Previous Article | Next Article 
Journal of Virology, November 1998, p. 9201-9207, Vol. 72, No. 11
0022-538X/98/$04.00+0
Copyright © 1998, American Society for Microbiology. All rights reserved.
Combination Gene Delivery of the Cell Cycle
Inhibitor p27 with Thymidine Kinase Enhances Prodrug
Cytotoxicity
Xavier
Danthinne,1,
Kazunori
Aoki,1,2
Akiko L.
Kurachi,1
Gary J.
Nabel,1,2,3 and
Elizabeth G.
Nabel1,*
Howard Hughes Medical
Institute2 and
Departments of Internal
Medicine1 and
Biological
Chemistry,3 University of Michigan Medical
Center, Ann Arbor, Michigan 48109-0644
Received 9 March 1998/Accepted 6 August 1998
 |
ABSTRACT |
Cytoxicity induced by the herpesvirus thymidine kinase (TK) gene in
combination with prodrugs is dependent on cell growth and leads to the
elimination of genetically modified cells, thus limiting the duration
of expression and efficacy of this treatment in vivo. Here, an effort
was made to enhance TK/prodrug efficacy by coexpression of a
cyclin-dependent kinase inhibitor (CKI), p27, to render cells resistant
to TK/prodrug by inhibiting DNA synthesis. Expression of p27 by
transfection substantially reduced cell cycle progression, and its
activity was enhanced by mutations designed to stabilize the protein.
Coexpression of p27 and TK or a p27/TK fusion protein led to greater
prodrug cytotoxicity than that produced by TK alone in the Renca cell
line, which is sensitive to bystander killing. Combination gene
transfer of this CKI with TK therefore sustained the synthesis of TK by
genetically modified cells to enhance the susceptibility of bystander
cells to prodrug cytotoxicity and increased the efficacy of this gene transfer approach.
| |
INTRODUCTION |
The thymidine kinase (TK) gene from
herpes simplex virus type 1 (HSV-1) is widely used as a cytotoxic gene
in combination with prodrugs in different mammalian gene transfer and
transgenic systems. HSV TK phosphorylates the guanosine analogs
ganciclovir (GCV) and acyclovir (ACV) more efficiently than cellular TK
does, and the monophosphate drugs are subsequently phosphorylated by cellular enzymes into their triphosphate forms (3), which
are incorporated into elongating DNA, leading to elongation arrest (ACV) or decreased DNA synthesis (GCV) (4, 5, 14, 17, 21,
39). Death usually ensues, through a mechanism identified in some
cases as apoptosis (7, 35), although the mechanism and
pathways that lead to cell death are not completely understood.
One feature of this gene transfer/prodrug approach is the generation of
bystander cytotoxicity that leads to the death of untransduced cells
adjacent to genetically modified cells. Several potential mechanisms
have been proposed to mediate this phenomenon. Freeman et al.
hypothesized that the uptake of phosphorylated GCV by bystander cells
occurs via the endocytosis of apoptotic vesicles, originating from the
TK-transduced cells and containing the toxic drug (12);
however, increasing evidence suggests that the bystander effect is
mediated via gap junctions that allow phosphorylated ganciclovir to
translocate from TK+ to TK
cells
intercellularly (2, 11). Although the bystander effect can
be observed in vitro, an immune component might be involved in some
tumor models since this phenomenon is impaired or even absent in
immunocompromised animals (7, 13).
The TK/GCV system has been successfully applied in cancer and
cardiovascular models in vivo (8, 10, 21, 22, 25, 29);
however, the efficiency of gene delivery in vivo remains low. Because
of their potential antitumor activity, cytokines have been combined
with TK. Ram et al. constructed retroviral vectors carrying both the
HSV TK and interleukin-2 (IL-2) genes, but no enhancement of tumor
eradication was observed upon transduction of rat 9L gliosarcoma
(31). Cotreatment of established tumors with TK- and
IL-2-expressing adenoviral vectors was shown to enhance eradication of
metastatic colon carcinoma in mouse liver (6) and head and
neck cancer in mice (26, 27). In nude mice, coinjection of
C6 glioma cells with retroviral producer cells expressing TK and IL-4
appeared to inhibit tumor growth more effectively than coinjection with
cells expressing TK only (1). In another approach, Rogulski
et al. fused the sequences encoding TK and Escherichia coli
cytosine deaminase and observed a slight synergistic toxicity and an
enhanced radiosensitivity in glioma cells (33).
In this study, we have explored an alternative strategy to increase
cell killing by TK/GCV. Since administration of GCV in the presence of
HSV TK leads to lysis caused by its effects on DNA replication, we
hypothesized that the growth arrest of gene-modified cells would render
them less sensitive to TK/GCV-mediated killing and prolong the duration
of TK expression, thereby sustaining local conversion of GCV and the
cytotoxic effect on adjacent cells. We have found that combination gene
transfer of TK and a cyclin-dependent kinase inhibitor (CKI) enhanced
bystander cell killing in the presence of GCV.
| |
MATERIALS AND METHODS |
Plasmids.
The cDNAs which encode human p21
(42), p16 (37), p27 (30), HSV-1 TK,
human alkaline phosphatase (hAP) (24), and human CD2 were
inserted in VR1012, a eukaryotic expression vector that contains a
cytomegalovirus (CMV) immediate-early gene promoter, enhancer, and
intron and a bovine growth hormone polyadenylation signal. A plasmid
expressing human immunodeficiency virus type 1 (HIV-1) Vpr under
control of the CMV immediate-early gene promoter and a simian virus 40 polyadenylation signal was a gift from E. Cohen (University of
Montreal, Montreal, Canada).
