Cationic gemini lipids with cyclen headgroups: Interaction with DNA and gene delivery abilities

Article abstract of DOI:10.1039/c4ra05974c

A series of novel 1,4,7,10-tetraazacyclododecane (cyclen)-based gemini cationic lipids were synthesized, and l-cystine was used as backbone between the two amphiphilic units. The liposomes formed from the lipids and DOPE could efficiently condense plasmid DNA into nanoparticles with suitable size and zeta-potentials, which might be suitable for gene transfection. These lipids were applied as non-viral gene delivery vectors, and their structure-activity relationship was studied. It was found that both the hydrophobic tails and the linking group could largely influence the transfection efficiency, and the oleylamine derived lipid gave the best transfection results, which were close to the commercially available transfection reagent lipofectamine 2000. The gemini structure would favor the gene transfection, and the transfection efficiency of the gemini lipid was much higher than the mono counterpart. Besides, these lipids have very low cytotoxicity, suggesting their good biocompatibility. Results indicate that such gemini lipids might be promising non-viral gene delivery vectors. This journal is

Full text of DOI:10.1039/c4ra05974c

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Cationic gemini lipids with cyclen headgroups:
interaction with DNA and gene delivery abilities†
Cite this: RSC Adv., 2014, 4, 44261
Yi-Mei Zhang, Yan-Hong Liu, Ji Zhang,* Qiang Liu, Zheng Huang and Xiao-Qi Yu*
A series of novel 1,4,7,10-tetraazacyclododecane (cyclen)-based gemini cationic lipids were synthesized,
and L-cystine was used as backbone between the two amphiphilic units. The liposomes formed from the
lipids and DOPE could efficiently condense plasmid DNA into nanoparticles with suitable size and zeta-
potentials, which might be suitable for gene transfection. These lipids were applied as non-viral gene
delivery vectors, and their structure–activity relationship was studied. It was found that both the
hydrophobic tails and the linking group could largely influence the transfection efficiency, and the
oleylamine derived lipid gave the best transfection results, which were close to the commercially
available transfection reagent lipofectamine 2000. The gemini structure would favor the gene
transfection, and the transfection efficiency of the gemini lipid was much higher than the mono
counterpart. Besides, these lipids have very low cytotoxicity, suggesting their good biocompatibility.
Results indicate that such gemini lipids might be promising non-viral gene delivery vectors.
Received 19th June 2014
Accepted 10th September 2014
DOI: 10.1039/c4ra05974c
www.rsc.org/advances
nanoparticles with smaller size, facilitating the endocytosis of
the lipid/DNA complex (lipoplex). Up to now, some gemini
1
. Introduction
9
The key factor toward efficient gene therapy is the development lipids have been found to exhibit higher TE than commercially
1
0,11
of ideal gene carriers, which may deliver therapeutic genes into available lipofectamine 2000.
They also showed good serum
1
target cells. Gene carriers are mainly divided into two types: tolerance in the transfection, suggesting their good
12
viral and non-viral. Although viral vectors have relatively high biocompatibility.
transfection efficiency (TE), their biosafety-related disadvan-
It has been demonstrated that the gene delivery efficiencies
tages such as immunogenicity, oncogenic effects and potential of gemini lipids mainly depend on the molecular architectures
mutagenesis limit their applications. On the other hand, non- of the polar head groups, hydrophobic tails,
viral vectors have been receiving numerous attentions over the of backbone as well as spacer.
2
13
14,15
and the nature
16
17,18
In our previous studies,
3
past decades for their safety, easy preparation and modica- 1,4,7,10-tetraazacyclododecane (cyclen) was rst applied as
tion, and reproducibility. Some new strategies to enhance the headgroup in the design of efficient non-viral gene delivery
2
4
19–21
gene transfection were also developed. As an important class of vectors.
It was found that the unique cyclic structure of
22
non-viral vectors, cationic lipids have many advantages such as cyclen may facilitate DNA binding and lower the cytotoxicity of
simple structures, good repeatability, biocompatibility, and cationic lipids. In this report, we would like to introduce a series
potential commercial value. Since the rst study of DOTMA as of novel cyclen-based cationic gemini lipids with cystine as
5
12
cationic lipid gene vector by Felgner in 1987, various types of backbone. Various hydrophobic tails were introduced via
6
lipids and liposomes for gene delivery were reported.
amide or ester bonds. For comparison, mono counterpart was
Gemini lipid, which was rst raised by Menger and Littau in also prepared. Their interactions with plasmid DNA together
7
1
991, refers to the amphiphilic molecules with two identical with the structure–activity relationship (SAR) in the gene
simple surfactants linked by a spacer. As a result, gemini lipids transfection process were studied.
