A Gemini Cationic Lipid with Histidine Residues as a Novel ... - MDPI

nanomaterials Article

A Gemini Cationic Lipid with Histidine Residues as a Novel Lipid-Based Gene Nanocarrier: A Biophysical and Biochemical Study María Martínez-Negro 1 , Laura Blanco-Fernández 2 , Paolo M. Tentori 1 , Lourdes Pérez 3 , Aurora Pinazo 3 , Conchita Tros de Ilarduya 2 , Emilio Aicart 1 and Elena Junquera 1, * 1

2

3

*

Grupo de Química Coloidal y Supramolecular, Departamento de Química Física, Facultad de Ciencias Químicas, Universidad Complutense de Madrid, 28040 Madrid, Spain; [email protected] (M.M.-N.); [email protected] (P.M.T.); [email protected] (E.A.) Departamento de Farmacia y Tecnología Farmacéutica, Facultad de Farmacia, Universidad de Navarra, Instituto de Investigación Sanitaria de Navarra (IdiSNA), 31008 Pamplona, Spain; [email protected] (L.B.-F.); [email protected] (C.T.d.I.) Dpto. Tecnología Química y Tensioactivos, IQAC-CSIC, 08034 Barcelona, Spain; [email protected] (L.P.); [email protected] (A.P.) Correspondence: [email protected]; Tel.: +34-91-3944-131

Received: 15 November 2018; Accepted: 12 December 2018; Published: 16 December 2018







Abstract: This work reports the synthesis of a novel gemini cationic lipid that incorporates two histidine-type head groups (C3 (C16 His)2 ). Mixed with a helper lipid 1,2-dioleoyl-sn-glycero3-phosphatidyl ethanol amine (DOPE), it was used to transfect three different types of plasmid DNA: one encoding the green fluorescence protein (pEGFP-C3), one encoding a luciferase (pCMV-Luc), and a therapeutic anti-tumoral agent encoding interleukin-12 (pCMV-IL12). Complementary biophysical experiments (zeta potential, gel electrophoresis, small-angle X-ray scattering (SAXS), and fluorescence anisotropy) and biological studies (FACS, luminometry, and cytotoxicity) of these C3 (C16 His)2 /DOPE-pDNA lipoplexes provided vast insight into their outcomes as gene carriers. They were found to efficiently compact and protect pDNA against DNase I degradation by forming nanoaggregates of 120–290 nm in size, which were further characterized as very fluidic lamellar structures based in a sandwich-type phase, with alternating layers of mixed lipids and an aqueous monolayer where the pDNA and counterions are located. The optimum formulations of these nanoaggregates were able to transfect the pDNAs into COS-7 and HeLa cells with high cell viability, comparable or superior to that of the standard Lipo2000*. The vast amount of information collected from the in vitro studies points to this histidine-based lipid nanocarrier as a potentially interesting candidate for future in vivo studies investigating specific gene therapies. Keywords: lipid-based gene nanocarrier; gemini cationic lipid with histidine residues; gene delivery; plasmid DNAs; transfection; cell viability; biophysical characterization

1. Introduction Amino acids play important roles in cell life, acting as cell signaling molecules, regulating gene expression, and synthesizing hormones, among others. These features make them attractive as components of amino acid-based gene carriers for gene therapy, or as a part of different kind of molecules, such as of cationic monovalent or multivalent lipids [1–6], polypeptides [7–11], polymers [12–15], polycations [16] and/or dendrimers [11,17–20]. Generally, gene therapy has focused on viral and non-viral vectors for the treatment of hereditary or acquired illnesses [21–23]. In the case of viral carriers, the positively charged amino acids in the virus capsids are the basis for the tremendous Nanomaterials 2018, 8, 1061; doi:10.3390/nano8121061

www.mdpi.com/journal/nanomaterials

Nanomaterials 2018, 8, 1061

2 of 18

success of this class of vectors in cell penetration [24]. However, they can easily trigger both innate and adaptive immune responses, which justifies the vast efforts made towards the synthesis of non-viral vectors [25,26]. Among other non-viral carriers, cationic lipids (CLs) constitute adequate platforms for effective gene compaction [27–30]. In addition, they show interesting features such as ease of manufacture, low cost, and biocompatibility; however, their low in vivo delivery efficiency has fueled the search for new cationic vectors. For instance, gemini cationic lipids (GCLs), structurally formed by two hydrophobic tails and two cationic head groups linked by a spacer, have attracted notable attention [27,31]. A variety of modifications to their structure can improve the transfection efficiency and reduce the cellular toxicity [30,32,33]. In this context, amino acid derivatives have been studied not only in cationic lipids forming lipoplexes [1,3,5,34–38], but also in polyplexes [5,12,39], to enhance the efficiency of gene delivery. For example, lysine-rich segments can facilitate the condensation of DNA [40]. The use of lysine units as cationic groups leads to interesting chiral nanostructures (i.e., ribbon-type and nanotubes) that can also thoroughly condense and compact the DNA, making them suitable for implementation as gene delivery vectors [40,41]. Arginine-based lipids also exhibit relevant DNA binding capability through parallel hydrogen bonding of the guanidinium group present in arginine [42–44]. Importantly, due to their ability to disrupt endosomes via protonation of the imidazole group at physiological pH, lipids comprising histidine residues yield high gene compaction rates, together with prominent transfection efficiencies [5,38,45–48]. Double-chain imidazolium surfactants [28,49], also exhibit effective siRNA delivery capacity, while gemini imidazolium surfactants efficiently condense DNA and deliver nucleotides inside human cells [33,50,51]. However, the use of a cationic nanocarrier of plasmid DNA, combining the capability of gemini-type lipids to transfect nucleic acids efficiently together with the high biocompatibility of the amino acid histidine head groups, remains unexplored and thus, was the objective of the present work. The study demonstrates the potential for gene delivery of a cationic gemini lipid, comprised of two N(τ),N(π)-bis(methyl)-histidine hexadecyl amides linked through a spacer chain of three methylene groups (referred to as C3 (C16 His)2 ) (see Scheme 1). In order to evaluate these features, the C3 (C16 His)2 gemini cationic lipid was mixed with 1,2-dioleoyl-sn-glycero-3-phosphatidyl ethanol amine (DOPE), a well-known fusogenic helper lipid. Firstly, physicochemical characterization was carried out for C3 (C16 His)2 /DOPE-pDNA lipoplexes containing two different DNA plasmids (pDNA), one encoding the green fluorescent protein (GFP; pEGFP-C3) and the other a luciferase (pCMV-Luc). Agarose gel electrophoresis and zeta potential analyses were used to evaluate the compaction of both plasmids, while their structure was characterized by small-angle X-ray scattering (SAXS) and the bilayer fluidity by fluorescence anisotropy. The transfection efficiency was evaluated in vitro using the COS-7 and HeLa cell lines, through fluorescence assisted cell sorting (FACS) and luminometry. More interestingly, the C3 (C16 His)2 /DOPE-pDNA lipoplexes were found to exhibit little cytotoxicity across all the formulations studied in the present work. With the aim to corroborate the potential of this GCL bearing histidine residues as a therapeutic vector, further in vitro experiments were performed with a therapeutic plasmid (pCMV-IL12) encoding IL-12, a heterodimeric cytokine linked with autoimmunity. The biological in vitro results demonstrate that the C3 (C16 His)2 /DOPE-pDNA lipoplexes may have an appreciable potential as efficient and biocompatible nucleic acid nanocarriers. Nonetheless, further in vivo experiments are necessary to confirm them as promising therapeutic vectors.

Nanomaterials 2018, 8, 1061 Nanomaterials 2018, 8, x FOR PEER REVIEW

3 of 18 3 of 18

Scheme 1. 1. Chemical procedure used to synthesize the gemini cationicthe lipidgemini C3(C16His) 2 from N(α)Scheme Chemical procedure used to synthesize cationic lipid Cbz-histidine: (a) (CH 3 ) 2 SO 4 in methanol, (b) hexadecyl amine, benzotriazol-1C3 (C His) from N(α)-Cbz-histidine: (a) (CH ) SO in methanol, (b) hexadecyl amine, 16 2 3 2 4 yloxy)tris(dimethylamino)phosphonium hexafluorophosphate (BOP), and benzotriazol-1-yloxy)tris(dimethylamino)phosphonium hexafluorophosphate (BOP), 1,4and diazabicyclo[2.2.2]octane (DABCO) in dimethyl formamide (DMF), (c) hydrogenation, Pd/C in 1,4-diazabicyclo[2.2.2]octane (DABCO) in dimethyl formamide (DMF), (c) hydrogenation, Pd/C in methanol, and glutaricacid acid(HOOC-CH (HOOC-CH22-CH -CH22-CH DCM. methanol, and (d)(d) glutaric -CH22-COOH), -COOH),BOP, BOP,and andDABCO DABCOinin DCM.

2. Materials and Methods 2. Materials and Methods 2.1.2.1. Materials Materials The divalent gemini cationic lipid, lipid, bis(N(τ),N(π)-bis(methyl)-histidine hexadecyl amide)amide) propane The divalent gemini cationic bis(N(τ),N(π)-bis(methyl)-histidine hexadecyl (C3propane (C16 His)(C was synthesized accordingaccording to a procedure fully detailed the Supplementary Materials. 16His) 2), was synthesized to a procedure fully in detailed in the Supplementary 2 ),3(C TheMaterials. remaining compounds, of the highest grade commercially available and available used without The remaining all compounds, all of the highest grade commercially and further used without further purification, supplied manufacturers: by the followinghexadecyl manufacturers: by purification, were supplied by were the following aminehexadecyl by Fluka amine (Bucharest, Fluka (Bucharest, Romania); N(α)-carbobenzyloxyLL -histidine L-His.HCl) by Novabiochem Romania); N(α)-carbobenzyloxyL -histidine (N-Cbz-His.HCl)(N-Cbzby Novabiochem AG (Laufelfingen, AG (Laufelfingen, Switzerland); trifluoroacetic acid (TFA) and palladium activated Switzerland); trifluoroacetic acid (TFA) and palladium on activated charcoalon (Pd/C, 10%)charcoal by Merck (Pd/C, 10%) by Merck (Darmstadt, Germany); 1,4-diazabicyclo[2.2.2]octane (DABCO) by Aldrich (Darmstadt, Germany); 1,4-diazabicyclo[2.2.2]octane (DABCO) by Aldrich (Saint Louis, MO, USA); (Saint Louis, MO, USA); (benzotriazol-1-yloxy)tris(dimethylamino)phosphonium (benzotriazol-1-yloxy)tris(dimethylamino)phosphonium hexafluorophosphate (BOP) by TCI Europe hexafluorophosphate (BOP) by TCI Europe (Zwijndrecht, deuterated solvents by Eurisotop (Zwijndrecht, Belgium); deuterated solvents by Eurisotop Belgium); (Saint-Aubin, France); the zwitterionic lipid (Saint-Aubin, France); the zwitterionic lipid 1,2-dioleoyl-sn-glycero-3-phosphatidyl ethanol amine 1,2-dioleoyl-sn-glycero-3-phosphatidyl ethanol amine (DOPE) by Avanti Polar Lipids, Inc., (Alabaster, (DOPE) by Avanti Polar Lipids, Inc., (Alabaster, AL, USA); and the sodium salt of calf thymus DNA AL, USA); and the sodium salt of calf thymus DNA (ctDNA) by Sigma-Aldrich (St. Louis, MI, USA). (ctDNA) by Sigma-Aldrich (St. Louis, MI, USA). The pEGFP-C3 plasmid DNA (4700 bp) encoding The pEGFP-C3 plasmid DNA (4700 bp) encoding GFP was extracted from competent Escherichia coli GFP was extracted from competent Escherichia coli bacteria previously transformed with pEGFP-C3 bacteria previously transformed with pEGFP-C3 using a GenElute HP Select Plasmid Gigaprep Kit using a GenElute HP Select Plasmid Gigaprep Kit (Sigma-Aldrich). The pCMV-Luc VR1216 plasmid (Sigma-Aldrich). The pCMV-Luc VR1216 plasmid DNA (6934 bp) encoding a luciferase (Clontech, Palo DNA (6934 bp) encoding a luciferase (Clontech, Palo Alto, CA, USA) was amplified in E. coli using a Alto, CA, USA) was amplified in E. GMBH, coli using a Qiagen Plasmid Giga Kit (Qiagen GMBH, Hilden, Qiagen Plasmid Giga Kit (Qiagen Hilden, Germany). The plasmid pCMV-interleukin-12 Germany). The plasmid pCMV-interleukin-12 (5500 bp) encoding interleukin-12 (pCMV-IL12) was (5500 bp) encoding interleukin-12 (pCMV-IL12) was kindly provided by Dr. Chen Qian, (University kindly provided by Dr. Chen Qian, (University Navarra, Pamplona, Solutions were prepared of Navarra, Pamplona, Spain). Solutions were of prepared with distilled Spain). water from a Milli-Q Millipore with distilled water from a Milli-Q Millipore system. system. 2.2.2.2. Characterization ofofthe Lipid CC33(C (C1616 His) Characterization theIntermediates Intermediatesand andthe the Gemini Gemini Cationic Cationic Lipid His) 2 2 13 C NMR (nuclear magnetic The structures of the compounds were were characterized by 1 Hby and1H The structures of target the target compounds characterized and 13C NMR (nuclear resonance Chemical shifts (δ) areshifts reported in parts per million (ppm) downfield from magneticspectroscopy). resonance spectroscopy). Chemical (δ) are reported in parts per million (ppm) tetramethylsilane (TMS), and specific details are given in the Supplementary Materials. Distortionless downfield from tetramethylsilane (TMS), and specific details are given in the Supplementary enhancement by polarization transfer (DEPT) spectra was recorded to phase up was and recorded down thetoCH/CH Materials. Distortionless enhancement by polarization transfer (DEPT) spectra phase 3 upCH and down the CH/CH 3 and CH 2 signals, respectively. Mass spectra with fast atom bombardment and signals, respectively. Mass spectra with fast atom bombardment (FAB) or electrospray 2 (FAB) or were electrospray techniques were recorded ondetermination a VG-QUATTRO. The determination of Fourier techniques recorded on a VG-QUATTRO. The of Fourier transformation infrared spectroscopy (FT-IR) was performed on a Nicolet FT-IR (Avatar 360 equipped with a Smart iTR

