Design and Evaluation of Gemini Surfactant-Based Lipoplexes Modified with Cell-Binding Peptide for Targeted Gene Therapy in Melanoma Model

Purpose Achieving successful gene therapy requires delivery of a gene vector specifically to the targeted tissue with efficient expression and a good safety profile. The objective of this work was to develop, characterize and determine if a novel gemini surfactant-based lipoplex systems, modified with a cancer-targeting peptide p18-4, could serve this role. Methods The targeting peptide p18-4 was either chemically coupled to a gemini surfactant backbone or physically co-formulated with the lipoplexes. The influence of targeting ligand and formulation strategies on essential physicochemical properties of the lipoplexes was evaluated by dynamic light scattering and small angle X-ray scattering techniques. In vitro transfection activity and cellular toxicity of lipoplexes were assessed in a model human melanoma cell line. Results All lipoplexes zeta potential and particle size were optimal for cellular uptake and physical stability of the system. The lipoplexes adopted an invertedhexagonal lipid arrangement. The lipoplexes modified with the peptide showed no significant changes in physicochemical properties or lipoplex assembly. The modification of the lipoplexes with the targeting peptide significantly enhanced protein expression 2-6 fold compared to non-modified lipoplexes. In addition, p18-4 modified lipoplexes significantly improved the safety of the lipoplexes. The ability of the p18-4 modified lipoplexes to selectively express the model protein was confirmed by using healthy human epidermal keratinocytes (HEKa). Conclusion The gemini surfactant-based lipoplexes modified with p18-4 peptide showed significantly higher efficiency and safety compared to the system that did not contain a cancer targeting peptide and provided evidence for their potential application to achieve targeted melanoma gene therapy. __________________________________________________________________________________________


INTRODUCTION
In gene therapy, the efficiency of non-viral vectors depends on the successful delivery and expression of the exogenous genetic material into the targeted tissue, while causing minimum toxicity. For cancer gene therapy, non-viral gene delivery systems, which usually assemble as nano-sized particles, have the ability to non-specifically target the tumor (passive targeting) by an enhanced permeability and retention (EPR) effect, extravasation through the leaky vasculature surrounding the tumors and entrapment in the tumor mass.(1) However, passive cancer targeting can be challenged by several factors including, but not limited to, the type of cancer, the heterogeneity of the tumor, blood supply, the interaction of the nanoparticles with the tumor and surrounding tissue, and the characteristics of the delivery system. (2,3). On the other hand, cancer-specific targeting can be achieved by decorating the non-viral delivery system with cancer-specific ligands (active targeting). Such ligands can be a small-molecule moiety, protein, peptide or antibody (monoclonal antibody or fragment) that can specifically bind to an overexpressed enzyme, a receptor or a cancerspecific antigen.(4) Such targeting can enhance the accumulation and ______________________________________ interaction of the ligand-modified nanoparticles which can improve their specific cellular internalization. As a result of specific uptake by cancer cells, off-target gene expression and toxicity can be avoided or at least minimized. (5).
In cancer gene and drug delivery, different ligands have been explored for cancer specific targeting. These ligands include monoclonal antibodies such as trastuzumab to target HER2 receptor in breast cancer, peptides such as arginylglycyl-aspartic acid (RGD) to target overexpressed integrins, and small molecules such as folic acid to target the folate-binding protein.
(3) Among different peptide ligands that have been explored for cancer targeting is p160, a linear dodecapeptide (VPWMEPAYQRFL) that was identified by phage display. It displayed a superior affinity for neuroblastoma cells and selective internalization in different breast cancer cell lines but not by normal umbilical vein/vascular endothelium cells (HUVEC). (6,7) Based on the structure of p160, several linear decapeptides have been engineered with higher affinity to cancer cells (MDA-MB-435, and MCF-7) and better enzymatic stability. (8) Among these targeting peptides, p18-4 (WxEAAYQrFL) [ Figure 1] demonstrated the highest affinity and selectivity toward breast cancer cells and showed resistance to proteolytic degradation. (8) Decoration of liposomaldoxorubicin with p18-4, as a cancer targeted drug delivery system, significantly improved the cellular uptake and cytotoxicity of doxorubicin selectivity in breast cancer. (9) In mice bearing MDA-MB-435 tumour, the anti-cancer activity of the p18-4 modified liposomal-doxorubicin was improved in comparison to the unmodified liposomal system. (10).
Cationic gemini surfactants are cationic lipids that have been comprehensively evaluated and characterized for gene delivery. (11,12) The versatility in the chemical structure of gemini surfactants makes it easy to produce highly efficient gene delivery systems with a good safety profile. (13)(14)(15) For example, chemical modification in the basic structure of gemini surfactant by introducing amino acid substituents in the spacer region, producing 12-7N(GK)-12 [ Figure 1], has led to the production of an efficient biocompatible carrier. The applicability of topical gene delivery by cationic gemini surfactants was evaluated previously for dermal, vaginal and ocular applications. (13,15,16) The positive outcome provided evidence for the viability of using gemini surfactant-based lipoplexes for topical and transdermal gene delivery. Recently, we have reported the development and evaluation of an RGD-modified gemini surfactant lipoplex system that shows specificity to melanoma. (17) The introduction of the RGD moiety into the gemini surfactant structure enhanced the transfection activity by more than two fold compared to nonmodified lipoplexes. (17) The successful targeting of overexpressed integrin in a melanoma model provided evidence for the potential of using gemini surfactant-based lipoplexes for targeted cancer gene therapy.
The objective of this work was to develop and characterize a novel cationic gemini surfactant-based lipoplex decorated with a specific cancer-targeting ligand (p18-4) and to evaluate its activity in a melanoma model. As well, we investigated the influence of the ligand modification on the essential physicochemical properties of the lipoplex system that govern the transfection activity. The specificity of the developed ligand-modified lipoplexes to target melanoma cells was also examined.

