Content uploaded by Dirk van Bekkum
Author content
All content in this area was uploaded by Dirk van Bekkum on Nov 24, 2014
Content may be subject to copyright.
Available via license: CC BY 4.0
Content may be subject to copyright.
BioMed Central
Page 1 of 13
(page number not for citation purposes)
BMC Molecular Biology
BMC Molecular Biology
2002,
3
x
Research article
A comparison of efficacy and toxicity between electroporation and
adenoviral gene transfer
Pierre Lefesvre*, Joline Attema and Dirk van Bekkum
Address: Crucell BV, PO BOX 2048, 2301CA, Leiden, The Netherlands
E-mail: Pierre Lefesvre* - p.lefesvre@crucell.com; Joline Attema - j.attema@crucell.com; Dirk van Bekkum - bekkum@crucell.com
*Corresponding author
Abstract
Background: Electroporation of skeletal muscle after injection of naked DNA was shown by
others to increase transgene expression. Information regarding tissue damage caused by
electroporation is conflicting. It is also not well known how plasmid electroporation compares with
transfection by adenoviral vectors. To investigate these questions the most used protocol for
muscle electroporation was used, i.e. 8 pulses of 200 V/cm and 20 ms at a frequency of 1 Hz.
Results: Intra-muscular DNA transfer of pLuciferase was increased by 2 logs after electroporation,
confirming data described by others. However, the blood levels of the encoded protein were still
lower than those obtained after injection of first generation adenoviral vectors. Also, the
electroporation procedure, on its own, caused severe muscle damage consisting of rhabdomyolysis
and infiltration, whereas the adenoviral vectors caused only a slight infiltration. As damage of
targeted tissue may be an advantage in the case of tumour transfection, we also compared the two
transfection methods in tumour tissue. In case of poorly permissive tumours, adenoviral vectors
cannot transfect more than 2% of the tumour tissue without inducing significant liver damage. In
contrast, the electroporation seems to offer a wider therapeutic window since it does not cause
any systemic toxicity and still induce's significant transfection.
Conclusions: Plasmid electroporation of the muscle induce severe local damage and is of no
advantage over adenoviral vectors for obtaining high blood levels of a vector encoded protein. In
contrast, electroporation of tumours might be safer than adenoviral gene transfer.
Background
Numerous diseases require treatment by systemic delivery
of a therapeutic protein. Repetitive or continuous injec-
tions are the only delivery method used in daily practice.
As a mean of reducing the inconvenience of multiple in-
jections or implantation of mini pumps, gene transfer
technology may offer certain advantages. In order to
achieve a high plasma concentration, whatever the vector
used, the transfection of a large tissue mass is required
(e.g. liver or muscle). In rodents, intravenous injection of
a non-targeted vector – viruses or plasmid preparations –
induces mainly transfection in the liver and the spleen
[1,2]. As of today, transfection of other large organs by vi-
ral vectors has not been accomplished in by way of sys-
temic delivery [3]. The main inconveniences of
intravenous administration of first generation adenoviral
vectors are their immunogenicity [4] and hepatic toxicity
[5,6]. The muscle as a large tissue mass is a candidate for
Published: 13 August 2002
BMC Molecular Biology 2002, 3:12
Received: 24 April 2002
Accepted: 13 August 2002
This article is available from: http://www.biomedcentral.com/1471-2199/3/12
© 2002 Lefesvre et al; licensee BioMed Central Ltd. This article is published in Open Access: verbatim copying and redistribution of this article are permitted
in all media for any non-commercial purpose, provided this notice is preserved along with the article's original URL.
BMC Molecular Biology 2002, 3 http://www.biomedcentral.com/1471-2199/3/12
Page 2 of 13
(page number not for citation purposes)
production of recombinant protein, but the transfection is
limited to the injection site. Intra muscular administra-
tion is an attractive option as it avoids liver damage, but
the levels reached are not satisfactory [7]. An alternative to
transfection with adenoviral vector is the injection of na-
ked DNA into the muscle followed by electroporation.
The electroporation setting that induced high blood levels
of excreted encoded proteins were defined by Bettan et al.
[8] and Mir et al. [9,10]. This setting of 8 pulses of 20 ms
and 200 V/cm at a frequency of 1 Hz increased the plas-
mid expression up to 2 logs in the C57Bl/6 mice. In this
study, first generation adenovirus transfection and plas-
mid electroporation of the muscle were compared to de-
termine which one is the most suitable for obtaining high
plasma levels of two proteins of interest, namely: mhATF-
BPTI and endostatin. mhATF-BPTI is a newly designed
chimeric protein that has angiostatic properties and en-
dostatin is an angiostatic undergoing clinical trials that
was used for comparison. As these two proteins are inter-
fering with tissue healing, the luciferase gene was used to
study the local tissue damage caused by the two transfec-
tion methods.
Another application of electroporation is the transfection
of tumours. Human as well as rodent tumours vary greatly
in expressing the surface receptors for adenovirus. Conse-
quently, the gene transfer with adenoviral vectors is very
low in tumours that poorly express those receptors. Plas-
mid DNA alone is not very efficient transfectant, therefore
the electroporation offers a possibility to transfect tu-
mours resistant to viral infection [11,12]. In this study, the
electroporation of a non-permissive adenocarcinoma of
the lung is compared to the adenoviral vector.
Results
Electroporation enhances plasmid gene expression in mus-
cle
The luciferase activity in C57Bl/6 mice muscle was com-
pared when transfected with plasmid alone or with plas-
mid followed by electroporation at various voltage. As
depicted on Fig 1 electroporation increased transfection to
a maximum of 2 logs which was attained at 180 V/cm.
