Article pubs.acs.org/Langmuir
Ca2+-Mediated Anionic Lipid−Plasmid DNA Lipoplexes. Electrochemical, Structural, and Biochemical Studies Ana L. Barrán-Berdón,† Belén Yélamos,∥ Marc Malfois,⊥ Emilio Aicart,† and Elena Junquera*,† †
Grupo de Química Coloidal y Supramolecular, Departamento de Química Física I, and ∥Departamento de Bioquímica y Biología Molecular I, Facultad de Ciencias Químicas, Universidad Complutense de Madrid, 28040 Madrid, Spain ⊥ Consorcio para la Construcción, Equipamiento y Explotación del Laboratorio de Luz de Sincrotrón (CELLS), 08290 Cerdanyola del Vallès, Barcelona, Spain S Supporting Information *
ABSTRACT: Several experimental methods, such as zeta potential, gel electrophoresis, small-angle X-ray scattering, gene transfection, fluorescence microscopy, flow cytometry, and cell viability/cytotoxicity assays, have been used to analyze the potential of anionic lipids (AL) as effective nontoxic and nonviral DNA vectors, assisted by divalent cations. The lipoplexes studied are those comprised of the green fluorescent protein-encoding plasmid DNA pEGFP-C3, an anionic lipid as 1,2-dioleoyl-sn-glycero-3-phospho-(1′-rac-glycerol) (DOPG) or 1,2-dioleoyl-sn-glycero-3-phospho-L-serine (DOPS), and a zwitterionic lipid, the 1,2-dioleoyl-sn-glycero-3-phosphatidylethanolamine (DOPE, not charged at physiological pH). The studies have been carried on at different liposome and lipoplex compositions and in the presence of a variety of [Ca2+]. Electrochemical experiments reveal that DOPG/DOPE and DOPS/ DOPE anionic liposomes may compact more effectively pDNA at low molar fractions (with an excess of DOPE) and at AL/ pDNA ratios ≈20. Calcium concentrations around 15−20 mM are needed to yield lipoplexes neutral or slightly positive. From a structural standpoint, DOPG/DOPE-Ca2+-pDNA lipoplexes are self-assembled into a HcII phase (inverted cylindrical micelles in hexagonal ordering with plasmid supercoils inside the cylinders), while DOPS/DOPE-Ca2+-pDNA lipoplexes show two phases in coexistence: one classical HcII phase which contains pDNA supercoils and one Lα phase without pDNA among the lamellae, i.e., a lamellar stack of lipidic bilayers held together by Ca2+ bridges. Transfection and cell viability studies were done with HEK293T and HeLa cells in the presence of serum. Lipoplexes herein studied show moderate-to-low transfection levels combined with moderate-to-high cell viability, comparable to those yield by Lipofectamine2000*, which is a cationic lipid (CL) standard formulation, but none of them improve the output of typical CL gen vectors, mostly if they are gemini or dendritic. This fact would be indicating that, nowadays, lipofection via anionic lipids and divalent cations as mediators still needs to enhance transfection levels in order to be considered as a real and plausible alternative to lipofection through improved CLs-based lipoplexes.
■
supramolecular devices, and so on),7−17 new formulations (nonviral in nature) capable of compacting, transporting, protecting, and delivering genetic material into the cells, all those actions being done as much efficiently as possible and also with the lowest possible level of toxicity to cells. Among the above-mentioned vehicles, lipofection (transfection with lipidic formulations) has been revealed as one of the most adequate methods to transfect DNA to a wide variety of cells. However, the use of cationic lipids (CLs) presents several inconveniences,18−21 such as cytotoxicity in vitro and in vivo and instability of their complexes with DNA in the presence of serum.22−24 In this respect, anionic lipids (ALs), which are present in the eukaryotic cell membrane, with less levels of phagocytosis by macrophages25 and higher biocompatibility,
INTRODUCTION Therapeutic gene transfer is not a new concept as it was first developed more than two decades ago.1,2 The treatment of human disease by gene therapy is promising in the medical field, because it is possible to revert the effects of many genetic disorders3,4 through the introduction of recombinant DNA into the cell with the aim of producing the necessary biologically active proteins to repair the cell damage.5 However, DNA internalization requires that the biopolymer must be first compacted by viral or nonviral vectors and then protected from degradation. Virus-based vectors, although more efficient to date as transfecting agents than nonvirus-based systems, may provoke immunogenic reactions.4,6 In addition, the nonviral systems are easier to prepare and can transfect larger plasmids into cells.4 In the past years, a great deal of effort has been devoted to find out, among the wide range of possibilities (polymers, cationic liposomes, zwitterionic liposomes, magnetic and nonmagnetic nanoparticles, amphiphilic cyclodextrins, © 2014 American Chemical Society
Received: April 11, 2014 Revised: September 8, 2014 Published: September 11, 2014 11704
dx.doi.org/10.1021/la502823z | Langmuir 2014, 30, 11704−11713
Langmuir
Article
Scheme 1. Molecular Structure of (a) DOPG, (b) DOPS, and (c) DOPE
may be pointed out as potentially safer DNA delivery systems. Their interaction with negatively charged DNA helixes has to be mediated by cations, mostly divalent cations, which may act as effective cationic bridges between the anionic liposomes and DNA. Several divalent cations have been tested for use in anionic lipoplex formulations, such as Ca2+, Mg2+, Mn2+, Co2+, Cd2+, and Zn2+.22,26 Among them, Ca2+ cations, which have proved to be the most effective for DNA compaction with anionic liposomes,5,22 not only may condense DNA molecules, but also may promote uptake of lipoplexes by the cells.27 Interesting studies related to the interaction of either ALs and/ or zwitterionic lipids, with a variety of cations in the absence of DNA, can be also found in the literature.28−32 Similarly to cationic lipoplexes, transfection efficiency (TE) may be improved by the incorporation of helper lipids, such as 1, 2dioleoyl-sn-glycero-3-phosphatidylethanolamine (DOPE), into the delivery system. It is well-known that, due to the fusogenic character of DOPE, it improves the stability of liposome, enhances tissue penetration, and helps initiate the breakout of DNA from endosomes. Some publications attributed these benefits to its capacity to induce the inverted hexagonal lipid phase HCII in the lipoplex, due to its characteristic packing parameter.33,34 With our previous background in the characterization of the role of CLs on the DNA transfection to a wide variety of cell lines, this work aims to confirm if lipofection where CLs are substituted by (ALs + Ca2+) may be established as a new route for gene transfection, checking if this approach may be outlined as an improved (more efficient and less cytotoxic) nonviral gene delivery system with respect to the well-established vectors, in general, and CLs in particular. Accordingly, this works reports the physicochemical and biophysical characterization of the mixed liposomes formed by an anionic lipid as 1,2-dioleoyl-sn-glycero-3-phospho-(1′-racglycerol) (DOPG) or 1,2-dioleoyl-sn-glycero-3-phospho-Lserine (DOPS) and the zwitterionic lipid (DOPE), the last one possessing null charge at physiological pH = 7.4, at different molar ratios, and of the lipoplexes that they formed with the green fluorescent protein-encoding plasmid DNA pEGFP-C3 (pDNA),35 in the presence of different Ca2+
concentrations. Similar systems have been previously reported in the literature,22,26,27,36−38 but none of these publications includes an analysis of the influence of lipoplex structure on TE and cytotoxicity levels (structure−activity relationship). In this work, the electrochemical behavior of liposomes and lipoplexes has been analyzed by zeta potential and gel electrophoresis. Small-angle X-ray scattering (SAXS) has been used to obtain information about the lipoplex structures, with the aim of correlating them with their capabilities as gene transfecting agents. Transfection efficiency experiments (fluorescence microscopy and FACS) have been done in two cell lines: human embryonic kidney 293T cells (HEK293T) and HeLa cells (cervical cancer cells). Cell viability studies have been carried out by MTT assay to establish the lipoplexes cytotoxicity. It is expected that the whole picture of results, both biophysical and biochemical, will clarify if lipofection protocols involving ALs in the presence of divalent cations may be considered as a promising alternative in gene therapy.
