Design, synthesis, and in vitro gene delivery efficacies of novel cholesterol-based gemini cationic lipids and their serum compatibility: A structure-activity investigation

Article abstract of DOI:10.1021/jm0611253

Five cholesterol-based gemini cationic lipids, which differ in the length of the spacer [-(CH₂)_n-] chain between the head groups, have been synthesized. These lipids are useful as nonviral gene delivery agents, and all cholesterol-based gemini lipids (2a-2e) are better transfecting agents than their monomeric lipid counterpart 1. Transfection efficiency of all the gemini lipids except lipid 2a [-(CH₂)₃-] was maintained even when the serum was present during the transfection conditions as compared to the monomeric lipid 1, with which a dramatic decrease in transfection efficiency was observed. With the increase in spacer chain length from propanediyl [-(CH₂)₃-] to pentanediyl [-(CH ₂)₅-], transfection efficiency increased both in the absence and presence of serum. However, transfection efficiency decreased with further increase in the length from the pentanediyl [-(CH₂) ₅-] to the dodecanediyl [-(CH₂)₁₂-] spacer. Among these gemini lipids 2c showed the highest transfection activity, which was also greater than that of the commercially available formulation.

Full text of DOI:10.1021/jm0611253

2432
J. Med. Chem. 2007, 50, 2432-2442
Design, Synthesis, and in Vitro Gene Delivery Efficacies of Novel Cholesterol-Based Gemini
Cationic Lipids and Their Serum Compatibility: A Structure-Activity Investigation
Avinash Bajaj,† Paturu Kondiah,‡ and Santanu Bhattacharya*,†,§
Department of Organic Chemistry, Department of Molecular Reproduction, DeVelopment and Genetics, Indian Institute of Science,
Bangalore 560 012, India, and Chemical Biology Unit of JNCASR, Bangalore 560 064, India
ReceiVed September 28, 2006
Five cholesterol-based gemini cationic lipids, which differ in the length of the spacer [-(CH2)n-] chain
between the head groups, have been synthesized. These lipids are useful as nonviral gene delivery agents,
and all cholesterol-based gemini lipids (2a-2e) are better transfecting agents than their monomeric lipid
counterpart 1. Transfection efficiency of all the gemini lipids except lipid 2a [-(CH2)3-] was maintained
even when the serum was present during the transfection conditions as compared to the monomeric lipid 1,
with which a dramatic decrease in transfection efficiency was observed. With the increase in spacer chain
length from propanediyl [-(CH2)3-] to pentanediyl [-(CH2)5-], transfection efficiency increased both in
the absence and presence of serum. However, transfection efficiency decreased with further increase in the
length from the pentanediyl [-(CH2)5-] to the dodecanediyl [-(CH2)12-] spacer. Among these gemini
lipids 2c showed the highest transfection activity, which was also greater than that of the commercially
available formulation.
1. Introduction
the head group and the long alkyl chains, shows greater in ViVo
transfection efficiency than the corresponding ester analogue
N-[1-(2,3-dioleoyloxy)propyl]-N,N,N-trimethylammonium chlo-
ride (DOTAP).16,17 We have shown that the use of the ether
linkage leads to a dramatic increase in transfection efficiencies
of cholesterol-based cationic lipids as compared to the ester-
or urethane-based cholesterol lipid analogues.18,19 Hydrocarbon-
chain-based lipids with varying oxyethylene linkages in cationic
pseudoglyceryl lipids have also been studied for gene transfec-
tion activities. It has been observed that incorporation of
oxyethylene units at linkages brought about significant enhance-
ment in the in Vitro gene delivery efficiencies of the cationic
lipids in mammalian cells.20,21 The presence of the ether linkage
makes these compounds hydrolytically stable, and their aqueous
suspensions were also found to have long shelf life.
Gene therapy, known as the fourth revolution in medicine,
is the new hope for curing various genetic disorders by
delivering the desired gene to the specific cells.1 Viruses are
the best means of delivering the genes,2 but recent deaths in
clinical trials have raised questions on the use of viruses as gene
delivery vehicles.3 Among nonviral vectors, cationic liposomes,4
although less efficient, have a great potential for delivery of
genetic therapeutic drugs, because of less toxicity, low immu-
nogenicity, high DNA carrying capacity, their ease of large-
scale production, and simplicity in preparation and administra-
tion.5-10 Ever since its discovery,11 cationic lipid-mediated gene
delivery remains at the forefront of present day scientific
research owing to its possible application in the field of gene
therapy.12
Most of the cationic lipid formulations studied for gene
delivery are based on either glycerol-13 or cholesterol-based
cytofectins.14 For instance, 3â-[N,(N,N-dimethylaminoethane)-
carbamoyl] cholesterol (DC-Chol),a a cationic cholesterol
derivative, has been successfully used as a coaggregate with
1,2-dioleoyl-L-R-glycero-3-phosphatidyl ethanolamine (DOPE)
to prepare liposomes that transfect mammalian cells efficiently.15
Cationic lipid suspensions readily form complexes with the
negatively charged DNA (gene) under ambient conditions. The
molecular structure of cationic lipids is an important parameter
that controls their DNA complexation and gene transfection
activity. For instance, the functional group that links the polar
head group and the hydrocarbon chains of such lipid molecules
plays a crucial role in their utilization in gene transfer events.
Thus, N-[1-(2,3-dioleyloxy)propyl]-N,N,N-trimethylammonium
chloride (DOTMA), which contains an ether linkage between
Clearly, design and syntheses of new lipid systems with
alternative structural types are crucial for the development of
potent synthetic vectors for such applications. Gemini lipids are
another class of lipids, where two lipid molecules are joined to
each other Via a spacer. We reported, for the first time, the
synthesis and aggregation properties of various glycerol-
backbone-based gemini lipids.22,23 The effect of the spacer length
on the aggregation properties of the gemini lipids was demon-
strated in these investigations. There are, however, very few
reports on the transfection properties of gemini lipids. Ahmad
and co-workers have shown transfection properties of cardio-
lipin-based gemini lipids.24,25 More recently, gemini surfactants
based on lipophilic pyrilium salts have also been examined for
gene delivery.26 Recently, we have reported the synthesis of
gemini cationic lipids based on an aromatic backbone between
the hydrocarbon chains and headgroup.27 However, there is no
systematic study until now that examines the effect of the spacer
on the gene transfection activities in a given series of gemini
lipids. To our knowledge, there is no report on cholesterol-based
gemini lipids as well.
* Corresponding author. Phone: (91)-80-2293-2664. Fax: (91)-80-2360-
0529. E-mail: sb@orgchem.iisc.ernet.in.
† Department of Organic Chemistry, Indian Institute of Science.
‡ Department of Molecular Reproduction, Development and Genetics,
Indian Institute of Science.
§ JNCASR.
Toward this end, we have designed five cholesterol-based
gemini cationic lipids (2a-2e) having a polymethylene spacer
between the cationic ammonium headgroups (Figure 1). These
gemini lipids differ in the length of the spacer chain, varying
from propanediyl to dodecanediyl. Though synthetically more
a Abbreviations: DC-Chol, 3â-[N,(N,N-dimethylaminoethane)-carbam-
oyl] cholesterol; DOPE, 1,2-dioleoyl-L-R-glycero-3-phosphatidylethanol-
amine; DOTMA, N-[1-(2,3-dioleyloxy)]-N,N,N-trimethylammonium chlo-
ride; DOTAP, N-[1-(2,3-dioleyl)]-N,N,N-trimethylammonium chloride; MFI,
mean fluorescence intensity; FACS, fluorescence-activated cell sorting.
10.1021/jm0611253 CCC: $37.00 © 2007 American Chemical Society
Published on Web 04/20/2007

Cholesterol-Based Gemini Lipids
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 10 2433
2.2. Aggregate Formation from Cationic Gemini Lipids.
Upon hydration, all the gemini lipids except 2a were found to
get dispersed in water easily. Lipid 2a could be also dispersed
albeit with difficulty. All the lipid molecules mentioned here
formed stable suspensions in water. The suspensions formed
from the lipids were found to be translucent. Generally, TEM
examination of air-dried, aqueous suspensions revealed the
existence of closed aggregate structures for all the gemini lipids
(not shown). The suspensions formed from 2a showed some
distinct aggregates. The aggregates formed from all the gemini
lipids ranged from 30 to 90 nm in size. When compared, all
gemini lipid aggregates were generally found to be smaller in
size as compared to the aggregates of monomeric lipid 1, which
is found to be 100 ( 30 nm in size (not shown).
