N -Isopropylacrylamide-Modified Polyethylenimines as Effective Gene Carriers

Full Paper

1680

N-Isopropylacrylamide-ModifiedPolyethylenimines as Effective Gene Carriers

Huayu Tian, Feifan Li, Jie Chen, Yubin Huang, Xuesi Chen*

25 kDa branched polyethylenimines are modified by N-isopropylacrylamide via Michaeladdition. An agarose gel retardation assay shows that all derivatives display good bindingaffinity toward plasmid DNA. The modified PEI-25K shows lower cytotoxicity in MTT assayand better transfection efficiency than unmodified PEI-25K in HeLa cells. The endocytosis efficiency of theoptimized complexes is determined to be 99.9% by flowcytometry. More interestingly, although the derivativesare not designed to conjugate with targeting ligands ornuclear localization signals, confocal laser scanningmicroscopy (CLSM) demonstrates that the optimizedderivative results in increased endocytosis and stronglyenhanced nuclear uptake compared with PEI-25K.

1. Introduction

Gene therapy holds big promise in treating inherited and

acquired diseases or disorders.[1,2] However, despite dec-

ades of research efforts, significant improvements are still

needed before the full potential of this therapeutic modality

can be realized.[3] Lack of safe, reliable, less toxic and highly

efficient gene carriers is still a bottleneck that limits the

further development of gene therapy. Although recombi-

nant viral carriers are proven to be highly efficient,[4] their

limited gene capacity, high cost, immunogenicity, and

pathogenicity have led to increased efforts to develop

nonviral gene vectors.[5,6] Synthetic nonviral vectors are an

attractive alternative owing to their low production cost,

definite physicochemical properties, large gene loading

capacity, flexible designability and potential safety. However

their transfection efficiency is still relatively low. Polyethyl-

enimine (PEI) has become a well-studied and commercially

Dr. H. Tian, F. Li, J. Chen, Prof. Y. Huang, Prof. X. ChenKey Laboratory of Polymer Ecomaterials, Changchun Institute ofApplied Chemistry, Chinese Academy of Sciences, Changchun130022, ChinaE-mail: [email protected]

Macromol. Biosci. 2012, 12, 1680–1688

� 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim wileyonline

readily available nonviral cationic polymer for plasmid

transfection since it was initially introduced for plasmid DNA

(pDNA) delivery and 25 kDa branched PEI (PEI-25K) has been

shown as one of the most efficient polymers for pDNA

delivery to date.[7–13] However, compared with viral vectors,

PEI is known to have a less desirable cytotoxicity profile and

relatively low gene delivery efficiency. Thus, improvements

are needed in order to bring this delivery vehicle successfully

to patient care.[14]

Past efforts in modifications of PEI have led to both

enhanced and diminished transfection efficiency. Zhang’s

group found that biotinylated PEI/avidin bioconjugate and

its DNA complexes demonstrated much lower cytotoxicity

and higher transfection efficacy in HepG2 cells.[15] Our

previous work showed that modification of PEI with

hydrophobic amino acid segments also led to improved

transfection efficiency.[16] Wagner et al. showed that simple

modifications of branched PEI by the introduction of ethyl

acrylate, acetyl functionality, or the introduction of a

negatively charged acid led to highly efficient small

interfering RNA (siRNA) delivery.[17] Bae’s work demon-

strated that coupling poly(e-caprolactone) to branched PEI

via an amide group improved gene transfection efficiency.[18]

Ramezani et al. reported that conjugation of a series of

library.com DOI: 10.1002/mabi.201200249

N-Isopropylacrylamide-Modified Polyethylenimines as Effective Gene Carriers

www.mbs-journal.de

alkyloligoamine derivatives to 10 kDa branched PEI

improved the transfection efficiency slightly.[19] Schat-

zlein’s work demonstrated that conjugation of palmitic acid

and methylation of PEI-25K significantly reduced its

transfection efficiency.[20] Klibanov’s group found that

permethylation, perethylation, and acylation of PEI-25K

significantly decreased the transfection efficiency and that

only some of the obtained derivatives maintained equiva-

lent transfection efficiency.[21] Park et al. showed that

conjugation of myristic acid to PEI-25K improved the

transfection efficiency by about 2-fold,[22] and Kim’s work

illustrated that conjugation of cholesterol to PEI-25K also

increased the transfection efficiency to a similar extent.[23]

Putnam et al. reported that acetylation, butylation, and

hexylation of PEI-25K improved the transfection efficiency

to nearly 4-fold in HeLa cells in vitro.[24] Almost all the

above-mentioned modifications reduced the cytoxicity to

some degree. At the same time, almost all of these

approaches emphasized modifications of only the primary

amines or largely the primary amines of PEI via amide

groups or reduction of non-protonated amines.

