1. Art and Diseño

Publication of the International Union Against Cancer
Int. J. Cancer: 106, 827– 835 (2003)
© 2003 Wiley-Liss, Inc.
IL-18 IS PRODUCED BY PROSTATE CANCER CELLS AND SECRETED IN
RESPONSE TO INTERFERONS
Sophie LEBEL-BINAY1,2, Nicolas THIOUNN3, Gonzague DE PINIEUX4, Annick VIEILLEFOND4, Bernard DEBRE´ 3, Jean-Yves BONNEFOY2,
Wolf-Herman FRIDMAN1,5 and Franck PAGE` S1,5*
1
INSERM U 255, Centre de Recherches Biome´dicales des Cordeliers, Paris, France
2
Centre d’Immunologie Pierre Fabre, Saint-Julien en Genevois Cedex, France
3
Service d’Urologie, Hoˆpital Cochin, AP-HP, Paris, France
4
Service d’Anatomie Pathologique, Hoˆpital Cochin, AP-HP, Paris, France
5
Service d’Immunologie Biologique, Hoˆpital Europe´en Georges Pompidou, AP-HP, Paris, France
Murine models have shown that IL-18 has antiangiogenic
and antitumor effects, but little is known about IL-18 production in human tumors. We investigated IL-18 expression in
clinically localized prostate cancers by immunohistochemistry and showed that 75% of the prostate cancers studied
(27/36 cases) presented with tumor cells producing IL-18.
Prostate tumor cell lines PC-3, DU 145 and LNCaP synthesized the immature form of IL-18 (p24). IFN-␥ produced in
prostate cancers induced caspase-1 mRNA and IL-18 secretion of tumor cell lines, which was inhibited by the cellpermeable Tyr-Val-Ala-Asp-aldehyde caspase-1 inhibitor
(YVAD-CHO). Interestingly, IFN-␣ also induced IL-18 secretion of the poorly differentiated cell line PC-3. PC-3 and DU
145, but not the well-differentiated cell line LNCaP, expressed IL-18R␣ (IL-1Rrp) protein and transcripts for IL18R␤ (AcPL). Exogenous IL-18 increased mitochondrial activity of both cell lines evaluated by the tetrazolium (MTT)
assay but did not influence their proliferation. This indicated
that prostate tumor cells could secrete IL-18 in response to
IFN-␥ in the tumor microenvironment and that IL-18 could
act as a autocrine/paracrine factor for the tumor. In the
cohort of patients studied, IL-18 expression in prostate cancers (with up to 10% of tumor cells stained) was associated
with a favorable outcome and equally predictive as pathologic stage on multivariate analysis (log rank test, p ‫ ؍‬0.02).
Tumor IL-18 production is a novel physiopathologic feature
of prostate cancer and appears to be a favorable event in the
course of the disease. Modulation of IL-18 production by
interferons could have a beneficial clinical effect, which deserves further investigation.
© 2003 Wiley-Liss, Inc.
Key words: IL-18; caspase-1; prostate cancer; prognosis; IFN-␥;
IFN-␣
IL-18 was first identified as an IGIF based on its ability to
induce high levels of IFN-␥ secretion by both NK cells and Th1
clones.1,2 IL-18 belongs to the IL-1 family,3 lacks a signal sequence2 and is processed into an 18 kDa mature form by caspase1.4 It is mainly produced by macrophages and dendritic cells.
However, we and others have shown that IL-18 is also synthesized
by nonimmune cells.5 IL-18 potentiates IL-12-induced Th1 development6,7 and plays an important role in T-cell proliferation.2 In
addition, it enhances FasL-mediated cytotoxicity of NK cells and
Th1 cells8,9 and has proinflammatory properties by inducing chemotactic molecules for macrophages, polymorphonuclear neutrophils and inflammatory cytokines such as TNF-␣ or IL-1␤.10 –12
Murine models have demonstrated the antitumor activity of
IL-18, either by systemic administration13,14 or in tumors genetically modified to constitutively express IL-18.15 Antitumor effects
are mediated by IL-18-induced IFN-␥, NK cells and T CD4ϩ
Fas-dependent cytotoxicity.13–16 In vitro, IL-18 added to cultures
of tumor cells, NK cells, dendritic cells and T cells induces tumor
cell death by enhancing NK cell cytotoxicity, which activates
dendritic cells to promote a specific CTL response.17 This could
position IL-18 as an important bridge between innate and adaptive
antitumor immune responses. Some antitumor effects of IL-18
may also be mediated via nonimmunologic mechanisms as IL-18
has antiangiogenic properties in vitro and immunohistochemistry
studies have revealed hypovascularization of IL-18-treated tumors.15,18
Little is known about the in vivo presence of IL-18 in human
tumors. In a previous study, we observed that IL-18 protein was
expressed at various levels in colon cancers and that loss of
transcripts at the tumor site of the 2 downstream targets of IL-18,
i.e., IFN-␥ and FasL, was associated with the concomitant presence of distant metastases.19 These results suggested a role of
IL-18 in the control of tumor spreading in humans, in line with the
antitumor properties of IL-18 observed in murine models.
Research in prostate cancer has focused on the role of IFN-␥ in
tumor progression. In vitro, IFN-␥ causes cycle arrest of prostate
tumor cell lines, induces the cyclin-dependent kinase inhibitor
p21WAF1, downregulates neu/HER-220 and decreases the metastatic potential of some prostate cancer cell lines.21 IFN-␥ is
produced by stromal and epithelial cells in normal prostate and
prostate cancers,22 and analysis of the transcriptional profile of
prostate cancers by Affymetrix (Santa Clara, CA) Genechip technology has shown that about 30% of malignant tumors present
downmodulation of IFN-␥-inducible molecules.23 This may indicate modulation of IFN-␥ in tumors, but no information is available about the existence of the IFN-␥-inducing cytokine IL-18 in
normal prostate and prostate cancers. We therefore investigated the
IL-18 status of prostate cancers and its association with clinical
outcome.
Abbreviations: CI, confidence interval; CTL, cytotoxic T lymphocyte;
DAB, diaminobenzidine; ECL, enhanced chemiluminescence; HRP, horseradish peroxidase; IGIF, IFN-␥-inducing factor; IL-18R, IL-18 receptor;
MAb, monoclonal antibody; NK, natural killer; PSA, prostate-specific
antigen; RR, relative risk; STAT, signal transducer and activator of transcription; TBS, TRIS-buffered saline; Th1, T-helper 1; TNF, tumor necrosis factor.
Grant sponsor: Centre d’Immunologie Pierre Fabre; Grant sponsor:
Association pour la Recherche sur le Cancer: Pˆole d’Etude du Microenvironnement Tumoral.
The first 2 authors contributed equally to this work.
*Correspondence to: Service d’Immunologie Biologique, Hˆopital Europ´een Georges Pompidou, 20-40 rue Leblanc, 75 908 Paris Cedex 15,
France. Fax: ϩ33-1-56-09-20-80.