Figure 1. Molecular structures of cholesterol-based cationic monomeric
lipid 1 and gemini lipids (2a-2e) synthesized and used for transfection
studies.
2.3. Mixed Liposome Formation with 1,2-Dioleoyl-L-r-
glycero-3-phosphatidyl Ethanolamine (DOPE). Liposomes
could be conveniently prepared from each of the gemini lipids
with naturally occurring helper lipid (1,2-dioleoyl-L-R-glycero-
3-phosphatidyl ethanolamine (DOPE)) by first subjecting the
films of the lipid mixtures to hydration and repeated freeze-
thaw cycles, followed by sonication at 60 °C for 15 min. All
the gemini lipids formed optically transparent suspensions.
Vesicles were prepared under sterile conditions and were
resonicated for 5 min at room temperature before transfection
experiments. The vesicular suspensions were sufficiently stable,
and no precipitation was observed within 3 months if stored at
4 °C.
demanding, an ether functionality was chosen between the
cholesterol and headgroup mainly due to the enhanced biological
activities of the cationic lipids having ether linkages compared
to those of ester analogues, as mentioned earlier, and also due
to the improved chemical stability of ether linkages.
Though unnatural, rigid cholesterol-based gemini lipids form
an important class. These gemini lipids are expected to have
different aggregation and transfection properties than those that
are based on fatty acid chains. In this series of gemini lipids
we have chosen a flexible hydrophobic polymethylene spacer
such that the spacer between headgroups varies in length. In
this paper, we describe the detailed synthetic procedures and
optimized transfection activities of these biologically active
compounds. To put our results in appropriate perspective, we
compared the transfection activities of these new gemini cationic
lipids (2a-2e) with those of the monomeric cholesterol-based
lipid 1. We have observed enhanced gene transfection activities
for all these gemini lipids as compared to those of the
monomeric one. Serum often inhibits the transfection efficiency
of cationic liposomes. However, four of these gemini lipids
except gemini lipid 2a manifest significantly improved trans-
fection activity even in the presence of serum.
2.4. Gel Electrophoresis. To characterize the electrostatic
binding interactions between the plasmid DNA and the mixed
cationic liposomes as a function of different N/P charge ratios
(or lipid/DNA mol ratios), we performed conventional electro-
phoretic gel retardation assays (Figure 2). All gemini lipid
suspensions were able to retard the plasmid DNA from the well
at N/P ratio of 1.0. It should be noted that every gemini molecule
possesses two hard charges, which indicates that the whole
plasmid DNA gets retarded at lipid/DNA mol ratio of 0.5.
Significant retardation of DNA was also observed at N/P ratio
of 0.5 (or lipid/DNA mol ratio of 0.25) as well.
2. Results and Discussion
2.5. Transfection Biology. 2.5.1. Optimization of Lipid/
DOPE Ratio. Naturally occurring lipids such as DOPE have
been known to increase the efficiency of transfection in lipid-
mediated gene transfer applications.16 In order to find out the
most effective formulations, transfections with identical lipid/
DNA mol ratios (or N/P ratios), varying the mol ratio of the
gemini lipids (2a-2e) in DOPE, were performed (Figure 3).
To find out the optimized transfection efficiency, both the
number of transfected cells and the mean fluorescence intensity
have been considered. The mean fluorescence intensities (MFI)
defined for GFP-positive cells reveal that the level of GFP
expression with a higher MFI value correlates positively with
a high GFP expression.29 These data were obtained from flow
cytometric analysis.
2.1. Chemistry. Five cholesterol-based gemini lipids with a
polymethylene spacer have been synthesized with different
spacer lengths. These gemini lipids have been synthesized from
the precursor cholest-5-en-3â-oxyethan-N,N-dimethylamine (7)
by reacting it with corresponding R,ω-dibromoalkanes (Scheme
1). Compound 7 was synthesized from cholest-5-en-3â-oxy-
ethane tosylate (6), which was synthesized by slightly modifying
a reported procedure.28 First, cholesterol (3) was tosylated using
p-toluenesulfonyl chloride in pyridine-chloroform (v/v: 1/1)
with a catalytic amount of DMAP for 6 h at 0 °C in 92% yield.
Cholesterol tosylate (4) was then subjected to a reaction with
ethylene glycol in dioxane under reflux for 4 h to afford cholest-
5-en-3â-oxyethan-2-ol (5) in 85% yield. Compound 5 was then
tosylated with p-toluenesulfonyl chloride in pyridine-chloro-
form (v/v: 1/1) for 6 h at 0 °C to get 6 in 90% yield. Reaction
of the tosylate (6) with dimethylamine in dry MeOH in a screw-
top pressure tube at 80 °C afforded the precursor 7 in
quantitative yields. Gemini lipids with flexible hydrophobic
spacers have been synthesized by reacting 7 with corresponding
R,ω-dibromoalkanes in 50-60% yields. All the gemini lipids
have been purified by repeated crystallizations from MeOH and
EtOAc. Lipids 2a and 2d were found to be quite hygroscopic
in nature. All the gemini lipids were characterized by 1H-NMR,
13C-NMR, ESI-MS, and elemental analysis as described in the
Experimental Section. Lipid 1 was synthesized as reported
previously.28
All the gemini lipids are most effective at a lipid/DOPE mol
ratio of 1:4, except lipid 2d, which was found to be most
effective at a lipid/DOPE mol ratio of 1:5. When the transfection
efficiencies at their optimal lipid/DOPE ratio were compared,
lipids 2c and 2d were found to be superior transfecting agents.
Although the numbers of transfecting cells were high in the
case of lipids 2a and 2b, the MFI was low. At this lipid/DOPE
ratio, although the number of transfected cells were very high,
they could transfect a considerably lesser amount of DNA.29
Therefore, lipids 2a and 2b transfected more cells with low MFI,
whereas lipids 2c and 2d were able to transfect a lesser number
of cells but with high MFI. In contrast, the gemini lipid 2e at

2434 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 10
Scheme 1. Synthesis of Cholesterol-Based Gemini Lipidsa
Bajaj et al.
a Reaction conditions: (a) p-TsCl/Py/CHCl3, 0 °C, 6 h; (b) ethylene glycol, 1,4-dioxane, 4 h, reflux; (c) p-TsCl/Py/CHCl3, 0 °C, 6 h; (d) dimethylamine,
MeOH, 80 °C, screw-top pressure tube; (e) MeOH-EtOAc, R,ω-dibromoalkanes, 80 °C, screw-top pressure tube, 48-72 h.
Figure 2. Electrophoretic gel patterns for lipoplex-associated DNA in gel retardation for gemini lipids: (A) 2a, (B) 2b, (C) 2c, (D) 2d, and (E)
2e. The N/P ratios are indicated at the top of each lane.
different lipid/DOPE mol ratios was found to be less effective
both in terms of the number of transfected cells and the MFI.
2.5.2. Optimization of N/P Ratio. After optimizing the lipid/
DOPE ratio for every lipid, all the gemini lipids were tested by
taking the identical amount of DNA (0.8 µg) and varying the
amount of lipid using the respective optimized lipid/DOPE mol
ratio (Figure 4). Lipid 2a was able to transfect to a maximum
extent of 40% of the cells with MFI of ∼125 at N/P ratio of
0.25, whereas lipid 2b could transfect approximately 50% of
the cells with a nearly identical MFI at the same N/P ratio.
Transfection efficiency was found to decrease at higher N/P
ratios, especially in terms of MFI. Gemini lipid 2c was able to
transfect only 40% of cells at N/P of 0.5 or 0.75, but the MFI
observed was found to be very high as compared to that showed
by either lipids 2a or 2b. The MFI observed with lipid 2c was
∼150, which was approximately 2 times greater than that
affected by gemini lipids 2a and 2b at N/P ratio of 0.5. Gemini
lipid 2d showed a maximum transfection efficiency of 40-60%
with an MFI of ∼120 at N/P ratio of 0.5. At N/P ratio of 0.25,
the maximum MFI of ∼145 was observed, but the number of
transfected cells observed was less as compared to that at N/P
ratio of 0.5. But with the dodecamethylene-spacer-based gemini
lipid (2e), the transfection efficiency decreased drastically.
