Carcinogenesis vol.25 no.3 pp.333--341, 2004
DOI: 10.1093/carcin/bgh014
Down-regulation of c-myc and Cyclin D1 genes by antisense oligodeoxy nucleotides
inhibits the expression of E2F1 and in vitro growth of HepG2 and Morris 5123
liver cancer cells
Maria M.Simile, Maria R.De Miglio, Maria R.Muroni,
Maddalena Frau, Giuseppina Asara, Silvia Serra, Maria
D.Muntoni, Maria A.Seddaiu, Lucia Daino, Francesco
Feo1 and Rosa M.Pascale
Department of Biomedical Sciences, Division of Experimental Pathology and
Oncology, University of Sassari, I-07100 Sassari, Italy
1
To whom correspondence should be addressed
Email: [email protected]
A number of genetic interactions are involved in the control of cell cycle, but their role and nature have not been
completely clarified. The knowledge of the behavior of
these interactions in hepatocellular carcinoma, could optimize preventive and therapeutic strategies based on cell cycle
restraint. We studied downstream events following c-MYC
and CYCLIN D1 gene inhibition, by lipoplex-delivered
MYC and CYCLIN D1 antisense oligodeoxy nucleotides
(aODNM , aODND1 ), in in vitro cultured human HepG2
and rat Morris 5123 hepatoma cells. 0.5--20 mM aODNM
and aODND1 inhibited in vitro growth of both cell types.
Scramble oligomer (SCR) and sense ODNs had no or relatively poor effect. Ten micromolar aODNM and aODND1 ,
but not SCR, also induced a significant increase in the
apoptotic index of HepG2 and 5123 cells, and inhibited
colony formation in soft agar by HepG2 cells. Treatment
of the cells with aODNM plus aODND1 had no additive
effect on growth and apoptosis. aODNM and aODND1
induced 450% decrease in c-MYC and CYCLIN D1 gene
expression, respectively, at both mRNA and protein level.
The inhibition of gene expression by aODNs was highly
specific, and SCR was without effect. The reduction in
c-MYC and CYCLIN D1 expression by aODNs, was associated with a 450% decrease in E2F1 mRNA and protein
production, without changes in CYCLIN A and CYCLIN E
expression. These results suggest the involvement of both
c-MYC and CYCLIN D1 on E2F1 gene function, and indicate that aODNM and aODND1 may inhibit hepatoma cell
growth through down-regulation of the E2F1 gene. The
inhibition of E2F1 gene expression by E2F1 aODN, was
associated with strong growth restraint of HepG2 cells.
Thus, interactions of c-MYC and CYCLIN D1 with E2F1
gene are essential for cell cycle activity in hepatoma cells,
and their inhibition may have a therapeutic effect.
Introduction
The cell cycle of mammalian cells is regulated through the
serial activation of cyclins and cyclin-dependent kinases
Abbreviations: aODN, antisense oligodeoxy nucleotide; aODND1 , Cyclin D1
aODN; aODNE , E2F1 aODN; aODNM , c-MYC aODN; CDK, cyclindependent kinase; D, DOTAP; FCS, fetal calf serum; HCC, hepatocellular
carcinoma; SCR, scramble oligomer; sODN, sense oligodeoxy nucleotide.
Carcinogenesis vol.25 no.3 # Oxford University Press; all rights reserved.
(CDKs). Cyclins D family, in complexes with CDK4/6, and
Cyclin E and Cyclin A, in complexes with CDK2, participate
in the phosphorylation of the pocket proteins, pRb, p107 and
p130, resulting in inactivation of their growth suppressor activities and increase in the amounts of E2F proteins that form
complexes with DP1, and activate DNA synthesis genes (1,2).
Deregulation of pRb--E2F pathway has been associated with
tumorigenesis in several animal and human tissues (3,4). Liver
pre-neoplastic and neoplastic lesions, in c-myc and c-myc/Tgfa
transgenic mice (5) exhibit over-expression of cyclin D1,
increase in Cyclin D1--Cdk4 complexes, pRb phosphorylation,
increase in E2f1--DP1 and E2f2--DP1 heterodimers and transcriptional induction of E2f target genes. In the rat model of
hepatocarcinogenesis, amplification and over-expression of
c-myc, and over-expression of cyclins D1, E and A, and E2f1
genes, in neoplastic nodules and hepatocellular carcinomas
(HCCs), is associated with rise in Cyclin D1--Cdk4, Cyclin E-Cdk2, Cyclin A--Cdk2 and E2F1--DP1 complexes, and pRb
hyper-phosphorylation (6--8). Interestingly, slowly growing
lesions, induced in genetically resistant Wistar and BN rats,
show low or no increases in c-myc, cyclin D1, cyclin E, cyclin A
and E2f1 expression and complex formation, associated with
p16INK4A over-expression and pRb hypo-phosphorylation.
Since the liver lesions of resistant rats are not prone to fast
growth, these observations link the activity of cell cycle
genes to hepatocarcinogenesis progression. Over-expression
of c-MYC, CYCLIN D1, CYCLIN E and CYCLIN A genes,
has also been found in human HCCs, and seems to have a
prognostic value (9--11).
Although the biochemical events controlling the activity of
cell cycle are well known (1,2), the nature and role of regulatory interactions between cell cycle genes are still largely
unclear. c-Myc has been shown to induce or repress cyclin D1
gene (12--14), and to regulate Cyclin D1 post-translationally
(15). Researches on transgenic mice have shown a cooperation
between E2F1 and c-Myc during mouse liver carcinogenesis
(16), but it has been suggested that these nuclear factors
control liver cell proliferation and carcinogenesis through
different mechanisms (17). E2F1 promoter contains c-MYC
responsive elements (18), and D-type cyclin-dependent kinase
activity specifically activates E2F1 promoter (19). The overexpression of c-MYC and CYCLIN D1 genes plays a central
role in the deregulation of cell cycle in HCC of rodents and
humans (6--11,20). The knowledge of the effects of the interactions of these genes with other cell cycle genes, may provide
new information to optimize preventive and therapeutic
strategies based on cell cycle restraint. To address this question, we studied downstream events induced by the inhibition
of c-MYC and CYCLIN D1 gene activity, in in vitro cultured
human HepG2 and rat Morris 5123 liver cancer cells. In these
experiments, c-MYC and CYCLIN D1 gene expression was
inhibited by lipoplex-delivered MYC and CYCLIN D1 antisense oligodeoxy nucleotides (aODNM , aODND1 ). aODNs
specifically inhibit gene expression (21,22) probably by RNA
333
M.M.Simile et al.
targeting followed by degradation via ribonuclease (23). This
work demonstrates that a regulation of E2F1 gene by both
c-MYC and CYCLIN D1 gene activities is essential for cell
cycle activity and growth of liver cancer cells, and the inhibition of these genetic interactions may have a therapeutic effect.
