Gene Therapy (2001) 8, 1149–1156
 2001 Nature Publishing Group All rights reserved 0969-7128/01 $15.00
www.nature.com/gt
RESEARCH ARTICLE
Transfer of bcl-xs plasmid is effective in preventing
and inhibiting rat hepatocellular carcinoma induced by
N-nitrosomorpholine
M Baba1, H Iishi2 and M Tatsuta1
1
Department of Gastrointestinal Oncology, and 2Department of Gastroenterology, Osaka Medical Center for Cancer and
Cardiovascular Diseases, Osaka, Japan
To examine the effect of the bcl-xs gene on the sequence
from hepatic precancerous lesions, foci and neoplastic nodules, to hepatocellular carcinomas, Sprague–Dawley rats
were given water containing 175 mg/l N-nitrosomorpholine
(NNM) for 8 weeks. At weeks 1, 4 and 7, the left lobe of the
rat liver was exposed and injected with the bcl-xs plasmid
(pCR3.1-rat bcl-xs cDNA) or pCR3.1 encapsulated in cationic empty liposomes each at a dose of 80 ␮g plasmid/kg
body weight. One minute later, low-field-strength, long-duration electric pulses were applied to the left lobe using a
pincette electrode with circular poles 1 cm in diameter. The
in vivo electroporation procedure significantly increased the
transfer of the reporter gene chloramphenicol acetyl transferase (CAT) plasmid via empty liposomes. Thus, CAT mRNA
was expressed not only at the sites of electrode contact but
at sites 0.5–1.0 cm away from the electrode, and expression
also increased with increasing doses of plasmid, meaning
that in vivo electroporation enabled the expression of plasmid DNA throughout an extensive area of the rat liver. By
week 11, the neoplastic nodules were significantly fewer and
smaller in the bcl-xs group than in the pCR3.1 group at the
two sites, one with and the other without electrode contact.
No hepatocellular carcinomas were found in the rats that had
received the bcl-xs plasmid, whereas these tumors were
observed in 30% of the rats given pCR3.1. Moreover, overexpression of the bcl-xs protein was detected, and apoptotic
activity was significantly increased in the neoplastic nodules,
foci and hepatocytes adjacent to the hepatic lesions. These
results indicate that the bcl-xs plasmid inhibits the occurrence and growth of rat hepatocellular carcinoma and may
thus be effective for the prevention and treatment of human
liver tumors. Gene Therapy (2001) 8, 1149–1156.
Keywords: bcl-xs plasmid; gene therapy; hepatocellular carcinoma; in vivo electroporation; apoptosis; rat
Introduction
The bcl-2 family of proteins can either promote or inhibit
cell death. It was proposed that the ratio of pro-survival
to pro-death members of this family determines the propensity of a cell to live or die.1,2 Because anti-apoptotic
proteins, bcl-2 and bcl-xL, are often enhanced in various
cancers, where they may play a pivotal role in tumor
initiation and progression,3,4 bcl-2 and bcl-xL genes were
suggested to be targets for cancer gene therapy. Administrations of bcl-2 and bcl-xL antisense oligonucleotides promoted apoptosis and inhibited cell growth in brest carcinoma cells5 and small-cell lung cancer cells,6 in which bcl2 or bcl-xL was detectable. In our previous report,7 we
examined the effects of in vivo electroporetic transfer of a
bcl-2 antisense oluigonucleotide phosphorothioate on rat
hepatocarcinogenesis, and showed that bcl-2 antisense
oligonucleotide inhibited expression of bcl-2 mRNA and
the occurrence and growth of rat hepatocellular carcinoma induced by NNM. The selective administration of
a bcl-2 antisense oligonucleotide may thus provide an
Correspondence: M Baba, Department of Gastrointestinal Oncology,
Osaka Medical Center for Cancer and Cardiovascular Diseases, 3–3, Nakamichi 1-chome, Higashinari-ku, Osaka 537–0025, Japan
Received 22 November 2000; accepted 23 May 2001
effective therapy for liver tumors. However, since not all
human liver tumors overexpress bcl-2 protein,8 bcl-2 antisense would probably be effective for all types of liver
tumors.
Among the pro-apoptotic bcl-2 family members, bcl-xs
is unique in that it lacks BH-1 and BH-2 domains, which
constitute part of the membrane spanning-domains of
bcl-2, bcl-xL and bax. Bcl-xs, however, has BH-3 as well as
BH-4 domains, whch enable it to interact outside the bcl2 family.9 Bcl-xs is thus thought to act either by binding
to and inactivating the pro-survival functions of bcl-2 and
bcl-xL, or by displacing pro-death bcl-2 family member
such as bax, from bcl-2 or bcl-xL, thereby allowing it to
kill cells.10 However, it was recently reported that bcl-xs
induces acute cell death in 3T3 cells by a novel cell death
pathway that does not utilize caspases.11,12 Infection with
a bcl-xs adenovirus vector effectively killed human mammary tumor cells and decreased their growth in transplanted murine tumors.13 In addition, it has been shown
that c-myc- or c-erb-2-transformed murine mammary cells
are extremely sensitive to apoptosis induced by bcl-xs
adenovirus vector as compared with non-transformed
cells.14 Most liver tumors overexpress c-myc,15,16 and may
be sensitive to apoptosis induced by exogenous bcl-xs.