are always consist of at least two hydrophilic headgroups and
8
two hydrophobic tails. Due to its special structure, gemini
lipids need a relatively lower amount to bind DNA compared to 2. Experimental section
their mono counterparts, and the lower dosage would lead to
lower toxicity. These lipids may condense DNA into
2
.1. Materials and methods
All chemicals and reagents were obtained commercially and
were used as received. Anhydrous acetonitrile, dichloromethane
Key Laboratory of Green Chemistry and Technology (Ministry of Education), College of
Chemistry, Sichuan University, Chengdu 610064, PR China. E-mail: xqyu@scu.edu.cn;
jzhang@scu.edu.cn; Fax: +86-28-85415886
(DCM) and chloroform (CHCl ) were dried and puried under
3
nitrogen by using standard methods and were distilled imme-
diately before use. [4,7,10-Tris(tert-butoxycarbonyl)-1,4,7,10-
tetraazacyclododecan-1-yl]acetic acid (tri-boc -cyclen- acetic
†
Electronic supplementary information (ESI) available. See DOI:
0.1039/c4ra05974c
1
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 44261–44268 | 44261

View Article Online
RSC Advances
Paper
23
acid, compound 5, Scheme 1), (9Z,12Z)-octadeca-9,12-dien-1- spectra data were recorded on a Bruker Daltonics BioTOF
24
amine(compound 3f) and (9Z,12Z)-octadeca-9,12-dien-1-ol mass spectrometer.
25
(
compound 3i) were synthesized according to the literature.
The supercoiled plasmid DNA (pUC-19) used in agarose-gel
assay was purchased from Takara (Dalian, China). The plas-
mids used in the study were pEGFP-N1 (Clontech, Palo Alto, CA,
USA). A Cell Counting Kit-8 (CCK-8) was purchased from
2
.2. Synthesis (detailed analysis data of novel compounds
are shown in ESI†)
2.2.1. Preparation of compound 2. To a solution of L-
12
Dojindo Laboratories (Kumamoto, Japan). The Dulbecco's Cystine (compound 1, 2.4 g, 10 mmol) in 22 mL of 1 N NaOH,
Modied Eagle's Medium (DMEM) and fetal bovine serum were BOC anhydride (4.36 g, 20 mmol) was added dropwise. The
purchased from Invitrogen Corp. MicroBCA protein assay kit resulting solution was le stirring at room temperature for
was obtained from Pierce (Rockford, IL, USA). Luciferase assay overnight. The reaction mixture was washed with hexane and
kit was purchased from Promega (Madison, WI, USA). Endo- adjusted to pH 2 with a saturated solution of potassium bisul-
toxin free plasmid purication kit was purchased from TIAN- fate, extracted with ethylacetate, dried over anhydrous MgSO4,
GEN (Beijing, China). HEK 293 human embryonic kidney cell and ltered. The ltrate was concentrated with a rotary evapo-
lines were purchased from Shanghai Institute of Biochemistry rator to yield 2.2 g. Yield: 49.4%.
and Cell Biology, Chinese Academy of Sciences. The NMR
2.2.2. Preparation of compound 4. 4a–4f: to a mixing
spectra were measured on a Varian INOVA-400 spectrometer solution of compound 2 (880 mg, 2 mmol) and N-methyl-
ꢀ
and the d scale in parts per million was referenced to residual morpholine (890 mg, 4.4 mmol) in CHCl , cooled down to 0 C,
3
solvent peaks or internal tetramethylsilane (TMS). MS-ESI isobutylchloroformate (1200 mg, 4.4 mmol) in CHCl was added
3
Scheme 1 Synthesis route of title lipids 7a–7l. Reagents and conditions: (a) 1 N NaOH, (Boc)
2
O; (b) N-methylmorpholine (NMM), isobutyl-
Cl ; (d) CF COOH, CH Cl ; (e) N,N-
chloroformate, CHCl ; (c) N,N-dicyclohexylcarbodiimide (DCC), 4-dimethylamino-pyridine (DMAP), CH
3
2
2
3
2
2
diisopropylethylamine (DIEA), 1-hydroxybenzotriazole (HOBt), (3-dimethylaminopropyl)ethyl-carbodiimidmonohydrochloride (EDCI); (f) NaOH,
MeOH.
44262 | RSC Adv., 2014, 4, 44261–44268
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
dropwise for activation of carboxyl. Aer 1 h, amine
RSC Advances
12
2.2.6. Preparation of compound 10. Step a: compound 9
(
compound 3a–3f) (4.8 mmol) was added, and the reaction was (0.7 g, 1.14 mmol) was dissolved in CH OH (50 mL), then 2 N
3
slowly warmed to room temperature. Aer 40 h of reaction, the NaOH was added to it. The reaction kept stirring at room
solvent was removed under reduced pressure, then CH Cl was temperature. Aer the reaction, the CH OH was removed under
2
2
3
2
added. The mixture was washed with H O, saturated aqueous reduced pressure. Then adding water to it and use 2 N HCl
NaHCO solution, and brine. The organic layer was dried over adjust the pH to acidity. Then, CH Cl was added, washed with
3
2
2
anhydrous sodium sulfate and evaporated the solvent to get brine twice. The organic layer was dried over anhydrous sodium
white solid. The residue was puried by silica gel column sulfate and evaporated the solvent to get white solid. Yield: 74%.
chromatography (PE/EA ¼ 5/1, v/v) to obtain compound 4a–4f.
