Characteristics and biodistribution of cationic liposomes and their DNA complexes

Journal of Controlled Release 69 (2000) 139–148 www.elsevier.com / locate / jconrel

Characteristics and biodistribution of cationic liposomes and their DNA complexes a, a a a Hideki Ishiwata *, Norio Suzuki , Shuichi Ando , Hiroshi Kikuchi , Takayuki Kitagawa b a

Pharmaceutical Formulation Research Laboratory, Daiichi Pharmaceutical Co., Ltd., Tokyo R& D Center, 16 – 13 Kita-Kasai 1 -Chome, Edogawa-ku, Tokyo 134 -8630, Japan b Department of Biochemistry and Cell Biology, National Institute of Infectious Diseases, Tokyo 162 -8640, Japan Received 5 April 2000; accepted 16 June 2000

Abstract We have developed some novel liposome formulations for gene transfection. The formulations consisting of O,O9ditetradecanoyl-N-(a-trimethyl ammonio acetyl) diethanolamine chloride (DC-6-14) as a cationic lipid, phospholipid and cholesterol showed effective gene transfection activity in cultured cells with serum and in vivo, i.e., intraperitoneal injection in mice. In this report, the physicochemical characteristics and biodistribution of the liposomes containing DC-6-14 (DC-6-14 liposomes) as a drug (gene) carrier for gene therapy were investigated in vitro and in vivo. DC-6-14 liposome–DNA complexes were usually thought to have positive surface charge. However, depending on the ratio of DNA to liposomes, zeta-potential of the complexes became negative. The diameter of the complexes also depended on the DNA–liposome ratio, and showed a maximum when their surface potential was neutral. When biodistribution of the complexes was determined after intravenous injection, positively charged complexes showed an immediate lung accumulation. On the other hand, negatively charged complexes did not show lung accumulation. These results have suggested that biodistribution of the DNA–liposome complexes, prepared with DC-6-14 liposomes, depends on their surface charge. Therefore, some surface modification of DC-6-14 liposomes may improve the biodistribution and hence the targetability of their DNA complexes.  2000 Elsevier Science B.V. All rights reserved. Keywords: Cationic liposome; Gene therapy; Biodistribution; Plasmid DNA; Blood cell aggregation

1. Introduction The importance of DNA therapeutics in gene therapy, in particular, has led to increased research and development in this area [1]. The current standard for gene therapy utilizes viral vectors as gene *Corresponding author. Tel.: 181-3-5696-8307; fax: 181-35696-8228. E-mail address: [email protected] (H. Ishiwata).

delivery systems [2]. Although high transfection efficiencies result from the use of viral vectors, there are some immune troubles occurring on repeated dosing. Compared to viruses, chemical DNA carriers have less immune reactions and numerous advantages. Particularly, there are a number of publications with convincing evidence that cationic liposomes can mediate gene delivery, by demonstrating detectable expression of a reporter gene in cultured cells and also in vivo via local injection [3]. However, trans-

0168-3659 / 00 / $ – see front matter  2000 Elsevier Science B.V. All rights reserved. PII: S0168-3659( 00 )00293-5

140

H. Ishiwata et al. / Journal of Controlled Release 69 (2000) 139 – 148

gene expression induced by cationic liposomes was usually low in serum-containing medium and was often toxic to cultured cells. We have developed new liposome formulations for effective gene transfection with newly synthesized, positively charged lipids. As a result of screening, a series of positively charged lipids was selected by their high transfection activity. The dimyristyl type, named DC-6-14, showed the highest transfection activity in this series. One of the liposome formulation prepared with DC-6-14, dioleoyl phosphatidylethanolamine (DOPE) and cholesterol (4:3:3 in a molar ratio; named TFL-08) showed effective gene transfection activity in cultured cells with serum and in vivo, i.e., intraperitoneal injection in nude mice [4]. The liposomes containing DC-6-14 in the DNA– liposome complexes retained their morphology in the presence of serum, and this property may account for the high activity in serum-containing media [5]. From these features of the liposomes, they are expected to be effective in gene transfection in vivo by systemic injection. For that purpose, the cationic liposomes as a drug carrier must deliver gene from blood stream to specific sites in the body without gene degradation and also as a vector to transport it inside the cells. Biodistribution of liposomes as drug carriers has been studied in the last 3 decades [6]. It has been generally recognized that negatively charged and neutral liposomes are cleared from the blood and accumulated in the liver and spleen after intravenous administration. Positively charged liposomes, containing stearylamine [7] or aminoglycolipids [8], were also shown to be accumulated in the liver and spleen after intravenous injection, though their blood circulating time was increased compared with the liposomes without such cationic lipids. During the last few years, several groups have reported that a significant level of gene expression in the lung was observed following systemic injection of complexes of DNA and liposomes containing cationic lipid such as DOTMA hN-[2,3-(dioleyloxy)propyl]-N,N,N-trimethylammonium bromidej [9,10] or DDAB (dimethyldioctadecylammonium bromide) [11]. The findings implied that these cationic liposomes delivered DNA effectively to the lung after intravenous injection. However, the mechanism underlying the biodistribution has remained

