Intramuscular SP1017-formulated DNA electrotransfer enhances transgene expression and distributes hHGF to different rat tissues

THE JOURNAL OF GENE MEDICINE R E S E A R C H A R T I C L EJ Gene Med 2004; 6: 111–118.Published online 1 October 2003 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jgm.463

Intramuscular SP1017-formulated DNAelectrotransfer enhances transgene expressionand distributes hHGF to different rat tissues

Marta Riera2,3

Miguel Chillon4,5

Josep M. Aran2

Josep M. Cruzado3

Joan Torras3

Josep M. Grinyo3

Cristina Fillat1,2*

1Genes and Disease Program, Centrede Regulacio Genomica (CRG),Barcelona, Spain2Medical and Molecular GeneticCenter, Institut de Recerca Oncologica(IRO)3Laboratory of ExperimentalNephrology, Campus Bellvitge,L’Hospitalet de Llobregat4Institucio Catalana de Recerca iEstudis Avancats (ICREA)5Department of Biochemistry andMolecular Biology, Centre deBiotecnologia Animal i TerapiaGenica (CBATEG), UniversitatAutonoma de Barcelona, Bellaterra,Spain

*Correspondence to: Cristina Fillat,Genes and Disease Program, Centrede Regulacio Genomica (CRG),Passeig Mar ı tim 37-49, 08003Barcelona, Spain. E-mail:[email protected]

Received: 25 March 2003Revised: 3 June 2003Accepted: 19 June 2003

Abstract

Background The systemic administration of a therapeutic protein is acommon approach for the treatment of multiple disorders. Intramuscular(i.m.) injection of plasmids, followed by electroporation, has been shown tofacilitate naked DNA gene transfer in skeletal muscle allowing proteins to beproduced and secreted at therapeutically relevant levels.

Methods Plasmid DNA, unformulated or formulated with the non-ioniccarrier SP1017, was injected at the rat tibialis anterior muscle followed by theapplication of electric pulses. Follow-up of protein expression was measured.

Results In our study we report that the non-ionic carrier SP1017significantly increases transgene expression in rat muscle after the i.m.injection of a formulated-pCMVβ plasmid followed by electroporation.Such increased expression was observed after delivering square-waveunipolar electric pulses at two different field strengths: low (110 V/cm)and high (175 V/cm). Moreover, elevated secreted placental alkalinephosphatase (SEAP) plasma levels were achieved with low-voltage(110 V/cm) electroporation. Our results also show that human hepatocytegrowth factor (hHGF) can be produced from rat muscle and delivered toblood circulation at a biologically active level after a single i.m. injection ofan SP1017-formulated plasmid (pCMV/hHGF) followed by electroporation.Tissue distribution studies mostly identified hHGF in the liver, but it was alsofound in the kidneys and lungs suggesting that here too the HGF could be oftherapeutic benefit.

Conclusions Our results indicate that SP1017 enhances the expression ofelectrotransferred genes. Such a delivery approach could prove an efficientmethod for the systemic production of therapeutic proteins. Copyright 2003John Wiley & Sons, Ltd.

Keywords electroporation; gene transfer; SP1017; skeletal muscle; rat;hepatocyte growth factor

Introduction

Numerous diseases require treatment with the systemic delivery of atherapeutic protein. Repetitive or continuous injections of a therapeuticprotein is the method currently employed in clinical practice to achievehigh plasma concentration levels. Gene transfer to muscle is a candidateapproach for the production of recombinant proteins, given the ease of access

Copyright 2003 John Wiley & Sons, Ltd.

112 M. Riera et al.

to the muscle, and the long-term duration of geneexpression [1,2]. Viral vector gene transfer has beenshown to be efficient but is not suitable for repetitiveinjection. On the other hand, direct intramuscular (i.m.)injection of naked plasmid DNA is feasible, but thelevels of therapeutic protein obtained are too low. Thecombination of plasmid DNA with non-ionic carriers,such as polyvinylpyrrolidone (PVP), has been shown toimprove plasmid delivery to rat muscle [3,4]. Moreover,the non-ionic carrier SP1017 has been shown to exhibitsignificantly higher efficacy than PVP. In fact, SP1017allowed a reduction in both carrier concentration andDNA to achieve equivalent gene expression levels [5].Nevertheless, it has been the application of DNAelectrotransfer that has greatly improved non-viral genetransfer efficiency. In vivo electroporation has shownthat consistently high levels of gene expression may beachieved for a large number of genes [6–8]. Moreover,electroporation has been applied in human clinical trials,for the delivery of chemotherapeutic agents, proving thatit is feasible, efficient and well tolerated [9–11].