There was no more enhancing effect when the voltage was
lowered to 100 V/cm. It remained to be determined if the
2 log increase in transgene expression was enough to ob-
tain high blood levels of a secreted transgenic protein.
Therefore, plasmids encoding mhATF-BPTI or human en-
dostatin were injected into the muscle with and without
electroporation and the plasma level of the encoded pro-
tein was monitored. To verify that the injection had been
properly performed, an equal amount of pAdapt Luc was
added to the plasmid preparation and the mixture was in-
jected in a separate group of animals. Both groups (with
and without pLuc), showing similar expression levels of
the secreted encoded protein were pooled. As shown in
Fig 2 the injection of plasmid alone – pATF-BPTI or pEn-
dostatin – did not lead to blood levels above the back-
ground. In contrast, plasmid injection followed by
electroporation resulted in detectable blood levels of
mhATF-BPTI (5 ng / ml) and of hEndostatin (25 ng / ml).
These concentrations are however, far below what one can
achieve with low dose adenovirus (10
9
iu) injected intra-
venously which results in protein production by the liver.
Plasmid electroporation does not compete with adenoviral
vectors
The volume of injection limits the quantity of adenovirus
or plasmid that can be injected into the gastrocnemius of
a mouse without overflow. The maximal volume of 50
µL
and the highest concentrations of the available batches of
either recombinant adenovirus (10
10
iu / 50 µl) or plas-
mid (50
µg / 50 µl) were employed to compare the two
gene transfer methods. The following electroporation
conditions i.e. 200 V/cm, 8 pulses of 20 ms at a frequency
of 1 Hz, were employed. The luciferase activity was meas-
ured in the gastrocnemius muscle of C57BL/6 mice 2 days
after the gene delivery. As depicted on figure 3 transgene
expression was roughly the same with adenoviral vector as
with plasmid electroporation. The leakage of adenovirus
out of the muscle induces a significant luciferase activity
in the liver and in the spleen. This level corresponds to an
intravenous virus dose that is far below that found to be
toxic for the liver. In contrast, the electroporation of the
muscle was not accompanied by any transgene expression
in the liver but a few mice showed a very low luciferase ac-
tivity in the spleen. It is noteworthy that the background
of luciferase is higher in the muscle than in the liver. This
was always observed and considered to be inherent to the
detection method. To further investigate the secretion ca-
pacity of the muscle, 10
10
iu Ad Adapt mhAB were injected
in the gastrocnemius of mice. The plasma levels of mhAB,
monitored for 3 weeks after the injection, showed a peak
at 10
± 4 ng /ml on day 10 after administration. In view of
the variation, transduction of the muscle with first gener-
ation Ad 5 is not significantly more efficient that with
electroporation.
Local injury caused by muscle electroporation
The damage induced by the electroporation was assessed
by clinical observation, histological examination of mus-
cle sections, and determination of creatine kinase in the
plasma. Clinical symptoms were not observed in any of
the mice during the observation period of 2 weeks. At au-
topsy, practiced 2, 7 and 14 days after the electroporation,
macroscopic examination of the treated areas showed lo-
cal abnormalities and an oedema of the whole gastrocne-
mius muscle. On day 2 the muscles were just a bit pale
whereas at day 7 and 14 the site of electroporation was
white / creamy and felt hard upon palpation (fig 4). Mi-
croscopic examination revealed severe muscle necrosis
BMC Molecular Biology 2002, 3 http://www.biomedcentral.com/1471-2199/3/12
Page 3 of 13
(page number not for citation purposes)
(fig 5) that was accompanied by a severe polynuclear eosi-
nophilic and mast cell infiltration. The quantification of
the pathological changes was assessed by determining the
percentage of necrotic fibers on transversal sections of the
muscle at the site of electroporation as well as by measur-
ing the creatine kinase (CK) concentration in the blood.
Histological scoring confirmed that the muscle damage
was maximal at 7 days after the electroporation (Fig 6).
Since the creatine kinase levels are not decreasing within
the first week after electroporation, the muscle lesions
must have continued over several days. In contrast, after
an acute muscle injury like an ischemic infarction, the
peak of CK occurs within few hours. This first peak may
have occurred after the electroporation but was not mon-
itored. Thus the elevation of the CK between day 7 and 14,
probably reflect the slow necrosis of the muscle fiber oc-
curring during the first week. More importantly, the elec-
tric pulse delivery caused on its own a severe
rhabdomyolysis and the prior injection of saline signifi-
cantly increased the muscle damage, but the addition of
naked DNA did not influence the muscle injury.
Other investigators showed that protein expression from
injected DNA was correlated with the voltage intensity up
to 250 v/cm. Thus, to determine a therapeutic window for
the electroporation procedure, the voltage intensity was
gradually diminished and the corresponding muscle dam-
age assessed as previously (Fig 7). The degree of necrosis
was indeed correlated with the voltage intensity and de-
creased to background at 50–100 V /cm. Unfortunately
the luciferase expression levels decreased as fast as the
Figure 1
Enhancement of plasmid expression in muscle by electroporation as a function of voltage. Luciferase expression in gastrocne-
mius muscle of C57BL/6 mice after injection of 50
µg pAdapt Luc followed by electroporation (black bars), injection of 50 µg
pAdapt Luc (striped bar), and injection of saline (white bar). The luciferase activity was determined 2 days after the treatment.
The values show the mean of three separate experiments. Data are expressed as means
± SD.
