Gene Transfer into Human Hepatoma Cells by Receptor-Associated

Receptor-associated protein (RAP) is a ligand for all members of low-density lipoprotein (LDL) receptor families. RAP is internalized into cells via ...
0 downloads 0 Views 114KB Size
Download PDF
326

Bioconjugate Chem. 2004, 15, 326−332

Gene Transfer into Human Hepatoma Cells by Receptor-Associated Protein/Polylysine Conjugates Tae-Gyun Kim,† Seog-Youn Kang,† Ju-Hye Kang,† Mi-Young Cho,† Joo-Il Kim,† Seung-Hee Kim,† and Jin-Seok Kim*,‡ Department of Pharmacology, National Institute of Toxicological Research, Korea Food and Drug Administration, Seoul 122-704, Republic of Korea, and College of Pharmacy, Sookmyung Women’s University, Seoul 140-742, Republic of Korea. Received February 24, 2003; Revised Manuscript Received January 19, 2004

Receptor-associated protein (RAP) is a ligand for all members of low-density lipoprotein (LDL) receptor families. RAP is internalized into cells via receptor-mediated endocytic trafficking, making it an attractive mechanism for efficient gene delivery. In this study, we have developed a gene delivery system using RAP as a targeting ligand. A RAP cDNA lacking a C-terminal heparin-binding domain was amplified by polymerase chain reaction (PCR) from a human liver cDNA library and was reamplified by using a primer containing a cysteine codon at its carboxyl end to facilitate its conjugation to polylysine (polyK). RAP was purified using a bacterial expression system and coupled to poly-Dlysine (PDL) or poly-L-lysine (PLL) of average MW 50 kDa via the heterobifunctional cross-linker SPDP. Using fluorescence-labeled RAP ligand, cellular uptake of the transfection complexes into HepG2 cells was shown to be highly efficient and more specific to PDL-conjugated RAP compared with PLLconjugated one. Plasmid DNA containing a luciferase reporter gene was condensed with either RAPPDL or RAP-PLL. In vitro transfection into HepG2 cells with RAP-PDL conjugate resulted in significantly higher luciferase expression levels in comparison to either nonconjugated PDL, or RAPPLL, or LipofecAMINE/DNA complexes in the presence of 10% fetal bovine serum. Luciferase expression was inhibited by the addition of excess RAP. Treatment of the cells with Lovastatin, which inhibits HMG-Co reductase and increases expression of LDL receptor, stimulates luciferase expression, suggesting that the gene delivery is specifically mediated by LDL receptor. Thus, RAP-PDL conjugates have the potential to be used as a new nonviral gene delivery vector.

INTRODUCTION

Receptor-mediated endocytosis (RME) represents a highly efficient internalization pathway (1) and offers an efficient cellular route for the delivery of gene therapy vectors. A number of gene transfer systems have been developed to deliver foreign DNA via RME using specific ligands such as asialoglycoprotein (2), transferrin (3), or anti-CD3 antibody (4). Gene transfer using these ligands has been accomplished by synthetic vectors termed molecular conjugates, which consist of two linked functional domains: a ligand domain and a DNA-binding domain to achieve binding and incorporation of the transgene into the vector complex. For many of these approaches, polylysine (polyK) provides a DNA-condensing carrier function, and chemical linkage methods are used to achieve linkage between the polyK and the ligand domain of the complex (5). For some cell types, receptormediated transport of DNA is a highly effective and physiological means of introducing DNA into the cells (3, 4, 6). * To whom correspondence should be addressed. Tel: +82-2-710-9574; Fax: +82-2-712-0032; E-mail: jskim@ sdic.sookmyung.ac.kr. † National Institute of Toxicological Research. ‡ Sookmyung Women’s University. 1 Abbreviations: RAP, receptor-associated protein; polyK, polylysine; PDL, poly-D-lysine; PLL, poly-L-lysine; RME.; receptor-mediated endocytosis; ILU, integrated light unit; LDL, lowdensity lipoprotein; LRP, LDL receptor-related protein; SPDP, 3-(2-pyridyldithio)propionic acid N-hydroxysuccinimide ester.

The low density lipoprotein (LDL) receptor gene family is a large family of endocytic receptors that bind and internalize various ligands, which include several apolipoproteins that are involved in cholesterol homeostasis (7). This receptor family is ubiquitously expressed in many tissues such as liver and brain (8). Moreover, LDL receptor activities are elevated in malignant cells, which have high LDL requirements as do rapidly dividing cells (9). The receptor-associated protein (RAP) is a specialized folding chaperone/escort protein that is associated with newly synthesized receptors in the endoplasmic reticulum (10) and can bind with high affinity to the members of the LDL receptor family such as the LDL receptor-related protein (LRP), the very low-density lipoprotein (VLDL) receptor, and megalin (gp330) (8). Exogenously added RAP binds to receptors on the cell surface in vitro and antagonizes all known ligand interactions with the receptors (10, 11). Interestingly, RAP is internalized into cells via receptor-mediated endocytic trafficking (12, 13), which is an attractive mechanism for efficient gene delivery. In this study, we explore the use of RAP as a targeting ligand for receptor-mediated gene transfer into human hepatoma cells (HepG2). The interaction, uptake, and gene expression of the transfection complexes were found to be due to the specific function of RAP ligand. As with other ligand-dependent gene delivery systems, we find that transfected gene expression is decreased by addition of unconjugated RAP ligand in the transfection medium and enhanced by treatment of Lovastatin, which inhibits cholesterol biosynthesis and increases the expression of

