N1,N12-diacyl spermines: SAR studies on non-viral lipopolyamine vectors for plasmid DNA and siRNA formulation

Article abstract of DOI:10.1007/s11095-008-9764-3

Purpose: To study the efficiency of novel synthetic N¹,N ¹²-diacyl spermines on DNA and siRNA binding, and to compare their transfection efficiency with viability in cell lines. Methods: Five long chain N¹,N¹²-diacyl lipopolyamines: N¹,N ¹²-[didecanoyl, dimyristoyl, dimyristoleoyl, distearoyl and dioleoyl]-1,12-diamino-4,9-diazadodecane were synthesized from the naturally occurring polyamine spermine. Their abilities to condense DNA and to form nanoparticles were characterized. Transfection efficiencies were studied in FEK4 primary skin cells and in an immortalized cancer cell line (HtTA), and compared with N⁴,N⁹-regioisomers. Also, the abilities of these novel compounds to bind to siRNA-forming nanoparticles were studied using a RiboGreen intercalation assay, and their abilities to deliver fluorescein-tagged siRNA into cells were quantified and compared with TransIT-TKO. Results: By incorporating two long aliphatic chains and varying their acylation position, length, and oxidation state in a stepwise manner, we show efficient p and siRNA formulation and delivery to primary skin and cancer cell lines. Although two C14 chains (both saturated or both mono-cis-unsaturated) were efficient transfecting agents, they were highly toxic. Conclusions: N¹,N ¹²-Dioleoyl spermine efficiently binds to and delivers pDNA and siRNA with high cell viability even in a primary skin cell line. It is a novel, efficient non-viral vector in the presence of serum.

Full text of DOI:10.1007/s11095-008-9764-3

#
Pharmaceutical Research, Vol. 27, No. 1, January 2010 ( 2008)
DOI: 10.1007/s11095-008-9764-3
Research Paper
N1,N12-Diacyl Spermines: SAR Studies on Non-viral Lipopolyamine Vectors
for Plasmid DNA and siRNA Formulation
Hassan M. Ghonaim,1 Shi Li,1 and Ian S. Blagbrough1,2
Received August 12, 2008; accepted October 21, 2008; published online October 30, 2009
Purpose. To study the efficiency of novel synthetic N1,N12-diacyl spermines on DNA and siRNA binding,
and to compare their transfection efficiency with viability in cell lines.
Methods. Five long chain N1,N12-diacyl lipopolyamines: N1,N12-[didecanoyl, dimyristoyl, dimyristoleoyl,
distearoyl and dioleoyl]-1,12-diamino-4,9-diazadodecane were synthesized from the naturally occurring
polyamine spermine. Their abilities to condense DNA and to form nanoparticles were characterized.
Transfection efficiencies were studied in FEK4 primary skin cells and in an immortalized cancer cell line
(HtTA), and compared with N4,N9-regioisomers. Also, the abilities of these novel compounds to bind to
siRNA-forming nanoparticles were studied using a RiboGreen intercalation assay, and their abilities to
deliver fluorescein-tagged siRNA into cells were quantified and compared with TransIT-TKO.
Results. By incorporating two long aliphatic chains and varying their acylation position, length, and
oxidation state in a stepwise manner, we show efficient p and siRNA formulation and delivery to primary
skin and cancer cell lines. Although two C14 chains (both saturated or both mono-cis-unsaturated) were
efficient transfecting agents, they were highly toxic.
Conclusions. N1,N12-Dioleoyl spermine efficiently binds to and delivers pDNA and siRNA with high cell
viability even in a primary skin cell line. It is a novel, efficient non-viral vector in the presence of serum.
KEY WORDS: gene delivery; primary skin cells; siRNA delivery; transfection; TransIT-TKO.
INTRODUCTION
only circumvent the drawbacks of viral vectors (limited
payloads, immune responses), but they are also designed to
Poly-nucleic acid delivery has many potential applica- have the advantages of simplicity of use based upon DNA
tions in medicine, and formulations to reach the goal of and siRNA anion titration and ease of large-scale production.
intracellular protein levels at therapeutic concentrations are There is a clear need for designed “intelligent materials” (24)
moving even more towards utilising non-viral gene therapy to achieve this nanoparticle self-assembly.
(NVGT) (1). We have recently reported on the relevant
The essential requirements for gene delivery are the
literature in this fast-moving and competitive research area transport of DNA through the cell membrane and ultimately
(2,3). Our non-viral gene delivery systems are centered upon to the nucleus. A first step in gene formulation is DNA
formulations of poly-nucleic acids complexed with synthetic condensation through masking the negative charges of the
cationic lipids and lipopolyamines and composed of two phosphate backbone by titration-forming nanoparticles. The
long-carbon chains (or a steroid) (4) covalently bound to a design of an efficient formula for the delivery of genetic
polyamine, e.g. spermine (1,12-diamino-4,9-diazadodecane) material requires a detailed understanding of the barriers that
(5,6). Structure–activity studies are required for non-viral hinder this process (20). These barriers include complex
vectors in siRNA delivery, as there is no immediate correla- formation between the DNA and the lipopolyamine leading
tion between the efficiency of a vector used for both DNA to DNA nanoparticle (lipoplex) formation by electrostatic
(7,8) and siRNA delivery (9–23). Our non-viral vectors not charge neutralisation (at least to 90%) and overall packing as
condensed DNA. Then, they are transported across cell-
1 Department of Pharmacy and Pharmacology, University of Bath,
membranes, either by adsorptive endocytosis or mediated by
cations, both routes leading to internalization of the lipoplex
(25). The site of action of siRNA is in the cytosol, so there is
no requirement to enter the nucleus (as for pDNA), but the
Bath, BA2 7AY, UK.
2 To whom correspondence should be addressed. (e-mail: prsisb@
bath.ac.uk)
ABBREVIATIONS: EGFP, enhanced green fluo-rescent protein;
siRNA lipoplex must efficiently afford protection from the
EMEM, Earle’s minimal essential medium; EthBr, ethidium bromide;
FCS, foetal calf serum; HRMS, high-resolution mass spectrometry;
high enzyme activity of RNase, which otherwise results in fast
hydrolysis with no chance of a therapeutic endpoint.
MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide;
Our aims are to design and develop efficient, non-toxic,
non-viral vectors for in vitro and possible in vivo applications,
using our novel spermine conjugates, based on change to the
NPC, nuclear pore complex; NVGT, non-viral gene therapy;
pEGFP, plasmid enhanced green fluorescent protein; t-BOC, t-
butoxycarbonyl.
#
17
0724-8741/10/0100-0017/0
2008 Springer Science + Business Media, LLC

18
Ghonaim, Li and Blagbrough
type, length, position, and number of the hydrophobic make changes to chain length from C18 (oleoyl and stearoyl)
anchors (5,6). These cationic lipids (lipospermines) probably to C14 (myristoyl and myristoleoyl), and that is in itself
assist in the self-assembling of polycationic scaffolds, as well significant, we also change the position of the amine groups
as facilitating absorptive endocytosis and/or fusion with cell and therefore the charge separation. These novel lipopoly-
membranes. We have therefore designed a series of novel amines are designed to incorporate the charge distribution of
regioisomers of diacyl spermines in order to prepare lipoplex putrescine (1,4-diaminobutane), a moiety found in both
formulations (without any pre-preparation of liposomes) of spermine and spermidine. We report the synthesis and
the circular plasmid DNA encoding for enhanced green characterisation of pEGFP and fluorescent-tagged siRNA
fluorescent protein (pEGFP) and a fluorescein-tagged siRNA lipoplex nanoparticles, and we assess them by physicochem-
for non-viral gene delivery and transfection of target cells as ical techniques, e.g. ethidium bromide (EthBr) fluorescence
they are designed to form nanoparticles which will efficiently quenching (28), and DNA gel shift assays, target cell pEGFP
enter cells by endocytosis (Fig. 1).
transfection (without and with serum—for nuclease activity),
Herein we report our investigations of the formulation and toxicity (cell viability, 3-(4,5-dimethylthiazol-2-yl)-2,5-
and delivery of pEGFP, a 4.7 kbp plasmid, with a molecular diphenyltetrazolium bromide, MTT assay) (29,30), and
weight of about 3.1 MDa (given an average of 330 Da per siRNA delivery with four new synthesized lipospermine
nucleotide, 660 Da per bp (26), carrying 9,400 negative formulations in both primary and cancer cell lines, and
charges, and a 21-mer of double stranded siRNA (27)). For compare our results with those obtained with the non-
our non-viral vectors, we vary the length and position of the liposomal lipospermine Lipogen® (N4,N9-dioleoyl spermine)
two fatty chains covalently bound to spermine. Although we (5,6) and TransIT-TKO.
S
S
75 Å
S
S
cytoplasm
NPC
nucleus
DNA release
transcription
mRNA
nucleus
migration
cytoplasm
translation
protein
endosome escape
H+
engulfing
endosome
N NH2
H2N N
condensed DNA
nanoparticles
Therapeutic DNA
Lipopolyamine
Fig. 1. The process of NVGT by endocytosis and the barriers that must be crossed by toroidal
condensed DNA nanoparticles, illustrated for lipopolyamines in the formulation of lipoplexes
leading to DNA delivery and ultimately to the goal of intracellular protein synthesis.