Step b: the compound just got (0.48 g, 0.78 mmol) in CH Cl
(70 mL) was cooled to 0 C. Sequentially HOBt (172 mg,
2
2
ꢀ
Yield: 33.0–50.3%
4
g–4k to the solution of compound 2 (880 mg, 2 mmol) and 1.12 mmol), EDC$HCl (215 mg, 1.12 mmol), N,N-diisopropyle-
ꢀ
4
-dimethylamino-pyridine (DMAP) (50 mg) in CH Cl at 0 C, thylamine (DIEA) (210 mg, 1.6 mmol) were gradually added and
2 2
ꢀ
N,N-dicyclohexyl-carbodiimide (DCC) (0.99 g, 4.8 mmol) in the reaction mixture was stirred for 1 h at 0 C. Then the mixed
CH Cl were added dropwise. Aer 30 min, the alcohol (3g–3k) solution was le for stirring at room temperature overnight. The
4.8 mmol) was added, and the reaction was slowly warmed to solvent was removed under reduced pressure, then EtOAc was
2
2
(
room temperature. Aer 40 h of reaction, the solution was added, the mixture was washed with water, saturated aqueous
ꢀ
cooled to 0 C, the precipitate formed was ltered off, evapo- NaHCO solution and saturated brine. The organic layer was
3
rated the solvent. The residue was puried by silica gel column dried over anhydrous sodium sulfate and evaporated the
chromatography (PE/EA ¼ 5/1, v/v) to obtain 4g–4k. Yield: 25.4– solvent to get white solid. The residue was puried by silica gel
3
8.2%
column chromatography (PE/EA ¼ 1/2, v/v) to obtain compound
2.2.3. Preparation of compound 6. Step a: Compound 4 10. Yield: 84.9%.
(
0.5 mmol) was dissolved in anhydrous CH
2
Cl
2
(5 mL), then
Cl
2
2.2.7. Preparation of compound 7l. Compound 10 (150 mg)
was dissolved in anhydrous CH Cl (1.5 mL), then triuoro-
triuoroacetic acid (CF COOH) (5 mL) in anhydrous CH
(
removed under reduced pressure. The residue was washed with was added at 0 C. Aer stirring 6 h, the solvent was removed
3
ꢀ
2
2
2
5 mL) was added at 0 C. Aer stirring 6 h, the solvent was acetic acid (CF COOH) (1.5 mL) in anhydrous CH Cl (1.5 mL)
3 2 2
ꢀ
anhydrous ether twice to get pure target compound. Yield: 85%. under reduced pressure. The residue was washed with anhy-
Step b: A solution of compound 5 (2.2 mmol) in CH Cl
50 mL) was cooled to 0 C. Sequentially HOBt (367 mg,
drous ether twice to get pure compound 7l. Yield: 80%.
2
2
ꢀ
(
2.4 mmol), EDC$HCl (460 mg, 2.4 mmol), N,N-diisopropyle-
thylamine (DIEA) (517 mg, 4 mmol) were gradually added. Aer
the reaction mixture was stirred for 1 h at 0 C, add the
2.3. Preparation of cationic liposome
ꢀ
The cationic lipid and the neutral lipid DOPE in a 1 : 2 mole
ratio were dissolved in 2.5 mL anhydrous chloroform in a glass
vial. The solvent was removed under reduced pressure with a
thin ow of moisture-free nitrogen gas. And the dried lipid lm
was then kept under high vacuum overnight. The lipid lm was
hydrated with 2.5 mL of Tris–HCl buffer (10 mM, pH 7.4) to the
compound (1 mmol) got from step a. Then the mixed solution
was le for stirring at room temperature overnight. The mixture
was washed with water, saturated aqueous NaHCO solution
3
and saturated brine. The organic layer was dried over anhydrous
sodium sulfate and evaporated the solvent to get white solid.
The residue was puried by silica gel column chromatography

nal lipid concentration of 1 mM. The samples were sonicated
(
PE/EA ¼ 1/2, v/v) to obtain compound 6a–6k. Yield: 37–52%
.2.4. Preparation of compound 7. Compound 6 (250 mg)
was dissolved in anhydrous CH Cl (2.5 mL), then triuoro-
in a bath sonicator to generate small unilamellar vesicles
2
26
according to previously described procedures.
2
2
acetic acid (CF
3
COOH) (2.5 mL) in anhydrous CH
2
Cl
2
(2.5 mL)
ꢀ
was added at 0 C. Aer stirring 6 h, the solvent was removed
2.4. Amplication and purication of plasmid DNA
under reduced pressure. The residue was washed with anhy-
drous ether twice to get pure compound 7a–7k. Yield: 82–90% pGL-3 and pEGFP-N1 plasmids were used. The former one was
2
.2.5. Preparation of compound 9. A solution of compound seed as the luciferase reporter gene, which was transformed in
ꢀ
5
(1.27 g, 2.4 mmol) in CH
2
Cl
2
(50 mL) was cooled to 0 C. M109 Escherichia coli, and the latter one was used as the
Sequentially HOBt (367 mg, 2.4 mmol), EDC$HCl (460 mg, enhanced green uorescent protein reporter gene, which was
2
.4 mmol), N,N-diisopropylethylamine (DIEA) (517 mg, 4 mmol) transformed in E. coli DH5a. Both plasmids were amplied in E.
