* Transplantation Immunology Department, Medicine Branch, Division of Clinical Sciences, National Cancer Institute, National Institutes of Health, and
Division of Therapeutic Proteins, Center for Biologics Evaluation and Research, Bethesda, MD; and
Laboratorio di Virologia, Istituto Superiore di Sanitá, Rome, Italy
Correspondence: Lei Yao, Transplantation Immunology Department, Medicine Branch, DCS, NCI, NIH, Bldg. 10, Room 12C07, Bethesda, MD 20892. E-mail: Yaol{at}mail.nih.gov
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(IFN-) and
the IFN--induced murine chemokine IFN--inducible
protein-10 (IP-10). The mechanisms by which EBV-LMP-1 promotes the
expression of IFN- has remained unclear. Here we show that murine
interleukin (IL)-18 was consistently expressed in regressing Burkitt
tumors but was either expressed at low levels or absent from
progressively growing Burkitt tumors. By immunohistochemical methods,
IL-18 protein was visualized in regressing but not in progressively
growing Burkitt tumors. In contrast, IL-12 p35 and IL-12 p40 were only
rarely expressed in regressing Burkitt tumors. In splenocyte cultures,
EBV-infected lymphoblastoid cells and LMP-1-transfected Burkitt cells
promoted the expression of IL-18 but not the expression of IL-12 p35
and IL-12 p40. A neutralizing antibody directed at murine IL-18 reduced
murine IP-10 expression induced by EBV-immortalized cells in splenocyte
cultures. These results provide evidence for IL-18 expression in
response to a viral latency protein and suggest that IL-18 may play an
important role as an endogenous inducer of IFN- expression, thereby
contributing to tumor regression.
Key Words: Burkitt lymphoma • interferon • angiogenesis • latent membrane protein-1 • tumor regression
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Previous studies have found that the EBV-encoded latency protein LMP-1 is responsible for the antitumor response elicited by EBV-immortalized cells in athymic mice [4 ]. In athymic mice, EBV-negative Burkitt cells transfected with the LMP-1 gene and expressing the protein at high levels give rise to tumors that spontaneously regress through necrosis and scarring, whereas the parental cells or control transfectants give rise to progressively growing malignant tumors.
Efforts to clarify the host response leading to tumor regression
induced by the EBV-LMP-1 protein have shown that two murine
chemokines—interferon-inducible protein 10 (IP-10) and the monokine
induced by interferon- (Mig)—serve as downstream mediators of the
antitumor response by inhibiting tumor angiogenesis [5
,
6
]. IP-10 and Mig, structurally related CXC chemokines
that share CXCR3 as a signaling receptor, are active as chemotactic
factors for T and natural killer (NK) cells, inhibit angiogenesis, and
exert antitumor effects [7
]. In earlier studies, we
found IFN-, IP-10, and Mig to be expressed at significantly higher
levels in mice having Burkitt tumors undergoing regression compared
with mice having Burkitt tumors growing progressively
[5
]. EBV-negative Burkitt cells transfected with IP-10
give rise to tumors that undergo massive spontaneous necrosis and
display widespread histological evidence of vascular damage
[5
]. When inoculated into mice with progressively
growing Burkitt tumors, IP-10 and Mig produce extensive tumor necrosis
and vascular damage [5
, 6
]. However, the
mechanism by which EBV-LMP-1 can elicit the expression of IFN-,
IP-10, and Mig, all of which mediate tumor tissue necrosis in this
murine model, has remained unclear.
The multifunctional cytokines IL-12 and IL-18 have been identified as
potent inducers of IFN- [8
, 9
];
therefore, we examined the potential contribution of IL-12 and IL-18 to
the antitumor responses elicited by EBV-LMP-1 in athymic mice.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Antibodies
Rabbit anti-murine IP-10 and rabbit anti-murine Mig were gifts
of J. M. Farber [Laboratory of Clinical Investigation, National
Institutes of Allergy and Infectious Diseases, National Institutes of
Health (NIH), Bethesda, MD]; mouse anti-rat IFN- monoclonal
antibody (clone DB-1; cross-reactive with mouse protein) was a gift of
D. Finbloom [Center for Biologics Evaluation and Research, U.S. Food
and Drug Administration (FDA), Bethesda, MD]; rabbit anti-murine IL-18
antisera were purchased from PeproTech (Rocky Hill, NJ) and R&D Systems
Inc. (Minneapolis, MI); a monoclonal anti-murine IL-18 antibody was
purchased from R&D Systems Inc.; and an antibody against murine IL-12
was a gift of Genetics Institute, Inc., Cambridge, MA.
