Low molecular weight PEI-based biodegradable lipopolymers as gene delivery vectors

Article abstract of DOI:10.1039/c2ob27211c

Non-viral gene vectors play an important role in the development of gene therapy. In this report, different hydrophobic chains were introduced into low molecular weight (LMW) PEI-based biodegradable oligomers to form a series of lipopolymers (LPs), and their structure-activity relationships were studied. Results revealed that the nine polymers can condense plasmid DNA well to form nanoparticles with appropriate sizes (120-250 nm) and positive zeta-potentials (+25-40 V). In vitro experiments were carried out and it was found that LP2 showed much higher transfection efficiency both in the presence and in the absence of serum under the polymer/DNA weight ratio of 0.8 in A549 cells.

Full text of DOI:10.1039/c2ob27211c

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Low molecular weight PEI-based biodegradable
lipopolymers as gene delivery vectors†
Cite this: Org. Biomol. Chem., 2013, 11,
1242
Miao-Miao Xun,‡ Xue-Chao Zhang,‡ Ji Zhang,* Qian-Qian Jiang, Wen-Jing Yi,
Wen Zhu* and Xiao-Qi Yu*
Non-viral gene vectors play an important role in the development of gene therapy. In this report,
different hydrophobic chains were introduced into low molecular weight (LMW) PEI-based biodegrad-
able oligomers to form a series of lipopolymers (LPs), and their structure–activity relationships were
studied. Results revealed that the nine polymers can condense plasmid DNA well to form nanoparticles
with appropriate sizes (120–250 nm) and positive zeta-potentials (+25–40 V). In vitro experiments were
carried out and it was found that LP2 showed much higher transfection efficiency both in the presence
and in the absence of serum under the polymer/DNA weight ratio of 0.8 in A549 cells.
Received 13th November 2012,
Accepted 11th December 2012
DOI: 10.1039/c2ob27211c
www.rsc.org/obc
electrostatic interaction between anionic DNA or RNA and the
cationic polymer. In addition to its DNA-condensing proper-
Introduction
Gene therapy, which is a new technology developed in the ties, this polymer possesses significant endosomolytic activity
modern medical research field, was used to treat diseases by because of its strong buffering capacity at acidic endosomal
delivering genes to specific cells. Since free oligonucleotides pH (5.1–7.4). High molecular weight (HMW) PEI has higher
and DNA are rapidly degraded by serum nucleases in the transfection efficiency (TE) than low molecular weight (LMW)
blood when injected intravenously, research efforts currently PEI, but its cytotoxicity is also higher.13,14 LMW PEI has been
focus on the design of effective carriers that can compact and demonstrated to have a relatively low cytotoxicity, but it also
protect oligonucleotides for gene delivery. Although viral cannot be used as a gene vector due to its low TE.
vectors1 have been proved to be very effective in achieving
PEI can be easily functionalized by a variety of functional
highly efficient gene delivery2 and expression in vitro, this group modifications to optimize its gene delivery activity
approach met severe limitations in terms of their immuno- because of the presence of amino groups. For instance, bio-
genicity and toxicity in clinical trials. Since the first perform- degradable molecules, such as polycaprolactone diacrylate
ance of polycation-mediated gene delivery in 1987,3 many linkage,15 cyclodextrin (CD),16 dextran,17 dithiobis(succinimi-
cationic polymers4,5 have been developed for gene transfection dylpropionate),18 poly(ethylene glycol) (PEG),19 and chitosan,20
applications. Most intensively studied examples include: poly- which can break down into low molecular weight segments,
ethyleneimine (PEI),6 poly(2-dimethylaminoethyl methacrylate) were introduced to modify LMW PEI. When our group also
(pDMAEMA),7 poly-L-lysine (pLL),8 chitosan,9 polyamidoamine introduced biodegradable ester-containing diglycidyl succinate
dendrimers (PAMAM),10 etc. Among these polymeric materials, linkages into LMW PEI, higher TE and lower cytotoxicity were
PEI is one of the most effective gene delivery agents and has obtained compared to 25 kDa PEI.21 However, in the presence
been extensively studied both in vitro and in vivo.11,12 It is an of serum, much lower TE was obtained, indicating the require-
attractive carrier for intracellular gene delivery because of its ment of further modifications.
well-established ability to condense nucleic acids via
Hydrophobic modifications of cationic polymers are receiv-
ing more and more attention. Introduction of hydrophobic
chains may improve the TE mainly due to the enhanced cellu-
lar uptake caused by the interaction of a hydrophobic chain
and a lipophilic membrane.22 Linear fatty acids, lithocholic
acid, cholesterol, tocopherol (vitamin E), etc. have been used
as hydrophobic parts to modify the cationic polymer to
improve the TE.23 In this report, we synthesized a series of
LMW PEI-based biodegradable low polymers with different
hydrophobic chains, and the effect of the long chains on gene
delivery was discussed.
Key Laboratory of Green Chemistry and Technology (Ministry of Education), College
of Chemistry, and State Key Laboratory of Biotherapy, West China Hospital, Sichuan
University, Chengdu 610064, People’s Republic of China. E-mail: ji_zhang@163.com,
xqyu@scu.edu.cn, zhuwen@scu.edu.cn; Fax: +86-28-85415886
†Electronic supplementary information (ESI) available: Additional experimental
information, including typical 1H NMR spectrum of LP2, GPC-monitored
polymer degradation, fluorescent quenching assays and GFP images involving
LP3 with different long chain concentrations. See DOI: 10.1039/c2ob27211c
‡These authors contributed equally to this work.
