Design of fusogenic liposomes using a poly(ethylene glycol) derivative having amino groups

Journal of Controlled Release 68 (2000) 225–235 www.elsevier.com / locate / jconrel

Design of fusogenic liposomes using a poly(ethylene glycol) derivative having amino groups Kenji Kono

a,b ,

*, Mitsugi Iwamoto a , Ryosuke Nishikawa b , Hironobu Yanagie c , Toru Takagishi a,b

a

Department of Applied Materials Science, Graduate School of Engineering, Osaka Prefecture University, 1 -1 Gakuen-cho, Sakai, Osaka 599 -8531, Japan b Department of Applied Bioscience, Research Institute for Advanced Science and Technology, Osaka Prefecture University, 1 -2 Gakuen-cho, Sakai, Osaka 599 -8570, Japan c Department of Clinical Surgery, Institute of Medical Science, University of Tokyo, 4 -6 -1 Shiroganedai, Minato-ku, Tokyo 108 -8639, Japan Received 6 March 2000; accepted 14 April 2000

Abstract As a novel fusogenic liposome, we designed liposomes modified with poly(glycidol) having b-alanine residues, which is a poly(ethylene glycol) derivative with positively charged groups. The polymer-modified liposomes of egg yolk phosphatidylcholine (EYPC) and dioleoylphosphatidylethanolamine (DOPE) were prepared by reverse phase evaporation. Fusion of the polymer-modified liposomes with anionic liposomes consisting of phosphatidic acid and DOPE was investigated. Fusion ability of the polymer-modified liposomes increased with increasing amount of the polymer fixed on the liposome. Also, inclusion of DOPE was necessary for the generation of the fusion ability of the polymer-modified liposomes. CV1 cells treated with the polymer-modified DOPE / EYPC liposomes containing calcein displayed diffuse fluorescence, suggesting that calcein was introduced into the cytoplasm. In contrast, only punctual fluorescence was observed in the cells treated with the polymer-modified EYPC liposomes containing calcein, indicating that calcein remained in the endosome and / or lysosome. In addition, COS1 cells were transfected efficiently by treatment with the polymer-modified EYPC / DOPE liposomes containing pSV2cat plasmid, whereas the transfection was not induced by treatment with the polymer-modified EYPC liposomes. Close correlation between fusion ability of the polymer-modified liposomes and their ability to deliver their contents to the cytoplasm implies that membrane fusion plays an important role in the liposome-mediated cytoplasmic delivery.  2000 Elsevier Science B.V. All rights reserved. Keywords: Fusogenic liposome; Cationic liposome; Poly(glycidol); Poly(ethylene glycol) derivative; Phosphatidylethanolamine; Cytoplasmic delivery

1. Introduction *Corresponding author. Tel.: 181-722-549-9330; fax: 181722-549-9913. E-mail address: [email protected] (K. Kono).

Because fusogenic liposomes can introduce their contents into the cytoplasm by fusing with cellular membranes such as plasma membrane and endosom-

0168-3659 / 00 / $ – see front matter  2000 Elsevier Science B.V. All rights reserved. PII: S0168-3659( 00 )00263-7

226

K. Kono et al. / Journal of Controlled Release 68 (2000) 225 – 235

al membrane, these liposomes are of great importance as delivery systems of membrane-impermeable molecules with biological activities, such as proteins, genes and oligonucleotides. Fusogenic liposomes have been designed according to various approaches. For example, they can be prepared using lipids capable of undergoing a bilayer-to-hexagonal II transition, such as dioleoylphosphatidylethanolamine (DOPE). pH-sensitive liposomes, which become fusogenic under weakly acidic conditions, have been made from DOPE in combination with titratable amphiphiles [1,2]. Also, liposomes with triggerable fusion activities have been prepared by incorporation of photo-polymerizable lipids [3] or an enzymatically cleavable peptide–lipid conjugate [4] into DOPE membranes. Another approach to the preparation of fusogenic liposome is conjugation of fusogenic molecules to liposome membranes. It has been shown that fusogenic activity can be given to stable liposomes by incorporation of viral fusion proteins [5,6], fusion peptides [7,8], and synthetic polymers [9,10] to liposome membranes. Since poly(ethylene glycol) is a well-known fusogenic polymer [11], it is expected that modification of liposomal membrane with this polymer gives fusion ability to stable liposomes. In fact, it has been reported that modification with poly(ethylene oxide)-bearing lipids could give egg yolk phosphatidylcholine (EYPC) liposomes with fusogenic activity [10], meanwhile poly(ethylene glycol) chains grafted to the liposome surface have been shown to stabilize the liposome and reduce its interaction with cells [12–15]. In earlier work, we synthesized a poly(ethylene glycol) derivative with carboxyl groups, succinylated poly(glycidol), which has a main chain structure similar to that of poly(ethylene glycol), and conjugated this polymer to EYPC liposomes [16,17]. We found that succinylated poly(glycidol)-modified liposomes were stable under neutral conditions, but exhibited fusion ability under acidic conditions where charged carboxylate groups become protonated, suggesting that partitioning of the polymer chain to the membrane rendered the liposome fusogenic [16]. In addition, the succinylated poly(glycidol)-modified liposomes could introduce their contents into cytoplasm of CV1 cells by fusing with endosome and / or lysosome which have acidic environments inside them [17].

