Download Pre-existing immunity to adenovirus does not prevent tumor

Gene Therapy (1997) 4, 1069–1076
 1997 Stockton Press All rights reserved 0969-7128/97 $12.00
Pre-existing immunity to adenovirus does not prevent
tumor regression following intratumoral administration
of a vector expressing IL-12 but inhibits virus
dissemination
JL Bramson 1,2, M Hitt1, J Gauldie2 and FL Graham1,2
Departments of 1Biology and 2 Pathology, McMaster University, Hamilton, Ontario, Canada
Adenovirus (Ad) vectors are being intensively studied as
vehicles for cancer gene therapy. We have been exploring
the benefits of direct intratumoral injection of Ads expressing cytokines for immunotherapy. Our previous work demonstrated that therapy using a vector expressing interleukin-12 (AdmIL-12.1) produced regressions in approximately 80% of treated tumors, supporting further preclinical
investigations. Recent reports have shown that immunity
to Ad can be a major limiting factor in Ad-mediated gene
transfer. As most animal studies with Ad vectors have
involved nonimmune hosts, it remains difficult to predict
how effective these treatments will be in humans, where
the majority of individuals have had previous exposure to
Ad. To address this question, we compared the effectiveness of the AdmIL-12.1 cancer therapy in naive and Adimmune mice. We found that both groups responded
equally well to treatment and that the response to AdmIL12.1 in both groups resulted in the generation of CTL
reactive against tumor antigen, indicating that antitumor
immunity was achieved. Peak transgene expression in the
tumor was only reduced by 2.4-fold in Ad-immune animals
compared with nonimmune mice. It was also observed that
in naive animals, the virus disseminated from the site of the
tumor following injection and by 72 h substantial transgene
expression was detected in peripheral organs, most
notably the liver. Transgene expression in the liver of Adimmune animals was reduced by greater than 1000-fold
relative to that in naive mice. These results strongly support the clinical utility of Ad-based cancer gene therapy and
suggest that Ad immunity may be advantageous in that it
is not a complete block to gene transfer in the tumor and
it greatly reduces virus dissemination.
Keywords: adenovirus; tumor immunotherapy; dissemination
Introduction
Adenovirus (Ad) vectors are being widely investigated
as vehicles for gene delivery in vivo. One particular application for which the vectors are ideally suited is the gene
therapy of cancer. The major advantages of Ad vectors
include: (1) a broad host range in terms of gene transfer;
(2) easy preparation of high titer stocks (1011–1012
p.f.u./ml); (3) a cloning capacity of approximately 8 kb
in vectors with deletions of early region 1 (E1) and E3
sequences; and (4) transient expression which limits longterm side-effects (reviewed by Hitt et al1). Studies in animal models indicate that efficient gene transfer using Ad
vectors in vivo may be limited by the development of
anti-Ad immunity.2–5 Humoral responses to the virus
produce high levels of neutralizing antibodies (Abs) preventing repeated vector administration3–5 and cellular
responses result in clearance of virally infected cells by
cytotoxic T cells (CTLs).2 As the majority of the human
population has been exposed to Ads during their lifetime,
protective immunity may be a potential obstacle to
Correspondence: FL Graham, 1280 Main Street W., Hamilton, Ontario,
Canada L8S 4K1
Received 27 March 1997; accepted 12 June 1997
effective Ad-mediated gene therapy in vivo. Much of the
evidence regarding the effects of virus immunity on gene
transfer was obtained from experiments investigating
lung and liver gene transfer. These tissue sites may be
particularly sensitive to humoral Ab responses due to
their direct contact with the systemic circulation and
mucosal surfaces. Our interest, on the other hand, focuses
on solid tumors which are often poorly vascularized and
somewhat detached from the systemic circulation. It is
therefore reasonable to speculate that intratumoral gene
transfer will not be as sensitive to Ad immunity as gene
transfer to other tissues.
Many groups, including ours, are investigating the utility of Ads for gene transfer to tumor tissue (reviewed in
Ref. 1). Our studies have involved cytokine gene transfer
and we have been successful in producing regressions of
established tumors using Ads which express either IL-2,
IL-4 or IL-12. 6–8 Others are using Ads to transfer prodrugactivating enzymes, such as the herpes simplex virus thymidine kinase gene (HSVtk) and E. coli cytosine deaminase (CDA), to render infected cells sensitive to the otherwise nontoxic prodrug (gancyclovir in the case of
HSVtk.9,10 As mentioned above, Ads are reasonable
choices as vectors for these therapies because expression
is transient and the virus is expected to infect primarily
the tissues surrounding the site of injection. However,
Immunity to adenovirus prevents vector dissemination
JL Bramson et al
1070
little work has been done to determine the extent to
which a replication-deficient Ad will disseminate from
the site of injection to infect peripheral organs. The dangers of disseminated virus are greatest with the drugactivating therapies, because this will result in organ
destruction in the presence of the prodrug. It is also a
major concern for immunotherapies because many of the
cytokines can cause profound systemic toxicities if
expression is not localized to the tumor.
