Transgene expression of human PON1 Q in mice protected the liver against CCl4-induced injury

THE JOURNAL OF GENE MEDICINE R E S E A R C H A R T I C L EJ Gene Med 2008; 10: 94–100.Published online 29 November 2007 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/jgm.1128

Transgene expression of human PON1 Q in miceprotected the liver against CCl4-induced injury

Chi ZhangWei PengXiaoling JiangBo ChenJie ZhuYuhui Zang*Junfeng ZhangTongyang ZhuJunchuan Qin

State Key Laboratory ofPharmaceutical Biotechnology, Schoolof Life Science, Nanjing University,Nanjing 210093, P.R. China

*Correspondence to: Yuhui Zang,State Key Laboratory ofPharmaceutical Biotechnology,School of Life Science, NanjingUniversity, Nanjing 210093,P.R. China.E-mail: zang [email protected]

Received: 18 July 2007Accepted: 26 September 2007

Abstract

Background Oxidative stress, often in association with decreased antioxi-dant defenses, plays a pathogenetic role in both initiation and progressionof liver injuries, leading to almost all clinical and experimental conditions ofchronic liver diseases. Human paraoxonase 1 (hPON1) is a liver-synthesizedenzyme possessing antioxidant properties. Here, we investigate the effects oftransgene-expressed hPON1 Q on alleviating lipid peroxidation and prevent-ing liver injury in a mouse model.

Methods The hPON1 Q gene was cloned into pcDNA3.0 plasmid andelectro-transferred into mouse skeletal muscle. After CCl4 had beenadministrated to induce liver injury, mice were monitored for serum levelsof alanine aminotransferase (ALT), aspartate aminotransferase (AST) andmalonyldialdehyde (MDA). The extent of CCl4-induced liver injury was alsoanalyzed through histopathological observations.

Results After gene delivery, hPON1 mRNA expression was detected inmouse muscle and serum PON1 activity was 1.5 times higher than that of thecontrol counterpart. In the PON1 Q gene transferred mice, protection againstCCl4-induced liver injury was reflected by significantly decreased serumALT, AST and MDA levels compared to those in control mice (P < 0.01).Histological observations also revealed that hepatocyte necrosis, hemorrhage,vacuolar change and hydropic degeneration were apparent in control miceafter CCl4 administration. In contrast, the damage was significantly prevented(P < 0.01) in the hPON1 Q transferred mice.

Conclusions Intramuscular electro-transfer of the hPON1 Q gene led toefficient expression of hPON1 in mice. Elevated levels of PON1, by virtueof its potency to alleviate oxidative stress, could protect mice from sufferingCCl4-induced liver damage. Copyright 2007 John Wiley & Sons, Ltd.

Keywords human paraoxonase 1 Q; skeletal muscle; gene transfer; oxidativestress; liver injury

Introduction

Liver injuries, leading to fibrosis, cirrhosis and other kinds of chronicliver diseases (CLD), occur in response to a variety of insults includingviral hepatitis, alcohol abuse, drugs and autoimmune attack of hepato-cytes. Whatever the etiology, reactive molecules originating from oxidativestress have been suggested to play a pathogenetic role in both initiationand progression of liver injuries [1–3]. Evidence of oxidative stress hasbeen detected in almost all the clinical and experimental conditions of CLD,

Copyright 2007 John Wiley & Sons, Ltd.

hPON1 Q Delivery Protects Against Liver Injury 95

often in association with decreased antioxidant defenses[4–6]. Therefore, antioxidation therapy is a promisingstrategy for treating or preventing oxidative stress relatedliver diseases.

Human serum paraoxonase 1 (hPON1), belonging toa family of enzymes that can catalyze the hydrolysisof a broad range of esters and lactones, is mainlyexpressed in liver and distributed in tissues such asliver, kidney, intestine, and mostly, serum [7]. Most ofserum PON1 is localized on the surface of high-densitylipoproteins (HDL) and HDL-associated PON1 is mainly anantioxidant protein that protects low-density lipoproteins(LDL) and HDL against oxidation and destroys biologicallyactive oxidized lipids in lipoproteins and arterial cells.This property, in turn, has led to numerous studiesinvestigating the role of PON1 genetic polymorphisms,level of PON1 expression, and PON1 activities undernormal and pathophysiological conditions, with a specialemphasis on coronary heart diseases [8,9]. The codingsequence of hPON1 has two common polymorphic sites,a Met(M)/Leu(L) substitution at position 55 and aGln(Q)/Arg(R) at position 192. Much attention has beenpaid to hPON1 Q since it shows higher affinity andcatalytic activity with a number of substrates than hPON1R [10].

