Dexamethasone-conjugated polyethylenimine as an efficient gene carrier with an anti-apoptotic effect to cardiomyocytes

THE JOURNAL OF GENE MEDICINE R E S E A R C H A R T I C L EJ Gene Med 2009; 11: 515–522.Published online 19 March 2009 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/jgm.1320

Dexamethasone-conjugated polyethylenimine asan efficient gene carrier with an anti-apoptoticeffect to cardiomyocytes

Hyunjung Kim1

Hyun Ah Kim1

Yun Mi Bae2

Joon Sig Choi2*Minhyung Lee1*

1Department of Bioengineering,College of Engineering, HanyangUniversity, Seoul, Korea2Department of Biochemistry, Collegeof Natural Science, ChungnamNational University, Daejeon, Korea

*Correspondence to: Joon Sig Choi,Department of Biochemistry,College of Natural Science,Chungnam National University,Daejeon 305-764, Korea.E-mail: [email protected]

*Correspondence to: Minhyung Lee,Department of Bioengineering,College of Engineering, HanyangUniversity, 17 Haengdang-dong,Seongdong-gu, Seoul 133-791,Korea.E-mail: [email protected]

Received: 10 October 2008Revised: 21 January 2009Accepted: 13 February 2009

Abstract

Background Dexamethasone is a potent glucocorticoid with anti-inflammatory effects. Dexamethasone can protect ischemic cardiomyocytesfrom apoptosis. To apply the anti-apoptotic effect of dexamethasone toischemic disease gene therapy, dexamethasone-conjugated polyethylenimine(PEI-Dexa) was synthesized and evaluated as an anti-apoptotic gene carrier.

Methods PEI-Dexa was synthesized with low molecular weight polyethylen-imine (PEI2K, 2 kDa). The transfection efficiency and cytotoxicity of PEI-Dexawere evaluated by luciferase assay and the MTT assay. To evaluate the anti-apoptotic effect, PEI-Dexa/DNA complex was transfected into cells and thecells were treated with H2O2. Cell viability and apoptosis level were measuredby the MTT assay and caspase-3 assay, respectively.

Results A transfection assay into H9C2 rat cardiomyocytes showed thatPEI-Dexa had the highest transfection efficiency at an 8 : 1 weight ratio (PEI-Dexa/DNA). At this ratio, PEI-Dexa had higher transfection efficiency thanhigh molecular polyethylenimine (PEI25K, 25 kDa) and PEI2K. In addition,the cytotoxicity of PEI-Dexa was lower than that of PEI25K. To evaluate theanti-apoptotic effect, PEI-Dexa/pSV-Luc or PEI2K/pSV-Luc was transfectedinto H9C2 cells and the cells were treated with H2O2. PEI-Dexa was found toreduce caspase-3 activity and increase cell viability compared to PEI2K. Hemeoxygenase-1 (HO-1) can protect ischemic cardiomyocytes from apoptosis.Therefore, pSV-HO-1 was cloned and transfected into H9C2 cells using PEI-Dexa. The cells transfected with PEI-Dexa/pSV-HO-1 complex had lowercaspase-3 activity and higher viability than the cells transfected with PEI-Dexa/pSV-Luc complex after the H2O2 treatment.

Conclusions PEI-Dexa is an efficient gene carrier with an anti-apoptoticeffect and may be useful for anti-apoptotic gene therapy in combination withpSV-HO-1. Copyright 2009 John Wiley & Sons, Ltd.

Keywords dexamethasone; gene carrier; heme oxygenase-1; hydrogen peroxide;polyethylenimine

Introduction

Polymeric gene carriers have been developed as an alternative to viralcarriers because of the intrinsic problems of viral carriers, such asimmunogenecity and insertional mutagenesis [1]. Polyethylenimine (PEI)

Copyright 2009 John Wiley & Sons, Ltd.

516 H. Kim et al.

is one of the most widely used polymeric gene carriers[2–4]. Many modifications to PEI have been tried forimproved gene delivery efficiency, low cytotoxicity, andtargeting gene delivery [2]. Recently, it was reportedthat conjugation of dexamethasone could improve thegene delivery efficiency of polyamidoamine or PEI [5,6].Dexamethasone is a potent glucocorticoid with anti-inflammatory activity. Dexamethasone binds to a glu-cocorticoid receptor after cellular entry, and the recep-tor/dexamethasone complex is subsequently translo-cated into the nucleus [7,8]. Therefore, dexamethasone-conjugated polymers/DNA complexes were efficientlydelivered into the nucleus with the glucocorticoid recep-tor, resulting in an increase of transgene expression[5,6]. In addition, the glucocorticoid receptor dilatesthe nuclear pore, which may be beneficial for nuclearentry of polymer/DNA complex [7]. One of the importantbiological activities of dexamethasone is a cytoprotec-tive effect of cardiomyocytes from apoptosis [9–11].Therefore, such a dexamethasone-conjugated polymermay also have an anti-apoptotic effect in hyperoxic car-diomyocytes as a dexamethasone derivative under stressconditions.

