Transfection of plasmid DNA by nanocarriers containing a gemini cationic lipid with an aromatic spacer or its monomeric counterpart

Colloids and Surfaces B: Biointerfaces 161 (2018) 519–527

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb

Full Length Article

Transfection of plasmid DNA by nanocarriers containing a gemini cationic lipid with an aromatic spacer or its monomeric counterpart María Martínez-Negro a , Ana L. Barrán-Berdón a , Clara Aicart-Ramos b , María L. Moyá c , Conchita Tros de Ilarduya d , Emilio Aicart a,∗ , Elena Junquera a a

Grupo de Química Coloidal y Supramolecular, Departamento de Química Física I, Universidad Complutense de Madrid, 28040 Madrid, Spain Dpto. Bioquímica y Biología Molecular I, Facultad de Ciencias Químicas, Universidad Complutense de Madrid, 28040 Madrid, Spain c Grupo de Química Coloidal y Catálisis Micelar, Departamento de Química Física I, Facultad de Farmacia, Universidad de Sevilla, 41012 Sevilla, Spain d Departamento de Farmacia y Tecnología Farmacéutica, Universidad de Navarra, E-31008 Pamplona, Spain b

a r t i c l e

i n f o

Article history: Received 16 July 2017 Received in revised form 16 October 2017 Accepted 7 November 2017 Available online 8 November 2017 Keywords: Complexation and protection of plasmid DNA Gemini cationic lipid with aromatic spacer Lipoplexes Structure and morphology Gene delivery

a b s t r a c t This study performed a biophysical characterization (electrochemistry, structure and morphology) and assessment of the biological activity and cell biocompatibility of GCL/DOPE-pDNA lipoplexes comprised of plasmid DNA and a mixed lipid formed by a DOPE zwitterionic lipid and a gemini cationic lipid NN -(1,3-phenylene bis (methylene)) bis (N,N-dimethyl-N-(1-dodecyl) ammonium dibromide (12PH12) containing an aromatic spacer or its monomeric counterpart surfactant, N-benzyl-N,N-dimethyl-N-(1dodecyl) ammonium bromide (12PH). Electrochemical results reveal that i) the gemini cationic lipid (12PH12) and the plasmid pDNA yield effective charges less than their nominal charges (+2 and −2/bp, respectively) and that ii) both vectors (12PH12/DOPE and 12PH/DOPE) could compact pDNA and protect it from DNase I degradation. SAXS and cryo-TEM experiments indicate the presence of a lamellar lyotropic liquid crystal phase represented as alternating layers of mixed lipid and plasmid. Transfection efficiency (by FACS and luminometry) and cell viability assay in COS-7 cells, performed with two plasmid DNAs (pEGFP-C3 and pCMV-Luc VR1216), confirm the goodness of the proposed formulations (12PH12/DOPE and 12PH/DOPE) to transport genetic material, with efficiencies and biocompatibilities comparable to or better than those exhibited by the control Lipofectamine 2000*. In conclusion, although major attention has been paid to gemini cationic lipids in the literature, due to the large variety of modifications that their structures may support to improve the biological activity of the resulting lipoplexes, it is remarkable that the monomeric counterpart surfactant with an aromatic group analyzed in the present work also exhibits good biological activity. The in vitro results reported here indicate that the optimum formulations of the gene vectors studied in this work efficiently transfect plasmid DNA with very low toxicity levels and, thus, may be used in forthcoming in vivo experiments. © 2017 Elsevier B.V. All rights reserved.

1. Introduction The goal of gene therapy is to accomplish the treatment of diseases that are caused by genetic disorders [1,2]. Among the different approaches, introduction of a nucleic acid in living cells to substitute for a certain damaged gene has been widely explored in recent decades [3–5]. Insertion and expression of exogenous DNA into cells require the use of carriers that must be able to condense the nucleic acid, protect it from degradation, promote its cellular uptake, and release it into the cytoplasm [6,7]. Firstly, viral

∗ Corresponding author. E-mail address: [email protected] (E. Aicart). https://doi.org/10.1016/j.colsurfb.2017.11.024 0927-7765/© 2017 Elsevier B.V. All rights reserved.

vectors were used to compact DNA; however, they had several drawbacks, such us immune responses, oncogenicity, and limitations on the size of the therapeutic gene [1,8]. Compared to viral vectors, cationic lipids are safe (there is no immune response), cost-effective and easier to fabricate, and for that, they have been explored as nanocarriers of nucleic acids in the last two decades, opening a wide range of possibilities [3,5,9,10]. Several factors on the rational design of the cationic lipid structure, i.e., nature of the cationic head, spacer group and tail length, are important for DNA delivery [11,12]. Among the wide variety of cationic lipids (CLs), gemini lipids (GCLs) are known to show better performance due to their greater ability to compact nucleic acids [5,13–15]. The GCLs, constituted by two equivalent cationic surfactants (one head-one tail) linked with a molecular spacer through their cationic head

520

M. Martínez-Negro et al. / Colloids and Surfaces B: Biointerfaces 161 (2018) 519–527

efficient transfection capabilities and adequate levels of cell viability that may be used in in vivo applications.