In this report we show that pre-existing Ad-immunity
does not change the outcome of intratumoral treatment
with an Ad-expressing IL-12 (AdmIL-12.1). During the
course of the studies, we observed that the virus does
disseminate in our model following intratumoral administration. It was then demonstrated that Ad-immunity
can inhibit vector dissemination indicating that immunity
is not a major barrier in this model and may actually
be beneficial.
Results
Effect of Ad immunity on tumor therapy with AdmIL-12.1
We have been studying the immunotherapeutic effects of
direct intratumoral injection of cytokine-expressing Ad
vectors in a model of polyoma middle T (PymT) induced
breast adenocarcinoma described by Guy et al.11 This
model is metastatic, phenotypically similar to human
breast adenocarcinoma and provides a defined tumor
associated antigen (PymT). To investigate the effect of
neutralizing Abs on the outcome of Ad-based immunotherapy, mice were immunized intranasally (i.n.) with 108
p.f.u. of wild-type Ad5. This route produced high levels
of circulating neutralizing Abs 4 weeks after immunization. The Ab titers of two randomly selected animals
were 264 and 358. Ten days following the immunization,
106 breast tumor cells were introduced subcutaneously
into the right hind flanks of both Ad-immune and agematched nonimmune mice and a further 21 days later,
the tumors were treated by intratumoral injection with
5 × 108 p.f.u. of AdmIL-12.1. This treatment has been
shown to induce regression of greater than 75% of treated
tumors.8 We have used two endpoints for analyzing the
effects of Ad-immunity on our tumor therapy: (1) tumor
growth and (2) long-term survival. Tumors in both Adimmune and nonimmune animals regressed following
treatment and responded equivalently to AdmIL-12.1
(Figure 1 and Table 1). This response was specific to the
AdmIL-12.1 virus because tumors treated with an E1deleted control virus, dl70-3, grew progressively in both
groups of mice. Long-term survival was also similar
between the two groups (Figure 2), although the Adimmune mice treated with both dl70-3 and AdmIL-12.1
appear to have slightly increased survival by comparison
to nonimmune controls. Thus, it appeared that preimmunity to Ad did not reduce the efficacy of the
AdmIL-12.1 treatment indicating that humoral immunity
to Ad may not present a serious barrier to the clinical
application of this treatment.
Tumor regression in immune animals is associated with
the development of antitumor immunity
As further evidence that antitumor immunity was
achieved in the Ad-immune animals treated with AdmIL12.1, animals which were cured by this treatment were
Figure 1 Effect of anti-adenovirus immunity on regression following
intratumoral injection with AdmIL-12.1. Tumor bearing animals (five per
group) were injected intratumorally with 5 × 108 p.f.u. of AdmIL-12.1 or
the control virus, dl70-3. Randomly selected animals were immunized
with wild-type Ad5 4 weeks before the AdmIL-12.1 therapy. Open symbols, non-Ad-immune; closed symbols, Ad-immune; circles, dl70-3 treated;
squares, AdmIL-12.1 treated.
tested for anti-PyMT CTL. Cytolytic effectors were prepared by co-culturing splenocytes from cured animals
with 516MT3 cells, a nontumor cell line which expresses
high levels of the PyMT antigen. A representative experiment is shown in Figure 3. Splenocytes were prepared
from two Ad-immune mice 21 days following treatment
with AdmIL-12.1 (the tumor on each animal had
regressed at this point). Both spleens contained high
levels of anti-PyMT CTL (.50% lysis at effector to target
ratios of 10:1) and the lytic activity was sensitive to the
presence of anti-CD3 Ab demonstrating that the
responding population was of T cell origin (data not
shown). A total of five animals from three different
experiments were assayed for the presence of Anti-PyMT
CTL and all exhibited similar levels of CTL (data not
shown). We also tested for CTL activity in the spleens
from Ad-immune tumor-bearing animals 21 days following treatment with dl70-3. Only weak anti-PyMT CTL
(.30% lysis at effector to target ratios of 90:1) was detectable in these animals (Figure 3b). The generation of antiPyMT CTL in these experiments was similar to that
observed in nonimmune animals treated with either
AdmIL-12.1 or dl70-3 (C Addison, manuscript in
preparation). The presence of CTL specific for tumor antigen confirms the development of antitumor immunity in
Ad5-immune animals following AdmIL-12.1 treatment.