A bicistronic construct expressing p27 and TK (pCMVp27citeTK) was made
by insertion of the EcoRI-NcoI fragment from
pCITE-1 (Novagen, Madison, Wis.) between an XbaI site
located immediately downstream from the p27 coding sequence and an
NcoI site containing the initiator codon of the TK gene.
This EcoRI-NcoI fragment ("CITE") contains a
copy of the encephalomyocarditis virus RNA 5' noncoding region, which
functions as an internal entry point for initiation of translation by
eukaryotic ribosomes. As a control for p27 activity, a vector
containing the p27 coding region but in reverse orientation with
respect to the CMV promoter, pCMVp27revcite TK, was prepared similarly.
To reduce the size of the expression cassette, a
SacII-EcoRV fragment containing the CMV intron
was deleted in both vectors.
Mutation of the Cdc2 kinase consensus phosphorylation site on p27 from
TPKK to AAGG was performed by using overlapping PCR-based
methods with
plasmid pCMVp27citeTK as a template. On one side,
sequences
corresponding to nucleotides 186 to 576 from the start
of the p27
coding region were amplified by using the oligonucleotides
26 (5'-CGATTTTCAGAATCACAAACCCC-3') and 24 (5'-GCCAGGC
CCC
CCGG
CCG
CCTGCTCCACAGAACC-3')
as primers. On the other side, sequences corresponding to
position
554 from the start of the p27 coding region to the
BglI site located
in the downstream CITE sequences were
amplified by using the oligonucleotides
23 (5'-GAGCAG
GCG
GCC
GGG
GGGCCTGGCCTCAGAAG-3')
and 27 (5'-TTTGGCCGCAGAGGCACCTGT-3'). Mutations in
oligonucleotides
23 and 24 are indicated in boldface type. Both PCR
products were
amplified in a single reaction by using oligonucleotides
26 and
27 as primers, with 6 cycles (94°C, 15 s; 45°C, 30 s; 72°C, 45
s) followed by 30 cycles (94°C, 15 s; 65°C,
30 s; 72°C, 45 s).
The resulting DNA fragment was digested
with
SacII and
XbaI and
inserted into
pCMVp27citeTK to replace the corresponding fragment.
The integrity of
the sequences was verified by sequencing.
A fusion protein between p27 and TK was made by deleting an
AatII-
NcoI fragment from pCMVp27citeTK, giving
rise to plasmid
pCMVp27TK. The resulting protein had the last four
amino acids
of p27 (RRQT) deleted, and an additional serine residue was
inserted
in front of the first methionine residue of the TK. A fusion
between
the NH
2-terminal part of the p27 coding region and
the sequences
encoding the TK was created by deleting the
SacII-
NcoI fragment
from pCMVp27citeTK, giving
rise to plasmid pCMVp27SNTK. Similarly,
the
SacII-
FspI and
MarI-
FspI
fragments were prepared by deletion
of this fragment from pCMVp27TK,
pCMVp27SFTK, and pCMVp27NFTK,
respectively. The open reading frame
between the
SacII and
FspI
sites was maintained
by inserting complementary oligonucleotides
(5'-GGTCGAC-3'
and 5'-GTCGACCGC-3'). Introduction of the
NH
2-terminal
part of p27 downstream of the cyclin-CDK2
binding domain of p27
was performed by ligating a
NcoI-
HindIII fragment from plasmid
VR1012-p21N between the
SacII and
FspI sites of
pCMVp27TK, giving
rise to plasmid pCMVp27Sp21FTK. VR1012/p21N contains
a copy of
the sequences coding for the first 75 amino acids of p21.
Similarly,
pCMVp27Np21FTK was constructed by inserting the same
NcoI-
HindIII
fragment between the
NarI and
FspI sites of pCMVp27TK.
293 cell transfections and fluorescence-activated cell sorter
(FACS) analysis.
293 cells were maintained in Dulbecco's modified
Eagle's medium supplemented with 10% fetal calf serum at 37°C and
5% CO2. Cells (2 × 106) inoculated the
previous day in 10-cm-diameter culture dishes were transfected with 15 µg of plasmid DNA by using CaPO4 transfection. For cell
cycle analysis, 293 cells were typically transfected with 3 µg of a
CD2 expression plasmid and 12 µg of the relevant CKI expression
plasmid.
One day after transfection, cells were detached from the tissue culture
dish with phosphate-buffered saline (PBS) containing
2 mM EDTA. Cell
clusters were disrupted by pipetting, and 10
6 cells were
plated in a 15-cm diameter dish. The next day, the
cells were
harvested, and the CD2 cells were analyzed for DNA
content by flow
cytometry as previously described (
36). Briefly,
10
6 cells were incubated with 50 µl of anti-CD2 mouse
hydridoma supernatant
(ATCC HB222) for 20 min on ice. The cells were
washed twice with
1 ml of PBS-2% fetal calf serum and incubated with
0.2 µg of fluorescein
isothiocyanate-conjugated sheep anti-mouse
immunoglobulin in 50
µl of PBS-2% fetal calf serum for 20 min on
ice. The cells were
washed with 1 ml of PBS-2% fetal calf serum and
fixed in 0.25%
paraformaldehyde-PBS for 1 h on ice. The fixed
cells were permeabilized
with 0.2% Tween 20-PBS for 15 min at 37°C.