10.1021/bc0340262 CCC: $27.50 © 2004 American Chemical Society Published on Web 02/28/2004

Receptor-Associated Protein/Polylysine Conjugates

LDL receptor (14, 15). Because LRP and LDL receptors are abundant on hepatoma cells, RAP-mediated gene delivery systems may be useful for the delivery of therapeutic genes into this cell type. EXPERIMENTAL PROCEDURES

Materials. Human liver cDNA library and LA Taq with GC buffer kit were purchased from TaKaRa Shuzo Company (Shiga, Japan). Pinpoint Xa protein purification system, pPinPoint Xa-3 vector, factor Xa protease, pRLCMV, and Dual-Luciferase reporter assay kit were from Promega (Madison, WI). Reagents including LB medium, ampicillin, N-[2-hydroxyethyl]piperazine-N′-[2-ethanesulfonic acid] (HEPES), 3-(2-pyridyldithio)propionic acid N-hydroxysuccinimide ester (SPDP), fetal bovine serum (FBS), streptavidin-fluorescene isothiocyanate (FITC), poly-D-lysine hydrobromide (PDL, molecular weight 30 000-70 000) and poly-L-lysine hydrobromide (PLL, molecular weight 30 000-70 000) were from Sigma Chemical Company (St. Louis, MO). PCR primers, restriction enzymes, isopropyl β-D-thiogalactopyranoside (IPTG), dithiothreitol (DTT), Eagle’s minimal essential medium (EMEM), LipofecAMINE reagent, and Opti-MEM medium were purchased from Life Technologies (Rockville, MD). Biotin, Sephacryl S-100 column, and Sepadex G-25 superfine were from Amersham Pharmacia Biotech (UK). Centriplus YM-10 filter device was from Millipore, Co. (Bedford, MA). BCA protein assay kit was from Pierce (Rockford, IL). QIAquick gel extraction kit, EndoFree plasmid Maxi kit, and column were from Qiagen (Germany). Lovastatin was kindly supplied by Choongwae Pharmaceutical Co. (Korea). Deionized-distilled water was produced by using Milli-Q water purification system (Millipore) and FI-STREEM III Bi-Distiller system (Barnstead, Dubuque, IA). All other chemicals and reagents were of tissue culture grade. Construction and Expression of RAP. The cDNA for partial RAP1-260 was amplified in polymerase chain reaction (PCR) from human liver cDNA library with a 5′ primer, GGAGAAGAACCAGCCCAAGC, and a 3′ primer, GCGCAGCCGCTCGCCCCAGGC. The RAP cDNA was reamplified with a 5′ primer containing a BamH I restriction site to facilitate its ligation and a 3′ primer containing a Kpn I site and an additional cysteine codon preceding a stop codon to facilitate its conjugation to polyK (6). The PCR product was cloned into BamH I-Kpn I-digested pPinPoint Xa-3, a bacterial expression vector with N-terminally biotin-tagged sequence. The plasmid structure was confirmed by DNA sequence analysis with ABI prism 3700 (Applied Biosystems), transformed into E. coli/JM109 cells, and expressed according to the manufacturer’s instructions. Purification and Reduction of RAP. RAP expressed with biotin tag was purified using PinPoint Xa protein purification system and digested with factor Xa protease in Xa buffer (20 mM Tris-HCl pH 7.4, 100 mM NaCl). The solution of digested RAP was incubated with avidin resin to remove the biotin tag and passed through Sephacryl S-100 column to remove factor Xa protease. Eluent fractions were concentrated with a centrifugal filter device (Centriplus YM-10). Purified RAP was analyzed using SDS-PAGE and assayed for protein concentration by BCA protein assay kit. A cysteine residue in the C-terminal of recombinant RAP was reduced by dithiothreitol (DTT) to generate free sulfhydryl groups. Briefly, to 1 mg of RAP in 0.8 mL of HEPESbuffered saline (HBS; 150 mM NaCl, 10 mM HEPES pH 7.4) was added 8 µL (4 µmol) of 0.5 M DTT solution. The

