Synthesis and DNA transfection properties of new head group modified malonic acid diamides

Article abstract of DOI:10.1016/j.ijpharm.2011.02.022

Malonic acid diamides with two long hydrophobic alkyl chains and a basic polar head group as a new class of non-viral gene transferring compounds have shown high transfection efficiency and moderate toxicity. Based on the results obtained with saturated and unsaturated alkyl residues new derivatives with a more complex head group structure have been synthesized. For this purpose, cationic respectively basic groups were introduced by one or two lysine residues bound via tris(aminoethyl)amine spacer to the malonic acid diamide backbone. By studying in vitro gene delivery an increase of transfection efficacy was observed when using lipids with at least one unsaturated alkyl chain. This leads to cationic lipids exhibiting comparable or even higher transfection efficacies compared to the commercially available transfection agents LipofectAmine and SuperFect. Phase transitions and phase structures of selected compounds have been analyzed and discussed in terms of transfection abilities. Particle size and zeta potential of liposomes and lipoplexes were also determined.

Full text of DOI:10.1016/j.ijpharm.2011.02.022

International Journal of Pharmaceutics 409 (2011) 46–56
Contents lists available at ScienceDirect
International Journal of Pharmaceutics
journal homepage: www.elsevier.com/locate/ijpharm
Synthesis and DNA transfection properties of new head group modified malonic
acid diamides
a
a,1
a
b
b
Christian Wölk , Martin Heinze , Patrick Kreideweiß , Matthias Dittrich , Gerald Brezesinski ,
a
a,∗
Andreas Langner , Bodo Dobner
Institute of Pharmacy, Department of Biochemical Pharmacy, Martin-Luther-University Halle-Wittenberg, Wolfgang-Langenbeck-Strasse 4, 06120 Halle (Saale), Germany
Max Planck Institute of Colloids and Interfaces, Science Park Potsdam-Golm, Am Muehlenberg 1, 14476 Potsdam, Germany
a
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Malonic acid diamides with two long hydrophobic alkyl chains and a basic polar head group as a new
class of non-viral gene transferring compounds have shown high transfection efficiency and moderate
toxicity. Based on the results obtained with saturated and unsaturated alkyl residues new derivatives
with a more complex head group structure have been synthesized. For this purpose, cationic respectively
basic groups were introduced by one or two lysine residues bound via tris(aminoethyl)amine spacer to
the malonic acid diamide backbone. By studying in vitro gene delivery an increase of transfection efficacy
was observed when using lipids with at least one unsaturated alkyl chain. This leads to cationic lipids
exhibiting comparable or even higher transfection efficacies compared to the commercially available
Received 4 November 2010
Received in revised form 14 February 2011
Accepted 17 February 2011
Available online 25 February 2011
Keywords:
Cationic lipids
Malonic acid diamides
Synthetic methods
Liposomes
TM
TM
transfection agents LipofectAmine and SuperFect . Phase transitions and phase structures of selected
compounds have been analyzed and discussed in terms of transfection abilities. Particle size and zeta
potential of liposomes and lipoplexes were also determined.
Lipoplexes
Non-viral gene transfer
© 2011 Elsevier B.V. All rights reserved.
1
. Introduction
are still associated with problems like immunogenic potential,
restrictions with regard to the size of delivered genetic material,
Gene therapy is a promising method not only for the treat-
ment of inherited diseases such as cystic fibrosis (Hyde et al.,
000; Noone et al., 2000) but also for acquired diseases like can-
cer, AIDS, and diabetes (Nazari and Jashi, 2009; Young et al., 2007;
Zhang et al., 2009). This therapy is also a great challenge to all
researchers dealing with the problem of back-door recruitment of
genetic material into cells and organisms. In general, gene deliv-
ery systems are mainly based on either viral or non-viral vectors
difficult production, and oncogenic potential (Kay et al., 2001;
Seow and Wood, 2009). Therefore, parallel to further experiments
with certain viruses the molecular design and synthesis of new
non-viral vectors was pushed and has now reached a remark-
able advance both in quantity and quality (Mintzer and Simanek,
2009). Especially the synthesis of cationic lipids has generated
new and interesting substances (Bhattacharya and Bajaj, 2009).
Nevertheless, the results achieved with the new compounds trans-
fecting polynucleotides into cells are not yet satisfactorily, mainly
because of the low transfection efficiency compared to viral vec-
tors. Therefore, one of primary tasks in the cationic lipid mediated
gene delivery consists in the design and synthesis of new vec-
tors, based on structural investigations of the lipoplex (cationic
lipid–DNA complex) morphology, to further improve the transfec-
tion efficacy (Ma et al., 2007). In general, the cationic lipids used
for gene transfer have the same schematic structure consisting of
three respectively four molecular parts: the hydrophobic chains
which are assumed to be responsible for the formation of lipo-
somes, in some lipids a backbone like glycerol, the spacer, and
the cationic head group. High efficiency in gene delivery relies
on an optimized interplay between these parts. Furthermore, the
co-lipid as an additional component plays an important role to
achieve high transfection efficacy (Ma et al., 2007; Zidovska et al.,
2009).
2
(
El-Aneed, 2004). Although the viral gene delivery systems have
a higher efficiency in gene transfer than the non-viral ones, they
Abbreviations:
mide ester; BOP, (benzotriazol-1-yloxy)-tris(dimethylamino)-phosphoniumhexa-
fluorophosphate; DOPE, 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine;
DSC, differential scanning
ethoxycarbonyl-1,2-dihydrochinoline;
-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
Boc-Lys(Boc)-OSu, N␣,N␧-di-boc-l-lysine-N-hydroxysuccini-
calorimetry;
FBS, fetal
EEDQ,
bovine
bromide;
2-ethoxy-1-
serum;
N/P
MTT,
ratio,
3
protonable nitrogen of cytofectines/phosphate groups of DNA - ratio; ONPG, O-
nitrophenyl-ˇ-galactopyranoside; PCS, photon correlation spectroscopy; PyBOP®
,
(
benzotriazol-1-yloxy)-tripyrrolidino-phosphoniumhexafluorophosphate; WAXS,
wide angle X-ray scattering.
∗ Corresponding author. Tel.: +49 345 55 25120; fax: +49 345 55 27018.
E-mail address: bodo.dobner@pharmazie.uni-halle.de (B. Dobner).
Present address: Universitätsklinikum Halle (Saale), Universitätsapotheke,
Ernst-Grube-Straße 40, 06092 Halle (Saale), Germany.
1
0
378-5173/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.ijpharm.2011.02.022

C. Wölk et al. / International Journal of Pharmaceutics 409 (2011) 46–56
47
Fig. 1. Structures and abbreviations of the six novel cationic lipids containing malonic acid diamide backbone.
Recently, we have reported on a new class of cationic lipids
2. Material and methods
based on a malonic acid diamide structure as the backbone with
two hydrophobic chains attached (Heinze et al., 2010). The amide
bonds, which connect one hydrophobic chain and the spacer as well
as the polar head group with the backbone, ensure biodegradabil-
ity which is necessary for low toxicity of the compounds (Lv et al.,
2.1. General procedures and materials
All materials and reagents were purchased from Sigma–Aldrich
Co. Ltd. unless stated otherwise. All solvents were analytically pure
and dried before use. Thin layer chromatography was carried out on
aluminum sheets coated with silica gel 60 F254 (Merck, Darmstadt,
Germany) and visualized with bromothymol blue. For column chro-
matography under normal pressure silica gel 60 (0.036–0.200 mm)
was used. Mass spectrometry analyses were performed with a
Finnigan MAT 710C (Thermoseparation Products, San Jose, USA)
for the ESI-MS spectra, and with a LTQ (linear ion trap)-Orbitrap
XL hybrid mass spectrometer (Thermo Fisher Scientific, Bremen,
Germany) for the HR-MS (high resolution mass spectrometry) spec-
2
006). Besides, amide bonds show higher stability with regard to
hydrolysis than ester bonds. Furthermore, a positive effect may
result from the planar geometry of the amide bonds which in
combination with the formation of a hydrogen bond network pro-
motes lamellar phases of the lipoplexes. Hence, we found good
serum stability compared to the commercially available transfec-
tion reagents. Also the synthesis used is a practicable method
using easily accessible materials. Starting from simple alkyl malonic
acid monoesters, the second hydrophobic chain, which is neces-
sary for effective transfection, was introduced by malonic acid
amide formation with long chain alkyl amines like hexadecylamine
or oleylamine. The cationic head group is formed via ethylene
diamine or tris(aminoethyl)amine groups or the amino acid lysine
using ethylene diamine as spacer receiving an oligoamide struc-
ture. These new compounds have shown transfection efficiencies
1
13
tra. H NMR and C NMR spectra were recorded on a Varian
Gemini 2000 and a Varian Inova 500. Elemental analyses were per-
formed with a CHNS-932 (Leco-corporation, St Joseph, Michigan,
TM
USA). SuperFect
and plasmid isolation kit were purchased from
QIAGEN (QIAGEN, Hilden, Germany). p-CMV-SPORT-␤-Gal plas-
TM
mid and LipofectAmine reagent were purchased from Invitrogen
TM
TM
comparable or even higher than LipofectAmine
or SuperFect
Life Technologies (Germany). A549 cells were purchased from Ger-
man Collection of Microoganisms and Cell Cultures (DSMZ GmbH,
Germany), and LLC PK1 cells (ATCC CL-11101) from the Ameri-
can Type Culture Collection (ATCC, USA). Fetal bovine serum, cell
culture media, and phosphate buffered saline were obtained from
PAA Laboratories (PAA Laboratories GmbH, Germany). Antibiotics
depending on the saturation degree of the hydrophobic chains and
on the size of the head group. As an extension of this work, we
report here about the synthesis and transfection characteristics of
six new multivalent cationic lipids with malonic acid diamide back-
bone and an enlarged head group by coupling with one or two lysine
molecules using tris(aminoethyl)amine as spacer (Fig. 1).
2
and MTT were purchased from Sigma (Sigma–Aldrich, Germany).
In addition, we have varied the alkyl chains by using differ-
ent combinations of hexadecyl and oleyl chains with the aim
to investigate structure–activity relationships. Lysine as a basic
proteinogenic amino acid enables the introduction of protonable
amino groups in the head group region of the lipids. Furthermore,
lysine has the ability to complex DNA as an element of histones
and poly-l-lysine (Luger et al., 1997; Park et al., 2006). The new
compounds were tested in cell cultures with regard to transfection
efficacy and toxicity properties. Furthermore, the cationic lipo-
somes and lipoplexes have been investigated with regard to their
size and surface charge by PCS2 and zeta potential measurements.
Selected lipids have been also investigated by DSC2 and WAXS2
experiments to determine phase transitions and phase structures.
The physical–chemical characterization has been performed with
the aim to obtain structure–function correlations of the used lipid
mixtures.