Fig. 1. Structure of b-AlaPG having anchors.

We expected that liposomes modified with a poly(ethylene glycol) derivative having positively charged groups will exhibit the ability to fuse with negatively charged membranes, such as cellular membranes. Positive charges of the polymer chain should enhance binding of the liposomes to the negatively charged target membranes. The polymer chains of a poly(ethylene glycol)-like structure existing between the closely contacting membranes would catalyze fusion of these membranes. In this study, we designed b-alanine-attached poly(glycidol) (b-AlaPG) as a poly(ethylene glycol) derivative having positively charged groups. Palmitoyl groups were also connected to the polymer chain, as anchor moieties to liposome membranes (Fig. 1). We investigated the ability of the b-AlaPGmodified liposomes to fuse with negatively charged model membranes. Also, we examined cytoplasmic delivery of calcein and plasmid DNA mediated by the polymer-modified liposomes. Close correlation between fusion ability of the b-AlaPG-modified liposomes and efficiency of cytoplasmic delivery of calcein and plasmid DNA mediated by these liposomes has been described.

2. Materials and methods

2.1. Materials Polyepichlorohydrin (average molecular weight ca. 700,000) was purchased from Aldrich Chemical (Milwaukee, WI, USA). b-Alanine was obtained from Kishida Chemicals (Osaka, Japan). Di-tertbutyl bicarbonate was supplied from Peptide Institute (Osaka, Japan). N,N9-Carbonyldiimidazole was from Wako Pure Chemical Industries (Osaka, Japan). N,NDimethylformamide (DMF) was obtained from Tokyo Kasei Kogyo Co. (Tokyo, Japan). EYPC and DOPE were kindly donated by NOF Co. (Tokyo, Japan). Phosphatidic acid and calcein were supplied by Sigma (St. Louis, MO, USA). N-(7-Nitro-2-1,3benzoxadiazol-4-yl)-dioleoyl phosphatidylethanolamine (NBD-PE) and lissamine rhodamine B-sul-

K. Kono et al. / Journal of Controlled Release 68 (2000) 225 – 235

fonyl phosphatidylethanolamine (Rh-PE) were from Avanti Polar Lipids (Alabaster, AL, USA). Plasmid pSV2cat was obtained from Stratagene (La Jolla, CA, USA). Dulbecco’s modified Eagle’s medium (DMEM) was from Nissui Pharmaceutical (Tokyo, Japan). Fetal bovine serum (FBS) was from Hyclone Laboratories (Logan, UT, USA).

2.2. Synthesis of N-tert-butyloxycarbonyl ( Boc)-b alanine b-Alanine (6.7 g) was dissolved in dioxane–water (2:1, v / v, 230 ml). Di-tert-butyl bicarbonate (18.2 g) and 1 M NaOH (76 ml) were added to the b-alanine solution and stirred for 24 h at room temperature. The solution was concentrated by evaporation, acidified with citric acid to pH 3, and extracted with ethyl acetate. The ethyl acetate extract was washed with water, dried over anhydrous Na 2 SO 4 , and evaporated. The residue was recrystallized from ethyl acetate–hexane.

2.3. Synthesis of N-palmitoyl-b -alanine b-Alanine benzyl ester p-toluenesulfonate salt was prepared by reaction of b-alanine (8.9 g) and ptoluenesulfonic acid monohydrate (19.0 g) in benzylalcohol (50 ml) and benzene (100 ml) under reflux for 5 h and subsequent recrystallization from ethanol–diethyl ether. Palmitoyl chloride (7.8 ml) was added to a solution of b-alanine benzyl ester p-toluenesulfonate salt (7.5 g) and triethylamine (8.0 ml) in chloroform at 08C and stirred for 5 h to afford N-palmitoyl-b-alanine benzyl ester. The reaction mixture was washed with 1 N HCl, 4% aqueous NaHCO 3 , and water, dried over anhydrous Na 2 SO 4 , and evaporated. The residue was recrystallized from chloroform–hexane. The obtained N-palmitoyl-balanine benzyl ester (8.4 g) and anisole (9.0 ml) were dissolved in 30% HBr in acetic acid (90 ml) and allowed to stand for 3 h. The solution was evaporated, washed with diethyl ether, and dried in vacuum.