pDNA and siRNA SAR Delivery Studies
19
MATERIALS AND METHODS
Materials
(MgSO4) and concentrated. Column chromatography of the
residue (CH2Cl2:MeOH; 120:1 to 100:1 to 50:1; v/v) gave the
desired diamide-dicarbamate as an oil (2.5 g, 85%). Rf=0.9
(CH2Cl2:MeOH:conc. NH3 aq; 50:10:1; v/v/v). 1H NMR, CDCl3:
Chemicals, including spermine, acyl chlorides, reagents, 1.48 (s, 18H, 2× NCOOC(CH3)3); 1.71–1.82 (m, 8H, H2, H6,
solvents, and buffers, were routinely purchased from Sigma- H7, H11); 3.15–3.22 (m, 12H, H1, H3, H5, H8, H10, H12); 8.0
Aldrich (Gillingham, UK) except where indicated, and cell cul- (br, N1H, N12H). 13C NMR, CDCl3: 25.8 (C6, C7); 27.1 (C2,
ture materials were from Life Technologies (Paisley, Scotland). C11); 28.2 (CH3); 35.8 (C1, C12); 42.8 (C3, C10); 46.8 (C5, C8);
80.1 (OC(CH3)3); 116.0 (NCOCF3, J=295); 157.0 (N4-CO, N9-
CO); 157.8 (N1-CO, N12-CO, J=46). MS, FAB+ found 594.6,
General Details
(17%, M++1), 56.8 (100%, NH2CH2CH=CH2), C24H40F6N4O6
requires (M+) 594. HRMS m/z, FAB+ found 595.2938, (M++Na),
Glassware, ninhydrin used for the detection of polyamines, C24H41F6N4O6 requires (M++1) 595.2925.
silica gel column and analytical chromatography were as
previously reported (31). All the synthesized lipopolyamines
were pure and homogenous on silica gel thin-layer chromatog-
raphy (CH2Cl2-MeOH-conc. aq. NH3 25:10:1, v/v/v) and were
characterized by 1H nuclear magnetic resonance (NMR) and
13C NMR spectroscopy recorded using a Varian Mercury 400
(operating at 400 MHz for 1H and 100.8 MHz for 13C)
spectrometer. Chemical shifts values are recorded in parts per
million (ppm) on the δ scale. Coupling constants (J) are absolute
values and recorded in Hz. 13C assignments were aided by 90°
and 135° DEPT pulse sequences, and other NMR assignments
follow from correlation spectroscopies; COSY, DEPT, HMBC,
HMQC spectra were all recorded using automated programmes.
All lipopolyamines showed satisfactory high- (HRMS) and low-
resolution mass spectrometric FAB data (reported in Da and
within 5 ppm).
N4,N9-Di(t-butoxycarbonyl)-1,12-diamino-4,9-diazadodecane
Potassium carbonate (295 mg, 2.1 mmol) was added to N1,
N12-ditrifluoroacetyl-N4,N9-di(t-butoxycarbonyl)-1,12-diamino-
4,9-diazadodecane (2.5 g, 4.2 mmol) in MeOH (10 mL) and
water (600 μL), and the reaction mixture was heated under
reflux (5 h), cooled to 20°C and then evaporated to dryness in
vacuo. Water (30 mL) was added to the residue, which was
extracted with CHCl3 (3×25 mL). The combined organic phase
was washed with water (15 mL), dried (MgSO4) and
concentrated. The aqueous layers were further basified with
ice-cold 5% NaOH (10 mL) and extracted with CHCl3 (4×
40 mL). After a water wash (25 mL), the combined organic
extracts were evaporated to dryness in vacuo. Column
chromatography of the oily residue (CH2Cl2:MeOH:conc. aq.
NH3; 200:10:1 to 100:10:1 to 50:10:1; v/v/v) gave the desired
dicarbamate as an oil (1.0 g, 59%). Rf=0.4 (CH2Cl2:MeOH:
N4,N9-Dioleoyl-1,12-diamino-4,9-diazadodecane 1
1
conc. aq. NH3; 50:10:1; v/v/v). H NMR, CDCl3: 1.40 (s, 18H,
Prepared via N1,N12-ditrifluoroacetyl-N4,N9-dioleoyl-1,
12-diamino-4,9-diazadodecane (32), the target lipopolyamine
displayed HRMS: found (m/z [M+H]+) 731.7115, C46H91N4O2
requires 731.7142 (Δ3.7 ppm), consistent with the data
previously reported (5,6).
2× NCOOC(CH3)3); 1.50–1.70 (m, 8H, H2, H6, H7, H11); 2.0
(br, 4H, NH2); 2.63 (s, 4H, H1, H12) 3.10–3.22 (m, 8H, H3, H5,
H8, H10). 13C NMR, CDCl3: 25.8 (C6, C7); 28.5 (CH3); 32.0
(C2, C11); 39.2 (C1, C12); 44.2 (C3, C10); 46.6 (C5, C8); 79.8
(NCOOC); 149.3 (NCO). MS, FAB+ found 402.7 (8%, M++1),
56.8 (100%, NH2CH2CH=CH2), C20H42N4O4 requires (M+)
402. HRMS m/z, FAB+ found 403.3275, (M++1), C20H43N4O4
requires (M++1) 403.3279.
N4,N9-Dimyristoleoyl-1,12-diamino-4,9-diazadodecane 2
Prepared via N1,N12-ditrifluoroacetyl-N4,N9-dimyristoleoyl-
1,12-diamino-4,9-diazadodecane (706 mg, 87%, Rf=0.9 CH2Cl2:
MeOH, 5:1, v/v) displayed MS, FAB+ found 811.2 (100%, m/z
[M+H]+), C42H73F6N4O4 requires 811, the target lipopolyamine
displayed HRMS: found (m/z [M+H]+) 619.5872, C38H75N4O2
requires 619.5890 (Δ2.9 ppm).
N1,N12-Dioleoyl-N4,N9-di(t-butoxycarbonyl)-1,
12-diamino-4,9-diazadodecane
N4,N9-Di(t-butoxycarbonyl)-1,12-diamino-4,9-diazadodecane
(500 mg, 1.2 mmol) was dissolved in pyridine (10 mL) under
anhydrous nitrogen, and oleoyl chloride (0.9 mL, 2.7 mmol)
was added. The reaction was continued for 16 h, the solution
was then concentrated in vacuo. Purification of the crude
compound was by column chromatography (CH2Cl2:MeOH;
N1,N12-Ditrifluoroacetyl-N4,N9-di(t-butoxycarbonyl)-1,
12-diamino-4,9-diazadodecane
Ethyl trifluoroacetate (1.3 mL, 10.9 mmol, 2.2 eq.) in MeOH 100:1 to 50:1; v/v) to give an oil (420 mg, 36%). 1H NMR,
(10 mL) was added dropwise to spermine (1.0 g, 4.9 mmol) in CDCl3: 0.83 (t, 6H, H18′, J=8); 1.12–1.36 (m, 58H, H4′–H7′,
MeOH in 0°C under anhydrous nitrogen; the reaction was stirred H12′–H17′, 2× NCOOC(CH3)3); 1.41 (m, overlap, 8H, H6,
at 20°C for 24 h to form N1,N12-ditrifluoroacetyl-1,12-diamino- H7, H3′); 1.58 (m, 4H, H2, H11); 1.96 (m, 8H, H8′, H11′);
4,9-diazadodecane, which was taken into the next step without 2.14 (t, 4H, H2′, J=7); 3.16 (m, 12H, H1, H3, H5, H8, H10,
purification. Triethylamine (1.0 mL, 7.4 mmol, 1.5 eq.) followed H12); 5.29 (m, 4H, H9′, H10′); 6.75 (br, NH). 13C NMR,
by BOC-ON (2.7 g, 10.9 mmol, 2.2 eq.) were added to the CDCl3: 14.1 (C18′); 28.4 (NCOOC(CH3)3); 22.7–31.9 (C2,
diprotected spermine solution. The resulting yellow solution was C6, C7, C11, C3′–C8′, C11′–C17′); 35.3 (C2′); 36.8 (C1, C12);
stirred at 20°C for 24 h then evaporated to dryness in vacuo and 43.2 (C3, C10); 46.6 (C5, C8); 79.7 (NCOOC); 129.8 (C9′,
the residue dissolved in EtOAc (25 mL). The organic phase was C10′); 156.3 (NCO); 173.2 (C1′). MS, FAB+ found 931.1 (4%,
washed with 5% aq. NaOH (4×25 mL), water (2×25 mL), dried M++1), 57.1 (100%, NH2CH2CH=CH2), 832.1 (M-BOC),