Materials and methods
Cell culture
Human HepG2 and rat Morris 5123 hepatocellular carcinoma cells were
maintained in Hams-F12 medium, supplemented with 10% fetal calf serum
(FCS) and antibiotics, at 37 C, in an atmosphere of 5 (HepG2) or 10% (Morris
5123) CO2 in air and constant humidity. The medium was changed every 48 h
and the cells were subcultured when they became confluent. After seeding, the
cultures were incubated 12 h at 37 C, synchronized for 48 h in Hams-F12
supplemented with 5% FCS, and then returned to complete medium before
adding oligodeoxy nucleotides.
Antisense and sense oligodeoxy nucleotides
Due to the homology between human and rat 50 end sequences of both c-MYC
and CYCLIN D1 genes, phosphorothioate nucleotides including the translation
initiation site of human CYCLIN D1 (28mer) and a 21mer sequence close to
this site of human c-MYC, were used with both cell types. The phosphorothioate oligonucleotide sequences used were: c-MYC antisense 50 -GTTAGCGAAGCTCACGTTGAG-30 , sense 50 -CTCAACGTGAGCTTCGCTAAC-30
(nucleotides 4169--4189); CYCLIN D1 antisense 50 -CTTCGCAGCACAGGAGCTGGTGTTCCAT-30 , sense 50 -ATGGAACACCAGCTCCTGTGCTGCGAAG-30 (nucleotides 241--268). The first 6 nt of the translation initiation
site of c-MYC gene were omitted to exclude the tetramer showing antiproliferative effects independent of its action as aODN (24). The MYC and
CYCLIN D1 oligomers used did not have any nucleotide sequence homologous to
the human and rat E2F1 gene sequences. The phospohorothioate oligonucleotide sequences of human E2F1 were: antisense 50 -TAGATCCGATCCAGCTCAGTGACA, sense 50 -TGTCACTGAGCTGGATCGGATCTA (nucleotides
355--379). The `scramble' oligomer (SCR) was: 50 -CAGGTCTTTCATCTAGAACGATGCGGG. Oligodeoxy nucleotides (Roche Diagnostic S.p.A.,
Monza, Italy), were purified by HPLC (25), evaporated to dryness, resuspended in sterile water and quantified spectrophotometrically. To prepare the
oligomer/N-[1-(2,3-dioleoyloxy)propyl]-N,N,N-trimethylammonium methylsulphate (DOTAP; Roche Diagnostic S.p.A.) lipoplex, aODN, sense oligodeoxy nucleotides (sODNs) or SCR solutions (1 mmol/ml) were added to equal
volumes of a solution containing 1 mg of DOTAP/ml of HEPES buffer
(pH 6.2), and gently mixed for 15 min at 37 C. Aliquots of the mixture were
added to synchronized cell cultures to obtain a final concentration of 10 mM
oligomers, unless otherwise stated. This was followed, after 24 h, by half
doses of nucleotides, and by a full dose in correspondence of each change of
medium.
In vitro proliferation and apoptosis assays
The cells were seeded at 50 000/cm2 in Hams-F12/FCS medium, and treated
with nucleotides as above. [3 H]Thymidine (1 mCi) was added to four replicates 18 h before detaching the cells with 0.06% trypsin. The cells were
harvested, at the times indicated, washed 4-fold in PBS and the radioactivity
(d.p.m.) was evaluated by using a liquid scintillation counter.
To determine the apoptotic index, the cells were seeded at 20 000/cm2 in
Hams-F12/FCS medium. After treating with nucleotides, the medium
was removed and the cells were washed with PBS, fixed with buffered
p-formaldehyde (pH 7.4) and stained with propidium iodide. Nuclear changes
representing apoptosis (chromatin condensation, margination and fragmentation), revealed by propidium iodide, were determined by scoring 5000 cells
and data were expressed as percentage of total hepatocytes (apoptotic index).
Cell culture in soft agar
For soft agar cultures, HepG2 cells were synchronized in low serum medium
and then grown for 24 h in the presence of 10 mM aODNs or SCR. Then the
medium was removed and the cells (25 000/ml) were suspended in Hams-F12
plus 10% FCS, containing 0.33% Noble agar (DIFCO Laboratories, Becton
Dickinson Italia, Milano, Italy). Cell suspensions were plated in triplicate in
six-well plates on top of 0.5% agar base. Twenty-four hours after plating, 1 ml
of medium/FCS, containing a half concentration of SCR or aODNs, was spread
out on the surface of the cultures. This addition was repeated every 24 h (with
full doses every 3 days). The number and size of colonies were evaluated
15 days after plating by computer assisted phase contrast microscopy.
334
Table I. Primers used for RT--PCR experiments
Gene
Human
c-MYC
Primers
Fw
Rev
Cyclin D1 Fw
Rev
CYCLIN E Fw
Rev
CYCLIN A Fw
Rev
E2F1
Fw
Rev
GAPDH
Fw
Rev
Rat
c-myc
Fw
Rev
cyclin D1 Fw
Rev
cyclin E
Fw
Rev
cyclin A
Fw
Rev
E2f1
Fw
Rev
gapdh
Fw
Rev
Base
pair
50 -AGAGTCTGGATCACCTTCTGCTGGA
50 -AGCCCCTGGTGCTCCATGAGGAGA
50 -CTGTGCTGCGAAGTGGAAACCAT
50 -TTCATGGCCAGCGGGAAGACCTC
50 -GATGACCGGGTTTACCCAAACTC
50 -GTGTCGCCATATACCGGTCAAAG
50 -AGTATCTCAGCTGGGAAGGAC
50 -AGTTAGCAGCCCTAGCACTGTC
50 -GCAAGTTACCAGGCTCGGCTCATA
50 -CAGAAGAGACCTGGCTTAAGGCT
50 -TTCCATGGCACCGTCAAGGCTGAG
50 -TGGTTCACACCCATGACGAACATGG
50 -AGGAACTATGACCTCGACTACG
50 -AGTAGCTCGGTCATCATCTCCAG
50 -CTTACTTCAAGTGCGTGCAGAGG
50 -GCTTGTTCACCAGAAGCAGTTCC
50 -CTGTCAGCTGACAGTGGAGAAGG
50 -AGGGTGCTACTTGACCCACTGGA
50 -CAGTGTGAAGATGCCCTGGCTT
50 -CAAGGATGGCCCGCATACTGTTA
50 -TGTGTAAGCCAGAGCCGGAGCT
50 -GATCCGATCCAGCTCAGTGACA
50 -GTATGACTCTACCCACGGCAAGTTC
50 -AGCCTTCTCCATGGTGGTGAAGAC
340
240
250
160
250
290
260
300
190
230
250
190
Comparative RT--PCR
At the times indicated (see Results) aODNs-treated cells and their control cells
were harvested by adding to each dish 1 ml of Tripure (Roche Diagnostic
S.p.A.) and total RNA and proteins were isolated and purified. Semiquantitative RT--PCR was performed by a modification (7) of the comparative
PCR method (26). In brief, master solutions, containing 4 mg/ml of total RNA
from different tissues, were prepared. The reaction mixture consisted of
RT--PCR reaction buffer containing 1:50 of volume of enzyme mix and 500
U/ml of ribonucelase inhibitor and 5 mM dithiothreitol (Titan One Tube RTPCR System, Boehringer, Mannheim). Ten microliter aliquots of the reaction
mixture, containing the appropriate primers (20 pmol each, Table I), and
dNTPs (200 mM each, including 5 mCi of [a-32 P]dCTP) were added to a series
of three tubes per tissue, followed by 10 ml aliquots of three appropriate
RNA dilutions. Calibration of mRNA/cDNA concentration was carried out
for each tissue, with primer pair specific for GAPDH reference gene. After
correction of the amounts of RNA used for RT--PCR analysis, a second
calibration was made, if necessary, in each series of tubes, followed by
enzymatic amplification. Cycling parameters for RT--PCR analyses were:
30 min at 55 C, for the synthesis cDNA first strand, followed by 2 min at
95 C, 1 min at 55 C, 1 min at 72 C, for 30 cycles in a GeneAmp PCR system
9700 (Applied Biosystem, Applera Italia, Monza, Italy). Size determination of
PCR products was performed by electophoresis in 2% agarose gel, using a
fragment size marker. PCR products were applied to a 6% polyacrylamide gel
containing 10% glycerol and Packard Instant Imager analysis of the gels
revealed single bands for all genes tested, which were quantified on the basis
of the radioactive counts on the membranes, and were reproduced by an image
analysis software on a personal computer.