2. Materials and methods 2.1. Materials

Scheme 1. Structure of a) monomeric (12PH) and b) gemini (12PH12) cationic lipids.

groups, offer a wide variety of potential structural modifications to improve their biological activity [3]. In this regard, different types of spacer groups (i.e., hydroxyethyl, oxyethylene, alkyl and hydroxyethyl moieties) and cationic heads (i.e., quaternary ammonium or imidazole groups) have been studied [5,16,17]. Previous studies reveal that determination of the optimum transfection efficiency is strongly influenced by the length and type of the spacer moiety. Thus, short groups as oligo-oxyethylene and hydrophobic rigid aromatic spacers have shown better performance in gene delivery [18–20]. With respect to the type and role of the positive charge on the surfactant head, better transfection efficiency and viability results were found with imidazolium groups in which the charge is known to be delocalized [17,21]. On the other hand, the length of the hydrophobic tail is another key parameter that affects crossing through the cellular membrane; for instance, GCLs with 12-carbon tails seem to show significantly better results than those with 14 carbons [22], although both the cationic head and hydrophobic tail are known to play an important role [23]. In any case, better transfection efficiencies are obtained when CLs are used together with a neutral or zwitterionic lipid, such as 1,2-dioleoyl-sn-glycero-3phosphatidylethanolamine (DOPE) or mono-oleoylglycerol (MOG), which, working as helper lipids, decrease the gel-transition temperature; i.e., the mixed lipid bilayer becomes, in turn, more fluid, and its fusion with the cell membrane is enhanced [24–27]. All these previous considerations indicate that the success of the transfection of nucleic acids strongly depends on the final structure and charge distribution in the nanocarrier formed by the GCL mixed with a neutral or zwitterionic lipid [3,5]. Consequently, in the present work, we report the rational design, physicochemical characterization, and transfection performances of a cationic gemini (two tails-two heads) lipid based on two 12-carbon hydrophobic tails and two quaternary dimethyl ammonium cationic heads linked via an aromatic ring (12PH12, see Scheme 1). Bearing in mind the high variety of parameters that influence the transfection efficiency, we have established a structure activity relationship (SAR) by comparing a gemini-type lipoplex (GCL/DOPE-pDNA) with its cationic monovalent (12PH, see Scheme 1) counterpart-type lipoplex (CL/DOPE-pDNA). Firstly, lipoplexes have been analyzed through zeta potential, agarose gel electrophoresis, small angle X-ray scattering (SAXS) and cryo-TEM to establish a relationship between the charge and the structure, as well as lipid vector-plasmid interaction. Secondly, to have a better comprehension of their behavior in in vitro biochemical studies, we have performed transfection experiments and determined the cell viability of those lipoplexes in COS-7 (African green monkey kidney) cells. The in vitro results of the present work provide interesting information to design gene lipid-type nanocarriers with

A zwitterionic lipid (1,2-dioleoyl-sn-glycero-3phosphatidylethanolamine, DOPE) with the best purity was purchased from Avanti Polar Lipids, Inc., Alabaster, USA. The synthesis of the GCL, N-N -(1,3-phenylene bis (methylene)) bis (N,N-dimethyl-N-(1-dodeyl) ammonium dibromide (12PH12), and its monomeric counterpart surfactant, N-benzyl-N,N-dimethylN-(1-dodecyl) ammonium bromide (12PH), was recently fully described [28]. Sodium salt of calf thymus DNA (ctDNA), provided by Sigma-Aldrich (St. Louis, USA) was used as linear DNA to determine the effective charge of the gemini cationic vector (12PH12). pEGFP-C3 plasmid DNA (4700 bp), used on biophysical and biological experiments, was extracted from competent Escherichia coli bacteria previously transformed with pEGFP-C3. The extraction was carried out using a GenElute HP Select Plasmid Gigaprep Kit (Sigma Aldrich). The plasmid pCMV-Luc VR1216 (6934 bp) encoding luciferase (Clontech, Palo Alto, USA), used for biological experiments, was amplified in E. coli and isolated and purified using a Qiagen Plasmid Giga Kit (Qiagen GMBH, Hilden, Germany). All the reagents and solvents, of the highest grade commercially available, were used without further purification. 2.2. Preparation of lipoplexes Appropriate amounts of cationic lipids, 12PH12 or 12PH, and helper lipid, DOPE, were dissolved in chloroform to obtain the desired CL molar fraction (␣) on the lipid mixtures. After briefly vortexing this solution, chloroform was removed to yield a dry lipid film. The resulting dry lipid films were then hydrated with 40 mM HEPES, pH 7.4, and homogenized by means of a combination of vortexing and sonication. The resulting multilamellar liposomes were transformed into the desired unilamellar liposomes by a sequential extrusion procedure that is widely explained elsewhere [16,19]. To prepare the lipoplex, appropriate amounts of a pDNA stock solution, prepared one day before, were added to lipid suspensions. pDNA concentrations in HEPES solution were chosen to fit the optimum conditions for each experimental technique as follows: 1 mg/mL for zeta potential, 1 mg/mL for cryo-TEM, 200 ␮g/capillar (≈5 mg/mL) for SAXS, and 1 ␮g/well (2 ␮g/mL) for biological studies. 2.3. Zeta potential and particle size A phase analysis light scattering technique (Zeta PALS, Brookhaven Instruments Corp., Holtsville, USA) was used to measure electrophoretic mobility, which was used to obtain the zeta potential (␨) of the nanoaggregates [19,29]. Particle size was determined by a dynamic light scattering (DLS) method using a particle analyzer (Zeta Nano Series; Malvern Instruments, Barcelona, Spain). In both studies, samples were prepared with buffer 40 mM HEPES, pH 7.4. Experimental conditions were as follows: 25 ◦ C, dispersant refractive index of 1.33 (water), viscosity of 0.9 cP, and dispersant dielectric constant of 78.5. Each zeta potential and particle size data point was taken as an average over 50 and 30 independent measurements, respectively. Measurements were carried out for the two studied lipoplexes as a function of the lipid/DNA mass ratio, (mL+ + mL 0 )/mDNA (mL+, mL 0 and mDNA being the masses of the cationic gemini lipid, of the zwitterionic helper lipid and of

