The effect of ionic strength on DNA condensation by cationic liposomes prepared as a mixture of ethane-1,2-diylbis(dodecyldimethylammonium bromide) (C2GS12) and dioleoylphosphatidylcholine (DOPC) was studied using fluorescence spectroscopy. The DNA condensation followed by changes in emission intensity of ethidium bromide shows a strong dependence on the ionic strength of the solution. At physiologically relevant ionic strength (0.15 mol/l NaCl), the amount of DNA condensed between lipid bilayers is approximately 40% lower compared to 0.005 mol/l NaCl. The structure of formed complexes was studied using small angle X-ray diffraction (SAXD). DNAC2GS12DOPC complexes form a condensed lamellar phase organisation, which is partially disrupted by the increase of ionic strength. Both the lamellar repeat distance and DNADNA distance show dependence on C2GS12/DOPC molar ratio, temperature and also on ionic strength. We found that the method of preparation significantly affects both the quality of organisation and the structural parameters of complexes as discussed in the paper. Metódou fluorescencnej spektroskopie sme studovali vplyv iónovej sily roztoku na DNA kondenzáciu v prítomnosti katiónových lipozómov pripravených zo zmesi etán-1,2-diylbis(dodecyldimetylamónium bromidu) (C2GS12) a dioleoylfosfatidylcholínu (DOPC). Kondenzácia DNA sledovaná prostredníctvom zmien v intenzite emisného ziarenia etídium bromidu vykazuje silnú závislos na iónovej sile roztoku. Pri fyziologicky relevantnej iónovej sile (0,15 mol/l NaCl) je mnozstvo DNA kondenzovanej medzi lipidovými dvojvrstvami o viac nez 40% nizsie nez v prostredí 0,005mol/l NaCl. Struktúra vzniknutých komplexov bola studovaná pomocou malouhlovej difrakcie RTG ziarenia (SAXD). DNAC2GS12DOPC komplexy vytvárajú kondenzovanú lamelárnu fázu, ktorej usporiadanie je ciastocne narusené pri zvýsenej iónovej sile. V závislosti od mólového pomeru C2GS12/DOPE, teploty a iónovej sily sme pozorovali zmeny periódy lamely ako aj vzdialenosti medzi DNA vláknami. Experimentálne výsledky sú diskutované vzhadom na spôsob prípravy komplexov. DNA, Gemini surfactants, dioleoylphosphatidylcholine, Fluorescence spectroscopy, Small angle X-ray diffraction. DNA, Gemini tenzidy, dioleoylfosfatidylcholín, fluorescencná spektroskopia, malouhlová RTG difrakcia Slovak abstract Keywords Kúcové slová: INTRODUCTION For the efficient and safe introduction of DNA into the cell nucleus (transfection) a suitable vector is needed. The most effective transfection agents are viruses (Robbins & Ghivizzani, 1998); however, there are serious safety concerns (Verma, 2000). As a result, the search for alternative non-viral vectors has intensified. One of the most promising groups of non-viral vectors are cationic liposomes. Since their first use by Felgner et al., (1987) for gene delivery to somatic cells, a wide variety of * email@example.com © Acta Facultatis Pharmaceuticae Universitatis Comenianae different liposomal transfection vectors has been developed. Cationic liposomes consisting of cationic lipid (Caracciolo et al., 2007; Mochizuki et al., 2013; Wasungu & Hoekstra, 2006) or cationic surfactant (Badea et al., 2005; Bombelli et al., 2005; Donkuru et al., 2012; Kirby et al., 2003) with or without a helper lipid with neutral net charge has been widely studied as potential vectors for in vitro and in vivo gene delivery. Cationic lipids or surfactants form complexes with the DNA polyanion what results in DNA condensation. The condensed DNA is partially protected from degradation by enzymes (Rolland, 1998) and can be transferred through endocytosis more effectively (Kirby et al., 2003). Helper lipid modifies the colloidal and structural properties of the complexes and supports their transport through cell membranes (HirschLerner et al., 2005). Relations between the structure, physicochemical properties and transfection efficiency of liposomal vectors are still not fully understood. Gemini surfactants (GS) were revealed as a promising group of cationic additive surfactants for gene delivery vectors by