European Journal of Pharmaceutics and Biopharmaceutics 89 (2015) 280–289

Contents lists available at ScienceDirect

European Journal of Pharmaceutics and Biopharmaceutics journal homepage: www.elsevier.com/locate/ejpb

Research paper

Polyene-based cationic lipids as visually traceable siRNA transfer reagents Emile Jubeli a,⇑, Liji Raju b, Nada Abdul Khalique b, Natalia Bilchuk c, Cory Zegel c, Agape Chen c, Howard H. Lou c, Christer L. Øpstad d, Muhammad Zeeshan d, Hans-Richard Sliwka d, Vassilia Partali d, Philip L. Leopold c, Michael D. Pungente e,⇑ a

Université Paris-Sud, EA 401, IFR 141, Faculté de pharmacie, Châtenay Malabry, France Research Division, Weill Cornell Medical College in Qatar, Doha, Qatar c Department of Chemistry, Chemical Biology & Biomedical Engineering, Stevens Institute of Technology, Hoboken, NJ, USA d Department of Chemistry, Norwegian University of Science and Technology (NTNU), Trondheim, Norway e Premedical Unit, Weill Cornell Medical College in Qatar, Doha, Qatar b

a r t i c l e

i n f o

Article history: Received 28 August 2014 Accepted in revised form 8 December 2014 Available online 20 December 2014 Keywords: Polyene-based lipids Carotenoids Cationic liposomes Gene delivery siRNA Knockdown Traceable

a b s t r a c t Cationic lipids are promising non-viral vectors for the cellular delivery of nucleic acids. Important considerations for the development of new delivery vectors are enhanced uptake efficiency, low toxicity and traceability. Traceable gene transfer systems however typically require the inclusion of a labeled excipient, and highly sensitive imaging instrumentation to detect the presence of the label. Recently, we reported the synthesis and characterization of colored, polyene cationic phospholipidoids composed of a rigid, polyenoic acid of predetermined dimension (C20:5 and C30:9) paired with flexible saturated alkyl chains of varying lengths (12:0, 14:0, 16:0, 18:0, 20:0 carbons). Herein, the potential of these cationic phospholipids as siRNA carriers was evaluated through standard liposomal formulations in combination with a neutral helper lipid DOPE. The polyene-based lipids were compared with a standard cationic lipid for siRNA-delivery into luciferase expressing HR5-CL11 cells. Within the series of lipids screened, knockdown results indicated that polyene cationic phospholipids paired with longer saturated alkyl chains are more effective as gene transfer agents, and perform comparably with the commercial lipid EPC. Furthermore, the chromophore associated with the polyene chain allowed tracking of the siRNA delivery using direct observation. The polyene lipoplexes were tracked on both a macroscopic and microscopic level either as a single-component or as a multi-component lipoplex formulation. When combined with a reference EPC, effective knockdown and tracking abilities were combined in a single preparation. Ó 2014 Elsevier B.V. All rights reserved.

1. Introduction Gene suppression using small interfering RNA (siRNA) is a promising strategy for the treatment of inherited or acquired diseases. The development of siRNA delivery systems has taken advantage of the established research towards antisense oligonucleotide and DNA administration acquired over the past few decades; however, siRNA transfer represents distinctly different challenges. Reagents suited for larger nucleic acid molecules

⇑ Corresponding authors. Université Paris-Sud, EA 401, IFR 141, Faculté de pharmacie, F-92296 Châtenay Malabry, France. Tel.: +33 146835774; fax: +33 146835965 (E. Jubeli). Premedical Unit, Weill Cornell Medical College in Qatar, PO Box 24144, Doha, Qatar. Tel.: +974 4492 8216; fax: +974 4492 8222 (M.D. Pungente). E-mail addresses: [email protected] (E. Jubeli), mdp2001@qatar-med. cornell.edu (M.D. Pungente). http://dx.doi.org/10.1016/j.ejpb.2014.12.011 0939-6411/Ó 2014 Elsevier B.V. All rights reserved.

such as DNA, for example, may not be effective for the administration of smaller siRNA biomolecules [1]. Furthermore, the mechanism associated with siRNA targeting by non-viral vectors remains to be fully elucidated and optimized. It is thought that the siRNA cargo must ‘‘unpack’’ from the lipid–nucleic acid complex prior to interacting with the RNA-induced silencing complex (RISC) [2]. Tracking of the lipid-siRNA lipoplex during its intracellular movement may offer greater insight into the precise mechanism of non-viral siRNA delivery. A variety of cationic lipids have been employed in the self-assembly of gene transfer vectors over the past several decades [3–5]. Initially, lipids utilized for gene transfer closely resembled the phospholipids found in cell membranes, often including fully saturated or mono-unsaturated fatty acyl tails although the length of the acyl chains has shown trends that vary directly or inversely with chain length depending on the cationic head groups involved [6]. Ensuing studies broadened the spectrum of lipid tails

E. Jubeli et al. / European Journal of Pharmaceutics and Biopharmaceutics 89 (2015) 280–289

used for nucleic acid delivery to include rigid, polyunsaturated hydrophobic domains such as tetraphenylethene lipids [12], diphenylethyne and triazine dendrimers [6,13,14] as well as cholesterol, steroids, or bile acids [7–11]. Cationic lipids showed a wide range of efficacies but, thus far, have defied a simple structure–activity relationship. We have previously reported the use of polyene-based lipids for the in vitro delivery of antisense oligonucleotides [15], plasmid DNA [16] and siRNA [17]. More recently we synthesized novel series cationic polyene lipids to determine whether a structure–activity relationship could be discerned within a focused group of molecules, all containing a rigid, C20:5 and 30:9 polyene chain paired with saturated alkyl chains of differing length (Fig. 1). The physical characterization as well as the biological efficacy of those molecules, as DNA transfer reagents once again failed to offer any clear structure activity relationship [18]. The problem with establishing a clear functional relationship likely relates to the number of biological steps required to transfer DNA to the nucleus [19]. A simpler target involves delivery of inhibitory RNA (iRNA) for sequence-selective knock-down of messenger RNAs.