E-mail: [email protected]
Received 9 August 2002; Revised 2 January, 24 February 2003; Accepted 3 March 2003
DOI 10.1002/ijc.11285
828
LEBEL-BINAY ET AL.
MATERIAL AND METHODS
Patients and clinical features
A total of 43 patients with localized prostate carcinoma treated
by radical prostatectomy at Cochin Hospital, Paris, were included.24 Gleason score and pathologic TNM stage were reevaluated in
each case by 2 pathologists and a urologist. Patients had newly
diagnosed tumors and did not receive any adjuvant treatment after
surgery. Bone metastases were assessed by bone X-ray and bone
scan. Extraosseous metastases were assessed by surgical biopsy.
Recurrence was defined as a significant elevation of PSA and/or
new symptoms due to local tumor recurrence. Of the 36 patients
selected (7 cases excluded for insufficient quality of IL-18 staining), 6 died from prostatic disease, 14 had a favorable clinical
outcome with no recurrence or metastases and 22 relapsed, with
bone metastases in 11 cases. Median follow-up was 95 months
(range 75–115).
Immunohistochemistry
Archival formalin-fixed, paraffin-embedded tissues were available for all 43 lesions. Ten specimens of normal prostate and 15
specimens of typical benign prostatic hyperplasia were also selected. A single morphologically representative block was selected
per case. Sections (5 ␮m) were air-dried, deparaffinized and heated
3 times for 5 min in a microwave oven in citrate buffer (pH 6.0).
Sections were first incubated for 5 min with 3% hydrogen peroxide
aqueous solution to quench endogenous peroxidase activity. Goat
polyclonal IgG anti-IL-18 was used for IL-18 detection (R&D
Systems, Abingdon, UK). This antibody recognized the inactive
precursor and mature form of IL-18. Sections were incubated for
60 min with the antibodies at a dilution of 1:50. Biotinylated
conjugate and streptavidin peroxidase were applied for 15 min, and
DAB chromogen was used as a peroxidase substrate complex (all
from the LSABϩ kit; Dako, Copenhagen, Denmark). Incubations
were performed at room temperature. Tissue sections were then
counterstained with Harris hematoxylin and mounted with aqueous
mounting medium (Dako). Intrinsic positive controls for immunoreactivity in each section were IL-18-stained cells in the stroma
(macrophage-like cells). A control using IgG from nonimmunized
goats (R&D Systems) at the same final concentration as that of
anti-IL-18 IgG was carried out for each case, to exclude nonspecific binding. Immunohistochemical staining was evaluated by 2
pathologists (A.V. and G.deP.) blinded to patient outcome, in each
case. The percentage of IL-18ϩ tumor cells was calculated as the
mean value obtained from 3 different fields randomly selected (at
ϫ400) within the same section. Results were expressed as 0%,
Յ10%, 11–33%, 34 – 66% and Ն67%. Staining cut-off for statistical evaluation was 10%.
Cells and cell cultures
Prostate cancer cell lines LNCaP, PC-3 and DU 145 (all from
the ATCC, Manassas, VA) were maintained in DMEM supplemented with 10% heat-inactivated FCS, 100 U/ml of penicillin, 50
␮g/ml of streptomycin (all from GIBCO, Grand Island, NY) and
50 ␮g/ml of Plasmocyn (Cayla, Toulouse, France).
RT-PCR
RNA was isolated from cells by the RNeasy Kit procedure and
treated with RNAse-free DNase (all from Qiagen, Valencia, CA).
RT-PCR was performed using 1 ␮g of total cellular RNA incubated with AMV reverse transcriptase, primer oligo(dT), dNTP
and RNase inhibitor for 1 hr at 42°C (cDNA synthesis kit; Boehringer-Mannheim, Indianapolis, IN). PCR amplification was performed as previously described.19 The sequences of the 5Ј and 3Ј
oligonucleotide primers and the sizes of their products were as
follows: IL-18, 5Ј-GCT TGA ATC TAA ATT ATC AGT C-3Ј,
5Ј-GAA GAT TCA AAT TGC ATC TTA T-3Ј, 342 bp; caspase-1,
5Ј-GGT CCT GAA GGA GAA GAG AA-3Ј, 5Ј-AGG CCT GGA
TGA TGA TCA CC-3Ј, 842 bp; IL-1Rrp, 5Ј-TTG GAG TGA
TGA CAG GAA CAC-3Ј, 5Ј-CAT CAG ATA GGT CGT TAC
TAC TAC C-3Ј, 223 bp; AcPL, 5Ј-GGT TAT TAC TCC TGC
GTG C-3Ј, 5Ј-CCA TTT TCT TCC CCG AAC ATC C-3Ј, 273 bp;
HPRT, 5Ј-TTC AAA TCC AAC AAA GTC TG-3Ј, 5Ј-AGC ACT
GAA TAG AAA TAG TGA TAG A-3Ј, 278 bp. The authenticity
of PCR products was verified by diagnostic restriction digests.
Amplification was conducted as follows: for caspase-1: step 1 was
1 cycle at 94°C for 5 min, step 2 was 40 cycles at 96°C for 2 min,
55°C for 1.5 min and 72°C for 2 min and step 3 was 1 cycle at
72°C for 10 min; for IL-18, IL-1Rrp, AcPL and HPRT: step 1 was
1 cycle at 94°C for 5 min, step 2 was 40 cycles at 94°C for 30 sec,
57°C for 30 sec and 72°C for 30 sec and step 3 was 1 cycle at 72°C
for 10 min. Half of the PCR products were electrophoresed on
1.5% agarose gel, stained with ethidium bromide and photographed under UV light.
Detection of IL-18 and IL-18R␣ by flow cytometry
Cells were detached from the bottom of dishes (Falcon, Becton
Dickinson Labware, Franklin Lakes, NJ) with cell dissociation
solution (Sigma, St. Louis, MO), using the procedure recommended by the manufacturer. IL-18 was detected with the 2-step
fixation and permeabilization Intrastain kit (Dako), according to
the manufacturer’s instructions. Briefly, following fixation and
permeabilization, 5 ϫ 105 cells were incubated with a mouse
antihuman proIL-18 MAb (R&D Systems) or with the isotypematched MAb (mouse IgG1, R&D Systems; final concentration 20
␮g/ml) for 20 min at room temperature. After washing, cells were
incubated for 20 min with antimouse IgG-FITC conjugate (1:50,
Sigma). Cells were then washed and analyzed using Cell Quest
software on a FACSCalibur (Becton Dickinson, San Jose, CA). To
analyze cell-surface IL-18R␣ expression, 5 ϫ 105 cells were
incubated with 20 ␮g/ml goat anti-IL-18R␣ polyclonal IgG (R&D
Systems) at 4°C for 30 min. As control, staining was performed
with nonspecific isotype-matched polyclonal antibody (R&D Systems). Cells were then washed twice and incubated for 30 min on
ice with biotinylated rabbit antigoat IgG (1:50; Vector, Burlingame, CA). After washing, cells were incubated at 4°C for 30 min
with streptavidin-FITC (1:50; Pharmingen, San Diego, CA) and
then analyzed using Cell Quest software on a FACSCalibur.