M. Martínez-Negro et al. / Colloids and Surfaces B: Biointerfaces 161 (2018) 519–527

the nucleic acid, respectively), and at several molar fractions (␣) of the cationic lipid in the cationic lipid/DOPE mixture. 2.4. Gel electrophoresis 2.4.1. DNA compaction assay Lipoplexes along with uncomplexed pDNA were loaded onto 1% agarose gels (with 0.7 ␮L of GelRed) and run for 30 min at 80 mV in 1 x TAE buffer. Fully complexed lipoplexes appeared as fluorescent bands in wells of the gel, while uncomplexed pDNA appeared outside the well. Spectra fluorescence conditions were excitation in the 302–312 nm and emission at 600 nm. Fluorescence intensity of each band was measured by using commercial Quantity One software and a Gel Doc XR instrument (Bio-Rad). 2.4.2. DNA protection assay DNase I (1 U/␮g of pDNA) was added to each mixed lipid sample and stirred for 30 min at 37 ◦ C. Then, 20 ␮L of 0.25 M EDTA was added to inactive DNase I, and the samples were incubated for 15 min. Next, 15 ␮L of 25% SDS was added and incubated for 5 min. Samples were electrophoresed for 40 min under 80 mV in 1% agarose gels (with 1 ␮L of ethidium bromide). Spectra fluorescence conditions were excitation at 482 nm and emission at 616 nm. The integrity of the plasmid in each composition was compared with untreated DNA as a control. 2.5. Small-angle X-ray scattering (SAXS) Experiments were carried out on a NCD11 Beamline at ALBA Synchrotron (Barcelona, Spain). The energy of the incident beam was 12.6 KeV (␭ = 0.995 Å). Samples were placed in sealed glass capillaries. The scattered X-ray was detected on a Quantum 201r CCD detector, converted to one-dimensional scattering by radial averaging, and represented as a function of the momentum transfer vector (q). SAXS experiments were performed for lipoplexes at several molar fractions (␣) of the mixed lipids and at various effective charge ratios (␳eff ) of the lipoplex. Measurements for each composition were run in duplicate in two independent capillaries. 2.6. Cryo-TEM Cryo-transmission electron microscopy experiments were carried out at Servei of Microscopy (Univ. Autónoma, Barcelona, Spain) following the standard procedure [30–32]. In these experiments, perforated Quantifoil R1.2/1.3 (hole diameter 1.2 ␮m) on a 400mesh copper grid was used. Images were obtained using a Jeol JEM 2011 cryo-electron microscope operated at 200 kV, under low-dose conditions, and using different degrees of defocus (500–700 nm) to obtain an adequate phase contrast [32]. Images were recorded on a Gatan 794 Multiscan digital camera. Finally, the CCD images were processed and analyzed with Digital Micrograph. 2.7. Cell culture COS-7 (African green monkey kidney) cells (American Type Collection, Rockville, MD, USA) were maintained at 37 ◦ C under 5% CO2 in complete medium constituted by Dulbecco’s modified Eagle’s medium-high glucose + glutaMAX (Gibco BRL Life Technologies) supplemented with 10% (v/v) heat inactivated fetal bovine serum (FBS), penicillin (100 U/mL) and streptomycin (100 ␮g/mL).

521

pDNA encoding GFP. In both methods, each measurement was carried out in triplicate in three wells from three independent cultures, and Lipofectamine (Lipo2000*) was used as the positive control (1.5 ␮L of Lipo2000*/␮g of DNA). 2.8.1. Luminometry Cells were seeded in complete medium in 48-well plates and incubated for 24 h at 37 ◦ C in 5% CO2 . The medium was removed and 0.3 mL of the same medium and 0.2 mL of complexes were added to each well. After 4 h incubation, the medium was removed, and the cells were further incubated for 48 h in complete medium. Cells were washed with phosphate-buffered saline (PBS) and lysate with 100 ␮L of reporter lysis buffer (Promega, Madison, WI, USA) at room temperature for 10 min, followed by two freeze-thaw cycles. The lysate cells were centrifuged for 2 min at 12,000 xg to pellet the debris. Then, 20 ␮L of the supernatant was assayed for total luciferase activity using the luciferase assay reagent (Promega). A luminometer (Sirius-2, Berthold Detection Systems, Innogenetics, Diagnóstica y Terapéutica, Barcelona, Spain) was used to measure luciferase activity. The protein content of the lysates was measured by the DC protein assay reagent (Bio-Rad, Hercules, CA, USA) using bovine serum albumin as the standard. Data were expressed as ng of luciferase per mg of protein. 2.8.2. Fluorescence assisted cell sorting (FACS) Cells were seeded in medium in 48-well plates and incubated for 24 h at 37 ◦ C in 5% CO2 . The medium was removed and 200 ␮L of trypsin 1X was added to each well. After two minutes incubation at 37 ◦ C, 400 ␮L of complete medium were added to neutralize the trypsin. Samples were put into cytometer tubes and centrifuged for 5 min at 1450 rpm at 4 ◦ C. Finally, the pellet was resuspended in 500 ␮L of a buffer containing PBS with 0.5% BSA and 2.5 mM EDTA. Then, cells were acquired using a flow cytometer. Fluorescence-activated cell sorting (FACS) analysis was performed using a Calibur 345 cytometer equipped with a 488 nm laser and with BD CellQuestTM Pro software. Cells were first gated using a forward scatter vs. side scatter (FSC vs. SSC) strategy to exclude debris (low events) and then were specifically analyzed for the 530 nm emission (FL1-H channel; the axis FL1-H shows relative intensity of GFP fluorescence). Data were analyzed using FlowJo LLC data software. Transfection efficiencies (TE) were quantified by means of% GFP cells, i.e., percentage of cells in which GFP expression is observed and the average intensity of fluorescence per cell (MFI, mean fluorescence intensity). 2.9. Cell viability Cell viability was quantified by a modified Alamar Blue assay. Briefly, 1 mL of 10% (v/v) Alamar Blue dye in Dulbecco’s modified Eagle’s medium, supplemented with 10% (v/v) FBS medium, was added to each well 48 h after transfection. After 2 h of incubation at 37 ◦ C, 200 ␮L of supernatant was assayed by measuring the absorbance at 570 nm and 600 nm. Wells containing medium and Alamar blue dye without cells were used as blanks. Cell viability was calculated according to the following formula: (A570 –A600 ) of treated cells x100/(A570 –A600 ) of control cells. Each sample was measured in three independent wells and Lipo2000* was taken as a positive control (1.5 ␮L of Lipo2000*/␮g of DNA). 3. Results and discussion 3.1. Biophysics of the lipoplexes in the absence of cells

