Department of Drug Delivery Research, Graduate School of Pharmaceutical Sciences, Kyoto University, Sakyo-ku, Kyoto 606-8501, Japan

Correspondence to: M Hashida

A novel mannosylated cholesterol derivative, cholesten-5-yloxy-N-(4-((1-imino-2--D-thiomannosyl-ethyl)amino)butyl) formamide (Man-C4-Chol), was synthesized in order to perform mannose receptor-mediated gene transfer with liposomes. Plasmid DNA encoding luciferase gene (pCMV-Luc) complexed with liposomes, consisting of a 6:4 mixture of Man-C4-Chol and dioleoylphosphatidylethanolamine (DOPE), showed higher transfection activity than that complexed with 3[N-(N',N'-dimethylaminoethane)-carbamoyl]cholesterol (DC-Chol)/DOPE(6:4) and N-[1-(2,3-dioleyloxy)propyl]-N,N,N-trimethylammonium chloride (DOTMA)/DOPE(1:1) liposomes in mouse peritoneal macrophages. The presence of 20 mM mannose significantly inhibited the transfection efficiency of pCMV-Luc complexed with Man-C4-Chol/DC- Chol/DOPE(3:3:4) and Man-C4-Chol/DOPE(6:4) liposomes. High gene expression of pCMV-Luc was observed in the liver after intravenously injecting mice with Man-C4-Chol/DOPE(6:4) liposomes, whereas DC-Chol/DOPE(6:4) liposomes only showed marked expression in the lung. The gene expression with Man-C4-Chol/DOPE(6:4) liposome/ DNA complexes in the liver was observed preferentially in the non-parenchymal cells and was significantly reduced by predosing with mannosylated bovine serum albumin. The gene expression in the liver was greater following intraportal injection. These results suggest that plasmid DNA complexed with mannosylated liposomes exhibits high transfection activity due to recognition by mannose receptors both in vitro and in vivo. Gene Therapy (2000) 7, 292-299.

gene delivery; cationic liposome; gene therapy; macrophages

Introduction

Macrophages play an important role in host immune functions such as antigen presentation.¹ Through transgene expression, attempts have been made to modulate the function and dysfunction of macrophages for the treatment of genetic metabolic diseases including Gaucher's disease,² or to inhibit HIV replication in these cells.³ Several strategies have been developed to transfer genes directly into macrophages but most of them use viral vectors.⁴ Despite the high transfection efficiency of viral vectors, safety concerns have been raised in clinical trials regarding their potential toxicity.⁵

The use of nonviral vectors is attractive for in vivo gene delivery because it is simpler than using viral systems and lacks some of the risks inherent in the latter. Application of DEAE-dextran is one of the methods used for gene delivery to macrophages in vitro.^6,7 However, this method is generally not suitable for in vivo use due to problems associated with cellular toxicity, low efficiency or non-specific biodistribution. Recently, complexes of polylysine linked to ligands such as immunoglobulin G⁸ and mannose^9,10 with DNA have been reported to enhance gene expression in macrophages. Nevertheless, the transfection efficiency of these vectors is limited due to endosomal or lysosomal degradation.

One of the most promising nonviral gene delivery systems developed so far involves cationic liposomes. Various kinds of cationic lipids have been synthesized and shown to be able to deliver genes into cells both in vitro and in vivo.^{11,12,13,14,15,16} DC-Chol liposomes have already been used in gene therapy applications in clinical settings.^17,18 However, cationic liposomes do not exhibit any cell specificity in vivo. We have previously reported that novel galactosylated cholesterol derivatives in combination with DOPE efficiently transferred a plasmid DNA into human hepatoma cells (HepG2) via an asialoglycoprotein receptor-mediated mechanism.¹⁹

In the present study, we synthesized a novel mannosylated cholesterol derivative, cholesten-5-yloxy-N-(4-((1-imino-2--D-thiomannosylethyl)amino)butyl)-formamide (Man-C4-Chol), for gene delivery to macrophages which are known to express large numbers of mannose receptors on their surface. First, the transfection efficiency of mannosylated liposome/DNA complexes was investigated in mouse peritoneal macrophages and their suitability was compared with two types of cationic liposomes, 3[N-(N',N'-dimethylaminoethane)-carbamoyl] cholesterol (DC-Chol)/DOPE(6:4) and N-[1-(2,3-dioleyloxy)propyl]-N,N,N-trimethylammonium chloride (DOTMA)/DOPE(1:1) liposomes. Next, in vivo gene expression following intravenous or intraportal injection of liposome/DNA complexes was investigated with respect to mannose receptor recognition.