Bioconjugate Chem., Vol. 15, No. 2, 2004 327

solution was kept under argon for 1 h, and the pH was adjusted to 5.2 by adding 3 M sodium acetate buffer. After gel filteration (Sepadex G-25 superfine), a solution of reduced RAP for preparing RAP-polyK conjugates was obtained (3). Preparation of RAP-polyK Conjugates. Two kinds of polyK were used to conjugate recombinant RAP; polyD-lysine (PDL) and poly-L-lysine (PLL), both of which have an average molecular weight of 50 kDa. Purified RAP was conjugated to each polyK, using a heterobifunctional cross-linker, 3-(2-Pyridyldithio)propionic acid N-hydroxysuccinimide ester (SPDP) (3). Briefly, each polyK, consisting of approximate 400 lysine residues, was mixed with SPDP at a molar ratio of 1:20 in PBS. After incubation for 1 h at room temperature, the reaction mixture was dialyzed against deionized-distilled water for 48 h to remove excess unreacted SPDP. This reaction theoretically yielded a polyK molecule with about 5% SPDP modification of its lysine residues. Each SPDPpolyK was then mixed with RAP and incubated for 24 h at room temperature. These RAP-PDL or RAP-PLL conjugates were used for preparing transfection complexes. Plasmid Vector. pRL-CMV (4079 bp), a plasmid expression vector containing a cDNA encoding Renilla reniformis (sea pansy) luciferase driven by the cytomegalovirus (CMV) promoter, was grown in E. coli DH5R cells in LB medium containing ampicillin (100 µg/mL) and was purified using EndoFree plasmid Maxi kit and column. Gel Retardation Assay. The RAP-PDL/DNA complexes were prepared by adding various amounts of RAP-PDL conjugate, ranging from 0 to 14 µg, in 20 µL of HBS to 1 µg of pRL-CMV DNA in 20 µL of HBS. The RAP-PDL solution was slowly added to the DNA solution and mixed gently. The reaction mixture was incubated for 30 min at room temperature and applied to a 0.8% agarose gel in Tris acetate EDTA buffer, pH 7.6, with 0.5 µg/mL ethidium bromide. Electrophoresis was carried out at 5 V/cm for 50 min. Cell Line. The human hepatoma cells (HepG2) were obtained from the American Type Culture Collection. HepG2 cells were maintained in T-75 culture flask with vented cap containing EMEM supplemented with heat inactivated 10% fetal bovine serum (FBS), penicillin (100 units/ml), streptomycin (100 µg/mL), 4 mM L-glutamine, and 1.0 mM sodium pyruvate at 37 °C, 5% CO2. Preparation of RAP-polyK/DNA Complexes and Transfection. Cells were plated in 24-well culture plates at 5 × 104 cells/well density and incubated for 24 h prior to transfection, by which time the cells were ∼30% confluent. One microgram of plasmid DNA and the various amounts of RAP-PDL were diluted to a final volume of 100 µL with HBS. The RAP-PDL solution was added to the DNA solution, gently mixed, and incubated at room temperature. RAP-PDL/DNA complexes were formed at various weight ratios, which were calculated as molar ratios of the charge of PDL nitrogen to DNA phosphate (N:P); 1 µg of PDL for a N:P ratio of 1.3:1, 3 µg of PDL for 3.8:1, 6 µg of PDL for 7.5:1, 9 µg of PDL for 11.3:1, and 12 µg of PDL for 15:1. Charge ratios were calculated as described (16). After 30 min incubation, the solutions of gene transfer complexes (200 µL) were mixed with 1.5 mL of 10% serum-containing medium and added to the cells (n ) 4). Forty-eight hours after incubation with transfection medium, the medium was removed and luciferase activity was assayed, by which time the cells were ∼90% confluent. RAP-PLL/DNA and PDL/DNA complexes were prepared in an identical manner as with

328 Bioconjugate Chem., Vol. 15, No. 2, 2004

RAP-PDL/DNA complexes. For the transfection by LipofecAMINE reagent, pRL-CMV DNA (1 µg) in OptiMEM medium was mixed with an equal volume of LipofectAMINE (14 µg) diluted appropriately in OptiMEM medium. Complexes were incubated for 30 min at room temperature before addition to cells. Cells were transfected in the presence of 10% FBS. Assay for Luciferase. Cell lysates were prepared by the addition of 100 µL of lysis buffer (Dual-Luciferase reporter assay kit) and incubation with agitation for 20 min at room temperature. After centrifugation at 12 000 g for 10 min, 10 µL of the supernatant of cell lysates were analyzed for luciferase activity using a Clinilumat LB9502 instrument (Berthold, Germany). These data were normalized for protein content assayed by BCA protein assay kit. Results shown are expressed as integrated light units (ILU)/mg of protein unless otherwise indicated. Detection of Internalized Gene Transfer Complexes. Complexes containing streptavidin-FITC were formed as follows: RAP containing N-terminal biotin tag was conjugated with PDL or PLL as described above. Ten micrograms of the conjugates was diluted with 100 µL of HBS and incubated with 5 µg of streptavidin-FITC for 10 min with occasional mixing. Plasmid DNA diluted in 100 µL of HBS was added to the mixture and incubated for 30 min at room temperature. Each complex was then diluted in medium containing 10% serum and added to the cells as above. Twenty-four hours after transfection, cellular uptake of FITC-labeled RAP-polylysine/DNA complexes was analyzed with fluorescence microscopy (Olympus IX). Competitive Inhibition by Free RAP. A competitive inhibition assay using free RAP was performed to confirm the specificity of RAP ligand-dependent gene delivery. Briefly, 105 HepG2 cells (24-well plate) were incubated with free recombinant RAP (0.2, 0.5, and 1 mg/mL in 1.5 mL of the medium) for 10 min in CO2 incubator, followed by the addition of gene transfer complexes in the presence of free RAP, which consisted of 1 µg of pRL-CMV and 9 µg of RAP-PDL with 200 µL of HBS. Transfection medium was removed after 24 h incubation, and luciferase activity of cell lysates was measured as described above. Treatment with Lovastatin. To further confirm receptor-mediated gene delivery, transfection efficiency of RAP-PDL conjugate was measured after treatment with Lovastatin, which inhibits a rate-limiting enzyme for cholesterol biosynthesis, hydroxymethylglutarylcoenzyme A (HMG-CoA) reductase (Ki ) 0.6 nM). The transfection and luciferase assay were performed as described in the above competition experiment except that Lovastatin was added to the medium with the transfection complexes simultaneously. RESULTS

Cloning and Expression of Recombinant RAP. We amplified a partial cDNA of the RAP gene (amino acids 1-260) lacking its C-terminal heparin-binding domain by PCR from human liver cDNA library. The PCR product was reamplified with primers that contain restriction sites for cloning and a C-terminal cysteine residue for conjugation (6) and then cloned into a bacterial expression vector. DNA sequencing analysis revealed that the cloned RAP gene showed 99.6% homology with the previously reported RAP gene (accession number: NM_002337, GI: 4505020) (17), and the predicted protein sequence was identical (data not shown). Purification and Reduction of Recombinant RAP. Recombinant RAP was produced in the form of a 47 kDa

Kim et al.

