Hydrophobic oxime ethers: A versatile class of pDNA and siRNA transfection lipids

Article abstract of DOI:10.1002/cmdc.201100259

The manipulation of the cationic lipid structures to increase polynucleotide binding and delivery properties, while also minimizing associated cytotoxicity, has been a principal strategy for developing next-generation transfection agents. The polar (DNA binding) and hydrophobic domains of transfection lipids have been extensively studied; however, the linking domain comprising the substructure used to tether the polar and hydrophobic domains has attracted considerably less attention as an optimization variable. Here, we examine the use of an oxime ether as the linking domain. Hydrophobic oxime ethers were readily assembled via click chemistry by oximation of hydrophobic aldehydes using an aminooxy salt. A facile ligation reaction delivered the desired compounds with hydrophobic domain asymmetry. Using the MCF-7 breast cancer, H1792 lung cancer and PAR C10 salivary epithelial cell lines, our findings show that lipoplexes derived from oxime ether lipids transfect in the presence of serum at higher levels than commonly used liposome formulations, based on both luciferase and green fluorescent protein (GFP) assays. Given the biological compatibility of oxime ethers and their ease of formation, this functional group should find significant application as a linking domain in future designs of transfection vectors.

Full text of DOI:10.1002/cmdc.201100259

DOI: 10.1002/cmdc.201100259
Hydrophobic Oxime Ethers: A Versatile Class of pDNA and
siRNA Transfection Lipids
Souvik Biswas,[a] Ralph J. Knipp,[a] Laura E. Gordon,[b] Seshagiri R. Nandula,[c] Sven-Ulrik Gorr,[c]
Geoffrey J. Clark,[b] and Michael H. Nantz*[a]
The manipulation of the cationic lipid structures to increase
polynucleotide binding and delivery properties, while also min-
imizing associated cytotoxicity, has been a principal strategy
for developing next-generation transfection agents. The polar
(DNA binding) and hydrophobic domains of transfection lipids
have been extensively studied; however, the linking domain
comprising the substructure used to tether the polar and hy-
drophobic domains has attracted considerably less attention as
an optimization variable. Here, we examine the use of an
oxime ether as the linking domain. Hydrophobic oxime ethers
were readily assembled via click chemistry by oximation of hy-
drophobic aldehydes using an aminooxy salt. A facile ligation
reaction delivered the desired compounds with hydrophobic
domain asymmetry. Using the MCF-7 breast cancer, H1792
lung cancer and PAR C10 salivary epithelial cell lines, our find-
ings show that lipoplexes derived from oxime ether lipids
transfect in the presence of serum at higher levels than com-
monly used liposome formulations, based on both luciferase
and green fluorescent protein (GFP) assays. Given the biologi-
cal compatibility of oxime ethers and their ease of formation,
this functional group should find significant application as a
linking domain in future designs of transfection vectors.
Introduction
Cationic lipids and their derived liposomes have become the
most well-studied and widely used synthetic, nonviral gene de-
livery vehicles since Felgner et al.[1] first disclosed DOTMA-
mediated gene transfer in 1987.[2,3] Due to the major limita-
tions of viral vectors, such as associated immune responses,
limited polynucleotide carrying capacity and high cost,[4,5] cat-
ionic lipids remain an attractive alternative. The advantages of
low immunogenicity, the ability to transfect RNA or DNA of
nearly unlimited size,[6,7] and the relative ease of cationic lipid–
plasmid DNA (pDNA) or RNA complex (lipoplex) formulation[8]
continue to attract interest aimed at developing safer and
more efficient cationic lipids for use as transfection agents.[9]
Cationic lipid molecules, such as the prototypical dual chain
lipids DOTMA[1] and DOTAP[10] (Figure 1), contain a polar, posi-
tively charged (DNA binding) head group connected to a hy-
drophobic domain via a linking functionality. These three prin-
cipal structural components of cationic glycerol-type lipids
have been extensively studied in efforts to improve lipid-medi-
ated intracellular delivery of polynucleotides to mammalian
cells.[11] Many structure–activity relationships have been deter-
mined,[12,13] particularly for the hydrophobic domain. Indeed,
the hydrophobic domain’s structural variables of chain length,
degree of unsaturation, and domain asymmetry are among the
strongest contributors to transfection efficacy.[14] Fewer direct
structural comparisons of changes in the cationic lipid back-
bone, or linking domain, have been reported, with the most
well-known comparison being that of the diether DOTMA
versus the diester DOTAP.[10] The linking functionality, and to a
lesser extent the cationic head group, seem to be the principal
determinants of toxicity.[15–17] The linker determines conforma-
tional flexibility, degree of stability, and biodegradability.
Among the most studied chemical functionalities comprising
the linking domain of transfection lipids are the ether, ester,
ortho ester,[18,19] carbamate,[20] amide,[21,22] and phosphono[23]
moieties.
[a] S. Biswas, R. J. Knipp, Dr. M. H. Nantz
Department of Chemistry, University of Louisville
2320 S. Brook Street, Louisville, Kentucky 40292 (USA)
E-mail: michael.nantz@louisville.edu
[b] L. E. Gordon, Dr. G. J. Clark
School of Medicine, Brown Cancer Center, University of Louisville
529 S. Jackson Street, Louisville, Kentucky 40292 (USA)
[c] Dr. S. R. Nandula, Dr. S.-U. Gorr
School of Dentistry, University of Minnesota, Moos Health Sciences Tower
515 Delaware Street SE, Minneapolis, Minnesota 55455 (USA)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201100259.