unelucidated. In this report, we investigated the characteristics of newly developed cationic liposomes containing DC-6-14 as a drug (gene) carrier. Then the influence of the ratio of DNA to the liposomes on the biodistribution of their DNA complexes after intravenous injection was further investigated. The possible mechanism of their biodistribution is also discussed.

2. Materials and methods

2.1. Materials Dioleoyl phosphatidylethanolamine (DOPE) was obtained from Nippon Oil and Fats (Japan). Cholesterol (Chol) was purchased from Sigma (USA). O,O9-Ditetradecanoyl-N-(a-trimethyl ammonio acetyl) diethanolamine chloride (DC-6-14) shown in Fig. 1 was supplied by Sogo Pharmaceutical (Japan). 3 H-Cholesteryl hexadecyl ether (CHE) and 14 Cinulin (Mr 5000–5500) were obtained from New England Nuclear (USA). Plasmid pGreen Lantern-1 was purchased from Gibco BRL (USA). All other chemicals were of special grade.

2.2. Preparation of liposomes and DNA–liposome complexes The lipid composition of the DC-6-14-containing liposome (DC-6-14 liposome) was DC-6-14 / DOPE / Chol (4:3:3, molar ratio). The procedure for the preparation of liposomes was essentially identical to that described by Ishiwata et al. [12]. These lipids were dissolved in chloroform and then dried to a thin film with a rotary evaporator under reduced pressure.

Fig. 1. Chemical structure of O,O9-ditetradecanoyl-N-(a-trimethyl ammonio acetyl) diethanolamine chloride (DC-6-14).

H. Ishiwata et al. / Journal of Controlled Release 69 (2000) 139 – 148

Remaining solvent was removed in vacuo for 2 h. Labeling of the liposome membranes was performed, using 3 H-CHE added to the appropriate lipid mixtures prior to solvent evaporation. The lipid film was then hydrated with 9% sucrose solution. The vesicles were extruded five times through polycarbonate filters (Nuclepore, Coster, USA) of 0.2 mm pore size. 14 C-Inulin entrapped liposomes were prepared using 9% sucrose solution containing 14 C-inulin. Untrapped inulin was removed by ultracentrifugation (SRP70AT, Hitachi, Japan). The dispersion of liposomes was diluted 20-fold with phosphate-buffered saline (PBS), pH 7.4, and centrifuged at 150 000 g for 30 min at 48C. The supernatant was discarded and the pellet resuspended in PBS and centrifuged. The final pellet obtained after three repeated centrifugation steps was resuspended in 9% sucrose solution. DNA–liposome complexes were prepared by the addition of DC-6-14 liposome suspension into 9% sucrose solution containing plasmid pGreen Lantern1. Mixing was performed with low and high plasmid concentrations of 15–20 mg / ml and 100–300 mg / ml, respectively. The low concentration was similar to that for in vitro use, while the high concentration condition was employed for in vivo experiments. The ratio of plasmid to lipid (P/L) was from 0 to 150 g plasmid per total lipid mol in liposomes.

2.3. Characterization of liposomes and DNA– liposome complexes The particle sizes of resulting liposomes and DNA–liposome complexes were determined by dynamic light scattering using a laser particle analyzer DLS 700 (Otsuka Electronics, Japan). Average diameter (weight average) was measured by scattered intensity at 258C. The shape of the particles was confirmed by electron microscopy. The electrophoretic mobilities of the particles in PBS were measured with an electrophoretic light scattering spectrophotometer (ELS 800, Otsuka Electronics, Japan) at 258C. The zeta-potential of those over about 100 nm in diameter was calculated from Smoluchowski’s equation [13].

