C

Pd

KD

a

ARRAA

KPGCHA

1

sctalPidsli

h0

ARTICLE IN PRESSG ModelOLSUB-6525; No. of Pages 9

Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces

j o ur nal ho me pa ge: www.elsev ier .com/ locate /co lsur fb

EGylated galactosylated cationic liposomes for hepatocytic geneelivery

ovashnee Naicker ∗, Mario Ariatti, Moganavelli Singhepartment of Biochemistry, Non-Viral Gene Delivery Laboratory, University of KwaZulu-Natal, Westville Campus, Durban 4000, South Africa

r t i c l e i n f o

rticle history:eceived 18 May 2014eceived in revised form 7 July 2014ccepted 8 July 2014vailable online xxx

eywords:EGylationlycotargetingationic liposomeepatocyteSGP-R

a b s t r a c t

The efficiency of liposome-mediated gene delivery is greatly enhanced by appropriate decoration ofvehicles with cell-specific targeting ligands. However, liposome–DNA complexes may still be opsonizedin serum thus ablating any advantage gained. A stealth aspect may therefore be conferred on com-plexes by poly(ethylene glycol) (PEG) grafting. Here, we examined the effect that degree of PEGylationhas on physicochemical properties, cytotoxicity and transfection activity of lipoplexes containingthe cytofectin 3�-[N-(N′, N′-dimethylaminopropane)-carbamoyl] cholesterol (Chol-T), the neutral co-lipid dioleoylphosphatidylethanolamine (DOPE), the asialoglycoprotein receptor (ASGP-R) targetedcholesteryl-�-d-galactopyranoside (Chol-�-Gal) ligand, and plasmid DNA in ASGP-R-negative (HEK293)and receptor-positive (HepG2) human cell lines. Lipoplexes were characterized by hydrodynamic sizing,electron microscopy, band shift, ethidium bromide (EtBr) intercalation and nuclease digestion assays.Cryo-TEM and DLS studies revealed that PEGylation generated smaller and more densely aggregatedlipoplexes than their non-PEGylated counterparts. MTT and AB reduction studies showed that thelipoplexes elicited a dose-dependent cytotoxic effect in both cell lines, with cell viability remaining above65% (MTT) and 50% (AB). The Ricinus communis (RCA120) agglutination test confirmed that the galactosylresidues on the targeted lipoplexes were well exposed and accessible. Transgene activity increased by 63%

and 77% when HepG2 was confronted by the 2 and 5 mole% PEGylated lipoplexes, respectively, comparedto their non-PEGylated counterparts. Furthermore, Chol-T Chol-�-Gal 5% PEG complexes were able toachieve a 164% increase in transfection level in the ASGP-R positive cell line (HepG2) compared to HEK293(ASGP-R negative). Results strongly indicate that PEGylation potentiates the activity of ASGP-R-targetedlipoplexes, highlighting their gene delivery potential.

© 2014 Elsevier B.V. All rights reserved.

. Introduction

“Stealth” or sterically stabilized liposomes are engineered byurface functionalization using polymers to control interfacial pro-esses. The strategy of using the polymer polyethylene glycol (PEG)o produce “stealth” liposomes is known as PEGylation. PEG isrguably the most common polymer employed to impart ontoiposomes a hydrophilic protective cloud encasing the liposome.EG acts as a protein-resistant surface layer due to its unlim-ted water solubility, large excluded volume, low toxicity and highegree of conformational entropy [1,2]. Sterically stabilized lipo-

Please cite this article in press as: K. Naicker, et al., PEGylated galactosSurf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014

omes are able to mitigate unsolicited interactions an unprotectediposome may have with charged serum proteins, thus increas-ng their circulatory time and avoiding rapid opsonization by the

∗ Corresponding author. Tel.: +27 71 884 9073.E-mail address: kovashneenaicker@gmail.com (K. Naicker).

ttp://dx.doi.org/10.1016/j.colsurfb.2014.07.010927-7765/© 2014 Elsevier B.V. All rights reserved.

reticuloendothelial system (RES). Indeed, there are PEGylated lipo-somes commercially available and several others under clinical andpreclinical development; however one perilous drawback of theseshielded liposomes is their inefficient specific interaction with tar-get cells leading to reduced intracellular delivery of genes. Thiscontradictory effect of PEGylation is usually referred to as the “PEGdilemma” [3]. PEGylated liposomes can therefore be actively tar-geted by surface modification with the appropriate targeting ligandto enhance selectivity and greater uptake by target cells.