Nearly 50% of all the cells get transfected with a low MFI of
∼60, when gemini lipid 2e was used at N/P ratio of 0.5 with
EGFP-c3. The number of transfected cells at the N/P ratio of
1.0 was high, but the MFI was found to be very low. All the
gemini lipids were found to be better transfecting agents than
the monomeric counterpart 1. The number of transfecting cells
was 60% in the case of monomeric lipid 1, but the MFI observed
was very less. All the gemini lipids followed a bell-shaped
graph, when transfection efficiencies were plotted against N/P
ratio. When all the gemini lipids were compared at N/P ratio
of 0.5 (Figure S2, Supporting Information), lipid 2c was found
to be the best transfecting agent in terms of MFI, although the
number of transfected cells was found be less. The transfecting
efficiency increased with the increase in the spacer chain length
from lipid 2a to lipid 2c and then decreased while going from
lipid 2c to 2d or 2e. Lipid 2c was found to have the maximum
MFI, which was in fact comparable to that of formulation E,
one of the best currently commercially available transfecting
agents.
2.5.3. Effect of Serum. One of the major drawbacks of the
known cationic lipids for their in ViVo use is the inhibition of

Cholesterol-Based Gemini Lipids
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 10 2435
Figure 3. Transfection efficiencies of EGFP-c3 DNA of cholesterol-based gemini lipids with various compositions of DOPE: (a) 2a, (b) 2b, (c)
2c, (d) 2d, and (e) 2e. The concentration of DNA ) 0.8 µg/well, and lipid was used at N/P ratio of 0.5. Data are expressed as the number of
transfected cells and MFI as obtained from flow cytometry analysis.
the transfection efficiency of cationic liposomes in the presence
of serum. Very few cationic lipids are known which demonstrate
transfection activity in the presence of serum,10,21,30 although
their efficiency is quite low. To our knowledge there is no
gemini lipid known in the literature that shows transfection
activity in the presence of serum. For instance, monomeric lipid
1, which is found to be a good transfecting agent as a coliposome
with DOPE, loses much of its efficiency as a transfection reagent
in the presence of serum. Maintenance of the transfection
efficiency in the absence of serum even for in Vitro experiments
is also cumbersome often due to the toxic nature of the synthetic
cationic lipid formulations.
Therefore, in order to investigate the effect of serum on the
gene transfection efficiencies of cholesterol-based gemini lipids,
we performed transfection experiments in the presence of 10%
serum with the optimized lipid/DOPE formulation at different
N/P ratios using plasmid EGFP-c3 and analyzed the data by
flow cytometry (Figure 4). The transfection efficiency of
monomeric lipid 1 decreased drastically, whereas in the case
of the commercially available reagent, formulation E, there was
∼70% decrease in the MFI of the transfected cells in the
presence of serum. In the case of gemini lipid 2a, there was no
change in the number of transfected cells, but the MFI decreased
in the presence of serum. This shows that lipid 2a is able to
transfect the same number of cells in the presence of serum,
but the amount of DNA delivered gets reduced in the presence
of serum, whereas in the case of lipid 2b the reduction in
transfection efficiency was not so appreciable as compared to
that of lipid 2a. Serum hardly inhibits the transfection efficiency
of lipid 2c both in terms of the percentage of transfected cells
and the MFI. As a matter of fact there was little enhancement
in the transfection efficiency in the presence of serum in case
of lipid 2c, both in terms of percentage of transfected cells and
MFI. At N/P ratio of 0.25, gemini lipid 2c showed an ∼1.5-
fold increase in MFI, and the transfection efficiency was found
to better than that of formulation E (Figure 4). Lipid 2c is able
to transfect nearly 50% of the cells with MFI of ∼175 in the
presence of serum at N/P ratio of 0.5-0.75, which is greater
than formulation E. In the case of the gemini lipid 2d, the
presence of serum did not inhibit the transfection efficiency at
all. In fact, there was some increase in the number of transfected
cells and the MFI in the presence of serum in this instance. To
our pleasant surprise, there was a significant increase in the
transfection efficiency of gemini lipid 2e in the presence of
serum as compared to the one carried out in the absence of
serum. Lipid 2e was able to transfect only to the extent of 35-
40% of the cells with MFI of ∼60 in the absence of serum, but
in the presence of serum, the transfection efficiency increased
to 60-70% with MFI of ∼110 at N/P ratio of 0.5. Similarly, in
the presence of serum, all gemini lipids followed a bell-shaped
curve, possessing maximum transfection at N/P ratio of 0.5-
0.75.
All the gemini lipids were found to be better transfecting
agents than the monomeric one even in the presence of serum.
Even the transfection efficiency of gemini lipid 2c was found
to be greater than one of the best commercially available
reagents, formulation E, which is also known to show trans-
fection in the presence of serum. Gemini lipid 2e showed a
2-fold increase in the transfection efficiency in the presence of
serum both in terms of the number of transfected cells and the
MFI as compared to that in the absence of serum. Flow
cytometric scans of transfection experiments using lipid the

2436 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 10
Bajaj et al.
Figure 4. Transfection efficiencies of gemini lipids using optimized lipid/DOPE formulations at various N/P ratios in the absence and the presence
of serum: (a) 2a, (b) 2b, (c) 2c, (d) 2d, and (e) 2e. The concentration of DNA ) 0.8 µg/well. Data are expressed as the number of transfected cells
and MFI as obtained from flow cytometry analysis.
2e/DOPE (1:4 mol ratio) formulation at N/P ratio of 0.5 in the
absence (-FBS-FBS) and presence (-FBS+FBS) of serum
are shown in Figure 5. Overall, there was a nearly 2-fold
increase in the number of transfected cells (M2 and M3 region),
whereas there was a 3-fold increase in number of highly
transfected cells (see M3 region) in the presence of serum. This
suggests that there are some serum components that must be
facilitating the efficient transfection activity with the liposome
prepared from this gemini lipid. This was found to happen only
in the case of the gemini lipid 2e, suggesting that there are some
structural features at the aggregation level in this lipid that
support the enhancement of gene transfection activities in the
presence of serum.
2.5.4. Effect of the Variation of the Amount of DNA. To
see whether variation in the amount of the DNA affects the
transfection efficiency of gemini lipids, we have performed
transfection experiments with all the gemini lipids at N/P ratios
of 0.25 and 0.5, varying the amount of the DNA from 0.4 to
2.0 µg/well. In the case of lipid 2a, there was no enhancement
in the transfection efficiency at high DNA amounts, although
there was a little improvement when 0.4 µg of DNA was used
(not shown). Similarly in case of lipid 2b, at both N/P ratios,
transfection efficiencies decreased at higher DNA amounts, but
at N/P ratio of 0.25, using 0.4 µg of DNA there was little
improvement in the gene transfection efficiency. There was an
increase in the transfection efficiency with an increase in the
amount of DNA at both N/P ratios in the case of lipid 2c, the
maximum being observed using 1.2-1.6 µg of DNA. At a N/P
ratio of 0.5, using 1.6 µg of DNA nearly 70% of the cells get
transfected with MFI of ∼160, which is comparable with
formulation E, both in terms of percentage of transfected cells
and MFI (Figure 6). As already shown, MFI comparable to that
of formulation E was observed using 0.8 µg of DNA at a N/P
ratio of 0.5, whereas the number of transfected cells was only
40%. Now using 1.6 µg of DNA at N/P ratio of 0.5 the number
of transfected cells increased to 70%, which is comparable to
that of formulation E without any change in MFI. In the case
of lipid 2d there was no enhancement in gene transfection
efficiency, except the little improvement in the MFI at N/P ratio
of 0.5 using 0.4 µg of DNA (not shown). We have shown that
lipid 2e shows better transfection efficiency in the presence of
serum; therefore, transfection studies with varying DNA amounts
were performed in the presence of serum for the lipid 2e/DOPE
(1:4) formulation. With the increase in the amount of DNA,

Cholesterol-Based Gemini Lipids
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 10 2437
Figure 5. Flow cytometric scans of transfection experiments using the lipid 2e/DOPE (1:4 mol ratio) formulation, when transfection experiments
were performed in the absence (-FBS-FBS) and presence (-FBS+FBS) of serum. Untransfected cells, used as the control, are also shown. M1
marks the population of cells having fluorescence under control, M2 marks the population of cells with medium fluorescence intensity, and M3
marks transfected cells with high fluorescence intensity.