Figure 1. Expression and purification of RAP. The cell lysates of RAP fusion protein expressing E. coli and purified RAP were analyzed by 4-20% gradient SDS-polyacrylamide gel and Coomassie staining. Lanes: 1, E. coli cell lysate before induction of the fusion protein; 2, lysate after induction; 3, lysate after binding with avidin resin; 4, eluant RAP with biotin tag (47 kDa); 5, RAP and biotin tag fragment incubated with factor Xa for 10 h; 6, RAP and factor Xa after incubaion with avidin resin; 7, purified RAP after column chromatography (30 kDa).

fusion protein that is biotinylated in the N-terminal tag (Figure 1, lane 4), of which approximated 10 mg was recovered using 30 mL of avidin resin. This purified fusion protein was digested with factor Xa protease for 10 h at 37 °C. Coomassie staining revealed a 30 kDa band corresponding to RAP; however, biotin tag (17 kDa) and factor Xa protease (44 kDa) contaminants were present (Figure 1, lane 5). To improve the purity, the solution of digested fusion protein was incubated with avidin resin (Figure 1, lane 6). After further purification by Sephacryl S-100 column chromatography, pooled fractions containing recombinant RAP displayed a single band at 30 kDa (Figure 1, lane 7), and the solution was concentrated. Approximately 500 µg of RAP was obtained from one liter of bacterial culture. The single cysteine residue at the C-terminal end of the purified RAP was reduced by DTT treatment and used in the following experiments. Synthesis of RAP-polyK Conjugates. A heterobifunctional cross-linker, SPDP, was used to introduce a dithiopyridine linker into polyK, where the NHS ester group of SPDP reacted with -amino groups of polyK to form an amide linkage (18). Ziady et al. reported that reaction with 20× molar excess of SPDP modifies 3.5% of the lysine residues in a polyK molecule with dithiopyridine linkers (19). These linkers will readily react with sulfhydryl groups of RAP, yielding disulfide linkages between RAP and polyK (6). Purified RAP-PDL or RAPPLL displayed reduced electrophoresis in SDS gradient polyacrylamide gel because of the basic amino acids in each polyK. We did not detect a significant amount of unconjugated free RAP in the reaction product (data not shown). Gel Retardation Assay. To determine the optimal RAP-PDL/DNA ratio by measuring the effect of complex formation and charge neutralization, agarose gel electrophoresis was performed (2). As shown in Figure 2, electrophoretic mobility of the RAP-PDL/DNA complex decreased gradually with increasing amounts of RAPPDL conjugates. DNA band retardation occurs due to the neutralization of DNA negative charges by the positive charges of RAP-PDL conjugates. At a weight ratio of 1 µg DNA:8 µg PDL coupled to RAP, plasmid DNA showed complete retardation, indicating that the DNA negative

Receptor-Associated Protein/Polylysine Conjugates

Bioconjugate Chem., Vol. 15, No. 2, 2004 329

Figure 2. Gel retardation assay of DNA complexed with RAPPDL. Lanes: M, size marker (1 kbp ladder); 1, plasmid DNA alone (1 µg); 2-8, DNA mixed with 2, 4, 6, 8, 10, 12, or 14 µg of PDL conjugated with RAP, respectively.

Figure 3. Transfection efficiency of RAP-polylysine/DNA complexes with human HepG2 hepatoma cells. The increasing amounts (µg) of polylysines shown on the ordinate (1, 3, 6, 9, 12) were used to complex 1 µg of DNA (pRL-CMV; 4079 bp). Each DNA complex (in 0.2 mL HBS) was mixed with 1.5 mL of culture medium (EMEM containing 10% FBS) and added to 5 × 104 cells per well in 24-well plates. After 48 h, the transfection medium was removed and cells were harvested. Luciferase activity is shown as means ( SD (n ) 4). Transfection efficiency of naked DNA (1 µg), LipofectAMINE-DNA (14 µg) and unconjugated PDL is shown for comparison.

charge was neutralized or slightly positively charged with RAP-PDL conjugate at this weight ratio (Figure 2, lane 5). Formulation Optimization of RAP-polyK/DNA Complexes. The RAP ligand-dependent transfection efficiencies of the complexes in five different weight ratios of polyK to DNA are shown in Figure 3. To detect gene expression, cells were transfected with particles containing the plasmid pRL-CMV, a vector with a luciferase reporter gene insert driven by the cytomegalovirus (CMV) promoter/enhancer. Plasmid DNA (1 µg) was either complexed with PDL alone, with RAP conjugated to PDL (RAP-PDL), or with RAP conjugated to PLL (RAPPLL). Plasmids complexed with RAP-PDL showed the highest transfection efficiencies (Figure 3, RAP-PDL). In addition to ligand specificity, transfection efficiencies were dependent on the cation:anion ratio. A weight ratio of 1 µg DNA:9 µg RAP-PDL yielded the highest luciferase expression level (up to 65-fold higher than those obtained with unmodified PDL and up to 25-fold higher than with LipofectAMINE, a commercial cationic lipid formulation). At the highest weight ratio (1 µg:12 µg), the viability of the cells was not significantly affected (viability g95%, data not shown) and the transfection efficiency decreased slightly (less than 14%). This decrease was observed in a previous study using polyK coupled to anti-CD3 antibody (4) and could be due to excess polyK relative to the amount of DNA used (1 µg). Interestingly, compared to their effectiveness at high cation:anion ratios, RAP-PDL/DNA complexes were more effective at lower cation:anion ratios than were ligand-free PDL/DNA complexes. Transfection efficiencies