Oxidative stress results from an imbalance betweenformation and neutralization of reactive oxygen species(ROS) [12]. ROS include the superoxide anion radical(O2

−·) [13]. The superoxide radicals are formed invarious biochemical reactions and cellular functions,such as inflammation, ischemia, infection and cancerformation [14]. Oxidative stress is a condition inducingapoptosis and involved in various diseases, includingheart failure [15–17]. Hydrogen peroxide (H2O2) is aROS and widely used as a cytotoxic reagent [18]. Inthe present study, H2O2 was used to induce apoptosis ofcardiomyocytes.

Heme oxygenases (HOs) are the rate-limiting enzymesin heme degradation. Three HO isoforms, HO-1, HO-2 andHO-3, have been identified. HO-1 expression in the heartis induced by stressful and inflammatory stimuli [19], andit has a protective effect on cardiomyocytes [20]. HO-1can protect cells from the angiotensin II-induced apoptosisby inhibiting the mitogen-activated protein kinase cascade[21,22], and gene therapy with the HO-1 gene has beeninvestigated as a treatment for ischemic heart disease[23,24].

In the present study, dexamethasone-conjugatedpolyethylenimine (PEI-Dexa) (Figure 1) was evaluated asa gene carrier with anti-apoptotic activity. Gene deliveryefficiency and cytotoxicity to H9C2 rat cardiomyocyteswere also evaluated. Furthermore, the anti-apoptoticactivity of PEI-Dexa was measured in the H2O2-treatedH9C2 cells. In addition, pSV-HO-1 was constructed andused as a model therapeutic gene for PEI-Dexa-mediatedgene delivery. The results obtained suggest that PEI-Dexa is an anti-apoptotic gene carrier and may be usefulfor anti-apoptotic gene therapy in combination withpSV-HO-1.

Figure 1. Chemical structure of PEI-Dexa

Materials and methods

Synthesis of PEI-Dexa

PEI-Dexa was synthesized as described previously [6].Briefly, PEI (PEI2K, 2 kDa) was dissolved in 1.8 ml ofanhydrous dimethyl sulfoxide (DMSO) with a two-foldmolar excess of Traut’s reagent and dexamethasone-21-mesylate. The reaction was allowed to proceed for4 h at room temperature and was quenched by theaddition of cold ethyl acetate. The precipitated productwas solubilized in water and dialysed for 1 day againstpure water using a dialysis membrane (molecular weightcut-off = 1000). The mixture was freeze-dried, and theproduct was obtained (46%, yield). From the 1H-NMRdata (300 MHz; Korea Basic Science Institute, Seoul,Korea), it was calculated that 1 mol of dexamethasonewas conjugated per mol of PEI.

Critical micelle concentrationmeasurement

The critical micelle concentration (CMC) of PEI-Dexawas determined by the fluorescence of pyrene monomer[25,26]. A stock solution of pyrene (6 × 10−6 M) wasprepared in acetone and added to deionized water togive a pyrene concentration of 6 × 10−7 M. The solutionwas heated at 65 ◦C for 3 h to evaporate acetone from thesolution and stored at 5 ◦C for 24 h. Then, the acetone-freepyrene solution was mixed with the same volume of PEI-Dexa solution. PEI-Dexa solution was prepared at variousconcentration ranges. The fluorescence emission wasmeasured using a luminescence spectrofluorophotometer(Perkin Elmer, Waltham, MA, USA). Pyrene was excitedat 336 nm and its emission was recorded at 373 nmand 383 nm, which correspond to the first and thirdvibrational peaks, the slit of excitation and emission was5 nm.

Acid-base titration

The proton buffering ability of PEI-Dexa was determinedby acid-base titration as previously described [27]. Briefly,

Copyright 2009 John Wiley & Sons, Ltd. J Gene Med 2009; 11: 515–522.DOI: 10.1002/jgm

PEI-Dexa as an anti-apoptotic gene carrier 517

1 mg of poly-L-lysine (PLL), PEI (2 kDa) or PEI-Dexawas dissolved in 10 ml of 150 mM NaCl. One hundredmicroliters of 1 N NaOH was added to this polymersolution, and the pH was measured. The polymer solutionwas titrated with an increasing volume of 0.1 N HCl, andthe pH was measured at every point.