BMC Molecular Biology 2002, 3 http://www.biomedcentral.com/1471-2199/3/12
Page 4 of 13
(page number not for citation purposes)
Figure 2
Blood concentrations achieved by muscle electroporation. Endostatin and mhATF-BPTI plasma levels (red lines) and muscle
luciferase activity (black lines) in C57BL/6 mice. The animals were sacrificed 4, 7, and 14 days after treatment to determine the
luciferase activity in the injected muscle and plasma levels of endostatin or ATF-BPTI. Five animals per group and data are
expressed as means
± SD. A: the mice received 50 µg pAdapt hEndo +/- 50 µg pAdapt Luc intramuscularly followed by no elec-
troporation (circles) or by a 200 V/cm pulse delivery (squares). B: the mice received pAdapt mhATF-BPTI +/- 50
µg pAdapt
Luc intramuscularly followed by no electroporation (circles) or by a 200 V/cm pulse delivery (squares). The plasma levels of
mhATF-BPTI injected without electroporation were below the detection limit. The injection of two plasmids together did not
influence the level of expression of either the luciferase plasmid or that of the plasmid encoding mhATF-BPTI or endostatin.
BMC Molecular Biology 2002, 3 http://www.biomedcentral.com/1471-2199/3/12
Page 5 of 13
(page number not for citation purposes)
muscle damage, and no therapeutic window could be
found with these settings.
Intra-tumoral electroporation as compared to transfection
with adenovirus
The permissiveness to adenoviral vectors of rodent as well
as of human tumours varies within a range of 3 logs. One
of the reasons is that viral vectors are dependant on mem-
brane receptors to enter the cells. Adenoviral transfection
can be dramatically impaired in tumours that lack the ap-
propriate receptors. Therefore, transfection of naked DNA
using electroporation might offer a possibility to transfect
adenoviral-resistant tumours. The L44 bronchial adeno-
carcinoma was selected from among ten rat tumours of
different histological origin, as being very poorly permis-
sive to adenovirus. The standard plasmid electroporation
was compared to adenoviral gene delivery in subcutane-
ously growing tumours. As depicted on fig 8 the electro-
poration of 50
µg pAdapt Luc and the injection of 5.10
9
iu
Ad Adapt Luc induce roughly similar luciferase expres-
sion. However, a non-negligible proportion of adenovi-
ruses leaks out of the tumour and reaches the liver. Such
leakage is 600 times higher with intra tumoral delivery
than after intra-muscular injection as judged by the luci-
ferase activity measured in the liver. In contrast, there is
no leakage of the luciferase activity from the tumour after
plasmid electroporation. In order to assess the toxicity as-
sociated with the viral leakage to the liver, graded doses of
luciferase and empty vectors were injected iv in Brown
Norway rats. The intravenous route was chosen on basis
of its consistent reproducibility. The luciferase activity in
the liver was measured and pathological changes were
scored in liver sections. To avoid interference by the trans-
gene, the liver damage were determined in animals inject-
ed with an empty vector. Figure 9 shows that the leakage
associated with the intra-tumoral injection of 5.10
9
iu and
10
10
iu is in the range of vector doses that induce a mild to
moderate liver damage. In an attempt to assess the thera-
peutic window of the adenoviral vector in this poorly per-
missive tumour, the number of transduced cells in the
tumour was determined by injection of vectors encoding
the
β-galactosidase gene (Table 1). It is assumed that no
significant effect is expected from a therapeutic transgene
(e.g thymidin kinase, mhATF-BPTI or endostatin) if the
percentage of transduced tumour tissue is below 5%. The
intra tumoral injection of 10
10
iu adenovirus resulted in
the transduction of only 1–2% of the tumour cells, mostly
limited around the needle track (Table 1). This vector dose
of 10
10
iu is already toxic for the liver (Fig. 9). Thus the
margin between efficacy and toxicity of the adenovirus in
non-permissive tumour is not wide enough for therapeu-
tic applications.
Figure 3
Adenoviral and electroporation gene transfer into the mus-
cle. Luciferase expression in gastrocnemius muscle, liver and
spleen of C57BL/6 mice after intra-muscular injection of
10
10
iu Ad Adapt Luc (grey bars, n = 5), 50 µg pAdapt Luc fol-
lowed by a 200 V/cm electroporation (black bars, n = 6),
10
10
iu Ad Adapt Empty (gray striped bars, n = 3), and 50 µg
pAdapt empty followed by a 200 V/cm electroporation (black
striped bars, n = 4). The luciferase activity was determined 2
days after injection. Data are expressed as means
± SD.
Table 1: Percentage of transduced cells in the L44 tumour
Dose Number of tumours Percentage of blue cells
10
10
iu 3 2 ± 0.5
2.5 10
9
iu 3 0.5 ± 0.2
5 10
8
iu 3 < 0.01
Percentage of cells expressing β-Galactosidase after intra-tumoral injection of Ad5 Adapt LacZ. The percentage of stained cells was determined on
25 × magnification field of a tumour section. Four sections per tumour. Data are expressed as mean ± SD.
BMC Molecular Biology 2002, 3 http://www.biomedcentral.com/1471-2199/3/12
Page 6 of 13
(page number not for citation purposes)
Discussion
Several investigators have used electroporation proce-
dures to increase the gene expression of naked DNA in-
jected in rodents muscles or in tumours [13,14]. There are
basically two types of muscle electroporation protocols:
(1) high voltage / short pulses and (2) low voltage / long
pulses. The highest plasma levels of an encoded protein
were obtained with low voltage (200 V /cm) long pulses
(20 ms) and 8 pulses at a frequency of 1 Hz, i.e. the blood
concentration of human secreted alkaline phosphatase
reached values around 2
µg / ml [8]. The plasmid prepara-
tions described by others are most of the time endotoxin
free and the transgene is under a CMV promoter without
any nuclear transport signal. Using these settings and the
BTX electroporation device, the transgene expression after
intra muscular plasmid injection could be enhanced by a
factor 100. This led to plasma serum levels of endostatin
around 10 – 50 ng /ml. The dose of plasmid per injected
muscle could not be easily increased for technical reasons.