Nanomaterials 2018, 8, 1061

4 of 18

accessory) with spectral range 4000–4500 cm−1 . Characterization of the C3 (C16 His)2 gemini cationic lipid by 1 H, 13 C, and 13 C-DEPT NMR spectra, and the FT-IR and UPLC-MS profiles, are also collected in the Supplementary Materials. 2.3. Preparation of Lipoplexes Firstly, in order to prepare a liposomal gene nanocarrier in (4-(2-hydroxyethyl)-1piperazineethanesulfonic acid) HEPES solution, following a procedure fully described earlier [32], appropriate amounts of C3 (C16 His)2 and DOPE were dissolved in chloroform to get the desired molar fractions of the cationic lipid (α) in the lipid mixture. Briefly, after vortexing the lipid solution, chloroform was removed and the dried films were hydrated with 40 mM HEPES (pH 7.4), which had been homogenized by vortexing and sonication to obtain a multilamellar liposome solution and, following a sequential extrusion method, converted to unilamellar liposomes of ~100 nm size. The lipoplexes were obtained by adding appropriate amounts of pDNA to each mixed lipid solution [32,33]. The final pDNA concentrations to fit the optimum conditions for each experimental technique used in this work were selected as follows: 1 mg/mL for zeta potential, 0.1 mg/mL for fluorescence anisotropy, 200 µg/capillary (~5 mg/mL) for SAXS, and 1 µg/well (2 µg/mL) for in vitro studies. 2.4. Experimental Methods Zeta potential and particle size: A phase analysis light scattering technique (Zeta PALS, Brookhaven Instruments Corp., Holtsville, NY, USA) was used to obtain the zeta potential of the samples, prepared with buffer 40 mM HEPES (pH 7.4) at 25 ◦ C [52]. The particle size was obtained by a dynamic light scattering (DLS) method using a particle analyzer (Zeta Nano Series; Malvern Instruments, Barcelona, Spain) [52]. Each zeta potential and particle size data point was taken as the average of 50 and 30 independent measurements, respectively. Measurements were carried out as a function of the total lipid/DNA mass ratio (mL+ + mL0 )/mpDNA (where mL+ , mL0 , and mDNA are the C3 (C16 His)2 , DOPE, and pDNA mass, respectively), and at various molar fractions (α) of the cationic lipid in the C3 (C16 His)2 /DOPE mixed lipid. DNA compaction assay by gel electrophoresis: Agarose gel electrophoresis experiments using a Gel Doc XR instrument (Bio-Rad) were carried out to analyze the compaction of pDNA by the mixed lipid nanocarrier. Details and experimental conditions of the procedure are reported previously [52]. Lipoplexes were loaded on 1% agarose gels (with 0.7 µL of GelRed). Samples were excited at 302–312 nm, and the fluorescence spectra emitted were collected at 600 nm. The fluorescence intensity was measured using the commercial Quantity One software. DNA protection assay by gel electrophoresis: The same Gel Doc XR instrument (Bio-Rad) was used in gel electrophoresis experiments [52] to analyze the pDNA protection degree against degradation by DNase I when pDNA was forming the C3 (C16 His)2 /DOPE-pDNA lipoplex. DNase I (1 U/µg of pDNA) was added to each lipoplex solution. Then, 20 µL of 0.25 M EDTA was added to inactive DNase I and the samples were incubated for 15 min. Next, 15 µL of 25% sodium dodecyl sulphate (SDS) was added to disrupt the lipoplexes, and later the samples were incubated for 5 min. The solutions were electrophoresed for 40 min under 80 mV, in 1% agarose gel with 1 µL of EtBr. Samples were excited at 482 nm, and the fluorescence spectra emitted were collected at 616 nm. Small-angle X-ray scattering: SAXS experiments were carried out on the beamline NCD11 at the ALBA Synchrotron (Barcelona, Spain). Details of the experiment and preparation of samples were fully described recently [52]. The energy of the incident beam was 12.6 KeV (λ = 0.995 Å). Samples were placed in sealed glass capillaries with an outside diameter of 1.5 mm. The scattered X-rays were detected on a Quantum 210r CCD detector, converted to one-dimensional scattering by radial averaging, and represented as a function of the momentum transfer vector (q = 4πsinθ/λ, where θ is half the scattering angle and λ is the wavelength of the incident X-ray beam). SAXS experiments were

Nanomaterials 2018, 8, 1061

5 of 18

performed for C3 (C16 His)2 /DOPE-pDNA lipoplexes at an effective charge ratio (C3 (C16 His)2 /DNA) of ρeff = 4, and at various cationic lipid molar fractions of the mixed lipid (α). Fluorescence anisotropy: Fluorescence anisotropy of the 1,6-diphenylhexatriene (DPH) probe was measured on a Perkin Elmer LS-50B luminescence spectrometer following an experimental protocol described elsewhere [53]. The probe, included in the mixed lipid bilayer, was excited at 360 nm, and its fluorescence emission was collected at 430 nm with slit widths of 2.5 nm. The anisotropy values, r = ( IVV − GIV H )/( IVV + 2GIV H ), were determined by measuring the intensity of the light emitted by the DPH probe, with the excitation and emission polarized following the modes: vertical–vertical (IVV ), vertical–horizontal (IV H ), horizontal–horizontal (IHH ), and horizontal–vertical (IHV ). The instrument grating factor ( G = IHV /IHH ) allowed for the correction of optical and electronic differences in the parallel and perpendicular channels was estimated as the average of 10 measurements for each solution. Cell culture: COS-7 (african green monkey kidney) and HeLa (human cervix adenocarcinoma) cells, supplied by American Type Collection (Rockville, MD, USA), were used to carry out in vitro biological experiments (transfection and cell viability). Experimental conditions of the complete medium, including fetal bovine serum (FBS), were similar to those detailed in earlier studies [52]. In vitro Transfection efficiency: The transfection efficiency of the C3 (C16 His)2 /DOPE-pDNA lipoplex formulations was analyzed through two complementary methods: luminometry for the pDNAs (pCMV-Luc and pCMV-IL12) encoding luciferase, and fluorescence assisted cell sorting (FACS) for the pDNA (pEGFP-C3) encoding GFP. In both methods [52], each measurement was carried out in triplicate in three wells, from three independent cultures (~50,000 cells/cm2 ). Lipofectamine (Lipo2000*) was used as the positive control. Luminometry: A luminometer (Sirius-2, Berthold Detection Systems, Innogenetics, Diagnóstica y Terapéutica, Barcelona, Spain) was used to determine the luciferase activity of the plasmid pCMV-Luc, using the luciferase assay reagent Promega. The experimental conditions and the followed procedure, using 48-well-plates, were fully described earlier [52]. The protein content was determined with the DC protein assay reagent (Bio-Rad, Hercules, CA, USA). The luciferase activity of the plasmid pCMV-IL12 was determined with an ELISA kit for murine IL12p70 (BD OptEIA ELISA sets, Pharmingen, San Diego, CA, USA) according to the manufacturer’s instructions. In both studies, data were obtained in RLU/mg protein, which have been converted to ng/mg protein using a standard calibrated curve. Fluorescence assisted cell sorting: The FACS study was carried out with a Calibur 345 flow cytometer equipped with a 488 nm laser and the BD CellQuestTM Pro software. Details of the apparatus and the experimental procedure, using 48-well-plates, were reported recently [52]. The data were analyzed using the FlowJo LLC data software. Initially, the cells were gated using a forward scatter vs. side scatter (FSC vs. SSC) strategy to exclude any debris (low events), and then specifically analyzed by their 530 nm emission (FL1-H channel; the axis FL1-H shows the relative intensity of the GFP fluorescence). Transfection efficiencies (TE) were quantified by means of the percentage of cells in which GFP expression was observed (%GFP), and by the average intensity of fluorescence per cell (MFI, mean fluorescence intensity). Cell viability: The cytotoxicity/cell viability was determined by a modified alamarBlue assay, following an experimental procedure reported previously [52]. Briefly, 1 mL of 10% (v/v) alamarBlue dye in Dulbecco’s modified Eagle medium, supplemented with 10% (v/v) FBS medium, was added to each well 48 h after transfection, and, after 2 h of incubation at 37 ◦ C, assayed by measuring the absorbance at 570 and 600 nm, The cell viability (%) was obtained according to the expression: [(A570 − A600 )treated cells /(A570 − A600 )untreated cells ] × 100, which correlates the absorbance data of treated and untreated cells [52]. Each sample was measured in three independent wells, and Lipo2000* was used as the positive control (1.5 µL of Lipo2000* per µg of DNA).

Nanomaterials 2018, 8, 1061

6 of 18

3. Results 3.1. Synthesis of the Gemini Lipid (C3 (C16 His)2 ) Gemini-based lipids are perfect candidates for new synthetic vectors for DNA and RNA transfection [54]. Amino acid-based surfactants offer some advantages compared to the classical bisQuats gemini surfactants, in which the polar head consists of two quaternary ammonium groups: They can be synthesized from renewable materials using economical procedures and, generally, they exhibit low cytotoxicity [55,56]. Moreover, it has been reported that, if the polar head and the hydrophobic moiety are connected by an amide bond, the structure of the surfactants becomes more stable under both chemical and thermal conditions and displays high biodegradability [57]. Taking into account these advantages, a novel gemini cationic lipid based on the amino acid histidine has been synthesized. This compound, referred to as C3 (C16 His)2 , consists of two N(τ),N(π)-bis(methyl)-histidine hexyl amides linked through a bridge of three –CH2 – groups (Scheme 1). The palmitoyl fatty chain (C16 ) has been selected based on the structure–activity results relationship reported in the literature, which suggests that surfactants with C16 –C18 alkyl chains exhibit superior transfection properties [58,59]. The general chemical structure of the target gemini lipid (C3 (C16 His)2 ), as well as the strategy adopted for its preparation, is outlined in Scheme 1. The synthesis of this new GCL was carried out in four steps. The first step involved the methylation of the two nitrogens of the imidazol ring of the N(α)-Cbz-histidine, using di-methyl sulfate (CH3 )2 SO4 dissolved in methanol as a methylating agent. The yield of the reaction was very high and the reaction mixture contained only the target intermediate and inorganic salts, which could be easily removed by filtration of the mixture in dry methanol. This bismethylated N(α)-Cbz-histidine was then reacted with hexadecyl amine to form a N(α)-Cbz-N(τ),N(π)-bis(methyl)-histidine hexadecyl amide intermediate. The synthesis involved treatment of the Z-protected histidine with BOP and DABCO. Our group has recently prepared single chain N(α)-Cbz-N(τ),N(π)-bis(methyl)-histidine alkyl amide surfactants using a different procedure [60]. The method used in the past provided very good yields; however, the purification of the C16 derivative was very problematic. The use of the coupling agent BOP to rapidly and efficiently activate the carboxylic group of the protected N(α)-Cbz-histidine led to the alkylated intermediate with 90% conversion in 3 h. Purification of this compound was then performed by washing the mixture with diethyl ether to remove the coupling agent. The Cbz group of the N-terminus was then removed via catalytic hydrogenation. Afterwards, the two carboxylic groups of tartaric acid were reacted with the free amine of the N(τ),N(π)-bis(methyl)-histidine hexadecyl amide using the same coupling agent, to form a gemini lipid. Preparative HPLC chromatography was employed to purify the obtained GCL. The chemical structure and purity of all the compounds, i.e., the intermediates and target GCL, was established by NMR spectroscopy (1 H, 13 C, and 13 C-DEPT), FT-IR, and mass spectrometry (UPLC-MS). The spectral characterization of the C3 (C16 His)2 gemini lipid, reported in Figure S1 in the Supplementary Materials, confirmed the designed structure. In the 13 C NMR spectrum, the signal for the terminal CH3 was observed at 14.5 ppm, and those of the two CH3 groups attached to the imidazolium nitrogen atoms at 34.1 and 36.4 ppm. The carbon of the asymmetric CH moiety was observed at 52.7, and those of the imidazolium ring at 123 and 138.4 ppm. The 13 C NMR spectrum shows two signals corresponding to CONH groups, in contrast to the single chain precursor bearing only one amide group. The 1 HNMR spectrum is also consistent with such a dimeric structure. The terminal CH3 appears as a triplet at 0.9 ppm, while the signals of the two CH3 groups attached to each imidazolium appear as two singlets at 3.8 and 3.9 ppm. The signal for the CH2 moiety of the spacer chain appears as a multiplet at 2.3 ppm, confirming the dimerization of the single chain intermediate. The FT-IR spectrum also agrees with the target structure, with the absorption frequencies for the C=O stretching band (amide I) around 1644 cm−1 . Moreover, a broad amide N–H band at 3200 cm−1 indicated the existence of hydrogen bonded amide groups. The FT-IR spectrum also contains the frequencies corresponding to aromatic groups and long alkyl chains (Figure S1).

