| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
2, Fas Ligand, and TNF-Related Apoptosis-Inducing Ligand1
,¶,
*
University of Pittsburgh Cancer Institute and Departments of
Pathology,
Otolaryngology, and
Surgery, University of Pittsburgh School of Medicine, Pittsburgh, PA 15231; and
¶ Institute for Oncology and Radiology, Belgrade, Yugoslavia
research fellow supported by 1-PO DE13059-01 National Institutes of Health Oral Cancer Center Grant.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1
2, Fas ligand,
and TNF-related apoptosis inducing ligand; while cancer cells express
the corresponding death receptors. Disruptions of interactions between
the four ligands expressed on DCs and corresponding death-signaling
receptors expressed on cancer cells using specific Abs or R:Fc fusion
proteins block the cytotoxic activity of DCs directed against cancer
cells. The novel findings suggest that DC killing of cancer cells is
mediated by the concerted engagement of four TNF family ligands of DCs
with corresponding death receptors of cancer cells. Overall, our data
demonstrate that DCs are fully equipped for an efficient direct
apoptotic killing of cancer cells and suggest that this mechanism may
play a critical role in both afferent and efferent anti-tumor
immunity. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In a companion manuscript (47), we have demonstrated that
human immature DCs may directly mediate apoptosis of a diverse array of
tumor cells, but not of normal cells. We have also shown that both cell
membrane-bound and -secreted proteins are mediators of this tumoricidal
activity. In the current study, we have evaluated the molecular
mechanisms by which tumor cell death results from interaction with DCs.
Based on a limited series of publications indicating that mouse
immature DCs and human-activated DCs may express functional Fas ligand
(FasL) and TNF-related apoptosis-inducing ligand (TRAIL; Refs.
19, 20, 21, 22), respectively, and our recent finding of the
activation of TNF family death receptor signaling-specific caspase-8 in
cancer cells exposed to DC killing (47), we presumed that
the newly described tumoricidal activity associated with human immature
DCs is mediated by cytotoxic TNF family ligands. In direct support of
this hypothesis, we obtained novel information that human immature DCs
express TNF, lymphotoxin (LT)-12, FasL,
and TRAIL, and kill cancer cells by the concerted engagement of all
four cytotoxic TNF family ligands. This direct effector function of
immature DCs provides the sentinel APCs with the ability to generate a
source of tumor Ags that may be subsequently processed and presented to
specific T cells by the same or additional host DCs.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following Abs were used in this study: anti-human TNF
(IgG1) mouse mAb (Endogen, Woburn, MA), AO.D12.10 anti-human LT-
(IgG1) mouse mAb (Biogen, Cambridge, MA), H398 anti-human TNFRp60
(IgG2a) mouse mAb (Alexis Biochemicals, San Diego, CA), anti-human
TNFRp80 (IgG2b) rat mAb (Genzyme, Cambridge, MA), B9.C9 anti-human
LT- (IgG1) mouse mAb (Biogen), BDA8 anti-human LT-R (IgG1)
mouse mAb (Biogen), anti-human LT-R goat polyclonal Abs (PCA;
University of California, Riverside, CA), NOK-1 anti-human FasL
(IgG1) mouse mAb (Caltag Laboratories, Burlingame, CA), M3 anti-human
Fas (IgG1) mouse mAb (Immunex, Seattle, WA), anti-human TRAIL
(anti-Apo-2L) rabbit PCA (Calbiochem, San Diego, CA), 32A1380
anti-human TRAILR1 (IgG1) mouse mAb (Imgenex, San Diego, CA),
54B1005 anti-human TRAILR2 mouse mAb (Imgenex), isotype control
mouse mAbs (Caltag Laboratories), normal rabbit serum (Endogen),
biotin-conjugated anti-mouse IgG (H + L) goat PCA (Jackson
ImmunoResearch, West Grove, PA), biotin-conjugated anti-rat IgG
(H + L) goat PCA (Jackson ImmunoResearch), biotin-conjugated
anti-goat IgG (H + L) rabbit PCA, (Vector Laboratories, Burlingame,
CA), CyChrome-conjugated anti-human HLA-DR (IgG1) mouse mAb (BD
PharMingen, San Diego, CA), and all previously described conjugated and
unconjugated Abs used for determination of DC phenotype and
purification of monocytes and DCs (47). PE-conjugated
streptavidin was obtained from Jackson ImmunoResearch.
Streptavidin-HRP (SA-HRP) was purchased from Vector Laboratories.
We also used the following fusion proteins in the performance of this
study: dimeric human TNFRp60:Fc fragment of human IgG1 (Biogen),
dimeric human TNFRp80:Fc fragment of human IgG1 (Immunex), dimeric
human LT-R:Fc fragment of human IgG1 (Biogen), dimeric mouse Fas:Fc
fragment of human IgG1 (R&D Systems, Minneapolis, MN), dimeric human
TRAILR2:Fc fragment of human IgG1 (Alexis Biochemicals), and dimeric
human IL-4R:Fc fragment of human IgG1 (Immunex).