ꢀ
were gradually added and the reaction mixture was stirred for coli grown in LB medium at 37 C and 220 rpm overnight. The
1
ꢀ
h at 0 C. Then compound 8 (280 mg, 2 mmol) was added to plasmids were puried by an EndoFree Tiangen™ Plasmid Kit.
the mixture solution, and le it stirring at room temperature Then, the puried plasmids were dissolved in TE (Tris + EDTA)
ꢀ
overnight. The mixture was washed with water, saturated buffer solution and stored at ꢁ80 C. The integrity of plasmids
3
aqueous NaHCO solution and saturated brine. The organic was conrmed by agarose gel electrophoresis. The purity and
layer was dried over anhydrous sodium sulfate and evaporated concentration of plasmids were determined by the ratio of
the solvent to get white solid. The residue was puried by silica ultraviolet (UV) absorbances at 260 nm/280 nm (ꢂ1.9). The
gel column chromatography (PE/EA ¼ 1/4, v/v) to obtain concentrations of plasmids were determined by ultraviolet (UV)
compound 9. Yield: 51.6%
absorbance at 260 nm.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 44261–44268 | 44263

View Article Online
RSC Advances
.5. Preparation of lipid/DOPE/DNA complexes (lipoplexes)
Paper
2
grid with Formvar lm into the freshly prepared nanoparticles
solution (10 mL). A few minutes aer the deposition, the
aqueous solution was blotted away with a strip of lter paper
and then the samples were dried for 2 min at room temperature.
The samples were stained with phosphotungstic acid (ATP)
aqueous solution and dried in air.
To prepare the lipid/DOPE/pDNA complexes (lipoplexes),
various amounts of cationic lipids were mixed with a constant
amount of DNA by pipetting thoroughly at various N/P ratios,
and the mixture was incubated for 30 min at room temperature.
The theoretical N/P ratio represents the charge ratio of cationic
lipid to nucleotide base (mole ratios) and was calculated by
considering the average nucleotide mass of 309.
2.10. Cell culture
Human embryonic kidney transformed 293 (HEK293) cells were
incubated in DMEM containing 10% fetal bovine serum (FBS)
2
.6. Agarose-gel retardation assay
ꢁ1
Lipid/DOPE/pDNA complexes at different N/P ratios (the amino and 1% antibiotics (penicillin-streptomycin, 10 000 U mL ) at
ꢀ
groups of lipids to phosphate groups of DNA) ranging from 0 to 37 C in a humidied atmosphere containing 5% CO
2
.
8
0
were prepared by adding an appropriate volume of lipids to
.125 mg pUC-19 DNA. The complexes were incubated at 37 C
ꢀ
2.11. MTT cytotoxicity assay
for 30 min. Then the complexes were electrophoresed on the
.0% (W/V) agarose gel containing gelred and with Tris-acetate
TAE) running buffer at 110 V for 30 min. DNA was visualized
with a UV lamp at wavelength of 312 nm by using BioRad
Universal Hood II.
Toxicity of lipoplexes toward HEK 293 cells was determined by
using a Cell Counting Kit-8 (CCK-8). About 8000 cells per well
1
(
27
were seeded into 96-well plates. Aer 24 h, optimized lipid/
DOPE formulations were complexed with 0.2 mg of DNA at
various N/P ratios for 30 min; 100 mL of lipoplexes were added to
the cells in the absence of serum. Aer 4 h of incubation, lip-
oplex solutions were removed, and 100 mL of media with 10%
FBS was added. Aer 24 h, 10 mL CCK-8 was added to each well
2.7. Ethidium bromide replacement assay
The ability of lipids 7 to condense DNA was studied using
ethidium bromide (EB) exclusion assays. Fluorescence spectra
were measured at room temperature in air by a Horiba Jobi-
nYvon Flu-oromax-4 spectrouorometer and corrected for the
ꢀ
and the plates were incubated at 37 C for another 1 h. Then, the
absorbance of each sample was measured using an ELISA plate
reader (model 680, BioRad) at a wavelength of 450 nm. The cell
viability (%) was obtained according to the manufacturer's
instruction. Blank control and Lipoplex prepared from Lip-
ofectamine 2000 were used as comparisons.