1242 | Org. Biomol. Chem., 2013, 11, 1242–1250
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
hydrophobic chains (LP1 and LP2), and this might be due to
the smaller steric hindrance of the shorter chain. For the ali-
Results and discussion
The preparation method of lipopolymers LP1–LP9 is shown in phatic chain with more than 10 carbon atoms (LP3–LP9), the
Scheme 1. Various aliphatic acids with different long chains molecular weight of the product was seldom affected by the
reacted with bis(β-hydroxylethyl)amine in the presence of 1- long chain structure. Further, the degradability of the poly-
ethyl-3-(3′-dimethylaminopropyl)carbodiimide hydrochloride meric materials was also studied by GPC. The Mw of LP2
(EDC·HCl) and N-hydroxybenzotriazole (HOBt) to give com- decreased from 31 109 to 18 767 Da within 24 h. Further exten-
pound 3. Subsequent reaction between 3 and epichlorohydrin sion of time resulted in a relatively slight decrease of Mw, and
in the presence of Bu4NBr and NaOH led to the diol glycidyl the Mw of 17 440 Da was detected after 120 h (Fig. S2, ESI†).
ethers 4, which were used as bridges in the cross-linking of These results affirmed the degradation of the lipopolymers,
LMW PEI. The cross-linking was simply processed by mixing especially in the first 24 h in solution.
the hydrophobic chain-containing linker 4, biodegradable
ester bond-containing linker 5 (mole ratio = 1 : 1) and PEI 600
Da at an appropriate amount ratio in refluxed ethanol. The
Table 1 Selective characteristics of polymers LP1–LP9
crude product was recrystallized 3 times to ensure its polydis-
persity. The proportion between the two different linkers in
Compounds
Mn
Mw
PDI
Ratio of linkers 4/5
the final product could be calculated from the ratio of 1H NMR
peak areas which belong to the methyl group (δ 0.89, 3H) on 4
and the ethylene group (δ 3.34, 4H) on 5. For typical com-
LP1
LP2
LP3
LP4
LP5
LP6
LP7
LP8
LP9
13 261
9414
4330
3413
4044
3565
2865
3969
3929
55 020
25 549
6196
4183
5139
4419
3282
5028
4974
4.149
2.714
1.431
1.226
1.271
1.240
1.146
1.267
1.266
1/2.18
1/1.46
1/1.55
1/3.35
1/1.62
1/1.55
1/1.69
1/1.41
1/5.67
1
pound LP2, the H NMR spectrum shows a peak area ratio of
3 (3H, –CH3 on 4) : 5.85 (4H, –COCH2CH2CO– on 5, Fig. S1,
ESI†), therefore the proportion between hydrophobic and
ester-containing linkers could be calculated as ∼1 : 1.5. Other
products showed similar correlations (Table 1), indicating that
the hydrophobic linkers 4 have slightly lower reaction activity
in the polymerization than the ester-containing linker 5. This
might be the result of the steric hindrance caused by the long
chains. The obtained polymers showed significant water solu-
bility, which was required for the subsequent studies. Further-
more, the buffering capacities of LP1–LP9 were estimated by
acid–base titration. PEI 25 kDa is well known for its strong
proton-sponge effect, leading to the disruption of endosome
in the transfection process.24 Data showed that the buffer
capabilities of LP1–LP9 were slightly lower than that of 25 kDa
PEI (Fig. 1). The lower buffer capability could be ascribed to
the participation of hydrophobic and biodegradable linkers,
which led to lower density of amino groups relative to PEI.
Gel permeation chromatography (GPC) was used to
measure the molecular weights of target compounds LP1–LP9.
Results (Table 1) show that a higher molecular weight product
would be obtained by using linker 4 with relatively short
Fig. 1 Determination of the buffering capacity of PEI and LP1–LP9 by acid–
base titration.
Scheme 1 Synthetic route of LP1–LP9. (i) EDC·HCl, HOBt, Et3N; (ii) epichlorohydrin, Bu4NBr, NaOH, H2O; (iii) ethanol, reflux.
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 1242–1250 | 1243

View Article Online
Paper
Organic & Biomolecular Chemistry
For non-viral gene vectors, DNA condensation capability is be due to the folding structure of the long chain caused by the
requisite. The formation of a polyplex can reduce the electro- carbon–carbon double bond. The zeta potentials shown in
static repulsion between DNA and the cell surface and can Fig. 3B gave an increasing trend along with the increase in
protect DNA against enzymatic degradation by nucleases in weight ratios. Charge reversal was observed when weight ratios
cytoplasm or serum.25 To investigate the DNA condensation increased to 0.4–0.8, which is in accord with the weight ratios
induced by LP1–LP9, agarose gel electrophoresis assays at for complete DNA retardation observed in gel electrophoresis
various LP/DNA weight ratios were performed. As shown in assays. Generally, the polyplexes formed from shorter hydro-
Fig. 2, a sufficient amount of pDNA was released from the phobic chain-containing lipopolymers, which have relatively
agarose gel with LP/DNA weight ratios ranging from 0.1 to 0.2.
However, pDNA was kept blocked in the well of the gel from
the ratio of 0.4, and almost no free pDNA could be observed
when the ratio came up to 0.8 (lane 4). Results suggested that
these materials can bind to DNA through electrostatic inter-
actions. Among the nine lipopolymers, LP7 shows lower DNA
condensation ability, which might be attributed to its lower
molecular weight (Table 1). In addition, fluorescent quenching
assays with ethidium bromide (EB) were used to further
demonstrate their DNA binding ability. The ethidium ion
shows a dramatic increase of fluorescence efficiency when it
intercalates into DNA. Results (Fig. S3, ESI†) showed that these
lipopolymers can displace EB from DNA upon polyplex for-
mation, leading to the quenching of the fluorescence. The
length of the hydrophobic chain has little effect on the
quenching abilities. Unlike the results of gel electrophoresis
assays, LP5 showed relatively weak DNA binding ability in flu-
orescent assays.
The appropriate size and zeta potential of polyplexes are of
critical importance for cationic polymers used as gene vectors.