The following cytokines and ligands were used: recombinant human GM-CSF
(Schering-Plough Research Institute, Kenilworth, NJ), recombinant human
IL-4 (Schering-Plough Research Institute), recombinant human trimeric
CD40L (Immunex), recombinant human trimeric FasL (Alexis Biochemicals),
recombinant human trimeric Flag-FasL (Alexis Biochemicals), recombinant
human trimeric TNF (Endogen), recombinant human
LT-12 (Biogen), and recombinant human
trimeric Apo-2L/TRAIL (Genentech, South San Francisco, CA). All
reagents from Immunex, Biogen, UC, Genentech, and Schering-Plough
Research Institute were generous gifts.
Polymerase chain reaction
Two sets of PCR primers were used. Primers of the first set had structures as previously described (23). The second set of primers was prepared to perform nested RT-PCR. The primers were designed using the Primer Express version 1.0 (Applied Biosystems, Foster City, CA). The gene sequences were obtained from the National Center for Biotechnology Information database (Bethesda, MD). The PCR primers and their product lengths were: TNF sense 5'-GGCAGTCAGATCATCTTCTCGA-3' and antisense 5'-TCAGCTTGAGGGTTTGCTACAA-3' (72 bp); FasL sense 5'-TGGCCCATTTAACAGGCAA-3' and antisense 5'-ACAATTCCATAGGTGTCTTCCCA-3' (70 bp); and TRAIL sense 5'-GGGCATTCATTCCTGAGCAA-3' and antisense 5'-CTCCTGAAATCGAAAGTATGTTTGG-3' (105 bp).
RT-PCR for detection of mRNA expression of -actin, TNF, LT-,
LT-, FasL, and TRAIL was performed using their previously
constructed sense/antisense primers and a previously described,
slightly modified technique (23). RNA was extracted from
sorted, highly purified (>98%) populations of monocytes and DCs
(0.5–1 x 106 of each), using the
acid-guanidinium phenol-chlorophorm method. RT-PCR was performed by
using the SuperScript preamplification system for first strand cDNA
synthesis (Life Technologies) and the AmpliTaq Gold (Applied
Biosystems) for PCR.
Nested RT-PCR was performed as follows. The PCR products, obtained as previously described (23), were diluted 10-fold in H2O. A total of 5 µl of the diluted PCR products were used as template for the second round of PCR, using the above described second set of primers. The nested PCR mixture (50 µl) contained 5.5 mM MgCl2, 0.5 µM of each primer, 0.3 mM dNTP, 5 µl of the diluted corresponding first-round PCR product, and 2.5 U AmpliTaq Gold DNA polymerase (Applied Biosystems). The amplification was performed on a PerkinElmer GeneAmp PCR 2400 thermal cycler (PerkinElmer, Norwalk, CT). The reaction was started with a 12-min incubation at 95°C, followed by 40 cycles of 15 s denaturation at 95°C, and finished by 60 s annealing and extension at 60°C. Eight-microliter aliquots of the each amplified product were separated by electrophoresis using 15% polyacrylamide gel (Bio-Rad, Hercules, CA) in Tris-borate EDTA buffer. The gels were stained with SYBR Green I (Sigma Aldrich, St. Louis, MO) diluted 1/10,000. Finally, to prove the specificity of RT-PCR, the amplicons were extracted from the gels and purified with the Concert rapid gel extraction/purification system (Life Technologies) and the obtained DNA was sequenced in an ABI Prism 310 Genetic Analyzer (Applied Biosystems).
Isolation of monocytes and DCs
Monocytes and DCs were purified from normal human peripheral
blood mononuclear leukocytes (PBMNL) as >95% pure populations of
CD5-CD3-TCR--CD19-CD56-CD16-CD14+
and
CD3-TCR--CD14-CD19-CD16-CD56-CD11b-CD4+HLA-DR+
cells, respectively, as described (47).
Generation of immature and mature DCs in culture
Immature and mature DC were generated as described (47). Briefly, immature DCs were produced by 5–7 day culturing of purified blood monocytes in the presence of GM-CSF + IL-4. Mature DCs were obtained by a 2-day incubation of immature DCs in the presence of CD40 ligand (CD40L).
Cell lines
Normal and cancer cell lines used in this study were of human origin and they were maintained in culture as previously described (24 , 47).
Cytotoxicity assays
MTT cytotoxicity assays were performed as previously described
(25 , 47). Killing mediated by soluble TNF
family ligands was assessed by addition of various dilutions of rTNF,
LT-12 and FasL, TRAIL, and CD40L to
target cells either individually or in various combinations, as
previously described (23). Blocking of the interactions
between the TNF family ligands and their receptors was performed as
previously described (23).