In the present study, PEI-25Ks were modified via Michael

addition with N-isopropylacrylamide (NIPA). Our approach

is distinctive compared to the previously reported and

above-mentioned ones in that: (1) modifications are largely

involved with secondary amines instead of primary amines,

which is consistent with the reactivity sequence of original

secondary amine>primary amine>> secondary amine

formed by reaction of primary amine;[25] and (2) the

content of protonated amines was not reduced through

the modification as no new amide groups were formed. As

such, the overall modification strategy turned secondary

amines into tertiary amines and primary amines into

secondary amines. We postulate that the benefit of this

approach is to enable the formation of stable complexes

between the PEI derivatives and pDNA through electro-

static interactions, while at the same time, such modifica-

tion improved hydrophobicity, which in turn, would

increase the interaction of carrier/DNA complexes with

cells membranes. More importantly, it is noteworthy

that the optimized PEI-25K derivative showed improved

endocytosis and enhanced nuclear uptake compared with

PEI-25K. To the best of our knowledge, this is the first

report on the preferential transportation of pDNA into

cell nuclei by employing NIPA modification and without

any nuclear localization signals, although the precise

mechanism remains to be explored.

2. Experimental Section

2.1. Materials

Hyperbranched PEI with a weight-average molecular weight of

25 kDa (PEI-25K) and NIPA were purchased from Aldrich (St. Louis,

www.MaterialsViews.com

Macromol. Biosci. 2012, 1

� 2012 WILEY-VCH Verlag GmbH

MO, USA). 2,4,6-Trinitrobenzenesulfonic acid (TNBS) was also

obtained from Aldrich as a 5% (w/v) aqueous solution. Chloroform

was purchased from Sinopharm Chemical Reagent Co., Ltd.

(Shanghai, China). Dulbecco’s modified Eagle medium (DMEM)

and fetal bovine serum (FBS) were purchased from Gibco (Grand

Island, USA). 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium

bromide (MTT) was purchased from Amresco (Solon, Ohio, USA).

The plasmid encoding firefly luciferase (pGL3), cell lysis buffer and

luciferin substrate were obtained from Promega (Madison, WI,

USA). Calf-thymus DNA and 40,6-diamidino-2-phenylindole dihy-

drochloride (DAPI) were purchased from Sigma (St. Louis, MO, USA).

Fluorescein isothiocyanate (FITC) was purchased from Dingguo

Bio-Technology Company (Beijing, China). Dialyses were carried

out by using SpectraPor CE dialysis membranes with a molecular

mass cut-off of 3500 Da (GreenBird, China). Other materials and

solvents were used as received without further purification.

2.2. Modification of PEI-25K by NIPA

About 1 g of PEI-25K was dissolved in 5 mL of chloroform, and then

the desired amount of NIPA based on molar ratios was added to

the solution with stirring. The solution was heated to 40 8C and

maintained for 96 h. Upon completion of the reaction, chloroform

was removed by rotary evaporation. The crude product was

dissolved in 5 mL of water, dialyzed against water, filtrated, and

lyophilized. The yield was 95%. Three derivatives, PEN88, PEN175

and PEN350 were prepared and named according to the feed molar

ratio of NIPA to PEI-25K (88:1, 175:1 and 350:1, respectively).

2.3. Determination of Primary Amine Content

The number of primary amines in the PEI-25K derivatives was

determined by the TNBS assay[26,27] and the number of grafted

NIPA via primary amines was calculated by the difference in the

number of primary amines between PEI-25K and the derivative. For

purpose of this calculation, the polydispersity of PEI-25K was

assumed as 1; the branched PEI should contain 25% primary

amines, 50% secondary amines, and 25% tertiary amines.[23] The

number of primary amines in the PEI-25K derivatives is equal to

that in the equivalent PEI-25K to get the following equation:

a

25000þmn� 25000

43� 4� x

� �¼ b

25000� 1

4

25000

43

where a is the weight of PEI-25K derivatives, m is the molecular

weight of NIPA, n is the number of NIPA in the PEI-25K derivatives

calculated from 1H NMR spectroscopy, x is the number of reduced

primary amines in the PEI-25K derivatives, b is the equivalent

weight of PEI-25K obtained by interpolation of the standard curve,

25 000 is the molecular weight of PEI-25K, 43 is the average

molecular weight of the repeating unit of PEI, and 1/4 is the

fraction of primary amines in the branched PEI.

2.4. Determination of Buffering Capacity

The buffering capacity of the PEI-25K and its derivatives in the pH

range of 12 to 2 was determined by acid-base titration.[28] Briefly,

20 mg of polymer material was dissolved in 20 mL of water to get a

1 mg �mL�1 solution, and the pH was adjusted to 12 by 4 M NaOH

2, 1680–1688

& Co. KGaA, Weinheim1681

1682

www.mbs-journal.de

H. Tian, F. Li, J. Chen, Y. Huang, X. Chen

solution. The solution was then titrated with 1 M HCl solution in

10 mL increments with stirring until pH¼2 was reached. Each data

point was collected at 10 min intervals with a pH meter after the

addition of acid solution.