Figure 4. Cellular uptake of fluorescence-labeled RAP-polyK/ DNA complexes into HepG2 cells. RAP fusion protein expressed with biotin tag was coupled to PDL or PLL, as described in Materials and Methods, and incubated with equal amounts of streptavidin-FITC. Twenty-four hours after addition of each fluorescence-labeled conjugate/DNA complex (9 µg:1 µg), endocytosed particles were visualized with fluorescence microscopy. The internalized gene transfer complexes of RAP-PDL (A) and RAP-PLL (B).

of RAP-PLL/DNA complexes were similar to those of ligand-free PDL/DNA complexes over all cation:anion ratios. Cellular Uptake of Fluorescence-Labeled RAPpolyK/DNA Complexes. The uptake of RAP-PDL gene transfer complexes into HepG2 cells was assessed by using fluorescence microscopy (Figure 4). RAP components of the complexes were labeled with streptavidinFITC. Following transfection with FITC-labeled RAPPDL/DNA, almost 100% of HepG2 cells contained endocytosed gene transfer particles. In each transfected cell, several FITC-labeled RAP-PDL/DNA complexes were observed (Figure 4A). In contrast, following transfection with FITC-labeled RAP-PLL/DNA complexes, only about 30% of the cells contained observable internalized gene transfer particles distributed in intracellular vesicles (Figure 4B). The number of fluorescent particles per cell was also lower in transfection with RAP-PLL/ DNA complexes, and the morphology of intracellular particles appeared to be more diffused than that of RAPPDL/DNA complexes. Endocytosed particles were only detected using RAP-PDL/DNA or RAP-PLL/DNA complexes; in cells incubated with the complexes containing unconjugated PDL, distinct internalized gene transfer particles were not detectable. Competitive Inhibition with Free RAPs. To confirm the specific interaction of RAP-PDL conjugates with hepatoma cells, competitive inhibition experiments were performed with excess free RAP, which was added to the cell culture medium during transfection (Figure 5). The addition of soluble RAP ligand at three concentrations decreased transfection efficiencies of RAP-PLD/DNA complexes by 10-30%. RAP concentrations of 0.2, 0.5, and 1 mg/mL were approximately 2, 5, and 10 molar equiv in excess of conjugated RAP. In transfections with unconjugated PDL/DNA complexes, coincubation with excess free RAP did not affect reporter gene expression in cells. These results indicate that the cell interaction,

330 Bioconjugate Chem., Vol. 15, No. 2, 2004

Figure 5. Free RAP competition experiment. Nine micrograms of RAP-PDL were used to complex 1 µg of DNA. DNA complexes in 200 µL of HBS were added to HepG2 cells (105 cells per well in 24-well plates) in 1.5 mL of EMEM containing 10% FBS in the absence or presence of the free RAP at three concentrations as indicated. The cells were harvested 24 h after transfection. Luciferase activity was measured as described in Materials and Methods and shown as means ( SD (n ) 4).

Kim et al.

Figure 7. Duration of transgene expression. The experiment was performed as described in the legend of Figure 5, except that luciferase activity was measured 1, 2, 3, 4, and 5 days after transfection. Data represented are a mean of four independent experiments. Table 1. Effects of Serum on the Transfection Efficiency of RAP-PDL/DNA Complexesa transfection conditions

luciferase activity: (ILU × 10-4/mg cell protein)

without serum 5% serum 10% serum 25% serum 50% serum

0.1 ( 0.3 1.8 ( 0.5 2.3 ( 0.7 2.5 ( 0.4 2.6 ( 0.7

a Note: HepG2 cells were transfected with RAP-PDL/DNA complexes (9 µg:1 µg). The experiment was performed as described in the legend of Figure 5, except that the complexes were diluted in medium with different amounts of fetal bovine serum, as indicated. Data are expressed as ILU × 10-4/mg cell protein. Results shown are mean values ( SD (n ) 4).

Figure 6. Enhanced transfection efficiencies of RAP-PDL conjugates by the treatment of Lovastatin, an inhibitor of HMG CoA reductase. The experiment was performed as described in the legend of Figure 5, except that transfections were carried out in the absence or presence of three concentrations of Lovastatin as indicated.

uptake, and gene expression of the RAP-PDL/DNA complexes in HepG2 cells is due to the specific interaction of the RAP-PDL conjugate with its receptor(s). The Effect of Lovastatin on RAP-Mediated Gene Transfer. Lovastatin (Mevinolin) is a potent competitive inhibitor of a rate-limiting enzyme of cholesterol biosynthesis, HMG-CoA reductase (Ki ) 0.6 nM) (20). Lovastatin activates LDL receptor gene expression in HepG2 cells (15). The presence of Lovastatin in RAP-PDL/DNA (1 µg:9 µg) transfections augmented reporter gene expression in a dose-dependent manner (Figure 6). The addition of 60 nM Lovastatin together with gene transfer complexes increased the transfection efficiency of RAPmediated gene delivery by a factor of 100 relative to that of Lovastatin nontreated cells. Duration of Transgene Expression. The timecourse of luciferase expression in HepG2 cells is shown in Figure 7. Cells were transfected when they were 2030% confluent and continued to grow for 2 days posttransfection, at which time the transfected cells approached confluence. The maximum level of luciferase activity was detected 2 days posttransfection. Luciferase activity decreased to ∼10% by the fifth day posttransfection. The degree of luciferase expression steeply declined 3 days after transfection and persisted for at least 5 days. A potential concern of gene transfer using RAP-PDL conjugates is the influence of cell proliferation. To assess

this potential influence, cells were retransfected at day 3 with the same complexes. This second transfection yielded marginal increases (∼5%) in gene expression levels at days 4 and 5 posttransfection. This result suggests that RAP-mediated gene delivery and expression may depend on cell proliferation. Effects of Serum on Transfection. Effects of serum on the transfection of RAP-PDL complexes was also investigated by comparing the expression level of reporter gene following transfection in the presence (final concentration 50%) or absence of serum (Table 1). Addition of fetal bovine serum (FBS) to the transfection medium slightly increased reporter gene expression. In the absence of FBS, the transfection efficiency was ∼25-fold lower than that in 10%, 25%, or 50% FBS. DISCUSSION