PON1 could reduce the lipid peroxidation and wasinactivated after hydrolysis of lipid peroxides [11].Previous studies described the relationship between serumPON1 activity and liver function, demonstrating thatdecrease in serum PON1 activity in CLDs was relatedto the degree of hepatic dysfunction [12]. Consequently,downgrade of PON1 activity weakened the antioxidantability of the serum and exacerbated liver cirrhosis andfibrosis at the same time. Moreover, active HDL-PON1complex also played an important role in preventinghepatocytes apoptosis and liver fibrosis through itsphysiological function to regulate the oxidative stress[13]. So, attempts to systematically increase PON1activity were considered to alleviate lipid peroxdiationand prevent liver injury.

In the present study, we first investigated the feasibilityof expressing rhPON1 in cultured mature myotubes andafterwards introduced the hPON1 Q cDNA gene intomouse skeletal muscle by electroporation (EP) to maintaina higher serum PON1 level. After carbon tetrachloride(CCl4) had been injected into mice to induce liver injury,evaluations of liver marker enzymes, extent of oxidativestress and liver histology were performed, revealing thatelevated levels of PON1 were effective in alleviating lipidperoxidation and protecting liver against CCl4-inducedliver damage.

Materials and methods

Plasmid

pBluescript-hPON1Q containing the hPON1 Q cDNAwas kindly provided by Professor Manji Sun from the

Academy of Military Medical Science, China. To constructpcDNA3.0-hPON1Q, the hPON1 Q gene was amplifiedby polymerase chain reaction (PCR) and inserted intopcDNA3.0 under the control of the cytomegalovirus(CMV) promoter. Plasmid construction was manipulatedby routine molecular cloning techniques. All chemicalswere from Sigma (USA) and enzymes from Takara(Japan).

Cell culture

Fragments of mouse skeletal muscle were obtainedfrom the crura of neonatal ICR mice (from theAnimal Centre of Nanjing Medical University, P. R.China). The mashed tissue was digested to sin-gle myoblasts by treating with trypsin and seededin high-glucose Dulbecco’s modified Eagle’s medium(DMEM) supplemented with 10% fetal calf serum(proliferation medium). Upon approaching confluence,myoblasts were induced to differentiate and fusedinto mature myotubes by changing the proliferationmedium to differentiation medium (DMEM with 2% horseserum).

Gene transfer and expression of hPON1Q in cultured myotubes

After trypsin and EDTA treatment, myotubes wereharvested and resuspended at a concentration of2 × 106 cells/ml in phosphate-buffered saline (PBS).The cell-DNA mixture (0.5 ml of cell suspension with10 µg plasmid) was placed into the electroporationchamber, and a single pulse was applied across theelectrodes (0.4 cm gap) using the Gene Pulser IIelectroporation system (BioRad). The optimized electricfield strength, time constant and capacitance for myotubeselectroporation were 700 V/cm, 20 ms and 1000 µF,respectively. After the pulse, cells were cultured onsix-well plates. To determine the PON1 activity in celllysate, cultured cells were released from plates andresuspended in 0.5 ml PBS. After ultrasonication on icefor 20 × 2 s pulses with 2 s intervals, supernatants of thecell lysate were collected by centrifuging for 5 min at12 000 rpm.

Arylesterase activity assay

Arylesterase activities towards phenyl acetate were mea-sured spectrophotometrically at 270 nm. E270 for thereaction was 1310 M−1 cm−1. One unit of arylesteraseactivity was equal to 1 µmol of phenyl acetatehydrolyzed/ml/min. Reaction mixtures contained 50 mMTris/HCl (pH 8.0), 2 mM CaCl2, 10 mM substrate andthe sample in a total volume of 1 ml. The reaction wascarried out for 10 min at 37 ◦C and stopped by adding100 µl 0.1 M EDTA. All chemicals were from Sigma(USA).