Immunoblotting
Protein extracts were opportunely diluted in a SDS-based buffer and total
protein concentration was determined according to Lowry et al. (27), using
bovine serum albumin as a standard. Diluted extracts were pre-cleared 15 min
with 50 ml/ml of Gamma Bind G Sepharose, centrifuged and incubated 30 min
at 4 C, after addition of appropriate control IgG (corresponding to the host
species of the primary antibody) and Protein G--agarose (8). After centrifugation, immunoprecipitation was performed by incubating supernatants overnight at 4 C with 4 mg of agarose-conjugated antibodies/mg of supernatant
protein (Santa Cruz Biotechnology, CA, USA). The following antibodies,
reacting with both human and rat epitopes, were used: mouse monoclonal
anti-MYC, anti-CYCLIN D1 and anti-E2F1 (sc-42, sc-246 and sc-251; Santa
Cruz Biotechnology); rabbit polyclonal anti-Cyclin E and anti-Cyclin A
Cell cycle gene interactions and antisense therapy
Fig. 1. Effect of different concentrations of antisense oligodeoxy
nucleotides against c-MYC and CYCLIN D1 on the growth of HepG2 and
Morris 56123 hepatoma cells. The cells were seeded at 50 000/cm2 ,
synchronized in low serum medium and, when indicated, treated with
0.5--20 mM aODNM , aODND1 , sODNs, SCR or DOTAP (D) alone. Six hours
after starting the treatment, [3 H]thymidine was added to the medium and the
incorporation into the cells was recorded 18 h later. Data (means SD,
n ˆ 6) represent the percentage of [3 H]thymidine incorporated with respect
to untreated controls. Multiple comparisons (TK test): HepG2, SCR (1--20
mM) and sODNs (5--20 mM) different from untreated control (at least
P 5 0.05). aODNM and aODND1 different from control at all concentrations
(at least P 5 0.001). 5123, SCR and sODND1 (5--20 mM) significantly
different from control (P 5 0.05). aODNM and aODND1 different from
control at all concentrations (P 5 0.0001).
(sc-481 and sc-751, Santa Cruz Biotechnology). Immunocomplexes were
absorbed on Gamma Bind G Sepharose beads, and washed 5-fold with
HNTG buffer (20 mM HEPES, pH 7.5, 150 mM NaCl, 0.1% Triton X-100
and 10% glycerol). The pellets were re-suspended in sample buffer (100 mM
Tris--Cl, pH 6.8, 5% SDS, 5% glycerol, 0.005% bromophenol blue, and 5% 2bmercaptoethanol), boiled for 5 min and centrifuged to separate immunoprecipitated protein. For negative controls primary antibodies were incubated, prior
to immunoprecipitation, with the respective immunogen peptide (1:20 w:w),
which resulted in inhibition of the immunoprecipitation. Aliquots of immunoprecipitated samples were separated by 10% SDS--polyacrylamide gel, and
transferred onto PVDF membranes (Millipore S.p.A., Vimodrone, Milano,
Italy). The membranes were washed for 10 min with 20% H2 O2 in 20%
methanol/TBS (10 mM Tris--Cl, pH 7.2, 150 mM NaCl), and then in TBS
alone. Non-specific binding was blocked by incubating 40 min at room temperature with Blocking Reagent (1:5 dilution in TBS; Boehringer, Mannheim,
Roche Diagnostics S.p.A., Monza, Italy). The membranes were washed 4-fold
in TBS/0.05% Tween 20 (TBS/ T), and then incubated 1 h, at 23 C with
biotinylated secondary antibody (Vector Laboratories, D.B.A. Segrate,
Milano, Italy) diluted 1:1000. After incubation with Streptavidin--horseradish
peroxidase conjugate and washing in TBS/ T, enhanced chemiluminescence
was used to visualize and quantify the immune complexes by Image Master
(Amersham Bioscences, Cologno Monzase, Milano, Italy). The specificity of
the reactions was tested by immunoblot analysis of control proteins with the
corresponding primary antibodies.
Statistical analysis
Data are expressed as means standard deviation (SD). The significance of
the differences between the means was evaluated by ANOVA, multiple comparisons were evaluated by the Tukey--Kramer (TK) test by using GraphPad
InStat 3 (www.graphpad.com). We selected P 5 0.05 as the minimum level of
significance.
Fig. 2. Effect of antisense oligodeoxy nucleotides against c-MYC and
CYCLIN D1 on the growth curve of HepG2 and Morris 56123 hepatoma
cells. The cells were seeded at 50 000/cm2 , synchronized in low serum
medium and, when indicated, treated with 10 mM aODNM , aODND1 ,
aODNM plus aODND1 (open circles) or SCR. [3 H]Thymidine was added to
the medium 18 h before harvesting the cells. Data are means SD, n ˆ 7.
ANOVA analysis showed significant differences among groups
(P 5 0.0001) in HepG2 and 5123 cells. Multiple comparisons (TK test):
SCR versus untreated control (C), P 5 0.05 at 72 and 96 h for HepG2
and 5123 cells. aODNM and aODND1 versus C and SCR at least P 5 0.05
at all time points for both cell types.
Results
In order to preliminarily establish the concentration of aODNs
inducing maximum inhibition of in vitro cell growth and
relatively low toxicity, HepG2 and Morris 5123 cells were
subcultured 24 h in the presence of increasing concentrations
of aODNM , aODND1 , sODNs, SCR or DOTAP alone. The data
in Figure 1 show a higher sensitivity to aODNs of 5123 cells
with respect to HepG2 cells. An ~60% decrease in labeled
thymidine incorporation occurred in 5123 cells, in the presence
of aODNM and aODND1 concentration as low as 0.5 mM,
compared with untreated control. An analogous inhibition of
HepG2 cell growth was only observed with 10 mM aODNs.