Figure 6. Effect of variation of the amount of DNA on the gene
transfection efficiency of the gemini lipid 2c/DOPE (1:4 mol ratio)
formulation at N/P ratio of 0.5.
Figure 7. Flow cytometric scans of gene transfection performed using
the lipid 2c/DOPE (1:4 mol ratio) formulation at N/P ratio of 2.0 and
formulation E in high percentages of serum: (a) 30% serum; (b) 50%
serum.
there is little increase in the number of transfected cells without
any significant change in MFI. Lipid 2e is able to transfect
nearly 65% of the cells with an MFI of ∼100 (not shown).
Therefore, DNA variation experiments suggest that among the
gemini lipids, lipid 2c shows the maximum transfection
comparable to that of formulation E, whereas lipid 2e shows a
very good transfection efficiency of 65% with an MFI of 100
in the presence of serum.
2.5.5. GFP Assay at High Serum Concentrations. To see
the effect of the high concentrations of serum on gene
transfection efficiencies, we have performed the gene transfec-
tion efficiencies using the best lipid 2c/DOPE (1:4) formulation.
Flow cytometric scans of transfection experiments using the
gemini lipid 2c/DOPE (1:4) formulation and the formulation E
in the presence of 30% and 50% serum concentrations are shown
in Figure 7. FACS profiles indicate that the lipid 2c/DOPE
formulation is able to transfect nearly 40% of the cells at a high
percentage of serum as well at N/P ratio of 2.0, whereas
formulation E is able to transfect only 25% of the cells at 50%
serum concentration. In terms of fluorescence intensity, formu-
lation E is able to transfect a lesser number of cells with high
MFI, whereas the lipid 2c/DOPE (1:4) formulation transfects
more cells with a low MFI. Here it should be noted that in
formulation-E-based transfection, an EC buffer and enhancer
are also used, whereas gemini lipid formulations are easy to
prepare. The low MFI intensity observed in the case of the lipid
2c/DOPE (1:4) formulation is expected as, at such high serum
concentrations, negatively charged serum proteins start to
compete with DNA for lipid complexation, therefore reducing
the number of copies of the plasmid delivered inside the cell.

2438 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 10
Bajaj et al.
Figure 8. Flow cytometric scans of transfection experiments using
the lipid 2c/DOPE (1:4 mol ratio) formulation and formulation E.
Untransfected cells, used as the control, are also shown. M1 marks the
population of cells having fluorescence under control, M2 marks the
population of cells with medium fluorescence intensity, and M3 marks
transfected cells with high fluorescence intensity.
Figure 9. Transfection efficiency of the gemini lipid 2c/DOPE (1:4
mol ratio) formulation at different N/P ratios as evidenced by luciferase
expression in the absence and presence of serum. Shown are the data
normalized with respect to formulation E.
nm. In the case of lipid 2d-based lipoplexes, globular aggregates
of 200-400 nm size were observed along with the presence of
some uncomplexed, naked DNA. Lipid 2e/DNA lipoplexes
showed the existence of irregular morphologies. As the lipid
2e/DOPE formulations were the least efficient transfecting
agents in the absence of serum and better transfecting agents in
the presence of serum, it is possible that these different
morphologies could as well be responsible for their high
transfection efficiency in the presence of serum.
2.8. Cytotoxicity Assay. MTT-based cell viability assays
were performed in HeLa cells across the entire range of lipid/
DNA charge ratios (N/P) used in the actual transfection
experiments. The percentage cell viabilities of all the gemini
lipids except lipid 2e were found to be very high at all the
concentrations used for the transfection experiments (Figure 11).
The most potent transfecting gemini lipid 2c was found to be
nontoxic at all concentrations studied.
2.9. Conclusions. For the first time, synthesis of five
cholesterol-based gemini cationic lipids with varying spacer
chain lengths between the cationic ammonium headgroups has
been achieved. These lipids form stable suspensions in water
easily. Electron microscopy revealed the presence of closed
membranous aggregates in these lipid aggregates. These gemini
lipids in the presence of helper lipid, DOPE, showed a
significant enhancement in the gene transfection activities as
compared to their monomeric lipid counterpart. With the
increase in length of the spacer from trimethylene to penta-
methylene, the transfection efficiency was found to increase,
whereas further increase in the spacer length to dodecamethylene
led to a decrease in the transfection activity. All the gemini
lipids were found to be more effective than monomeric lipid 1.
Gemini lipid 2c was found to be the most effective lipid, which
showed transfection activity greater than that of formulation E,
one of best-known commercially available transfecting reagents.
Lipid 2c was nearly 2 times more effective than formulation E.
All the gemini-lipid-based DOPE formulations did not possess
any toxicity at the concentrations at which transfections were
performed.
2.6. Comparison of Gemini Lipid 2c and Formulation E.
2.6.1. GFP Expression. Flow cytometric scans of transfection
experiments using the gemini lipid 2c/DOPE (1:4) formulation
and formulation E are shown as an overlay in Figure 8. These
profiles have been divided into three regions: M1 presents the
number of cells under control (untransfected), M2 presents the
transfected cells with medium fluorescence intensity, whereas
M3 presents highly fluorescent transfected cells. It has been
observed that the number of transfected cells using formulation
E and the gemini lipid 2c is almost the same, but the number
of highly fluorescent cells (see region M3) is almost 2 times
greater in the case of the lipid 2c formulation as compared to
formulation E, whereas the number of low fluorescent cells (see
region M2) is high in the case of formulation E as compared to
that of the gemini lipid 2c. This proves that formulation E is
able to transfect a comparable number of cells as gemini lipid
2c, but gemini lipid 2c transfects more cells with high
fluorescent intensity.
2.6.2. Luciferase Assay. To quantify the transgenic expres-
sion for the gemini lipid 2c, we used luciferase gene expression
assay in HeLa cells. Transfection efficiencies using the lipid
2c/DOPE (1:4) formulation at different N/P ratios as obtained
from luciferase assay, normalized with respect to formulation
E, are shown in Figure 9. Upon transfection with the luciferase
gene, pGL3-control, it was observed that the total amount of
the luciferase (protein) formed, or the average transgene
expression, was more than 2 times higher than the formulation
E. Serum does not inhibit the luciferase activity of the lipid
2c/DOPE formulation.
2.7. Characterization of Lipid-DNA Complexes by TEM.
The morphology of the lipid/plasmid complexes was visualized
under electron microscope after negative staining using uranyl
acetate. The complexes formed with different gemini lipids at
their optimized transfection ratios with plasmid DNA revealed
some distinct morphological changes depending on the length
of the spacer in between the cationic ammonium headgroups
of the gemini lipid molecules (Figure 10). The lipoplexes formed
from gemini lipid 2a/DNA had the tendency to form elongated
organizations like a network of rods. Gemini lipid 2b- and 2c-
based lipoplexes showed aggregated structures, but these
lipoplex aggregates were larger in size ranging from 100 to 200
Serum is known to be a major impediment to the applications
of cationic lipids for their use in ViVo. So there is always a
challenge to design such cationic lipids, which can be useful
for transfection studies in the presence of serum. All the gemini

Cholesterol-Based Gemini Lipids
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 10 2439
Figure 10. Transmission electron micrographs for the lipoplexes prepared from the most effective gemini lipid/DOPE liposome/DNA complexes
at the optimized N/P ratio: (a) 2a, (b) 2b, (c) 2c, (d) 2d, and (e) 2e.
novel cholesterol-based gemini lipids should be of interest to
researchers working in the field of gene therapy using nonviral
vectors.
3. Experimental Section
3.1. Materials and Methods. All reagents, solvents, and
chemicals used in this study were of the highest purity available.