Results

Four types of liposomes (DOTMA/DOPE(1:1), DC-Chol/DOPE(6:4), Man-C4-Chol/DOPE(6:4), Man-C4-Chol/DC-Chol/DOPE(3:3:4)) were prepared with various molar ratios and their particle size was confirmed to be about 200 nm. The cationic liposomes were complexed with a plasmid DNA encoding a luciferase gene, pCMV-Luc. Figure 1 shows the zeta potentials of various ratios of pCMV-Luc/liposome complexes. The complex at a ratio of 1:2.5 was negatively charged, whereas the complexes at ratios of 1:5, 1:10 and 1:15 were positively charged. MTT assay revealed that all the complexes have low toxicity against macrophages (data not shown).

Figure 2 shows the expression of luciferase gene in macrophages transfected with the pCMV-Luc/cationic liposome complexes. The optimal DNA/liposome ratio for DOTMA/DOPE(6:4), DC-Chol/DOPE(6:4), Man-C4-Chol/DOPE(6:4), and Man-C4-Chol/DC-Chol/DOPE (3:3:4) liposomes was found to be 1:15, 1:10, 1:5 and 1:2.5, respectively. When compared at the optimal ratio, the in vitro transfection activity of Man-C4-Chol/DOPE(6:4) was significantly higher than that of DOTMA/DOPE(1:1) (P < 0.001) and DC-Chol/DOPE(6:4) liposomes (P < 0.01), which are known to act as highly potent vectors in various cell types. The presence of 20 mM mannose significantly inhibited gene expression with Man-C4-Chol/DC-Chol/DOPE(3:3:4) and Man-C4-Chol/DOPE(6:4) liposomes, but not that with DOTMA/DOPE(1:1) and DC-Chol/DOPE(6:4) liposomes (Figure 3).

The transfection activity of the DNA/liposome complexes at the ratio of 1:5 (w/w) was investigated in NIH3T3 cells for comparison (Figure 4). Gene expression with DC-Chol/DOPE(6:4) liposomes was significantly higher than that with Man-C4-Chol/DOPE(6:4) and Man-C4-Chol/DC-Chol/DOPE(3:3:4) liposomes (P < 0.01). Unlike the case of macrophages, the presence of 20 mM mannose did not inhibit gene expression with Man-C4-Chol containing liposomes.

Figure 5 shows the in vivo gene expression after intravenous injection of pCMV-Luc/liposome complexes. Plasmid DNA was complexed with various cationic lipids at a ratio of 1:7.0 (g/nmol). The luciferase activity in the liver after administration of pCMV-Luc complexed with Man-C4-Chol/DOPE(6:4) and Man-C4-Chol/DC-Chol/DOPE(3:3:4) liposomes was 16 and seven times higher than that with DC-Chol/DOPE(6:4) liposomes, respectively. In addition, pCMV-Luc with Man-C4-Chol/DOPE(6:4) exhibited four-fold higher gene expression in the spleen than DC-Chol/DOPE(6:4) liposomes. When mannosylated bovine serum albumin (Man-BSA)^20,21,22 was injected intravenously 5 min before intravenous injection of pCMV-Luc complexed with Man-C4-Chol/DOPE(6:4) liposomes, gene expression in the liver and spleen was significantly reduced (Figure 6). In contrast, predosing of Man-BSA did not affect gene expression of pCMV-Luc complexed with DC-Chol/DOPE(6:4) liposomes.

Figure 7 shows the distribution of expressed luciferase activity between the liver parenchymal cells (PC) and liver non-parenchymal cells (NPC) after intravenous injection of pCMV-Luc/liposome complexes. Gene expression of Man-C4-Chol/DOPE(6:4) liposomes was preferentially observed in NPC and the activity ratio of NPC to PC (NPC/PC ratio) on a cell-number basis was calculated to be 25.2. On the other hand, DC-Chol/DOPE(6:4) liposomes gave NPC and an PC/NPC ratio of 4.8. The cell specificity in gene expression with mannosylated liposomes was further evaluated by histochemical means using pCMV-LacZ as a reporter gene (Figure 8). The gene expression was mainly found in Kupffer cells. DC-Chol/DOPE(6:4) liposomes showed only slight gene expression.