■
EXPERIMENTAL SECTION
Materials. Lipids. The lipids (see Scheme 1), both anionic (sodium salts of 1,2-dioleoyl-sn-glycero-3-phospho-(1′-rac-glycerol), DOPG, and of 1,2-dioleoyl-sn-glycero-3-phospho-L-serine, DOPS) and zwitterionic (1,2-dioleoyl-sn-glycero-3-phosphatidylethanolamine, DOPE) with the best purity, were purchased from Avanti Polar Lipids. The calcium chloride, with 99.5% purity, was provided from Merck. All the reagents and solvents used for the present study were of highest grade available commercially. Plasmid DNA. pEGFP-C3 plasmid DNA (pDNA) was extracted from competent Escherichia coli bacteria previously transformed with pEGFP-C3. The extraction was carried out using GenElute HP Select plasmid Gigaprep Kit (Sigma-Aldrich) following a protocol previously described (see Supporting Information for more details).39,40 Preparation of Mixed Liposomes and Lipoplexes. Appropriate amounts of anionic lipid (AL), L−, and DOPE, L0, were dissolved in chloroform to obtain the desired AL composition, α, of the mixed liposomes. After this solution was briefly vortexed, chloroform was removed to yield a dry lipid film. The resulting dry lipid films were then hydrated with HEPES 10 mM, pH = 7.4, and homogenized by means of a combination of vortexing, sonication, and moderate heat. The resulting MLVs were transformed into the desired unilamellar liposomes, LUVs, by a sequential extrusion procedure widely explained 11705
dx.doi.org/10.1021/la502823z | Langmuir 2014, 30, 11704−11713
Langmuir
Article
Figure 1. Plot of zeta potential, ζ, as a function of Ca2+ concentration for (a) DOPG/DOPE-Ca2+-pDNA lipoplexes and (b) DOPS/DOPE-Ca2+pDNA lipoplexes at α = 0.20 and AL/pDNA mole ratio of 10 (black squares), 15 (red circles), and 20 (blue triangle). Also included in the figures are zeta potential values for the mixed liposomes in the absence of pDNA (control, open symbols). All the measurements are in HEPES buffer (10 mM) at 298.15 K, pH = 7.4. Errors are within ±5%, [pDNA] = 0.05 mg/mL. elsewhere.39,40 Appropriate amounts of a stock solution of pDNA, prepared 1 day before, were first mixed with adequate volumes of calcium solution and subsequently added to liposomal suspensions in order to obtain the lipoplexes with the desired final Ca2+ concentration and AL/pDNA ratios. pDNA concentrations were chosen to fit the optimum conditions for each experimental technique (see Supporting Information for more details) Zeta Potential. The phase analysis light scattering technique (Zeta PALS, Brookhaven Instrum. Corp., U.S.) was used to measure electrophoretic mobility (and from it, zeta potential) (see Supporting Information for more details).41,42 Zeta potential values were measured for lipoplexes as a function of calcium concentration, at several mixed lipid compositions, α, and at several AL/pDNA mole ratios. Gel Electrophoresis. Lipoplexes along with uncomplexed plasmid DNA were loaded onto 1% agarose gel and run for 30 min at 80 mV in 1× TAE (Tris-HCl, acetate, and EDTA) buffer (see Supporting Information for further details). Fully complexed lipoplexes appeared as fluorescent bands in wells of the gel, while uncomplexed pDNA appeared outside of the well. Fluorescence intensity of each band was measured by using commercial Quantity One software provided with Gel Doc XR instrument (Bio-Rad). The band intensity for the free pDNA was considered as 100%, and other intensities were estimated accordingly. Small-Angle X-ray Scattering. SAXS experiments were carried out on the beamline NCD11 at ALBA Synchrotron Barcelona (Spain). The energy of the incident beam was 12.6 keV (λ = 0.995 Å). Samples were placed in sealed glass capillaries. The scattered X-ray was detected on CCD detector Quantum 210r, converted to onedimensional scattering by radial averaging, and represented as a function of the momentum transfer vector (see Supporting Information for additional details). SAXS experiments were run for both the liposomes and the lipoplexes. Transfection of pDNA. Transfection of pEGFP-C3 plasmid DNA across HEK293T cells (human embryo kidney transformed cancer) and HeLa cells (cervical cancer cells) using DOPG/DOPE and DOPS/DOPE lipid mixtures was performed in the presence of serum, at several [Ca2+]. Each sample is prepared in duplicate, and experiments are repeated twice independently for each composition. Transfection efficiency, TE, was evaluated by means of both fluorescence microscopy and FACS experiments (see more details in the Supporting Information). Control experiments were performed using commercial transfection reagent Lipofectamine2000* (Lipo2000*). Cell Viability Assay. The cytotoxicity or cell viability of each lipid formulation, as well as of Ca2+ solutions alone, toward HEK293T and HeLa cells was determined by using 3-(4,5-dimethylthiazole-2-yl)-2,5diphenyl tetrazolium bromide reduction method (MTT assay) following literature procedure43,44 (see Supporting Information for more details). Lipoplexes were prepared using 50 ng of pDNA per sample with the appropriate amount of mixed liposome and Ca2+ in order to obtain the desired AL/pDNA and pDNA:Ca2+ ratios. As fully
detailed in the Supporting Information, the percent cell viability was calculated from readings obtained from ELISA reader.