DOPE2 was purchased from Fluka (Fluka Chemie AG, Switzerland),
and cholesterol from Roth (Carl Roth GmbH, Germany).
2.2. Synthesis of the compounds
The synthesis of 1a,b; 2a,b; 3a–c and 4a–c is already described
(Heinze et al., 2010).
2.2.1. Coupling of lipophilic malonic acid monoamides with
tris(aminoethyl)amine
Reaction of compound 4a–c with tris(aminoethyl)amine to
5a–c:
General procedure: A solution of the acid 4a, 4b or 4c (1 mmol) and
PyBOP®2 (1 mmol, 520 mg) in CH Cl (40 mL) was added dropwise
2
2
to a mixture of tris(aminoethyl)amine (20 mmol, 2.93 g) and tri-

4
8
C. Wölk et al. / International Journal of Pharmaceutics 409 (2011) 46–56
m.p. 42–44 C; ESI-MS: m/z: 1060.6 [M+H] , 1094.5 [M+Cl] ; 1H
◦
+
−
ethylamine (1.5 mmol, 152 mg) under stirring. The mixture was
stirred at room temperature for 4 h. Then the solution was fil-
tered, washed with water and brine, dried over sodium sulfate,
filtered, and the solvent was evaporated. The crude compound
◦
3
NMR (CDCl , 400 MHz, 27 C) ı (ppm): 0.86 (t, JH,H = 6.3 Hz, 6H,
3
2[–CH CH ]), 1.24–1.82 (m, 74H, [–(CH ) CH (CH) CH (CH ) CH ]
7
2
3
2
2
2
2
2
6
3
[–(CH ) CH (CH) CH (CH ) CH ]
2[–C(CH ) ]
[–(CH2)3
2
6
2
2
2
2
6
3
3 3
5
a–c was purified by column chromatography using silica gel 60
CH NHBoc]), 1.96–1.99 (m, 8H, 2[–CH (CH) CH –]), 2.47–2.82
2
2
2
2
and CHCl /methanol/NH3 with gradient technique.
(m, 8H, [–NHCH CH N(CH CH NH–)(CH CH NH )]), 3.08–3.44
2 2 2 2 2 2 2
3
(
m, 9H, 3[–CONHCH –] [–CH NHBoc] [–COCHCO–]), 4.20–4.29
2 2
ꢀ
N-{2-[N,N-Bis(2-aminoethyl)amino]ethyl}-N ,2-
(m, 1H, [–COCHNHBoc–]), 5.30–5.36 (m, 4H, 2[–CH CH–]),
4.77–4.84/6.76–6.79/7.97–7.99/8.19–8.22/9.02–9.17 (5 × m, 5H,
2[–NHBoc] 3[–NHCO–]); Anal. calcd. for C61H117N O : C, 69.08; H,
dihexadecylpropane diamide (5a): Yield: 71.2%.
ꢀ
N-{2-[N,N-Bis(2-aminoethyl)amino]ethyl}-2-hexadecyl-N -[(9Z)-
7
7
octadec-9-enyl]propane diamide (5b): Yield: 64.4%.
11.12; N, 9.24; found: C, 69.27; H, 11.21; N, 9.46.
ꢀ
N-{2-[N,N-Bis(2-aminoethyl)amino]ethyl}-N ,2-[(9Z)-octadec-9-
enyl]propane diamide (5c): Yield: 85.5%.
2.2.3. Coupling of malonic acid diamides with two Boc-protected
lysine moieties
The experimental data of the compounds 5a–c are in agreement
with the literature (Heinze et al., 2010).
2.2.3.1. Reaction 5a–c to compounds 7a–c. Procedure 1: The reac-
tion was performed as described for the synthesis of 6a–c with the
difference that 1 mmol of compound 5a was treated with 2 mmol
(887 mg) of N ,N -di-boc-l-lysine-N-hydroxysuccinimide ester.
2
.2.2. Coupling of malonic acid diamides with one Boc-protected
lysine moiety
.2.2.1. Reaction 5a–c to Compounds 6a–c. General procedure:
Compound 5a, 5b or 5c (1 mmol) and triethylamine (1 mmol,
01 mg) were dissolved in CH Cl2 (20 mL). N˛,Nε-di-boc-l-lysine-
˛
ε
Procedure 2: The reaction was performed as described for the
synthesis of 6a–c with the difference that 1 mmol of compound 5b
or 5c was treated with 1.5 mmol (665 mg) of N ,N -di-boc-l-lysine-
2
˛
ε
1
N-hydroxysuccinimide ester.
2
N-hydroxysuccinimide ester (0.5 mmol, 220 mg) was given to this
solution. The mixture was stirred for 1 h at room temperature.
Following the solvent was evaporated. The residue was dissolved
in CHCl3 (50 mL), the organic solution was washed with a potas-
sium carbonate solution (15% m/m), dried over sodium sulfate,
filtered, and the solvent was evaporated. The crude products 6a–c
were purified by column chromatography using silica gel 60 and
2.2.3.2. N-(2-{N,N-Bis[2-(N-{2,6-di[N-(tert.-
butoxycarbonyl)amino]-1-oxohexyl}amino)ethyl]amino}ethyl)-
ꢀ
N ,2-dihexadecylpropane diamide (7a). Yield: 92% (procedure
1);
glass-like
colorless
substance;
C73H141N O12;
M:
9
−
1
m.p. 44–46 C; ESI-MS: m/z:1338.0 [M+H]+,
◦
1336.95 g mol
;
−
1
◦
1371.3 [M+Cl] ; H NMR (CDCl , 400 MHz, 27 C) ı (ppm):
0.86 (t, JH,H = 6.7 Hz, 6H, 2[–CH CH ]), 1.23–1.77 (m, 106H,
3
3
CHCl /methanol/NH3 with gradient technique.
3
2
3
[
–(CH2)15CH ] [–(CH ) CH ] 4[–C(CH ) ] 2[–(CH ) CH NHBoc]),
3 2 14 3 3 3 2 3 2
2
.2.2.2. N-(2-{N-(2-Aminoethyl)-N-[2-(N-{2,6-di[N-(tert.-
2.46–2.52
3.08–3.47
[–COCHCO–]),
(m,
(m,
6H,
13H,
4.21–4.30
[–NHCH CH N(CH CH NH–) ]),
2 2 2 2 2
butoxycarbonyl)amino]-1-oxohexyl}amino)ethyl]amino}ethyl)-
4[–CONHCH –]
2[–CH NHBoc]
2
2
ꢀ
N ,2-dihexadecylpropane diamide (6a). Yield: 41.7%; glass-like
(m,
2H,
2[–COCHNHBoc–]),
−
1
colorless substance; C57H113N7O7; M: 1008.55 g mol
4
NMR (CDCl , 400 MHz, 27 C) ı (ppm): 0.86 (t, JH,H = 6.6 Hz,
;
m.p.
4.76–4.84/5.61–5.63/6.95–6.99/7.41–7.46/7.63–7.66/7.82–7.92
◦
+
−
1
7–50 C; ESI-MS: m/z: 1009.5 [M+H] , 1043.2 [M+Cl] ;
H
(6 × m, 8H, 4[–NHBoc] 4[–NHCO–]); Anal. calcd. for C73H141N O12:
9
◦
3
3
C, 65.58; H, 10.63; N, 9.43; found: C, 65.76; H, 10.26; N 9.31.
6
H,
2[–CH CH ]),
1.08–1.80
(m,
82H,
[–(CH2)15CH3]
2
3
[
2
3
–CH (CH )14CH3]
.54–2.95 (m, 8H, [–NHCH CH N(CH CH NH–)(CH CH NH )]),
2[–C(CH ) ]
[–(CH ) CH NHBoc]),
2.2.3.3. N-(2-{N,N-Bis[2-(N-{2,6-di[N-(tert.-
2
2
3
3
2
3
2
2
2
2
2
2
2
2
butoxycarbonyl)amino]-1-oxohexyl}amino)ethyl]amino}ethyl)-
ꢀ
.09–3.72
(m,
9H,
3[–CONHCH –]
[–CH NHBoc]
2-hexadecyl-N -[(9Z)-octadec-9-enyl]propane diamide (7b). Yield:
2
2
[
4
5
–COCHCO–]),
.69–4.76/6.47–6.51/7.99–8.06/8.43–8.56/9.09–9.19
H, 2[–NHBoc] 3[–NHCO–]); Anal. calcd. for C57H113N7O7·2H O:
4.38–4.39
(m,
1H,
[–COCHNHBoc–]),
70.2% (procedure 2); glass-like substance; C75H143N O ; M:
9
12
− ◦
1
+
(5 × m,
1362.99 g mol ; m.p. 35–37 C; ESI-MS: m/z: 1363.8 [M+H] ,
−
1
◦
2
1397.2 [M+Cl] ; H NMR (CDCl , 400 MHz, 27 C) ı (ppm):
3
3
C, 65.54; H, 11.29; N, 9.39; found: C, 65.65; H, 11.16; N, 9.35.
0.86 (t, JH,H = 6.7 Hz, 6H, 2[–CH CH ]), 1.24–1.77 (m, 102H,
2
3
[
–(CH2)15CH3] [–(CH ) CH (CH) CH (CH ) CH ] 4[–C(CH ) ]
2 6 2 2 2 2 6 3 3 3
2
.2.2.3. N-(2-{N-(2-Aminoethyl)-N-[2-(N-{2,6-di[N-(tert.-
2[–(CH ) CH NHBoc]), 1.94–1.99 (m, 4H, [–CH (CH) CH –]),
2.45–2.57 (m, 6H, [–NHCH CH N(CH CH NH–) ]), 2.91–3.51 (m,
2 2 2 2 2
2 3 2 2 2 2
butoxycarbonyl)amino]-1-oxohexyl}amino)ethyl]amino}ethyl)-
ꢀ
2
-hexadecyl-N -[(9Z)-octadec-9-enyl]propane
diamide
(6b).
13H, 4[–CONHCH –] 2[–CH NHBoc] [–COCHCO–]), 4.22–4.27
2 2
Yield: 38.5%; glass-like colorless substance; C59H115N7O7; M:
(m, 2H, 2[–COCHNHBoc–]), 5.32–5.36 (m, 2H, [–CH CH–]),
4.76–4.84/5.60–5.62/6.94–6.98/7.42–7.47/7.62–7.66/7.84–7.91
−1
◦
+
1
1
0
034.58 g mol ; m.p. 36–39 C; ESI-MS: m/z: 1035.5 [M+H] ,
−
1
◦
069.3 [M+Cl] ; H NMR (CDCl , 500 MHz, 27 C) ı (ppm):
(6 × m, 8H, 4[–NHBoc] 4[–NHCO–]); Anal. calcd. for C75H143N O12:
3
9
3
.86 (t, JH,H = 6.9 Hz, 6H, 2[–CH CH ]), 1.22–1.82 (m, 78H,
C, 66.09; H, 10.57; N, 9.25; found: C, 66.15; H, 10.91; N, 9.48.