Given the maximal achievable plasmid concentration, the
injection volume would have become too large for the
muscle. Bettan et al. have shown that increasing the dose
from 30
µg to 300 µg by multiple injections resulted in 7
to 10 fold higher plasma levels. Thus, the intra muscular
injection of naked DNA followed by electroporation is
considered by some as a means to obtain therapeutic plas-
ma levels of proteins. However this range of plasma con-
centration can also be achieved by intramuscular or
systemic administration of first generation Ad5 vectors.
The toxicity of these vectors is well documented [6,5]. Sys-
Figure 4
Macroscopic change of the muscle after electroporation. The
gastrocnemius muscle of two C57BL/6 mice, 7 days after
injection of 50
µg pAdapt Empty followed by a 200 V/cm
electric pulse delivery. The arrows indicate the whitish part
of the muscle corresponding to fibre necrosis and inflamma-
tory infiltrate seen under magnification.
Figure 5
Pathological changes of the muscle after electroporation. The
photographs show representative histological sections of the
gastrocnemius of C57BL/6 mice. Magnification
×100. A1-3:
different sections at 7 days after plasmid electroporation
(200 v / cm). The sections were performed in the middle of
the white lesion seen macroscopically. They reveal a severe
rhabdomyolysis and infiltration by mast cells, eosinophils and
mononuclear cells. At the periphery of the lesion, the dam-
ages were less severe, the myocytes were not completely
necrotic but were pale. In the part of the muscle that was
macroscopically normal, there was only a diffuse mild to
moderate inflammatory infiltrate. The black arrows indicate
necrosis and white arrows indicate inflammatory infiltrate. B:
muscle section of a healthy mouse.
BMC Molecular Biology 2002, 3 http://www.biomedcentral.com/1471-2199/3/12
Page 7 of 13
(page number not for citation purposes)
temic delivery of adenovirus vectors induces mild liver
damage at low dose (10
9
iu in mice) and severe hepatitis
and bone marrow necrosis at high dose (10
11
iu in a
mice). Intramuscular injection of 10
10
iu induces moder-
ate local inflammation [15]. However, the local and sys-
temic toxicity of the adenoviruses can be dramatically
reduced by using gutless viruses [16]. Notably, we ob-
served severe muscle damage after the electric pulse deliv-
ery, which was worsened by the pre injection of saline but
was not influenced by the presence of plasmid. Most au-
thors dealing with electroporation gene transfer did not
investigate the toxicity of electroporation. All differences
between those published protocols and our parameters
were tested to verify if they were responsible for the toxic-
ity observed in our hands, but that was not the case. Sim-
ilar damage occurred after using a simple anaesthesia
(hypnorm) or in combination with muscle relaxants dur-
ing electroporation. Different positions of the leg between
the electrodes were tested with no influence on the toxic-
ity. No difference in muscle damage was observed be-
tween male and female mice. The influence of the type of
electrodes-needle or calliper – was not tested in our exper-
iments. However, there are indications from previous
publications that both electrodes would induce similar
muscle damage. Indeed, Gehl et al. established that the
toxicity of electroporation was correlated with the degree
of cell permeabilisation [17] and that these was similar
when using needles or calliper [18]. Thus, it may be as-
sumed that the electrode type is not of crucial influence
on the toxicity induced by the electroporation. Therefore
we conclude that the rhabdomyolysis developed in those
mice was not influenced by other parameters than the
electric pulse settings and the pre-injection of buffer. Only
two investigators have reported muscle damage after plas-
mid electroporation. Mathiesen described muscle necrosis
after plasmid electroporation that increased with the cu-
mulative duration of the pulses and that the necrotic fib-
ers never expressed the reporter gene. Hartikka et al.
attributed the muscle necrosis to the presence of plasmid,
but this conclusion is difficult to interpret since the eleva-
tion of serum CPK was not influenced by the plasmid. The
lesions described by these authors and us have much in
common with those induced by electrocution chocks. For
instance, Block et al. [19], in an attempt to mimic non-
thermally mediated muscle injury in electrical trauma,
used similar electric field strength and pulse length as for
plasmid electroporation. This resulted in quantitatively
and qualitatively very similar muscle damage to what we
found with the electroporation procedure. Thus, there is
strong evidence that muscle electroporation, as it usually
performed, is severely damaging and reduces plasmid ex-
pression to the survival cells. Furthermore, even if electri-
cal injuries generally involve much higher field strength
[20], it is noteworthy that some pathologic changes found
after electric shock resemble the one observed in muscle
after electroporation i.e. the decolouration in bands of the
muscle fibbers described by Morita et al. [21].
Bureau et al. have made attempts to improve the electro-
poration protocol by introducing new settings, consisting
in the combination of high voltage and low voltages puls-
es, that are inducing less permeabilisation of the myocytes
[22]. Even if these combined protocols seems to be slight-
ly less toxic, there is still much improvement needed to
compete with the new generations of adenoviral vectors as
a method to transfect muscle in patients.