Synthesis of p18-4 modified gemini surfactant
Peptide p18-4 on the chlorotrityl resin (with all side chain groups protected with Boc) was provided by Dr. Kaur's group. (8) The synthesis of gemini surfactant backbone [3-(bis 3-(N-dodecyl-N,Ndimethylamino)propyl)carbamoyl propanoic acid] 2+ 2Clwas described previously. (19) In brief, under a N2 atmosphere using standard Schlenk techniques, a Schlenk flask equipped with a magnetic stir bar was charged with 3,3′ -iminobis ( N , Ndimethylpropylamine) (1.682 g, 8.989 mmol) and succinic anhydride (0.988 g, 9.873 mmol) in 15 mL of DMF to form a homogeneous solution. After 3 days of stirring at ambient temperature, the reaction mixture was concentrated under vacuum to obtain an orange-yellow oily substance as the desired compound in quantitative yield. The compound was charged in a round-bottom flask with propanoic acid, (800.0 mg, 2.785 mmol) and 1-iodododecane (2.063 g, 1.720 mL, 6.933 mmol) in DMF (20 mL) to form a yellow homogeneous solution. After 12 h of stirring at ambient temperature, the solvent was removed under vacuum and the residue washed several times with diethyl ether to remove the excess iodododecane. The sample was dried, yielding an orange oily substance that was dissolved in 10 mL of distilled water and 2.5 equivalents of Amberlite® IRA-400(Cl) (2.400 g, 6.933 mmol) was added to the solution. After stirring for 1 h, the Amberlite resin was removed by filtration and the residue washed with water. An orange, oily substance [ 3 -( bis 3 -( N -dodecyl -N , N dimethylamino)propyl)carbamoyl propanoic acid] 2+ 2Clwas obtained in almost quantitative yield after removing excess water by freeze-drying.
The gemini backbone was activated under a N2 atmosphere in a round bottom flask, by reacting [ 3 -

Lipoplex formulations
The construction of the plasmid pGThCMV.IFN-GFP, used as a model for a robust plasmid, was described previously.(13) Plasmid DNA (coded as P in lipoplexes) was isolated and purified using QIAGEN Plasmid Giga Kit (Mississauga, ON, Canada) following the manufacturer's protocol. Aqueous solutions of 3 mM gemini surfactant/peptide were used to prepare all lipoplex formulations evaluated in this work.
Lipoplexes, Table 1, were formulated using a plasmid to gemini surfactant (phosphate/nitrogen) charge ratio of 1:10 in the presence of (DOPE) as a helper-lipid (coded as L in lipoplexes) creating plasmid/gemini surfactant/lipid lipoplexes [P.G.L], as described previously. (17) In brief, The DOPE were vesicles formed by sonication in 9.25% sucrose solution (pH 9) at a final concentration of 1 mM DOPE and filtered through Acrodisc ® 0.45 µm syringe filters (Pall Gelman, Ann Arbor, MI). The [P/G] lipoplexes were prepared by mixing an aliquot of 200 µg pDNA aqueous solution with an appropriate amount of 3 mM gemini surfactant solution and incubated at room temperature for 20 min. The [P.G.L] systems were prepared by mixing [P/G] lipoplexes with the DOPE vesicles at gemini surfactant to DOPE molar ratio of 1:10 and incubated at room temperature for 20 min.