The decrease in [3 H]thymidine incorporation, induced by
0.5--20 mM sODNs and SCR, ranged between 10 and 29%, in
HepG2 cells, and between 8 and 20% in 5123 cells. A 12--20%
decrease occurred in both cell types, treated with 10 mM SCR
or sODNs, and was considered an acceptable toxic effect
allowing an accurate determination of growth inhibition by
aODNs, and all other experiments were performed with this
nucleotide concentration. DOTAP alone had no significant
effects on cell growth.
Figure 2 shows the kinetics of the inhibition of in vitro
growth of HepG2 and Morris 5123 cells by MYC and
CYCLIN D1 aODNs. aODNM induced 42--61% decreases in
the growth of HepG2 cells, between 24 and 120 h, with respect
to untreated cells. aODNM -induced decrease in 5123 cell
growth was 48--60%, between 24 and 96 h. In aODND1-treated
cultures, the inhibition was 46--67%, between 24 and 120 h, for
335
M.M.Simile et al.
Fig. 3. Effect of antisense oligodeoxy nucleotides against c-MYC and
CYCLIN D1 on colony formation by HepG2 cells in soft agar. The cells
were grown for 24 h in Ham-F12/FCS containing, when indicated, 10 mM
aODNs or SCR, and then suspended in a medium containing 0.33% agar plus
aODNs or SCR and transferred on top of a medium containing 0.5% agar.
The colonies were counted after 15 days of subculture during which aODN or
SCR re-feeding was performed every 24 h. (Upper panel) (A) Untreated
control (C); (B) SCR-treated; (C and D) treated with aODNM and aODND1 ,
respectively. Magnification 8. (Lower panel) Mean values SD, n ˆ 6, of
the number/cm2 and surface area of colonies. TK test: SCR versus C, not
significant; aODNM and aODND1 versus C and SCR, P 5 0.0001.
HepG2 cells, and 46--75%, between 24 and 96 h, for 5123 cells.
Maximum inhibition occurred 24 h after starting the treatment
with aODNM , for both HepG2 and 5123 cells, and 72 and 24 h
after adding aODND1 , for HepG2 and 5123 cells, respectively.
The data in Figure 2 also show that treatment of HepG2 and
5123 cells for 24 and 72 h with aODN anti-c-MYC together
with anti-CYCLIN D1 did not result in any additive effect.
Relatively low inhibition of growth occurred in both types
of cells throughout the duration of the experiments, as a
consequence of SCR treatment, whereas DOTAP alone was
without significant effect (not shown). Growth inhibition
by aODNs was confirmed by the determination of colony
formation by HepG2 cells in soft agar: there occurred
(Figure 3) 48 and 32% decreases in number and 67 and 73%
decreases in size of colonies formed by cells treated with
aODNM and aODND1 , respectively, compared with untreated
control. SCR induced insignificant changes of colony number
and size.
336
Fig. 4. Apoptosis in HepG2 cells treated with antisense oligodeoxy
nucleotides against c-MYC and CYCLIN D1. When indicated, the cells were
treated for 72 h, after synchronization, with 10 mM aODNM , aODND1 ,
aODNM plus aODND1 , SCR or DOTAP (D) alone. (Upper panel) HepG2
cells stained with propidium iodide showing a few apoptotic cells in cultures
treated with DOTAP (A) or SCR (B), and clusters of cells with chromatin
margination, condensation and fragmentation in the cultures treated with
aODNM (C) or aODND1 (D). Magnification 325. (Lower panel)
Quantitative analysis showing the percentage of apoptotic cells (apoptotic
index) in untreated controls (C) and in the cultures treated with DOTAP (D),
SCR and aODNS (single or in combination). Data are means SD, n ˆ 5.
TK test: D versus C, not significant; SCR versus C, P 5 0.01; aODNM;
aODND1 and aODNM plus aODND1 versus SCR and C, at least P 5 0.001.
Growth restraint of HepG2 cells was associated with an
apoptogenic effect of both c-MYC and CYCLIN D1 aODNs.
Numerous apoptotic cells could be seen in the cultures treated
for 72 h with these aODNs (Figure 4C and D), compared with
DOTAP or SCR-treated cells (Figure 4A and B). Quantitative
analysis revealed 153 and 267% increases in apoptotic index of
HepG2 cells, cultured in the presence of aODNM and aODND1 ,
respectively, compared with untreated control. No additive
effect of MYC and CYCLIN D1 aODNs was found when
HepG2 cells were subcultured in the presence of both
aODNs. Relatively poor effect of SCR and no effect of
DOTAP alone on apoptotic index were observed.
The effects of aODNs on c-MYC and CYCLIN gene expression are shown in Figure 5. Forty to forty-four percent
decreases in c-MYC mRNA were found in HepG2 and 5123
cells, respectively, subcultured for 24 h in the presence of
aODNM , compared with untreated controls. aODNM did not
induce any change in CYCLIN D1 mRNA, in both types of
cells. Treatment of HepG2 and 5123 cells with aODND1 , for
72 and 24 h, respectively, resulted in 64 and 50% decreases in
CYCLIN D1 mRNA, without changes in c-MYC mRNA levels.
No significant changes of c-MYC and CYCLIN D1 mRNA
levels were found in HepG2 and 5123 cells treated with SCR
(Figure 5), or with c-MYC and CYCLIN D1 sODNs or DOTAP
alone (not shown).
Cell cycle gene interactions and antisense therapy
Fig. 5. Effect of antisense oligodeoxy nucleotides against c-MYC and
CYCLIN D1 on the expression of c-MYC and CYCLIN D1 genes in HepG2
and Morris 5123 hepatoma cells treated for 72 and 24 h after
synchronization, respectively, with 10 mM aODNM , aODND1 or SCR.
PCR products were separated by electrophoresis into 6% polyacrylamide
gel/10% glycerol. Size of PCR products was evaluated by comparing
ethidium bromide stained RT--PCR products with fragment size marker, after
electrophoresis in 2% agarose gel. (Upper panel) Representative
reproduction by Instant Imager software of RT--PCR radiolabeled products.
(Lower panel) Quantitative Instant Imager analysis showing mean
radioactive counts SD, n ˆ 5--7, normalized to GAPDH values (relative
amounts). ANOVA analysis showed significant differences among groups
(P 5 0.0001 for c-MYC and CYCLN D1) in HepG2 and 5123 cells. Multiple
comparisons (TK test), HepG2 and 5123 cells: c-MYC, aODNM versus
untreated control (C) and SCR, at least P 5 0.01; aODND1 versus C and
SCR, not significant. CYCLIN D1, aODND1 versus C and SCR, P 5 0.0001;
aODNM versus C and SCR, not significant.
The inhibition of c-MYC and CYCLIN D1 gene expression by
specific aODNs was confirmed at protein level (Figure 6).
Treatment of HepG2 and 5123 cells for 72 and 24 h, respectively, with aODNM resulted in 54 and 62% decreases in
c-MYC protein, with respect to untreated controls, without
changes of CYCLIN D1. Fifty to fifty-five percent decreases
in CYCLIN D1 occurred in HepG2 and 5123 cells incubated
72 and 24 h, respectively, in the presence of aODND1 . No
changes in c-MYC occurred in these conditions. SCR did not
affect the production of both c-MYC and CYCLIN D1 proteins.