The solvents were dried prior to use. Column chromatography was
performed using 60-120 mesh silica gel. NMR spectra were
recorded using a Jeol JNM λ-300 (300 MHz for 1H and 75 Hz for
13C) spectrometer. The chemical shifts (δ) are reported in ppm
downfield from the internal standard, TMS, for 1H-NMR and 13C-
Figure 11. MTT assay-based cellular cytotoxicity of gemini lipids
(2a-2e) against HeLa cells. The percent viability values shown are
the average of triplicate experiments performed on the same day.
NMR. Mass spectra were recorded on a Kratos PCKompact SEQ
V1.2.2 MALDI-TOF spectrometer, a MicroMass ESI-TOF spec-
trometer, or a Shimadzu table-top GC-MS or ESI-MS (HP1100LC-
MSD). Infrared (IR) spectra were recorded on a Jasco FT-IR 410
spectrometer using KBr pellets or neat. Gemini lipids were
synthesized as described below and were characterized fully by
their 1H-NMR, 13C-NMR, mass spectra, and elemental analysis.
3.2. Synthesis. 3.2.1. Cholest-5-en-3â-tosylate (4). To an ice-
cooled solution of cholesterol (3) (5.0 g, 0.013 mol) in dry pyridine
(5 mL) and dry chloroform (5 mL), tosyl chloride (3.7 g, 0.02 mol)
was added. A catalytic amount of DMAP was also added. The
reaction mixture was then allowed to stir at 0 °C for 6 h. To the
reaction mixture, chloroform (35 mL) was added, and the reaction
mixture was washed with 1 N HCl (2 × 50 mL), water (50 mL),
and brine (50 mL); the organic layer was separated and dried over
anhydrous Na2SO4. Chloroform from this solution was evaporated
to leave a residue. From the residue cholest-5-en-3â-tosylate (4)
was recrystallized using chloroform and methanol. Yield: white
solid, 6.49 g, 0.012 mmol, 92.8%. Mp: 133 °C. Lit mp: 132-133
°C.32 IR (neat) (cm-1): 2949, 2867, 1598, 1495, 1467, 1366, 1188,
lipids except lipid 2a showed enhanced transfection activity in
the presence of serum as well, whereas transfection by mono-
meric lipid 1 got diminished dramatically in the presence of
serum. Lipid 2e showed an unusual feature of enhanced gene
transfection in the presence of serum as compared to that in
the absence of serum. To see the generality of the transfection
by gemini lipids, transfection experiments performed in the
U373 cell line, using the best gemini lipid 2c, show 60-70%
transfection efficiency with an MFI of 100 (Figure S2, Sup-
porting information). Transmission electron micrographs of the
lipoplexes showed distinct morphological changes depending
upon the length of the spacer between the cationic headgroups.
We have been able to reduce their cytotoxicity and enhance
the transfection efficiency significantly. The simplicity of the
use of molecular building blocks like cholesterol and their high
chemostability and shelf life make these formulations particu-
larly attractive.
1
and 1174. H-NMR (CDCl3, 300 MHz): δ: 0.65 (s, 3H), 0.84-
2.29 (m, 41H), 2.44 (s, 3H), 4.32 (m, 1H), 5.30 (d, 1H, J ) 4.5
Hz), 7.31-7.34 (d, 2H, J ) 8.1 Hz), 7.78-7.81 (d, 2H, J ) 8.1
Hz).
Although the DNA delivery to eukaryotic cells involves
several steps31 and the mechanisms of many of these steps are
not very clear at the molecular level, it would be important to
note some correlation between the molecular structures of these
new gemini lipids, the membrane level properties, and the
transfection properties. Whatever may be the actual mechanism
for transfection mediated by this class of gemini lipid formula-
tions, the interesting and meaningful results obtained with these
3.2.2. Cholest-5-en-3â-oxyethane (5). Chlolest-5-en-3â-tosylate
(4) (3.5 g, 6.5 mmol) was taken in anhydrous dioxane. To this dry
ethylene glycol (10 g, 0.16 mol) was added, and the mixture was
refluxed under nitrogen for 4 h. The solution was cooled, and the
solvent was removed under vacuum. A white residue was obtained
which was dissolved in chloroform (50 mL) and washed with water.
The organic layer was separated, washed with NaHCO3 (50 mL),

2440 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 10
Bajaj et al.
water (50 mL), and brine (50 mL), and dried over anhydrous Na2-
SO4. Finally, this solvent was removed in vacuo and the product,
cholest-5-en-3â-oxyethane (5), was purified by column chroma-
tography over silica gel using a mixture of petroleum ether and
ethyl acetate. Yield: white solid, 2.3 g, 5.3 mmol, 85%. Mp: 99-
100 °C. Lit mp: 97-98 °C.32 IR (neat) (cm-1): 3438, 2936, 2868,
3.2.7. Lipid 2b. 1H-NMR (300 MHz, CDCl3): δ: 0.67 (s, 6H),
0.85-2.35 (m, 86H), 3.32 (m, 2H), 3.34 (br s, 12H), 3.74 (br s,
4H) 3.93 (br s, 4H), 4.02 (br s, 4H), 5.36 (d, J ) 4.5 Hz, 2H).
13C-NMR (CDCl3, 75 MHz): δ: 11.86, 18.70, 19.36, 19.94,
21.019 22.53, 22.80, 23.85, 24.28, 28.00, 28.14, 28.21, 31.83, 31.89,
35.78, 36.17, 36.75, 36.90, 38.75, 39.49, 39.72, 42.29, 50.04, 56.16,
56.67, 61.09, 61.80, 65.29, 79.99, 122.43, 139.77. ESI-MS: 485.9
(M+2/2). Anal. (C66H120N2O2Br2) C, H, N.
1
1466, and 1376. H-NMR (CDCl3, 300 MHz): δ: 0.67 (s, 3H),
0.85-2.35 (m, 41H), 3.20 (m, 1H), 3.57-3.60 (t, 2H), 3.72 (t, 2H),
5.34 (d, 1H). ESI-MS: 453 (M + Na+).
3.2.8. Lipid 2c. 1H-NMR (300 MHz, CDCl3): δ: 0.67 (s, 6H),
0.85-2.35 (m, 88H), 3.20 (m, 2H), 3.39 (br s, 12H), 3.81 (br. s,
4H) 3.90 (br s, 8H), 5.36 (d, J ) 4.5 Hz, 2H). 13C-NMR (CDCl3,
75 MHz): δ: 11.83, 18.69, 19.34, 21.04, 21.81 22.52, 22.77, 23.82,
24.30, 27.98, 28.14, 28.21, 31.81, 31.88, 35.76, 36.12, 36.75, 36.88,
38.74, 39.48, 39.68, 42.27, 50.02, 56.13, 56.67, 61.87, 63.69, 65.45,
79.91, 122.43, 139.76. ESI-MS: 492.4 (M+2/2), 1064.7, 1066.9
(M+2 + Br-). Anal. (C67H122N2O2Br2) C, H, N.
3.2.3. Cholest-5-en-3â-oxyethane tosylate (6). To an ice-cooled
solution of cholest-5-en-3â-oxyethan-2-ol (5) (2.2 g, 5.1 mmol) in
dry pyridine (5 mL) and dry chloroform (5 mL), tosyl chloride
(1.5 g, 7.86 mmol) was added. The reaction mixture was allowed
to stir at 0 °C for 6 h. To the reaction mixture chloroform (40 mL)
was added, and then the reaction mixture was washed with 1 N
HCl (2 × 50 mL), water (50 mL), and brine (50 mL). Finally, the
organic layer was separated and dried over anhydrous Na2SO4. The
solvent was removed using a rotary evaporator, and the product,
cholest-5-en-3â-oxyethane tosylate (6), was purified by column
chromatography over silica gel using a mixture of petroleum ether
and ethyl acetate. Yield: 2.68 g, 7.04 mmol, 90.0%. IR (neat)
(cm-1): 2935, 2886, 1598, 1465, 1360, 1189, and 1177. 1H-NMR
(CDCl3, 300 MHz): δ: 0.67(s, 3H), 0.85-2.25 (m, 41H), 2.44 (s,
3H), 3.06-3.13 (m, 1H), 3.63-3.66 (t, 2H), 4.13-4.16 (t, 2H),
5.31 (d, 1H), 7.32-7.35 (d, 2H), 7.79-7.82 (d, 2H). ESI-MS: 607
(M + Na+).33
3.2.9. Lipid 2d. 1H-NMR (300 MHz, CDCl3): δ: 0.67 (s, 6H),
0.85-2.31 (m, 90H), 3.20 (m, 2H), 3.40 (br s, 12H), 3.81-3.92
(br m, 12H), 5.36 (d, J ) 4.5 Hz, 2H). 13C-NMR (CDCl3, 75
MHz): δ: 11.79, 18.64, 19.31, 20.99, 21.76, 22.49, 22.73, 23.80,
24.23, 24.67, 27.91, 28.08, 31.76, 31.85, 35.73, 36.12, 36.70, 36.85,
38.69, 39.43, 39.66, 42.23, 49.97, 56.09, 56.62, 61.90, 63.51, 65.70,
79.78, 122.27, 139.82. ESI-MS: 499.2 (M+2/2). Anal. (C68H124N2O2-
Br2‚2.5H2O) C, H, N.
