| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
*
Department of Experimental Transplantation and Immunology, Division of Clinical Sciences, National Cancer Institute-Frederick Cancer Research and Development Center (NCI-FCRDC);
Intramural Research Support Program, Science Applications International Corporation (IRSP, SAIC) Frederick;
Data Management Services, NCI-FCRDC; and
§
Laboratory of Experimental Immunology, Division of Basic Sciences, NCI-FCRDC, Frederick, MD 21702
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and TNF-
(10, 11, 12), and enhanced
proliferation of activated T cells (13, 14). Through the induction of
inflammatory cytokines, IL-12 promotes the development of a Th1
phenotype in naive T cells that is necessary for the eradication of
intracellular pathogens (15, 16) and is thought to play an important
role in the immune-mediated killing of tumor cells (1, 2, 3). IL-12 is
produced and secreted primarily by dendritic cells,
monocytes/macrophages, and neutrophils in response to several types of
infections (17, 18, 19). High-affinity receptors for IL-12 are found only
on T cells and NK cells (20, 21). In addition to its role in infection, IL-12 has been shown to possess potent antitumor activity in a wide variety of murine tumor models (1, 3). Activity has been demonstrated against tumors of various histologies, including carcinomas arising from the colon (CT26, MC38), kidney (Renca), or lung (3LL); carcinogen-induced sarcoma cell lines (including the methylcholanthrene-induced series); and melanoma (B16F10 and derivatives) (22, 23, 24, 25, 26, 27). Significant antitumor activity has been demonstrated against established cutaneous tumor deposits up to 1 cm in diameter when IL-12 is administered either systemically or by peritumoral injection. In addition, IL-12 has demonstrated efficacy in reducing the number of experimental metastases (22, 24) and spontaneous metastases (25, 28), in some cases resulting in prolonged survival. Various methods have been used to deliver either systemic or local IL-12, including the use of rIL-12 protein, plasmid DNA expression vectors encoding the IL-12 subunits (29, 30), and recombinant retroviruses (31), pox viruses (32), or adenoviruses (33, 34). Gene therapy approaches have included the modification of fibroblasts (27, 35) or tumor cells (36, 37) to act as IL-12 delivery vehicles, and the direct injection of high-titer viruses into tumor deposits (32, 33, 34). Furthermore, IL-12 has demonstrated adjuvant activity in both peptide (38) and DNA-vaccine models (39). Thus, IL-12 has diverse potential applications for the treatment of cancer, and it remains to be determined how best to exploit its antitumor activity in a particular clinical situation.
The mechanisms through which IL-12 elicits its potent antitumor
activity remain unclear. T cells play a critical role in these events,
since in many model systems depletion of CD4+ and/or
CD8+ T cells can prevent IL-12-induced tumor regressions
(22, 24). In addition, animals cured of established disease by IL-12
therapy often develop immune memory for the regressed tumor, indicating
that activation of T cells occurs during the rejection process. In
contrast, in some experimental tumor models, IL-12 retains partial
activity in athymic nude mice (24, 40). Therefore, depending on the
tumor system employed, T cells, NK cells, and/or macrophages could be
involved in the antitumor immune responses generated by IL-12. IL-12
has also been reported to promote anti-angiogenic activity, a
process thought to occur by the IFN--mediated induction of the CXC
chemokines IP-10 and Mig (41, 42, 43). IP-10 and Mig are also chemotactic
for activated T cells (44), and thus if chemokine activity is critical
for IL-12 effects, the mechanism could involve inhibition of
neovascularization or stimulation of T cell migration to tumor sites.
Furthermore, IFN- production leads to macrophage activation and the
induction of iNOS (45), thus raising the possibility that reactive
oxygen metabolites (such as nitric oxide) contribute to the antitumor
activities of IL-12 (46, 47). Clearly, many potential mechanisms are
known through which IL-12 might exert its antitumor activity.
The preclinical efficacy studies have resulted in the initiation of phase I and phase II clinical trials of rIL-12 in human cancer. However, administration of IL-12 has been accompanied by dose- and schedule-related toxicities (48, 49). Furthermore, when IL-12 is used in combination with other cytokines (e.g., IL-2), there can be a regimen-dependent enhancement of systemic toxicity (25). Localization of IL-12 production to the tumor site could have a distinct advantage over systemic treatments by inducing an immune response against regressing tumor deposits without the need for high circulating levels of IL-12. We have tested this hypothesis by constructing a recombinant adenovirus that encodes the murine IL-12 p40 and p35 cDNAs linked in a single transcription unit by the encephalomyocarditis internal ribosome entry site (IRES).2 Previous studies have demonstrated that i.v.-administered recombinant adenovirus is taken up by the reticuloendothelial system, resulting in high levels of infection and recombinant gene expression within the liver (50, 51). We have used this observation to examine the possibility of targeting hepatic metastases with the AdCMV-IL-12 virus, and to examine possible mechanisms for the antitumor effect observed. Our data demonstrate that AdCMV-IL-12 possesses significant antitumor activity against day 3 established Renca metastases. This treatment is accompanied by a perivascular leukocyte infiltrate surrounding most intrahepatic blood vessels, and the induction of chemokine gene expression in the liver. The inhibition of metastatic growth occurs in the absence of CD4+ and CD8+ T cells, and NK cells, suggesting that nonclassical immune mechanisms are responsible for the therapeutic effects observed. These data demonstrate the efficacy of targeting hepatic metastatic lesions with i.v.-administered adenovirus.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
BALB/c mice were obtained from the Animal Production Area of the National Cancer Institute-Frederick Cancer Research and Development Center and were maintained in a specific pathogen-free environment. Mice were used between 7 and 10 wk of age. Renca is a BALB/c renal cell adenocarcinoma of spontaneous origin that was maintained in mice by serial i.p. passage. Animal care was provided in accordance with procedures outlined in the Guide for the Care and Use of Laboratory Animals (National Institutes of Health publication No. 86-23, 1985).