Size and zeta potential measurements
Formulations (800 µL of each) were transferred into a special cuvette (DTS1061, Malvern Instruments, Worcestershire, UK) for size distribution and zetapotential measurements using a Zetasizer Nano ZS instrument (Malvern Instruments, Worcestershire, UK). Each sample was measured four times, and the results are expressed as an average ± standard deviation (SD) of a triplicate with corresponding polydispersity index (PDI) value (Table 2).
Small-angle X-ray Scattering (SAXS) analysis Small-angle X-ray Scattering (SAXS) measurements were carried out at the Stanford Synchrotron Radiation Lightsource, Menlo Park, CA, California. Lipoplexes prepared as described above were concentrated (10x) by speed vacuuming at 35 ˚C. A wavelength of 1.1271 Å (11KeV energy) was utilized. Samples were loaded in 1.5mm boron-rich glass capillaries (Charles Supper Company, USA). The scattered X-ray was detected on MAR225-HE (225 mm x 225 mm (3072 x 3072 pixels, pixel size 73.24 μm) at 20 s exposure time and at a sample to detector distance of 1.1 m. The SAXS detector was calibrated with silver behenate. GSASII software was used to plot diffraction intensity versus 2θ (where θ is the diffraction angle) or q (scattering vector) by radial integration of the 2D patterns.  Table 2 Physiochemical properties (size and zeta potential) of different p18-4 modified lipoplexes formulations. Size and zeta potential measurements: values are an average of n = 3 ± standard deviation (SD). The polydispersity (PDI) index is indicated for the size measurements as an average of three measurements ± STD.
Cytotoxicity assay MTT assay was used to evaluate the cellular toxicity of lipoplex systems in A375 and HEKa cell lines. A sterile solution of 4 mg/mL of MTT in PBS buffer was prepared. Cell lines were seeded on 96-well plates and transfected with lipoplexes (as described above). After 48 h, the cell lines were evaluated for cytotoxicity. The supplemented media was removed from the wells and replaced with 0.5 mg/mL MTT in supplemented media and incubated at 37 C in 5% CO2 and 95% air for 3 h. The supernatant was removed and each well washed with PBS. The purple formazan crystal that formed was dissolved in DMSO. Plates were incubated for 10 min at 37 C. Absorbance was measured at 580 nm using BioTek microplate reader (Bio-Tek Instruments, VT, USA). The cellular toxicity is expressed as a percentage of the non-transfected control cells ± SD.

STATISTICAL ANALYSIS
Statistical analyses were performed using SPSS software (Version 23.0). The results are expressed as the average of three independent experiments ± SD. One way analysis of variance (ANOVA, Dunnett's test) was used for statistical analyses. Significant differences were considered at p < 0.05 level.

RESULTS
The influence of p18-4-modification on the physicochemical characteristics of the lipoplexes Physicochemical properties that control the transfection activity and stability of lipoplexes are: particle size, surface charge and lipid assembly. (21,22) All lipoplexes formed particles with an average diameter of less than 180 nm. The average particle size of the non-modified lipoplexes (L1: [P.G.L]) was 157±2.5 nm ( Table 2 Table 2).
The influence of the addition of the targeting peptide on the lipid phase arrangement and transitions of the lipoplexes was examined using SAXS [ Figure 2]. All lipoplexes exhibited Bragg's peaks with relative ratios of 1, √3 and √4, characteristic of an inverted hexagonal phase arrangement (HII). The preparation method and presence or absence of targeting peptide-modified gemini surfactant or free targeting peptide in the lipoplex system did not cause any alteration in the lipid phase arrangement.