Figure 1. Structural domains of common transfection lipids.
ChemMedChem 2011, 6, 2063 – 2069
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2063

M. H. Nantz et al.
MED
The use of oxime ether linkages for applications in chemical
biology and medicinal chemistry research has increased dra-
matically in the past several years as the benefits of chemose-
lectivity have become better understood. Chemoselective click
ligation[24] of an aldehyde or ketone carbonyl group with an
aminooxy counterpart to form an oxime ether linkage (i.e., oxi-
mation, Scheme 1) is the key step in the synthesis of numerous
major diastereomer has (E)-stereochemistry about each oxime
C=N bond (as depicted). The minor diastereomer has an (E)-
oxime ether as well as a (Z)-oxime ether linkage. The (E,E):(E,Z)
ratio was readily measured by integration of the well-separated
oximyl proton shifts in the 1H NMR spectra,[36] and the ratio
was found to be susceptible to a variety of factors including
exposure to mild acid and heat. The stereochemical integrity
of the oxime ether lipids is inconsequential in that isomeriza-
tion can be expected in vivo. Of greater interest, however, is
that the oximation approach used to prepare these com-
pounds delivers the lipids with intrinsic asymmetry in the hy-
drophobic domain. We have recently noted the beneficial in-
fluence of unsymmetrical hydrophobic domains in cationic-
lipid-mediated DNA transfection.[32] Consequently, we used the
oxime ether lipids as obtained directly in transfection experi-
ments. To probe further the influence of an unsymmetrical hy-
drophobic domain, we also prepared unsymmetrical analogue
4 (Scheme 1) by successively condensing 1 with dodecanal
and tetradecanal. Hydrophobic oxime ethers 2–5 were formu-
lated as cationic liposomes using equimolar quantities of co-
lipid dioleoylphosphatidylethanolamine (DOPE) for subsequent
studies.
Scheme 1. Oximation route to hydrophobic oxime ethers. Reagents and con-
ditions: a) RCHO, MeOH, RT, 4 h.
Size and charge characterization
Particle size and zeta potential measurements on the oxime
ether liposome suspensions revealed that the particles were
net positively charged with electrostatic potentials in the
range of 45–66 mV (Figure 2). Liposomes derived from bis-C18
lipid 5 were the largest, with particle diameters exceeding
200 nm, and the most highly charged. Not unexpectedly, the
liposomes derived from the shortest oxime ether lipids 2 and
4—i.e., lipids prepared using dodecanal—had the smallest
mean diameters, near 150 nm. The trend in particle size versus
particle charge correlates well for this series of lipids.
bioconjugates.[25–27] Oxime ether linkages are also used as a
mechanism for prodrug generation,[28] and are present in US
Food and Drug Administration (FDA)-approved pharmaceuti-
cals, such as fluvoxamine.[29] Although oximation has been
used to attach ligands to transfection lipids, such as the liga-
tion of carbohydrates,[30] and although an oxime-ether-based
assembly for siRNA delivery has been reported,[31] no studies
have examined the use of the robust yet biodegradable oxime
ether linkage as the backbone feature of cationic lipids for
DNA delivery.
Of significance as a predictor of transfection activity is the
net charge of the lipoplex formed on mixing a liposome sus-
pension with pDNA.[13,37] Complexation of the oxime ether/
DOPE liposomes with pDNA (pCMV-Luc) at different ammoni-
Our program in developing lipid-based gene delivery vec-
tors[32] and interest in click chemistry[33,34] inspired us to use ox-
imation as a facile approach to prepare low toxicity transfec-
tion-active oxime ethers (Scheme 1). Here, we report the physi-
cochemical characterization of this new class of lipids and in
vitro evaluation of a hydrophobic oxime ether panel in pDNA
and siRNA delivery experiments using epithelial cell lines, in-
cluding two human cancer cell lines.
Results and Discussion
Synthesis of hydrophobic oxime ethers
We prepared a panel of oxime ethers by reacting bis-
(aminooxy) ammonium salt 1[35] with hydrophobic aldehydes
dodecanal, tetradecanal, or (Z)-octadec-9-enal (oleyl aldehyde)
in methanol at room temperature to afford oxime ethers 2, 3
and 5, respectively (Scheme 1). In each case, the lipid was ob-
tained as a mixture containing two stereoisomers (diastereo-
meric ratio ranging from approximately 2.3:1 to 5.3:1). The
Figure 2. Zeta potential (&) and mean particle size (&) measurements for
oxime ether: DOPE liposomes (0.1 mg cationic lipid per mL H2O).