Dissemination of Ad following intratumoral administration
To monitor virus dissemination following intratumoral
injection in our model, we used a replication-defective
vector-expressing luciferase (AdDK2) as a marker to
detect infected cells in various tissues. Six animals
received an intratumoral injection of AdDK2 (5 × 108
p.f.u.), tissues were harvested from two animals at 24, 72
and 124 h after injection and assayed for luciferase content (Figure 4). Twenty-four hours following virus inoculation, very high levels of luciferase could be detected
within the tumor (.1000 ng). Luciferase activity was also
detectable in other organs at 24 h indicating that the virus
had disseminated, however, the extratumoral luciferase
Immunity to adenovirus prevents vector dissemination
JL Bramson et al
1071
Table 1 Summary of responses following intratumoral injection of AdmIL-12.1
Immune status
Ad-immune (%)
Non-immune (%)
Complete regressiona
Partial regressionb
Growth delayc
No response
4 (24)
0 (0)
6 (35)
11 (69)
5 (29)
3 (19)
2 (12)
2 (13)
Animals with established tumors were treated by intratumoral administration of 5 × 108 p.f.u. AdmIL-12.1 and the resultant changes
in tumor volume were measured regularly. The Table summarizes the results of four experiments involving four to five mice each.
Randomly selected animals were immunized by intranasal delivery of 108 p.f.u. of Ad5 4 weeks before the AdmIL-12.1 therapy.
a
Tumors regressed completely and did not return.
b
Tumors initially regressed to less than 50% of the original volume but were not eliminated and eventually returned.
c
Tumors displayed no change in size for at least 2 weeks following treatment.
Figure 2 Long-term survival following treatment with AdmIL-12.1. The
percentage of surviving animals in each experimental group (13 animals,
three experiments) is plotted as a function of time. Open symbols, nonAd immune; closed symbols, Ad-immune; circles, dl70-3 treated; squares,
AdmIL-12.1 treated.
represented less than 1% of the total luciferase in the animal. The highest expression of luciferase outside the
tumor at this time was found in the liver (5–6 ng) and
the fatpad adjacent to the tumor (approximately 2–3 ng).
By 72 h, luciferase expression was lower in the tumor
(800–900 ng) and expression in the fatpad was reduced
as well (0.005 ng). In contrast, the quantity of luciferase
in the liver had increased to .1000 ng, indicating that a
substantial amount of virus had disseminated from the
tumor and was expressing in peripheral organs. High
levels of expression seen in the liver are consistent with
previous studies showing that the liver is the major site
of infection following administration of Ad intravenously12 or by intraperitoneal injection. 13 Similarly
expression of luciferase was increased in the lung, spleen
and kidney but these were not major sites of dissemination since the combined activity in these organs was
,1% of the activity in the liver. By day 6, luciferase
expression was reduced in all organs. The kinetics of
luciferase expression in the tumor was consistent with
previous results from our laboratory obtained by monitoring expression of mIL-12 in tumor tissue.8 To determine if dissemination is a general effect following intratumoral virus administration, we have analyzed a total
Figure 3 Anti-polyoma middle T (PymT) specific CTL are present in the spleens of Ad-immune animals cured of tumor by AdmIL-12.1 treatment.
Spleens were removed from the mice following regression of the tumor and anti-PymT effectors were prepared as described in the Materials and methods.
The effectors were tested in a standard 51Cr-release assay on PyMT-expressing 516MT3 cells (PyMT+, solid lines) and the parental cells, PT0516 (mock,
dashed lines). (a) Anti-PymT CTL from two Ad-immune animals 21 days after treatment with AdmIL-12.1. (b) Anti-PymT CTL from two Ad-immune
animals 21 days after treatment with dl70-3.
Immunity to adenovirus prevents vector dissemination
JL Bramson et al
1072
Figure 4 Time-course of AdDK2 dissemination following intratumoral injection. Tumor-bearing mice received an intratumoral injection of 5 × 108 p.f.u.
of AdDK2 on day 0. On days 1, 3 and 6, two animals were killed and various tissues were removed, homogenized and assayed for luciferase activity
as an indication of viral dissemination.
of seven nonimmune animals for luciferase expression in
the tumor and liver on day 3. The mean expression of
luciferase in the tumor was 766 ± 250 and the mean
expression in the liver was 762 ± 265. Our results are
similar to those of Toloza et al14 who demonstrated that
Ad vectors can disseminate following intratumoral injection and infect peripheral organs, although they used a
higher virus load (2 × 109 p.f.u.).