2.5. Particle Size and Zeta Potential Measurements

The calf thymus DNA was used to form complexes with the cationic

polymers for particle size and zeta potential measurements. The

complex solutions were prepared by adding an equal volume of the

DNA solution to a polymer solution at various carrier/DNA ratios and

mixing byvortex. Then the solution was incubated for 20 min atroom

temperature. The particle size and Zeta potential were measured by

using a Zetasizer Nano ZS90 analyzer (Malvern Instruments, UK).

2.6. Gel Retardation Assay

The plasmid DNA (pGL3) solution was diluted to 0.1 mg �mL�1 for

the gel retardation assay. The complex solutions were prepared by

vortex mixing of various carrier/DNA weight ratios. After 20 min

incubation at room temperature, 10 mL of each complex solution

was analyzed by 1% (w/v) agarose gel electrophoresis (100 V, 1 h).

2.7. Cell Culture

Cells were cultured in the DMEM medium with high glucose,

supplemented with 10 vol% heat-inactivated FBS, 100 units �mL�1

penicillin, and 100 mg �mL�1 streptomycin in a 5% CO2 incubator at

37 8C under 95% humidity.

2.8. Cytotoxicity Assay

The cytotoxicity of the derivatives was assessed in comparison

with PEI-25K by an MTT assay. In brief, HeLa cells were seeded at

1.0� 104 cells per well in 96-well plates, and then cultured for

24 h. Various concentrations of PEI-25K and its derivatives were

prepared with phosphate-buffered saline (PBS) (pH¼ 7.4). Followed

by the addition of 20 mL of each polymer solution to each well, the

plate was incubated for an additional 24 h. At the end of the

experiments, 20 mL of the MTT solution (5 mg �mL�1 in PBS) was

added to each well and the plate was returned to the incubator. After

4 h, the MTT solution was carefully removed from each well, and

200 mL of dimethyl sulfoxide (DMSO) was added to dissolve the MTT

formazan crystals. The plate was incubated for an additional 10 min

before the absorbance at 492 nm was recorded by a microplate reader

(Bio-Rad). The cell viability (%) was calculated according to

cell viability ð%Þ ¼ ðAsample=AcontrolÞ � 100

where Asample is the absorbance of the polymers treated cells and

Acontrol is the absorbance of the untreated cells. Each experiment

was done in triplicate.

2.9. In vitro Transfection

Cells were seeded in 96-well plates at an initial density of 1.0�104

cells per well with 200 mL of DMEM. The plate was incubated at

37 8C in 5% CO2 until it reached 80% confluence, and then the

Macromol. Biosci. 201

� 2012 WILEY-VCH Verlag Gmb

culture medium was aspirated from each well and replaced with

180 mL of fresh DMEM containing 10% FBS, and immediately, 20 mL

of a complex solution (0.2 mg pGL3) per well at various carrier/DNA

ratios was added. The cells were then incubated for additional 48 h.

Then the medium was removed and the cells were gently washed

3 times with PBS. After thorough lysis of the cells with a cell lysis

buffering solution, the luciferase activity was determined by

detecting the light emission from an aliquot of the cell lysate

incubated with 100 mL of a luciferin substrate in a luminometer

(GloMaxTM 20/20, Promega). The protein content of the cell lysate

was determined by using a bicinchonidinic acid (BCA) protein

assay kit (Pierce). All experiments were carried out in triplicate

to ensure reproducibility.

2.10. Endocytosis Assay

Endocytosis assays for PEI-25K and PEN175 were conducted by flow

cytometry at the optimized carrier/DNA weight ratio determined

by fluorescence microscopy and quantitative transfection assay.

HeLa cells were seeded in 6-well plates at an initial density of

2.0�105 cells per well with DMEM containing 10% FBS. The plate

was incubated at 37 8C in 5% CO2 until it reached 80% confluence.

Then the culture medium was replaced with fresh DMEM

containing 10% FBS and complexes prepared from FITC labeled

PEN175 or PEI-25K with pGL3 at the respective optimal carrier/

pDNA weight ratio were added. The cells were incubated for

additional 48 h. Then the culture medium was aspirated from each

well and the plate was washed once with PBS. The cells were

detached, gathered and finally resuspended in PBS (pH¼ 7.4).

Endocytosis efficiency was evaluated as the percentage of cells

with FITC using a FACS Calibur System from Becton–Dickinson

(San Jose, CA).