2064
www.chemmedchem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2011, 6, 2063 – 2069

pDNA and siRNA Transfection Lipids
um nitrogen/DNA phosphate ratios (N:P ratio) gave the corre-
sponding lipoplexes. Zeta potential measurements of the
oxime-ether-derived lipoplexes at N:P ratios of 3, 5 and 7
showed high positive values in the range of 31–53 mV
(Figure 3). As can be expected at these electrostatic potentials,
Figure 3. Lipoplex zeta potential measurements. Lipoplexes were formulated
at 0.1 mg cationic lipid per mL H2O with pDNA (pCMV-Luc) at different N:P
charge ratios: N/P=3 (&); NP=5 (&); N/P=7 (<?Ql>). Results are ex-
pressed as the mean values of n=3 measurements.
minimal lipoplex aggregation ensued—the oxime ether/DNA
lipoplex suspensions were stable at concentrations up to
0.33 mg oxime ether lipid per mL beyond one week. Particle
size measurements of the lipoplexes (data not shown) indicat-
ed that the mean effective diameter for all N:P formulations re-
mained under 250 nm. Lipoplex size mirrored closely the pre-
cursor liposome, with the lipoplex generally having an increase
in diameter by 20–50 nm.
Figure 4. Transfection of cells using lipoplexes formulated at different N:P
charge ratios: a) MCF-7 cells (lipid 3: &; lipid 4: &; lipid 5: &) and b) H1792
cells (lipid 5: &; lipid 4: &; control: &). Results are expressed as total rela-
tive light units (RLU). Transfections were performed in 24-well tissue culture
plates using luciferase reporter construct (pCMV-Luc; 0.025 mg per well) with
18 h transfection time. Each data point represents the mean value of three
separate transfections. Error bars show the standard deviation (SD) from the
mean. LFT=lipofectamine 2000, jetP=jetPRIME.
On a scale of 1 mg DNA per well, transfection of MCF-7 cells
in a six-well format using lipoplexes derived from lipid 3 at an
N:P ratio of 7:1 gave approximately a 70-fold higher luciferase
expression (as assessed by RLU per mg protein) than lipofecta-
mine 2000-derived lipoplexes (see Supporting Information).
This result shows that the oxime ether lipid 3 is effective over
a wide range of DNA doses.
Transfection activity
We evaluated the transfection activity of the dual-chain oxime
ether lipid panel in human breast cancer (MCF-7) and human
lung cancer (H1792) cells using a CMV-luciferase reporter plas-
mid in the presence of serum (Figure 4). Whereas the shortest-
chain lipid 2 was essentially ineffective at all N:P formulations
examined in MCF-7 cells (data not shown), we were gratified
to find that the transfection efficiency of lipid 3 in this cell line
at an N:P ratio of 7 was notably superior (greater than 2.5
orders of magnitude higher activity) to the commercial trans-
fection standard lipofectamine 2000 (3500Æ990 RLU) as well
as to the popular[38] transfection formulation jetPrime (2500Æ
209 RLU) (Figure 4a). The unsaturated lipid 5 was also found to
be efficacious in MCF-7 cells, although we observed greater
deviation in the measurement of transgene expression at the
higher N:P ratio using this lipid (Figure 4a). We did not ob-
serve an additional boost in transfection activity on further am-
plifying the hydrophobic domain asymmetry by using dissym-
metric lipid 4, relative to the activities of lipids 2 and 5 in this
cell line.
Only lipids 4 and 5 led to high transgene expression in
H1792 cells (Figure 4b). In agreement with previous studies on
lung cell transfections,[14,39] the presence of unsaturation in the
hydrophobic domain, such as is present in lipid 5, proved to
be beneficial for activity. Lipoplexes derived from lipid 5 effec-
tively transfected H1792 cells relative to the commercial con-
trols (>60-fold higher activity at N:P=7). Interestingly, the en-
hanced asymmetry of lipid 4 resulted in improved transfection
of this cell line relative to the other saturated lipids examined.
To further examine the gene-transfer capability of lipids with
an oxime ether linker domain, we used a green fluorescent
protein (GFP) reporter vector to estimate the transfection effi-
ciency in MCF-7 cells (Figure 5). Examination of cells by fluores-
cence microscopy using an EVOS fluorescent microscope de-
ChemMedChem 2011, 6, 2063 – 2069
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
2065

M. H. Nantz et al.
MED
Figure 6. GFP expression in PAR C10 cells. a) Cells were transfected (0.2 mg
GFP reporter construct) with lipoplex formulations derived from lipids 2–5
(N:P=1.5) and excess lipofectamine 2000 (LFT; positive control; 1.6 mL);
b) Comparison of lipid 3 formulations at the indicated N:P ratio with LFT at
the vendor-recommended dose (0.5 mL). The transfection percentage of GFP
expressing cells was analyzed by flow cytometry 48 h post-transfection. Each
data point reflects the mean value Æ standard deviation (SD) of three inde-
pendent experiments.
ofectamine was slightly more effective in this cell line when
used in excess (>3ꢁ recommended dose; Figure 6a) and com-
parable to lipid 3 when used at the vendor-recommended
dose (Figure 6b). These results suggest that lipids with an
oxime ether linker domain are suitable for the pDNA transfec-
tion of epithelial cell lines and, depending on hydrophobic
domain composition, competitive with the transfection stan-
dard lipofectamine.