Cell lines
Cell lines were maintained at 37°C in a 5% CO2 incubator and grown in DMEM (Life Technologies, Grand Island, NY) supplemented with 5% FBS and 100 U/ml penicillin/streptomycin. The murine cell lines B16F10 (melanoma) and NIH3T3 (fibroblast) were provided by Dr. Robert Wiltrout (National Cancer Institute). The murine melanoma cell line K1735P was obtained from Dr. Margaret Kripke (M. D. Anderson Cancer Center). The human melanoma cell lines SK-MEL-28 and DM13 were obtained from the American Type Culture Collection (Rockville, MD). The fibroblast-like primary cell line HEL was obtained from BioWhittaker (Walkersville, MD). HUVECs were obtained from, and maintained in media supplied by, Clonetics (San Diego, CA). Adenoviruses were propagated on 293 cells kindly provided by Dr. Toren Finkel (National Institutes of Health). 293 and HEL cells were propagated in MEM (Life Technologies) supplemented with 10% FBS and 100 U/ml penicillin/streptomycin.
Construction of adenoviruses
The AdCMV-IL-12 recombinant adenovirus was constructed and
purified using the procedures described previously (52). Briefly, an
expression plasmid was constructed containing the murine IL-12 p40 and
p35 cDNAs interrupted by the encephalomyocarditis IRES and under
transcriptional control of the CMV promoter. A fragment of this
expression plasmid was excised using NarI and
SalI restriction enzymes and subcloned into the
ClaI and SalI sites of p
E1sp1A adenoviral
shuttle vector (Microbix Biosystems, Toronto, Canada). This shuttle
vector was cotransfected into 293 cells with the pBHG10 adenoviral
genome vector that contains a deletion in the E1 and E3 regions.
Primary plaques were isolated and screened by PCR to ensure the
presence of the p40 and p35 cDNAs. Positive clones were plaque purified
twice and a single clone was chosen for further amplification.
Large-scale amplification was performed by infecting 293 cells grown in
T150 culture flasks. Infected cells were harvested and the recombinant
virus was purified by cesium chloride gradient centrifugation. The
isolated virus was dialyzed to remove cesium chloride and frozen at
-70°C. The viral titer was determined by plaque assay on 293 cells.
The AdCMV-ßgal virus was a gift from Toren Finkel (National
Institutes of Health).
Infection of cells with recombinant viruses
One day before infection, cells were seeded in six-well tissue culture plates at 4 x 104 cells/well. At the time of infection, media was aspirated from each well and the amount of AdCMV-IL-12 required to give a multiplicity of infection of 50 was diluted in infection media (DMEM plus 2% FBS) and added to each well. Cells were incubated at 37°C for 60 min and 2 ml of complete media was added to each well. Supernatants were collected 24, 72, and 120 h after infection and fresh media was added to each well. Supernatants were frozen at -70°C and the level of IL-12 secretion was determined by ELISA.
Determination of levels of IL-12 production
An ELISA was performed that detects p35/p40 heterodimers. Purified hamster and rat anti-murine Abs directed against the p35 chain of IL-12 (PharMingen, San Diego, CA) were diluted to 6 µg/ml in 0.1 M sodium bicarbonate (pH 8.2) and added to the wells of an enhanced binding ELISA plate (Corning, Corning, NY). The plate was incubated at 4°C overnight, washed two times with PBS/0.05% Tween 20 (PBST) and blocked at room temperature with PBS/10% FBS for 2 h. Recombinant murine IL-12 (PharMingen) was serial twofold diluted from 2 ng/ml to 31.25 pg/ml in PBS/10% FBS and added to the plate as standards. Samples were serial diluted 1:7 and added to the plate in triplicate. After an overnight incubation at 4°C, the plate was washed four times with PBST. Biotinylated rat anti-murine p40 Ab (PharMingen) was diluted to 1 µg/ml and 100 µl was added to each sample for 45 min at room temperature. Samples were then washed six times in PBST. Avidin-peroxidase (Sigma, St. Louis, MO) was diluted 1:400 from a 1 mg/ml stock and 100 µl was added to each sample. The plate was incubated at room temperature for 30 min and washed eight times with PBST. Fresh ABTS substrate buffer (Sigma) was thawed and 10 µl of 30% hydrogen peroxide was added for every 10 ml of substrate buffer used. The ABTS substrate buffer was added to each well and incubated at room temperature for 2 h. Plates were read on a Ceres 900 (Bio-Tek Instruments, Winooski, VT) plate reader at an OD of 405 nm.