20
Ghonaim, Li and Blagbrough
C56H106N4O6 requires (M+) 930. HRMS m/z, FAB+ found 1.59 (m, 8H, H6, H7, H3′); 1.61 (m, 4H, H2, H11); 2.15 (t,
953.8081, (M++Na), C56H107N4O6 requires (M++1) 931.8185. 4H, H2′, J=7); 2.28 (m, 8H, H3, H5, H8, H10); 3.18 (t, 4H,
H1, H12, J=8); 6.71 (br, NH). 13C NMR, CDCl3: 14.1 (C14′);
N1,N12-Distearoyl-N4,N9-di(t-butoxycarbonyl)-1,
22.6 (C13′); 25.2–27.7 (C2, C6, C7, C11, C3′); 28.4 (2×
NCOOC(CH3)3); 29.0–31.8 (C4′–C12′); 35.3 (C2′); 36.9 (C1,
C12); 43.3 (C3, C10); 46.7 (C5, C8); 79.7 (NCOOC); 156.3
(NCO); 173.3 (C1′). MS, FAB+ found 823.0 (14%, M++1),
56.9 (100%, NH2CH2CH=CH2), C48H94N4O6 requires (M+)
822. HRMS m/z, FAB+ found 845.7095, (M++Na),
12-diamino-4,9-diazadodecane
N4,N9-Di(t-butoxycarbonyl)-1,12-diamino-4,9-diazadodecane
(500 mg, 1.2 mmol) was dissolved in pyridine (10 mL)
under anhydrous nitrogen, and stearoyl chloride (0.9 mL,
2.7 mmol) was added. The reaction was continued for 16 h;
the solution was then concentrated in vacuo. Purification of
C
48H94N4O6Na requires (M++Na) 845.7066.
the crude compound was by column chromatography N1,N12-Didecanoyl-N4,N9-di(t-butoxycarbonyl)-1,
(CH2Cl2:MeOH; 100:1 to 50:1; v/v) to give an oil (840 mg, 12-diamino-4,9-diazadodecane
1
73%). H NMR, CDCl3: 0.96 (t, 6H, H18′, J=8); 1.29–1.33
(m, 56H, H4′–H17′); 1.40 (s, 18H, 2× NCOOC(CH3)3); 1.55–
N4,N9-Di(t-butoxycarbonyl)-1,12-diamino-4,9-diazadodecane
1.70 (m, 12H, H2, H6, H7, H11, H3′); 2.18 (m, 4H, H2′); 2.31 (500 mg, 1.2 mmol) was dissolved in pyridine (10 mL)
(m, 8H, H3, H5, H8, H10); 3.18 (m, 4H, H1, H12); 6.78 (br, under anhydrous nitrogen, and decanoyl chloride (0.9 mL,
NH). 13C NMR, CDCl3: 14.1 (C18′); 22.8–27.7 (C2, C6, C7, 2.7 mmol) was added. The reaction was continued for 16 h;
C11, C3′, C17′); 28.5 (OCH3); 29.0–31.9 (C4′–C16′); 35.8 the solution was then concentrated in vacuo. Purification of
(C2′); 37.2 (C1, C12); 43.2 (C3, C10); 46.7 (C5, C8); 79.8 the crude compound was by column chromatography
(NCOOC); 156.3 (NCO); 173.2 (C1′). MS, FAB+ found (CH2Cl2:MeOH; 100:1 to 50:1; v/v) to give an oil (644 mg,
1
935.0 (8%, M++1), 56.8 (100%, NH2CH2CH=CH2), 73%). H NMR, CDCl3: 0.83 (t, 6H, H10′, J=7); 1.21–1.36
C
56H110N4O6 requires (M+) 934. HRMS m/z, FAB+ found (m, 28H, H6, H7, H4′–H9′); 1.41 (s, 18H, 2× NCOOC
935.8498, (M++1), C56H111N4O6 requires (M++1) 935.8516.
(CH3)3); 1.51–1.59 (m, 8H, H2, H11, H3′); 2.12 (t, 4H, H2′,
J=8); 3.42 (m, 12H, H1, H3, H5, H8, H10, H12); 7.25 (br,
NH). 13C NMR, CDCl3: 14.0 (C10′); 22.6 (C9′); 25.0 (C6,
C7); 25.7 (C3′); 27.5 (C2, C11); 28.3 (CH3); 29.0–29.4 (C4′–
C7′); 31.7 (C8′); 35.3 (C1, C12); 36.8 (C2′); 43.2 (C3, C10);
46.6 (C5, C8); 77.0 (NCOOC); 156.3 (NCO); 173.4 (C1′).
MS, FAB+ found 711.4 (32%, M++1), 57.0 (100%,
NH2CH2CH=CH2), 611.4 (38%, M-BOC), C40H78N4O6
requires (M+) 710. HRMS m/z, FAB+ found 733.5820,
(M++Na), C40H78N4O6Na requires (M++Na) 733.5994.
N1,N12-Dimyristoleoyl-N4,N9-di(t-butoxycarbonyl)-1,
12-diamino-4,9-diazadodecane
Myristoleoic acid (0.7 mL, 2.9 mmol), HOBt (70 mg,
0.6 mmol) and DCC (587 mg, 4.3 mmol) were added in the
solvent of N4,N9-di(t-butoxycarbonyl)-1,12-diamino-4,9-
diazadodecane (520 mg, 1.3 mmol) in CH2Cl2 (10 mL). The
reaction was continued for 16 h; the solution was then filtered
to remove the urea and concentrated in vacuo. Purification of
the crude compound was by column chromatography (CH2Cl2: N1,N12-Dioleoyl-1,12-diamino-4,9-diazadodecane 3
1
MeOH; 100:1 to 50:1; v/v) to give an oil (800 mg, 76%). H
NMR, CDCl3: 0.85 (t, 6H, H14′, J=8); 1.13–1.35 (m, 42H, H4′–
To a stirring solution of N1,N12-dioleoyl-N4,N9-di(t-
H7′, H12′, H13′, 2× NCOOC(CH3)3); 1.40–1.57 (m, overlap, butoxycarbonyl)-1,12-diamino-4,9-diazadodecane (420 mg,
12H, H2, H6, H7, H11, H3′); 1.96 (m, 8H, H8′, H11′); 2.12 (t, 0.5 mmol) in CH2Cl2 (180 mL), under nitrogen, at 20°C was
4H, H2′, J=6); 3.15 (m, 12H, H1, H3, H5, H8, H10, H12); 5.28 added TFA (20 mL). After 2 h, the solution was concentrated
(m, 4H, H9′, H10′); 6.73 (br, NH). 13C NMR, CDCl3: 13.9 in vacuo to yield the title compound as a white semi-solid
(C14′); 28.3 (2× NCOOC(CH3)3); 22.2–31.8 (C2, C6, C7, C11, (320 mg, 97%, di-TFA salt). 1H NMR, d6-DMSO: 0.96 (t, 6H,
C3′–C8′, C11′–C13′); 35.3 (C2′); 36.8 (C1, C12); 43.2 (C3, C10); H18′, J=7); 1.22–1.64 (m, 52H, H2, H6, H7, H11, H3′–H7′,
46.6 (C5, C8); 79.6 (NCOOC); 129.8 (C9′, C10′); 156.3 (NCO); H12′–H17′); 2.00–2.55 (m, 14H, C2′, H8′, H11′, NH); 3.00 (m,
173.2 (C1′). MS, FAB+ found 819.3 (16%, M++1), 56.9 (100%, 12H, H1, H3, H5, H8, H10, H12); 5.32 (m, 4H, H9′, H10′);
NH2CH2CH=CH2), 719.2 (36%, M-BOC), C48H90N4O6 8.62 (br, NH). 13C NMR, d6-DMSO: 14.1 (C18′); 22.8–36.1
requires (M+) 818. HRMS m/z, FAB+ found 841.6768, (C1, C2, C6, C7, C11, C12, C2′–C8′, C11′–C17′); 45.2 (C3,
(M++Na), C48H90N4O6Na requires (M++Na) 841.6753.
C10); 47.3 (C5, C8); 130.0 (C9′, C10′); 180.0 (C1′). MS, FAB+
found 730.9 (100%, M++1), C46H90N4O2 requires (M+) 730.
N1,N12-Dimyristoyl-N4,N9-di(t-butoxycarbonyl)-1,
12-diamino-4,9-diazadodecane
N1,N12-Distearoyl-1,12-diamino-4,9-diazadodecane 4
To a stirring solution of N1,N12-distearoyl-N4,N9-di(t-
N4,N9-Di(t-butoxycarbonyl)-1,12-diamino-4,9-diazadodecane
(500 mg, 1.2 mmol) was dissolved in pyridine (10 mL) butoxycarbonyl)-1,12-diamino-4,9-diazadodecane (840 mg,
under anhydrous nitrogen, and myristoyl chloride (0.9 mL, 0.9 mmol) in CH2Cl2 (180 mL), under nitrogen, at 20°C was
2.7 mmol) was added. The reaction was continued for 16 h; added TFA (20 mL). After 2 h, the solution was concentrated in
the solution was then concentrated in vacuo. Purification of vacuo to yield the title compound as a white solid (620 mg,
1
the crude compound was by column chromatography 94%, di-TFA salt), m.p. 125–127°C. H NMR, d6-DMSO: 0.85
(CH2Cl2:MeOH; 100:1 to 50:1; v/v) to give an oil (824 mg, (t, 6H, H18′, J=8); 1.23–2.07 (m, 72H, H2, H6, H7, H11, H2′–
1
81%). H NMR, CDCl3: 0.85 (t, 6H, H14′, J=7); 1.22–1.37 H17′); 3.10 (m, 12H, H1, H3, H5, H8, H10, H12); 7.96, 8.61 (br,
(m, 40H, H4′–H13′); 1.43 (s, 18H, 2× NCOOC(CH3)3); 1.55– NH). 13C NMR, d6-DMSO: 13.7 (C18′); 21.9–29.1 (C2, C6, C7,