These properties of LP1–LP9/DNA complexes were measured at
various LP/DNA weight ratios, and the results are shown in
Fig. 3. The particle sizes tend to be constant with the increase
of weight ratios (>2), and mean diameters of 120–250 nm were
found for these particles (Fig. 3A). Polyplex particles formed
between DNA and lipopolymers with shorter hydrophobic
chains would have smaller sizes. For oleoyl and linoleoyl-sub-
stituted compounds LP8 and LP9, they could form polyplexes
with distinctly smaller sizes compared to those formed from
Fig. 3 Average particle size (A) and zeta-potential (B) of LP1–LP9/DNA at
LP7, which has the same amount of carbon atoms. This might
weight ratios of 0.2, 0.4, 0.8, 1.6, 3.2 and 6.4 (mean SD, n = 3).
Fig. 2 Agarose gel electrophoresis retardation assays of LP/DNA complexes. Plasmid DNA was mixed with LP1–LP9 at different weight ratios: lane 0: pDNA control;
lanes 1–5: LP/DNA weight ratios of 0.1, 0.2, 0.4, 0.8 and 1.6, respectively.
1244 | Org. Biomol. Chem., 2013, 11, 1242–1250
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
high charge (N atoms) density, showed higher zeta potential at
Compared to classic cationic polymers, the hydrophobic
the weight ratio of ∼3.2. On the other hand, polyplexes formed long chain-containing lipopolymers would be liable to the
from LP8 and LP9 showed lowest positive zeta-potential, which membrane fusion with cells, making the endocytosis and
might be attributed to the existence of electron-rich double further transfection easier. The transfection efficiencies of the
bonds.
polyplexes formed from lipopolymers LP1–LP9 were investi-
The cytotoxicity of cationic polymers is thought to be gated in A549 cells both in the presence and in the absence of
caused by damage from the interaction with the plasma mem- serum. PEI 25 kDa was used for comparison because of its
brane or other cellular compartments. MTT-based cell viability high TE and easy availability. PEI/DNA polyplexes were pre-
assays were performed in A549 cells at various LP/DNA ratios pared at an N/P ratio of 10 (weight ratio of 1.39), and polymer
with or without serum, and the results are shown in Fig. 4. LP1–LP9/DNA complexes were prepared at the optimized
The lipopolymers showed similar cytotoxicities to 25 kDa PEI, weight ratio of 0.8.26 It was interesting that not the long chain-
especially under higher weight ratios (Fig. 4A). Besides the modified materials, but LP2 with relatively short hydrophobic
positive charges, the relatively high cytotoxicity may be due to chains (decoyl) gave the best transfection results, no matter in
the membrane fusion ability caused by the long hydrophobic the experiments without or with serum. LP2 showed 5 times
chains. LP8 and LP9 showed higher cell viability, which might higher TE compared to that of the PEI/DNA complex in the
be the results of their lower charge density. The cytotoxicities absence of serum (Fig. 5A). However, the transfection efficien-
of polyplexes were also estimated in the presence of serum cies of LPs in the presence of serum were unfortunately lower
(Fig. 4B). Because of their higher molecular weight, LP1 and than PEI, which might be the result of their higher cytotoxicity
LP2 showed higher cytotoxicity than PEI, especially under the in serum medium. As shown in Fig. 5B, LP1 and LP2 gave
weight ratio of 1.6. Polyplexes with shorter hydrophobic chains higher TE compared to others. Since LP1 only gave lower
have higher positive charge, leading to lower cell viability. In efficiency in the serum free transfection, we speculate that for
contrast, polyplexes formed from longer chain modified lipo- this type of lipopolymer, the structure of the hydrophobic
polymers (LP5–LP9) showed good biocompatibility and led to chain but not molecular weight affects the TE most.
higher cell viability.
To directly visualize the infected cells expressing green fluo-
rescent protein reporter gene pEGFP-Nl, enhanced green fluo-
rescent protein expression in A549 cells was observed by an
Fig. 4 Relative cell viabilities caused by LPs (A) and LP/DNA polyplexes (B)
toward A549 cells, 25 kDa PEI and PEI/DNA complex were used as control. In
figure B, experiments were processed with serum. For comparison with figure A,
the concentrations of LPs were 1.2 and 2.4 for weight ratios of 0.8 and 1.6,
respectively.
Fig. 5 Luciferase expression in A549 cells (A: serum-free; B: with serum) trans-
fected by 25 kDa PEI/DNA (N/P = 10) and LP/DNA complexes at w/w of 0.8
(mean SD, n = 3).
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 1242–1250 | 1245

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 6 Fluorescent microscope images of pEGFP-transfected in A549 cells (A: serum-free B: serum). N/P of PEI/DNA was 10 (weight ratio of 1.39), LP/DNA weight
ratio was 0.8.
inverted fluorescent microscope. The weight ratios were the potential, and subsequently higher cytotoxicity. However,
same as those used in the luciferase assay, and the pictures higher cytotoxicity did not always lead to lower TE, and LP2
are shown in Fig. 6. The images showed that compared to with short decoyl groups exhibited the best transfection
25 kDa PEI/DNA, the density of transfected cells caused by results. Further modification of lipopolymers and relative
LP2/DNA was enhanced both in the presence and absence of mechanism studies are now in progress.
serum. Meanwhile, LP4 and LP8 showed weaker GFP
expression, which was in accordance with the results in lucifer-
ase assays processed in the same cell line. It is worth to
mention that we also prepared the lipopolymers with different
Experimental section
long chain densities by varying the 4/5 ratios in the polymeriz- Materials and methods
ation process (Scheme 1). These products were also applied to
All chemicals and reagents were obtained commercially and
the GFP assays, and the results (typical images are shown in
were used as received. Anhydrous ethanol was dried and puri-
Fig. S4, ESI†) showed that the lipopolymer prepared with a
fied under nitrogen by using standard methods and was dis-
tilled immediately before use. LMW PEI (branched, average
molecular weight 600 Da, 99%) was purchased from Aladdin
(Shanghai, China). 25 kDa PEI (branched, average molecular
1 : 1 mole ratio of hydrophobic linker 4 and ester-containing
linker 5 could give the best results. Meanwhile, reducing the
long chain density led to distinct decrease of GFP gene trans-
fection (the first row in Fig. S4†).