Flow cytometry
The analysis was performed using a FACScan (BD Biosciences, Mountain View, CA) flow cytometer. Phenotypes of monocytes and DCs were determined following their direct two-color staining with fluorochrome-conjugated mAbs specific for the lineage markers of T cells, B cells, NK cells, monocytes, and DCs as described (26). The cell surface or intracellular protein expression of TNF family ligands and their receptors was assessed using a previously described, highly sensitive, three-step flow cytometry technique (23). To assess expression of TNF family ligands on circulating DCs, three-color flow cytometry was performed with total PBMNL following their specific staining with Abs. First, the TNF family ligands were labeled with nonconjugated specific Abs, biotin-conjugated secondary Abs, and PE-conjugated streptavidin as described (23). Second, PBMNL were simultaneously labeled with CyChrome-conjugated anti-HLA-DR mAb and FITC-conjugated anti-CD3, anti-CD56, anti-CD14, and anti-CD19 mAbs in the presence of excess of nonconjugated normal mouse Ig (200 µg/ml). Following the staining procedures, cells were fixed with 1% paraformaldehyde and analyzed. One- and two-color flow cytometry analyses were performed as previously described (23, 26). Three-color flow cytometry was performed by gating DCs (HLA-DR+/lineage marker- PBMNL) and their analyzing for the expression of TNF family ligands. Specific reactivity of an Ab and levels of a cellular Ag expression were determined by comparative analyses of cell fluorescence and ratios between the mean fluorescence intensities of cells stained with specific Abs and isotype-matched control Abs, respectively.
Amplified immunocytofluorescence and confocal microscopy
DCs were resuspended in PBS containing 1% FCS and sequentially stained in cell suspensions with primary Abs against TNF family ligands (1 µg/ml), biotin-conjugated goat anti-mouse Ig Abs (1 µg/ml), SA-HRP (1/100 dilution), and cyanine 3-conjugated tyramide (NEN, Boston, MA). After each of these stainings, DCs were washed twice in PBS plus 1% FCS. Following the staining with tyramide and subsequent washing, DCs were resuspended in 25 µl of PBS containing 5 µg/ml of acridine orange (Sigma Aldrich), applied on poly-L-lysine-treated slides, and immediately examined by a Leica TCS-NT laser confocal microscope (Leica, Allendale, NJ). Two-color analysis was performed. The specimens were illuminated at 514 nm, the emitted image was split by a 565-nm beam splitter and collected simultaneously by two photomultipliers through a 535-nm (acridin orange) and 600-nm (cyanine 3) final barier filter. Optical sections were cut with a 50 µm confocal pinhole that ensures maximal Z resolution. In reconstructed series, images were collected every 1 µm. Reconstruction was performed with ImageSpace Software (Molecular Dynamics).
Statistical analysis
Statistical analyses of the results were performed using the Wilcoxon’s signed-rank pair tests. To determine whether data provide evidence for differences in profiles of percent killing as a function of E:T ratio or ligand concentration, multivariate permutation methods were used. Differences were considered significant when p < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In a companion manuscript (47), we have demonstrated
that immature DCs can efficiently induce death of cancer cells in vitro
using an apoptotic mechanism that involves both cell membrane-bound and
-secreted cytotoxic molecules. Previous studies have shown that mouse
and human DCs may express functional FasL and TRAIL
(19, 20, 21, 22), respectively. In addition, our recent findings
have indicated that the TNF family death receptor signaling-specific
caspase-8 is activated in cancer cells exposed to DC cytotoxic activity
(47). Based on these findings, we hypothesized that human
immature DCs might express and use to kill cancer cells cytotoxic TNF
family ligands. Therefore, to determine the molecules which potentially
mediate killing of cancer cells by DCs, we first examined whether
freshly isolated blood immature DCs as well as in vitro generated,
monocyte-derived, immature and mature DCs express mRNAs encoding TNF,
LT-, LT-, FasL, and TRAIL. Freshly isolated blood monocytes,
precursors of DCs, were also tested. Monocytes (Fig. 1
A) and blood immature DCs
(Fig. 1B) as well as both in vitro-generated immature (Fig. 1C) and mature (Fig. 1D) DCs consistently
expressed mRNAs encoding each of these ligands. However, while the
mRNAs could be detected in monocytes and in in vitro-generated
monocyte-derived DCs using a standard RT-PCR, they could be found in
blood DCs only by using a more sensitive nested RT-PCR. This finding
indicates that blood DCs might express lower levels of the TNF family
ligand mRNAs than monocytes and their in vitro-generated DC
derivatives.
|
, intact). In addition, monocytes of
some individuals exhibited low levels of cell surface FasL. In
contrast, both in vivo- (Fig. 3) and in
vitro- (Fig. 4, intact) generated immature DCs
consistently expressed all five TNF
family ligands on their cell surface. Furthermore, based on different
levels of expression of HLA-DR, blood-circulating DCs could be
classified into two subsets, HLA-DRlow
("immature") and HLA-DRint ("mature")
DCs, which also displayed different levels of cell membrane-bound TNF
family ligands. Thus, the former subset expressed higher levels of the
ligands than the latter (Fig. 3). Monocyte-derived mature DCs, like
monocytes, expressed only low levels of TNF on their cell surface (Fig. 5, intact). The cell surface levels of
TNF family ligands were relatively low, even on immature DCs, and could
be consistently detected only by using an amplified flow cytometry
method. In contrast, when the cells were permeabilized, both types of
DCs as well as monocytes were strongly and equivalently stained with
the five ligand-specific probes (Figs. 2, 4, and 5, permeabilized).