Preparation of plasmid DNA

The human heme oxygenase-1 cDNA (Genebank acces-sion number: NM002133) was cloned by reversetranscriptase-polymerase chain reaction (RT-PCR) usingtotal RNA from 293 cells as a template. Thesequences of the primers were: forward primer, 5′-CCCAAGCTTATGGAGCGTCCGCAACCCG-3′, backwardprimer, 5′-GCTCTAGAGCATTCACATGGCATAAAGC-3′.HinDIII and XbaI sites were incorporated into the for-ward and backward primers, respectively, for cloningconvenience (enzyme sites are underlined). The amplifiedHO-1 cDNA was inserted into pSV-Luc (pGL3-promoter;Promega, Madison, WI, USA) at the place of the luciferasecDNA, resulting in the construction of pSV-HO-1. Theconstruction of the plasmid was confirmed by restrictionenzyme analysis and direct sequencing.

The plasmids was transformed in Escherichia coli DH5α

and amplified in Terrific Broth media at 37 ◦C overnight.The plasmid was purified using the Maxi plasmidpurification kit (Qiagen, Valencia, CA, USA). Purifiedplasmid was dissolved in Tris-ethylenediaminetetraaceticacid buffer, and its purity and concentration weredetermined by ultraviolet absorbance at 260 nm. Theoptical density ratios at 260–280 nm were in the rangeof 1.7–1.8.

Cell culture and transfection

H9C2 rat cardiomyocytes were purchased from ATCC(Manassas, VA, USA). The cells were maintained in Dul-becco’s modified Eagle’s medium (DMEM) supplementedwith 10% fetal bovine serum (FBS) in a 5% CO2 incubator.For the transfection assays, the cell was seeded at a densityof 1 × 105 cells/well in six-well flat-bottomed microassayplates (Falcon Co., Becton Dickenson, Franklin Lakes,NJ, USA) 24 h before the transfection. PEI-Dexa/pSV-Luccomplexes were prepared at various weight ratios in 5%glucose solution. PEI2K/pSV-Luc and high molecular PEI(PEI25K, 25 kDa)/pSV-Luc complexes were prepared at a40 : 1 and 5 : 1 N/P ratio in 5% glucose solution, respec-tively, based on previous reports [27]. Before transfection,the medium was replaced with 2 ml of fresh DMEM with-out FBS. Then, the polymer/pSV-Luc complexes wereadded to the cells. The amount of plasmid was fixed at2 µg/well. The cells were then incubated for 4 h at 37 ◦Cin a 5% CO2 incubator. After 4 h, the transfection mix-tures were removed, and 2 ml of fresh DMEM mediumcontaining FBS was added. The cells were incubated foran additional 44 h at 37 ◦C.

Luciferase assay

After transfection, the cells were washed twice withphosphate-buffered saline (PBS), and 200 µl of reporterlysis buffer (Promega) was added to each well. After15 min of incubation at room temperature, the cellswere harvested and transferred to microcentrifuge tubes.After 15 s of vortexing, the cells were centrifuged at13 000 r.p.m. 10 000 g for 5 min. The extracts weretransferred to fresh tubes and stored at −70 ◦C until use.The protein concentration of the extract was determinedwith a BCA protein assay kit (Pierce, Iselin, NJ, USA).Luciferase activity was measured in terms of relativelight units (RLU) using a 96-well plate luminometer(Berthold Detection System GmbH, Pforzheim, Germany).The luciferase activity was monitored and integrated overa period of 20 s The final luciferase values were reportedin terms of RLU/mg total protein.

Cytotoxicity assay

Evaluation of cytotoxicity was performed by the MTTassay. H9C2 cells were seeded at a density of 5 × 104

per well in 24-well plates and incubated for 24 hbefore transfection. The PEI2K/pSV-Luc complex andPEI25K/pSV-Luc were prepared at a 5 : 1 N/P ratio in5% glucose solution. The PEI-Dexa/pSV-Luc complex wasprepared at an 8 : 1 weight ratio (PEI-Dexa/plasmid). Theamount of plasmid was fixed at 0.5 µg/well. The mediumwas replaced with fresh DMEM medium without FBSbefore transfection, and the polymer/pSV-Luc complexeswere added to the cells. After incubation at 37 ◦C for4 h, the transfection mixture was replaced with 500 µlof fresh DMEM medium supplemented with 10% FBS,and the cells were incubated for an additional 24 or72 h at 37 ◦C. After incubation, MTT solution in PBS wasadded. The cells were incubated for an additional 4 h at37 ◦C. After the incubation, MTT-containing medium wasaspirated off and 750 nm of DMSO was added to dissolvethe formazan crystal formed by live cells. Absorbance wasmeasured at 570 nm. The cell viability (%) was calculatedaccording to the equation:

Cell viability (%) = OD570(sample)/OD570(control) × 100

where the OD570(sample) represents the measurement fromthe well treated with polymer/plasmid DNA complex andthe OD570(control) represents the measurements from thewells treated with 5% glucose.