■
RESULTS AND DISCUSSION To figure out if anionic lipofection mediated by divalent cations may be a plausible alternative via to the widely explored cationic vectors, a complete physicochemical characterization, involving both electrochemical and structural evaluation, carefully correlated with biochemical tests to check both the transfection efficiency and the cytotoxicity, is necessary. This section will present these three studies: electrochemical, structural, and biochemical, in this order. The negative charge of either DOPG/DOPE or DOPS/ DOPE liposomes is expected to be reduced upon interaction with pDNA in the presence of Ca2+ cations. Figure 1a and b shows zeta potential data for DOPG/DOPE-Ca2+-pDNA and DOPS/DOPE-Ca2+ lipoplexes (α = 0.20, as an example), respectively, at several AL/pDNA mole ratios and as a function of [Ca2+]. Also included are the control values for both AL/ DOPE-Ca2+ systems in the absence of pDNA, corresponding with the experiments at AL/pDNA ratio = 20. It is remarkable that the typical sigmoidal fit found for lipoplexes constituted by cationic lipids and DNA35,40,42 is not seen herein. In fact, at low [Ca2+], zeta potential shows a sharp rise (from approximately −50 or −60 mV to around 0 or −10 mV) upon increasing [Ca2+], but it tends to level off at around electroneutrality (ζ ∼ 0 mV) for [Ca2+] around 15−20 mM, irrespectively of either the type of anionic lipid used (DOPG or DOPS) or the AL/ pDNA ratio. It means that the negative charge of anionic liposomes can be balanced with the addition of calcium salts, but once the charge is almost neutral (null zeta potential), an excess of Ca2+ remains in the bulk and does not contribute to zeta potential of the nanoaggregate. In other words, these findings are likely to be due to a balance in charge ratios within the lipoplexes, and, accordingly, a small increase in lipid or Ca2+ concentrations above this point scarcely contributes to the overall zeta potential. A subtle effect of AL/pDNA mole ratio, however, can be noticed; thus, for a given [Ca2+], zeta potential is slightly higher upon increasing lipid/DNA ratio, this effect being more notorious as long as [Ca2+] increases. This feature could be indicating that as long as AL content increases, more Ca2+ can be sandwiched between the negative surfaces of AL/ DOPE liposomes and the available negative charges of pDNA, the whole charged entity resulting with a net zeta potential around zero or slightly positive. Anyway, note that the content in AL is low for the sample shown in Figure 1 (α = 0.20); this means that a high amount of Ca2+ is not necessary to balance 11706
dx.doi.org/10.1021/la502823z | Langmuir 2014, 30, 11704−11713
Langmuir
Article
words, Ca2+ cations make possible the interaction among two negatively charged entities, i.e., pDNA plasmid molecules and the anionic membranes, thus favoring pDNA complexation and hindering pDNA electrophoretic mobility. This fact occurs at [Ca2+] around 20−25 mM in both cases, in very good agreement with zeta potential experiments. Structure, size, and morphology of lipoplexes are factors that clearly affect transfection efficiency. The average hydrodynamic diameters of mixed liposomes, determined with the PALS technique, were (119 ± 10) nm and (117 ± 9) nm for the AL/ DOPE mixed liposomes, where AL = DOPG or DOPS, respectively. These sizes are consistent with the extrusion protocol used,45 which is based on a sequential procedure that ends with a final pass of the mixed liposome solution through a polycarbonate membrane with pores of 100 nm diameter. Structural studies of DOPG/DOPE-Ca2+-pDNA and DOPS/ DOPE-Ca2+-pDNA lipoplexes, as well as the corresponding controls, i.e., AL/DOPE/Ca2+ systems in the absence of pDNA, have been carried out by using SAXS experiments. Figure 3 in the main text and Figure SI-2 of Supporting Information show SAXS diffractograms, with the Miller indices being included, for the lipoplexes and the mixed liposomes (controls) studied in this work, respectively. It is remarkable that, while SAXS peaks in Figure 3a (DOPG/DOPE-Ca2+-pDNA) index well with a HcII phase (inverted cylindrical micelles in hexagonal ordering), diffractograms in Figure 3b (DOPS/DOPE-Ca2+-pDNA) show the coexistence of two phases, one lamellar Lα type (stack of lipidic bilayers) and another HcII phase. The peaks corresponding to these phases can be more clearly seen in Figure 4 that shows, as an example, the diffractograms for the lipoplexes studied herein, at α = 0.20, AL/pDNA ratio = 20, and [Ca2+] = 25 mM, in a logarithmic scale. The reason for this different structural behavior, i.e., only HcII phases in DOPG/DOPE-Ca2+pDNA lipoplexes and coexistence of HcII + Lα phases in DOPS/ DOPE-Ca2+-pDNA lipoplexes, despite the fact that the three lipids are structurally identical with respect to the hydrophobic part, must be due to the differences on their polar heads, which, in turn, imply a different packing parameter, P, of the lipids used. In this respect, given that DOPG, as well as DOPE, has a slightly smaller polar head than DOPS (P > 1), it will have a higher tendency to form hexagonal reverse aggregates which, in turn, would justify the structural pattern found in the lipoplexes studied herein. Furthermore, note that the mixed liposomes used have a high DOPE content. It must be remarked that the coexistence of phases, also known as lipid demixing, is not new because such behavior was already found elsewhere39,40,46,47 and is documented to be related to a better transfection level,
the negative charge. Note as well that almost no differences in zeta potential curves have been found in the control done in the absence of pDNA as a function of calcium concentration, with respect to the experiments done in the presence of pDNA. This experimental evidence was expected given that zeta potential is an interfacial property; i.e. it measures the potential (or net charge) of the particles in the shear plane when they moved by the action of an electric field; when mixed liposomes, Ca2+, and pDNA are present, pDNA is compacted in a variety of structures within which pDNA is condensed. Agarose gel electrophoresis experiments confirm all the above-mentioned features. These studies are very useful to analyze the efficiency of these AL-Ca2+ formulations in complexing pDNA. Experiments were done at α = 0.20, 0.25, and 0.5, each of them at AL/pDNA ratios of 10, 15, and 20, with [Ca2+] ranging from 10 to 100 mM (see Supporting Information Figure SI-1), but results revealed that pDNA was compacted only at α = 0.20 and AL/pDNA ratio = 20, which were fixed as the optimum experimental conditions for all the studies herein included. Figure 2a and b resumes these results
Figure 2. Gel electrophoresis results for (a) DOPG/DOPE-Ca2+pDNA lipoplexes and (b) DOPS/DOPE-Ca2+-pDNA lipoplexes at α = 0.20 and AL/pDNA mole ratio of 20. Lane 1: pDNA. Lanes 2−6: lipoplexes in the presence of different [Ca2+]: 5 mM (lane 2), 10 mM (lane 3), 25 mM (lane 4), 50 mM (lane 5), and 75 mM (lane 6).
