoagulant responses, is
only beginning to be understood. The critical involvement
of TNFR60 in many TNF-a responses was first evident from
in vitro antibody inhibition studies" and later in vivo studies
with receptor-gene-knock-out animals fully confirmed these
predictions." A physiologic function for TNFR80 was less
apparent until it was shown that there is a reduced efficacy
of human as compared with mouse TNF-a in mouse tumor
modelsz8that could be related to the fact that human TNFa does not bind to the murine TNFR80.29It was initially
suggested that TNFR80 only functions to increase local concentrations of TNF-a at the cell membrane in the vicinity
of TNFR60 because of a high off-rate of soluble TNF-a
from TNFR80 (ligand passing model'5). Meanwhile, an independent signal capacity of TNFR80 has been clearly estab]ishedlh.22.30and an invivo function, eg, in lipopolysaccharide
(LPS)-induced shock syndrome and tissue necrosis, is apparent from receptor-knock-out mice.3' The data presented here
now show that TNFR80 also plays an active signaling role,
independent of TNFR60, in tissue factor induction in endothelial cells. The potential physiological significance of this
findingcanbe deduced from the above-mentioned mouse
tumor model^,'^.^^ suggesting an involvement of TNFR80, if
one considers the role of the tumor vasculature critical in
the initiation of necrosis, by induction of intravascular coagulation. This view of TNFR80 function is fully supported
by recent data from TNFR80 knock-out mice that exerted a
strongly reduced sensitivity to TNF-mediated tissue necrosis3'
Our results are at variance with those obtained in a similar
model by Paleolog et al," who could not show endothelial
cell activation, including tissue factor production, via
TNFRSO by itself. Two, mutually nonexclusive explanations
might be considered to account for these discrepant findings:
first, these authors used later passage cells, which, in
agreement with our own observations, already have a reduced TNFR80 membrane expression as compared with
freshly isolated or earlier passage HUVECS (Fig 1). Second,
the functional activities of the antibodies used here (polyclonal IgG, recognizing multiple TNFR80 epitopes) and in
their study (MoAb) are clearly different. It is our own experience (Grell et al, submitted), as well as those of
that the agonistic activity of TNFR-specific MoAbs can differ considerably in different experimental models. In addition to receptor activation by agonistic antibodies, we have
presented here two other independent lines of evidence for
an active signaling of TNFR80 in HUVECs and a cooperation with TNFR60 atan intracellular signal pathway level
(1) inhibition of TNF-a-induced tissue factor production
by NDGA, a substance that has been previously shown to
selectively interfere with TNFR80 and not TNFR60 signaling in a model of tumor cell cytotoxicity involving both
TNFR typesI6; (2) induction of tissue factor production by
a nonsecretable, membrane-bound TNF-a, which formally
rules out ligand passing as a mechanism of TNFR80 function, and inhibition of this response by TNFR80-specific
antagonistic antibodies. Of note is that on activation with
membrane bound TNFa, tissue factor production was particularlyhighand each of the receptor-specific, antagonistic
antibodies caused a marked inhibition of tissue factor production. These data strongly support the view that, in HUVECs, synergistic signal pathways of the twoTNFR are
operative. Although further studies are required to elucidate
in detail the molecular nature of the pathways used by each
of the TNFRs in endothelial cells, both the antibody inhibition studies as well as those employing inhibitors of signal
pathways suggest that for optimal tissue factor production
by HUVECs, a cooperation of both TNFRs at a signal pathway level is necessary.
Our observation that membrane TNF-a is a very efficient
module for induction of tissue factor gene expression points
tothe importance of cell-cell interactions and juxtacrine
mechanisms for regulating endothelial functions in a local
environment, such as the microvessels of a vascularized tumor or atherosclerotic lesions. We have previously shown
that leukocyte adhesion to HUVECs induces endothelial tissue factor via endogenous TNF-a gene expression," which
is initiated by a combined activation of ICAM-l and Eselectin induced signal pathways.'"" All cells capable of
TNF-a secretion, including endothelial cells, produce TNFa via the membrane-expressed 26-kD prec~rsor.'~.~'
Accordingly, it is conceivable that a juxtacrine positive-feedback
loop is initiated on leukocyte adhesion to the endothelium
promoting and expanding the procoagulant state to neighboring endothelial cells by a mutual interaction of membrane
TNF-a/TNFRs and stimulation of tissue factor synthesis.