weight 25 kDa) and MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide) were purchased from Sigma-
Aldrich (St. Louis, MO, USA). The plasmids used in the study
were pGL-3 (Promega, Madison, WI, USA, coding for luciferase
Conclusions
DNA) and pEGFP-N1 (Clontech, Palo Alto, CA, USA, coding for
In summary, a new class of LMW PEI-based biodegradable EGFP DNA). 1640 Medium and fetal bovine serum were pur-
polymeric compounds with different hydrophobic chains were chased from Invitrogen Corp. The MicroBCA protein assay kit
synthesized. These so-called lipopolymers were expected to was obtained from Pierce (Rockford, IL, USA). The luciferase
have both good DNA condensation ability and good membrane assay kit was purchased from Promega (Madison, WI, USA).
fusion ability. LP1–LP9 could bind DNA to form nanoparticles The endotoxin free plasmid purification kit was purchased
with proper sizes and zeta-potentials. Among the synthesized from TIANGEN (Beijing, China).
materials, decoyl-modified LP2 showed the highest TE both in
1H-NMR spectra were obtained on a Bruker AV400 spec-
the presence and in the absence of serum, and about 5 times trometer. CDCl3 or D2O was used as the solvent and TMS as
higher TE compared to that of the PEI/DNA complex was the internal reference. The molecular weight of polyamine was
found. Their structure–activity relationships were studied, and determined by gel permeation chromatography (GPC) (Waters
it was suggested that lipopolymers with higher molecular 515 pump, Waters 2410 Refractive Index Detector, 25 °C, incor-
weight could be obtained by using starting material with porating Shodex columns OHPAK KB-803). A filtered mixture
shorter hydrophobic chains. The polyplexes formed from these of 0.2 mol L−1 HAc/NaAc buffer which contained 20% CH3CN
polymers may have smaller particle size and higher zeta- (volume ratio) was used as the mobile phase with a flow rate of
1246 | Org. Biomol. Chem., 2013, 11, 1242–1250
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
0.5 mL min−1. Molecular weights were calculated against poly 69.93, 69.63, 50.83, 50.79, 48.96, 46.49, 44.30, 44.05, 33.20,
(ethylene glycol) standards of number average molecular 31.96, 29.59, 29.56, 29.38, 25.41, 22.74, 14.18. MALDI-HRMS:
weights ranging from 200 to 80 000.
m/z 394.2550 ([M + Na]+), C20H37NO5Na+, calc. 394.2570.
4d: Yield 55.2%. H NMR (600 MHz, CDCl3) δ: 0.81 (t, J =
6.4 Hz, 3H, CH3CH2), 1.15–1.29 (m, 16H, CH3(CH2)8CH2),
1
Synthesis and characterization of linker 4
In a typical procedure, a mixture of aliphatic acid 1 (24 mmol), 1.53–1.58 (m, 2H, CH2CH2CH2), 2.30 (t, J = 7.8 Hz, 2H,
dihydroxyethylamine (3 g, 29 mmol), N-hydroxybenzotriazole OCCH2CH2), 2.50–2.55 (m, 2H, OCH2CH), 2.69–2.75 (m, 2H,
(3.21 g, 24 mmol) and triethylamine (2.41 g, 24 mmol) was dis- OCH2CH), 3.03–3.09 (m, 2H, OCHCH2), 3.27–3.32 (m, 2H,
solved in dichloromethane, the solution was stirred for 15 min OCH2CH), 3.48–3.61 (m, 8H, N(CH2)2O), 3.66–3.73 (m, 2H,
in an ice bath. 1-Ethyl-3-(3′-dimethylaminopropyl)carbodii- OCH2CH). 13C NMR (100 MHz, CDCl3) δ: 173.83, 71.95, 71.76,
mide (EDC) hydrochloride was slowly added dropwise. The sol- 69.90, 69.60, 50.79, 50.75, 48.93, 46.46, 44.26, 44.01, 33.17,
ution was stirred in an ice bath for 30 min and at room 31.96, 29.69, 29.67, 29.59, 29.56, 29.53, 29.38, 25.38, 22.73,
temperature for another 24 h. The solution was washed with 14.16. MALDI-HRMS: m/z 422.2882 ([M + Na]+), C22H41NO5Na+,
saturated aqueous NaCl solution three times. The organic calc. 422.2883.
layer was dried (anhydrous Na2SO4) and the solvent was
4e: Yield 37.3%. 1H NMR (400 MHz, CDCl3) δ: 0.86 (t, J =
removed by rotary evaporation. This crude product was puri- 6.8 Hz, 3H, CH3CH2), 1.20–1.32 (m, 20H, CH3(CH2)10CH2),
fied by column chromatography (silica, CH2Cl2/CH3OH: v/v = 1.55–1.65 (m, 2H, CH2CH2CH2), 2.35 (t, J = 7.2 Hz, 2H,
20/1) to give colorless liquid 3. Compound 3 (9.76 mmol), epi- OCCH2CH2), 2.55–2.62 (m, 2H, OCH2CH), 2.73–2.79 (m, 2H,
chlorohydrin (5.37 g, 58.03 mmol), sodium hydroxide pellets OCH2CH), 3.08–3.13 (m, 2H, OCHCH2), 3.31–3.38 (m, 2H,
(2.32 g, 58.03 mmol), water (0.487 g, 27.08 mmol), and tetra- OCH2CH), 3.56–3.64 (m, 8H, N(CH2)2O), 3.71–3.79 (m, 2H,
butylammonium chloride (0.16 g, 0.484 mmol) were stirred in an OCH2CH). 13C NMR (100 MHz, CDCl3) δ: 173.82, 71.93, 71.74,
ice bath for 30 min and at room temperature for another 12 h. 69.88, 69.59, 50.78, 50.74, 48.92, 46.45, 44.24, 43.99, 33.16,
The solid produced in the reaction was filtered off and washed 31.95, 29.71, 29.68, 29.59, 29.56, 29.52, 29.39, 25.37, 22.72,
with dichloromethane. The combined organic layer was dried 14.15. MALDI-HRMS: m/z 450.3186 ([M + Na]+), C24H45NO5Na+,
with anhydrous magnesium sulfate. The solvent and excess calc. 450.3196.