Figure 9 shows the gene expression after intraportal injection of pCMV-Luc/liposome complexes. Luciferase activities in the liver and spleen after injection of pCMV-Luc complexed with Man-C4-Chol/DOPE(6:4) and Man C4-Chol/DC-Chol/DOPE(3:3:4) liposomes were higher than that with DC-Chol/DOPE(6:4) liposomes. The luciferase activity in the lung was markedly reduced compared with that following intravenous injection (Figures 5 and 9). Prior injection of Man-BSA significantly reduced gene expression in the liver and spleen (Figure 10), but this was not the case for gene expression with pCMV-Luc complexed with DC-Chol/DOPE(6:4) liposomes.

Discussion

Macrophages are known to be very little susceptible to gene transfection by nonviral vectors.²³ In macrophages, the present study showed that the gene expression and uptake of plasmid DNA complexed with mannosylated liposomes was higher than with DOTMA/DOPE(1:1) and DC-Chol/DOPE(6:4) liposomes, and significantly inhibited by excess free mannose (Figure 3). In NIH3T3 cells lacking mannose receptors,²⁴ on the other hand, the transfection efficiency of DNA with mannosylated liposomes was less than that with DC-Chol/DOPE(6:4) liposomes and not inhibited by excess free mannose (Figure 4). These results suggest that the complex of DNA and mannosylated liposomes is recognized and taken up by mannose receptors. In our previous study,¹⁹ lipid carriers with galactosylated cholesterol derivatives demonstrated cell-specific and highly efficient gene transfection in HepG2 cells expressing asialoglycoprotein receptors. Thus, the present approach is very effective in introducing cell-specific ligand structures to liposomes.

Erbacher et al¹⁰ investigated the suitability of various glycosylated polylysine derivatives for introducing plasmid DNA into human monocyte-derived macrophages and found that mannosylated polylysine exhibited high transfection activity. They also demonstrated that the transfection activity was markedly enhanced in the presence of chloroquine suggesting that prevention of endosomal and lysosomal degradation enhances gene delivery using receptor-mediated endocytosis. On the other hand, DOPE is known to accelerate the endosomal escape of plasmid DNA due to its pH-sensitivity¹¹ and the high transfection efficiency obtained in our gene carrier systems could, at least partly, be explained by the use of DOPE. This confirms the advantage of using liposomes which exhibit high flexibility as far as introducing various functional components is concerned.

Introduction of ligands for cell-surface receptors into liposomes has been attempted to improve transfection efficiency: as far as macrophages are concerned, the transferrin receptor-mediated transfer of liposome/DNA complexes has been successful.²³ In most cases, however, liposomes were coated with macromolecular ligands such as transferrin, immunoglobulins and asialoglycoproteins. In these cases, carriers are relatively easy to prepare, although there may be some problems involving factors such as reproducibility. Taking this into account, we have developed a low-molecular weight lipidic ligand for asialoglycoprotein¹⁹ and mannose receptors.

Intravenous injection of DNA/cationic liposome complexes results in gene expression in many tissues including the heart, lung, liver, kidney and spleen.²⁵ Although the level of gene expression varies from study to study, the lung invariably shows the highest level of expression.^{12,13,14,26,27,28,29} In addition, Li et al²⁹ compared intravenous and intraportal injection of cationic liposome/ protamine/DNA (LPD) and demonstrated that the gene expression in the lung was still higher than that in liver, even when the intraportal route was used. In contrast to this, intravenous and intraportal administration of Man-C4-Chol/DOPE(6:4) and Man-C4-Chol/DC-Chol/DOPE (3:3:4) liposome/DNA complexes resulted (with one exception) in higher gene expression in the liver than in the lung and spleen, suggesting that selective and efficient delivery of plasmid DNA to liver is achieved by these carriers (Figures 5 and 9). In the exceptional case of intravenous injection of Man-C4-Chol/DC-Chol/DOPE(3:3:4) liposome/plasmid DNA complex, gene expression in the lung was equal to that in liver (Figure 5). In addition, predosing of Man-BSA significantly reduced gene expression in the liver and spleen using mannosylated liposome/DNA complexes (Figures 6 and 10). These results support the participation of mannose receptors in the uptake of the mannosylated liposome/DNA complexes by liver Kuppfer and/or endothelial cells. Thus, splenic macrophages may also be targeted in this approach.