These results suggest that both c-MYC and CYCLIN D1
genes support the in vitro growth of liver cancer cells, even
in the apparent absence of reciprocal interactions. We thus
Fig. 6. Detection by immunoprecipitation (IP) of c-MYC, and CYCLIN D1
in homogenates from HepG2 and 5123 cells treated for 72 and 24 h after
synchronization, respectively, with 10 mM aODNM , aODND1 or SCR.
Sample proteins and blocking peptides were co-IP by antibodies against
MYC, and CYCLIN D1, separated by 10% SDS--PAGE. (Upper panel)
Representative IP analysis, with control proteins (blocking peptides) in the
last lane. (Lower panel) Chemiluminescence analysis showing mean
values SD (n ˆ 5), normalized to control proteins (arbitrary units).
ANOVA analysis showed significant differences among groups (P 5 0.0001
for c-MYC and CYCLN D1) in HepG2 and 5123 cells. Multiple comparisons
(TK test), HepG2 and 5123 cells: c-MYC, aODNM versus untreated control
(C) and SCR, P 5 0.0001; aODND1 versus C and SCR, not significant.
CYCLIN D1, aODND1 versus C and SCR, P 5 0.0001; aODNM versus C and
SCR, not significant. SCR versus C, not significant.
explored the possibility that c-MYC and CYCLIN D1 genes
contribute to regulate the expression of some key components
of cell cycle, such as the genes encoding CYCLINS A and E,
and E2F1. It appears clearly (Figure 7) that a 24 and 72 h
subculture of HepG2 cells in the presence of aODNM and
aODND1 did not significantly influence the expression of
CYCLIN A and CYCLIN E genes, whereas it induced 55--60%
decreases in E2F1 mRNA levels, with respect to untreated
cells. Analogous results were found with 5123 cells: a 24 h
treatment with aODNM or aODND1 led to 52--66% decreases in
E2f1 mRNA, without affecting Cyclin A and Cyclin E mRNAs.
No significant effects of SCR on all mRNAs were found.
Immunoprecipitation experiments with HepG2 cells grown
for 24 h in the presence of aODNM or aODND1 , confirmed
these results (Figure 8) showing an ~50% decrease in E2F1
with respect to the untreated cells. SCR had insignificant
337
M.M.Simile et al.
Fig. 7. Effect of antisense oligodeoxy nucleotides against c-MYC and CYCLIN D1 on the expression of CYCLIN A, CYCLIN E and E2F1 genes in HepG2 and
Morris 5123 hepatoma cells treated for 24 and 72 (HepG2) and 24 h (Morris 5123) after synchronization, with 10 mM c-MYC, CYCLIN D1 aODNs and SCR. PCR
products were separated by electrophoresis into 6% polyacrylamide gel/10% glycerol. Size of PCR products was evaluated by comparing ethidium bromide
stained RT--PCR products with fragment size marker, after electrophoresis in 2% agarose gel. (Upper panels) Representative reproduction by Instant Imager
software of RT--PCR radiolabeled products. Numbers on the top of lanes represent treatment times (hours). (Lower panels) Quantitative Instant Imager analysis
showing mean radioactive counts SD (n ˆ 5), normalized to GAPDH values (relative amounts). Numbers on the abscissa indicate the treatment time (hours).
ANOVA analysis showed significant differences among groups only for E2F1 (P 5 0.0001) in HepG2 and 5123 cells. Multiple comparisons (TK test), E2F1:
aODNM and aODND1 versus untreated control (C) and SCR, at least P 5 0.001 at any time point for both types of cells; C versus SCR, not significant.
effects. No effect of aODNM and aODND1 were found on
CYCLIN A and CYCLIN E levels.
These results indicate that E2F1 gene is a target of both
c-MYC and CYCLIN D1 in liver cancer cells, and suggest that
its inhibition is involved in the growth restraint induced by
MYC and CYCLIN D1 aODNs. In keeping with this, we found
that a 24 h subculture of HepG2 cells, in the presence of aODN
directed to E2F1 (aODNE ) resulted in an ~80% decrease
in [3 H]thymidine incorporation (Figure 9A). SCR induced a
26% decrease, whereas a lower (15%) decrease occurred
with sODN (not shown). Moreover, aODNE induced a 75%
decrease in E2F1 mRNA, with respect to untreated cells
(Figure 9B). No significant decrease occurred in SCR-treated
cells.
Discussion
Increasing evidence indicates that a complex and not yet
completely known regulatory network controls the cell cycle
(1,2), and different studies have envisaged the existence of a
number of genetic interactions whose effects are still the object
of debate (4,14,28--31). Previous research on the cell cycle in
HCC of rodents and humans (5--11), suggested the existence of
altered interactions among cell cycle regulatory proteins. For
instance, c-MYC amplification and/or over-expression is a
common alteration of early stages of hepatocarcinogenesis
(5--8) that may influence the expression of various cell cycle
338
genes trans-regulated by MYC protein. In this study we have
investigated, by using the antisense strategy, some molecular
interactions between cell cycle controlling genes in in vitro
growing human HepG2 and rat Morris 5123 liver cancer cells.
According to our results, both c-MYC and CYCLIN D1 genes
are involved, even in the apparent absence of reciprocal interactions, in the control of cell cycle, by affecting a common
target. Indeed, strong inhibition (450%) of either MYC or
CYCLIN D1 protein production by c-MYC or CYCLIN D1
aODNs, was associated with a marked growth restraint of
both HepG2 and 5123 cells. Since the treatment of cells
with aODNM and aODND1 specifically inhibited MYC and
CYCLIN D1 genes, and no cross-inhibition occurred, our
results seem to deny the existence, in our experimental system,
of interactions between MYC and CYCLIN D1, described previously at the level of gene activity or post-transcriptionally, in
other experimental systems (12--15). Alternatively, relatively
low c-MYC or CYCLIN D1 gene product, in liver cancer cells
treated with aODNM or aODND1 , respectively, could be sufficient for trans-activation of that of the two genes not inhibited
by the non-specific aODN. However, CYCLIN D1 transactivation by c-MYC has been denied (3,13,14), and to our
knowledge c-MYC transactivation by CYCLIN D1 has never
been demonstrated. According to our results E2F1 could represent a target of c-MYC and CYCLIN D1, whose inhibition by
MYC and CYCLIN D1 aODNs is associated with growth
restraint of liver cancer cells. MYC responsive elements are
Cell cycle gene interactions and antisense therapy
Fig. 8. Detection by IP of CYCLIN A, CYCLIN E and E2F1 in homogenates
from HepG2 cells treated for 24 h after synchronization with 10 mM c-MYC,
CYCLIN D1 aODNs or scramble oligomer (S). Sample proteins and blocking
peptides were co-immunoprecipitated by antibodies against CYCLIN A,
CYCLIN E and E2F1, and separated by 10% SDS--PAGE. (Upper panel)
Representative IP analysis, with control proteins (blocking peptides) in the
last lane. (Lower panel) Chemiluminescence analysis showing mean
values SD (n ˆ 5), normalized to control proteins (arbitrary units). ANOVA
analysis showed significant differences among groups for E2F1
(P 5 0.0001). Multiple comparisons (TK test), E2F1: aODNM and aODND1
versus untreated control (C) and SCR, P 5 0.001; C versus S, not
significant.
present in E2F1 promoter (18), and MYC transcription factor
induces the transcription of E2F1, E2F2 and E2F3 genes (32).