Copyright 2007 John Wiley & Sons, Ltd. J Gene Med 2008; 10: 94–100.DOI: 10.1002/jgm

96 C. Zhang et al.

Transgene expression of hPON1 Q inICR mice

Female ICR mice (4 weeks old) were obtained from theAnimal Centre of Nanjing Medical University, China.After anesthetization, mice were injected with 50 µg ofpcDNA3.0-hPON1Q plasmid in 50 µl PBS into their righttibialis anterior (TA) muscle. Thirty seconds after DNAinjection, electroporation was applied through two 0.5 cmstainless steel plate electrodes placed on either side of thesurgically exposed TA. Square wave electric pulses weregenerated by an electropulsator (model ECM-830; BTX,San Diego, CA, USA) with output voltage of 200 V/cm,eight pulses each of 20 ms duration, and 1 Hz frequencyof pulse delivery. Another two groups of mice were usedas control. One group was injected with 50 µg pcDNA3.0in TA muscles with electroporation, and the other wasinjected with 50 µl PBS without electroporation. Bloodsamples were collected from the tails every other dayafter electroporation and 1 µl serum was separated tomeasure PON1 activities as previously described. All micewere fed at room temperature with sufficient water andfood.

Isolation of total RNA and reverse-transcription (RT)-PCR

Total RNA was isolated from 0.1 g mouse liver ormuscle using the RNA Isolation & Purification kit(Waston BitoTech, Shanghai, China) and digestedwith DNAse I (Takara) to avoid DNA contamination.First-strand cDNA was synthesized from 2 mg totalRNA using 10 pmol oligo-dT primer in a 25 µlreaction containing 1 µl RNasin (Takara), 1 mmol dNTP,10 mmol dithiothreitol, and 15 U reverse transcriptase(Invitrogen). Synthesized cDNA was diluted tenfold andused as templates in subsequent PCR. The hPON1 Qprimers were as follows, forward: cgatggcgaagctgattgcg,reverse: cgttagagctcacagtaaag. Mouse glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as aninternal control based on its conserved region (forward:ttagcccccctggccaagg, reverse: ggaggccatgatggccatg). ThePCR program was as follows: denaturation at 94 ◦C for3 min, followed by 30 cycles of 30 s at 94 ◦C, 30 s at 51 ◦C,90 s at 72 ◦C, and a final polymerization step at 72 ◦C for7 min. PCR products were analyzed by electrophoresis on1.0% agarose gel containing ethidium bromide.

CCl4 treatment

CCl4 (99.8% purity) was dissolved in corn oil (1 : 1v/v). Mice were intraperitoneally injected with 0.8 ml/kgbody weight CCl4 to elicit liver injury. Control micereceived an equal volume of corn oil only. Mice weresacrificed and blood was collected via the inferior venacava 24 h after CCl4 treatment. Livers and TA muscleswere removed immediately for further processing. The

animal experiments were conducted according to the ‘Useand Care Guidelines of Experimental Animals’ of JiangsuProvince, China.

Serum aminotransferase activity

Serum alanine aminotransferase (ALT) and aspartateaminotransferase (AST) activities were measured usingan Hitachi 7020 biochemical analyzer.

Measurement of malonyldialdehyde

Lipid peroxidation was spectrophotometrically detectedmeasuring the serum level of malonyldialdehyde (MDA),the end product derived from the breakdown ofpolyunsaturated fatty acids and related esters, with theclassical thiobarbituric acid (TBA) method as previouslydescribed [14].

Histological analysis

The TA muscles were embedded in tissue freezing medium(LEICA) and cut into frozen sections using a rapidsectioning cryostat (LEICA CM1900). The excised liverwas fixed in buffered formalin, embedded in paraffin, cutinto 5-µm-thick sections, and examined with hematoxylin-eosin (H&E) staining. All chemicals were from Sigma(USA).

Statistical analysis

The data were analyzed for statistical significance usingone-way analysis of variance (ANOVA) and Student’s ttest. Statistical significance was determined at P < 0.05.