Antibodies to adenovirus prevent virus dissemination
We have shown that Ad immunization before therapy
does not prevent regression following intratumoral injection of AdmIL-12.1. It would be expected that the circulating Abs generated following immunization could protect sites like the liver from disseminated virus which
probably gains access to the organs through the bloodstream. To test this hypothesis, we repeated the experiment outlined in the previous paragraph using Adimmune mice. Tumor-bearing animals, both naive and
Ad-immune, received an inoculum of 5 × 108 p.f.u.
AdDK2 intratumorally. Twenty-four and 72 h later,
tissues were harvested, homogenized and assayed for
luciferase activity. At 24 h, when we observed peak transgene expression in the tumor tissue, total luciferase
within the tumors of immunized mice was only reduced
by an average of 2.4-fold compared to nonimmune animals (compare 1282 ± 273 ng luciferase in nonimmune
animals (n = 4) to 544 ± 195 in Ad-immune animals
(n = 4); Figure 5a). At 72 h, when the expression within
the liver was maximal, transgene expression in the liver
was reduced by .1000-fold (compare 762 ± 265 ng in
nonimmune animals (n = 7) to 0.29 ± 0.09 ng in Adimmune animals (n = 4); Figure 5b) while tumor-associated luciferase was only 6.8-fold lower in the Ad-immune
mice (compare 762 ± 246 in nonimmune animals (n = 7)
to 113 ± 71 in Ad-immune animals (n = 4)). Thus, it
appeared that the inhibitory effect of the neutralizing
antibodies was much greater in the liver than in the
tumor. Similarly, dissemination of AdDK2 to other
organs (spleen, lung, kidney) in the Ad-immune mice
was also reduced (data not shown). Thus, Ad-immunity
appears to be beneficial in this setting by inhibiting the
dissemination of virus from the site of injection and protecting peripheral tissue from any adverse effects of
gene transfer.
Discussion
Preclinical studies have clearly demonstrated that the
presence of neutralizing Abs may be a major limiting factor in effective Ad-mediated gene delivery in vivo.3–5 This
has led to speculation regarding the effectiveness of Ad5based gene therapy, as the majority of the population has
been exposed to this virus during their lifetime. However, these studies have focused on repetitive administration of Ad to tissues like the liver and the lungs which
are in direct contact with the systemic circulation and
mucosal surfaces and consequently are readily protected
by anti-Ad Abs. Solid tumors, on the other hand, are
often poorly vascularized and therefore may not be penetrated by Abs to as high a degree as other tissues, so it
is reasonable to postulate that the effects of neutralizing
Abs would be reduced within the microenvironment of
the tumor.
To examine the effects of Ad immunity on tumor gene
therapy using an Ad expressing IL-12 (AdmIL-12.1), we
immunized animals before inoculation with tumor cells.
Once the tumors had established, the vector was administered directly into the tumor. The efficacy of the treatment did not appear to be diminished in the immune
animals, supporting our hypothesis that the neutralizing
Abs may not be a major limitation in the context of the
tumor nodule. The incidence of tumor regression in both
Ad-immune and nonimmune groups was between 60
and 70%, consistent with our previous observations.8 The
results obtained with the nonimmune group (no permanent cures) differ from results reported in a previous
publication in which a cure rate of approximately 30%
was obtained for nonimmune mice treated with AdmIL12.1. The previous study involved a larger cohort (36
mice) compared to the current study (16 mice) and we
have found that although the overall cure rate with
AdmIL-12.1 is approximately 30–40%, this parameter can
fluctuate from one experiment to the next. In some
Immunity to adenovirus prevents vector dissemination
JL Bramson et al
1073
Figure 5 Neutralizing antibodies to Ad inhibit virus dissemination to the liver. Tumor-bearing mice received an intratumoral injection of 5 × 108 p.f.u.
of AdDK2 on day 0. Randomly selected animals were immunized intranasally with Ad5 before tumor establishment. Twenty-four (a) and 72 h (b)
following AdDK2 administration, tissues were harvested and assayed for luciferase activity. Each bar is representative of four to seven mice. Shaded
bars, non-immune mice; closed bars, Ad-immune mice.
experiments we failed to achieve permanent cures while
in others 100% of the treated tumors underwent complete
and permanent regression (J Bramson, unpublished
observations). Although it may seem that the results of
the AdmIL-12.1 are variable, it should be noted that in
both the present and previous reports the incidence of
tumor regression was between 60 and 80% of the AdmIL12.1-treated animals and these animals survived at least
twice as long as the dl70-3 controls. Thus, while there was
variability in the frequency of complete cures without
relapse, the outcome of the AdmIL-12.1 therapy in terms
of tumor regression and increased survival is highly
reproducible and is not impaired when animals are
immunized against Ad before therapy.