2.11. Confocal Laser Scanning Microscope (CLSM)

Observations

HeLa cells were seeded on 11 mm glass cover slips in 6-well plates at

an initial density of 2.0� 105 cells per well with DMEM containing

10% FBS. The plate was incubated at 37 8C in 5% CO2 until it reached

80% confluence. Then the culture medium was replaced with fresh

DMEM containing 10% FBS and complexes prepared from FITC

labeled PEN175 or PEI-25K with pGL3 at the respective optimal

carrier/pDNA weight ratio were added. Cells were allowed to

internalize complexes for 2.5 h and 5 h prior to being fixed with

3.7% paraformaldehyde (PFA). The cover slips were then washed

3 times with PBS, and the cell nuclei were stained with 1 mL of

DAPI (1 mg �mL�1) for 10 min. The cover slips were then washed

several times with PBS and enclosed in glycerol and visualized by

CLSM (Leica TCS SP2).

3. Results and Discussion

3.1. Synthesis and Characterization of PEI Derivatives

The synthetic procedures for the modifications of PEI-25K

are presented in Figure 1. The modifications were carried

out through a Michael addition and the reaction was

2, 12, 1680–1688

H & Co. KGaA, Weinheim www.MaterialsViews.com

H2NHN

NNH2

NH2

m +HN

OPEI-25K

n

HNO

H2N NH

NNH

NH

ab

NNH2c

NH

O

NH

Figure 1. Synthesis of the derivatives.

4.0 3.5 3.0 2.5 2.0 1.5 1.0

Derivativesd c

b

HNHN

O

abc

da

a

ppm

H2NHN

NNH2

NH2

nm

PEI

Figure 2. 1H NMR spectra of PEI and its derivatives in CDCl3.


www.mbs-journal.de

straightforward as it does not need any strict conditions,

catalysts or activator reagents such as N-(3-dimethylami-

nopropyl)-N0-ethylcarbodiimide hydrochloride, dicyclo-

hexylcarbodiimide or N-hydroxysuccinimide. The 1H NMR

spectra provided structural information and established

the formation of the NIPA derivatives (Figure 2).

The polydispersity of PEI-25K was presumed to be 1.

The branched PEI should contain 25% primary amines, 50%

secondary amines, and 25% tertiary amines.[23] Therefore,

there were 145.4 primary amines, 290.7 secondary amines

and 145.4 tertiary amines in each PEI-25K molecule.

The number of grafted NIPA in each derivative was

calculated from its 1H NMR spectrum (Table 1). The number

of primary amines in each polymer was determined by

the TNBS assay, whereas the number of grafted NIPA via

primary amines was calculated by the difference in the

number of primary amines between PEI-25K and the

derivative.[26,27] The number of reacted primary amines is

equivalent to the number of grafted NIPA via primary

amines. The number of reacted secondary amines was

calculated by the difference between the number of grafted

NIPA PEI-25K and reacted primary amines. As far as we

know, amine groups’ reactivity sequence for the Michael

Table 1. Characterization of derivatives of PEI-25K.

Derivative NIPA to PEI-25K

feed molar ratio

No. of grafted

NIPAa)

PEN88 88:1 79.8

PEN175 175:1 128.6

PEN350 350:1 131.1

a)The number of the grafted NIPA units in each derivative molecule w

reacted primary amines was determined by a TNBS assay; c)Compare




addition follows in the order of: original secondary (28)amine>primary (18) amine>> secondary amine formed

by the reaction of primary amine.[25] Therefore, the

number of secondary amines undergoing the Michael

addition was much higher than that of primary amines

even at a high feed molar ratio of NIPA to PEI-25K. As a

result, the content of tertiary amines significantly increased

in the derivatives, but the content of secondary amines

decreased while the content of primary amines did not

change significantly. It is noteworthy that the number of

reacted amines and primary amines for PEN175 was not

much different from that for PEN350 in spite of the large

difference in feed molar ratios of NIPA to PEI-25K. Up to a

certain level, there seemed to be a limit on the conjugation

efficiency of NIPA with PEI-25K, presumably caused by

the steric effect between isopropyl groups of NIPA and

PEI’s closer structure.

3.2. Buffering Capacity of PEI-25K and its Derivatives

Buffering capacity is considered as one of the most

important properties for the release of carrier/DNA

complexes from endosomes or lysosomes. Acid-base

titration experiments were carried out at the same polymer

weight concentration to evaluate the buffering capacity of

No. of reacted

1- aminesb)

2- Amine

contentc)

1- Amine

contentc)

3.6 74% 98%

12.4 60% 91%

17.6 61% 88%

as derived from data of 1H NMR spectroscopy; b)The amount of

d with PEI-25K.

2, 1680–1688


0 100 200 300 400 500 600

2

4

6

8

10

12 PEI-25K PEN175 PEN350

pH

HCl (1M)

Figure 3. Titration curves for aqueous solutions of PEI-25K and itsderivatives.