This indicates that the proteins of five cytotoxic TNF family ligands
were expressed at high and similar quantities in the cytoplasm of all
three types of APCs. The finding of coexpression of LT- and - on
the cell surface of DCs indicates that the membrane-bound
LT-12 heterotrimer might be present on
these cells. These data are in a full accordance with those for mRNA
expression (Fig. 1
, B and C), and provide clear
evidence that human immature DCs express on their cell surface at least
four cytotoxic TNF family ligands, including TNF,
LT-12, FasL, and TRAIL.
|
|
|
|
, FasL was not evenly
distributed on the cell surface, but it appeared to be principally
arrayed on the protrusions and dendrites of the cell membrane. Similar
data were obtained in analyses of the other cell-surface-expressed TNF
family ligands (data not shown). The selective expression of such
ligands on DC dendrites might increase the focal density and lethality
of these molecules in intimate contact points of DC-tumor cell
conjugates.
|
Based on our previous finding that DC conditioned media contain
tumoricidal proteins (47), we analyzed such media for the
presence of soluble TNF and FasL, using specific ELISAs. We observed
the presence of substantial levels of both ligands in the supernatants
of both immature and mature DCs (Table I). However, significantly higher levels
of these ligands were present in the conditioned culture media of
mature vs immature DCs.
|
TNF family ligands mediate death in susceptible target cells by
interacting with their cell surface-bound TNF family death receptors.
Therefore, if this ligand-receptor interaction is a physiologically
relevant molecular mechanism used by DCs to kill cancer cells, then
cancer cells susceptible to killing by DCs should express at least a
subset of these TNF family receptors. In support of this model, we have
previously demonstrated (23), using flow cytometry, that
cells of a variety of human cancer cell lines express multiple TNF
family receptors, including TNFR1, TNFR2, LT-R, and Fas. In the
present study, we extended this investigation and demonstrated that
cancer cell lines also expressed TRAILR1 and TRAILR2 (Fig. 7). However, while TNFR1, LT-R, Fas,
and TRAILR1 were consistently expressed on all 20 cancer cell lines
tested, TNFR2 and TRAILR2 were significantly expressed only on a
proportion of these lines. Interestingly, cancer cell lines, which did
not significantly express TNFR2 and TRAILR2 on their cell surface,
generally showed lower susceptibility to killing by DCs than those
which significantly expressed these receptors. Such examples were lung
cancer 226 and Daudi compared with lung cancer 596 and Jurkat,
respectively (47) (data not shown). These data indicate
that TNFR2 and TRAILR2 may have a significant role in DC-mediated
killing of cancer cells. In addition, we determined that freshly
isolated noncultured tumor cells (e.g., squamous cell carcinoma of head
and neck) expressed all the TNF family receptors found on cultured
cancer cells. The pattern of expression of the death receptors on
freshly-isolated cancer cells was similar to that described above for
cultured cancer cells (data not shown). In contrast to the large
majority of cancer cell lines and freshly isolated tumor cells, which
consistently showed both susceptibility to DC-mediated killing
(47) and cell surface expression of multiple TNF family
death receptors, normal cells such as DCs, T cell blasts, fibroblasts,
mammary epithelial cells, melanocytes, and keratinocytes were resistant
to killing by DCs (47) and consistently expressed only
LT-R and Fas, but either did not express or expressed at very low
levels, TNFR1, TNFR2, TRAILR1, and TRAILR2 on their cell
surfaces (Fig. 8, intact). However,
significant levels of these molecules were observed in the
permeabilized cells (Fig. 8, permeabilized). Among several cancer cell
lines tested, only the Pittsburgh Cancer Institute-30 SCCHN line
showed a pattern of expression of TNF family receptors
(data not shown) and resistance to killing by DCs (47),
which was similar to that of normal cells. These data demonstrate that
cancer cells express a more diverse set and/or higher levels of TNF
family death receptors on their cell surface than normal cells.
Therefore, DCs and cancer cells might be able to effectively interact
via a range of complementary TNF family ligands and receptors,
respectively, resulting in tumor cell apoptosis.
|
|
To directly test the above possibility, we performed blocking
experiments in which we attempted to interfere with the killing of
cancer cells by DCs using blocking Abs or R:Fc fusion proteins specific
for the TNF family ligands or their receptors. Immature DCs were
preincubated with individual or combined Abs or R:Fc fusion proteins
specific for TNF, LT-, LT-, LT-12,
FasL, and TRAIL, while cancer cells were preincubated with individual
or combined Abs specific for TNFR1, TNFR2, LT-R, Fas, TRAILR1, and
TRAILR2. The effects of these treatments on DC-mediated killing of
cancer cells were assessed, using 3-h MTT cytotoxicity assays.