Results

Expression of hPON1 Q in culturedmature myotubes

The recombinant plasmid pcDNA3.0-hPON1Q containinghPON1 Q cDNA under the control of the CMV promotorwas constructed as described in the Materials and Methodssection. To investigate whether gene delivery in musclecould result in efficient expression and secretion ofhPON1 Q in vivo, mouse myotube cells were isolated andtransferred with pcDNA3.0-hPON1Q plasmid. Followingplasmid delivery, arylesterase activities in the culturemedium were assayed to determine hPON1 Q expressionlevel. As expected, the arylesterase activity of transfectedmyotube cells kept increasing for more than 72 h and thehighest enzyme activity in culture medium was 0.971 ±0.043 U/ml, whereas there was no detectable arylesteraseactivity in untreated cells. The total arylesterase activitiesin culture medium and cell lysate were also evaluated



Figure 1. RT-PCR analysis of PON1 gene expression in variousmice tissues. Total RNA were isolated from muscle or liverand subjected to RT-PCR analysis. PCR products were analyzedby electrophoresis on 1.0% agarose gel containing ethidiumbromide. Lane 1, liver of control mouse; lane 2, liver of mousewith hPON1 Q gene delivery; lane 3, muscle of control mouse;lane 4, muscle of mouse with hPON1 Q gene delivery

respectively to determine the ratio of secretion, showingthat about 84.7% of the expressed hPON1 Q was secretedinto culture medium.

Transgene expression of hPON1 Q inmouse skeletal muscle

pcDNA3.0-hPON1 Q plasmid was introduced into mouseskeletal muscle by electroporation as described in theMaterials and Methods section. To determine whetherhPON1 Q was expressed efficiently after gene delivery,we performed RT-PCR with the total RNA isolated frommuscle and liver. As shown in Figure 1, transfer of thehPON1 Q gene by electroporation (EP) led to a moderatePON1 mRNA expression in mouse muscle. In contrast,there was no obvious variation in the level of liversynthesized PON1 mRNA after hPON1 Q gene transfer.

Mice serum arylesterase activity started to increasefrom 4 days after hPON1 Q gene delivery and peakedat 10 days later. The highest mean activity in serumwas 91.7 ± 8.7 U/µl, about 1.5 times higher thanthose of the control groups, which were treated withpcDNA3.0 plasmid (56.3 ± 8.1 U/µl) or PBS (56.1 ±4.6 U/µl) (P < 0.01). The difference between the twocontrol groups was not significant. The higher PON1levels in the hPON1 Q gene transferred mice weresustained for more than 16 days after EP (Figure 2).Frozen section examination of TA muscles showedthat there was no obvious histological distinctionbetween electro-transferred and normal TA muscle,indicating that injury to skeletal muscle by EP couldbe recovered in a short period of time (data notshown).

Protective effect of transgene-expressed hPON1 against CCl4-inducedliver injury

To examine the protective effects of transgene-expressedhPON1 Q against CCl4-induced liver injury, one group

Figure 2. Effects of pcDNA3.0-hPON1Q plasmid delivery onmice serum PON1 activities. The serum arylesterase activitywas determined according to its ability to hydrolyze phenylacetate as described in the Materials and Methods section.Data were means of ten different samples of a group. Serumarylesterase activities of mice treated with pcDNA3.0-hPON1Qwere significantly higher than those of two control groups(P < 0.01) treated with PBS or pcDNA3.0 from 6 days aftergene delivery. (ž: pcDNA3.0-hPON1Q with electroporation; �:PBS without electroporation; �: pcDNA3.0 with electroporation)

of mice were electro-transferred with pcDNA3.0-hPON1Q plasmid in skeletal muscle and then intraperitoneallyinjected with 0.8 ml of CCl4/kg body weight whenserum PON1 level reached its peak at 10 days aftergene delivery (EP group). Another two groups withouthPON1 Q gene transfer were injected with the samedose of CCl4 (CCl4 group) or corn oil (control group),respectively. Serum PON1 activities of the EP groupand the CCl4 group started to drop quickly afterCCl4 injection. The absolute decreased values in theEP group and the CCl4 group at 24 h after CCl4injection were similar. Nevertheless, the EP groupstill sustained a higher serum arylesterase activity(52.2 ± 7.8 U/ml) than that of the CCl4 group (32.9 ±10.1 U/ml), and was approximately equivalent to thatof the control group without CCl4 treatment (57.1 ±3.1 U/ml) (Table 1).