1684

www.mbs-journal.de


the prepared derivatives. A larger amount of HCl solution

is required to alter the pH of cationic polymer solutions

with higher buffering capacity. As seen in Figure 3, different

trends of changes in the buffering capacities of the

derivatives were observed over different pH ranges. After

Figure 4. Results of gel retardation assay for PEI-25K and PEN polym



modification, the buffering capacity of the derivatives was

improved in the pH range from 12.0 to 8.1, while it was

reduced over the pH range from 8.1 to 2.0. This could be a

result, respectively, of an increase in the tertiary amines

content and decreases in secondary and primary amines

contents. However, the overall buffering capacity of the

polymers slightly declined. In addition, PEN175 and PEN350

showed very similar buffering capacity despite of large

differences in the feed ratios. This suggests that there is

a plateau effect in structural and functional changes to

PEI-25K by NIPA when exceeding a certain level of NIPA.

3.3. Gel Retardation Assays

The ability to bind and condense DNA into nano-sized

complex particles is a necessary prerequisite for efficient

gene delivery vectors. The binding affinity was determined

by their electrophoretic mobility on agarose gels at various

carrier/pDNA weight ratios when PEI-25K was used as

the positive control. Complexation was inferred from the

retardation of pDNA mobility and the critical carrier/pDNA

weight ratios for completely retardation-reflected binding

affinity to pDNA. As shown in Figure 4, the critical weight

ers.

2, 12, 1680–1688


0

50

100

150

200

660

PEI-25K PEN88 PEN175 PEN350

Parti

cle

size

(nm

)

[Carrier/DNA] weight ratios

0.2 0.4 0.8 2.0 0.2 0.4 0.8 2.0 0.2 0.4 0.8 2.0 0.2 0.4 0.8 2.0

Figure 5. Particle sizes of complexes formed from PEI-25K and itsderivatives with calf thymus DNA in water.

-40-30-20-10

0102030405060

0.2 0.2 0.2 0.4 0.8 2.0 0.4 0.8 2.0 0.4 0.8 2.0 0.2 0.4 0.8 2.0

Zeta

Pot

entia

l (m

v)


[Carrier/DNA] Weight Ratios

Figure 6. Zeta potentials of complexes formed of PEI-25K and itsderivatives with calf thymus DNA in water.


www.mbs-journal.de

ratio for PEI-25K was 0.2 and that for the PEN polymers was

0.3. This small difference suggests that the binding affinity

and condensation ability of the PEN polymers changed only

slightly with increasing grafted NIPA contents. Therefore,

all the PEN polymers appear to have good binding affinity to

pDNA and could condense pDNA into compact complex

particles.

3.4. Particle Size and Zeta Potential of Complexes

Polycations can condense negatively charged DNA into

positively charged nano-sized particles through electro-

static interactions. However, transport of the complexes

into target cells requires passage across different barriers,

ranging in sizes from micrometers to nanometers.[29,30] The

moderate surface positive charge and proper particle size of

carrier/DNA complexes are closely related to endocytosis

and further transfection. As shown in Figure 5, complexes of

PEI-25K or PEN polymers with calf thymus DNA in aqueous

solution became more compact with increasing carrier/

DNA weight ratios and finally had a size of about 140 nm.

Particles in the micrometer range are too large for cell

internalization. The desirable complexes in vitro should be

small, with sizes from tens to hundreds of nanometers,

and be moderately compact for protecting and releasing

gene cargo.[31] Moderately compact complexes are bene-

ficial to protecting gene cargo from destruction by enzymes

in cells and also good for systematic circulation and

effective endocytosis. However, too compact complexes

would be a disadvantage for effective release of gene cargo

in the cytoplasm. Data shown in Figure 5 indicate that the

particle sizes of complexes prepared from PEI-25K and its

derivatives with DNA were just in the desirable range for

gene delivery.

In addition, zeta potentials also play an important role in

endocytosis and material toxicity. A moderately positively

charged complex would be in favor of effective endocytosis




because of the existence of negatively charged cell

membranes. On the other hand, a highly positively charged

complex or one with a high charge density could contribute

to cell cytotoxicity.[32] As shown in Figure 6, with the

increase in carrier/pDNA weight ratios, the charge of

complexes gradually shifted from negative to positive, and

then gradually increased and eventually stabilized in the

þ30 to þ40 mV range. Some derivative/DNA complexes

showed higher zeta potentials than PEI-25K/DNA, and

some showed lower ones. There was no consistent trend of

potential change with change in grafted NIPA contents of

the derivatives.

3.5. Cell Toxicity Assays

Cationic polymers have been reported to undermine cell

membranes via electrostatic interactions between polyca-

tions and the plasma membranes.[32] A high positive charge

and a high positive charge density are generally regarded as

the major contributors to cellular toxicity. The cytotoxicity

of PEI-25K and its derivatives after 24 h incubation was

determined by MTT assays in HeLa cells. The cytotoxicity

profiles of the derivatives and PEI-25K shown in Figure 7

demonstrated that the cell viability for the derivatives at

various concentrations were higher than that for PEI-25K,

indicating lower cellular toxicity of the derivatives. It was

observed that the cell viability values were more than

50% for all the derivatives with a concentration lower than

20 mg �mL�1, but it was less than 25% for PEI-25K. It was also

found that the cell viability for derivatives increased with

increasing grafted NIPA contents. The reduced cytotoxicity

might be due to the increased gap between the protonated

amines and plasma membranes through the introduction

of NIPA. They weakened the interactions between the

polycations and the plasma membranes. In conclusion, the

derivatives in this study were less toxic than PEI-25K at

available concentrations and that would be beneficial to

effective transfection.

