PS exposure increases the susceptibility of cells to fusion with DOTAP liposomes

Katarzyna Stebelska a,1, Paulina Wyrozumska a, Aleksander F. Sikorski a,b,∗
a Laboratory of Cytobiochemistry, Institute of Biochemistry and Molecular Biology, University of Wroclaw,
Przybyszewskiego 63/77, 51-148 Wroclaw, Poland
b Academic Centre for Biotechnology of Lipid Aggregates, Przybyszewskiego 63/77, 51-148 Wroclaw, Poland
Received 10 October 2005; received in revised form 8 January 2006; accepted 11 January 2006
Available online 17 February 2006

Abstract

Cationic liposomes are used as efficient carriers for gene delivery into mammalian cells due to their ability to bind nucleic acids, adsorb onto the cell surface and fuse with negatively charged membranes. This last property enables the release and escape of their cargo from endosomal compartments. The efficiency of this fusion mainly depends on the surface charge of the target membranes. Here, we report that cells of two different lines, epithelial adenocarcinoma HeLa and lymphocytic leukemia Jurkat T, which externalize PS, are more susceptible to fusion with DOTAP liposomes than control cells. We compared the ability to undergo fusion of untreated and apoptotic cells. Apoptosis was induced by various pro-apoptotic agents and treatments, namely: incubation in the presence of MnCl2, cytostatic drugs fludarabine and mitoxantrone, staurosporine and serum depletion in the case of HeLa cells. Jurkat T cells were treated similarly except apoptosis was additionally induced by incubation in the presence of 4% EtOH. Epithelial cells fused with the highest efficiencies of lipid mixing, when pretreated with staurosporine. Jurkat T cells were less susceptible to fusion, but they also displayed an increase in fusion efficiency after the induction of apoptosis. Alternatively, we treated the cells with metabolic inhibitors causing ATP-depletion in order to inactivate aminophospholipid translocase. After ATP-depletion, HeLa and Jurkat T cells fused with DOTAP liposomes with higher efficiencies than control cells. Our conclusion is that the lipid asymmetry of natural membranes may limit fusion with cationic liposomes.

Keywords: Cationic liposomes; Fusion; Phosphatidylserine; Aminophospholipid translocase

1. Introduction

Fusion between natural and cationic liposomes/ lipoplexes is thought to be a key event in the mechanism of transfection mediated by cationic liposomes. Fusion may occur at the stage of interactions with the plasma membrane [1,2]. However, adsorptive endocy- tosis is thought to be the main way for lipoplexes to enter the cell, and it is suggested that fusion with endosomal membranes leads to lipoplex cargo release, which enables lysosomal degradation to be avoided [3–11]. It is known that electrostatic interactions of oppositely charged membranes not only enable initial contact and adhesion, but also lead to the generation of negatively curved lipid structures, which promotes fusion [12–14]. Ability of cationic liposomes to fuse with cellular membranes greatly enhances their trans- fection activity. On the other hand, efficient fusion may lead to serious perturbance of natural membranes with possible effects on their biological functioning (enrich- ment in cationic lipid molecules after fusion, rupture, loss of bilayer integrity associated with an increase of its permeability and possibility of cellular responses to changes of intracellular concentrations of signal ions, for example, Ca2+). Therefore, fusion between cationic liposomes, commonly used as systems of DNA deliv- ery, and biological or artificial membranes deserves special attention and has been intensively studied [15–20].

The negative surface charge of the cell originates from the anionic sugar residues of glycoproteins and glycolipids. Lipids that could be involved in fusion with cationic liposomes, PS and PE, normally localize within the inner monolayer of the plasma membrane, which is controlled by ATP-dependent aminophospho- lipid translocase [21]. Thus, cellular membranes are normally asymmetric with an expected lower suscep- tibility to fusion. The establishment of fusion capa- bility of biological membranes of maintained or dis- rupted lipid asymmetry is important because cationic lipid molecules incorporated within natural membrane as a result of fusion may be involved in interactions with negatively charged macromolecules (e.g. DNA) prompting their cytosolic entry—this particular aspect of cationic lipids/surfactants interactions with biologi- cal membranes is rather rarely taking into account [22], although fusion between cationic liposomes and neg- atively charged biological membranes is proposed to enhance DNA release from lipoplexes [4,7,8,14]. As was previously found a change of the surface charge may be an important factor modulating the fusion sus- ceptibility [16–18]. There is only one report dealing with binding of cationic liposomes to apoptotic cells [23].

The aim of the present study was to test whether the PS exposure in living cells induces detectable changes in their susceptibility to fuse with DOTAP liposomes. Several ways to promote PS externaliza- tion such as induction of apoptosis [24,25] and ATP- depletion was applied. Another way to experimentally disturb membrane asymmetry was to increase mobil- ity of the lipid components of the membrane by incubation of cells with liposomes at 45 ◦C. Our results indicate that cells in which asymmetry was disturbed fused with DOTAP liposomes with considerably higher efficiencies.

2. Materials and methods

2.1. Reagents and cells

1,2-Dioleoyl-3-trimethylammoniumpropane (DOT- AP) was purchased from Northern Lipids (Vancouver, British Columbia, Canada). N-(7-nitrobenz-2-oxa-1, 3-diazol-4-yl)-1,2-dihexadecanoyl-sn-3-phosphoetano- lamine (NBD-PE) and N-(lissamineTM rhodamine B sulfonyl)-1,2-dihexadecanoyl-sn-glycero-3-phosphoet- hanolamine (Rh-PE) were purchased from Molecular Probes (Eugene, OR, USA), sodium dodecyl sulphate (SDS) was purchased from Roth (Karlsruhe, Germany). The kit for apoptosis detection, containing FITC- annexin V, was obtained from Oncogene Research Products (Cambridge, MA, USA).Human epithelial adenocarcinoma HeLa cells and human lymphocytic leukemia Jurkat T cells were obtained from the Institute of Immunology and Experi- mental Therapy, Wroclaw, Poland. MEMα, RPMI 1640 and trypsin/EDTA were also purchased from the same source.