The cells were washed
again with 1 ml of PBS-2% fetal calf serum and
incubated for 1
h at 37°C in 1 ml of PBS containing 30 µg of
propidium iodide
and 2 U of DNase-free RNase (Boehringer Mannheim) per
ml. Fluorescence
was analyzed on a FACScan (Becton Dickinson) flow
cytometer. Data
represent at least 10,000 events corresponding to the
cells expressing
the highest CD2 levels. The DNA profiles were analyzed
by using
Modfit LT software (Verity Software House, Inc.).
Assays of proliferation and bystander effect.
Renca cells
were maintained in RPMI medium supplemented with 10% fetal calf serum
at 37°C and 5% CO2 in 10-cm-diameter dishes until they
reached 90% confluence; they were then transfected with 25 µg of DNA
complexed with 100 µg of Lipofectamine (Gibco BRL). For bystander
experiments, Renca cells were typically transfected with 5 µg of CD2
and 20 µg of TK expression vector plasmids.
One day after transfection, cells were harvested and diluted with
increasing amounts of untransfected cells. A total of 10
4
cells were plated per well in a 96-well microtiter plate and
incubated
for 6 h at 37°C to permit cell adherence to the plate.
The
medium was then changed to fresh medium containing 5 µM GCV.
Cultures
were terminated at 5 days, and cell proliferation was
measured by using
a colorimetric cell proliferation assay (
23).
To determine
transfection efficiencies, 0.5 × 10
6 cells in a
10-cm-diameter culture dish were incubated for 2 days
at 37°C,
harvested, and analyzed by FACS for CD2 expression as
described above.
| |
RESULTS |
Transfection of p27 plasmid induces growth arrest more effectively
than p21, p16, or Vpr.
Because inhibition of DNA synthesis by
TK/GCV has been implicated in cell death, we postulated that arrest of
cell cycle progression may render transfected cells less sensitive to
TK/GCV and prolong TK expression to enhance potency. Inhibition of cell
cycle progression can be achieved by expression of several gene
products, including the p16, p21, and p27 CKIs, which arrest cell cycle
progression in G0 or at the G1/S boundary.
HIV-1 Vpr inhibits the activity of cyclin B-Cdc2 and arrests the cell
cycle at the G2/M checkpoint (9).
The first step was to identify the most potent CKI that could be
combined subsequently with TK. CKIs were expressed under
the control of
the same CMV enhancer-promoter and were cotransfected
with a plasmid
expression vector encoding the cell surface marker
CD2 (pCMV-CD2) into
a highly transfectable cell line, 293 (
15),
or into Renca
cells, a murine epithelial carcinoma cell line.
Cells which expressed
CD2 as a marker were analyzed for their
DNA content by flow cytometry.
In p21- and p27-transfected 293
cells, the proportions of cells in
G
1 were 58 and 76%, respectively,
in comparison to 27%
for the control (Fig.
1). p16 did not
show
activity in 293 cells, although it was readily detected by Western
blot analysis (data not shown) and inhibited Cdk4 kinase activity
(
40). Expression of Vpr led to an accumulation of about 42%
of the cells in the G
2 phase. In p21-, p27-, and
p16-transfected
Renca cells, the percentages of the CD2-positive cells
in G
1 were
66, 76, and 63%, respectively, in comparison to
44% for the control.
Interestingly, Vpr did not show any activity in
Renca cells. Because
p27 showed the greatest reduction in cell cycle
progression in
both cell lines, it was examined further for its effect
on TK/GCV
cytotoxicity in cotransfected cells.
View larger version (27K):
[in this window]
[in a new window]
|
FIG. 1.
Overexpression of p27 arrests the cell cycle more
strongly than overexpression of p21, p16, or Vpr. A vector expressing
p21, p27, p16, Vpr, or human alkaline phosphatase (hAP) under the
control of a CMV enhancer-promoter was transfected into 293 or Renca
cells together with a CD2-expressing plasmid. After 2 days of
expression, the cells were harvested and stained simultaneously with an
anti-CD2 antibody and propidium iodide. The histograms represent the
DNA profiles of the cells expressing the highest CD2 levels. The
fraction of cells in each phase of the cell cycle is indicated above
the corresponding peak. In 293 cells, the percentages of CD2-positive
cells were 62, 60, 67, 60 and 63% for panels from top to bottom,
respectively. In Renca cells, these percentages were 5.7, 6.8, 6.2, 7.8, and 7.4%, respectively.
|
|
A vector which coexpressed p27 and TK was prepared by inserting both
coding sequences in a single transcription unit, with
p27 inserted
downstream from the CMV promoter, followed by a
cis-acting
internal translational entry site (cite) and the HSV-1 TK gene,
giving
rise to pCMVp27citeTK. As a negative control, the p27 coding
sequence
was inserted in antisense orientation to generate pCMVp27revciteTK.
The
ability of these vectors to induce G
1 growth arrest in 293
cells was tested by cotransfection of pCMVp27citeTK and pCMV-CD2.