Representative data of 1 of 10 similar experiments performed are shown
in Fig. 9. We observed that each
individual reagent which blocks interactions between either TNF-TNFRs,
LT-12-LT-R, FasL-Fas, or TRAIL-TRAILRs
consistently, but partially inhibited the induction of cell death in
tumor cells mediated by immature DCs. The levels of inhibition were
generally similar for different ligand-receptor pairs, and using
optimal blocking reagents such as R:Fc fusion proteins, consistently
ranged between 40 and 50% (Fig. 9B). Importantly,
simultaneous disruption of the all four ligand-receptor pairs by the
combined treatment of DCs with the TNFR1:Fc, LT-R:Fc, Fas:Fc, and
TRAILR2:Fc fusion proteins (Fig. 9B) or cancer cells with
the anti-TNFR1, anti-TNFR2, anti-LT-R, anti-Fas,
anti-TRAILR1, and anti-TRAILR2 mAbs (Fig. 9C) more
efficiently blocked DC-mediated killing of cancer cells than the
individual disruption of the corresponding ligand-receptor pairs, and
decreased this activity by 
75%. Control isotype-matched
(nonreactive) mAbs, normal human Ig, or IL-4R:Fc fusion protein failed
to suppress DC-mediated killing of cancer cells. These data show that
most of DC tumoricidal activity is mediated by a coordinate engagement
of TNF, LT-12, FasL, and TRAIL, and that
all four cytotoxic TNF family ligands are critical mediators and equal
contributors in this activity. The results may also indicate a minor
participation of additional, yet unknown, cytotoxic molecules in DC
tumoricidal activity.
|
12 and Apo-2L/TRAIL) or three (i.e.,
TNF, LT-12, and FasL; TNF,
LT-12, and Apo-2L/TRAIL; or TNF, FasL,
and Apo-2L/TRAIL), able to engage at least three death receptors.
However, tumor cell death was not effectively induced when these
reagents were applied individually or in combinations of two (TNF and
FasL, TNF and LT-12, or FasL and
LT-12), capable of engaging only one or
two death receptors (Ref. 23 , data not shown). The ligands
showed similar activity in a wide range of concentrations (from 0.01 to
50.00 ng/ml). These data show that cancer cells express functional
TNFR1, TNFR2, LT-R, Fas, TRAILR1, and TRAILR2, which are able to
signal apoptosis when they are simultaneously and coordinately engaged.
They also complement the above results of blocking experiments and
additionally support the possibility that DCs mediate the apoptotic
killing of cancer cells via a simultaneous and coordinate engagement of
their multiple TNF family ligands with corresponding death receptors of
cancer cells. The findings also indicate that normal cells are spared
from killing by DCs, because they express only one or two TNF family
death receptors which engagement is not sufficient to efficiently
signal apoptosis. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
TNF, LT-12, FasL, and TRAIL are potent
cytotoxic TNF family ligands with the unique function of inducing
apoptosis in susceptible cancer cells (27, 28, 29). All four
ligands are expressed and used to kill cancer cells by professional
cytotoxic cells, such as CD8+ CTLs and NK cells
(23, 27, 28, 29, 30, 31, 32, 33, 34, 35). FasL, TNF, and TRAIL are also expressed by
activated CD4+ T cells, B cells, and macrophages
(27, 28, 31, 36, 37, 38, 39, 40, 41, 42). In addition, it has been shown that
FasL may be expressed by mouse immature DCs (19, 20),
while TRAIL and TNF by human-activated DCs (21, 22, 43).
Here, we first demonstrate that human immature DCs (both freshly
isolated from the blood and in vitro generated by GM-CSF plus IL-4
stimulation of monocytes) express TNF,
LT-12, FasL, and TRAIL on their cell
membrane. In contrast, monocytes and mature DCs (generated by CD40L
stimulation of immature DCs) express only significant levels of TNF on
their cell surface. However, high levels of all four ligands were
detected in the cytoplasm of each of these cell types. This observation
suggests that appropriate stimulation of the APCs might rapidly induce
the translocation to the cell surface and/or the release into
extracellular microenvironment of these intracellular molecules.
Similar to in vitro-generated immature and mature DCs, freshly isolated
blood HLA-DRlow (immature) DCs express higher
cell surface levels of the cytotoxic TNF family ligands than their
HLA-DRint (mature) counterparts. In addition, in
vitro-generated mature DCs produced higher levels of soluble TNF and
FasL than immature DCs. Therefore, the cell surface expression and
secretion of these cytotoxic ligands appears to be tightly regulated,
and differentiation stage-dependent within the DC "lineage". The
possible regulatory mechanism might include activities of cytokines
and/or metalloproteases (27, 28). This pathway has been
currently scrutinized in our laboratory.