To make a better assessment of the effect of Adimmunity on transgene expression in the tumor, we used
a luciferase expression vector (AdDK2) to allow quantification of transgene expression. By analyzing peak
luciferase levels in the tumor, we found that expression
was only reduced an average of 2.4-fold in tumors of Adimmune animals compared to naive mice. It is interesting
that the IL-12 therapy was not affected by the state of
immunity of the animal even though expression was
reduced in Ad-immune animals. There are at least two
possible explanations for this: first, the dose of virus we
are using is above the minimal amount required to
induce an effective response. We have found that
2.5 × 108 p.f.u. AdmIL-12.1 is as effective as the dose we
used in these studies, 5 × 108 p.f.u.8 Second, administration of Ad5 to the airway mucosae would result in the
development of antiviral CTL,3 and T cell responses
against the Ad vector in the tumor could make up for
the loss of IL-12 expression. Immune cells responding to
Ad-infected cells could clear the infected cells and secrete
cytokines such as IL-2 and IFN-g into the microenvironment of the tumor which could act as an adjuvant to the
IL-12 being expressed by the AdmIL-12.1 vector. Consistent with this hypothesis, tumor-bearing Ad-immune
mice treated with the control virus (dl70-3) survived
longer than the respective nonimmune mice. The anti-Ad
response alone, however, was not sufficient to cause
tumor regression.
Previously, Toloza et al,14 using a viral load of 2 × 109
p.f.u. per tumor, have demonstrated that an Ad vector
can disseminate from the site of the tumor and infect
peripheral organs. We found that even with a lower dose
(5 × 108 p.f.u. per tumor), the virus can still disseminate
and infect peripheral organs such as the liver and the
spleen. Quite strikingly, the transgene expression in the
liver at day 3 was almost equal to that in the tumor and
in certain animals it was greater. In the context of using
Ad vectors for gene therapy of solid tumors, it had
always been hoped that the virus would primarily infect
the tissues in the vicinity of the site of inoculation. However, it is quite clear from both sets of studies that the
virus can enter the circulation following intratumoral
administration and infect peripheral organs. The
observed dissemination is probably due to incomplete
adsorption at the site of injection since we have used an
E1-deleted replication-defective Ad in a species which is
nonpermissive for viral replication. One danger of circulating virus is that it will infect the liver giving rise to
inflammation and damage. We have found that liver
damage is apparent following in vivo delivery of Ad vectors regardless of the route of administration (footpad,
intramuscular or subcutaneous).15 This trauma to the
liver can then be compounded by the nature of the transgene, which in the case of cancer gene therapy may
encode a cytokine or prodrug activating enzyme. Thus,
disseminated virus may be a concern to the health of
the patient.
As we discussed above, the presence of neutralizing
Abs has been found to be inhibitory to successful Admediated gene transfer to the liver. When we examined
dissemination of AdDK2 by monitoring transgene
expression in the liver, we found drastically reduced levels in the livers of Ad-immune mice. While peak transgene expression was only reduced by 2.4-fold in the
tumors of immune animals, peak expression in the liver
was reduced by .1000-fold. Therefore, the presence of
Immunity to adenovirus prevents vector dissemination
JL Bramson et al
1074
neutralizing Abs to Ad was found to be protective to the
host by inhibiting the dissemination of virus and spurious expression in peripheral organs. Our observations
indicate that if organ toxicity is observed during the
course of the clinical trials, it may be beneficial to immunize patients undergoing Ad-based therapies before the
initiation of the clinical protocol. Alternatively, techniques for passive immunity may be considered which
would simply involve administration of anti-Ad polyserum during the first few days or weeks of treatment.