As far as the design of carriers for active targeting using receptor-mediated endocytosis is concerned, the density and stereochemistry of the ligand seem to be important. As far as the molecular design of galactosylated macromolecules is concerned, we demonstrated that galactosylated protein is recognized by liver cells in a manner directly related to the estimated surface density of the galactose residues.³⁰ It is likely that the same strategy applies to liposomes. The chemical structure and physicochemical characteristics of Man-C4-Chol seem to satisfy the conditions for transfection in macrophages by offering a cationic charge and being recognized by the mannose structure on the liposomal surface.

In summary, Man-C4-Chol, a novel mannosylated cholesterol derivative, exhibited a higher transfection activity than DC-Chol liposomes in macrophages based on a receptor-mediated mechanism. Since this compound itself has a positive charge, a high density of mannose residues can be deposited on the liposome surface without adversely affecting the binding ability of cationic liposomes to DNA. These characteristics of liposomes with a mannosylated cholesterol derivative are reflected in their superior in vivo gene transfection.

Materials and methods

Chemicals

N-(4-aminobuthyl) carbamic acid tert-buthyl ester was purchased from Tokyo Chemical Industry (Tokyo, Japan). DC-Chol was synthesized as reported.¹⁷ DOTMA was purchased from Tokyo Kasei Organic Chemicals (Tokyo, Japan). pGL3-Control Vector and pcDNA3 vector were purchased from Promega (Madison, WI, USA) and Invitrogen (Carlsbad, CA, USA), respectively. Cholesteryl chloroformate was purchased from Sigma Chemicals (St Louis, MO, USA) and Opti-MEM was purchased from Gibco BRL (Grand Island, NY, USA). Man-BSA was synthesized as reported previously.^20,21,22 All other chemicals were obtained commercially as reagent-grade products.

Construction and preparation of plasmid DNA

pCMV-Luc and pCMV-LacZ were constructed by subcloning the HindIII/XbaI firefly luciferase and bacterial -galactosidase (-gal) cDNA fragments from pGL3-control vector into the polylinker of pcDNA3 vector, respectively. Plasmid DNA was amplified in E. coli strain DH5a and then isolated and purified using a QIAGEN Plasmid Giga Kit (Qiagen, Hilden, Germany). Purity was confirmed by 1% agarose gel electrophoresis followed by ethidium bromide staining and the DNA concentration was measured by UV absorption at 260 nm.

Synthesis of cholesten-5-yloxy-N-(4-((1-imino-2--D-thiomannosyl-ethyl)amino)butyl)formamide (Man-C4-Chol)

As reported previously,¹⁹ N-(4-aminobutyl)-(cholesten-5-yloxyl)formamide was obtained from cholesteryl chloroformate and N-(4-aminobuthyl) carbamic acid tert-butyl ester. The product was reacted with 5 equivalents of 2-imino-2-methoxyethyl-1-thiomannoside in pyridine containing 1.1 equivalents of triethylamine for 24 h. After evaporation of the reaction mixture in vacuo, the resultant material was suspended in water and dialyzed against water for 48 h. After the dialyzate was lyophilized, the crude compound was suspended three times in ethylacetate and separated by decantation to purify it. ¹H-NMR (200 MHz, CDCl₃): 0.5-2 (m, cholesteryl skeleton), 2.5 (m, N-CH₂), 3.0 (broad s, N-CH₂), 3.7 (broad s, S-CH₂), 5.9 (s, C-NH). FAB-MS m/z: 736 (M⁺).

Harvesting and culture of macrophages

Male ICR mice weighing 20-25 g were obtained from the Shizuoka Agricultural Cooperative Association for Laboratory Animals, Shizuoka, Japan. Elicited macrophages were harvested from mice 4 days after intraperitoneal injection of 1 ml 2.9% thioglycolate medium (Nissui Pharmaceutical, Tokyo, Japan). The washed cells were suspended in RPMI 1640 medium supplemented with 10% heat-inactivated fetal bovine serum (FBS, Flow Laboratories, Irvine, UK), penicillin G (100 U/ml), and streptomycin (100 g/ml) and were plated on six- or 12-well culture plates (Falcon, Becton Dickinson, Lincoln Park, NJ, USA) at a density of 3 ´ 10⁵ cells/cm². After incubation for 24 h at 37°C in 5% CO₂-95% air, non-adherent cells were washed off with culture medium and cells were cultivated for another 48 h.