Caspase-3 assay

Apoptosis of the transfected cells was measured usingCaspase-Glo 3/7 Assay reagent (Promega, Madison, WI,USA). For transfection, the H9C2 cells were seeded ata density of 5 × 103 cells/well in 96-well flat-bottomedmicroassay plates (Falcon Co., Becton Dickenson) 24 h


518 H. Kim et al.

before the transfection. PEI-Dexa/pSV-Luc complexeswere prepared at various weight ratios. PEI2K/pSV-Luccomplexes were prepared at a 5 : 1 N/P ratio. Beforetransfection, the medium was replaced with 200 µl offresh medium without FBS. Then, the polymer/pSV-Luccomplexes were added to the cells. The amount of plasmidwas fixed at a 0.2 µg/well. The cells were then incubatedfor 4 h at 37 ◦C in a 5% CO2 incubator. After 4 h, thetransfection mixtures were removed, and 100 µl of freshDMEM medium containing H2O2 (Junsei, Tokyo, Japan).The final H2O2 concentration was 5 µM. The cells wereincubated for an additional 12 h at 37 ◦C in a 5% CO2 incu-bator. After the incubation, 50 µl of Caspase-Glo reagentwas added to each well, and samples are incubated atroom temperature for 1 h. The luminescence of each sam-ple was measured in terms of RLU, using a 96-well plateluminometer (Berthold Detection System GmbH).

Enzyme-linked immunosorbent assay(ELISA) of HO-1

ELISA was performed using a human HO-1 ELISA Kit(Assay Designs, Ann Arbor, MI, USA) to measure thehuman heme oxygenase-1 in cell lysates. Briefly, thecells were harvested and lysed using extraction reagent(Assay Designs). Fifty microliters of the samples wereadded to the designated wells. One hundred microlitersof anti-human HO-1 rabbit polyclonal antibody wasadded to each well, and the plate was incubated atroom temperature for 1 h. After washing, 100 µl ofthe horseradish peroxidase conjugated to anti-rabbitimmunoglobulin G was added into each well, and theplate was incubated at room temperature for 30 min. Afterwashing, 100 µl of the stabilized tetramethylbenzidinesubstrate was added to each well, and the platewas incubated at room temperature for 15 min. Afterincubation, the stop solution was added to the wells, andthe absorbance was measured at 450 nm.

Statistical analysis

Results are reported as the mean ± SD. The comparisonof luciferase activity or HO-1 concentration was madeby Student’s t-test. p < 0.05 was considered statisticallysignificant.

Results

Physical characterization of PEI-Dexa

The CMC of PEI-Dexa was determined by the fluorescenceof pyrene monomer [25,26]. The CMC value of PEI-Dexawas observed to be 1.2 mg/ml.

PEI has been reported to have a proton-buffering effect[28]. This effect facilitates disruption of the endosomeand endosomal escape of the polymer/plasmid complex

in the transfection process. PEI-Dexa was synthesizedwith PEI2K and dexamethasone (Figure 1). Therefore,PEI-Dexa may have a proton-buffering effect due to thePEI segment. To confirm the buffering effect of PEI-Dexa, acid-base titration was performed. The acid-basetitration profile was obtained for PLL, PEI2K and PEI-Dexa(Figure 2). The initial pH of these polymers was adjustedto approximately 9.4, and 0.1 N HCl was gradually addedto the polymer solutions. As a result, the PLL solutionshowed a sudden drop of pH after the addition of 0.1 ml of0.1 N HCl. This result suggests that PLL has little bufferingeffect. However, PEI and PEI-Dexa showed considerablebuffering capacity, although the buffer capacity of PEI-Dexa was slightly lower than that of PEI.

Transfection efficiency of PEI-Dexa tocardiomyocytes H9C2 cells

To evaluate the transfection efficiency, an in vitrotransfection assay was performed with PEI-Dexa. Tooptimize the transfection conditions of PEI-Dexa, PEI-Dexa/pSV-Luc complexes were prepared at various weightratios and transfected into H9C2 cells. As shown inFigure 3, the highest transfection efficiency of PEI-Dexawas obtained at an 8 : 1 weight ratio (PEI-Dexa/plasmid).Therefore, PEI-Dexa/plasmid complex was prepared at a8 : 1 weight ratio in the subsequent experiments.