REFERENCES
1. Adolf GR, Grell M, Scheurich P: Tumor necrosis factor, in
Luger TA, Schwarz T (eds): Epidermal Growth Factors and Cytokines, New York, NY, Marcel Dekker, 1994, p 63
2. Jevnikar AM, Brennan DC, Singer GG, Heng JE, Maslinsk W,
Wuthrich RP, Glimcher LH, Kelley VER: Stimulated kidney tubular
epithelial cells express membrane associated and secreted TNFa.
Kidney Int 40:203, 1991
3. Kahaleh MB, Zhou S: Induction of tumor necrosis factor synthesis by endothelial cells upon exposure to rTNF. Arthritis Rheum
32:124, 1989
4. Pober JS, Cotran RS: Cytokines and endothelial cell biology.
Physiol Rev 70:427, 1990
5 . Nawroth P, Stem D: Modulation of endothelial cell hemostatic
properties by tumor necrosis factor. J Exp Med 164:740, 1986
6. Clauss M, Murray JC, Vianna M, de Waal R, Thurston G,
Nawroth P, Gerlach H, Gerlach M, Bach R, Familletti PC, Stem D:
A polypeptide factor produced by fibrosarcoma cells that induces
endothelial tissue factor and enhances the procoagulant response to
tumor necrosis factorkachectin. J Biol Chem 265:7078, 1990
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
TNF RECEPTOR SIGNALING
7. Mulder AB, Hegge-Paping KSM, Magielse CPE, Blom NR,
Smit JW, van der Meer J, Halie MR, Blom VJJ: Tumor necrosis
factora induced endothelial tissue factor is located on the cell surface
rather than in the subendothelial matrix. Blood 84:1559, 1993
8. Kirchhofer D, Tschopp TB, Hadvary P, Baumgaxtner HR: Endothelial cells stimulated with tumor necrosis factor alpha express
varying amounts of tissue factor resulting in inhomogenous fibrin
deposition in a native blood flow system. Effects of thrombin inhibitors. J Clin Invest 93:2073, 1994
9. Kirchhofer D, Sakariassen K, Clozel M, Tschopp T, Hadvary
P, Nemerson Y, Baumgartner H: Relationship between tissue factor
expression and deposition of fibrin, platelets and leukocytes on cultured endothelial cells under venous blood flow conditions. Blood
81:2050, 1993
10. Warr T, Rao L, Rapaport S: Disseminated intravascular coagulation in rabbits induced by administration of endotoxin or tissue
factor: Effect of anti-tissue factor antibodies and measurement of
plasma extrinsic pathway inhibitor activity. Blood 75:148, 1990
loa. Schmid E, Muller TH, Budzinski R-M, Binder K, Pfizenmaier K Signaling by E-selectin and ICAM-l induces endothelial
tissue factor production via autocrine secretion of PAF and TNFa.
J Interferon Cytokine Res 1995 (in press)
11. Schmid E, Muller TH, Budzinsky RM, Pfizenmaier K, Binder
K: Lymphocyte adhesion to human endothelial cells induces tissue
factor expression via a juxtacrine pathway. Thromb Haemost 73:421,
1995
12. Mackay F, Loetscher H, Stueber D, Gehr G, Lesslauer W:
Tumor necrosis factor alpha (TNF-alpha)-induced cell adhesion to
human endothelial cells is under dominant control of one TNF receptor type, TNF-R55. J Exp Med 177:1277, 1993
13. Paleolog EM, Delasalle SAJ, Buurman WA, Feldmann M:
Functional activities of receptors for tumor necrosis factor alpha on
human vascular endothelial cells. Blood 84:2578, 1994
14. Slowik MR, De Luca LG, Fiers W, Pober JS: Tumor necrosis
factor activates human endothelial cells through the p55 tumor necrosis factor receptor but the p75 receptor contributes to activation at
low tumor necrosis factor concentration. Am J Pathol143:1724,
1993
15. Tartaglia LA, Pennica D, Goeddel DV: Ligand passing: The
75-kDa tumor necrosis factor (TNF) receptor recruits TNF for signaling by the 55-kDa TNF receptor. J Biol Chem 268:18542, 1993
16. Grell M, Zimmermann G, Hulser D, Pfizenmaier K, Scheurich
P: TNF receptors TR60 and TR80 can mediate apoptosis via induction of distinct signal pathways. J Immunol 153:1963, 1994
17. Heller RA, Song K, Fan N, Chang DJ: The p70 tumor necrosis
factor receptor mediates cytotoxicity. Cell 70:47, 1992
18. Tartaglia LA, Weber RF, Figari IS, Reynolds C, Palladino
