Synthesis, characterization and transfection activity of new saturated and unsaturated cationic lipids

Article abstract of DOI:10.1016/j.farmac.2004.06.007

We synthesized new cationic lipids, analogue to N-[1-(2,3-dioleoyloxy) propyl]-N,N,N-trimethylammonium chloride (DOTMA) and 1,2-dimyristyloxypropyl-3- dimethyl-hydroxyethylammonium bromide (DMRIE), in order to compare those containing a dodecyl chain with those having a relatively long chain with two or five double bonds, such as squalenyl and dihydrofarnesyl derivatives, or complex saturated structures, such as squalane derivatives. The fusogenic helper lipid dioleoylphosphatidylethanolamine (DOPE) was added to cationic lipids to form a stable complex. Liposomes composed of 50:50 w/w cationic lipid/DOPE were prepared and incubated with plasmidic DNA at various charge ratios and the diameter and zeta potential of the complexes were measured. The surface charge of the DNA/lipid complexes can be controlled by adjusting the cationic lipid/DNA ratio. Finally, we tested the in vitro transfection efficiency of the cationic lipid/DNA complexes using different cell lines. The transfection efficiency was highest for the dodecyloxy derivative containing a single hydroxyethyl group in the head, followed by the dodecyloxy and the farnesyloxy trimethylammonium derivatives. Instead the C27 squalenyl and C27 squalanyl derivatives resulted inactive.

Full text of DOI:10.1016/j.farmac.2004.06.007

IL FARMACO 59 (2004) 869–878
http://france.elsevier.com/direct/FARMAC/
Original article
Synthesis, characterization and transfection activity of new saturated
and unsaturated cationic lipids
Silvia Arpicco a, Silvana Canevari b, Maurizio Ceruti a, Enrico Galmozzi b,
Flavio Rocco a, Luigi Cattel a,*
a Dipartimento di Scienza e Tecnologia del Farmaco, Università di Torino, Via Pietro Giuria 9, 10125, Torino, Italy
b Unit of Molecular Therapies, Department of Experimental Oncology, Istituto Nazionale Tumori, 20133 Milan, Italy
Received 9 April 2004; accepted 15 June 2004
Available online 30 September 2004
Abstract
We synthesized new cationic lipids, analogue to N-[1-(2,3-dioleoyloxy)propyl]-N,N,N-trimethylammonium chloride (DOTMA) and
1,2-dimyristyloxypropyl-3-dimethyl-hydroxyethylammonium bromide (DMRIE), in order to compare those containing a dodecyl chain with
those having a relatively long chain with two or five double bonds, such as squalenyl and dihydrofarnesyl derivatives, or complex saturated
structures, such as squalane derivatives. The fusogenic helper lipid dioleoylphosphatidylethanolamine (DOPE) was added to cationic lipids to
form a stable complex. Liposomes composed of 50:50 w/w cationic lipid/DOPE were prepared and incubated with plasmidic DNA at various
charge ratios and the diameter and zeta potential of the complexes were measured. The surface charge of the DNA/lipid complexes can be
controlled by adjusting the cationic lipid/DNA ratio. Finally, we tested the in vitro transfection efficiency of the cationic lipid/DNA complexes
using different cell lines. The transfection efficiency was highest for the dodecyloxy derivative containing a single hydroxyethyl group in the
head, followed by the dodecyloxy and the farnesyloxy trimethylammonium derivatives. Instead the C27 squalenyl and C27 squalanyl
derivatives resulted inactive.
© 2004 Elsevier SAS. All rights reserved.
Keywords: Cationic lipids; Transfection activity; Liposomes; Plasmidic DNA
1. Introduction
formulations have been described [10]. Cationic liposomes
normally contain a cationic amphiphile and a helper lipid
which is usually dioleoylphosphatidylethanolamine
(DOPE). DOPE may be required for the generation of stable
cationic liposomes, and also provides membrane fusion acti-
vity. The most common model for cationic lipids contain the
following domains: (1) a cationic or polycationic centre able
to bind to DNA; (2) a lipidic part (generally composed of two
hydrocarbon chains between C12 and C20 or of a cholesteryl
moiety); (3) a spacer linking these two parts [11].
The first cationic lipid was N-[1-(2,3-dioleoyloxy)
propyl]-N,N,N-trimethylammonium chloride (DOTMA) [7]
which consisted of a trimethylammonium moiety and two
oleoyl chains linked by a glycerol spacer. The cationic part is
generally a simple [7,12] or functionalized ammonium as
DMRIE [13,14], a polyamine [9,15] or guanidine [16,17].
The lipidic part consists of a flexible saturated fatty acid acyl
chain or fatty acid chains containing a cis double bond (i.e.
oleoyl) and/or a cholesterol ring [7,18]. Fatty acid chains give
the lipoplexes membrane fluidity and high lipid-mixing ca-
Over the last several years, various gene delivery systems
have been investigated for gene therapy approaches of both
inherited and acquired disease [1,2]. Different viral and non-
viral systems are currently under intensive investigations [3].
Retroviral, adenoviral, and adeno-associated viral vectors
produce highly efficient transfections regarding the number
of transfected cells [4]. However drawbacks caused by the
viral element, such as immune or toxic reactions, recombina-
tions with the wild-type virus and insertional mutagenesis,
have stimulated efforts to improve synthetic non-viral trans-
fer systems based on cationic lipids/liposomes or cationic
polymers [5,6].
Since the initial pioneering works of Felgner et al. in 1987
[7,8] and also of Beehr et al. in 1989 [9], many cationic lipid
* Corresponding author. Tel.: +39-011-6707693; fax: +39-011-6707695.
E-mail address: luigi.catttel@unito.it (L. Cattel).
0014-827X/$ - see front matter © 2004 Elsevier SAS. All rights reserved.
doi:10.1016/j.farmac.2004.06.007

870
S. Arpicco et al. / IL FARMACO 59 (2004) 869–878
pacity; in contrast, a cholesterol ring offers rigidity. More
interestingly it has been reported that cationic cholesterol-
based liposomes display a high membrane fusion capacity
while being stable in the presence of various substances
(ions, mucus) [12,19]. Liposomes containing the cationic
cholesterol derivative were the first synthetic vectors to be
used in clinical trials [20,21].
Cholesterol, as many steroid-like natural products is deri-
ved biogenetically through a complex mechanism that star-
ting from the C30 acyclic precursor (3S)-2,3-oxidosqualene,
leads to lanosterol [22]. In order to cyclize it to lanosterol, the
semi-rigid squalene moiety containing all trans double bonds
10.7 mmol) was added. The solution was cooled to 0 °C and
methanesulfonyl chloride (1.3 ml, 16 mmol) was added dro-
pwise under a dry nitrogen atmosphere.After stirring for 12 h
at 0 °C, heptane was added and the white solid formed was
filtered off. The solution was concentrated and then purified
by flash chromatography with elution in petroleum
ether/diethyl ether 95:5 v/v. The pure product 2 was obtained
as a white solid (2.62 g, 93%). 1H-NMR (CDCl3) d 1.1 (t, 3H,
CH3), 1.5 [m, 20H, (CH2)10], 3.25 (s, 3 H, CH3SO2), 4.4 (t,
2 H, CH2OSO2). MS–CI (isobutane) 265 (100). Satisfactory
EI mass spectra has not been obtained. Anal. (C13H28O3S) C,
H, O, S.
must assume
a
pre-boat–pre-chair–pre-boat–pre-chair
conformation, similar to that found in cyclized molecule.