2.2. Cell culture

Epithelial cells of the HeLa line were grown at 37 ◦C, under 5% CO2 to 80% confluence in MEMα medium with 10% heat-inactivated fetal bovine
serum (FBS, Gibco-Invitrogen, Paisley, UK) containing 2 mM glutamine and penicillin 100 U/ml/streptomycin 100 µg/ml/neomycin 100 µg/ml, passaged every three days with 0.25% trypsin/EDTA, and seeded at a 1:6 dilu- tion.Lymphocytic leukemia Jurkat T cells were main- tained at 37 ◦C, under 5% CO2 in RPMI 1640 medium supplemented with 10% heat-inactivated fetal bovine serum, 2 mM glutamine, and 100 U/ml penicillin, 100 µg/ml streptomycin and 100 µg/ml neomycin and passaged every three days by 1:10 dilution.

2.3. Preparation of cationic liposomes

Chloroform solutions of lipids were dried under a stream of nitrogen followed by the removal of the solvent under a high vacuum for 2 h. The lipids were hydrated in a buffer containing 20 mM Hepes and 150 mM NaCl, pH 7.4. The obtained suspension of MLVs was sonicated for 30 s and subsequently extruded 10 times through a 100 nm pore-size polycarbonate filter. The DOTAP lipo- somes contained the fluorescent probes Rh-PE and NBD- PE each at a 1 mol%/mol concentration. The size dis- tributions of the liposomes were determined by photon correlation spectroscopy (PCS) using a Zeta-Sizer 5000, Malvern Instruments (Grovewood Lane, UK).

2.4. Lipid mixing assay

Jurkat T cells or trypsinized HeLa cells were washed and resuspended in a cold test buffer composed of 20 mM Hepes, 140 mM NaCl, 3.5 mM KCl, 1 mM MgCl2,1 mM CaCl2 and 100 µg/ml glucose, pH 7.4, and counted using a Bu¨rcker’s chamber. Fusion between membranes of opposite charge was monitored as the lipid mixing degree by measuring the decrease in energy transfer after the dilution of the two fluorescent probes NBD-PE and Rh- PE in unlabeled membranes [26]. Liposomes containing 1 mol%/mol of both probes were added at a concentration of 15 µM (lipid) to a cell suspension (1 × 106 of trypsinized HeLa cells or 2 × 106 of Jurkat T cells) in 1 ml of test buffer. The samples were incubated for 1 h at a temperature of 37 or 45 ◦C. After that, they were kept on ice until measurements of fluorescence were performed.

Fluorescence was measured as the emission of NBD-PE at 536 nm using an excitation wavelength of 463 nm in a Kontron SFM 25 spectrofluorimeter. The degree of lipid mixing was calculated as:apoptosis-inducing agent, and incubated at a temperature of 37 ◦C. We used the following treatments:
- incubation in the presence of 2 mM MnCl2 for 8 h [28];
- incubation in the presence of fludarabine and mitox- antrone, each at a concentration of 1 µM, for 8 h;
- incubation in a medium without serum for 12 or 24 h [29];
- incubation in the presence of staurosporine at a con- centration of 1 µM for 4 h [30].

After incubation, the cells were rinsed with PBS and trypsinized with 0.25% trypsin/EDTA.
Centrifuged Jurkat T cells were resuspended at a den- sity of 1 × 106 cells/ml in 20 ml of medium containing an apoptotic agent, and incubated at 37 ◦C. The following procedures were used:

- incubation in the presence of 4% ethanol for 2 h [23];
- incubation in the presence of fludarabine and mitox- antrone, each at a concentration of 1 µM for 8 h;
- incubation in the presence of serum-depleted medium for 24 h;
- incubation in the presence of staurosporine at a concentration of 1 µM for 4 h.

2.7. ATP-depletion

where F0 is the initial fluorescence and Fmax is the max- imal fluorescence measured after the addition of 20 µl of 10% SDS to a sample.

2.5. K+ depletion

Cells depleted in potassium ions treated as described in [27] lose the ability to form coated pits. Trypsinized HeLa cells were washed with a buffer composed of 20 mM Hepes, 140 mM NaCl, 1 mM MgCl2, 1 mM CaCl2 and 100 µg/ml glucose, pH 7.4. Centrifuged cells were resuspended in the above-mentioned buffer, which had been diluted 1/1 with sterile H2O, and were subsequently incubated for 5 min at a temperature of 37 ◦C. The cells were centrifuged, washed with an isotonic buffer without K+, and finally resuspended in the buffer and incubated for 30 min at a temperature of 37 ◦C. The control cells were treated the same way but using a buffer containing 10 mM KCl.

2.6. Induction of apoptosis

HeLa cells were grown to confluence in 75 cm2 culture flasks. Cell monolayers were rinsed with PBS and covered with 20 ml of fresh medium containing an
HeLa cells seeded in a confluent monolayer in 75 cm2 culture flasks were rinsed twice with PBS and incubated for 0.5, 1, or 6 h in a buffer composed of 20 mM Hepes, 140 mM NaCl, 3.5 mM KCl, 1 mM MgCl2, 1 mM CaCl2,100 µg/ml glucose, 25 mM deoxyglucose and 5 mM sodium azide, pH 7.4, at temperature of 37 ◦C. Then they were treated with 0.25% trypsin/EDTA, washed, and resuspended in the test buffer. Centrifuged Jurkat T cells,after the removal of the culture medium, were washed and resuspended in a buffer containing antimetabolites and incubated at 37 ◦C for 1 or 6 h. Fusion experiments were performed as described above, except that the cells treated with antimetabolites for 1 h were incubated with DOTAP liposomes in the presence of antimetabolites.