ꢁ1
system response. EB (5 mL, 1.0 mg mL ) was put into quartz
cuvette containing 2.5 mL of 10 mM 4-(2-hydroxyethyl)-1-
piperazinee-thanesulfonic acid (HEPES) solution (pH7.4). Aer
shaking, the uorescence intensity of EB was measured. Then
ꢁ
1
CT-DNA (10 mL, 1.0 mg mL ) was added to the solution and
mixed symmetrically, and the measured uorescence intensity
is the result of the interaction between DNA and EB. Subse-
2.12. In vitro transfection procedure
Gene transfection of a series of complexes was investigated in
HEK293 cells. In order to obtain about 80% conuent cultures at
the time of transfection, 24-well plates were seeded with 85 000
cell per well in 500 mL of serum-free media 24 h before trans-
fection. For the preparation of lipoplexes applied to cells, various
amounts of liposomes and DNA were serially diluted separately in
antibiotic-free DMEM culture medium; then, the DNA solutions
were added into liposome solutions and mixed briey by pipetting
up and down several times, aer which the mixtures were incu-
bated at room temperature for about 30 min to obtain lipoplexes
of desired N/P ratios, the nal lipoplexes volume was 100 mL, and
the DNA was used at a concentration of 1 mg per well. Aer 30 min
of complexation, old cell culture medium was removed from the
wells, cells were washed once with serum-free DMEM, and the
above 100 mL lipoplexes was added to each well. The plates were
ꢁ
1
quently, the solutions of lipid 7a–7k (0.5 mmol L , 4 mL for
ꢁ
1
each addition) and 7l (1 mmol L , 4 mL for each addition) were
added to the above solution for further measurement. All the
samples were excited at 520 nm and the emission was measured
at 600 nm. The pure EB solution and DNA/EB solution without
cationic liposome were used as negative and positive controls,
respectively. The percent relative uorescence (%F) was deter-
mined using the equation %F ¼ (F ꢁ FEB)/(F
EB and F denote the uorescence intensities of pure EB
solution and DNA/EB solution, respectively.
0
ꢁ FEB), wherein
F
0
2.8. Lipoplex particle sizes and zeta potentials
Sizes and zeta potentials of the lipid/pDNA lipoplexes at various
N/P ratios were analyzed at room temperature on a dynamic
light scattering system (Zetasizer Nano ZS, Malvern instruments
ꢀ
then incubated for 4 h at 37 C in a humidied atmosphere
2
containing 5% CO . At the end of incubation period, medium was
ꢀ
Led) at 25 C. The lipoplexes particle solutions were rst
removed, and 500 mL of fresh DMEM medium containing 10%
FBS was added to each well. Plates were further incubated for a
period of 24 h before checking the reporter gene expression.
For uorescent microscopy assays, cells were transfected by
complexes containing pEGFP-N1. Aer 24 h incubation, the
ꢁ
1
prepared by mixing 7/DOPE lipids and pUC-19 (2 mg mL
)
under diverse N/P ratios in 1 mL deionized water.
2.9. Transmission electron microscopy (TEM)
TEM images were obtained on a JEM-100CX (JEOL) trans- microscopy images were obtained at the magnication of 100
mission electron microscope at an acceleration voltage of and recorded using Viewnder Lite (1.0) soware. Control
100 kV. The TEM samples were prepared by dipping a copper transfection was performed in each case using a commercially
44264 | RSC Adv., 2014, 4, 44261–44268
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
available transfection reagent Lipofectamine 2000™ based on DOPE ratio of 1 : 2 was used herein. Agarose gel electrophoresis
the standard conditions specied by the manufacture. was rst performed to investigate their DNA condensation
For luciferase assays, cells were transfected by complexes ability, which is known as a primary factor for the cationic lipids
containing pGL-3. For a typical assay in a 24-well plate, 24 h serving as gene vectors. As shown in Fig. 1. All of the lipids
posttransfection as described above, the old medium was showed good DNA binding abilities. Full DNA retardation on
removed from the wells, and the cells were washed twice with the gel could be observed from the N/P ratio of 2 except the
5
00 mL of prechilled PBS. According to Luciferase assay kit experiment involving 7f, which completely retarded DNA from
promega) manufacture, 100 mL of 1 ꢃ cell lysis buffer diluted higher N/P ratio. Further, ethidium bromide (EB) dye replace-
with PBS was then added to each well, and the cells were lysed ment assay was also carried out to evaluate their DNA binding
(
21
for 30 min in a horizontal rocker at room temperature. The cell abilities . Results in Fig. 2 show the relative uorescence
lysate was transferred completely to Eppendorf tubes and intensities of EB in HEPES solution (pH 7.4, 10 mM) in the
centrifuged (4000 rpm, RT) for 2 min; the supernatant was presence of various amounts of liposomes. It was observed that
transferred to Eppendorf tubes and stored in ice. For the assay, the uorescent intensity dramatically decreased with the
20 mL of this supernatant and 100 mL of luciferase assay increase of N/P ratio, indicating that the liposomes have strong
substrate (Promega) were used. The lysate and the substrate DNA-binding abilities to replace EB. At the N/P ratios of 4, all
were both thawed to RT before performing the assay. The the liposomes could quench at last 50% of original uorescence
substrate was added to the lysate, and the luciferase activity was intensity. For the amide contained gemini lipids 7a–7e, the
measured in a luminometer (Turner designs, 20/20, Promega, original uorescent intensities were quenched to 0–15% at the
USA) in standard single-luminescence mode. The integration N/P ratio of 3 (Fig. 2A). Meanwhile, for the ester contained
time of measurement was 10 000 ms. A delay of 2 s was given gemini lipids 7g–7k, the uorescent quenching abilities were
before each measurement. The protein concentration in the cell relatively lower (Fig. 2B). Such difference might be attributed to
lysate supernatant was estimated in each case with Lowry the more negatively charged ester bonds. The chain length
protein assay kit (PIERCE, Rockford, IL, USA). Comparison of might also affect the binding ability, and liposomes with longer
the TEs of the individual lipids was made based on the data for hydrophobic chains exhibited better uorescence quenching
ꢁ1
luciferase expressed as relative light units (RLU) mg
of ability, indicating tighter DNA binding (7a–7d). Higher unsa-
protein. All the experiments were done in triplicates, and results turation degree of the hydrophobic chain would not benet the
presented are the average of at least two such independent binding, and 7f-derived liposome showed distinctly weaker
experiments done on the same days.
uorescence quenching ability. Moreover, the mono counter-
part 7l showed obviously lower binding ability than the gemini
lipids, suggesting that the conguration of gemini liposomes
might benet their interaction toward DNA.