Nanomaterials 2018, 8, 1061

7 of 18

The structure of this gemini was also established by UPLC-MS mass spectrometry, which showed a [M-Cl]+2 peak corresponding to the dipositive molecular ion lacking both chloride ions. 3.2. Biophysics of the C3 (C16 His)2 /DOPE-pDNA Lipoplexes Prior to biological studies to evaluate whether the presence of amino acid residues as cationic heads in the gemini cationic lipid (C3 (C16 His)2 ) was able to promote the transfection efficiency and bioavailability of plasmid DNA, a detailed physicochemical characterization of the C3 (C16 His)2 /DOPE-pDNA lipoplexes was carried out. In this context, zeta potential and agarose gel electrophoresis (compaction assay) are fundamental tools to assess the electroneutrality of such a lipoplex. Figure 1 (inset) shows the electrophoresis agarose gel image for C3 (C16 His)2 /DOPE-pDNA lipoplexes at two C3 (C16 His)2 molar compositions, (α) = 0.2 and 0.5, at various mL /mDNA = (mL+ + mL0 )/mDNA mass ratios. In the first lane, used as the control, coiled and supercoiled forms of plasmid DNA are observed as two fluorescent bands. The absence of these fluorescent bands in the other lanes indicates that a full compaction of pDNA remained in the well. The minimum mL /mDNA mass relation at which all pDNA was effectively compacted was 1 and 0.6, at molar fractions of C3 (C16 His)2 in the C3 (C16 His)2 /DOPE mixed lipid (α) of 0.2 and 0.5, respectively. In order to determine with better accuracy the mass relation value at which charge inversion of the C3 (C16 His)2 /DOPE-pDNA lipoplex occurs, measurements of the zeta potential (ζ) were carried out at several mass ratios. Figure 1 reports the zeta potential data against mL /mDNA at different molar fractions (α = 0.2, 0.4, 0.5, 0.6, and 0.7); the fitted curves present the typical sigmoidal profile, where the lipoplexes display charge inversion at specific (mL /mDNA )φ values. The effective charges of the C3 (C16 His)2 cationic lipid (qeff,C3 (C16 His)2 = 1.8 ± 0.1) and plasmid DNA (qeff,pDNA = −0.3 ± 0.1) were determined from the electroneutrality data at varying molar fractions (α), using a procedure reported by us [27,51,61] and fully described in the Supplementary Materials. These values indicate that, while the C3 (C16 His)2 lipid presented almost the whole positive nominal charge (+2), pDNA, at physiological pH it remained in a supercoiled conformation [62,63] and exhibited only around 14% of its negative nominal charge (−2/bp). This fact is very advantageous, since a smaller amount of cationic vector is necessary to obtain a positive complex, thus reducing its potential cytotoxicity. The effective charge ratio of the C3 (C16 His)2 /DOPE-pDNA lipoplex (ρeff ) between the positive C3 (C16 His)2 /DOPE mixed lipid (n+ ) and the negative pDNA (n− ) charges required for lipoplexes to be appropriate for transfection can be easily determined by using the C3 (C16 His)2 and pDNA effective charges previously obtained through the relation: q+ n+ e f f ,L ( m L+ /M L+ ) ρe f f = − = − (1) n qe f f ,DN A (m DN A /MDN A ) where ML+ and MDNA are the molecular weight of the GCL and pDNA, respectively.

Nanomaterials 2018, 8, 1061 Nanomaterials 2018, 8, x FOR PEER REVIEW

8 of 18 8 of 18

Figure 1.1.Plots Plots of the zeta (potential (ζ) magainst the mLratio /mDNA mass in the Figure of the zeta potential ) against the mass in the C3(C16ratio His)2/DOPEL / m DNA C3 (C16 His)2 /DOPE-pDNA lipoplexes at various mole fractions of the C3 (C16 His)2 cationic lipid in pDNA lipoplexes at various mole fractions of the C3(C16His)2 cationic lipid in the C3(C16His)2/DOPE the C (C16 His)2 /DOPE mixed lipid (α = 0.2, 0.4, 0.5, 0.6 and 0.7; red, blue, pink, orange and green, mixed3 lipid (α = 0.2, 0.4, 0.5, 0.6 and 0.7; red, blue, pink, orange and green, respectively) and of the respectively) and of the C (C16 His)2 /DOPE-ctDNA lipoplex at α = 0.5 (black). Inset: Electrophoresis C3(C16His)2/DOPE-ctDNA3 lipoplex at α = 0.5 (black). Inset: Electrophoresis agarose gel of the agarose gel of the C3 (C16 His)2 /DOPE-pDNA lipoplexes at various mL /mDNA mass ratios and α = 0.2 C3(C16His)2/DOPE-pDNA lipoplexes at various m L / m DNA mass ratios and α = 0.2 and 0.5. and 0.5.

In In addition, addition, the the size size and and structure structure of of the the lipoplexes, lipoplexes, key key parameters parameters related related to to their their effective effective biological were determined determined by by DLS DLS and and SAXS SAXS analyses. analyses. The The hydrodynamic hydrodynamic size size (D (Dhh)) and biological activity, activity, were and polydispersity index (PDI) (PDI) values values for for C C33(C (C16 16His) polydispersity index His)22/DOPE-pDNA /DOPE-pDNAcomplexes complexeswith withtwo twodifferent different pDNA pDNA plasmids plasmids (pEGFP-C3 (pEGFP-C3 and and pCMV-Luc) pCMV-Luc) are are collected collected in in Table Table 11 at at two two different different molar molar fractions fractions of of the the mixed ratios of of the the lipoplex. lipoplex. The obtained sizes sizes (D (Dhh == 135–185 mixed lipid, lipid, and and at at two two effective effective charge charge ratios The obtained 135–185 nm), nm), together together with with the the low low polydispersity polydispersity indexes, indexes, indicate indicate that that the the prepared prepared lipoplexes lipoplexes are are suitable suitable to to potentially cross the cell membrane. potentially cross the cell membrane. Table 1. Size Size (D (Dhh)) and His) 2/DOPE-pDNA lipoplex at Table 1. and polydispersity polydispersity index index (PDI) (PDI) values valuesof ofthe theCC33(C (C1616 His) lipoplex 2 /DOPE-pDNA at two effective charge ratios(ρ(ρ andatattwo twomolar molarfractions fractionsof of the the C C33(C (C1616His) His)2 2cationic cationiclipid lipid in in the the two effective charge ratios eff), and eff ), C33(C (C1616His) His)2/DOPE mixed lipid, = 0.2 and = 0.5. C mixed lipid, αα = 0.2 and α =α 0.5. 2 /DOPE Lipoplex Lipoplex

ρeff =ρ4eff = 4

Dh (nm) Dh (nm)PDIPDI

α = 0.2 α = 0.2 (C16His)2/DOPE-pEGFP-C3 C3 (C16 His)C23/DOPE-pEGFP-C3 170 C3 (C16 His)C23/DOPE-pCMV-Luc 177 (C16His)2/DOPE-pCMV-Luc α = 0.5 α = 0.5 C3 (C16 His)C23/DOPE-pEGFP-C3 147 (C16His)2/DOPE-pEGFP-C3 C3 (C16 His)C23/DOPE-pCMV-Luc 154 (C16His)2/DOPE-pCMV-Luc

ρeff = 10 ρeff = 10 DhD(nm) PDI PDI h (nm)

170 0.170.17 177 0.270.27

136 136 119 119

0.11 0.15

0.11 0.15

147 0.120.12 154 0.190.19

177 177 185 185

0.40 0.26

0.40 0.26

Errorsare are less less than Errors than10%. 10%.

SAXS diffractograms lipoplexes, with with the the SAXS diffractograms were were also also recorded recorded for forCC 3(C His)22/DOPE-pDNA /DOPE-pDNA lipoplexes, 3 (C 1616His) pEGFP-C3 plasmid at a wide range of molar compositions (α = 0.2, 0.4, 0.5, and 0.7) and effective pEGFP-C3 plasmid at a wide range of molar compositions (α = 0.2, 0.4, 0.5, and 0.7) and effective charge ratios ratios of of the thelipoplex lipoplex(ρ (ρeffeff==1.5, 1.5,2.5, 2.5,and and4).4). Figure where intensity is plotted charge Figure 2, 2, where thethe intensity at ρateffρ=eff4 =is4plotted vs. vs. the momentum transfer vector (q), shows the Bragg peaks alldiffractograms the diffractograms correlate the momentum transfer vector (q), shows that that the Bragg peaks in allinthe correlate well well with the Miller indexes (100) (200), which typicalofofa amultilamellar multilamellar(L (Lαα) )lyotropic lyotropic crystal crystal with the Miller indexes (100) andand (200), which areare typical phase (the (theplots plotsfor forρeff ρeff= = and provided in Figure inSupplementary the Supplementary Materials). phase 1.51.5 and 2.52.5 areare provided in Figure S2 inS2the Materials). This This lamellar structure be explained a sandwich-typephase, phase,with withalternating alternating bilayers bilayers of of lamellar structure can can be explained as as a sandwich-type C (C His) /DOPE mixed lipid and the aqueous monolayer where the plasmid DNA and counterions C33(C1616His)2/DOPE mixed lipid and the aqueous monolayer where the plasmid DNA and counterions 2 are located located (Scheme (Scheme 2). 2). The The qq factors factors of of the the Bragg Bragg peaks peaks allow allow calculation of the the interlayer periodic are calculation of interlayer periodic distance ( d  2πn / qn 00 ), which results from the sum of the lipid bilayer (dm) and aqueous monolayer

Nanomaterials 2018, 8, 1061

9 of 18

Nanomaterials 2018, 8, x FOR PEER REVIEW Nanomaterials 2018, 8, x FOR PEER REVIEW