2.4. Liposome stability in mouse blood 14

C-Inulin (Mr 5000–5500) entrapped DC-6-14

141

liposomes were incubated in mouse heparinized blood, plasma and blood cell suspension. Mouse heparinized blood was collected from male BALB / c mice from SLC (Japan). Plasma and blood cells were separated by centrifugation (3000 rpm, 10 min, 48C). Blood cells were washed three times with PBS by centrifugation and resuspended at 100% (v / v) in PBS. One volume of liposome suspension (5 mM total lipid) was added to nine volumes each of pre-warmed blood, plasma and blood cell suspension, and incubated at 378C with gentle stirring. Samples taken from the incubation mixtures at selected times were diluted 10-fold with PBS. 14 CInulin released from liposomes to external medium was collected by ultrafiltration in a disposable unit (Centrisart I; Mr cut-off, 20 000; Sartorius, Germany) immediately after sampling [14]. The radioactivities of inulin in sample solution and filtrate were measured in a liquid scintillation analyzer (LSC 900, Aloka). 14 C-Inulin retention was calculated from the following equation [12]: % Retention 5 100 3 [1 2 (DPMt / DPM) ? (DPMx 2 DPMo) /(DPMt 2 DPMo)] where DPMo is the radioactivity of released inulin from liposomes incubated in PBS at 378C at time zero. DPMt is the total radioactivity of inulin in liposomes incubated in PBS. DPMx is the radioactivity of released inulin from liposomes incubated in blood, plasma or blood cell suspension at 378C at time x. DPM is the total radioactivity of inulin in liposomes incubated in blood, plasma or blood cell suspension at 378C at time x. DPMt / DPM is a correction term which is used to minimize the pipetting and other systematic errors in the measurement. The value of DPMt / DPM ranged from 0.9 to 1.1.

2.5. Biodistribution of liposomes and DNA– liposome complexes Male BALB / c mice weighing 20–25 g from SLC were administered the samples via tail vein. For kinetic studies of DC-6-14 liposomes, the injected dose was 2.6 mmol lipid per body and double ( 3 HCHE and 14 C-inulin) labeled liposomes were used. Biodistribution of DNA–liposome complexes was

142

H. Ishiwata et al. / Journal of Controlled Release 69 (2000) 139 – 148

investigated with a dose of 0.5 mmol lipid per body. Complexes were prepared by the addition of 3 HCHE labeled DC-6-14 liposomes into the 9% sucrose solution containing 100–300 mg / ml of plasmid pGreen Lantern-1. At a specified time, mice were sacrificed for collecting blood, lung, liver, spleen and kidney. Collected organs were homogenized with saline and associated radioactivities were measured in a liquid scintillation analyzer (LSC 900, Aloka). The results were expressed in percent of injected radioactivities accumulated in each organ. Plasma volume was assumed to be 4.88% of body weight [15].

ml) was modeled on the procedure to make them for in vitro gene transfection experiments. The average diameter of each complex increased in comparison with that of DC-6-14 liposomes and showed a peak at the P/L of about 85 g / mol in both concentration conditions (Fig. 2a). However, the maximum value

2.6. Count of blood cell after liposome injection DC-6-14 liposomes (2.6 mmol) were injected into mice via the tail vein and the animals were sacrificed to collect the blood at 10 min or 2 h after the injection. Blood cells in the blood samples containing EDTA were counted using an automatic cell counter (H-1 system, Technicon). The results were expressed as percent of control cell count, obtained from non-treated normal mice. Blood cell counts of the mice receiving 9% sucrose solution, i.e., solvent of liposome suspension, were investigated at 10 min post-injection.

3. Results

3.1. Characteristics of DC-6 -14 liposomes and their DNA complexes The physicochemical characteristics of the complexes were investigated with various mixing ratios of DNA to liposomes. The diameter of DC-6-14 liposomes and DNA–liposome complexes were measured by dynamic light scattering. The diameter of DC-6-14 liposomes was homogeneous, being about 150 nm. This size was reproducible by manufacture on milliliter and liter scales (data not shown). DNA– liposome complexes were prepared by the addition of liposomes to plasmid DNA solution. Mixing was performed under high and low plasmid concentration conditions (see Materials and Methods). The high concentration condition (100–300 mg / ml) was used to prepare the complexes for later animal experiments. The low concentration condition (15–20 mg /

Fig. 2. Characteristics of DNA–liposome complexes as a function of the ratio of plasmid to lipids. DNA–liposome complexes were prepared by the addition of DC-6-14 liposomes to plasmid pGreen Lantern-1. Mixing was performed under two concentration conditions; concentrations of plasmid in the mixtures were 15–20 mg / ml and 100–300 mg / ml for low (triangles) and high (diamond) concentration, respectively. The high concentration condition was used to prepare the complexes for in vivo experiments. The average sizes (a) of resulting DNA–liposome complexes were measured by dynamic light scattering. Zeta-potential (b) was calculated using Smoluchowski’s equation [13] from the electrophoretic mobilities in PBS at 258C. The mobilities were obtained from electrophoretic light scattering measurements.