epichlorohydrin were distilled off to give a yellow oil. The
4f: Yield 41.4%. 1H NMR (400 MHz, CDCl3) δ: 0.86 (t, J = 6.8
residue was purified by silica gel column chromatography Hz, 3H, CH3CH2), 1.19–1.32 (m, 24H, CH3(CH2)12CH2),
(silica, ethyl acetate/petroleum ether: v/v = 3/1) to give product 1.55–1.62 (m, 2H, CH2CH2CH2), 2.34 (t, J = 15.6 Hz, 2H,
4.
OCCH2CH2), 2.53–2.59 (m, 2H, OCH2CH), 2.74–2.79 (m, 2H,
4a: Yield 43.8%. H NMR (400 MHz, CDCl3) δ: 0.88 (t, J = 6 OCH2CH), 3.08–3.13 (m, 2H, OCHCH2), 3.31–3.37 (m, 2H,
1
Hz, 3H, CH3CH2), 1.23–1.37 (m, 4H, CH3(CH2)2CH2), 1.60–1.65 OCH2CH), 3.55–3.65 (m, 8H, N(CH2)2O), 3.68–3.77 (m, 2H,
(m, 2H, CH2CH2CH2), 2.35 (t, J = 7.2 Hz, 2H, OCCH2CH2), OCH2CH). 13C NMR (100 MHz, CDCl3) δ: 173.90, 71.99, 71.80,
2.55–2.65 (m, 2H, OCH2CH), 2.75–2.82 (m, 2H, OCH2CH), 69.95, 69.64, 50.84, 50.80, 48.97, 46.51, 44.31, 44.06, 33.22,
3.08–3.20 (m, 2H, OCHCH2), 3.32–3.37 (m, 2H, OCH2CH), 32.01, 29.79, 29.75, 29.65, 29.62, 29.58, 29.45, 25.43, 22.78,
3.58–3.64 (m, 8H, N(CH2)2O), 3.71–3.78 (m, 2H, OCH2CH). 13C 14.21. MALDI-HRMS: m/z 478.3516 ([M + Na]+), C26H49NO5Na+,
NMR (100 MHz, CDCl3) δ: 173.91, 72.02, 71.82, 69.97, 69.66, calc. 478.3509.
50.86, 50.83, 48.99, 46.53, 44.34, 44.09, 33.18, 31.75, 25.11,
4g: Yield 26.9%. 1H NMR (400 MHz, CDCl3) δ: 0.84 (t, J =
22.65, 14.09. MALDI-HRMS: m/z 338.1888 ([M
+
Na]+), 6.4 Hz, 3H, CH3CH2), 1.17–1.29 (m, 28H, CH3(CH2)14CH2),
1.52–1.62 (m, 2H, CH2CH2CH2), 2.33 (t, J = 7.2 Hz, 2H,
C16H29NO5Na+, calc. 338.1944.
1
4b: Yield 31.5%. H NMR (400 MHz, CDCl3) δ: 0.85 (t, J = 6 OCCH2CH2), 2.52–2.58 (m, 2H, OCH2CH), 2.72–2.77 (m, 2H,
Hz, 3H, CH3CH2), 1.21–1.35 (m, 8H, CH3(CH2)4CH2), 1.55–1.65 OCH2CH), 3.04–3.11 (m, 2H, OCHCH2), 3.28–3.35 (m, 2H,
(m, 2H, CH2CH2CH2), 2.34 (t, J = 7.6 Hz, 2H, OCCH2CH2), OCH2CH), 3.52–3.64 (m, 8H, N(CH2)2O), 3.67–3.76 (m, 2H,
2.52–2.59 (m, 2H, OCH2CH), 2.74–2.79 (m, 2H, OCH2CH), OCH2CH). 13C NMR (100 MHz, CDCl3) δ: 173.82, 71.96, 71.77,
3.06–3.14 (m, 2H, OCHCH2), 3.29–3.37 (m, 2H, OCH2CH), 69.92, 69.62, 50.81, 50.77, 48.94, 46.48, 44.27, 44.02, 33.19,
3.54–3.61 (m, 8H, N(CH2)2O), 3.69–3.77 (m, 2H, OCH2CH). 13C 31.99, 29.76, 29.72, 29.62, 29.59, 29.55, 29.42, 25.39, 22.75,
NMR (100 MHz, CDCl3) δ: 173.87, 71.97, 71.79, 69.93, 69.63, 14.15. MALDI-HRMS: m/z 506.3818 ([M + Na]+), C28H53NO5Na+,
50.83, 50.79, 47.80, 46.07, 44.30, 44.05, 33.19, 31.82, 29.50, calc. 506.3822.
1
29.24, 25.40, 23.43, 14.16. MALDI-HRMS: m/z 366.2261 ([M +
4h: Yield 28.3%. H NMR (400 MHz, CDCl3) δ: 0.87 (t, J =
4.0 Hz, 3H, CH3CH2), 1.17–1.35 (m, 20H, (CH2)10), 1.55–1.65
Na]+), C18H33NO5Na+, calc. 366.2257.