pDNA and siRNA SAR Delivery Studies
21
C11, C3′–C15′, C17′); 31.2–35.4 (C1, C12, C2′, C16′); 47.0 (C3, coli JM 109 bacterial strain (Promega) and grown on in Luria–
C10); 49.1 (C5, C8); 172.0 (C1′). MS, FAB+ found 734.9 (100%, Bertani (LB) broth and analysed as previously reported (31).
M++1), C46H94N4O2 requires (M+) 734. HRMS m/z, FAB+
found 735.7441, (M++1), C46H95N4O2 requires (M++1) 735.7450.
DNA Condensation (Ethidium Bromide Fluorescence
Quenching Assay)
N1,N12-Dimyristoleoyl-1,12-diamino-4,9-diazadodecane 5
Each concentration of the pEGFP stock solutions
(approximately 1 μg/μL, 1 mL) was determined spectroscopi-
To a stirring solution of N1,N12-dimyristoleoyl-N4,N9-di
cally (Milton Roy Spectronic 601 spectrometer, 1 cm path
length, 3 mL cuvette) (26) with 6 μg (approximately 6 μL)
pDNA diluted to 3 mL with buffer (20 mM NaCl, 2 mM
HEPES, 10 μM EDTA, pH 7.4) and condensed with lip-
opolyamine aliquots as previously reported (31).
(t-butoxycarbonyl)-1,12-diamino-4,9-diazadodecane (800 mg,
1.0 mmol) in CH2Cl2 (180 mL), under nitrogen, at 20°C was
added TFA (20 mL). After 2 h, the solution was concentrated
in vacuo to yield the title compound as a white semi-solid
(600 mg, 99%, di-TFA salt). 1H NMR, d6-DMSO: 0.96 (t, 6H,
H14′, J=8); 1.38–2.04 (m, 46H, H2, H6, H7, H11, H11′–H13′,
H3′–H8′, N4H, N9H); 2.24 (t, 4H, H2′, J=7); 2.98 (m, 8H, H3,
H5, H8, H10); 3.32 (m, 4H, H1, H12); 5.40 (m, 4H, H9′,
H10′); 8.20 (br, NH). 13C NMR, d6-DMSO: 13.2 (C14′); 22.0–
35.9 (C2′–C8′, C11′–13′, C1, C2, C6, C7, C11, C12); 45.1 (C3,
C10); 47.3 (C5, C8); 129.0 (C9′, C10′); 176.0 (C1′). MS, FAB+
found 618.9 (100%, M++1), C38H74N4O2 requires (M+) 618.
HRMS m/z, FAB+ found 619.5885, (M++1), C38H75N4O2
requires (M++1) 619.5892.
Gel Electrophoresis
Each sample of plasmid DNA (0.5 μg), either free or
complexed with different lipopolyamine concentrations, was
analyzed by gel electrophoresis for about 1 h under 75 V/cm,
through an agarose gel (1%) containing EthBr (1 μg/mL) in
Tris–acetate–EDTA 1× (40 mM Tris–acetate and 1 mM
EDTA) buffer as previously reported (31).
N1,N12-Dimyristoyl-1,12-diamino-4,9-diazadodecane 6
Lipoplex Particle Size
To a stirring solution of N1,N12-dimyristoyl-N4,N9-di(t-
butoxycarbonyl)-1,12-diamino-4,9-diazadodecane (824 mg,
1.0 mmol) in CH2Cl2 (180 mL), under nitrogen, at 20°C was
added TFA (20 mL). After 2 h, the solution was concentrated
in vacuo to yield the title compound as a white solid (620 mg,
99%, di-TFA salt), m.p. 92–94°C. 1H NMR, d6-DMSO: 0.91 (t,
6H, H14′, J=7); 1.22–1.37 (m, 40H, H4′–H13′); 1.55–1.59 (m,
8H, H6, H7, H3′); 1.61 (m, 4H, H2, H11); 2.15 (t, 4H, H2′, J=
8); 2.28 (m, 8H, H3, H5, H8, H10); 2.40 (br, NH); 3.18 (m, 4H,
H1, H12); 8.62 (br, NH). 13C NMR, d6-DMSO: 13.9 (C14′);
22.6 (C13′); 25.2–27.7 (C2, C6, C7, C11, C3′); 29.0–31.8 (C4′–
C12′); 35.3 (C2′); 36.9 (C1, C12); 43.3 (C3, C10); 46.7 (C5, C8);
172.9 (C1′). MS, FAB+ found 623.8 (100%, M++1),
C38H78N4O2 requires (M+) 622. HRMS m/z, FAB+ found
623.6199, (M++1), C38H79N4O2 requires (M++1) 623.6198.
The average particle size for each lipoplex formed (at a
highly efficient charge ratio of transfection) was determined
using a NanoSight LM10 (NanoSight Ltd, Salisbury, UK). All
measurements were carried out on lipoplexes prepared from
1 μg pEGFP or 25 pmol of siRNA (fluorescein labelled RNAi
Delivery Controls (Label IT®, Mirus) in HEPES buffer
(0.2 mL) at pH 7.4 as previously reported (31).
ζ-Potential Measurements
The ζ-potential measurements for the lipoplexes formed
at an efficient charge ratio of transfection, after mixing with a
vortex mixer, were determined using a Delsa™Nano Zeta
Potential (Beckman Coulter, Buckinghamshire, UK) as
previously reported (31).
N1,N12-Didecanoyl-1,12-diamino-4,9-diazadodecane 7
Cell Culture and Transfection Experiments
To a stirring solution of N1,N12-didecanoyl-N4,N9-di(t-
butoxycarbonyl)-1,12-diamino-4,9-diazadodecane (644 mg,
0.9 mmol) in CH2Cl2 (180 mL), under nitrogen, at 20°C was
added TFA (20 mL). After 2 h, the solution was concentrated
in vacuo to yield the title compound as a white solid (455 mg,
Two cell lines were used in the transfection experiments:
human primary skin fibroblast cells FEK4 (33,34) derived
from a foreskin explant and human cervix carcinoma, HeLa
derivative and transformed cell line (HtTA) (35,36). The
HtTA cells being stably transfected with a tetracycline-
controlled transactivator (tTA) consisting of the tet repressor
fused with the activating domain of virion protein 16 of the
herpes simplex virus (HSV). Cells were cultured as previously
reported (31).
1
98%, di-TFA salt), m.p. 67–69°C. H NMR, d6-DMSO: 0.83
(t, 6H, H10′, J=8); 1.21–1.59 (m, 36H, H2, H6, H7, H11, H3′–
H9′); 2.02–2.14 (m, 14H, H3, H5, H8, H10, H2′, N4H, N9H);
3.42 (m, 4H, H1, H12); 7.25 (br, NH). 13C NMR, CDCl3: 14.0
(C10′); 22.6–29.4 (C2, C6, C7, C11, C3′–C7′, C9′); 31.7–36.8
(C1, C12, C2′, C8′); 43.2 (C3, C10); 46.6 (C5, C8); 173.4
(C1′). MS, FAB+ found 511.6 (100%, M++1), C30H62N4O2
requires (M+) 510. HRMS m/z, FAB+ found 511.4948,
(M++1), C30H63N4O2 requires (M++1) 511.4946.
DNA Transfection Experiments in the Absence or Presence
of Serum
For the transfection (gene delivery) and the resultant
gene activity (transfection efficiency), FEK4 and HtTA cells
were seeded at 50,000 cells/well in 12-well plates in Earle’s
Minimal Essential Medium (EMEM, 2 mL) containing foetal
calf serum (FCS) for 24 h to reach a plate confluency of 50–
60% on the day of transfection. Then the media were
Amplification and Purification of Plasmid DNA (pEGFP)
DNA plasmid (26) encoding-enhanced green fluorescent
protein (pEGFP) (Clontech) was transformed into Escherichia

22
Ghonaim, Li and Blagbrough
replaced by 0.4 mL Opti-MEM (serum free media, Gibco MTT solution (10 μL, 5 mg/mL) to reach a final concen-
BRL) for 4 h at 37°C in 5% CO2. The lipoplex was prepared tration of 0.5 mg/mL (29). Then the plates were incubated
by mixing pEGFP (1 μg in 50 μL) with the cationic lipopoly- for a further 4 h at 37°C in an atmosphere of 5% v/v CO2.
amine in Opti-MEM (typically 10 μg in 50 μL) according to The percent viability was detected as previously reported
the N/P charge ratio at 20°C for 30 min and then incubated (31,37).
with the cells (final volume of 0.5 mL) for 4 h at 37°C in 5%
CO2 in Opti-MEM (in the absence of serum). Then the cells
were washed and cultured for a further 44 h in full growth
medium at 37°C in 5% CO2 before the assay.
Confocal Microscopy Visualization
Following the siRNA delivery protocols, cells were
seeded on a sterile cover-slip at the bottom of each well.
After 48 h, media were aspirated and cells were fixed with
freshly prepared 4% formaldehyde solution in PBS (1 mL/
well) for 15 min at 37°C. After formaldehyde fixing, cells
adhering to cover-slips were labelled with labelling solution
according to the manufacturer’s protocol. Labelling solution
contained both Alexa Fluor 594 wheat germ agglutinin for
cell membrane labelling, λex=591 nm and λem=618 nm, and
For transfection in the presence of serum, as in the above
procedure, but on the day of transfection, the media were
replaced by fresh EMEM (0.4 mL) containing FCS. The
lipoplex was prepared as above and then incubated with the
cells for 4 h at 37°C in 5% CO2 in full growth medium (in
the presence of serum). Then the cells were washed and
cultured for a further 44 h in full growth medium at 37°C in
5% CO2 before the assay.
Levels of enhanced green fluorescent protein (EGFP
positive cells) in the transfected cells were detected and
corrected for background fluorescence of the control cells
using a fluorescence-activated cell sorting (FACS) machine
(Becton Dickinson FACS Vantage dual Laser Instrument,
argon ion laser 488 nm) as previously reported (31).
Hoechst 33342 for nuclei labelling, λex=350 nm and λem
=
461 nm, mixed in one solution purchased from Invitrogen
(Image-iT live plasma membrane and nuclear labelling kit),
cell labelling with Alexa Fluor WGA (5 μg/mL) and Hoechst
33342 (2 μL, 2 μΜ). After that, labelled cells were mounted
using mounting liquid (20 μL, Mowiol, Merck) and left for
16 h. Then, mounted cover-slips were viewed on a confocal
laser scanning microscope (LSM510META, Zeiss, Jena, Ger-
many) under the ×60 oil immersion objective, with filters: red
siRNA Binding (RiboGreen Intercalation Assay)
RiboGreen solution (Invitrogen, 50 μL diluted 1 to 20)
λex=543 nm and λem=560–615 nm, blue λex=405 nm and λem=
was added to each well of a 96-well plate (opaque bottom) 420–480 nm, and green λex=488 nm and λem=505–530 nm.
containing free non-targeting siRNA1 (50 ng) (Dharmacon,
Thermo Fisher Scientific Biosciences) or complexed with
lipopolyamines at different ratios in TE buffer (50 μL, 10 mM
RESULTS AND DISCUSSION
Tris–HCl, 1 mM EDTA, pH 7.5, in DEPC-treated water)
using FLUOstar Optima Microplate Reader (BMG-LAB-
Synthesis of Lipospermines
TECH), λex=480 nm, λem=520 nm as previously reported
(31,37).
Spermine was used as the starting material for the
synthesis of the seven desired lipopolyamines: N4,N9-
dioleoyl spermine 1, N4,N9-dimyristoleoyl spermine 2, N1,
N12-dioleoyl spermine 3, N1,N12-distearoyl spermine 4, N1,
N12-dimyristoleoyl spermine 5, N1,N12-dimyristoyl spermine
6, and N1,N12-didecanoyl spermine 7 (Fig. 2).
siRNA Delivery Experiments
For the siRNA delivery, we use fluorescein-labelled RNAi
Delivery Controls (Label IT®, Mirus). FEK4 and HtTA cells
were seeded at 50,000 cells/well in 12-well plates in EMEM
(2 mL) containing FCS for 24 h to reach a plate confluency of
50–60% on the day of transfection. Then the media were
replaced by 437.5 μL fresh media. The lipoplex was prepared
by mixing siRNA (12.5 pmol in 12.5 μL) with the different
amounts of the cationic lipopolyamines in Opti-MEM (typi-
cally 2–20 μg in 50 μL) at 20°C for 30 min and then incubated
with the cells (final volume of 0.5 mL) for 4 h at 37°C in 5%
CO2 in full growth medium. Then the media were replaced by
2 mL of fresh media for 44 h at 37°C in 5% CO2 before the
assay. Levels of fluorescein-labelled siRNA in the transfected
cells were detected as previously reported (31,37).
The tetra-amine was chemo-selectively protected on both
the primary amino functional groups with ethyl trifluoroacetate
(2.2 eq.) in methanol (38,39). Fatty acyl chloride (oleoyl or
myristoleoyl) (2.2 eq.) and triethylamine (2.5 eq.) were then
added to the diprotected spermine solution in CH2Cl2 and
methanol (1:1, v/v). Deprotection by treatment with potassium
carbonate in aq. methanol (40,41) and flash column chroma-
tography afforded the two target N4,N9-lipospermines,
homogenous on silica gel thin-layer chromatography e.g. N4,
N9-dioleoyl spermine 1 Rf=0.44 (CH2Cl2:MeOH:conc. aq. NH3
25:10:1 v/v/v) (4–6). Alternatively, t-BOC protection of the two
secondary amines was achieved with BOC-ON, which yielded
the diamide-dicarbamate (39), and potassium carbonate
catalyzed deprotection of the primary amines afforded N4,N9-
di-t-BOC spermine (39,42). Acylation of the two primary
amines in pyridine, followed by t-BOC deprotection using
In Vitro Cytotoxicity (MTT) Assay
FEK4 and HtTA cells were seeded in 96-well plates at TFA in CH2Cl2 (1:9 v/v) afforded the five target N1,N12-
8,000 cells/well and incubated for 24 h at 37°C in 5% CO2. lipospermines as their di-trifluoroacetate salts. The structures
Lipoplexes, e.g. N1,N12-dioleoyl spermine complexed with of these novel target N1,N12-lipospermines were established by
pEGFP or siRNA, were added in the same way as in the 1H- (at 400 MHz) and 13C- (at 100.8 MHz) NMR spectroscopy
1
transfection protocol. After incubation for 44 h, the media and confirmed by HRMS. The characteristic H NMR signals
were replaced with fresh media (90 μL) and sterile filtered included δH 0.8–0.9 methyl protons, 1.2–1.3 many methylene