Figure 5. MCF-7 cells were transfected with a GFP expression plasmid
(0.1 mg per well in a 24-well plate). Figure are presented as an overlay of
green fluorescence (visualized by fluorescent microscopy after 20 h of trans-
fection) over the phase contrast image: a) cells treated with pDNA (negative
control); b) cells transfected with lipoplexes derived from lipofectamine
2000 (positive control; vendor-recommended dose=0.25 mL); c) cells trans-
fected with lipoplexes derived from lipid 3 at N:P=7.
termined that the results were in agreement with those of the
luciferase assay.
Transfection with siRNA
MCF-7 and H1792 cells are both cancer cell lines. To test the
lipids in a noncancer epithelial cell line, GFP was expressed in
the salivary gland cell line PAR C10.[40] Lipoplexes were formu-
lated at an N:P ratio of 1.5, the optimal ratio determined for
lipids 2–5 in this cell line. GFP expression was evaluated 48 h
post-transfection. The fraction of transfected cells was quanti-
tated by flow cytometry. Lipid 3 transfected approximately
40% of the cells, lipid 2 transfected about 20%, while lipids 4
and 5 exhibited less than 10% cell transfection (Figure 6a). Lip-
In addition to pDNA transfection, we tested the ability of lipid
3 to transfect epithelial cells with siRNA. The cellular “house-
keeping” gene glyceraldehyde-3-phosphate dehydrogenase
(GAPDH) was selected as a target, and transfection efficiency
was determined by measuring the reduction in GAPDH activity.
Lipid 3 reduced GAPDH activity by 55Æ9% (meanÆSEM),
while the control lipid, lipofectamine RNAiMAX, reduced
GAPDH activity by 71Æ6% (n=9). This difference was not stat-
2066
www.chemmedchem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2011, 6, 2063 – 2069

pDNA and siRNA Transfection Lipids
istically significant (p=0.23). These results suggest that lipid 3
effectively introduces siRNA into epithelial cells.
Cytotoxicity measurements
To measure the cytotoxicity of the lipids in MCF-7 and H1792
cells, we performed cell viability studies using the vital dye
trypan blue to stain dead cells. The number of live cells re-
maining after treatment was calculated as a percentage over
the untreated control (Figure 7). Our lipids were found to be
less toxic than lipofectamine and jetPrime in both cell lines,
with cell viability comparable to untreated cells at all N:P for-
mulations examined.
Figure 8. Cell viability assay. PAR C10 cells were transfected (N:P=1.5) with
lipoplex formulations derived from lipids 2–5 and lipofectamine 2000 (LFT;
1.6 mL). After 48 h, the cell viability was measured by calculating the percent-
age reduction of alamar blue dye. The data are shown as the mean value Æ
standard deviation (SD) of three independent experiments.
The ability to transfect epithelial cells could have clinical ap-
plications, as epithelial surfaces are readily accessible and attra-
tive targets for gene therapy protocols that aim to express
theraputic proteins in vivo.[41] The ability to simultaneously
transfect pDNA and siRNA with a single reagent now opens
potential applications to express therapeutic proteins in cells
that are simultaneously modified by gene knockdown.
Figure 7. Relative cytotoxicity. MCF-7 (&) and H1792 cells (&) were trans-
fected using either lipofectamine 2000 (LFT), jetPrime (jP) or oxime ether lip-
oplexes at the specified N:P ratio (value in parenthesis). Transfections were
performed in a 96-well format. Dead and live cells were counted 18 h post-
transfection using a trypan blue stain (n=3).
Experimental Section
Chemistry
All 1H NMR (400 MHz) and 13C NMR (100 MHz) spectra were re-
corded in CD3OD (Varian 400 MR NMR spectrometer). High-resolu-
tion mass spectrometry (HRMS) was performed by the CREAM fa-
cility at the University of Louisville (Kentucky, USA) using electro-
spray ionization (ESI) (Thermo LTQ-FT). Lipid 3 was prepared as pre-
viously described.[35]
For PAR C10 cells, the viability was determined by reduction
of alamar blue (Figure 8). None of the lipids reduced cell viabil-
ity significantly compared with untreated cells, confirming the
low toxicity of lipids with an oxime ether linker domain.
N,N-Dimethyl-N-2-((dodecylideneamino)oxy)-(2-((dodecylidene-
amino)oxy)ethyl)ethanaminium iodide (2): Compound
1
(26.5 mg, 0.091 mmol) and dodecanal (35.4 mg, 0.19 mmol) were
dissolved in MeOH (3 mL) and stirred at RT. After 3.5 h, the solvent
was removed in vacuo, and the residue was purified by column
chromatography (SiO2; MeOH:CH2Cl2, 3%) to afford oxime ether
lipid 2 as a white solid (34.6 mg, 61%; (E, E):(E, Z), 5.3:1): Rf =0.52
Conclusions
We have demonstrated that the facile aminooxy carbonyl click
reaction can be used to rapidly assemble transfection lipids
furnished with asymmetric hydrophobic domains. The data
show that lipids containing an oxime ether linking domain are
well tolerated by both cancerous and noncancerous epithelial
cells. Furthermore, with appropriate tuning of hydrophobic
domain composition and N:P formulation, lipoplexes derived
from oxime ether lipids readily transfect representative epithe-
lial cell lines with excellent activity relative to established trans-
fection agents. The present study shows that oxime ethers are
a suitable choice for linker domain. Given the biological com-
patibility of oxime ethers and their ease of formation, this func-
tional group should find much application in the future design
of transfection vectors.