2.8. Proliferation test

One day before the experiment, HeLa cells were seeded in 24-well plates at a density of 1 × 105 cells per well. The cells were rinsed twice with PBS and cov- ered with 1 ml of a buffer composed of 20 mM Hepes, 140 mM NaCl, 3.5 mM KCl, 1 mM MgCl2, 1 mM CaCl2,100 µg/ml glucose, 25 mM deoxyglucose and 5 mM sodium azide, pH 7.4, and incubated at 37 ◦C for 0.5, 2 and 6 h. The control cells were incubated with a simi- lar buffer without antimetabolites. After incubation, the cells were washed twice with PBS, covered with 1 ml of culture medium and left for 24 h at a temperature of 37 ◦C. The next day, the cells were trypsinized, carefully removed from the wells, resuspended in 0.2 ml of PBS, and counted using a Bu¨rcker’s chamber.

2.9. PS exposure and fusion detection using fluorescence microscopy

Microscopic observations were performed to detect PS exposure. One day before the experiment, HeLa cells were seeded on cover slips inserted in wells of a 24-well plate at 7 × 105 cell per well. The cell monolayers were subjected to ATP-depletion or apoptosis induction. The FITC-annexin V-based assay was used to detect PS on the cell surface of the apoptotic cells. The cells were washed twice with PBS and covered with a “binding buffer” containing 2.5 mM Ca2+. FITC-annexin and pro- pidium iodide were added to a concentration of 2.5 and 0.5 µg/ml, respectively. After 20 min incubation, micro-
scopic observations were performed using an Olympus B211 fluorescent microscope.

Microscopic observations were also performed to detect fusion between ATP-depleted HeLa cells and DOTAP liposomes labeled with the fluorescent probes NBD-PE and Rh-PE. Cells were preincubated for 1h in the presence of antimetabolites: 25 mM deoxyglucose and 5 mM NaN3 at 37 ◦C. Liposomes were added to a concentration of 25 µM (lipid), and incubation was prolonged for an additional 60 min in the presence of the antimetabolites. After that, the cells were washed twice with PBS and observed under blue light to excite the flu- orescence of NBD-PE in the case of fused liposomes, or both probes in the case of unfused liposomes.All experiments were performed in several series as indicated in figure legends. Statistical significancy was assessed by using Student’s t-test.

3. Results

3.1. Apoptotic HeLa cells in fusion with DOTAP liposomes

In our study, we used trypsinized cells. Trypsinization is reported to affect the cytoskeleton [31] and glyco- protein receptors; therefore, after such treatment, the ability of cells to undergo endocytosis is expected to be reduced. The aim of the experiment described below was to evaluate if fusion observed upon interactions of cationic liposomes and trypsinized HeLa cells occurs mainly within plasma membrane. As can be seen in Fig. 1, the kind of experimental treatment performed in our study (see Section 2) probably eliminates fusion with endosomal membranes, as cells after trypsinization fused with almost the same efficiencies as cells addition- ally subjected to K+ depletion, which was reported to eliminate endocytosis [27]. Control cells which partially recovered ability for endocytosis (incubated for 30 min at a temperature of 37 ◦C after trypsinization) fused with significantly higher efficiencies than trypsinized cells kept before the fusion experiment at 4 ◦C or K+-depleted cells incubated for 30 min at a temperature of 37 ◦C. The cells treated similarly as K+-depleted cells but using hypotonic buffer containing 5 mM KCl and subsequently incubated for 30 min at a temperature of 37 ◦C fused with DOTAP liposomes with slightly higher efficiencies then control cells. Therefore, trypsinization may be assumed as a treatment which allows to observe fusion on the level of interactions mainly within the plasma membrane.

Fig. 1. Fusion between DOTAP liposomes and trypsinized HeLa cells or trypsynized HeLa cells subjected to K+ depletion in comparison to the efficiencies of fusion observed for control cells with a resumed ability for endocytosis. Untreated cells ( ), cells subjected to K+ deple- tion ( ), cells incubated before fusion experiment at a temperature of 37 ◦C for 30 min, with a partially recovered ability for endocytosis ( ), cells subjected to a similar procedure as cells depleted in potas- sium ions but using a buffer containing K+ ( ) (for details see also Section 2). The values represent the mean ± S.D. (n = 10, three inde- pendent experiments were performed—two of them in triplicates, and one in quadruplicates), *P < 0.05, **P < 0.001 compared to untreated cells ( ).

We compared the susceptibility of the control and apoptotic cells to fuse with DOTAP liposomes (Fig. 2). The induction of apoptosis was carried out in various ways. We chose a suitable treatment based on data in the literature, and did not study the progress of apop- tosis in detail. As our purpose was to expose PS on the cell surface, we only qualitatively tested PS pres- ence by FITC-annexin V staining (results not shown). In the fusion experiments, we additionally tested the effect of temperature on fusion efficiency comparing results observed after incubation at 37 and 45 ◦C. We supposed that incubation at 45 ◦C should inactivate the enzymes responsible for maintaining lipid asymmetry and elevate transmembrane lipid mobility—both of these factors should facilitate fusion. In any case, except for the cells treated with a mixture of mitoxantrone and flu- darabine, the levels of fusion with DOTAP liposomes were significantly higher at 45 ◦C than those measured after incubation at 37 ◦C, indicating a possible role of increased lipid mobility in prompting fusion (see Fig. 2).

Fig. 2. Apoptotic HeLa cells are more susceptible to fusion with DOTAP liposomes than control cells. To induce apoptosis, cells were treated with combined cytostatic drugs: fludarabine and mitoxantrone ( ), MnCl2 (u), serum depletion for 12 h ( ) or 24 h ( ), and stau- rosporine ( ) (for details of each procedure see Section 2). The fusion experiments were performed as described in Section 2. The values shown represent the mean ± S.D. (n = 10, three independent experi- ments were performed—two of them in triplicates, and one in quadru- plicates), *P < 0.05 or **P < 0.001 compared to the control cells ( ).