3. Results and discussion
3
.1. Synthesis of target cationic gemini lipids 7
The particle size, surface charge and morphology of the
liposome/DNA complexes (lipoplexes) were important factors
for their large inuences on gene transfection, expecially on the
apparent cytotoxicity, cellular uptake/trafficking, and the
release of the encapsulated gene. Dynamic light scattering
The synthetic route of the cyclen-based gemini lipids 7 is shown
in Scheme 1. L-Cystine was chosen as backbone for its
symmetric and modiable structure. The condensation
between butyloxycarbonyl (Boc) protected cystine 2 and various
amines or alcohols in the presence of N-methylmorpholine
(DLS) assay was applied to measure the sizes and zeta-potentials
of the formed lipoplexes, and the results are shown in Fig. 3.
The average diameters were observed in a range of 100–650 nm,
which strongly depended on the N/P ratios (Fig. 3A). With the
rise of N/P ratio, the particle sizes reduced and stabilized at
about 100 nm, indicating the full condensation of DNA. As the
exceptions, lipoplexes formed from 7f and 7l gave larger
(
NMM) or dicyclohexylcarbodiimide (DCC) gave relative
diamides or diesters 4, respectively. Compound 4 was then
23
deprotected and reacted with tri-Boc-cyclenylacetic acid 5 to
give the protected precursors 6. Target gemini lipids 7 could be
smoothly obtained by removing the Boc groups with triuoro-
acetic acid, and the resulting triuoroacetates were used in
subsequent studies. In addition, for comparison, the mono
counterpart lipid 7l, which has only one headgroup and one tail,
was also prepared via similar method from glycine methyl ester
8. The structures of all new compounds were characterized and
conrmed by NMR and HRMS.
3.2. Interactions between the liposomes and plasmid DNA
Cationic liposomes can be formed from either individual
cationic lipid or more frequently from a combination of cationic
lipid and neutral lipids such as DOPE. Each gemini lipid was
mixed in different molar ratios (1 : 1, 1 : 2, and 1 : 3) with DOPE
to determine the optimal combination. According to the best
Fig. 1 Electrophoretic gel retardation assays of plasmid DNA with the
presence of liposomes at various N/P ratios. The molar ratio of lipid/
behavior in the transfection experiment (data not shown), the 7/ DOPE was 1 : 2.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 44261–44268 | 44265

View Article Online
RSC Advances
Paper
Fig. 2 Ethidium bromide displacement assay by 7a–7f, 7l (A) and 7g–7k (B) under various N/P ratios in 10 mM Hepes buffer solution. The molar
ratio of lipid/DOPE was 1 : 2.
Fig. 3 Mean particle sizes (A) and zeta-potentials (B) of the lipoplexes formed from 7 at N/P ratio of 1, 2, 4, 6, 8.
particles under N/P of 8, and this might be due to their relatively largest N/P ratio, indicating that more liposome was needed for
lower DNA binding ability, which is in accordance with the the condensation of DNA.

uorescence quenching results. In addition, the stability of the
Transmission electron microscopy (TEM) was further per-
lipoplexes was also studied, and it was shown that the sizes of formed to get direct information about the shape and
the particles involving lipids 7d, 7e and 7l seldom changed morphology of lipoplexes. Representative electron micrograph
during 24 h incubation, while the lipofectamine 2000-contained of lipoplexes formed from 7e and lipofectamine 2000 are shown
particles exhibited obvious size increase (Fig. S1†). Such results in Fig. 4 (A and C respectively), which reveal that the lipoplex
suggest the good stability of the lipoplexes. Zeta potentials of all show almost spherical particles with the diameters in the range
the lipoplexes showed similar trends. With the increase of N/P of 50–100 nm. Further, the 7e/DNA lipoplex in serum environ-
ratio, the zeta potentials reversed from negative to positive at ment was also investigated by the same assay. Result in Fig. 4B
N/P of 2 or 4, and nally reached around +35 mV. At N/P ratio of shows that the complex is stable in the presence of serum, and
2, the surface charges of lipoplexes derived from 7a–7e turned to no obvious change of particle size or morphology was observed.
positive, while other lipoplexes still remained negatively
charged. Such results also suggest that the amide-contained
gemini lipids 7a–7e have stronger DNA binding ability, which
was similar to the EB displacement results (Fig. 2A). Addition-
ally, the mono counterpart 7l was again proved to have lower
DNA binding ability. Among the lipoplexes studied, the zeta-
potential of the one derived from 7l turned to positive at the
3
.3. Cytotoxicity
Low cytotoxicity is necessary for efficient gene carriers. The cell
viability of the lipoplexes formed from cationic gemini lipids as
a function of N/P ratio was evaluated in HEK 293 cells by CCK-8,
and commercially available lipofectamine 2000 was used for
comparison. As shown in Fig. 5, even at high N/P ratios, the
44266 | RSC Adv., 2014, 4, 44261–44268
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
Fig. 4 TEM image of lipoplexes. (A & B) 7e/DOPE/DNA lipoplex at N/P ratio of 12 in deionized water (A) and 50% of FBS (B); (C) lipofectamine
000/DNA lipoplex in deionized water.