Animal studies
Four- to 6-week-old BALB/c nu/nu mice (National
Cancer Institute, Bethesda, MD) maintained in pathogen-limited
conditions were used throughout. Mice received 400 Rads of total body
irradiation and 24 h later were injected subcutaneously in the
right abdominal quadrant with 107 exponentially growing
CA46 Burkitt lymphoma cells. Some of the mice received weekly
intratumor injections of buffer alone (0.1 mL), whereas other mice
received weekly intratumor inoculations of 107
exponentially growing VDS-O lymphoblastoid cells (0.1-mL total
injection volume). All animals were observed twice weekly, and tumor
sizes were recorded as the product of two-dimensional caliper
measurements. Tumors were removed 4 weeks after initial cell
inoculation.
Reverse transcriptase-polymerase chain reaction
Total cellular RNA was isolated from tumors and splenocytes by
standard methods (using RNAzol solution; GIBCO-BRL, Gaithersburg, MD).
Four milligrams of RNA were reverse transcribed using the SuperScript
preamplification system (GIBCO-BRL). Complementary DNA (cDNA)
corresponding to 80 ng of total RNA was subjected to polymerase chain
reaction (PCR) amplification in a 50-µg reaction mixture containing
20 mmol/L of Tris-HCl (pH 8.4), 50 mmol/L of KCl, 1.5 mmol/L of
MgCl2, 200 µmol/L of each deoxynucleotide triphosphate,
2.5 U of Taq DNA polymerase (GIBCO-BRL), and 0.2 µmol of
each primer pair. The primers for amplification of murine IFN-,
IP-10, Mig, IL-12 p35, IL-12 p40, and granzyme B and the conditions for
amplification were as previously described [5
,
10
]. The primer pair for amplification of murine IL-18
was 5' ACTGTACAACCGCAGTAATACGG and 3' AGTGAACATACAGATTTATCCC
[11
]. The primer pair for amplification of the murine
housekeeping gene encoding cytochrome oxidase subunit II (MCOII)
included 5' TGGCCTACCCATTCCAACTT and 3' GGTTAACGCTCTTAGCTTCA. All
primer pairs were specific for murine RNA and discriminated human RNA.
For semiquantitative results, the amount of RNA from each sample was
selected on the basis of equivalent amounts of MCO II cDNA amplified
from each sample, and the number of amplification cycles was selected
experimentally for each primer pair to fit the linear part of the
sigmoid curve reflecting the relationship between the number of
amplification cycles and the amount of PCR product. PCR products were
detected with 32P-labeled nucleotides
([
-32P]dCTP [specific activity, 
3,000 Ci/mmol;
Amersham, Arlington Heights, IL] on 8% acrylamide [Long Ranger; AT
Biochem, Malvern, PA])-Tris-borate ethylenediamine tetraacetate gels,
followed by autoradiography.
Immunohistochemistry
Paraffin-embedded tissue sections were deparaffinized twice in
xylene and twice in 100% ethanol, incubated in 3% hydrogen peroxide
in methanol, and rehydrated through graded ethanol washes followed by
Tris-buffered saline (TBS). After blocking with 3% serum in TBS,
sections were incubated overnight at 4°C with dilutions of selected
primary antibodies (1:200 dilution of rabbit anti-murine IP-10, 1:200
dilution of rabbit anti-murine Mig, 1:200 dilution of rabbit anti-mouse
IL-18, or 1:100 dilution of rat anti-mouse IFN- monoclonal
antibody). After washing, biotinylated goat anti-rabbit, goat anti-rat,
or horse anti-mouse secondary antibody (2 µg/mL; Vector Labs,
Burlingame, CA) was applied followed by VECTASTAIN ABC peroxidase
complex (Elite ABC kit; Vector Labs). The sections were developed using
peroxidase 3,3'-diaminobenzidine substrate and counterstained with
hematoxylin. All sections were mounted, dehydrated, and examined by
light microscopy.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, granzyme B
(Fig. 1)
and the chemokines IP-10 and Mig (data not shown) were found
to be more abundant in tumor tissues from regressing as opposed to
progressively growing tumors. The murine IL-12 p35 and IL-12 p40 PCR
products were generally undetected or detected at low levels in these
tumors. In contrast, the PCR products for murine IL-18 were more
abundant in tumor tissues from regressing as opposed to progressively
growing tumors (Fig. 1) .