These findings indicate that bcl-xs gene is more effective
for various types of liver tumors than antisense oligonucleotides of the anti-apoptotic bcl-2 family.
Prevention and inhibition of rat liver tumor by bcl-xs plasmid
M Baba et al
1150
In humans, liver cirrhosis is known to predispose to
liver tumors and also to frequently be associated with
recurrence of liver tumors.17,18 Therefore, in this experiment, we observed the effect of bcl-xs gene on the
sequence from hepatic precancerous lesions, foci and
neoplastic nodules, to hepatocellular carcinomas, which
was obtained by giving rats NNM in drinking for 8
weeks.19
Since antisense oligonucleotides are cleaved in vivo
within a short time, despite chemical modification of
phosphodiester linkages,20 a higher dose than that of the
plasmid or viral vectors is necessary for treatment,
resulting in toxicity. However, viral vectors have the limitations such as the possibility of insertional mutagenesis
or induction of a host immune response, despite inducing
high long-term gene expression.21 On the other hand,
plasmid DNA is neither replicated nor integrated into the
host cell genome, but remains in its episomal form,22 and
gene therapy with plasmid DNA is a rapidly expanding
field.23–25 Moreover, in vivo electroporation can transfer
DNA into most cells including skin, liver and tumor cells,
as well as to both proliferating and nonproliferating
cells.26–28
Therefore, to examine the effectiveness of the bcl-xs
gene in preventing the occurrence and inhibiting the
growth of liver tumors, we encapsulated bcl-xs plasmids
in cationic empty liposomes, which assured particles
with a constant diameter, and injected them directly into
the NNM-treated rat liver, and then performed in vivo
electroporation.
Figure 1 (a) A schematic representation of the site in the left lobe of the
rat liver sandwiched by the pincette electrode during the in vivo electroporation procedure and the site 0.5–1.0 cm away from the site of electrode
contact. Sections (2–3 mm thick) were taken from the sites and used for
histological examination and analysis of apoptotic activity and bcl-xs protein. (b) A schematic diagram of the three segments (a, b, c) in a liver
section from both the site of electrode contact and the site away from
electrode. These segments were analyzed for expression of the reporter
gene, CAT mRNA, by RT-PCR.
(Figure 2a). CAT mRNA was also expressed at sites 0.5–
1.0 cm from the point of electrode contact in rat liver
tissue that had received 80 or 120 ␮g plasmid/kg (Figure
2b). The expression of CAT mRNA at sites some distance
from the electrode was weaker than at sites of electrode
contact. However, both at sites of electrode contact and
at those some distance from the electrode, expression
Results
CAT expression in rat liver
The pCR3.1-CAT plasmid (Invitrogen), containing the
chloramphenicol acetyl transferase gene (CAT cDNA
fragment, 659 bp) downstream from the CMV promoter
in pCR3.1 as a reporter gene, was used to examine the
extent of the area into which the plasmid had been transferred, in the left hepatic lobe and in which bcl-xs mRNA
was expressed. Sprague–Dawley rats were randomly divided into two groups of six rats each and each group was
then subdivided into three subgroups of two rats. All rats
were injected with 15 mg/kg sodium pentobarbital intraperitoneally. After complete anesthesia, the left lobe of
the liver was exposed and then injected with pCR3.1-CAT
at doses of 40, 80 or 120 ␮g/kg body weight, as plasmids
encapsulated in empty liposomes which assured particles
with a constant diameter. One minute later, in three of
the subgroups, the site where the plasmid had been
injected was sandwiched in a pincette electrode with circle poles 1 cm in diameter (positive and negative) (Figure
1a). Eight electrical pulses, with a current of 0.5 A, electric-field strength of 70–100 V/cm and a pulse duration
of 50 ms, were delivered to the left lobe. However, in
the other three subgroups injected with plasmids, in vivo
electroporation was not performed. Three days later,
three segments (20–30 mg) (a, b, c) were taken from each
section of the left lobe of the liver, ie a sandwiched section and a non-sandwiched section (Figure 1b). The segments were examined by RT-PCR for CAT mRNA. CAT
mRNA was found to be expressed in all three segments
(a, b, c) at the sites of electrode contact in rat livers that
had been injected with 40, 80 or 120 ␮g CAT plasmid/kg
Gene Therapy
Figure 2 CAT mRNA expression in rat liver tissue. Twelve rats were
divided into two groups (each six rats). Each group was then subdivided
into three subgroups of two rats each whose left lobe of the liver was then
injected with pCR3.1-CAT encapsulated in empty liposomes, at doses of
40, 80 or 120 ␮g/kg body weight, respectively. One minute later, in vivo
electroporation was performed with a pincette electrode in the three subgroups at sites of plasmid injection and not in the other three subgroups.
Control rats were injected with 80 ␮g/kg of pCR3.1 encapsulated in liposomes. Three days later, three segments (a, b, c) were taken from each
section from both the sites that were sandwiched between the pincette
electrode and those that were not (Figure 1b). Total RNA was extracted
from them, and expression of CAT mRNA was analyzed by RT-PCR. (a)
CAT mRNA expression at the site of electrode contact (electric transfer)
and at the corresponding site in rat liver injected with pCR3.1-CAT
(lipofection). (b) CAT mRNA expression at a site 0.5–1.0 cm away from
the electrode contact (electric transfer) and at a site corresponding to that
site (lipofection). M, DNA marker band of 603 bp; C, the PCR product
(475 bp) as positive control, which was amplified by PCR using 100 pg
of pCR3.1-CAT.