Two
days after transfection, CD2
+ cells were analyzed for DNA
content. pCMVp27citeTK was comparable
to pCMVp27 in its ability to
cause G
1/S growth arrest, with 55
and 51% of cells in
G
1, respectively (Fig.
2A and
B). The effect
of TK on cell growth was
measured in the presence or absence of
GCV in these expression vector
plasmids. Cells transfected with
the vectors expressing TK did not
proliferate for up to 6 days
after transfection in the presence of GCV
(Fig.
2C). Comparable
levels of TK expression were observed from the
pCMVTK and pCMVp27citeTK
plasmids by Western blot analysis (Fig.
2D),
indicating that the
differences in their activity are due to the
expression of p27.
In addition, because 293 cells do not show
significant bystander
killing and were nearly completely transfected
(data not shown),
this result suggested that most cells expressed
sufficient TK
to be lysed in the presence of GCV. Taken together, these
results
suggest that p27 and p27/TK expression vectors were comparably
effective in arresting cell cycle progression and that TK remained
functional in this vector. To confirm that cells which express
p27 and
TK were more viable in the presence of GCV than those
with TK alone,
293 cells were transfected with a set of p27/TK
expression plasmids and
a CD2 expression plasmid. GCV was added
1 day after transfection, and
the cells were harvested 4 days
later. The percentage of
CD2
+ cells was determined by FACS analysis. Twenty to 22%
of the strongly
positive CD2 cells transfected with pCMVTKcitep27 or
pCMVp27citeTK
were detected, compared to
12% for the cells
transfected with
pCMVTKcitep27rev or pCMVp27revciteTK. The cells
expressing the
highest levels of CD2 were also most efficiently
arrested by p27
(when CD2 expression was increased, G
1
growth arrest was higher).
These results suggested that cells arrested
by p27 survive GCV
treatment better than growing cells. Before testing
the efficacy
of combination gene transfer in bystander killing, the
growth
arrest activity of p27 was optimized further.
View larger version (32K):
[in this window]
[in a new window]
|
FIG. 2.
Coexpression of p27 and TK in 293 cells. (A) Schematic
representation of the constructs expressing p27 and TK. (B) DNA
profiles of 293 cells transfected with the respective plasmids
illustrated in panel A and a CD2-expressing plasmid. All of the
CD2-expressing cells were included in the analysis. The fraction of
cells in G1 phase of the cell cycle is indicated above the
corresponding peak. The fractions of CD2-positive cells for each
transfection (from left to right) were 65, 61, 68 and 64%,
respectively. (C) Growth of 293 cells transfected with the respective
plasmids illustrated in panel A in the presence or in the absence of 5 µM GCV. Proliferation was measured by using a colorimetric assay.
Data represent the average of three measurements.
OD570-650, optical density at 570 to 650 nm. (D)
Comparable expression of TK in different pCMVTK (lanes 1 and 2) and
pCMVp27citeTK (lanes 3 and 4) plasmid expression vectors. Western blot
analysis was performed by standard methods (40) with a
polyclonal rabbit antiserum to HSV TK, kindly provided by William
Summers (Yale University).
|
|
Optimization of G1/S growth arrest by p27 expression
vectors.
The structure of p27 includes an NH2-terminal
region similar to p21 and contains a CDK binding region (amino acids 28 to 79) (30, 41). Expression of this domain of p27 is
sufficient to inactivate cyclin A- or E-Cdk2 complexes and cause
G1/S growth arrest. The function of the COOH-terminal
region of p27 has not been established, although it binds to E1A and
contains a putative nuclear localization signal (amino acids 153 to
169) and a consensus Cdc2 phosphorylation site (amino acid T187).
To increase p27 activity, the consensus Cdc2 phosphorylation site TPKK
was mutated to AAGG in pCMVp27citeTK, giving rise to
pCMVp27
cdcciteTK. Indeed, it was recently shown that cyclin E-Cdk2
phosphorylates p27 on T187, promoting the degradation of p27 and
subsequent transit from G
1 to S phase (
38).
Expression of this
vector resulted in more cells arrested in
G
1 than expression of
wild-type p27 did (Fig.
3). An alternative p27 mutant was made
by
fusion of the COOH terminus of p27 to the NH
2 terminus of
TK
(pCMVp27TK). This fusion gene product provided the advantage of
a
single open reading frame which would allow its combination
with a
third gene that might further enhance the action of p27
and TK.
Transfection of 293 cells with the p27/TK fusion protein
plasmid showed
that it was comparably active to p27 in the bicistronic
vector (Fig.
4). In the absence of GCV, cells
transfected with
pCMVp27TK, pCMVp27citeTK, or pCMVp27 had similar
growth curves.
In the presence of GCV, cell proliferation was strongly
inhibited
in cells transfected with pCMVp27citeTK and pCMVp27revcite TK
and to a lesser extent with pCMVp27TK. Results obtained with a
colorimetric proliferation assay were comparable to those observed
by
determination of cell numbers (Fig.
4B and C).
View larger version (19K):
[in this window]
[in a new window]
|
FIG. 3.
Effect of Cdc2 consensus site mutation on the activity
of p27. DNA profiles of 293 cells transfected with plasmid
pCMVp27revciteTK (A), pCMVp27citeTK (B), or pCMVp27cdcciteTK (C)
together with pCMV-CD2. All of the CD2-expressing cells were included
in the analysis. The percentages of these transfected cells in the
entire cell population were 82, 81 and 74%, respectively. The fraction
of cells in each phase of the cell cycle is indicated above the
corresponding peak.
|
|
View larger version (13K):
[in this window]
[in a new window]
|
FIG. 4.