for DOPG/DOPE-Ca2+-pDNA and DOPS/DOPE-Ca2+-pDNA lipoplexes at α = 0.20 and AL/pDNA ratio = 20 (lanes 2−6), as an example, together with those for uncomplexed pDNA (lane 1), as a positive control. Ca2+ concentration increases from 5 mM (lane 2) to 75 mM (lane 6). As it can be observed, at [Ca2+] up to 5−10 mM (i.e., lanes 2 and 3), there is still free (uncomplexed) pDNA that moves along the lane. Increasing Ca2+ concentration decreases the amount of free pDNA; in fact, a complete loss of pDNA bands along the lane in lanes 4−6 is a clear indication of a full compaction of pDNA (that remains in the well) by Ca2+-bridged AL/DOPE liposomes. In other
Figure 3. SAXS diffractograms of (a) DOPG/DOPE-Ca2+-pDNA lipoplexes and (b) DOPS/DOPE-Ca2+-pDNA lipoplexes at α = 0.20, AL/pDNA mole ratio = 20, and several [Ca2+]. 11707
dx.doi.org/10.1021/la502823z | Langmuir 2014, 30, 11704−11713
Langmuir
Article
Figure 4. SAXS diffractograms of (a) DOPG/DOPE-Ca2+-pDNA lipoplexes and (b) DOPS/DOPE-Ca2+-pDNA lipoplexes at α = 0.20, AL/pDNA mole ratio = 20, and [Ca2+] = 25 mM, showing an inverted hexagonal HcII structure in (a) and the coexistence of inverted hexagonal HcII (black) and lamellar Lα (blue) structures in (b).
maybe either linear (ctDNA) or circular (pDNA),33,40,48,49 indicates that hexagonal phases are less compacted than lamellar ones; (ii) Ca2+ concentration seems to have an almost negligible effect on the structure of the resulting lipoplexes, since almost no variation of d vs [Ca2+] is observed (see Figure 5); and (iii) d is the same, within the associated error dLα = (5.3 ± 0.2) nm, in the case of Lα phases of DOPS/DOPE-Ca2+ in both the absence and presence of pDNA. This value is in very good agreement with the literature.28 In the case of the HcII phases, however, d values, very similar in both systems, are slightly lower for DOPG/DOPE-Ca2+ (dHcII ≈ 7.1) and DOPS/ DOPE-Ca2+ (dHcII ≈ 7.0 nm) in the absence of pDNA than those found in the presence of pDNA (dHcII ≈ 7.6−7.7 nm). Scheme 2 shows both 2D and 3D views of the abovementioned structures. In both HcII and Lα phases, the interlayer distance d can be expressed as the sum of the thickness of the lipidic region, dm, containing the hydrophobic tails of the lipids and that of the aqueous region, dw, hypothetically containing supercoiled pDNA and divalent cations (d = dm + dw). Notice, however, that HcII is a reverse phase while Lα is a direct one; i.e., in HcII structures, d is the distance between the centers of two adjacent cylinders, dw is the inner diameter of cylindrical micelles, and dm is the distance between the surfaces of adjacent cylinders, while in Lα structures d is the distance between lipidic bilayers, dw is the aqueous monolayer between two lipidic bilayers, and dm is the thickness of the lipidic bilayer. Considering that the three lipids of this work show an identical hydrophobic region (see Scheme 1), a value of ≈4.5 nm can be estimated for dm in all the cases, which means in the case of HcII structures (dHcII ≈ 7.6 nm) that the inner diameter of the cylinders is ≈3.1 nm, while in Lα structure (dLα ≈ 5.5 nm) the aqueous region is ≈1 nm thick. Considering, additionally, that an hydrated Ca2+ cation has a radius of ≈0.4 nm,50 the structural data of HcII phase reveal that the inner aqueous region of reversed micelles in hexagonal order may allocate and compact the pDNA negatively charged supercoils, the interaction among the biopolymer and the anionic lipidic surfaces being mediated by a layer of Ca2+ cations in an ioncoated lipid tubes fashion. This picture would explain why dHcII values found for AL/DOPE-Ca2+-pDNA lipoplexes studied herein are slightly higher (see Figure 5 and Supporting Information Table SI-1) than those for AL/DOPE-Ca2+
although this conclusion must be confirmed later with biochemical tests. It is remarkable that the same scenario of structures has been found for the mixed liposomes in the absence of pDNA (Figure SI-2 of the Supporting Information). Accordingly, DOPG/DOPE-Ca2+ and DOPS/DOPE-Ca2+ mixed systems show only HII phases in the first system and coexistence of both HII and Lα phases in the second one. From the data shown in the SAXS diffractograms, the interlayer distance, d, directly related to the q factor (d = 2π/ q100 in Lα phase or d = a = dpDNA = 4π/√3q10 in HcII phase), can be determined (see Table SI-1 of Supporting Information). Figure 5 shows a plot of d as a function of [Ca2+ ] for the HcII
Figure 5. Plots of the periodic distance, d, as a function of [Ca2+] for DOPG/DOPE-Ca2+-pDNA lipoplexes (HcII phase, black squares) and for DOPS/DOPE-Ca2+-pDNA lipoplexes (HcII phase, red circles; Lα phase, blue triangles). α = 0.20, AL/pDNA mole ratio = 20. Also included in the figure are d values for the mixed liposomes in the absence of pDNA (controls, corresponding open symbols).
phases of both DOPG/DOPE-Ca2+-pDNA (black solid squares) and DOPS/DOPE-Ca2+-pDNA (red solid circles) lipoplexes, and also for the Lα phase of DOPS/DOPE-Ca2+pDNA lipoplexes (blue solid triangles). Open symbols correspond to d values as a function of [Ca2+ ] for the corresponding mixed liposomes in the absence of pDNA. Notice that (i) d values in HcII phases are similar in both lipoplexes (dHcII ≈ 7.6−7.7 nm), and higher than those obtained in the lamellar one (dLα ≈ 5.5 nm). This feature, which has been also found in CL/DOPE-DNA systems, where CL is either a typical one head−two tails CL or a gemini CL, and DNA 11708
dx.doi.org/10.1021/la502823z | Langmuir 2014, 30, 11704−11713
Langmuir
Article
Scheme 2. Schematic Drawings of 3D and 2D Views of HCII and Lα Phases of AL/DOPE-Ca2+-pDNA Lipoplexes, Where AL May Be DOPG or DOPS
Figure 6. Fluorescence micrographs showing GFP expression in HEK293T cells (panels a−d and f) and HeLa cells (panels e, g, and h) obtained from DOPG/DOPE-Ca2+-pDNA lipoplexes at [Ca2+] = 25 mM (panel a for HEK293T and e for Hela) and 50 mM (panel b for HEK293T)); DOPS/DOPE-Ca2+-pDNA lipoplexes at [Ca2+] = 25 mM (panel c for HEK293T and g for Hela) and 50 mM (panel d). In all the cases, α = 0.20 and AL/pDNA ratio = 20. Panel f and h shows the results obtained with Lipofectamine2000* as a positive control for HEK293T and HeLa cell, respectively.