2
3
[
–(CH2)15CH3] [–(CH ) CH (CH) CH (CH ) CH ] 2[–C(CH ) ]
2 6 2 2 2 2 6 3 3 3
[
–(CH ) CH NHBoc]), 1.93–1.99 (m, 4H, [–CH (CH) CH –]),
2.2.3.4. N-(2-{N,N-Bis[2-(N-{2,6-di[N-(tert.-
2
3
2
2
2
2
2
3
4
4
2
6
.44–2.94 (m, 8H, [–NHCH CH N(CH CH NH–)(CH CH NH )]),
butoxycarbonyl)amino]-1-oxohexyl}amino)ethyl]amino}ethyl)-
2
2
2
2
2
2
2
ꢀ
.08–3.75 (m, 9H, 3[–CONHCH –] [–CH NHBoc] [–COCHCO–]),
N ,2-di[(9Z)-octadec-9-enyl]propane diamide (7c). Yield: 61%
2
2
.39–4.41 (m, 1H, [–COCHNHBoc–]), 5.31–5.36 (m, 2H, [–CH CH–]),
.71–4.79/6.46–6.55/7.99–8.04/8.29–8.52/9.07–9.16 (5 × m, 5H,
(procedure 2); glass-like colorless substance; C77H145N O12;
9
−
1
◦
M: 1389.03 g mol
;
m.p. 61–62 C; ESI-MS: m/z: 1389.5
+
−
1
[–NHBoc] 3[–NHCO–]); Anal. calcd. for C59H115N7O7·1.5H O: C,
[M+H] , 1424.8 [M+Cl] ;
H
NMR (CDCl3, 400 MHz,
2
◦
3
6.75; H, 11.20; N, 9.24; found: C, 66.67; H, 11.02; N, 9.10.
27 C)
ı
(ppm): 0.86 (t,
(m, 98H,
[–(CH ) CH (CH) CH (CH ) CH ]
JH,H = 6.8 Hz, 6H, 2[–CH CH ]),
[–(CH )7CH (CH) CH (CH ) CH ]
2 2 2 2 2 6 3
2
3
1
.24–1.75
2
.2.2.4. N-(2-{N-(2-Aminoethyl)-N-[2-(N-{2,6-di[N-(tert.-
4[–C(CH ) ]
2
6
2
2
2
2
6
3
3 3
butoxycarbonyl)amino]-1-oxohexyl}amino)ethyl]amino}ethyl)-
2[–(CH ) CH NHBoc]), 1.96–1.99 (m, 8H, 2[–CH (CH) CH –]),
2 3 2 2 2 2
ꢀ
N ,2-di[(9Z)-octadec-9-enyl]propane diamide (6c). Yield: 36.5%;
2.45–2.56 (m, 6H, [–NHCH CH N(CH CH NH–) ]), 2.91–3.36 (m,
2 2 2 2 2
−
1
glass-like colorless substance; C61H117N7O7; M: 1060.64 g mol
;
13H, 4[–CONHCH –] 2[–CH NHBoc] [–COCHCO–]), 4.22–4.25
2 2

C. Wölk et al. / International Journal of Pharmaceutics 409 (2011) 46–56
49
(
4
(
m, 2H, 2[–COCHNHBoc–]), 5.27–5.36 (m, 4H, 2[–CH CH–]),
.76–4.84/5.61–5.64/6.95–6.98/7.42–7.47/7.64–7.67/7.83–7.92
[–COCHCO–] [–COCHNH –]), 5.32–5.36 (m, 4H, 2[–CH CH–]),
2
13
7.01–7.06/7.82–7.86/8.11–8.16 (3 × m, 3H, 3[–NHCO–]); C NMR
◦
6 × m, 8H, 4[–NHBoc] 4[–NHCO–]), Anal. calcd. for C77H145N O12:
(CDCl , 100 MHz, 27 C) ı (ppm): 175.50, 175.43, 171.39, 171.30,
9
3
C, 66.58; H, 10.52; N, 9.08; found: C, 66.61; H, 10.23; N, 9.03.
170.99, 129.89, 129.68, 56.18, 55.23, 54.88, 54.51, 53.72, 41.70,
3
9.57, 39.44, 37.82, 37.20, 37.14, 35.25, 35.13, 32.82, 32.74, 32.67,
2
2
.2.4. Deprotection of amino groups
.2.4.1. Removal of the Boc group from 6a–c and 7a–c to synthesize
32.58, 31.89, 29.76, 29.69, 29.58, 29.54, 29.51, 29.44, 29.39, 29.31,
29.18, 27.68, 27.23, 26.97, 23.12, 22.66, 14.07.
compounds 8a–c and 9a–c. General procedure: Compound 6a–c or
a–c (1 mmol) was dissolved in CH Cl2 (20 mL). The solution was
7
2
2
.2.4.5. N-[2-(N,N-Bis{2-[N-(2,6-diamino-1-
treated with trifluoro acetic acid (5 mL) under stirring. The mix-
ture was allowed to stir for 5 h, then diluted with CHCl3 (50 mL),
and washed with diluted ammonia (3% m/m) until the water layer
shows a pH-value above 7. The organic layer was separated, dried
over sodium sulfate, filtered, and the solvent was evaporated. The
products 8a–c and 9a–c were purified by column chromatography
ꢀ
oxohexyl)amino]ethyl}amino)ethyl]-N ,2-dihexadecylpropane
diamide (9a). Yield: 89.6%; white crystalline substance;
−
1
◦
C53H109N O ; M: 936.49 g mol ; m.p. 89–91 C; ESI-MS: m/z:
9
found: 936.8673; H NMR (CDCl , 400 MHz, 27 C) ı (ppm):
0
9
4
+
2+
+
37.0 [M+H] , 469.4 [M+2H] ; HR-MS calcd.: 936.8675 [M+H]
1
◦
3
3
.84 (t, JH,H = 6.6 Hz, 6H, 2[–CH CH ]), 1.22–1.47 (m, 70H,
2
3
using silica gel 60 and CHCl /methanol/NH3 with gradient tech-
3
[
–(CH2)15CH ] [–(CH ) CH ] 2[–(CH ) CH NH ]), 2.34 (s, 8H,
3 2 14 3 2 3 2 2
[–NH ]), 2.49–2.53 (m, 6H, [–NHCH CH N(CH CH NH–) ]), 2.7 (t,
2 2 2 2 2 2
nique.
4
3
JH,H = 6.5 Hz, 4H, 2[–CH NH ]), 3.14–3.41 (m, 11H, 4[–CONHCH –]
2
2
2
2
.2.4.2. N-[2-(N-(2-Aminoethyl)-N-{2-[N-(2,6-diamino-1-
ꢀ
[–COCHCO–] 2[–COCHNH2–]), 7.21/7.49/7.69/7.78/7.97/8.05 (6 × t,
oxohexyl)amino]ethyl}amino)ethyl]-N ,2-dihexadecylpropane
13
◦
C NMR (CDCl /CD OD, 100 MHz, 27 C) ı
3 3
4
H, 4[–NHCO–]);
diamide (8a). Yield: 87%; white crystalline substance;
−
1
◦
(ppm): 175.57, 170.97, 170.82, 54.38, 53.56, 53.43, 53.23, 53.11,
40.41, 39.07, 37.17, 37.01, 34.43, 31.81, 31.42, 31.20, 29.19, 29.14,
29.01, 28.84, 28.76, 27.05, 26.49, 22.41, 22.16, 13.34.
C47H97N7O ; M: 808.32 g mol ; m.p. 74–75 C; ESI-MS: m/z:
8
found: 808.7727; H NMR (CDCl , 400 MHz, 27 C) ı (ppm):
0
3
+
2+
+
08.8 [M+H] , 405.6 [M+2H] ; HR-MS calcd.: 808.7726 [M+H] ,
1
◦
3
3
.85 (t, JH,H = 6.7 Hz, 6H, 2[–CH CH ]), 1.22–1.55 (m, 64H,
2
3
[
15
–(CH2) CH3]
14
[–(CH2) CH3]
[–(CH ) CH NH ]), 2.55–2.56
2 3 2 2
2.2.4.6. N-[2-(N,N-Bis{2-[N-(2,6-diamino-1-
(
m, 6H, [–NHCH CH N(CH CH NH–)(CH CH NH )]), 2.75 (t,
ꢀ
2
2
2
2
2
2
2
oxohexyl)amino]ethyl}amino)ethyl]-2-hexadecyl-N -[(9Z)-octadec-
3
JH,H = 6.0 Hz, 4H, 2[–CH NH ]), 3.12–3.41 (m, 8H, 3[–CONHCH –]
2
2
2
9-enyl]propane diamide (9b). Yield: 73%; white crystalline
[
(
–COCHCO–]
[–COCHNH –]),
7.27–7.38/7.85–7.93/8.16–8.23
−1 ◦
; m.p. 79–81 C;
2
substance; C55H111N O ; M: 962.53 g mol
9
4
3 × m, 3H, 3[–NHCO–]); 13C NMR (CDCl /CD OD, 100 MHz, 27 C)
◦
+
2+
3
3
ESI-MS: m/z: 963.6 [M+H] , 482.4 [M+2H] ; HR-MS calcd.:
62.8832 [M+H] , found: 962.8829; H NMR (CDCl , 500 MHz,
7 C) ı (ppm): 0.85 (t, JH,H = 6.8 Hz, 6H, 2[–CH CH ]), 1.23–1.78
ı (ppm): 175.64, 171.22, 171.04, 170.96, 54.36, 53.78, 53.60, 53.47,
4
2
+
1
9
2
3
0.08, 39.35, 39.23, 38.03, 37.31, 37.06, 34.41, 31.57, 29.85, 29.35,
9.30, 29.16, 28.99, 28.90, 27.15, 26.63, 22.31, 13.66.