Prevention and inhibition of rat liver tumor by bcl-xs plasmid
M Baba et al
increased with increasing doses of CAT plasmid (Figure
3a and b).
By contrast, in rat livers injected with CAT plasmid
encapsulated in liposomes, but not exposed to in vivo
electroporation, CAT mRNA was expressed only in liver
tissues injected with 80 or 120 ␮g plasmid/kg, but not in
liver tissues injected with 40 ␮g plasmid/kg. Moreover,
expressions in the three segments were not at the same
levels (Figures 2a and 3a). No CAT mRNA was expressed
in any segments corresponding to sites 0.5–1.0 cm from
the point of electrode contact (Figure 2b). There was no
CAT expression in rat livers injected with pCR3.1
regardless of whether or not in vivo electroporation was
performed (Figure 2a).
In addition, to examine the effect of empty liposomes
on in vivo electroporation, four Sprague–Dawley rats
were divided into two groups and then injected with 80
␮g/kg of pCR3.1-CAT (encapsulated in empty
liposomes) and naked pCR3.1-CAT, respectively. Next, in
vivo electroporation was perfomed in the two groups. In
the two groups, CAT mRNA was expressed not only at
sites of electrode contact but also at sites away from electrode. However, the expression of CAT mRNA was
approximately two times higher in the liver tissues given
plasmid encapsulated in liposomes than those given
naked plasmid (Figure 4a and b).
These results show that the reporter gene was
expressed not only in tissues exposed to electric pulses,
but also in tissues some distance from the point of electrode contact, and that empty liposomes increased electroporetic transfer of CAT plasmid. Plasmids encapsulated in empty liposomes have been transferred into a
significant wide area of rat liver by in vivo electroporation
with the pincette electrode.
Effect of bcl-xs gene transfer on liver tumor growth
To assess the therapeutic effect of the bcl-xs gene on liver
tumors, plasmid pCR3.1-bcl-xs was electroporetically
transferred into the livers of rats treated with NNM. Rat
bcl-xs cDNA (576 bp) was produced by RT-PCR using
total RNA extracted from Sprague–Dawley rat spleens
and was subcloned by insertion into the downstream
CMV promoter of pCR3.1. Rats were divided randomly
into three groups of 20 rats each and given drinking
water containing 175 mg/l NNM for 8 weeks. In this
experimental model, the progression of hepatocellular
carcinoma can be observed as a sequence from precancerous lesions to hepatocellular carcinoma. Histological
Figure 3 Linear arbitrary unit (LAU) tracing of CAT mRNA expression.
The levels of CAT mRNA by RT-PCR were examined using FLA-2000
Science Imaging System (exitation 473 nm, emission more than 580 nm),
normalized by ␤-actin and represented as percent of LAU relative to control (PCR product using 100 pg of pCR3.1-CAT). (a) LAU tracing of
Figure 2a; (b) LAU tracing of Figure 2b.
Figure 4 Effect of empty liposomes on the transfer of CAT plasmid by in
vivo electroporation. Four rats were divided into two groups and injected
with 80 ␮g/kg of pCR3.1-CAT encapsulated in empty liposomes and naked
pCR3.1-CAT, respectively, and in vivo electroporation was performed in
the two groups. (a) expression of CAT mRNA by RT-PCR in three segments at both sites of electrode contact and sites away from electrode. (b)
LAU tracing of (a) showing effect of empty liposomes on transfer of the
plasmid using in vivo electroporation.
1151
Gene Therapy
Prevention and inhibition of rat liver tumor by bcl-xs plasmid
M Baba et al
1152
typing of hepatic lesions was performed according to
Institute of Laboratory Animal Resources (1980) guidelines29 (Figure 5). There were no precancerous lesions,
foci or neoplastic nodules, for the 2 weeks starting at the
beginning of the experiment. At week 3, foci, areas of
cellular alterations, appeared and began to grow, and at
week 7 neoplastic nodules, which were composed of
more alternative cells than foci, manifested from foci and
began to grow. When administration of NNM was
stopped at week 8, more than 95% of precancerous
lesions had undergone remodeling and only less than 1%
of residual irreversible lesions had progressed to hepatocellular carcinoma. Tumor-bearing rats survived until
more than 20 weeks.
At weeks 1, 4 and 7, pCR3.1-bcl-xs encapsulated in
empty liposomes (group 1), pCR3.1 encapsulated in liposomes (group 2), or liposomes only (group 3) were
injected into the left hepatic lobes of rats, at a dose of 80
␮g/kg body weight, and these injections were followed
by in vivo electroporation. At week 11, there were no differences in the numbers and size of the foci at the sites
that received the electric pulses in groups 1 (pCR3.1-bclxs), 2 (pCR3.1) and 3 (liposomes), and also there was no
difference in the size of the foci at sites away from electrode in groups 1, 2 and 3. However, neoplastic nodules,
were significantly fewer and smaller in group 1 than in
groups 2 and 3 (Figure 6a and b). Moreover, the incidences of hepatocellular carcinoma revealed bcl-xs to
have completely inhibited the development of hepatocellular carcinoma both at sites of electrode contact and at
sites away from the electrode though hepatocellular carcinomas were present in groups 2 and 3 to the same
extent (Table 1).