The present studies have further substantiated the
approach of using Ad vectors for cancer gene therapy. It
is clear that, at least for solid tumors, the presence of Adimmunity may not be the impediment that was originally
expected. Although peak expression was slightly reduced
in the tumors of Ad-immune individuals, it should be
possible to increase expression levels by boosting the
virus inoculum. The immune response against Ad antigens in virally infected cells could also limit therapies
where long-term expression is important. Fortunately,
the advent of nth generation viruses, which are fully
deleted of viral genes, should overcome that obstacle.16–19
In the absence of viral gene expression, Ad-immune individuals should not clear the transduced tumor cells any
more rapidly than naive individuals. Of course further
studies are required to determine whether the results
presented here apply to other therapies such as those
involving HSVtk and CDA. Thus, continued preclinical
experimentation should yield improved treatment strategies that can be applied effectively in a clinical setting.
Materials and methods
Animals, cell culture and reagents
Six- to 8-week-old FVB/n female mice were obtained
from Taconic Farms (Germantown, NY, USA) and
housed in a specific pathogen-free facility until use. The
293 cells (adenoviral E1 transformed human embryonic
kidney cells20) were maintained in F-11 medium supplemented with 10% newborn bovine serum. The target
lines for the CTL assay, PT0516 and 516MT3, are nontumor kidney fibroblast lines derived from an FVB
mouse (C Addison, manuscript in preparation). The
516MT3 cell line was generated by stably transfecting the
PT0516 cells with the polyoma middle T cDNA (PyMT).
The PT0516, 516MT3 and Hela cells were cultured in aMEM supplemented with 10% fetal calf serum (FCS).
Splenocyte culture was performed in RPMI-1640 with
10% FCS, 20 mm HEPES, 50 mm b-mercaptoethanol. All
cell culture medium was supplemented with 2 mm l-glutamine, 100 mg/ml penicillin and 100 U/ml streptomycin.
Cell culture media, reagents and Nunc tissue culture dishes were obtained from Life Technologies (Gaithersburg,
MD, USA).
Adenoviral vectors
Viruses were propagated on 293 cells and purified by cesium chloride gradient centrifugation as described.21 Construction and characterization of an Ad vector
(AdmIL12.1) expressing murine IL-12 has been described
previously.22 The Ad5-based recombinant system used to
produce these vectors allows for the insertion of
expression cassettes in either the E1 or E3 region of Ad5.23
The expression cassette for the p35 subunit of IL-12 is in
the E1 region of AdmIL12.1 and the IL-12 p40 subunit
cassette is in the E3 region. Expression of each IL-12
cDNA is driven by the human cytomegalovirus (HCMV)
immediate–early gene promoter and terminated by the
polyadenylation signal of SV40 and transcription is in the
same direction as the E1 and E3 transcription units they
replace. The luciferase expressing virus, AdDK2, contains
an expression cassette in E1 consisting of the murine
cytomegalovirus immediate–early promoter, the cDNA
for firefly luciferase, and the polyadenylation signal of
SV40.24 AdLacZ is a replication-competent virus which
contains an expression cassette encoding the E. coli bgalactosidase enzyme in E3.25 The control adenovirus,
dl70-3, is a mutant of Ad5 deleted in E1 and having a
deletion/substitution in E3.23
Neutralizing antibody assay
Serum samples were tested for neutralizing antibodies
using an assay developed by L Prevec at McMaster University (unpublished). Serial five-fold dilutions of serum
were prepared in PBS with Mg++ and Ca++ (PBS++). An
aliquot of 150 ml of diluted serum was mixed with a
150 ml aliquot of AdLacZ diluted to 7.5 × 105 p.f.u./ml in
PBS++ and incubated at 37°C for 1 h. An aliquot of 100 ml
of the serum/AdLacZ mixture was added to one well of
a 24-well plate seeded 6 h earlier with 2 × 105 Hela cells.
After incubating for 1 h at 37°C, nonadsorbed virus was
washed out of the wells and the plates were incubated a
further 24 h. The cells in each well were then lysed by
the addition of 100 ml of 250 mm Tris (pH 7.8), 0.5% NP40 and incubated on ice for 20 min. An 875 ml aliquot of
10 mm KCl, 1 mm MgSO4, 50 mm b-mercaptoethanol in
100 mm sodium phosphate buffer (pH 7.5) was added to
each well and the plates were incubated at 37°C for 5
min. A 330 ml aliquot of the substrate solution (4 mg/ml
O-nitrophenol-b-d-galactopyranoside (Sigma, Oakville,
Ontario, Canada) in 100 mm sodium phosphate buffer,
pH 7.5), was then added to each well and incubated at
37°C for 1 h. The reaction was terminated by the addition
of 430 ml of 1 m Na2CO3 and the optical density of the
solution in each well was measured at 450 nm. The relative absorbance for each dilution of anti-serum is
presented as a fraction of the absorbance produced using
a corresponding dilution of control serum. The titer of
neutralizing Ab was expressed as the inverse of the
dilution required to produce 50% reduction in lacZ
expression as measured by absorbance at 450 nm.