1
(CH2Cl2/MeOH, 9:1); mp: 1558C (dec); H NMR (400 MHz, CD3OD):
(major diastereomer) d=0.86 (t, J=6.8 Hz, 6H), 1.20–1.39 (m, 28H),
1.43–1.48 (m, 4H), 2.17 (dd, J=6.8, 7.2 Hz, 4H), 3.20 (s, 6H), 3.73–
3.75 (m, 4H), 4.40–4.41 (m, 4H), 7.46 ppm (t, J=6.0 Hz, 2H);
13C NMR (100 MHz, CD3OD): d=14.6, 23.9, 27.1, 27.3, 27.7, 30.4,
30.6, 30.9, 33.2, 53.4, 64.9, 68.0, 68.3, 155.0, 155.5 ppm; IR (neat):
~
v=1735, 1468, 966, 920 cmÀ1
;
HRMS: m/z [M]+ calcd for
C30H62N3O2+: 496.4837, found: 496.4839.
N,N-Dimethyl-N-2-((dodecylideneamino)oxy)-(2-((tetradecyli-
deneamino)oxy)ethyl)ethanaminium iodide (4): Compound
1
(22.6 mg, 0.078 mmol) and dodecanal (7.2 mg, 0.034 mmol) were
dissolved in MeOH (1 mL) and stirred at RT. After 14 h, the solvent
was removed in vacuo, and the residue was purified by column
chromatography (SiO2; MeOH:CH2Cl2, 5%) to afford the intermedi-
ChemMedChem 2011, 6, 2063 – 2069
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
2067

M. H. Nantz et al.
MED
ate mono-oxime ether as a yellow oil (14.0 mg, 39%; (E, E):(E, Z),
1.7:1): Rf =0.25 (CH2Cl2/MeOH, 9:1); 1H NMR (400 MHz, CD3OD):
(major diastereomer) d=0.85 (t, J=6.8 Hz, 3H), 1.20–1.38 (m, 16H),
1.45 (dd, J=6.4, 6.8 Hz, 2H), 2.16 (dd, J=6.4, 7.4 Hz, 2H), 3.17–3.20
(m, 6H), 3.65–3.74 (m, 4H), 4.04 (m, 2H), 4.40 (m, 2H), 7.45 ppm (t,
J=6.0 Hz, 1H); 13C NMR (100 MHz, CD3OD): d=14.6, 23.9, 27.1,
USA). H1792 cells were cultured in RPMI (Invitrogen), 1% Penn-
strep, 10% FBS.
The PAR C10 cell line was immortalized from rat parotid acinar
cells.[40] Experiments were performed on cells of passage number
40–60. The cultures were grown to 50–60% confluency in DMEM/
Ham’s F12 media (1:1; GIBCO BRL, Gaithersburg, MD, USA) contain-
ing 2.5% (v/v) FBS (GIBCO BRL, Gaithersburg, MD, USA) and the fol-
lowing supplements: 0.1 mm retinoic acid, 80 ngmLÀ1 epidermÀa1l
growth factor, 2 nm triiodothyronine, 5 mm glutamine, 0.4 mgmL
hydrocortisone, 5 mgmLÀ1 insulin, 5 mgmLÀ1 transferrin, 5 ngmLÀ1
sodium selenite and 50 mgmLÀ1 gentamicin (all from Sigma, St.
Louis, MO, USA). Cells were cultured at 378C in a humidified at-
mosphere of 95% air/5% CO2.
27.3, 27.6, 30.4, 30.6, 30.8, 30.9, 33.2, 53.3, 64.7, 68.0, 68.4, 70.0,
134.0, 155.0, 155.5 ppm; HRMS: m/z [M]+ calcd for C18H40N3O2
:
+
330.3115, found: 330.3118.
The C12 mono-oxime ether intermediate (13.1 mg, 0.029 mmol) and
tetradecanal (9.1 mg, 0.043 mmol) were dissolved in MeOH (1 mL)
and stirred at RT. After 14 h, the solvent was removed in vacuo,
and the residue was purified by column chromatography (SiO2;
MeOH:CH2Cl2, 3%) to afford unsymmetrical lipid 5 as a yellow oil
(13 mg, 69%; (E, E):(E, Z), 2.3:1): Rf =0.48 (CH2Cl2/MeOH, 9:1);
1H NMR (400 MHz, CD3OD): (major diastereomer) d=0.84 (t, J=
6.8 Hz, 6H), 1.23–1.26 (m, 36H), 1.44 (dd, J=6.8, 7.2 Hz, 4H), 2.15
(dd, J=6.4, 7.2 Hz, 4H), 3.18 (s, 6H), 3.69–3.72 (m, 4H), 4.38–4.40
(m, 4H), 7.44 ppm (t, J=6.2 Hz, 2H); 13C NMR (100 MHz, CD3OD):
d=14.6, 23.9, 27.1, 27.3, 27.6, 30.4, 30.6, 30.7,.30.8, 30.9, 33.2, 53.3,
Liposome formulation: An equimolar amount of DOPE from a stock
solution (10 mg DOPE per mL CHCl3) was added to a solution of
oxime ether lipid (1 mg) in CHCl3 (0.20 mL). The solvent was evapo-
rated, and the resultant thin lipid film was dried under vacuum
(4 h). Ultra-pure water (3 mL) was added to the dry lipid film, and
the suspension was sonicated (bath sonicator) for 1 min at RT to
furnish the liposome formulation (0.33 mg oxime ether lipid per
mL).