These results show that three electroporetic transfers of
bcl-xs plasmids encapsulated in liposomes inhibited the
conversion of foci into neoplastic nodules and the development of hepatocellular carcinoma, and that the effect
of bcl-xs extended even to sites 0.5–1.0 cm away from the
point of electrode contact.
Western blotting for bcl-xs protein
At week 11, bcl-xs protein was expressed at sites of electrode contact in group 1 (pCR3.1-bcl-xs), and also
expressed at sites away from electrode in the liver
although less than at sites of electrode contact. However,
there was no expression of bcl-xs protein in group 2
(pCR3.1) at both sites of electrode contact and sites away
from electrode. On the other hand, at both sites of electrode contact and sites away from electrode, bcl-xL protein was expressed in groups 1 and 2 to the same degree
(Figure 7).
Figure 6 Number and size of foci, neoplastic nodules, and hepatocellular
carcinomas in NNM-treated rats. Every 3 weeks, a total of three times,
directly the rats given NNM were injected into the left lobe of the liver
with 80 ␮g/kg of bcl-xs plasmids, pCR3.1 encapsulated in liposomes and
liposomes only, and electric pulses were then applied. At week 11, all rats
were killed, and liver sections from both sites of electrode contact (a) and
sites 0.5–1.0 cm away from electrode (b) were taken and analyzed histologically. *Significantly different from the values for groups 2 and 3 (P
⬍ 0.05).
Table 1 Incidence of hepatocellular carcinoma in NNM-treated
rats
Group
No.
1
2
3
Figure 5 Staining of focus, neoplastic nodule and hepatocellular carcinoma in left hepatic lobe at week 11. In this experimental model, foci
and neoplastic nodules appeared from weeks 3 and 7, respectively, and
hepatocellular carcinomas manifested from week 11. Scale bar, 100 nm.
Gene Therapy
a
Treatmenta
pCR3.1-bcl-xs
pCR3.1
Liposomes only
No. of
rats
20
20
20
No. of rats bearing
hepatocellular carcinoma (%)
Sites of
electrode
contact
Away from
electrodes
0 (0)b
6 (30)
8 (40)
0 (0)b
6 (30)
6 (30)
For an explanation of the procedure, see Figure 6.
Significantly different from the values for groups 2 and 3 (P ⬍
0.02).
b
Prevention and inhibition of rat liver tumor by bcl-xs plasmid
M Baba et al
1153
Figure 7 Western blotting analysis of bcl-xs protein. For an explanation
of the procedure, see Figure 6. At week 11, the sections at both site of
electrode contact and site away from electrode were analyzed. Lanes 1 and
2, protein expression at the site of electrode contact and at a site away
from electrode in group 1 (pCR3.1-bcl-xs), respectively; lanes 3 and 4,
protein expression at the site of electrode contact and at a site away from
electrode in group 2 (pCR3.1). Values represented the percentage of protein expression quantified using Eagle Sight Image Capture and Analysis
Software (Stratagene) and corrected for ␤-actin.
These results show that bcl-xs has overexpressed by
transfer of the bcl-xs plasmid and that exogenous bcl-xs
plasmid did not affect the transcription of bcl-xL, an alternative splicing product of bcl-xs.
Apoptotic index
In situ labeling of new 3⬘-OH of the DNA fragments generated by apoptosis-associated endonuclease was used to
identify apoptotic cells at sites of electrode contact and
at sites away from electrode. In group 1 (pCR3.1-bcl-xs),
the apoptotic index in the preneoplastic nodules, foci and
hepatocytes adjacent to the hepatic lesions was significantly higher than in groups 2 (pCR3.1) and 3
(liposomes), but there was no difference in the apoptotic
index in the neoplastic nodules, foci and hepatocytes
adjacent to the hepatic lesions in groups 2 and 3 at both
the sites of electrode contact and away from electrode
(Figure 8a and b).
These results show that inhibition of the development
of hepatocellular carcinomas by bcl-xs plasmid caused the
pro-apoptotic activity of bcl-xs protein.
Discussion
Transfer of the bcl-xs plasmids by in vivo electroporation
into the livers of rats that had consumed NNM induced
overexpression of the bcl-xs protein in liver tissues and
increased the apoptotic index of hepatocytes adjacent to
hepatic lesions, foci and neoplastic nodules. The development of neoplastic nodules was also decreased, resulting
in complete inhibition of hepatocellular carcinoma. These
findings show bcl-xs to decrease precancerous hepatic
lesions and inhibit the occurrence of hepatocellular carcinoma in rats by means of its pro-apoptotic activity. Since
in humans, liver cirrhosis is known to predispose toward
hepatocellular carcinoma and to frequently be associated
with recurrence of liver tumors,17,18 bcl-xs gene therapy
may be effective in both preventing and inhibiting the
occurrence and recurrence of liver tumors.