Hepatocyte growth factor (HGF) is a multifunctionalcytokine that has been shown to have therapeuticproperties in a variety of pathological conditions such asischemic tissue disfunction, toxic insults, tissue fibrosisand in the progression of chronic diseases [12–16].However, exogenous HGF is extremely unstable in bloodwith a half-life of 3–5 min [17]. This makes it extremelydifficult to achieve high, sustained levels of the proteinin blood circulation. Thus, i.m. gene transfer may provea promising strategy permitting the persistent expressionof HGF in vivo.

In the present paper we evaluate the ability to improvei.m. plasmid gene transfer efficiency by combiningin vivo electroporation with SP1017 poloxamers. Thiscombination resulted in increased gene expression inmuscle and elevated levels of protein secretion. Moreover,the combination of electroporation with the SP1017-formulated plasmid encoding hHGF led to high proteinlevels into blood circulation at the best electroporationconditions tested. Interestingly, significant levels of hHGFwere also detected in liver, lung and kidney.

Materials and methods

Plasmid DNA

In this study we employed three plasmid DNAs: pCMVβ

plasmid (Clontech, Palo Alto, CA, USA) containing theβ-galactosidase gene under the control of the CMVimmediate-early promoter, pSEAP plasmid (Clontech)under the control of the same promoter and drivingan optimized coding sequence for the modified humanalkaline phosphatase protein, and pCMV/hHGF plasmidencoding for the human hepatocyte growth factor (HGF)gene. The hHGF expression vector used was constructedas follows: a 2.2 kb fragment containing the hHGF cDNAwas excised from pBluescript-SK by digestion with Sal I

and Sac I and cloned at the aforementioned restrictionsites of pSP72 (Promega, Madison, WI, USA). A FLAGsequence from pCMV-tag4 (Stratagene, Amsterdam, TheNetherlands) was introduced into the Sal I site of thepSP72hHGF. To generate a fused hHGF/FLAG proteinthe stop codon of the hHGF was previously removed. Anadaptor containing the hHGF KpnI-SalI fragment withoutthe stop codon was cloned to replace the original KpnI-SalIfragment. Finally, the hHGF/FLAG sequence was removedfrom the pSP72hHGF/FLAG by Not I digestion andsubcloned to the pCIneo plasmid (Promega). The resultantvector was designated as pCMV/hHGF. The integrity ofthe sequence was determined by DNA sequencing.

Plasmids were expanded into E. coli strain JM-109and purified with the EndoFree plasmid Giga kit(Qiagen GmbH, Hilden, Germany) in accordance withthe supplier’s protocol. DNA was dissolved in Endofree TEbuffer and kept frozen in aliquots at a concentration of4 µg/µl.

Cell culture

L6 rat skeletal muscle myoblasts (L6.G8, ECACC) culturedin DMEM supplemented with 10% FBS, penicillin(100 U/mL), streptomycin (100 mg/mL) and 2 mMglutamine (Gibco, Invitrogen Corp, Paisley, UK) weretransfected with 10 µg of pCMV/hHGF plasmid withSuperfect transfection reagent (Qiagen) following themanufacturer’s instructions. Cells were plated in 100-mm2 cell culture dishes and subjected to G418 (Sigma,Poole, UK) selection (800 µg/ml) for 14 days. Then theG418-resistant clones were isolated by ring cloning.

Five hundred thousand L6 cells were seeded in thecomplete medium in 60-mm2 Petri dishes. Twenty-fourhours later, the cells were changed to serum-free media(1.5 ml) and cultured for the next 24 h. The culture mediawas then collected and kept frozen at −80 ◦C until hHGFprotein levels were assayed with an ELISA kit (QuantikineImmunoassay human HGF; R&D Systems, Minneapolis,MN, USA).

Reagents

SP1017 is a non-ionic carrier, comprising two amphiphilicblock copolymers, pluronics L61 and F127. The copoly-mers, when combined, constitute an effective carriersystem with high thermodynamic stability and low toxic-ity [18]. The SP1017 solution used was a kind gift fromSupratek Pharma Inc. (Laval, QC, Canada).

Formulation of plasmid DNA withSP1017

Equal volumes of plasmid DNA and SP1017 (0.02%) orsaline solution were gently mixed to achieve a final DNAconcentration of 2 µg/µl and 0.01% w/v of SP1017. Theformulation was injected immediately after preparation.

Copyright 2003 John Wiley & Sons, Ltd. J Gene Med 2004; 6: 111–118.