lipidic bilayers held together by Ca2+ bridges, and it has been previously found for DOPG/DOPC-divalent cations-ctDNA lipoplexes.26 Another evidence that confirms that pDNA is not present within these lamellar phases is supported by the fact that the d value found for the Lα phase of DOPS/DOPE-Ca2+ mixed lipidic system is the same, within the associated error, as that obtained for DOPS/DOPE-Ca2+-pDNA lipoplexes. In conclusion, DOPG/DOPE-Ca2+-pDNA lipoplexes are selfassembled into a HcII phase, while DOPS/DOPE-Ca2+-pDNA lipoplexes show two phases in coexistence: one classical HcII phase which contains all or part of pDNA supercoils and one Lα phase without pDNA among the lamellae. See Scheme 2 for a 3D representation of both phases. With respect to pDNA−pDNA correlation (pDNA−pDNA distance), it must be noticed that in HcII phase the interlayer distance, d, obtained directly from q values on diffractograms is as well dpDNA (=7.6 nm), since plasmid molecules are within the
mixed lipidic systems (controls), as commented before; assuming that dm is not affected by the absence or presence of pDNA, it is expected that pDNA provokes a slight expansion of the inner cylinder, with the corresponding increase in d values. In contrast, the structural results brought by Lα phase could not justify a classical lamellar lipoplex structure, with lipidic bilayers sandwiching pDNA molecules by means of each monolayer of Ca2+ cations, since an aqueous monolayer of ≈1 nm cannot provide enough space for such arranging. Moreover, the obtained lamellar spacing (≈5.5 nm) is characteristic of anionic membranes stacked by divalent anions without DNA.26 Accordingly, these results could be explained if one considers that a different lamellar phase is formed, without supercoiled pDNA within the aqueous monolayer between the lipid bilayers. This phase, which has no counterpart in CLs-pDNA lipoplexes by now, would be represented by a lamellar stack of 11709
dx.doi.org/10.1021/la502823z | Langmuir 2014, 30, 11704−11713
Langmuir
Article
Figure 7. Transfection of pEGFP-C3 to HEK293T cells (a) and HeLa cells (b) in the presence of serum (−FBS+FBS) by using the lipoplexes: (1) DOPG/DOPE-Ca2+-pDNA complexes, [Ca2+] = 25 mM; (2) DOPG/DOPE-Ca2+-pDNA complexes, [Ca2+] = 50 mM; (3) DOPS/DOPE-Ca2+pDNA complexes [Ca2+] = 25 mM; (4) DOPS/DOPE-Ca2+-pDNA complexes [Ca2+] = 50 mM. Lipofectamine 2000* is used as positive control. Experiments were performed using 0.8 μg pDNA per well. AL/DOPE α = 0.20 and AL/pDNA mole ratio = 20.
for the two lipoplexes studied in this work. Comparing GFP expression levels of panels b and d, one can also conclude that the anionic lipidic formulation DOPG/DOPE exhibits slightly better transfection levels than DOPS/DOPE. Nevertheless, it must be noted that any of them exceed the transfection efficiency shown by the commercial control (Lipofectamine2000*, panel f), which is also a lipidic formulation but made of a mixture of CLs. Similar conclusions can be extracted from micrographs for HeLa cells (panels e and g), although slightly lower levels of expression can even be seen in this case. These qualitatively results can be confirmed with percent GFP and MFI values obtained from FACS experiments, shown in Figure 7 for both cell lines. Note that percent GFP values are moderate-to-low in all the cases, always below 20%, and lower than the values obtained for the control Lipofectamine2000*. Again, it seems that TE efficiency is slightly better for HEK293T cells than for HeLa cells, and DOPG/DOPE-Ca2+ lipoplexes seem to have slightly better transfection performances than DOPS/DOPE-Ca2+ ones. Better TE levels have been reported by Burgess et al.22 on COS7 cells for DOPG/ DOPE-Ca2+-pEGFP-N3 lipoplexes; note, however, that both the plasmid and the cell line are different than those used in this work. MTT assays, which inform about the percentage of cells that remained viable to grow and divide after the lipoplex delivery in cells 48 h post-transfection, were performed on HEK293T (see Figure 8) and HeLa cells (Supporting Information Figure SI-3) to check the biocompatibility of DOPG/DOPE-Ca2+-pDNA
aqueous interior of micellar cylinders. This value is quite similar to the one reported for other CL/DOPE-ctDNA lipoplexes where CL is either one head−two tails type (DOEPC/DOPEctDNA49 and DC-Chol/DOPE-ctDNA41) or gemini CL/ DOPE-pDNA.40 In the case of Lα structures, in contrast, dpDNA correlation must be also inferred from diffractograms, from the corresponding peak for pDNA. It must be remarked that in the present work, a careful analysis of diffractograms of Figure 3b for DOPS/DOPE-Ca2+-pDNA lipoplexes reveals the absence of a clear pDNA peak. This feature, also found in previously reported AL/ctDNA systems,26 may be due to a weak in-plane pDNA ordering or could be also explained if pDNA molecules are not present in-between lipidic bilayers. The second reason would reinforce the above hypothesized Lα phase consisting of anionic lipidic bilayers bridged by Ca2+ cations without pDNA. Once the lipoplexes studied herein are electrochemically and structurally analyzed, it is necessary to check the following: (i) their capacity to really transfect the complexed and protected pDNA to cells and (ii) the toxicity levels that these lipidic gene vectors provoke in the cells (cell viability). Accordingly, transfection efficiency (TE) of the lipoplexes in HEK273T and HeLa cells was evaluated using both fluorescence microscopy and FACS experiments. The transfection studies were carried out at α = 0.20 and 0.25 and, for each of these compositions, at AL/pDNA ratio =15 and 20 and [Ca2+] = 10, 25, 50, and 75 mM. Keep in mind that lipoplexes with α = 0.20 