Taking this mechanism into account we synthesized many
inhibitors of cholesterol biosynthesis, based on azasqualene
derivatives as acyclic molecules mimicking the natural subs-
trate [23].
In this study, we developed new series of cationic lipids
analogue to DOTMA and DMRIE by introducing one or
more E double bonds in the hydrophobic chain, in order to
obtain a semi-rigid hydrophobic anchor placed in between
the flexible saturated chain and the cyclic rigid cholesterol
moiety. We also wanted to evaluate the transfection efficien-
cies of the new unsaturated, squalene-like cationic lipids on a
panel of normal and tumor cell lines, by comparison with
analogues characterized by C12 saturated fatty acid chains
and commercial cationic lipids.
2.1.2. (6E)-3,7,11-trimethyl-6,10-dodecadienyl
methanesulfonate (7, Scheme 1)
Farnesol 3 (Scheme 2) (3 g, 13.5 mmol) was oxidized to
(6E)-3,7,11-trimethyl-2,6,10-dodecatrienal as a mixture of
2E and 2Z isomers (4, Scheme 2) by reaction with pyridi-
nium chlorochromate (5.8 g, 27 mmol) in 30 ml of anhydrous
dichloromethane. The reaction was stirred for 1 h at room
temperature under nitrogen atmosphere. The mixture was
then evaporated under reduced pressure and the precipitate
washed several times with diethyl ether; the ethereal phase
was evaporated and purified by flash chromatography using
petroleum ether as eluant. Yield 2.3 g, 78%.1H-NMR
(CDCl3) d 1.58–1.65 (m, 12 H, allylic CH3), 2.00–2.12 (m,
8 H, allylic CH2), 5.02–5.15 (t, 2 H, vinylic CH), 5.67 (m,
1 H, E and Z C=CHCHO), 9.94 (m, 1 H, E and Z
C=CHCHO).MS-EI m/z 220 (15), 205 (20), 177 (18), 136
(40), 69 (100). Anal. (C15H24O) C, H, O.
2. Experimental procedures
(6E)-3,7,11-trimethyl-2,6,10-dodecatrienal
4
(2 g,
The 1H nuclear magnetic resonance (NMR) spectra were
recorded on a Bruker AC 200 instrument (Karlsruhe, Ger-
many) in CDCl3 solution at room temperature, with SiMe4 as
internal standard. Mass spectra were obtained on a Finnigan-
MAT TSQ 700 spectrometer (San Jose, CA). Microanalyses
were performed on an elemental analyzer 1106 (Carlo Erba
Strumentazione, Milan, Italy). The reactions were monitored
by thin-layer chromatography (TLC) on F254 silica gel pre-
coated sheets (Merck, Milan, Italy); after development the
sheets were exposed to iodine vapour. Flash-column chroma-
tography was performed on 230–400 mesh silica gel
(Merck). Petroleum ether refers to the fraction boiling in the
40–60 °C range. All solvents were distilled prior to flash
chromatography.
9 mmol), and Pd(PPh3)4 (104 mg, 0.09 mmol) were added to
a mixture of acetic acid (0.52 ml, 9 mmol) and benzene
(30 ml mixed under argon atmosphere). Bu3SnH (2.9 ml,
10.8 mmol) was then added dropwise, and the mixture was
allowed to react for 1 h at room temperature. The resulting
brown solution was poured into iced water–dichloromethane
(3 × 20 ml), washed with saturated brine (2 × 20 ml), dried
with anhydrous magnesium sulfate and evaporated under
reduced pressure. The product was purified by flash chroma-
tography with petroleum ether/diethyl ether 99.5:0.5 v/v af-
fording 1.3 g of (6E)-3,7,11-trimethyl-6,10-dodecadienal (5,
1
Scheme 2) (65% yield). H-NMR (CDCl3) d 0.90 [d, 3 H,
CH(CH3)], 1.25–1.50 [m, 3H, CH2CH(CH3)CH2CHO],
1.55–1.60 (d, 9 H, allylic CH3), 1.98–2.30 (m, 8 H, allylic
CH2 and CH2CHO), 5.00–5.12 (m, 2 H, vinylic CH), 9.70 (t,
1 H, CHO). MS-EI m/z 222 (10), 189 (5), 179 (95), 69 (100).
Anal. (C15H26O) C, H, O.
Aldehyde 5 was converted to the corresponding alcohol
6 by reaction with NaBH4 (0.334 g, 8.9 mmol) in 30 ml of
methanol. After stirring for 30 min at room temperature
under a nitrogen atmosphere, the solvent was evaporated in
vacuo and the solid washed with dichloromethane and eva-
porated under reduced pressure. The crude product was puri-
fied by flash chromatography with petroleum ether/diethyl
Dioleoylphosphatidylethanolamine (DOPE) and Escort™
transfection reagent were purchased from Sigma Chemical
Co. (St. Louis, MO), Lipofectin™ was from GIBCO BRL
(Paisley, UK). All other reagents were of analytical grade.
2.1. Preparation of methanesulfonyl derivatives (2,7,9,11)
2.1.1. Dodecyl methanesulfonate (2, Scheme 1)
To dodecanol 1 (2 g, 10.7 mmol), dissolved in 20 ml of
anhydrous dichloromethane, triethylamine (1.5 ml,

S. Arpicco et al. / IL FARMACO 59 (2004) 869–878
871
Scheme 1. Preparation of cationic lipids (16–23).
ether 99:1 v/v, giving (6E)-3,7,11-trimethyl-6,10-
dodecadien-1-ol (6, Scheme 2) (1.09 g, 83%). 1H-NMR
(CDCl3) d 0.92 [d, 3 H, CH(CH3)], 1.23–1.48 (m, 5H,
CH2CH(CH3) CH2CH2OH), 1.50–1.60 (m, 9 H, allylic
CH3), 2.00–2.25 (m, 6 H, allylic CH2), 3.67 (m, 2 H,
CH2OH), 5.02–5.12 (m, 2 H, vinylic CH). MS-EI m/z 224
(20), 181 (100), 163 (60), 123 (90), 69 (95). Anal. (C15H28O)
C, H, O.
2.1.3. 1,1′,2-Trisnorsqualenyl methanesulfonate:
(4E,8E,12E,16E)-4,8,13,17,21-pentamethyl-4,8,12,16,20-
docosapentaenyl methanesulfonate (9, Scheme 1)
1,1′,2-trisnorsqualene alcohol 8 (1 g, 2.6 mmol) pre-
viously prepared in our laboratory [24] was reacted with
methanesulfonyl chloride (0.306 ml, 3.9 mmol) and triethy-
lamine (0.362 ml, 2.6 mmol) in 20 ml of anhydrous dichlo-
romethane and purified by flash chromatography with elution
in petroleum ether/diethyl ether 97:3 v/v to give compound 9.