From a practical standpoint, low-field-strength long-
Figure 8 (a) Apoptotic index of hepatic lesions and hepatocytes adjacent
to hepatic lesions. For an explanation of the procedure, see Figure 6. At
week 11, liver sections from both sites of electrode contact and sites away
from electrode in the rat livers injected with liposomes only and 80 ␮g/kg
of bcl-xs plasmid and pCR3.1, following in vivo electroporation were taken
and fixed. Hepatocellular carcinomas, neoplastic nodules, foci and hepatocytes adjacent to the hepatic lesions were examined in 10 rats from each
group. Five hundred cells from each hepatic lesions and 1000 hepatocytes
from the area adjacent to the hepatic lesions were examined. The apoptotic
index in foci, neoplastic nodules, hepatocellular carcinomas and hepatocytes adjacent to the hepatic lesions is expressed as percent of labeled cells
relative to the total number of cell examined. *Significantly different from
the values for groups 2 and 3 (P ⬍ 0.05). **Significantly different from
the value for group 3 (P ⬍ 0.05). (b) Staining of neoplastic nodules in
groups 1 (pCR3.1-bcl-xs), 2 (pCR3.1) and 3 (liposomes). Scale bar, 50 nm.
duration electric pulses increase in vivo plasmid DNA
transfer to tumor cells (10 pulses; 800 V/cm; 4 ms per
pulses),26 normal liver cells (8 pulses; 250 v/cm; 50 ms
per pulses)27 and normal skeletal muscle (6 pulses; 200
v/cm; 50 ms per pulses).30 In the present experiment,
transfer of the bcl-xs plasmid encapsulated in liposomes
by in vivo electroporation with a pincette electrode under
conditions of 8 pulses, 70–100 v/cm, and 50 ms per pulse
significantly inhibited the development of rat liver
tumors at both sites of electrode contact and sites 0.5–1.0
cm away from the electrode. Moreover, CAT mRNA of
the reporter gene was expressed at sites of contact with
the electrode, as well as 0.5–1.0 cm away from the electrode, although less than at contact sites, meaning that
expression of the reporter gene extends beyond the border of the electrode (a circle 1 cm diameter). It was
reported that when plasmid pCMV-luciferase was transferred into skeletal muscle by electric square-wave pulses
of low field strength (less than 300 V/cm) and long duration (more than 1 ms), the electric pulses increased gene
transfer not only via cell permeabilisation but also by a
Gene Therapy
Prevention and inhibition of rat liver tumor by bcl-xs plasmid
M Baba et al
1154
direct effect on the DNA molecule, promoting DNA
migration and cellular uptake,30 and the electrotransfer
resulted in long-lasting expression. For example, stable
expression of fibroblast growth factor 1 plasmid in skeletal muscle was observed for at least 9 months.30 It has
also been speculated that target cells are genetically
modified by the electrotransfer of plasmids, which are
subsequently expressed for an extended period.31 It
appeared that bcl-xs plasmids were transferred into hepatocytes from the site of electrode contact via cell permeabilisation, and that they then migrated away from the
electrode contact site as the result of a direct modifying
effect of the pulses on the plasmid and that bcl-xs protein
was expressed for several weeks at least, an adequate
time for bcl-xs to inhibit the development of hepatocellular carcinoma.
Cationic empty liposomes (EL-C-01) enhanced the
expression of CAT plasmids via in vivo electroporation in
normal rat livers. Our previous experiment also indicated
that encapsulating bcl-2 antisense with that EL-C-01, forming particles with a constant diameter (approximately
100 nm), enhances the electric transfer of DNA. It appears
that DNA particles encapsulated in liposomes may pass
through pores produced by electroporation, and that
DNA uptake can be increased by the combination of in
vivo electroporation and empty liposomes.7
Although transfer of bcl-xs plasmids via in vivo electroporation can transfer the bcl-xs gene selectively to the
desired site, ie, hepatocellular carcinoma or a precancerous lesion, if the bcl-xs plasmid is taken up by normal
hepatocytes, the plasmid becomes toxic. However, it was
reported that c-myc- or c-erb-2-transformed murine mammary cells are extremely sensitive to apoptosis induced
by the bcl-xs adenovirus vector, whereas immortalized
nontransformed murine mammary cells are relatively
resistant to apoptosis induced by this vector.13 Given that
hepatic precancerous cells and hepatocellular carcinoma
overexpress c-myc and that the expression of c-myc in normal hepatocytes is minimal, many tumor cells and precancerous cells were killed while normal hepatocytes
appeared to be protected from apoptosis by bcl-xs.
Electroporetic transfer of the bcl-xs plasmid immediately after the surgical resection of liver tumors apprears
to prevent liver cirrhosis and decreases the recurrence of
liver tumors. In a previous study,7 the electroporetic
transfer of bcl-2 antisense oligonucleotides decreased bcl2 mRNA expression and inhibited the development of
hepatocellular carcinoma at sites of electrode contact, but
not at sites away from the electrode. However, with the
administration of the bcl-xs plasmid, rat hepatocarcinogenesis was inhibited at both the sites of electrode contact
and sites away from the electrode. The use of in vivo electroporetic transfer of plasmid DNA may be effective for
treating disease which has spread over a wide area, such
as liver cirrhosis.