Renca hepatic metastases model
To generate hepatic metastases, 5 x 104 Renca cells were injected intrasplenically into BALB/c or SCID mice. The number of Renca cells injected had been experimentally determined such that 100 to 300 liver metastases would develop. After injection of Renca cells, the spleen was removed. Three days later, mice were injected i.v. with HBSS or 5 x 107 plaque-forming units (pfu) of either the AdCMV-ßgal or the AdCMV-IL-12 virus in a 200-µl volume. Virus was administered once, or four times at 4-day intervals. Liver tissue was obtained at the indicated time points and processed for RNA isolation or histologic analysis. Five days after the final virus injection, mice were euthanized, the livers were harvested, and the number of surface metastases was counted with the aid of a dissecting microscope.
In vivo depletion experiments
T cell subsets were depleted in vivo by i.p. injection of mAbs directed against CD4+ (GK1.5) or CD8+ (2.43) T cells under conditions that led to greater than 95% depletion of each subset. Three doses of Ab (250 µg/dose) were administered before the beginning of therapy and were continued three times weekly during the course of the experiment. NK cells were depleted by i.v. injection of anti-asGM1 administered 3 days before beginning therapy and maintained with a weekly injection. Control depletions were performed with isotype-matched rat or rabbit Ig as indicated in the text.
Determination of hepatic NK activity
SCID mice bearing Renca hepatic metastases were depleted of NK cells and treated with recombinant adenoviruses as described above. Two days after the first and second virus treatments, livers were removed from animals receiving HBSS, AdCMV-ßgal, or AdCMV-IL-12. Livers were mechanically dissociated and inflammatory cells isolated at the interphase of a Lympholyte-M (Accurate Chemical and Scientific, Westbury, NY) gradient. These effector cells were plated at ratios of 100:1, 33:1, 11:1, and 3:1 with 51Cr-labeled YAC-1 target cells, and percentage lysis and lytic units (20% lysis per 0.1 million cells) determined by standard methods.
Histology
Tissues were fixed in neutral, buffered formalin and embedded in paraffin. For immunoperoxidase staining of paraffin-embedded tissues, sections were warmed to 60°C for 10 min, deparaffinized, and rehydrated. Slides were digested with protease VIII 0.5% for 3 min at 37°C. Endogenous peroxidase activity was quenched with 3% hydrogen peroxide and the slides rinsed in PBS/0.5% Tween 20 (PBST). Nonspecific binding was blocked by incubation in PBS containing 1% BSA and 1.5% normal goat serum for 20 min. Primary Ab and isotype-matched control Ab were diluted with PBST and added to tissues for 1 h. Slides were washed in PBST, and reacted for 30 min with biotinylated goat anti-rabbit secondary Ab (Vector, Burlingame, CA), and then reacted with the ABC Elite reagent (Vector) for 30 min. Diaminobenzidene was applied for 4 min as substrate for the peroxidase reaction. Slides were counterstained with hematoxylin, dehydrated, and mounted with Permount. For anti-CD3 staining, primary Ab (rabbit anti-CD3; Dako, Carpinteria, CA) was diluted 1:200. For lysozyme staining, primary Ab (rabbit anti-human lysozyme; Dako) was used at a dilution of 1:100.
Changes in intrahepatic gene expression during therapy
At the times indicated in the text, small pieces of liver from control and treated mice were snap frozen on dry ice. RNA was isolated by mechanical dissociation in Trizol reagent (Life Technologies) as suggested by the manufacturer. For Northern blot analysis, 15 µg of each RNA sample was denatured in buffer containing 50% formamide and 20% formaldehyde, separated by gel electrophoresis on a 1.5% agarose-formaldehyde gel, and blotted onto nitrocellulose membranes. 32P-labeled probes were generated by using a random prime labeling kit (Life Technologies). The blots were prehybridized for 1 h at 42°C in 12 ml of Hybrisol solution (50% formaldehyde, 10% dextran sulfate, and 1% SDS; Oncor, Gaithersburg, MD). Hybridization was performed at 42°C for 16 h using 2 x 107 cpm of denatured probe and 400 µg/ml salmon sperm DNA. The filters were rinsed four times in 2x SSC, 0.1% SDS at 42°C, and then washed in this buffer twice for 20 min at 68°C. A final wash in 0.5x SSC and 0.025% SDS was performed at 68°C for 20 min. The blots were exposed using XAR-5 x-ray film (Eastman Kodak, Rochester, NY) and intensifying screens at -70°C. Equal loading of RNA samples was demonstrated by reprobing stripped filters with a glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA.