Yield 0.6 g, 50%. 1H-NMR (CDCl3) d 1.58–1.70 (m, 20 H,
allylic CH3 and CH2CH2OSO2), 1.98–2.10 (m, 18 H, allylic
CH2), 2.98 (s, 3 H, CH3SO2), 4.19 (t, 2 H, CH2OSO2),
5.02–5.18 (m, 5 H, vinylic CH). MS-EI m/z 464 (5), 395 (10),
136 (45), 95 (100). Anal. (C28H48O3S) C, H, O, S.
(6E)-3,7,11-trimethyl-6,10-dodecadien-1-ol
6
(1 g,
4.5 mmol) was transformed to the corresponding methane-
sulfonate 7 (Scheme 1) using the method described for com-
pound 2. The crude product was purified by flash chromato-
graphy (petroleum ether/diethyl ether, 98:2 v/v). Yield
1
1.08 g, 80%. H-NMR (CDCl3) d 0.95 [d, 3 H, CH(CH3)],
1.25–1.48 (m, 4H, CH2CH(CH3)CH2CH2O), 1.55–1.65 [m,
10 H, allylic CH3 and CH(CH3)], 2.2 (m, 6 H, allylic CH2),
3.1 (s, 3 H, CH3SO2), 4.3 (m, 2 H, CH2OSO2), 5.1 (t, 2 H,
vinylic CH). MS-EI m/z 302 (5), 259 (30), 123 (65), 81 (75),
69 (100). Anal. (C16H30O3S) C, H, O, S.
2.1.4. 1,1′,2-Trisnorsqualanyl methanesulfonate:
4,8,13,17,21-pentamethyldocosyl methanesulfonate
(11, Scheme 1)
1,1′,2-trisnorsqualene alcohol 8 (1 g, 2.6 mmol) was ad-
ded to a mixture of Pd/C (10% Pd, 200 mg) in 100 ml of

872
S. Arpicco et al. / IL FARMACO 59 (2004) 869–878
purified by flash chromatography. Compounds 12, 13 and
14 were purified by elution in diethyl ether, while compound
15 was purified by elution in diethyl ether/methanol 97:3 v/v.
2.2.1. N,N-dimethyl-2,3-didodecyloxypropilamine
(12, Scheme 1)
0.346 g, 20% yield.1H-NMR (CDCl3) d 0.86 (t, 6 H, CH3),
1.27 (m, 36 H, CH2), 1.56 (m, 4 H, 2 CH2CH2O), 2.24 [s, 6 H,
N(CH3)2], 2.37 (d, 2H, CH2N), 3.45–3.60 (m, 7 H,
CH2OCHCH2OCH2). MS-EI, m/z 455 (1), 396 (1), 270 (40),
58 (100). Anal. (C29H61NO2) C, H, N,O.
2.2.2. N,N-Dimethyl-2,3-bis [(6E)-3,7,11-trimethyl-6,10-
dodecadienyloxy]propylamine (13, Scheme 1)
1
0.263 g, 13% yield. H-NMR (CDCl3) d 0.90 [d, 6 H,
2 CH(CH3)], 1.25–1.48 (m, 8 H, 2CH2CH(CH3)CH2CH2O),
1.50–1.75 [m, 20 H, allylic CH3 and 2 CH (CH3)], 1.98–2.10
(m, 12 H, allylic CH2), 2.28 [s, 6 H, N(CH3)2], 2.41 (d, 2 H,
CH2N), 3.46–3.62 (m, 7 H, CH2OCHCH2OCH2), 5.00–5.12
(m, 4 H vinylic CH). MS-EI, m/z 532 (20), 462 (8), 395 (8),
308 (75), 123 (92), 109 (100).Anal. (C35H65NO2) C, H, N,O.
2.2.3. N,N-Dimethyl-2,3-bis[(4E,8E,12E,16E)-
4,8,13,17,21-pentamethyl-4,8,12,16,20-
Scheme 2. Synthesis of (6E)-3,7,11-trimethyl-6,10-dodecandien1-ol (6)
starting from farnesol (3).
docosapentaenyloxy]propylamine (14, Scheme 1)
absolute ethanol, under H2 atmosphere. The reaction procee-
ded for 24 h under stirring under H2 atmosphere, the catalyst
was removed by filtration and the resulting solution was
concentrated under reduced pressure. 1,1′,2-trisnorsqualane
alcohol 10 was purified by flash chromatography (petroleum
ether/diethyl ether, 98:2 v/v). Yield 67%, 0.825 g. The cor-
responding methanesulfonyl derivative 11 was obtained by
procedure described above. The crude product was purified
by flash chromatography (petroleum ether/diethyl ether 98:2
v/v) to give compound 11 (0.690 g, 70%).1H-NMR (CDCl3)
d 0.80–0.90 (m, 18 H, CH3), 1.20–1.38 (m, 30 H, CH2),
1.68–1.80 (m, 5 H, CH), 3.01 (s, 3H, CH3–SO2), 4.20 (t, 2 H,
CH2–O–SO2). MS-EI m/z 474 (4), 351 (8), 125 (33), 97
(100). Anal. (C28H58O3S) C, H, O, S.
1
0.325 g, 10% yield. H-NMR (CDCl3) d 1.35 (m, 4 H,
2 CH2CH2O), 1.58–1.70 (m, 36 H, allylic CH3), 1.96–2.08
(m, 36 H, allylic CH2), 2.27 [s, 6 H, N(CH3)2], 2.54 (d, 2 H,
CH2N), 3.45–3.62 (m, 7 H, CH2OCHCH2OCH2), 5.00–5.15
(m, 10 H, vinylic CH). MS-EI, m/z 856 (16), 787 (14), 719
(25), 651 (28), 583 (85), 515 (22), 471 (38), 205 (100). Anal.
(C59H101NO2) C, H, N,O.
2.2.4. N,N-Dimethyl-2,3-bis(4,8,13,17,21-pentamethyldo-
cosyloxy)propylamine (15, Scheme 1)
0.6 g, 18% yield. 1H-NMR (CDCl3) d 0.82–0.90 (m, 36 H,
CH3), 1.25–1.60 (m, 60 H, CH2), 1.70–1.80 (m, 10 H, CH),
2.27 [s, 6 H, N(CH3)2], 2.40 (d, 2 H, CH2–N), 3.45–3.60 (m,
7 H, CH2OCHCH2OCH2). MS-EI, m/z 876 (20), 861 (5), 764
(10), 480 (100). Anal. (C59H121NO2) C, H, N,O.
2.2. Preparation of diethers (12–15)
For the preparation of compounds 12–15 the method des-
cribed for the synthesis of DOTMA was followed, with
minor modifications [7].