The results of this experiment show bcl-xs to inhibit
the occurrence and growth of rat hepatocellular carcinoma, and that transfer of the bcl-xs gene may be effective
in preventing liver tumors as well as inhibiting their
recurrence.
Materials and methods
Production of bcl-xs plasmid
Total RNA was isolated from the livers of 6-week-old
Sprague–Dawley rats and reverse transcribed into firstGene Therapy
strand cDNA by using random primers (Promega, Madison, WI, USA) and RNase H-reverse transcriptase
(Superscript II; Gibco, BRL, Rockville, MD, USA). The following oligonucleotide primers based on human bcl-xs
cDNA sequence10 were synthesized (OligoExpress,
Amersham Pharmacia Biotech, Tokyo, Japan).
Forward: TTGGACAATGGACTGGTTGA,
Reverse: GTAGAGTGGATGGTCAGTG.
The first-strand cDNA was subjected to 40 cycles of PCR
amplification (1-min denaturation at 94°C, 2-min
annealing at 55°C, and 2-min extention 72°C), with a final
extension phase of 7 min at 72°C, and the amplified products were resolved through 1.5% agarose gels. The 576
bp-band of bcl-xs cDNA was obtained, and no nonspecific
production was detected. A 15 ng sample of bcl-xs cDNA
was inserted into the multiple cloning site downstream
of the CMV promoter of the expression vector, pCR3.1
(Invitrogen, San Diego, CA, USA) by TA-cloning.
pCR3.1-bcl-xs was also digested by BamHI, and the bcl-xs
cDNA obtained was inserted into the BamHI site of pUC
18 (Invitrogen) and transformed into TOP10F⬘-competent
cells. pUC 18-bcl-xs was extracted, purified with a Wizard
Plus Midipreps DNA Purification System (Promega,
Madison, WI, USA), and the cDNA sequence was analyzed by ABI PRISMS 377 (Applied Biosystems Foster
City, CA, USA). The resulting sequence was the same as
that of the bcl-x short cDNA of Sprague–Dawley rats submitted in the gene bank (accession AF 136230).
pCR3.1-bcl-xs was amplified in E. coli strain TOP10F⬘
and then isolated and purified with the Wizard Plus Midipreps DNA Purification System (Promega). Saline (2.3
ml) containing 500 ␮g of purified pCR3.1-bcl-xs was combined with 0.2 ml of cationic empty liposome solution
(Coatsome-EL-C-01; Nihon Yushi, Tokyo, Japan), gently
mixed, and incubated for 30 min at room temperature
and fresh saline was added at a final volume of 12.5 ml.
Coatsome-EL-C-01 are cationic empty liposomes, which
can form particles encapsulating DNA with a constant
diameter, and consists of l-␣-dipalmitoylphosphatidyl
cholin (52 ␮mol), cholesterol (40 ␮mol), and stearylamine
(8 ␮mol) in 3 ml saline solution. Under these conditions,
the liposomes encapsulate DNA in cationic particles of
approximately 100 nm diameter with 80% encapsulative efficiency.7
In vivo gene transfer
Sixty 4-week-old male Sprague–Dawley rats were purchased from Japan SLC (Shizuoka, Japan). After housing
them in metal cages in animal quarters with controlled
temperature (21–22°C), humidity (30–50%) and light (12h cycle) for 2 weeks, the animals were randomly divided
into three groups of 20 rats each and given drinking
water containing NNM 175 mg/l for 8 weeks. From week
9 until the end of the experiment, the rats were given
only ordinary tap water to drink. At weeks 1, 4 and 7,
all rats were subcutaneously injected with 0.25 mg/ml
atropine sulfate (Sigma, St Louis, MO, USA). Ten minutes
later, the rats were injected with 15 mg/kg sodium pentobarbital i.p. (Abbott Laboratory, North Chicago, IL, USA).
After the animals were completely anesthetized, the liver
was exposed by making a transverse incision in the abdomen. The left lobe of the liver was exposed by drawing
it through the incision, and pCR3.1-bcl-xs plasmid encapsulated in liposomes (group 1), pCR3.1 encapsulated in
Prevention and inhibition of rat liver tumor by bcl-xs plasmid
M Baba et al
liposomes (group 2), or liposomes only (group 3),
respectively, were injected into it at a dose of 80 ␮g
plasmid/kg body weight in saline (0.5 ml). One minute
later, the site where the plasmid was injected, was sandwiched in a pincette electrode with circular poles 1 cm
in diameter (positive and negative). Eight electrical pulses with a current of 0.5 A, electric-field strength of 70–
100 V/cm, and a pulse duration of 50 ms were delivered
to the left lobe with a BTX 500 optimizer and a BTX T820
standard square wave electroporator (BTX, San Diego,
CA, USA). This procedure did not increase the temperature of the liver by more than 1°C.
At week 11, all rats were killed by deep ether anesthesia. Sections (2 or 3 mm thick) were taken from the
two lesions occurring where the electrode poles had
made contact, and 0.5–1.0 cm away from the site of electrode contact in the left lobe (Figure 1a). The sections
were fixed with cold Zamboni’s solution, embedded in
paraffin, and two serial 3-␮m-thick sections were cut. For
histological analysis, each section from the different
groups was stained with hematoxylin and eosin, and the
number and mean area of foci, neoplastic nodules and
hepatocellular carcinomas in the liver were measured
with ‘Image Hyper’ software (InterQuest, Osaka, Japan).