The development of a safe and efficient method for the delivery of transgenes to target cells and tissues still represents a major challenge. It has been reported that polycation conjugated with ligand for asialoglycoprotein receptor (2), transferrin receptor (3), c-kit receptor (21), or EGFR receptor (22, 23) resulted in specific gene delivery to cells expressing these receptors. The LDL receptor gene family is well-characterized. LDL-receptor-ligand interactions lead to endocytosis, providing an attractive mechanism for gene transfer. The LDL receptor is abundant in liver tissue, which is the principle organ of several genetic, acquired, and viral diseases amenable to gene therapy (24). Furthermore, all actively dividing cells require cholesterol which is taken up by the cells as a ligand-lipoprotein complex through receptor-mediated endocytosis (RME) (8). The expression of LDL receptors is particularly high on proliferating malignant cells as compared with the corresponding normal

Receptor-Associated Protein/Polylysine Conjugates

tissues (25, 26, 27), making the LDL receptor an attractive target for cancer therapy. For this reason, there are many studies on the selective delivery of cytotoxic compounds to cancer cells using LDL as a vehicle through the RME pathway (9). However, very few studies on gene transfer through LDL receptormediated endocytosis have been reported. Sipehia et al. developed a new method for DNA transfection of cultured endothelial cells using apolipoprotein E (Apo E) adsorbed lipofection reagent-DNA complexes (28). The transfection efficiencies with this method was found to be more efficient than conventional methods including calcium phosphate precipitation and lipofection. Furthermore, Kim et al. demonstrated another gene delivery vehicle (Terplex DNA) that is generated through electrostatic and hydrophobic interactions between LDL, stearyl-PLL, and DNA (29). However, receptor-mediated gene delivery of ligand-polycation conjugates via LDL receptor gene family members has not been reported previously. To exploit this ubiquitous and efficient transport mechanism for introducing transgene into cells, human RAP was purified at high yields using a bacterial expression system and was conjugated with a DNA-binding polycation, polyK. The affinity of RAP for LDL receptor gene family is equivalent to that of apolipoprotein E, a high affinity ligand for the receptors (7). The intact 39kDa RAP (323 amino acids) contains several structural and functional domains (7). In this study, we used a truncated version of RAP (N-terminal 260 amino acids), which contains at least two high affinity binding domains for LDL receptor gene families (30, 31) but does not contain RAP’s C-terminal heparin binding domain. Because this truncated RAP lacks a cysteine residue, we engineered our recombinant RAP to contain a C-terminal cysteine residue. RAP-containing complexes were prepared by coupling SPDP-polyK to the cysteine residue of recombinant RAP. Recently, Gupta et al. (6) developed a similar recombinant ligand for receptor-mediated gene delivery: a single chain Fv of a monoclonal antibody (directed against human polymeric immunoglobulin receptor [plgR]), containing a C-terminal cysteine residue conjugated to polyK. We chose two DNA-binding polycations for conjugating to RAP that had previously been used in the RME-based gene transfer system: PLL (2-4, 32) and PDL (33). The results of in vitro experiments clearly showed that the conjugation of the targeting ligand RAP to PDL results in a 65-fold increase in the transfection efficiency in HepG2 cells (Figure 3). For the RAP-PDL/DNA complexes, the best transfection efficiency was found at a ratio of 1 µg DNA:9 µg RAP-PDL. Corresponding results indicated complete retardation of the complexes at a ratio of 1 µg DNA:8 µg in a gel retardation assay (Figure 2). The increase of transfection efficiency was blocked by addition of free RAP (Figure 5), indicating that RAP ligand dependent uptake of the transfection complex is responsible for the higher transfection efficiency. This result is consistent with previous reports of receptormediated gene delivery (4, 6). Uptake of gene transfer particles into HepG2 cells via RAP-PDL is highly efficient: almost all cells contained several endocytosed particles following transfer of the FITC-avidin labeled complexes (Figure 4). In contrast to the RAP-PDL system, transfection efficiencies of RAP-PLL/DNA complexes were similar to that of ligand-free PDL/DNA complexes (Figure 3). These data (both Figures 3 and 4) indicate that the PDL complex is more stable than the PLL complex in a lysosomal environment as well as in transfection media.