In contrast to intramuscular delivery, the electroporation
of tumours seems to offer a wider therapeutic window. In
case of tumours that are poorly permissive to adenovirus,
it is difficult to transduce a large number of tumour cells
with adenoviral vectors without inducing severe liver
damage. In contrast, the systemic toxicity is not a risk with
plasmid electroporation as it is with adenoviral transfec-
tion. The possible damage induced by the electroporation
could not be easily determined due to the high back-
ground of necrosis in the tumour tissue, but if it occurs it
could only add to the therapeutic effect. In our experi-
ments the concentration of plasmid preparations limited
the maximal dose that could be delivered into the tumour,
but If this technical obstacle of producing highly concen-
trated plasmid was solved, electroporation would offer an
applicable transfection technique for viral-resistant tu-
mours that are accessible with an electroporation device
i.e. head and neck or skin cancers.
Methods
Animals
Pathogen-free inbred male C57Bl/6 mice, weighing 20 to
30 gr and Brown Norway rats, weighing 300 to 350 gr
were purchased from Harlan, The Netherlands. All ani-
mals were fed ad libitum with laboratory chow and water
and were kept under standard laboratory conditions. For
assay of plasma creatin phosphokinase (CPK), hEndosta-
tin, and mhATF-BPTI, mice were anaesthetised with iso-
flurane, bled by tail vein cut and the blood was collected
in EDTA tubes. All animal procedures were performed in
accordance with the official guidelines after obtaining per-
mission of the animal welfare committee. Measurements
of CPK were performed according to standard clinical pro-
cedures.
Plasmid preparation
The plasmid pAdapt Luc, pAdpat LacZ, pAdapt hEndo and
pAdapt mhAB encoding respectively for the luciferase,
β-
galactosidase, human endostatine and murinised human
ATF-BPTI genes were constructed as described previously
[23]. The different genes are under the control of the cy-
tomegalovirus (CMV) immediate-early promotor and ter-
minated by the simian virus (SV40) late poly(A) signal.
BMC Molecular Biology 2002, 3 http://www.biomedcentral.com/1471-2199/3/12
Page 8 of 13
(page number not for citation purposes)
Figure 6
Quantification of muscle damage over time. The gastrocnemius muscle was injected with 50
µL (1 µg / µL) of pAdapt Empty
followed by a 200 V/cm electric pulse delivery (black bars), 50
µL of PBS and 200 v/cm electroporation (striped bars), electro-
poration alone (white bars), 50
µL pAdapt Empty (red bars), 50 µL of PBS (blue bars), or were not treated (grey bars). The
muscle and blood were collected 2, 7, and 14 days after treatment. Four animals per group. Data are expressed as means
± SD.
A: Creatine kinase plasma levels. B: percentage of necrotic fibers determined in the electroporated part of the muscle. Statisti-
cal analysis (Scheffe's test) showed a significant difference of the muscle damage at day 2, 7 and 14 between the mice treated by
electroporation without pre-injection and those pre-injected with PBS or plasmid and then treated with electroporation.
There is not a significantly different damage between the muscle pre injected with PBS and the muscle pre injected with plas-
mid prior to the electroporation. The p values between the pre injected groups (with PBS or with plasmid) and the "electropo-
ration only" group are indicated above the bars.
BMC Molecular Biology 2002, 3 http://www.biomedcentral.com/1471-2199/3/12
Page 9 of 13
(page number not for citation purposes)
Figure 7
Quantification of muscle damage by electroporation using different voltage. The gastrocnemius muscle was injected with 50
µL
(1
µg / µL) of pAdapt Empty (black bars), or 50 µL of PBS (striped bars) followed by the electric pulse delivery. The control
mice received only the electric pulse delivery (white bars). The muscle and blood were collected 2 days after treatment. Four
animals per group. Data are expressed as means
± SD. A: plasma levels of the Creatine kinase. B: percentage of necrotic fibers
determined in the electroporated part of the muscle. Statistical analysis (Scheffe's test) showed a significant difference of the
muscle damage at the highest voltage (200 and 250 V/cm) between the mice treated by electroporation without pre-injection
and those pre-injected with PBS or plasmid and then treated with electroporation. There is not a significantly different damage
between the muscle pre injected with PBS and the muscle pre injected with plasmid prior to the electroporation. The p values
between the pre injected groups (with PBS or with plasmid) and the "electroporation only" group are indicated above the bars.
BMC Molecular Biology 2002, 3 http://www.biomedcentral.com/1471-2199/3/12
Page 10 of 13
(page number not for citation purposes)
The p Adapt mhAB encodes for a murinised form of the
human ATF-BPTI (mhAB) [24]. In this construct the
mhATF-BPTI, is preceded by the native secretion signal
peptide of the human urokinase. The human endostatin
coding sequence (InvivoGen, CA, USA) was cloned in the
Ad Adapt shuttle vector. The encoded endostatin corre-
sponds to the 183 residue of the human endostatin de-
scribed by O'Reilly et al. (1997) with an intact N-terminus
(HSHRDFQ...), preceded by the secretion signal peptide
of the human IL-2. The p Adapt empty is identical to p
Adapt Luc except that it does not encode any transgene.
All plasmid DNA preparations were prepared using Qia-
gen Mega Kits and purified using ENDOfree kits (Qiagen,
Valancia, CA).
Adenoviral vectors
Recombinant adenovirus vectors were generated in
PER.C6™ cells by homologous recombination between an
adapter plasmid (pAdapt) and the E1 deleted Ad 5 DNA
plasmid as described elsewhere [23]. As a result of the ab-
sence of sequence overlap between the Adapt plasmid and
the Ad5 E1 sequences integrated into the genome of
PER.C6, the vector stocks used in this study did not con-
tain replicative competent adenovirus (RCA) [25]. All vec-
tors were produced on PER.C6™ using standard
procedures [25]. Infectious units (iu)/ml were determined
by end point cytopathogenic effect (CPE) assay on 911
cells [26]. Viral particles were determined by HPLC [27].