2, 1680–1688


0 10 20 30 40 500

20

40

60

80

100C

ell V

iabi

lity

(%)

Polymer Concentation (µg/mL)


Figure 7. Cellular toxicity of PEI-25K and its derivatives in HeLacells.

1686

www.mbs-journal.de


3.6. Transfection Activity and Intracellular Uptake

Transfection efficiency of PEI-25K and the derivatives

for plasmid DNA was investigated in HeLa cells in vitro.

Gene transfection efficiency was measured as the luciferase

enzyme activity and the relative light units (RLU) were

normalized to the total protein content. Transfection results

for PEI-25K and its derivatives were provided in Figure 8. It

showed that the PEN polymers had much better transfection

efficiency than PEI-25K at their optimized weight ratios.

The optimized transfection efficiency of PEN175 as the

most efficient PEN polymer was 18 fold higher than that of

PEI-25K which was established as one of the most efficient

commercial polymers for plasmid delivery. In addition, the

optimized carrier/pDNA weight ratio was 2.5 for PEI-25K

but 10 for the PEN polymers. It is generally accepted that the

high transfection efficiency of PEI-25K is derived from the

‘‘proton sponge effect’’.[8] The critical endosomal pH range is

about 7.0 to 5.5, in which buffering capacity caused the

rupture of endosomes by osmotic swelling. The buffering

capacity in that pH range has been shown to play an

important role in high transfection efficiency. However, the

106

107

108

109

1010

2.5 5 10 20 2.5 5 10 20

RLU

/mg

prot

ein

PEN175PEN88

PEI-25K

5 2.5 1.25

[Carriers/pDNA] weight ratios

Figure 8. Transfection efficiency of PEI-25K and the PEN polymers.



buffering capacity of the PEN polymers in that pH range

decreased. In order to get enough buffering capacity to

rupture endosomes for high transfection, more PEN

polymers than PEI-25K would be required, thus, it is not

surprising that the optimized carrier/pDNA weight ratios

for the PEN polymers were higher than that of PEI-25K.

In order to reveal the cause for the high transfection

efficiency of the PEN polymers, two key factors influencing

transfection, namely, the endocytosis efficiency and the

ability to enter nuclei were investigated. First, PEN175 and

PEI-25K were labeled with FITC for tracing their trails in

cells. Then the complex solutions were prepared at their

own optimized weight ratios, which were 10 for PEN175

and 2.5 for PEI-25K. Finally, endocytosis efficiency for the

complex solutions was evaluated by flow cytometry. As

seen in Figure 9, the endocytosis efficiency of the complexes

Figure 9. Flow cytometry quantified endocytosis efficiency inHeLa cells.

2, 12, 1680–1688



www.mbs-journal.de

from PEN175 was much higher than that of PEI-25K.

The quantified endocytosis efficiencies for blank, PEI-25K/

pDNA, and PEN175/pDNA were respectively 0.1, 52.1 and

99.9%. Okano et al. reported that corona poly(N-isopropyl-

acrylamide) derivatives significantly increased intra-

cellular uptake above its lower critical solution temperature

for the enhancement of interactions between NIPA and cell

membranes.[33] After the introduction of NIPA into PEI-25K,

the hydrophobicity of the PEN polymers was improved

compared with that of PEI-25K because of the introduction

of the isopropyl groups. Furthermore, the hyper-branched

PEI-25K had an approximately global structure, so the

hydrophobic periphery of the PEN polymers might have a

bigger contact area for membrane than a linear structure.

As a result, the interactions between the PEN polymers and

the cellular membranes were enhanced, which further

improved intracellular uptake.

In addition, the intracellular distribution of complex

particles in HeLa cells was visually investigated by CLSM. As

shown in Figure 10, the complexes of PEI-25K/pDNA mainly

stayed in the peripheral areas around the cells and only a

few of them were internalized when treated for 2.5 h.