281

Fig. 2. Structures of co-lipid, DOPE, and commercial cationic lipid EPC.

This present report describes the formulation of C30:9 lipids in the presence of co-lipid dioleoylphosphatidylethanolamine (DOPE) (Fig. 2) together with siRNA, and transfer efficiency of these lipid-siRNA lipoplexes to cells as assessed by in vitro protein knockdowns. The relative efficiency of siRNA delivery by the polyene lipids was compared with references 1,2-dimyristoyl-snglycero-3-ethylphosphocholine (EPC), [20] and Lipofectamine2000 (Invitrogen). Throughout the studies, the characteristic orange color of the C30:9 chain was readily apparent; hence, we demonstrate how the incorporation of a polyene chain into a cationic lipid formulation affords additional advantages in following the stages of lipoplex delivery in cells and in cell lysates during sample preparation and processing. 2. Materials and methods 2.1. Materials Control cationic lipid 1,2-dimyristoyl-sn-glycero-3-ethylphosphocholine (EPC) and the neutral co-lipid 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE) were obtained from Avanti Polar Lipids (Alabaster, AL). Lipofectamine 2000 was obtained from Invitrogen Ltd. (Paisley, UK). HR5-CL11 cells were obtained from the American Type Culture Collection (ATCC, Manassas, VA). Unless otherwise stated, all other chemicals were obtained from Sigma– Aldrich Chemical Co. (St. Louis, MO). Duplex siRNA targeted to the GL2 luciferase gene (Target Sequence: 50 -CGT ACG CGG AAT ACT TCG A-30 ) and siGENOME non-targeting siRNAs (negative control) were used in this study. They were both obtained from ThermoFisher Scientific (Lafayette, CO). Synthetic cationic lipids included C30-12, C30-14, C30-16, C30-18 and C30-20, synthesized and fully characterized as previously reported [18]. 2.2. General methods The mean hydrodynamic diameter of liposome and lipoplex at various N/P (+/) molar charge ratios was measured by quasi-elastic light scattering with a Zetasizer APS (Malvern Instruments, UK). For each sample, the measurement was performed three times for 2 different batches at a temperature of 25 °C with a detection angle of 90°. The commercially available transfection reagent lipofectamine 2000 was used as a control formulation. The co-lipid used for the formulation of liposomes and lipoplexes was DOPE. As a zwitterion, the concentration of DOPE was not included in the final calculation of charge ratios when forming lipoplexes between cationic lipids and DNA. 2.3. Preparation of lipid stock solutions

Fig. 1. Structures of the synthesized cationic polyene lipids, C30-12, C30-14, C3016, C30-18 and C30-20. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

All lipids were dissolved in dichloromethane (DCM) in round bottom flasks, followed by evaporation of the organic solvent under reduced pressure resulting in thin films. Ethanol was then

282

E. Jubeli et al. / European Journal of Pharmaceutics and Biopharmaceutics 89 (2015) 280–289

added to achieve stock solutions of 1 mM. Care was taken to protect the lipids from air and ambient light. All stock solutions were stored in amber vials under an inert atmosphere (N2 or Ar) and kept at 80 °C.

incubation period at room temperature, 50 lL of Bright Glo™ working solution, prepared according to the manufacturer’s directions (Promega) were added to each well and mixed by pipetting. Luminescence was then read on a Victor Envision, high throughput plate reader.

2.4. Liposome preparation 2.8. Total protein (BCA) assay Aliquots of the ethanolic stock solutions of the cationic lipid were mixed with a neutral helper co-lipid (DOPE) in a molar ratio of 3:2 respectively, and the solvent was removed at room temperature under reduced pressure. Sterile deionized water was then added to give a final concentration of 2 mM for cationic lipid, followed by vortexing before storing the lipid mixture overnight at 4 °C to allow liposome formation. After overnight hydration, the liposome solutions were kept for 5 min in a 37° water bath and then sonicated for 30 min at 40 °C. Multicomponent liposomes containing mixtures of EPC/C30-20/ DOPE were prepared as described above. Briefly, C30-20 was combined with EPC (during the ethanolic mixture preparation) in increasing molar proportions (0%, 10%, 25% and 50%) while the overall cationic (EPC + C30-20) to neutral co-lipid (DOPE) molar ratio was maintained at 3:2. 2.5. Lipoplex formulation Lipid/siRNA lipoplexes were formulated by adding 54 ll of OPTI-MEMÒ (Gibco/life technologies, Grand Island, NY) with 6 lL of siRNA (20 lM of either GL2 or control) to give siRNA aliquots. Liposomes were diluted in OPTI-MEMÒ to get 60 lL aliquots of desired molar concentration. SiRNA aliquots were added to the microcentrifuge tubes containing the diluted liposomes, and mixtures were pipetted thoroughly and incubated for 20 min at room temperature before adding 204 lL of OPTI-MEMÒ to each formulation and applying them on the cells as described in the assay section.