Immunoblot analysis
Cell pellets of the human prostate cancer cell lines were lysed in
1% Triton buffer containing a cocktail of protease inhibitors (Boehringer-Mannheim). For each cell line, 100 ␮g of protein, estimated by the DC protein assay (Bio-Rad, Hercules, CA), were
solubilized with Laemmli sample buffer and subjected to SDSPAGE (15%) using reduction conditions. Proteins were transferred
by a semidry transblot system; the blot was blocked for 2 hr with
5% w/v nonfat dry milk and 0.05% Tween-20 in TBS and incubated with 1 ␮g/ml of anti-hIL-18 MAb (R&D Systems) for 6 hr
at room temperature. Blots were washed and then incubated for 45
min with HRP-conjugated antimouse IgG at 1:6,000 (Bio-Rad).
After washing, IL-18 was detected using the ECL detection kit
(Amersham, Arlington Heights, IL). The Mr of the proteins was
estimated by comparison with the position of a standard (Kaleidoscope 39-3 kDa, Bio-Rad), using recombinant IL-18 (Chemicon,
Temecula, CA) as control.
Cell treatment for caspase-1 induction and IL-18 production
Caspase-1 induction. Cell lines (1.5 ϫ 106 cells) were seeded in
20 ml of culture medium in a 75 cm2 flask (Falcon, Becton
Dickinson). After incubation for 1 day at 37°C, medium was
replaced and IFN-␥ (Abcys; Valbiotech, Paris, France) added to
the culture at a final concentration of 1,000 U/ml. After 24 hr of
incubation at 37°C, cells were collected by trypsinization and
submitted to RNA extraction and RT-PCR for caspase-1, as described above.
IL-18 secretion induced by interferons. Cell lines were seeded at
1.5 ϫ 106 cells (for collection at 24, 48 and 72 hr) and at 1 ϫ 106
cells (for collection at 96 hr) in 20 ml of culture medium in a 75
cm2 flask. After 1 day of incubation at 37°C, medium was replaced
and IFN-␥ (1,000 U/ml Abcys, Valbiotech) or IFN-␣ (1,000 and
IL-18 PRODUCTION BY PROSTATE CANCER
5,000 U/ml Roferon-A; Roche, Mannheim, Germany) was added
to the culture. After incubation for 24, 48, 72 or 96 hr, supernatants
were collected, centrifuged at 4,000g for 15 min to remove cells
and debris and stored at –20°C until analysis. Specimens were then
thawed at room temperature and concentrated by Centricon-Plus
20 (Millipore, Bedford, MA) at 4,000g for 30 min. The IL-18 level
was then recorded with a human IL-18 ELISA kit for determination of active IL-18 (Medical & Biological Laboratories, Nagoya,
Japan).25 Values were weighted by the concentration factors obtained and normalized for the total number of tumor cells initially
seeded to allow a valid comparison between all conditions. Experiments were performed 3 times.
Inhibition of caspase-1. DU 145 cells were seeded in 75 cm2
flasks and preincubated with 10 ␮M of the cell-permeable TyrVal-Ala-Asp-aldehyde caspase-1 inhibitor (YVAD-CHO; Calbiochem, Darmstadt, Germany) for 1 hr at 37°C. Cells were then
stimulated with 1,000 U/ml of IFN-␥. Caspase-1 inhibitor was
added to the culture every 24 hr. Supernatants were collected at 48
and 72 hr, centrifuged and frozen until IL-18 determination. Experiments were performed 3 times.
Tetrazolium (MTT) assay
Cells were seeded on a flat-bottomed 96-well plate at 5 ϫ 104
cells/well in RPMI-1640 with 10% FCS. Twenty-four hours later,
cultures were downshifted to serum-free medium, to which recombinant IL-18 (0 –100 ng/ml, Chemicon) was added. After 4 days of
culture, mitochondrial activity was assessed by adding 50 ␮g of
the vital dye MTT (Sigma) to culture. The blue dye taken up by the
cells after 4 hr of incubation was dissolved in 0.04 N HClisopropanol (100 ␮l/well), and absorbance at 550 nm was read on
an automated microplate reader. Data are means of 3 separate
experiments.
Proliferation assay
Cells were seeded at 5,000 cells/well in 96-well flat-bottomed
microtiter plates, and IL-18 (Chemicon) at various concentrations
was added at the time of plating. Plates were incubated at 37°C in
5% CO2 for 38 hr and then labeled for 10 hr with 1 ␮Ci/well
3
H-thymidine. Cells were harvested onto glass fiber filter paper,
and the incorporated radioactivity was measured by liquid scintillation counting. All samples were measured in triplicate in at least
2 independent experiments.
Statistical analysis
The association between IL-18 staining and clinicopathologic
variables was tested using the ␹2 test. The primary outcome was
disease-specific relapse, which was determined from the date of
radical prostatectomy. Data were evaluated for disease relapse
using univariate and multivariate analyses in a Cox proportional
hazards model for IL-18 staining and other clinical and pathologic
predictors of outcome. The statistical significance of differences
between means of IL-18 secreted was evaluated by Student’s
unpaired t-test. p Ͻ 0.05 was required for significance.
RESULTS
IL-18 is expressed in normal prostate and prostate cancers
Prostate cancers and distant normal prostate tissue of 36 patients
with newly diagnosed localized prostate carcinoma treated by
radical prostatectomy24 were subjected to IL-18 immunostaining.
In normal prostate tissue, the IL-18 signal was consistently observed in basal cells and some scattered cells in the stroma with the
shape and size of macrophages, whereas epithelial cells did not
express IL-18 (Fig. 1a). Figure 1b shows a characteristic IL-18
immunoreactivity in areas of basal cell hyperplasia. No IL-18
staining was observed in epithelial cells of typical adenomatous
hyperplasia. Positive tumor immunostaining for IL-18 was detected in 27 of the 36 prostate cancers studied. The IL-18 signal
was located in the cytoplasm of tumor cells, and some stromal cells
also presented IL-18 reactivity (Fig. 1c,d). The pattern of tumor
829
IL-18 expression was heterogeneous. Foci of tumor cells situated
away from zones of IL-18 staining (Fig. 1e) did not express IL-18
(Fig. 1f). Percentages of IL-18ϩ tumor cells in each tumor were as
follows: 0% (9 cases), Յ10% (8 cases), 11–33% (8 cases), 34 –
66% (8 cases) and Ն67% (3 cases). Overall, 75% of the tumors
analyzed (27/36 cases) presented Ͼ10% of tumor cells producing
IL-18.