Tumor studies
The tumor model used in this study is based on a transgenic mouse strain which harbors the PyMT gene under
the control of the mouse mammary tumor virus long terminal repeat.18 Female transgenic mice develop multifocal mammary carcinomas by 8–10 weeks of age at which
point the tumors are excised and transplanted ectopically
to syngeneic FVB mice (eg by subcutaneous injection of
single cell suspensions into the hind flanks), where they
will induce tumors and grow indefinitely (typically the
animals must be killed at 40–60 days). Briefly, tumors
were established in syngeneic mice as follows: tumors
were excised from dead transgenic mice and a single cell
suspension was prepared as described.6 An aliquot of 106
tumor cells was injected subcutaneously in the right hind
flank of normal FVB mice. Randomly selected animals
were immunized intranasally with 10 ml of wild-type
Immunity to adenovirus prevents vector dissemination
JL Bramson et al
Ad5 (1 × 108 p.f.u.) 7–10 days before tumor cell transplantation. Approximately 21 days after tumor cell inoculation, visible tumors had developed in all the recipients
and at this time the mice were injected intratumorally
with the appropriate concentration of virus in a volume
of 50 ml. Tumors were measured biweekly using calipers
and the tumor volume was calculated from the longest
diameter and average width assuming a prolate spheroid. The animals were killed when the longest diameter
was greater than 20 mm or when any two measurements
were greater than 10 mm.
Virus dissemination was determined by harvesting
tumors and various tissues at 24, 72 and 144 h after intratumoral injection of AdDK2. The tissues were snapfrozen in liquid nitrogen and maintained at −70°C. To
measure luciferase expression in these tissues, the frozen
samples were homogenized in 0.1 m potassium phosphate buffer (pH 7.8) containing 100 mm phenylmethylsulfonyl fluoride and 10 mg/ml aprotonin (Sigma).
Homogenized samples were then sonicated three times
for 10 s. Supernatants were cleared by centrifugation in
a microfuge at maximum speed and used immediately in
luciferase assays as described.13
CTL assay
The spleens from treated animals were removed aseptically and unicellular suspensions were prepared by passing the spleens through a metal mesh. Red blood cells
were lysed by suspending the splenocytes in 5 ml of
0.83% ammonium chloride. The splenocytes were cultured in the presence of irradiated (5000 rad) PyMT
expressing 516MT3 cells at a ratio of 100 splenocytes to
one stimulator cell. Five days later, the cells were harvested and tested for cytolytic activity against 516MT3
cells and the parental PT0516 cells which were both labeled with 51Cr (100 mCi/106 cells; Mandel Scientific,
Guelph, Ontario, Canada). Effector:target ratios ranging
from 90:1 to 3.3:1 were prepared in a final volume of
250 ml in V-bottomed 96-well plates. The background and
maximum release were determined by adding the target
cells (in 50 ml) to 200 ml of medium alone and 1 N hydrochloric acid, respectively. A 3 ml aliquot of anti-CD3
ascitic fluid (145-2C11) was added to certain wells to control for T cell mediated killing. The % specific lysis was
evaluated as:
c.p.m. experimental − c.p.m. background
× 100 .
c.p.m. maximum − c.p.m. background
Acknowledgements
We thank Uma Sankar, Duncan Chong and John Rudy
for their expert technical assistance. This work was supported by grants from the National Cancer Institute of
Canada (NCIC), the Medical Research Council of Canada
(MRC), Baxter Healthcare and London Life Insurance. FL
Graham is a Terry Fox Research Scientist of the NCIC
and JL Bramson is supported by a MRC fellowship.
References
1 Hitt M, Addison CL, Graham FL. Human adenovirus vectors
for gene transfer into mammalian cells. In: August T et al (eds).
Gene Therapy (Advances in Pharmacology, vol. 40). Academic Press:
San Diego, 1997, pp 137–206.
2 Yang Y et al. Cellular immunity to viral antigens limits E1deleted adenoviruses for gene therapy. Proc Natl Acad Sci USA
1994; 91: 4407–4411.
3 Yang Y, Li Q, Ertl HCJ, Wilson JM. Cellular and humoral
immune responses to viral antigens create barriers to lungdirected gene therapy with recombinant adenoviruses. J Virol
1995; 69: 2004–2015.
4 Dai Y et al. Cellular and humoral immune responses to adenoviral vectors containing factor IX gene: tolerization of factor IX
and vector antigens allows for long-term expression. Proc Natl
Acad Sci USA 1995; 92: 1401–1405.