Total protein content was measured using PierceÒ BCA Protein Assay (Pierce Biotechnology, Rockford, IL). Forty-eight hours after the application of lipoplexes, treated cells in the transparent 96well plate were washed with PBS, then 10 lL of passive lysis buffer (Promega) was added and the plate was incubated at room temperature for 5 min with gentle shaking. BCA working reagent (200 lL), prepared according to the manufacturer’s directions, was then added to each well and gently mixed by pipetting. Plates were incubated at room temperature for 1 h then read at 562 nm on a Victor Envision, high-throughput plate reader. A calibration curve realized from a bovine serum albumin standard solution was used to determine cellular protein content per well. 2.9. Cytotoxicity assay The cellular toxicity of the various lipoplex formulations was evaluated using the HR5-CL11 cell line and MTS Assay. Forty-eight hours after the application of lipoplexes, treated cells in the transparent 96-well plate were washed with PBS, and 20 lL of CellTiter96Ò solution (CellTiter96Ò Aqueous One Solution Cell Proliferation Assay) was added and the cells were incubated further for 1 h at 37 °C. The absorbance of converted dye, which correlates with the number of viable cells, was measured at 490 nm using a Victor Envision, high throughput plate reader. The percentage of viable cells was calculated as the absorbance ratio of treated to untreated cells. 2.10. Cell lysate preparation

2.6. Cell culture HR5-CL11, human cervix carcinoma, HeLa derivative cells stably transfected with the luciferase reporter via a tetracycline controlled transcriptional trans-activator were employed in this study. Cationic lipid mediated transfection of siRNA duplex (GL2) for specific knockdown of the luciferase transcript as well as of validated control siRNA, was performed using HR5-CL11 cells following standard methods. Briefly, cells were grown in DMEM media supplemented with 10% fetal calf serum and 100 U/mL of penicillin/streptomycin and the equivalent of 1 lg/mL doxycycline which highly stimulates Luciferase expression in the HR5-CL11 cells. Cells were seeded 24 h before transfection onto opaque and transparent 96-well plate at a density of 104 cells per well and incubated with a 5% CO2 atmosphere at 37 °C. Cells were grown to 80% confluence before being washed with PBS and incubated with 50 lL of each lipid-siRNA complex in triplicate for 4 h at 37 °C and each experiment was repeated three times. The complexes were then removed by aspiration and the cells were washed with PBS before adding 100 lL of DMEM media containing 2 lg/mL of doxycycline to each well. Cells were left to incubate for an additional 44 h. Following the incubation, cells were used for the assays of luciferase, total proteins and cytotoxicity following the protocols mentioned below. 2.7. Luciferase knockdown assay Forty-eight hours after the application of lipoplexes, treated cells in the opaque 96-well plates were washed with PBS, and lysed by adding 50 lL of Glo-Lysis™ buffer to each well. After a 15 min

For macroscopic tracking experiments cells were seeded in 6well plates and different formulations of lipoplexes prepared as mentioned above were applied on cells. Forty-eight hours after the application of lipoplexes, treated cells were washed twice with PBS, incubated for 5 min with 500 lL trypsin at 37 °C before adding 4.5 mL of fresh media, cells were then spun down to obtain a cell pellet. The trypsin-laden media was aspirated and cells were lysed by adding 1 mL of Glo-Lysis™ buffer to the pellet obtained for each N/P (+/) molar charge ratio. After 15 min of incubation at room temperature, the lysate was spun down to remove the cell debris. The lysate supernatant was then transferred into a clean, clear-welled Eppendorf tube. The UV–Vis spectra of the lysates at different charge ratios were taken on a Beckman Coulter DU 730 Life Science UV–Vis Spectrophotometer. 2.11. Light and epifluorescence microscopy For microscopic tracking experiments, lipoplex formulations of varying N:P (+/) molar charge ratios composed of C30-20/DOPE (in an overall cationic lipid:co-lipid molar ratio of 3:2, respectively) were prepared, as were lipoplexes composed of EPC/DOPE (3:2 M ratio) at N:P (+/) molar charge ratio of 1.5 as a positive control. In order to co-localize the colored lipids and the cargo siRNA, the GL2 siRNA was replaced in these formulations with Block–iT™ Alexa FluorÒ Red Fluorescent Oligo (Invitrogen/life technologies, Grand Island, NY) (kex = 555 and kem = 565). HR5-CL11 cells were plated in a transparent 96-well plate as described previously and incubated with the lipoplexes for 4 h and 24 h. Live imaging was realized using an inverted light micro-

E. Jubeli et al. / European Journal of Pharmaceutics and Biopharmaceutics 89 (2015) 280–289

scope (Olympus 1X51 – Olympus Corp., Tokyo, Japan) equipped with LED light source modules for EPI fluorescence illumination and a 510–550 nm excitation filter. Cells were also visualized using phase contrast microscopy that utilizes transmitted light in order to detect cells, and thus were capable of detecting the inherent color of the lipids. Images were captured using a 40 objective lens with a digital camera (DP72; Olympus Corp., Tokyo, Japan) and were processed with image analysis software (DP2-BSW; Olympus Corp., Tokyo, Japan). 2.12. Confocal microscopy HR5-CL11 cells were seeded on sterile round coverslips in the bottom of 24-well plate at a density of 5  104 cells/well and grown for 48 h prior to the application of lipoplexes. Formulations composed of C30-20/DOPE or EPC/DOPE at a 3:2 M ratio and 1.5 N:P (+/) charge ratio were prepared as mentioned previously and 300 lL of each formulation were added to the well and incubated for 4 h. In order to label the cell membrane after rinsing with PBS, cells were incubated for 10 min at 37 °C with a 300 lL solution (5.0 lg/mL) in PBS of Oregon GreenÒ 488 conjugate of wheat germ agglutinin (WGA) which binds to sialic acid and N-acetylglucosaminyl residues (kex = 496 nm and kem = 526 nm) (Invitrogen, United States). After rinsing again with PBS, cells were fixed with a 4% solution of paraformaldehyde (Sigma–Aldrich). Confocal laser scanning microscopy (Zeiss LSM-710, Carl Zeiss, Germany) was used for in vitro imaging. Images were acquired with Ar-laser (458 nm) for the Oregon green and DPSS-laser (561 nm) for Block–iT™ Alexa FluorÒ Red Fluorescent Oligo using Plan-Apochromat 63  1.4 NA oil immersion objective lens. Fluorescence was collected by using a 505–550 nm band pass filter for Oregon green and a long pass filter beginning at 560 nm for Block–iT™ Alexa FluorÒ Red Fluorescent siRNA. 3. Results Five cationic lipids differing in the length of a second saturated alkyl chain, represent a new class of visible cationic polyene lipids [18]. The aim of this study was to investigate their siRNA binding and delivery efficiency as a function of varying the length of the saturated alkyl side-chain, and their traceability throughout the formulation and delivery process.