IL-18 is produced by prostate tumor cell lines
To confirm and extend this finding of the capacity of prostate
tumor cells to produce IL-18, we evaluated IL-18 expression of 2
poorly differentiated malignant prostate cell lines (PC-3 and DU
145) and the well-differentiated LNCaP cell line. IL-18 amplification products were detected in all cell lines (Fig. 2a). Intracellular staining for pro-IL-18 by flow cytometry was detected in the
representative experiment shown in Figure 2b for the 3 cell lines,
PC-3, DU 145 and LNCaP. The only form of IL-18 detected in cell
lines by Western blot analysis was the 24 kDa inactive precursor
form. LNCaP always displayed the less intense signal for IL-18.
IFN-␥ modulates caspase-1 mRNA and IL-18 secretion of
prostate tumor cell lines
Only small amounts of IL-18 (about 20 pg/ml for 106 cells) were
inconsistently detected by ELISA for PC-3, DU 145 and LNCaP (5
separate experiments, data not shown). As IL-18 secretion is
preceded by activation of caspase-1, we looked for caspase-1
transcripts in the cell lines. Caspase-1 PCR signals were inconsistently observed in PC-3 and DU 145 despite 40 amplification
cycles, while LNCaP did not express transcripts for caspase-1 (Fig.
3a). The caspase-1 mRNA profile of the cell lines could therefore
account for the weak and inconsistent IL-18 secretion.
IFN-␥ produced in prostate cancers22 has been shown to induce
caspase-1 in tumor cell lines derived from other cancers.26 –28 We
therefore wondered whether IFN-␥ could modulate caspase-1 in
prostate tumor cell lines. Incubation with 1,000 U/ml of IFN-␥ for
24 hr markedly increased caspase-1 mRNA in PC-3 and induced
caspase-1 mRNA in DU 145 and LNCaP (Fig. 3a). In addition,
1,000 U/ml of IFN-␥ added to cultures induced IL-18 secretion in
a dose- (not shown) and time-dependent manner (Fig. 3b). PC-3
secreted about 4-fold more IL-18 than DU 145, while only very
small amounts of IL-18 were detected in the supernatant of LNCaP
cells. The effect of IFN-␥ on cell viability was monitored in
parallel and did not reveal any increase in cell death as detected by
Trypan blue exclusion (data not shown). Evidence of the role of
caspase-1 in IFN-␥-induced IL-18 secretion was strengthened by
the capacity of a caspase-1 peptide inhibitor, YVAD-CHO, to
inhibit about 70% of the IL-18 secreted by DU 145 when stimulated for 72 hr with 1,000 U/ml of IFN-␥ (Fig. 4). This indicates
that prostate tumor cells synthesize pro-IL-18 and can secrete the
cytokine in response to IFN-␥ in the tumor microenvironment.
Prostate tumor cell lines express IL-18R␣
IL-18 can be secreted by prostate tumor cell lines in response to
IFN-␥. We looked for expression on tumor cell lines of the 2
chains of IL-18R, i.e., IL-18R␣ the ligand binding chain (IL1Rrp), and IL-18R␤ (AcPL), the receptor signaling chain. The
well-differentiated tumor cell line LNCaP did not express transcripts for the 2 chains of IL-18R (Fig. 5a). In contrast, PC-3 and
DU 145 presented with IL-1Rrp transcripts, and a faint but consistent RT-PCR signal for AcPL (Fig. 5a). Membrane staining for
IL-18R␣ was detected by flow cytometry for PC-3 and DU 145
(Fig. 5b). Tumor cell line reactivity to IL-18 was evaluated for
PC-3 and DU 145 under serum-free conditions using a tetrazolium
(MTT) assay. PC-3 showed an increase of cellular mitochondrial
activity under IL-18 stimulation, with a maximum obtained for a
final concentration of 1 ng/ml of IL-18 (Fig. 5c), as evaluated after
4 days of culture. IL-18 stimulated slightly but reproductively DU
145, with a less pronounced effect than that observed for PC-3
(Fig. 5c). Data were in accordance with a functional IL-18R
expressed on tumor cells and indicated that IL-18 could act in an
830
LEBEL-BINAY ET AL.
FIGURE 1 – Photomicrographs
represent IL-18 immunoreactivity
patterns in (a) normal prostate,
with IL-18 immunostaining detected in basal cells and some stromal cells (arrows) of normal
glands; (b) positive staining in areas of basal cell hyperplasia; (c,d)
2 prostate cancers with tumor cells
stained for IL-18; (e,f) heterogeneity of IL-18 tumor staining in 2
different areas of the same prostate
cancer. Control using IgG from
nonimmunized goat at the same final concentration as that of antiIL-18 IgG was carried out for each
case and excluded nonspecific
binding [magnification ϫ120 for
(a,c,d), ϫ50 for (b) and ϫ100 for
(e,f)].
autocrine/paracrine way. We evaluated whether IL-18 influenced
the growth of the cell lines and did not observe proliferative
activity of the cytokine by cell counts (data not shown) and
proliferation assay using 3H-thymidine incorporation (Fig. 6). Proliferation assay was also performed under serum-free conditions
with similar results (data not shown). Incubation of tumor cell lines
PC-3 and DU 145 with IL-18 neutralizing antibody or with antiIL-18R␣ antibody, selected for its ability neutralize human IL18R-mediated biologic activity, did not influence tumor growth
(data not shown).
IL-18 synthesis by prostate tumor cells and clinical outcome
Since IL-18 is known to have antitumor activity, we asked
whether it might influence the clinical outcome of patients with
prostate cancer. In the cohort of 36 patients studied, no correlation
was observed between IL-18 (staining cut-off 10%) and Gleason
score or pathologic stage (Table I) even after reanalysis of the data
using different cut-off values for tumor IL-18 positivity at 33%
and 66%. Univariate analysis of disease relapse in patients stratified on the basis of the level of IL-18 expression (staining cut-off
10%) showed a better outcome of patients who had tumors with
high IL-18 expression (RR ϭ 0.4, 95% CI 0.2–1; p ϭ 0.049). This
result was confirmed in multivariate analysis, where the p value
even increased after adjustment for Gleason score and pathologic
stage (RR ϭ 0.3, 95% CI 0.1– 0.8; p ϭ 0.02) (Table II). This
supports the notion that tumor staining for IL-18 might be prog-
nostic for better outcome, independently of clinicopathologic features.
Effect of IFN-␣ on IL-18 secretion by prostate tumor cell lines
To look for other immune molecules that could modulate IL-18
in prostate cancers, we evaluated the effect of IFN-␣, which shares
some biologic activities with IFN-␥ on prostate tumor cells. Interestingly, 5,000 U/ml of IFN-␣-induced time-dependent IL-18 secretion by the poorly differentiated cell line PC-3 (Fig. 7). This
phenomenon was also present at 1,000 U/ml with about 300 ng of
IL-18 /106 cells detected at 96 hr (data not shown). In contrast,
IL-18 secretion was weak for LNCaP and DU 145 (Fig. 7). These
results indicate that IFN-␣ is also capable of modulating IL-18
secretion of prostate tumor cell lines, with a probably less pronounced effect than that observed for IFN-␥.