 View larger version (53K): [in a new window] |
Figure 1. Patterns of cytokine and IFN- gene expression in experimental
Burkitt tumors. BALB/c nu/nu mice were injected
subcutaneously with the EBV-negative Burkitt lymphoma CA46 cell line
(106 cells/mouse). A group of mice (n = 6)
was treated weekly with intratumor inoculations of 107
VDS-O lymphoblastoid cells (100 µL), whereas control mice
(n = 6) were injected weekly into the tumor with buffer
alone (100 µL). Four weeks after the Burkitt cells were inoculated,
all tumors were removed and measured, and RNA was extracted from the
regressing and progressively growing Burkitt tumors. Total RNA was
subjected to semiquantitative reverse transcriptase-PCR analysis.
|
. Whereas tumor tissues from progressively growing Burkitt
tumors were generally negative for all tested proteins, tumor tissues
from regressing Burkitt tumors invariably stained for murine IL-18,
IFN-, and IP-10. No staining for IL-12 was detected (data not
shown). In representative samples (Fig. 2
), murine IL-18 was present mostly in patchy areas at the
interphase that demarcates live from necrotic tumor tissue. IL-18
resided in large cells with the morphology of macrophages, whereas
IFN- was diffuse throughout the tumor and resided in small cells
with the morphology of lymphocytes (Fig. 2)
. These experiments
demonstrated that tumor regression in this model is associated with a
murine host response that includes increased tumor tissue expression of
murine IL-18 but not IL-12.
 View larger version (74K): [in a new window] |
Figure 2. Immunohistochemical detection of IL-18 and IFN- in regressing
Burkitt tumors. Representative tissues from regressing Burkitt tumors
were stained for murine IL-18 and IFN- and counterstained with
hematoxylin. Original magnification x40.
|
from splenocytes cultured with the
EBV-infected lymphoblastoid B95-8 and VDS-O cell lines. Thus, the
pattern of murine cytokine and chemokine gene expression induced in
splenocytes by EBV-positive cells was similar to that induced in BALB/c
nu/nu mice bearing regressing Burkitt tumors.
 View larger version (59K): [in a new window] |
Figure 3. Cytokine and IFN- gene expression in splenocytes cultured with
EBV-negative, EBV-positive, and LMP-1-expressing cell lines. BALB/c
nu/nu splenocytes (106/mL) were cultured for
48 h with the indicated cell lines (106 cells/mL). At
the end of incubation, total RNA was extracted, and cytokine gene
expression was evaluated by semiquantitative reverse
transcriptase-PCR.
|
Effects of IL-18 neutralization on IP-10 expression induced by
EBV-immortalized cells in splenocytes
IL-18 has previously been shown to induce expression of IFN-
[9
]. In turn, IFN- is known to be a potent inducer of
IP-10 and other chemokines [7
]. To examine whether IL-18
serves as a mediator of IP-10 expression induced by EBV-immortalized
cells, we used a neutralizing monoclonal antibody against murine IL-18
(10 µg/mL; R&D Systems) in cultures of BALB/c nu/nu
splenocytes culture with the lymphoblastoid VDS-O cell line. In
parallel, we tested the effects of a neutralizing antibody against
murine IL-12 (10 µg/mL; Genetics Institute) alone or in conjunction
with the neutralizing antibody against murine IL-18 (10 µg/mL).