283

Liposomes composed of EPC were smaller than all of the rigid cationic polyene lipids used in this study. Polydispersity index (PdI) was smaller than 0.3 for all the liposomes (except for C30-12). Lipoplexes were formed by combining the positively charged liposomes with negatively charged siRNA in defined nitrogen/ phosphorus (N/P, or +/) molar charge ratios of 0.5:1, 1.5:1, 3.0:1 and 5.0:1. For low molar charge ratio formulations (0.5:1 and 1.5:1) the added siRNA resulted in an increase in the lipoplex size compared with the parent liposomes, likely due to a limited irregular compacting and the formation of aggregates giving rise to PdI values, as reported in Table 2. It should be mentioned that in the highly diluted samples used in this study, DLS measurements are influenced by the presence of few large aggregates. On the other hand, higher molar charge ratio formulations were associated with smaller lipoplex size and PdI values, indicating better siRNA compacting efficiency with an increased amount of positively charged lipid available to interact electrostatically with the negatively charged siRNA. 3.2. Lipid/siRNA delivery (luciferase knockdown and cytotoxicity) The efficiency of siRNA delivery by the lipoplexes was investigated using a luciferase knockdown assay in HR5-CL11 cells, stably transfected with the luciferase reporter. While formulations based on C30-12 and C30-14 showed minor reduction in luciferase activity (10–30%) at charge ratios of 0.5 and/or 1.5, higher proportions of lipid failed to achieve luciferase knockdown. Lipids C30-16, C30-18 and C30-20 displayed significant and charge ratio-dependent knockdown, with the highest N/P ratio of 5 being most effective in each case. Among the five polyene lipids, C30-20 was the most effective, resulting in luciferase knockdown ranging between 50% and 70% of control levels. The cytotoxicity associated with the polyene lipoplexes and those of EPC (with N/P molar charge ratio ranging from 0.5 to 5), as well as with Lipofectamine2000, was determined on the HR5CL11 cell line after 48 h incubation at 37 °C with 5% CO2 using the MTS assay. The results reported in Fig. 3 reveal a charge ratio-dependent cytotoxicity associated with all lipoplex formulations. For many of the formulations that gave high knockdown activity, high cytotoxicity was also observed. Of the five novel C30:9 lipids, C30-20 exhibited the lowest levels of cytotoxicity (Fig. 3). In fact, the two lowest N/P ratio formulations containing C30-20 revealed minimal toxicity coupled with 50–60% inhibition of luciferase expression.

3.1. Liposome and lipoplex formation 3.3. Cationic polyene lipids as tracers of nucleic acid transfer Liposomes were prepared using a conventional method upon hydration of thin lipid films. DOPE was included as a neutral zwitterionic co-lipid and was paired with cationic lipids at a constant molar ratio of 3:2 (cationic lipid:co-lipid). The liposome particle size data from dynamic light scattering (DLS) (Table 1) reveal a range in average liposome diameter between 60 and 410 nm.

Table 1 Particle size (nm) and polydispersity index (PdI) of liposomes formed with cationic lipid to co-lipid ratio of 3:2.a Formulation

Size

PdI

EPC C30-12 C30-14 C30-16 C30-18 C30-20

61 409 166 146 162 291

0.3 0.5 0.2 0.2 0.3 0.3

a Mean values of size (in nm) and PdI reported from n = 3 determinations per cationic lipid as measured by DLS.

While these unique cationic polyene lipids in general exhibited a moderate ability to act as siRNA transfer reagents, with the best performance approaching the activity of the well-characterized cationic lipid EPC, their unmistakably recognizable color provides another potential advantage as a gene transfer reagent, namely, the ability to distinguish treated cell cultures with little or no technological assistance. C30-20 was subsequently chosen in tracking studies since it exhibited the greatest level of luciferase knockdown and the best cell tolerance. EPC formulated with DOPE was used as a control. 3.3.1. Macroscopic lipid tracking For the macroscopic lipid tracking, lipoplexes were prepared and applied to HR5-CL11 cells using the previously mentioned transfection protocol. The yellow-orange color of the polyene lipids lipoplexes could be observed in each set of wells that received C3020-containing lipoplexes. In contrast, wells receiving EPC/DOPE were colorless (Fig. 4A). In order to confirm the internalization of lipoplex into cells, transfection was performed with C30-20/DOPE

284

E. Jubeli et al. / European Journal of Pharmaceutics and Biopharmaceutics 89 (2015) 280–289

Table 2 Particle sizes (nm) and PdI of lipoplexes (from liposomes prepared in lipid/co-lipid mixtures of 3:2 M ratio combined with siRNA).a

a

Molar ratio

EPC/DOPE

C30-12/DOPE

C30-14/DOPE

Size

PdI

Size

PdI

Size

0.5 1.5 3 5

1681 814 2595 244

0.5 0.4 0.5 0.2

971 1065 473 409

0.6 0.7 0.1 0.2

704 1043 1916 706

C30-16/DOPE

C30-18/DOPE

C30-20/DOPE

PdI

Size

PdI

Size

PdI

Size

PdI

0.8 0.8 0.8 0.5

1144 717 166 215

0.6 0.6 0.43 0.46

1019 1036 772 1229

0.6 0.7 0.6 0.3

861 1148 789 477

0.4 0.6 0.6 0.5

Mean values of size (in nm) and PdI reported from n = 3 determinations per cationic lipid as measured by DLS.