~
64.9, 68.0, 68.4, 155.0, 155.5 ppm; IR (neat): v=1728, 1467, 964,
942 cmÀ1; HRMS: m/z [M]+ calcd for C32H66N3O2+: 524.5150, found:
524.5172.
Luciferase expression: Luciferase transfections in MCF-7 and H1792
cells were performed in triplicate using 0.025 mg of pDNA (pCMV-
Luc) per well. Cells were seeded up to 1ꢁ105 cells per well in a 24-
well plate to give 50–60% confluency, and 400 mL of media con-
taining 10% FBS was added to each well. Lipoplexes (lipid–pDNA
complex) were prepared at N:P charge ratios of 3, 5 and 7 by
adding the required volume of liposome solution to a pDNA solu-
tion (3 mL, 0.025 mg DNA per mL) and incubated for 5 min. Serum-
free DMEM (100 mL) was then added to each lipoplex solution fol-
lowed by incubation for 20 min at RT with occasional gentle vortex
mixing. The lipoplex solutions were diluted to 600 mL with serum-
free DMEM, and then an aliquot of the final lipoplex formulation
(200 mL) was added directly to each well. No additional media con-
taining FBS was added to the cells during transfection. After 18 h
incubation at 37.58C, the cells were lysed and luciferase expression
was quantified using a commercial kit (Promega Corp., Madison,
WI, USA) and luminometer according to the vendor’s protocol. Lip-
ofectamine 2000 (Invitrogen) and jetPrime (Polyplus Transfection
SA, Illkirch, France) lipoplexes were prepared at the vendor-recom-
mended stoichiometry (2 mL stock per mg DNA).
N,N-Dimethyl-N-2-((((Z)-octadec-9-en-1-ylidene)amino)oxy)-(2-
((((Z)-octadec-9-en-1-ylidene)amino)oxy)ethyl)ethanaminium
iodide (5): Compound 1 (19.7 mg, 0.068 mmol) and cis-9-octadece-
nal (37.8 mg, 0.142 mmol) were dissolved in MeOH (3 mL) and
stirred at RT. After 3.5 h, the solvent was removed in vacuo, and
the residue was purified by column chromatography (SiO2;
MeOH:CH2Cl2, 3%) to afford oxime ether lipid 4 as a yellow oil
(45.8 mg, 90.0%; (E, E):(E, Z), 3.0:1): Rf =0.51 (CH2Cl2/MeOH, 9:1);
1H NMR (400 MHz, CD3OD): (major diastereomer) d=0.84 (t, J=
6.8 Hz, 6H), 1.23–1.27 (m, 40H), 1.42–1.47 (m, 4H), 1.96–1.97 (m,
8H), 2.15 (dd, J=6.4, 7.2 Hz, 4H), 3.16 (s, 6H), 3.68–3.71 (m, 4H),
4.38–4.39 (m, 4H), 5.27–5.29 (m, 4H), 7.43 ppm (t, J=6.0 Hz, 2H);
13C NMR (100 MHz, CD3OD): d=14.6, 23.9, 27.6, 28.3, 30.4, 30.5,
30.6, 30.8, 31.0, 33.2, 53.3, 64.9, 68.0, 130.9, 131.1, 155.0 ppm; IR
(neat): v=1730, 1464, 964, 942 cmÀ1; HRMS: m/z [M]+ calcd for
~
C42H82N3O2+: 660.6401, found: 660.6395.
Zeta potential and particle size: Measurements of liposome and
lipoplex zeta potential and particle size were performed using a
ZetaPALS dynamic light scattering detector (Brookhaven Instru-
ments Corporation; Model 90 Plus). All measurements were taken
in water. Lipoplexes were formulated at a final concentration of
0.1 mg oxime ether lipid per mL water. To formulate lipoplexes at
different N:P ratios, 100 mL of the liposome formulation was added
to the required quantity of pDNA (taken from a stock solution of
0.1 mg pDNA per mL) in an Eppendorf tube. After incubating each
lipoplex for 15 min, the solutions were diluted to 1 mL by adding
ultra-pure water. Measurements were taken within minutes of for-
mulation.
GFP expression: Transfection studies were performed using a GFP
reporter gene in MCF-7 cells using 0.1 mg of pEGFP DNA (Clontech
Laboratories Inc., Madison, WI, USA) per well in a 24-well plate
with a similar transfection protocol as described for the luciferase
assays. Cells were examined 40 h post-transfection by fluorescence
microscopy (EVOS, Advanced Microscopy Group, Bothell, WA, USA)
and photographed.