◦
3
2
3
(m,
66H,
[–(CH2)15CH3]
[–(CH ) CH (CH) CH (CH ) CH ]
2 6 2 2 2 2 6 3
2
2
[–(CH ) CH NH ]), 1.94–1.99 (m, 4H, [–CH (CH) CH –]),
.47–2.55 (m, 6H, [–NHCH CH N(CH CH NH–) ]), 2.69 (t,
2 2 2 2 2
2
3
2
2
2
2
2
2
.2.4.3. N-[2-(N-(2-Aminoethyl)-N-{2-[N-(2,6-diamino-1-
ꢀ
oxohexyl)amino]ethyl}amino)ethyl]-2-hexadecyl-N -[(9Z)-octadec-
3
JH,H = 6.2 Hz, 4H, 2[–CH NH ]), 3.06–3.44 (m, 11H, 4[–CONHCH –]
2
2
2
9-enyl]propane diamide (8b). Yield: 97%; white crystalline
[
–COCHCO–] 2[–COCHNH –]), 5.32–5.35 (m, 2H, [–CH CH–]),
2
−1
◦
substance; C49H99N7O ; M: 834.36 g mol , m.p. 59–61 C; ESI-
13
3
+
7.98/7.89/7.72/7.66/7.44/7.11 (6 × t, 4H, 4[–NHCO–]);
C NMR
2+
MS: m/z: 834.8 [M+H] , 418.4 [M+2H] ; HR-MS calcd.: 832.7882
◦
(CDCl , 125 MHz, 27 C) ı (ppm): 175.38, 175.14, 171.84, 171.39,
3
+
1
◦
M+H] , found: 832.7886; H NMR (CDCl /CD OD, 400 MHz, 27 C)
3 3
[
171.14, 170.98, 129.93, 129.70, 55.23, 54.93, 54.70, 54.53, 54.43,
54.31, 53.72, 41.68, 41.62, 39.47, 37.97, 37.88, 37.24, 37.14, 35.38,
35.19, 33.01, 32.91, 32.79, 32.56, 31.87, 31.85, 29.72, 29.66, 29.63,
29.61, 29.47, 29.44, 29.36, 29.31, 29.27, 29.26, 29.14, 29.09, 27.64,
27.17, 26.95, 26.91, 23.09, 23.07, 22.64, 14.06.
ı
(ppm): 0.78 (t, 3JH,H = 6.4 Hz, 6H, 2[–CH CH ]), 1.15–1.69
2 3
(
m,
60H,
15
[–(CH2) CH3]
[–(CH ) CH (CH) CH (CH ) CH ]
2 6 2 2 2 2 6 3
[
–(CH ) CH NH ]), 1.87–1.92 (m, 4H, [–CH (CH) CH –]), 2.44–2.48
2
3
2
2
2
2
2
(
m, 6H, [–NHCH CH N(CH CH NH–)(CH CH NH )]), 2.64 (t,
2 2 2 2 2 2 2
JH,H = 6.0 Hz, 4H, 2[–CH NH ]), 2.89 (s, 6H, 3[–NH ]), 3.02–3.33
3
2
2
2
(
m, 8H, 3[–CONHCH –] [–COCHCO–] [–COCHNH –]), 5.24–5.28
2
2
(
m, 2H, [–CH CH–]), 7.46–7.50/7.90–7.93/8.01–8.12 (3 × m, 3H,
2.2.4.7. N-[2-(N,N-Bis{2-[N-(2,6-diamino-1-
1
3
◦
ꢀ
3
1
5
3
2
2
[–NHCO–]); C NMR (CDCl /CD OD, 100 MHz, 27 C) ı (ppm):
oxohexyl)amino]ethyl}amino)ethyl]-N ,2-di[(9Z)-octadec-9-
3
3
75.82, 171.55, 171.29, 129.89, 129.65, 55.43, 54.79, 54.26, 53.98,
3.85, 40.96, 40.94, 39.65, 39.52, 38.69, 37.76, 37.65, 37.36, 37.26,
4.95, 34.90, 32.56, 32.37, 31.89, 31.87, 31.75, 31.57, 31.51, 29.74,
9.68, 29.65, 29.63, 29.49, 29.46, 29.36, 29.32, 29.29, 29.28, 29.15,
7.54, 27.19, 26.95, 22.83, 22.81, 22.65, 14.04.
enyl]propane diamide (9c). Yield: 94.6%; white waxy substance;
−
1
◦
C57H113N9O4; M: 988.57 g mol
,
m.p. 83–85 C; ESI-MS:
+
2+
m/z: 988.7 [M+H] 495.2 [M+2H] ; HR-MS calcd.: 988.8988
+
1
[M+H] , found: 988.8987;
H
NMR (CDCl3, 400 MHz,
◦
3
27 C)
ı
(ppm): 0.81 (t,
(m, 62H,
[–(CH2)6CH2(CH)2CH2(CH2)6CH3]
JH,H = 6.9 Hz, 6H, 2[–CH2CH3]),
[–(CH )7CH (CH) CH (CH ) CH ]
1
.20–1.52
2
2
2
2
2
6
3
2
.2.4.4. N-[2-(N-(2-Aminoethyl)-N-{2-[N-(2,6-diamino-1-
2[–(CH2)3CH2NH2]),
1.73
ꢀ
oxohexyl)amino]ethyl}amino)ethyl]-N ,2-di[(9Z)-octadec-9-
(s, 8H, 4[–NH2]), 1.89–1.94 (m, 8H, 2[–CH2(CH)2CH2–]),
2.43–2.58 (m, 6H, [–NHCH2CH2N(CH2CH2NH–)2]), 2.63 (t,
enyl]propane diamide (8c). Yield: 91%; white waxy substance;
−1
◦
3
C51H101N7O3; M: 860.39 g mol
m/z: 861.0 [M+H] , 431.4 [M+2H] ; HR-MS calcd.: 860.8039
;
m.p. 62–65 C; ESI-MS:
JH,H = 6.1 Hz, 4H, 2[–CH2NH2]), 3.00–3.35 (m, 11H, 4[–CONHCH2–]
+
2+
[–COCHCO–] 2[–COCHNH2–]), 5.24–5.31 (m, 4H, 2[–CH CH–]),
+
1
13
[
M+H] , found: 860.8021;
H
NMR (CDCl3, 400 MHz,
7.13/7.45/7.61/7.66/7.84/7.92 (6 × t, 4H, 4[–NHCO–]);
C NMR
◦
3
◦
2
1
7 C)
ı
(ppm): 0.86 (t,
(m, 56H,
JH,H = 6.6 Hz, 6H, 2[–CH CH ]),
[–(CH )7CH (CH) CH (CH ) CH ]
2 2 2 2 2 6 3
(CDCl3, 100 MHz, 27 C) ı (ppm): 175.36, 175.16, 171.68, 171.26,
171.06, 170.92, 129.91, 129.68, 55.37, 55.35, 55.01, 54.81, 54.59,
54.49, 54.32, 53.82, 41.85, 41.82, 39.55, 38.03, 37.94, 37.38, 37.28,
35.44, 35.29, 33.23, 33.18, 32.96, 32.59, 31.89, 29.80, 29.77, 29.69,
29.59, 29.56, 29.51, 29.45, 29.38, 29.32, 29.29, 29.18, 29.13, 28.97,
27.67, 27.24, 27.01, 26.98, 23.19, 23.17, 22.66, 14.09.
2
3
.25–1.83
[
–(CH ) CH (CH) CH (CH ) CH ]
[–(CH ) CH NH ]),
2
6
2
2
2
2
6
3
2
3
2
2
1
.97–2.00
(m,
8H,
2[–CH (CH) CH –]),
2.49–2.57
2.70
2
(m,
(t,
2
2
2
6
H,
[–NHCH CH N(CH CH NH–)(CH CH NH )]),
2
2
2
2
2
2
3
JH,H = 6.4 Hz, 4H, 2[–CH NH ]), 3.05–3.43 (m, 8H, 3[–CONHCH ]
2
2
2

5
0
C. Wölk et al. / International Journal of Pharmaceutics 409 (2011) 46–56
2.3. Protocols for in vitro transfection experiments, viability tests,
five times on different days. For the experiments in the presence
and physical–chemical characterizations
of FBS lipoplexes were incubated with the cells as described above,
certainly the concentration of serum was adjusted to 10% from the
outset of experiments and kept for 48 h.
2.3.1. Preparation of liposomes
A CHCl /methanol solution (9/1, v/v) of the pure cationic lipid
3
was combined with a CHCl solution of DOPE or cholesterol, respec-
2.3.5. Toxicity assay
3
tively. The mixtures were dried under reduced pressure to remove
the organic solvent, and the dried film was vacuum-desiccated for
Cytotoxicity of the lipoplexes was assessed by the MTT
reduction assay (Mosmann, 1983). The cytotoxicity assays were
performed in 96 well plates by maintaining the same ratio
and number of cells and lipoplexes as used in the transfec-
tion experiments. MTT was added after refreshing the medium
(24 h after addition of lipoplexes to the cells) and incubated for
3
h. The lipid film was dissolved in distilled water at room temper-
−1
ature to a final concentration of 2 mg mL . The film was allowed
to swell in a water bath at 40 C for at least 15 min. Subsequently,
◦
the lipid mixtures were vortexed and sonicated to clarity in a bath
◦
◦
sonicator. The samples were stored in the refrigerator at 4 C before
3 h at 37 C. Afterward, the cells were lysed with a mixture of
experiments.
10% sodium dodecyl sulfate in acetic acid/dimethyl sulfoxide.
Absorption was measured at 570 nm. The results were expressed
as percent viability = [A570 (treated cells) − background]/[A570
(untreated cells) − background] × 100.
2.3.2. Preparation of plasmid DNA
p-CMV-SPORT-␤-Gal was isolated from Escherichia coli DH5a
(
Invitrogen Life Technologies) using a Quiagen plasmid mega
kit (Quiagen) following the manufacturer’s instructions (Quiagen
handbook). DNA with an OD260/OD280 = 1.94 was used for exper-
iments.
2.3.6. Differential scanning calorimetry (DSC)
Lipids were dispersed in MilliQ water to a concentration
−
1
of 1 mg mL . After hydration, the dispersions were vortexed,
sonicated and heated above the phase transition temperature
for several times. DSC measurements were performed on a
microcalorimetry system (MCS DSC, MicroCal Inc., Northampton,
MA). Sample and reference were degassed for at least 20 min
by slow stirring under vacuum. After filling, the DSC cells were
kept under 2 bar of nitrogen pressure to prevent evaporation at
2.3.3. Cell cultures
LLC PK1 cells (porcine kidney epithelial cells) were cultured in
7
2
5 cm2 tissue culture flasks in Medium 199 adjusted to contain
−
1
2
−1
gen-
.2 g L
sodium bicarbonate, 10% FBS , and 0.05 mg mL
◦
tamycin at 37 C and 5% CO . The cells were grown confluently
2
−
1
and were regularly split twice a week. Only cells that had under-
gone fewer than 30 passages were used for transfection and MTT
experiments. A549 cells (human lung carcinoma cells) were cul-
higher temperatures. The heating rate was 60 K h . Samples were
scanned for at least six times after being cooled down to 10 C in
order to check for reproducibility.
◦
2
tured in 75 cm tissue culture flasks in Dulbecco’s modified eagle
−
1
medium (DMEM) adjusted to contain 4.5 g L
glucose, 10% FBS,
2.3.7. Wide angle X-ray diffraction (WAXS)
Lipids were prepared as described above for DSC experiments.
Lipid concentrations of 15–20 wt% in MilliQ water were used. All
−1
◦
and 0.05 mg mL gentamycin at 37 C and 5% CO . The cells were
2
grown confluently and were regularly split twice a week. For exper-
iments, only cells in the range of passages 20–60 were used.
samples were transferred into glass capillaries (inner diameter
◦
1
.5 mm, GLAS, Germany) and stored for 7–10 days at 4 C before
2.3.4. Transfection biology
the measurements.