2.8. In vitro transfection efficiency Two methods were used to evaluate transfection efficiency (TE): luminometry for the pDNA encoding luciferase and FACS for the

To obtain information about the structure-activity relationship (SAR), the physicochemical characteristics of the 12PH12/DOPEpDNA lipoplex containing the gemini cationic lipid (12PH12)

522

M. Martínez-Negro et al. / Colloids and Surfaces B: Biointerfaces 161 (2018) 519–527

Fig. 1. Agarose gel electrophoresis of lipoplexes: a) 12PH12/DOPE-pDNA and b) 12PH/DOPE-pDNA at several mass ratios (mL + + mL 0 )/mDNA of the lipoplex. Free pDNA (lane 1 in both gels) was used as a control. Solid and dashed white arrows correspond to the coiled and supercoiled pDNA, respectively.

were firstly evaluated and compared to those obtained with the 12PH/DOPE-pDNA lipoplex containing its monomeric counterpart (12PH) (Scheme 1). Initially, the level of pDNA compaction by lipidic nanovectors was determined by means of agarose gel electrophoresis, which provides us with rough information about the minimum (mL + + mL 0 )/mDNA ratio needed to efficiently compact the pDNA. Fig. 1 shows the results obtained for the 12PH12/DOPE-pDNA and 12PH/DOPE-pDNA lipoplexes at three different (mL + + mL 0 )/mDNA ratios at a molar fraction of the mixed lipid ␣ = 0.2. The fluorescent bands, corresponding to the coiled forms (pointed with solid arrows) and supercoiled forms (pointed with the dashed arrows) of the plasmid, disappear across the lanes as the (mL + + mL 0 )/mDNA mass ratio increases. This loss of fluorescence is correlated with an adequate level of pDNA compaction. As seen, the pDNA is compacted at (mL + + mL 0 )/mDNA = 0.77 and 0.91 for the 12PH12/DOPE-pDNA and 12PH/DOPE-pDNA lipoplexes, respectively. The zeta potential, however, provides more precise information since it allows the obtaining of the electroneutrality ratio ((mL + + mL 0 )/mDNA ) of the lipoplexes, defined as the mass ratio at which there is a charge compensation and the zeta potential is cancelled. This value is important because the lipoplexes must have a net positive charge to interact efficiently with the negatively charged cell membranes. From the electroneutrality ratios, effective charges of the cationic lipid (q+ + ) and plasmid

respectively. The second step is the determination of the effective charge of the plasmid DNA (using Eq. (2)) for the complex constituted by the same cationic lipid (with its effective charge previously determined) but now with the plasmid pDNA instead of the linear DNA (linearDNA), q+ = q− eff,pDNA

eff,L+

m + + m 0   L L mDNA

␾

˛Mbp ␣ML+ + (1 − ˛)ML0



(2)

Fig. 2 (dashed curve) shows typical sigmoidal behavior of the zeta potential vs. (mL + + mL 0 )/mDNA for the 12PH12/DOPE-ctDNA at ␣ = 0.5. The described procedure permits the obtaining of the electroneutrality ratio ((mL + + mL 0 )/mDNA ) for the 12PH12/DOPEctDNA lipoplex and, from this ratio, the effective charge of the 12PH12 lipid, (qeff,12PH12 + = 1.6 ± 0.1), which is 20% lower than its nominal charge (+2). Fig. 2 also shows zeta potential data vs. (mL + + mL 0 )/mDNA for 12PH12/DOPE-pDNA lipoplexes at different molar compositions, ␣ = 0.2, 0.4, 0.5 and 0.7 (Fig. S1 reports the zeta potential study for the 12PH/DOPE-pDNA lipoplex). Afterward,

eff,L

) may be obtained. It is widely reported in the litDNA (q− eff,pDNA erature that linear DNA, such as calf thymus DNA (linearDNA), has an effective charge that matches its nominal value (−2/bp) [29,33], while plasmid DNA and polycationic gene vectors have effective charges that are usually lower than their nominal ones [16,17,34,35]. Accordingly, in a first step, the determination of the effective charge of the GCL (q+ + ) of the lipoplex is achieved foleff,L

lowing a fully reported procedure [16,29] based on eq. 1, in which the lipoplex is constituted by this cationic lipid and a stranded linear DNA with a known charge of −2/bp,



q+

eff,L+

= q− eff,linearDNA

␣ML+ + (1 − ␣)ML0



Mbp ␣ (m

L+

+,

0

+ mL0 )/mDNA





(1) ␾

In this equation, mL mL and mDNA are the masses of cationic lipid, DOPE and DNA, respectively, while ML + , ML 0 and Mbp are molar masses of cationic lipid, DOPE and plasmid per base pair,

Fig. 2. Plots of the zeta potential vs. (mL + + mL 0 )/mDNA for the 12PH12/DOPE-ctDNA lipoplex (pink dashed line at ␣ = 0.5) and the 12PH12/DOPE-pDNA lipoplex (black, red, green and blue solid lines at ␣ = 0.2, 0.4, 0.5 and 0.7, respectively). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

M. Martínez-Negro et al. / Colloids and Surfaces B: Biointerfaces 161 (2018) 519–527

523

Fig. 3. Gel electrophoresis of DNA protection by the 12PH12/DOPE mixed lipids for the pEGFP plasmid (a) and for the pCMV-Luc plasmid (b) at effective charge ratios ␳eff = 4 and 10 of the lipoplex. Lane 1, coiled and supercoiled pDNA; lane 2, pDNA-DNase I; and the rest of the lanes, 12PH12/DOPE-pDNA lipoplexes at different molar compositions of the cationic lipid: ␣ = 0.2, lanes 3–4, and ␣ = 0.5, lanes 5–6. Solid and dashed white arrows correspond to the coiled and supercoiled pDNA, respectively.