pDNA and siRNA SAR Delivery Studies
23
H
N
NH2
H2N
N
H
i
H
N
NHCOCF3
F3COCHN
F3COCHN
N
H
ii
iii
R
N
NHCOCF3
N
R
O
O
NH2
N
N
H2N
NH2
N
H2N
N
O
O
1
2
O
O
NH2
N
H2N
N
O
O
ii, iv
O
O
H
N
H
N
NH
NH
N
H
N
HN
HN
H
O
O
4
6
3
O
O
H
N
H
N
NH
NH
N
H
HN
HN
O
N
H
O
5
O
H
NH
N
N
H
HN
O
7
Fig. 2. Synthetic scheme for spermine-based cationic lipids. Reagents: i ethyl trifluoroacetate, 24 h at 20°C; ii fatty
acyl chloride, fatty acid (with DCC and cat. HOBt) or BOC-ON (R=oleoyl, myristoleoyl, or t-BOC), pyridine,
16 h at 20°C; iii potassium carbonate in aq. methanol, reflux (5 h); iv TFA:CH2Cl2 (1:9 v/v), 2 h at 20°C.
chain protons, and 5.3–5.4 ppm for the alkene protons (where
required). Relative to spermine (free base in CDCl3) where δC
C3 47.6 and C5 49.8 (Δ2.2 ppm), whose chemical shifts are
predicted to decrease by 2 ppm on being α- to a protonated
ammonium ion due to a slight decrease in deshielding of the
carbon atoms in the vicinity of the ammonium ion, C3 was
found to resonate typically at 43.2–45.2 and C5 at 46.6–47.3
(Δ2.1–3.4 ppm) where C3 was predicted (in CDCl3) to be at
45.6 and C5 47.7 (Δ2.1 ppm) (43). Other spermine resonances
N4,N9-Dioleoyl spermine
N1,N12-Dioleoyl spermine
N1,N12-Distearoyl spermine
N1,N12-Dimyristoleoyl spermine
N1,N12-Dimyristoyl spermine
100
80
60
40
20
0
0
1
2
3
4
5
6
N/P charge ratio
Fig. 4. Agarose gel (1%) assay of fluorescent EthBr intercalated in
pEGFP, lipoplex formation with N1,N12-dimyristoyl spermine 6 at N/
P ratios 2–12 progressively retards gel permeation.
Fig. 3. Plot of EthBr fluorescence quenching assay of pEGFP
complexed with different lipospermines.

24
Ghonaim, Li and Blagbrough
Table I. Particle Size (mean SD, n=9) pEGFP and siRNA Complexes with Selected Lipopolyamines
Lipospermine
DNA-lipoplex N/P charge ratio
DNA-lipoplex diameter (nm)
siRNA-lipoplex diameter (nm)
N4,N9-dioleoyl spermine 1
2.5
3
30
12
12
150 (12)
170 (30)
210 (35)
250 (49)
190 (39)
110 (12)
150 (28)
N1,N12-dioleoyl spermine 3
N1,N12-distearoyl spermine 4
N1,N12-dimyristoleoyl spermine 5
N1,N12-dimyristoyl spermine 6
170 (34)
are found as expected: C1 36–38, C2 24–26, C6 23–24 ppm. complexes ranged from 170 nm (N1,N12-dioleoyl spermine 3)
The characteristic 13C NMR signals also included 13.2–14.1 to 250 nm (N1,N12-dimyristoleoyl spermine 5), compared with
methyls, 21.9–36.8 ppm methylenes (a long alkyl chain typically N4,N9-dioleoyl spermine 1 pEGFP lipoplex nanoparticles of
starts 14.1, 22.8, 32.1, 29.5), 129.0–130.0 cis-alkenes C9′ and 150 nm (Table I). Two siRNA lipoplexes were measured:
C10′ (where required), and 172.0–180.0 ppm the amide 150 nm (N1,N12-dioleoyl spermine 3) and 170 nm (N1,N12-
carbonyl groups.
dimyristoleoyl spermine 5), compared with N4,N9-dioleoyl
spermine 1 siRNA lipoplex nanoparticles of 110 nm (Table I).
Two representative ζ-potential measurements for the
lipoplexes formed (at their optimum charge ratio of
DNA Condensation
In Fig. 3, we show the DNA condensation ability of the transfection with 3 μg/mL pEGFP) were measured with a
synthesized lipopolyamines in an EthBr fluorescence quench- Delsa™Nano Zeta Potential instrument showing N4,N9-
ing assay. The results show that N1,N12-dioleoyl spermine 3 dioleoyl spermine 1 +2.17 mV (at N/P=2.5) and N1,N12-
and N1,N12-dimyristoleoyl spermine 5 have the best DNA dioleoyl spermine 3 +3.69 mV (at N/P=3).
condensing ability, more than 90% EthBr fluorescence
quenching at N/P charge ratio 6, where DNA is defined as pEGFP Transfection and In Vitro
condensed (44), and N1,N12-dimyristoyl spermine 6 shows Cytotoxicity
75% fluorescence quenching, but N1,N12-distearoyl spermine
4 only achieves 55% at the same N/P ratio. Typical gel
In preliminary experiments, we readily established that
electrophoresis data (Fig. 4) show that N1,N12-dimyristoyl saturated lipid chains were more toxic than unsaturated ones,
spermine 6 was able to condense pEGFP DNA progressively, and that shorter chain analogues were more toxic than longer
as a result of neutralization of DNA phosphate negative ones. We therefore did not make any further detailed studies
charges on titration with the two ammonium positive charges of N1,N12-didecanoyl spermine 7. The transduction of EGFP
on the lipopolyamine, to its optimised N/P charge ratio of into a primary skin cell line FEK4 and a cancer cell line (HeLa-
transfection (N/P=12) at which it inhibited the electrophoretic derived HtTA) was investigated in presence and absence of
mobility of plasmid DNA.
serum (Fig. 5), and we found that there was no significant
difference in the transfection levels in the presence of serum
for N4,N9-dioleoyl spermine 1 (Lipogen) and N1,N12-dioleoyl
spermine 3. Encouraged by these positive results, all our
Lipoplex Particle Size and ζ-Potential
The particle size measurements were carried out on the subsequent transfection experiments are performed in the
lipoplexes at a concentration which gave a high transfection presence of serum. The optimum concentrations (in a final
efficiency. Particle size characterization by laser diffraction volume of 0.5 mL) and their corresponding N/P charge ratios
showed that the nanoscale particle size of the formed DNA for transfection were experimentally determined by using
FEK4 Transfection
FEK4 Viability
HtTA Transfection
HtTA Viability
100
80
60
40
20
0
100
80
60
40
20
0
FEK4 Serum Free
HtTA Serum Free
FEK4 Serum
HtTA Serum
100
80
60
40
20
0
Lipogen
Oleoyl
Stearoyl
Myristoleoyl Myristoyl
Fig. 6. Lipofection and cytotoxicity effects of pEGFP (1 μg) com-
plexed with Lipogen (N4,N9-dioleoyl spermine 1) (2.77 μg), Oleoyl
(N1,N12-dioleoyl spermine 3) (3.32 μg), Stearoyl (N1,N12-distearoyl
spermine 4) (33.36 μg), Myristoleoyl (N1,N12-dimyristoleoyl spermine
5) (11.24 μg) and Myristoyl (N1,N12-dimyristoyl spermine 6)
(11.31 μg) on the primary skin cell line FEK4 and the HeLa-derived
cancer cell line HtTA.
N4,N9-dioleoyl
spermine
N1,N12-dioleoyl
spermine
Fig. 5. Lipofection effects of pEGFP (1 μg) complexed with Lipogen
(N4,N9-dioleoyl spermine 1 (2.77 μg) and N1,N12-dioleoyl spermine 3
(3.32 μg) on the primary skin cell line FEK4 and the HeLa-derived
cancer cell line HtTA in the absence and presence of serum.