MA, Goeddel DV: The two different receptors for tumor necrosis
factor mediate distinct cellular responses. Proc Natl Acad Sci U S
A 88:9292, 1991
19. Fiers W: Tumor necrosis factor, characterization at the molecular, cellular and in vivo level. FEBS Lett 285:199, 1991
20. Brouckaert P, Libert C, Everaerdt B, Takahashi N, Cauwels
A, Fiers W: Tumor necrosis factor, its receptors and the connection
with Interleukin 1 and Interleukin 6. Immunobiology 187:317, 1993
21. Thoma B, Grell M, Pfizenmaier K, Scheurich P: Identification
1841
of a 60-kD tumor necrosis factor (TNF) receptor as the major signal
transducing component in TNF responses. J Exp Med 172:1019,
1990
22. Grell M, Scheurich P, Meager A, Pfizenmaier K: TR60 and
TR80 tumor necrosis factor (TNF) receptors can independently mediate cytolysis. Lymphokine Cytokine Res 12:143, 1993
23. Brockhaus M, Schoenfeld HJ, Schlaeger HE, 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
24. Jaffe EA, Nachman RL, Becker CG, Minick CR: Culture of
human endothelial cells derived from umbilical veins. J Clin Invest
52:2745, 1973
25. Perez C, Albert I, DeFay K, Zachariades N, Gooding L,
Kriegler M: A nonsecretable cell surface mutant of tumor necrosis
factor (TNF) kills by cell-to-cell contact. Cell 63:251, 1990
26. Schiitze S, Potthoff K, Machleidt T, Berkovic D, Wiegmann
K, Kronke M: TNF activates NF-kappa B by phosphatidylcholinespecific phospholipase C-induced “acidic” sphingomyelin breakdown. Cell 71:765, 1992
27. Pfeffer K, Matsuyama T, Kundig TM, Wakeham A, Kishihara
K, Shahinian A, Wiegmann K, Ohashi PS, Kronke M, Mak T W :
Mice deficient for the 55 kd tumor necrosis factor receptor are resistant to endotoxic shock, yet succumb to L. monocytogenes infection.
Cell 73:457, 1993
28. Brouckaert P, Everaedt B, Shaw A, Fiers W: Selective species
specificity of tumor necrosis factor for toxicity in the mouse. Lymphokine Cytokme Res 11:193, 1992
29. Lewis M, Tartaglia LA, Lee A, Bennett GL, Rice GC, Wong
GH, Chen EY, Goeddel DV: Cloning and expression of cDNAs for
two distinct murine tumor necrosis factor receptors demonstrate one
receptor is species specific. Proc Natl Acad Sci USA 88:2830, 1991
30. Tartaglia LA, Goeddel DV, Reynolds C, Figari IS, Weber
RF, Fendly BM,Palladino MA: Stimulation of human T cell proliferation by specific activation of the 75-kDa tumor necrosis factor
receptor. J Immunol 151:4637, 1993
31. Erickson S, de Sauvage FJ, Kikly K, Carver-Moore K, PittsMeek S, Gillett N, Sheehan KCF, Schreiber RD, Goeddel DV, Moore
MW: Decreased sensitivity to tumour-necrosis factor but normal Tcell development in TNF receptor-2-deficient mice. Nature 372:560,
1994
32. Bigda J, Beletsky I, Brakebusch C, Varfolomeev Y, Engelmann H, Bigda J, Holtmann H, Wallach D: Dual role of the p75
tumor necrosis factor (TNF) receptor in TNF cytotoxicity. J Exp
Med 180:445, 1994
33. Totpal K, LaPushin R, Kohno T, Damay BG, Aggarwal BB:
TNF and its receptor antibody agonist differ in mediation of cellular
responses. J Immunol 153:2248, 1994
34. Kriegler M, Perez C, DeFay K, Albert I, Lu SD: A novel
form of TNFkachectin is a cell surface cytotoxic transmembrane
protein: ramifications for the complex physiology of TNF. Cell
53:45, 1988
35. Crowe PD, Walter BN, Mohler KM, Otten-Evans C, Black
RA, Ware C: A metalloprotease inhibitor blocks shedding of the 80
kD TNF receptor and TNF processing in T lymphocytes. J Exp Med
181:1205,1995
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
1995 86: 1836-1841
Both tumor necrosis factor receptors, TNFR60 and TNFR80, are
involved in signaling endothelial tissue factor expression by
juxtacrine tumor necrosis factor alpha
EF Schmid, K Binder, M Grell, P Scheurich and K Pfizenmaier
Updated information and services can be found at:
http://www.bloodjournal.org/content/86/5/1836.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.