Serum aminotransferase activities and concentrationsof MDA, a molecule generated during lipoperoxidation,were also assayed at 24 h after CCl4 injection to investi-gate the degrees of liver injury and lipid peroxidation indifferent groups. The results are summarized in Table 1.In the CCl4 group, serum AST and ALT activities weresignificantly higher than the corresponding values inthe control group, indicating the obvious injury effectsof CCl4 on liver function. In contrast, the serum ASTand ALT activities of the EP group were significantlylower than those of the CCl4 group (P < 0.01), show-ing that transgene expression of hPON1 Q could protectthe hepatocytes against CCl4-induced injury. After CCl4administration, the serum MDA level in the CCl4 groupwas about 3.7 times higher than those in the controlgroup. However, serum MDA contents of the EP group


98 C. Zhang et al.

Table 1. Serum PON1, AST, ALT and MDA levels in various groupsof mice

GroupaPON1

(U/ml)bALT(KU)

AST(KU)

MDA(nmol/ml)

Controlgroup

57.1 ± 3.1 37 ± 9 110 ± 17 7.48 ± 1.84

CCl4group

32.9 ± 10.1∗∗ 481 ± 129∗∗ 1034 ± 292∗∗ 27.58 ± 3.85∗∗

EP group 52.2 ± 7.8## 269 ± 63## 637 ± 123## 7.18 ± 2.37##

Blood samples were collected at 24 h after CCl4 treatment. Values aremeans ± SD of ten blood samples from different mice.aControl group, mice were treated with oil; CCl4 group, mice were treatedwith CCl4; EP group, mice were treated with CCl4 after hPON1 Q genedelivery.bThe serum PON1 activity was determined according to its ability tohydrolyze phenyl acetate as described in Materials and Methods.∗∗Denotes a significant (P < 0.01) difference between the CCl4 group andthe control group in the same column.##Denotes a significant (P < 0.01) difference between the EP group andthe CCl4 group in the same column.

remained at nearly the same level as those in the controlgroup, demonstrating that hPON1Q over-expression wascapable of lowering the mice oxidative state induced byCCl4 treatment.

Histological analysis

The extent of CCl4-induced liver injury was further ana-lyzed through histopathological observations. For eachgroup, ten different sections were stained with H&E andexamined under light microscopy. In the CCl4 group,

hepatocyte necrosis was the predominant histopatholog-ical lesion and the affected livers displayed hemorrhage,vacuolar change, and hydropic degeneration of hepato-cytes. In contrast, only spotty areas of cell death werepresent in the hPON1 Q gene transferred mice. Accord-ing to statistical analysis results, the degree of liverinjury in the EP group was significantly lower than thatin the CCl4 group (P < 0.01), revealing that the micepossessing higher PON1 potentials suffered less injurywhen exposed to CCl4. Representative sections and hep-atocyte necrosis statistics of each group are shown inFigure 3.

Discussion

Administration of CCl4 is an established experimentalmodel of severe toxic liver injury involving generation ofoxidative stress and frequently used for the screeningof anti-hepatotoxic and/or hepatoprotective activitiesof drugs [15]. In vivo, CCl4 is metabolized into thetrichloromethyl radical and other oxidant species, whichcan lead to the disruption of structural and functionalintegrity in liver. After injury is initiated from theblockage of lipoprotein secretion and the accumulationof lipids in the liver by the trichloromethyl radical, lipidperoxidation is triggered. In the present study, the declinein serum PON1 activity upon CCl4 administration mightrelate to both the reduction of PON1 synthesis and theabnormalities of HDL in the primary injury process [16].

Figure 3. Protective effects of hPON1 Q gene delivery against liver injury in CCl4-treated mice. H&E staining was performedon paraffin-embedded sections of liver tissues. Sections from normal mouse liver were used as a control (A). Representativestained histological section from mouse of the CCl4 group (B) without hPON1 Q gene delivery exhibited extensive centrilobularhepatocellular necrosis, hemorrhage, hydropic degeneration and vacuolar change, whereas the counterpart from the EP group(C) with hPON1 Q gene delivery exhibited much less hepatocellular injury. After the staining, evaluation of hepatocyte necrosiswas digitalized (D). Necrosis was evaluated according to necrotic foci in mm2. Values were means ± standard deviation (SD) of tensections from different mice. The difference between the EP group (B) and the CCl4 group (C) was very significant (P < 0.01)