Statistical methods
The number of liver metastases for animals in treatment groups was compared with that in control groups with a priori pairwise Wilcoxon rank sum tests. Statistical analysis of lytic activity of hepatic NK cells with or without in vivo depletion by anti-asGM1 was performed by analysis of covariance of 51Cr-release against the logarithm of the effector/target ratio. Briefly, the method involves fitting a series of straight lines in a comparative regression procedure. If the hypothesis of a single common line is rejected, more restricted hypotheses regarding the commonality of intercepts or slopes (or both) of lines through the separate data sets are then performed.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Soluble IL-12 has shown therapeutic efficacy in promoting tumor
regression in several tumor models (1, 2, 3). However, sustained systemic
levels of some cytokines, including IL-12, have been shown to produce
toxic side effects in clinical trials (49). Therefore, we examined the
possibility that high localized levels of cytokine expression at the
tumor site may prove to be more beneficial. We constructed a
recombinant adenovirus encoding the murine IL-12 p40 and p35 cDNAs. To
express both chains of IL-12 in a single virus, we utilized the IRES
element from encephalomyocarditis virus to link p40 and p35 cDNA in a
single transcription unit under control of the CMV promoter (Fig. 1
).
|
, IL-12 was efficiently synthesized and
secreted by a variety of murine and human cell lines. Maximal IL-12
secretion occurred 72 h after infection of murine cell lines, with
levels approaching 1 µg/ml and declining thereafter. Infected human
cell lines, including two melanoma cell lines, and primary fibroblasts
(HEL) and HUVEC, secreted very high levels of IL-12 throughout the
course of the experiment. The higher level of IL-12 production by human
cells may reflect the greater efficacy of infection of human cells by
adenoviruses. The secreted IL-12 was shown to be biologically active,
as demonstrated by its ability to stimulate T cell proliferation in a
PHA blast assay (data not shown).
|
Previous studies have demonstrated that i.v. injection of
adenovirus leads to high levels of infection and expression in the
reticuloendothelial system, especially the liver (50, 51). To test the
hypothesis that i.v.-administered IL-12-expressing adenovirus could be
used to exploit this biology, we used a hepatic metastatic model, in
which Renca tumor cells were injected into the spleens of syngeneic
BALB/c mice. This results in hepatic metastases that are seeded via the
portal circulation. Three days later, mice were injected i.v. with
5 x 107 pfu of either AdCMV-ßgal (to control
for nonspecific host responses directed against the virus) or
AdCMV-IL-12. Mice received either a single injection or four injections
administered at 4-day intervals. Five days after the last injection,
livers were harvested from euthanized mice, and the number of surface
metastases was determined. As shown in Figure 2, control mice that received i.v. HBSS
exhibited a large number of metastases with a mean of 272 metastases
per mouse. Mice that received either one or four injections of
AdCMV-ßgal virus also developed a large number of hepatic metastases
(mean of 209 and 163, respectively). In contrast, treatment with
AdCMV-IL-12 proved to be very effective. Administration of a single
injection of AdCMV-IL-12 greatly reduced the number of metastases (mean
of 16), while mice treated four times had a mean of 12 metastases.
Similar therapeutic efficacy was observed when the dose of AdCMV-IL-12
was lowered to 1 x 107 pfu (data not shown).
|
IL-12 has been proposed to activate several antitumor processes,
including the production of Th1-type cytokines resulting in the
development of cell-mediated immune responses directed against tumor
cells. To begin to elucidate the mechanisms responsible for inhibition
of metastatic growth in the hepatic metastases model, histologic
analysis was performed. Mice were treated with four i.v. doses of HBSS
or 5 x 107 pfu of either AdCMV-ßgal or
AdCMV-IL-12 virus. Liver samples were obtained after two doses of virus
or after completion of therapy and embedded for sectioning. Sections
from HBSS-injected animals showed normal liver histology (Fig. 3A). Sections from mice
treated with two doses of the AdCMV-ßgal virus demonstrated a minimal
mononuclear cell infiltrate located primarily around hepatic blood
vessels, with little parenchymal infiltrate (Fig. 3B). This pathologic process was clearly distinct
from control animals and suggested a localized response to AdCMV-ßgal
virus that had infected endothelial cells or perivascular hepatocytes.
Samples from AdCMV-IL-12-treated mice demonstrated a dramatic increase
in this perivascular mononuclear cell infiltrate (Fig. 3C), which surrounded virtually all intrahepatic
blood vessels, but not sinusoids. The degree and localized nature of
this inflammatory response has not been previously described for IL-12
therapy. Immunohistochemical staining demonstrated that the
infiltrating cells were in part composed of CD3+ T cells
(Fig. 3D). Sections were also stained with an Ab
directed against lysozyme, which is expressed by macrophages and
neutrophils. Lysozyme-positive cells were abundant in liver sections
from AdCMV-IL-12-treated animals, both in the perivascular and
intraparenchymal areas (Fig. 3F), but not in control
livers (Fig. 3E). Visualization at a higher power
indicated that the lysozyme-positive cells in the perivascular areas
were neutrophils and macrophages, while the parenchymal
lysozyme-expressing cells were thought to be Kupffer cells. As expected
from the degree of infiltrate in AdCMV-IL-12-treated livers, the level
of CD3+ T cells and lysozyme-positive cells was much
greater than in AdCMV-ßgal-treated control animals (data not shown).
Of note, the level of lysozyme staining of Kupffer cells is much
greater in AdCMV-IL-12-treated livers, suggesting an activation of
these cells, or a recruitment of monocytes to this tissue compartment.