H. Ishiwata et al. / Journal of Controlled Release 69 (2000) 139 – 148

143

varied with the conditions and was about 400 nm and 800 nm for the low and high concentration conditions, respectively. On the other hand, the dependence of zeta-potential on the P/L ratio indicated similar behavior in both high and low plasmid concentration conditions (Fig. 2b). At P/L ratios of 0 to 75 g / mol, the surface charge of the complexes was positive and was comparable to that of DC-6-14 liposomes. At about 85 g / mol the surface charge became neutral, and they had negative surface charge at ratios over 107 g / mol. The complexes at a P/L ratio of 125 g / mol exhibited negative charge potential, but showed an effective gene transfection in vitro [5].

3.2. Interaction of liposomes with blood cells in vitro Before in vivo application of the DNA complexes, prepared with DC-6-14 liposomes, in vitro characteristics of the cationic liposomes were investigated first. Liposome stability in blood was determined using the marker leakage index. DC-6-14 liposomes, containing 14 C-inulin as an internal aqueous phase maker, were incubated in blood at 378C with gentle stirring. The leakage of the substance would indicate lysis of the liposomes. A gradual marker release was observed from liposomes, and the retention of the marker was about 60–70% even after a 2-h incubation (Fig. 3). Similar retention kinetics were observed on incubation in plasma and blood cell suspension. Positively charged liposomes like stearylaminecontaining ones were reported to usually enhance the permeability of biomembranes and damage the cells [16]. DC-6-14 liposomes can interact with both plasma proteins and blood cells, leading to a gradual liposome lysis (Fig. 3). However, hemolytic activity of DC-6-14 liposomes was weaker than that of stearylamine-containing liposomes and was comparable to that of TMAG [N-(a-trimethyl ammonio acetyl)-didodecyl-D-glutamate chloride]-containing liposomes (data not shown).

3.3. Biodistribution of DC-6 -14 liposomes and their DNA complexes in mice In vivo fate of the liposomes after intravenous injection in mice was investigated (Fig. 4), for in

Fig. 3. Stability of DC-6-14 liposomes in mouse blood. DC-6-14 liposomes containing 14 C-inulin were incubated in blood (diamond), plasma (upper triangles) and blood cell suspension (lower triangles) at 378C with gentle stirring. One volume of liposome suspension was added to nine volumes each of prewarmed blood, plasma and blood cell suspension; the lipid concentration during the incubation was 500 mM. At selected times, samples were taken from incubation mixtures and released inulin from liposomes to external medium, was collected by ultrafiltration in a disposable unit [14].

vitro characteristics such as hemolytic activity (data not shown) and stability in blood (Fig. 3) of the DC-6-14 liposomes were found to be suitable for in vivo application as a drug (gene) carrier. Liposomes were labeled with 3 H-CHE in the lipid bilayer and 14 C-inulin in the interior aqueous phase, respectively. Over 60% of injected radioactivities of 3 H-CHE and 14 C-inulin were found in the lung at 3 min postinjection. Thereafter, the amount of these labels in the lung decreased and that in the liver coincidently increased. However, only small amounts of 3 H and 14 C radioactivities were observed in the plasma, spleen and kidney in this period. Similar kinetics were observed with a lower lipid dose of about 0.5 mmol (data not shown). Because DC-6-14 liposomes were noted to show the accumulation in the lung immediately after intravenous injection, lung accumulation of the liposomes complexed with DNA was also investigated at 3 min post-intravenous injection as a function of the plasmid to liposome ratio (Fig. 5). DNA–liposome complexes were made by mixing 3 H-CHE-labeled DC-6-14 liposomes into plasmid pGreen Lantern-1. The mixing was performed under the high plasmid

144

H. Ishiwata et al. / Journal of Controlled Release 69 (2000) 139 – 148

Fig. 5. Biodistribution of DNA–liposome complexes in mice. The amounts of 3 H-CHE in plasma and various tissues were determined at 3 min post-injection in mice receiving 0.5 mmol of lipids via the tail vein. Plasma (diamond hollowed), lung (diamond filled) and liver (round filled) were collected. The results were expressed as percent of total injected radioactivities6S.D. Plasma volume was assumed to be 4.88% of body weight [15].