the electroneutrality ratios ((mL + + mL 0 )/mDNA ) and, from these ratios, the effective charge of plasmid DNA are obtained for the 12PH12/DOPE-pDNA and 12PH/DOPE-pDNA lipoplexes at several molar fractions (␣). These values, reported in Table S1, show that the effective charge of pDNA (qeff,pDNA − ) in the lipoplexes formed by 12PH12/DOPE-pDNA and 12PH/DOPE-pDNA are 15–20% and 40% of the nominal values (-2/bp), respectively. The fact that the real charge of pDNA is lower is crucial to obtain positively charged systems with less of the cationic vector and, in turn, potentially less cytotoxic. Once the effective charges of GCL and pDNA are deter-

mined, it is possible to calculate the effective charge ratio (␳eff ) of the lipoplex following Eq. (3),

␳eff =

q+ + (mL+ /ML+ ) n+ eff,L = − n− qeff,pDNA (mpDNA /MpDNA/bp )

(3)

The slopes in Figs. 2 and S1 are indicative of the electrostatic interactions in the complex formed by the gene vector and the DNA, which are generally related to the effective charge of the cationic lipid [33]. Thus, the picture that emerges from the plots in those figures indicates that the slopes and, in turn, the compaction level

Fig. 4. SAXS diffractograms of (a) 12PH12/DOPE-pDNA and (b) 12PH/DOPE-pDNA lipoplexes at several molar compositions of the cationic lipid (␣ = 0.2, 0.4, 0.5 and 0.7) and an effective charge ratio of ␳eff = 4 of the lipoplex. Schematic drawing (c) of 12PH12, DOPE, plasmid pDNA, and the multilamellar phase (L␣ ) that forms the lipoplex. Arrows indicate the pDNA-pDNA correlation peak.

524

M. Martínez-Negro et al. / Colloids and Surfaces B: Biointerfaces 161 (2018) 519–527

Fig. 5. Cryo-TEM micrographs of 12PH12/DOPE-pDNA lipoplex at ␳eff = 4 and at molar compositions of the cationic lipid: ␣ = 0.2 (a) and ␣ = 0.5 (b), and 12PH/DOPE-pDNA lipoplex at molar compositions of cationic lipid: ␣ = 0.2 (c) and ␣ = 0.5 (d). Asterisks, mixed lipid liposomes; arrows, lamellar structure with multilamellar pattern. Scale bar is 200 nm.

and stability of the lipoplexes follow the order 12PH12/DOPE-pDNA > 12PH/DOPE-pDNA. DNA is known to be degraded by DNases existing in human serum and, therefore, some vectors that yield good in vitro outcomes, are not efficient agents for in vivo experiments. For instance, it is important to evaluate the ability of the gene vector to protect the plasmid against DNase I degradation. To confirm that the gene vectors used in the present work protect the nucleic acid, DNA protection assays (based on gel electrophoresis experiments) were carried out with two different plasmids, one encoding GFP (pEGFPC3) and the other encoding luciferase (pCMV-Luc). Fig. 3 reports the results obtained at two molar fractions of the mixed lipid (␣ = 0.2 and 0.5) and at two effective charge ratios (␳eff = 4 and 10) of the 12PH12/DOPE-pDNA lipoplex (results for the 12PH12/DOPE-pDNA lipoplex are shown in Fig. S2). Naked plasmid DNA, digested within the first minutes following DNase I addition, is used as a control (lanes 2). The DNA bands in lanes 3–6 of both gels indicate the presence of intact DNA, confirming that the 12PH12/DOPE-pDNA and 12PH/DOPE-pDNA lipoplexes compact the plasmid and avoid the DNase I access. This significant protection of pDNA from degra-

dation points to these nanocarriers as potential non-viral gene vectors. The structural characterization was achieved by DLS, SAXS and cryo-TEM. Particle sizes obtained by DLS at molar compositions (␣ = 0.2 and 0.5) of the mixed lipid and at two effective charge ratios (␳eff = 4 and 10) for lipoplexes (with each one of the two plasmids) are reported in Table S2. All the nanoaggregates have positive zeta potential and particle sizes ranging in most cases from 130 to 200 nm, with low values of polydispersity (PDI). The positive zeta potential promotes an attractive interaction with negative glycoproteins and phospholipids of the cell membrane while a size <200 nm is known to favor an endocytosis mechanism to cross the cell membrane. Both characteristics point to these systems as appropriate nanocarriers for transfection of nucleic acids in biological processes [36,37]. Previous results, reported by our group and others, have confirmed that lipoplexes forming hexagonal and/or cubic structures transfect nucleic acids with higher efficiency than those forming lamellar structures [19,20,38–40]. Accordingly, the results are relevant to explore the presence of one or even several of these lipoplex structures in the 12PH12/DOPE-pDNA and 12PH/DOPE-pDNA

M. Martínez-Negro et al. / Colloids and Surfaces B: Biointerfaces 161 (2018) 519–527

525

Fig. 6. Transfection efficiency (%GFP cells and MFI) in COS-7 cells at two molar compositions of the cationic lipid (␣ = 0.2 and 0.5) for the 12PH12/DOPE-pEGFP and 12PH/DOPEpEGFP lipoplexes. Experiments were performed with 10% serum (FBS). Orange and green bars correspond to effective charge ratios ␳eff = 4 and 10 of the lipoplex, respectively. Black bar, Lipo2000*, as a positive control. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