2.3. Preparation of quaternary ammonium salts (16–23)
The methanesulfonates (2, 7, 9 and 11) (3.8 mmol),
Compounds 16–19 were obtained by reaction of the cor-
responding diethers 12–15 (0.6 mmol) with methyl iodide
(6 mmol) and potassium carbonate (12 mmol) in 20 ml of
absolute ethanol. After 15 h under reflux, the reaction mix-
ture was cooled and ethanol evaporated; water was then
added and the mixture was extracted with dichloromethane
(3 × 50 ml). The organic layers were pooled and washed with
saturated brine, dried over anhydrous MgSO4 and concentra-
ted in vacuo.
dissolved in xylene, were added to
a mixture of
3-(dimethylamino)-1,2-propanediol (1.3 mmol) and potas-
sium tert-butoxide (5.7 mmol) in 15 ml of xylene, under a
nitrogen atmosphere. The reaction mixture was stirred under
nitrogen at 50 °C for 30 min and then heated for 12 h under
reflux. After the mixture had cooled to room temperature,
heptane was added and washed with water and saturated
brine; the organic layer was dried over anhydrous MgSO4
and concentrated in vacuo to give a crude product, which was

S. Arpicco et al. / IL FARMACO 59 (2004) 869–878
873
2.3.1. (2,3-Didodecyloxypropyl)trimethylammonium iodide
(16, Scheme 1)
2.3.6. [2,3-Bis (6E)-(3,7,11-trimethyl-6,10-
dodecadienyloxy)propyl](2-hydroxyethyl)
1
0.329 g, 92% yield. H-NMR (CDCl3) d 0.88 (t, 6 H,
CH3), 1.26 (m, 36 H, CH2), 1.57 (m, 4 H, 2 CH2CH2O),
3.46–3.68 (m, 16 H, N+(CH3)3 and CH2OCHCH2OCH2),
4.06 (d, 2 H, CH2N+). MS-EI, m/z 455 (55), 411 (25), 397
(45), 300 (100). MS-CI (isobutane) 471 (100). Anal.
(C30H64INO2) C, H, I, N,O.
dimethylammonium bromide (21, Scheme 1)
1
0.224 g, 57% yield. H-NMR (CDCl3) d 0.90 [d, 6 H,
CH(CH3)], 1.23–1.48 (m, 8 H, 2 CH2CH(CH3)CH2CH2O),
1.50–1.72 [d, 20 H, allylic CH3 and 2 CH(CH3)], 1.98–2.18
(m, 12 H, allylic CH2), 3.17 (s, 1 H, OH), 3.45–3.67 (m,
13 H, (CH3)2N+ and CH2OCHCH2OCH2), 3.93 (t, 2 H,
CH2OH), 4.04–4.12 (m, 4 H, CH2N+CH2), 5.00–5.12 (m,
4 H vinylic CH). MS-EI, m/z 532 (25), 531(14), 494 (55),
424 (62), 308 (60), 294 (20), 270 (100). Anal.
(C37H70BrNO3) C, H, Br, N, O.
2.3.2. {2,3-Bis[(6E)-3,7,11-trimethyl-6,10-
dodecadienyloxy]propyl}trimethylammonium iodide
(17, Scheme 1)
1
0.234 g, 58% yield. H-NMR (CDCl3) d 0.88 [d, 6 H,
2 CH(CH3)], 1.22–1.47 (m, 8 H, 2 CH2CH(CH3)CH2CH2O),
1.50–1.75 [d, 20 H, allylic CH3 and 2 CH(CH3)], 1.96–2.18
(m, 12 H, allylic CH2), 3.48–3.70 (m, 16 H, N+(CH3)3 and
CH2OCHCH2OCH2), 4.08 (d, 2 H, CH2–N+), 5.02–5.13 (m,
4 H vinylic CH). MS-EI, m/z 547 (0.5), 532 (40), 463 (70),
308 (100). Anal. (C36H68INO2) C, H, I, N,O.
2.3.7. {2,3-Bis[(4E,8E,12E,16E)-4,8,13,17,21-pentamethyl-
4,8,12,16,20-docosapentaenyloxy]propyl}(2-
hydroxyethyl)dimethylammonium bromide (22, Scheme 1)
1
0.571 g, 97% yield. H-NMR (CDCl3) d 1.35 (m, 4 H,
2 CH2CH2O), 1.50–1.68 (m, 36 H, allylic CH3), 1.95–2.10
(m, 36 H, allylic CH2), 2.45 (s, 1 H, OH), 3.35–3.60 [m,
13 H, (CH3)2N+ and CH2OCHCH2OCH2], 3.88 (t, 2 H,
CH2OH), 4.02–4.13 (m, 4 H, CH2N+CH2), 5.00–5.14 (m,
10 H vinylic CH). MS-CI, m/z 901 (10), 857 (100). Anal.
(C61H106BrNO3) C, H, Br, N,O.
2.3.3. {2,3-Bis[(4E,8E,12E,16E)-4,8,13,17,21-pentamethyl-
4,8,12,16,20-
docosapentaenyloxy]propyl}trimethylammonium iodide
(18, Scheme 1)
1
2.3.8. [2,3-Bis (4,8,13,17,21-pentamethyldocosyloxy)
propyl](2-hydroxyethyl)dimethylammonium bromide
(23, Scheme 1)
0.419 g, 70% yield. H-NMR (CDCl3) d 1.36 (m, 4 H,
2 CH2-CH2-O), 1.58–1.72 (m, 36 H, allylic CH3), 1.95–2.10
(m, 36 H, allylic CH2), 3.44–3.72 [m, 16 H, N+(CH3)3 and
CH2OCHCH2OCH2), 4.08 (d, 2 H, CH2N+), 5.02–5.18 (m,
10 H vinylic CH). MS-CI (isobutane) 851 (100), 789 (22),
721 (16). A satisfactory EI mass spectrum has not been
obtained. Anal. (C60H104INO2) C, H, I, N,O.
1
0.558 g, 93% yield. H-NMR (CDCl3) d 0.80–0.92 (d,
36 H, CH3), 1.22–1.60 (m, 60 H, CH2), 1.62–1.75 (m, 10 H,
CH), 2,45 (s, 1H, OH), 3.38–3.62 [m, 13 H, (CH3)2N+ and
CH2OCHCH2OCH2], 3.88 (t, 2 H, CH2–OH), 4.05–4.15 (m,
4 H, CH2N+CH2). MS-CI, m/z 877 (100), 863 (15). Anal.
(C61H126BrNO3) C, H, Br, N, O.
2.3.4. [2,3-Bis(4,8,13,17,21-pentamethyl-docosyloxy) pro-
pyl]trimethylammonium iodide
2.4. Preparation of liposomes
(19, Scheme 1)
1
0.357 g, 60% yield. H-NMR (CDCl3) d 0.80–0.90 (m,
Compounds (16–23) were prepared either alone or in
combination with the neutral colipid DOPE (1:1 w/w). The
cationic lipids and DOPE were dissolved in chloroform and
an appropriate volume of solution was transferred to glass
vials. The solvent was evaporated under reduced pressure to
give the dried lipid films. Residual traces of chloroform were
removed by placing vials under high vacuum overnight. The
lipid films were re-suspended in sterile pyrogen-free distilled
water at a concentration of 1 mg/ml by vortexing for 1 min at
room temperature or sonicating in a bath type sonicator
(Bransonic 12) to clarity (in the case of the less soluble
compounds 18, 19, 22 and 23).
We further prepared other two different types of liposo-
mes by adding a solution of Tween 80 in acetone. The first
kind of liposomes were obtained by adding Tween 80, in
order to obtain a mixture of cationic lipid/DOPE/Tween
80 1:1:1 w/w. The other liposomes were prepared by adding
Tween 80 to the cationic liposomes in a 1:1 ratio.