Fig. 4. Kinetics of biodistribution of DC-6-14 liposomes injected into mice via the tail vein. The amounts of 3 H-CHE (a) and 14 C-inulin (b) in plasma and various tissues were determined in mice receiving 2.6 mmol of lipids. At selected times, mice were sacrificed and plasma (diamond hollowed), lung (diamond filled), liver (round filled), spleen (lower triangles filled) and kidney (upper triangles hollowed) were collected. The results were expressed as percent of total injected radioactivities6S.D. Plasma volume was assumed to be 4.88% of body weight [15].

concentration condition (see Materials and Methods). The complexes with P/L ratios of 0 to 83.3 g / mol had positive charge and those with P/L ratios of 107 to 150 g / mol had negative charge (Fig. 2b). Their average diameter was from 150 to 800 nm, depending on the P/L ratio (Fig. 2a). Significant accumulation in the lung was observed with the complexes of 0 to 83.3 g / mol, that is, positively charged complexes. In contrast, the negatively charged complexes of

107 to 150 g / mol decreased lung accumulation and increased liver uptake were observed. Thus, lung accumulation of the complexes was dependent on the surface charge of the complexes. Accumulation in the plasma, spleen and kidney was not remarkable with any charged complexes (data not shown). The size of complexes did not affect their biodistribution, since positively or negatively charged complexes of about 150 nm and those of about 700 nm in diameter showed similar biodistribution.

3.4. Blood cell counts in blood stream Considering that the blood cells have a negatively charged cell surface, lung accumulation of DC-6-14 liposomes and their DNA complexes may be caused by their positive charge. One possibility was that the capillaries in the lung were blocked by aggregates formed via electrostatic interaction between blood cells and positively charged liposomes. In such case, blood cell count might decrease following injection of positively charged particles. Blood cells in normal mice were enumerated to be 5.0310 3 / ml, 9.1310 6 / ml and 1.3310 6 / ml for leukocytes, erythrocytes and platelets, respectively, which were comparable to the previous publication

H. Ishiwata et al. / Journal of Controlled Release 69 (2000) 139 – 148

Fig. 6. Blood cell counts after intravenous injection of DC-6-14 liposomes. Mice received 2.6 mmol of lipids via the tail vein and were sacrificed at 10 min (gray) or 2 h (dotted) post-injection to collect their blood. After EDTA treatment, blood cells in the blood samples were counted using an automatic cell counter. The results were expressed as percent of control. Control (filled) cell counts obtained in non-treated normal mice were 5.0310 3 / ml, 9.1310 6 / ml and 1.3310 6 / ml for leukocytes, erythrocytes and platelets, respectively. Blood cell counts of the mice receiving sucrose solution (hollowed), as solvent of liposome suspension, are also indicated.

[15]. At 10 min post-injection of the liposomes, the number of platelets and leukocytes dramatically decreased to about 60%, and they recovered to over 80% after 2 h (Fig. 6). The decrease in numbers of leukocytes and platelets was confirmed by microscopic observation (data not shown). Erythrocyte count was not influenced by the DC-6-14 injection. The number of platelets and leukocytes in rats also decreased to about 20% at 3 min post-intravenous injection of DC-6-14 liposomes, and recovered to 100% within 30 min, while erythrocyte count was not influenced. Neutral and negatively charged liposomes did not induce such blood cell decrease in rats (unpublished data).

4. Discussion New liposome formulations were developed for effective gene transfection with newly synthesized positively charged lipids. As a result of screening, a series of cationic lipids was selected because of their high transfection activity in cultured human cell lines. This series of derivatives differs only in the structure of hydrophobic acyl chains. DC-6-14,