SP1017 Enhances Expression of Electrotransferred Genes 113

Intramuscular injection

Male Sprague-Dawley rats (200–250 g) were anaes-thetized intramuscularly with a combination anaestheticcontaining ketamine (30 mg/kg), atropine (0.25 mg/kg)and diazepam (0.625 mg/kg). The legs were shaved andmoistened with 70% ethanol. Plasmid was injected intothe tibialis anterior (TA) muscle with a 28-gauge needleinserted in a proximal to distal direction. The injected vol-ume was 200 µl. Rats were fed ad libitum and maintainedunder a 12-h light/dark cycle. Animal procedures wereperformed in accordance with recommendations for theproper care and use of laboratory animals.

In vivo electroporation

Following the i.m. injection of plasmid DNA an electricalfield was applied to the area around the injection.Muscles were held by caliper electrodes composed of two1.5-cm2 steel plates and electric pulses were deliveredto the muscle using a pulse generator (BTX ECM830electroporator; Genetronics Inc., San Diego, CA, USA).In all the experiments we delivered eight 20-ms pulsesat a frequency of 2 Hz. Depending on the group, twodifferent output voltages were applied, 110 or 175 V/cm.A conductive gel applied to the shaved leg ensuredelectrical contact with the skin. To improve plasmidDNA diffusion, 25 units of bovine hyaluronidase (Sigma)in 60 µl saline solution were injected into the muscle2 h before the administration of the plasmid DNA andelectroporation [19].

β-Galactosidase assay

To analyze total β-galactosidase expression activity the TAmuscles were homogenized in 3 ml of a lysis buffer froma Galacto-Light Plus kit (Tropix, Bedford, MA, USA)supplemented with 0.2 mM PMSF. Following 50 min ofincubation at 48 ◦C, the samples were centrifuged at2000 g for 10 min. Supernatants were stored at −70 ◦Cuntil assayed. For each assay, 2 µl of the supernatantwere placed in luminometric tubes containing 200 µlof reaction buffer. After 30 min of incubation at roomtemperature, light emission was measured for 10 s afteradding accelerator to terminate enzyme activity andtrigger light emission. For this detection, a TD-20/20luminometer (Turner Designs Inc., Sunnyvale, CA, USA)was used.

Histology

X-Gal staining for β-galactosidase was performed asfollows: 8-µm cryostat sections were fixed in 2%formaldehyde and 0.3% Nonidet P-40 in PBS (pH 7.4) for2 h at 4 ◦C and then washed in PBS. Slides were incubatedfor 1.5 h at 37 ◦C in X-gal staining solution (1 mg/ml X-gal, 5 mM K4Fe(CN)6, 5 mM K3Fe(CN)6, 2 mM MgCl2

in PBS, pH 7.4). Sections were counterstained in 1%alcoholic eosin, dehydrated through graded alcohols,cleared in xylene and mounted in DPX. Quantification ofblue fiber intensity was performed by the manual taggingof a total of 300 fibers at 100× magnification.

Alkaline phosphatase assay

At different times after DNA electrotransfer, bloodsamples, for alkaline phosphatase quantification, werecollected from the rat tail vein in SST tubes. Thedetection of human secreted alkaline phosphatase wasperformed with the Phospha-Light (Tropix) commercialchemiluminiscent reporter gene assay system. In 25 µlof plasma endogenous alkaline phosphatase activity wasinhibited by a mixture of differential alkaline phosphataseinhibitors from the kit and by heat inactivation (65 ◦C)for 30 min. Light detection was performed at the peakof light emission, 20 min after adding Phospha-Lightreaction buffer in a TD-20/20 luminometer.

ELISA determination of hHGF levels

At different time points after DNA electrotransfer, bloodsamples were collected for human HGF quantificationfrom the tail vein in SST tubes. The protein was quantifiedby a commercial ELISA kit (Quantikine immunoassayhuman HGF; R&D Systems) in accordance with themanufacturer’s instructions.

A homogenization buffer containing 20 mM Tris-HCl,pH 7.5, 2 M NaCl, 0.1% Tween-80, 1 mM EDTA, and1 mM PMSF was used to measure hHGF levels in tissue.Variable buffer volumes were added to the tissue at aratio of 1 : 4 (w/v). After centrifuging at 19 000 g for30 min at 4 ◦C, the supernatant was recovered for hHGFdetermination with the same ELISA kit. The proteinconcentration of the supernatant was determined withthe BCA assay, and the final results were expressed aspg/mg or ng/mg protein.

Statistical analysis

The significance of the effects of the treatment wasassessed by a one-way ANOVA test. AUCs (area underthe curve) were calculated with the PK-calc program.Student’s t-test was used for comparisons between twogroups.