36 H, CH3), 1.24–1.65 (m, 60 H, CH2), 1.68–1.78 (m, 10 H,
CH), 3.48 [s, 9 H, N+(CH3)3], 3.59–3.78 (m, 7 H,
CH2OCHCH2OCH2), 4.05 (d, 2 H, CH2N+). MS-CI (isobu-
tane) 892 (1), 877 (80), 776 (100). A satisfactory EI mass
spectrum has not been obtained. Anal. (C60H124INO2) C, H,
I, N,O.
For the preparation of compounds 20–23, the diethers
12–15 (0.6 mmol) were reacted with 2-bromoethanol
(3 mmol) in 15 ml of acetonitrile, under reflux for 12 h. After
evaporation of the solvent, the crude product was treated as
described for compounds 16–19.
2.3.5. (2,3-Didodecyloxypropyl)
(2-hydroxyethyl)dimethylammonium bromide
(20, Scheme 1)
1
0.306 g, 88% yield. H-NMR (CDCl3) d 0.89 (t, 6 H,
CH3), 1.28 (m, 36 H, CH2), 1.56 (m, 4 H, 2 CH2CH2O), 2.87
(s, 1 H, OH), 3.48–3.69 [m, 13 H, N+(CH3)2 and
CH2OCHCH2OCH2), 3.87 (t, 2 H, CH2OH), 4.06–4.16 (m,
4 H, CH2N+CH2). MS-EI, m/z 455 (10), 300 (8), 270 (100).
Anal. (C31H66BrNO3) C, H, Br, N,O.
Liposome size was monitored by photon correlation spec-
troscopy using a Coulter Model N4SD submicron particle
analyzer (Coulter Electronics Inc., Hialeah, FL).

874
S. Arpicco et al. / IL FARMACO 59 (2004) 869–878
The zeta potential of the different preparations was also
plexes obtained by incubating DNA and cationic
lipid/DOPE/Tween 80 1:1:1 or DNA and cationic
lipid/Tween 80 1:1.
evaluated (Coulter DELSA model 440, Coulter Electronics
Inc.). Light scattering data were collected for 60 s at four
angles simultaneously using a frequency range of 500 Hz and
a current range equal to or lower than the conductivity of the
sample.
3. Results
3.1. Chemistry
2.5. Preparation of DNA/cationic lipid complexes
Different cationic lipids based on an hydrophobic and a
cationic portion, linked by an ethereal bond similar to the
commercially-available reagent DOTMA [7], were prepared.
As shown in Scheme 1, in order to evaluate the importance of
chain length and number of insaturations present in the hy-
drophobic portion for transfection capacity, we developed
various compounds with dodecyl, 2,3-dihydrofarnesyl, tris-
norsqualenyl and trisnorsqualanyl ethereal chains, bearing
methyl and 2-hydroxyethyl substituents linked to the nitro-
gen atom.
The preparation of the compounds is shown in Scheme 1;
initially the methanesulfonyl derivatives of dodecanol 2,
1,1′,2-trisnorsqualene alcohol 9, 1,1′,2-trisnorsqualane alco-
hol 11 and of a derivative of farnesol, obtained by selective
reduction of the double bond vicinal to the alcohol group 7
[25], were prepared. For this purpose, the different alcohols
1, 6, 8 and 10 were reacted with methanesulfenyl chloride in
the presence of triethylamine.
For the preparation of the farnesol derivative it was not
possible to use farnesol directly, because the double bond in
position 2 with respect to the alcohol function renders it too
reactive with methanesulphenyl chloride and various degra-
dation products are formed; we consequently reduced this
double bond (Scheme 2).
Reaction between 3-(dimethylamino)-1,2-propanediol
and the different methanesulfonates 2, 7, 9 and 11 in the
presence potassium tert-butoxide then afforded the diethers
12–15. Transformation to the corresponding quaternary am-
monium salts was achieved by reacting the diethers with
methyl iodide [23] in the case of derivatives 16–19 or with
2-bromoethanol [13] in the case of derivatives 20–23. The
compounds were characterized by 1H-NMR and MS spectra.
Plasmid DNA and the different cationic lipid/DOPE (1:1
w/w) preparations were mixed together by vortexing at room
temperature and the complex formation was allowed to pro-
ceed for 30 min at room temperature. Different amounts of
cationic liposomes were used in order to obtain complexes at
ratio of 1:2, 1:1, 2:1, 3:1, 5:1 and 8:1 lipid/DNA (ratio
w/w).The size and the zeta potential values of the complexes
thus prepared were determined as detailed above.
2.6. Cell cultures
Cell lines that are derived from different origin were used
for this study. HEK 293 (human foetal kidney), CHO (chi-
nese hamster ovary), A431 (human epidermoid carcinoma)
and KB (human nasopharyngeal carcinoma) cell lines were
purchased from ATCC (Rockville, MD) and IGROV1 (hu-
man ovarian carcinoma) cell line was purchased from J.
Benard (Institute G. Roussy, Villejuif, France). All cell lines
were maintained in RPMI 1640 (Gibco BRL, Paisley, UK)
supplemented with 5% FCS, 2 mM L-glutamine and gentalyn
(100 U/ml) at 37 °C in a humidified atmosphere of 5% CO2 in
air.
2.7. Transfection
Cells (1 × 105) were seeded in 12-well plates one day
before transfection at a density resulting in 40–50%
confluence on the day of transfection. Before transfection,
cells were washed once with PBS, and 1 ml of fresh medium
without FCS was added. Supercoiled pCAT basic vector
(Promega, Madison, WI) (2.5 µg), used as carrier for DNA,
was co-transfected with CMV-Luc (0.5 µg), containing the
luciferase gene downstream from a CMV promoter (a gift
from Dr. Maria Zajac-Kaye, Bethesda, MD) using 5 µl of
each cationic liposome (Lipofectin™, Escort™ or the new
synthesized products). After 10 min incubation at room tem-
perature, the mixture was added dropwise to the cells fol-
lowed by overnight incubation at 37 °C in a 5% CO2 atmos-
phere. Medium was replaced with fresh medium containing
10% FCS and incubation continued for an additional 48 h.
Cell lysates were prepared using 100 µl of Reporter Lysis
Buffer (Promega) per well. To determine transfection effi-
ciency, 100 µl of Luciferase Assay Buffer (Promega) was
added to 10 µg of cell lysate and luciferase activity was
evaluated using a TD Luminometer (Promega). The final
values of luciferase are reported in terms of RLU/mg total
protein. We also evaluated the transfection efficiency of com-
3.2. Preparation and characterization of liposomes
The cationic lipids were dissolved in chloroform and
mixed in 1:1 ratio (w/w) with the colipid DOPE and the dried
film was resuspended by vortexing or by sonicating it to
clarity. In the preparations of lipid vesicles with the com-
pounds containing the derivative of squalene and squalane in
the hydrophobic chains (compounds 18, 19, 22 and 23) it
was difficult to resuspend these derivatives in water, due to
their low solubility. In the case of these same derivatives, to
prepare liposomes it was necessary to sonicate in a bath-type
sonicator.
The different preparations showed mean diameters of
320–400 nm for the vortexed liposomes and of 250–300 nm
for the sonicated ones.

S. Arpicco et al. / IL FARMACO 59 (2004) 869–878
875
Cationic liposomes displayed a net positive zeta potential
3.4. In vitro transfection effıciency
value ranging from +40 to +55 mV both for the cationic
liposomes alone and for their mixture with DOPE. These
values were similar to those observed for commercially
available cationic lipids such as Lipofectin™ and Escort™.