It should be noted that the effect of incubation at elevated temperature is smaller in the case of cells of disturbed asymmetry due to the various treatment than in the con- trol cells. Moreover, fusion efficiencies observed for normal and apoptotic cells after incubation with the liposomes at the temperature of 45 ◦C were similar, what suggest that the treatment induced distortion of lipid membrane asymmetry.

Cells treated with the combined cytostatic drugs mitoxantrone and fludarabine fused with DOTAP lipo- somes with similar efficiencies to those of the control cells after incubation at 37 ◦C. As mentioned above, after incubation at elevated temperature of 45 ◦C, the level of fusion was not higher. A possible explanation could be the involvement of positively charged mitoxantrone in the neutralization of cell surface charge and/or neutral- ization/immobilization of PS lipid molecules.

For the induction of apoptosis, we incubated HeLa cells with 2 mM MnCl2 for 8 h. Cells treated this way were only moderately more susceptible to fusion with DOTAP liposomes after incubation at 37 ◦C than the control cells (see Fig. 2). Cells incubated without serum for 12 and 24 h, which is also known to induce apopto- sis [29], underwent fusion with DOTAP liposomes with significantly higher efficiencies than the control cells (Fig. 2).The highest levels of fusion were observed after treat- ment with a very potent apoptotic agent, staurosporine [32]. After 4 h incubation with 1 µM staurosporine, adherent HeLa cells in the monolayer underwent char- acteristic morphological changes: shrinking and detach- ment from each other. They also exposed PS as detected microscopically (about 80% of the cells were stained with FITC-Annexin V). After trypsinization, they fused with DOTAP liposomes with efficiencies of approxi- mately 40% which was about 2.5 times higher than those of the control cells.

3.2. ATP-depleted HeLa cells in fusion with DOTAP liposomes

To decrease the intercellular level of ATP, we used treatment with the antimetabolites deoxyglucose and sodium azide. We tested the susceptibility of cells treated with 25 mM deoxyglucose and 5 mM sodium azide for 0.5, 2 and 6 h to fusion with DOTAP lipo- somes (Fig. 3). Cells which were incubated with these metabolic inhibitors for 0.5 h and after trypsinization, were washed and resuspended in the same test buffer as the control cells, were found to fuse with DOTAP liposomes similarly to the control cells. They also pro- liferated in the same way as the control cells, indicating the low toxicity level of such a treatment and no effect on its ability to adhere to the surface of the culture dish (Fig. 4). Cells in a monolayer treated for 1 and 2 h with the antimetabolites began to undergo morphological changes and to detach from each other. After trypsiniza- tion, they did not round up. The proliferation rate of cells treated for 2 h with antimetabolites was twice as low as that of the control cells (Fig. 4). As could be expected, after a prolonged, 6 h treatment with deoxyglu- cose and sodium azide, the cells easily detached from the surface of the culture dish. The proliferation test showed that about 5% of the cells survived the incu- bation (Fig. 4). The percentage of cells exposing PS, as observed microscopically with FITC-annexin V, was: 23.0 ± 0.11 after 2 h treatment and 36.5 ± 4.5 after 6 h treatment (see also Fig. 5(A)). Cells incubated for 2 and 6 h with antimetabolites fused with DOTAP liposomes with very high efficiencies (60–65%)—about four times higher than those of the control cells or cells treated with deoxyglucose and sodium azide for 0.5 h (Fig. 3). For the cells treated with antimetabolites for 2 and 6 h there was no statistically significant effect of elevated temper- ature of 45 ◦C. Similarly high efficiencies of fusion were observed for the cells which were incubated in hypotonic buffer and were not allowed to reseal after the treatment (results not shown).

Fig. 3. HeLa cells depleted in ATP fuse efficiently with DOTAP lipo- somes. Cells were incubated in the presence of 25 mM deoxyglucose and 5 mM sodium azide for 0.5 h ( ), 2 h ( ) and 6 h (u). The fusion experiments were performed as described in Section 2. The values represent the mean ± S.D. (n = 10, two independent experiments were performed, each in five multiplicates), *P < 0.05, **P < 0.001 compared to the control cells ( ).

Fig. 4. A proliferation test of HeLa cells incubated for 0.5 h ( ), 2 h ( ) and 6 h (u) in the presence of antimetabolites: 25 mM deoxyglu- cose and 5 mM sodium azide. The results are presented as a % of the control culture results obtained for cells which were not treated with antimetabolites (for details of the experimental procedure, see Section 2). The bars represent ± S.D. (n = 3, an experiment was per- formed in triplicate).

From microscopic observations of HeLa cells adhered on the surface of the culture dish, treated for 2 h with 25 mM deoxyglucose and 5 mM NaN3 and incubated for 1 h with DOTAP liposomes, it could be concluded that fusion is very efficient (Fig. 5(B)). There was a predom- inance of green fluorescence emitted by NBD-PE, with only a few unfused yellow–orange liposomes observ- able.