Kirby et al, (2003). GS consists of two hydrophobic chains and two ionic (polar) groups linked by a spacer (Menger & Keiper, 2000). GS as non-viral vectors for gene therapy have become a major focus of research because of the unique solution properties imparted by their molecular structure. Their critical micelle concentration is, in general, an order of magnitude or more lower and the surface activity is an order of magnitude greater than it is for comparable single-chain surfactants, but at comparable or lower levels of toxicity (Menger & Keiper, 2000; Wettig et al., 2008). One of the most studied type of GS are alkane-,-diyl-bis(alkyldimethylammonium bromide)s (CnGSm, where n is the number of spacer carbons and m is the number of carbons in the alkyl chains). CnGSm were found to increase the efficiency of DNA transfer into bacterial cells (Horniak et al., 1989). In combination with helper lipid dioleoylphosphatidylethanolamine (DOPE) they have shown good transfection efficiency both in vitro (Cardoso et al., 2014; Foldvari et al., 2006; Muñoz-Úbeda et al., 2012) and in vivo (Badea et al., 2005). The physicochemical properties of GSDNA complexes or without helper lipids were also intensively investigated (García et al., 2014; Grueso et al., 2013; Pietralik et al., 2013; Uhríková et al., 2005a). Our group has studied systematically the structure and polymorphic behaviour of complexes DNACnGSmneutral phospholipid (Pullmannová et al., 2008, 2012a, 2012b; Uhríková et al., 2002, 2005b, 2007). Our experiments showed that besides the composition of complexes their microstructure is influenced also by other factors such as the ionic strength of the aqueous medium or the method of preparation (Pullmannová et al., 2012b). This work extends our study focused on the effect of the ionic strength on DNA condensation and the structure of the complexes formed in respect to the used method of preparation. The complexes were prepared by direct mixing of dioleoylphosphatidylcholine (DOPC) unilamellar liposomes with a solution of C2GS12 and DNA in the aqueous medium at two concentrations of NaCl, 0.005 and 0.15 mol/l, respectively. Fluorescence spectroscopy was employed to follow DNA condensation by cationic liposomes. The structure of complexes was studied using small angle X-ray diffraction (SAXD). The results are compared and discussed with respect to our previous study (Pullmannová et al., 2012b) where different methods of complexes preparation were used. MATERIALS AND METHODS Materials Highly polymerised calf thymus DNA (sodium salt) Type I (average Mr of nucleotides = 308) was purchased from Sigma Chemicals, St. Louis, Missouri, USA; ethidium bromide (EtBr) was purchased from Merck, Germany and neutral phospholipid dioleoylphosphatidylcholine (DOPC) was purchased from Avanti Polar Lipids, Alabaster, Alabama, USA. Ethane-1,2diyl-bis(dodecyldimethylammonium bromide), C2CS12, was prepared as described in (Imam et al., 1983) and purified by manifold crystallisation from a mixture of acetone and methanol. The NaCl of analytical purity was purchased from Lachema, Brno, Czech Republic. Preparation of DNA solutions DNA was dissolved at concentration 5 mg/ml in 0.005 mol/l NaCl or 0.15 mol/l NaCl, respectively. The precise value of DNA concentration was determined spectrophotometrically (Hewlett Packard 8452A Diode array spectrophotometer), according to cDNA = A260. 47×10-6 [g/ml], where A260 is the absorbance at wavelength = 260 nm. The concentration of DNA is referred as molar concentration of DNA bases. The purity of DNA was checked by measuring the absorbance A at = 260 and 280 nm. We obtained the value of A260 /A280 = 1.81. Preparation of cationic liposomes Dispersions of DOPC multilamellar liposomes were prepared by hydration of dry lipid in 0.005 mol/l and 0.15 mol/l NaCl solutions and their homogenisation by vortexing. DOPC unilamellar liposomes were prepared by extrusion of the lipid dispersion through polycarbonate filters with pores of diameter 100 nm. The DOPC unilamellar liposomes were mixed with the solution of