DISCUSSION
In our study, we demonstrated that prostate cancer cells produce
IL-18 and that IL-18 secretion was modulated by interferons in
prostate tumor cell lines. Production of IL-18 by the tumor was
associated with a better clinical outcome.
Little is known about the capacity of tumor cells to produce
IL-18. Pancreatic tumor cells and Burkitt lymphoma cell lines
produce IL-18 (F. Page`s et al., unpublished data), and malignant
skin tumor cell lines and ovarian carcinomas also secrete IL-
IL-18 PRODUCTION BY PROSTATE CANCER
831
18,29,30 suggesting that IL-18 may be produced by a wide range of
tumors.
In the prostate, although normal epithelial cells do not produce
IL-18, about 75% of prostate cancers studied (27/36 cases) presented tumor cells (derived from normal epithelial cells) producing
IL-18. IL-18 tumor staining was not associated with Gleason grade
or pathologic stage and was heterogeneous, ranging from 10% to
66% of tumor cells producing IL-18. Prostate cancers often present
areas with different Gleason grades, indicating varying degrees of
differentiation of tumor cells and deteriorating cancer cell architecture. This intratumoral heterogeneity may account for the heterogeneous tumor IL-18 expression.
Anti-IL-18 antibodies used in the immunohistochemical study
do not discriminate between the immature and active forms of the
cytokine, which raises the question of IL-18 secretion by the
tumor. At the very least, the staining pattern of IL-18 in prostate
FIGURE 2 – (a) RT-PCR analysis of mRNA expression of IL-18
transcripts in PC-3, DU 145 and LNCaP cell lines. HPRT control PCR
was performed to monitor RT-PCR amplification efficiency. The myelomonocytic KG-1 cell line is shown as positive control. (b) Intracytoplasmic expression of pro-IL-18 on the 3 prostate tumor cell lines
detected by flow cytometry. Shaded and open histograms represent staining with anti-proIL-18 antibody and isotype-matched irrelevant MAb,
respectively. (c) Western blot analysis of IL-18 protein expression. AntiIL-18 antibody detected a 24 kDa protein in all prostate cell lines analyzed. Recombinant human IL-18 was used as positive control.
FIGURE 3 – (a) RT-PCR analysis
of mRNA expression of caspase-1
transcripts in PC-3, DU 145 and
LNCaP nonstimulated or stimulated cell lines with 1,000 U/ml
IFN-␥ for 24 hr. The myelomonocytic KG-1 cell line is shown as
positive control. (b) Means Ϯ SD
of IL-18 detected by ELISA in culture supernatants of PC-3, DU 145
and LNCaP cell lines after stimulation with 1,000 U/ml of IFN-␥
for 0, 24, 48, 72 and 96 hr.
FIGURE 4 – Means Ϯ SD of IL-18 measured by ELISA of culture
supernatant of DU 145 stimulated with IFN-␥ (1,000 U/ml) in the
absence (solid bar) or presence (hatched bar) of caspase-1 inhibitor
YVAD-CHO for 48 and 72 hr. *Student’s t-test.
832
LEBEL-BINAY ET AL.
FIGURE 6 – Effect of recombinant IL-18 (0 –250 ng/ml) on prostate
tumor cell lines. Plates were incubated for 38 hr and then labeled for
10 hr with 3H-thymidine. Each point is the mean Ϯ SD of triplicate
experiments.
TABLE I – DISTRIBUTION OF IL-18 STAINING AS FUNCTION OF GLEASON
SCORE AND PATHOLOGIC STAGE
IL-18 immunoreactivity
FIGURE 5 – (a) RT-PCR analysis of mRNA expression of IL-18R␣
(IL-1Rrp) and IL-18R␤ (AcPL) transcripts in PC-3, DU 145 and
LNCaP. (b) Reactivity of anti-IL-18R␣ (IL1-Rrp) antibody with prostate tumor cell lines was analyzed by flow cytometry. Shaded and open
histograms represent staining with anti-IL-18R␣ antibody and isotypematched irrelevant antibody, respectively. (c) Effect of recombinant
IL-18 on mitochondrial activity of prostate carcinoma cells. Cells were
grown in DMEM with 10% FCS. Twenty-four hours later, cells were
downshifted to serum-free conditions, and recombinant IL-18 was
added at the concentrations indicated on the abscissa. Four days later,
mitochondrial activity was assessed by MTT assay. Results are expressed as relative ratios to the IL-18-free controls. Data are means Ϯ
SD of 3 separate experiments.
cancers indicates that large amounts of mature IL-18 could be
rapidly processed and secreted by the tumor in response to tumor
cell stimulation. Certain arguments suggest secretion of IL-18 by
at least a subset of tumor cells: (i) caspase-1, required for cleavage
of pro-IL-18 into an active form, is detected in about 75% of
prostate tumors from patients with no endocrine therapy prior to
surgery,31 with a heterogeneous pattern of expression within tumor
cells,32 and patients evaluated in our series had newly diagnosed
pathologic stage TXN0M0 tumors and were untreated before surgery; (ii) using an ELISA kit that specifically detects the mature
form of IL-18,25 we detected low concentrations of IL-18 in the
culture supernatants of prostate tumor cell lines; (iii) the IFN-␥
produced in prostate tumors22 was able to increase or induce
Gleason score
3–6
7–9
Pathologic stage
T1–T2
T3–T4
n (%)
Յ10%
n (%)
Ͼ10%
n (%)
13 (36.1)
23 (63.9)
4 (23.5)
13 (76.5)
9 (47.4)
10 (52.6)
19 (52.8)
17 (47.2)
9 (52.9)
8 (47.1)
10 (52.6)
9 (47.4)
p*
0.14
0.99
*p values were determined by the ␹2 test. p Ͻ 0.05 was required for
significance.
caspase-1 mRNA of prostate tumor cell lines, with a subsequent
increase of IL-18 secretion in the medium. Taken together, these
results suggest that at least a subset of tumor cells within prostate
cancers secrete active IL-18.
The induction or increase of transcripts for caspase-1 by IFN-␥
has also been observed in cell lines derived from ovarian,26 hepatic27 and colon28 carcinomas. IFN-␥ induced cell apoptosis via
activation of caspase-126,27 or sensitized the cells to killing by
numerous proapoptotic stimuli.28 We did not detect an increase of
cell death following IFN-␥ stimulation in prostate tumor cell lines,
as previously published.33,34
Interestingly, the hormone-sensitive cell line LNCaP displayed
a different pattern of IL-18 production under IFN-␥ stimulation
and IL-18R expression compared to the hormone-independent cell
lines PC-3 and DU 145. Upon stimulation by interferons, LNCaP
cells reacted with a strong induction of caspase-1 mRNA but
nevertheless reproducibly secreted the lowest amounts of IL-18
833
IL-18 PRODUCTION BY PROSTATE CANCER
TABLE II – UNIVARIATE AND MULTIVARIATE ANALYSES FOR IL-18 STAINING AND CLINICOPATHOLOGIC VARIABLES
WITH RELAPSE-FREE SURVIVAL
Variable
n
IL-18 staining
Ͻ10%
Ն10%
Gleason score
3–6
7–9
Pathologic stage
T1–T2
T3–T4
Univariate analysis
RR1
95% CI
Multivariate analysis
p2
RR
95% CI
0.049
17
19
1
0.4
0.02
1
0.3
0.2–1.0
0.1–0.8
0.049
13
23
1
2.5
0.7
1
1.2
1.0–6.4
0.4–4.0
0.01
19
17
1
3.5
1.4–8.8
p
0.02
1
4.1
1.3–12.7
Risk ratio.– Performed using a Cox proportional hazards model. P Ͻ 0.05 was required for significance and
is presented in bold.