Figure 10. Intracellular distribution of complexes prepared from FITor PEI-25K with plasmid DNA at their own optimized weight ratiPEI-25K/pDNA complexes were treated for (a) 2.5 h and (b) 5 h. PEN175/were treated for (c) 2.5 h and (d) 5 h. Nuclei were stained with DAPI




Though there were some complexes were seen in the cells

after 5 h, none entered the nuclei. However, for PEN175

there were already some complexes internalized only after

2.5 h, but hardly any of them were in the nuclei; many of

them were endocytosed into cells after 5 h and a consider-

able amount of complexes entered nuclei. These results

indicated that the PEN polymers not only have much higher

endocytosis efficiency but also are highly effective for

nuclear uptake compared with PEI-25K. Although the

mechanism remains to be investigated, it is likely that

complex particles, still bound to pDNA, may be able to enter

cell nuclei based on previous work.[34–36] Because synthetic

gene vectors are largely known for their lack of nuclear

entry characteristics, almost all previous reports focused on

the chemical conjugation of nuclear localization signals

(NLSs)[37–40] or nuclear factor kappa B (NFkB)[41] to improve

the nuclear entry. Wang et al. reported that 10% dimethyl

sulfoxide markedly improved TAT or TAT fusion proteins’

penetration into cells without knowing the precise

mechanism.[42] To the best of our knowledge, our study is

the first report of transportation of pDNA preferentially into

cell nuclei by NIPA modification of PEI without any nuclear

C labeled PEN175os in HeLa cells.pDNA complexes

.

2, 1680–1688

& Co. KGaA, Weinheim

localization signals.

In summary, the high transfection

efficiency of PEN175 is attributed to four

aspects: (1) the modification reduced

cytotoxicity by increasing the gap

between the protonated amines and

the cell membranes because introduction

of NIPA on PEI; (2) the original ‘‘proton

sponge effect’’ of PEI-25K was also in

favor of rupturing endosomes and lyso-

somes in its derivatives, which would be

beneficial to high transfection; (3) the

enhanced interactions of complexes with

cell membranes led to improved endo-

cytosis efficiency because of improved

hydrophobicity and a bigger contact area;

(4) the complexes formed by PEN175 and

pDNA showed high nuclear localization

and in favor of further transcription.

4. Conclusion

In our study, PEI-25K was successfully

modified by NIPA via Michael addition.

The modification was mainly through

the reaction of the secondary amines.

The buffering capacity of the polymers

changed in different trends within

different pH ranges. It was higher in

the pH¼ 12.0 to 8.1 range and lower

in the pH¼ 8.1 to 2.0 range. All of the

1687

1688

www.mbs-journal.de


modified derivatives had good binding affinity to DNA and

could condense plasmid DNA into particles with a size of

about 140 nm and a zeta potential of þ30 to þ40 mV. The

derivatives all showed reduced cell toxicity after modifica-

tion because of the increased gap between the protonated

amines and cell membranes, resulting in reduced interac-

tions. All the PEN polymers had better transfection

efficiency than PEI-25K. The optimized transfection effi-

ciency of PEN175 was 18 fold higher than that of PEI-25K

which is one of the most efficient commercial polymers for

plasmid transfection. Furthermore, PEN175/pDNA com-

plexes had very high endocytosis efficiency and much

enhanced ability to enter nuclei. In summary, simple

modification of PEI-25K with NIPA afforded plasmid

delivery vectors with not only low cytotoxicity but also

high efficiency for transfection and nuclei entrance.

Acknowledgements: This work was supported by the NationalNatural Science Foundation of China (Grant Nos. 21074129,50903009, 51021003, 51222307), the Ministry of Science andTechnology of China (International cooperation and communica-tion program 2010DFB50890 and 2011RI0001), Jilin provincescience and technology development program (20100115,20120306).

Received: July 17, 2012; Revised: August 28, 2012; Publishedonline: November 5, 2012; DOI: 10.1002/mabi.201200249

Keywords: gene carriers; hyperbranched; modification; nucleientrance; polyethylenimine

[1] E. D. Zanjani, W. F. Anderson, Science 1999, 285, 2084.[2] A. Abbott, Nature 2001, 411, 410.[3] I. M. Verma, N. Somia, Nature 1997, 389, 239.[4] R. C. Mulligan, Science 1993, 260, 926.[5] E. Marshall, Science 1999, 286, 2244.[6] M. D. Brown, A. G. Schatzlein, I. F. Uchegbu, Int. J. Pharm. 2001,

229, 1.[7] O. Boussif, F. Lezoualc’h, M. A. Zanta, M. D. Mergny, D. Scherman,

B. Demeneix, J. P. Behr, Proc. Natl. Acad. Sci. USA 1995, 92, 7297.[8] J. P. Behr, Chimia 1997, 51, 34.[9] G. Backliwal, M. Hildinger, V. Hasija, F. M. Wurm, Biotechnol.

Bioeng. 2008, 99, 721.[10] M. Bertschinger, A. Schertenleib, J. Cevey, D. L. Hacker, F. M.

Wurm, Mol. Biotechnol. 2008, 40, 136.[11] Y. Rajendra, D. Kiseljak, L. Baldi, D. L. Hacker, F. M. Wurm,

J. Biotechnol. 2011, 153, 22.



[12] J. S. Remy, B. Abdallah, M. A. Zanta, O. Boussif, J. P. Behr,B. Demeneix, Adv. Drug Delivery Rev. 1998, 30, 85.