DISCUSSION
Gene therapy for cancer is the most extensively explored amongst worldwide gene therapy clinical trials as 64.6% ( 1,590 trials) are investigating the potential of cancer gene therapy. (23) Creating an efficient and targeted gene delivery agent with effective therapeutic outcomes and good safety profile are required for successful clinical applications. Targeting tumors with lipid-based gene delivery systems can be achieved by chemically modifying the lipoplexes with tumor specific ligands. (24)(25)(26) Gemini surfactant-based lipoplexes have been widely explored for gene delivery as they show good transfection efficiency and low cellular toxicity. The glycyl-lysine modified gemini surfactant used in this work, 12-7N(GK)-12, showed potential application as a non-invasive topical transfection agent for development of vaginal genetic vaccination. (15) In addition, by chemically modifying the 12-7N(GK)-12 with cRGD, to target over-expressed integrin receptor on melanoma, led to a significant increase of gene expression. (17) The ability of the RGD-modified lipoplexes to specifically target overexpressed α3/β1 integrin in melanoma was confirmed by using a positive competitor peptide (free RGD) and false-negative control peptide (RAD). As well, the RGD-modified gemini surfactant lipoplex did not enhance the transfection activity in normal human epidermal keratinocytes (HEKa), indicating the selectivity of the system toward the target. (17).
The nature of the nano-carriers, such as modification of the chemical structure, and formulation strategies can influence the physicochemical properties of the lipoplex system, ultimately affecting the transfection efficiency. The size, shape and overall charge of the assembled lipoplexes can influence the physicochemical stability of the gene delivery system, its biodistribution, cellular uptake, gene expression, and cellular toxicity. (27) Gemini surfactant-based lipoplexes [P.L.G] can be internalized by mammalian cells via clathrin-mediated or caveolaemediated endocytosis pathways. (28) Modification of the chemical structure of the gemini surfactant influenced the shape of the lipoplexes and hence the main endocytosis uptake mechanism. For example, lipoplexes build with 12-7NH-12 gemini surfactant formed spherical-shape nanoparticles and were internalized mainly through a clathrin-mediated pathway. On the other hand, the modification of the gemini surfactant with a dipeptide moiety in the spacer region (i.e., 12-7N(GK)-12) resulted in the formation of cylindrically shaped lipoplexes with larger particle size. This lipoplex system demonstrated pH sensitive characteristics and internalized the cells through caveolae-mediated endocytosis. (28) In the current work, co-formulation of lipoplex systems with free p18-4 (L4:  The heterogeneity of the size of these lipoplexes might have contributed to alterations in cellular uptake mechanisms, leading to an improvement in gene expression. (29) The modifications of the lipoplexes with targeting p18-4 peptide did not influence the supramolecular assembly, as all lipoplexes evaluated in this work arranged into inverted hexagonal structures (HII) [ Figure 2]. This lipid phase arrangement is believed to contribute to the ability of the gemini based lipoplexes to interact with the cell membrane and facilitate the endosomal escape of the pDNA upon endocytosis. (13,30) The use of peptide penetration enhancers has been widely explored to enhance the uptake of highly hydrophobic drugs and in gene delivery. For skin delivery, these peptides interact with skin proteins (namely keratins which are highly expressed in the skin) and enhance the transcellular uptake of drug molecules. (31) In normal skin conditions, keratins (a family of intermediate filaments) are the main structural proteins in the epidermis (mainly type II keratin KRT1 and KRT10 which mainly expressed in keratinocytes) and they show important role in maintaining skin integrity and in intracellular signaling pathways. (32) Mutations in keratins expressions are contributed to several skin hereditary disorders. (32,33) In addition, several keratins showed distinctive expression (upregulated or downregulated) in certain cancers (such as lung, breast and skin cancers) and can be used for disease diagnosis and monitoring. (32) The linear dodecapeptide peptide p18-4 is able to bind preferably to cancerous cells (MDA-MB-435 and MCF-7) and internalized upon binding through endocytosis.(9, 10) In a MDA-MB-435 xenograft mouse model, p18-4 conjugated doxorubicin showed a 5-fold reduction in tumor size compared to the free drug.(9, 10) p18-4 specifically interacts with overexpressed KRT-1 in MDA-MB-435 and MCF-7 cells. (34) In the current work, we evaluated the ability of p18-4 modified lipoplexes to target melanoma using a model cell line (A375) showing overexpression of keratins. (35,36) The increase in gene expression by the p18-4 modified lipoplexes could be attributed to the specific interaction of p18-4 with keratins and lead to improvement of cellular uptake of the lipoplexes and higher transfection activity [ Figure 3 A]. We believe that the combination of heterogeneity of particle size and the availability of free p18-4 (not covalently bound to gemini surfactant) on the surface of the lipoplexes to interact with keratins in formulations L4 and L5 can explain the significant increase in transfection compared to non-modified (L1) and chemically modified lipoplexes (L2 and L3). The results of the control study (the transfection in HEKa cells which express a normal level of keratins) indicate the targeting specificity of the p18-4 modified lipoplexes to melanoma [ Figure 4 A].

CONCLUSIONS
The results of this study show that gemini surfactant based lipoplexes modified with targeting peptide (p18-4) are able to deliver pDNA to melanoma and achieve higher gene expression with a good safety profile. The specificity of melanoma targeting for the lipoplexes was demonstrated by the higher gene expression compared to non-malignant keratinocytes. The ability of the targeted lipoplexes to interact specifically to overexpressed keratins in melanoma and to enhance their penetration into the tissue indicate the potential application of the system for topical gene delivery.