C2GS12 at various molar ratios C2GS12/DOPC and stored at 4°C for 24 hours. Fluorescence experiments The samples of DNAC2GS12DOPC complexes for fluorescence experiments were prepared in the range of C2GS12/ DNA = 02 mol/base mol for both studied NaCl concentrations. DNA solutions were mixed with fluorescence probe EtBr at DNA/EtBr = 12 base mol/mol. After 5 minutes, mixtures of unilamellar liposomes at wished molar ratio C2GS12/DOPE were added into the samples and the volume completed to 3000 l by appropriate NaCl solution. The florescence of samples was measured 30 minutes after the preparation using the Fluoromax-4 spectrofluorometer (Jobin Yvon, France). The emission fluorescence intensity of EtBr was measured at em = 596 nm, using exciting wavelength ex = 520 nm. The emission intensity of each sample was corrected for the background fluorescence of EtBr in the absence of DNA and then normalised to the EtBr fluorescence of sample containing DNA without any C2GS12DOPC liposomes (C2GS12/ DNA = 0). Small-angle X-ray diffraction experiments The samples of DNAC2GS12DOPC complexes for SAXD experiments were prepared in 0.005 and 0.15 mol/l NaCl solutions by mixing the dispersions of C2GS12DOPC unilamellar liposomes prepared at the range of molar ratios C2GS12/DOPC = 0.1-0.5 and the DNA solution. The samples were prepared at theoretical isoelectric point (DNA/C2GS12 = 2 base mol/mol). SAXD experiments were performed at the soft condensed matter beamline A2 at HASYLAB at the Deutsches Elektronen Synchrotron (DESY) in Hamburg (Germany), using a monochromatic radiation of wavelength = 0.15 nm. The evacuated double-focusing camera was equipped with linear delay line readout detector. The samples were measured at 20°C and equilibrated at selected temperature 5 min before measurement. Temperature scans were performed at a scan rate 1°C/ min and the diffractograms were recorded for 10 s every minute. The data were normalised against the incident beam intensity using the signal intensity measured in the ionisation chamber. The SAXD detector was calibrated using rat tail collagen (Roveri et al., 1980). Each diffraction peak of SAXD region was fitted with a Lorentzian function above a linear background. negative charge of DNA phosphate groups leads to compaction and condensation of the DNA molecules and their condensation (Eastman et al., 1997). The condensation of DNA leads to displacement of intercalated EtBr, that presents itself as the decrease of fluorescence intensity (Eastman et al., 1997; Izumrudov et al., 2002; Wiethoff et al., 2003). Fig. 1 shows the dependence of the emission intensity of EtBr on C2GS12/DNA molar ratio at two ionic strength of solutions, in 0.005 mol/l and 0.15 mol/l of NaCl, respectively. The concentrations of DNA (3 mol/l), EtBr (0.25 mol/l) and DOPC (12 mol/l) were kept constant, while the concentration of C2GS12 varied depending on C2GS12/DNA molar ratio. The dependence of normalised fluorescence emission intensity of EtBr on the CnGS12/DNA molar ratio has a sigmoidal course and the minimal intensity reaches at C2GS12/DNA 1.2 mol/ base mol in both used NaCl solutions. Above this molar ratio the decrease of emission intensity is insignificant. The main difference between the two used NaCl solutions is in the observed minimum of normalised emission intensity of EtBr. At low ionic strength (0.005 mol/l NaCl), the minimum is achieved at 21.89 ± 0.01%, while at high ionic strength (0.15 mol/l NaCl), we determined 62.71 ± 0.01% of total intensity. The structure of DNAC2GS12DOPC complexes DNA condensation We studied the structure of DNAC2GS12DOPC complexes DNA condensation was indicated by a decrease of emission hydrated by 0.005 mol/l and 0.15 mol/l NaCl solutions as a intensity of fluorescence probe EtBr. The free molecules of function of C2GS12/DOPC molar ratio. All samples were preEtBr in a solution follow a nonradiative decay pathway that pared at the theoretical isoelectric point based on nominal involves donation of an amino group proton to solvent. When charges of each species, corresponding