[13] C. Raymond, R. Tom, S. Perret, P. Moussouami, D. L’Abbe,G. St-Laurent, Y. Durocher, Methods 2011, 55, 44.

[14] M. L. Forrest, G. E. Meister, J. T. Koerber, D. W. Pack, Pharm. Res.2004, 21, 365.

[15] X. Zeng, Y.-X. Sun, X.-Z. Zhang, S.-X. Cheng, R.-X. Zhuo, Pharm.Res. 2009, 26, 1931.

[16] H. Tian, W. Xiong, J. Wei, Y. Wang, X. Chen, X. Jing, Q. Zhu,Biomaterials 2007, 28, 2899.

[17] A. Zintchenko, A. Philipp, A. Dehshahri, E. Wagner, Bioconj.Chem. 2008, 19, 1448.

[18] L. Y. Qiu, Y. H. Bae, Biomaterials 2007, 28, 4132.[19] A. Dehshahri, R. K. Oskuee, W. T. Shier, A. Hatefi, M. Ramezani,

Biomaterials 2009, 30, 4187.[20] A. Brownlie, I. F. Uchegbu, A. G. Schatzlein, Int. J. Pharm. 2004,

274, 41.[21] M. Thomas, A. M. Klibanov, Proc. Natl. Acad. Sci. USA 2002, 99,

14640.[22] S. Kim, J. S. Choi, H. S. Jang, H. Suh, J. Park, Bull. Korean Chem.

Soc. 2001, 22, 1069.[23] D.-A. Wang, A. S. Narang, M. Kotb, A. O. Gaber, D. D. Miller,

S. W. Kim, R. I. Mahato, Biomacromolecules 2002, 3, 1197.[24] A. M. Doody, J. N. Korley, K. P. Dang, P. N. Zawaneh, D. Putnam,

J. Controlled Release 2006, 116, 227.[25] D. Wang, Z. Zheng, C. Hong, Y. Liu, C. Pan, J. Polym. Sci., Part A:

Polym. Chem. 2006, 44, 6226.[26] T. W. Johnson, I. M. Klotz, Macromolecules 1974, 7, 149.[27] W.-C. Tseng, C.-H. Tang, T.-Y. Fang, J. Gene Med. 2004, 6, 895.[28] O. Boussif, F. Lezoualc’h, M. A. Zanta, M. D. Mergny,

D. Scherman, B. Demeneix, J. P. Behr, Proc. Natl. Acad. Sci.USA 1995, 92, 7297.

[29] M. R. Park, K. O. Han, I. K. Han, M. H. Cho, J. W. Nah, Y. J. Choi,C. S. Cho, J. Controlled Release 2005, 105, 367.

[30] Y. Liu, T. M. Reineke, J. Am. Chem. Soc. 2005, 127, 3004.[31] L. Chen, H. Tian, J. Chen, X. Chen, Y. Huang, X. Jing, J. Gene

Med. 2010, 12, 64.[32] S. Choksakulnimitr, S. Masuda, H. Tokuda, Y. Takakura,

M. Hashida, J. Controlled Release 1995, 34, 233.[33] J. Akimoto, M. Nakayama, K. Sakai, T. Okano, Biomacro-

molecules 2009, 10, 1331.[34] T. Kim, M. Lee, S. W. Kim, Biomaterials 2010, 31, 1798.[35] K. Huang, B. Voss, D. Kumar, H. E. Hamm, E. Harth, Bioconj.

Chem. 2007, 18, 403.[36] T. Schroder, N. Niemeier, S. Afonin, A. S. Ulrich, H. F. Krug,

S. Brase, J. Med. Chem. 2008, 51, 376.[37] K. M. Park, H. C. Kang, J. K. Cho, I. J. Chung, S. H. Cho, Y. H. Bae,

K. Na, Biomaterials 2009, 30, 2642.[38] T. Kanazawa, Y. Takashima, M. Murakoshi, Y. Nakai, H. Okada,

Int. J. Pharm. 2009, 379, 187.[39] R. Misra, S. K. Sahoo, Eur. J. Pharm. Sci. 2010, 39, 152.[40] M. A. Zanta, P. Belguise-Valladier, J. P. Behr, Proc. Natl. Acad.

Sci. USA 1999, 96, 91.[41] A. Mesika, I. Grigoreva, M. Zohar, Z. Reich, Mol. Ther. 2001, 3,

653.[42] H. Wang, C. Y. Zhong, J. F. Wu, Y. B. Huang, C. B. Liu,

J. Controlled Release 2010, 143, 64.

2, 12, 1680–1688


N -Isopropylacrylamide-Modified Polyethylenimines as Effective Gene Carriers

Documents