145

dimyristyl type (diC14:0), is significantly more active than any other acyl chain types. One of the liposome formulation prepared with DC-6-14, DOPE and Chol (4:3:3 in a molar ratio, named TFL-08) showed effective gene transfection activities in cultured human cells in the presence of serum and in the peritoneal cavity of nude mice [4]. The optimal mixing ratio of DNA to liposomes was investigated in vitro by Serikawa et al. [5] and was found to be 1.0 mg plasimd DNA to 8 nmol of liposomal lipids (P/L; 125 g / mol). Because of the high transfection activity in serum-containing medium [4,5], these cationic liposomes are expected to be usable in vivo by systemic injection. Usually, cationic liposomes containing stearylamine have been thought to be inadequate for therapeutic application mainly due to their toxicity [17,18]. However, DC-6-14 liposomes were much improved in respect of hemolytic activity and stability in blood in vitro. The diameter of DC-6-14 liposomes was homogeneous, about 150 nm, and reproducible by manufacture on small (milliliter) and large (liter) scales (unpublished data). Moreover, it was also found that this liposome formulation can serve as freeze–dried empty liposomes [14,19]. These results have suggested potential utility of DC6-14 liposomes for in vivo application, as a vector for genes and drugs. In this report, DNA–liposome complexes were prepared with DC-6-14 liposomes by the addition to plasmid DNA. The diameter of each complex increased in comparison with that of DC-6-14 liposomes. This increase in diameter of the complexes suggested that negative charge of plasmid induced liposome bridging by electrostatic interaction with positive charge on the liposome surface membranes. Because of the accessibility of plasmid and liposomes, larger complexes in size would be observed when they are prepared with the solution of higher plasmid concentration. The maximum complex diameter was observed with the complex of around 85 g / mol. At this P/L ratio, net neutral zeta-potentials of the complexes were observed. With decreasing P/L ratio, the zeta-potential became positive, and it was negative at the higher P/L ratios. These results were consistent with the case of DMRIE (1,2dimyristyloxypropyl-3-dimethyl-hydroxyethyl ammonium bromide)-containing liposomes [20]. Nega-

146

H. Ishiwata et al. / Journal of Controlled Release 69 (2000) 139 – 148

tively charged complexes were suggested to show DNA protruding from the surface of the complexes. It was proposed that a portion of DNA was displaced from the vesicle surface and, therefore, the DNA shielded the positive charges of the cationic lipid without being neutralized by the lipid, giving the complex a more negative zeta potential than DNA tightly binding to the surface [20]. Generally, a positive charge on the liposomes was important for the interaction of liposomes with DNA as well as with cell surface, and the binding of the complexes to cells was caused by electrostatic interaction between positive charge on the complexes and negative charge on cell surface. The cationic complexes were supposed to bind to glycosaminoglycans on the cell surfaces by electrostatic interactions [21,22]. However, in the case of DC-6-14 liposome, the complexes at a P/L ratio of 125 g / mol had negative charge potential (Fig. 2b), and showed an effective gene transfection by cultured cells in the medium containing 5% fetal calf serum. Serikawa et al. recently showed the effective expression of GFP gene using DC-6-14 liposomes in a Hela cell line with P/L ratio from 40 to 400, and revealed that the P/L ratios around 125 were most effective [5]. Another group reported that negatively charged complexes showed an effective transfection [23,24]. These negatively charged complexes would interact with different sites of the cell surface from those interacting with positively charged ones. In addition, DC-6-14 liposome has a unique property for gene delivery to human cells in serum-containing medium [4,5]. In spite of numerous publications about in vivo fate of neutral and negatively charged liposomes, those concerning cationic liposomes have been scarce. Positively charged liposomes containing stearylamine [7] or aminoglycolipids [8] were reported to show longer circulation time in blood, effected by escaping liver uptake. These liposomes were aimed at their circulating in blood and reaching specific desired sites passively when injected intravenously. However, cationic DC-6-14 liposomes did not show any evidence of circulation in blood; they accumulated in the lung immediately after injection and were thereafter taken up by the liver (Fig. 4). In the case of DNA complexes with DC-614 liposomes, immediate lung accumulation after