2
3.4. In vitro gene transfection
Studies on the structure–activity relationship of cationic lipo-
some are quite important and helpful for further design of ideal
gene carriers. Among the twelve lipids, six of them (7a, 7e, 7g,
7h, 7k, 7i) were rst selected for the transfection of luciferase
reporter gene. Fig. 6A gives the quantitatively transfection
results in HEK 293 cells at N/P ratios of 4, 8 and 12 with lip-
ofectamine 2000 and naked DNA as controls. These six lipids
have large structural difference on the hydrophobic moiety, and
this led to distinct diversity of their TE. Lipid 7e with oleylamine
as the hydrophobic group gave the best TE, which was only
slightly less than lipofectamine 2000. For the lipids with same
aliphatic chains, the amide-contained lipids (7a and 7e) gave
several times higher TE than their ester-contained analogs (7g
and 7h). We also found similar amide preference in our
28
previous studies on the lipidic gene vectors. Tocopherol has
under lipid/DOPE ratio of 1 : 2. Data represent mean ꢄ SD (n ¼ 3). Each been found to be a potential candidate for the hydrophobic
Fig. 5 Cytotoxicity of the lipoplexes prepared at various N/P ratios
12,19
lipoplex were at N/P ratio of 2, 4, 8, 12, 24, respectively.
group of lipidic gene vectors.
However, the tocopherol-
contained lipid 7k was much less effective compared to 7e
with aliphatic long chain. Besides, by comparison with 7a, the
relative cell viabilities treated with the lipoplexes were distinctly TE of mono counterpart 7l decreased for an order of magnitude,
higher than those involving lipofectamine 2000. Nearly no indicating the advantage of gemini structure. Further, the effect
cytotoxicity was observed for lipoplexes at N/P ratios lower than of unsaturation degree of the hydrophobic chains on the
1
2, which was the highest N/P ratio used in the transfection transfection was then investigated. As shown in Fig. 6B, N/P
experiments. The results suggest that the cyclen-based gemini ratios of 8 and 12, at which better TEs were obtained in
lipids are biocompatible and suitable for gene delivery.
previous studies, were applied in the transfection experiments
Fig. 6 Transfection efficiencies of lipoplexes in HEK 293 cells at various N/P ratios under lipid/DOPE ratio of 1 : 2.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 44261–44268 | 44267

View Article Online
RSC Advances
Paper
using 7d, 7e, 7f and 7i. The oleyl group was found to be the best
choice for the transfection, and oleyl-contained lipids 7e and 7h
gave the best TEs. The linoleyl-contained 7f and 7i gave much
decreased TEs, which were even lower than the lipid with
saturated chains (7d). As many reported and some commer-
cially available transfection reagents also have oleyl moiety in
their structures, such results might be reasonable. The intro-
ducing of monounsaturated hydrophobic chain would facilitate
membrane disruption and DNA escape from endosome by
increasing the membrane uidity. Further increase of unsa-
turation degree may lead to difficulty of liposome forming and
subsequent DNA binding, which would hamper the transfection
process. The above structure–activity relationships may afford
us clues for further optimization of gemini lipid gene delivery
materials.
4 (a) Y. Su, R. Mani and M. Hong, J. Am. Chem. Soc., 2008, 130,
8856–8864; (b) Y. Su, S. Li and M. Hong, Amino Acids, 2013,
44, 821–833.
5 T. R. G. P. L. Felgner, M. Holm, R. Roman, H. W. Chan,
M. Wenz, J. P. Northrop, G. M. Ringold and M. Danielsen,
Proc. Natl. Acad. Sci. U. S. A., 1987, 84, 7413–7417.
6 S. Bhattacharya and A. Bajaj, Chem. Commun., 2009, 4632–
4656.
7 F. M. Menger and C. A. Littau, J. Am. Chem. Soc., 1993, 115,
10083–10090.
8 F. M. Menger, J. S. Keiper and V. Azov, Langmuir, 1999, 16,
2062–2067.
9 M. Kumar, K. Jinturkar, M. R. Yadav and A. Misra, Crit. Rev.
Ther. Drug Carrier Syst., 2010, 27, 237–278.
10 V. Floch, N. Legros, S. Loisel, C. Guillaume, J. Guilbot,
T. Benvegnu, V. Ferrieres, D. Plusquellec and C. Ferec,
Biochem. Biophys. Res. Commun., 1998, 251, 360–365.