and AL/pDNA ratio = 20 were optimum wih respect to the level of compaction of pDNA, as found in gel electrophoresis experiments. In fact, no significant levels of GFP expression were found at α = 0.25. Figure 6 shows the fluorescence micrographs of both HEK293T and HeLa cells for the optimum formulations (α = 0.20 and AL/pDNA = 20) of DOPG/DOPE-Ca2+-pDNA and DOPS/DOPE-Ca2+-pDNA lipoplexes and Lipofectamine2000*, as a positive control, in the presence of serum (−FBS+FBS). The green color implies that pDNA has been efficiently transfected to the nucleus of cells, thus provoking the expression of GFP. Accordingly, a first sight of fluorescence micrographs reveals that both DOPG/ DOPE and DOPS/DOPE anionic liposomes seem to be able to compact, protect, and transfect the green fluorescent protein encoding plasmid DNA pEGFP-C3 (pDNA) to HEK293T cells, in the presence of Ca2+ cations as effective mediators of the process, at the ionic concentrations used in this work. Furthermore, a more careful analysis would point to [Ca2+] = 50 mM (panels b and d) as the optimum ionic concentration
Figure 8. Cell viability assay of DOPG/DOPE-Ca2+-pDNA and DOPS/DOPE-Ca2+-pDNA lipoplexes at several [Ca2+] in HEK293T cells. Lipo2000* is used as a positive control. 11710
dx.doi.org/10.1021/la502823z | Langmuir 2014, 30, 11704−11713
Langmuir
Article
and DOPS/DOPE-Ca2+-pDNA lipoplexes at α = 0.20, AL/ pDNA = 20 and at different [Ca2+]. The effect of Ca2+ alone on the viability of these cells was also evaluated at the same concentrations used for TE experiments (see Supporting Information Figure SI-4). In these assays, Lipofectamine2000* has been used as a positive control, as well. As can be seen in Figure 8, DOPG/DOPE-Ca2+-pDNA lipoplexes are slightly more toxic for the HEK293T cells than Lipofectamine2000*. In fact, although an appreciable effect of [Ca2+] is not observed, it is remarkable that the formulation with better transfection (i.e. [Ca2+] = 50 mM) is, unfortunately, the most cytotoxic. However, in the case of DOPS/DOPE-Ca2+-pDNA lipoplexes, cell viability results are comparable to that of Lipofectamine2000*, or even better, as the formulation with [Ca2+] = 15 mM which is more biofriendly than the control, with a cell viability reaching almost 100%. Furthermore, note that even higher concentrations of calcium yield also acceptable cell viability levels, around 80% for [Ca2+] = 25 mM and around 75% for [Ca2+] = 50 mM. On the other hand, both lipoplexes do not show significant differences in their cytotoxicity levels (around 70−75% in all the cases) against HeLa cells as a function of [Ca2+] (Supporting Information Figure SI-3), with respect to those of Lipofectamine2000*. Also remarkable is that, although the Ca2+ concentrations that have allowed the best performances for AL formulations are much higher than in vivo cellular levels, cell viability values are still high (around 100%) when either HEK293T and/or HeLa cells are in the presence of only Ca2+ solutions at concentrations of 25 and 50 mM, as can be seen in Supporting Information Figure SI-4. Nonetheless, keep in mind that these results are obtained under in vitro conditions, not always applicable in vivo. Consequently, among the lipoplexes herein reported, it seems that DOPS/DOPE-Ca2+-pDNA lipoplexes at [Ca2+] ranging from 25 mM to 50 mM, characterized by HcII phases (Lα phase does not contain pDNA), may be a reasonably good compromise solution (considering both TE and cytotoxicity results) with respect to choosing AL-Ca2+ vector with moderate biochemical performances. However, one should keep in mind that even the best results herein reported do not exceed those obtained by improved CLs-based formulations, mostly considering that the Ca2+ concentrations that allow the best performances for AL formulations are much higher than in vivo cellular levels.
lipoplexes. In addition, a lamellar Lα phase, without pDNA sandwiched in between the lipidic lamellae, has been also found in the DOPS/DOPE-Ca2+-pDNA lipoplex; it is remarkable that this phase does not have an analogue in the CL-based lipoplexes. The Ca2+-mediated lipoplexes reported in this work show moderate-to-low transfection levels combined with moderate-to-high cell viability, with comparable values to those yield by the control Lipofectamine2000*, which is a cationic lipid standard formulation. However, none of the ALCa2+-pDNA lipoplexes studied in this work improve the output of recently synthesized CLs gen vectors, mostly considering that the Ca2+ concentrations at which AL-based vectors yield the best performances are significantly higher than in vivo cellular levels. It would accordingly indicate that, nowadays, lipofection via anionic lipids and divalent cations as mediators still needs to enhance transfection levels to be considered as a plausible general alternative to lipofection through improved CLs-based lipoplexes.
■
ASSOCIATED CONTENT
* Supporting Information S
Fully detailed experimental procedures and some complementary figures of agarose gel electroforesis, SAXS, and cell viability experiments; table with structural information on both liposomes and lipoplexes. This material is available free of charge via Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*Fax: +34913944135. E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS The authors ackowledge MICINN of Spain project no. CTQ2012-30821 and project no. BFU2010-22014. SAXS experiments were performed at NCD11 beamline at ALBA Synchrotron Light Facility with the collaboration of ALBA staff. The authors also thank C. Aicart for carrying out amplification of plasmid DNA at the Biochemistry and Molecular Biology I Department of the UCM of Spain.