RT-PCR
Total RNA was extracted with an RNA isolation kit
(Purescript; Gentra System, Minneapolis, MN, USA),
treated with DNase, extracted with phenol/chloroform
(1:1) and precipitated with 1/10 volume of 3 m sodium
acetate and two volumes ethanol. The total RNA was suspended in 35 ␮l of DEPC-treated water containing 10
units RNasin and measured spectrophotometrically. The
total RNA (1 ␮g) from different groups was used for
cDNA synthesis with avian myeloblastosis virus (AMV)
reverse transcriptase XL (RNA PCR kit, ver. 2, Takara,
Shiga, Japan). First-strand synthesis was performed at
55°C for 60 min. PCR was performed in PCR buffer (10
mm Tris-HCl, pH 8.3, 50 mm KCl, 1.5 mm MgCl2 and
0.1% (W/V) gelatin) containing 200 ␮m of each dNTP and
1.25 units of Taq polymerase and the upstream primer
(5⬘ACCGTTCAGCTGGATATTACG3⬘) and the downstream primer (5⬘ CATGATGAACCTGAATCGCC 3⬘).
Forty cycles of denaturation (94°C, 1 min), annealing
(55°C, 2 min), and extention (72°C, 1 min) were conducted, with a final 7-min extension. PCR products (475 bp)
were electrophoresed on a 1.5% agarose gel containing
ethidium bromide. The levels of CAT cDNA were
determined with FLA 2000 Science Imaging System (Fuji
Film, Tokyo, Japan) and normalized relatively by ␤actin levels.
Western blotting analysis
At week 11, the tissues from both sites that was sandwiched between the electrode and sites 0.5–1.0 cm away
from electrode in the left lobe were lysed in a lysis buffer
20 mm Tris HCl buffer (pH 7.4), 1% Triton, 0.1% sodium
deoxycholate, 10 ␮g/ml aprotinin, 10 ␮g/ml leupeptin
and 1 mm phenylmethylsulfonyl fluoride. After removing the cell debris by centrifugation at 14 000 g for 30 min,
the protein concentration was estimated with a Micro
Protein Assay Reagent kit (Pierce, Rockford, IL, USA). A
50 ␮g sample of each protein lysate was resolved in 14%
SDS-polyacrylamide gel and transferred on to a PVDF
menbrane (Millipore, Bedford, MA, USA) by semidry
Western blotting. After preblocking in 5% nonfat skim
milk in Tris-buffered saline Tween (pH 7.6) (T-TBS) at
room temperature for 1 h, the membrane was incubated
with a rabbit anti-rat bcl-x polyclonal antibody (1:150)
(Santa Cruz, Santa Cruz, CA, USA). After washing three
times with T-TBS, the membrane was incubated with horseradish peroxidase-conjugated goat anti-rabbit IgG antiserum (1:1500) (Amersham Pharmacia Biotech, Little
Chalfont, UK) for 1 h. After three washes, immunoblots
were developed with the ECL chemiluminescence system
(Amersham Pharmacia Biotech) to visualize bcl-x protein
bands. The bcl-xs and bcl-xL bands were distingushed by
molecular weight markers. The levels of bcl-xs and bclxL proteins were determined with Eagle Sight Image
Capture and Analysis Soft Ware (Stratagene, La Jolla, CA,
USA) and normalized by ␤-actin levels.
1155
Apoptotic index
Sections were taken from both sites of electrode contact
and sites away from electrode, and fixed with 4% paraformaldehyde, embedded in paraffin, and cut into serial
3-␮m-thick sections. Each serial paraffin section was
dewaxed in xylene, rehydrated in ethanol, and incubated
with 20 ␮g/ml proteinase K in phosphate-buffered saline
(PBS, pH 7.2) at room temperature for 15 min to quench
endogenous peroxidase. The slides were then incubated
with the mixture of terminal transferase and digoxigenin11-UTP and ATP, and the anti-digoxigenin conjugated
peroxidase solution (Apop. Tag in situ apoptosis detection kit: Oncor, Gaithersburg, MD, USA)32 was then
applied to the slide. Color was developed with diaminobenzidine.
Statistics
Results are expressed as mean ± s.e. Standard descriptive
statistics, Student’s t test and Fisher’s exact probability
test, were used. A P value of ⬍0.05 was considered to
indicate a significant difference between groups.
Acknowledgements
We thank Meiwa Shoji Company (Osaka, Japan) for lending a BTX 500 optimizer and a BTX 820 standard sequare
wave electroporator (BTX, San Diego, CA, USA).
References
1 Chao DT, Korsmeyer SJ. Bcl-2 family: regulations of cell death.
Ann Rev Immunol 1998; 16: 395–419.
2 Yang E, Korsmeyer SJ. Molecular thanatopsis: a discourse on
the bcl-2 family and cell death. Blood 1996; 88: 386–401.
3 Desoize B. Anticancer drug resistance and inhibition of
apoptosis. Anticancer Res 1994; 14: 2291–2294.
4 Olopade OJ et al. Overexpression of bcl-x protein in primary
breast cancer is associated with high tumor grade and nodal
metastases. Cancer J Sci Am 1997; 3: 230–237.