Bioconjugate Chem., Vol. 15, No. 2, 2004 331

Furthermore, the transfection of fluorescence-labeled RAP-PLL/DNA complexes showed that the internalized fluorescence particles, which had a diffused morphology (Figure 4), could be observed in about 30% of the cells. In a study of the endocytic trafficking of RAP/gp330, a member of the LDL receptor gene family, RAP dissociated from the receptor complex in late endosomes and was then rapidly delivered to lysosomes and degraded (13). The data from our study show that a portion of the endocytosed RAP-PLL/DNA complexes might remain unstable and be degraded in lysosomal compartments, whereas RAP-PDL is resistant to such degradation and released into the cytoplasm, entering the nucleus. Thus, media instability and lysosomal degradation of PLL and DNA may explain why, as observed in the PLL-based system, inclusion of membrane-active agents such as adenovirus capsid or membrane-active compounds into the gene transfer complexes greatly enhances transfection efficiency (22, 34, 35). DNA encoding the luciferase gene complexed to RAPPDL conjugate was specifically and efficiently transfected into HepG2 cells bearing LDL receptor and LRP. Gene transfer by this method is ligand dependent and receptor mediated. DNA complexes containing RAP were shown to be effectively internalized into HepG2 cells by ligandreceptor interaction. HepG2 cells, which are abundant in LRP and LDL receptor, were efficiently transfected with the RAP-PDL/DNA complexes in a ligand-dependent manner. Nontargeted complexes that lacked the RAP ligand did not efficiently transfect HepG2 cells. Transfection of a luciferase reporter gene by the RAP-PDL conjugate could be inhibited by addition of an excess of free RAP. Although we did not test competition by the other ligands for the receptors such as Apo E or Apo B, they are likely to compete for the following reason: The transfection efficiency of the conjugates into HepG2 cells was greatly enhanced in the presence of Lovastatin (a cholesterol biosynthesis inhibitor), which results in the overexpression of the LDL receptors in HepG2 cells. The transfection efficiencies of RAP-PDL conjugates in Lovastatin treated cells increased up to 100-fold relative to that in nontreated cells. In addition to enhancing transfection efficiency, the incorporation of ligands offers the opportunity for specific targeting of gene delivery. Since rapidly dividing cells express the LDL receptor at a high level, RAP is a promising ligand in cancer gene therapy when proliferating cells such as malignant cells are to be targeted. However, specificity of target tissue and/or organ may be difficult to achieve using RAP due to the broad expression of the RAP binding receptor, the LDL receptor gene family, in the human body. This difficulty may be overcome by incorporating a tissue specific enhancer and/ or promoter sequence in the expression vector. In conclusion, we have developed a novel gene delivery system via the RME pathway using RAP conjugated with PDL as a gene delivery vector. Gene delivery using PDL itself or RAP-PLL was not efficient. Our data demonstrate that synthetic RAP-PDL is an effective reagent facilitating formation of DNA-ligand complexes for subsequent receptor-mediated gene transfer. RAP-PDL is thus a promising tool for enabling targeted transfer of genes in vivo. LITERATURE CITED (1) Schwartz, A. L. (1995) Receptor cell biology: Receptormediated endocytosis. Pediatr. Res. 38, 835-843.

332 Bioconjugate Chem., Vol. 15, No. 2, 2004 (2) Wu, G. Y., and Wu, C. H. (1987) Receptor-mediated in vitro gene transformation by a soluble DNA carrier system. J. Biol. Chem. 262, 4429-4432. (3) Wagner, E., Zenke, M., Cotton, M., Beug, H., and Birnstiel, M. L. (1990) Transferrin-polycation conjugates as carriers for DNA uptake into cells. Proc. Natl. Acad. Sci. U.S.A. 87, 34103414. (4) Buschle, M., Cotton, M., Kirlappos, H., Mechtler, K., Schaffner, G., Zauner, W., Birnstiel, M. L., and Wagner, E. (1995) Receptor-mediated gene transfer into human T lymphocytes via binding of DNA/CD3 antibody prticles to the CD3 T cell receptor complex. Hum. Gene Ther. 6, 753-761. (5) Cristiano, R. J., and Curiel, D. T. (1996) Strategies to accomplish gene delivery via the receptor-mediated endocytosis pathway. Cancer Gene Ther. 3, 49-57. (6) Gupta, S., Eastman, J., Silski, C., Ferkol, T., and Davis, P. B. (2001) Single chain Fv: a ligand in receptor-mediated gene delivery. Gene Ther. 8, 586-592. (7) Gliemann, J. (1998) Receptors of the low density lipoprotein (LDL) receptor family in man. Multiple functions of the large family members via interaction with complex ligands. Biol. Chem. 379, 951-964. (8) Hussain, M. M., Strickland, D. K., and Bakillah, A. (1999) The mammalian low-density lipoprotein receptor family. Annu. Rev. Nutr. 19, 141-172. (9) Firestone, R. A. (1994) Low-density lipoprotein as a vehicle for targeting antitumor compounds to cancer cells. Bioconjugate Chem. 5, 105-113. (10) Bu, G. (1998) Receptor-associated protein: a specialized chaperone and antagonist for members of the LDL receptor gene family. Curr. Opin. Lipidol. 9, 149-155. (11) Willnow, T. E. (1998) Receptor-associated protein (RAP): A specialized chaperone for endocytic receptors. Biol. Chem. 379, 1025-1031. (12) Willnow, T. E., and Herz, J. (1994) Genetic deficiency in low density lipoprotein receptor-related protein (LRP) confers cellular resistance to pseudomonas exotoxin A - evidence that LRP is required for uptake and degradation of multiple ligands. J. Cell Sci. 107, 719-726. (13) Czekay, R., Orlando, R. A., Woodward, L., Lundstrom, M., and Farquhar, M. G. (1997) Endocytic trafficking of megalin/ RAP complexes: Dissociation of the complexes in late endosomes. Mol. Biol. Cell 8, 517-532. (14) Traber, M. G., and Kayden, H. J. (1984) Inhibition of cholesterol synthesis by mevinolin stimulates low-density lipoprotein receptor activity in human monocyte-derived macrophages. Atherosclerosis 52, 1-11. (15) Qin, W., Infante, J., Wang, S. R., and Infante, R. (1992) Regulation of HMG-CoA reductase, apoprotein-B and LDL receptor gene expression by the hypocholesterolemic drugs simvastatin and ciprofibrate in HepG2, human and rat hepatocytes. Biochim. Biophys. Acta 1127, 57-66. (16) Felgner, P. L., Barenholz, Y., Behr, J. P., Cheng, S. H., Cullis, P., Huang, L., Jessee, J. A., Seymour, L., Szoka, F., Thierry, A. R., Wagner, E., and Wu, G. (1997) Nomenclature for synthetic gene delivery systems (editorial). Hum. Gene Ther. 8, 511-512. (17) Furukawa, T., Ozawa, M., Huang, R. P., and Muramatsu, T. (1990) A heparin binding protein whose expression increase during differentiation of embryonal carcinoma cells to parietal endoderm cells: cDNA cloning and sequence analysis. J. Biochem. 108, 297-302. (18) Hermanson, G. T. (1996) Bioconjugate Techniques, pp 6465, Academic Press, San Diego, CA. (19) Ziady, A., Ferkol, T., Gerken, T., Dawson, D. V., Perlmutter, D. H., and Davis, P. B. (1998) Ligand substitution of receptor targeted DNA complexes affects gene transfer into hepatoma cells. Gene Ther. 5, 1685-1697.