lipoplexes studied herein, and with this aim, SAXS experiments are performed that cover the whole range of the mixed lipid composition (␣) (␣ = 0.2, 0.4, 0.5 and 0.7) at several lipid/DNA charge ratios (␳eff = 1.5, 2.5 and 4).Figs. 4 and S3 show the diffractograms (intensity vs. q factor) in which the Bragg reflections (Miller indexes are included in the figures) can be indexed to a lamellar lyotropic liquid crystal phase (L␣ ), with the interlayer distance (d) directly related to the qhkl factor (2␲n/qhkl ) and n as the diffraction order. Table S3 shows these SAXS parameters. This lamellar structure may be represented as alternating layers of mixed lipids and supercoiled plasmid DNA in a sandwich-type fashion (Fig. 2c), where d = dm + dw , dm and dw being the thickness of the mixed lipid bilayer and the aqueous monolayer that allocates the plasmid, respectively. Values of this interlayer periodicity (d), reported in Fig. S4 as a function of ␳eff , range from 7 to 5.5 nm, indicating that a) at a constant ␣, d remains constant or slightly decreases with ␳eff for both 12PH12/DOPE-pDNA and 12PH/DOPE-pDNA lipoplexes and that b) at a constant ␳eff , d slightly decreases with ␣. These d values, together with a value of dm of approximately 4.5 nm for the 12PH12/DOPE or 12PH/DOPE lipid bilayer [5], allow the estimation of a dw value of 1.8-2.5 nm for the aqueous monolayer, which is enough to allocate the plasmid DNA. As the SAXS results indicate that lipoplexes in this study form only lamellar structures, independent of the molar composition (␣) of the mixed lipid and of the effective charge ratio (␳eff ) of the lipoplex; for the rest of the biophysical and biological studies, only two (low and intermediate) molar compositions (␣) of the mixed lipid (␣ = 0.2 and 0.5) and two (medium and high) effective charges ratios (␳eff ) of the lipoplex (␳eff = 4 and 10) are selected. Cryo-TEM, together with SAXS, is a powerful tool to obtain structural and morphological information of the nanoaggregates formed by the lipoplexes. Fig. 5 reports a selection of micrographs of 12PH12/DOPE-pDNA (a-b) and 12PH/DOPE-pDNA (c-d) with the pEGFP-C3 plasmid at a charge ratio of ␳eff = 4 and at molar compositions of the mixed lipids of ␣ = 0.2 and 0.5. These images and others not shown reveal the presence in both systems of i) vesicles of mixed lipids without pDNA (indicated with asterisks) and ii) lamellar nanoaggregates (see white arrows), in good agreement with SAXS experiments, with a multilamellar pattern formed from spherical or deformed liposomes with pDNA compacted in each aqueous monolayer in between the lamellae (see the scheme of Fig. 2c). This multilamellar pattern has also been found in a wide variety of lipoplexes formed by different types of gemini cationic lipids mixed with helper lipids when compacting nucleic acids [16,20,34]. Although, in previous studies, two types of multilamel-

lar patterns (cluster-type and fingerprint type) were found, the lipoplexes studied in the present work only show a cluster-type multilamellar pattern.

3.2. Biological activity (in vitro) of the lipoplexes in the presence of cells Once the physicochemical characterization has shown that the 12PH12/DOPE-pDNA and 12PH/DOPE-pDNA lipoplexes are capable of compacting pDNA and protecting it from DNase I degradation by forming nanoaggregates with structures potentially appropriate for transfecting nucleic acids to cells, their efficiency and biocompatibility as gene vectors are evaluated. Thus, the transfection efficiency (TE) was tested in COS-7 cells in the presence of 10% of serum (FBS) at molar compositions of the cationic lipid ␣ = 0.2 and 0.4 of the mixed lipid, each one at effective charge ratios ␳eff = 4 and 10 of the lipoplex. In the experiments, two types of plasmids were used, one encoding GFP (pEGFP-C3) and the other encoding luciferase (pCMV-Luc VR1216). As shown in Fig. 6, the TE results are evaluated by FACS and reported in terms of% of the population of cells expressing GFP protein (%GFP) and the mean fluorescence intensity (MFI). The results reveal that both mixed lipid-type vectors, 12PH12/DOPE and 12PH/DOPE, could transfect the plasmid pEGFPC3. In terms of%GFP and MFI, the efficacy is better at moderate molar composition (␣ = 0.5), with the optimum formulations for both nanocarriers being even better than Lipo2000*. Additionally, there are no major differences in TE regarding the effective charge ratio of the lipoplex, and the efficacy is similar or even slightly better at a moderate charge ratio (␳eff = 4). The transfection efficiency is also evaluated in terms of ng of luciferase per mg of protein in lipoplexes containing the plasmid pCMV-Luc VR1216 (Fig. S5). As seen, the efficiencies obtained in this study with the formulations of the two mixed lipids, 12PH12/DOPE and 12PH/DOPE, are comparable (in some cases, even better) than Lipo2000*. Finally, the biocompatibility of 12PH12/DOPE-pDNA and 12PH/DOPE-pDNA lipoplexes containing either pEGFP-C3 or pCMV-Luc VR1216 is evaluated by an Alamar Blue assay on COS-7 cells. The results, shown in Fig. 7, at several molar fractions of the cationic lipid (␣) and effective charge ratios (␳eff ) of the lipoplex are also compared with that obtained when using Lipo2000* as a positive control. It is evident in the figure that cell viability is similar regardless of the plasmid used, which confirms that the potential toxicity comes from the cationic lipid. At ␣ = 0.2, all the lipoplex formulations present cell viabilities over 85%, which may be considered an optimum value; in several of these formulations, the cell