5 Paula A et al. Bcl-xL antisense treatment induced apoptosis in
breast carcinoma cells. Int J Cancer 2000; 87: 582–590.
6 Ziegler A et al. Induction of apoptosis in small-cell lung cancer
cells by an antisense oligonucleotide targeting the bcl-2 coding
sequence. J Natl Cancer Inst 1997; 89: 1027–1036.
7 Baba M, Iishi H, Tatsuta M. In vivo electroporetic transfer of bcl2 antisense oligonucleotide inhibits the development of hepatocellular carcinoma in rats. Int J Cancer 2000; 85: 260–266.
8 Fiorentino M et al. High levels of bcl-2 messenger RNA detected
by in situ hybridyzation in human hepatocellular carcinomas.
Dia Mol Pathol 1999; 8: 189–194.
Gene Therapy
Prevention and inhibition of rat liver tumor by bcl-xs plasmid
M Baba et al
1156
9 Reed JC. Double identity for proteins of the bcl-2 family. Nature
1997; 387: 773–776.
10 Boise LH et al. Bcl-x, a bcl-2-related gene that functions as a domant regulator of apoptotic cell death. Cell 1993; 74: 597–608.
11 Fridman JS, Benedict MA, Maybaum J. Bcl-xs-induced cell death
in 3T3 cells does not require or induce caspase activation. Cancer
Res 1999; 59: 5999–6004.
12 Fridman JS et al. Cytchrome c depletion upon expression of bclxs. J Biol Chem 2001; 276: 4205–4210.
13 Clarke MF et al. A recombinant bcl-xs adenovirus selectively
induces apoptosis in cancer cells but not in normal bone marrow
cells. Proc Natl Acad Sci USA 1995; 92: 11024–11028.
14 Sumantran VN et al. A bcl-xs adenovrus selectively induced
apoptosis in transformed cells compared to normal mammary
cells. Neoplasia 2000; 2: 251–260.
15 Abou-Elella A, Gramlich T, Fritsch C, Gansler T. c-Myc amplification in hepatocellular carcinoma predicts unfavorable prognosis. Mol Pathol 1996; 9: 95–98.
16 Pascale RM et al. c-Myc amplification in pre-malignant and
malignant lesions induced in rat liver by the resistant hepatocyte
model. Int J Cancer 1996, 68: 136–142.
17 Macias Rodriguez MA et al. Risk factors for hepatocellular carcinoma in patients with liver cirrhosis. Rev Esp Enferm Dig 2000;
92: 458–469.
18 Hanazaki K et al. Survival and recurrence after hepatic resection
of 386 consecutive patients with hepatocellular carcinoma. J Am
Coll Surg 2000; 191: 381–388.
19 Moore MA, Mayer D, Bannasch P. The dose dependence and
sequential appearance of putative preneoplastic populations
induced in the rat liver by stop experiments with N-nitrosomorpholine. Carcinogenesis 1982; 3: 1429–1436.
Gene Therapy
20 Dagle JM, Andracki ME, Devine RJ, Walder JA. Physical properties of oligonucleotides containing phosphoramidate modified
internucleoside linkages. Nucleic Acids Res 1991; 19: 1805–1810.
21 Kay MA, Liu D, Hoogerbrugge PM. Gene therapy. Proc Natl
Acad Sci USA 1991; 94: 12744–12746.
22 Jiao S et al. Direct gene transfer into nonhuman primate myofibers in vivo. Hum Gene Ther 1992; 3: 21–33.
23 Ledley FD. Nonviral gene therapy: the promise of genes as
pharmaceutical products. Hum Gene Ther 1995; 6: 1129–1144.
24 Lee RJ, Huang L. Lipidic vector systems for gene transfer. Crit
Rev Ther Drug: Carr Syst 1997; 14: 173–206.
25 Niven R, Zhang Y, Smith J. Toward development of a non-viral
gene therapeutic. Adv Drug Del Rev 1997; 26: 135–150.
26 Rols M-P et al. In vivo electrically mediated protein and gene
transfer in murine melanoma. Nat Biotechnol 1998; 16: 168–171.
27 Suzuki T, Shin BC, Fujikura K, Matsuzaki T. Direct gene transfer
into rat liver cells by in vivo electroporation. FEBS Lett 1998; 425:
436–440.
28 Heller L et al. Electrically mediated plasmid DNA delivery to
hepatocellular carcinomas in vivo. Gene Therapy 2000; 7: 826–829.
29 Institute of Laboratory Animal Resources, Histologic typing of
liver tumors of the rat. J Natl Cancer Inst 1980; 64: 177–207.
30 Mir LM et al. High-efficiency gene transfer into skeletal muscle
mediated by electric pulses. Proc Natl Acad Sci USA 1999; 96:
4262–4267.
31 Baer A, Schubeler D, Bode J. Transcriptional properties of genomic transgene integration sites marked by electroporation or
retroviral infection. Biochemistry 2000; 39: 7041–7049.
32 Gavrieli Y, Sheman Y, Ben-Sasson SA. Identification of programmed cell death in situ via specific labeling of nuclear DNA
fragmentation. J Cell Biol 1992; 119: 493–501.