The particle to infectious unit ratio was always lower than
5.
Electric pulse delivery and intramuscular injection
The animals were anaesthetised by intra peritoneal injec-
tion of fentanyl / fluanisone (Hypnorm, Janssen Animal
Health, The Netherlands). Rear legs and the flank skin
were shaved for the muscle and tumoral electroporation
respectively. Fifty microliter of plasmid (50
µg) or of ade-
Figure 8
Gene transfer in the L44 tumour and leakage to the liver. Luciferase activity in the L44 tumour and in the liver of Brown Nor-
way rats after intra tumoral injection of 10
10
iu Ad Adapt Luc (black bars, n = 4), 5.10
9
iu Ad Adapt Luc (grey bars, n = 4), 50 µg
pAdapt Luc followed by a 200 V/cm electroporation (striped bars, n = 4), or 50
µg pAdapt Luc (white bars, n = 3). The luci-
ferase activity was determined 2 days after the treatment. Data are expressed as means
± SD.
BMC Molecular Biology 2002, 3 http://www.biomedcentral.com/1471-2199/3/12
Page 11 of 13
(page number not for citation purposes)
novirus (10
10
iu) were injected slowly to prevent back
flow in the gastrocnemius with a 29 Gauge needle. When
2 plasmids were injected simultaneously, the volume of
the mixture was 100
µl. The tumours had a volume of 500
cm3 and the injection volume was 100
µl. Two minutes
after plasmid injection, transcutaneous electric pulses
were applied through two stainless steel plate electrodes
of a dimension of 5
× 10 mm (BTX, calliper electrodes;
Westburg, The Netherlands) placed on each side of the leg
or the tumour. Electrode jelly was used on the electrode
plates to ensure good electrical contact. The distance be-
tween the electrodes were usually around 5–6 mm. The
electrodes were applied on the muscle until they came in
complete contact with the skin, but without compressing
the muscle. Eight 20-ms pulses were delivered at a fre-
quency of 1 Hz using a BTX ECM 830 electroporator. Var-
iation between 50 and 200 V /cm were investigated in
different experiments.
Luciferase activity assay in tissue
Mice were sacrificed by an overdose of isoflurane and
whole organs were dissected out, frozen in liquid nitrogen
and stored at -80
°C. Organs were homogenised in phos-
phate buffered saline pH 7.8 using a blender. To lyse the
Figure 9
Liver damage in relation with the dose of adenoviral vectors. On the left Y axis is plotted the luciferase activity measured in the
liver of Brown Norway rats 2 days after iv injection of graded doses of Ad Adapt Luc (squares). On the right Y axis is plotted
the liver damage score observed 7 days after iv injection of graded doses of Ad Adapt Empty (circles). The vp / iu ratio of both
vectors were similar. Therefore, for a given luciferase activity in the liver, one can read on the graph the corresponding liver
damage (the bleu and red lines). The values A (bleu) and B (red) are the luciferase activity in the liver after the intratumoral
injection of 10
10
iu and 5.10
9
iu Ad adapt Luc respectively (see figure 8). The value A and B correspond to intravenous doses of
3.5 10
9
iu and 1.2 10
9
iu respectively. These doses induce a liver damage score between 2 and 4 (circle line).
BMC Molecular Biology 2002, 3 http://www.biomedcentral.com/1471-2199/3/12
Page 12 of 13
(page number not for citation purposes)
cells, DTT (SIGMA, The Netherlands) (1 mM) and Triton
x-100 (0.1%) (Merck, The Netherlands) were added. After
centrifugation at 10,000 rpm for 10 min, 20
µl of the su-
pernatant was added to 100
µl of luciferase assay substrate
(Promega, The Netherlands). Relative light units (RLU)
were determined for 30 s using a luminometer (Lumat
951, Wallac, Belgium). The amount of protein in the ex-
tracts was determined with a commercial kit (Bio-Rad lab-
oratories, The Netherlands) based on the Coomasie
brilliant blue G250 binding assay [28]. The level of luci-
ferase activity in the tissue homogenates was expressed in
RLU / mg protein.
Elisa assays
The mhATF-BPTI ELISA was developed by P. Quax et al.
(TNO-Prevention and Health, Leiden The Netherlands),
using a monoclonal antibody specific for the ATF as the
capture antibody and a polyclonal antibody directed
against BPTI as the detector antibody [24]. As a standard
we used either urokinase or mhATF-BPTI. The levels of hu-
man endostatin in mice plasma were determined with an
ELISA kit (InvivoGen, CA USA) according to the manufac-
turer procedure.
Cell line
The L44 adenocarcinoma was induced in Brown Norway
rats by local irradiation of the thorax [29]. The L44 carci-
noma was serially passaged on syngeneic rats. For the ex-
periments, the tumours were established in the flank of
rats by implanting small pieces (3
× 3 × 3 mm) of tumour
tissue sub-cutaneously as described before [30].