1
2
FIGURE 7 – Means Ϯ SD of IL-18 measured by ELISA in culture
supernatant of PC-3, DU 145 and LNCaP cells stimulated with 5,000
U/ml of IFN-␥ for 0, 24, 48, 72 and 96 hr.
with the slowest kinetics among the 3 cell lines analyzed. While
induction of caspase-1 is known to precede IL-18 secretion, these
data indicate that it is not sufficient and that additional regulatory
steps are involved. Caspase-1 is synthesized as an inactive proenzyme that is selectively cleaved after an aspartate residue to
produce the active enzyme.35 The apparent kinetic discrepancies
between the cell lines may be attributable to different states of
caspase-1 activation. LNCaP cells were cultured without androgen. This induces a strong induction of the proto-oncogene Bcl2,36 which may have antagonistic effects on caspase-1 activation.37
In addition, previous reports have demonstrated the absence of
reactivity to IFN-␥ of LNCaP in terms of proliferation, induction
of STAT-138 and MHC class I modulation.39 Our data show that
LNCaP cells also differ from the hormone-independent cell lines
for the IFN-␥-inducing cytokine IL-18.
Expression of the IL-18 binding moiety of the receptor and the
presence of transcripts for the ␤ chain required for optimal transduction signal in PC-3 and DU 145 suggested possible autocrine/
paracrine activity of IL-18 on tumor cells. This was strengthened
by the observation of a reproducible increase in mitochondrial
activity of the cell lines following IL-18 stimulation. The effect of
IL-18 on tumor cells will have to be carefully and extensively
evaluated as the IL-18 produced by BF16F10, a murine melanoma
cell line expressing IL-18R, is a survival factor promoting FasL
expression and decreases the susceptibility of tumor cells for
killing by NK cells.40 In a first attempt, we evaluated the proliferation of tumor cells under IL-18 stimulation and did not detect
any influence of IL-18 on tumor growth. Furthermore, IL-18R
expression by prostate tumor cells may lead to underestimation of
the capacity of tumor cells to secrete IL-18, as evaluated by ELISA
on culture supernatants. Free IL-18 detected by ELISA could only
represent the amount of cytokine not bound to IL-18R. This could
also explain why IL-18 in supernatants was only detected 48 hr
following IFN-␥ stimulation, whereas caspase-1 transcripts were
observed at 24 hr.
Apart from our study in colon cancer,19 no publication has
addressed the clinical significance of IL-18 expression in tumors.
In the present study, prostate cancer patients with little or no tumor
IL-18 staining (Ͻ10% of tumor cells stained) were identified to be
a population with a high risk of recurrence. Statistical significance
was observed despite the small sample size, but the p value was
close to the limit required for significance. Reanalysis of the data
using a multivariate model adjusted for Gleason score and pathologic stage increased the degree of significance, suggesting a
robust difference and indicating that IL-18 staining could be an
independent prognostic factor, just as predictive as pathologic
stage. This prognostic value now needs to be confirmed and
extended in a larger patient cohort.
Identification of the mechanisms underlying the beneficial action of IL-18 in terms of prognosis was beyond the scope of our
study but needs to be investigated. Several studies have provided
evidence that a Th1-type tumor microenvironment, in accordance
with the in situ presence of activated cytotoxic T cells or NK cells,
favors the control of tumor spreading.41,42 IL-18, a Th1-promoting
cytokine, may promote cytotoxic responses in prostate cancers but
has also been shown to induce Th2 cytokines in the absence of
IL-12.43 This is, however, improbable in prostate cancers as, apart
from the locally produced IL-6,44 2 Th1 cytokines potentially
induced by IL-18, IL-245 and IFN-␥,22 are increased at the tumor
site compared to normal prostate or benign prostatic hyperplasia.
An alternative explanation for our data would be that IL-18 could
interfere with angiogenesis or the proliferative or metastatic capacity of tumor cells. However, we have to keep in mind that IL-18
has also proinflammatory properties and that proinflammatory
cytokines, including IL-1␤46 and TNF-␣,47 promote cancer cell
adhesion and metastasis. IL-18 increases expression of VCAM-1
by the hepatic sinusoidal endothelium, which favors adherence of
melanoma cells and a role of the cytokine has been demonstrated
in the development of hepatic metastases of B16M in vivo.48,49
This could in part explain the detrimental role of high serum
concentration of IL-18 in patients with gastric50 or breast51 carcinomas. IL-18 could then have a beneficial effect when produced at
the tumor site and be detrimental when its concentration increases
in the serum.
In conclusion, this study shows that IL-18 is produced by
prostate tumor cells and its secretion is increased by the interferons. In patients, production of IL-18 by the tumor could be
associated with a better outcome. Data also support the notion that
immune molecules such as interferons could influence the local
environment in a way that would be beneficial for the patients.
ACKNOWLEDGEMENTS
We thank Drs. T. Flam and M. Zerbib for providing clinical
data, M. Carton for help with statistical analysis and Ms. O.
Carrie`re for helpful technical assistance.
834
LEBEL-BINAY ET AL.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
Okamura H, Nagata K, Komatsu T, Tanimoto T, Nukata Y, Tanabe F,
Akita K, Torigoe K, Okura T, Fukuda S, Kurimoto M. A novel
costimulatory factor for gamma interferon induction found in the
livers of mice causes endotoxic shock. Infect Immun 1995;63:3966 –
72.
Okamura H, Tsutsi H, Komatsu T, Yutsudo M, Hakura A, Tanimoto
T, Torigoe K, Okura T, Nukada Y, Hattori K, Kurimoto M. Cloning
of a new cytokine that induces IFN-gamma production by T cells.
Nature 1995;378:88 –91.
Ushio S, Namba M, Okura T, Hattori K, Nukada Y, Akita K, Tanabe
F, Konishi K, Micallef M, Fujii M, Torigoe K, Tanimoto T, et al.
Cloning of the cDNA for human IFN-gamma-inducing factor, expression in Escherichia coli, and studies on the biologic activities of the
protein. J Immunol 1996;156:4274 –9.
Gu Y, Kuida K, Tsutsui H, Ku G, Hsiao K, Fleming MA, Hayashi N,
Higashino K, Okamura H, Nakanishi K, Kurimoto M, Tanimoto T, et
al. Activation of interferon-gamma inducing factor mediated by interleukin-1beta converting enzyme. Science 1997;275:206 –9.