15
4
. Conclusion
11 T. Zhou, A. Llizo, P. Li, C.-x. Wang, Y. Guo, M. Ao, L. Bai,
C. Wang, Y. Yang and G. Xu, J. Phys. Chem. C, 2013, 117,
In summary, a series of cyclen-based cationic gemini lipids with
two symmetrical head groups and two hydrophobic tails were
designed and synthesized. Cystine was used as backbone
between the two amphiphilic units. Liposomes could be
smoothly formed by these lipids and helper lipid DOPE. The
interaction between the liposome and plasmid DNA was
studied. Results reveal that these liposomes have strong DNA
binding ability, and full DNA condensation could be achieved at
the N/P ratio of 2–4. The lipoplexes have proper sizes and zeta-
potentials, which might be suitable for gene transfection. These
materials were applied as non-viral gene delivery vectors, and
their structure–activity relationship was investigated. Subtle
changes in the structure of the lipids would lead to large effect
on the TE. The oleyl amine derived lipid 7e gave the best TE,
which was close to the commercially available transfection
reagent lipofectamine 2000. Besides, these lipids have very low
cytotoxicity, suggesting their good biocompatibility. Further
modications and extended application of such type of gemini
lipids are currently underway.
26573–26581.
1
1
2 B. Kedika and S. V. Patri, Mol. Pharm., 2012, 9, 1146–1162.
3 D. Zhi, S. Zhang, S. Cui, Y. Zhao, Y. Wang and D. Zhao,
Bioconjugate Chem., 2013, 24, 487–519.
4 R. Koynova, B. Tenchov, L. Wang and R. C. MacDonald, Mol.
Pharm., 2009, 6, 951–958.
5 D. Zhi, S. Zhang, B. Wang, Y. Zhao, B. Yang and S. Yu,
Bioconjugate Chem., 2010, 21, 563–577.
6 B. Kedika and S. V. Patri, Bioconjugate Chem., 2011, 22, 2581–
1
1
1
1
1
2592.
7 M. Rajesh, J. Sen, M. Srujan, K. Mukherjee, B. Sreedhar and
A. Chaudhuri, J. Am. Chem. Soc., 2007, 129, 11408–11420.
8 R. Mukthavaram, S. Marepally, M. Y. Venkata, G. N. Vegi,
R. Sistla and A. Chaudhuri, Biomaterials, 2009, 30, 2369–
2384.
1
9 Q. Liu, Q.-Q. Jiang, W.-J. Yi, J. Zhang, X.-C. Zhang, M.-B. Wu,
Y.-M. Zhang, W. Zhu and X.-Q. Yu, Bioorg. Med. Chem., 2013,
21, 3105–3113.
2
2
2
2
2
2
2
2
2
0 B.-Q. Liu, W.-J. Yi, J. Zhang, Q. Liu, Y.-H. Liu, S.-D. Fan and
X.-Q. Yu, Org. Biomol. Chem., 2014, 12, 3484–3492.
1 Q.-D. Huang, G.-X. Zhong, Y. Zhang, J. Ren, Y. Fu, J. Zhang,
W. Zhu and X.-Q. Yu, PLoS One, 2011, 6, e23134.
2 R. M. Izatt, K. Pawlak, J. S. Bradshaw and R. L. Bruening,
Chem. Rev., 1991, 91, 1721–2085.
Acknowledgements
This work was nancially supported by the National Program on
Key Basic Research Project of China (973 Program,
2
2
012CB720603), the National Science Foundation of China (no.
1232005), and the Specialized Research Fund for the Doctoral
3 J. W. Jeon, S. J. Son, C. E. Yoo, I. S. Hong, J. B. Song and
J. Suh, Org. Lett., 2002, 4, 4155–4158.
Program of Higher Education (20120181130006) in China. J. Z.
thanks the Program for New Century Excellent Talents in
University (NCET-11-0354). We also thank Analytical & Testing
Center of Sichuan University for the TEM analysis.
4 C. Fong, D. Wells, I. Krodkiewska, P. G. Hartley and
C. J. Drummond, Chem. Mater., 2006, 18, 594–597.
5 M. Sodeoka, R. Sampe, S. Kojima, Y. Baba, T. Usui, K. Ueda
and H. Osada, J. Med. Chem., 2001, 44, 3216–3222.
6 Q. Liu, W.-J. Yi, Y.-M. Zhang, J. Zhang, L. Guo and X.-Q. Yu,
Chem. Biol. Drug Des., 2013, 82, 376–383.
References
7 K. Luo, C. Li, L. Li, W. She, G. Wang and Z. Gu, Biomaterials,
1
(a) G. M. Rubanyi, Mol. Aspects Med., 2001, 22, 113–142; (b)
A. El-Aneed, Eur. J. Pharmacol., 2004, 498, 1–8.
2012, 33, 4917–4927.
8 B. Wang, W.-J. Yi, J. Zhang, Q.-F. Zhang, M.-M. Xun and
X.-Q. Yu, Bioorg. Med. Chem. Lett., 2014, 24, 1771–1775.
2
3
A. El-Aneed, J. Controlled Release, 2004, 94, 1–14.
X. Guo and L. Huang, Acc. Chem. Res., 2011, 45, 971–979.
44268 | RSC Adv., 2014, 4, 44261–44268
This journal is © The Royal Society of Chemistry 2014