■
■
CONCLUSIONS Several general conclusions can be drawn regarding lipofection using AL vectors for plasmid DNA in the presence of Ca2+, from both the physicochemical and biochemical results reported in this work. DOPG/DOPE and DOPS/DOPE anionic liposomes are able to compact pDNA by means of Ca 2+ bridges. This compaction is fulfilled at a Ca 2+ concentration ≈15−20 mM, as confirmed by both zeta potential and agarose gel electrophoresis. Zeta potential profile is quite different with respect to the one found in CL-based lipoplexes, characterized by a typical sigmoidal fit. In the present case, lipoplexes are never positively charged, the net charge of the complex being around zero when pDNA is totally compacted; an excess of Ca2+ remains in the bulk and does not contribute to zeta potential of the nanoaggregates. From a structural point of view, when pDNA is compacted, it is allocated, together with the divalent cations, inside the inverted micellar cylinders that characterize the HcII phases found in both DOPG/DOPE-Ca2+-pDNA and DOPS/DOPE-Ca2+-pDNA
REFERENCES
(1) Miller, A. D. Human gene therapy comes of age. Nature 1992, 357, 455−60. (2) Anderson, W. F. Human gene therapy. Science 1992, 256, 808− 13. (3) Lasic, D. D. Liposomes in Gene Delivery; CRC Press: Boca Raton, FL, 1997. (4) Foldvari, M.; Badea, I.; Wettig, S.; Verrall, R.; Bagonluri, M. Structural characterization of novel gemini non-viral DNA delivery systems for cutaneous gene therapy. J. Exp. Nanosci. 2006, 1, 165−176. (5) Kulkarni, V. I.; Shenoy, V. S.; Dodiya, S. S.; Rajyaguru, T. H.; Murthy, R. R. Role of calcium in gene delivery. Expert Opin. Drug Delivery 2006, 3, 235−45. (6) Verma, I. M.; Somia, N. Gene therapyPromises, problems, and prospects. Nature 1997, 389, 239−242. (7) Bilalov, A.; Olsson, U.; Lindman, B. DNA−lipid self-assembly: Phase behavior and phase structures of a DNA−surfactant complex mixed with lecithin and water. Soft Matter 2011, 7, 730−742. (8) Bilalov, A.; Olsson, U.; Lindman, B. Complexation between DNA and surfactants and lipids: Phase behavior and molecular organization. Soft Matter 2012, 8, 11022−11033. 11711
dx.doi.org/10.1021/la502823z | Langmuir 2014, 30, 11704−11713
Langmuir
Article
(9) Balazs, D. A.; Godbey, W. Liposomes for use in gene delivery. J. Drug Delivery 2011, 2011. (10) McManus, J. J.; Radler, J. O.; Dawson, K. A. Phase behavior of DPPC in a DNA−calcium−zwitterionic lipid complex studied by small-angle X-ray scattering. Langmuir 2003, 19, 9630−9637. (11) Donkuru, M.; Badea, I.; Wettig, S.; Verrall, R.; Elsabahy, M.; Foldvari, M. Advancing nonviral gene delivery: Lipid- and surfactantbased nanoparticle design strategies. Nanomedicine 2010, 5, 1103− 1127. (12) Davila-Ibanez, A. B.; Salgueirino, V.; Martinez-Zorzano, V.; Marino-Fernandez, R.; Garcia-Lorenzo, A.; Maceira-Campos, M.; Munoz-Ubeda, M.; Junquera, E.; Aicart, E.; Rivas, J.; Javier Rodriguez-Berrocal, F.; Legido, J. L. Magnetic silica nanoparticle cellular uptake and cytotoxicity regulated by electrostatic polyelectrolytes-DNA loading at their surface. ACS Nano 2012, 6, 747−759. (13) Lai, W. F. Cyclodextrins in non-viral gene delivery. Biomaterials 2014, 35, 401−11. (14) Rodik, R. V.; Klymchenko, A. S.; Jain, N.; Miroshnichenko, S. I.; Richert, L.; Kalchenko, V. I.; Mely, Y. Virus-sized DNA nanoparticles for gene delivery based on micelles of cationic calixarenes. Chemistry 2011, 17, 5526−38. (15) Alatorre-Meda, M.; Taboada, P.; Krajewska, B.; Willemeit, M.; Deml, A.; Klosel, R.; Rodriguez, J. R. DNA−poly(diallyldimethylammonium chloride) complexation and transfection efficiency. J. Phys. Chem. B 2010, 114, 9356−9366. (16) Yang, P.; Singh, J.; Wettig, S.; Foldvari, M.; Verrall, R. E.; Badea, I. Enhanced gene expression in epithelial cells transfected with amino acid-substituted gemini nanoparticles. Eur. J. Pharm. Biopharm. 2010, 75, 311−320. (17) Alatorre-Meda, M.; Taboada, P.; Sabin, J.; Krajewska, B.; Varela, L. M.; Rodriguez, J. R. DNA−chitosan complexation: A dynamic light scattering study. Colloids Surf., A 2009, 339, 145−152. (18) Felgner, J. H.; Gadek, T. R.; Holm, M.; Roman, R.; Chan, H. W.; Wenz, M.; Northrop, J. P.; Ringold, G. M.; Danielsen, M. Lipofection: A highly efficient, lipid-mediated DNA-transfection procedure. Proc. Natl. Acad. Sci. U.S.A. 1987, 84, 7413−7417. (19) Bhattacharya, S.; Bajaj, A. Advances in gene delivery through molecular design of cationic lipids. Chem. Commun. 2009, 4632−4656. (20) Masotti, A.; Mossa, G.; Cametti, C.; Ortaggi, G.; Bianco, A.; Del Grosso, N.; Malizia, D.; Esposito, C. Comparison of different commercially available cationic liposome−DNA lipoplexes: Parameters influencing toxicity and transfection efficiency. Colloids Surf., B 2009, 68, 136−144. (21) Wasungu, L.; Hoekstra, D. Cationic lipids, lipoplexes, and intracellular delivery of genes. J. Controlled Release 2006, 116, 255− 264. (22) Srinivasan, C.; Burgess, D. J. Optimization and characterization of anionic lipoplexes for gene delivery. J. Controlled Release 2009, 136, 62−70. (23) Caracciolo, G.; Amenitsch, H. Cationic liposome/DNA complexes: From structure to interactions with cellular membranes. Eur. Biophys. J. 2012, 41, 815−29. (24) Caracciolo, G.; Callipo, L.; De Sanctis, S. C.; Cavaliere, C.; Pozzi, D.; Lagana, A. Surface adsorption of protein corona controls the cell internalization mechanism of DC-Chol-DOPE/DNA lipoplexes in serum. Biochim. Biophys. Acta 2010, 1798, 536−543. (25) Roerdink, F.; Wassef, N. M.; Richardson, E. C.; Alving, C. R. Effects of negatively charged lipids on phagocytosis of liposomes opsonized by complement. Biochim. Biophys. Acta 1983, 734, 33−9. (26) Liang, H.; Harries, D.; Wong, G. C. L. Polymorphism of DNA− anionic liposome complexes reveals hierarchy of ion-mediated interactions. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 11173. (27) Patil, S. D.; Rhodes, D. G.; Burgess, D. J. Anionic liposomal delivery system for DNA transfection. AAPS J. 2004, 6. (28) Nadler, M.; Steiner, A.; Dvir, T.; Szekely, O.; Szekely, P.; Ginsburg, A.; Asor, R.; Resh, R.; Tamburu, C.; Peres, M.; Raviv, U. Following the structural changes during zinc-induced crystallization of charged membranes using time-resolved solution X-ray scattering. Soft Matter 2011, 7, 1512−1523.