Kim et al. (20) Alberts, A. W., Chen, J., Kuron, G., Hunt, V., Huff, J., Hoffman, C., Rothrock, J., Lopez, M., Joshua, H., Harris, E., Patchett, A., Monaghan, R., Currie, S., Stapley, E., AlbersSchonberg, G., Hensens, O., Hirshfield, J., Hoogsteen, K., Liesch, J., and Springer, J. (1980) Mevinolin: a highly potent competitive inhibitors of hydroxymethylglutaryl-coenzyme A reductase and a cholesterol-lowering agent. Proc. Natl. Acad. Sci. U.S.A. 77, 3957-3961. (21) Schwarzenberger, P., Spence, S. E., Gooya, J. M., Michiel, D., Curiel, D. T., Ruscetti, F. W., and Keller, J. R. (1996) Targeted gene transfer to human hematopoietic progenitor cell lines through the c-kit receptor. Blood 87, 472-478. (22) Chen, J., Gamou, S., Takayanagi, A., and Shimizu, N. (1994) A novel gene delivery system using EGF receptormediated endocytosis. FEBS Lett. 338, 167-169. (23) Lee, T. K., Han, J., Fan, S., Liang, Z., Tian, P., Gu, J., and Ng, I. O. (2001) Gene delivery using a receptor-mediated gene transfer system targeted to hepatocellular carcinoma cells. Int. J. Cancer 93, 393-400. (24) Ledley, F. D. (1993) Hepatic Gene therapy: Present and Future. Hepatology 18, 1263-1273. (25) Rudling, M. J., Stahle, L., Peterson, C. O., and Skoog, L. (1986) Content of low-density lipoprotein receptors in breast cancer tissue related to survival of patients. Br. Med. J. 292, 580-582. (26) Niendorf, A., Nagele, H., Gerding, D., Meyer-Pannwitt, U., and Gebhardt, A. (1995) Increased LDL receptor mRNA expression in colon cancer is correlated with a rise in plasma cholesterol levels after curative surgery. Int. J. Cancer 61, 461-464. (27) Juliusson, G., and Vitols, S. (1988) Impaired low-density lipoprotein receptor activity in chronic B-lymphocytic leukemia cells. Eur. J. Hematol. 40, 18-24. (28) Sipehia, R., and Martucci, G. (1995) High-efficiency transformation of human endothelial cells by Apo E-mediated transfection with plasmid DNA. Biochem. Biophys. Res. Commun. 214, 206-211. (29) Kim, J., Maruyama, A., Akaike, T., and Kim, S. W. (1997) In vitro gene expression on smooth muscle cells using a terplex delivery system. J. Controlled Release 47, 51-59. (30) Orlando, R. A., and Farquhar, M. G. (1994) Functional domains of the receptor-associated protein (RAP). Proc. Natl. Acad. Sci. U.S.A. 91, 3161-3165. (31) Rall, S. C., Ye, P., Jr., Bu, G., and Wardell, M. R. (1998) The domain structure of human receptor-associated protein: Protease sensitivity and guanidine HCl denaturation. J. Biol. Chem. 273, 24152-24157. (32) Perales, J. C., Ferkol, T., Molas, M., and Hanson, R. W. (1994) An evaluation of receptor-mediated gene transfer using synthetic DNA-ligand complexes. Eur. J. Biochem. 226, 255266. (33) Hisayasu, S., Miyauchi, M., Akiyama, K., Gotoh, T., Satoh, S., and Shimada, T. (1999) In vivo targeted gene transfer into liver cells mediated by a novel galactosyl-D-lysine/D-serine copolymer. Gene Ther. 6, 689-693. (34) Gao, L., Wagner, E. O., Cotton, M., Agarwal, S., Harris, C., Romer, M., Miller, L., Hu, P. C., and Curiel, D. (1993) Direct in vivo gene transfer to airway epithelium employing adenovirus-polylysine-DNA complexes. Hum. Gene Ther. 4, 17-24. (35) Plank, C., Oberhauser, B., Mechtler, K., Koch, C., and Wagner, E. (1994) The influence of endosome-disruptive peptides on gene transfer using synthetic virus-like gene transfer systems. J. Biol. Chem. 17, 12918-12924.

BC0340262