526

M. Martínez-Negro et al. / Colloids and Surfaces B: Biointerfaces 161 (2018) 519–527

Fig. 7. Cell viability of COS-7 cells at two molar compositions of the cationic lipid (␣ = 0.2 and 0.5) of the mixed lipids for the lipoplexes 12PH12/DOPE-pDNA and 12PH/DOPEpDNA. The plasmids that were used were pEGFP (a) and pCMV-Luc (b). Experiments were performed with 10% serum (FBS). Orange and green bars correspond to effective charge ratios ␳eff = 4 and 10 of the lipoplex, respectively. Black bar, Lipo2000*, as a positive control. The results are normalized to those obtained for untreated cells (100%). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

viability levels are even better than that shown by the cells transfected with the positive control Lipo2000*. However, at a moderate CL composition (␣ = 0.5), the viability remains at high levels (over 80%) only for the formulation containing the gemini lipid 12PH12, while it decreases to values of 60–70% or even lower for the formulation containing the 12PH lipid, reinforcing that there is higher toxicity in the one cationic head-one tail lipids compared to their gemini counterparts (two cationic heads-two tails). Accordingly, bearing in mind both transfection efficiency and cell viability experiments, it is possible to conclude that among the two lipid-type gene vectors tested in this study, the gene vector constituting a mixture of the gemini lipid 12PH12 and the helper zwitterionic lipid DOPE shows the best outcomes at ␣ = 0.5 and ␳eff = 4 on transfecting pEGFP and ␣ = 0.2 and ␳eff = 4 on transfecting pCMV-Luc; in both cases, the performances are comparable or even better than those reached with the standard Lipo2000*. Nevertheless, although the attention in the literature has been primarily focused on GCLs, possibly due to the high variety of potential structural modifications to improve their biological activity on transfecting pDNA to living cells, it must be stated that the monomeric counterpart of the GCL containing an aromatic group near the cationic head, as 12PH shows in the present work, also exhibits good biological activity, comparable to that of the corresponding GCL (12PH12), which was also found in earlier works with other cationic lipids [41,42]. The presence of the -electrons from the aromatic group of the CL possibly allows additional - interactions with the DNA bases that favor the transfection efficiency. The in vitro results of the present work provide information that could help with designing future lipid-based gene nanocarriers with efficient transfection and cell viability that, in turn, may be used in forthcoming in vivo experiments.

degradation. SAXS and cryo-TEM results indicate the presence of a lamellar lyotropic liquid crystal structure at the studied mixed lipid compositions and effective charge ratios of the lipoplexes; this structure, with a multilamellar pattern, can be represented as alternating bilayers of the CL/DOPE mixed lipids and aqueous monolayers of plasmid DNA. Regarding transfection efficiency and cell viability results in COS-7 cells, we conclude that the gene vector constituted by a mixture of the gemini lipid (12PH12) and DOPE helper lipid shows the best outcomes at ␣ = 0.5 and ␳eff = 4 on transfecting pEGFP and at ␣ = 0.2 and ␳eff = 4 on transfecting pCMV-Luc; in both cases, the performances are comparable or even better than those of the standard Lipo2000*. Nevertheless, it must be noted that the monomeric surfactant (12PH) also exhibits good biological activity, even close to the activity reached with the gemini counterpart (12PH12). All these results allow us to conclude that these formulations present adequate characteristics to be evaluated in future in vivo experiments. Conflict of interest Authors declare no competing financial interest. Acknowledgments This work was supported by grants from MINECO of Spain (contract numbers CTQ2012-30821, CTQ2015-65972-R, CTQ2015-64425-C2-2-R and CTQ2014-55208-P), Xunta de Galicia (2007/085), and University Complutense of Madrid (Spain) (project no. UCMA05-33-010). SAXS experiments were performed with the NCD11 beamline at ALBA Synchrotron Light Facility with the collaboration of ALBA staff. Cryo-TEM experiments were performed at the Servei de Microscopia of Univ. Autónoma of Barcelona (Spain).

4. Conclusions Appendix A. Supplementary data Biophysical and biological characterizations of lipoplexes formed by a plasmid DNA and lipid-based gene nanocarrier, constituted by a DOPE zwitterionic helper lipid and a gemini cationic lipid with an aromatic spacer (12PH12) or its monomeric counterpart (12PH), have been reported. Several conclusions may be established: i) the gemini cationic lipid (12PH12) yields an effective charge that is 20% less than its nominal one (+2), while the effective charge of plasmid DNA in the lipoplexes formed by 12PH12/DOPEpDNA and 12PH/DOPE-pDNA are 15–20% and 40% of its nominal one (-2/bp), respectively. Electrochemical experiments reveal that both systems could compact pDNA and protect it from DNase I

Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.colsurfb.2017.11. 024. References [1] [2] [3] [4]

I.M. Verma, N. Somia, Nature 389 (1997) 239–242. N. Somia, I.M. Verma, Nat. Rev. Genet. 1 (2000) 91–99. E. Junquera, E. Aicart, Adv. Colloid Interface Sci. 233 (2016) 161–175. R.S. Dias, B. Lindman, DNA Interaction with Polymers and Surfactants, Wiley & Sons, Hoboken NJ, 2008.

M. Martínez-Negro et al. / Colloids and Surfaces B: Biointerfaces 161 (2018) 519–527 [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15]

[16]

[17] [18] [19]

[20]

[21] [22] [23] [24] [25]