to the molar ratio EtBr is intercalated into DNA, the ethidium cation is isolated C2GS12/DNA = 0.5 mol/base mol. Fully hydrated DOPC was from the solvent and the proton transfer pathway between measured as a control sample. At 20°C DOPC forms a liquid EtBr and the solvent is blocked. This leads to increase of fluo- crystalline lamellar L phase (Wiener & White, 1992). The difrescence intensity about 20-fold (Izumrudov et al., 2002). DNA fractogram of DOPC at 20°C (Fig. 2) shows two peaks, L(1) and interacts with cationic surfactants or cationic liposomes due L(2), related to the first and the second order of the lamellar to electrostatic attraction between Fig. 1 Dependences of normalised fluorescence intensity Inorm distance d = S1 = 6.13 ± 0.01 cationic agent and nega- phase. We determined the repeat of DNAEtBrC2GS12DOPC complexes on C2GS12/DNA nm, where NaCl concentration 0.005 mol/l () and 0.15 mol/l tively charged phosphate groups of DNA. Neutralisation of the molar ratio ats1 is the position of maximum of the first order (). RESULTS Fig. 2 Diffractogram of complexes on ependences of normalised fluorescence intensity Inorm of DNAEtBrC2GS12DOPCfully hydrated DOPC at 20°C. Intensity is in logarithmic scale. Fig. 1 Dependences of normalised fluorescence intensity Inorm of Fig. 2 Diffractogram of fully hydrated DOPC at 20°C. Intensity is C2GS12/DNA molar ratio at NaCl concentration 0.005 mol/l () and 0.15 mol/l (). DNAEtBrC2GS12DOPC complexes on C2GS12/DNA molar in logarithmic scale. ratio at NaCl concentration 0.005 mol/l () and 0.15 mol/l (). peak. The repeat distance d includes the thickness of the phospholipid bilayer dL and the water layer thickness, dW , thus d = dL + dW. In the complexes dw contains a monolayer of hydrated DNA strands. In our experimental protocol, we mixed DOPC unilamellar liposomes with a solution of C2GS12 to get the wished C2GS12/ DOPC molar ratio. Due to the surfactantlipid interaction, the hydrophobic alkyl substituents of C2GS12 molecules intercalate between the lipid acyl chains of DOPC bilayer. Polar headgroups of C2GS12 molecules create positively charged surface of the bilayer, and the C2GS12/DOPC molar ratio determines the surface charge density. Cationic C2GS12DOPC liposomes interact with DNA and form complexes. Fig. 3 shows diffractograms of DNAC2GS12DOPC complexes prepared at molar ratios 0.1 C2GS12/DOPC 0.5 hydrated by 0.005 mol/l NaCl and measured at 20°C. Diffractograms are typical for a condensed lamellar phase (LC) with DNA strands regularly ordered between the lipid bilayers (Lasic et al., 1997; Rädler et al., 1997). We observed two peaks characteristic for lipid bilayer stacking and a small broad peak related to a regular DNA packing. The structural parameters, the repeat distance dLC of LC phase and DNADNA distance dDNA = 1/sDNA, of DNAC2GS12DOPC complexes hydrated by 0.005 mol/l NaCl are shown in Fig. 4. Incorporation of DNA between the lipid bilayers results in the increase d of LC phase compared to the d of pure DOPC (6.13 nm). We observed a decrease of the repeat distance of LC phase from dLC = 6.63 ± 0.01 nm at C2GS12/DOPC = 0.1 mol/ mol to dLC = 6.10 ± 0.01 nm at C2GS12/DOPC = 0.5 mol/mol. This decrease of d is typical for a lamellar lipid system with incorporated amphiphilic molecules with shorter alkyl chains compared to the length of acyl chains of the lipid. This mismatch results in a higher incidence of gauche-conformation of the lipid chains, leads to a lateral expansion of phospholipid bilayer and the decrease in its thickness (Balgavý & Devínsky, 1996; King & Marsh, 1986). The C2GS12/DOPC molar ratio also influences the arrangement of the DNA strands. The higher amount of C2GS12 in the lipid mixture increases the surface charge density of liposomes. For complexes prepared at isoelectric point, the surface charge density is considered a key parameter influencing the dDNA (Koltover et al., 1999). The repeat distance dDNA as a function of C2GS12/DOPC molar ratio shows a decrease from dDNA = 4.54 ± 0.02 nm (C2GS12/DOPC = 0.2 mol/mol) to dDNA = 3.46 ± 0.03 nm (C2GS12/DOPC = 0.4 mol/ mol). For complexes formed at molar ratio C2GS12/DOPC = 0.1 and 0.5, respectively, the DNA peak is not observed due to its overlap with peaks of lamellar phase. We have studied thermally induced changes of the structure of DNAC2GS12DOPC complexes at C2GS12/DOPE = 0.25 mol/mol in 0.005 mol/l NaCl. Fig. 5 shows the dependence of repeat distances d and dDNA of the complexes in the range 2060°C. For comparison, Fig. 5 shows also the temperature dependence of the DOPC repeat distance. With increasing temperature, we observe a small systematic decrease in d of DOPC due to thermally induced lateral expansion of the Fig. 3 Diffractograms of bilayer. Similarly to the pure lipid, we 0.005 mol/l NaCl at of DNAC2GS12DOPC complexes in found the decrease different C2GS1 C the repeat distance 20°C. (Intensities are in logarithmic scale). molar ratios measured atof L phase. dLC decreases from 6.47 to 6.25 nm when the temperature increased gradually from 20 to Fig. 4 Dependence of repeat distance dLC of repeatDNA repeat() and DNA repeat on C2GS12/DOP Fig. 4 Dependence () and distance dLC distance dDNA () ratio in 0.005 mol/l NaCl at 20°C. The dashed line represents the repeat distance of pure DO distance dDNA () on C2GS12/DOPC molar ratio in 0.005 mol/l 3 Diffractograms ofFig. 3 Diffractograms of DNAC2GS12DOPC complexes in at different C2GS12/DOPC DNAC2GS12DOPC complexes in 0.005 mol/l NaCl 0.005 mol/l NaCl at at 20°C. (Intensities are in logarithmic NaCl molar ratios measured different C2GS12/DOPC molar ratios mea- scale). at 20°C. The dashed line represents the repeat distance of sured at 20°C. (Intensities are in logarithmic scale). pure DOPC. 60°C. DNA follows thermally induced lateral expansion of the membrane. We determined the increase of dDNA from 4.09 ± 0.03 nm at 20°C to 4.23 ± 0.03 nm at 60°C. Diffractograms of complexes with the same composition, however, prepared in 0.15 mol/l of NaCl (Fig. 6) are different in comparison to those in Fig. 3. In addition to the first and second peaks of LC phase, we observe smaller and broader peaks of another lamellar phase (L2). The overlap of the peaks of LC and L2 phase indicates their close periodicities. The proportion of the LC and L2 phases change with the C2GS12/DOPC molar ratio: at C2GS12/DOPC = 0.1 mol/mol, the L2 phase is only present structure. As the fraction of C2GS12 increases, the portion of L2 phase decreases and the LC phase becomes dominant. The difference in the shapes of the peaks indicates the smaller positional order of L2 phase compared to LC. The higher ionic strength influences also the ordering of the DNA. The intensities of the DNA peak are lower compared to those observed in Fig. 3 for complexes prepared in 0.005 mol/l NaCl. This may indicate that at higher ionic strength, a lower fraction of the DNA is bound in the complexes or that the strands are less ordered. Structural parameters of DNAC2GS12DOPC complexes prepared in 0.15 mol/l NaCl are plotted as a function of C2GS12/ DOPC molar ratio in Fig. 7. The increase in C2GS12/DOPE molar ratio leads to similar changes in the structural parameters of LC phase as observed for complexes prepared at lower ionic strength. We observe a decrease of dLC from 6.75 ± 0.01 nm (at C2GS12/DOPC = 0.15 mol/mol) to dLC = 6.27 ± 0.01 nm (at C2GS12/DOPC = 0.5 mol/mol) and the decrease of dDNA 4.06 ± 0.04 nm (C2GS12/DOPC = 0.15 mol/mol) to 3.21 ± 0.02 (C2GS12/DOPC = 0.35 mol/mol). Thus at high ionic strength, the repeat distance of LC phase increased slightly, while dDNA decreased when compared to the structural parameters of complexes hydrated by 0.005 mol/l NaCl. The repeat distance of L2 phase (d2) is slightly lower compared to LC phase. At C2GS12/DOPC = 0.1 mol/mol, the only observed phase is L2 with, d2 = 6.59 ± 0.01 nm. Changes in the C2GS12/DOPC molar ratio are reflected also in d2, however, in less extensive manner compared to dLC. As such, the values of dLC and d2 converge, and finally at C2GS12/DOPC = 0.5 mol/mol, they are almost equal (dLC d2). Repeat distances of LC