1
6–24 h before transfection, cells were seeded into a 96 well
Wide angle X-ray scattering (WAXS) experiments were carried
out at the Soft Condensed Matter beamline A2, HASYLAB, DESY
(Hamburg, Germany). WAXS data were collected by a linear detec-
tor with delay line readout. The incoming beam had a wavelength
of 0.15 nm, and the exposure time was 30 s. The temperature was
fixed during the measurements. After each heating step, a waiting
time of 5 min was used to ensure that the sample is in equilibrium.
Polyethylene terephthalate was used for calibration. Positions of
the Bragg peaks were converted into reciprocal spacings s = 1/d with
d being the real space repeat distances of the lattice planes.
plate at a density of 8000–10,000 cells/well by usage of LLC PK1
and18,000–20,000 in thecase of A549cells. Lipoplex mixtureswere
prepared by combining plasmid DNA (0.1 ␮g per well, 2.5 ␮g mL
with varying amounts of cationic liposome suspension in the
absence or presence of 10% serum in Medium 199 or DMEM. The
samples were incubated for 15 min at room temperature (total vol-
ume 40 ␮L/well). The N/P ratios were varied from 1:1 to 4:1 or
even higher in cases when the results of experiments foreshadow
increasing transfection efficacies at higher charge ratios. During
this time, cells were washed once with phosphate buffered saline
−
1
)
(
PBS). Then the complexes were added to the cells. After 4 h of incu-
2.3.8. Size measurements
bation, 160 ␮L of Medium 199 or DMEM was added to the cells in a
way that the final concentration of FBS reached 10%. The medium
was refreshed after 24 h, and the reporter gene activity was esti-
mated after 48 h. Therefore, the cells were washed with PBS and
lysed for 15 min in lysis buffer (5 mM Chaps in 50 mM Hepes buffer).
The solution was taken out of the wells, reunited in safe-lock tubes
The sizes of liposomes and lipoplexes were measured in dis-
tilled water by photon correlation spectroscopy on a HPPSsET
(Malvern, U.K.) with sample refractive index of 1.33 and a viscosity
◦
of 0.8872 mPa s at 25 C. Lipoplexes were prepared as for transfec-
tion experiments and measured with 2.5 ␮g DNA/mL sample. The
z-average and PDI were calculated by using the automatic mode.
Every sample was measured three times. Each spectroscopic run
consists of 10 consecutive scans with a duration of 10 s. Results
shown are the average of these three values.
(
Eppendorf), centrifuged, and stored on ice. It was pipeted again
−
1
2
in a 96 well plate and substrate solution (1.33 mg mL of ONPG ,
MgCl2 2 mM, ˇ-mercaptoethanol 100 mM in 0.2 M sodium phos-
◦
phate pH 7.3) was added and incubated for 30 min at 37 C. The
reaction was stopped with 1 M sodium carbonate. Absorption at
2.3.9. Zeta potential measurements
4
05 nm was converted to ˇ-galactosidase units using a calibration
Zeta potential (ꢀ) measurements of liposomes and lipoplexes
were performed with a Zetasizer Nano ZS (Malvern, U.K.) in dis-
tilled water. Based on the Hemholtz–Smoluchowski relation, the ꢀ
potential was automatically calculated by dispersions software 4.0
(Malvern, U.K.). Liposomes and lipoplexes, prepared as for trans-
fection experiments and measured with 2.5 ␮g DNA/mL sample,
curve constructed with pure commercial ˇ-galactosidase enzyme.
Protein concentration was detected with bichinonic acid reaction
(
Rosenberg, 1996). Absorption at 570 nm was converted by using a
calibration curve made of bovine serum albumin. All experiments
were carried out sixfold. Every experiment was repeated three to

C. Wölk et al. / International Journal of Pharmaceutics 409 (2011) 46–56
51
were measured with defined viscosity (0.8872 mPa s) and refractive
◦
index (1.33) at 25 C by three runs. Results shown are the average
of these three values.
2.3.10. Statistical analysis
All measurements were repeated several times (n ≥ 3–9) and
expressed as means ± standard deviation. One-way ANOVA was
used in conjunction with Bonferoni’s multiple comparison test to
assess the statistical significance.
3
. Results and discussion
3.1. Synthesis of malonic acid diamides
The synthesis of the compounds is shown in Scheme 1. The
Fig. 2. Representative DSC (heat flow dQ/dT versus temperature T) and WAXS (inten-
sity versus reciprocal spacing s) data of lipids 1 and 6. The WAXS patterns are taken
at temperatures marked with circles in the corresponding thermograms.
starting material was the malonic acid diethyl ester, which was
converted into monoalkylated malonic acid ester 1a,b via treat-
ment with hexadecyl bromide or oleyl methansulfonate using
published methods (Schmidt and Jankowski, 1996). This reaction
allows a simple variation of the alkyl chain by using differ-
ent alkyl halides or methansulfonates. The next reaction step
was a selective saponification with the aim that one ester func-
tion remains preserved as a protective group for the following
amide coupling. This mono-saponification was performed using
described methods (Breslow et al., 1944). A second alkyl chain
was introduced by coupling the alkyl malonic acid monoethyl
esters 2a,b with hexadecylamine or oleylamine to prepare the
3.2. In vitro transfection efficiency and cell toxicity of new
malonic acid diamides
To enhance gene transfection efficiency of cationic lipids, they
are usually mixed with co-lipids and used as liposomes. Most fre-
quently, cholesterol and DOPE are used as co-lipids (Kim et al.,
2
004; Ma et al., 2007; Mukherjee et al., 2005). These lipids help
to enable liposome formation, furthermore, they contribute to the
structure and physicochemical characteristics of the lipoplexes,
which are formed after incubation of cationic liposomes with DNA,
resulting in an influence on the transfection efficiency (Koltover
et al., 1998; Simberg et al., 2001; Zidovska et al., 2009; Zuhorn et al.,
2
-alkyl malonic acid ester amides 3a–c. For this reaction EEDQ2
was used owing to the good experience made in earlier works
Heinze et al., 2010). Following, the second carboxylic acid ester
(
group was transformed into the free acid group of 4a–c by a
further saponification step using potassium hydroxide as saponi-
fication agent. The received carbonic acids were coupled with
tris(aminoethyl)amine building the malonic acid diamides 5a–c
which were already reported (Heinze et al., 2010). The described
synthesis was modified by using PyBOP instead of BOP2 due
to the cancerogenic potential described for hexamethylphospho-
ramide, a byproduct of the BOP coupling (Coste et al., 1990).
To enlarge the cationic head group and increase the number of
protonable amino groups, the proteinogenic amino acid lysine
was chosen. For coupling the amino acid via amide bond to
2
002, 2005). Although the great influence of co-lipids for the trans-
fection efficiency is indisputable, the mechanism of interaction is
not fully understood and depends on many parameters. Currently
it is not possible to predict the ideal co-lipid, therefore a screening
for the optimal helper lipid for transfection experiments is nec-
essary. DOPE and cholesterol were also used as co-lipids in the
present case and mixed in different molar ratios with the tested
®
cytofectines to find the most effective lipid/co-lipid ratio. To find
the most effective N/P ratio2 for transfection, the DNA was mixed
with different amounts of these lipid mixtures. Only the data of the
in vitro transfection tests using the lipid/co-lipid ratios respectively
N/P ratios, which gave the best results, are shown. Surprisingly,
lipid 1 and lipid 2 do not form liposomes neither with DOPE nor
with cholesterol as co-lipid. Hence, no information about the in vitro
transfection efficiency of these lipids could be obtained.
A strong self-association could be responsible for the inability of
these lipids to form liposomes. Also the observation of crystalline
aggregates during the liposome preparation is in agreement with
such an assumption. Therefore, DSC and WAXS experiments were
used to show that lipid 1 and lipid 2 are in a highly ordered gel state,
whereas lipids 3, 4, 5 and 6 are in the liquid-crystalline state at room
temperature. In Fig. 2, representative DSC and WAXS data of lipid 1
are compared with those of lipid 6. The other lipids are still under
examination, and a detailed discussion of the thermodynamical and
structural data will follow soon elsewhere.
2
the tris(aminoethyl)amine spacer Boc-Lys(Boc)-OSu , as an acti-
vated Boc-protected derivative of lysine, was used. The activation
of the carbonyl group as N-hydroxysuccinimide ester enables
a racemisation-free amide coupling reaction (McDermott and
Benoiton, 1973). Therefore, we can assume that the stereogenic
centre in the lysine moiety is S-configured. First, we tried to syn-
thesize the compounds 6a–c and 7a–c in one reaction by using
a reaction ratio of lipid 5a–c to Boc-Lys(Boc)-OSu of 1:1.5 (n/n).
However, using this ratio the compounds 6b and 6c were isolated
in a yield lower than 5% so that only the bislysine derivatives 7b
and 7c yielded in satisfactory amounts. Because of this observa-
tion, the bislysine derivative 7a was synthesized with a reaction
ratio of 1:2 (n/n), with the aim not to waste the reactant 5a. The
monolysine derivatives 6a–c could only be obtained in an ade-
quate yield by using a reaction ratio of 1:0.5 (n/n). Due to the
fact that the yield is at most 50% using the reactants 5a–c, one
should attempt to improve this reaction step by using a protected
tris(aminoethyl)amine derivative. Finally, the cationic lipids 8a–c
and 9a–c could be obtained by treating the Boc-protected deriva-
tives 6a–c respectively 7a–c with trifluoro acetic acid. The new
lipids were chemically characterized and tested with regard to their
transfection efficiency.
The WAXS pattern of lipid 1 reveals an orthorhombic chain
−1
−1
◦
lattice (s11 = 2.38 nm , s02 = 2.59 nm ) at 20 C. The unit cell per-
˚ 2
pendicular to the chain axis is 5.0 × 7.7 A and therefore very
˚ 2
close to the well known herringbone arrangement (5.0 × 7.5 A ) of
hydrocarbon chains in organic crystals. The packing of the chains is
−1
extremely tight. Additionally, a Bragg peak at s = 2.10 nm repre-
sents hydrogen bonds between the head groups with a fixed repeat
distance of ∼0.476 nm. The Bragg peak at 2.80 nm belongs also
−1

5
2
C. Wölk et al. / International Journal of Pharmaceutics 409 (2011) 46–56
◦
2
◦
Scheme 1. Synthesis of novel cytofectines starting from alkyl malonic acid diethyl ester: (a) KOH, EtOH, rt; (b) HCl, H2O, 0 C; (c) R NH2, EEDQ, EtOH, 50 C; (d) KOH, EtOH,
8
◦
◦
®
0 C; (e) HCl, H2O, 0 C; (f) N(CH2CH2NH2)3, TEA, PyBOP , CH2Cl2, rt; (g) Boc-Lys(Boc)-OSu, TEA, CH2Cl2, rt; (h) TFA, CH2Cl2, rt; (i) NH3 (3% m/m), CHCl3.
to the proposed head group lattice, which is still under investiga-
tion. The phase formed at room temperature is therefore a sub-gel
phase.
attached to the tris(aminoethyl)amine spacer. As shown, lipid 6
achieves a statistically significant higher transfection efficiency (*)
than lipid 3. Lipid 3 is comparable with the reference substances
TM
TM
Strong attractive forces (van der Walls interactions of the lipid
chains and hydrogen bonding of the head groups) will reduce the
degree of hydration of the head groups. Water molecules are not
able to penetrate in between the heads. The broad transition that
LipofectAmine
and SuperFect
with regard to the transfection
efficiencies. These results show that the introduction of a second
lysine moiety in the head group area of the malonic acid diamides
increases the transfection efficiency in A549 cells. The results of
the MTT based cytotoxicity experiments of lipid 6 in A549 cells are
shown in Fig. 4.