Particle size and zeta potential measurements

The particle sizes of liposome/plasmid DNA complexes were measured in a dynamic light scattering spectrophotometer (LS-900, Otsuka Electronics, Osaka, Japan). The zeta potential of liposome/plasmid DNA complexes was determined with a laser electrophoresis zeta-potential analyzer (LEZA-500T, Otsuka Electronics).

In vitro cytotoxicity test

The cytotoxicity of liposomes was evaluated by the 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) assay. After 3 days in culture, the cells plated on a 96-well plate at a density of 5.6 ´ 10⁴ cells/0.28 cm² were incubated with liposome/DNA complexes in 100 l OptiMEM I for 6 h. The cells were incubated for an additional 4 h following the addition of 0.5 mg/ml MTT solution and lysed overnight at 37°C with 10% SDS solution. The absorbance was measured at test and reference wavelengths of 570 and 660 nm, respectively, in a two-wavelength microplate photometer (Bio-Rad Model 450, Hercules, CA, USA).

Preparation of DNA/liposome complexes for in vitro experiments

Man-C4-Chol, DC-Chol, DOTMA, or a 1:1 mixture of Man-C4-Chol and DC-Chol was mixed with DOPE in chloroform and the mixture was dried, vacuum desiccated, and resuspend in 1 ml sterile 20 mM HEPES buffer (pH 7.8) in a sterile test tube. After hydration, the dispersion was sonicated for 5 min in a bath sonicator to form liposomes. Cationic liposomes and plasmid DNA (0.5 mg/ml) in a 12 ´ 75 mm polystyrene tube were diluted with Opti-MEM or Hank's balanced solution (HBSS) (pH 7.4) buffer at various ratios (w/w) before carrying out the transfection experiment.

In vitro transfection experiment

Macrophages were seeded in 10.5 cm² dishes at a density of 1.1 ´ 10⁶ cells/cm² in RPMI 1640 supplemented with 10% fetal calf serum. After 3 days in culture, the culture medium was replaced with Opti-MEM I containing 0.5 g/ml plasmid DNA and cationic liposomes. Six hours later, the incubation medium was replaced again with RPMI 1640 supplemented with 10% FBS and incubated for an additional 18 h. Then, the cells were scraped and suspended in 200 l pH 7.4 phosphate-buffered saline (PBS). One hundred microliters cell suspension was subjected to three cycles of freezing (liquid N₂ for 3 min) and thawing (37°C for 3 min), followed by centrifugation at 10 000 g for 3 min. The supernatants were stored at -20°C until the luciferase assay was performed. Ten microliters supernatant was mixed with 100 l luciferase assay buffer (Picagene, Toyo Ink, Tokyo, Japan) and the light produced was immediately measured in a luminometer (Lumat LB 9507, EG&G Berthold, Bad Wildbad, Germany). The activity is indicated as the relative light units per milligram protein. The protein content of the cell suspension in PBS was determined by the modified Lowry method using BSA as a standard. Experiments with NIH3T3 cells were carried out in a similar way.

Preparation of DNA/liposome complexes for in vivo experiments

DNA/liposome complexes for in vivo experiments were prepared as described by Templeton et al.¹² Equal volumes of DNA and stock liposome solution diluted with 5% dextrose to produce various ratios of DNA/ liposomes were mixed in 1.5 ml Eppendorf tubes at room temperature. Then, the DNA solution was added rapidly to the surface of the liposome solution using a Pipetman pipet and the mixture was agitated rapidly by pumping it up and down twice in the pipet tip.