3.3. Apoptotic and ATP-depleted Jurkat T cells in fusion with DOTAP liposomes

The properties, morphology and endocytotic activ- ity of Jurkat T non-adherent lymphocytic leukemia cells are different from these of endothelial HeLa cells, which depends on the way of organization of the cellular mem- brane/cytoskeleton system. As the properties of cellu- lar membranes and their susceptibility to fusion are thought to be controlled by an interplay in the state of organization of the plasma membrane/cellular membranes
/cytoskeleton, we decided to test the ability of these cells to fuse upon interactions with DOTAP liposomes. Untreated Jurkat T cells fused with DOTAP lipo- somes rather poorly at 37 ◦C (Fig. 6), although elevation of the temperature gave a two-fold increase in the effi- ciency of fusion. Cells in which apoptosis was induced by treatment with 4% EtOH, staurosporine, a mixture of mitoxantrone and fludarabine, or serum depletion were clearly more susceptible to fusion with DOTAP lipo- somes, but the levels of lipid mixing were still quite low when compared to the results for HeLa cells (see Figs. 2 and 6). It should be mentioned that about 40% of Jurkat T cells grown for 8 h in the presence of the cytostatic mixture exposed PS [Dubielecka, personal communication]. Moreover, cells treated for 2 h with the antimetabolites deoxyglucose and sodium azide fused with DOTAP liposomes with the same efficiencies as the control cells. A detectable increase in the level of fusion level was found for cells treated with antimetabo- lites for 6 h but only to the level of 15–20% (Fig. 7). The efficiencies of fusion between control cells and DOTAP liposomes after the incubation at elevated temperature of 45 ◦C were similar to those observed for apoptotic cells.

Fig. 5. HeLa cells depleted in ATP expose PS and undergo efficient fusion with DOTAP liposomes. (A) The presence of PS on the surface of the plasma membrane of ATP-depleted HeLa cells demonstrated as binding FITC-annexin V. Cells were incubated for 2 h in the presence of 25 mM deoxyglucose and 5 mM sodium azide, and then treated with FITC-annexin V and propidium iodide (for details see Section 2). Magnification 300×.(B) HeLa cells depleted in ATP in fusion with DOTAP liposomes. HeLa cells were preincubated for 1 h in the presence of 25 mM deoxyglucose and 5 mM sodium azide at 37 ◦C, and then incubated with DOTAP liposomes labeled with NBD-PE and Rh-PE for the next hour in the presence of the antimetabolites. Cells were observed under blue light. Orange spots represent unfused liposomes. The green flourescense of cellular membranes originating from NBD-PE indicates fusion with DOTAP liposomes and a decrease in the level of energy transfer between fluorescent probes upon dilution. Magnification 900×.

Fig. 6. Control and apoptotic Jurkat T cells in fusion with DOTAP liposomes. Cells were treated with the following procedures to induce apoptosis: 2 h incubation in the presence of 4% EtOH ( ), 24 h serum depletion (u), 8 h incubation with the combined cytostatic drugs flu- darabine and mitoxantrone ( ), 4 h incubation with staurosporine ( ). Fusion experiments were performed as described in Section 2 at tem- peratures of 37 and 45 ◦C. The values represent the mean ± S.D. (n = 10, two independent experiments were performed, each in five multiplicates), *P < 0.05, **P < 0.001 compared to the control cells ( ).

Fig. 7. ATP-depleted Jurkat T cells in fusion with DOTAP liposomes. Cells were treated with antimetabolites (25 mM deoxyglucose and 5 mM sodium azide) for 2 h ( ) and 6 h ( ). The fusion experiments were performed as described in Section 2 at temperatures of 37 and 45 ◦C. The values represent the mean ± S.D. (n = 10, two independent experiments were performed, each in five multiplicates), *P < 0.05,**P < 0.001 compared to the control cells ( ).

4. Discussion

Fusion between lipid membranes is mediated by non- lamellar lipid structures of negative curvature. There- fore, the occurrence of fusion depends mainly on the lipid composition of interacting membranes. Cationic liposomes fuse with target membranes because of the presence of negatively charged lipid species and/or neu- tral helper lipid molecules, which due to their tendency to organize into nonlamellar structures of negative cur- vature, promote fusion [13,14]. Such lipids, namely PS and PE, are components of natural membranes, for exam- ple, the plasma membrane, but they are normally located within the inner monolayer of the membrane, and there- fore they do not contact with positively charged lipo- somes after initial adsorption.
In this study, we show that a disturbance of the normal asymmetry of the plasma membrane resulting in enrich- ment of outer monolayer with aminophospholipids may prompt fusion with cationic liposomes. Liposomes com- posed of cationic DOTAP without the addition of helper lipids, for example, DOPE, were used as we wanted to analyze a single effect of loss of plasma membrane lipid asymmetry. We measured the susceptibility to fusion with DOTAP liposomes of cells of two different cell lines: adherent epithelial HeLa and lymphocytic Jurkat T.

Epithelial HeLa cells and lymphocytic Jurkat T cells were clearly different in their abilities to fuse, which may be partially the effect of the experimental proce- dure we used—although the total surface area of the number of cells used in the fusion test with DOTAP liposomes (a 15 µM concentration of lipid) (which was 1 × 106 trypsinized HeLa cells and 2 × 106 Jurkat T cells) was comparable, we did not take into account the exact total surface area of cellular membranes which could fuse with liposomes. Nevertheless, for both of the cell lines tested, the expected elevation in fusion efficien- cies after the induction of apoptosis or ATP-depletion was observed. The increase in the ability to fuse with DOTAP liposomes obtained after PS exposure in apop- totic cells supports the assumption that what we observe is mainly fusion with the plasma membrane.

We found that various treatment leading to apoptosis induction and PS exposure promote fusion, which con- curs with literature data on artificial phospholipids vesi- cles or erythrocyte membranes [16–18,20,33]. Apoptotic Jurkat T cells were already reported to bind cationic lipo- somes more intensively than control cells [23]. “Nonspecific” elevation of lipid mobility by incubation of cells at 45 ◦C increased fusion capability in control cells. This increase in the cells in which the PS exposure had been first induced by various treatment was smaller than for untreated cells. It is known that the elevated temperature could be by itself apoptosis inducer [34,35].

