1,2-Dioleoyl-3-trimethylammonium-propane methyl sulfate salt (DOTAP) was obtained from Avanti Polar Lipids Inc. 1,2-Dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE, COATSOME; cat. no. ME-8181) was obtained from NOF Co., Ltd. L-Alanine, L-arginine hydrochloride, L-asparagine monohydrate, L-cysteine, L-glutamine, glycine, L-histidine, L-isoleucine, L-leucine, L-lysine, L-methionine, L-proline, L-serine, L-threonine and L-valine were obtained from Wako Pure Chemical Industries, Ltd. All the other chemicals used were of the highest available grade, and were obtained from commercial sources.

siRNAs targeting nucleotides of firefly luciferase (Luc siRNA) and non-targeting siRNA control (Cont siRNA) as a negative control for Luc siRNA were synthesized by Sigma Genosys. The siRNA sequences of the Luc siRNA were as follows: Sense strand, 5'-CCGUGGUGUUCGUGUCUAAGA-3' and antisense strand, 5'-UUAGACACGAACACCACGGUA-3' (19), and the siRNA sequences of the Cont siRNA were as follows: Sense strand, 5'-GUACCGCACGUCAUUCGUAUC-3' and antisense strand, 5'-UACGAAUGACGUGCGGUACGU-3' (20).

To compare the cake volumes after freeze-drying, 5 ml 100 mM alanine, arginine, asparagine, cysteine, glutamine, glycine, histidine, isoleucine, leucine, lysine, methionine, proline, serine, threonine and valine solution were transferred into a 5 ml vial, and then frozen at -80˚C, followed by drying under a high vacuum pressure (10-20 Pa) using a FDU-540 freeze-dryer (Tokyo Rikakikai Co.) equipped with a DRC-2L dry chamber (Tokyo Rikakikai Co.).

Cationic liposomes were prepared from DOTAP and DOPE at a molar ratio of 1:1(21). For the preparation of cationic liposomes using the thin-film hydration method, DOTAP and DOPE were dissolved in chloroform, and the chloroform was evaporated under vacuum on a rotary evaporator at 60˚C to obtain a thin film. The thin film was hydrated with water at 60˚C through vortexing. The hydrated liposomes were placed in an eggplant flask and sonicated using a bath-type sonicator (Bransonic^® 2510 J-MTH, 42 kHz, 100 W; Branson UL Trasonics Co.) for 5-10 min at room temperature.

To prepare cationic liposome/siRNA complexes (siRNA lipoplexes), each liposome preparation was added to siRNA at a charge ratio (+:-) of 4:1 via vortex mixing for 10 sec, leaving the solution at room temperature for 15 min. The charge ratio (+:-) of liposomes:siRNA is expressed as the molar ratio of cationic lipid to siRNA phosphate.

To measure the size of siRNA lipoplexes, siRNA lipoplexes were formed through the addition of cationic liposomes to 5 µg Cont siRNA with vortex-mixing for 10 sec and leaving the solution at room temperature for 15 min. In the preparation of freeze-dried siRNA lipoplexes, each lipoplex containing 5 µg Cont siRNA was diluted in 933 µl 100 mM amino acid solution sterilized using a 0.45-µm filter [125 µl amino acid solution per 50 pmol (0.67 µg) siRNA], and then transferred to a 6-well plate, followed by freezing at -80˚C. The frozen plates were dried under a high vacuum using a freeze-dryer. Freeze-dried siRNA lipoplexes were reconstituted to an appropriate volume (~1 ml) with water, and the particle size (cumulant average particle size) and polydispersity index (PDI) of the siRNA lipoplexes were measured using the cumulant method with a ELS-Z2 light-scattering photometer (Otsuka Electronics Co., Ltd.) at 25˚C.

Human breast cancer MCF-7 cells stably expressing firefly luciferase (MCF-7-Luc), constructed via transfection with a pcDNA3 plasmid containing the firefly luciferase gene from the plasmid psiCHECK2 (Promega Corporation), were donated by Dr Kenji Yamato of the University of Tsukuba. MCF-7-Luc cells were grown in RPMI-1640 medium (Wako Pure Chemical Industries, Ltd.) supplemented with 10% heat-inactivated FBS (Invitrogen; Thermo Fisher Scientific, Inc.) and 1.2 mg/ml G418 (Santa Cruz Biotechnology, Inc.) at 37˚C in a 5% CO₂ humidified atmosphere.