E. Junquera, E. Aicart, Curr. Topics Med. Chem. 14 (2014) 649–663. I.S. Zuhorn, J. Engberts, D. Hoekstra, Eur. Biophys. J. 36 (2007) 349–362. C.M. Wiethoff, C.R. Middaugh, J. Pharm. Sci. 92 (2003) 203–217. H. Sakurai, K. Kawabata, F. Sakurai, S. Nakagawa, H. Mizuguchi, Int. J. Pharm. 354 (2008) 9–15. S.D. Wettig, R.E. Verrall, M. Foldvari, Curr. Gene Ther. 8 (2008) 9–23. M.A. Mintzer, E.E. Simanek, Chem. Rev. 109 (2008) 259–302. T. Ahmed, A.O. Kamel, S.D. Wettig, Nanomedicine 11 (2016) 289–306. D. Zhi, S. Zhang, S. Cui, Y. Zhao, Y. Wang, D. Zhao, Bioconjugate Chem. 24 (2013) 487–519. S. Bhattacharya, A. Bajaj, Chem. Commun. (2009) 4632–4656. A. Bajaj, P. Kondaiah, S. Bhattacharya, Biomacromolecules 9 (2008) 991–999. M. Damen, J. Aarbiou, S.F.M. van Dongen, R.M. Buijs-Offerman, P.P. Spijkers, M. van den Heuvel, K. Kvashnina, R.J.M. Nolte, B.J. Scholte, M.C. Feiters, J. Controlled Release 145 (2010) 33–39. ˜ S.K. Misra, M. Munoz-Úbeda, S. Datta, A.L. Barrán-Berdón, C. Aicart-Ramos, P. Castro-Hartmann, P. Kondaiah, E. Junquera, S. Bhattacharya, E. Aicart, Biomacromolecules 14 (2013) 3951–3963. ˜ A.L. Barrán-Berdón, S.K. Misra, S. Datta, M. Munoz-Úbeda, P. Kondaiah, E. Junquera, S. Bhattacharya, E. Aicart, J. Mater. Chem. B 2 (2014) 4640–4652. A. Bajaj, P. Kondaiah, S. Bhattacharya, J. Med. Chem. 51 (2008) 2533–2540. ˜ M. Munoz-Úbeda, S.K. Misra, A.L. Barrán-Berdón, S. Datta, C. Aicart-Ramos, P. Castro-Hartmann, P. Kondaiah, E. Junquera, S. Bhattacharya, E. Aicart, Biomacromolecules 13 (2012) 3926–3937. M. Martínez-Negro, K. Kumar, A.L. Barrán-Berdón, S. Datta, P. Kondaiah, E. Junquera, S. Bhattacharya, E. Aicart, A.C.S Appl. Mater. Interfaces 8 (2016) 22113–22126. P. Midoux, C. Pichon, J.J. Yaouanc, P.A. Jaffrès, Brit. J. Clin. Pharm. 157 (2009) 166–178. S. Perrone, M. Usai, P. Lazzari, S.J. Tucker, H.M. Wallace, M. Zanda, Bioconjugate Chem. 24 (2013) 176–187. E.R. Lee, J. Marshall, C.S. Siegel, C. Jiang, N.S. Yew, M.R. Nichols, J.B. Nietupski, R.J. Ziegler, M.B. Lane, K.X. Wang, Hum. Gene Ther. 7 (1996) 1701–1717. J. Gustafsson, G. Arvidson, G. Karlsson, M. Almgren, Biochim. Biophys. Acta 1235 (1995) 305–312. S. Simões, V. Slepushkin, R. Gaspar, M.C.P. de Lima, N. Düzgünes, Gene Ther. 5 (1998) 955–964.

527

[26] A.C.N. Oliveira, K. Raemdonck, T. Martens, K. Rombouts, R. Simon-Vazquez, C. Botelho, I. Lopes, M. Lucio, A. Gonzalez-Fernandez, M. Oliveira, A.C. Gomes, K. Braeckmans, Acta Biomater. 25 (2015) 216–229. [27] G. Caracciolo, H. Amenitsch, Eur. Biophys. J. 41 (2012) 815–829. [28] V.I. Martín, A. Rodríguez, M.M. Graciani, I. Robina, A. Carmona, M.L. Moyá, J. Colloid Interface Sci. 363 (2011) 284–294. ˜ [29] M. Munoz-Úbeda, S.K. Misra, A.L. Barrán-Berdón, C. Aicart-Ramos, M.B. Sierra, J. Biswas, P. Kondaiah, E. Junquera, S. Bhattacharya, E. Aicart, J. Am. Chem. Soc. 133 (2011) 18014–18017. [30] J. Bednar, C.L. Woodcock, Chromatin, Academic Press Inc., San Diego, 1999. [31] J. Dubochet, M. Adrian, J.J. Chang, J.C. Homo, J. Lepault, A.W. McDowall, P. Schultz, Q. Rev. Biophys. 21 (1988) 129–228. [32] J. Dubochet, B. Zuber, M. Eltsov, C. Bouchet-Marquis, A. Al-Amoudi, F. Livolant, How to read a vitreous section, Methods Cell. Biol. (2007) 385–406. [33] V.D. Badwaik, E. Aicart, Y.A. Mondjinou, M.A. Johnson, V.D. Bowman, D.H. Thompson, Biomaterials 84 (2016) 86–98. ˜ [34] K. Kumar, A.L. Barrán-Berdón, S. Datta, M. Munoz-Úbeda, C. Aicart-Ramos, P. Kondaiah, E. Junquera, S. Bhattacharya, E. Aicart, J. Mater. Chem. B. 3 (2015) 1495–1506. [35] A. Ahmad, H.M. Evans, K. Ewert, C.X. George, C.E. Samuel, C.R. Safinya, J. Gene Med. 7 (2005) 739–748. [36] D. Liu, A. Mori, L. Huang, Biochim. Biophys. Acta: Biomembr. 1104 (1992) 95–101. [37] A. Gabizon, D. Papahadjopoulos, Biochim. Biophys. Acta: Biomembr. 1103 (1992) 94–100. [38] A.L. Barran-Berdon, M. Martinez-Negro, L. Garcia-Rio, O. Domenech, C.T. de Ilarduya, E. Aicart, E. Junquera, J. Mater Chem B. 5 (2017) 3122–3131. ˜ [39] M. Munoz-Úbeda, A. Rodríguez-Pulido, A. Nogales, A. Martín-Molina, E. Aicart, E. Junquera, Biomacromolecules 11 (2010) 3332–3340. [40] K.K. Ewert, H.M. Evans, A. Zidovska, N.F. Bouxsein, A. Ahmad, C.R. Safinya, J. Am. Chem. Soc. 128 (2006) 3998–4006. [41] G. Vulugundam, K. Kumar, P. Kondaiah, S. Bhattacharya, Org. Biomol. Chem. 13 (2015) 4310–4320. [42] J. Biswas, S.K. Mishra, P. Kondaiah, S. Bhattacharya, Org. Biomol. Chem. 9 (2011) 4600–4613.