LacZ expression assay
Forty-eight hours after Ad5 Adapt LacZ administration
rats were sacrificed and tumours were removed and cut in
2 mm sections. Sections were fixed in 10% phosphate
buffered formalin (pH 7.0) for 60 min at room tempera-
ture and incubated overnight in 0.5 M sucrose. The sam-
ples were subsequently frozen in liquid nitrogen. Ten
µm
thick frozen sections were prepared and stained with 5-
bromo-4-chloro-3-indolyl-
β-galactopyranoside (X-gal)
solution (Molecular Probes, The Netherlands) overnight
at 37
°C. Finally, sections were counterstained with hae-
matoxylin and eosin. The percentage of transduced cells
in an histological section was assessed by the surface of
stained cells divided by the whole section surface. Cells
were considered positive when a blue staining was seen in
the nucleus. The surface of stained cells in each section
was determined on digitalised photographs of 25
× mag-
nification fields. The surface measurements were per-
formed with the NIH Image 1.62 software.
Pathology
The microscopic examination of the gastrocnemius was
performed on paraffin section of formalin-fixated muscle.
Ten transversal sections form one extremity to the other
were performed and stained with haematoxylin and
eosin. All sections were examined but the quantification
of the rhabdomyolysis was determined on 3 sections, 1
mm apart from each other, cut through the part of the
muscle that was electroporated. The percentage of necrotic
fibers was quantified on photographs of 25
× magnifica-
tion field. The surface of the necrotic areas was measured
with the NHI image analyser software and reported to the
total surface of the whole sections.
The scoring of the liver damage was performed in Brown
Norway rats 7 days after the intravenous injection of
10
10
iu Ad Adapt Empty. The liver pieces were fixed in for-
malin 10%, embedded in paraffin and sections of 10
µm
were stained with haematoxylin and eosin. The total dam-
age score is a compilation of scores of apoptosis, vacuolar
changes, nuclear condensation, anisonucleosis, megalo-
cytosis, mitosis, and inflammation. Normal control liver
score is zero and the maximal is nine.
List of abbreviations
ATF: Amino Terminal Fragment
BPTI: Bovine Pancreatic Trypsine Inhibitor
ELISA: human enzyme-linked immunoabsorbent assay
iu: Infectious units
iv: Intra venous
mhATF-BPTI: murinised human ATF-BPTI
vp: virus particle
Acknowledgment
We thank Dr Chris Zurcher for the scoring of the liver damage and Ger-
maine Penders for her technical support.
References
1. Alemany R, Curiel DT: CAR-binding ablation does not change
biodistribution and toxicity of adenoviral vectors. Gene Ther
2001, 8:1347-1353
2. Paielli DL, Wing MS, Rogulski KR, Gilbert JD, Kolozsvary A, Kim JH,
Hughes J, Schnell M, Thompson T, Freytag SO: Evaluation of the bi-
odistribution, persistence, toxicity, and potential of germ-
line transmission of a replication-competent human adeno-
virus following intraprostatic administration in the mouse.
Mol Ther 2000, 1:263-274
3. Leissner P, Legrand V, Schlesinger Y, Hadji DA, van Raaij M, Cusack
S, Pavirani A, Mehtali M: Influence of adenoviral fiber mutations
on viral encapsidation, infectivity and in vivo tropism. Gene
Ther 2001, 8:49-57
4. Vos A, Neubert A, Pommerening E, Muller T, Dohner L, Neubert L,
Hughes K: Immunogenicity of an E1-deleted recombinant hu-
man adenovirus against rabies by different routes of admin-
istration. J Gen Virol 2001, 82:2191-2197
5. Christ M, Louis B, Stoeckel F, Dieterle A, Grave L, Dreyer D, Kintz J,
Ali Hadji D, Lusky M, Mehtali M: Modulation of the inflammatory
properties and hepatotoxicity of recombinant adenovirus
BMC Molecular Biology 2002, 3 http://www.biomedcentral.com/1471-2199/3/12
Page 13 of 13
(page number not for citation purposes)
vectors by the viral E4 gene products. Hum Gene Ther 2000,
11:415-427
6. Ji L, Bouvet M, Price RE, Roth JA, Fang B: Reduced toxicity, atten-
uated immunogenicity and efficient mediation of human p53
gene expression in vivo by an adenovirus vector with deleted
E1-E3 and inactivated E4 by GAL4-TATA promoter replace-
ment. Gene Ther 1999, 6:393-402
7. Hauser MA, Robinson A, Hartigan-O'Connor D, Williams-Gregory
DA, Buskin JN, Apone S, Kirk CJ, Hardy S, Hauschka SD, Chamberlain
JS: Analysis of muscle creatine kinase regulatory elements in
recombinant adenoviral vectors. Mol Ther 2000, 2:16-25
8. Bettan M, Emmanuel F, Darteil R, Caillaud JM, Soubrier F, Delaere P,
Branelec D, Mahfoudi A, Duverger N, Scherman D: High-level pro-
tein secretion into blood circulation after electric pulse-me-
diated gene transfer into skeletal muscle. Mol Ther 2000, 2:204-
210
9. Mir LM, Bureau MF, Gehl J, Rangara R, Rouy D, Caillaud JM, Delaere
P, Branellec D, Schwartz B, Scherman D: High-efficiency gene
transfer into skeletal muscle mediated by electric pulses. Proc
Natl Acad Sci U S A 1999, 96:4262-4267
10. Mir LM, Bureau MF, Rangara R, Schwartz B, Scherman D: Long-