A p27-TK fusion protein which retains both activities.
293 cells were transfected with plasmids pCMVp27TK (fusion) ( ),
pCMVp27citeTK ( ), pCMVp27revciteTK ( ), pCMVp27 ( ), and
pCMVp27rev ( ). One day after transfection, cells were seeded in a
96-well plate and cultured in the absence (A) or presence (B and C) of
5 µM GCV. Cell proliferation was measured by use of a colorimetric
assay (A and B) or by counting viable cells (trypan blue staining) (C).
OD570-650, optical density at 570 to 650 nm.
|
|
Since the activity of p27 is regulated by protein degradation via the
ubiquitin-proteasome pathway (
28), we reasoned that
deletion
of the recognition sequence for the ubiquitination apparatus,
similar
to mutations of the destruction boxes of cyclin B (
18),
would increase p27 stability and activity. Internal deletions
were made
in the COOH-terminal region of p27, leaving the
NH
2-terminal
region which binds to CDK intact. Analysis of
these mutants revealed
that they induced G
1/S growth arrest
more effectively than wild-type
p27 did (Fig.
5). Eighty-two to 85% of cells were
found in G
1 when the sequences between
NarI and
FspI sites or
SacII and
FspI
sites
were deleted, in comparison to 73% for the complete protein
or 77%
with a protein with the COOH-terminal region completely
deleted. Thus,
the activity of p27 could be increased by deletion
of regions
implicated in the degradation of this protein and was
also retained in
the p27/TK fusion protein which retained both
activities at levels
comparable to each individual wild-type protein.
View larger version (22K):
[in this window]
[in a new window]
|
FIG. 5.
Effect of deletions in the p27 coding region on p27
activity. 293 cells were transfected with plasmids pCMVp27TK (A),
pCMVp27NFTK (B), pCMVp27SFTK (C), pCMVp27SNTK (D), and pCMVTK (E),
together with pCMV-CD2. The transfection efficiencies were 65, 71, 68, 68, and 67% CD2+ cells, respectively. The histograms
represent the distribution of the CD2+ cells throughout the
cell cycle 2 days after transfection.
|
|
Combination p27 and TK gene transfer enhances bystander cell
killing.
To analyze its efficacy in prodrug-mediated killing, we
transfected Renca cells, a murine epithelial carcinoma cell line, with
pCMVp27citeTK, pCMVp27TK, and pCMVp27-revciteTK, together with
pCMV-CD2, which served as a control reporter to standardize transfection efficiency. One day after transfection, transfection efficiencies were determined, and cells were harvested and mixed with
increasing numbers of nontransfected cells. GCV was added to the tissue
culture medium, and cell proliferation was analyzed 4 days later.
GCV-induced killing was undetectable in cells cocultured with ~8%
pCMVp27revciteTK-transfected cells, probably reflecting the replacement
of dead cells by the fast-growing untransfected cells. In contrast,
cell death was readily observed at these ratios for the other
expression vectors, with the pCMVp27TK fusion plasmid being more potent
than pCMVp27citeTK (Fig. 6). Since a
minority of cells were transfected but each vector had a similar TK
activity in 293 cells (Fig. 4), it is likely that this difference is
due to the bystander effect previously documented to occur in Renca cells (25).
View larger version (22K):
[in this window]
[in a new window]
|
FIG. 6.
Coexpression of p27 with TK enhances cell killing in a
murine sarcoma cell line. Renca cells were transfected with the
plasmids indicated above the graphs, together with pCMV-CD2. The
transfection efficiencies (left to right) were 8.0, 8.2, and 7.7%
CD2+ cells, respectively. One day after transfection, the
cells were mixed with untransduced cells at different ratios, 5 µM
GCV was added, and cell proliferation was measured 4 days later. The
data represent the average and standard deviation of three
measurements. OD570-650, optical density at 570 to 650 nm.
 ,  GCV; , +GCV.
|
|
| |
DISCUSSION |
We have investigated whether expression of a CKI in combination
with TK can increase the efficacy of prodrug-mediated bystander cell
killing. We hypothesized that growth arrest would render cells
resistant to TK/GCV, allowing for more sustained prodrug conversion and
diffusion to neighboring cells. This approach is well suited for
diseases of cell proliferation, including restenosis, hyperplasia, and
localized malignancies. We show that coordinate expression of p27 and
TK increases cell killing and confers a more potent bystander effect.
Among the inhibitors of cell cycle progression that were tested,
including p21, p27, p16, and Vpr, p27 showed the most effective growth
arrest, despite the fact that 293 cells express the viral E1A
oncoprotein (15) shown previously to inactivate p27
(20). E1A was shown to bind more strongly to the COOH- than
the NH2-terminal region of p27 (20). Here, the
activity of pCMVp27N, which expresses only the N-terminal part of p27,
was not higher than that of pCMVp27, which expresses the full-size
protein (data not shown), suggesting that expression of p27 in this
highly transfectable line was sufficient to saturate limited amounts of
constitutively expressed E1A. A fusion protein between p27 and TK
retained both growth arrest and cytotoxic activities at levels
comparable to those of the individual proteins. This fusion protein
retains a nuclear localization sequence and suggests that prodrug
conversion is likely equally effective whether in the nucleus or
cytoplasm, where it has been localized previously (16).