3.2.10. Lipid 2e. 1H-NMR (300 MHz, CDCl3): δ: 0.67 (s, 6H),
0.85-2.30 (m, 102H), 3.20 (m, 2H), 3.39 (br s, 12H), 3.68 (br m,
4H), 3.89-3.92 (br m, 8H), 5.34 (d, J ) 4.5 Hz, 2H). 13C-NMR
(CDCl3, 75 MHz): δ: 11.83, 18.67, 19.31, 21.01 22.52, 22.72,
22.77, 23.80, 24.25, 25.94, 27.96, 28.09, 28.18, 28.59, 31.81, 31.86,
35.74, 36.14, 36.73, 36.85, 38.74, 39.46, 39.68, 42.26, 50.01, 51.83,
56.11, 56.65, 61.93, 63.41, 66.23, 79.93, 122.45, 139.74. ESI-MS:
485.9 (M+2/2). Anal. (C74H136N2O2Br2) C, H, N.
3.3. Liposome Preparation. Individual lipid or its mixture with
DOPE in the desired mol ratio was dissolved in chloroform in
autoclaved Wheaton glass vials. Thin films were made by evapora-
tion of the organic solvent under a steady stream of dry nitrogen.
The last traces of organic solvent were removed by keeping these
films under vacuum overnight. Freshly autoclaved water (Milli-Q)
was added to the individual film such that the final concentration
of the cationic lipid was 0.5 mM. The mixtures were kept for
hydration at 4 °C for 10-12 h and were repeatedly freeze-thawed
(ice-cold water to 60 °C) with intermittent vortexing to ensure
hydration. Sonication of these suspensions for 15 min in a sonicator
bath at 60 °C afforded closed, cationic liposomes as evidenced from
transmission electron microscopy. Liposomes were prepared and
kept under sterile conditions. Formulations were stable and, if stored
frozen, possessed long shelf life.
3.4. Transmission Electron Microscopy. Freshly prepared
aqueous suspensions of each cationic lipid (0.5 mM) or lipoplex
were examined under transmission electron microscopy by negative
staining using 1% uranyl acetate. A 10 µL sample of the suspension
was loaded on Formvar-coated, 400 mesh copper grids and allowed
to remain for 1 min. Excess fluid was wicked off the grids by
touching their edges to filter paper, and 10 µL of 1% uranyl acetate
was applied on the same grid for 1 min, after which the excess
stain was similarly wicked off. The grid was air-dried for 30 min,
and the specimens were observed under TEM (JEOL 200-CX)
operating at an acceleration voltage of 120 keV. Micrographs were
recorded at a magnification of 5000-20 000X.
3.5. Plasmid DNA. pEGFP-c3 (Clontech, U.S.A.), which
encodes for an enhanced green fluorescence protein (GFP) under
a CMV promoter, was amplified in Escherichia coli (DH5R) and
purified using Qiagen Midi Prep plasmid purification protocol
(Qiagen, Germany). The purity of the plasmid was checked by
electrophoresis on 1.0% agarose gel. The concentration of DNA
was estimated spectroscopically by measuring the absorption at 260
nm and confirmed by gel electrophoresis. Plasmid preparations
showing a value of OD260/OD280 > 1.8 were used.
3.2.4. Cholest-5-en-3â-oxyethan-N,N-dimethylamine (7). Di-
methylamine was dissolved in dry ethanol in a screw-top pressure
tube at 0 °C. The cholest-5-en-3â-oxyethane tosylate (6) (250 mg,
0.5 mmol) dissolved in dry MeOH (2 mL) was carefully added to
the dimethylamine solution at 0 °C. The reaction mixture inside
the pressure tube was screw capped and heated at 80 °C for 24 h.
The remaining dimethylamine was evaporated, and the solvent was
removed by rotary evaporation to afford a yellowish mass. The
crude product was dissolved in CHCl3 (25 mL) and washed with
saturated NaHCO3 (25 mL), water (25 mL), and then brine solution
(25 mL). The final product was a yellowish solid (206 mg, 0.5
mmol, 99%). Mp: 145 °C. Rf ) 0.3 (CHCl3). FT-IR (neat) (cm-1):
1
2936, 2866, 1465, 1377, 1274, 1195, 1110, and 1078. H-NMR
(300 MHz, CDCl3): δ: 0.67 (s, 3H), 0.85-2.20 (m, 41H), 2.26
(s, 6H), 2.38 (m, 1H), 2.47-2.51 (t, J ) 6.0 Hz, 2H), 3.11-3.18
(m, 1H), 3.54-3.59 (t, J ) 6.0 Hz, 2H), 5.34 (d, J ) 4.5 Hz, 1H).
13C-NMR (CDCl3, 75 MHz): δ: 11.66, 18.54, 19.18, 20.87, 22.40,
22.65, 23.69, 24.10, 27.80, 28.05, 28.13, 31.68, 31.73, 35.61, 36.01,
36.65, 37.06, 38.87, 39.33, 39.59, 42.09, 45.63, 49.97, 55.99, 56.55,
59.05, 65.72, 79.15, 122.27, 140.66. ESI-MS: 458 (M+ + 1).
3.2.5. General Method for the Synthesis of Gemini Lipids
(2a-2e). A solution of cholest-5-en-3â-oxyethan-N,N-dimethyl-
amine (7) (0.2 mmol) and an appropriate R,ω-dibromoalkane (0.07
mmol) in dry MeOH-EtOAc (4 mL, v/v: 1/1) was refluxed over
a period of 48-72 h in a screw-top pressure tube, until TLC
indicated complete disappearance of the dibromide. After that, the
reaction mixture was cooled and the solvent was evaporated to
furnish a crude solid. This was repeatedly washed with ethyl acetate
to remove any of the cholest-5-en-3â-oxyethan-N,N-dimethylamine
(7), and the residue was finally subjected to repeated crystallizations
from a mixture of MeOH and ethyl acetate. This afforded a white
solid in each case. The product yields ranged from 50% to 60%.
The purities of these lipids were ascertained from TLC; the Rf
ranged from 0.2 to 0.3 in 10:1 CHCl3/MeOH. All the new gemini
lipids were fully characterized by 1H-NMR, 13C-NMR, mass
spectrometry, and C, H, N analysis. Pertinent spectroscopic and
analytical data are given below.
1
3.2.6. Lipid 2a. H-NMR (300 MHz, CDCl3 + CD3OD): δ:
0.68 (s, 6H), 0.85-2.34 (m, 84H), 3.32 (br s, 14H), 3.74 (br s,
8H) 3.92 (br s, 4H), 5.37 (d, J ) 4.5 Hz, 2H). 13C-NMR (CDCl3
+ CD3OD, 75 MHz): δ: 11.74, 18.60, 19.23, 21.01, 22.45, 22.68,
23.74, 24.18, 27.91, 28.03, 28.11, 31.75, 31.80, 35.69, 36.09, 36.68,
36.85, 38.64, 39.41, 39.64, 39.41, 39.64, 42.21, 49.99, 56.06, 56.62,
3.6. Cell Culture. Cells (HeLa, U373) were cultured in Dul-
becco’s modified Eagle’s medium (DMEM, Sigma) supplemented
with 10% fetal bovine serum (FBS) in T25 culture flasks (Nunc,
61.54, 62.10, 64.56, 79.90, 122.40, 139.66. ESI-MS: 478.5 (M+2
/
2), 1035.4 (M+2 + Br-). Anal. (C65H118N2O2Br2‚4.5H2O) C, H, N.