Previous studies have indicated that FasL (19, 20) and
TRAIL (21, 22) may be used by mouse immature DCs and
IFN-activated human DCs, respectively, to kill rare susceptible target
cells. Our study shows that a diverse array of human cancer cell lines
as well as freshly isolated cancer cells express multiple TNF family
death receptors (i.e., TNFR1 and/or TNFR2, LT-R, Fas, TRAILR1 and/or
TRAILR2) and are efficiently killed by DCs. However, the susceptibility
of cancer cells to killing by DCs appears to be dependent on the
multiplicity and levels of cell membrane expression of their deadly
receptors. In contrast, normal cells either do not express or express
very low levels of TRAILR1, TRAILR2, TNFR1, and/or TNFR2 and are
resistant to killing mediated by DCs (47). This suggests
that cell surface expression of multiple TNF family death receptors on
target cells may define their susceptibility to killing by DCs.
Individually applied rTNF, LT-12, FasL,
or TRAIL are minimally cytotoxic to most cancer cell lines that we have
evaluated (data not shown). In addition, monocytes, which express only
TNF, are not discernibly cytotoxic against cancer cells
(47), while immature DCs, which express multiple cell
membrane-bound TNF family cytotoxic ligands, are cytotoxic for most
cancer cell lines. This indicates that cell surface expression of
multiple cytotoxic TNF family ligands may determine the tumoricidal
activity of DCs. Individual and especially combined disruption of
interactions between TNF, LT-12, FasL, or
TRAIL and their receptors notably inhibits DC-mediated killing of
cancer cells. In contrast, simultaneous application of two or more
recombinant cytotoxic TNF family ligands, which can engage at least
three death receptors, rapidly induces apoptosis of cancer cells. Taken
together, these findings show that human immature DCs kill cancer cells
(47) by the simultaneous and coordinate engagement of
their multiple TNF family ligands with the corresponding multiple death
receptors on cancer cells.
These findings also point out that cancer cells are relatively resistant to apoptosis mediated through individual TNF family ligands and suggest that tumors may use anti-apoptotic mechanisms as survival skills. These mechanisms may be related to the low levels of cell surface expression of the receptors and/or to the activity of intracellular anti-apoptotic molecules (27, 43). Indeed, we observed that the TNF family death receptors are expressed on the cell surface of cancer cells at low levels, which are unfavorable for their ligand-mediated homotrimerization necessary for generation of an effective apoptotic signal. Under such conditions, the coordinate engagement of at least three different death receptors, sharing a domain of homology involved in generation of common proapoptotic signals, may result in their heterotrimerization leading to triggering of tumor cell apoptosis (23). This hypothesis is circumstantially supported by the presence of common death domain in four of the six death receptors expressed on cancer cells (TNFR1, Fas, TRAILR1, and TRAILR2; Refs. 27, 28).
Cell membrane-bound forms of cytotoxic TNF family ligands have been observed to be more cytotoxic than their soluble forms, that are frequently nonfunctional and may serve as functional antagonists of their cell membrane-bound forms (27, 28). In our study, mature DCs produced higher levels of the secreted forms of TNF and FasL than immature DCs and expressed lower levels or lack the cell membrane-bound forms. Consistent with these findings, mature DCs were less effective at inducing apoptosis against cancer cells than immature DCs (47).
This maturation-dependent change in expression of cytotoxic TNF family ligands (in terms of forms and function) might be also relevant to the maturation-dependent changes of Ag-presenting function of DCs. In this way, immature DCs most effectively prepare (by inducing apoptosis of tumor cells), capture, and process exogenous Ags (such as dying tumor cells), while mature Ag-loaded DCs are most predisposed to present tumor Ags and to promote the activation and expansion of Ag-specific T cells via direct cell-to-cell contact. Depending on the signals they receive (complete or incomplete), Ag-specific T cells are believed to differentiate into two different cell populations, one that is relatively resistant, and one that is relatively sensitive to TNF family ligand-induced apoptosis, respectively (44, 45, 46). Therefore, the low apoptosis-inducing ability of mature DCs might serve as a safeguard against the inappropriate death of clinically-important Ag-specific T cells (39).
We have shown that not only in vitro generated, monocyte-derived, but also freshly isolated peripheral blood immature DCs express the four major cytotoxic TNF family ligands and exhibit a marked apoptosis-inducing anticancer activity (47). Thus, this cytotoxic activity might be operative in vivo and may be critically and directly involved in the elimination of evolving tumors as well as in the control of tumor growth in situ. In addition to these direct anticancer activities, the ability of immature DCs to promote tumor apoptosis clearly suggests their pivotal role in the initiation of processing and, subsequently, cross-presentation of cellular tumor Ags and generation of specific anti-tumor T effector cells believed to be critical to effective immunotherapy of cancer (7, 17).
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Nikola L. Vujanovic, University of Pittsburgh Cancer Institute, Biomedical Science Tower W1046, 211 Lothrop Street, Pittsburgh, PA 15213. E-mail address: vujanovicnl{at}msx.upmc.edu
3 Abbreviations used in this paper: DC, dendritic cell; LT, lymphotoxin; TRAIL, TNF-related apoptosis-inducing ligand; FasL, Fas ligand; PCA, polyclonal Abs; PBMNL, peripheral blood mononuclear leukocytes; SA-HRP, streptavidin HRP; int, intermediate; CD40L, CD40 ligand.