For Rev-transfection, siRNA lipoplexes were formed through the addition of cationic liposomes to 50 pmol Cont siRNA or Luc siRNA by vortexing for 10 sec, and then leaving the solutions to stand at room temperature for 15 min. Each lipoplex was diluted in a 125 µl solution containing different concentrations of amino acids (10, 25, 50, 100 or 150 mM), transferred to a 12-well plate (50 pmol siRNA/well), followed by freezing at -80˚C. The frozen plates were dried under high vacuum using a freeze-dryer and stored at room temperature in a desiccator until required. For Rev-transfection using freeze-dried siRNA lipoplexes, MCF-7-Luc cells (1x10⁵ cells) were suspended in 1 ml culture medium supplemented with 10% FBS, and this suspension was added to each well (final siRNA concentration, 50 nM). The medium, after the rehydration of siRNA lipoplexes freeze-dried with 10, 25, 50, 100 or 150 mM amino acid solutions, contained 1.25, 3.125, 6.25, 12.5 or 18.75 mM extra amino acids, respectively. A total of 48 h after transfection, the cells were lysed through the addition of 125 µl cell lysis buffer (Pierce Luciferase Cell Lysis Buffer; Thermo Fisher Scientific Inc.) and subjected to one cycle of freezing (-80˚C) and thawing (37˚C), followed by centrifugation at 15,000 x g for 10 sec at 4˚C. A total of 10 µl aliquots of the cell lysate supernatants were mixed with 50 µl PicaGene MelioraStar-LT Luminescence Reagent (Toyo Ink Mfg. Co., Ltd.), and the luminescence was measured as counts per second (cps) with a chemoluminometer (ARVO X2; Perkin Elmer, Inc.). The protein concentration of the supernatants was determined using BCA reagent (Pierce BCA Protein assay kit; Thermo Fisher Scientific Inc.), with bovine serum albumin (BSA; Sigma-Aldrich; Merck KGaA) as a standard, and the luciferase activity (cps/µg protein) was calculated. Luciferase activity (%) was calculated relative to the luciferase activity (cps/µg protein) of untransfected cells.

To investigate the effect of amino acids on luciferase expression in MCF-7-Luc cells, 125 µl 100 mM amino acid solution was transferred to a 12-well plate, followed by freeze-drying. MCF-7-Luc cells (1x10⁵) were suspended in 1 ml medium supplemented with 10% FBS and the resulting cell suspension was added to the wells. Luciferase activity was measured as described above.

Each lipoplex containing 5 pmol Cont siRNA was diluted with a 12.5 µl solution containing 100 mM amino acid, and then transferred to each well of a 96-well plate (5 pmol siRNA/well). After freezing at -80˚C, the plates were dried under a high vacuum using a freeze-dryer. MCF-7-Luc cells (1x10⁴) were suspended in 100 µl culture medium supplemented with 10% FBS, and the suspension was added to each well (final siRNA concentration of 50 nM). After 24 h of incubation, cell viability was determined using a Cell Counting Kit-8 assay (Dojindo Laboratories). The relative cell viability (%) was calculated as the percentage of the cells added to the wells without freeze-dried lipoplexes.

Data are presented as the mean ± standard deviation of at least three repeats. The statistical significance of the differences between mean values was determined using Student's t-test or a one-way ANOVA followed by a post-hoc Tukey's test in GraphPad Prism version 4.0 (GraphPad Software Inc.). P<0.05 was considered to indicate a statistically significant difference.

In the present study, 15 amino acids (alanine, arginine, asparagine, cysteine, glutamine, glycine, histidine, isoleucine, leucine, lysine, methionine, proline, serine, threonine and valine) were used as cryoprotectants during the freeze-drying of siRNA lipoplexes, and their effects on gene knockdown efficiency after Rev-transfection using freeze-dried siRNA lipoplexes were investigated. As aspartic acid, glutamic acid, phenylalanine, tyrosine and tryptophan did not completely dissolve in water at a concentration of 100-150 mM, these were excluded as potential cryoprotectants. The appearance of cakes (dry powder) after freeze-drying may be an indicator of product quality. Thus, first, a 100 mM amino acid solution was prepared and freeze-dried in a vial. Histidine and threonine formed shrunken cakes (Fig. 1B) after freeze-drying; arginine, glycine and proline formed collapsed cakes (Fig. 1C); and lysine formed puffing cakes (Fig. 1D). In contrast, the other amino acids formed large cakes (Fig. 1A), similar to previously reported disaccharides (7,8). This result suggests that different amino acid types affect the appearance of cakes after freeze-drying.