9 of 18 9 of 18

distance (d = 2πn/qn00 ), which results from the sum of the lipid bilayer (dm ) and aqueous monolayer (dw) thicknesses. The obtained values for the periodicity (d) ranged from 5.7 to 6.7 nm, in agreement (d thicknesses. The The obtained obtained values values for for the the periodicity (d) (d) ranged ranged from from 5.7 5.7 to to 6.7 6.7 nm, nm, in in agreement agreement (dw w)) thicknesses. with those previously reported [52,64] for periodicity other mixed lipid systems formed by DOPE (18C) and a with those previously reported [52,64] [52,64] for mixed lipid formed by by DOPE DOPE (18C) (18C) and aa with those lipid previously reported for other other mixedlength lipid systems systems formed cationic containing an alkyl chain of similar (16C). Interestingly, a decreaseand of the cationic lipid containing anan alkyl chain of similar lengthlength (16C). (16C). Interestingly, a decrease of the interlayer cationic lipid containing alkyl chain of similar Interestingly, a decrease of interlayer distance d was observed (from 6.7 to 5.7 nm) with decreasing GCL molar fraction αthe (i.e., distance d distance was observed (from 6.7 to 5.7 nm) with GCL molar fraction (i.e., when the interlayer d was observed (from 6.7 to 5.7 decreasing nm) with decreasing GCL molarα fraction α (i.e., when the C3(C16His) 2 content decreases and that of DOPE increases). This fact can be explained as a C3 (C16the His)C23(C content and that ofand DOPE increases). This fact canThis be explained as a combination when 16His)decreases of DOPE fact can as a combination of: (a)2 content a thinnerdecreases bilayer (i.e., athat smaller dm) asincreases). α increases, because C3be (C16explained His)2 is slightly of: (a) a thinner bilayer (i.e., a smaller d ) as α increases, because C (C His) is slightly shorter than m a smaller dm) as α increases, 3 16 combination (a) a thinner (i.e., because C3(C16His)of 2 is slightly shorter thanof:DOPE; and/or bilayer (b) a higher compaction of pDNA, leading to a2 reduction dw with the DOPE; and/or (b) a and/or higher (b) compaction of pDNA, leading to aleading reduction ofreduction dw with the increasing shorter than DOPE; a higher compaction of pDNA, to a of d with the increasing α. Assuming Tanford’s model [65–67] and keeping in mind the partial whydrophilic α. Assuming Tanford’s model [65–67] and keeping in mind the partial hydrophilic character of the increasing Assuming Tanford’s model [65–67] and keeping in mind the partial hydrophilic character α. of the C3(C16His) 2 spacer, which contains -NH and C=O groups as well as 16 carbon atoms C (C His) spacer, which contains -NH and C=O groups as well as 16 carbon atoms in each of the 3 16 character 3(C16His)2 spacer, which contains -NH and C=O groups as well as 16 carbon atoms in each of of2 the the Ctwo hydrophobic chains, the mixed lipid bilayer (C3(C16His)2/DOPE) should have a twoeach hydrophobic chains, the mixed lipid the bilayer (C3lipid (C16 His) should have a thickness d of 2 /DOPE) in of the two hydrophobic chains, bilayer (C3(Cmonolayer 16His)2/DOPE) a thickness dm of around 3.8 nm, thus leavingmixed 2–3 nm for the aqueous (dw), should which ishave a mspace around 3.8 nm, thus leaving 2–3 nm for the aqueous monolayer (d ), which is a space suitable to house w thickness m of around 3.8 nm, thus leaving 2–3 nm for the aqueous monolayer (dw), which is a space suitabledto house the pDNA. the pDNA. suitable to house the pDNA. (100) (100)

I(a.u.) I(a.u.)

(100) (100)

(100) (100)

(100) (100)

1 1

eff = 4 eff = 4 (200)  = 0.7 (200)  = 0.7

(200)  = 0.5 (200)  = 0.5

(200) (200)

(200) (200)

 = 0.4  = 0.4

 = 0.2  = 0.2

2

q 2(nm-1) q (nm-1)

3 3

Figure 2. SAXS diffractograms of C3(C16His)2/DOPE-pDNA lipoplexes at an effective charge ratio ρeff Figure 2. 2. SAXS diffractograms diffractogramsof ofCC33(C (C1616 His)2 /DOPE-pDNA lipoplexes at an effective charge ratio Figure His) lipoplexes at an effective charge ratio ρeff = 4 andSAXS several molar fractions of the C3(C2/DOPE-pDNA 16His)2 cationic lipid in the C3(C16His)2/DOPE mixed lipid = 4 and several molar fractions of the C16 in C the C163His) (C162His) =ρeff4(α). and several molar fractions of the C3(C His) 2 cationic lipidlipid in the 3(C /DOPE mixedmixed lipid 3 (C 16 His) 2 cationic 2 /DOPE lipid (α). (α).

Scheme Lamellar structure of Cthe 3(C 16His) 2/DOPE-pDNA lipoplex showing the phase, Scheme 2. 2. Lamellar structure of the His) lipoplex showing (a)(a) the gelgel phase, 3 (CC16 2 /DOPE-pDNA Scheme 2.the Lamellar structure the C3(C 16His)2/DOPE-pDNA lipoplex showing (a) the gel phase, and (b)(b) the fluid phase of the C3 of (C mixed reaching thethe gel-to-fluid and fluid phase of the C316 (CHis) 16His) 2/DOPE mixedlipid lipidbilayer bilayerafter after reaching gel-to-fluid 2 /DOPE and (b) the fluid phase(Tof C 3(C16His)2/DOPE mixed lipid bilayer after reaching the gel-to-fluid transition temperature InIn the the interlayer periodic distance (d)(d) results from thethe sum of of transition temperature (T).the m). thefigure, figure, the interlayer periodic distance results from sum m transition temperature (T m). In the figure, the interlayer periodic distance (d) results from the sum of thethe lipid bilayer (d ) and aqueous monolayer (d ) thicknesses. lipid bilayer m(dm) and aqueous monolayer w (dw) thicknesses. the lipid bilayer (dm) and aqueous monolayer (dw) thicknesses.

Fluorescence anisotropy may bebe correlated with the fluidity ofof a lipid bilayer [53,68], a property Fluorescence anisotropy may correlated with the fluidity a lipid bilayer [53,68], a property Fluorescence anisotropy may be correlated with the fluidity of a lipid bilayer [53,68], a property closely related to the transfection activity, since the fluidity of a lipoplex is directly related totoitsits closely related to the transfection activity, since the fluidity of a lipoplex is directly related closely related to the the transfection activity, since the fluidity of a lipoplex isdetermined directly related to its capability toto cross cellular membrane. The fluorescence anisotropy (r)(r) was byby rotation capability cross the cellular membrane. The fluorescence anisotropy was determined rotation capability to cross the cellular membrane. The fluorescence anisotropy (r) was determined by rotation of a fluorescent probe, DPH, allowed in the bilayer of the lipoplexes. The degree of rotation of the ofDPH a fluorescent probe, DPH, allowed in thebilayer bilayerfluidity of the and, lipoplexes. The degree of rotation of the probe should increase with the lipid thus, the probe anisotropy should also DPH probe should increase with the lipid bilayer fluidity and, thus, the probe anisotropy should also

Nanomaterials 2018, 8, 1061

10 of 18

of a fluorescent probe, DPH, allowed in the bilayer of the lipoplexes. The degree of rotation of the of 18 lipid bilayer fluidity and, thus, the probe anisotropy 10 should also decrease. Figure 3 reports the anisotropy data for the DPH probe allowed in the lipid bilayer decrease. Figure 3 reports the anisotropy data for the DPH probe allowed in the lipid bilayer of of C3 (C16 His) 2 /DOPE-pDNA lipoplexes containing the pEGFP-C3 plasmid at two different molar Cfractions 3(C16His)of 2/DOPE-pDNA lipoplexes containing plasmid at two molar the mixed lipid bilayer (α = 0.2 and the 0.5) pEGFP-C3 and an effective charge ratiodifferent of the lipoplex fractions of the mixed lipid bilayer (α = 0.2 and 0.5) and an effective charge ratio of the lipoplex of ρeff of ρeff = 4, as a function of the temperature. The obtained r values indicate that the fluorescence =anisotropy 4, as a function of thewith temperature. The obtained values to indicate that the anisotropy decreased the temperature, whichr points an increase in fluorescence the lipid bilayer fluidity decreased with the temperature, which points to an increase in the lipid bilayer fluidity (Scheme (Scheme 2). The gel-to-fluid transition temperature, determined as shown in the inset of Figure2).3, The gel-to-fluid transition temperature, determined inboth the inset 3, following thea following the Phillips method [69], was Tm = (26 as ± shown 1) ◦ C at α = of 0.2Figure and 0.5, suggesting ◦ Phillips method [69], was Tmtemperature = (26 ± 1) °C both α = 0.2 0.5, that, suggesting a fluid bilayer at fluid bilayer at physiological (37at C). It must be and pointed in the whole temperature physiological temperature (37 °C). It must be pointed that, in the whole temperature range range analyzed, the anisotropy values were lower than 0.2, which is the typical threshold analyzed, for a fluid the anisotropy values were than 0.2, which the typical low threshold for a fluidtemperature, membrane. membrane. Therefore, theselower anisotropy values wereisparticularly at physiological Therefore, these rendering anisotropythe values were low at physiological temperature, r430 = 0.05– r430 = 0.05–0.08, C3 (C16 His)2particularly /DOPE nanovector potentially attractive for application as 0.08, rendering the C 3 (C 16 His) 2 /DOPE nanovector potentially attractive for application as a gene a gene nanocarrier in vitro and in vivo assays. nanocarrier in vitro and in vivo assays. Nanomaterials 8, x FOR PEER REVIEW DPH probe2018, should increase with the

0.0001

0.16 d3r/dT3

T m = 26.5 ºC 0.0000

-0.0001

0.12

20

T(ºC)

30

40

r430

10

0.08

10

20

30

40

T(ºC)

Figure 3. Fluorescence anisotropy at 430 nm (r430 ) of the DPH fluorescent probe against the temperature for the C at an effective charge ρeff = 4 and two against C3 (C16 His) Figure 3.3 (C Fluorescence anisotropylipoplexes at 430 nm (r430 ) of the DPHratio fluorescent probe the2 16 His)2 /DOPE-pDNA cationic lipidfor molar (ρ2/DOPE-pDNA = 0.2 in red andlipoplexes α = 0.5 in black) in the C3 (C His) /DOPE mixed lipid. temperature the fractions C3(C16His) at an effective charge ratio ρ eff = 4 and two 16 2 The inset determination of the gel-to-fluid (Tin following the 2Phillips C 3(C16 His)shows 2 cationic lipid molar fractions (ρ = 0.2 intransition red and αtemperature = 0.5 in black) C3(C16His) /DOPE m ) the method. Errors lessshows than 3%. mixed lipid. Theare inset determination of the gel-to-fluid transition temperature (Tm) following the Phillips method. Errors are less than 3%.

Bearing in mind that nucleic acids are susceptible to degradation by the DNases present in serum, the ability of this gene vectoracids to efficiently compacttoand protect pDNA against degradation by Bearing in mind that nucleic are susceptible degradation by the DNases present in DNase I was evaluated. Figure 4 reports the gel electrophoresis results for C (C His) /DOPE-pDNA 3 16against 2 degradation serum, the ability of this gene vector to efficiently compact and protect pDNA lipoplexes plasmidsFigure (pEGFP-C3 and pCMV-Luc), at two molarresults compositions = 0.22/DOPEand 0.5) by DNase Iwith wastwo evaluated. 4 reports the gel electrophoresis for C3(C(α16His) and two effective charge ratios (ρ = 4 and 10). The naked plasmid was used as the control in the(α first eff pDNA lipoplexes with two plasmids (pEGFP-C3 and pCMV-Luc), at two molar compositions = lane (notice the characteristic fluorescent bands of DNA), whereas the second lane shows the DNA 0.2 and 0.5) and two effective charge ratios (ρeff = 4 and 10). The naked plasmid was used as the control digested DNase I addition (no bands fluorescent bands are present). Accordingly, the in the first(and lanedegraded) (notice theafter characteristic fluorescent of DNA), whereas the second lane shows presence of DNA bands in the other lanes (3–6) confirms that C (C His) /DOPE-pDNA successfully 2 the DNA digested (and degraded) after DNase I addition 3(no16fluorescent bands are present). limited the access of DNaseofI to the compacted DNA. Notice this protection efficient for both Accordingly, the presence DNA bands in the other lanesthat (3–6) confirms thatwas C3(C 16His)2/DOPEplasmids, mostly at limited ρeff = 10.the access of DNase I to the compacted DNA. Notice that this protection pDNA successfully was efficient for both plasmids, mostly at ρeff = 10.