To compare the transfection efficiency of PEI-Dexa withPEI2K and PEI25K, the transfection into H9C2 cell wasperformed with PEI-Dexa, PEI2K and PEI25K Based on aprevious study [27], PEI2K/plasmid and PEI25K/plasmidcomplexes were prepared at 40 : 1 and 5 : 1 N/P ratios,respectively. PEI-Dexa/plasmid complexes were preparedat an 8 : 1 weight ratio (PEI-Dexa/plasmid). As a result,PEI-Dexa showed higher transfection efficiency thanPEI2K and PEI25K (Figure 4). This higher transfectionefficiency may be due to efficient nuclear translocation ofthe PEI-Dexa/plasmid complex, which is mediated by aglucocorticoid receptor [5,6].

Figure 2. Acid-base titration. One milligram of PLL, PEI2K,or PEI-Dexa was dissolved in 10 ml of 150 mM NaCl. Twentymicroliters of 1 N NaOH were added to each polymer solution.The polymer solution was titrated with an increasing volume of0.1 M HCl



Figure 3. Transfection efficiency of PEI-Dexa into H9C2 car-diomyocytes depending on weight ratio. PEI-Dexa/pSV-Luc com-plexes were prepared at various weight ratios and transfectedinto H9C2 cells. Transfection efficiency was measured by aluciferase assay. The data are expressed as the mean ± SD of sixexperiments. ∗p < 0.01 compared to 4 : 1 and 6 : 1 weight ratios

Figure 4. Transfection efficiency of PEI2K, PEI25K or PEI-Dexainto H9C2 cells. Polymer/pSV-Luc complexes were preparedas described in the Materials and methods. The transfectionefficiency of each complex was measured by a luciferase assay.The data are expressed as the mean ± SD of six experiments.∗p < 0.01 compared to PEI2K and PEI25K

Cytotoxicity of PEI-Dexa

To evaluate the cytotoxicity of PEI-Dexa, PEI-Dexa/plasmid complex was transfected into H9C2 cells. Thecytotoxicity of PEI-Dexa was compared with that ofPEI2K or PEI25K. The complexes were transfected intoH9C2 cells, and the cytotoxicity was evaluated by MTTassay at 24 or 72 h after transfection. PEI2K showed anapproximately 100% cell viability compared to the control(Figures 5A and 5B). Although PEI-Dexa had a highertoxicity than PEI2K, PEI-Dexa showed a lower cytotoxicityto H9C2 cells than PEI25K (Figures 5A and 5B).

Figure 5. Cytotoxicity of PEI2KD, PEI25KD or PEI-Dexa to H9C2cells. Polymer/pSV-Luc complexes were prepared as describedin the Materials and methods and transfected into H9C2 cells.At 24 h (A) or 72 h (B) after transfection, cell viability wasmeasured by the MTT assay. The data are expressed as themean ± SD of five experiments. ∗p < 0.05 compared to control,PEI2K and PEI25K

Anti-apoptotic effect of PEI-Dexa

To assess the anti-apoptotic effect of PEI-Dexa, PEI-Dexa/pSV-Luc complex was transfected into H9C2 cellsand the cells were treated with H2O2. Apoptosis levelwas measured by caspase-3 assay and cell viabilitywas determined by the MTT assay. PEI2K/pSV-Luccomplex was also transfected as a control. The cellstransfected with PEI-Dexa/pSV-Luc complex had highercell viability than the cells transfected with PEI/pSV-Luccomplex (Figure 6A). Furthermore, the cells transfectedwith PEI-Dexa/pSV-Luc had lower caspase-3 activity thanPEI2K/pSV-Luc complex (Figure 6B). This result indicatesthat PEI-Dexa has anti-apoptotic effect.

Anti-apoptotic effect of HO-1 incombination with PEI-Dexa

The human HO-1 gene was delivered to H9C2 cells asa model gene. The human HO-1 cDNA was cloned by


520 H. Kim et al.

Figure 6. The effect of PEI-Dexa on the H2O2 treated H9C2 cells.Polymer/pSV-Luc complexes were prepared as described in theMaterials and methods. (A) Cell viability was determined bythe MTT assay. (B) Apoptosis level was measured by caspase-3activity. The data are expressed as the mean ± SD of fiveexperiments. ∗p < 0.05 compared to PEI2K/pSV-Luc