In order to improve the solubility of compounds 18, 19,
22 and 23, we prepared other liposomes composed either of
cationic lipid/DOPE/Tween 80 in a ratio of 1:1:1 or of catio-
nic lipid/Tween 80 1:1 w/w. We observed an increased solu-
bility only for the preparation composed of cationic liposo-
mes and Tween 80. There was no difference in terms of size
and charge.
The transfection efficiency of the different cationic lipids
in mixture with DOPE was evaluated on two normal and
three malignant cell lines with different transfection capaci-
ties: HEK 293 (human foetal kidney), CHO (chinese hamster
ovary), A431 (human epidermoid carcinoma), KB (human
nasopharyngeal carcinoma) and IGROV1 (human ovarian
carcinoma) cell lines. We compared the capabilities of our
cationic lipids to those of the commercially-available Lipo-
fectin™ and Escort™.
As shown in Fig. 3, two groups of cell lines could be
distinguished within the five cell lines tested. HEK and CHO
cells belong at the first group; they showed high transfection
efficiency with dodecyl and farnesyl derivatives 16, 20 and
17 but also with the commercial cationic lipids, although the
latter had a transfection capability about 10-fold lower. The
carcinoma cell lines belong to the second group; they showed
a transfection efficiency with commercial cationic lipid in an
order of magnitude between 1 and 10, while the transfection
capability reached 10 and 102 of magnitude with the dodecyl
and farnesyl derivatives 16, 20 and 17. This increment was
particularly evident in IGROV1 cells, which showed a trans-
fection efficiency of a magnitude around 103. Note that with
compound 20, the dodecyloxypropyl hydroxyethyldimethyl
ammonium derivative, all five cell lines were found to be
transfected with an efficiency from four to one thousand
times higher than that of Lipofectin™.
3.3. Complexes with DNA
Complexes between plasmidic DNA and various amounts
of the different preparations of cationic lipids, in 1:1 mixture
with DOPE, were prepared and analyzed. The mean diameter
of the complexes was determined using photon correlation
spectroscopy.
At low lipid/DNA ratios, the complexes were small with a
mean diameter similar to that of the corresponding vesicles
alone. The size of the vesicles increased on increasing the
amount of lipid used (lipid/DNA ratio 2:1 and 5:1), then the
complexes were smaller for higher excesses of lipid (8:1).
Fig. 1 shows the size of the complexes with different exces-
ses of lipids 16, 17, 20 and 21.
The squalene and squalene derivatives did not provide
satisfactory results in that, owing to their poor solubility in an
aqueous environment, it was not possible to incubate them
with known quantities of DNA. We carried out other trans-
fection experiments preparing liposomes composed of catio-
nic lipid/DOPE/Tween 80 1:1:1 and cationic lipid/Tween
80 1:1, but none of these compounds showed a better trans-
fection efficiency in all cell lines tested (data not shown).
We further evaluated the zeta potential values of these
complexes: results are in Fig. 2. We observed that the com-
plexes showed negative zeta potentials until 5:1 lipid/DNA
ratio, while we obtained a net positive value for excesses of
lipids over DNA of 8:1.
The results were satisfactory for all the cationic lipids,
except for the squalene and squalane derivatives 18, 19,
22 and 23. For these, it was not possible to establish the exact
concentration in the mixtures, given their poor solubility in
water. On account of this problem, we did not succeed in
verifying for what lipid excess the mixture incubated with
DNA would have had a positive zeta potential.
4. Discussion
Cationic lipid molecules contain four different functional
domains: a positively charged head group, a spacer of va-
Fig. 1. Size of complexes between DNA and lipid. Cationic lipid vesicles
were prepared as 1:1 (w/w) of cationic lipid:DOPE and mixed with DNA at
various ratios. Error bars represent standard deviations.
Fig. 2
. Zeta potentials of cationic lipids alone and cationic lipid:DNA
complexes. Error bars represent standard deviations.

876
S. Arpicco et al. / IL FARMACO 59 (2004) 869–878
Fig. 3
.
Evaluation of the transfection efficiency. Cells were transiently transfected as reported in Materials and methods. For KB cells the transfection efficiency
with Escort™ was not evaluated. Data (mean S.D.) are representative of three independent experiments for each cell line.
rying length, a linker bond and a hydrophobic anchor. Taking
the cationic lipid series as a model [7,13], the positively
charged head group consisted of a quaternary ammonium
with substitution of methyl or hydroxyethyl groups respon-
sible for interactions between liposome-DNA complex and
cell membrane. A spacer arm containing only one atom
spacer, and the hydrophobic portion made up of an alkoxy
chain, are important for stability, biodegradability and trans-
fection activity.
The fatty acid chains, which are usually monounsaturated
(oleoyl, C18) or shorter saturated (C16 or C14), are generally
important since they are responsible for membrane fluidity
and good lipid mixing within the bilayer. Felgner et al. [13]
have shown that, as the length of the saturated aliphatic chain
increases from 14 to 16 or 18 carbon atoms, the transfection
activity of the resulting cationic lipids progressively decli-
nes. By contrast, it has also been shown that the highest
activity is either for palmitoyl or for oleoyl chains. In cationic
lipids, increasing the number of double bonds in the fatty
acid chain corresponds to a decrease of both the phase tran-
sition temperature and the bilayer stiffness of the resulting
vesicles [27,28]. Bilayer rigidity may interferes with mem-
brane fusion, which is the postulated step in the DNA deli-
very mechanism [11]. Lipid fusion is very important for the
escape of DNA from the endosome to the nucleus. In the
preliminary formulation study we determined the phase tran-
sition temperature of the different cationic ammonium lipids
by DSC. The C12 saturated derivatives showed an endother-
mic transition characterized by a sharp peak, respectively, at
53.37 and 79.77 °C. By introducing one or more double
bonds in the fatty chain we expected to obtain a decrease of
phase-transition temperature. Indeed, the bis unsaturated
C15 dihydro farnesyl derivative showed a peak at lower
temperature (29.75 °C). The C27 squalenyl ammonium deri-
vative, characterized by the presence of four E double bonds
showed a broad peak, from 30 to 42 °C, at DSC. The proxi-
mity of the phase transition temperature to 37 °C, tempera-
ture of the cells during transfection, may also influence the
efficiency of the fusion step.
Cationic lipids form a stable complex with the fusogenic
helper lipid DOPE, which has been reported to have a high
tendency to form an inverted hexagonal phase (HII) at acidic
pH [29]. Formation of the HII by DOPE probably destabilizes
the endosomes, resulting in an efficient escape from lysoso-
mal degradation and cytoplasmatic release of DNA [30]. The
preparation of MLV cationic liposome was strongly affected

S. Arpicco et al. / IL FARMACO 59 (2004) 869–878
877
by the solubility of each lipid; thus whereas the preparation
of liposomes containing a short lipophilic tail, such as 16 and
20, and the farnesyl derivatives (17, 21) was easy, those
incorporating the highly lipophilic derivatives (18, 19, 22,
and 23) were little soluble in water. For these compounds,
long sonication times were necessary. By adding DOPE to
liposomes we obtained particles whose diameter was inver-
sely related to the amount of DOPE. DOPE probably inte-
racts with cationic lipids, forming a stable complex through
ion pair formation between the negatively charged phosphate
on DOPE and the quaternary ammonium group on the catio-
nic lipid, as well as favourable intermolecular hydrogen bond
interactions [31,32].