2.4. Synthesis of b -AlaPG Poly(glycidol) was prepared by conversion of polyepichlorohydrin according to the method of Cohen [18]. Boc-b-alanine (12.1 g), N-palmitoyl-b-

227

alanine (1.3 g), and N,N9-carbonyl diimidazole (10.9 g) were dissolved in DMF (30 ml) and stirred at 08C for 2 h. Then poly(glycidol) (5.0 g) dissolved in DMF (150 ml) was added to the solution at 08C. The mixed solution was stirred at 08C for 8 h and then at room temperature for 2 days. The polymer was purified by gel permeation chromatography on a Sephadex LH20 column (Pharmacia) eluting with DMF. The polymer in DMF solution was evaporated and dried in vacuum. The polymer was dissolved in dioxane (8 ml) and 4 N HCl in dioxane solution (35 ml) was added to the polymer solution. The mixed solution was allowed to stand at room temperature for 1 h, evaporated, washed with diethyl ether, and dried under vacuum.

2.5. Liposome preparation Liposomes were prepared by reverse phase evaporation [19]. The b-AlaPG-modified liposomes used for fusion assay were prepared as follows. A lipid (7 mg) in chloroform solution (0.7 ml) containing NBD-PE (1 mol.%) and Rh-PE (1 mol.%) was mixed with a given volume of a polymer in methanol solution (10 mg / ml). The solution was evaporated and dried under vacuum for 3 h to afford a dry thin membrane. The membrane was dissolved in diethyl ether (1.3 ml). To the solution was added an aqueous 10 mM Tris–HCl and 100 mM NaCl solution (pH 7.4, 0.4 ml) and the mixed solution was sonicated using a bath-type sonicator, forming a homogeneous emulsion. Diethylether was removed from the emulsion by evaporation. The liposome suspension was extruded through a polycarbonate membrane with a pore size of 100 nm. The liposomes were purified by gel permeation chromatography on a Sephadex G200 column using 10 mM Tris–HCl and 100 mM NaCl solution (pH 7.4). PA / DOPE liposomes were prepared via the above procedure using a dry membrane of a mixture of DOPE and PA (1:1, mol / mol). The b-AlaPG-modified liposomes containing calcein were prepared as follows. Calcein-loaded liposomes were obtained via the above procedure using a dry membrane of a mixture of EYPC and DOPE and an aqueous calcein solution (63 mM, pH 7.4). Free calcein was removed by gel permeation chromatography on a Sephadex G-50 column. Calcein-loaded liposomes (3.0 mM) were incubated in 30 mM Tris–HCl and 120 mM NaCl solution containing

228

K. Kono et al. / Journal of Controlled Release 68 (2000) 225 – 235

b-AlaPG at the polymer / lipid (w / w) ratio of 0.025 or 0.1 for 1 h. Then, the b-AlaPG-modified liposomes were purified by floating the liposomes on sucrose gradients twice. Sucrose contained in the liposome suspension was removed by dialysis against 30 mM Tris–HCl and 120 mM NaCl solution (pH 7.4) for 1 h. Plasmid DNA-encapsulating liposomes were prepared via the above procedure using lipid (12 mmol) and plasmid (80 mg) dissolved in an aqueous 10 mM Tris–HCl (pH 8.0) solution (0.2 ml). Plasmid-loaded liposomes were extruded through a polycarbonate filter with a pore size of 200 nm. Free plasmid was removed by floating the liposomes on sucrose gradients. The liposomes were incubated in the b-AlaPG solution at the polymer / lipid (w / w) ratio of 0.025 or 0.1 and then purified by floating the liposomes on sucrose gradients twice. Sucrose contained in the liposome suspension was removed by dialysis against 30 mM Tris–HCl and 120 mM NaCl solution (pH 7.4) for 1 h.

2.6. Fusion of liposomes Fusion of liposomes was detected by resonance energy transfer between NBD-PE and Rh-PE as reported by Struck et al. [20]. An aliquot of dispersion of the b-AlaPG-modified liposomes containing NBD-PE and Rh-PE was added to 2 ml of 10 mM Tris–HCl and 100 mM NaCl solution (pH 7.4) in a quartz cell at 358C (final concentration of lipid, 25 mM). Then, an aliquot of PA / DOPE (1:1, mol / mol) liposomes was added to the cell (final concentration of lipid, 25 mM) and the fluorescence intensities of NBD-PE and Rh-PE in the liposome suspension was monitored using a spectrofluorometer (Shimadzu RF5000). The excitation wavelength for NBD-PE was 450 nm and monitoring wavelengths for NBD-PE and Rh-PE were 525 and 586 nm, respectively. The R value [16,21], the ratio of fluorescence intensity at 525 nm to that at 586 nm, was calculated. This ratio was converted into the apparent concentration of these fluorophores in the membrane using a standard curve of the concentration of the fluorophores vs. R value [16,22]. The percent fusion was defined as % Fusion 5 100 3 (C t 2 C 0 ) /(C f 2 C 0 ) where C 0 and C t mean the initial and intermediary

concentrations of the fluorophores in the membrane, respectively. C f, which equals 0.5 mol.% of each fluorophore, represents the concentration when the complete fusion of the liposomes occurs.

2.7. Cell culture CV1 cell, an African green monkey kidney cell line, and COS1 cell, a derivative of CV1 transformed with a mutant of simian virus 40 (SV40) [23], were cultured in DMEM supplemented with 10% FBS in a humidifier incubator (5% CO 2 ) at 378C.