In vivo delivery and gene expression

Five-week-old ICR mice were injected intravenously with 300 l of DNA/liposome complexes using a 30-gauge syringe needle. Intraportal injection was carried out under ether anesthesia. Six hours after injection, mice were killed and lung, liver, kidney, spleen and heart were removed and assayed for gene expression. For the competitive experiment, mannosylated bovine serum albumin (Man-BSA), at a dose of 20 mg/kg, was injected intravenously into mice. Five minutes later, liposome/DNA complexes were injected. The organs were washed twice with cold saline and homogenized with lysis buffer (0.05% Triton X-100, 2 mM EDTA, 0.1 M Tris, pH 7.8). The lysis buffer was added in a weight ratio of 5 l/mg for liver samples or 4 l/mg for other organ samples. After three cycles of freezing and thawing, the homogenates were centrifuged at 10 000 g for 10 min at 4°C and 20 l supernatant was analyzed to determine the luciferase activity using a luminometer (Lumat LB9507, EG&G Berthold). The protein concentration of each tissue extract was determined by the modified Lowry method. Luciferase activity in each organ was normalized to relative light units (RLU) per milligram extracted protein.

X-gal staining

Six hours following the intravenous injection of liposome/plasmid DNA complexes, the mice were killed and livers were fixed by infusing cold 1% glutaraldehyde containing 1 mM MgCl₂ dissolved in phosphate-buffered saline (PBS) and then immersed in the fixative for 40 min at room temperature. After rinsing with cold PBS, the tissue was subsequently embedded in Tissue-Tek OCT compound (Miles, Elkhart, IN, USA) and frozen in cold isopentan. Frozen sections, 5 m thick, were prepared on a Cryostat (Leica, Germany). The liver section was immersed in X-gal solution (1 mg/ml 5-bromo-4-chloro-3-indolyl--d-galactoside, 5 mM K₃[Fe(CN)₆], 5 mM K₂[Fe(CN)₆], 2 mM MgCl₂ in PBS, pH 7.4) and incubated overnight at 37°C. The tissue was mounted on glass slides, and couterstained with Nuclear Fast Red (Wako Pure Chemical Industries, Osaka, Japan).

Cellular localization of luciferase activity in liver

Six hour after intravenous injection of plasmid DNA/liposome complexes, each mouse was anesthetized with pentobarbital sodium (40-60 mg/kg) and the liver was perfused with perfusion buffer (Ca²⁺, Mg²⁺-free HEPES solution, pH 7.2) for 10 min and then with HEPES solution containing 5 mM CaCl₂ and 0.05% (w/v) collagenase (type I) (pH 7.5) for 10 min. Immediately after the start of perfusion, the vena cava and aorta were cut and the perfusion rate maintained at 3-4 ml/min. Following discontinuation of the perfusion, the liver was excised and the capsule membranes removed. The cells were dispersed in ice-cold Hank's-HEPES buffer containing 0.1% BSA by gentle stirring. The dispersed cells were filtered through the cotton mesh sieves, followed by centrifugation at 50 g for 1 min. The pellets containing PC were washed twice with Hank's-HEPES buffer by centrifuging at 50 g for 1 min. The supernatant containing NPC was similarly centrifuged twice more. The resulting supernatant was then centrifuged twice at 200 g for 2 min. PC and NPC were resuspended separately in ice-cold Hank's-HEPES buffer (2 ml for PC and 2 ml for NPC). The cell numbers and viability were determined by the trypan blue exclusion method. After three cycles of freezing and thawing, the homogenates were centrifuged at 10 000 g for 10 min at 4°C and 20 l supernatant was analyzed for luciferase activity using a luminometer.

Acknowledgements

This work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, and Culture of Japan.

1 Dietrich G et al. Delivery of antigen-encoding plasmid Dna into the cytosol of macrophages by attenuated suicide Listeria monocytogenes. Nature Biotechnol 1998; 16: 181-185,

2 Ohtani T et al. Efficient transfer and sustained high expression of the human glucocerebrosidase gene in mice and their functional macrophages following transplantation of bone marrow transduced by a retroviral vector. Proc Natl Acad Sci USA 1992; 89: 11332-11336, MEDLINE

3 Kohn DB, Sarver N. Gene therapy for HIV-1 infections. Adv Exp Biol Med 1996; 395: 421-428,

4 Schneider SD, Rusconi S, Seger RA, Hossle JP. Adenovirus-mediated gene transfer into monocyte-derived macrophages of patients with X-linked chronic granulomatous disease: ex vivo correction of feficient respiratory burst. Gene Therapy 1997; 4: 524-532, MEDLINE

5 Mahato RI, Takakura Y, Hashida M. Nonviral vectors for in vivo gene delivery: physicochemical and pharmacokinetic conditions. Crit Rev Ther Drug 1997; 14: 133-172,