As shown previously, we have found that the 65-amino-acid
NH2-terminal region of p27 is necessary and sufficient for
binding to cyclin-CDK complexes (19, 30). In addition, our
results show that the addition of sequence downstream of p27 does not interfere with cell cycle inhibition. These findings are consistent with the nonglobular, extended structure of the
NH2-terminal region of p27 proposed by Russo et al.
(34). In 293 cells, pCMVp27citeTK and pCMVp27revciteTK
showed comparable TK activities (Fig. 2 and 4). Cell death was not
observed upon transfection of Renca cells with pCMVp27revciteTK, in
contrast to pCMVp27citeTK and pCMVp27TK (Fig. 6), likely because the
transfection efficiency in this cell line is lower and lysis occurs
largely through the bystander effect (25). This finding is
therefore consistent with the expectation that coexpression of p27 with
TK would sustain TK activity and increase its potency and subsequent
bystander effect.
The p27/TK fusion protein offered an alternative approach by which to
combine expression of these gene products and could be used in
combination with a third gene product. For example, the p15 CKI would
be expected to cooperate with p27 to induce cell cycle arrest
(32). Alternatively, cytosine deaminase could be used as an
independent cytotoxic prodrug which could complement the activity of
TK/GCV (33). Synergy has also been demonstrated between IL-2
and TK in tumor models (1, 6). The possibility of enhancing
bystander cell killing at the same time that cell proliferation is
inhibited suggests that this approach may also be applicable to
diseases of cell proliferation, such as restenosis, or to localized
malignancies such as head and neck carcinoma or sarcomas, where
complete local resection is often not successful.
| |
ACKNOWLEDGMENTS |
We thank Anne Mraunac, Donna Gschwend, and Nancy Barrett for
manuscript preparation.
This work was supported in part by a grant from the National Institutes
of Health (HL53466). X. Danthinne was supported by a fellowship from
the D. Collen Foundation. E.G.N. is an Established Investigator of the
American Heart Association.
| |
FOOTNOTES |
*
Corresponding author. Mailing address: Department of
Internal Medicine, University of Michigan Medical Center, 1150 W. Medical Center Dr., 7220 MSRB III, Ann Arbor, MI 48109-0644. Phone:
(734) 763-5103. Fax: (734) 763-4851. E-mail: enabel{at}umich.edu.
Present address: VA Medical Center, Research Service 151, Boise, ID
83702.
| |
REFERENCES |
| 1.
|
Benedetti, S.,
F. Dimeco,
B. Pollo,
N. Cirenei,
B. M. Colombo,
M. G. Bruzzone,
E. Cattaneo,
A. Vescovi,
S. Didonato,
M. P. Colombo, and G. Finocchiaro.
1997.
Limited efficacy of the HSV-TK/GCV system for gene therapy of malignant gliomas and perspectives for the combined transduction of the interleukin-4 gene.
Hum. Gene Ther.
8:1345-1353[Medline].
|
| 2.
|
Bi, W. L.,
L. M. Parysek,
R. Warnick, and P. J. Stambrook.
1993.
In vitro evidence that metabolic cooperation is responsible for the bystander effect observed with HSV tk retroviral gene therapy.
Hum. Gene Ther.
4:725-731[Medline].
|
| 3.
|
Boehme, R. E.
1984.
Phosphorylation of the antiviral precursor 9-(1,3-dihydroxy-2-propoxymethyl)guanine monophosphate by guanylate kinase isoenzymes.
J. Biol. Chem.
259:12346-12349[Abstract/Free Full Text].
|
| 4.
|
Borrelli, E.,
R. Heyman,
M. Hsi, and R. M. Evans.
1988.
Targeting of an inducible toxic phenotype in animal cells.
Proc. Natl. Acad. Sci. USA
85:7572-7576[Abstract/Free Full Text].
|
| 5.
|
Breakefield, X. O., and N. A. DeLuca.
1991.
Herpes simplex virus for gene delivery to neurons.
New Biol.
3:203-218[Medline].
|
| 6.
|
Chen, S. H.,
K. Kosai,
B. Xu,
K. Pham-Nguyen,
C. Contant,
M. J. Finegold, and S. L. Woo.
1996.
Combination suicide and cytokine gene therapy for hepatic metastases of colon carcinoma: sustained antitumor immunity prolongs animal survival.
Cancer Res.
56:3758-3762[Abstract/Free Full Text].
|
| 7.
|
Colombo, B. M.,
S. Benedetti,
S. Ottolenghi,
M. Mora,
B. Pollo,
G. Poli, and G. Finocchiaro.
1995.
The "bystander effect": association of U-87 cell death with ganciclovir-mediated apoptosis of nearby cells and lack of effect in athymic mice.
Hum. Gene Ther.
6:763-772[Medline].
|
| 8.
|
Culver, K. W.,
Z. Ram,
S. Wallbridge,
H. Ishii,
E. H. Oldfield, and R. M. Blaese.
1992.
In vivo gene transfer with retroviral vector-producer cells for treatment of experimental brain tumors.
Science
256:1550-1552[Abstract/Free Full Text].
|
| 9.
|
Di Marzio, P.,
S. Choe,
M. Ebright,
R. Knoblauch, and N. R. Landau.
1995.