4c: Yield 35.7%. 1H NMR (400 MHz, CDCl3) δ: 0.87 (t, J = 6.4 (m, 2H, CH2CH2CH2), 1.95–2.05 (m, 4H, CH2CH2CH), 2.37 (t,
Hz, 3H, CH3CH2), 1.19–1.31 (m, 12H, CH3(CH2)6CH2), J = 8.0 Hz, 2H, OCCH2CH2), 2.54–2.59 (m, 2H, OCH2CH),
1.55–1.64 (m, 2H, CH2CH2CH2), 2.36 (t, J = 7.6 Hz, 2H, 2.67–2.78 (m, 2H, OCH2CH), 3.07–3.16 (m, 2H, OCHCH2),
OCCH2CH2), 2.58–2.65 (m, 2H, OCH2CH), 2.75–2.82 (m, 2H, 3.31–3.36 (m, 2H, OCH2CH), 3.57–3.62 (m, 8H, N(CH2)2O),
OCH2CH), 3.07–3.18 (m, 2H, OCHCH2), 3.32–3.38 (m, 2H, 3.70–3.78 (m, 2H, OCH2CH), 5.27–5.40 (m, 2H,
OCH2CH), 3.56–3.66 (m, 8H, N(CH2)2O), 3.71–3.79 (m, 2H, CH2CHCHCH2). 13C NMR (100 MHz, CDCl3) δ: 174.02, 130.41,
OCH2CH). 13C NMR (100 MHz, CDCl3) δ: 173.86, 71.97, 71.79, 130.25, 72.21, 72.01, 70.17, 69.86, 51.05, 51.01, 49.17, 46.72,
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 1242–1250 | 1247

View Article Online
Paper
Organic & Biomolecular Chemistry
44.52, 44.27, 33.42, 31.82, 30.07, 30.00, 29.96, 29.83, 29.80, Agarose gel electrophoresis
29.74, 29.62, 29.52, 25.61, 22.99, 14.42. MALDI-HRMS: m/z
LP1–LP9/DNA complexes at different weight ratios ranging
from 0.1 to 1.6 were prepared by adding an appropriate
volume of LP1–LP9 (in 150 mM NaCl solution) to 0.8 mL of
pEGFP-N1 DNA (120 ng mL−1 in 40 mM Tris–HCl buffer sol-
ution). The complexes were diluted by 150 mM NaCl solution
to a total volume of 6 mL, and then the complexes were incu-
bated at 37 °C for 30 min. After that the complexes were elec-
trophoresed on the 0.7% (W/V) agarose gel containing EB and
with Tris-acetate (TAE) running buffer at 150 V for 30 min.
DNA was visualized with a UV lamp using a Bio-Rad Universal
Hood II.
504.3669 ([M + Na]+), C28H51NO5Na+, calc. 504.3665.
4i: Yield 35.9%. 1H NMR (400 MHz, CDCl3) δ: 0.87 (t, J = 6.4
Hz, 3H, CH3CH2), 1.23–1.30 (m, 14H, (CH2)7), 1.57–1.66 (m,
2H, CH2CH2CH2), 1.96–2.07 (m, 4H, CH2CH2CH), 2.35 (t, J =
7.6 Hz, 2H, OCCH2CH2), 2.55–2.59 (m, 2H, OCH2CH),
2.73–2.79 (m, 2H, OCH2CH), 3.07–3.13 (m, 2H, OCHCH2),
3.32–3.37 (m, 2H, OCH2CH), 3.52–3.67 (m, 8H, N(CH2)2O),
3.71–3.78 (m, 2H, OCH2CH), 5.28–5.37 (m, 4H,
CH2CHCHCH2). 13C NMR (100 MHz, CDCl3) δ: 174.06, 130.55,
130.44, 128.33, 128.26, 72.22, 72.04, 69.87, 68.30, 51.07, 51.04,
49.20, 46.74, 44.55, 44.29, 33.44, 31.85, 29.99, 29.76, 29.67,
29.54, 27.56, 27.52, 25.63, 22.90, 14.39. MALDI-HRMS: m/z
502.3512 ([M + Na]+), C28H49NO5Na+, calc. 502.3509.
Ethidium bromide displacement assay
The ability of LP1–LP9 to condense DNA was studied using
ethidium bromide (EB) exclusion assays. Fluorescence spectra
were measured at room temperature in air by a Horiba Jobin
Yvon Fluoromax-4 spectrofluorometer and corrected for the
system response. EB (5 mL, 1 mg mL−1) was put into a quartz
cuvette containing 2.5 mL of 150 mM NaCl solution. After
shaking, the fluorescence intensity of EB was measured. Then
CT DNA (4.08 mL, 6.24 mg mL−1) was added to the solution
and mixed symmetrically, and the measured fluorescence
intensity is the result of the interaction between DNA and EB.
Subsequently, the solutions of LP1–LP9 (1 mg mL−1, 2.5 mL
for each addition) were added to the above solution for further
measurements.
Synthesis and characterization of target lipopolymers LP1–LP9
In a typical procedure, PEI 600, compound 4 and compound 5
(prepared according to ref. 21) and 3 mL of anhydrous ethanol
were mixed in a flask with magnetic stirring and refluxed for
48 h in an oil bath. After the reaction, the mixture was diluted
with 3 mL of anhydrous methanol, and the crude product was
precipitated by the addition of anhydrous dichloromethane–
cyclohexane (v/v = 1/1). The precipitate was collected and dried
in a vacuum to get the product as a colorless oil. The molecu-
lar weights of compounds LP1–LP9 were measured by GPC.