Nanomaterials Nanomaterials 2018, 8,2018, 10618, x FOR PEER REVIEW

11 of 18

11 of 18

Figure 4. pDNA protection assay against degradation by DNase I (gel electrophoresis experiments):

Figure 4. pDNA protection assay against degradation by DNase I (gel electrophoresis experiments): (a) pCMV-Luc plasmid, and (b) pEGFP-C3 plasmid. In both experiments: lane 1, pDNA control; lane (a) pCMV-Luc plasmid, and (b) pEGFP-C3 plasmid. In both experiments: lane 1, pDNA control; lane 2, pDNA–DNase I; lanes 3–6, C3(C16His)2/DOPE-pDNA lipoplexes at two molar fractions of the 2, pDNA–DNase I; lanes 3–6, C3 (C lipoplexes molar 16 His) 2 /DOPE-pDNA C3(C16His)2 cationic lipid in the mixed lipid (lanes 3–4, α = 0.2; lanes 5–6, αat= two 0.5) and two fractions effective of the C3 (C16charge His)2 cationic lipid in the mixed lipid (lanes 3–4, α = 0.2; lanes 5–6, α = 0.5) and two effective ratios of the lipoplex lanes 3 and 5, ρeff = 4; lanes 4 and 6, ρeff = 10). charge ratios of the lipoplex lanes 3 and 5, ρeff = 4; lanes 4 and 6, ρeff = 10). 3.3. In Vitro Biological Activity of C3(C16His)2/DOPE-pDNA Lipoplexes

3.3. In Vitro Biological Activity of C3 (C16 His)2 /DOPE-pDNA Lipoplexes

The C3(C16His)2/DOPE-pDNA lipoplexes were then evaluated in in vitro experiments, in order to assess the factors influencing their efficiency and safety and establish the optimum conditions for The C3 (C16 His) 2 /DOPE-pDNA lipoplexes were then evaluated in in vitro experiments, in order in vivo experiments. The transfection efficiency was evaluated on the COS-7 and HeLa cell lines in to assess the factors influencing their efficiency and safety and establish the optimum conditions for the presence of 10% of serum. Figure 5a shows the luciferase expression (for plasmid pCMV-Luc) in in vivoterms experiments. The transfection efficiency was evaluated on the COS-7 and HeLa cell lines in of ng of luciferase/mg of protein, as obtained from luminometry. Figures 5b and S3 report the the presence of 10% of serum. Figure pEGFP-C3), 5a shows the luciferase expression (for plasmid pCMV-Luc) in transfection efficiency (for plasmid obtained by FACS and expressed in terms of mean terms of ng of luciferase/mg of protein, obtained luminometry. Figures 5b and S3for report the fluorescence intensity per cell (MFI), in as both cell linesfrom (see Figure S3 of Supplementary Materials the FACS results expressed in terms of %GFP).obtained The experiments were carried out at two molarof mean transfection efficiency (for plasmid pEGFP-C3), by FACS and expressed in terms fractions of the mixed lipid, α = 0.2 and 0.5, and at two effective charge ratios of the lipoplex, ρ eff = 4 fluorescence intensity per cell (MFI), in both cell lines (see Figure S3 of Supplementary Materials for the and 10. Additionally, all the results were compared against the universal positive control FACS results expressed in terms of %GFP). The experiments were carried out at two molar fractions Lipofectamine 2000 (Lipo2000*). The results in Figure 5a show that, when transfecting plasmid of the mixed lipid, α = 0.2 and 0.5, was andclearly at twobetter effective charge ratios ofthe thecontrol lipoplex, ρeff =at4 and 10. pCMV-Luc, transfection efficacy than that obtained with Lipo2000* Additionally, theαresults theatuniversal positive 2000 ρeff = 4 atall both = 0.2 andwere 0.5 incompared COS-7 cellsagainst (especially α = 0.5), and at α = 0.2control in HeLaLipofectamine cells; on the other hand, results at ρshow eff = 10 that, remain comparable to that ofplasmid the control Lipo2000* intransfection the (Lipo2000*). The the results inobtained Figure 5a when transfecting pCMV-Luc, and better considerable in the HeLa cells,the at the two molar fractions. efficacyCOS-7 was cells clearly than lower that obtained with control Lipo2000* atOtherwise, ρeff = 4 atFigure both α = 0.2 5b shows that, when assessing transfection efficiency using plasmid pEGFP-C3: (a) in the COS-7 cells, and 0.5 in COS-7 cells (especially at α = 0.5), and at α = 0.2 in HeLa cells; on the other hand, the all MFI results were comparable to those of the control, and interestingly, the MFI outcome at ρeff = 4 results and obtained at ρeff = 10 better remain comparable to that of the control Lipo2000* in the COS-7 cells and α = 0.2 was slightly than that obtained with the control Lipo2000*; and (b) in the HeLa cells, considerable lower in the HeLa cells, theattwo fractions. Otherwise, Figure 5b shows that, MFI outcomes were higher at ρeff = 4at than ρeff =molar 10 for both molar fractions (α = 0.2 and 0.5), but when assessing transfection efficiency using pEGFP-C3: (a) in the COS-7 all that MFI results comparatively lower to those obtained for plasmid COS-7 and for the control Lipo2000*. Thecells, picture emerges from results Figure 5 is that transfection efficiency is clearly dependent on= the were comparable tothe those of in the control, and interestingly, the MFI outcome at ρeff 4 plasmid, and α = 0.2 was the cell line, the effective charge ratio of the lipoplex, and the molar fraction of the mixed lipid slightly better than that obtained with the control Lipo2000*; and (b) in the HeLa cells, MFI outcomes constituting the gene carrier. In any case, the optimum formulations, i.e., with performances superior were higher at ρeff = 4 than at ρeff = 10 for both molar fractions (α = 0.2 and 0.5), but comparatively to those of Lipo2000*, were: (a) for plasmid pCMV-Luc, ρeff = 4 at both molar fractions (α = 0.2 and lower to obtained forρCOS-7 and for the control Lipo2000*. The picture that emerges from the 0.5)those in COS-7 cells, and eff = 4 at α = 0.2 in HeLa cells; and (b) for plasmid pEGFP-C3, ρeff = 4 at α = 0.2 results ininCOS-7 Figure 5 is that transfection efficiency is clearly achieved dependent the plasmid, the cell line, cells. Additionally, the higher levels of transfection at theon optimum formulations the effective charge ratio oflipids the lipoplex, molar fraction of interesting the mixedresult lipidinconstituting the by the new synthetized compared and to thethe control Lipo2000* is an terms of improving non-viral vectors for gene delivery, Lipo2000* is one of the best gene carrier. In the anyefficacy case, of the optimum formulations, i.e., given with that performances superior to those of controls used compare transfection efficacy. In4an to correlate these Lipo2000*, were: (a)tofor plasmid pCMV-Luc, ρeff = atattempt both molar fractions (αbiological = 0.2 andactivity 0.5) in COS-7 results with the structural information previously commented, it is remarkable that the lipoplexes cells, and ρeff = 4 at α = 0.2 in HeLa cells; and (b) for plasmid pEGFP-C3, ρeff = 4 at α = 0.2 in COS-7 herein studied are organized in multilamellar (Lα) lyotropic liquid phases at the two molar fractions

cells. Additionally, the higher levels of transfection achieved at the optimum formulations by the new synthetized lipids compared to the control Lipo2000* is an interesting result in terms of improving the efficacy of non-viral vectors for gene delivery, given that Lipo2000* is one of the best controls used to compare transfection efficacy. In an attempt to correlate these biological activity results with the structural information previously commented, it is remarkable that the lipoplexes herein studied are organized in multilamellar (Lα ) lyotropic liquid phases at the two molar fractions checked (i.e., α = 0.2 and 0.5), that show a similar fluid bilayer given the very low values of fluorescence anisotropy

Nanomaterials 2018, 8, 1061 Nanomaterials 2018, 8, x FOR PEER REVIEW Nanomaterials 2018, 8, x FOR PEER REVIEW

12 of 18 12 of 18 12 of 18

checked = 0.2All and that show fluid givenfractions the very(0.2 lowand values of obtained in (i.e., both α cases. this0.5), evidence seemsatosimilar reinforce thatbilayer both molar 0.5) drive checked (i.e.,anisotropy α = 0.2 and 0.5), that show a similar bilayer given the very that lowboth values of fluorescence obtained in both cases. All thisfluid evidence seems to reinforce molar to formulations that are potentially valid for transfection purposes. fluorescence obtained in both cases. evidence seems reinforce that both molar fractions (0.2anisotropy and 0.5) drive to formulations thatAll are this potentially valid forto transfection purposes. fractions40 (0.2 and 0.5) drive to formulations that are potentially valid for transfection purposes. a) a)

40

b) b)

40 40

30 30

20

MFIMFI

ng ng luciferase/mg protein luciferase/mg protein

30 30

20

10

10

10

0 0

20 20

10

Lipo2000* eff eff   Lipo2000* eff  = 0.2eff  = 0.2

0

eff   eff eff   eff  = 0.5

0

Lipo2000* eff eff   Lipo2000* eff  = 0.2eff  = 0.2

 = 0.5

eff   eff eff   eff  = 0.5  = 0.5

Figure 5. Transfection efficiency levels of C3 (C16 His)2 /DOPE-pDNA lipoplexes in HeLa (solid Figure Transfection efficiency levels of Cmolar 3(C16His) 2/DOPE-pDNA in HeLa (solid bars) bars) and 5.COS-7 cells (dashed bars) at two fractions of the Clipoplexes 3 (C16 His)2 cationic lipid in the Figure 5. Transfection efficiency levels of C 3(C16His)2/DOPE-pDNA lipoplexes in HeLa (solid bars) and His) COS-7 cells (dashed bars) at two molar fractions of the C3(C16His)2 cationic lipid in the C3 (C 16 2 /DOPE mixed lipid (α = 0.2 and 0.5): (a) in terms of ng of luciferase/mg of protein for and COS-7 cells (dashed bars) at two molar fractions of the C3(C16His)2 cationic lipid in the C3(C16 His) 2/DOPE mixed lipid (α = 0.2 and 0.5): (a) in terms of ng of luciferase/mg of protein for plasmid pCMV-Luc, and (b) in terms of mean fluorescence intensity (MFI) for plasmid pEGFP-C3. C 3(C16His)2/DOPE mixed lipid (α = 0.2 and 0.5): (a) in terms of ng of luciferase/mg of protein for plasmid pCMV-Luc, and (b) in terms mean fluorescence intensity (MFI) for plasmid pEGFP-C3. The experiments were performed in theofpresence of 10% of fetal bovine serum (FBS). The green and plasmid pCMV-Luc, and (b) in terms of presence mean fluorescence intensity (MFI) for(FBS). plasmid pEGFP-C3. The experiments were performed in the of 10% of fetal bovine serum The green and blue bars correspond to effective charge ratios ρ = 4 and 10 of the lipoplex, respectively. Gray eff The wereto performed in the presence fetal bovine serumrespectively. (FBS). The green blueexperiments bars correspond effective charge ratios ρeffof= 10% 4 andof10 of the lipoplex, Gray and bar: bar: Lipo2000* as the positive control. The data represent the mean ± SD of three wells and are blue bars correspond to effective charge eff = 4 and 10 of the lipoplex, respectively. Gray bar: Lipo2000* as the positive control. Theratios data ρrepresent the mean ± SD of three wells and are representative of three independent experiments. Lipo2000* as the positive control. experiments. The data represent the mean ± SD of three wells and are representative of three independent representative of three independent experiments.