Mutational analysis of cell cycle arrest, nuclear localization, and virion packaging of human immunodeficiency virus type 1 Vpr.
J. Virol.
69:7909-7916[Abstract].
|
| 10.
|
Ezzeddine, Z. D.,
R. L. Martuza,
D. Platika,
M. P. Short,
A. Malick,
B. Choi, and X. O. Breakefield.
1991.
Selective killing of glioma cells in culture and in vivo by retrovirus transfer of the herpes simplex virus thymidine kinase gene.
New Biol.
3:608-614[Medline].
|
| 11.
|
Fick, J.,
F. G. Barker,
P. Dazin,
E. M. Westphale,
E. C. Beyer, and M. A. Israel.
1995.
The extent of heterocellular communication mediated by gap junctions is predictive of bystander tumor cytotoxicity in vitro.
Proc. Natl. Acad. Sci. USA
92:1071-11075.
|
| 12.
|
Freeman, S. M.,
C. N. Abboud,
K. A. Whartenby,
C. H. Packman,
D. S. Koeplin,
F. L. Moolten, and G. N. Abraham.
1993.
The "bystander effect": tumor regression when a fraction of the tumor mass is genetically modified.
Cancer Res.
53:5274-5283[Abstract/Free Full Text].
|
| 13.
|
Freeman, S. M.,
K. A. Whartenby,
J. L. Freeman,
C. N. Abboud, and A. J. Marrogi.
1996.
In situ use of suicide genes for cancer therapy.
Semin. Oncol.
23:31-45[Medline].
|
| 14.
|
Gordon, J. W.,
G. A. Scangos,
D. J. Plotkin,
J. A. Barbosa, and F. H. Ruddle.
1980.
Genetic transformation of mouse embryos by microinjection of purified DNA.
Proc. Natl. Acad. Sci. USA
77:7380-7384[Abstract/Free Full Text].
|
| 15.
|
Graham, F. L.,
J. Smiley,
W. C. Russel, and R. Nairn.
1977.
Characteristics of a human cell line transformed by DNA from human adenovirus type 5.
J. Gen. Virol.
36:59-74[Abstract/Free Full Text].
|
| 16.
|
Haar, H., and T. Flatmark.
1987.
Evidence that deletion of coding sequences in the 5' end of the thymidine kinase gene of herpes simplex virus type 1 affects the stability of the gene products.
J. Gen. Virol.
68:2817-2829[Abstract/Free Full Text].
|
| 17.
|
Heyman, R. A.,
E. Borrelli,
J. Lesley,
D. Anderson,
D. D. Richman,
S. M. Baird,
R. Hyman, and R. M. Evans.
1989.
Thymidine kinase obliteration: creation of transgenic mice with controlled immune deficiency.
Proc. Natl. Acad. Sci. USA
86:2698-2702[Abstract/Free Full Text].
|
| 18.
|
Luca, F. C.,
E. K. Shibuya,
C. E. Dohrmann, and J. V. Ruderman.
1991.
Both cyclin A60 and B97 are stable and arrest cells in M-phase, but only cyclin B97 turns on cyclin destruction.
EMBO J.
10:4311-4320[Medline].
|
| 19.
|
Luo, Y.,
J. Hurwitz, and J. Massagué.
1995.
Cell-cycle inhibition by independent CDK and PCNA binding domains in p21Cip1.
Nature
375:159-161[Medline].
|
| 20.
|
Mal, A.,
R. Y. C. Poon,
P. H. Howe,
H. Toyoshima,
T. Hunter, and M. L. Harter.
1996.
Inactivation of p27Kip1 by the viral E1A oncoprotein in TGF -treated cells.
Nature
380:262-265[Medline].
|
| 21.
|
Moolten, F. L.,
J. M. Wells,
R. A. Heyman, and R. M. Evans.
1990.
Lymphoma regression induced by ganciclovir in mice bearing a herpes thymidine kinase transgene.
Hum. Gene Ther.
1:125-134[Medline].
|
| 22.
|
Moolten, F. L., and J. M. Wells.
1990.
Curability of tumors bearing herpes thymidine kinase genes transferred by retroviral vectors.
J. Natl. Cancer Inst.
82:297-300[Abstract/Free Full Text].
|
| 23.
|
Mosmann, T.
1983.
Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays.
J. Immunol. Methods
65:55-63[Medline].
|
| 24.
|
Muller, D. W. M.,
D. Gordon,
H. San,
Z.-Y. Yang,
V. J. Pompili,
G. J. Nabel, and E. G. Nabel.
1994.
Catheter-mediated pulmonary vascular gene transfer and expression.
Circ. Res.
75:1039-1049[Abstract/Free Full Text].
|
| 25.
|
Ohno, T.,
Z.-Y. Yang,
L. Xu,
M. Jaffe,
E. G. Nabel,
D. Normolle, and G. J. Nabel.
1997.
Combination gene transfer to potentiate tumor regression.
Gene Ther.
4:361-366[Medline].
|
| 26.
|
O'Malley, B. W.,
K. A. Cope,
S. H. Chen,
D. Li,
M. R. Schwartz, and S. L. Woo.
1996.
Combination gene therapy for oral cancer in a murine model.
Cancer Res.