As already proposed for atherosclerosis, oxidativestress related molecules were found to act as mediatorsto modulate tissue and cellular events responsible forthe progression of liver fibrosis. Several experimentalreports have stated that antioxidant treatment in vivo waseffective in preventing or reducing liver fibrosis. Sometraditional drugs, such as vitamin E and L-carnitine, wereused to protect liver against CCl4-induced damage in theinitial phase by preventing lipid peroxidation [17]. PON1was able to hydrolyze specific oxidized phospholipids aswell as cholesteryl linoleate hydroperoxide. A peroxidase-like activity of PON1 may eliminate such lipid peroxidesand prevent the initiation of LDL oxidation inducedby cations and other oxidants. PON1 can also interactwith and remove specific lipid peroxides by actingstoichiometrically as a ‘suicidal enzyme’ with oxidizedlipids as a ‘biological buffer’. Consequently, the loweringof lipid peroxidation by PON1 can lead to alleviation ofhepatic damage in the initial phase.

The aim of this study was to confirm in vivo that PON1possesses antioxidant properties and over-expression ofhuman PON1 could protect liver against CCl4-induceddamage in a mouse model. Using an intramuscularelectro-transfer method, we demonstrated that elevatedlevels of PON1 significantly reduced the extent of injurycaused by CCl4 administration, as indicated by bothhistological observation and aminotransferase evaluation.In addition, the hPON1 Q gene transferred mice alsoexhibited lower degrees of oxidative stress, reflectedby the significantly reduced serum MDA concentrationin contrast to their control counterparts. Moreover,correlations between enhanced expression of PON1 andreduced generation of oxidative stress-related moleculeswere observed (Table 1). These data provide compellingand mechanistic evidence for the importance of PON1 inregulation of liver injury.

Nonviral gene delivery methods offer an attractive alter-native to viruses for gene therapy because they offer genetransfer without the potential risks of immunogenicityand cell transformation to patients associated with viralvectors. However, most of them generally suffer frompoor transfection efficiency and only short-term expres-sion [18–20]. Electroporation can efficiently introducenaked plasmid DNA into a broad range of cells and tissuesin vitro and in vivo [21]. Skeletal muscle is easily acces-sible, well vascularized and multinucleate, which makesit a suitable platform to provide a source of systemicallydistributed therapeutic proteins [22]. Previous reportsdemonstrated that intramuscular electro-transfer couldincrease the expression of reporter or therapeutic geneby 2 to 4 orders of magnitude in contrast to the nakedDNA injection and the expression could last for up to9 months [23]. Electroporation could also achieve similartransduction efficiencies as viral transduction deliveredby intramuscular injection [24,25], whereas it was safer,less expensive and easier for manipulation than viralvectors [26]. In previous reports, the plasmid-containingdystrophin or insulin precursor gene has already been

transferred intramuscularly into model mice by electro-poration and the target proteins were detected in theskeletal muscle cells or serum [27]. Consequently, themuscle-synthesized protein fulfilled therapeutic functionsand the symptoms of model animals were alleviated.Therefore, transferring genes into muscle by electropora-tion might be an effective method for gene therapy andhave great potential in clinical applications. In the currentstudy, we demonstrated the feasibility of intramuscularelectro-transfer for delivery and long-term expression ofthe hPON1 Q gene carried by a pcDNA3.0 recombinantplasmid. Expression of hPON1 Q in mouse muscle wasverified by RT-PCR.

In summary, our work opens the way for somaticdelivery and expression of the hPON1 Q gene in vivofor the purposes of gene therapy in an animal model ofliver injury. The present findings supported that muscle-expressed hPON1 Q, by virtue of its potency to alleviateoxidative stress, could protect mice from suffering CCl4-induced liver damage. Mechanism of the protection mightexist in attenuation of lipid peroxidation of injuredliver microsomes by HDL-associated PON1. Nevertheless,the practical results of our study suggest a potentiallytherapeutic method of nonviral delivery for liver injuries.

Acknowledgements

We thank the Center for Disease Control of Jiangsu Provincefor technical support. The study was supported by the NationalNatural Science Foundation of China (Grant No. 30670858).

References

1. Friedman SL. Molecular regulation of hepatic fibrosis, anintegrated cellular response to tissue injury. J Biol Chem 2000;275: 2247–2250.

2. Friedman SL, Maher JJ, Bissell DM. Mechanisms and therapy ofhepatic fibrosis: report of the AASLD single topic basic researchconference. Hepatology 2000; 32: 1403–1408.

3. Pinzani M, Gentilini P. Biology of hepatic stellate cells and theirpossible relevance in the pathogenesis of portal hypertension incirrhosis. Semin Liver Dis 1999; 19: 397–410.