The MLV liposomes composed of a 50:50 w/w cationic
lipid/DOPE were mixed with plasmid DNA at various charge
ratios and the diameter and zeta potential of the resulting
complexes (lipoplexes) were measured. As expected, the
absolute value of the mean diameter of the lipoplexes was a
function of the excess of lipid/DNA ratio, varying from
300 nm (lipid/DNA ratio 1:1) to a maximum of 750–900 nm
at liposome/DNA ratio of 5:1, and returning to a smaller size
(500–600 nm) at a lipid/DNA ratio 8:1.
The surface charge of a DNA/lipid complex is expected to
influence its interaction with various biological components
as well as its distribution, access and entry to the target cells.
As observed using light scattering analysis, the lipoplex zeta
potential changed from negative to positive when the mixing
charge ratio changes from excess negative to excess positive.
For instance the zeta potential changed from a negative value
(–25 mV) for a 1:2 lipid/DNA ratio to a positive ratio (+55
mV) for a 8:1 lipid/DNA ratio. The surface charge of the
DNA/lipid complexes can therefore be controlled by adjus-
ting the ratio of cationic lipid to DNA ratio.
To improve the water solubility of squalene and squalane
cationic lipids, we decided to use the neutral surfactant
Tween 80 with or without DOPE as helper co-lipid. It has
been found that the inclusion of small amounts of Tween
80 into the lipid formulation (such as DC-cholesterol) provi-
des better stability for DNA/lipid complexes, as well as high
transfection activity [33].
rase gene placed downstream from a cytomegalovirus pro-
moter. The different transfection capacities were evaluated
on five types of cell lines: HEK, CHO, A431, IGROV1 and
KB.
Initially we tested the transfection efficiencies of all the
novel transfection lipids using a 1:1 w/w mixture with
DOPE. Interestingly, the amphipile having both a C12 satu-
rated chain and a single hydroxyethyl group in the head
group showed the highest activity compared with the com-
mercial Lipofectin™, followed by the C12 ammonium ana-
logue of DOTMA. If the hydrophobic moiety was increased
to a C15 chain containing two double bonds, as in the farne-
syl derivative (17), the activity, albeit less, was still satisfac-
tory. C27 fully saturated squalane derivatives (19 and 23), or
those containing three double bonds (18 and 22), retained no
activity. The addition of Tween 80, as in formulations 2 and
3, slightly increased the acqueous solubility but did not
increase transfection activity.
Thus, for our series of lipids, Tween 80 was not able to
replace DOPE as a helper lipid. Indeed, when the most active
cationic lipid was complexed with Tween 80, the activity
dramatically decreased. Taking Lipofectin (RLU=1) as refe-
rence, 20 had RLU=1016 when formulated as a 1:1 DOPE
complex, RLU=7 when 1 mg of Tween 80 was added (for-
mulation 3) and RLU was negative when DOPE is replaced
by the surfactant.
These finding suggest the fundamental role of DOPE in
the activity of all synthesized lipid compounds, as has
already been reported [32]. A proper DOPE association with
the cationic lipid may increase DNA cell uptake, facilitate the
escape of DNA from endosomes and increase the ability of
DNA to dissociate from the lipid complex. We also showed
that the activity may vary depending on the target cells,
which may include differences in cell surface composition
that effect complex binding, and differences in the complex
uptake mechanism. The results also showed that, for best
transfection activity, a shorter and saturated alkyl chain has a
much more beneficial effect than a longer unsaturated fatty
chain.
The C12 cationic lipid derivatives 16 and 20 have been
reported in a patent, with no description of synthesis or
biological activity [34]. Compound 20 was also cited, within
a list of different lipoplexes, in a review not reporting any
detailed investigations [35].
The C15 farnesil derivative, having an unusual branched
structure, albeit less active than the C12 saturated lipids, is
very interesting, being more active than commercial lipids,
such as DOTMA or ESCORT™ in almost all cell lines tested.
Starting from this lead, further work will be necessary to
better elucidate the structural–relationship, taking into consi-
deration new series of lipid derivatives.
Similarly to DOPE, the addition of Tween 80 to cationic
lipids progressively reduced the diameter of the particles. In
addition, due to the absence of electric charge, the addition of
Tween 80 to the lipid did not change the absolute value of the
original positive charge. Thus, if we only considered the
problems related to physical stability of the colloidal formu-
lation, the surfactant Tween 80 behaved like the helper lipid
DOPE.
Lastly, we tested the in vitro transfection efficiency of the
cationic lipid/DNA complexes using different formulations:
(1) cationic lipid/DOPE 1:1 w/w; (2) cationic lipid/Tween
80 1:1 w/w; (3) cationic lipid/DOPE: Tween 80 1:1:1.
For comparison we also tested the commercial lipids Es-
cort™ (DOTAP/DOPE; 1:1) and Lipofectin™ (DOTMA/
DOPE; 1:1). The differently formulated lipid/DNA comple-
xes were co-transfected with a plasmid containing a lucife-
Acknowledgements
This work was supported by MIUR 40–60%.

878
S. Arpicco et al. / IL FARMACO 59 (2004) 869–878
[18] H. Farhood, R. Bottega, R.M. Epand, L. Huang, Effect of cationic
References
cholesterol derivatives on gene transfer and proteine kinase C activity,
Biochimica et Biophysica Acta 1111 (1992) 239–246.
[1] A.P. Rolland, From genes to gene medicines: recent advances in
nonviral gene delivery, Critical Reviews in Therapeutic Drug Carrier
Systems 15 (1998) 143–198.
[2] H.M. Blau, M.L. Springer, Gene therapy—a novel form of drug
delivery, The New England Journal of Medicine 2 (1995) 1204–1207.
[3] R.J. Mrsny, Special feature: a survey of the recent patent literature on
the delivery of genes and oligonucleotides, Journal of Drug Targeting
7 (1999) 1–10.
[4] E.G. Hanania, J. Kavanagh, G. Hortobagyi, R.E. Giles, R. Champlin,
A.B. Deisseroth, Recent advances in the application of gene therapy to
human disease, The American Journal of Medicine 99 (1995) 537–
552.
[5] A.D. Miller, Cationic liposomes for gene therapy, Angewandte Che-
mie International Edition 37 (1998) 1768–1785.
[19] R. Leventis, R.J. Silvius, Interaction of mammalian cells with lipid
dispersions containing novel metabolizable cationic amphiphiles,
Biochimica et Biophysica Acta 1023 (1990) 124–132.
[20] N.J. Caplen, E.W.F.W. Alton, P.G. Middleton, J.R. Dorin, B.J. Steven-
son, X. Gao, S.R. Durham, P.K. Jeffrey, M.E. Hodson, C. Coutelle,
L. Huang, D.J. Porteous, R. Williamson, D.M. Geddes, Liposomes-
mediated CFTR gene transfer to the nasal epithelium of patients with
cystic fibrosis, Nature Medicine 1 (1995) 39–46.