2.8. Association of liposome with cells Association of liposomes with CV1 cells were measured as previously reported [17,24]. Liposomes containing 5 mol.% NBD-PE were prepared via the above method. The cells (5310 5 ) were plated in a 25-cm 2 -flask containing 5 ml of DMEM supplemented with 10% FBS 48 h prior to the experiment. The cells were washed three times with phosphate-buffered saline (PBS) containing 0.36 mM CaCl 2 and 0.42 mM MgCl 2 (PBS-CM). Then, liposome dispersion (0.05 mM, 3 ml) was added to the cells and incubated for 2 h at 378C. The cells were washed three times with PBS-CM and incubated in 3 ml of DMEM with 10% FBS at 378C for 20 h. The cells were washed with PBS-CM three times and dislodged by treatment with PBS containing 1 mM EDTA. Fluorescence intensity of NBD-PE associated with the cell at 525 nm was measured at excitation wavelength of 450 nm. The cell number was determined from the cell protein concentration using BCA protein assay reagent (Pierce, IL, USA).

2.9. Fluorescence microscopy of cells treated with liposomes CV1 cells (5310 5 ) were plated in a 25-cm 2 -flask containing 5 ml of DMEM supplemented with 10% FBS 24 h prior to the experiment. The cells were washed three times with PBS-CM. Calcein-loaded liposome dispersion (0.05 mM, 3 ml) was added to the cells and incubated for 2 h at 378C. The cells were washed three times with PBS-CM and incubated in 3 ml of DMEM with 10% FBS at 378C for

K. Kono et al. / Journal of Controlled Release 68 (2000) 225 – 235

20 h. The cells were washed with PBS-CM three times and viewed with a microscope (Olympus IMT2) equipped with a phase contrast and epifluorescence with an excitation filter set that produces excitation in the range 470–500 nm and allows observation of fluorescence emission in the range 515–540 nm with a long wave pass dichroic mirror and barrier filter.

2.10. Plasmid delivery COS1 cells (1310 6 ) were plated in a 25-cm 2 flask containing 5 ml of DMEM supplemented with 10% FBS 24 h prior to the experiment. The cells were washed three times with serum-free DMEM. Plasmid-loaded liposomes suspended in 30 mM Tris–HCl and 120 mM NaCl (0.05 mM, pH 7.4, 2 ml) were added to the cells and incubated for 5 h at 378C. The cells were washed with serum-free DMEM and incubated in 5 ml of DMEM–10% FBS at 378C for 84 h before harvesting. Chloramphenicol acetyltransferase (CAT) activity in the cells was assayed as described elsewhere [25]. Conversion of [dichloroacetyl-1,2- 14 C]chloramphenicol (Amersham) to acetylated products was assayed by autoradiography of thin-layer chromatograms and scintillation counting of spots cut from the chromatoplates.

2.11. Other methods Phospholipid concentration was determined by the method of Bartlett [26]. Nuclear magnetic resonance spectra were taken with a JEOL JNM-GX-270 MHz instrument. Molecular weight of poly(glycidol) was estimated by gel permeation chromatography on a Shodex KD-803 column (Showa Denko) with differential refractive index detection (Jasco, RI-930) using N,N-dimethylformamide at 0.2 ml / min as an eluent. Polyethylene glycol standards were used for calibration.

3. Results and discussion

3.1. Characterization of b -AlaPG Poly(glycidol) was prepared by the conversion of

229

polyepichlorohydrin according to the method of Cohen [18]. The weight average and number average molecular weights of poly(glycidol) were estimated to be 7600 and 4700, respectively. While polyepichlorohydrin used as the starting material is shown to have an average molecular weight of ca. 700,000, the poly(glycidol) obtained had much smaller molecular weights. Probably, partial cleavage of the polymer chain took place during the conversion reaction. The b-AlaPG having anchors (Fig. 1) was synthesized by reaction of poly(glycidol) with Boc-b-alanine and N-palmitoyl-b-alanine and subsequent deprotection of Boc groups. The polymer obtained was estimated by its 1 H-NMR spectrum to have unreacted, balanine-attached and N-palmitoyl-b-alanine-attached glycidol units at the ratio of 42.5:55.0:2.5 (mol / mol / mol). On the basis of the number average molecular weight of poly(glycidol), b-AlaPG prepared in this study is calculated to have ca. 1.6 palmitoyl groups per polymer chain. The acid–base titration of the b-AlaPG showed that the polymer changes its charge density between pH 6.5 and 10.5 and its pKa was 8.7.