Acid–base titration
Polymer degradation study
In this assay, briefly, LP1–LP9 (0.25 mmol of amino groups)
were dissolved in 5 mL of 150 mM NaCl aqueous solution, and
1 N HCl was added to adjust the pH to 2.0. Aliquots (50 mL for
each) of 0.1 M NaOH were added, and the solution pH was
measured with a pH meter (pHS-25) after each addition. For
comparison, PEI (25 kDa) was used under the same experi-
mental conditions.
For the polymer degradation study, LP2 (Mw = 31 109, smaller
than that used for transfection) was dissolved in a single-
strength (1×) phosphate-buffered saline (PBS) solution and was
constantly shaken in a 37 °C incubator at 100 rpm. The LP2
solution was withdrawn at different time points and then lyo-
philized. The relative molecular mass of degraded products
was determined by GPC using a Waters 515 pump and a
Waters 2410 refractive index detector (25 °C).
Cell culture
Human non-small-cell lung carcinoma A549 cells were incu-
bated, respectively, and 1640 medium containing 10% fetal
bovine serum (FBS) and 1‰ antibiotics (penicillin–strepto-
mycin, 10 000 U mL−1) at 37 °C in a humidified atmosphere
containing 5% CO2.
Particle size and zeta potential measurements
Particle size and zeta potential measurements of polyplexes
were carried out using a Nano-ZS 3600 (Malvern Instruments,
USA) with a He–Ne laser beam (633 nm, fixed scattering angle
of 901°) at 25 °C. LP1–LP9/DNA polyplexes at weight ratios
ranging from 0.2 to 6.4 were prepared by the same method
with above-mentioned agarose gel electrophoresis. After
30 min incubation in 100 mL ultrapure water, polyplex sol-
utions were diluted to a final volume of 1 mL before
measurements.
Amplification and purification of plasmid DNA
pGL-3 and pEGFP-N1 plasmids were used. The former one as
the luciferase reporter gene was transformed in E. coli JM109
and the latter one as the enhanced green fluorescent protein
gene was transformed in E. coli DH5α. Both plasmids were
amplified in Luria–Bertani broth at 37 °C overnight. The plas-
mids were purified by an EndoFree Tiangen TM Plasmid Kit.
Cell viability assay
Then the purified plasmids were dissolved in TE buffer sol- Toxicity of LP1–LP9 toward A549 cells was determined by
ution and stored at −20 °C. The integrity of plasmids was con- using MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazo-
firmed by agarose gel electrophoresis. The purity and lium bromide) reduction assay following literature procedures.
concentration of plasmids were determined by ultraviolet (UV) The A549 cells (5000 cells per well) were seeded into 96-well
absorbance at 260 and 280 nm.
plates. The cells were then incubated in a culture medium
1248 | Org. Biomol. Chem., 2013, 11, 1242–1250
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
containing LP1–LP9 with a particular concentration for 48 h. final volume of 100 mL. Then the complexes were incubated at
After that, the medium was replaced with 200 μL of fresh room temperature for 30 min. The plates were washed with
medium, and 20 μL of sterile filtered MTT (5 mg mL−1) stock PBS twice, and 100 mL LP1–LP9/DNA complexes were then
solution in PBS was added to each well. After 4 h, the added to a well with an additional 150 mL of medium without
unreacted dye was removed by aspiration. The formazan crys- FBS. The final concentration of DNA in the complexes was cal-
tals were dissolved in 150 μL DMSO per well and measured culated to be 6 mg mL−1. After 4 h of incubation, the LP1–LP9/
spectrophotometrically in an ELISA plate reader (model 550, DNA complex containing medium was replaced with 0.5 mL of
Bio-Rad) at a wavelength of 570 nm. The cell survival was fresh medium containing 10% FBS, and the cells were further
expressed as follows: cell viability = (ODtreated/ODcontrol) × incubated for 24 h at 37 °C. The cells were directly observed by
100%.
an inverted microscope (Nikon ECLIPSE Ti-U). The microscopy
Cytotoxicity tests of the complexes were carried out by images were obtained at a magnification of 100× and recorded
determining the percentage of cell viability using human non- using Viewfinder Lite (1.0) software (NIS-Element s F 3.2).
small-cell lung carcinoma A549 cells. 100 μL A549 cells were
seeded at 1.0 × 104 cells per well in 96-well plates (Falcon, BD).
Complexes containing 0.15 μg pDNA in 15 Ml Opti-MEM® I
Acknowledgements
medium were added to each well. After incubation for 48 h,
10 μL CCK-8 was added to each well and the plates were incu-
bated at 37 °C for another 4 h. Absorbance was measured at a
wavelength of 490 nm and a reference wavelength of 630 nm
using a microplate reader 550 (Bio-Rad).
This work was financially supported by the National Program
on Key Basic Research Project of China (973 Program,
2012CB720603, 2013CB328900) and the National Science
Foundation of China (Nos. 20902062, 21232005). J. Z. thanks
the Program for New Century Excellent Talents in University
(NCET-11-0354). We also thank Analytical & Testing Center of
Sichuan University for structural analysis of the compounds.
In vitro transfection
Luciferase assay. The 25 kDa PEI was used as the positive
control due to its high TE in vitro and in vivo. The plasmid
pGL-3 was used as a reporter gene. Transfections of pGL-3
plasmid mediated by LP1–LP9 in A549 cells were studied as
compared with 25 kDa PEI. A549 cells were seeded at a density
of 6 × 104 cells per well in the 24-well plate with 0.5 mL of
medium containing 10% FBS and incubated at 37 °C for 24 h.