GFP expression also was assessed in PAR C10 cells using the plas-
mid pZsGreen1-N1 (Clontech). Transfection of PAR C10 cells pro-
ceeded by plating the cells at a density of 5ꢁ104 cellscmÀ2 in 24-
well plates in complete medium (described above) with pDNA
(200 ng per well). Lipolexes were formed at an N:P charge rato of
1.5 in 200 mL of complete medium containg 2.5% FBS. Media was
removed, and cells were transfected with lipoplexes. After 3 h,
each well was replenished with 250 mL of complete media. Lipo-
fectamine was used as a positive control. After 48 h, GFP-express-
ing cells were analyzed using a FACS Calibur flow cytometer
(Becton Dickinson, San Jose, CA, USA) and CellQuest software (ver-
sion 3.3).
Biology
Tissue culture: Human breast cancer cells (MCF-7) and lung tumor
cells (H1792) were purchased from American Type Culture Collec-
tion (Manassas, VA, USA). Cells were grown to 50–60% confluency
prior to transfection. MCF-7 cells were cultured in Dulbecco’s modi-
fied Eagle’s medium (DMEM; Invitrogen, Carlsbad, CA, USA), 1%
Pennstrep (cat. no.: 30-002-CI, Mediatech Inc., Manassas, VA, USA),
10% fetal bovine serum (FBS; Valley Biomedical, Winchester, VA,
2068
www.chemmedchem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2011, 6, 2063 – 2069

pDNA and siRNA Transfection Lipids
[13] K. Ewert, N. L. Slack, A. Ahmad, H. M. Evans, A. J. Lin, C. E. Samuel, C. R.
Safinya, Curr. Med. Chem. 2004, 11, 133–149.
[14] R. P. Balasubramaniam, M. J. Bennett, A. M. Aberle, J. G. Malone, M. H.
Nantz, R. W. Malone, Gene Ther. 1996, 3, 163–172.
[15] A. M. Aberle, F. Tablin, J. Zhu, N. J. Walker, D. C. Gruenert, M. H. Nantz,
Biochemistry 1998, 37, 6533–6540.
[16] A. Masotti, G. Mossa, C. Cametti, G. Ortaggi, A. Bianco, N. Del Grosso, D.
Malizia, C. Esposito, Colloids Surf. B 2009, 68, 136–144.
[17] A. Fraix, T. Montier, N. Carmoy, D. Loizeau, L. Burel-Deschamps, T.
Le Gall, P. Giamarchi, H. Couthon-Gourvꢂs, J.-P. Haelters, P. Lehn, P.-A.
Jaffrꢂs, Org. Biomol. Chem. 2011, 9, 2422–2432.
[18] J. Zhu, R. J. Munn, M. H. Nantz, J. Am. Chem. Soc. 2000, 122, 2645–2646.
[19] H. Chen, H. Zhang, C. M. McCallum, F. C. Szoka, X. Guo, J. Med. Chem.
2007, 50, 4269–4278.
Cytotoxicity assay: Lipid cytotoxicity was evaluated using the vital
dye trypan blue to distinguish between live and dead cells. MCF-7
and H1792 cells were seeded at 1ꢁ104 cells per well and then
grown to 50–60% confluency in 96-well plates. Transfections were
performed as described above. The media was removed 18 h post-
transfection. Cells were then trypsinized by adding 100 mL trypsin
to each well and incubated for 5 min at 37.58C. Trypsin was neu-
tralized by adding 1X phosphate-buffered saline (1X PBS), and then
10 mL of extract from each well was mixed with 10 mL of trypan
blue. Dead and live cells were counted under an inverted micro-
scope to determine the percentage cell viability using a hemocy-
tometer.
Cell viability in PAR C10 cells was assessed using the alamarBlue
cell viability reagent (Invitrogen). The PAR C10 cells were transfect-
ed as described above using lipoplexes formulated from lipids 2–5
at the optimal N:P charge ratio of 1.5, and lipofectamine at 1.6 mL.
At 48 h post-transfection, the viability assay was performed as per
the manufacturer’s protocol. Absorbance was measured at 600 nm
and 570 nm using a Synergy HT multimode microplate reader (Bio-
Tek, Winooski, VT, USA). The reduction of alamar blue is proportion-
al to the proliferation of cells and is expressed as a percentage.
[20] D. Liu, W. Qiao, Z. Li, Y. Chen, X. Cui, K. Li, L. Yu, K. Yan, L. Zhu, Y. Guo, L.
Cheng, Chem. Biol. Drug Des. 2008, 71, 336–344.
[21] G. Byk, C. Dubertret, V. Escriou, M. Frederic, G. Jaslin, R. Rangara, B.
Pitard, J. Crouzet, P. Wils, B. Schwartz, D. Scherman, J. Med. Chem. 1998,
41, 224–235.
[22] M. Savva, P. Chen, A. Aljaberi, B. Selvi, M. Spelios, Bioconjugate Chem.
2005, 16, 1411–1422.
[23] E. Guꢄnin, A.-C. Hervꢄ, V. V. Floch, S. Loisel, J.-J. Yaouanc, J.-C. Clꢄment,
C. Fꢄrec, H. des Abbayes, Angew. Chem. 2000, 112, 643–645; Angew.
Chem. Int. Ed. 2000, 39, 629–631.
[24] H. C. Kolb, M. G. Finn, K. B. Sharpless, Angew. Chem. 2001, 113, 2056–
2075; Angew. Chem. Int. Ed. 2001, 40, 2004–2021.
Acknowledgements
[25] L. A. Marcaurelle, Y. Shin, S. Goon, C. R. Bertozzi, Org. Lett. 2001, 3,
3691–3694.