Figure 7. The structure of pSV-Luc and pSV-HO-1

RT-PCR using total RNA from 293 cells as a template. pSV-HO-1 was constructed with the HO-1 cDNA (Figure 7). Aprevious study [29] suggested that dexamethasone mightinhibit the HO-1 promoter activity, resulting in reductionof the endogenous HO-1 expression. Therefore, HO-1gene delivery to cardiomyocytes using PEI-Dexa maycompensate for the possible drawback of dexamethasone.In the transfection assay, HO-1 was expressed in the pSV-HO-1 transfected H9C2 cells (Figure 8). PEI-Dexa/pSV-HO-1 complex was transfected into the H9C2 cells andthe cells were incubated with 5 µM H2O2. To evaluatethe anti-apoptotic effect of pSV-HO-1 gene, caspase-3 assay and the MTT assay were performed afterthe transfection with PEI-Dexa/pSV-HO-1 complex. ThePEI-Dexa/pSV-Luc complex was also transfected as acontrol. After incubation with 5 µM H2O2, PEI-Dexa/pSV-HO-1 complex had a higher cell viability (Figure 9A)and a lower caspase-3 activity (Figure 9B) than PEI-Dexa/pSV-Luc. Therefore, the results suggest that pSV-HO-1 in combination with PEI-Dexa can enhance theanti-apoptotic effect.

Discussion

In the present study, we demonstrate that PEI-Dexa isa gene carrier with an anti-apoptotic effect on H9C2

Figure 8. HO-1 expression after transfection of pSV-HO-1 intoH9C2 cells. pSV-Luc and pSV-HO-1 were transfected into H9C2cells using PEI-Dexa as a gene carrier. HO-1 expression levelswere measured by HO-1 ELISA. The data are expressed as themean ± SD of four experiments. ∗p < 0.01 compared to pSV-Luc

Figure 9. Effect of HO-1 on the H2O2 treated H9C2 cells.PEI-Dexa/cDNA complexes were prepared as described in theMaterials and methods section. (A) Cell viability was determinedby the MTT assay. (B) Apoptosis level was measured by caspase-3activity. The data are expressed as the mean ± SD of fiveexperiments. ∗p < 0.05 compared to PEI-Dexa/pSV-Luc

cardiomyocytes. Previously, we showed that PEI-Dexadelivered plasmid efficiently to 293 or HepG2 cells dueto an efficient nuclear translocation effect mediatedby the glucocorticoid receptor [6]. The glucocorticoidreceptor translocates into the nucleus upon binding to itsligand, such as dexamethasone [8]. Indeed, the nucleartranslocation of the dexamethasone-conjugated polymerwas previously confirmed using confocal microscopy ora glucocorticoid receptor inhibitor [5,6]. In addition,the glucocorticoid receptor has the effect of nuclearpore dilation when it binds to its ligand, which is alsobeneficial for nuclear entry of polymer/DNA complex [7].Furthermore, dexamethasone has been widely used asan anti-inflammatory drug and it was recently reportedthat dexamethasone reduced apoptotic cell death underhypoxia [9–11]. Based on these results, we evaluatedPEI-Dexa as a gene carrier to cardiomyocytes with ananti-apoptotic effect.



PEI-Dexa has some advantages as a gene carrier tocardiomyocytes. First, PEI-Dexa has higher transfectionefficiency than PEI25K. Previously, the PEI2K/plasmidratio was optimized for gene delivery [27]. PEI2K hadmuch lower cytotoxicity than PEI25K [27]. Due to thesmaller size of PEI2K, the PEI2K/plasmid complex is not astight as the PEI25K/plasmid complex. Therefore, a largeramount of PEI2K was required for gene delivery comparedto PEI25K. PEI2K showed its maximum transfection at a40 : 1 N/P ratio in our previous research [27]. At this ratio,the transfection efficiency of PEI2K was much lower thanthat of PEI25K, suggesting that modification of PEI2K isrequired for practical application. One of the examplesis a cholesterol-conjugated PEI, which was named as awater-soluble lipopolymer (WSLP) [27]. WSLP has ashigh transfection efficiency as PEI25K without significanttoxicity into various cells [27,30]. This high transfectionefficiency of WSLP may be due to efficient cellular uptakemediated by the low-density lipopolymer receptor [27].In the present study, the conjugation of dexamethasonealso increased transfection efficiency significantly, andPEI-Dexa had higher transfection efficiency to H9C2cells than PEI25K. This higher transfection efficiency ofPEI-Dexa may be due to an efficient translocation ofPEI-Dexa/plasmid into the nucleus.