56:1737-1741[Abstract/Free Full Text].
|
| 27.
|
O'Malley, B. W., Jr.,
D. A. Sewell,
D. Li,
K. Kosai,
S. H. Chen,
S. L. Woo, and L. Duan.
1997.
The role of interleukin-2 in combination adenovirus gene therapy for head and neck cancer.
Mol. Endocrinol.
11:667-673[Abstract/Free Full Text].
|
| 28.
|
Pagano, M.,
S. W. Tam,
A. M. Theodoras,
P. Beer-Romero,
G. Del Sal,
V. Chau,
P. R. Yew,
G. F. Draetta, and M. Rolfe.
1995.
Role of the ubiquitin-proteasome pathway in regulating abundance of the cyclin-dependent kinase inhibitor p27.
Science
269:682-685[Abstract/Free Full Text].
|
| 29.
|
Plautz, G.,
E. G. Nabel, and G. J. Nabel.
1991.
Selective elimination of recombinant genes in vivo with a suicide retroviral vector.
New Biol.
3:709-715[Medline].
|
| 30.
|
Polyak, K.,
M.-H. Lee,
H. Erdjument-Bromage,
A. Koff,
J. M. Roberts,
P. Tempst, and J. Massagué.
1994.
Cloning of p27Kip1, a cyclin-dependent kinase inhibitor and a potential mediator of extracellular antimitogenic signals.
Cell
78:59-66[Medline].
|
| 31.
|
Ram, Z.,
S. Walbridge,
J. D. Heiss,
K. W. Culver,
R. M. Blaese, and E. H. Oldfield.
1994.
In vivo transfer of the human interleukin-2 gene: negative tumoricidal results in experimental brain tumors.
J. Neurosurg.
80:535-540[Medline].
|
| 32.
|
Reynisdottir, I.,
K. Polyak,
A. Iavarone, and J. Massague.
1995.
Kip/Cip and Ink4 Cdk inhibitors cooperate to induce cell cycle arrest in response to TGF-beta.
Genes Dev.
9:1831-1845[Abstract/Free Full Text].
|
| 33.
|
Rogulski, K. R.,
J. H. Kim,
S. H. Kim, and S. O. Freytag.
1997.
Glioma cells transduced with an Escherichia coli CD/HSV-1 TK fusion gene exhibit enhanced metabolic suicide and radiosensitivity.
Hum. Gene Ther.
8:73-85[Medline].
|
| 34.
|
Russo, A. A.,
P. D. Jeffrey,
A. K. Patten,
J. Massagué, and N. P. Pavletich.
1996.
Crystal structure of the p27Kip1 cyclin-dependent-kinase inhibitor bound to the cyclin A-CDK2 complex.
Nature
382:325-331[Medline].
|
| 35.
|
Samejima, Y., and D. Meruelo.
1995.
`Bystander killing' induces apoptosis and is inhibited by forskolin.
Gene Ther.
2:50-58[Medline].
|
| 36.
|
Schmid, I.,
C. H. Uitteenbogaart, and J. V. Giorgi.
1991.
A gentle fixation and permeabilization method for combined cell surface and intracellular staining with improved precision in DNA quantification.
Cytometry
12:279-285[Medline].
|
| 37.
|
Serrano, M.,
G. J. Hannon, and D. Beach.
1993.
A new regulatory motif in cell-cycle control causing specific inhibition of cyclin D/CDK4.
Nature
366:704-707[Medline].
|
| 38.
|
Sheaff, R. J.,
M. Groudine,
M. Gordon,
J. Roberts, and B. E. Clurman.
1997.
Cyclin E-CDK2 is a regulator of p27-Kip1.
Genes Dev.
11:1464-1478[Abstract/Free Full Text].
|
| 39.
|
St. Clair, M. H.,
C. U. Lambe, and P. A. Furman.
1987.
Inhibition by ganciclovir of cell growth and DNA synthesis of cells biochemically transformed with herpesvirus genetic information.
Antimicrob. Agents Chemother.
31:844-849[Abstract/Free Full Text].
|
| 40.
|
Tanner, F. C.,
Z.-Y. Yang,
D. Gordon,
G. J. Nabel, and E. G. Nabel.
1998.
Expression of cyclin-dependent kinase inhibitors in cardiovascular disease.
Circ. Res.
82:396-403[Abstract/Free Full Text].
|
| 41.
|
Toyoshima, H., and T. Hunter.
1994.
p27, a novel inhibitor of G1 cyclin-cdk protein kinase activity, is related to p21.
Cell
78:67-74[Medline].
|
| 42.
|
Yang, Z.-Y.,
N. D. Perkins,
T. Ohno,
E. G. Nabel, and G. J. Nabel.
1995.
The p21 cyclin-dependent kinase inhibitor suppresses tumorigenicity in vivo.
Nat. Med.
1:1052-1056[Medline].
|
Journal of Virology, November 1998, p. 9201-9207, Vol. 72, No. 11
0022-538X/98/$04.00+0
Copyright © 1998, American Society for Microbiology. All rights reserved.
This article has been cited by other articles:
-
Ganesh, L., Yoshimoto, T., Moorthy, N. C., Akahata, W., Boehm, M., Nabel, E. G., Nabel, G. J.
(2006). Protein Methyltransferase 2 Inhibits NF-{kappa}B Function and Promotes Apoptosis. Mol. Cell. Biol.
26: 3864-3874
[Abstract]
[Full Text]
Top
Abstract
Introduction