pDNA and siRNA SAR Delivery Studies
25
Oleoyl(1,12)FEK4
Myristoleoyl(1,12)FEK4
Oleoyl(1,12)HtTA
Myristoleoyl(1,12)HtTA
100
80
60
40
20
0
ascending N/P ratios of lipospermines from 2, 4, 6, etc. until we
reached around 80% transfection and there was not a further
step-up in transfection efficiency at the next highest N/P ratio.
Therefore, the optimum amounts (and corresponding N/P
charge ratios) for transfection with pEGFP (1 μg in 50 μL)
were found to be N4,N9-dioleoyl spermine 1 (Lipogen,
2.77 μg, N/P=2.5), N1,N12-dioleoyl spermine 3 (3.32 μg, N/P=
3), N1,N12-distearoyl spermine 4 (33.36 μg, N/P=30.0), N1,N12-
dimyristoleoyl spermine 5 (11.24 μg, N/P=12.0), and N1,N12-
dimyristoyl spermine 6 (11.31 μg N/P=12.0). The results
(Fig. 6, histograms) indicate that N1,N12-dioleoyl spermine 3,
which typically shows 70% and 65% transfection ability, and
2
4
6
8
10
12
µg
Fig. 8. Cellular delivery to the primary skin cell line FEK4 and the
cancer cell line HtTA of fluorescein-tagged siRNA (12.5 pmol)
complexed with N1,N12-dioleoyl spermine
3 -
and N1,N12
70% and 60% cell viability for FEK4 and HtTA cells, dimyristoleoyl spermine 5 at increasing concentrations.
respectively (Fig. 6, lines), is the best of all the tested N1,N12-
diacyl spermine vectors and can be compared with its N4,N9-
higher), but also with a view to minimal toxicity (high cell
viability), minimum effective concentrations. These optimum
delivery concentrations were experimentally determined by
using ascending amounts of lipopolyamines (Fig. 8). Flow
cytometric FACS analysis of both cell lines, gated for live cells
(Fig. 9), after 48 h transduction with fluorescein-tagged siRNA
complexed with N1,N12-dioleoyl spermine 3, clearly shows high
siRNA delivery efficiency. The siFection of FEK4 primary cells
(hard to transfect) and a cancer cell line (HtTA) were compared
with a market leader TransIT-TKO (Mirus). The siRNA
delivery results (Fig. 10, histograms) show that N4,N9-dioleoyl
spermine 1 (Lipogen, 80% FEK4 and 87% HtTA), N1,N12-
dioleoyl spermine 3 (77% FEK4 and 85% HtTA), N4,N9-
dimyristoleoyl spermine 2 (84% FEK4 and 91% HtTA) and N1,
N12-dimyristoleoyl spermine 5 (81% FEK4 and 78% HtTA) are
regioisomer (Lipogen 1). N1,N12-Distearoyl spermine 4, the
saturated analogue of N1,N12-dioleoyl spermine 3, whilst
maintaining the same (C-18) chain length, was not a good
transfecting agent and was far too toxic. Shortening the acyl
chains to C14 gave mixed pEGFP transfection results with
certainly no improvement over C-18, and the toxicity remained
high, with the saturated analogue N1,N12-dimyristoyl spermine
6 more toxic than the unsaturated N1,N12-dimyristoleoyl
spermine 5.
siRNA Delivery and In Vitro Cytotoxicity
The RiboGreen intercalation assay was performed on two
selected non-viral delivery vectors from the N1,N12-diacyl
spermine series and compared with the corresponding N4,N9-
regioisomers. These vectors were both from the unsaturated
series as they showed the highest cell viability in the pEGFP
transfection assays. The siRNA binding results from the
RiboGreen intercalation assay show that efficient fluorescence
quenching occurs by N/P charge ratio 4 for all four
lipopolyamines (Fig. 7). However, the N4,N9-regioisomer
pattern is preferred, as N4,N9-dioleoyl spermine (Lipogen) 1
and N4,N9-dimyristoleoyl spermine 2 achieved 80% fluorescence
quenching by N/P=2.5.
The transduction of fluorescein-tagged siRNA into a
primary skin cell line FEK4 and a cancer cell line (HeLa-derived
HtTA) was investigated in the presence of serum (RNase).
We aimed for cellular delivery efficiency about 80% (or
N4,N9-dimyristoleoyl spermine
N1,N12-dimyristoleoyl spermine
N4,N9-dioleoyl spermine
100
N1,N12-dioleoyl spermine
80
60
40
20
0
0
1
2
3
4
Fig. 9. Gated FACS analysis of FEK4 (above) and HtTA (below)
N/P charge ratio
after 48 h transfection of fluorescein-tagged siRNA complexed with
Fig. 7. Plot of RiboGreen intercalation assay of siRNA complexed N1,N12-dioleoyl spermine 3 (12 μg): shaded in violet untransduced
with different lipospermines.
cells, shaded in white fluorescein-positive cells.

26
Ghonaim, Li and Blagbrough
FEK4 Transfection
FEK4 Viability
HtTA Transfection
HtTA Viability
successfully be able to give the desired protein through
transcription and translation (Fig. 1). As siRNA has its site of
action in the cytosol, escape from the endosome is therefore a
key parameter, without the need to traffic to the nucleus.
The position, length and degree of unsaturation play an
important role in their design and formulation where the lipid
moiety must be considered in shape (volume) and substituent
pattern, as well as the polyamine moiety and its pKa values (4–
6,45–47). The first key step in this gene formulation is DNA
condensation into nanoparticles by masking the negative
charges of the phosphate backbone (44). This titration with a
lipopolyamine causes alleviation of charge repulsion between
remote phosphates along the DNA helix leading to collapse
into a more compact structure that facilitates cell entry. The
pEGFP condensation results (Fig. 3), investigated using the
EthBr fluorescence quenching assay, revealed that of our
synthetic lipopolyamines N1,N12-dioleoyl spermine 3 and N1,
N12-dimyristoleoyl spermine 5 were able to condense pDNA to
less than 10% EthBr fluorescence at N/P charge ratio 6, while
N1,N12-dimyristoyl spermine 6 and N1,N12-distearoyl spermine
4 show only 75% and 55% condensation, respectively, at that
N/P ratio. The lipid chains make a significant contribution
to the pDNA condensation; the differences are clear (Fig. 3)
with respect to position and unsaturation, where pDNA
condensation efficiency follows: saturated—1,12<unsaturated—
1,12<unsaturated—4,9 regioisomers. Thus, N4,N9-dioleoyl
spermine 1 (Lipogen) is the best pDNA condensing agent at
N/P charge ratios 4 and 5. The RiboGreen siRNA intercalation
assay results showed efficient fluorescence quenching for the
four chosen lipopolyamines (Fig. 7), the N4,N9-regioisomeric
pattern being preferred, achieving 80% fluorescence quenching
by N/P=2.5, which only occurred at N/P=4 for the N1,N12-
analogues. Particle size of the final pDNA formulation is also
an important factor in improving gene delivery (48,49). Particle
size results (Table I) for pDNA nanoparticle lipoplexes were
around 205 nm, and for two siRNA lipoplexes were 150 and
170 nm. On the relationship between particle size and
transfection efficiency, there are no definite limits to the
nanoparticle size that is suitable for transfection (50).
Nanoparticles have relatively higher intracellular uptake than
microparticles (22). Also, on the nanoscale, smaller-size
100
100
80
60
40
20
0
80
60
40
20
0
TransIT
Oleoyl(4,9)
Oleoyl(1,12) Myristoleoyl(4,9)Myristoleoyl(1,12)
Fig. 10. Cellular delivery and cytotoxicity effects of siRNA
(12.5 pmol) complexed with TransIT-TKO (4 μL), Oleoyl (4,9)
(Lipogen, N4,N9-dioleoyl spermine 1) (8.0 μg), Oleoyl (1,12) (N1,
N
12-dioleoyl spermine 3) (12.0 μg), Myristoleoyl (4,9) (N4,N9-
dimyristoleoyl spermine 2) (10.0 μg) and Myristoleoyl (1,12) (N1,
N
12-dimyristoleoyl spermine 5) (12.0 μg) on the primary skin cell line
FEK4 and the HeLa-derived cancer cell line HtTA.
just lower in efficiency for both cell lines than TransIT-TKO
(91% FEK4 and 93% HtTA). The cell viability (MTT assay)
results (Fig. 10, lines) show that N4,N9-dioleoyl spermine 1
(Lipogen) and N1,N12-dioleoyl spermine 3 have better viability
of FEK4 cells (76% and 65% respectively) over HtTA cells
(46% and 47% respectively); next is N1,N12-dimyristoleoyl
spermine 5. Whilst N4,N9-dimyristoleoyl spermine 2 achieves
high siFection results, it is too toxic. However, N1,N12-dioleoyl
spermine 3 and N1,N12-dimyristoleoyl spermine 5 are efficient
new non-viral siRNA delivery vectors (Figs. 8, 9, 10 and 11),
displaying satisfactory cell viability.
In this study, we have investigated changes to the position,
length and degree of unsaturation of the symmetrical diacyl fatty
chain formulations of lipospermine on pDNA and siRNA
binding and cellular delivery. The lipid moiety in our cationic
lipids interacts with the phospholipid bilayer of the cell
membrane, and that facilitates cell entry, either in crossing the
membrane bilayer and/or in helping to weaken the endosomal
bilayer and thereby aiding escape into the cytosol (Fig. 1). DNA
must now traffic to the nucleus and cross the nuclear
membrane, which is thought to occur through the nuclear pore
complex (NPC) or by direct association with the chromatin
during mitosis. After nuclear entry, the payload DNA should
Fig. 11. The FEK4 primary cell line (left) and HtTA cell line (right) showing delivery of a
fluorescein-siRNA tag inside the cells, delivered with N1,N12-dioleoyl spermine 3. The tag
fluoresces green, the nuclei fluoresce blue from the Hoechst 33342, and the cell lipid bilayers
fluoresce red from the Alexa Fluor 594 (LSM510META, under the ×60 oil immersion objective).