Received for publication July 16, 2001. Accepted for publication December 5, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
v5 and CD36, and cross-present antigens to cytotoxic T lymphocytes. J. Exp. Med. 188:1359.-related apoptosis-inducing ligand in the enhanced cytotoxicity of IFN--stimulated human dendritic cells to tumor cells. J. Immunol. 166:5407. receptor. J. Biol. Chem. 272:30835. induces both cell lysis and apoptosis in susceptible target cells. Cell. Immunol. 171:102.[Medline]
are involved in the cytotoxic activity of B cells from tumor-bearing mice against tumor targets. J. Immunol. 152:3333.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  M. Terme, G. Mignot, E. Ullrich, M. Bonmort, V. Minard-Colin, A. Jacquet, J. L. Schultze, G. Kroemer, C. Leclerc, N. Chaput, et al. The Dendritic Cell-like Functions of IFN-Producing Killer Dendritic Cells Reside in the CD11b+ Subset and Are Licensed by Tumor Cells Cancer Res., August 15, 2009; 69(16): 6590 - 6597. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Himoudi, M. Yan, G. Bouma, D. Morgenstern, R. Wallace, B. Seddon, J. Buddle, A. Eddaoudi, S. J. Howe, N. Cooper, et al. Migratory and Antigen Presentation Functions of IFN-Producing Killer Dendritic Cells Cancer Res., August 15, 2009; 69(16): 6598 - 6606. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Strauss, J. A. Lindquist, N. Arhel, E. Felder, S. Karl, T. L. Haas, S. Fulda, H. Walczak, F. Kirchhoff, and K.-M. Debatin CD95 co-stimulation blocks activation of naive T cells by inhibiting T cell receptor signaling J. Exp. Med., June 8, 2009; 206(6): 1379 - 1393. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Chyou, E. H. Ekland, A. C. Carpenter, T.-C. J. Tzeng, S. Tian, M. Michaud, J. A. Madri, and T. T. Lu Fibroblast-Type Reticular Stromal Cells Regulate the Lymph Node Vasculature J. Immunol., September 15, 2008; 181(6): 3887 - 3896. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. S. Hill, F. Errington, L. P. Steele, A. Merrick, R. Morgan, P. J. Selby, N. T. Georgopoulos, D. M. O'Donnell, and A. A. Melcher OK432-Activated Human Dendritic Cells Kill Tumor Cells via CD40/CD40 Ligand Interactions J. Immunol., September 1, 2008; 181(5): 3108 - 3115. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Chauvin and R. Josien Dendritic Cells as Killers: Mechanistic Aspects and Potential Roles J. Immunol., July 1, 2008; 181(1): 11 - 16. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Wei, H. Wang, B. Lu, B. Li, S. Hou, W. Qian, K. Fan, J. Dai, J. Zhao, and Y. Guo Cancer Immunotherapy Using In vitro Genetically Modified Targeted Dendritic Cells Cancer Res., May 15, 2008; 68(10): 3854 - 3862. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Papewalis, B. Jacobs, M. Wuttke, E. Ullrich, T. Baehring, R. Fenk, H. S. Willenberg, S. Schinner, M. Cohnen, J. Seissler, et al. IFN-{alpha} Skews Monocytes into CD56+-Expressing Dendritic Cells with Potent Functional Activities In Vitro and In Vivo J. Immunol., February 1, 2008; 180(3): 1462 - 1470. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. M. Srivastava, C. Varalakshmi, and A. Khar The Ischemia-Responsive Protein 94 (Irp94) Activates Dendritic Cells through NK Cell Receptor Protein-2/NK Group 2 Member D (NKR-P2/NKG2D) Leading to Their Maturation J. Immunol., January 15, 2008; 180(2): 1117 - 1130. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Stary, C. Bangert, M. Tauber, R. Strohal, T. Kopp, and G. Stingl Tumoricidal activity of TLR7/8-activated inflammatory dendritic cells J. Exp. Med., June 11, 2007; 204(6): 1441 - 1451. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. M. Srivastava, Ch. Varalakshmi, and A. Khar Cross-linking a mAb to NKR-P2/NKG2D on dendritic cells induces their activation and maturation leading to enhanced anti-tumor immune response Int. Immunol., May 1, 2007; 19(5): 591 - 607. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Xu, A. K. Chakrabarti, J. L. Tan, L. Ge, A. Gambotto, and N. L. Vujanovic Essential role of the TNF-TNFR2 cognate interaction in mouse dendritic cell-natural killer cell crosstalk Blood, April 15, 2007; 109(8): 3333 - 3341. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Hoene, M. Peiser, and R. Wanner Human monocyte-derived dendritic cells express TLR9 and react directly to the CpG-A oligonucleotide D19 J. Leukoc. Biol., December 1, 2006; 80(6): 1328 - 1336. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Huang, T. Tatsumi, E. Pizzoferrato, N. Vujanovic, and W. J. Storkus Nitric Oxide Sensitizes Tumor Cells to Dendritic Cell-Mediated Apoptosis, Uptake, and Cross-Presentation Cancer Res., September 15, 2005; 65(18): 8461 - 8470. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Trinite, C. Chauvin, H. Peche, C. Voisine, M. Heslan, and R. Josien Immature CD4-CD103+ Rat Dendritic Cells Induce Rapid Caspase-Independent Apoptosis-Like Cell Death in Various Tumor and Nontumor Cells and Phagocytose Their Victims J. Immunol., August 15, 2005; 175(4): 2408 - 2417. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Schmitz, S. Zhao, Y. Deuse, K. Schakel, R. Wehner, H. Wohner, K. Holig, F. Wienforth, A. Kiessling, M. Bornhauser, et al. Tumoricidal Potential of Native Blood Dendritic Cells: Direct Tumor Cell Killing and Activation of NK Cell-Mediated Cytotoxicity J. Immunol., April 1, 2005; 174(7): 4127 - 4134. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Makarenkova, A. K. Chakrabarti, J. A. Liberatore, P. Popovic, G. Lu, S. Watkins, and N. L. Vujanovic Dendritic cells and natural killer cells interact via multiple TNF family molecules J. Leukoc. Biol., March 1, 2005; 77(3): 408 - 413. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Mocellin, M. Provenzano, C. R. Rossi, P. Pilati, R. Scalerta, M. Lise, and D. Nitti Induction of Endothelial Nitric Oxide Synthase Expression by Melanoma Sensitizes Endothelial Cells to Tumor Necrosis Factor-Driven Cytotoxicity Clin. Cancer Res., October 15, 2004; 10(20): 6879 - 6886. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Akasaki, G. Liu, N. H. C. Chung, M. Ehtesham, K. L. Black, and J. S. Yu Induction of a CD4+ T Regulatory Type 1 Response by Cyclooxygenase-2-Overexpressing Glioma J. Immunol., October 1, 2004; 173(7): 4352 - 4359. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Okada, T. Tsugawa, H. Sato, N. Kuwashima, A. Gambotto, K. Okada, J. E. Dusak, W. K. Fellows-Mayle, G. D. Papworth, S. C. Watkins, et al. Delivery of Interferon-{alpha} Transfected Dendritic Cells into Central Nervous System Tumors Enhances the Antitumor Efficacy of Peripheral Peptide-Based Vaccines Cancer Res., August 15, 2004; 64(16): 5830 - 5838. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Krueger and K. Callis Potential of Tumor Necrosis Factor Inhibitors in Psoriasis and Psoriatic Arthritis Arch Dermatol, February 1, 2004; 140(2): 218 - 225. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Tatsumi, J. Huang, W. E. Gooding, A. Gambotto, P. D. Robbins, N. L. Vujanovic, S. M. Alber, S. C. Watkins, H. Okada, and W. J. Storkus Intratumoral Delivery of Dendritic Cells Engineered to Secrete Both Interleukin (IL)-12 and IL-18 Effectively Treats Local and Distant Disease in Association with Broadly Reactive Tc1-type Immunity Cancer Res., October 1, 2003; 63(19): 6378 - 6386. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Stebbing, B. Gazzard, S. Portsmouth, F. Gotch, L. Kim, M. Bower, S. Mandalia, R. Binder, P. Srivastava, and S. Patterson Disease-associated dendritic cells respond to disease-specific antigens through the common heat shock protein receptor Blood, September 1, 2003; 102(5): 1806 - 1814. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Hoves, S. W. Krause, D. Halbritter, H.-G. Zhang, J. D. Mountz, J. Scholmerich, and M. Fleck Mature But Not Immature Fas Ligand (CD95L)-Transduced Human Monocyte-Derived Dendritic Cells Are Protected from Fas-Mediated Apoptosis and Can Be Used as Killer APC J. Immunol., June 1, 2003; 170(11): 5406 - 5413. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
98% pure. Monocytes and blood immature DCs were
obtained by cell sorting of CD14+ or
HLA-DR+CD3-CD56-CD14-CD19-
PBMNL, respectively. Immature and mature DCs were generated in vitro by
culturing of purified blood monocytes in the presence of GM-CSF + IL-4
or GM-CSF + IL-4 + CD40L, respectively. Expression of the mRNAs in
monocytes and monocyte-derived DCs was assessed by a standard RT-PCR,
while the mRNA expression in blood DCs was examined using a nested
RT-PCR. The data are from a representative experiment of five
performed. The presented results are amplified cDNA products separated
by PAGE and detected by SYBR Green fluorescent dye. A,
Monocytes; B, freshly-isolated blood immature DCs;
C, in vitro generated, monocyte-derived, immature DCs;
D, in vitro generated, monocyte-derived, mature
DCs.