3.2. Fusion ability of b -AlaPG-modified liposomes The polymer-modified liposomes were prepared by reverse phase evaporation using mixtures of the polymer and the lipids and subsequent extrusion through a polycarbonate membrane with a pore size of 100 nm. To evaluate the effect of modification of b-AlaPG on the ability of the liposome to fuse, we examined fusion between the b-AlaPG-modified EYPC / DOPE (1:1, mol / mol) liposomes with varying polymer contents and PA / DOPE (1:1, mol / mol) liposomes (Fig. 2). The time course of liposome fusion is shown in Fig. 2A. The unmodified liposomes hardly fused with PA / DOPE liposomes. However, the polymer-modified liposomes exhibited the ability to fuse with PA / DOPE liposomes. In general, these liposomes fused quickly in the initial 2 min after mixing of these liposomes. But after this period, the fusion proceeded more slowly with time and finally reached a constant level. Fig. 2B represents percent fusion between the b-AlaPG-modified liposomes and PA / DOPE liposomes after 10 min incubation as a function of the polymer content. It is apparent that the fusion ability

230

K. Kono et al. / Journal of Controlled Release 68 (2000) 225 – 235

Fig. 2. Fusion of EYPC / DOPE (1:1, mol / mol) liposomes with varying amounts of b-AlaPG with PA / DOPE (1:1, mol / mol) liposomes. (A) Time-course of fusion. b-AlaPG / lipid (w / w) ratios of the liposomes were 3 / 7 (♦), 2 / 8 (m), 1 / 9 (d), and 0 / 10 (j). (B) Percent fusion after 10 min as a function of b-AlaPG content of the liposomes. Data are shown as mean6S.D. (n53).

of the polymer-modified liposomes increases with increasing polymer content. Because amino groups on the side chains of the polymer are positively charged under the neutral condition, partitioning of the polymer chain to the liposomal membrane is suppressed. Thus the liposomes were stable under that condition. However, when the anionic liposomes are present, the polymer-modified liposome associates with the anionic liposome through electrostatic interaction. This association causes charge

neutralization of both of the polymers and the anionic liposomes, and close apposition of these liposomal membranes. Because of the amphiphilic nature of the polymer’s main chain, it should exist near the interface of these membranes and generate structural defects in the contacting area of both membranes, resulting in fusion of the liposomes. Thus, the liposomes with a higher polymer content exhibited more intensive fusion with the anionic liposomes. The ability of liposome to fuse depends on its lipid composition. It is well known that inclusion of phosphatidylethanolamine having unsaturated chains in liposomal membranes enhances their fusion ability [27–29]. Thus, we examined the effect of lipid composition of the polymer-modified liposomes on their fusion ability. As shown in Fig. 3, EYPC liposomes modified with the polymer did not fuse with the anionic liposomes. However, when DOPE was included in the membrane of the polymer-modified liposomes, the liposomes exhibited fusion ability against the anionic liposomes. In addition, their fusion ability was enhanced with increasing DOPE content. DOPE has a high tendency to form a nonbilayer structure under physiological conditions. Also, its head group is hydrated to a lower degree than that of phosphatidylcholines [2,30]. Thus, it is likely that inclusion of a higher amount of DOPE reduces repulsive hydration forces arising between closely approached membranes of these liposomes and facilitates their fusion. While fusogenic activity of the polymermodified liposomes was elevated significantly by increasing DOPE content up to 50%, further increase in DOPE content was less effective for the enhancement of fusion ability of the polymer-modified liposomes. As is seen in Fig. 3A, the polymer-modified liposomes with DOPE content of 70 mol.% fused with the anionic liposomes more slowly than those with DOPE contents of 50 and 30% for the first several minutes. While phosphatidylcholine can stabilize phosphatidylethanolamine bilayers by decreasing intermolecular hydrogen bondings of phosphatidylethanolamine head groups and increasing hydration of the membrane surface, more than 20 mol.% of phosphatidylcholine is needed for the stabilization of phosphatidylethanolamine bilayers [2]. Although the membrane of the polymer-modified

K. Kono et al. / Journal of Controlled Release 68 (2000) 225 – 235

Fig. 3. Fusion of b-AlaPG-modified liposomes of varying DOPE / EYPC ratios with PA / DOPE (1:1, mol / mol) liposomes. (A) Timecourse of fusion. DOPE / EYPC (mol / mol) ratios of b-AlaPGmodified liposomes were 7 / 3 (♦), 5 / 5 (m), 3 / 7 (d), and 0 / 10 (j). (B) Percent fusion after 10 min as a function of DOPE content of b-AlaPG-modified liposomes. Data are shown as mean6S.D. (n53). b-AlaPG / lipid (w / w) ratios of the liposomes was 3 / 7.

liposomes with DOPE content of 70 mol.% was stabilized by 30 mol.% of EYPC, such high DOPE content may allow phosphatidylethanolamine-enriched domains in the membrane, resulting in a fusion property of the liposome different from that of the liposomes with lower DOPE contents.