Cholesterol-Based Gemini Lipids
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 10 2441
Denmark) and were incubated at 37 °C in a humidified atmosphere
containing 5% CO2. Cells were regularly passaged by trypsinization
with 0.1% trypsin (EDTA 0.02%, dextrose 0.05%, and trypsin 0.1%)
in PBS (pH 7.2).
3.7. Cytotoxicity. The toxicity of each cationic lipid formulation
toward HeLa cells in the presence of 10% FBS was determined
using 3-(4,5-dimethylthiazole-2-yl)-2,5-diphenyltetrazolium bromide
reduction assay following literature procedures.34,35
removed from the wells; the cells were washed with PBS and
trypsinized by adding 100 µL of 0.1% trypsin. To each well 200
µL of PBS containing 20% FBS was added. Duplicate cultures were
pooled and analyzed by flow cytometry immediately using a Becton
and Dickinson flow cytometer equipped with a fixed laser source
at 488 nm.
3.11. Luciferase Assay. The efficiency of transfection was
determined by checking the activity of luciferase protein expressed
using a single-luciferase assay kit provided by Promega (U.S.A.)
following manufacturer’s protocol. For a typical assay in a 24-
well plate, ∼48 h post-transfection, the old medium was removed
from the wells and the cells were washed twice with 200 µL of
PBS. To each well 100 µL of cell lysis buffer was then added, and
the cells were lysed for 30 min in a horizontal rocker at room
temperature (RT). The cell lysate was transferred completely to
Eppendorf tubes and centrifuged (4000 rpm, RT) for 2 min; the
supernatant was transferred to Eppendorf tubes and stored in ice.
For the assay, 10 µL of this supernatant and 10 µL of luciferase
assay substrate (Promega) were used. The lysate and the substrate
were both thawed to RT before performing the assay. The substrate
was added to the lysate, and the luciferase activity was measured
in a luminometer (Turner designs, 20/20, Promega, U.S.A.) in
standard single-luminescence mode. The measurement was per-
formed for 10 s. A delay of 2 s was given before each measurement.
The protein concentration in the cell lysate supernatant was
estimated in each case using Bradford’s method with bovine serum
albumin as a standard.37 Comparison of the transfection efficiencies
of the individual lipids was made based on data for luciferase
expressed as either relative light units (RLU) or relative light units
(RLU)/µg of protein.
3.8. Transfection Procedure. All transfection experiments were
carried out in HeLa Cells in antibiotic-free media unless specified
otherwise. In a typical experiment, 24-well plates were seeded with
45 000 cells/well in antibiotic-free media 24 h before transfection
such that they were at least ∼70% confluent at the time of
transfection. For transfection, lipid formulation and DNA were
serially diluted separately in DMEM containing no serum to have
the required working stocks. DNA was used at a concentration of
0.8 µg/well unless specified otherwise. The lipid and DNA were
complexed in a volume of 200 µL by incubating the desired amount
of lipid formulation and DNA together at room temperature for
about 30 min. The lipid concentrations were varied so as to obtain
the required lipid/DNA (N/P) charge ratios. Charge ratios here
present the ratio of charge on the cationic lipid (in mol) to nucleotide
base molarity and were calculated by considering the average
nucleotide mass of 330. After 30 min of complexation, 200 µL of
media was added to the complexes (final DNA concentration )
12.12 µM). Old medium was removed from the wells, the cells
were washed with DMEM, and the lipid-DNA complexes in 200
µL media were added to the cells. The plates were then incubated
for 6 h at 37 °C in a humidified atmosphere containing 5% CO2.
At the end of the incubation period, the medium was removed and
the cells were washed with DMEM; 500 µL of DMEM containing
10% FBS was added per well. Plates were further incubated for a
period of 42 h before checking for the reporter gene expression.
GFP expression was examined by fluorescence microscopy and was
quantified by flow cytometry analysis. Control transfections were
performed in each case by using commercially available transfection
reagent “Effectene” using standard conditions specified by the
manufacturer. Here, “Effectene” is represented by formulation E.
All the experiments were done in duplicate, and the results presented
are the average of at least two such independent experiments done
on two different days. For comparison, transfections using lipid 1
were performed using the optimized lipid/DOPE ratio19 of 1:1 at a
N/P ratio of 1.0, using 0.8 µg of plasmid DNA.
Acknowledgment. This work was supported by the Depart-
ment of Biotechnology, Government of India, New Delhi, India.
A.B. thanks CSIR for a senior research fellowship. We thank
Dr. Omana Joy and Ms. Padmini for their help during the work.
Supporting Information Available: Elemental analysis values,
structure-activity graph, and flow cytometric scans of transfection
experiments. This material is available free of charge via the Internet
at http://pubs.acs.org.
References
For transfections in the presence of serum, lipid and DNA were
separately diluted in serum-free media as already mentioned and
the complexation was done in serum-free media (200 µL) for 30
min. The complex was then diluted to 400 µL with DMEM
containing 20% FBS so as to achieve a final serum concentration
of 10%. The cells were then incubated with this complex for 6 h.
At the end of the incubation period, the medium was removed and
the cells were washed with DMEM; 500 µL of DMEM containing
10% FBS was added per well. For transfections at 30% and 50%
of serum concentrations, complexes were diluted with DMEM
containing 60% FBS or with neat FBS, respectively.
3.9. Gel Electrophoresis. To examine the complexation of DNA
with cationic lipid suspensions at different lipid/DNA ratios, we
prepared lipid-DNA complexes at different lipid/DNA charge
ratios in an identical manner as was done with the transfection
experiments. After 30 min of incubation, these complexes were
electrophoretically run on a 1.0% agarose gel. The uncomplexed
DNA moved out of the well, but the DNA that was complexed
with lipid remained inside the well.36
(1) Friedmann, T. Human gene therapysan immature genie, but certainly
out of the bottle. Nat. Med. 1996, 2, 144-147.
(2) Kay, M. A.; Glorioso, J. C.; Naldini, L. Viral vectors for gene
therapy: the art of turning infectious agents into vehicles of
therapeutics. Nat. Med. 2001, 7, 33-40.
(3) Marshall, E. Gene therapy: what to do when clear success comes
with an unclear risk? Science 2002, 298, 510-511.
(4) Miller, A. Cationic liposomes in gene therapy. Angew. Chem., Int.
Ed. 1998, 37, 1768-1785.
(5) Ronsin, G.; Perrin, C.; Guedat, P.; Kremer, A.; Camilleri, P.; Kirby,
A. J. Novel spermine-based cationic gemini surfactants for gene
delivery. Chem. Commun. 2001, 2234-2235.
(6) Ewert, K.; Ahmad, A.; Evans, H. M.; Schmidt, H.-W.; Safinya, C.
R. Efficient synthesis and cell-transfection properties of a new
multivalent cationic lipid for nonviral gene delivery. J. Med. Chem.
2002, 45, 5023-5029.
(7) Hulst, R.; Muizebelt, I.; Oosting, P.; Van der Pol, C.; Wagenaar, A.;
Smisterova, J.; Bulten, E.; Driessen, C.; Hoekstra, D.; Engberts, J.
B. F. N. Sunfish amphiphiles: conceptually new carriers for DNA
delivery. Eur. J. Org. Chem. 2004, 835-849.
(8) Bhattacharya, S.; Bajaj, A. Recent advances in lipid molecular design.
Curr. Opin. Chem. Biol. 2005, 9, 647-655.