Pages F, Lazar V, Berger A, Danel C, Lebel-Binay S, Zinzindohoue
F, Desreumaux P, Cellier C, Thiounn N, Bellet D, Cugnenc PH,
Fridman WH. Analysis of interleukin-18, interleukin-1 converting
enzyme (ICE) and interleukin-18-related cytokines in Crohn’s disease
lesions. Eur Cytokine Netw 2001;12:97–104.
Robinson D, Shibuya K, Mui A, Zonin F, Murphy E, Sana T, Hartley
SB, Menon S, Kastelein R, Bazan F, O’Garra A. IGIF does not drive
Th1 development but synergizes with IL-12 for interferon-gamma
production and activates IRAK and NFkappaB. Immunity 1997;7:
571– 81.
Adachi O, Kawai T, Takeda K, Matsumoto M, Tsutsui H, Sakagami
M, Nakanishi K, Akira S. Targeted disruption of the MyD88 gene
results in loss of IL-1- and IL-18-mediated function. Immunity 1998;
9:143–50.
Dao T, Ohashi K, Kayano T, Kurimoto M, Okamura H. Interferongamma-inducing factor, a novel cytokine, enhances Fas ligand-mediated cytotoxicity of murine T helper 1 cells. Cell Immunol 1996;173:
230 –5.
Tsutsui H, Matsui K, Kawada N, Hyodo Y, Hayashi N, Okamura H,
Higashino K, Nakanishi K. IL-18 accounts for both TNF-alpha- and
Fas ligand-mediated hepatotoxic pathways in endotoxin-induced liver
injury in mice. J Immunol 1997;159:3961–7.
Puren AJ, Razeghi P, Fantuzzi G, Dinarello CA. Interleukin-18 enhances lipopolysaccharide-induced interferon-gamma production in
human whole blood cultures. J Infect Dis 1998;178:1830 – 4.
Puren AJ, Fantuzzi G, Gu Y, Su MS, Dinarello CA. Interleukin-18
(IFNgamma-inducing factor) induces IL-8 and IL-1beta via TNF
alpha production from non-CD14ϩ human blood mononuclear cells.
J Clin Invest 1998;101:711–21.
Fehniger TA, Shah MH, Turner MJ, VanDeusen JB, Whitman SP,
Cooper MA, Suzuki K, Wechser M, Goodsaid F, Caligiuri MA.
Differential cytokine and chemokine gene expression by human NK
cells following activation with IL-18 or IL-15 in combination with
IL-12: implications for the innate immune response. J Immunol 1999;
162:4511–20.
Micallef MJ, Yoshida K, Kawai S, Hanaya T, Kohno K, Arai S,
Tanimoto T, Torigoe K, Fujii M, Ikeda M, Kurimoto M. In vivo
antitumor effects of murine interferon-gamma-inducing factor/interleukin-18 in mice bearing syngeneic Meth A sarcoma malignant
ascites. Cancer Immunol Immunother 1997;43:361–7.
Osaki T, Peron JM, Cai Q, Okamura H, Robbins PD, Kurimoto M,
Lotze MT, Tahara H. IFN-gamma-inducing factor/IL-18 administration mediates IFN-gamma- and IL-12-independent antitumor effects.
J Immunol 1998;160:1742–9.
Tasaki K, Yoshida Y, Maeda T, Miyauchi M, Kawamura K, Takenaga
K, Yamamoto H, Kouzu T, Asano T, Ochiai T, Sakiyama S, Tagawa
M. Protective immunity is induced in murine colon carcinoma cells by
the expression of interleukin-12 or interleukin-18, which activate type
1 helper T cells. Cancer Gene Ther 2000;7:247–54.
Hashimoto W, Osaki T, Okamura H, Robbins PD, Kurimoto M,
Nagata S, Lotze MT, Tahara H. Differential antitumor effects of
administration of recombinant IL-18 or recombinant IL-12 are mediated primarily by Fas-Fas ligand- and perforin-induced tumor apoptosis, respectively. J Immunol 1999;163:583–9.
Tanaka F, Hashimoto W, Okamura H, Robbins PD, Lotze MT, Tahara
H. Rapid generation of potent and tumor-specific cytotoxic T lymphocytes by interleukin 18 using dendritic cells and natural killer
cells. Cancer Res 2000;60:4838 – 44.
Coughlin C, Salhany KE, Wysocka M, Aruga E, Kurzawa H, Chang
AE, Hunter CA, Fox, JC, Trinchieri G, Lee WM. Interleukin-12 and
interleukin-18 synergistically induce murine tumor regression which
involves inhibition of angiogenesis. J Clin Invest 1998;101:1441–52.
Pages F, Berger A, Henglein B, Piqueras B, Danel C, Zinzindohoue F,
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
Thiounn N, Cugnenc PH, Fridman WH. Modulation of interleukin-18
expression in human colon carcinoma: consequences for tumor immune surveillance. Int J Cancer 1999;84:326 –30.
Kominsky SL, Hobeika AC, Lake FA, Torres BA, Johnson HM.
Down-regulation of neu/HER-2 by interferon-gamma in prostate cancer cells. Cancer Res 2000;60:3904 – 8.
Street SE, Cretney E, Smyth MJ. Perforin and interferon-gamma
activities independently control tumor initiation, growth, and metastasis. Blood 2001;97:192–7.
Royuela M, de Miguel MP, Ruiz A, Fraile B, Arenas MI, Romo E,
Paniagua R. Interferon-gamma and its functional receptors overexpression in benign prostatic hyperplasia and prostatic carcinoma:
parallelism with c-myc and p53 expression. Eur Cytokine Netw 2000;
11:119 –27.
Shou J, Soriano R., Hayward SW, Cunha GR, Williams PM, Gao WQ.
Expression profiling of a human cell line model of prostatic cancer
reveals a direct involvement of interferon signaling in prostate tumor
progression. Proc Natl Acad Sci USA 2002;99:2830 –5.
De Pinieux G, Flam T, Zerbib M, Taupin P, Bellahcene A, Waltregny
D, Vieillefond A, Poupon MF. Bone sialoprotein, bone morphogenetic protein 6 and thymidine phosphorylase expression in localized
human prostatic adenocarcinoma as predictors of clinical outcome: a
clinicopathological and immunohistochemical study of 43 cases.
J Urol 2001;166:1924 –30.
Kikkawa S, Shida K, Okamura H, Begum NA, Matsumoto M, Tsuji
S, Nomura M, Suzuki Y, Toyoshima K, Seya T. A comparative
analysis of the antigenic, structural, and functional properties of three
different preparations of recombinant human interleukin-18. J Interferon Cytokine Res 2000;20:179 – 85.
Kim EJ, Lee JM, Namkoong SE, Um SJ, Park JS. Interferon regulatory factor-1 mediates interferon-gamma-induced apoptosis in ovarian
carcinoma cells. J Cell Biochem 2002;85:369 – 80.