3.3. Liposome-mediated delivery of calcein to cells Since surface modification with the b-AlaPG gave

231

DOPE-containing liposomes with an ability to fuse with the negatively charged liposomes, next we examined cytoplasmic delivery of calcein mediated by the polymer-modified liposomes. Because the calcein molecule has negative charges under neutral conditions, it may associate with the polymer via electrostatic interaction. Thus, we prepared the polymer-modified liposomes encapsulating calcein by incubation of calcein-loaded liposomes with aqueous polymer solutions and subsequent purification by floating the liposomes on sucrose gradients. Fig. 4 shows fluorescence and phase contrast micrographs of CV1 cells treated with the calceinloaded EYPC or EYPC / DOPE (1:1) liposomes modified with varying amounts of the polymer. When the cells were incubated with the bare DOPE / EYPC (1 / 1) liposomes, the fluorescence of calcein observed in the cells was very weak, suggesting that these liposomes were hardly taken up by the cells or that calcein molecules were still entrapped in the liposome, because calcein was self-quenched in the liposome [17]. In contrast, when the cells were treated with the polymer-modified DOPE / EYPC (1 / 1) liposomes prepared at the polymer / lipid (w / w) ratio of 0.025, somewhat stronger fluorescence of calcein was seen in the cells, indicating that calcein was released from the liposome. Since the cells displayed punctual fluorescence in this case, calcein molecules might exist inside of endosome and / or lysosome [31]. When the cells were treated with the DOPE / EYPC (1 / 1) liposome prepared at the polymer / lipid (w / w) ratio of 0.1, more intensive fluorescence of calcein was observed in the cells. In this case, diffuse fluorescence was seen as well as punctual. This result suggests that calcein molecules were partly transferred to cytoplasm [17,31]. When the polymer-modified EYPC liposomes were used for cell treatment, diffuse fluorescence became weaker and punctual fluorescence became stronger, compared to the cells treated with the EYPC / DOPE liposome prepared at the same lipid / polymer ratio. Because the liposomes with higher polymer and higher DOPE contents exhibited stronger fusion ability, these results imply that fusion ability of the liposomes plays a role in the transfer of calcein from endosomal and / or lysosomal compartments to cytoplasm. We expected that modification of the liposome with the b-AlaPG will increase their affinity to cell

232

K. Kono et al. / Journal of Controlled Release 68 (2000) 225 – 235

Fig. 4. Fluorescence micrographs (left) and phase contrast micrographs (right) of CV1 cells treated with calcein-loaded b-AlaPG-modified DOPE / EYPC (1:1, mol / mol) liposome prepared at polymer / lipid (w / w) ratios of 0 (A), 0.025 (B), and 0.1 (C), and b-AlaPG-modified EYPC liposomes prepared at polymer / lipid (w / w) ratio of 0.1 (D).

K. Kono et al. / Journal of Controlled Release 68 (2000) 225 – 235

Fig. 5. Association of various b-AlaPG-modified EYPC (open bars) and DOPE / EYPC (1:1, mol / mol) (hatched bars) liposomes with CV1 cells after 2 h incubation. The abscissa represents the polymer / lipid (w / w) ratio used for preparation of the liposomes. Data are shown as mean6S.D. (n53).

surface. However, as shown in Fig. 5, the amount of the liposome associated with CV1 cell decreased after the modification with the polymer. Because positive charges were introduced to the liposome surface, its affinity to the cell should be enhanced. At the same time, however, highly hydrated polymer chains attached to the liposome surface might suppress its uptake by the cell. As a result, the amount of liposome associated with the cell was decreased to some extent. In comparison between the bare EYPC liposome and the bare EYPC / DOPE liposome, apparently the latter was taken up by the cell more efficiently than the former. Similarly, enhanced uptake of phosphatidylethanolamine-containing liposomes was also reported by Chu et al. [32]. Because the head group of phosphatidylethanolamine is less hydrated than that of phosphatidylcholine, the difference in hydration of these lipids might affect the liposome uptake by the cells [32]. In contrast, the polymer-modified liposomes were taken up by the cells to similar extents, irrespective of their lipid composition. This

233

fact suggests that their surfaces were covered efficiently with the polymer chains. To confirm transfer of the liposomal contents to cytoplasm, we examined delivery of pSV2cat, which contains a standard marker gene coding for the E. coli CAT, by the b-AlaPG-modified liposomes. COS1 cells were used for the evaluation of the transfection activity of the liposomes, because they produce SV40 large tumor (T) antigen, which allows replication of pSV2cat [23]. COS1 cells were incubated with the DNA-containing liposomes of varying compositions for 5 h. Then the cells were washed and cultured for 84 h in fresh serum-containing medium before the CAT assay. Fig. 6 illustrates the CAT activity in the extract of the cells treated with the plasmid-containing liposomes of varying compositions. When the cells were treated with the bare EYPC liposomes, the cell extract hardly exhibited CAT activity. While the CAT activity of the cell extract increased slightly with increasing amount of polymer coating the EYPC liposomes, the activity was low even for the cell incubated with the liposome having the highest amount of polymer. For the cells treated with bare DOPE / EYPC (1:1, mol /

Fig. 6. Transfection of COS1 cells mediated by b-AlaPG-modified EYPC (open bars) and DOPE / EYPC (1:1, w / w) (hatched bars) liposomes. The abscissa represents the polymer / lipid (w / w) ratio used for preparation of the liposomes. Data are shown as mean6S.D. (n52 or 3).