Results

SP1017 enhances electro-gene transferefficiency

The SP1017 carrier system has previously been shown toincrease gene expression of plasmid DNA in rat skeletal


114 M. Riera et al.

naked DNA DNA+SP1017

naked DNA DNA+SP1017

β-ga

lact

osid

ase

(rlu

/mg

prot

)

A B C

G H

I J

F

∗

∗∗

D E

0 5 10 15 20 25 30 35

days after injection

K

∗∗

∗∗ ∗∗

108

107

106

105

104

β-ga

lact

osid

ase

(rlu

/mg

prot

)

108

107

106

105

104

β-ga

lact

osid

ase

(rlu

/mg

prot

)

108

107

106

105

104

Figure 1. β-Galactosidase expression in rat TA muscles at two different output voltages: 110 V/cm (A, B, C, D, E) and 175 V/cm(F, G, H, I, J, K). (A, F) β-Galactosidase activity measured in muscles injected with unformulated DNA (black bars) or formulatedwith SP1017 (grey bars) at day 7. (K) β-Galactosidase activity measured at the indicated time points: ♦ pCMVβ-unformulated; �pCMVβ-formulated. Values are presented as means ± SEM (n = 5). ∗p < 0.05, ∗∗p < 0.001. X-Gal staining was performed at day 7after injection. (B, D, G, I) Muscles injected with pCMVβ-unformulated; (C, E, H, J) muscles injected with pCMVβ-formulated. (B,C, G, H) 20× original magnification; (D, E, I, J) 100× original magnification



muscle [5]. The aim of our study was to evaluate theability of the non-ionic carrier SP1017 to enhance plasmidDNA expression in electroporated rat skeletal muscle.

Rat tibialis anterior muscles were injected with 400 µgof unformulated plasmid DNA pCMVβ or 400 µg ofpCMVβ formulated with SP1017. Immediately afterinjection, eight 20-ms electric pulses at 2 Hz wereapplied. Two different output voltage conditions wereused: (A) 110 V/cm and (B) 175 V/cm. Quantitation oftransgene expression was determined by measuring β-galactosidase activity in homogenates from the entiretibialis at day 7. The presence of SP1017 significantlyincreased β-galactosidase gene expression as comparedwith unformulated plasmid in the two different electro-poration regimes studied (Figures 1A and 1F). Moreover,the histological analysis indicated that the positive areasfor X-gal staining were larger in the muscles of ani-mals from the SP1017 group (Figures 1B, 1C, 1G, 1H)and that the intensity of the staining of individual fiberswas increased at both voltages: 110 V/cm and 175 V/cm(Figures 1D, 1E, 1I, 1J). After quantifying 300 blue-stained fibers from muscles subjected to 175 V/cm, theSP1017 group showed a 30 ± 3% rate of very deep bluestaining, while the saline group only 15 ± 2% (p < 0.05).

We next studied whether the effects of SP1017 wouldpersist for an extended period. A time course of theexpression of electrotransferred pCMVβ-formulated and-unformulated was compared. The β-galactosidase levelsof the SP1017 group were maintained 3- to 7-foldhigher than those of the saline group at all timepoints examined when a field strength of 175 V/cm wasapplied (Figure 1K). The AUCs for pCMVβ-formulatedwith SP1017 and unformulated over the 30-day periodwere 41579404 and 10108059 RLU/day/mg protein,respectively (p < 0.05).

Therefore, these results indicate that i.m. injectionof plasmid DNA formulated with SP1017 followed byelectroporation is an efficient gene transfer approach.

High-level protein secretion afterin vivo electroporation in combinationwith SP1017

Muscle has been shown to efficiently secrete the productof a transgene encoding for an extracellularly exportedprotein into the blood circulation [20,21]. We nextinvestigated the ability of the system to efficiently delivera protein systemically. To this end, 800 µg of pSEAPformulated with SP1017 were injected into the rat TAmuscle followed by electroporation at an output voltageof 110 V/cm. High levels of circulating human secretedalkaline phosphatase were found with peak levels ofplasma SEAP protein (1800 ng/ml) at day 5, fallingoff up to day 14 when protein became undetectable(Figure 2). These results indicate that, even when theSP1017 is administered in combination with low-voltageelectroporation, marked levels of SEAP activity can bedetermined in plasma.


SE

AP

(ng

/ml)

0

300

600

900

1200

1500

1800

2100

0 2 4 6 8 10 12 14 16 18 20 22 24

Figure 2. Plasma SEAP levels in rats after i.m. injection of 800 µgof pSEAP formulated with SP1017 followed by electroporationat 110 V/cm. Serum was collected from the tail vein at theindicated time points after injection. Values are presented asmeans ± SEM (n = 5)

High hHGF expression levels afterin vivo electroporation in combinationwith SP1017

We studied whether the therapeutic protein hHGF couldbe efficiently delivered in vivo with the combined system:SP1017-formulated plasmid and electroporation.