Furthermore, as vectors in biological studies are required to be both effective and safe for cells, the Furthermore, as vectors in biological studies are required to be both effective and safetested for cells, toxicityFurthermore, of the C3 (C16as His) lipoplexes forrequired the COS-7 and HeLa cell lines by an 2 /DOPE-pDNA vectors in biological studies are to be both effective andwas safe for cells, the toxicity of the C 3(C16His)2/DOPE-pDNA lipoplexes for the COS-7 and HeLa cell lines was tested alamarBlue assay. The results are reported in Figure 6 (and Figure S4 of Supplementary Materials) for the toxicity of the Cassay. 3(C16His)2/DOPE-pDNA lipoplexes for the COS-7 and HeLa cell lines was tested by an alamarBlue The results are reported in Figure 6 (and Figure S4 of Supplementary both plasmids (pCMV-Luc and All formulations exceeded a viability of 80%, by an alamarBlue The pEGFP-C3). results are reported in FigureAll 6 (and Figure S4exceeded ofpercentage Supplementary Materials) for bothassay. plasmids (pCMV-Luc and pEGFP-C3). formulations a viability typically considered the threshold for cell safety, with values that in some cases reached 95% viability, Materials) both typically plasmidsconsidered (pCMV-Luc pEGFP-C3). formulations exceeded a viability percentagefor of 80%, theand threshold for cellAll safety, with values that in some cases even greater than those shownconsidered by the control Lipo2000*, thus confirming that the that C3 (Cin16some His)2 /DOPE percentage of 80%, typically the threshold for cell safety, with values reached 95% viability, even greater than those shown by the control Lipo2000*, thus confirmingcases that gene carrier is viability, a promising forthose the shown transfection both Lipo2000*, DNA plasmids, pEGFP-C3 and reached evencandidate greater than by the of control thus confirming that the C3(C95% 16His)2/DOPE gene carrier is a promising candidate for the transfection of both DNA pCMV-Luc, in both cell lines, COS-7 and HeLa. It should be noted that the lack of toxicity indicates the C 3(C16His) 2/DOPE gene carrier is a promising candidate for the transfection of both DNA plasmids, pEGFP-C3 and pCMV-Luc, in both cell lines, COS-7 and HeLa. It should be noted that the plasmids, pEGFP-C3 and that pCMV-Luc, inobserved both cell lines, COS-7 andlipoplexes HeLa. It should beboth noted thatcontrol the that theofdifferent transfection efficiency between both and the positive lack toxicity indicates the different transfection efficiency observed between lipoplexes lack of toxicity indicates that the different transfection efficiency observed between both lipoplexes Lipo2000* can be attributed to a different behavior in terms of interaction with the cell, which could and the positive control Lipo2000* can be attributed to a different behavior in terms of interactionbe and positive control Lipo2000* can be attributed to into alipid. different in terms of interaction related to entry the cell byimproved the novelentry GCL withthe theimproved cell, which couldinto be related to the cell behavior by the novel GCL lipid. 100

a)

100

100

a)

100

80

80

80

80

% Viability % Viability

% Viability % Viability

with the cell, which could be related to improved entry into the cell by the novel GCL lipid.

60 60 40 40

60 60 40 40

20

20

20

20

0 0

b)

b)

0

Lipo2000* Lipo2000*

 = 0.2  = 0.2

 = 0.5  = 0.5

0

Lipo2000* Lipo2000*

 = 0.2  = 0.2

 = 0.5  = 0.5

Figure 6. 6.Cell cells in inthe thepresence presence C316(C lipoplexes 16 His) 2 /DOPE-pDNA Figure Cellviability viability of of COS-7 COS-7 cells of of C3(C His) 2/DOPE-pDNA lipoplexes at two at Figure 6. Cell viability of COS-7 cells in the presence of C 3 (C 16 His) 2 /DOPE-pDNA lipoplexes at two two molar fractions of the C163His) (C162His) lipid in3(C the C3 (C mixed lipid = 0.2 molar fractions of the C3(C cationic lipid in the C 16His) 2/DOPE lipid (α = 0.2 and(α0.5): 2 cationic 16 His)mixed 2 /DOPE molar fractions of the C 3 (C 16 His) 2 cationic lipid in the C 3 (C 16 His) 2 /DOPE mixed lipid (α = 0.2 and 0.5): and with pCMV-Luc plasmid and (b) withplasmid. pEGFP-C3 and blue bars (a)0.5): with(a) pCMV-Luc plasmid and (b) with pEGFP-C3 Theplasmid. green andThe bluegreen bars correspond to (a) with pCMV-Luc plasmid with plasmid. TheGray greenbar: andLipo2000* blue barsbar: correspond to as correspond to effective charge ratios 4pEGFP-C3 and 10 of respectively. the lipoplex, respectively. Gray Lipo2000* effective charge ratios ρeff = 4and and(b) 10ρeff of =the lipoplex, as the positive ratios eff = 4 and 10 of the lipoplex, respectively. Gray bar: Lipo2000* as the positive theeffective positivecharge control. Theρdata represent the mean ± SD of three wells and are representative of three independent experiments.

Nanomaterials 2018, 8, x FOR PEER REVIEW

13 of 18

control. The data represent the mean ± SD of three wells and are representative of three independent experiments. Nanomaterials 2018, 8, 1061 13 of 18

Several works [1,3,18,22,37,38,46,47] reported in the literature have involved lipoplexes that Several works [1,3,18,22,37,38,46,47] reported in the and literature have involved lipoplexes that consist of a cationic lipid with one or two alkyl chains monovalent or multivalent cationic consist of a cationic lipid with one or two alkyl chains and monovalent or multivalent cationic imidazole or amino acid head groups, but gemini cationic lipids with histidine-based head groups imidazole or amino been acid head groups, but gemini cationic lipidsSince with the histidine-based head groups have not previously explored as potential non-viral vectors. biological activities in the have not have previously been explored as potential non-viral vectors. Since biological in literature been reported with plasmids and/or cell lines different thanthe those used in activities the present the literature have been reported with plasmids and/or cell lines different than those used in the work, the performances of the gene vector determined herein are compared with those of a lipoplex present work, the performances of the was geneused vector areplasmid compared those previously reported by us [33], which to determined transfect theherein pEGFP intowith HeLa cellsofata lipoplexmolar previously reported by us [33], which was used to transfect thethe pEGFP plasmid into HeLa several fractions of the mixing lipids and charge effective ratios of lipoplex, similar to those cells at several fractions the gemini mixing lipids andvector chargeofeffective ratios system of the lipoplex, similar selected in themolar present work.ofThe cationic that mixed (bis(hexadecyl to those selected in the present work. gemini cationic vector of that mixed system (bis(hexadecyl imidazolium) propane), referred as CThe 3(C16 Im)2), had a comparable gemini structure to that of the imidazolium) as Cwas Im)2 ), had comparablelipid gemini structure to that of the the present work Cpropane), 3(C16His)2,referred and DOPE as aacoadjuvant in both systems. Thus, 3 (C16present present work C3 (C16 His)lipoplex wasMFI present as α a coadjuvant in both Thus, the C 3(C16Im) 2/DOPE-pDNA showed = 20 (at = 0.2 and ρefflipid = 4) and 75% systems. cell viability in the 2 , and DOPE C3 (C16cell Im)2line /DOPE-pDNA showed MFI 20 (at α obtained = 0.2 and in ρeff the = 4)present and 75%work cell viability HeLa in presencelipoplex of serum, while the= results for the in3(C the HeLa cell line in presence the present for the C 16His) 2/DOPE-pDNA lipoplex of areserum, MFI = while 25 (at the α = results 0.2–0.5 obtained and ρeff = in 4) and 95% cellwork viability. In C3 (C16 His)2the /DOPE-pDNA are MFI = head 25 (atgroups α = 0.2–0.5 andhistidine ρeff = 4) and 95% cell viability. conclusion, substitution lipoplex of two imidazole by two residues in the GCL In conclusion, the substitution of two imidazole head groups by two histidine residues in the structure provokes a slight improvement of transfection efficiency and a moderate increase ofGCL cell structure(in provokes a slight improvement of transfection efficiency and a and moderate increase of cell viability HeLa cells), confirming that amino acid residues in general, histidine residues in viability (incontribute HeLa cells), that amino in general, residues in particular, to confirming a better cellular uptakeacid andresidues endosomal releaseand of histidine DNA to cytoplasm, particular,due contribute to a better cellular uptake and endosomal releaseatof DNA to cytoplasm, probably probably to the protonation of the imidazole group of histidine cellular pH [5]. due to the considering protonationthese of theresults, imidazole of to histidine at cellular pHnew [5]. formulations could also After and group in order prove whether these After results, and in order to prove whether gene, these experiments new formulations enhance theconsidering transfectionthese activity of complexes carrying a therapeutic with acould very also enhance the transfection of complexes carrying awere therapeutic gene, experiments with a potent anti-tumoral cytokineactivity gene (pCMV-interleukin-12) performed. Figure 7 provides very potent on anti-tumoral (pCMV-interleukin-12) were Figure provides information the activitycytokine of C3(C16gene His)2/DOPE-DNA (pCMV-IL12) inperformed. the COS-7 cell line,7expressed information onpCMV-IL12 the activity of (C16At His) (pCMV-IL12) in theefficacy COS-7 cell line, are expressed in terms of ng perC3mL. α 2=/DOPE-DNA 0.2 and ρeff = 10 the transfection results below in terms of ng pCMV-IL12 per however, mL. At α =those 0.2 and ρeff = 10 results are below those of the Lipo2000* control; obtained at the α = transfection 0.5 and both efficacy ρeff = 4 and 10, and at α = those ofρthe control; those obtained = 0.5 and both ρeffresults = 4 and 10,not and at α = 0.2 0.2 and eff =Lipo2000* 4, are greater thanhowever, or comparable to those at of αLipo2000*. These are surprising andcomparison ρeff = 4, are with greater or obtained comparable to those of Lipo2000*. results given are notthat surprising in in thethan data using the reporter gene These pCMV-Luc, different comparison with the adata obtained reporter gene similar pCMV-Luc, giveninthat different plasmids, plasmids, even with similar size, using do notthe always display behavior terms of transfection even with a similar size, do not always display similar behavior in terms of transfection activity. activity.

ng IL12 / mL

4

3

2

1

0

Lipo2000*

= 0.2

= 0.5

Figure 7. Transfection activity His)22/DOPE-pDNA /DOPE-pDNAlipoplexes lipoplexesininCOS-7 COS-7cells cellsat at two two molar molar Figure activity of of CC33(C (C16 16His) fractionsof ofthe theCC lipid in Cthe His)2 /DOPE mixed (α =0.5) 0.2 carrying and 0.5) fractions 3(C 16His) 2 cationic lipid in the 3(CC 16His) 2/DOPE mixed lipid (α lipid = 0.2 and 3 (C 16 His) 2 cationic 3 (C16 carryingpCMV-IL12. plasmid pCMV-IL12. and blue bars correspond tocharge effective charge plasmid The greenThe andgreen blue bars correspond to effective ratios ρeff =ratios 4 andρeff 10=of4 and 10 of the lipoplex, respectively. Gray bar: Lipo2000* as the positive control. The data represent the lipoplex, respectively. Gray bar: Lipo2000* as the positive control. The data represent the meanthe ± mean ± SDwells of three and are representative of three independent experiments. SD of three andwells are representative of three independent experiments.

The overall overall in in vitro conclusion that that The vitro results results reported reported in in the the present present work work allow allow the the conclusion C33(C (C1616His) His) lipoplexes maybebeinteresting interestingasasnanocarriers nanocarriersfor fornon-viral non-viral gene gene therapy therapy 2 /DOPE-DNA C 2/DOPE-DNA lipoplexes may applications, as illustrated through transfection efficiency with pEGFP-C3 and pCMV-Luc plasmids, and for nucleic acid immunotherapy with pCMV-IL12. Likewise, this system affords high cell viability