234

K. Kono et al. / Journal of Controlled Release 68 (2000) 225 – 235

mol) liposomes, the CAT activity was very low as was the case of the EYPC liposomes. However, the cells treated with the polymer-coated DOPE / EYPC liposomes exhibited ca. 30-fold stronger activity, indicating that these liposomes achieved much more efficient transfer of their contents to the cytoplasm. As shown before, only liposomes consisting of EYPC and DOPE and modified with the polymer exhibited fusion ability against the anionic liposomes and delivered calcein to the cytoplasm of CV1 cell. The result of plasmid delivery mediated by these liposomes also demonstrates correlation between efficiency of the liposome for cytoplasmic delivery and its fusion ability. At present, the pathway through which the liposomes deliver their contents into cytoplasm is not clear. We first expected that the polymer-modified liposomes fuse with the plasma membrane. As mentioned above, the cells treated with the polymermodified EYPC / DOPE liposomes containing calcein displayed diffuse fluorescence of calcein. However, punctuate fluorescence coexisted in the same cells. This observation suggests that the liposomes were taken up by the cell through an endocytic pathway. Probably, some fraction of these liposomes fused with endosomal and / or lysosomal membranes and transferred their contents into the cytoplasm. In conclusion, we prepared liposomes modified with b-AlaPG, which is a cationic polymer with poly(ethylene glycol)-like structure, as a novel fusogenic liposome. It was found that the fusion ability of the polymer-modified liposome was affected by its polymer content as well as its lipid composition. The liposomes consisting of DOPE and EYPC (1:1, mol / mol) and modified with the polymer exhibited a strong ability to fuse with anionic PA / DOPE (1:1, mol / mol) liposomes, whereas the polymer-modified EYPC liposomes did not have fusion ability. The polymer-modified liposomes having fusion ability achieved cytoplasmic delivery of calcein and plasmid DNA, but the liposomes which do not have fusion ability could not deliver these molecules into the cytoplasm. Close correlation between fusion ability of the polymer-modified liposomes and their ability of cytoplasmic delivery implied that membrane fusion plays an important role in the liposome-mediated cytoplasmic delivery. We are currently investigating the mechanism by

which the polymer-modified liposomes deliver their content into the cytoplasm as well as comparison of efficiency of cytoplasmic delivery between these liposomes and other systems.

References [1] C.J. Chu, F.C. Szoka Jr., pH-Sensitive liposomes, J. Liposome Res. 4 (1994) 61–395. [2] D.C. Litzinger, L. Huang, Phosphatidylethanolamine liposomes: drug delivery, gene transfer and immunodiagnostic applications, Biochim. Biophys. Acta 1113 (1992) 201–227. [3] D.E. Bennet, D.F. O’Brien, Photoactivated enhancement of liposome fusion, Biochemistry 34 (1995) 3102–3113. [4] C.C. Pak, S. Ali, A.S. Janoff, P. Meer, Triggerable liposomal fusion by enzyme cleavage of a novel peptide–lipid conjugate, Biochim. Biophys. Acta 1372 (1998) 13–27. [5] A. Loyter, D.J. Volsky, Reconstituted Sendai virus envelopes as carriers for the introduction of biological material into animal cells, in: G. Poste, G.L. Nicolson (Eds.), Membrane Reconstitution, Elsevier Biomedical Press, Amsterdam, 1982, pp. 215–266. [6] M. Nakanishi, T. Uchida, H. Sugawa, M. Ishiura, Y. Okada, Efficient introduction of contents of liposomes into cells using HVJ (Sendai virus), Exp. Cell Res. 159 (1985) 399– 409. [7] A.L. Bailey, M.A. Monck, P.R. Cullis, pH-Induced destabilization of lipid bilayers by a lipopeptide derived from influenza hemagglutinin, Biochim. Biophys. Acta 1324 (1997) 232–244. ¨ J. Philippot, [8] C. Puyal, L. Maurin, G. Miruel, A. Bienvenue, Design of a short membrane-destabilizing peptide covalently bound to liposomes, Biochim. Biophys. Acta 1195 (1994) 259–266. [9] N. Oku, S. Shibamoto, F. Ito, H. Gondo, M. Nango, Low pH induced membrane fusion of lipid vesicles containing protonsensitive polymer, Biochemistry 26 (1987) 8145–8150. [10] N. Higashi, M. Yamauchi, Y. Okumura, M. Nakanishi, J. Sunamoto, Fusion between Jurkat cell and PEO-lipid modified liposome, Biochim. Biophys. Acta 1285 (1996) 183– 191. [11] Q.F. Ahkong, D. Fisher, W. Tampion, J.A. Lucy, Mechanisms of cell fusion, Nature 253 (1975) 194–195. [12] A.L. Klibanov, K. Maruyama, A.M. Beckerleg, V.P. Torchilin, L. Huang, Activity of amphipathic poly(ethylene glycol) 5000 to prolong the circulation time of liposomes depends on the liposome size and is unfavorable for immunoliposome binding to target, Biochim. Biophys. Acta 1062 (1991) 142–148. [13] T.M. Allen, C. Hansen, F. Martin, C. Redemann, A. YauYoung, Liposomes containing synthetic lipid derivatives of poly(ethylene glycol) show prolonged circulation half-lives in vivo, Biochim. Biophys. Acta 1066 (1991) 29–36. [14] J. Senior, C. Delgado, D. Fisher, C. Tilcock, G. Gregoriadis,