4. Pietrangelo A. Iron, oxidative stress and liver fibrogenesis. JHepatol 1998; 28(Suppl 1): 8–13.

5. Poli G, Parola M. Oxidative damage and fibrogenesis. FreeRadical Biol Med 1997; 22: 287–305.

6. Parola M, Bellomo G, Robino G, et al. 4-Hydroxynonenal asa biological signal: molecular bases and pathophysiologicalimplications. Antioxidant Redox Signaling 1999; 1: 255–284.

7. Richard WJ, Sara PD. The importance of high-densitylipoproteins for paraoxonase-1 secretion, stability, and activity.Free Radical Biol Med 2004; 37: 1986–1994.

8. Aviram M. Introduction to the serial review on paraoxonases,oxidative stress, and cardiovascular diseases. Free Radical BiolMed 2004; 37: 1301–1303.

9. Zhu J, Ze Y, Zhang C, et al. High level expression of recombinanthuman paraoxonase1 Q in silkworm larvae (Bombyx mori). ApplMicrobiol Biotechnol 2006; 72: 103–108.

10. Li H, Liu D, Liang C. Paraoxonase gene polymorphisms,oxidative stress, and diseases. J Mol Med 2003; 81: 766–769.

11. Aviram M, Rosenblat M, Billecke S, et al. Human serumparaoxonase (PON1) is inactivated by oxidized low densitylipoprotein and preserved by antioxidants. Free Radical Biol Med1999; 26: 892–904.


100 C. Zhang et al.

12. Ferre N, Camps J, Prats E, et al. Serum paraoxonase activity: anew additional test for the improved evaluation of chronic liverdamage. Clin Chem 2002; 48: 261–268.

13. Ferre N, Marsillach J, Camps J, et al. Paraoxonase-1 is associatedwith oxidative stress, fibrosis and FAS expression in chronic liverdiseases. J Hepatol 2006; 45: 51–59.

14. Ohkawa H, Ohishi N, Yagi K. Assay for lipid peroxides in animaltissues by thiobarbituric acid reaction. Anal Biochem 1979; 95:351–358.

15. Lee KJ, Woo ER, Choi CY, et al. Protective effect of acteoside oncarbon tetrachloride-induced hepatotoxicity. Life Sci 2004; 74:1051–1064.

16. Ferre N, Camps J, Cabre M, et al. Hepatic paraoxonase activityalterations and free radical production in rats with experimentalcirrhosis. Metabolism 2001; 9: 997–1000.

17. Demirdag K, Bahcecioglu IH, Ozercan IH, et al. Role of L-carnitine in the prevention of acute liver damage induced bycarbon tetrachloride in rats. J Gastroentrol Hepatol 2004; 19:333–338.

18. Neu M, Fischer D, Kissel T. Recent advances in rational genetransfer vector design based on poly(ethylene imine) and itsderivatives. J Gene Med 2005; 7: 992–1009.

19. White RE, Wade-Martins R, Hart SL, et al. Functional delivery oflarge genomic DNA to human cells with a peptide-lipid vector. JGene Med 2003; 5: 883–892.

20. Wells DJ. Gene therapy progress and prospects: electroporationand other physical methods. Gene Ther 2004; 11:1363–1369.

21. Gehl J. Electroporation: theory and methods, perspectives fordrug delivery, gene therapy and research. Acta Physiol Scand2003; 177: 437–447.

22. McMahon JM, Wells DJ. Electroporation for gene transfer toskeletal Muscles. Biodrugs 2004; 18: 155–165.

23. 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.

24. Mir LM, Bureau MF, Rangara R, et al. High level in vivo geneexpression after electric pulsemediated gene transfer intoskeletal muscle. C. R. Acad. Sci. III 1998; 321: 893–899.

25. 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.

26. Wong SH, Lowes KN, Quigley AF, et al. DNA electroporationin vivo targets mature fibres in dystrophic mdx muscle. FEBSLett 2001; 495: 16–20.

27. Dong Y, Tang JG. Gene therapy for streptozotocin-induceddiabetic mice by electroporational transfer of naked humaninsulin precursor DNA into skeletal muscle in vivo. FEBS Lett2001; 495: 16–20.


Documents

Transgene expression of human PON1 Q in mice protected the liver against CCl4-induced injury