Fig. 3. Luciferase knockdown (% of luciferase activity) and % cell viability for polyene lipoplex formulations, C30-12, C30-14, C30-16, C30-18 and C30-20, and control EPC, both formulated with co-lipid DOPE, 48 h after transfection with various N/P (+/) molar charge ratios. Relative luciferase expression is the ratio of luciferase expression for the GL2 treated cells to the control siRNA treated cells, both normalized to total protein. Data are the average of triplicates ± SE. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

B

C30-20/DOPE 0.5 C30-20/DOPE 1.5 C30-20/DOPE C30 20/DOPE 3 C30-20/DOPE 5

C

C30-20/DOPE 1.5 Cells Alone

D Opcal density (abso orbance units)

A

C30-20/DOPE = 1.5 C30-20/DOPE = 5

C30-20/DOPE 0.5 C30-20/DOPE = 3 EPC/DOPE = 5

1.2 1 0.8 0.6 0.4 0.2 0

300

350

400

450

500

550

600

Wavelength (nm) Fig. 4. Confirmation of transfection by macroscopic observation. The orange color of C30-20/DOPE formulations could be observed in each set of wells that received it, whereas in contrast, wells receiving EPC/DOPE were colorless (A). HR5-CL11 cells were transfected with lipid/DNA complexes at N/P (+/) molar charge ratios of 0.5, 1.5, 3.0 and 5.0 formed with C30-20 or EPC using DOPE as a co-lipid. Forty-eight hours post-transfection, the plates were washed twice with PBS, and cell lysates were obtained from pooled triplicate experiments, as displayed in tubes for C30-20 (B) and EPC (C) arranged from lowest to highest N/P ratio from left-to-right. Finally, the absorbance of lysates was obtained for the C30-20/DOPE transfected cells at all N/P charge ratios, and the absorbance of lysates for EPC/DOPE transfected cells at an N/P (+/) molar charge ratio of 5.0 (D). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

lipoplexes in a 6-well plate. Forty-eight hours post-transfection, cells were washed, trypsinized, pelleted and cell lysates were obtained. The lysates from the C30-20 delivery experiments revealed recovered lipids at concentrations consistent with the relative N/P charge ratios of formulations initially applied to cells (Fig. 4B). No color was detected with EPC/DOPE treated cells (Fig. 4C), and the relative absorptions obtained from the UV–Vis spectra of the lysates also proved consistent (Fig. 4D). 3.3.2. Microscopic lipid tracking For microscopic tracking, formulations were incubated in 96well plates with cells for 4 h as described above. The visualization

of lipoplexes associated with cells was performed through phase contrast microscopy 4 h and 24 h post-transfection (Fig. 5, columns B and C). To determine whether the siRNA remained associated with the carrier, a red fluorescent siRNA was employed, allowing the use of fluorescence microscopy (Fig. 5, columns A and D). Importantly, the different charge ratios were prepared using the same concentration of siRNA (same total amount of fluorescence was applied to the cells) but with increasing concentrations of cationic lipid in the formulation, i.e. higher charge ratio contains higher amount of cationic lipid. The phase contrast images taken 4 h post-transfection after washing with PBS (Fig. 5, panels B1–B4 and Fig. 6) reveal

E. Jubeli et al. / European Journal of Pharmaceutics and Biopharmaceutics 89 (2015) 280–289

285

Fig. 5. Fluorescence and phase contrast imaging of HR5-CL11 cells incubated for 4 h and 24 h with different formulations of lipoplexes (green areas in panel 5A correspond to a false color indication of camera saturation during image acquisition). Scale bar = 20 mm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

yellow-orange aggregates associated with organelles, as well as large clusters of yellow-orange aggregates outside cells. The latter structures were particularly visible at N/P (+/) molar charge ratios 3:1 and 5:1. The increased intensity of yellow-orange color associated with the higher N/P (+/) molar charge ratios is due to the greater amounts of lipid available to interact with cells. As expected, cells transfected with EPC-based lipoplexes (Fig. 5, panel B5) show a normal appearance without any distinguishable color. When the same cells were visualized using epifluorescence microscopy (Fig. 5, panels A1–A4), red fluorescence was co-localized with the orange lipids in the form of bright red concentrated inclusions indicating aggregated, cell-associated lipoplexes. It is noteworthy that cells incubated with higher charge ratios are rounded and some are dying, this is a result of the greater toxicity associated with the lipids at these concentrations, as quantified by the MTS assay (Fig. 3). Images taken 24 h post-transfection with the phase contrast microscope (Fig. 5, panels C1–C4) revealed a reduction in the

amount of visible yellow-orange clusters around cells. In this case, the cells were not re-washed before imaging. As seen in the fluorescence images (Fig. 5, panels D1–D4), the red concentrated spots seen at 4 h gave rise after 24 h mainly to a more diffuse background throughout the cytoplasm, although a few aggregates could be seen in some cells. The EPC/DOPE containing formulation (Fig. 5, panel D5) revealed an excellent diffusion in the cytoplasm of all cells with a complete disappearance of red clusters of lipoplexes, indicating a full release from the endosomes. 3.4. Confocal microscopy Oregon GreenÒ 488 conjugate of wheat germ agglutinin (WGA) which binds to sialic acid and N-acetylglucosaminyl residues was used to label the plasma membrane and thus assist to identify lipid vesicles inside cells. For confocal imaging, only cells transfected for 4 h using C30-20/DOPE and EPC/DOPE at N/P (+/) 1.5:1 were visualized. These images were obtained to confirm that vesicles