Second, PEI-Dexa can protect cardiomyocytes fromapoptosis. The glucocorticoid receptor can induce endoge-nous gene expression after ligand binding. Previousstudies [9,11] showed that dexamethasone induced theexpression of anti-apoptotic Bcl-xL and hsp72. The induc-tion of Bcl-xL may reduce the release of cytochrome cfrom the mitochondrial inner membrane and inhibit theactivation of caspases. As demonstrated in the presentstudy, the addition of the PEI-Dexa/plasmid complexto the H2O2-treated cardiomyocytes reduced caspase-3activity (Figures 7 and 9). This can also serve as evidencethat PEI-Dexa retained its physiological effect as a ligandto the glucocorticoid receptor. Therefore, PEI-Dexa mayhave dual functions as (i) an efficient gene carrier and (ii)an anti-apoptotic reagent. Considering that dexametha-sone has been widely used as an anti-inflammatory drug,PEI-Dexa can be applied to various inflammatory diseases,such as rheumatoid arthritis. The intra-articular injectionof dexamethasone was shown to have a therapeutic effecton rheumatoid arthritis [31]. Therefore, the dual func-tions of being a gene carrier and an anti-inflammatorydrug may improve the therapeutic effect of gene therapyfor inflammatory diseases.

Third, PEI-Dexa has lower cytotoxicity than PEI25K.Although PEI2K showed lower cytotoxicity than PEI-Dexa, PEI2K showed much lower transfection efficiencycompared to PEI-Dexa. A previous study suggested thatthe cytotoxicity of a cationic polymer such as PEI is due tothe high charge density of the polymer [32]. Therefore, areduction of charge density would be beneficial for safegene delivery. The lower cytotoxicity of PEI-Dexa maybe due to lower charge density of PEI-Dexa compared toPEI25K.

Dexamethasone was reported to interfere with HO-1 promoter activity and reduce endogenous HO-1expression levels in microvessel endothelial cells [29].This possible drawback of the dexamethasone-conjugatedpolymer can be compensated for by the delivery of theexogenous HO-1 gene.

In summary, PEI-Dexa is an efficient gene carrier intoH9C2 cardiomyocytes with a lower cytotoxicity thanPEI25K. In addition, PEI-Dexa has an anti-apoptoticeffect in the H2O2-treated H9C2 cells. PEI-Dexa-mediateddelivery of pSV-HO-1 into H9C2 cells increased cellviability and decreased caspase-3 activity. Therefore, PEI-Dexa in combination with the HO-1 expression plasmidmay be useful in the development of ischemic myocardiumgene therapy.

Acknowledgements

This research received financial support from the Ministry ofEducation, Science and Technology (M10534030003-08N3403-00310) in Korea.

References

1. Lee M, Kim SW Polyethylene glycol-conjugated copolymers forplasmid DNA delivery. Pharm Res 2005; 22: 1–10.

2. Kang HC, Lee M, Bae YH Polymeric gene carriers. Crit RevEukaryot Gene Expr 2005; 15: 317–342.

3. Han S, Mahato RI, Sung YK, et al. Development of biomaterialsfor gene therapy. Mol Ther 2000; 2: 302–317.

4. Lee M, Kim SW Polymeric gene carriers. Pharm News 2002; 9:407–415.

5. Choi JS, Ko KS, Park JS, et al. Dexamethasone conjugated poly(amidoamine) dendrimer as a gene carrier for efficient nucleartranslocation. Int J Pharm 2006; 320: 171–178.

6. Bae YM, Choi H, Lee S, et al. Dexamethasone-conjugated lowmolecular weight polyethylenimine as a nucleus targetinglipopolymer gene carrier. Bioconjug Chem 2007; 18: 2029–2036.

7. Shahin V, Albermann L, Schillers H, et al. Steroids dilate nuclearpores imaged with atomic force microscopy. J Cell Physiol 2005;202: 591–601.

8. Adcock IM, Caramori G Cross-talk between pro-inflammatorytranscription factors and glucocorticoids. Immunol Cell Biol2001; 79: 376–384.

9. Sun L, Chang J, Kirchhoff SR, et al. Activation of HSFand selective increase in heat-shock proteins by acutedexamethasone treatment. Am J Physiol Heart Circ Physiol 2000;278: H1091–H1097.

10. Reeve JL, Szegezdi E, Logue SE, et al. Distinct mechanisms ofcardiomyocyte apoptosis induced by doxorubicin and hypoxiaconverge on mitochondria and are inhibited by Bcl-xL. J Cell MolMed 2007; 11: 509–520.

11. Chen QM, Alexander D, Sun H, et al. Corticosteroids inhibit celldeath induced by doxorubicin in cardiomyocytes: inductionof antiapoptosis, antioxidant, and detoxification genes. MolPharmacol 2005; 67: 1861–1873.