Confirming the results above, murine IL-18 and murine IP-10 were
undetectable in the culture supernatants of splenocytes cultured alone
or in the presence of VDS-O cells (data not shown), presumably due to
assay sensitivity. We therefore examined IP-10 expression by
semiquantitative reverse transcriptase-PCR. As shown (Fig. 4
), the PCR product for murine IP-10 was absent from splenocytes
cultured in medium alone but was detected in splenocytes cultured with
VDS-O cells. By comparison with splenocytes cultured with VDS-O cells
alone, the PCR product for murine IP-10 was reduced from splenocytes
cultured with neutralizing antibody against murine IL-18 either alone
or in conjunction with neutralizing antibody against murine IL-12. By
contrast, the IP-10 PCR product was only minimally reduced in
splenocytes cultured with neutralizing antibody to murine IL-12 alone
compared with cultures without antibody. These results indicate that
IL-18, not IL-12, mediates, at least in part, IP-10 expression induced
by EBV-immortalized cells in splenocytes.
 View larger version (42K): [in a new window] |
Figure 4. IL-18 dependency of IP-10 expression induced by EBV-immortalized cells
in splenocytes. BALB/c nu/nu splenocytes
(106/mL) were cultured for 48 h with the
lymphoblastoid cell line VDS-O (106 cells/mL) in medium
alone, with neutralizing monoclonal antibody against murine IL-18 (10
µg/mL), with neutralizing antibody against murine IL-12 (10 µg/mL),
or with anti-IL-18 (10 µg/mL) plus anti-IL-12 (10
µg/mL)-neutralizing antibodies together. At the end of incubation,
total RNA was extracted, and cytokine gene expression was evaluated by
semiquantitative reverse transcriptase-PCR. After normalization for the
murine MCO II housekeeping gene, the results of phosphorimage analysis
were plotted as arbitrary units (pixels). Lanes (left to right): 1,
splenocytes in medium alone; 2, splenocytes plus VDS-O; 3, splenocytes
plus VDS-O plus anti-IL-18; 4, splenocytes plus VDS-O plus anti-IL-12;
5, splenocytes plus VDS-O plus anti-IL-18 plus anti-IL-12.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and the IFN--inducible chemokines IP-10 and Mig
[2
3
4
5
]. The observation reported here that IL-18 but not
IL-12 is consistently expressed in tumors that regress raises the
possibility that IL-18 may represent a critical endogenous mediator of
IFN- production in the context of antitumor responses.
Two cytokines, IL-12 and IL-18, are presently regarded as the principal
inducers of IFN- during inflammatory responses, but their relative
contribution to IFN- production in various settings remains
undefined [8
, 9
]. Many studies support the
notion that IL -12 and IL-18 act synergistically as IFN- inducers.
This synergy has been explained on the basis of the observations that
IL-12 can stimulate the expression of the IL-18 receptor
[12
, 13
], that IL-12 and IL-18 can regulate
the production of each other [14
, 15
], and
that IL-12 and IL-18 can regulate the IFN- promoters but act at
different levels [16
]. By itself, IL-12 is a potent
inducer of IFN- [17
]. When given to
IL-1-converting-enzyme (ICE) knockout mice, IL-12 induced substantial
IFN- production independently of IL-18, which remained low in the
circulatory system [18
]. In IL-18 knockout mice, IL-12
promoted NK cell cytotoxicity [19
]. However, recent
observations suggest that endogenous IL-18 may contribute significantly
to IFN- stimulation by IL-12, both in vitro and in vivo
[18
].
IL-18 was originally described as an endotoxin-induced serum factor
that stimulates IFN- production [20
]. Both in vivo
and in vitro, neutralizing antibodies to murine IL-18 markedly reduced
LPS-induced IFN- production, suggesting that endogenous IL-18 is
required for IFN- production induced by microbial agents
[11
, 18
]. However, the ability of IL-18 to
induce IFN- is mostly dependent on the contribution of a second
signal provided by mitogens, microbial agents, or IL-12
[21
22
23
].