Therefore, after two doses of AdCMV-IL-12, there ensues a T cell and
myeloid perivascular infiltrate and activation or increase of
intraparenchymal Kupffer cells.
|
T cells are not required for the therapeutic efficacy of AdCMV-IL-12
The previous data suggested that the perivascular mononuclear cell
infiltrate observed after therapy with AdCMV-IL-12 was involved in the
antitumor process. The relative importance of CD4+ or
CD8+ T cells was evaluated by T cell subset depletion
experiments, in which the injection of depleting Abs was initiated 5
days before virus therapy (Fig. 4).
HBSS-treated mice develop several hundred metastases (mean of 276).
Mice treated with i.v. injections of AdCMV-IL-12 virus again showed a
dramatic decrease in the number of metastases observed (mean of 22).
Surprisingly, depletion of CD4+ or CD8+ T cells
had no significant effect on the efficacy of AdCMV-IL-12 therapy (means
of 35 and 26, respectively). FACS analysis on lymph nodes from
subset-depleted animals demonstrated that depletion was >95%
effective (data not shown). Immunohistochemical staining of liver
sections from CD4+ or CD8+ T cell-depleted
animals revealed an approximate 50% decrease in the number of
CD3+ T cells in the perivascular infiltrates compared with
AdCMV-IL-12-treated control mice (data not shown). These data suggested
that neither T cell subset alone was sufficient to inhibit the
formation of Renca hepatic metastases. This was surprising given the
abundant CD3+ T cell infiltrate observed in the previous
experiment in nondepleted AdCMV-IL-12-treated mice (Fig. 3
D).
|
The possibility remained that a small but important population of
effector T cells was not eliminated by Ab depletion and was responsible
for the antitumor activity. To fully rule out a role for T cells in
this treatment model, subsequent experiments were performed in SCID
mice. As shown in Figure 5A,
treatment of SCID mice with the AdCMV-IL-12 virus greatly diminished
the number of Renca hepatic metastases (mean of 12) compared with
either control vehicle alone (mean of 230) or AdCMV-ßgal treatment
(mean of 125). These data corroborate the Ab depletion experiments, and
prove that T cells are not required for the inhibition of growth of
Renca hepatic metastases.
|
B). Although mice treated with anti-asGM1 and
either HBSS alone or AdCMV-ßgal displayed significant variability in
the number of hepatic metastases that developed, the mean of both
samples (127 and 116, respectively) was significantly higher than mice
treated with the AdCMV-IL-12 virus (mean of 10). The depletion of
hepatic NK activity by anti-asGM1 was very effective (Table II). In mock-depleted animals, there was
significant induction of hepatic NK activity when assayed 2 days after
a single dose of AdCMV-IL-12; only residual NK activity was present on
day 7 after the second dose of virus. In the anti-asGM1-depleted
mice, no induction of NK activity was observed at either time
point. Thus, therapeutic efficacy of this hepatic metastasis model is
maintained in the absence of residual T cell or NK cell
function.
|
Given the previous data, it was of great interest to determine the
pattern and composition of perivascular infiltrate in livers isolated
from asGM1-depleted SCID mice after treatment with AdCMV-IL-12. As
predicted from the lack of T and NK cells, the degree of infiltrate was
significantly less on H + E-stained sections (Fig. 6) than that observed in liver sections
from AdCMV-IL-12-treated BALB/c mice (Fig. 3). The absence of T cells
was confirmed as no CD3+ cells were observed in these
sections (data not shown). However, staining with antilysozyme did
demonstrate a retention of a significant infiltrate in the perivascular
areas of AdCMV-IL-12-treated mice compared with controls (Fig. 6, D vs C). As with immunocompetent mice,
this appears to be composed of macrophages, neutrophils, and myeloid
extramedullary hematopoiesis. These data suggest that the
nonlymphocytic component of this AdCMV-IL-12-induced inflammatory
response may be important for the antitumor activity observed in the
SCID system.
|
IL-12 treatment induces the expression of several cytokines,
including IFN- and TNF- (10, 11, 12), which then function to activate
the expression of downstream genes. To further examine the mechanism(s)
through which the AdCMV-IL-12 virus may be exerting its antitumor
effects, we examined the level of expression of several genes that
could be important in reducing tumor growth. These included Mig, IP-10
Fas, FasL, and iNOS. Animals bearing Renca hepatic metastases were
treated with four injections of HBSS, AdCMV-ßgal, or AdCMV-IL-12 as
described for previous experiments, and liver specimens were isolated 2
days after each virus injection. RNA was extracted and Northern blot
analysis was performed. As shown in Figure 7, basal levels of IP-10 and Mig
expression were very low in control HBSS-injected animals. High levels
of IP-10 and Mig RNAs were induced within 2 days of the initial
injection of AdCMV-IL-12, whereas in AdCMV-ßgal-treated animals this
induction was noted only after the second injection. In both cases,
expression decreased after the third dose, and was nearly gone 2 days
after the fourth dose of virus. These data suggested that if these
chemokines were responsible for the antitumor activity observed in this
model, the more rapid and higher levels of chemokine expression induced
by AdCMV-IL-12 must account for much of the therapeutic efficacy
observed. Further, it appeared that multiple doses of virus led to
down-regulation of chemokine response, perhaps by tachyphylaxis to
IL-12, or more likely due to immune responses directed against the
virus (50, 53). Treatment with AdCMV-IL-12 did not induce changes in
the expression levels of RNA encoding iNOS, Fas, or FasL (data not
shown).