Reverse transfection is a method in which siRNA lipoplexes are attached to the bottom of cell culture plates through freeze-drying, then, at the time of transfection, a cell suspension is added to the culture plate (7,8). To assess the gene-knockdown effects of siRNA lipoplexes freeze-dried in the presence of amino acids, freeze-dried siRNA lipoplexes in 12-well plates were reconstituted with MCF-7-Luc cells suspended in culture medium. Here, DOTAP was used as a cationic lipid and DOPE as a helper lipid, and cationic liposomes were prepared at a molar ratio of 1:1(8). In our previous study, it was reported that cationic liposomes composed of DOTAP and DOPE could efficiently deliver siRNA into the cells via conventional transfection (forward transfection) and strongly suppress the expression of target genes (>80% knockdown) (7). siRNA lipoplexes were formed at a charge ratio (+:-) of 4:1, diluted with solutions containing different concentrations of amino acids, and then added into the wells of a 12-well plate, followed by freeze-drying (Fig. S1). When siRNA lipoplexes were freeze-dried without amino acids, the lipoplexes with Luc siRNA did not suppress luciferase activity after Rev-transfection (Fig. 2A-C). However, increasing concentrations of amino acids used in the freeze-drying of siRNA lipoplexes were associated with increased gene-knockdown activity after Rev-transfection. Freeze-drying of lipoplexes containing Luc siRNA in the presence of 50-100 mM amino acids strongly suppressed luciferase activity regardless of the amino acid type. Interestingly, luciferase activity after Rev-transfection with lipoplexes of Cont siRNA differed depending on the amino acid used. The results could be divided into three groups: The presence of amino acids in freeze-drying that exhibited increased luciferase activity (Fig. 2A), decreased luciferase activity (Fig. 2B) or no notable change in luciferase activity (Fig. 2C). Increasing concentrations of alanine, glutamine, glycine, serine, threonine or valine in the freeze-drying of siRNA lipoplexes increased luciferase activity (~200% compared with the untreated cells) after Rev-transfection with lipoplexes of Cont siRNA (Fig. 2A). In contrast, the presence of arginine, cysteine or lysine during freeze-drying decreased luciferase activity at increasing concentrations, whereas at concentrations >100 mM, luciferase activity decreased to ~20% compared with the untreated cells (Fig. 2B). However, the presence of asparagine, histidine, isoleucine, leucine, methionine or proline during freeze-drying did not notably affect luciferase activity (Fig. 2C). To confirm the effects of amino acids on luciferase activity, MCF-7-Luc cells were added to each well of a multi-well plate, which was freeze-dried with 100 mM amino acids without siRNA lipoplexes (Fig. 3). As a result, the extra amino acids affected luciferase activity similarly to those after Rev-transfection with the lipoplexes of Cont siRNA freeze-dried with amino acids (Fig. 2), indicating that the presence of extra amino acids in the culture medium may affect luciferase expression or luciferase activity.

Cell viability was measured 24 h after Rev-transfection with freeze-dried siRNA lipoplexes into MCF-7-Luc cells. The siRNA lipoplexes freeze-dried with arginine, isoleucine and leucine showed slightly increased cytotoxicity (67-74% cell viability), whereas those freeze-dried with lysine exhibited strong cytotoxicity (11% cell viability) (Fig. 4). In contrast, the siRNA lipoplexes freeze-dried with the other amino acids did not induce a significant cytotoxic effect.

To examine whether the amino acids used during freeze-drying affected the size of siRNA lipoplexes after rehydration, freeze-dried siRNA lipoplexes in the presence of 100 mM amino acid solutions were prepared, and the sizes of the resulting siRNA lipoplexes after rehydration were measured. The cationic liposomes were ~100 nm in size, and the siRNA lipoplexes were ~180 nm (data not shown) before freeze-drying. After freeze-drying and rehydration, the sizes of the siRNA lipoplexes ranged from 190-2,900 nm, although the lipoplexes freeze-dried with arginine, histidine, leucine and lysine were aggregated (Table I). The siRNA lipoplexes freeze-dried with cysteine, glycine, isoleucine, methionine or valine had larger sizes (800-2,900 nm). In contrast, siRNA lipoplexes freeze-dried with alanine, asparagine, glutamine, proline, serine or threonine were relatively smaller (~340, ~370, ~360, ~540, ~420 and ~190 nm, respectively) with a monodisperse distribution (0.17-0.24 in PDI). Amongst these, polar but uncharged hydrophilic amino acids (asparagine, glutamine, serine and threonine) appeared effective for production of stable siRNA lipoplexes in size after freeze-drying (200-400 nm), indicating that amino acids that possessed an -OH or -CONH₂ moiety in the side chain may interact with the polar head groups on the surface of the siRNA lipoplexes, and prevent the aggregation of siRNA lipoplexes caused by freeze-drying. From these results, the types of amino acids used during freeze-drying affected the size of the siRNA lipoplexes after rehydration.