pDNA and siRNA SAR Delivery Studies
27
polyplexes are more able to enter cells and thereby increase the amines from the (significant base-weakening) transient
efficiency of transfection (51). charges on the secondary amines, and the primary amines
From the combination of both the transfection and viability may therefore be assumed to have pKa values typical of
results (Fig. 6), we found that N1,N12-dioleoyl spermine 3 primary amines, 10.6–10.9, i.e. N4,N9-diacyl analogues will
(transfection efficiency FEK4 70%, HtTA 65% and cell carry two positive charges at physiological pH (7.4). Moving
viability FEK4 70%, HtTA 60%) was a better vector than the diacyl chains to the N1,N12-position means that these
the other three N1,N12-analogues based on the two factors of analogues will still carry two positive charges at pH=7.4, but
transfection efficiency and cell viability. It was superior to the their pKa values are likely to follow those of putrescine (1,4-
saturated analogue N1,N12-distearoyl spermine 4 and the diaminobutane) (10.8 and 9.4 or 10.7 and 9.2) (59,60), and we
shorter analogues, unsaturated N1,N12-dimyristoleoyl assume that there will not be any significant effect from the γ-
spermine 5 and saturated N1,N12-dimyristoyl spermine 6 and amide functional groups. Thus, these novel conjugates are
comparable to its N4,N9-regiosisomer N4,N9-dioleoyl spermine dibasic for N/P charge ratio calculations, and their efficient
1 (Lipogen), which displayed transfection efficiency FEK4 charge neutralisation by phosphate anion titration will be a
85%, HtTA 78% and cell viability FEK4 83%, HtTA 65%. function of steric more than electronic effects when compared
Based on these DNA transfection and cell viability results, we with the 12 atoms (10 C and 2 N) separating the two positive
investigated the change in the position and length of the charges in the N4,N9-series. We conclude from the above
symmetrical diacyl fatty chain formulation of lipospermine on results that N1,N12-dioleoyl spermine 3 is the best of N1,N12-
siRNA binding (Fig. 7) and cellular delivery (Figs. 8, 9, 10 and substituted series for both DNA and siRNA delivery. It is
11). Following from a similar combination of both the siRNA comparable with its commercially available N4,N9-
delivery and the cell viability results, we found that the two regioisomer N4,N9-dioleoyl spermine 1 (Lipogen) for DNA
dioleoyl lipospermine regioisomers 1 and 3 are of comparable transfection efficiency where Lipogen 1 achieves typically
activity, only slightly below that achieved with TransIT and 81% transfection and N1,N12-dioleoyl spermine 3 achieves
more efficient than the two dimyristoleoyl regioisomeric 67% (both with good cell viability), and for siRNA delivery,
analogues 2 and 5 (Fig. 10).
typically 83% with Lipogen 1 and 81% with our novel
Whilst in the earlier part of these studies, we delivered analogue 3 (with similar cell viability).
pDNA and measured transfection by assaying the produced
EGFP, in the latter part we have presented preliminary evidence
for efficient siRNA delivery using a fluorescently tagged probe,
a ds RNA 21-mer designed not to interfere with normal cell
protein biosynthesis (31). Barriers to efficient siRNA delivery
include excretion through lysosomes, serum nuclease (RNase)
digestion, non-specific distribution (in vivo), and tissue barriers
that must be crossed, e.g. into the cell and out of the endosome.
To support further our gated FACS evidence (Fig. 9) that there
is siRNA delivery to the cells, we have carried out confocal
laser scanning microscopy studies on both cell lines (Fig. 11).
Using one labelling solution (Invitrogen) containing both
Alexa Fluor 594 wheat germ agglutinin (5 μg/mL) for cell
membrane labelling, and Hoechst 33342 (2 μΜ) for nuclei
labelling, we have shown (Fig. 11) successful delivery of the
(green) fluorescent tag to both FEK4 and HtTA cell-lines with
N1,N12-dioleoyl spermine 3. We have also previously reported
a successful protein (β-actin) knockdown experiment in HEK
293 cells (30% as assayed by Western blot protein analysis)
where the siRNA was delivered with N4,N9-dioleoyl spermine
1 (Lipogen) (52).
The DNA compacting properties of polyamines (espe-
cially spermine) are well-known, hence the use of spermine as
the cationic part in several synthetic DNA carriers. Clement
and his research group recently reported the use lipophos-
phoramidate as the lipidic part of lipospermines for gene
delivery (53). Due to their high net charge at physiological
pH, polyamines effectively charge neutralize the phospho-
diester backbone of DNA (54). Many pKa measurements
have been made on spermidine, spermine, and their deriva-
tives (55,56), which all indicate their high number of positive
charges at physiological pH. Spermidine typically shows pKa
values of 11.1, 10.0 and 8.6 or 10.9, 9.8 and 8.4 (57) and
spermine pK1 10.9, pK2 10.1, pK3 8.9 and pK4 8.1 (43) and
10.9, 10.0, 8.8 and 8.4 (58). However, in preparing the N4,N9-
regioisomeric analogues, we have separated the two primary
ACKNOWLEDGEMENTS
We acknowledge the financial support of an Egyptian
Government studentship to H.M.G. We are grateful to Prof.
R.M. Tyrrell for the FEK4 and HtTA cell lines and to Dr. C.
Pourzand (both University of Bath) for helpful advice in cell
biology. We thank NanoSight Ltd (Salisbury, UK) for the
NanoSight LM10 and Beckman Coulter (High Wycombe,
UK) for the Delsa™Nano.
REFERENCES
1. W. J. Li, and F. C. Szoka. Lipid-based nanoparticles for nucleic
acid delivery. Pharm. Res. 24:438–449 (2007). doi:10.1007/s11095-
006-9180-5.
2. I. S. Blagbrough and H. M. Ghonaim. Polyamines and their
conjugates for gene and siRNA delivery. In G. Dandrifosse (Ed),
Biological aspects of biogenic amines, polyamines and conjugates,
Research Signpost, India, 26 pp. (2008) In Press.
3. I. S. Blagbrough and C. Zara. Animal models for target diseases
in gene therapy—using DNA and siRNA delivery strategies.
Pharm. Res. 26:In Press (2009).
4. N. Adjimatera, T. Kral, M. Hof, and I. S. Blagbrough. Lipopoly-
amine-mediated single nanoparticle formation of calf thymus
DNA analyzed by fluorescence correlation spectroscopy. Pharm.
Res. 23:1564–1573 (2006). doi:10.1007/s11095-006-0278-6.
5. O. A. A. Ahmed, N. Adjimatera, C. Pourzand, and I. S.
Blagbrough. N4,N9-Dioleoyl spermine is a novel nonviral
lipopolyamine vector for plasmid DNA formulation. Pharm.
Res. 22:972–980 (2005). doi:10.1007/s11095-005-4592-1.
6. O. A. A. Ahmed, C. Pourzand, and I. S. Blagbrough. Varying the
unsaturation in N4,N9-dioctadecanoyl spermines: Nonviral
lipopolyamine vectors for more efficient plasmid DNA
formulation. Pharm. Res. 23:31–40 (2006). doi:10.1007/s11095-
005-8717-3.
7. J. S. Remy, C. Sirlin, P. Vierling, and J. P. Behr. Gene-transfer
with a series of lipophilic DNA-binding molecules. Bioconjug.
Chem. 5:647–654 (1994). doi:10.1021/bc00030a021.