|
, animals treated with two doses of
AdCMV-IL-12 express abundant mRNA for both chemokines compared with
HBSS- or AdCMV-ßgal-treated mice. Of note, an increase in expression
of Mig and IP-10 mRNA levels is detected in one of two HBSS control
animals. This has been noted in other experiments, and is paradoxical
considering the potent induction of these genes by IFN-, which is
produced primarily by cells that are depleted from these mice.
Nonetheless, the induction of hepatic expression of Mig and IP-10 by
AdCMV-IL-12 in the SCID system is consistent with a possible role for
these chemokines in the antitumor efficacy observed. The mechanism of
CXC-chemokine induction in this model is under under investigation.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The inhibition of Renca metastatic growth was accompanied by a dramatic perivascular mononuclear cell infiltrate surrounding hepatic blood vessels that, during early stages of therapy (after two doses of virus), was composed of T cells, macrophages, and neutrophils. Both the extent of the perivascular infiltrate and the intensity of staining of lysozyme-positive intraparenchymal Kupffer cells was significantly increased in the AdCMV-IL-12-treated mice compared with mice treated with AdCMV-ßgal. After four doses of virus, the perivascular infiltrate appeared to have altered subtly, with fewer CD3+ cells, similar levels of lysozyme-positive perivascular cells and Kupffer cells, and an increase in parenchymal islands of cells that appear to represent IL-12-induced extramedullary hematopoiesis. Previous histologic analysis of livers from mice receiving systemically administered soluble IL-12 have demonstrated a focal mononuclear cell infiltrate scattered throughout the hepatic parenchyma; these cells consisted of NK cells, CD8+ T cells, and monocytes (9). The unique perivascular nature of the infiltrate observed in this study, which surrounded virtually all hepatic blood vessels, may be caused by a host response to adenovirus infection of endothelial cells, subendothelial hepatocytes, and/or uptake by Kupffer cells. This would explain the low-level perivascular infiltrate observed in Adcmv-ßgal-treated mice. Furthermore, the local production of IL-12 would be expected to amplify this antiviral host response. Thus, virus-encoded IL-12 may synergize with a host antiviral immune response to create a novel inflammatory milieu within the liver. From our histologic examination of early metastases arising in control (HBSS- or AdCMV-ßgal-treated) livers, it is apparent that Renca metastases arise by attachment to and penetration of hepatic blood vessels, followed by local proliferation and expansion into the liver parenchyma. Thus, tumor cells and the cells composing the perivascular infiltrate are brought into close anatomic proximity. This may partially account for the therapeutic efficacy of this model. Very few metastases were observed in histologic sections of AdCMV-IL-12-treated mice, even after two doses of virus. This makes it very difficult to experimentally dissect the possible anti-angiogenic mechanisms of AdCMV-IL-12 treatment on the developing Renca metastases, and suggests that tumor growth is blocked at a very early stage, perhaps due to inhibition of neovascularization. The few metastases that were observed are likely those that had already escaped the antitumor mechanism of this form of therapy.
Surprisingly, the therapeutic efficacy of AdCMV-IL-12 was maintained in
mice depleted of CD4+ or CD8+ T cells, and in
SCID mice, even after depletion of NK cells. However, histologic
analysis demonstrated that a perivascular infiltrate persisted in SCID
mice treated with AdCMV-IL-12, even after NK cell depletion with
anti-asGM1. Most of these cells stain positive for lysozyme, and
morphologically consisted of macrophages and neutrophils. In addition,
there were many more lysozyme-positive Kupffer cells observed in the
AdCMV-IL-12-treated animals than in controls. These data suggest that
it is the myeloid arm of the induced infiltrate, perhaps activated by
IL-12 and its downstream cytokine cascade (IFN-, TNF-, and
others), that is critical for antitumor activity either by directly
lysing tumor cells, or through other mechanisms. Our gene expression
data suggest that induction of macrophage iNOS gene expression is not
the tumoricidal mechanism, although low-level paracrine nitric oxide
production is difficult to exclude. Other effector cells with as yet
uncharacterized phenotypes may be important in the local cellular
environment of the liver. For instance, an NK1.1+ cell that
expresses intermediate levels of the ß TCR (NK1+
TCRint) has been identified in the liver of immunocompetent
and nude mice (40). Systemic administration of soluble IL-12 greatly
enhances the cytotoxicity of these NK T (NKT) cells against
NK-sensitive and -resistant tumor targets. NKT cells express a single
TCR -chain (formed by V14 and J281 segments), and develop in a
thymus-independent manner through positive selection on nonclassical
MHC CD1b molecules (54). Recently, it was demonstrated that NKT cells
were required for the IL-12-mediated rejection of metastatic tumor
cells in murine models. The NKT cell phenotype, including that of being
CD4-CD8- and resistant to
anti-asGM1-depletion, would make them prime candidates for
IL-12-dependent effector cells in our system (54). However, this cannot
explain the therapeutic efficacy observed in SCID mice, since
rearrangement of the TCR gene segments would be precluded. This may
suggest the presence of a novel IL-12-responsive cell that mediates the
antitumor activity we have described, either by direct cytotoxic
mechanisms or through the induction of secondary mediators such as Mig
and IP-10.