Nanomaterials 2018, 8, 1061

14 of 18

independent of the compacted plasmid, being therefore a potential candidate for study in future in vivo experiments. 4. Conclusions A novel biocompatible gemini cationic lipid (C3 (C16 His)2 ) bearing histidine residues was synthesized, and the mixed lipid together with the DOPE helper lipid was used as a gene vector for three DNA plasmids. The C3 (C16 His)2 /DOPE-pDNA lipoplexes were characterized by biophysical and biological studies, and the delivery efficiency was evaluated by in vitro experiments. The zeta potential results revealed that, while the C3 (C16 His)2 lipid presented an effective charge (qeff,C3 (C16 His)2 = 1.8 ± 0.1) close to the whole positive nominal charge (+2), the pDNA plasmid only exhibited around 14% of its negative nominal charge (−2/bp). Agarose gel electrophoresis assays confirmed that the C3 (C16 His)2 /DOPE vector satisfactorily compacted and protected pDNA against DNase I degradation. DLS measurements showed that the formed C3 (C16 His)2 /DOPE-pDNA complexes presented sizes ranging 120–190 nm, being thus very suited for efficient gene delivery. The SAXS diffractograms of the C3 (C16 His)2 /DOPE-pDNA lipoplexes were consistent with a lamellar structure based in a sandwich-type phase, with alternating layers of C3 (C16 His)2 /DOPE mixed lipids and an aqueous monolayer containing the pDNA and counterions. The low fluorescence anisotropy values indicated a fluid bilayer at physiological temperature, making the C3 (C16 His)2 /DOPE nanovector potentially attractive for application as a gene carrier in in vitro and in vivo studies. The cytotoxicity studies showed that the C3 (C16 His)2 /DOPE-pDNA complexes were well tolerated by both COS-7 and HeLa cells. The in vitro biological experiments allowed us to conclude that the optimum formulations of the C3 (C16 His)2 /DOPE-pDNA lipoplexes are highly efficient and biocompatible as nucleic acid nanocarriers in COS-7 and HeLa cells, with results comparable or superior to those of the universal positive control Lipo2000*. In summary, results reported in this work prove that the C3 (C16 His)2 /DOPE-pDNA lipoplex constitutes an interesting platform for in vitro gene delivery, with high cell bioavailability, and accordingly is a potential candidate for future in vivo applications. Supplementary Materials: The following information is available online at http://www.mdpi.com/2079-4991/ 8/12/1061/s1. Details of the synthesis of intermediates and the gemini cationic lipid (C3 (C16 His)2 ); Figure S1: Characterization of the C3 (C16 His)2 gemini cationic lipid: 1 H, 13 C, and 13 C-DEPT NMR spectra, and UPLC-MS profile; determination of the effective charges of the gemini cationic lipid and plasmid DNA; and additional results: zeta potential, electrophoresis gel assays for DNA protection; Figure S2: SAXS diffractograms of C3 (C16 His)2 /DOPE-pDNA lipoplexes at effective charge ratios ρeff = 1.5 and 2.5 and several molar compositions of the C3 (C16 His)2 cationic lipid in the C3 (C16 His)2 /DOPE mixed lipid (α); Figure S3: Cell Viability of HeLa cells in the presence of C3 (C16 His)2 /DOPE-pDNA lipoplexes, at two molar compositions of the cationic lipid in the mixed lipid (α = 0.2 and 0.5) with pCMV-Luc plasmid. Green and blue bars correspond to effective charge ratios ρeff = 4 and 10 of the lipoplex, respectively. Gray bar: Lipo2000*, as positive control. The data represent the mean ± s.d. of three wells and are representative of three independent experiments; and Figure S4: Cell Viability of HeLa cells in the presence of C3 (C16 His)2 /DOPE-pDNA lipoplexes, at two molar compositions of the cationic lipid in the mixed lipid (α = 0.2 and 0.5) with pCMV-Luc plasmid. Green and blue bars correspond to effective charge ratios ρeff = 4 and 10 of the lipoplex, respectively. Gray bar: Lipo2000*, as positive control. The data represent the mean ± s.d. of three wells and are representative of three independent experiments. Author Contributions: M.M.-N. performed the whole biophysical and biological in vitro experiments and wrote the first draft of the MS. P.M.T. performed a part of the zeta potential study. L.P. and A.P. carried out the synthesis of the gemini cationic lipid (C3 (C16 His)2 ) at the IQAC-CSIC, Barcelona (Spain). L.B.-F. helped M.M.-N. to carry out some of the biological in vitro experiments (DNA protection, luminometry and cell viability). C.T.d.I. designed and supervised the biological experiments carried out in her laboratory at the University of Navarra (Spain). E.A. and E.J. (corresponding author) conceived the concept, designed the experiments, supervised the biophysical and biological results, wrote the analysis and discussion, and prepared the final version of the MS. All authors have given their approval to the final version of the manuscript. Funding: This work was supported by grants from the MINECO of Spain (contract numbers CTQ2015-65972-R, CTQ2017-88948-P, and CTQ2015-64425-C2-2-R) and the Universidad Complutense de Madrid (Spain) (project no. UCMA05-33-010). Acknowledgments: SAXS experiments were performed using the NCD11 beamline at the ALBA Synchrotron Light Facility with the collaboration of ALBA staff. Authors also thank C. Aicart-Ramos for performing the plasmid DNA amplification at the Departmento de Bioquímica y Biología Molecular I, Facultad de Química (Universidad

Nanomaterials 2018, 8, 1061

15 of 18

Complutense de Madrid, Spain) and H. Lana-Vega for his assistance in the luminometry experiments at the Departamento de Farmacia y Tecnología Farmacéutica, Facultad de Farmacia (Universidad de Navarra, Spain). Conflicts of Interest: The authors declare no conflict of interest.

Abbreviations C3 (C16 His)2 DOPE DPH FBS Lipo2000* pCMV-Luc pCMV-IL12 pEGFP-C3 q+ L+ q− pDN A

bis(N(τ),N(π)-bis(methyl)-histidine hexadecyl amide) propane 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine Diphenylhexatriene fluorescent probe Fetal Bovine Serum Control Lipofectamine 2000 in presence of serum FBS Plasmid DNA encoding Luciferase Plasmid DNA encoding Interleukin-12 Plasmid DNA encoding Green Fluorescent Protein Positive effective charge of the cationic lipid Negative effective charge of plasmid DNA per base pair (bp)

References 1. 2. 3. 4.

5.

6. 7.

8. 9.

10.

11.

12.

13.

14.

Zheng, L.T.; Yi, W.J.; Su, R.C.; Liu, Q.; Zhao, Z.G. Reducible Amino Acid Based Cationic Lipids as Highly Efficient and Serum-Tolerant Gene Vectors. Chempluschem 2016, 81, 125–134. [CrossRef] Su, R.C.; Liu, Q.; Yi, W.J.; Zheng, L.T.; Zhao, Z.G. Lipoic acid functionalized amino acids cationic lipids as gene vectors. Bioorg. Med. Chem. 2016, 26, 4692–4697. [CrossRef] [PubMed] Obata, Y.; Suzuki, D.; Takeoka, S. Evaluation of cationic assemblies constructed with amino acid based lipids for plasmid DNA delivery. Bioconjugate Chem. 2008, 19, 1055–1063. [CrossRef] Damen, M.; Cristobal-Lecina, E.; Sanmarti, G.C.; van Dongen, S.F.M.; Garcia Rodriguez, C.L.; Dolbnya, I.P.; Nolte, R.J.M.; Feiters, M.C. Structure-delivery relationships of lysine-based gemini surfactants and their lipoplexes. Soft Matter 2014, 10, 5702–5714. [CrossRef] [PubMed] Midoux, P.; Pichon, C.; Yaouanc, J.J.; Jaffres, P.A. Chemical vectors for gene delivery: A current review on polymers, peptides and lipids containing histidine or imidazole as nucleic acids carriers. Br. J. Pharmacol. 2009, 157, 166–178. [CrossRef] [PubMed] Dias, R.S.; Lindman, B. DNA Interaction with Polymers and Surfactants; Wiley & Sons: Hoboken, NJ, USA, 2008. Douat, C.; Aisenbrey, C.; Antunes, S.; Decossas, M.; Lambert, O.; Bechinger, B.; Kichler, A.; Guichard, G. A cell-penetrating foldamer with a bioreducible linkage for intracellular delivery of DNA. Angew. Chem. Int. Ed. 2015, 54, 11133–11137. [CrossRef] [PubMed] Kanazawa, T.; Yamazaki, M.; Fukuda, T.; Takashima, Y.; Okada, H. Versatile nuclear localization signal-based oligopeptide as a gene vector. Biol. Pharm. Bull. 2015, 38, 559–565. [CrossRef] Tai, Z.G.; Wang, X.Y.; Tian, J.; Gao, Y.; Zhang, L.J.; Yao, C.; Wu, X.; Zhang, W.; Zhu, Q.G.; Gao, S. Biodegradable stearylated peptide with internal disulfide bonds for efficient delivery of siRNA in vitro and in vivo. Biomacromolecules 2015, 16, 1119–1130. [CrossRef] Wang, X.Y.; Tai, Z.G.; Tian, J.; Zhang, W.; Yao, C.; Zhang, L.J.; Gao, Y.; Zhu, Q.G.; Gao, J.; Gao, S. Reducible chimeric polypeptide consisting of octa-D-arginine and tetra-L-histidine peptides as an efficient gene delivery vector. Int. J. Nanomed. 2015, 10, 4669–4690. Okuda, T.; Sugiyama, A.; Niidome, T.; Aoyagi, H. Characters of dendritic poly((L)-lysine) analogues with the terminal lysines replaced with arginines and histidines as gene carriers in vitro. Biomaterials 2004, 25, 537–544. [CrossRef] Bogacheva, M.; Egorova, A.; Slita, A.; Maretina, M.; Baranov, V.; Kiselev, A. Arginine-rich cross-linking peptides with different SV40 nuclear localization signal content as vectors for intranuclear DNA delivery. Bioorg. Med. Chem. Lett. 2017, 27, 4781–4785. [CrossRef] [PubMed] Chou, S.T.; Hom, K.; Zhang, D.; Leng, Q.; Tricoli, L.J.; Hustedt, J.M.; Lee, A.; Shapiro, M.J.; Seog, J.; Kahn, J.D.; et al. Enhanced silencing and stabilization of siRNA polyplexes by histidine-mediated hydrogen bonds. Biomaterials 2014, 35, 846–855. [CrossRef] [PubMed] Huang, Z.; Zhao, D.M.; Deng, X.; Zhang, J.; Zhang, Y.M.; Yu, X.Q. Functionalized asymmetric bola-type amphiphiles for efficient gene and drug delivery. Nanomaterials 2018, 8, 115. [CrossRef] [PubMed]

Nanomaterials 2018, 8, 1061

15. 16.

17.

18. 19. 20. 21. 22.

23.

24. 25. 26.

27. 28.

29.

30. 31.

32.

33.

16 of 18

Leng, Q.X.; Chou, S.T.; Scaria, P.V.; Woodle, M.C.; Mixson, A.J. Increased tumor distribution and expression of histidine-rich plasmid polyplexes. J. Gene Med. 2014, 16, 317–328. [CrossRef] [PubMed] Stevenson, M.; Ramos-Perez, V.; Singh, S.; Soliman, M.; Preece, J.A.; Briggs, S.S.; Read, M.L.; Seymour, L.W. Delivery of siRNA mediated by histidine-containing reducible polycations. J. Controll. Release 2008, 130, 46–56. [CrossRef] Kamaruzaman, K.A.; Moyle, P.M.; Toth, I. Peptide-based multicomponent oligonucleotide delivery systems: Optimisation of poly-L-lysine dendrons for plasmid DNA delivery. Int. J. Pept. Res. Ther. 2017, 23, 119–134. [CrossRef] Kim, J.W.; Lee, J.J.; Choi, J.S.; Kim, H.S. Electrostatically assembled dendrimer complex with a high-affinity protein binder for targeted gene delivery. Int. J. Pharm. 2018, 544, 39–45. [CrossRef] Wang, F.; Wang, Y.T.; Wang, H.; Shao, N.M.; Chen, Y.Y.; Cheng, Y.Y. Synergistic effect of amino acids modified on dendrimer surface in gene delivery. Biomaterials 2014, 35, 9187–9198. [CrossRef] Zhou, J.S.; Li, Y.; Dong, H.Q.; Yuan, H.; Ren, T.B.; Li, Y.Y. Effect of monomer sequence of poly(histidine/lysine) catiomers on gene packing capacity and delivery efficiency. RSC Adv. 2015, 5, 14138–14146. [CrossRef] Junquera, E.; Aicart, E. Cationic lipids as transfecting agents of DNA in gene therapy. Curr. Top. Med. Chem. 2014, 14, 649–663. [CrossRef] Shigeta, K.; Kawakami, S.; Higuchi, Y.; Okuda, T.; Yagi, H.; Yamashita, F.; Hashida, M. Novel histidine-conjugated galactosylated cationic liposomes for efficient hepatocyte-selective gene transfer in human hepatoma HepG2 cells. J. Controll. Release 2007, 118, 262–270. [CrossRef] [PubMed] Verma, S.K.; Mani, P.; Sharma, N.R.; Krishnan, A.; Kumar, V.V.; Reddy, B.S.; Chaudhuri, A.; Roy, R.P.; Sarkar, D.P. Histidylated lipid-modified Sendai viral envelopes mediate enhanced membrane fusion and potentiate targeted gene delivery. J. Biol. Chem. 2005, 280, 35399–35409. [CrossRef] [PubMed] Vannucci, L.; Lai, M.; Chiuppesi, F.; Ceccherini-Nelli, L.; Pistello, M. Viral vectors: A look back and ahead on gene transfer technology. New Microbiol. 2013, 36, 1–22. [PubMed] Mintzer, M.A.; Simanek, E.E. Nonviral vectors for gene delivery. Chem. Rev. 2008, 109, 259–302. [CrossRef] [PubMed] Raper, S.E.; Yudkoff, M.; Chirmule, N.; Gao, G.-P.; Nunes, F.; Haskal, Z.J.; Furth, E.E.; Propert, K.J.; Robinson, M.B.; Magosin, S. A pilot study of in vivo liver-directed gene transfer with an adenoviral vector in partial ornithine transcarbamylase deficiency. Hum. Gene Ther. 2002, 13, 163–175. [CrossRef] [PubMed] Junquera, E.; Aicart, E. Recent progress in gene therapy to deliver nucleic acids with multivalent cationic vectors. Adv. Colloid Interface Sci. 2016, 233, 161–175. [CrossRef] [PubMed] Martínez-Negro, M.; Kumar, K.; Barrán-Berdón, A.L.; Datta, S.; Kondaiah, P.; Junquera, E.; Bhattacharya, S.; Aicart, E. Efficient cellular knockdown mediated by siRNA nanovectors of gemini cationic lipids having delocalizable headgroups and oligo-oxyethylene spacers. ACS Appl. Mater. Interfaces 2016, 8, 22113–22126. [CrossRef] Barrán-Berdón, A.L.; Misra, S.K.; Datta, S.; Muñoz-Úbeda, M.; Kondaiah, P.; Junquera, E.; Bhattacharya, S.; Aicart, E. Cationic gemini lipids containing polyoxyethylene spacers as improved transfecting agents of plasmid DNA in cancer cells. J. Mater. Chem. B 2014, 2, 4640–4652. [CrossRef] Bhattacharya, S.; Bajaj, A. Advances in gene delivery through molecular design of cationic lipids. Chem. Commun. 2009, 31, 4632–4656. [CrossRef] Kirby, A.J.; Camilleri, P.; Engberts, J.; Feiters, M.C.; Nolte, R.J.M.; Soderman, O.; Bergsma, M.; Bell, P.C.; Fielden, M.L.; Rodriguez, C.L.G.; et al. Gemini surfactants: New synthetic vectors for gene transfection. Angew. Chem. Int. Ed. 2003, 42, 1448–1457. [CrossRef] Muñoz-Úbeda, M.; Misra, S.K.; Barrán-Berdón, A.L.; Datta, 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. [CrossRef] [PubMed] Misra, S.K.; Muñoz-Úbeda, M.; Datta, S.; Barrán-Berdón, 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 nano-aggregates directly formed from plasmid DNA. Biomacromolecules 2013, 14, 3951–3963. [CrossRef] [PubMed]