and L2 phase of complexes prepared at molar ratio C2GS12/DOPC = 0.25 in 0.15 mol/l NaCl as a function of temperature are shown in Fig. 8. The repeat distances of both lamellar phases (LC and L2) consecutively decrease in the temperature range 2042°C. Above 42°C, the values of lattice parameters of LC and L2 phases are similar and increase slightly with the temperature. Above 55°C, LC phase was the only phase observed in the diffractograms. Fig. 5 Dependences of the repeat distance dFig. 6of L phase and of DNAC2GS12DOPC complexes at various molar ratios C2GS12/DOPC in () Diffractograms LC C g. 5 Dependences of DNA repeatdistance dd () of Ltemperature at molar ratio distance dDNA () on of DNAC2GS12DOPC complexes at varithe the repeat distance LC () on phase and the DNA repeat Fig. 6 Diffractograms (Intensities are in logarithmic scale). mol/l NaCl at 20°C DNA mperature at molar ratio C2GS12/DOPC = 0.25 and the repeat distance d () of pure DOPC in 0.005 C2GS12/DOPC = 0.25 and the repeat distance d () of pure DOPC ous molar ratios C2GS12/DOPC in 0.15 mol/l NaCl at 20°C (Intenmol/l NaCl. in 0.005 mol/l NaCl. sities are in logarithmic scale). In 0.005 mol/l NaCl, the complexes form typical condensed lamellar phase (LC). When we increased concentration of NaCl Fluorescence spectroscopy experiments show clearly the in- to physiologically relevant values (0.15 mol/l), we observed fluence of the ionic strength on the condensation of DNA by a coexistence of LC and a second lamellar phase L2. The porC2GS12DOPC liposomes. While in 0.005 mol/l NaCl, the ob- tion of both phases, L2 and LC, is dependent on the C2GS12/ served minimum of EtBr emission intensity is at approximately DOPC molar ratio and temperature. At low content of C2GS12, 22%, in 0.15 mol/l NaCl the minimum is at approximately 63% the dominating phase is L2 while at C2GS12/DOPC = 0.5 mol/ of the total emission intensity of DNA without any cationic li- mol, the major structure is LC phase. The closeness of lattice posomes. This minimum was achieved at the same C2GS12/ parameters of both phases, dLC ~ d2 , suggests that the L2 phase DNA molar ratios (1.2 mol/base mol) for both studied ionic is most probably a condensed lamellar phase too, formed at strengths. This suggests that regardless of the ionic strength slightly different C2GS12/DOPC molar ratios due to non-hoof the solution, the formation of DNAC2GS12DOPC com- mogeneous mixing of the two solutions (DOPC and C2GS12) plexes ends at the same point. However, the experiments at the cationic liposomes preparation. revealed a big difference in the minimum of EtBr emission in- In our previous work Pullmannová et al. (2012b), the structure tensity between the two ionic strengths used. This indicates of DNACnGS12DOPC complexes (n = 2-4) have shown difa decrease in the efficiency of C2GS12/DOPC liposomes for ferences when complexes were prepared by two different proDNA condensation at high ionic strength. The lower ability of cedures: DOPC was mixed with CnGS12 in an organic solvent cationic liposomes to condense DNA at higher ionic strength and then dried under a stream of gaseous nitrogen followed is caused by the screening effect of small ions present in solu- by vacuum. The dry lipid films were hydrated by NaCl solution tion on the electrostatic interaction between DNA polyanion at different concentration (0.0050.200 mol/l), and multilaand cationic liposomes (Jing et al., 2004). Our results are in mellar liposomes were prepared. DNA solution was added to good agreement with the work of Eastman et al., (1997) where the dispersion of liposomes by two different methods, either it was observed a decrease in efficiency of DNA condensation by drop-by-drop addition or by addition of all appropriate by cationic lipid 1,2-dimyristyloxypropyl-3-dimethylhydroxy- amount of DNA in one step. At low ionic strength, all complexethyl ammonium bromide at higher ionic strength. Increase es have shown LC phase. At high ionic strength, the structure of the NaCl concentration to 1.5 mol/l lead to total suppres- of complexes