6 Rupprecht AP, Coleman DL. Transfection of adherent murine peritoneal macrophages with a reporter gene using DEAE-dextran. J Immunol Meth 1991; 2: 157-163,

7 Mack KD et al. A novel method for DEAE-dextran mediated transfection of adherent primary cultured human macrophages. J Immunol Meth 1998; 211: 79-86,

8 Rojanasakul Y et al. Targeted gene delivery to alveolar macrophages via fc-receptor-mediated endocytosis. Pharmacol Res 1994; 11: 1731-1736,

9 Ferkol T, Perales JC, Mularo F, Hanson R. Receptor-mediated gene transfer into macrophages. Proc Natl Acad Sci USA 1996; 93: 101-105, MEDLINE

10 Erbacher P et al. Gene transfer by DNA/glycosylated polylysine complexes into human blood monocyte-derived macrophages. Hum Gene Ther 1996; 7: 721-729, MEDLINE

11 Felgner JH et al. Enhanced gene delivery and mechanism studies with a novel series of cationic lipid formulations. J Biol Chem 1994; 269: 2550-2561, MEDLINE

12 Templeton NS et al. Improved DNA:liposome complexes for increased systemic delivery and gene expression. Nature Biotechnol 1997; 15: 647-652, MEDLINE

13 Song YK, Liu F, Chu S, Liu D. Characterization of cationic liposome-mediated gene transfer in vivo by intravenous administration. Hum Gene Ther 1997; 8: 1585-1594, MEDLINE

14 Hong K, Zheng W, Baker A, Papahadjopoulos D. Stabilization of cationic liposome-plasmid DNA complexes by polyamine and poly(ethylene glycol)-phospholipid conjugates for efficient in vivo gene delivery. FEBS Lett 1997; 400: 233-237, Article MEDLINE

15 Okayama R, Noji M, Nakanishi M. Cationic cholesterol with a hydroxyethylamino head group promotes significantly liposome-mediated gene transfection. FEBS Lett 1997; 408: 232-234, MEDLINE

16 Oudrhiri N et al. Gene transfer by guanidinium-cholesterol cationic lipids into airway epithelial cells in vitro and in vivo. Proc Natl Acad Sci USA 1997; 94: 1651-1656, MEDLINE

17 Gao X, Huang L. A novel cationic liposome reagent for efficient transfection of mammalian cells. Biochem Biophys Res Commun 1991; 179: 280-285, MEDLINE

18 Nabel GJ et al. Direct gene transfer with DNA-liposome complexes in melanoma: expression, biologic activity, and lack of toxicity in humans. Proc Natl Acad Sci USA 1993; 90: 11307-11311, MEDLINE

19 Kawakami S et al. Asialoglycoprotein receptor-mediated gene transfer using novel galactosylated cationic liposomes. Biochem Biophys Res Commun 1998; 252: 78-83, MEDLINE

20 Nishikawa M et al. Pharmacokinetics of receptor-mediated hepatic uptake of glycosylated albumin in mice. Int J Pharmacol 1992; 85: 75-85,

21 Fujita T et al. Targeted delivery of human recombinant superoxide dismutase by chemical modification with mono- and polysaccharide derivatives. J Pharmacol Exp Ther 1992; 263: 971-978, MEDLINE

22 Fujita T et al. Therapeutic effects of superoxide dismutase derivatives modified with mono- or polysaccharides on hepatic injury induced by ischemia/reperfusion. Biochem Biophys Res Commun 1992; 189: 191-196, MEDLINE

23 Simoes S et al. Transfection of human macrophages by lipoplexes via the combined use of transferrin and pH-sensitive peptides. J Leuk Biol 1999; 65: 270-279,

24 Hebert E, Monsigny M. Galectin-3 mRNA level depends on transformation phenotype in ras-transformed NIH 3T3 cells. Biol Cell 1994; 81: 73-76, MEDLINE

25 Mahato RI, Kawabata K, Takakura Y, Hashida M. In vivo disposition characteristics of plasmid DNA complexed with cationic liposomes. J Drug Target 1995; 3: 149-157, MEDLINE

26 Zhu N, Liggitt D, Liu Y, Debs R. Systemic gene expression after intravenous DNA delivery into adult mice. Science 1993; 261: 209-211, MEDLINE