Interestingly, ATP-depleted HeLa cells appeared to be very susceptible to fusion with DOTAP liposomes: they fused with higher efficiencies than apoptotic cells, even those which had been treated with the most potent apop- totic agent used in this study, staurosporine, despite the lower extent of PS exposure. High efficiencies of fusion between ATP-depleted HeLa cells and DOTAP lipo- somes would suggest that what we observe is a dilution of fluorescent probes within large amounts of membranes, possibly all cellular membranes. Microscopic observa- tions support this hypothesis: the fluorescent probe used in our study to detect fusion is dispersed within all cel- lular membranes, including the plasma membrane and nuclear membrane, implying that lipid mixing occurred due to an increased susceptibility of all the membranes to undergo fusion. The reason of the observed differ- ence in susceptibility to fusion between the staurosporine treated and ATP-depleted epithelial HeLa cells, but not lymphocytic Jurkat T cells, which could be interpreted as a lack of correlation between PS exposure and fusion susceptibility, is not known—we can only speculate that some other changes in cellular membranes organiza- tion/composition developed upon ATP-depletion, apart from PS exposure, contribute to this effect. It should be noted that in the case of HeLa cells a progress of PS expo- sure upon prolonged incubation with antimetabolites observed as an increase of a percent of PS-externalizing cells did not cause statistically significant elevation of cellular susceptibility to fusion with DOTAP liposomes. Moreover, the proliferation test revealed that prolonged exposure to the metabolic inhibitors used in our study alters the survival rate of HeLa cells, indicating induction of necrosis. ATP-depletion, considered as pro-necrotic treatment producing changes in biological membranes, not only causes PS exposure due to the inactivation of aminophospholipid translocase [36], but also activates phospholipases A2, which leads to the enrichment of cellular membranes in lysophospholipids [37,38]. More- over cytosolic acidification, which is known to occur upon ATP-depletion [39], and its effect on cellular mem- branes may prompt fusion. Local discontinuities in the lipid bilayers, in this case possibly induced by the pres- ence of lysophospholipids, were suggested to increase capability to fusion [40,41]. It should be mentioned here that although lysophospholipids were reported to prevent formation of negatively curved structures of fusion intermediates (stalks) [42,43], some other studies demonstrate their activity as fusion promoters [44–47]. This is mainly due to their ability to induce membrane defects and their involvement in formation of positively curved fusion pores while present in the inner mono- layer [42,48]. Therefore, we propose that the observed high efficiencies of fusion upon ATP-depletion, particu- larly for epithelial cells, may be a result of both plasma membrane lipid asymmetry disturbance and membrane enrichment in lysophospholipids, in this case considered as fusion promoters.
Additionally, we observed that staurosporine treated HeLa cells quickly underwent typical for apoptosis, clearly visible shrinkage. In contrast, ATP-depleted HeLa cells did not undergo this kind of changes. More- over, HeLa cells exposed to metabolic inhibitors were less susceptible to trypsin-induced shape alternations. ATP-depleted HeLa cells displayed a lower ability to adhere to a culture dish. All these observations indicate changes in cytoskeleton organization, probably increase in the level of actin polymerization typical to ATP- depletion [49,50]. We cannot exclude that a decrease in surface area of plasma membrane associated with an increase in density of cytoplasm [51] may affect capabil- ity of staurosporine treated apoptotic cells to fusion. This may additionally explain lower susceptibility of these cells to fusion with DOTAP liposomes, when compared to ATP-depleted cells.
In our study we showed evidence that it is possible to induce such changes in natural membranes (PS exposure, temperature controlled increase of lipid mobility), which enhance their ability to fuse with cationic liposomes. In particular, while the effect of lipid asymmetry distur- bance or target membrane lipid composition on suscepti- bility to fusion was reported by others [16–18,34,52–55] the observation of significant increase in ability of cul- tured cells to fuse with cationic liposomes after induction of apoptosis or ATP-depletion was not reported pre- viously in the literature. It should be also mentioned that our assumption on possible role of PS exposure in prompting fusion, although was easily predictable, appeared to be more complex when complicated biolog- ical system was tested experimentally. It is likely that cellular susceptibility to fusion with cationic liposomes may also depend on other then simply a loss of plasma membrane lipid asymmetry cellular responses to applied treatments and the manner of cytoskeleton
/cellular mem- branes organization, variable for different cell lines and their energy status. However, our main conclusion is that the lipid asymmetry of natural membranes limits fusion with cationic liposomes composed of DOTAP. Although most of the treatments used in our study to expose PS irreversibly lead to cell death we suppose that alternative methods of lipid scrambling would find at least limited application in prompting of fusion by means of enrichment of the outer surface plasma membrane in negatively charge compound, for example, treatment with aminophospholipid translocase inhibitors.

Acknowledgment

The study was supported by grant No. 3P04B01325 from the State Committee for Scientific Research (KBN), Poland.