[26] J. Gajewiak, S. Cai, X. Z. Shu, G. D. Prestwich, Biomacromolecules 2006,
7, 1781–1789.
[27] G. Surkau, K. J. Bçhm, K. Mꢃller, H. Prinz, Eur. J. Med. Chem. 2010, 45,
3354–3364.
[28] I. C. Choong, J. A. Ellman, J. Org. Chem. 1999, 64, 6528–6529.
[29] D. P. Figgitt, K. J. McClellan, Drugs 2000, 60, 925–954.
[30] E. Perouzel, M. R. Jorgensen, M. Keller, A. D. Miller, Bioconjugate Chem.
2003, 14, 884–898.
[31] S. Carmona, M. R. Jorgenson, S. Kolli, C. Crowther, F. H. Salazar, P. L.
Marion, M. Fujino, Y. Natori, M. Thanou, P. Arbuthnot, A. D. Miller, Molec.
Pharm. 2009, 6, 706–713.
We thank the US National Institutes of Health (NIH)
(5R21DE019271, 1P20 RR18733) for support of this work. R.J.K.
thanks the University of Louisville (Kentucky, USA) for a graduate
research fellowship. We also thank Dr. Olga Baker (University at
Buffalo, New York, USA) for donating PAR C10 cells, Mr. Xu Han
and Dr. Mahsa Abdolhosseini (University of Minnesota, Minneap-
olis, USA) for contributions to the PAR C10 experiments.
Keywords: aminooxy salts · gene transfer · nonviral vectors ·
oximation · transfection agents
[32] M. H. Nantz, C. W. Dicus, B. Hilliard, S. Yellayi, S. Zou, J. G. Hecker, Mole-
cules Molec. Pharm. 2010, 7, 786–794.
[33] J. G. Hecker, G. O. Berger, K. A. Scarfo, S. Zou, M. H. Nantz, ChemMed-
Chem 2008, 3, 1356–1361.
[34] S. Biswas, L. E. Gordon, G. J. Clark, M. H. Nantz, Biomaterials 2011, 32,
2683–2688.
[35] S. Biswas, X. Huang, W. R. Badger, M. H. Nantz, Tetrahedron Lett. 2010,
51, 1727–1729.
[1] P. L. Felgner, T. R. Gadek, M. Holm, R. Roman, H. W. Chan, M. Wenz, J. P.
Northrop, G. M. Ringold, M. Danielsen, Proc. Natl. Acad. Sci. USA 1987,
84, 7413–7417.
[2] L. Wasungu, D. Hoekstra, J. Controlled Release 2006, 116, 255–264.Refer-
ence [2].
[36] For example, the oximyl proton shift for the (E)-oxime ether side chain
in 3 occurs at d=7.58 ppm, while the (Z)-isomer shift appears at d=
6.91 ppm.
[3] A. Mountain, Trends Biotechnol. 2000, 18, 119–128.
[4] C. E. Thomas, A. Ehrhardt, M. A. Kay, Nature Rev. Genetics 2003, 4, 346–
358.
[37] J.-Y. Je, Y.-S. Cho, S.-K. Kim, Biomacromolecules 2006, 7, 3448–3451.
[38] P. D. Allaire, A. L. Marat, C. Dall’Armi, G. Di Paolo, P. S. McPherson, B.
Ritter, Mol. Cell 2010, 37, 370–378.
[39] S. Fletcher, A. Ahmad, E. Perouzel, A. Heron, A. D. Miller, M. R. Jorgen-
sen, J. Med. Chem. 2006, 49, 349–357.
[40] D. O. Quissell, K. A. Barzen, R. S. Redman, J. M. Camden, J. T. Turner, In
Vitro Cell. Dev. Biol. Anim. 1998, 34, 58–67.
[41] Y. Samuni, B. J. Baum, Biochim. Biophys. Acta, Mol. Basis Dis. 2011, in
press; DOI: 10.1016/j.bbadis.2011.06.014.
[5] L. S. Young, P. F. Searle, D. Onion, V. Mautner, J. Pathol. 2006, 208, 299–
318.
[6] S. Li, L. Huang, Gene Ther. 2000, 7, 31–34.
[7] K. Kodama, Y. Katayama, Y. Shoji, H. Nakashima, Curr. Med. Chem. 2006,
13, 2155–2161.
[8] A. Elouahabi, J.-M. Ruysschaert, Mol. Ther. 2005, 11, 336–347.
[9] T. Montier, T. Benvegnu, P. A. Jaffrꢂs, J. J. Yaouanc, P. Lehn, Curr. Gene
Ther. 2008, 8, 296–312.
[10] R. Leventis, J. R. Silvius, Biochim. Biophys. Acta, Biomembr. 1990, 1023,
124–132.
[11] C. Tros de Ilarduya, Y. Sun, N. Dꢃzgꢃnes, Eur. J. Pharm. Sci. 2010, 40,
159–170.
Received: May 25, 2011
[12] D. Niculescu-Duvaz, J. Heyes, C. J. Springer, Curr. Med. Chem. 2003, 10,
1233–1261.
Published online on August 31, 2011
ChemMedChem 2011, 6, 2063 – 2069
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
2069