intravenous injection was also observed with positively charged ones, while negatively charged ones did not (Fig. 5). DC-6-14 liposomes can interact with blood cells in blood stream because aggregates of blood cells were observed following the addition of the liposomes to blood in vitro. Blood cell counts decreased when DC-6-14 liposomes were in the lung and returned to normal levels as liposomes left the organ (see Figs. 4 and 6). Therefore, it was suggested that intravenously-injected DC-6-14 liposomes form aggregates with blood cells and these aggregates were trapped in lung capillaries temporally, and as the liposomes left the lung, aggregate breakage might occur. The interaction between the cells and liposomes might be relatively weak as to induce degradation of aggregates in lung capillaries. Then, blood cells and liposomes were gradually redistributed to blood stream and to liver, respectively. Glycolipid liposomes were suggested to be adsorbed on blood cell surfaces to escape liver uptake [25]. Interaction of stearylamime liposomes with blood cells induced cell lysis [16]. This diversity might be caused by the properties of cationic lipids, though the interaction itself would be induced by electrostatic interaction. Recently, DNA–liposome complexes injected intravenously were reported to show gene expression in the lung [9–11] but the charge of the complexes was unknown. We have demonstrated in this study that the surface charge of the complexes, prepared with DC-6-14 liposomes, varied from positive to negative depending on the DNA–liposome ratio, and only the positively charged DNA–liposome complexes showed lung accumulation. Therefore, positively charged complexes could show gene expression in lung after intravenous injection, but negatively charged ones could not. The lung might be targeted by these positively charged complexes according to the biodistribution properties of DC-614 liposomes. Negatively charged complexes prepared with DC-6-14 liposomes showed liver accumulation at 3 min post-injection similarly to negatively charged liposomes [6] and naked plasmid [26]. However, recovery of the complexes from the collected organs was decreased (about 55%). Biodistribution of the rest of the injected complexes remained uncertain; they might be delivered to other tissues. To achieve the gene targeting by the DNA–liposome

H. Ishiwata et al. / Journal of Controlled Release 69 (2000) 139 – 148

complexes, some modification of the complexes will be necessary to permit the complex to circulate in blood for a relatively long time in order to interact with the desired sites within the body.

[3] [4]

5. Conclusion [5]

In this report, we investigated the characteristics of cationic DC-6-14 liposomes and their DNA complexes in vitro and in vivo from the viewpoint of a gene delivery system. Most in vitro gene transfection experiments have been performed with positively charged complexes. However, negatively charged complexes can be made depending on the DNA– liposome ratio, and they also showed remarkable gene transfection in serum-containing medium [5]. After systemic injection, positively charged complexes showed immediate lung accumulation, while negatively charged ones did not. This phenomenon was thought to be induced by entrapment in lung capillaries of aggregates formed from blood cells and positively charged complexes by electrostatic interaction, because blood cell counts decreased following the injection of the cationic liposomes. Therefore, these positively charged complexes might induce effective gene expression in the lung. Specific site targeting of gene by this cationic liposome via intravenous injection will be achieved with some modifications to improve the surface of the complexes and their circulation in blood.

[6] [7]

[8]

[9]

[10]

[11]

[12]

[13]

Acknowledgements We wish to thank Miho Takahashi, Mayumi Shibano and Kiyoshi Ebihara for their technical assistance. This study was supported in part by a grant from the Human Science Foundation of Japan (No. 31093).

[14]

[15] [16]

References [1] R.C. Mulligan, The basic science of gene therapy, Science 260 (1993) 926–932. [2] T. Kitada, S. Asakawa, N. Hattori, H. Matsumine, Y. Yamamura, S. Minoshima, M. Yokochi, Y. Mizuno, N.

[17]

[18]