Other clues to possible mechanisms of antitumor activity in this model
system evolved from studies of changes in the pattern of hepatic gene
expression during treatment. The levels of IP-10 and Mig RNA were
increased in liver specimens from immunocompetent animals treated with
either AdCMV-ßgal or AdCMV-IL-12; however, the kinetics of induction
were distinct. AdCMV-IL-12 induced high levels of IP-10 and Mig
expression within 2 days of the first virus dose, while increased
levels were only detected after the second dose of AdCMV-ßgal. This
suggests that the initial dose of AdCMV-ßgal primed the recipients
for a response on subsequent doses. One possibility is that antiviral T
cells were induced by the initial dose, with subsequent T cell
activation and IFN- secretion after virus rechallenge leading to
chemokine gene induction. Similarly, systemic infection of mice with
vaccinia virus induces hepatic Mig and IP-10 production after 5 days
(55). In the case of AdCMV-IL-12, the initial dose could lead to direct
IL-12-mediated activation of NK cells with high levels of IFN-
secretion and the induction of downstream chemokines. Since inhibition
of micrometastasis may be most effective early in the course of their
establishment, the rapid chemokine induction by AdCMV-IL-12 could
explain its therapeutic efficacy compared with AdCMV-ßgal, and also
explain why a single dose of AdCMV-IL-12 was as therapeutic as four
doses. The induction of hepatic NK activity by AdCMV-IL-12 was also
greatest after the initial dose of virus, further supporting the notion
that important biologic responses occur after the first dose of virus.
For both viruses, the level of chemokine gene expression began to
decrease by the third and fourth doses, a result that strongly suggests
a suppression of viral gene expression by the immune system (50, 53).
Induction of Mig and IP-10 RNA by AdCMV-IL-12 treatment was also
observed in NK-depleted SCID mice, consistent with a role for these
chemokines in the antitumor activity observed. The mechanism of this
induction is unclear given the lack of NK and T cells, major sources of
IFN- production.
In summary, i.v. injection of AdCMV-IL-12 is a highly effective therapy against experimental hepatic metastasis. Although the mechanism of antitumor activity is not clear, treatment is associated with a rapid increase in hepatic Mig and IP-10 RNA levels concomitant with the development of a novel perivascular infiltrate surrounding most intrahepatic blood vessels. T cells and NK cells are not required for activity; however, the continued presence of neutrophils and macrophages within the infiltrates of SCID mice suggests an important role for cells of the myeloid system. The early induction of chemokine gene expression could lead to the therapeutic benefit witnessed here, perhaps by inhibiting neovascularization of establishing tumor metastases (41, 56), or by acting as a chemotactic factor for macrophages (44). Intravenous injection of recombinant adenoviruses may represent a novel approach for the targeted treatment of hepatic metastases. This model may be relevant to a number of human cancers, such as colon cancer and ocular melanoma, in which hepatic metastases play an important and often devastating clinical role. Targeted treatment of regional metastases could be of value in the adjuvant setting, i.e., after chemotherapy in Dukes B and C colon cancer, or after surgical resection of isolated hepatic metastases. Intravenous administration of AdCMV-IL-12 could represent a mechanistically novel organ-targeted approach to diseases for which other therapies are badly needed.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Abbreviations used in this paper: IRES, internal ribosome entry site; pfu, plaque-forming units; H + E, hematoxylin and eosin; NKT, natural killer T (cells); asGM1, asialo GM1.
Received for publication November 26, 1997. Accepted for publication January 30, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
in vivo. Int. Immunol. 6:157. production by natural killer cell stimulatory factor: characterization of the responder cells and synergy with other inducers. J. Exp. Med. 173:869. by neonatal human CD4 T cells. J. Immunol. 151:1938.[Abstract]
in mediating the antitumor efficacy of interleukin-12. J. Immunother. 17:71.