First we tested whether the pCMV/hHGF constructled to high levels of expression of the hHGF proteinin vitro. L6 muscle-derived rat cells were transfectedwith the pCMV/hHGF expression vector and subjectedto G418 selection. Clones were grown and assayed forhHGF production. The hHGF that accumulated in theconditioned media from L6 cells over 24 h was measuredand found to reach a maximum value of 703 pg/ml.

To evaluate the in vivo expression of hHGF we directlyinjected the TA muscle with pCMV/hHGF formulated withSP1017 followed by eight 20-ms electropulses at 2 Hz, atboth low (110 V/cm) and high voltage (175 V/cm). hHGFexpression was analyzed in the TA muscle and increasedprotein levels were found at day 7 after injection for bothregimes of electroporation. Remarkably, hHGF expressionachieved levels as high as 300 ng of hHGF protein permg of total tissue protein under high voltage conditions(Figure 3B). Although the expression of hHGF declines atday 15, a significant amount of hHGF protein was stillfound to come from the muscles that received 175 V/cm.In fact the hHGF levels at day 15 were very similar(10 ng/mg prot.) to those found at day 7 when the lowervoltage was applied (9 ng/mg prot.) (Figures 3A and 3B).

hHGF production was then measured in plasma over14 days. At low voltage an increase of hHGF levels inplasma was already observed at day 2 after injection andpeak values (28–35 pg/ml) were achieved after 3–5 days.hHGF expression was undetectable by day 14 (Figure 4A).When the higher voltage was applied a very large increasein hHGF plasma levels was observed peaking at day 6(460 pg/ml, Figure 4B). Moreover, hHGF levels remainedhigher than those achieved at the lower voltage through today 11 (80 vs. 35 pg/ml). However, as previously found,on day 14 no protein expression was detected. These


116 M. Riera et al.

0

2

4

6

8

10

12

14

16

7 day 15 day

hHG

F (

ng/m

g pr

ot)

0

100

200

300

400

7 day 15 day

hHG

F (

ng/m

g pr

ot)

A B

Figure 3. hHGF expression in rat TA muscles injected with 800 µg of pCMV/hHGF formulated with SP1017 followed byelectroporation. Protein levels were determined at the indicated time points. (A) 110 V/cm, (B) 175 V/cm. Values are presented asmeans ± SEM (n = 5)

A B

0

10

20

30

40

50

0

100

200

300

400

500

600


hHG

F (

pg/m

l)

hHG

F (

pg/m

l)

0 2 4 6 8 10 12 14 16


0 2 4 6 8 10 12 14 16

Figure 4. hHGF protein plasma levels in rats injected with 800 µg of pCMV/hHGF formulated with SP1017 followed byelectroporation. Protein levels were determined at the indicated time points. (A) 110 V/cm, (B) 175 V/cm. Values are presented asmeans ± SEM (n = 8)

0

100

200

300

400

500

600

7 day 15 day

hHG

F (

pg/m

g pr

ot)

0

10

20

30

40

50

60

7 day 15 day

hHG

F (

pg/m

g pr

ot)

A B

Figure 5. hHGF protein levels in from rats peripheral tissues injected with 800 µg of pCMV/hHGF formulated with SP1017 followedby electroporation. Protein levels were determined at the indicated time points in the kidney , liver and lung . (A) 110 V/cm,(B) 175 V/cm. Values are presented as means ± SEM (n = 5)

results indicate that the loss of expression in plasma wasindependent of the peak levels of gene expression.

In a therapeutic application of hHGF gene transferit would be necessary to ensure that the hHGF proteinreaches the target tissue. To address this point we assessedwhether the presence of high hHGF in plasma could resultin protein localization in peripheral tissues. hHGF proteinwas detected in lung, liver and kidney at day 7 afterinjection with both regimes of electroporation, although

the highest levels were found with the higher voltageof 175 V/cm (Figures 5A and 5B). In the kidney, hHGFhad virtually disappeared by day 15 after treatment atthe lower voltage whereas, at the higher voltage, similarlevels were found on days 7 and 15. In the lung, thehHGF levels did not vary from day 7 to day 15 for eitherof the regimes of electroporation tested. Interestingly, inthe liver, hHGF levels were higher on day 15 than on day7 when treated at the highest voltage.



Discussion

With the aim of defining an optimal approach for sys-temic production of therapeutic proteins we studied theeffectiveness of the non-ionic carrier SP1017 in com-bination with electroporation to enhance intramusculargene transfer efficiency. For the first time we report thatSP1017 substantially improves gene expression in rats,after the i.m. injection of formulated plasmid followed byelectroporation.