Nanomaterials 2018, 8, 1061

34.

35.

36.

37.

38.

39.

40.

41.

42. 43.

44. 45.

46.

47.

48. 49.

50.

51.

17 of 18

Chang, J.; Xu, X.; Li, H.; Jian, Y.; Wang, G.; He, B.; Gu, Z. Components simulation of viral envelope via amino acid modified chitosans for efficient nucleic acid delivery: In vitro and in vivo study. Adv. Funct. Mater. 2013, 23, 2691–2699. [CrossRef] Kono, K.; Ikeda, R.; Tsukamoto, K.; Yuba, E.; Kojima, C.; Harada, A. Polyamidoamine dendron-bearing lipids as a nonviral vector: Influence of dendron generation. Bioconjugate Chem. 2012, 23, 871–879. [CrossRef] [PubMed] Zhang, Y.; Chen, J.; Xiao, C.; Li, M.; Tian, H.; Chen, X. Cationic dendron-bearing lipids: Investigating structure–activity relationships for small interfering RNA delivery. Biomacromolecules 2013, 14, 4289–4300. [CrossRef] [PubMed] Karmali, P.P.; Majeti, B.K.; Sreedhar, B.; Chaudhuri, A. In vitro gene transfer efficacies and serum compatibility profiles of novel mono-, di-, and tri-histidinylated cationic transfection lipids: A structure-activity investigation. Bioconjugate Chem. 2006, 17, 159–171. [CrossRef] [PubMed] Kumar, V.V.; Pichon, C.; Refregiers, M.; Guerin, B.; Midoux, P.; Chaudhuri, A. Single histidine residue in head-group region is sufficient to impart remarkable gene transfection properties to cationic lipids: Evidence for histidine-mediated membrane fusion at acidic pH. Gene Ther. 2003, 10, 1206–1215. [CrossRef] Hwang, H.S.; Hu, J.; Na, K.; Bae, Y.H. Role of polymeric endosomolytic agents in gene transfection: A comparative study of poly(L-lysine) grafted with monomeric L-histidine analogue and poly(L-histidine). Biomacromolecules 2014, 15, 3577–3586. [CrossRef] Aldawsari, H.; Edrada-Ebel, R.; Blatchford, D.R.; Tate, R.J.; Tetley, L.; Dufès, C. Enhanced gene expression in tumors after intravenous administration of arginine-, lysine-and leucine-bearing polypropylenimine polyplex. Biomaterials 2011, 32, 5889–5899. [CrossRef] Barrán-Berdón, A.L.; Muñoz-Úbeda, M.; Aicart-Ramos, C.; Pérez, L.; Infante, M.R.; Castro-Hartmann, P.; Martín-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. [CrossRef] Kim, T.I.; Baek, J.U.; Bai, C.Z.; Park, J.S. Arginine-conjugated polypropylenimine dendrimer as a non-toxic and efficient gene delivery carrier. Biomaterials 2007, 28, 2061–2067. [CrossRef] [PubMed] Vigneron, J.P.; Oudrhiri, N.; Fauquet, M.; Vergely, L.; Bradley, J.C.; Basseville, M.; Lehn, P.; Lehn, J.M. Guanidinium-cholesterol cationic lipids: Efficient vectors for the transfection of eukaryotic cells. Proc. Nat. Acad. Sci. USA 1996, 93, 9682–9686. [CrossRef] [PubMed] Li, L.; Nie, Y.; Zhu, R.; Shi, S.; Luo, K.; He, B.; Yang, Y.; Yang, J.; Gu, Z. Preparation and gene delivery of alkaline amino acids-based cationic liposomes. Arch. Pharm. Res. 2008, 31, 924–931. [CrossRef] [PubMed] Kichler, A.; Leborgne, C.; März, J.; Danos, O.; Bechinger, B. Histidine-rich amphipathic peptide antibiotics promote efficient delivery of DNA into mammalian cells. Proc. Nat. Acad. Sci. USA 2003, 100, 1564–1568. [CrossRef] [PubMed] Heyes, J.A.; Niculescu-Duvaz, D.; Cooper, R.G.; Springer, C.J. Synthesis of novel cationic lipids: Effect of structural modification on the efficiency of gene transfer. J. Med. Chem. 2002, 45, 99–114. [CrossRef] [PubMed] Jiang, Q.; Yue, D.; Nie, Y.; Xu, X.H.; He, Y.Y.; Zhang, S.Y.; Wagner, E.; Gu, Z.W. Specially-Made Lipid-Based Assemblies for Improving Transmembrane Gene Delivery: Comparison of Basic Amino Acid Residue Rich Periphery. Mol. Pharm. 2016, 13, 1809–1821. [CrossRef] [PubMed] Ju, J.; Huan, M.L.; Wan, N.; Qiu, H.; Zhou, S.Y.; Zhang, B.L. Novel cholesterol-based cationic lipids as transfecting agents of DNA for efficient gene delivery. Int. J. Mol. Sci. 2015, 16, 5666–5681. [CrossRef] Dobbs, W.; Heinrich, B.; Bourgogne, C.; Donnio, B.; Terazzi, E.; Bonnet, M.E.; Stock, F.; Erbacher, P.; Bolcato Bellemin, A.L.; Douce, L. Mesomorphic imidazolium salts: New vectors for efficient siRNA transfection. J. Am. Chem. Soc. 2009, 131, 13338–13346. [CrossRef] Zhou, T.; Llizo, A.; Li, P.; Wang, C.; Guo, Y.; Ao, M.; Bai, L.; Wang, C.; Yang, Y.; Xu, G. High transfection efficiency of homogeneous DNA nanoparticles induced by imidazolium gemini surfactant as nonviral vector. J. Phys. Chem. C 2013, 117, 26573–26581. [CrossRef] Kumar, K.; Barrán-Berdón, A.L.; Datta, S.; Muñoz-Úbeda, M.; Aicart-Ramos, C.; Kondaiah, P.; Junquera, E.; Bhattacharya, S.; Aicart, E. A delocalizable cationic headgroup together with an oligo-oxyethylene spacer in gemini cationic lipids improves their biological activity as vectors of plasmid DNA. J. Mater. Chem. B 2015, 3, 1495–1506. [CrossRef]

Nanomaterials 2018, 8, 1061

52.

53.

54. 55. 56. 57.

58.

59.

60.

61.

62. 63. 64.

65. 66. 67. 68. 69.

18 of 18

Martinez-Negro, M.; Guerrero-Martinez, A.; Garcia-Rio, L.; Domenech, O.; Aicart, E.; de Ilarduya, C.T.; Junquera, E. Multidisciplinary approach to the transfection of plasmid DNA by a nonviral nanocarrier based on a gemini-bolaamphiphilic hybrid lipid. ACS Omega 2018, 3, 208–217. [CrossRef] [PubMed] Rodriguez-Pulido, A.; Aicart, E.; Junquera, E. Electrochemical and spectroscopic study of octadecyltrimethylammonium bromide/DNA surfoplexes. Langmuir 2009, 25, 4402–4411. [CrossRef] [PubMed] Ahmed, T.; Kamel, A.O.; Wettig, S.D. Interactions between DNA and Gemini surfactant: Impact on gene therapy: Part I. Nanomedicine 2016, 11, 289–306. [CrossRef] [PubMed] Pinazo, A.; Manresa, M.; Marques, A.; Bustelo, M.; Espuny, M.; Perez, L. Amino acid–based surfactants: New antimicrobial agents. Adv. Colloid Interface Sci. 2016, 228, 17–39. [CrossRef] [PubMed] Pinazo, A.; Angelet, M.; Pons, R.; Lozano, M.; Infante, M.R.; Perez, L. Lysine-bisglycidol conjugates as novel lysine cationic surfactants. Langmuir 2009, 25, 7803–7814. [CrossRef] [PubMed] Colomer, A.; Pinazo, A.; García, M.T.; Mitjans, M.; Viardell, M.P.; Infante, M.R.; Martínez, V.N.; Pérez, L. pH-sensitive surfactants from lysine: Assessment of their cytotoxicity and environmental behavior. Langmuir 2012, 28, 5900–5912. [CrossRef] [PubMed] Cardoso, A.M.; Morais, C.M.; Silva, S.G.; Marques, E.F.; de Lima, M.C.P.; Jurado, M.A.S. Bis-quaternary gemini surfactants as components of nonviral gene delivery systems: A comprehensive study from physicochemical properties to membrane interactions. Int. J. Pharm. 2014, 474, 57–69. [CrossRef] [PubMed] Satyal, U.; Draghici, B.; Dragic, L.L.; Zhang, Q.; Norris, K.W.; Madesh, M.; Brailoiu, E.; Ilies, M.A. Interfacially engineered pyridinium pseudogemini surfactants as versatile and efficient supramolecular delivery systems for DNA, siRNA, and mRNA. ACS Appl. Mater. Interfaces 2017, 9, 29481–29495. [CrossRef] Bustelo, M.; Pinazo, A.; Manresa, M.; Mitjans, M.; Vinardell, M.; Perez, L. Monocatenary histidine-based surfactants: Role of the alkyl chain length in antimicrobial activity and their selectivity over red blood cells. Colloids Surf. A 2017, 532, 501–509. [CrossRef] Muñoz-Úbeda, M.; Misra, S.K.; Barrán-Berdón, A.L.; Aicart-Ramos, 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. [CrossRef] 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. [CrossRef] Lyubchenko, Y.L.; Shlyakhtenko, L.S. Visualization of supercoiled DNA with atomic force microscopy in situ. Proc. Natl. Acad. Sci. USA 1997, 94, 496–501. [CrossRef] [PubMed] Barran-Berdon, A.L.; Martinez-Negro, M.; Garcia-Rio, L.; Domenech, O.; de Ilarduya, C.T.; Aicart, E.; Junquera, E. A biophysical study of gene nanocarriers formed by anionic/zwitterionic mixed lipids and pillar-5-arene polycationic macrocycles. J. Mater. Chem. B 2017, 5, 3122–3131. [CrossRef] Tanford, C. The Hydrophobic Effect: Formation of Micelles and Biological Membranes; Wiley & Sons: New York, NY, USA, 1980. Tanford, C. Micelle shape and size. J. Phys. Chem. 1972, 76, 3020–3024. [CrossRef] Tanford, C. Theory of micelle formation in aqueous solutions. J. Phys. Chem. 1974, 78, 2469–2479. [CrossRef] Borenstain, V.; Barenholz, Y. Characterization of liposomes and other lipid assemblies by multiprobe fluorescence polarization. Chem. Phys. Lipids 1993, 64, 117–127. [CrossRef] Phillips, J.N. The energetics of micelle formation. Trans. Faraday Soc. 1955, 51, 561–569. [CrossRef] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).