[21] G.J. Nabel, E.G. Nabel, Z.Y. Yang, B.A. Fox, G.E. Plautz, X. Gao,
L. Huang, S. Shu, D. Gordon, A.E. Chang, Direct gene transfer with
DNA liposomes complexes in melanoma: expression, biological
activity, lack of toxicity in humans, Proceedings of the National
Academy of Sciences of the United States of America 90 (1993)
11307–11311.
[22] L. Cattel, M. Ceruti, Inhibitors of 2,3-oxidosqualene cyclase as tools
for studying the mechanism and function of enzyme, in: E.J. Parish,
W.D. Nes (Eds.), Biochemistry and function of sterols, American Oil
Chemists’ Society, Champaign, IL, 1997, pp. 1–21.
[23] M. Ceruti, G. Balliano, F. Viola, L. Cattel, N. Gerst, F. Schuber,
Synthesis and biological activity of azasqualenes, bis-azasqualenes
and derivatives, European Journal of Medicinal Chemistry 22 (1987)
199–208.
[24] S.E. Sen, G.D. Prestwich, Trisnorsqualene alcohol, a potent inhibitor
of vertebrate squalene epoxidase, Journal of American Chemical
Society 111 (1989) 1508–1510.
[6] S. Chesnoy, L. Huang, DNA condensed by polications and lipids for
gene transfer, S.T.P. Pharma Science 9 (1999) 5–12.
[7] P.L. Felgner, T.R. Gadek, M. Holm, R. Roman, H.W. Chan, M. Wenz,
J.P. Northrop, G.M. Ringold, M. Danielsen, Lipofection: a highly
efficient, lipid-mediated DNA-tranfection procedure, Proceedings of
the National Academy of Sciences of the United States of America 84
(1987) 7413–7417.
[8] P.L. Felgner, G.M. Ringold, Cationic liposomes mediated transfec-
tion, Nature 337 (1989) 387–388.
[9] J.P. Behr, B. Demeneix, J.P. Loeffler, J. Perez-Mutul, Efficient gene
transfer into mammalian primary endocrine cells with lipopolyamine-
coated DNA, Proceedings of the NationalAcademy of Sciences of the
United States of America 86 (1989) 6982–6986.
[25] M. Ceruti, F. Rocco, F. Viola, G. Balliano, P. Milla, S. Arpicco,
L. Cattel, 29-methylidene-2,3-oxidosqualene derivatives as stereospe-
cific mechanism-based inhibitors of liver and yeast oxidosqualene
cyclase, Journal of Medicinal Chemistry 41 (1998) 540–554.
[26] J.S. Remy, C. Sirlin, P. Vierlining, J.P. Behr, Gene transfer with a
series of lipophilic DNA-binding molecules, Bioconjugate Chemistry
5 (1994) 647–654.
[27] G. Cevc, How membrane chain-melting phase-transition temperature
is affected by the lipid chain asymmetry and degree of unsaturation:
an effective chain-length model, Biochemistry 30 (1991) 7186–7193.
[28] K.M.W. Keough, Influence of chain unsaturation and chain position
on thermotropism and intermolecular interactions in membranes, Bio-
chemical Society Transactions 18 (1990) 835–840.
[29] J.M. Seddon, Structure of inverted hexagonal (HII) phase, and non-
lamellar phase transitions of lipids, Biochimica et Biophysica Acta
1031 (1990) 1–69.
[30] D.C. Litzinger, L. Huang, Phosphatidyl ethanolamine liposomes:
drug delivery, gene transfer and immunodiagnostic applications, Bio-
chimica et Biophysica Acta 1113 (1992) 201–227.
[10] R.J. Lee, L. Huang, Lipidic vector system for gene transfer, Critical
Reviews in Therapeutic Drug Carrier Systems 14 (1997) 173–206.
[11] X. Gao, L. Huang, Cationic liposomes mediated gene transfer, Gene
Therapy 2 (1995) 710–722.
[12] X. Gao, L. Huang, A novel cationic liposomes reagent for efficient
transfection of mammalian cells, Biochemical and Biophysical
Research Communications 179 (1991) 280–285.
[13] J.H. Felgner, R. Kumar, C.N. Sridhar, C.J. Wheeler, Y.J. Tsai, R. Bor-
der, P. Ramsey, M. Martin, P.L. Felgner, Enhanced gene delivery and
mechanism studies with a novel series of cationic lipid formulations,
The Journal of Biological Chemistry 269 (1994) 2550–2561.
[14] C.J. Wheeler, P.L. Felgner,Y.J. Tsai, J. Marshall, L. Sukhu, S.G. Doh,
J. Hartikka, J. Nietupski, M. Manthorpe, M. Nichols, M. Plewe,
X. Liang, J. Norman, A. Smith, S.H. Cheng, A novel cationic lipid
greatly enhances plasmid DNA delivery and expression in mouse
lung, Proceedings of the National Academy of Sciences of the United
States of America 93 (1996) 11454–11459.
[15] E.R. Lee, J. Marshall, C.S. Siegel, C. Jiang, N.S. Yew, M.R. Nichols,
J.B. Nietupski, R.J. Ziegler, M.B. Lane, K.X. Wang, N.C. Wan,
R.K. Scheule, D.J. Harris, A.E. Smith, S.H. Cheng, Detailed analysis
of structures and formulations of cationic lipids for efficient gene
transfer to the lung, Human Gene Therapy 7 (1996) 1701–1717.
[16] J.P. Vigneron, N. Oudrhiri, M. Fauquet, L. Vergely, J.C. Bradley,
M. Basseville, P. Lehn, J.M. Lehn, Guanidinium-cholesterol cationic
lipids: efficient vectors for the transfection of eukaryotic cells, Pro-
ceedings of the National Academy of Sciences of the United States of
America 93 (1996) 9682–9686.
[31] E. Tomlinson, A.P. Rolland, Controllable gene therapy pharmaceutics
of non-viral gene delivery systems, Journal of Controlled Release 39
(1996) 357–372.
[32] A. Fasbender, J. Marshall, T.O. Moninger, T. Grunst, S. Cheng,
M.J. Welsh, Effect of colipids in enhancing cationic lipid-mediated
gene transfer in vitro and in vivo, Gene Therapy 4 (1997) 716–725.
[33] F. Liu, J. Yang, L. Huang, D. Liu, New cationic lipid formulations for
gene tranfer, Pharmaceutical Research 13 (1996) 1856–1860.
[34] C.J. Wheeler, Complex cationic lipids, WO 97/19675.
[35] P.L. Felgner,Y.J. Tsai, L. Sukhu, C.J. Wheeler, M. Manthorpe, J. Mar-
shall, S.H. Cheng, Improved cationic lipid formulations for in vivo
gene therapy, Annals New York Academy of Sciences 775 (1995)
126–139.
[17] B. Pitard, N. Oudrhiri, J.P. Vigneron, M. Hauchecorne, O. Aguerre,
R. Toury, M. Airiau, R. Ramasawmy, D. Scherman, J. Crouzet,
J.M. Lehn, P. Lehn, Structural characteristics of supramolecular
assemblies formed by guanidinium-cholesterol reagents for gene
transfection, Proceedings of the National Academy of Sciences of the
United States of America 96 (1999) 2621–2626.