A cooperation between c-myc and E2f1 in inducing various
tumors, including HCC, has been described in transgenic
mouse models (16,17). Induction of E2f1 by c-myc was
shown to have a role in the onset of mammary cancer in
mice (33), and E2F activity is essential for the survival of
human cancer cell over-expressing MYC (34). Our results
also suggest a possible interaction between CYCLIN D1 and
E2F1 genes. The nature of this interaction is presently not
completely clear. E2F1 promoter can interact with D-type
CDKs (19). It is not known if this involves CYCLIN
D1--CDK complex, whose activity is regulated by CYCLIN
D1 levels in liver cancer cells (8). The presence of elements
reacting with MYC and CYCLIN D1--CDK, in E2F1 gene
promoter, and the observation that MYC and CYCLIN D1
aODNs are equally effective in inhibiting E2F1 expression,
indicate the need of a control of E2F1 expression by both cMYC and CYCLIN D1. Our results seem to exclude that aODN
effects documented in the present study depend on ODNs
toxicity. Even if we used a relatively high aODN dose (10
mM) in most experiments, growth inhibition by aODNs was
also observed with doses as low as 0.5 mM, whereas 10--20 mM
sODNs and SCR had no effect or relatively low effects on cell
growth, cell death and gene expression. The possibility of
aspecific inhibition of E2F1 mRNA by MYC and CYCLIN
D1 aODNs is ruled out by the absence of homologies between
aODNM and aODND1 and E2F1 gene sequence, that excludes
aspecific binding of aODNs to E2F1 mRNA. Furthermore, the
aODNs directed to c-MYC and CYCLIN D1 inhibited only the
correspondent gene, and did not affect CYCLIN E, CYCLIN A
Fig. 9. Effect of antisense oligodeoxy nucleotides against E2F1 on the
growth of HepG2 cells and E2F1 gene expression. (A) The cells were seeded
at 50 000/cm2 , synchronized in low serum medium and, when indicated,
treated with 10 mM E2F1 aODN (aODNE ) or SCR. Six hours after starting
the treatment, [3 H]thymidine was added to the medium and the incorporation
into the cells was recorded 18 h later. Data are means SD, n ˆ 5. TK test:
SCR versus untreated control (C) not significant. aODNE versus SCR, P 5
0.0001. (B) Gene expression. PCR products were separated by
electrophoresis into 6% polyacrylamide gel/10% glycerol. Size of PCR
products was evaluated by comparing ethidium bromide stained RT--PCR
products with fragment size marker, after electrophoresis in 2% agarose gel.
Data are means SD, n ˆ 5. ANOVA analysis showed significant
differences among groups for E2F1 (P 5 0.0001). Multiple comparisons
(TK test): SCR versus C, not significant. aODNE versus C and SCR,
P 5 0.001.
and housekeeping genes. Finally, a number of observations in
different experimental systems have shown a high specificity
of aODNs against MYC, CYCLIN D1 and various other genes
(21,22,35,36).
pRb hyperphosphorylation and consequent release of its
suppressor association with the E2F family, leads to the
increase in E2F transcriptional activity and up-regulation of
cell cycle genes, including CYCLIN A, CYCLIN D1 and
CYCLIN E, DNA synthesis genes (DNA polymerase a, dihydropholate reductase), transcriptional factors, including
c-MYC, E2F1 and E2F2 (1,37--40). Thus, decreased E2F1
expression by MYC and CYCLIN D1 aODNs, could result in
decreased transactivation of c-MYC and CYCLIN D1 genes by
E2F1, and consequent accretion of the aODNs effects. However, according to our findings, MYC and CYCLIN D1 aODNs
apparently did not affect CYCLIN A and CYCLIN E expression
in HepG2 and 5123 cells, although they induced 450%
decrease in E2F1 production. A possible explanation may be
that residual E2F1 protein levels, in liver cancer cells treated
with MYC and CYCLIN D1 aODNs, are sufficient to activate
339
M.M.Simile et al.
CYCLIN A and CYCLIN E genes. Another possibility is the
absence of a regulatory loop between E2F1 and CYCLIN A and
CYCLIN E in in vitro growing liver cancer cells. Nonetheless,
the inhibition of E2F1 expression, either by MYC, CYCLIN D1
or E2F1 aODNs should result in a decrease in the formation of
E2F1--DP1 complex (1--3), and, consequently, in relatively
low activation of DNA synthesis genes. This may explain the
observed marked decrease in DNA synthesis, even in the
presence of active CYCLIN E and CYCLIN A genes, suggesting
that E2F1 is a sensitive target for antisense therapy of HCC.
This conclusion apparently contrasts with the reported unique
ability of E2f1 gene to function in mice as both oncogene and
tumor suppressor gene (41). It has been found that the development of skin papillomas and head and neck tumors is accelerated by crossing c-Myc transgenic mice with E2f1 null mice.
Various other observations, however, indicate that c-Myc and
E2F1 over-expression in transgenic mice promotes hepatocarcinogenesis, that is further accelerated by the co-expression
of the two genes in double transgenic mice (16,17). Contrasting results could be likely attributable to the E2f1 property of
signaling p53-dependent apoptosis (42), and to the requirement of distinct activities of E2F family genes for S phase
and apoptosis induction by c-Myc (32). Thus, the effect of
changes of E2f1 activity on the growth of tumors overexpressing c-Myc could depend on differences in the balance
between the relative activities of different E2F family members and, consequently, between apoptosis and cell growth in
different experimental systems.
Although over-expression of both c-MYC and E2F1 may
sensitize host cells to apoptosis (43), the treatment of HepG2
cells with MYC and CYCLIN D1 aODNs induced cell death by
apoptosis. Inhibition of growth and induction of apoptosis by
c-MYC or CYCLIN D1 aODNs have already been observed in
human, woodchuck and rat hepatoma cell lines (35,36,44). The
inhibition of c-myc expression in neoplastic liver nodules of
rats treated with S-adenosyl-l-methionine, is associated with
growth restraint and enhanced apoptosis (45,46). The mechanisms leading to apoptosis in cells with down-regulated c-MYC,
CYCLIN D1 and E2F1 genes are not clear. High apoptosis in
HCC cells, treated with c-MYC aODN has been attributed to
Bcl-2 under-regulation (36). However, this mechanism has not
been evaluated in our experimental conditions. Further work is
needed to examine the changes of different genes involved in
apoptosis, including the members of E2F family, and cell death
signaling pathways, during MYC, CYCLIN D1 and E2F1
aODNs therapy, to understand the mechanisms underlying
the double effect of these treatments on cell growth and
apoptosis.