(29) Petrache, H. I.; Tristram-Nagle, S.; Gawrisch, K.; Harries, D.; Parsegian, V. A.; Nagle, J. F. Structure and fluctuations of charged phosphatidylserine bilayers in the absence of salt. Biophys. J. 2004, 86, 1574−86. (30) Moshe, L.; Saper, G.; Szekely, O.; Linde, Y.; Gilon, C.; Harries, D.; Raviv, U. Modulating the structure and interactions of lipid− peptide complexes by varying membrane composition and solution conditions. Soft Matter 2013, 9, 7117−7126. (31) Szekely, O.; Steiner, A.; Szekely, P.; Amit, E.; Asor, R.; Tamburu, C.; Raviv, U. The structure of ions and zwitterionic lipids regulates the charge of dipolar membranes. Langmuir 2011, 27, 7419− 38. (32) Harries, D.; Podgornik, R.; Parsegian, V. A.; Mar-Or, E.; Andelman, D. Ion induced lamellar−lamellar phase transition in charged surfactant systems. J. Chem. Phys. 2006, 124, 224702. (33) Koltover, I.; Salditt, T.; Rädler, J. O.; Safinya, C. R. An inverted hexagonal phase of cationic liposome−DNA complexes related to DNA release and delivery. Science 1998, 281, 78−81. (34) Ewert, K.; Slack, N. L.; Ahmad, A.; Evans, H. M.; Lin, A. J.; Samuel, C. E.; Safinya, C. R. Cationic lipid−DNA complexes for gene therapy: Understanding the relationship between complex structure and gene delivery pathways at the molecular level. Curr. Med. Chem. 2004, 11, 133−149. (35) Muñoz-Ubeda, M.; Misra, S. K.; Barran-Berdon, A. L.; AicartRamos, C.; Sierra, M. B.; Biswas, J.; Kondaiah, P.; Junquera, E.; Bhattacharya, S.; Aicart, E. Why is less cationic lipid required to prepare lipoplexes from plasmid DNA than linear DNA in gene therapy? J. Am. Chem. Soc. 2011, 133, 18014−18017. (36) Patil, S. D.; Rhodes, D. G.; Burgess, D. J. Biophysical characterization of anionic lipoplexes. Biochim. Biophys. Acta 2005, 1711, 1−11. (37) Martín-Molina, A.; Luque-Caballero, G.; Faraudo, J.; QuesadaPérez, M.; Maldonado-Valderrama, J. Adsorption of DNA onto anionic lipid surfaces. Adv. Colloid Interface Sci. 2014, 206, 172−185. (38) Luque-Caballero, G.; Martin-Molina, A.; Sanchez-Trevino, A. Y.; Rodriguez-Valverde, M. A.; Cabrerizo-Vilchez, M. A.; MaldonadoValderrama, J. Using AFM to probe the complexation of DNA with anionic lipids mediated by Ca2+: The role of surface pressure. Soft Matter 2014, 10, 2805−2815. (39) Muñoz-Ubeda, M.; Misra, S. K.; Barran-Berdon, A. L.; Data, S.; Aicart-Ramos, C.; Castro-Hartmann, P.; Kondaiah, P.; Junquera, E.; Bhattacharya, S.; Aicart, E. How does the spacer length of cationic gemini lipids influence the lipoplex formation with plasmid DNA? Physicochemical and biochemical characterizations and their relevance in gene therapy. Biomacromolecules 2012, 13, 3926−3937. (40) Misra, S. K.; Munoz-Ubeda, M.; Datta, S.; Barran-Berdon, A. L.; Aicart-Ramos, C.; Castro-Hartmann, P.; Kondaiah, P.; Junquera, E.; Bhattacharya, S.; Aicart, E. Effects of a delocalizable cation on the headgroup of gemini lipids on the lipoplex-type nanoaggregates directly formed from plasmid DNA. Biomacromolecules 2013, 14, 3951−63. (41) Muñoz-Ubeda, M.; Rodriguez-Pulido, A.; Nogales, A.; MartinMolina, A.; Aicart, E.; Junquera, E. Effect of lipid composition on the structure and theoretical phase diagrams of DC-Chol/DOPE-DNA lipoplexes. Biomacromolecules 2010, 11, 3332−3340. (42) Rodriguez-Pulido, A.; Martin-Molina, A.; Rodriguez-Beas, C.; Llorca, O.; Aicart, E.; Junquera, E. A theoretical and experimental approach to the compaction process of DNA by dioctadecyldimethylammonium bromide/zwitterionic mixed liposomes. J. Phys. Chem. B 2009, 113, 15648−15661. (43) Hansen, M. B.; Nielsen, S. E.; Berg, K. Re-examination and further development of a precise and rapid dye method for measuring cell-growth cell kill. J. Immunol. Methods 1989, 119, 203−210. (44) Bajaj, A.; Kondiah, P.; Bhattacharya, S. Design, synthesis, and in vitro gene delivery efficacies of novel cholesterol-based gemini cationic lipids and their serum compatibility: A structure−activity investigation. J. Med. Chem. 2007, 50, 2432−2442. (45) Rodriguez-Pulido, A.; Aicart, E.; Llorca, O.; Junquera, E. Compaction process of calf thymus DNA by mixed cationic− 11712
dx.doi.org/10.1021/la502823z | Langmuir 2014, 30, 11704−11713
Langmuir
Article
zwitterionic liposomes: A physicochemical study. J. Phys. Chem. B 2008, 112, 2187−2197. (46) Barran-Berdon, A. L.; Muñoz-Ubeda, M.; Aicart-Ramos, C.; Perez, L.; Infante, M. R.; Castro-Hartmann, P.; Martin-Molina, A.; Aicart, E.; Junquera, E. Ribbon-type and cluster-type lipoplexes constituted by a chiral lysine based cationic gemini lipid and plasmid DNA. Soft Matter 2012, 8, 7368−7380. (47) Koltover, I.; Salditt, T.; Safinya, C. R. Phase diagram, stability, and overcharging of lamellar cationic lipid−DNA self-assembled complexes. Biophys. J. 1999, 77, 915−924. (48) Safinya, C. R. Structures of lipid−DNA complexes: Supramolecular assembly and gene delivery. Curr. Opin. Struct. Biol. 2001, 11, 440−448. (49) Muñoz-Ubeda, M.; Rodriguez-Pulido, A.; Nogales, A.; Llorca, O.; Quesada-Perez, M.; Martin-Molina, A.; Aicart, E.; Junquera, E. Gene vectors based on DOEPC/DOPE mixed cationic liposomes: A physicochemical study. Soft Matter 2011, 7, 5991−6004. (50) Rattan, L. Encyclopedia of Soil Science, 2nd ed.; CRC Press: Boca Raton, FL, 2006; Vol. 1, p 198.
11713
dx.doi.org/10.1021/la502823z | Langmuir 2014, 30, 11704−11713