◦
precedes the peak at 56 C in the DSC thermogram could be due
to a stepwise breaking of the hydrogen bonds. Only afterwards,
◦
the hydrocarbon chains can melt. At 60 C, the disappearance of
all peaks in the WAXS pattern evidences the liquid-crystalline
state with molten chains of lipid 1. The DSC scans are reversible
after cooling. The poor hydration behavior due to strong hydrogen
bonding networks can be a first explanation for the observation
of crystalline aggregates during the liposome preparation. The
homogeneous incorporation of helper lipids seems to be extremely
difficult.
In contrast, lipid 6 is already in the liquid-crystalline state
◦
at 20 C. The WAXS patterns show no Bragg peaks, and the DSC
thermogram does not show any transition peak up to high temper-
atures. Therefore, the chains are disordered, the molecules occupy
larger areas, and the heads are fully hydrated. The homogeneous
incorporation of proper helper lipids can be easily performed, and
the liposomes are therefore more stable.
Fig. 3 summarizes the relative in vitro gene delivery efficien-
cies of lipid 6 mixed with the co-lipid DOPE in a lipid/co-lipid
ratio of 2:1 (n/n) after 48 h in A549 cells using different N/P ratios.
The figure shows increasing transfection efficiency by increasing
Fig. 3. In vitro transfection efficacies of lipid 6 in A549 cells using DOPE as co-
lipid (at lipid/DOPE molar ratio of 2:1). Micro units of ˇ-galactosidase related to cell
protein concentration were plotted against the varying N/P ratios. The transfection
efficiencies of the liposome samples were compared to those of lipid 3/DOPE 1:2
the N/P ratio. Using a N/P ratio of 3.33, the observed transfection
TM
efficiency surpasses those of the references LipofectAmine
and
TM
TM
(n/n) with a N/P ratio of 4 and of the commercially available LipofectAmine and
SuperFect . These results are statistical significant (*). Further-
more, in Fig. 3 lipid 6 is compared with lipid 3 which contains
the same lipophilic molecule part but only one lysine moiety
TM
SuperFect . The transfection values shown are the average of at least three experi-
ments carried out on three different days. Statistical analyses were carried out with
p < 0.05, one-way ANOVA, and Bonferoni’s multiple comparison tests (*).

C. Wölk et al. / International Journal of Pharmaceutics 409 (2011) 46–56
53
Fig. 6. In vitro transfection efficacies of lipid 6 and lipid 3 in LLC PK1 (gray) and
TM
A549 (white) cells in the presence of serum compared with LipofectAmine and
TM
SuperFect . Concentration of FBS was adjusted to 10% at the outset and kept for
4
8 h. N/P ratios: lipid 6, 3.33:1; lipid 3, 4:1. The transfection values shown are the
Fig. 4. MTT-based cellular cytotoxicity of lipid 6 using DOPE as co-lipid (at a lipid/co-
lipid molar ratio of 2:1) in A549 cells. The percent cell viability was plotted against
the varying N/P ratios. The results are compared to those of lipid 3/DOPE 1:2 (n/n)
with a N/P ratio of 4, LipofectAmine , and SuperFect . The experiments were
performed as described in the text. The cytotoxicity values shown are the average
of at least three experiments carried out on different days.
average of leastwise three experiments carried out on three different days. Statistical
analyses were carried out with p < 0.05, one-way ANOVA, and Bonferoni’s multiple
comparison tests (*, ).
∧
TM
TM
better transfection results with cholesterol as co-lipid (Heinze et al.,
2010). Since the polar head group of DOPE is larger in comparison
to cholesterol, stronger interactions with the large head group of
the tested malonic acid amides might be responsible for this effect.
In contrast to this, lipid 4 which contains two hexadecyl chains
shows better transfection efficacy with cholesterol. It is known
that cholesterol has a fluidizing effect on lipid bilayers by disturb-
ing the packing of the alkyl chains (Oldfield and Chapman, 1972),
what seems to be the most reasonable explanation for this obser-
vation. However, precise explanations cannot be made without
further physicochemical investigations of the lipoplexes of mal-
onic acid diamides with co-lipids. Interestingly, Lipid 6/DOPE 2:1
(n/n) shows a statistically significant better transfection efficacy
It can be seen that increasing lipid concentration (higher N/P
ratio) results in decreasing cell viability. Nevertheless, the observed
viability in all tested N/P ratios is higher than that observed for
TM
LipofectAmine and lipid 3/DOPE 1:2 (n:n).
Fig. 5 summarizes the results of in vitro gene delivery of lipid
4, lipid 5, and lipid 6 in LLC PK1 cells after 48 h. The graph shows
the most efficient N/P ratios for the used lipids in combination with
either cholesterol or DOPE as co-lipid. The three tested lipids have
the same hydrophilic protonable head group but differ in the length
(
two CH -units) and the saturation degree of the alkyl chains. It can
2
be seen, that an increase of the number of unsaturated alkyl chains
results in higher transfection efficacies, regardless of using DOPE or
cholesterol as co-lipid. The fact that an increase of the alkyl chain
fluidity, e.g. by using unsaturated chains, leads to increasing trans-
fection efficacies has been already observed in earlier works in the
research group (Heinze et al., 2010) and is also described in the liter-
ature (Bennett et al., 1998; Felgner et al., 1994; Spector and Schnur,
TM
than LipofectAmine (data not shown) also observed in A549 cells
(see Fig. 3).
3.3. In vitro transfection in presence of serum
1
997). Furthermore, it can be seen that DOPE is the more effective
co-lipid compared to cholesterol (except lipid 4). This observa-
tion is in contrast to results achieved earlier with malonic acid
diamides containing a smaller head group, which often obtained
With the aim to realize in vivo application it is necessary to use
lipids forming lipoplexes which are stable in the presence of serum.
To get first evidence for the serum compatibility of the selected
lipids, cells were incubated with the lipoplexes in the presence of
1
0% FBS for 48 h. Admittedly, total protein concentration of human
−
1
serum, which is between 66 and 83 g l (Schaller et al., 2008), is
just reached with a level of 60%, but first information about serum
compatibility of the used lipoplexes can still be obtained.
Fig. 6 summarizes the transfection results achieved with lipid 6
and lipid 3 in LLC PK1 cells (gray) and A549 cells (white) in pres-
TM
TM
ence of serum compared with LipofectAmine
or SuperFect
.
First it should be mentioned that a decrease of the gene transfer was
observed for nearly all tested lipids compared to gene transfer with-
out serum over a period of 4 h from the beginning of the incubation.
TM
But the decrease was not as strong as observed for LipofectAmine
which shows very low transfection efficacy in A549 cells when
serum was added from the beginning. All lipids are significantly
TM
better than LipofectAmine
(
with regard to transfection efficacy
*, ). In A549 cells, lipid 3 surpasses LipofectAmineTM extensively.
∧
Surprisingly, the mixture lipid 3/DOPE 1:3 (n/n) with a N/P ratio
of 4 shows better transfection efficacies with 10% FBS used from
the beginning of the incubation compared with a 4 h serum free
interval at the beginning (data not shown).
Fig. 5. In vitro transfection efficacies of lipid 4, lipid 5, and lipid 6 in LLC PK1 cells.
N/P ratios: lipid 4, 2.5:1; lipid 5, 1.67:1 (Chol), 3.33:1 (DOPE); and lipid 6, 2.5:1
(
Chol), 1.67:1 (DOPE). The transfection values shown are the average of leastwise
three experiments carried out on three different days.

5
4
C. Wölk et al. / International Journal of Pharmaceutics 409 (2011) 46–56
Table 1
Size measurements (z-ave. = z-average, PDI values), and ꢀ potentials of lipid 3, lipid 4, lipid 5, lipid 6, and LipofectAmine using DOPE and cholesterol (Chol.) as co-lipid
at different lipid/co-lipid molar ratios). The results are given for the pure liposomes and the corresponding lipoplexes. The given z-average diameters, PDI values, and ꢀ
TM
(
potentials were determined by three runs.
Liposomes
Lipoplexes
z-Ave.± s (nm)
PDI ± v
ꢀ ± s (mV)
z-Ave.± s (nm)
PDI ± v
ꢀ ± s (mV)
Lipid 3/DOPE 1:2 (n/n) N/P 4:1
Lipid 3/DOPE 1:3 (n/n) N/P 4:1
Lipid 4/DOPE 1:1 (n/n) N/P 2.5:1
Lipid 4/Chol. 1:1 (n/n) N/P 2.5:1
Lipid 5/DOPE1:2 (n/n) N/P 3.33:1
Lipid 5/Chol.1:1 (n/n) N/P 1.67:1
Lipid 6/DOPE2:1 (n/n) N/P 1.67:1
Lipid 6/DOPE2:1 (n/n) N/P 3.33:1
68 ± 1
103 ± 3
232 ± 6
85 ± 5
0.33 ± 6%
0.25 ± 8%
0.46 ± 5%
0.76 ± 3%
0.72 ± 25%
0.22 ± 5%
0.16 ± 42%
0.16 ± 42%
0.37 ± 17%
0.54 ± 25%
0.54 ± 25%
32.1 ± 2.5
28.6 ± 2.7
19.5 ± 5.7
38.3 ± 2.8
30.6 ± 1.7
47.6 ± 2.2
39.3 ± 1.9
39.3 ± 1.9
38.7 ± 3.2
48.4 ± 3.2
48.4 ± 3.2
109 ± 1
184 ± 1
93 ± 1
0.16 ± 10%
0.18 ± 1%
0.35 ± 1%
0.71 ± 4%
0.07 ± 45%
0.23 ± 7%
0.31 ± 1%
0.32 ± 29%
0.22 ± 12%
0.20 ± 7%
0.20 ± 1%
23.5 ± 0.86
22.5 ± 1.66
9.6 ± 0.51
8.1 ± 1.31
8.2 ± 1.56
8.9 ± 0.88
6.3 ± 0.50
14.5 ± 2.17
6.6 ± 0.71
30.7 ± 0.9
39.3 ± 4.2
244 ± 4
853 ± 29
103 ± 2
450 ± 1
1415 ± 53
992 ± 33
120 ± 1
107 ± 1
623 ± 67
91 ± 1
1592 ± 98
1592 ± 98
142 ± 6
154 ± 10
154 ± 10
Lipid 6/Chol. 1:2 (n/n) N/P 2.5:1
TM
LipofectAmine
LipofectAmine
N/P 3:1
N/P 9:1
TM
3
.4. Size measurements of liposomes and lipoplexes
the two effective mixtures of lipid 3 in A549 cells which form only
very small particles (109 nm and 184 nm). The results of lipid 3 con-
tradict statements made in literature earlier (Ma et al., 2007; Turek
et al., 2000), proposing that lipoplexes of less than 250 nm in size
show efficient transfection only in the absence of serum. As men-
tioned above, lipid 3/DOPE1:3(n/n) withN/Pratio of4 shows better
transfection results in presence of serum, compared with serum
free tests. The measurements in serum were made with regard of
a possible application in vivo. For successful in vivo gene transfer,
particles should have a size less than 150 nm, because larger ones
are quickly cleared from blood (Kayser and Kiderlen, 2004; Li and
Szoka, 2007). Therefore, the particle sizes of lipid 3 are preferential
for in vivo application, whereas the size of lipid 6/DOPE 2:1 (n/n)
N/P 3.33:1 only allows ex vivo application. But, one has to take into
account that the size measurements were carried out under ideal-
ized conditions with water as medium. Salt containing medium or
serum could influence the particle size drastically. Because of this,
the measured z-averages must not reflect the true particle sizes
during transfection experiments. We tried to investigate the par-
ticle size in serum, but because of the serum proteins this is not
possible by PCS.