5 Gilgenkrantz H et al. Transient expression of genes transferred
in vivo into heart using first-generation adenoviral vectors: role
of the immune response. Hum Gene Ther 1995; 6: 1265–1274.
6 Addison CL et al. Intratumoral injection of an adenovirus
expressing interleukin 2 induces regression and immunity in a
murine breast cancer model. Proc Natl Acad Sci USA 1995; 92:
8522–8526.
7 Addison CL, Gauldie J, Muller WJ, Graham FL. An adenoviral
vector expressing interleukin-4 modulates tumorigenicity and
induces regression in a murine breast cancer model. Int J Oncol
1995; 7: 1253–1260.
8 Bramson J et al. Direct intratumoral injection of an adenovirus
expressing IL-12 induces regression and long-lasting immunity
which is associated with highly localized expression of IL-12.
Hum Gene Ther 1996; 7: 1995–2002.
9 Chen S-H et al. Gene therapy for brain tumors: regression of
experimental gliomas by adenovirus-mediated gene transfer in
vivo. Proc Natl Acad Sci USA 1994; 91: 3054–3057.
10 Hirschowitz EA et al. In vivo adenovirus-mediated gene transfer
of the Escherichia coli cytosine deaminase gene to human colon
carcinoma-derived tumors induces chemosensitivity to 5fluorocytosine. Hum Gene Ther 1995; 6: 1055–1063.
11 Guy CT, Cardiff RD, Muller WJ. Induction of mammary tumors
by expression of polyoma middle T oncogene: a transgenic
mouse model for metastatic disease. Mol Cell Biol 1992; 12:
954–961.
12 Huard J et al. The route of administration is a major determinant
of the transduction efficiency of rat tissues by adenoviral recombinants. Gene Therapy 1995; 2: 107–115.
13 Mittal SK et al. Monitoring foreign gene expression by a human
adenovirus-based vector using the firefly luciferase gene as a
reporter. Virus Res 1993; 28: 67–90.
14 Toloza EM et al. In vivo cancer gene therapy with a recombinant
interleukin-2 adenovirus vector. Cancer Gene Ther 1996; 3: 11–17.
15 Wan Y et al. Dendritic cells transduced with an adenovirus vector encoding a model tumor-associated antigen for tumor vaccination. Hum Gene Ther 1997; 8: 1355–1360.
16 Kochanek S et al. A new adenoviral vector: replacement of all
viral coding sequences with 28 kb of DNA independently
expressing both full-length dystrophin and beta-galactosidase.
Proc Natl Acad Sci USA 1996; 93: 5731–5736.
17 Fisher KJ et al. Recombinant adenovirus deleted of all viral genes
for gene therapy of cystic fibrosis. Virol 1996; 217: 11–22.
18 Parks RJ et al. A helper-dependent adenovirus vector system:
removal of helper virus by Cre-mediated excision of the viral
packaging signal. Proc Natl Acad Sci USA 1996; 93: 13565–13570.
19 Hardy S et al. Construction of adenovirus vectors through Crelox recombination. J Virol 1997; 71: 1842–1849.
20 Graham FL, Smiley J, Russell WC, Nairn R. Characteristics of a
human cell line transformed by DNA from human adenovirus
5. J Gen Virol 1977; 36: 59–72.
21 Hitt M, Bett AJ, Prevec L, Graham FL. Construction and propagation of human adenovirus vectors: In: Celis JE (ed). Cell
Biology: A Laboratory Handbook. Academic Press: San Diego, 1994,
pp 479–490.
22 Bramson J et al. Construction of a double recombinant adenovirus vector expressing a heterodimeric cytokine: in vitro and in
vivo production of biologically active interleukin-12. Hum Gene
Ther 1996; 7: 333–342.
1075
Immunity to adenovirus prevents vector dissemination
JL Bramson et al
1076
23 Bett AJ, Haddara W, Prevec L, Graham FL. An efficient and
flexible system for construction of adenovirus vectors with
insertions or deletions in early regions 1 and 3. Proc Natl Acad
Sci USA 1994; 91: 8802–8806.
24 Addison CL, Hitt M, Kunsken D, Graham FL. Comparison of
the human versus murine cytomegalovirus immediate early gene
promoters for transgene expression by adenoviral vectors. J Gen
Virol 1997; 78: 1653–1661.
25 Bett AJ, Prevec L, Graham FL. Packaging capacity and stability
of human adenovirus type 5 vectors. J Virol 1993; 67: 5911–5921.