The injection of pCMVβ/SP1017 at two differentregimes of electroporation resulted in increased β-galactosidase activity, as determined at day 7. Moreover,the expression of β-galactosidase in the SP1017 grouppersisted for at least 30 days.

How SP1017 contributed to increase gene expressionis not fully understood but the histological experimentsperformed with the formulated pCMVβ showed thatSP1017 enhanced both, the extent of gene expressionin the tissue and the levels of expression in individualfibers, thus suggesting that the number of DNA plasmidcopies introduced into the muscle fibers may be higherin the presence of SP1017. This could be the result ofthe increased cellular uptake of plasmid DNA and/ora reduction in DNA degradation [21]. In fact, themechanism of action of SP1017 remains to be exploredalthough some studies have indicated that the effect of theSP1017 is achieved by increasing membrane permeability[5,22]. Interestingly, it has been proposed that the non-ionic polymer PVP increases dispersion, protection andfacilitates the uptake of the plasmid in the muscle[23]. Our results seem to favor the hypothesis thatSP1017/plasmid formulation will lead to an increase incellular uptake and protection from nuclease degradation,in similar way to what has been proposed with respectto PVP.

SP1017 is composed of two block copolymers, pluronicsL61 and F127, also known as poloxamers. Interestingly,it has recently been shown that the poloxamer 188 whencoinjected with plasmid DNA followed by very high-voltage electroporation results in a substantial reductionin serum creatine phosphokinase, but no inhibitionin the development of muscle damage was observed.Nevertheless, the authors propose that p188 might beuseful in counteracting the effects of electrical traumain the electroporated muscles [24]. With the regimesof electroporation employed in the present study, thehistological analysis revealed a degree of lymphocyteinfiltration at day 7, especially at 175 V/cm (data notshown). Tissue damage after electroporation was similarin muscles injected with SP1017-formulated plasmid andin muscles injected with plasmid prepared in salinesolution. However, we cannot rule out the possibilitythat SP1017, in a similar way to p188, will contribute toa reduction in serum creatine phosphokinase activity. Infact, the proinflammatory response found is more likelyrelated to the presence of a large number of plasmidDNA containing hypomethylated CpG residues than tothe regimes of electroporation studied [25,26].

HGF is a growth factor with multiple biologicalproperties and several studies have tested the efficiencyof the recombinant hHGF protein as a therapeutic agentin preventing tissue fibrosis and dysfunctions of manydifferent organs [15,27–30].

HGF has a very short half-life (t1/2 3–5 min) [17]as it is rapidly cleared by the liver in vivo. Thus, thedelivery of recombinant hHGF protein only producesa transient elevation of HGF in blood circulation andthus requires the administration of repetitive injections inorder to achieve a biological response, a very inconvenientand extremely costly process. Thus, hHGF gene transferbased strategies may prove a feasible approach to certainkinds of dysfunctions. Our results, based on the in vivoelectroporation of hHGF/SP1017-formulated plasmid,showed that efficient expression of hHGF in plasma wasachieved, attaining a maximum level of 0.45 ng/ml atday 7. Previous studies by Ueki et al. have demonstratedthat the intramuscular administration of the HGF genewith liposomes containing the hemagglutinating virus ofJapan (HVJ liposomes) induced plasma levels of hHGFin a rat model of liver cirrhosis. In this work levels ofonly 0.05 ng/ml were reached after a single injectionand up to 0.1 ng/ml of hHGF were achieved after asecond injection. Interestingly, at such low levels, thefibrosis in the cirrhotic liver was observed to resolve,thereby improving the survival rate amongst the ratsaffected by the disease [28]. In mice, concentrationsof 2 ng/ml HGF were achieved in plasma after muscleelectroporation under regimes slightly different to thoseemployed in our study. However, this 2 ng/ml representedtotal HGF, including the endogenous basal levels of miceHGF already at 0.5 ng/ml. Thus, after electroporation,only a 4-fold increase in HGF levels was obtained [31].Under these conditions, the authors have shown that HGFcan attenuate acute liver injury and prevent changes inrat kidneys in subtotally nephrectomized rats [32].

We observed a decline in the expression in plasmaafter 2 weeks. This could most likely be accounted for thereduction of HGF expression in muscle due to a drop inprotein production and to the short stability of the proteinin circulation. In fact, in a recent study, using hHGF-encoding adenoviral vector, transient levels of serumhHGF were also observed. In this study the authorsclaim that such a transient effect could be explainedby the pharmacokinetic profile of rapid clearance of thehepatocyte growth factor [33].