12. McCord JM Oxygen-derived free radicals in postischemic tissueinjury. N Engl J Med 1985; 312: 159–163.

13. Markesbery WR Oxidative stress hypothesis in Alzheimer’sdisease. Free Radic Biol Med 1997; 23: 134–147.

14. Dypbukt JM, Ankarcrona M, Burkitt M, et al. Differentprooxidant levels stimulate growth, trigger apoptosis, orproduce necrosis of insulin-secreting RINm5F cells. The role ofintracellular polyamines. J Biol Chem 1994; 269: 30553–30560.

15. Crow MT, Mani K, Nam YJ, et al. The mitochondrial deathpathway and cardiac myocyte apoptosis. Circ Res 2004; 95:957–970.


522 H. Kim et al.

16. Murdoch CE, Zhang M, Cave AC, et al. NADPH oxidase-dependent redox signalling in cardiac hypertrophy, remodellingand failure. Cardiovasc Res 2006; 71: 208–215.

17. Singal PK, Khaper N, Farahmand F, et al. Oxidative stress incongestive heart failure. Curr Cardiol Rep 2000; 2: 206–211.

18. Halliwell B, Clement MV, Long LH Hydrogen peroxide in thehuman body. FEBS Lett 2000; 486: 10–13.

19. Kacimi R, Chentoufi J, Honbo N, et al. Hypoxia differentiallyregulates stress proteins in cultured cardiomyocytes: role of thep38 stress-activated kinase signaling cascade, and relation tocytoprotection. Cardiovasc Res 2000; 46: 139–150.

20. Yet SF, Tian R, Layne MD, et al. Cardiac-specific expression ofheme oxygenase-1 protects against ischemia and reperfusioninjury in transgenic mice. Circ Res 2001; 89: 168–173.

21. Foo RS, Siow RC, Brown MJ, et al. Heme oxygenase-1 genetransfer inhibits angiotensin II-mediated rat cardiac myocyteapoptosis but not hypertrophy. J Cell Physiol 2006; 209: 1–7.

22. Hu CM, Chen YH, Chiang MT, et al. Heme oxygenase-1 inhibitsangiotensin II-induced cardiac hypertrophy in vitro and in vivo.Circulation 2004; 110: 309–316.

23. Tang YL, Qian K, Zhang YC, et al. A vigilant, hypoxia-regulatedheme oxygenase-1 gene vector in the heart limits cardiac injuryafter ischemia-reperfusion in vivo. J Cardiovasc Pharmacol Ther2005; 10: 251–263.

24. Tang YL, Tang Y, Zhang YC, et al. Protection from ischemicheart injury by a vigilant heme oxygenase-1 plasmid system.Hypertension 2004; 43: 746–751.

25. Niu L, Nagarajan R, Guan F, et al. Biocatalytic synthesis of multi-block copolymer composed of poly (tetrahydrofuran) and poly(ethylene oxide). J Macromol Sci A 2006; 43: 1975–1981.

26. Dominguez A, Fernandez A, Gonzalez N, et al. Determinationof critical micelle concentration of some surfactants by threetechniques. J Chem Educ 1997; 74: 1227–1231.

27. Lee M, Rentz J, Han SO, et al. Water-soluble lipopolymer as anefficient carrier for gene delivery to myocardium. Gene Ther2003; 10: 585–593.

28. Akinc A, Thomas M, Klibanov AM, et al. Exploringpolyethylenimine-mediated DNA transfection and theproton sponge hypothesis. J Gene Med 2005; 7:657–663.

29. Deramaudt TB, da Silva JL, Remy P, et al. Negative regulationof human heme oxygenase in microvessel endothelial cells bydexamethasone. Proc Soc Exp Biol Med 1999; 222: 185–193.

30. Lee M, Rentz J, Bikram M, et al. Hypoxia-inducible VEGFgene delivery to ischemic myocardium using water-solublelipopolymer. Gene Ther 2003; 10: 1535–1542.

31. Makrygiannakis D, af Klint E, Catrina SB, et al. Intraarticularcorticosteroids decrease synovial RANKL expression ininflammatory arthritis. Arthritis Rheum 2006; 54: 1463–1472.

32. Fischer D, Bieber T, Li Y, et al. A novel non-viral vector forDNA delivery based on low molecular weight, branchedpolyethylenimine: effect of molecular weight on transfectionefficiency and cytotoxicity. Pharm Res 1999; 16: 1273–1279.


Documents

Dexamethasone-conjugated polyethylenimine as an efficient gene carrier with an anti-apoptotic effect to cardiomyocytes