28
8. J. P. Behr, B. Demeneix, J. P. Loeffler, and J. P. Mutul. Efficient
Ghonaim, Li and Blagbrough
29. T. Mosmann. Rapid colorimetric assay for cellular growth and
survival: application to proliferation and cytotoxicity assays. J.
Immunol. Methods. 65:55–63 (1983). doi:10.1016/0022-1759(83)
90303-4.
gene-transfer into mammalian primary endocrine-cells with
lipopolyamine-coated DNA. Proc. Natl. Acad. Sci. U.S.A.
86:6982–6986 (1989). doi:10.1073/pnas.86.18.6982.
9. M. A. Behlke. Progress towards in vivo use of siRNAs. Mol. 30. D. Fischer, T. Bieber, Y. X. Li, H. P. Elsasser, and T. Kissel. A
Ther. 13:644–670 (2006). doi:10.1016/j.ymthe.2006.01.001.
10. A. Doody, and D. Putnam. RNA-interference effectors and their
delivery. Crit. Rev. Ther. Drug Carrier Syst. 23:137–164 (2006).
11. Y. Ikeda, and K. Taira. Ligand-targeted delivery of therapeutic
novel non-viral vector for DNA delivery based on low molecular
weight, branched polyethylenimine: Effect of molecular weight
on transfection efficiency and cytotoxicity. Pharm. Res. 16:1273–
1279 (1999). doi:10.1023/A:1014861900478.
siRNA. Pharm. Res. 23:1631–1640 (2006). doi:10.1007/s11095- 31. H. M. Ghonaim, S. Li, and I. S. Blagbrough. Very long chain N4,
006-9001-x.
N9-diacyl spermines: Non-viral lipopolyamine vectors for
efficient plasmid DNA and siRNA delivery. Pharm. Res. 25:In
Press (2008).
12. C. X. Li, A. Parker, E. Menocal, S. L. Xiang, L. Borodyansky,
and J. H. Fruehauf. Delivery of RNA interference. Cell Cycle.
5:2103–2109 (2006).
13. Z. Racz, and P. Hamar. Can siRNA technology provide the tools
for gene therapy of the future? Curr. Med. Chem. 13:2299–2307
(2006). doi:10.2174/092986706777935186.
32. G. Ronsin, C. Perrin, P. Guedat, A. Kremer, P. Camilleri, and
A. J. Kirby. Novel spermine-based cationic gemini surfactants for
gene delivery. Chem. Commun. 2234–2235 (2001). doi:10.1039/
b105936j
14. K. A. Howard, and J. Kjems. Polycation-based nanoparticle
delivery for improved RNA interference therapeutics. Exp.
Opin. Biol. Therapy. 7:1811–1822 (2007). doi:10.1517/14712598.
7.12.1811.
33. R. M. Tyrrell, and M. Pidoux. Quantitative differences in host-
cell reactivation of ultraviolet-damaged virus in human-skin
fibroblasts and epidermal-keratinocytes cultured from the same
foreskin biopsy. Cancer Res. 46:2665–2669 (1986).
15. H. Takahashi, D. Letourneur, and D. W. Grainger. Delivery of 34. J. L. Zhong, A. Yiakouvaki, P. Holley, R. M. Tyrrell, and C.
large biopharmaceuticals from cardiovascular stents: A review.
Biomacromolecules. 8:3281–3293 (2007). doi:10.1021/bm700540p.
16. S. B. Zhang, B. Zhao, H. M. Jiang, B. Wang, and B. C. Ma.
Cationic lipids and polymers mediated vectors for delivery of
Pourzand. Susceptibility of skin cells to UVA-induced necrotic
cell death reflects the intracellular level of labile iron. J. Invest.
Dermatol. 123:771–780 (2004). doi:10.1111/j.0022-202X.2004.
23419.x.
siRNA. J. Control. Release. 123:1–10 (2007). doi:10.1016/j.jcon- 35. M. Gossen, and H. Bujard. Tight control of gene-expression in
rel.2007.07.016.
mammalian-cells by tetracycline-responsive promoters. Proc.
Natl. Acad. Sci. U.S.A. 89:5547–5551 (1992). doi:10.1073/
pnas.89.12.5547.
17. A. R. De Fougerolles. Delivery vehicles for small interfering
RNA in vivo. Hum. Gene Ther. 19:125–132 (2008). doi:10.1089/
hum.2008.928.
18. K. F. Pirollo, and E. H. Chang. Targeted delivery of small
interfering RNA: Approaching effective cancer therapies. Can-
cer Res. 68:1247–1250 (2008). doi:10.1158/0008-5472.CAN-07-
5810.
36. E. Kvam, V. Hejmadi, S. Ryter, C. Pourzand, and R. M. Tyrrell.
Heme oxygenase activity causes transient hypersensitivity to
oxidative ultraviolet A radiation that depends on release of iron
from heme. Free Radic. Biol. Med. 28:1191–1196 (2000).
doi:10.1016/S0891-5849(00)00205-7.
19. S. A. Cryan, R. Donohue, B. J. Ravo, R. Darcy, and C. M.
O’Driscoll. Cationic cyclodextrin amphiphiles as gene delivery
vectors. J. Drug. Del. Sci. Technol. 14:57–62 (2004).
20. M. Walsh, M. Tangney, M. O’Neill, J. Larkin, D. Soden, S.
McKenna, R. Darcy, G. O’Sullivan, and C. O’Driscoll. Evalua-
37. M. K. Soltan, H. M. Ghonaim, M. El Sadek, M. Abou Kull, L.
Abd El-aziz, and I. S. Blagbrough. Design and synthesis of N4,
N9-disubstituted spermines for non-viral siRNA delivery—
structure–activity relationship studies of siFection efficiency
versus toxicity. Pharm. Res. 26:In Press (2009).
tion of cellular uptake and gene transfer efficiency of pegylated 38. A. J. Geall, and I. S. Blagbrough. Homologation of polyamines
poly-L-lysine compacted DNA: Implications for cancer gene
therapy. Mol. Pharm. 3:644–653 (2006). doi:10.1021/mp0600034.
21. S. A. Cryan, and C. M. O’Driscoll. Mechanistic studies on
in the rapid synthesis of lipospermine conjugates and related
lipoplexes. Tetrahedron. 56:2449–2460 (2000). doi:10.1016/S0040-
4020(99)01082-0.
nonviral gene delivery to the intestine using in vitro differenti- 39. M. C. O’Sullivan, Q. B. Zhou, Z. L. Li, T. B. Durham, D.
ated cell culture models and an in vivo rat intestinal loop. Pharm.
Res. 20:569–575 (2003). doi:10.1023/A:1023286413666.
22. S. C. De Smedt, J. Demeester, and W. E. Hennink. Cationic
polymer based gene delivery systems. Pharm. Res. 17:113–126
(2000). doi:10.1023/A:1007548826495.
23. B. G. Stubbe, S. C. De Smedt, and J. Demeester. “Programmed
polymeric devices” for pulsed drug delivery. Pharm. Res. 21:1732–
1740 (2004). doi:10.1023/B:PHAM.0000045223.45400.01.
24. K. Remaut, N. N. Sanders, B. G. De Geest, K. Braeckmans, J.
Demeester, and S. C. De Smedt. Nucleic acid delivery: Where
Rattendi, S. Lane, and C. J. Bacchi. Polyamine derivatives as
inhibitors of trypanothione reductase and assessment of their
trypanocidal activities. Bioorg. Med. Chem. 5:2145–2155 (1997).
doi:10.1016/S0968-0896(97)00157-0.
40. R. J. Bergeron, and J. S. McManis. Total synthesis of ( )-15-
deoxyspergualin. J. Org. Chem. 52:1700–1703 (1987).
doi:10.1021/jo00385a010.
41. R. J. Bergeron, and J. S. McManis. Reagents for the stepwise
functionalization of spermine. J. Org. Chem. 53:3108–3111
(1988). doi:10.1021/jo00248a037.
material sciences and bio-sciences meet. Mater. Sci. Eng. R-Rep. 42. T. Blessing, J. S. Remy, and J. P. Behr. Template oligomerization
58:117–161 (2007). doi:10.1016/j.mser.2007.06.001.
25. J. Y. Zhou, J. W. Yockman, S. W. Kim, and S. E. Kern.
Intracellular kinetics of non-viral gene delivery using polyethy-
lenimine carriers. Pharm. Res. 24:1079–1087 (2007). doi:10.1007/
s11095-006-9229-5.
of DNA-bound cations produces calibrated nanometric particles.
J. Am. Chem. Soc. 120:8519–8520 (1998). doi:10.1021/ja980620o.
43. A. J. Geall, R. J. Taylor, M. E. Earll, M. A. W. Eaton, and I. S.
Blagbrough. Synthesis of cholesteryl polyamine carbamates: pKa
studies and condensation of calf thymus DNA. Bioconjug. Chem.
11:314–326 (2000). doi:10.1021/bc990115w.
26. P. L. Felgner, Y. Barenholz, J. P. Behr, S. H. Cheng, P. Cullis, L.
Huang, J. A. Jessee, L. Seymour, F. Szoka, A. R. Thierry, E. 44. V. A. Bloomfield. DNA condensation by multivalent cations.
Wagner, and G. Wu. Nomenclature for synthetic gene delivery
systems. Hum. Gene Ther. 8:511–512 (1997). doi:10.1089/
hum.1997.8.5-511.
Biopolymers. 44:269–282 (1997). doi:10.1002/(SICI)1097-0282
(1997)44:3<269::AID-BIP6>3.0.CO;2-T.
45. A. J. Geall, M. A. W. Eaton, T. Baker, C. Catterall, and I. S.
Blagbrough. The regiochemical distribution of positive charges
along cholesterol polyamine carbamates plays significant roles in
modulating DNA binding affinity and lipofection. FEBS Lett.
459:337–342 (1999). doi:10.1016/S0014-5793(99)01262-4.
27. R. M. Schiffelers, M. C. Woodle, and P. Scaria. Pharmaceutical
prospects for RNA interference. Pharm. Res. 21:1–7 (2004).
doi:10.1023/B:PHAM.0000012145.49054.6c.
28. A. J. Geall, and I. S. Blagbrough. Rapid and sensitive ethidium
bromide fluorescence quenching assay of polyamine conjugate- 46. I. S. Blagbrough, D. Al-Hadithi, and A. J. Geall. Cheno-, urso-
DNA interactions for the analysis of lipoplex formation in gene
therapy. J. Pharm. Biomed. Anal. 22:849–859 (2000). doi:10.1016/
S0731-7085(00)00250-8.
and deoxycholic acid spermine conjugates: Relative binding
affinities for calf thymus DNA. Tetrahedron. 56:3439–3447
(2000). doi:10.1016/S0040-4020(00)00265-9.

pDNA and siRNA SAR Delivery Studies
29
47. I. S. Blagbrough, A. J. Geall, and A. P. Neal. Polyamines and
novel polyamine conjugates interact with DNA in ways that can
be exploited in non-viral gene therapy. Biochem. Soc. Trans.
31:397–406 (2003). doi:10.1042/BST0310397.
54. A. S. Snygg, M. Hung, and S. K. C. Elmroth. The poly-
amines spermidine and spermine retard the platination rate
of single-stranded DNA oligomers and plasmid. J. Inorg.
Biochem. 101:1153–1164 (2007). doi:10.1016/j.jinorgbio.
2007.04.017.
48. C.-F. Hung, T.-L. Hwang, C.-C. Chang, and J.-Y. Fang. Physico-
chemical characterization and gene transfection efficiency of 55. C. Frassineti, S. Ghelli, P. Gans, A. Sabatini, M. S. Moruzzi,
lipid emulsions with various co-emulsifiers. Int. J. Pharm.
289:197–208 (2005). doi:10.1016/j.ijpharm.2004.11.008.
49. D. G. Anderson, A. Akinc, N. Hossain, and R. Langer.
Structure/property studies of polymeric gene delivery using a
and A. Vacca. Nuclear magnetic resonance as a tool for
determining protonation constants of natural polyprotic bases
in solution. Anal. Biochem. 231:374–382 (1995). doi:10.1006/
abio.1995.9984.
library of poly(β-amino esters). Mol. Ther. 11:426–434 (2005). 56. Y. Takeda, K. Samejima, K. Nagano, M. Watanabe, H. Sugeta,
doi:10.1016/j.ymthe.2004.11.015.
and Y. Kyogoku. Determination of protonation sites in thermo-
spermine and in some other polyamines by 15N and 13C nuclear
magnetic resonance spectroscopy. Eur. J. Biochem. 130:383–389
(1983). doi:10.1111/j.1432-1033.1983.tb07164.x.
50. X. Gao, and L. Huang. Potentiation of cationic liposome-
mediated gene delivery by polycations. Biochemistry. 35:1027–
1036 (1996). doi:10.1021/bi952436a.
51. J. Panyam, and V. Labhasetwar. Biodegradable nanoparticles for 57. M. M. Kimberly, and J. H. Goldstein. Determination of pKa
drug and gene delivery to cells and tissue. Adv. Drug Deliv. Rev.
55:329–347 (2003). doi:10.1016/S0169-409X(02)00228-4.
52. D. McLaggan, N. Adjimatera, K. Sepcic, M. Jaspars, D. J.
values and total proton distribution pattern of spermidine by
carbon-13 nuclear magnetic resonance titrations. Anal. Chem.
53:789–793 (1981). doi:10.1021/ac00229a011.
MacEwan, I. S. Blagbrough, and R. H. Scott. Pore forming 58. E. Rowatt, and R. J. P. Williams. The effect of multivalent
polyalkylpyridinium salts from marine sponges versus synthetic
lipofection systems: distinct tools for intracellular delivery of
cDNA and siRNA. BMC Biotechnol. 6:6 (2006). doi:10.1186/
1472-6750-6-6.
ions on the inactivation of bacteriophage ϕX174 by lipopoly-
saccharide from Escherichia coli C. Biochem. J. 231:765–768
(1985).
59. J. F. Coetzee, and G. R. Padmanabhan. Properties of bases in
acetonitrile as solvent. IV. Proton acceptor power and homo-
conjugation of mono- and diamines. J. Am. Chem. Soc. 87:5005–
5010 (1965). doi:10.1021/ja00950a006.
53. F. Lamarche, M. Mevel, T. Montier, L. Burel-Deschamps, P.
Giamarchi, R. Tripier, P. Delepine, T. Le Gall, D. Cartier, P.
Lehn, P.-A. Jaffrees, and J.-C. Clement. Lipophosphoramidates
as lipidic part of lipospermines for gene delivery. Bioconjug. 60. A. Albert and E. P. Serjeant. In: The Determination of Ionization
Chem. 18:1575–1582 (2007). doi:10.1021/bc070070w. Constants. Chapman and Hall Ltd., London. pp. 91–95 (1971).