However, we cannot rule out the possibility that therapid decrease in gene expression could also derive froman immunoresponse caused by the presence of hHGFprotein in rats although a single plasmid injection wasapplied, and the human and rat HGF proteins showed88% similarity at the protein level.

It is important to note that we have observed thepresence of hHGF in peripheral tissue such as the kidneys,lungs and liver, with the latter showing the highestlevels of HGF protein. Such tissue distribution couldbe accounted for a potential balance of the circulationcircuits through which the HGF protein is travelling, and


118 M. Riera et al.

by the levels of expression of the HGF receptor, c-met, inthe different organs [34].

A hydrodynamic-based in vivo transfection procedureemploying the systemic administration of naked HGFexpression plasmid has also been studied with high,persistent levels of hHGF expression achieved in thekidneys and liver [35]. Moreover, when this strategy wasapplied to mouse models of different renal dysfunctions,an improvement in the course of the disease was reported[16,29,30]. This system has been able to achieve the mostimpressive hHGF levels in plasma (3–4 ng/ml). However,such high levels were also seen to have certain adverseeffects such as overall body growth and the enlargementof several organs [35]. Moreover, it is a highly aggressiveapproach and far from use in clinical practice.

We believe that the results presented here constitutean efficient gene therapy approach capable of achievingeither the local or systemic production of therapeuticproteins with the added advantages of simplicity, safetyand lack of toxicity.

Acknowledgements

We thank Giovanni Gaudino (Amedeo Avogadro University,Eastern Piedmont, Novara, Italy) for kindly providing us with thehHGF cDNA and Supratek Pharma Inc. (Laval, QC, Canada) forproviding us with the SP1017 carrier. This work was supportedby grants FIS 00/0017-02 and FEDER 2FD1997-2109-C02-02.

References

1. Rissanen TT, Vajanto I, Yla-Herttuala S. Gene therapy fortherapeutic angiogenesis in critically ischaemic lower limb – onthe way to the clinic. Eur J Clin Invest 2001; 31: 651–666.

2. Prud’homme GJ, Lawson BR, Chang Y, et al. Immunotherapeu-tic gene transfer into muscle. Trends Immunol 2001; 22:149–155.

3. Mumper RJ, Wang J, Klakamp SL, et al. Protective interactivenoncondensing (PINC) polymers for enhanced plasmiddistribution and expression in rat skeletal muscle. J ControlRelease 1998; 52: 191–203.

4. Anwer K, Shi M, French MF, et al. Systemic effect of humangrowth hormone after intramuscular injection of a single doseof a muscle-specific gene medicine. Hum Gene Ther 1998; 9:659–670.

5. Lemieux P, Guerin N, Paradis G, et al. A combination ofpoloxamers increases gene expression of plasmid DNA in skeletalmuscle. Gene Ther 2000; 7: 986–991.

6. Mir LM, Bureau MF, Gehl J, et al. High-efficiency gene transferinto skeletal muscle mediated by electric pulses. Proc Natl AcadSci U S A 1999; 96: 4262–4267.

7. Fewell JG, MacLaughlin F, Mehta V, et al. Gene therapy forthe treatment of hemophilia B using PINC-formulated plasmiddelivered to muscle with electroporation. Mol Ther 2001; 3:574–583.

8. Martinenghi S, Cusella De Angelis G, Biressi S, et al. Humaninsulin production and amelioration of diabetes in mice byelectrotransfer-enhanced plasmid DNA gene transfer to theskeletal muscle. Gene Ther 2002; 9: 1429–1437.

9. Mir LM, Glass LF, Sersa G, et al. Effective treatment of cutaneousand subcutaneous malignant tumours by electrochemotherapy.Br J Cancer 1998; 77: 2336–2342.

10. Gehl J, Geertsen PF. Efficient palliation of haemorrhagingmalignant melanoma skin metastases by electrochemotherapy.Melanoma Res 2000; 10: 585–589.

11. Rodriguez-Cuevas S, Barroso-Bravo S, Almanza-Estrada J, et al.Electrochemotherapy in primary and metastatic skin tumors:

phase II trial using intralesional bleomycin. Arch Med Res 2001;32: 273–276.

12. Hojo S, Fujita J, Yoshinouchi T, et al. Hepatocyte growth factorand neutrophil elastase in idiopathic pulmonary fibrosis. RespirMed 1997; 91: 511–516.

13. Kaido T, Imamura M. Hepatocyte growth factor: clinicalimplications in hepatobiliary pancreatic surgery. J HepatobiliaryPancreat Surg 2001; 8: 65–75.

14. Mizuno S, Kurosawa T, Matsumoto K, et al. Hepatocyte growthfactor prevents renal fibrosis and dysfunction in a mouse modelof chronic renal disease. J Clin Invest 1998; 101: 1827–1834.