Table 1 summarizes the results of size measurements of lipo-
somes and lipoplexes. The polydispersity index (PDI) is also shown.
The PCS measurements of the cationic liposomes respectively
lipoplexes show that the particle sizes differ very much. Since the
liposome size strongly depends on the method of liposome prepa-
ration, and furthermore, the liposome structures change during
lipoplex formation, we will focus on lipoplex sizes in the following
part. Lipoplexes from 100 nm to nearly 1500 nm (comparing the
different lipids at different lipid/co-lipid and N/P ratios) have been
observed. The main question was whether a correlation between
particle size and transfection efficiency exists.
Fig. 7 shows the dependence of the particle size on the N/P ratio
in case of lipid 6/DOPE 2:1 (n/n) lipoplexes. Decreasing N/P ratio
leads to a decrease of the particle size, a phenomenon that has been
described in literature earlier (Eastman et al., 1997). Comparing
the particle size (Fig. 7) with the transfection efficiencies (Fig. 3)
shows that larger lipoplexes are more effective in gene transfer,
an observation which was made by other researches too (Sakurai
et al., 2000). Zuhorn proposed that the particle size may reflect
the structural instability needed for perturbing intracellular mem-
brane barriers to allow cytosolic escape of DNA resulting in higher
transfection efficiency (Shi et al., 2002; Zuhorn et al., 2007). Other
arguments for the observed higher transfection efficiency of larger
lipoplexes are the dependence of endocytosis pathways from par-
ticle sizes (Zuhorn et al., 2007), and a better sedimentation of larger
particles during lipoplex incubation of cells in vitro (Ma et al., 2007).
Furthermore, this figure illustrates the drastic size change by mix-
ing liposomes with DNA. Starting from liposomes with a z-average
of 1592 nm, the decrease of the N/P ratio decreases continuously
the lipoplex size to values below 400 nm. In contrast, Table 1 shows
that in most cases the lipoplexes have larger diameters compared
to the corresponding liposomes.
Comparing the transfection results (Fig. 5) with lipoplex sizes
shows a tendency, that larger lipoplexes are more efficient in gene
transfer. But it is difficult to make clear correlations between par-
ticle sizes and transfection efficacy because the lipoplexes differ
in the cationic lipid, the type and amount of co-lipid, and the N/P
ratio. These three factors influence either transfection efficiency
or particle size. It is an interesting question whether the particle
size influences directly the transfection efficiency or can be taken
only as an indicator, resulting from other factors enhancing gene
transfer directly. The influence of particle size is still controversially
discussed (Ma et al., 2007; Zuhorn et al., 2007).
Fig. 7. Photon correlation spectroscopy analyses and ꢀ potential measurements of
liposomes and lipoplex mixtures of lipid 6 using DOPE as co-lipid (at lipid/DOPE
molar ratio of 2:1). First y-axis (circles): changes of liposome and lipoplex size as a
function of decreasing lipid concentration (concentration of DNA = 2.5 ␮g/mL). The
average diameters of liposomes and lipoplexes were determined by three spectro-
scopic runs comprising ten consecutive scans, each of 10 s duration. Second y-axis
The comparison of Fig. 6 with Table 1 is interesting. The most
effective lipid–DNA mixture of lipid 6/DOPE 2:1 (n/n) N/P 3.33:1 in
LLC PK1 cells forms very large lipoplexes (1415 nm) compared to
(
squares): changes of ꢀ potentials of liposomes and lipoplexes as a function of
decreasing lipid concentration (concentration of DNA = 2.5 ␮g/mL). The ꢀ potentials
of liposomes and lipoplex particles were determined by three runs.

C. Wölk et al. / International Journal of Pharmaceutics 409 (2011) 46–56
55
3
.5. Zeta potential measurements of liposomes and lipoplexes
Surprisingly, predominantly DOPE turns out to be the more
effective co-lipid in contrast to the observation made for the other
malonic acid diamides, which contain a smaller head group. For
them cholesterol was found to be more efficacious. In summary, a
series of new cationic malonic acid diamides was synthesized and
chemically characterized. The testing of these new compounds in
cell culture with regard to transfection efficacy and cytotoxicity
shows that they are able to transfer genetic material with low tox-
icity. In some cases, the tested lipids were considerably better than
Both liposomes and lipoplexes were investigated with regard
to their surface charge by ꢀ potential measurements in water. This
allows to the direct comparison with the PCS measurements. The
obtained ꢀ potentials are shown in Table 1. The charges measured
for the liposomes are in the range between +20 mV and +50 mV
depending on the co-lipid and the lipid/co-lipid ratio (Perrie
TM
and Gregoriadis, 2000). Also the potential of LipofectAmine
TM
TM
(
+48.4 mV) is in this range. Comparing the ꢀ potentials with the
the references LipofectAmine
and SuperFect . Lipid 6 has the
gene transfer efficiency, no direct correlation could be observed.
For example, the efficient lipid mixture (lipid 6/DOPE 2:1 (n/n)) and
the ineffective mixture (lipid 4/cholesterol 1:1 (n/n)) show nearly
the same ꢀ potential (Table 1 and Fig. 5).
highest transfection potential of the six new lipids in both cell lines.
Finally, lipid 3 and lipid 6 show promising transfection results
in the presence of serum. Although in most cases a decreasing of
the transfection efficacies was observed in serum in comparison
with serum free tests, these lipids show significantly better results
For all lipid mixtures used, the ꢀ potential decreases after addi-
tion of DNA (Table 1). This shows that DNA successfully couples to
the liposomes to form the lipoplexes. Interestingly, the ꢀ potential
of all lipoplexes investigated is still positive showing that the DNA
coupling does not lead to a charge reversal. There are different the-
ories in literature explaining the positive ꢀ potential of lipoplexes,
either by an entrapment of DNA by the surrounding cationic lipids
or a partial adsorption of DNA at the surface (Eastman et al., 1997;
Ma et al., 2007; Mevel et al., 2008). Further investigations of the
lipoplex morphology are needed to get a clear answer.
TM
than LipofectAmine . Surprisingly, lipid 3/DOPE 1:3 (n/n) with
a N/P ratio of 4 shows even better transfection results in pres-
ence of serum compared to tests using a serum free interval at the
beginning.
First DSC and WAXS experiments show a clear correlation
between the phase transition parameters as well as phase struc-
tures and the ability of the new lipids to form stable liposomes
as a precondition for successful transfection. The investigations
of the particle size and ꢀ potential gives interesting results for
first discussion, but do not generate clear correlations between
these parameters and the transfection efficiency. Therefore, fur-
ther investigations are need to find the most important parameters
for such a correlation, which would make the time intensive and
costly cellular transfection screening experiments unnecessary.
Fig. 7 shows the change of the ꢀ potential for the lipid 6/DOPE
2
:1 (n/n) mixture in dependence of the N/P ratio. A decrease of the
positive charge after mixing the liposomes with DNA can be seen. It
seems that the decrease of the ꢀ potential levels off around +7 mV
below a N/P ratio of 2.5:1. Again, comparing the ꢀ potential (Fig. 7)
with the transfection efficiency (Fig. 3), no clear correlation can be
observed.
Acknowledgments
The charges of the other lipoplexes formed with lipid 4, lipid
5
and lipid 6 (by varying N/P ratios and co-lipids) are all in the
We thank the Max Planck Institute of Colloids and Interfaces
Potsdam for financial support, and HASYLAB at DESY, Hamburg,
Germany, for beamtime and excellent support. Furthermore, the
authors wish to thank Dr. Christian Ihling and Prof. Dr. Andrea Sinz
range of +6 mV to +15 mV, and do also not correlate with the gene
transfer efficiency (comparing Table 1 and Fig. 5). Only lipid 3 and
LipofectAmineTM show in their lipoplexes higher positive values
of the ꢀ potential. In the case of lipid 3, a smaller decrease of the
(Institute of Pharmacy, Martin-Luther-University, Halle (Saale),
ꢀ
potential after lipoplex formation has been observed (compared
Germany) for the high resolution mass spectrometry analyses.
with the values for lipids 4–6). This is maybe a hint for efficient
DNA entrapment in multilamellar lipoplex structures (Ma et al.,
References
2
007) explaining the good transfection results in serum. However,
the high N/P ratio of 4 can also explain this observation, because
Bennett, C.F., Mirejovsky, D., Crooke, R.M., Tsai, Y.J., Sridkar, C.N., Wheeler, C.J.,
Felgner, P.L., 1998. Structural requirements for cationic lipid mediated phospho-
rothioate oligonucleotides delivery to cells in culture. J. Drug Target. 5, 149–162.
Bhattacharya, S., Bajaj, A., 2009. Advances in gene delivery through molecular design
of cationic lipids. Chem. Commun., 4632–4656.
LipofectAmineTM shows high charge values at high N/P ratios too
(
Table 1). Interestingly, the values measured for the lipoplexes of
lipid 3/DOPE with an N/P ratio of 4 at different lipid/co-lipid ratios
1:2 or 1:3) are nearly the same, which supports the idea of good
(
Breslow, D.S., Baumgarten, E., Hauser, C.R., 1944. A new synthesis of ␤-keto esters
of the type RCOCH2COOC2H5. J. Am. Chem. Soc. 66, 1286–1288.