Shin EC, Ahn JM, Kim CH, Choi Y, Ahn YS, Kim H, Kim SJ, Park
JH. IFN-gamma induces cell death in human hepatoma cells through
a TRAIL/death receptor-mediated apoptotic pathway. Int J Cancer
2001;93:262– 8.
Ossina NK, Cannas A, Powers VC, Fitzpatrick PA, Knight JD, Gilbert
JR, Shekhtman EM, Tomei LD, Umansky SR, Kiefer MC. Interferongamma modulates a p53-independent apoptotic pathway and apoptosis-related gene expression. J Biol Chem 1997;272:16351–7.
Park H, Byun D, Kim TS, Kim YI, Kang JS, Hahm ES, Kim SH, Lee
WJ, Song HK, Yoon DY, Kang CJ, Lee C, et al. Enhanced IL-18
expression in common skin tumors. Immunol Lett 2001;79:215–9.
Wang ZY, Gaggero A, Rubartelli A, Rosso O, Miotti S, Mezzanzanica
D, Canevari S, Ferrini S. Expression of interleukin-18 in human
ovarian carcinoma and normal ovarian epithelium: evidence for defective processing in tumor cells. Int J Cancer 2002;98:873– 8.
Sasaki Y, Ahmed H, Takeuchi T, Moriyama N, Kawabe K. Immunohistochemical study of Fas, Fas ligand and interleukin-1beta converting enzyme expression in human prostatic cancer. BJU Int 1998;
81:852–5.
Winter RN, Kramer A, Borkowski A, Kyprianou N. Loss of caspase-1
and caspase-3 protein expression in human prostate cancer. Cancer
Res 2001;61:1227–32.
Hobeika AC, Etienne W, Cruz PE, Subramaniam PS, Johnson HM.
IFN-gamma induction of p21WAF1 in prostate cancer cells: role in cell
cycle, alteration of phenotype and invasive potential. Int J Cancer
1998;77:138 – 45.
Nakajima Y, DelliPizzi A, Mallouh C, Ferreri NR. Effect of tumor
necrosis factor-alpha and interferon-gamma on the growth of human
prostate cancer cell lines. Urol Res 1995;23:205–10.
Cohen GM. Caspases: the executioners of apoptosis. Biochem J
1997;326:1–16.
Gleave M, Tolcher A, Miyake H, Nelson C, Brown B, Beraldi E,
Goldie J. Progression to androgen independence is delayed by adjuvant treatment with antisense Bcl-2 oligodeoxynucleotides after castration in the LNCaP prostate tumor model. Clin Cancer Res 1999;5:
2891– 8.
Bruckheimer EM, Kyprianou N. Bcl-2 antagonizes the combined
apoptotic effect of transforming growth factor-beta and dihydrotestosterone in prostate cancer cells. Prostate 2002;53:133– 42.
Kominsky SL, Hobeika AC, Lake FA, Torres BA, Johnson HM.
Down-regulation of neu/HER-2 by interferon-gamma in prostate cancer cells. Cancer Res 2000;60:3904 – 8.
Bander NH, Yao D, Liu H, Chen YT, Steiner M, Zuccaro W, Moy P.
MHC class I and II expression in prostate carcinoma and modulation
by interferon-alpha and -gamma. Prostate 1997;33:233–9.
Cho D, Song H, Kim YM, Houh D, Hur DY, Park H, Yoon D, Pyun
KH, Lee WJ, Kurimoto M, Kim YB, Kim YS, et al. Endogenous
interleukin-18 modulates immune escape of murine melanoma cells
IL-18 PRODUCTION BY PROSTATE CANCER
41.
42.
43.
44.
45.
46.
by regulating the expression of Fas ligand and reactive oxygen intermediates. Cancer Res 2000;60:2703–9.
Pages F, Vives V, Sautes-Fridman C, Fossiez F, Berger A, Cugnenc
PH, Tartour E, Fridman WH. Control of tumor development by
intratumoral cytokines. Immunol Lett 1999;68:135–9.
Tartour E, Gey A, Sastre-Garau X, Lombard Surin I, Mosseri V,
Fridman WH. Prognostic value of intratumoral interferon gamma
messenger RNA expression in invasive cervical carcinomas. J Natl
Cancer Inst 1998;90:287–94.
Hoshino T, Wiltrout RH, Young HA. IL-18 is a potent coinducer of
IL-13 in NK and T cells: a new potential role for IL-18 in modulating
the immune response. J Immunol 1999;162:5070 –7.
Hobisch A, Rogatsch H, Hittmair A, Fuchs D, Bartsch G Jr, Klocker
H, Bartsch G, Culig Z. Immunohistochemical localization of interleukin-6 and its receptor in benign, premalignant and malignant prostate
tissue. J Pathol 2000;191:239 – 44.
Royuela M, De Miguel MP, Bethencourt FR, Fraile B, Arenas MI,
Paniagua R. IL-2, its receptors, and bcl-2 and bax genes in normal,
hyperplastic and carcinomatous human prostates: immunohistochemical comparative analysis. Growth Factors 2000;18:135– 46.
Lauri D, Bertomeu MC, Orr FW, Bastida E, Sauder D, Buchanan MR.
Interleukin-1 increases tumor cell adhesion to endothelial cells
47.
48.
49.
50.
51.
835
through an RGD dependent mechanism: in vitro and in vivo studies.
Clin Exp Metastasis 1990;8:27–32.
Orosz P, Echtenacher B, Falk W, Ruschoff J, Weber D, Mannel DN.
Enhancement of experimental metastasis by tumor necrosis factor. J
Exp Med 1993;177:1391– 8.
Vidal-Vanaclocha F, Fantuzzi G, Mendoza L, Fuentes AM, Anasagasti MJ, Martin J, Carrascal T, Walsh P, Reznikov LL, Kim SH,
Novick D, Rubinstein M, et al. IL-18 regulates IL-1beta-dependent
hepatic melanoma metastasis via vascular cell adhesion molecule-1.
Proc Natl Acad Sci USA 2000;97:734 –9.
Mendoza L, Carrascal T, De Luca M, Fuentes AM, Salado C, Blanco
J, Vidal-Vanaclocha F. Hydrogen peroxide mediates vascular cell
adhesion molecule-1 expression from interleukin-18-activated hepatic
sinusoidal endothelium: implications for circulating cancer cell arrest
in the murine liver. Hepatology 2001;34:298 –310.
Kawabata T, Ichikura T, Majima T, Seki S, Chochi K, Takayama E,
Hiraide H, Mochizuki H. Preoperative serum interleukin-18 level as a
postoperative prognostic marker in patients with gastric carcinoma.
Cancer 2001;92:2050 –5.
Gunel N, Coskun U, Sancak B, Gunel U, Hasdemir O, Bozkurt S.
Clinical importance of serum interleukin-18 and nitric oxide activities
in breast carcinoma patients. Cancer 2002;95:663–7.