27 Liu F, Qi H, Huang L, Liu D. Factors controlling the efficiency of cationic lipid-mediated transfection in vivo via intravenous administration. Gene Therapy 1997; 4: 517-523, MEDLINE

28 Li S, Rizzo MA, Bhattacharya S, Huang L. Characterization of cationic lipid-protamine-DNA (LPD) complexes for intravenous gene delivery. Gene Therapy 1998; 5: 930-937, MEDLINE

29 Li S, Huang L. In vivo gene transfer via intravenous administration of cationic lipid-protamine-DNA (LPD) complexes. Gene Therapy 1997; 4: 891-900, MEDLINE

30 Nishikawa M et al. Galactosylated proteins are recognized by the liver according to the surface density of galactose moieties. Am J Physiol 1995; 268: G849-G856, MEDLINE

Figure 1 Zeta potential of DNA/liposome complexes at various ratios (w/w). Plasmid DNA (0.5 g/ml) was complexed with cationic liposomes at ratios of 1:2.5, 1:5, 1:10 and 1:15. Each value represents the mean ± s.d. values (n = 3).

Figure 2 Transfection activity of DNA/liposome complexes at various ratios (w/w) in macrophages. Plasmid DNA concentration was fixed at 0.5 g/ml in all experiments. Each value represents the mean ± s.d. values (n = 3).

Figure 3 Effect of copresence of 20 mM mannose on the transfection activity of DNA/liposome complexes in macrophages. Plasmid DNA (0.5 g/ml) was complexed with cationic liposomes at a ratio of 1:5 (w/w). Cells were transfected with DNA/liposome complexes in the presence and absence of 20 mM mannose. Each value represents the mean ± s.d. values (n = 3). Statistical analysis was performed by analysis of variance (

Figure 4 Transfection activity of DNA/liposome complexes in NIH 3T3 cells. Plasmid DNA (0.5 g/ml) was complexed with cationic liposomes at a ratio of 1:5. Cells were transfected with DNA/liposome complexes in the presence and absence of 20 mM mannose. Each value represents the mean ± s.d. values (n = 3). Statistical analysis was performed by analysis of variance (N.S., not significant).

Figure 5 Transfection activity of DNA/liposome complexes after intravenous administration in mice. Plasmid DNA (50 g) was complexed with cationic lipids at a ratio of 1:7.0 (g/nmol). Luciferase activity was determined 6 h after injection in the lung, liver, kidney, spleen and heart. Each value represents the mean ± s.d. values (n = 3).

Figure 6 Inhibition of gene expression of DNA/liposome complexes after intravenous injection by co-administration with Man-BSA in mice. Plasmid DNA (50 g) was complexed with cationic lipids at a ratio of 1:7.0 (g/nmol). Luciferase activity was determined at 6 h. DNA/liposome complexes were injected with or without Man-BSA. Each value represents the mean ± s.d. values (n = 3). Statistical analysis was performed by analysis of variance (

Figure 7 Hepatic cellular localization of luciferase activity after intravenous administration of pCMV-Luc/liposome complexes in mice. Luciferase activity was determined 6 h after injection in PC and NPC. Each value represents the mean ± s.d. values (n = 3).

Figure 8 -Galactosydase staining microphotograph of the liver frozen section after intravenous injection of pCMV-LacZ/liposome complexes in mice. Plasmid DNA (50 g) was complexed with cationic lipids at a ratio of 1:7.0 (g/nmol). -Galactosydase activity was determined 6 h after injection. There was localized expression in Kuppfer cells (arrows). Original magnification, 200´.

Figure 9 Transfection activity of DNA/liposome complexes after intraportal injection in mice. Plasmid DNA (50 g) was complexed with cationic lipids at a ratio of 1:7.0 (g/nmol). Luciferase activity was determined 6 h after injection in the lung, liver, kidney, spleen, and heart. Each value represents the mean ± s.d. values (n = 3).

Figure 10 Inhibition of gene expression of DNA/liposome complexes after intraportal injection by co-administration with Man-BSA in mice. Plasmid DNA (50 g) was complexed with cationic lipids at a ratio of 1:7.0 (g/nmol). Luciferase activity was determined at 6 h. DNA/liposome complexes were injected with or without Man-BSA. Each value represents the mean ± s.d. values (n = 3). Statistical analysis was performed by analysis of variance (