differed depending on the method used to DNA sion of DNA condensation by cationic liposomes. Eastman et addition. While complexes prepared by one-step addition of al., (1997) assumed that the high ionic strength leads to for- DNA have shown LC phase; in the complexes prepared by mation of complexes that even at the excess of cationic lipid step-by-step addition of DNA, the phase separation was obcontains uncompacted DNA without full compensation of served. Similar to our system, an additional lamellar phase has its anionic charges. This partially uncompacted DNA would been detected, particularlyCat low C2GS12/DOPC molar ratios Fig. 7 Dependences of the repeat distance of L phase (dLC ()) and L2 phase (d2 ()) and DNA therefore still be accessible to EtBr even when it is bound in and high ionic strength of solution. However, contrary to our distance dDNA () on C2GS12/ DOPC molar ratio in 0.15 mol/l NaCl at 20°C. The dashed line re complexes with cationic liposomes. samples, SAXD has shown only a minor volume fraction of this the repeat distance of pure DOPC. SAXD measurements revealed that the ionic strength influ- lamellar phase through all CnGS12/DOPC molar ratios, and ences also the structure of DNA-C2GS12-DOPC complexes. the LC phase was the predominant structure. Generally, DNA DISCUSSION Fig. 8 DNA repeat 7 Dependences of Fig. 7 Dependences of theLC phase (dLC of LC and L2 phase (d2 of the repeat distances of Lrepeat distances of and L2 phase()) ()) on temper the repeat distance of repeat distance ()) phase (dLC ()) Fig. 8 Dependences ()) and Dependences of the C phase (dLC ()) LC phase (dLC (d2 and 2 phase (d2 ()) at temperature in 0.15 mol/l NaCl at molar and L2 DOPC ()) and DNA repeat mol/l NaCl at on 0.15 mol/l NaCl on nce dDNA () on C2GS12/phase (d2 molar ratio in 0.15distance dDNA ()20°C. The dashedLline represents molar ratio C2GS12/DOPC = 0.25. ratio C2GS12/DOPC = 0.25. C2GS12/ DOPC molar ratio in 0.15 mol/l NaCl at 20°C. the repeat distance of pure DOPC. The dashed line represents the repeat distance of pure DOPC. CnGS12DOPC complexes prepared by the method used in Pullmannová et al. (2012b) has shown better long-range ordered structure, manifested by higher intensities and sharper peaks on diffractograms. Thus preparation of complexes by direct mixing of components dispersed in aqueous solutions results in a formation of less organised structures. At high ionic strength, our method of the complex preparation also supports phase separation. We detected much higher portion of the L2 phase compared to Pullmannová et al. (2012b). The revealed dissimilarities result from a different origin of the L2 phase. In our samples, the L2 phase was formed due to nonuniform incorporation of C2GS12 into DOPC bilayers during the components mixing. In the study of Pullmannová et al. (2012b), the phase separation is caused by drop-by-drop addition of DNA when in the first stages positively overcharged complexes were formed. In this stage of preparation, domains of lipid enriched with C2GS12 are formed (due to lateral diffusion of surfactants molecules in the lipid bilayer) on interaction with DNA. Both phases, the LC and additional lamellar phase, are in coexistence within one structure and they cannot be separated macroscopically as proven by Pullmannová et al. (2012b). Similar microscopic phase separation induced by the high ionic strength was observed also in DNA complexes with mixture of cationic lipid dioleoyl trimethylammonium propane and DOPC (Koltover et al., 1999), with mixture of C4GS12 and dilauroylphosphatidylcholine (Uhríková et al., 2004) or with zwitterionic phospholipids in the presence of divalent cations (McManus et al., 2003; Uhríková et al., 2005a). Our observation of the thermally induced changes in the structure of the complexes also supports this assumption. With increasing temperature, the L2 phase gradually merges with the LC phase. This is enabled by a lateral mixing of domains in the lipid bilayer driven by the increased kinetic energy of heated molecules. Structural parameters of