3.10. Flow Cytometry. The reporter gene expression was
examined by fluorescence microscopy at regular intervals and was
quantified 48 h post-transfection by flow cytometry. The percentage
of transfected cells was obtained by determining the statistics of
the cells fluorescing above the control level wherein nontransfected
cells were used as the control. Approximately 10 000 cells were
analyzed to achieve the statistical data, which have been presented
as the average of at least two independent measurements. For flow
cytometry analysis, ∼48 h post-transfection, the old medium was
(9) Ilies, M. A.; Johnson, B. H.; Makori, F.; Miller, A.; Seitz, W. A.;
Thompson, E. B.; Balaban, A. T. Pyridinium cationic lipids in gene
delivery: an in vitro and in vivo comparison of transfection efficiency
versus a tetraalkylammonium congener. Arch. Biochem. Biophys.
2005, 435, 217-226.
(10) Karmali, P. P.; Majeti, B. K.; Sreedhar, B.; Chaudhuri, A. In vitro
gene transfer efficacies and serum compatibility profiles of novel
mono-, di-, and tri-histidinylated cationic transfection lipids:
a
structure-activity investigation. Bioconjugate Chem. 2006, 17, 159-
171.

2442 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 10
Bajaj et al.
(11) Felgner, P. L.; Gadek, T. R.; Holm, M.; Roman, R.; Chan, H. W.;
Wenz, M.; Northrop, J. P.; Ringold, G. M.; Danielsen, M. Lipofec-
tion: a highly efficient, lipid-mediated DNA-transfection procedure.
Proc. Natl. Acad. Sci. U.S.A. 1987, 84, 7413-7417.
(12) Wang, J.; Guo, X.; Xu, Y.; Barron, L.; Szoka, F. C., Jr. Synthesis
and characterization of long chain alkyl acyl carnitine esters.
Potentially biodegradable cationic lipids for use in gene delivery. J.
Med. Chem. 1998, 41, 2207-2215.
(13) Behr, J. P.; Demeneix, B.; Loeffler, J. P.; Perez-Mutul, J. Efficient
gene transfer into mammalian primary endocrine cells with li-
popolyamine-coated DNA. Proc. Natl. Acad. Sci. U.S.A. 1989, 86,
6982-6986.
(14) Okayama, R.; Noji, M.; Nakanishi, M. Cationic cholesterol with a
hydroxyethylamino head group promotes significantly liposome-
mediated gene transfection. FEBS Lett. 1997, 408, 232-234.
(15) Gao, X.; Huang, L. A. novel cationic liposome reagent for efficient
transfection of mammalian cells. Biochem. Biophys. Res. Commun.
1991, 179, 280-285.
(16) Felgner, J. H.; Kumar, R.; Sridhar, C. J.; Wheeler, Y. J.; Tsai, R.;
Border, P.; Ramsey, M.; Martin, C. N.; Felgner, P. L. Enhanced gene
delivery and mechanism studies with a novel series of cationic lipid
formulations. J. Biol. Chem. 1994, 269, 2550-2561.
(17) Song, Y. K.; Liu, F.; Chu, S.; Liu, D. Characterization of cationic
liposome-mediated gene transfer in vivo by intravenous administra-
tion. Hum. Gene Ther. 1997, 8, 1585-1594.
(18) Ghosh, Y. K.; Visweswariah, S. S.; Bhattacharya, S. Nature of linkage
between the cationic headgroup and cholesteryl skeleton controls gene
transfection efficiency. FEBS Lett. 2000, 473, 341-344.
(19) Ghosh, Y. K.; Visweswariah, S. S.; Bhattacharya, S. Advantage of
the ether linkage between the positive charge and the cholesteryl
skeleton in cholesterol-based amphiphiles as vectors for gene delivery.
Bioconjugate Chem. 2002, 13, 378-384.
(20) Dileep, P. V.; Antony, A.; Bhattacharya, S. Incorporation of
oxyethylene units between hydrocarbon chain and pseudoglyceryl
backbone in cationic lipid potentiates gene transfection efficiency in
the presence of serum. FEBS Lett. 2001, 509, 327-331.
(21) Bhattacharya, S.; Dileep, P. V. Cationic oxyethylene lipids. Synthesis,
aggregation, and transfection properties. Bioconjugate Chem. 2004,
15, 508-519.
(25) Chien, P. Y.; Wang, J.; Carbonaro, D.; Lei, S.; Miller, B.; Sheikh,
S.; Ali, S. M.; Ahmad, M. U.; Ahmad, I. Novel cationic cardiolipin
analog-based liposome for efficient DNA and small interfering RNA
delivery in vitro and in vivo. Cancer Gene Ther. 2005, 12, 321-
328.
(26) Ilies, M. A.; Seitz, W. A.; Johnson, B. H.; Ezell, E. L.; Miller, A.
L.; Thompson, E. B.; Balaban, A. T. Lipophilic pyrylium salts in
the synthesis of efficient pyridinium-based cationic lipids, gemini
surfactants, and lipophilic oligomers for gene delivery. J. Med. Chem.
2006, 49, 3872-3887.
(27) Paul, B.; Bajaj, A.; Indi, S. S.; Bhattacharya, S. Synthesis of novel
dimeric cationic lipids based on an aromatic backbone between the
hydrocarbon chains and headgroup. Tetrahedron Lett. 47, 8401-
8405.
(28) Bhattacharya, S.; Ghosh, Y. K. Vesicle formation from oligo-
(oxyethylene)-bearing cholesteryl amphiphiles: site-selective effects
of oxyethylene units on the membrane order and thickness. Langmuir
2001, 17, 2067-2075.
(29) Chabaud, P.; Camplo, M.; Payet, D.; Serin, G.; Molreau, L.;
Barthelemy, P.; Grinstaff, M. W. Cationic nucleoside lipids for gene
delivery. Bioconjugate Chem. 2006, 17, 466-472.
(30) Matsui, K.; Sando, S.; Sera, T.; Aoyama, Y.; Sasaki, Y.; Komatsu,
T.; Terashima, T.; Kikuchi, J.-I. Cerasome as an infusible, cell-
friendly, and serum-compatible transfection agent in a viral size. J.
Am. Chem. Soc. 2006, 128, 3114-3115.
(31) Xu, Y.; Szoka, F. C., Jr. Mechanism of DNA release from cationic
liposome/DNA complexes used in cell transfection. Biochemistry
1996, 35, 5616-5623.
(32) Davis, S. C.; Szoka, F. C., Jr. Cholesterol phosphate derivatives:
synthesis and incorporation into a phosphatase and calcium-sensitive
triggered release. Bioconjugate Chem. 1998, 9, 783-792.
(33) Bhattacharya, S.; Krishnan-Ghosh, Y. Membrane formation from
oxyethylene bearing cationic cholesterol derivatives. Indian J. Chem.,
Sect. B 2001, 40B, 891-894.
(34) Mosmann, T. Rapid colorimetric assay for cellular growth and
survival: application to proliferation and cytotoxicity assays. J.
Immunol. Methods 1983, 65, 55-63.
(35) Hansen, M. B.; Nielsen, S. E.; Berg, K. Re-examination and further
development of a precise and rapid dye method for measuring cell
growth/cell kill. J. Immunol. Methods 1989, 119, 203-210.
(36) Choi, J. S.; Lee, E. J.; Jang, H. S.; Park, J. S. New cationic liposomes
for gene transfer into mammalian cells with high efficiency and low
toxicity. Bioconjugate Chem. 2001, 12, 108-113.
(22) Bhattacharya, S.; De, S.; George, S. K. Synthesis and vesicle
formation from novel pseudoglyceryl dimeric lipids. Evidence of
formation of widely different membrane organizations with excep-
tional thermotropic properties. Chem. Commun. 1997, 2287-2288.
(23) Bhattacharya, S.; De, S. Synthesis and vesicle formation from dimeric
pseudoglyceryl lipids with (CH2)m spacers: pronounced m-value
dependence of thermal properties, vesicle fusion, and cholesterol
complexation. Chem.sEur. J. 1999, 5, 2335-2347.
(37) Zor, T.; Selinger, Z. Linearization of the Bradford protein assay
increases its sensitivity: theoretical and experimental studies. Anal.
Biochem. 1996, 236, 302-308.
(24) Kasireddy, K.; Ali, S. M.; Ahmad, M. U.; Choudhury, S.; Chien, P.
Y.; Sheikh, S.; Ahmad, I. Synthesis of cationic cardiolipin analogs.
Bioorg. Chem. 2005, 33, 345-362.
JM0611253