286

E. Jubeli et al. / European Journal of Pharmaceutics and Biopharmaceutics 89 (2015) 280–289

C30-20 (upper panel) and EPC (lower panel) based formulations, and are consistent with data obtained previously using phase contrast and epifluorescence microscopy. 3.5. Cationic polyene lipids as traceable additives when co-formulated with other transfection agents

Fig. 6. Higher magnification fluorescence and phase contrast images of HR5-CL11 cells incubated for 4 h with C30-20/DOPE lipoplexes in a 1.5 N/P molar charge ratio comparing localization of cell-associated polyene lipid particles (A) with cellassociated fluorescent siRNA (B). Scale bar = 20 mm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

seen with phase contrast microscopy and epifluorescence microscopy had successfully delivered siRNA to the interior of the cell rather than simply being associated with the cell surface. All images were taken at mid-height (depth) of the cells. Fig. 7 shows that the lipid-associated red fluorescent siRNA is distributed in part within the cell cytoplasm and in part on the periphery of cells prior to internalization. These observations are common for both

To determine whether the tracking properties of the polyenebased lipids could be exploited by co-formulating these colored lipids with other, more effective cationic lipid gene carriers, formulations were made including a ratio of (N/P) 1.5:1. The commercial cationic lipid EPC was co-formulated in a multi-component mixture with varying amounts of C30-20, specifically, 0%, 10%, 25% and 50% mole fraction with the balance contributed by EPC. Fig. 8 reveals that the C30-20-containing lipoplexes associated with cells were visible using phase contrast microscopy. Thus, using simple detection methods such as light microscopy and direct observation, transfected samples could be identified. Most importantly, perhaps, is the fact that the incorporation of the colored C30-20 lipid as a ‘‘tracking additive or agent’’ up to 25% of the total cationic lipid content did not impact the knockdown performance or the cell tolerance of EPC (Fig. 9). 4. Discussion We had previously reported the use of polyene lipids for siRNA delivery [17] as well as a cationic polyene lipids incorporating a short second ethyl chain (synthesized using an alternative synthetic pathway) [16] for the delivery of DNA. The preliminary results obtained with those novel lipids encouraged us to develop a new series of polyene lipids and fully characterize them including their surface properties, molecular areas, molecular structure, interaction with plasmid DNA and the microstructure of the best transfecting compounds using small angle X-ray diffraction

Fig. 7. Confocal images of HR5-CL11 cells incubated for 4 h with C30-20/DOPE lipoplexes in a 1.5 charge ratio (upper panel) and with EPC/DOPE lipoplexes in a 1.5 charge ratio (lower panel). The green channel shows the plasma membrane labeled with an Oregon green 488 conjugate of wheat germ agglutinin (WGA) while the red channel shows the lipoplexes delivering Block-iT™ Alexa FluorÒ Red Fluorescent Oligo siRNA. Scale bar = 20 mm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

E. Jubeli et al. / European Journal of Pharmaceutics and Biopharmaceutics 89 (2015) 280–289

287

Fig. 8. Fluorescence and phase contrast images of HR5-CL11 cells incubated for 4 h with lipoplexes formed from EPC co-formulated in a multi-component mixture with varying amounts of C30-20: 10% (left-hand, upper and lower images) and 25% (right-hand, upper and lower images). Note that the green areas in the lower two panels represent fluorescence saturation, as described previously in the caption for Fig. 5. Scale bar = 20 mm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 9. Knockdown and cytotoxicity results for lipoplexes formed from EPC co-formulated in a multi-component mixture with 0%, 10%, 25% or 50% added C30-20 with HR5CL11 cells. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

(SAXRD) [18]. These lipids possess two chains in the hydrophobic domain, one rigid and the other flexible, as the literature clearly indicates that lipid gene carriers with two hydrophobic chains are generally more active than those with a single chain [21]. In our previous work these polyene lipids associated with DNA giving lipoplexes with lamellar structure determined by SAXRD and were able to transfect CHO-K1 cells. In the light of these findings, the goal of the present study was to compare the siRNA delivery properties of these novel cationic lipids incorporating a rigid 30:9 polyene chain balanced by a saturated, flexible alkyl chain of varying length. The unique properties of these colored compounds led to a series of valuable macroscopic and microscopic tracking properties that are useful for liposomal preparation, lipid-nucleic acid lipoplex formation, the application onto cultured cells, and ultimately, for subcellular in gene delivery tracing. No trend was observed between liposome size and increasing length of the saturated side-chain within the polyene lipid series. Polydispersity index was lower than 0.3 for all the liposomes

(except for C30-12). This PDI is expected for liposomes prepared without extrusion and without added surfactant. Generally, lipoplex formulations of higher N/P molar charge ratios were associated with smaller particles (Table 2), higher cytotoxicity and higher knockdown efficiency (Figs. 3 and 5). Increased charge ratios were obtained by using a constant siRNA concentration with increasing concentrations of lipids. High N/P formulations contained higher amounts of particles and had more homogenous size distribution. Consequently, these formulations resulted in better luciferase knockdown (>50% knockdown), but displayed higher cytotoxicity possibly due to increased accumulation of lipids inside cells. Among the polyene-based cationic lipids, greater knockdown efficiency was generally observed with lengthening of the flexible saturated chain. The flexibility introduced by the saturated chain likely enabled more efficient packaging of the lipoplexes. However, the polyene lipids did not outperform the well-established EPC cationic lipid ruling out this simple conclusion. Instead, a more