G1 and S cyclins, through interaction with their catalytic
subunits, induce phosphorylation of pRb, that leads to the
activation of E2F family genes (47). E2F1 interacts with DP1
and induces S-phase genes by binding to promoters. According
to our results, E2F1 expression is supported by both MYC and
CYCLIN D1. Therefore, increase in CYCLIN D1 gene activity
in HCC may lead to both, decreased interaction of E2F1 with
pRb and increased expression of E2F1. This latter phenomenon also underlies MYC action. This could have some relevance for HCC prevention and therapy. The inhibition of E2F1
gene expression and regulatory interactions of MYC and/or
CYCLIN D1 with E2F1, in association with the pharmacological inhibition of CDKs (48,49), could enhance the therapeutic effect of these inhibitors, which could be used consequently
at relatively low and non-toxic doses.
340
Acknowledgements
Supported by grants from Associazione Italiana Ricerche sul Cancro, Compagnia S. Paolo, Torino, Ministero della Istruzione (PRIN and FIRB), and
Assessorato Igiene e SanitaÁ RAS.
References
1. Gra~
na,X., Garriga,J. and Mayol,X. (1998) Role of the retinoblastoma
protein family, pRB, p107 and p130 in the negative control of cell growth.
Oncogene, 24, 3365--3383.
2. Adams,P.D. (2001) Regulation of retinoblastoma tumor suppressor protein
by cyclin/cdks. Biochim. Biophys. Acta, 1471, 123--133
3. Bartek,J., Lukas,J. and Bartkova,J. (1999) Perspective: defects in cell
cycle control and cancer. J. Pathol., 187, 95--99.
4. Michalides,R.J.A.M. (1999) Cell cycle regulators: mechanisms and their
role in aetiology, prognosis, and treatment of cancer. J. Clin. Pathol., 52,
555--568.
5. Santoni-Rugiu,E., Jensen,M.R. and Thorgeirsson,S.S. (1998) Disruption of
pRb/E2F pathway and inhibition of apoptosis are major oncogenic events
in liver constitutively expressing c-myc and Transforming Growth
Factor a. Cancer Res., 58, 123--134.
6. Pascale,R.M., De Miglio,M.R., Muroni,M.R., Simile,M.M., Daino,L.,
Seddaiu,M.A., Nufris,A., Gaspa,L., Deiana,L. and Feo,F. (1996) c-myc
amplification in pre-malignant and malignant lesions induced in rat liver
by the resistant hepatocyte model. Int. J. Cancer, 68, 136--142.
7. De Miglio,M.R., Simile,M.M., Muroni,M.R. et al. (1999) c-myc overexpression and amplification correlate with progression of preneoplastic
liver lesions to malignancy in the poorly susceptible Wistar rat strain. Mol.
Carcinogen., 25, 21--29.
8. Pascale,R.M., Simile,M.M., De Miglio,M.R., Muroni,M.R., Calvisi,D.F.,
Asara,G., Casabona,D., Frau,M., Seddaiu,M.A. and Feo,F. (2000) Cell
cycle deregulation in liver lesions of rats with and without genetic
predisposition to hepatocarcinogenesis. Hepatology, 35, 1341--1350.
9. Zhang,Y.J., Jiang,W., Chen,C.J., Lee,C.S., Kahn,S.M., Santella,R.M. and
Weinstein,I.B. (1993) Amplification and overexpression of cyclin D1 in
human hepatocellular carcinoma. Biochem. Biophys. Res. Commun., 196,
1010--1016.
10. Paterlini,P., Flejow,J.F., De Mitri,M.S., Pisi,E., Franco,D. and Brechot,C.
(1995) Structure and expression of the cyclin A gene in human primary
liver cancer. Correlation with flow cytometric parameters. J. Hepatol., 23,
47--52.
11. Ohashi,R., Gao,C., Miyazaki,M., Hamazaki,K., Tsuji,T., Inoue,Y.,
Uemura,T., Hirai,R., Shimizu,N. and Namba,M. (2001) Enhanced
expression of cyclin E and cyclin A in human hepatocellular carcinoma.
Anticancer Res., 21, 657--662.
12. Daksis,J.I., Lu,R.Y., Facchini,L.M., Marhin,W.W. and Penn,L.J.Z. (1994)
Myc induces cyclin D1 expression in the absence of de novo protein
synthesis and links mitogen-stimulated signal transduction to the cell
cycle. Oncogene, 9, 3635--3645.
13. Philipp,A., Schneider,A., Vasrik,I., Finke,K., Xiong,Y., Beach,D.,
Alitalo,K. and Eilers,M. (1994) Repression of cyclin D1: a novel function
of MYC. Mol. Cell. Biol., 14, 4032--4043.
14. Obaya,A.J., Mateyak,M.K. and Sedivy,J.M. (1994) Mysterious liaisons:
the relationship between c-Myc and the cell cycle. Oncogene, 18,
2934--2941.
15. Perez-Roger,I., Kim,S.H., Griffiths,B., Sewing,A. and Land,H. (1999)
Cyclins D1 and D2 mediate Myc-induced proliferation via sequestration of
p27KIP1 and p21CIP1. EMBO J., 18, 5310--5320.
16. Conner,E.A., Lemmer,E.R., Omori, M., Wirth,P.J., Factor,V.M. and
Thorgeirsson,S.S. (2000) Dual functions of E2F-1 in transgenic mouse
model of liver carcinogenesis. Oncogene, 19, 5054--5062.
17. Conner,E.A., Lemmer,E.R., Sanchez,A., Factor,V.M. and Thorgeirsson, S.S.
(2003) E2F1 blocks and c-Myc accelerates hepatic ploidy in transgenic
mouse models. Biochim. Biophys. Res. Commun., 302, 114--120.
18. Oswald,F., Lovec,H., Moroy,T. and Lipp,M. (1994) E2F-dependent
regulation of human MYC: trans-activation by cyclins D1 and A overrides
tumour suppressor protein functions. Oncogene, 9, 2029--2036.
19. Johnson,D.G. (1995) Regulation of E2F1 gene expression by p130 and
D-type cyclin kinase activity. Oncogene, 11, 1685--1692.
20. Sherr,C.J. (2000) The Pezcoller Lecture: cancer cell cycles revisited.
Cancer Res., 60, 3689--3695.
21. Helene,C. (1994) Control of oncogene expression by antisense nucleic
acids. Eur. J. Cancer, 30A, 1721--1726.
Cell cycle gene interactions and antisense therapy
22. Weiss,B., Davidkova,G. and Zhou,L.-W. (1999) Antisense RNA gene
therapy for studying and modulating biological processes. Cell. Mol. Life
Sci., 55, 334--358.
23. Curcio,L.D., Bouffard,D.Y. and Scanlon,K.J. (1997) Oligonucleotides as
modulators of cancer gene expression. Pharmacol. Ther., 74, 317--332.