K. Kono et al. / Journal of Controlled Release 68 (2000) 225 – 235

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

Influence of surface hydrophilicity of liposomes on their interaction with plasma protein and clearance from the circulation: studies with poly(ethylene glycol)-coated vesicles, Biochim. Biophys. Acta 1062 (1991) 77–82. M.C. Woodle, K.K. Matthay, M.S. Newman, J.E. Hidayat, L.R. Collins, C. Redemann, F.J. Martin, D. Papahadjopoulos, Versatility in lipid compositions showing prolonged circulation with sterically stabilized liposomes, Biochim. Biophys. Acta 1105 (1992) 193–200. K. Kono, K. Zenitani, T. Takagishi, Novel pH-sensitive liposomes: liposomes bearing a poly(ethylene glycol) derivative with carboxyl groups, Biochim. Biophys. Acta 1193 (1994) 1–9. K. Kono, T. Igawa, T. Takagishi, Cytoplasmic delivery of calcein mediated by liposomes modified with a pH-sensitive poly(ethylene glycol) derivative, Biochim. Biophys. Acta 1325 (1997) 143–154. H.L. Cohen, The conversion of halogen-containing polymers to the corresponding hydroxyl derivatives, J. Polym. Sci. Polym. Chem. Ed. 13 (1975) 1993–2003. F. Szoka Jr., D. Papahadjopoulos, Procedure for preparation of liposomes with large internal aqueous space and high capture by reverse-phase evaporation, Proc. Natl. Acad. Sci. USA 75 (1978) 4194–4198. D.K. Struck, D. Hoekstra, R.E. Pagano, Use of resonance energy transfer to monitor membrane fusion, Biochemistry 20 (1981) 4093–4099. K. Kono, S. Kimura, Y. Imanishi, pH-Dependent interaction of amphiphilic polypeptide poly(Lys-Aib-Leu-Aib) with lipid bilayer membrane, Biochemistry 29 (1990) 3631–3637. K. Kono, H. Nishii, T. Takagishi, Fusion activity of amphiphilic polypeptide having acidic amino acid residues: generation of fusion activity by a-helix formation and charge neutralization, Biochim. Biophys. Acta 1164 (1993) 81–90.

235

[23] Y. Gluzman, SV40-transformed Simian cells support the replication of early SV40 mutants, Cell 23 (1981) 175–182. [24] K. Kono, R. Nakai, K. Morimoto, T. Takagishi, Temperature-dependent interaction of thermo-sensitive polymermodified liposomes with CV1 cells, FEBS Lett. 456 (1999) 306–310. [25] C.M. Gorman, L. Moffat, B.H. Howard, Recombinant genomes which express chloramphenicol acetyltransferase in mammalian cells, Mol. Cell. Biol. 2 (1982) 1044–1051. [26] G.R. Bartlett, Phosphorus assay in column chromatography, J. Biol. Chem. 234 (1959) 466–468. ¨ ¨ [27] R. Sundler, N. Duzgunes, D. Papahadjopoulos, Control of membrane fusion by phospholipid head groups II: the role of phosphatidylethanolamine in mixtures with phosphatidate and phosphatidylinositol, Biochim. Biophys. Acta 649 (1981) 751–758. [28] M.J. Hope, P.R. Cullis, The role of nonbilayer lipid structures in the fusion of human erythrocytes induced by lipid fusogens, Biochim. Biophys. Acta 640 (1981) 82–90. [29] A.J. Verkleij, Role of nonbilayer lipids in membrane fusion, in: J. Wilschut, D. Hoekstra (Eds.), Membrane Fusion, Marcel Dekker, New York, 1991, pp. 155–166. [30] J. Wilschut, Membrane fusion in lipid vesicle systems, in: J. Wilschut, D. Hoekstra (Eds.), Membrane Fusion, Marcel Dekker, New York, 1991, pp. 89–126. [31] J. Connor, L. Huang, Efficient cytoplasmic delivery of a fluorescent dye by pH-sensitive immunoliposomes, J. Cell Biol. 101 (1985) 582–589. [32] C.J. Chu, J. Dijkstra, M.-Z. Lai, K. Hong, F.C. Szoka, Efficiency of cytoplasmic delivery by pH-sensitive liposomes to cells in culture, Pharm. Res. 7 (1990) 824–834.