DNAC2GS12DOPC complexes reported by Pullmannová et al. (2012b) were determined for complexes prepared by step-by-step DNA addition at C2GS12/ DOPC = 0.2 mol/mol through a range 0.0050.200 mol/l of NaCl concentrations. Comparing structural parameters of complexes at the same composition but prepared by different methods, surprisingly, differences are small and more apparent at low ionic strength. In 0.005 mol/l NaCl, the dLC of LC phase is reduced by approximately 0.4 nm and the dDNA by approximately 0.5 nm using our method of preparation, while in 0.15 mol/l NaCl, we detected differences smaller than 0.2 nm in dLC , and ~ 0.3 nm in dDNA. The importance of the systematic study of changes in polymorphic behaviour of DNAcationic liposomes complexes caused by the used preparation method is underlined by a recent work of Cardoso et al. (2014), which suggests that the way of preparation strongly affects also the transfection efficiency. Authors found that complexes DNA with C2GS12 or C2GS16 and mixture of DOPE and cholesterol has shown higher transfection efficiency when they were prepared by direct mixing of DNA, GS and helper lipid liposomes in the aqueous medium compared to the delivery vectors prepared by mixing of GS, DOPE and cholesterol in an organic solvent and, consecutively, DNA adding to GS/DOPE/cholesterol liposomes. Authors assumed that the difference in transfection efficiency could result from differences in structures of formed complexes. The structure of these complexes was not studied in the referred work. CONCLUSIONS We found that the ability of C2GS12-DOPC liposomes to condense DNA at physiologically relevant ionic strength is significantly reduced. While at low ionic strength (0.005 mol/l NaCl), the decrease in EtBr emission intensity indicates almost 80% of the total DNA condensed by the cationic lipid bilayer, less than 40% of DNA was bound in complexes prepared at 0.15 mol/l NaCl. Experiments revealed that the condensation process ends at the same ratio, C2GS12/DNA 1.2 mol/base mol, at both studied ionic strengths in spite of the difference in the condensation efficiency. Our results confirmed that the method of preparation of DNAcationic liposomes complexes affects their structure significantly. DNAC2GS12DOPC complexes prepared by direct mixing of DOPC unilamellar liposomes with a solution of C2GS12, and consecutively with DNA solution form a condensed lamellar phase showing a small shift in its structural parameters when compared to the complexes prepared by mixing the lipid components in an organic solvent and applying hydration method as used in Pullmannová et al., (2012b). We revealed significant structural differences when complexes were formed at high ionic strength. The complexes hydrated by 0.15 mol/l NaCl have shown a lower degree of long-range order and two-phase coexistence in a large range of C2GS12 DOPC molar ratios and temperature. The transfection efficiency of delivery vectors depends on their structure as indicate recently published experiments of Cardoso et al. (2014). During our many years of research in relation to the structure and methods of DNAcationic liposomes complexes preparation, we observed that a condensed lamellar phase is formed when the lipid components are mixed in an organic solvent and dried under vacuum. Consecutively, the dry lipid film is hydrated by aqueous medium and mixed with a solution of DNA in one step. Other ways of the complexes preparation frequently result in a coexistence of two or more phases in their structure, particularly when complexes are prepared at low-surface charge density and in high-ionic strength of aqueous medium. ACKNOWLEDGEMENT The research leading to these results has received funding from the European Community's Seventh Framework Programme (FP7/20072013) under grant agreement n° 226716 (HASYLAB project II-20100372 EC), by the JINR project 04-41069-2009/2014 and grants APVV 0212-10; APVV 0516-12; VEGA 1/1224/12 and FaF UK/27/2014.
Acta Facultatis Pharmaceuticae Universitatis Comenianae – de Gruyter
Published: Dec 30, 2014