References

[1] M. Pedroso de Lima, S. Simo˜es, P. Pires, R. Gaspar, V. Slepushkin,
N. Du¨zgu¨nes¸, Gene delivery mediated by cationic liposomes: from biophysical aspects to enhancement of transfection, Mol. Membr. Biol. 16 (1999) 103–109.
[2] P. Pires, S. Simo˜ es, S. Nir, R. Gaspar, N. Du¨zgu¨nes, M.C. Pedroso de Lima, Interaction of cationic liposomes and their DNA com- plexes with monocytic leukemia cells, Biochim. Biophys. Acta 1418 (1999) 71–84.
[3] D.S. Friend, D. Papahadjopoulos, R.J. Debs, Endocytosis and intracellular processing accompanying transfection mediated by cationic liposomes, Biochim. Biophys. Acta 1278 (1996) 41–
50.
[4] O. Zelpathi, F.C. Szoka Jr., Mechanism of oligonucleotide release of cationic liposomes, Proc. Natl. Acad. Sci. U.S.A. 93 (1996) 11493–11498.
[5] T. Stegmann, J.-Y. Legendre, Gene transfer mediated by cationic lipids: lack of correlation between lipid mixing and transfection, Biochim. Biophys. Acta 1325 (1997) 71–79.
[6] A. El Ouahabi, M. Thiry, V. Pector, R. Fuks, J.M. Ruysschaert,
M. Vandenbranden, The role of endosome destabilizing activity in the gene transfer process mediated by cationic lipids, FEBS Lett. 414 (1997) 187–192.
[7] I. Koltover, T. Salditt, J.O. Radler, C.R. Sa¨finya, An inverted hexagonal phase of cationic liposomes-DNA complexes related to DNA release and delivery, Science 281 (1998) 78–81.
[8] A. Noguchi, T. Furuno, C. Kawaura, M. Nakanishi, Membrane fusion plays important role in gene transfection mediated by cationic liposomes, FEBS Lett. 433 (1998) 169–173.
[9] S. Simo˜es, V. Slepushkin, P. Pires, R. Gaspar, M.C. Pedroso de Lima, N. Du¨zgu¨nes¸, Human serum albumin enhances DNA transfection by lipoplexes and confers resistance to inhibition by serum, Biochim. Biophys. Acta 1463 (2000) 459–469.
[10] T. Gira˜o da Cruz, S. Simo˜es, P. Pires, S. Nir, M. Pedroso de Lima, Kinetic analysis of the initial steps involved in lipoplex- cell interactions: effect of various factors that influence trans- fection activity, Biochim. Biophys. Acta 1510 (2001) 136–
151.
[11] I.S. Zuhorn, R. Kalicharan, D. Hoekstra, Lipoplex-mediated transfection of mammalian cells occurs through the cholesterol- dependent clathrin-mediated pathway of endocytosis, J. Biol. Chem. 277 (2002) 18021–18028.
[12] Y.S. Tarahovsky, A.L. Arsenault, R. MacDonald, T.J. McIntosh,
R.M. Epand, Electrostatic control of phospholipid polymorphism, Biophys. J. 79 (2000) 3193–3200.
[13] I.M. Hafez, P.R. Cullis, Roles of lipid polymorphism in intracel- lular delivery, Adv. Drug Deliv. Rev. 47 (2001) 139–148.
[14] Y.S. Tarahovsky, R. Koynova, R.C. MacDonald, DNA release from lipolexes by anionic lipids: correlation with lipid mesomorphism, interfacial curvature, and membrane fusion, Biophys. J. 87 (2004) 1054–1064.
[15] L. Stamatatos, R. Leventis, M.J. Zuckermann, J.R. Silvius, Inter- action of cationic lipid vesicles with negatively charged phospho- lipid vesicles and biological membranes, Biochemistry 27 (1988) 3917–3925.
[16] N. Du¨zgu¨nes¸, J.A. Goldstein, D.S. Friend, P.L. Felgner, Fusion of liposomes containing a novel cationic lipid, N- [2,3-(dioleyloxy)propyl]-N,N,N-trimethylammonium: induction by multivalent anions and asymmetric fusion with acidic phos- pholipid vesicles, Biochemistry 28 (1989) 9179–9184.
[17] D.P. Pantazatos, R.C. MacDonald, Directly observed membrane fusion between oppositely charged phospholipid bilayers, J. Membr. Biol. 170 (1999) 27–38.
[18] D.P. Pantazatos, S.P. Pantazatos, R.C. MacDonald, Bilayer mix- ing, fusion, and lysis following the interaction of populations of cationic and anionic phospholipid bilayer vesicles, J. Membr. Biol. 194 (1999) 129–139.
[19] S.P. Pantazatos, R.C. MacDonald, Real-time observation of lipoplex formation and interaction with anionic bilayer vesicles,
J. Membr. Biol. 191 (2003) 99–112.
[20] L. Wang, R.C. MacDonald, New strategy for transfection: mix- tures of medium-chain and long-chain cationic lipids synergisti- cally enhance transfection, Gene Ther. 11 (2004) 1358–1362.
[21] P. Devaux, Static and dynamic lipid asymmetry in cell mem- branes, Biochemistry 30 (1991) 1163–1173.
[22] J.P. Clamme, S. Bernacchi, C. Vuilleumier, G. Duportail, Y. Me´ly, Gene transfer by cationic surfactants is essentially limited by the trapping of the surfactant/DANN complexes onto the cell mem- brane: a fluorescence investigation, Biochim. Biophys. Acta 1467 (2000) 347–361.
[23] S. Bose, I. Tuunainen, M. Parry, O. Penate Medina, G. Mancini,
P.K.J. Kinnunen, Binding of cationic liposomes to apoptotic cells, Anal. Biochem. 331 (2004) 385–394.
[24] D.L. Bratton, V.A. Fadok, D.A. Richter, J.M. Kailey, L.A. Guthrie, Appearance of phosphatidylserine on apoptotic cells requires calcium-mediated nonspecific flip-flop and is enhanced by loss of the aminophospholipid translocase, J. Biol. Chem. 272 (1997) 26159–26165.
[25] P. Williamson, R.A. Schlegel, Transbilayer phospholipid move- ment and the clearance of apoptotic cells, Biochim. Biophys. Acta 1585 (2002) 53–63.