Then the complexes were prepared at different weight ratios by
adding 2 μg of plasmid DNA to an appropriate volume of the
LP1–LP9 solution. Before transfection, the cells were washed
with a serum-free or serum medium, and then the LP1–LP9/
DNA complexes were added with the serum-free or serum
medium for 4 h at 37 °C. Then the serum-free or serum
medium was replaced by the flash medium containing 10%
FBS, and the cells were further incubated for 24 h. After that,
the protein was extracted. The luciferase assay was performed
according to manufacturer’s protocols (Promega). Relative
light units (RLUs) were measured with a chemiluminometer
(FLOROSKAN ASCENT FL). The total protein was measured
according to a BCA protein assay kit (Pierce). Luciferase
activity was expressed as RLU per mg protein. Data are shown
as mean standard deviation (SD) based on 3 independent
measurements. The statistical significance between two sets of
data was calculated using Student’s t-test. A P value of 0.05 was
considered statistically significant.
Notes and references
1 X. Zhang and W. T. Godbey, Adv. Drug Delivery Rev., 2006,
58, 515–534.
2 S. C. De Smedt, J. Demeester and W. E. Hennink, Pharm.
Res., 2000, 17, 113–126.
3 G. Y. Wu and C. H. Wu, J. Biol. Chem., 1987, 262, 4429–
4432.
4 M. A. Mintzer and E. E. Simanek, Chem. Rev., 2009, 109,
259–302.
5 R. Laga, R. Carlisle, M. Tangney, K. Ulbrich and
L. W. Seymour, J. Controlled Release, 2012, 161, 537–553.
6 O. Boussif, F. Lezoualc’h, M. A. Zanta, M. D. Mergny,
D. Scherman, B. Demeneix and J.-P. Behr, Proc. Natl. Acad.
Sci. U. S. A., 1995, 92, 7297–7301.
7 F. Dai and W. Liu, Biomaterials, 2011, 32, 628–638.
8 R. Qi, X. Hu, L. Yan, X. Chen, Y. Huang and X. Jing, J. Con-
trolled Release, 2011, 152, E167–E168.
9 T.-H. Kim, H.-L. Jiang, D. Jere, I.-K. Park, M.-H. Cho,
J.-W. Nah, Y.-J. Choi, T. Akaike and C.-S. Cho, Prog. Polym.
Sci., 2007, 32, 726–753.
Green fluorescent protein assay. Transfections of the 10 J. F. KukowskaLatallo, A. U. Bielinska, J. Johnson,
pEGFP-N1 plasmid mediated by LP1–LP9 in A549 cells were
also evaluated. The best weight ratios in A549 cells determined
R. Spindler, D. A. Tomalia and J. R. Baker, Proc. Natl. Acad.
Sci. U. S. A., 1996, 93, 4897–4902.
from the luciferase assay were used. A549 cells were incubated 11 R. Kircheis, L. Wightman and E. Wagner, Adv. Drug Delivery
at a density of 80 000 cells per well in 24-well plates, respect- Rev., 2001, 53, 341–358.
ively, 24 h prior to transfection. LP1–LP9/DNA complexes were 12 T. G. Park, J. H. Jeong and S. W. Kim, Adv. Drug Delivery
prepared by adding an appropriate volume of LP1–LP9 solu- Rev., 2006, 58, 467–486.
tion (in 5% glucose solution) to 50 mL pEGFP-N1 DNA solu- 13 W. T. Godbey, K. K. Wu and A. G. Mikos, J. Biomed. Mater.
tion (30 mg mL−1 in 40 mM Tris–HCl buffer solution) with a
Res., 1999, 45, 268–275.
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 1242–1250 | 1249

View Article Online
Paper
Organic & Biomolecular Chemistry
14 K. Kunath, A. von Harpe, D. Fischer, H. Peterson, U. Bickel, 21 S. Li, Y. Wang, J. Zhang, W.-H. Yang, Z.-H. Dai, W. Zhu and
K. Voigt and T. Kissel, J. Controlled Release, 2003, 89, 113–
125.
15 X. T. Shuai, T. Merdan, F. Unger, M. Wittmar and T. Kissel,
Macromolecules, 2003, 36, 5751–5759.
16 S. H. Pun, N. C. Bellocq, A. J. Liu, G. Jensen, T. Machemer,
E. Quijano, T. Schluep, S. F. Wen, H. Engler, J. Heidel and
M. E. Davis, Bioconjugate Chem., 2004, 15, 831–840.
17 Y.-X. Sun, X. Wang, S.-X. Cheng, X.-Z. Zhang and
R.-X. Zhuo, J. Controlled Release, 2008, 128, 171–178.
18 M. A. Gosselin, W. J. Guo and R. J. Lee, Bioconjugate Chem.,
2001, 12, 989–994.
X.-Q. Yu, Mol. BioSyst., 2011, 7, 1254–1262.
22 Z. Liu, Z. Zhang, C. Zhou and Y. Jiao, Prog. Polym. Sci.,
2010, 35, 1144–1162.
23 C. Fu, L. Lin, H. Shi, D. Zheng, W. Wang, S. Gao, Y. Zhao,
H. Tian, X. Zhu and X. Chen, Biomaterials, 2012, 33, 4589–
4596.
24 J. Tu, T. Wang, W. Shi, G. Wu, X. Tian, Y. Wang, D. Ge and
L. Ren, Biomaterials, 2012, 33, 7903–7914.
25 I. Y. Farber and A. J. Domb, Mater. Sci. Eng., C., 2007, 27,
595–598.
26 The effect of LPs/DNA weight ratios on the TE was pre-
investigated by green fluorescent protein assays, and
results showed that highest density of transfected cells was
found under the ratio of 0.8 (images not shown, also see
Fig. S4†).
19 C. Brus, H. Petersen, A. Aigner, F. Czubayko and T. Kissel,
Bioconjugate Chem., 2004, 15, 677–684.
20 K. Wong, G. B. Sun, X. Q. Zhang, H. Dai, Y. Liu, C. B. He
and K. W. Leong, Bioconjugate Chem., 2006, 17, 152–158.
1250 | Org. Biomol. Chem., 2013, 11, 1242–1250
This journal is © The Royal Society of Chemistry 2013