ß+, TCR-
+ T lymphocytes, and NK cells. J. Immunol. 149:3495.[Abstract]
production. J. Immunol. 153:1697.[Abstract]
production by anti-tumor T cells. Int. Immunol. 7:1135.ß T cells activated by IL-12 as a major effector in inhibition of experimental tumor metastasis. J. Immunol. 156:3366.[Abstract]
-mediated tumor growth inhibition induced during tumor immunotherapy with rIL-12. Int. Immunol. 8:855.14 NKT cells in IL-12-mediated rejection of tumors. Science 278:1623. with patterns of tissue expression that suggest nonredundant roles in vivo. J. Immunol. 157:4511.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  S. M. Meeran, S. Katiyar, C. A. Elmets, and S. K. Katiyar Interleukin-12 Deficiency Is Permissive for Angiogenesis in UV Radiation-Induced Skin Tumors Cancer Res., April 15, 2007; 67(8): 3785 - 3793. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Meeran, S. K. Mantena, S. Meleth, C. A. Elmets, and S. K. Katiyar Interleukin-12-deficient mice are at greater risk of UV radiation-induced skin tumors and malignant transformation of papillomas to carcinomas. Mol. Cancer Ther., April 1, 2006; 5(4): 825 - 832. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Zhou, E. A. Lafleur, N. V. Koshkina, L. L. Worth, M. S. Lester, and E. S. Kleinerman Interleukin-12 Up-Regulates Fas Expression in Human Osteosarcoma and Ewing's Sarcoma Cells by Enhancing Its Promoter Activity Mol. Cancer Res., December 1, 2005; 3(12): 685 - 692. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Harada, M. Shimada, S. Okano, T. Suehiro, Y. Soejima, Y. Tomita, and Y. Maehara IL-12 Gene Therapy Is an Effective Therapeutic Strategy for Hepatocellular Carcinoma in Immunosuppressed Mice J. Immunol., December 1, 2004; 173(11): 6635 - 6644. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Kozar, R. Kaminski, T. Switaj, T. Oldak, E. Machaj, P. J. Wysocki, A. Mackiewicz, W. Lasek, M. Jakobisiak, and J. Golab Interleukin 12-based Immunotherapy Improves the Antitumor Effectiveness of a Low-Dose 5-Aza-2'-Deoxycitidine Treatment in L1210 Leukemia and B16F10 Melanoma Models in Mice Clin. Cancer Res., August 1, 2003; 9(8): 3124 - 3133. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Nakamori, M. Iwahashi, M. Nakamura, K. Ueda, X. Zhang, and H. Yamaue Intensification of Antitumor Effect by T Helper 1-dominant Adoptive Immunogene Therapy for Advanced Orthotopic Colon Cancer Clin. Cancer Res., June 1, 2003; 9(6): 2357 - 2365. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Zhang and L. J. DeGroot Gene Therapy of a Rat Follicular Thyroid Carcinoma Model with Adenoviral Vectors Transducing Murine Interleukin-12 Endocrinology, April 1, 2003; 144(4): 1393 - 1398. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. R. Ortaldo and H. A. Young Expression of IFN-{gamma} Upon Triggering of Activating Ly49D NK Receptors In Vitro and In Vivo: Costimulation with IL-12 or IL-18 Overrides Inhibitory Receptors J. Immunol., February 15, 2003; 170(4): 1763 - 1769. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Li, X. Zhang, and X. Xia Regression of Tumor Growth and Induction of Long-Term Antitumor Memory by Interleukin 12 Electro-Gene Therapy J Natl Cancer Inst, May 15, 2002; 94(10): 762 - 768. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. J. DeGroot and R. Zhang CLINICAL REVIEW 131: Gene Therapy for Thyroid Cancer: Where Do We Stand? J. Clin. Endocrinol. Metab., July 1, 2001; 86(7): 2923 - 2928. [Full Text] [PDF]
|
|
|
|
|
|
E. A. Lafleur, S.-F. Jia, L. L. Worth, Z. Zhou, L. B. Owen-Schaub, and E. S. Kleinerman Interleukin (IL)-12 and IL-12 Gene Transfer Up-Regulate Fas Expression in Human Osteosarcoma and Breast Cancer Cells Cancer Res., May 1, 2001; 61(10): 4066 - 4071. [Abstract] [Full Text]
|
|
|
|
 |
|
 W. C. Russell Update on adenovirus and its vectors J. Gen. Virol., November 1, 2000; 81(11): 2573 - 2604. [Full Text]
|
|
|
|
|
|
S. A. Ali, C. S. McLean, M. E. G. Boursnell, G. Martin, C. L. Holmes, S. Reeder, C. Entwisle, D. M. Blakeley, J. G. Shields, S. Todryk, et al. Preclinical Evaluation of "Whole" Cell Vaccines for Prophylaxis and Therapy Using a Disabled Infectious Single Cycle-Herpes Simplex Virus Vector to Transduce Cytokine Genes Cancer Res., March 1, 2000; 60(6): 1663 - 1670. [Abstract] [Full Text]
|
|
|
|
|
|
R. M. Minter, J. E. Rectenwald, K. Fukuzuka, C. L. Tannahill, D. La Face, V. Tsai, I. Ahmed, E. Hutchins, R. Moyer, E. M. Copeland III, et al. TNF-{alpha} Receptor Signaling and IL-10 Gene Therapy Regulate the Innate and Humoral Immune Responses to Recombinant Adenovirus in the Lung J. Immunol., January 1, 2000; 164(1): 443 - 451. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Watanabe, R. G. Fenton, J. M. Wigginton, K. L. McCormick, K. M. Volker, W. E. Fogler, P. G. Roessler, and R. H. Wiltrout Intradermal Delivery of IL-12 Naked DNA Induces Systemic NK Cell Activation and Th1 Response In Vivo That Is Independent of Endogenous IL-12 Production J. Immunol., August 15, 1999; 163(4): 1943 - 1950. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