DNA entrapment into the lipoplex structure. But, as mentioned
above further structural investigations are need.
Coste, J., Le-Nguyen, R., Castro, B., 1990. PyBOP®: a new peptide coupling reagent
devoid of toxic by-product. Tetrahedron Lett. 31, 205–208.
Eastman, S.J., Siegel, C., Tousignant, J., Smith, A.E., Cheng, S.H., Scheule, R.K., 1997.
Biophysical characterisation of cationic lipid: DNA complexes. Biochim. Biophys.
Acta 1325, 41–62.
4
. Conclusions
El-Aneed, A., 2004. An overview of current delivery systems in cancer gene therapy.
J. Control. Rel. 94, 1–14.
By designing new lipids for non-viral gene delivery it is impor-
tant to investigate structure–activity relationships to expedite the
development of cytofectines. This work extends the effort to find
such structure–activity relationships for the new class of cyto-
fectines with malonic acid diamide backbone by enlarging the head
group region. In agreement with the results of earlier works we
found that the insertion of at least one unsaturated alkyl chain leads
to higher transfection efficacies. Furthermore, in the case of lipid 3
and lipid 6 tested in A549 cells, it was shown that an enlargement
of the cationic head group, by coupling with two lysine molecules
instead of only one, results in an enhancement of the transfection
efficiency. Unfortunately, this observation could not be supported
by using the other malonic acid diamides which contain one or two
saturated alkyl chains in the lipophilic molecule part, because of
the inability of lipid 1 and lipid 2 to form liposomes.
Felgner, J.H., Kumar, R., Sridhar, C.N., Wheeler, C.J., Tsai, Y.J., Border, R., Ramsey, P.,
Martin, M., Felgner, P.L., 1994. Enhanced gene delivery and mechanism studies
with a novel series of cationic lipid formulations. J. Biol. Chem. 269, 2550–2561.
Heinze, M., Brezesinski, G., Dobner, B., Langner, A., 2010. Novel cationic lipids based
on malonic acid amides backbone: transfection efficacy and cell toxicity prop-
erties. Bioconjugate Chem. 21, 696–708.
Hyde, S.C., Southern, K.W., Gileadi, U., Fitzjohn, E.M., Mofford, K.A., Waddell, B.E.,
Gooi, H.C., Goddard, C.A., Hannavy, K., Smyth, S.E., Egan, J.J., Sorgi, F.L., Huang,
L., Cuthbert, A.W., Evans, M.J., Colledge, W.H., Higgins, C.F., Webb, A.K., Gill,
D.R., 2000. Repeat administration of DNA/liposomes to the nasal epithelium of
patients with cystic fibrosis. Gene Ther. 7, 1156–1165.
Kay, M.A., Glorioso, J.C., Naldini, L., 2001. Viral vectors for gene therapy: the art of
turning infectious agents into vehicles of therapeutics. Nat. Med. 7, 33–40.
Kayser, O., Kiderlen, A.F., 2004. Non viral gene delivery systems. In: Pharmaceutical
Biotechnology. Wiley-VCH, Weinheim.
Kim, H.S., Moon, J., Kim, K.S., Choi, M.M., Lee, J.E., Heo, Y., Cho, D.H., Jang, D.O., Park,
Y.S., 2004. Gene-transferring efficiencies of novel diamino cationic lipids with
varied hydrocarbon chains. Bioconjugate Chem. 15, 1095–1101.

5
6
C. Wölk et al. / International Journal of Pharmaceutics 409 (2011) 46–56
Koltover, I., Salditt, T., Radler, J.O., Safinya, C.R., 1998. An inverted hexagonal phase of
cationic liposome-DNA complexes related to DNA release and delivery. Science
Rosenberg, I.M., 1996. Protein Analysis and Purification: Benchtop Techniques.
Birkenhäuser, Boston. 0-8176-3717-6.
2
81, 78–81.
Sakurai, F., Inoue, R., Nishino, Y., Okuda, A., Matsumoto, O., Taga, T., Yamashita, F.,
Takakura, Y., Hashida, M., 2000. Effect of DNA/liposome mixing ratio on the
physicochemical characteristics, cellular uptake and intracellular trafficking of
plasmid DNA/cationic liposome complexes and subsequent gene expression. J.
Control. Rel. 66, 255–269.
Li, W., Szoka Jr., F.C., 2007. Lipid-based nanoparticles for nucleic acid delivery. Pharm.
Res. 24, 438–449.
Luger, K., Mäder, A.W., Richmond, R.K., Sargent, D.F., Richmond, T.J., 1997. Crys-
tal structure of the nucleosome core particle at 2.8 A˚ resolution. Nature 389,
2
51–260.
Schaller, J., Gerber, S., Kaempfer, U., Lejon, S., Trachsel, C., 2008. Human Blood Plasma
Proteins. John Wiley & Sons Ltd., Chichester-West Sussex, p. 17.
Schmidt, R.R., Jankowski, K., 1996. New types of nonionic surfactants with sugar
head groups. Liebigs Ann., 867–879.
Seow, Y., Wood, M.Y., 2009. Biological gene delivery vehicles: beyond viral vectors.
Mol. Ther. 17, 767–777.
Lv, H., Zhang, S., Wang, B., Cui, S., Yan, J., 2006. Toxicity of cationic lipids and cationic
polymers in gene delivery. J. Control. Rel. 114, 100–109.
Ma, B., Zhang, S., Jiang, H., Zhao, B., Lv, H., 2007. Lipoplex morphologies and
their influences on transfection efficiency in gene delivery. J. Control. Rel. 123,
1
84–194.
McDermott, J.R., Benoiton, N.L., 1973. N-Methylamino acids in peptide synthe-
Shi, F., Wasungu, L., Nomden, A., Stuart, M.C.A., Polushki, E., Engberts, J.B.F.N., Hoek-
stra, D., 2002. Interference of poly(ethylene glycol)-lipid analogues with cationic
lipid-mediated delivery of oligonucleotides; role of lipid exchangeability and
non-lamellar transitions. Biochem. J. 366, 333–341.
Simberg, D., Danino, D., Talmon, Y., Minsky, A., Ferrari, M.E., Wheeler, C.J., Barenholz,
Y., 2001. Lipids and lipoproteins. J. Biol. Chem. 276, 47453–47459.
Spector, M.S., Schnur, J.M., 1997. DNA ordering on a lipid membrane. Science 275,
791–792.
sis. IV. Racemization and yields in peptide-bond formation. Can. J. Chem. 51,
2
562–2570.
Mevel, M., Breuzard, G., Yaouanc, J.J., Clement, J.C., Lehn, P., Pichon, C., Jaffres, P.A.,
Midoux, M., 2008. Synthesis and transfection activity of new cationic phospho-
ramidate lipids: high efficiency of an imidazolium derivative. ChemBioChem 9,
1
462–1471.
Mintzer, M.A., Simanek, E.E., 2009. Nonviral vectors for gene delivery. Chem. Rev.
09, 259–302.
1
Turek, J., Dubertret, C., Jaslin, G., Antonakis, K., Scherman, D., Pitard, B., 2000. Formu-
lations which increase the size of lipoplexes prevent serum associated inhibition
of transfection. J. Gene Med. 2, 32–40.
Mosmann, T., 1983. Rapid colorimetric assay for cellular growth and survival:
application to proliferation and cytotoxicity assays. J. Immunol. Methods 65,
5
5–63.
Young, Z.R., Wang, H.F., Zhao, J., Peng, Y.Y., Wang, J., Guinn, B.A., Huang, L.Q., 2007.
Recent developments in the use of adenoviruses and immunotoxins in cancer
gene therapy. Cancer Gene Ther. 14, 599–615.
Zhang, J., Jia, D., Olson, B.E., Campbell, A.G., Thule, P.M., 2009. Hepatic insulin gene
therapy diminishes liver glycogen despite insulin responsive transcriptional
effects in diabetic CD-1 mice. J. Gene Med. 11, 588–597.
Mukherjee, K., Sen, J., Chaudhuri, A., 2005. Common co-lipids, in synergy, impart
high gene transfer properties to transfection-incompetent cationic lipids. FEBS
Lett. 579, 1291–1300.
Nazari, R., Jashi, S., 2009. Recent developments in the use of adenoviruses and
immunotoxins in cancer gene therapy. Curr. Gene Ther. 9, 20–25.
Noone, P.G., Hohneker, K.W., Zhou, Z., Johnson, L.G., Foy, C., Gipson, C., Jones, K.,
Noah, T.L., Leigh, M.W., Schwartzbach, C., Efthimiou, J., Pearlman, R., Boucher,
R.C., Knowles, M.R., 2000. Safety and biological efficacy of a lipid–CFTR complex
for gene transfer in the nasal epithelium of adult patients with cystic fibrosis.
Mol. Ther. 1, 105–114.
Zidovska, A., Evans, H.M., Ahmad, A., Ewert, K.K., Safinya, C.R., 2009. The
role of cholesterol and structurally related molecules in enhancing trans-
fection of cationic liposome–DNA complexes. J. Phys. Chem.
5208–5216.
B 113,
Zuhorn, I.S., Oberle, V., Visser, W.H., Engberts, J.B., Bakowsky, U., Polushkin, E., Hoek-
stra, D., 2002. Phase behavior of cationic amphiphiles and their mixtures with
helper lipid influences lipoplex shape, DNA translocation, and transfection effi-
ciency. Biophys. J. 83, 2096–2108.
Oldfield, E., Chapman, D., 1972. Molecular dynamics of cerebroside-cholesterol and
sphingomyelin—cholesterol interactions: implications for myelin membrane
structure. FEBS Lett. 21, 303–306.
Park, T.G., Jeong, J.H., Kim, S.W., 2006. Current status of polymeric gene delivery
systems. Adv. Drug Deliv. Rev. 58, 467–486.
Zuhorn, I.S., Bakowsky, U., Polushkin, E., Visser, W.H., Stuart, M.C., Engberts, J.B.,
Hoekstra, D., 2005. Nonbilayer phase of lipoplex–membrane mixture deter-
mines endosomal escape of genetic cargo and transfection efficiency. Mol. Ther.
11, 801–810.
Zuhorn, I.S., Engberts, J.B.F.N., Hoekstra, D., 2007. Gene delivery by cationic lipid
vectors: overcoming cellular barriers. Eur. Biophys. J. 36, 349–362.
Perrie, Y., Gregoriadis, G., 2000. Liposome-entrapped plasmid DNA: characterisation
studies. Biochim. Biophys. Acta 1475, 125–132.
®
Quiagen Plasmid Purification Handbook, 2011, www.kirschner.med.harvard.edu/
files/protocols/QIAGEN QIAGENPlasmidPurification EN.pdf.