147

Shimizu, Mutations in the parkin gene cause autosomal recessive juvenile parkinsonism, Nature 392 (1998) 605– 608. X. Gao, L. Huang, Cationic liposome-mediated gene transfer, Gene Ther. 2 (1995) 710–722. A. Kikuchi, Y. Aoki, S. Sugaya, T. Serikawa, K. Takakuwa, K. Tanaka, N. Suzuki, H. Kikuchi, Development of novel cationic liposomes for efficient gene transfer into peritoneal disseminated tumor, Hum. Gene Ther. 10 (1999) 947–955. T. Serikawa, N. Suzuki, H. Kikuchi, K. Tanaka, T. Kitagawa, A new cationic liposome for efficient gene delivery with serum into cultured human cells: a quantitative analysis using two independent fluorescent probes, Biochim. Biophys. Acta 1467 (2000) 419–430. G. Gregoridis, Liposomes as Drug Carriers – Recent Trends and Progress, Wiley, Chichester, 1988. G. Gregoriadis, D.E. Neeruniun, Control of the rate of hepatic uptake and catabolism of liposome-entrapped proteins injected into rats. Possible therapeutic applications, Eur. J. Biochem. 47 (1974) 179–185. H. Aoki, C. Sun, K. Fuji, K. Miyajima, Disposition kinetics of liposomes modified with synthetic aminoglycolipids in rats, Int. J. Pharm. 115 (1995) 183–191. N. Zhu, D. Liggitt, Y. Liu, R. Debs, System gene expression after intravenous DNA delivery in adult mice, Science 261 (1993) 209–211. Y.K. Song, D. Liu, Free liposomes enhance the transfection activity of DNA / lipid complexes in vivo by intravenous administration, Biochim. Biophys. Acta 1372 (1998) 141– 150. K. Hong, W. Zheng, A. Baker, D. Papahadjopoulos, Stabilization of cationic liposome–plasmid DNA complexes by polyamines and poly(ethylene glycol)–phospholipid conjugates for efficient in vivo gene delivery, FEBS Lett. 400 (1997) 233–237. H. Ishiwata, A. Vertut-Doi, T. Hirose, K. Miyajima, Physical–chemistry characteristics and biodistribution of poly(ethylene glycol)-coated liposomes using poly(oxyethylene) cholesteryl ether, Chem. Pharm. Bull. 43 (1995) 1005–1011. A.W. Adamson, Physical Chemistry of Surface, 2nd Edition, Interscience, New York, 1967. K. Yachi, H. Harashima, H. Kikuchi, R. Sudo, H. Yamauchi, K. Ebihara, H. Matsuo, K. Funato, H. Kiwada, Biopharmaceutical evaluation of the liposomes prepared by rehydration of freeze–dried empty liposomes (FDELs) with an aqueous solution of a drug, Biopharm. Drug Dispos. 17 (1996) 589–605. Y. Tanimoto, Hematology – Intersection of Man and Experimental Animals, Seishi Shoin, Tokyo, 1982. E. Yoshihara, T. Nakae, Cytolytic activity of liposomes containing stearylamine, Biochim. Biophys. Acta 854 (1986) 93–101. D.H. Adams, G. Joyce, V.J. Richardson, B.E. Ryman, H.M. Wisniewski, Liposome toxicity in the mouse central nervous system, J. Neur. Sci. 31 (1977) 173–179. E. Mayhew, M. Ito, R. Lazo, Toxicity of non-drug-containing liposomes for cultured human cells, Exp. Cell Res. 171 (1987) 195–202.

148

H. Ishiwata et al. / Journal of Controlled Release 69 (2000) 139 – 148

[19] H. Kikuchi, N. Suzuki, K. Ebihara, H. Morita, Y. Ishii, A. Kikuchi, S. Sugaya, T. Serikawa, K. Tanaka, Gene delivery using liposome technology, J. Control. Release 62 (1999) 269–277. [20] S.J. Eastman, C. Siegel, J. Tousignant, A.E. Smith, S.H. Cheng, R.K. Scheule, Biophysical characterization of cationic lipid:DNA complexes, Biochim. Biophys. Acta 1325 (1997) 41–62. [21] K. Mislick, J.D. Baldeschwieler, Evidence for the role of proteoglycans in cation-mediated gene transfer, Proc. Natl. Acad. Sci. USA 93 (1996) 12349–12354. [22] L. Mounkes, W. Zhong, G. Cipres-Palacin, T.D. Heath, R.J. Debs, Proteoglycans mediate cationic liposome–DNA complex-based gene delivery in vitro and in vivo, J. Biol. Chem. 273 (1998) 26164–26170. [23] E.R. Lee, J. Marshall, C.S. Siegel, C. Jiang, N.S. Yew, M.R.

Nichols, J.B. Nietupski, R.J. Ziegler, M.B. Lane, K.X. Wang, N.C. Wan, R.K. Scheule, D.J. Harris, A.E. Smith, S.H. Cheng, Detailed analysis of structures and formulations of cationic lipids for efficient gene transfer to the lung, Hum. Gene Ther. 7 (1996) 1701–1717. [24] A.J. Fasbender, J. Zabner, M.J. Welsh, Optimization of cationic lipid-mediated gene transfer to airway epithelia, Am. J. Physiol. 269 (1995) L45–L51. [25] H. Aoki, T. Tottori, F. Sakurai, K. Fuji, K. Miyajima, Effects of positive charge density on the liposomal surface on disposition kinetics of liposomes in rats, Int. J. Pharm. 156 (1997) 163–174. [26] R.I. Mahato, K. Kawabata, Y. Takakura, M. Hashida, In vivo disposition characteristics of plasmid DNA complexed with cationic liposomes, J. Drug Targeting 3 (1995) 149–157.