In our experimental system, we found no evidence of increased
expression of IL-12 p35 and IL-12 p40 mRNA or increased levels of IL-12
protein in tumor tissue in which IFN- was detected. We cannot
exclude that small levels of IL-12 might be present and serve as a
second signal for IFN- production induced by IL-18 or that another
molecule might substitute for IL-12. Another possibility is that the
EBV-LMP-1-positive cells serve roles both as inducers of IL-18
secretion and as a source of a second signal for IFN- production by
IL-18. It is noteworthy that the EBV-positive cells promoted increased
IL-18 mRNA expression in tumor tissue and splenocytes and increased
accumulation of cell-associated IL-18 protein in tumor tissue. In
addition, a neutralizing antibody to murine IL-18 reduced levels of
murine IP-10 expression induced by EBV-immortalized cells in
splenocytes, whereas an antibody to murine IL-12 had minimal effect in
this system. Because mature IL-18 is produced from its inactive
precursor by the ICE caspase-1, the expression of IL-18 mRNA is not
always associated with the secretion of a functional IL-18 protein
[24
]. Indeed, mice deficient in caspase-1 fail to
produce IFN- in response to endotoxin due to the absence of mature
IL-18 in these mice [18
, 25
,
26
]. However, the presence of IFN- mRNA and protein in
the absence of detectable IL-12 p35 or IL-12 p40 mRNA supports the
notion that biologically active IL-18 protein was being induced in this
system.
A murine mammary carcinoma cell line transfected with the IL-18 gene
was found to be less tumorigenic and to form tumors more slowly than
control cells [27
]. The antitumor effect of IL-18 in
this system required IFN- and was attributed to inhibition of
angiogenesis by IL-18. We have previously demonstrated that inhibition
of tumor angiogenesis is central to the antitumor effects of
EBV-positive cells, and we have identified the IFN--inducible
chemokines IP-10 and Mig as downstream mediators of this antiangiogenic
effect. Furthermore, we have proposed that NK cells, recruited to the
tumor site by the chemokines IP-10 and Mig and locally activated by
IFN-, are required for the antitumor effect [28
]. By
consistently detecting IL-18 expression in regressing Burkitt
tumors, the current experiments provide evidence that IL-18
represents a missing link in the antitumor response induced by
EBV-LMP-1-positive cells.
In addition to its more clearly defined role in inflammation, IL-18 may
play an important role in natural responses to cancer. An analysis of
patterns of cytokine and chemokine expression in human
lymphoproliferative diseases demonstrated that IL-18, IFN-, and Mig
are present at significantly higher levels in lymphoid tissues from
patients diagnosed with acute EBV-induced infectious mononucleosis as
opposed to tissues with post-transplant lymphoproliferative disease
[29
]. Although both conditions are associated with EBV
infection, acute infectious mononucleosis is mostly a self-limited
illness, whereas post-transplant lymphoproliferative disease is
associated with the unbridled expansion of EBV-infected cells. In the
EBV-associated lymphoid granulomatosis and nasal or nasal-type T/NK
cell lymphomas, IP-10 and Mig have also been found to be expressed at
significantly higher levels than in tissues with lymphoid hyperplasia
[30
]. Recently, we found IL-18 and IFN- also to be
elevated in lymphoid granulomatosis and nasal or nasal-type T/NK cell
lymphoma tissues (L. Yao, J. Setsuda, C. Sgadari, B. Cherney, and G.
Tosato, unpublished data). The increased expression of IL-18, IFN-,
IP-10 and Mig in certain EBV-positive lymphoproliferative diseases with
a benign outcome or prominent tumor tissue necrosis raises the
possibility that IL-18 may participate in critical host responses
designed to limit excessive or abnormal lymphoid cell growth.
Received September 11, 2000; revised January 3, 2001; accepted January 5, 2001.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  F. Pages, J. Galon, G. Karaschuk, D. Dudziak, M. Camus, V. Lazar, S. Camilleri-Broet, C. Lagorce-Pages, S. Lebel-Binay, G. Laux, et al. Epstein-Barr virus nuclear antigen 2 induces interleukin-18 receptor expression in B cells Blood, February 15, 2005; 105(4): 1632 - 1639. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Gracie, S. E. Robertson, and I. B. McInnes Interleukin-18 J. Leukoc. Biol., February 1, 2003; 73(2): 213 - 224. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