term, high level in vivo gene expression after electric pulse-
mediated gene transfer into skeletal muscle. C R Acad Sci III
1998, 321:893-899
11. Lohr F, Lo DY, Zaharoff DA, Hu K, Zhang X, Li Y, Zhao Y, Dewhirst
MW, Yuan F, Li CY: Effective tumor therapy with plasmid-en-
coded cytokines combined with in vivo electroporation. Can-
cer Res 2001, 61:3281-3284
12. Wells JM, Li LH, Sen A, Jahreis GP, Hui SW: Electroporation-en-
hanced gene delivery in mammary tumors. Gene Ther 2000,
7:541-547
13. Smith LC, Nordstrom JL: Advances in plasmid gene delivery and
expression in skeletal muscle. Curr Opin Mol Ther 2000, 2:150-154
14. Tamura T, Nishi T, Goto T, Takeshima H, Dev SB, Ushio Y, Sakata T:
Intratumoral delivery of interleukin 12 expression plasmids
with in vivo electroporation is effective for colon and renal
cancer. Hum Gene Ther 2001, 12:1265-1276
15. Yang Y, Haecker SE, Su Q, Wilson JM: Immunology of gene ther-
apy with adenoviral vectors in mouse skeletal muscle. Hum
Mol Genet 1996, 5:1703-1712
16. Adam E, Nasz I: [In Process Citation]. Orv Hetil 2001, 142:2061-
2070
17. Gehl J, Mir LM: Determination of optimal parameters for in
vivo gene transfer by electroporation, using a rapid in vivo
test for cell permeabilization. Biochem Biophys Res Commun 1999,
261:377-380
18. Gehl J, Sorensen TH, Nielsen K, Raskmark P, Nielsen SL, Skovsgaard
T, Mir LM: In vivo electroporation of skeletal muscle: thresh-
old, efficacy and relation to electric field distribution. Biochim
Biophys Acta 1999, 1428:233-240
19. Block TA, Aarsvold JN, Matthews KL 2nd, Mintzer RA, River LP,
Capelli-Schellpfeffer M, Wollmann RL, Tripathi S, Chen CT, Lee RC:
The 1995 Lindberg Award. Nonthermally mediated muscle
injury and necrosis in electrical trauma. J Burn Care Rehabil 1995,
16:581-588
20. Lee RC, Zhang D, Hannig J: Biophysical injury mechanisms in
electrical shock trauma. Annu Rev Biomed Eng 2000, 2:477-509
21. Morita M, Kaneko M, Onodera K, Mizuguchi M, Ohtani S: [A histo-
logical study of striated muscle and nerve fiber in the human
caused by a high voltage electric current]. Nippon Hoigaku
Zasshi 1989, 43:40-45
22. Bureau MF, Gehl J, Deleuze VV, Mir LM, Scherman D: Importance
of association between permeabilization and electrophoret-
ic forces for intramuscular DNA electrotransfer. Biochim Bio-
phys Acta 2000, 1474:353-359
23. Esandi MC, Verlinden SFF, Vries de MW, Pfeffer G, Bekkum van DW:
Neutralizing antibodies against Adeno 5 virus open the ther-
apeutic window for interleukin-3 gene therapy of tumors
with adenovirus 5 based vectors. submitted to gene therapy 2001
24. Lamfers ML, Lardenoye JH, de Vries MR, Aalders MC, Engelse MA,
Grimbergen JM, van Hinsbergh VW, Quax PH: In vivo suppression
of restenosis in balloon-injured rat carotid artery by adeno-
virus-mediated gene transfer of the cell surface-directed
plasmin inhibitor ATF.BPTI. Gene Ther 2001, 8:534-541
25. Fallaux FJ, Bout A, van der Velde I, van den Wollenberg DJ, Hehir KM,
Keegan J, Auger C, Cramer SJ, van Ormondt H, van der Eb AJ, Valerio
D, Hoeben RC: New helper cells and matched early region 1-
deleted adenovirus vectors prevent generation of replica-
tion-competent adenoviruses. Hum Gene Ther 1998, 9:1909-1917
26. Fallaux FJ, Kranenburg O, Cramer SJ, Houweling A, Van Ormondt H,
Hoeben RC, Van Der Eb AJ: Characterization of 911: a new
helper cell line for the titration and propagation of early re-
gion 1-deleted adenoviral vectors. Hum Gene Ther 1996, 7:215-
222
27. Shabram PW, Giroux DD, Goudreau AM, Gregory RJ, Horn MT,
Huyghe BG, Liu X, Nunnally MH, Sugarman BJ, Sutjipto S: Analytical
anion-exchange HPLC of recombinant type-5 adenoviral
particles. Hum Gene Ther 1997, 8:453-465
28. Bradford MM: A rapid and sensitive method for the quantita-
tion of microgram quantities of protein utilizing the princi-
ple of protein-dye binding. Anal Biochem 1976, 72:248-254
29. Kal HB, Meijnders PJ, Van Berkel AH, Van Bekkum DW: Response
to chemotherapy of non-small cell bronchial rat tumours
growing subcutaneously or in the lung. In Vivo 1991, 5:301-306
30. Esandi MC, van Someren GD, Bout A, Mulder AH, van Bekkum DW,
Valerio D, Noteboom JL: IL-1/IL-3 gene therapy of non-small
cell lung cancer (NSCLC) in rats using 'cracked' adenopro-
ducer cells. Gene Ther 1998, 5:778-788
Publish with BioMed Central and every
scientist can read your work free of charge
"BioMedcentral will be the most significant development for
disseminating the results of biomedical research in our lifetime."
Paul Nurse, Director-General, Imperial Cancer Research Fund
Publish with BMC and your research papers will be:
available free of charge to the entire biomedical community
peer reviewed and published immediately upon acceptance
cited in PubMed and archived on PubMed Central
yours - you keep the copyright
editorial@biomedcentral.com
Submit your manuscript here:
http://www.biomedcentral.com/manuscript/
BioMedcentral.com