24. Saijo,Y., Uchiyama,B., Abe,T., Satoh,K. and Nukiwa,T. (1997)
Contiguous four-guanosine sequence in c-myc antisense phosphorothioate
oligonucleotides inhibits cell growth on human lung cancer cells: possible
involvement of cell adhesion inhibition modulators of cancer gene
expression. Jpn. J. Cancer Res., 88, 26--33.
25. Capaccioli,S., Di Pasquale,G., Mini,E., Mazzei,T. and Quattrone,A. (1993)
Cationic lipids improve antisense oligonucleotide uptake and prevent
degradation in cultured cells and in human serum. Biochem. Biophys. Res.
Commun., 197, 818--825.
26. Brass Nheckel,D. and Meese,E. (1998) Comparative PCR: An improved
method to determine gene amplification. BioTechniques, 24, 22--26.
27. Lowry,O.K., Rosebrough,N.J., Farr,A.L. and Randall,R.J. (1951) Protein
measurement with folinphenol reagent. J. Biol. Chem., 293, 265--275.
28. Fan,J. and Bertino,J.R. (1997) Functional roles of E2F in cell cycle
regulation. Oncogene, 14, 1191--1200.
29. Coller,H.A.,
Grandori,C.,
Tamayo,P., Colbert,T.,
Lander,E.S.,
Eisenman,N.R. and Golub,T.R. (2000) Expression analysis with oligonucleotide microarrays reveals that MYC regulates genes involved in growth,
cell cycle, signaling, and adhesion. Proc. Natl Acad. Sci. USA, 97,
3260--3265.
30. Nasi,S., Ciarapica,R., Jucker,R., Rosati,J. and Souceck,L. (2001) Making
decisions through Myc. FEBS Lett., 490, 153--162.
31. Calbo,J., Parre~no,M., Sotillo,E., Yong,T., Mazo,A., Garriga,J. and
Gra~na,X. (2002) G1 cyclin/cyclin-dependent kinase-coordinated phosphorylation of endogenous pocket proteins differentially regulates their
interactions with E2F4 and E2F1 and gene expression. J. Biol. Chem., 277,
50263--50274.
32. Leone,G., Sears,R., Huang,E. et al. (2001) Myc requires distinct E2F
activities to induce S phase and apoptosis. Mol. Cell, 8, 105--113.
33. Liao,D.J., Natarajan,G., Deming,S.L., Jamerson,M.H., Johnson,M.,
Chepko,G. and Dickson,R.B. (2000) Cell cycle basis for the onset and
progression of c-Myc-induced, TGFalpha-enhanced mouse mammary
gland carcinogenesis. Oncogene, 19, 1307--1317.
34. Santoni-Rugiu,E., Duro,D., Farkas,T., Mathiasen,I.S., Jaattela,M.,
Bartek,J. and Lukas,J. (2002) E2F activity is essential for survival of
Myc-overexpressing human cancer cells. Oncogene, 21, 6498--6509.
35. Uto,H., Ido,A., Moriuchi,A., Onaga,Y., Nagata,K., Onaga,M., Tahara,Y.,
Hori,T., Hirono,S., Hayashi,K. and Tsubouchi,H. (2001) Transduction of
antisense cyclin D1 using two-step gene transfer inhibits the growth of rat
hepatoma cells. Cancer Res., 61, 4779--4783.
36. Ebinuma,H., Saito,H., Kosuga,M., Wakabayashi,K., Saito,Y., Takagi,T.,
Nakamoto,N., Okuyama,T. and Ishii,H. (2001) Reduction of c-myc
expression by an antisense approach under Cre/loxP switching induces
apoptosis in human liver cancer cells. J. Cell. Physiol., 188, 56--66.
37. DeGregori,J., Kowalik,T. and Nevis,J.R. (1995) Cellular targets for
activation by E2F1 transcription factor include DNA synthesis and G1 /Sregulatory genes. Mol. Cell. Biol., 15, 4215--4224.
38. Nevins,J.RT. (1998) Toward an understanding of the functional
complexity of the E2F and retinoblastoma families. Cell Growth Differ.,
9, 585--593.
39. Jacks,T. and Weinberg,R.A. (1996) Cell-cycle control and its watchman.
Nature, 381, 643--644.
40. Johnson,D.G. and Scheneider-Broussard,R. (1998) Role of E2F in cell
cycle control and cancer. Front. Biosci., 3, d447--d448.
41. Rounbehler,R.J., Rogers,P.M., Conti,C.J. and Johnson,G.G. (2002)
Inactivation of E2f1 enhances tumorigenesis in a Myc transgenic model.
Cancer Res., 62, 3276--3281.
42. Pan,H., Yin,C., Dyson,N.J., Harlow,E., Yamasaki,L. and Van Dyke,T.
(1998) Key roles for E2F1 in signaling p53-dependent apoptosis and in
cell division within developing tumors. Mol. Cell. Biol., 2, 283--292.
43. Tanaka,H., Matsumura,I., Ezoe,S., Satoh,Y., Sakamaki,T., Albanese,C.,
Machii,T., Pestell,R.G. and Kanakura,Y. (2002) E2F1 and c-Myc
potentiate apoptosis through inhibition of NF-kappaB activity that
facilitates MnSOD-mediated ROS elimination. Mol. Cell., 9, 1017--1029.
44. Wang,H.P., Zhang,L., Dandri,M. and Rogler,C.E. (1998) Antisense
downregulation of N-myc1 in woodchuck hepatoma cells reverses the
malignant phenotype. J. Virol., 72, 2192--2198.
45. Garcea,R., Daino,L., Pascale,R.M., Simile,M.M., Puddu,M., Frassetto,S.,
Cozzolino,P., Seddaiu,M.A., Gaspa,L. and Feo,F. (1989). Inhibition of
promotion and persistent nodule growth by S-adenosyl-l-methionine in rat
liver carcinogenesis: role of remodeling and apoptosis. Cancer Res., 49,
1850--1856.
46. Garcea,R., Daino,L., Pascale,R., Simile,M.M., Puddu,M., Ruggiu,M.E.,
Seddaiu,M.A., Satta,G., Sequenza,M.J. and Feo,F. (1989) Protooncogene
methylation and expression in regenerating and preneoplastic liver nodules
induced in the rat by diethylnitrosamine: effect of variations of
S-adenosylmethionine:S-adenosylhomocysteine ratio. Carcinogenesis,
10, 1183--1192.
47. Sherr,C.J. (1996) Cancer cell cycles. Science, 274, 1672--1677.
48. Ruetz,S., Fabro,D., Zimmermann,J., Meyer,T. and Gray,N. (2003)
Chemical and biological profile of dual Cdk1 and Cdk2 inhibitors. Curr.
Med. Chem. Anti-Canc. Agents, 3, 1--14.
49. Yu,C., Rahmani,M., Dai,Y., Conrad,D., Krystal,G., Dent,P. and Grant,S.
(2003) The lethal effects of pharmacological cyclin-dependent kinase
inhibitors in human leukemia cells proceed through a phosphatidylinositol
3-kinase/Akt-dependent process. Cancer Res., 63, 1822--1833.
Received August 8, 2003; revised October 15, 2003;
accepted October 28, 2003
341