15. Matsumoto K, Nakamura T. Hepatocyte growth factor:renotropic role and potential therapeutics for renal diseases.Kidney Int 2001; 59: 2023–2038.

16. Yang J, Liu Y. Blockage of tubular epithelial to myofibroblasttransition by hepatocyte growth factor prevents renal interstitialfibrosis. J Am Soc Nephrol 2002; 13: 96–107.

17. Kawaida K, Matsumoto K, Shimazu H, et al. Hepatocyte growthfactor prevents acute renal failure and accelerates renalregeneration in mice. Proc Natl Acad Sci U S A 1994; 91:4357–4561.

18. Alakhov V, Klinski E, Lemieux P, et al. Block copolymericbiotransport carriers as versatile vehicles for drug delivery.Expert Opin Biol Ther 2001; 1: 583–602.

19. McMahon JM, Signori E, Wells KE, et al. Optimisation ofelectrotransfer of plasmid into skeletal muscle by pretreatmentwith hyaluronidase – increased expression with reduced muscledamage. Gene Ther 2001; 8: 1264–1270.

20. Bettan M, Emmanuel F, Darteil R, et al. High-level proteinsecretion into blood circulation after electric pulse-mediatedgene transfer into skeletal muscle. Mol Ther 2000; 2: 204–210.

21. Nicol F, Wong M, MacLaughlin FC, et al. Poly-L-glutamate,an anionic polymer, enhances transgene expression forplasmids delivered by intramuscular injection with in vivoelectroporation. Gene Ther 2002; 9: 1351–1358.

22. Kabanov AV, Batrakova EV, Alakhov VY. Pluronic blockcopolymers as novel polymer therapeutics for drug and genedelivery. J Control Release 2002; 82: 189–212.

23. Mumper RJ, Duguid JG, Anwer K, et al. Polyvinyl derivativesas novel interactive polymers for controlled gene delivery tomuscle. Pharm Res 1996; 13: 701–709.

24. Hartikka J, Sukhu L, Buchner C, et al. Electroporation-facilitateddelivery of plasmid DNA in skeletal muscle: plasmid dependenceof muscle damage and effect of poloxamer 188. Mol Ther 2001;4: 407–415.

25. McMahon JM, Wells KE, Bamfo JE, et al. Inflammatoryresponses following direct injection of plasmid DNA into skeletalmuscle. Gene Ther 1998; 5: 1283–1290.

26. Krieg AM. Direct immunologic activities of CpG DNA andimplications for gene therapy. J Gene Med 1999; 1: 56–63.

27. Miyagawa S, Sawa Y, Taketani S, et al. Myocardial regenerationtherapy for heart failure: hepatocyte growth factor enhancesthe effect of cellular cardiomyoplasty. Circulation 2002; 105:2556–2561.

28. Ueki T, Kaneda Y, Tsutsui H, et al. Hepatocyte growth factorgene therapy of liver cirrhosis in rats. Nat Med 1999; 5: 226–230.

29. Yang J, Dai C, Liu Y. Systemic administration of naked plasmidencoding hepatocyte growth factor ameliorates chronic renalfibrosis in mice. Gene Ther 2001; 8: 1470–1479.

30. Dai C, Yang J, Liu Y. Single injection of naked plasmid encodinghepatocyte growth factor prevents cell death and amelioratesacute renal failure in mice. J Am Soc Nephrol 2002; 13: 411–422.

31. Xue F, Takahara T, Yata Y, et al. Attenuated acute liver injuryin mice by naked hepatocyte growth factor gene transfer intoskeletal muscle with electroporation. Gut 2002; 50: 558–562.

32. Tanaka T, Ichimaru N, Takahara S, et al. In vivo gene transferof hepatocyte growth factor to skeletal muscle prevents changesin rat kidneys after 5/6 nephrectomy. Am J Transplant 2002; 2:828–836.

33. Hwang T-H, Yoon B-C, Jeong J-S, et al. A single administrationof adenoviral-mediated HGF cDNA permits survival of mice fromacute hepatic failure. Life Sci 2003; 72: 851–861.

34. Liu Y, Tolbert EM, Sun AM, et al. In vivo and in vitro evidencefor increased expression of HGF receptor in kidney of diabeticrat. Am J Physiol 1996; 271: F1202–1210.

35. Yang J, Chen S, Huang L, et al. Sustained expression of nakedplasmid DNA encoding hepatocyte growth factor in micepromotes liver and overall body growth. Hepatology 2001; 33:848–859.


Documents

Intramuscular SP1017-formulated DNA electrotransfer enhances transgene expression and distributes hHGF to different rat tissues