[26] D.K. Struck, D. Hoekstra, R.E. Pagano, Use of resonance energy transfer to monitor membrane fusion, Biochemistry 20 (1981) 4093–4099.
[27] J.M. Larkin, M.S. Brown, J.L. Goldstein, R.G. Anderson, Deple- tion of intracellular potassium arrests coated pit formation and receptor-mediated endocytosis in fibroblasts, Cell 33 (1983) 273–285.
[28] H. Oubrahim, E.R. Stadtman, P. Boon Chock, Mitochondria play no roles in Mn(II)-induced apoptosis in HeLa cells, Proc. Natl. Acad. Sci. U.S.A. 98 (2001) 9505–9510.
[29] Y. Ishizaki, L. Cheng, A.W. Mudge, M.C. Raff, Programmed cell death by default in embryonic cells, and cancer cells, Mol. Biol. Cell 6 (1995) 1443–1458.
[30] E. Maeno, Y. Ishizaki, T. Kanaseki, A. Hazama, Y. Okada, Nor- motonic cell shrinkage because of disordered volume regulation is an early prerequisite to apoptosis, Proc. Natl. Acad. Sci. U.S.A. 97 (2000) 9487–9492.
[31] I.K. Buckley, T.R. Raju, M. Stewart, Claims that intermediate filaments contain F-actin are unwarranted, J. Cell Biol. 90 (1981) 309–311.
[32] X. Dong Zhang, S.K. Gillespie, P. Heresy, Staurosporine induces apoptosis of melanoma by both caspase-dependent and -independent apoptotic pathways, Mol. Cancer Ther. 3 (2004) 187–197.
[33] A.L. Bailey, P.R. Cullis, Membrane fusion with cationic lipo- somes: effects of target membrane lipid composition, Biochem- istry 36 (1997) 1628–1634.
[34] C. Mauz-Korholz, S. Dietzsch, P. Schippel, U. Banning, D. Korholz, Molecular mechanisms of hyperthermia- and cisplatin- induced cytotoxicity in T cell leukemia, Anticancer Res. 23 (2003) 2643–2647.
[35] H. Sook Chung, S. Ro Park, E. Kyung Choi, H.-J. Park, R.J. Griffin, C.W. Song, H.J. Park, Role of sphingomyelin-MAPKs pathway in heat-induced apoptosis, Exp. Mol. Med. 35 (2003) 181–188.
[36] B. Gleiss, V. Gogvadze, S. Orrenius, B. Fadeel, Fas-triggered phosphatidylserine exposure is modulated by intracellular ATP, FEBS Lett. 519 (2002) 136–158.
[37] D.C. Harrison, J.J. Lamasters, B. Herman, A pH-dependent phos- pholipase A2 contributes to loss of plasma membrane integrity during chemical hypoxia in rat hepatocytes, Biochem. Biophys. Res. Commun. 174 (1991) 654–659.
[38] F.F. Sun, W.E. Fleming, B.M. Taylor, Degradation of membrane phospholipids in the cultured human astrogial cell line UC- 11MG during ATP-depletion, Biochem. Pharmacol. 45 (1993) 1149–1155.
[39] M. Karmazyn, The myocardial sodium-hydrogen exchanger (NHE) and its role in mediating ischemic and reperfusion injury, Keio J. Med. 47 (1998) 65–72.
[40] T.J. Aldwinckle, Q.F. Ahkong, A.D. Bangham, D. Fisher, J.A. Lucy, Effects of poly(ethylene glycol) on liposomes and ery- throcytes permeability changes and membrane fusion, Biochim. Biophys. Acta 689 (1982) 548–560.
[41] D.S. Dimitrov, R.K. Jain, Membrane stability, Biochim. Biophys. Acta 779 (1984) 437–468.
[42] L. Chernomordik, Non-bilayer lipids and biological fusion inter- mediates, Chem. Phys. Lipids 81 (1996) 203–213.
[43] J. Zimmerberg, L.V. Chernomordik, Membrane fusion, Adv. Drug Deliv. Rev. 38 (1999) 197–205.
[44] J.I. Howell, J.A. Lucy, Cell fusion induced by lysolecythin, FEBS Lett. 4 (1969) 147–150.
[45] A. Agostini, M. Stromer, W. Hasselbach, Effect of lipid modifica- tion on fusion of sarcoplasmic reticulum vesicles, Z. Naturforsch. C 33 (1978) 428–436.
[46] T. Nagao, T. Kubo, R. Fujimoto, H. Nishio, T. Takeuchi, F. Hata, Ca2+-independent fusion of secretory granules with phospholi- pase A2-treated plasma membranes in vitro, Biochem. J. 307 (1995) 563–569.
[47] S. Rufini, J.Z. Pedersen, A. Desideri, P. Luly, Beta-bungarotoxin- mediated liposome fusion: spectroscopic characterization by fluorescence and ESR, Biochemistry 29 (1990) 9644–9651.
[48] G. Cevc, H. Richardsen, Lipid vesicles and membrane fusion, Adv. Drug Deliv. Rev. 38 (1999) 207–232.
[49] J.C. Seagrave, S.W. Burchiel, Interactions between benzo[a]pyrene and UVA light affecting ATP levels, cytoskeletal organization, and resistance to trypsinization, Toxicol. Lett. 117 (2000) 11–23.
[50] S.J. Atkinson, M.A. Hosford, B.A. Molitoris, Mechanism of actin polymerization in cellular ATP depletion, J. Biol. Chem. 7 (2003) 5194–5199.
[51] K.H. Sit, R. Paramanantham, B.H. Bay, K.P. Wong, Reduced sur- face area in apoptotic rounding of human Chang liver cells from serum deprivation, Anat. Rec. 240 (1994) 456–468.
[52] S.J. Eastman, M.J. Hope, K.F. Wong, P.R. Cullis, Influence of phospholipid asymmetry on fusion between large unilammellar vesicles, Biochemistry 31 (1992) 4262–4268.
[53] A.L. Bailey, P.R. Cullis, Modulation of membrane fusion by asymmetric transbilayer distribution of aminolipids, Biochem- istry 33 (1994) 12573–12580.
[54] J.M. Baldwin, O’Reilly, M. Whitney, J.A. Lucy, Surface expo- sure of phosphatidylserine is associated with the swelling and osmotically-induced fusion of human erythrocytes in the presence of Ca2+, Biochim. Biophys. Acta 1028 (1990) 14–20.
[55] M. Schewe, P. Muller, T. Korte, A. Herrman, The role of phospholipid DOTAP chloride asymmetry in calcium phosphate-induced fusion of human erythrocytes, J. Biol. Chem. 267 (1992) 5910–5915.