288

E. Jubeli et al. / European Journal of Pharmaceutics and Biopharmaceutics 89 (2015) 280–289

complex relationship between packaging and unpacking is likely involved. It is clear that these rigid, linear, poly-unsaturated lipids are likely to have packaging characteristics distinct from those of saturated, mono-unsaturated, or polycyclic lipids commonly used for the formation of lipoplexes. A comparison of the knockdown results for a single N/P molar charge ratio (1.5 for example) among lipids C30-12 to C30-20 reveals different levels of efficiency although lipoplexes have similar dimension, indicating that size is not the main factor influencing knockdown efficiency. More work is needed to optimize the formulation in order to have greater control of liposomes size, and more importantly the size of lipoplexes. This perhaps could be achieved via extrusion or the use of polyethylene glycol-modified lipids. Images taken using light and epifluorescence microscopy demonstrated the association of lipoplexes with cells (Figs. 5 and 6) while confocal microscope images provided a visual indication of the presence of lipoplexes inside cells (Fig. 7). In our previous work [18] cationic polyene lipids revealed limited efficiency for the concerning in vitro delivery of plasmid DNA. C30-20 was best performing. Lipoplex formulations containing C30-20 combined with DOPE and plasmid DNA realized gene expression up to 50% of that induced by EPC/DOPE/DNA lipoplexes. Similarly, C30-20/siRNA lipoplexes showed the best performance among the family of polyene-based lipids, with RNA knockdown ranging from 40% to 70% while EPC/DOPE yielded 50–90% knockdowns, depending on N/P ratio. Thus, C30-20 in formulations with both plasmid DNA and siRNA accomplished the most successful in vitro nucleic acid delivery. Besides the ability of these cationic polyene lipids to act as gene transfer reagents, their strong recognizable color provides a great advantage through the ability to distinguish treated cell cultures and track cell-associated materials. Although polyunsaturated cationic compounds have previously been utilized in gene transfer [22,23], the potential utility of the characteristic color associated with such compounds for tracking has been underappreciated. The cationic polyene lipids used in this study were particularly useful to contribute structurally to the lipoplex permitting inclusion of a 25% mole fraction with EPC, thus conferring visual detection without interfering with the ability of EPC to yield efficient nucleic acid transfer. The use of a fluorescent siRNA confirmed the co-localization of the cationic lipids with the delivered siRNA inside and outside the cell (Figs. 5 and 6). Confocal microscopy offered a visual proof of the presence of lipoplexes inside cells. The application of these colored lipids provides a very simple tracking tool applied on live cells using light microscope without the need of prior fixing or the use of fluorescent tags and fluorescence microscopy. Hence, transfection can be positively confirmed by simple observation of treated cell culture plates, by observation of cell lysates derived from treated cells, or by microscopic observation of treated cells. Use of these lipids to track delivery in vivo is also a likely benefit given the low toxicity of some of the formulations tested. Taken together, these factors constitute a novel and useful advance. These findings suggest that the investigated cationic polyene lipids combine advantage of nucleic acid delivery capacity to eukaryotic cells with inherent tracking properties through the presence of the hydrophobic polyene chain.

5. Conclusion When this series of novel polyene nucleic acid carrier reagents, was applied to HR5-CL11 cells a structure–function relationship emerged, revealing greater knockdown efficiency with lengthening of the flexible saturated chain. The optimum in vitro siRNA

knockdown and biocompatibility were achieved with C30-20. Although siRNA delivery using C30-20 lipoplexes in HR5-CL11 cells was not superior to known gene carriers, the potential utility of a rigid chain was demonstrated. The orange color associated with the polyene chain, provides confident handling of the lipids throughout formulation as well as instant visual confirmation of treated versus non-treated cell cultures. The utility of direct visualization was demonstrated by detection of the transfection reagent macroscopically in tissue culture plates and microscopically in intact cells. By combining C30-20 with a more effective gene transfer reagent, the benefits of both compounds were realized in a single formulation. Investigations modifying the rigid hydrophobic domain (shorter or longer chain), lengthening the flexible chain and changing the lipid backbone are in progress. Conflict of interest The authors declared no conflicts of interest. Acknowledgment This publication was made possible by a Grant from the Qatar National Research Fund under the National Priorities Research Program, award NPRP08-705-3-144 (LPI M. Pungente). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the Qatar National Research Fund. Muhammad Zeeshan thanks the Higher Education Commission of Pakistan for a scholarship. References [1] A. Schroeder, C.G. Levins, C. Cortez, R. Langer, D.G. Anderson, Lipid-based nanotherapeutics for siRNA delivery, J. Internal Med. 267 (1) (2009) 9–21. [2] E. Bernstein, A.A. Caudy, S.M. Hammond, G.J. Hannon, Role for a bidentate ribonuclease in the initiation step of RNA interference, Nature 409 (6818) (2001) 363–366. [3] P.L. Felgner, T.R. Gadek, M. Holm, R. Roman, H.W. Chan, M. Wenz, J.P. Northrop, G.M. Ringold, M. Danielsen, Lipofection: a highly efficient, lipid-mediated DNA-transfection procedure, Proc. Natl. Acad. Sci. Sci. 84 (21) (1987) 7413– 7417. [4] S. Bhattacharya, A. Bajaj, Advances in gene delivery through molecular design of cationic lipids, Chem. Commun. (31) (2009) 4632–4656. [5] G.Y. Wu, C.H. Wu, Receptor-mediated in vitro gene transformation by a soluble DNA carrier system, J. Biol. Chem. 262 (10) (1987) 4429–4432. [6] X. Gao, L. Huang, A novel cationic liposome reagent for efficient transfection of mammalian cells, Biochem. Biophys. Res. Commun. 179 (1) (1991) 280–285. [7] J.K. Guy-Caffey, V. Bodepudi, J.S. Bishop, K. Jayaraman, N. Chaudhary, Novel polyaminolipids enhance the cellular uptake of oligonucleotides, J. Biol. Chem. 270 (52) (1995) 31391–31396. [8] J.S. Choi, E.J. Lee, H.S. Jang, J.S. Park, New cationic liposomes for gene transfer into mammalian cells with high efficiency and low toxicity, Bioconjug. Chem. 12 (1) (2001) 108–113. [9] A. Aissaoui, N. Oudrhiri, L. Petit, M. Hauchecorne, E. Kan, M. Sainlos, S. Julia, J. Navarro, J.P. Vigneron, J.M. Lehn, P. Lehn, Progress in gene delivery by cationic lipids: guanidinium-cholesterol-based systems as an example, Curr. Drug Targets 3 (1) (2002) 1–16. [10] P.E. Kish, Y. Tsume, P. Kijek, T.M. Lanigan, J.M. Hilfinger, B.J. Roessler, Bile acidoligopeptide conjugates interact with DNA and facilitate transfection, Mol. Pharm. 4 (1) (2006) 95–103. [11] W.P. Goldring, E. Jubeli, R.A. Downs, A.J. Johnston, N. Abdul Khalique, L. Raju, D. Wafadari, M.D. Pungente, Novel macrocyclic and acyclic cationic lipids for gene transfer: synthesis and in vitro evaluation, Bioorg. Med. Chem. Lett. 22 (14) (2012) 4686–4692. [12] C. Zhang, S. Jin, S. Li, X. Xue, J. Liu, Y. Huang, Y. Jiang, W.-Q. Chen, G. Zou, X.-J. Liang, Imaging intracellular anticancer drug delivery by self-assembly micelles with aggregation-induced emission (AIE Micelles), ACS Appl. Mater. Interfaces 6 (7) (2014) 5212–5220. [13] D. Joester, M. Losson, R. Pugin, H. Heinzelmann, E. Walter, H.P. Merkle, F. Diederich, Amphiphilic dendrimers: novel self-assembling vectors for efficient gene delivery, Angew. Chem. Int. Ed. 42 (13) (2003) 1486–1490. [14] O.M. Merkel, M.A. Mintzer, D. Librizzi, O. Samsonova, T. Dicke, B. Sproat, H. Garn, P.J. Barth, E.E. Simanek, T. Kissel, Triazine dendrimers as nonviral vectors for in vitro and in vivo RNAi: the effects of peripheral groups and core structure on biological activity, Mol. Pharm. 7 (4) (2010) 969–983. [15] L.J. Popplewell, A. Abu-Dayya, T. Khana, M. Flinterman, N. Abdul Khalique, L. Raju, C.L. Opstad, H.R. Sliwka, V. Partali, G. Dickson, M.D. Pungente, Novel

E. Jubeli et al. / European Journal of Pharmaceutics and Biopharmaceutics 89 (2015) 280–289 cationic carotenoid lipids as delivery vectors of antisense oligonucleotides for exon skipping in Duchenne muscular dystrophy, Molecules 17 (2) (2012) 1138–1148. [16] C.L. Opstad, H.R. Sliwka, V. Partali, A. Elgsaeter, P. Leopold, E. Jubeli, N.A. Khalique, L. Raju, M.D. Pungente, Synthesis, self-assembling and gene delivery potential of a novel highly unsaturated, conjugated cationic phospholipid, Chem. Phys. Lipids 170–171 (2013) 65–73. [17] M.D. Pungente, E. Jubeli, C.L. Opstad, M. Al-Kawaz, N. Barakat, T. Ibrahim, N. Abdul Khalique, L. Raju, R. Jones, P.L. Leopold, H.R. Sliwka, V. Partali, Synthesis and preliminary investigations of the siRNA delivery potential of novel, singlechain rigid cationic carotenoid lipids, Molecules 17 (3) (2012) 3484–3500. [18] C.L. Opstad, M. Zeeshan, A. Zaidi, H.-R. Sliwka, V. Partali, D.G. Nicholson, C. Surve, M.A. Izower, N. Bilchuk, H.H. Lou, P.L. Leopold, H. Larsen, A. Liberska, N.A. Khalique, L. Raju, M. Flinterman, E. Jubeli, M.D. Pungente, Novel cationic polyene glycol phospholipids as DNA transfer reagents: lack of a structure– activity relationship due to uncontrolled self-assembling processes., Chem. Phys. Lipids 183 (2014) 117–136.

289

[19] R.W. Horobin, V. Weissig, A QSAR-modeling perspective on cationic transfection lipids. 1. Predicting efficiency and understanding mechanisms, The J. Gene Med. 7 (8) (2005) 1023–1034. [20] R. Koynova, B. Tenchov, L. Wang, R.C. MacDonald, Hydrophobic moiety of cationic lipids strongly modulates their transfection activity, Mol. Pharm. 6 (3) (2009) 951–958. [21] Y.-P. Zhang, D. Reimer, G. Zhang, P. Lee, M. Bally, Self-assembling DNA-lipid particles for gene transfer, Pharmaceut. Res. 14 (2) (1997) 190–196. [22] I. Blanc, M. Bueno Da Costa, J. Bolard, M. Saint-Pierre Chazalet, Oligonucleotide delivery by a cationic derivative of the polyene antibiotic amphotericin B. I: interaction oligonucleotide/vector as studied by optical spectroscopy and electron microscopy, Biochim. Biophys. Acta 1464 (2) (2000) 299–308. [23] C. Garcia-Chaumont, O. Seksek, B. Jolles, J. Bolard, A cationic derivative of amphotericin B as a novel delivery system for antisense oligonucleotides, Antisense Nucleic Acid Drug Develop 10 (3) (2000) 177–184.