Liver cancer is the sixth most commonly diagnosed solid tumourin the world, with the vast majority of primary liver cancers(75–90%) being hepatocellular carcinomas (HCC). It is projectedthat 560,000 new HCC cases are diagnosed around the world onan annual basis [4]. Asialoglycoprotein receptor (ASGP-R) mediated

ylated cationic liposomes for hepatocytic gene delivery, Colloids.07.010

targeting of cationic liposome–DNA constructs to hepatocytes hasbeen the backbone of hepatocytic targeting efforts [5–7], since thisreceptor is reported to be richly and almost exclusively expressedon liver parenchymal cells [8]. This hepatic lectin exhibits high

ING ModelC

2 aces B

apn[aactTitti

w�tPgcl

2

2

[r(bIcLP2MbE(Lf(w[PTrLp

2

ha(aC

2

t([

ARTICLEOLSUB-6525; No. of Pages 9

K. Naicker et al. / Colloids and Surf

ffinity and a rapid internalization rate, therefore representing aotential target for HCC treatment. Carbohydrates and their cog-ate lectins have played a central role in liver specific targeting9]. Lectins are multi-domain proteins that are able to recognizend specifically bind sugar or glycan moieties attached to proteinsnd lipids [9]. Direct lectin targeting employs nanocarriers surface-onjugated with carbohydrates that are able to recognize and bindo lectins that are overexpressed on the surface of the target cell.his strategy of facilitating receptor-mediated endocytic internal-zation of the carrier via carbohydrate residue-lectin interaction isermed glycotargeting [10,11]. ASGP-Rs have proven to be versa-ile and robust portals by which glycotargeted lipoplexes gain entrynto hepatocytes for effective gene therapy [12–15].

Based on the above, cationic liposomes were functionalizedith the targeting ligand cholesteryl-�-d-galactopyranoside (Chol--Gal) for direct targeting of the ASGP-R and were PEGylated

o produce “stealth” liposomes. Specific ligand modification ofEGylated liposomes was applied here to explore hepatocyticlycotargeting and the effect PEGylation (at two different mole per-entages) may have on receptor-mediated targeted transfection byipoplexes.

. Materials and methods

.1. Materials

1,2-Distearoyl-sn-glycero-3-phosphoethanolamine-N-methoxy(polyethylene glycol)-2000] (DSPE-PEG2000), hereaftereferred to as PEG, was acquired from Avanti Polar LipidsAlabaster, AL, USA). Dioleoylphosphatidylethanolamine (DOPE),icinchoninic acid (BCA) assay kit, asialofetuin (AF) from FCS Type, and the lectin from Ricinus communis (castor bean) agglutininalled RCA120 were sourced from Sigma–Aldrich Chemical Co. (St.ouis, MO, USA). Luciferase assay reagent was purchased fromromega Corp. (Madison, WI, USA). 3-(4,5-Dimethyl-2-thiazolyl)-,5-diphenyl-2H-tetrazolium bromide (MTT) was acquired fromerck (Darmstadt, Hesse, Germany). AlamarBlue® (AB) cell via-

ility reagent was obtained from Invitrogen (Carlsbad, CA, USA).agle’s Minimum Essential Medium (EMEM) with l-glutamine4.5 g L−1), and trypsin–versene mixture was purchased fromonza BioWhittaker (Verviers, Liège, Belgium). Gamma-irradiatedoetal calf serum (FCS) was purchased from Highveld BiologicalPty) Ltd. (Kelvin, Gauteng, RSA). The plasmids used in this studyere pCMV-luc [coding for firefly luciferase (luc)] and pCMV-GFP

coding for green fluorescent protein (GFP)], purchased fromlasmid Factory (Bielefeld, North Rhine-Westphalia, Germany).urboFectTM (TF), a control in vitro cationic polymer transfectioneagent, was purchased from Fermentas Life Sciences (Vilnius,ithuania). All other chemicals and reagents were of analyticalurity grade or higher, and purchased commercially.

.2. Cell culture

Human embryonic kidney cells (HEK293, ASGP-R negative) anduman liver carcinoma cells (HepG2, ASGP-R positive) were prop-gated in EMEM supplemented with 10% (v/v) FCS and antibiotics100 U mL−1 penicillin, 100 �g mL−1 streptomycin, 0.25 �g mL−1

mphotericin B) at 37 ◦C under a humidified atmosphere with 5%O2.

.3. Liposome and lipoplex preparation

Please cite this article in press as: K. Naicker, et al., PEGylated galactosSurf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014

Liposomes were prepared according to the thin-film evapora-ion method as described elsewhere [16]. The cytofectin 3�-[N-N′,N′-dimethylaminopropane)-carbamoyl] cholesterol (Chol-T)17] and the targeting ligand cholesteryl-�-d-galactopyranoside

PRESS: Biointerfaces xxx (2014) xxx–xxx

(Chol-�-Gal) [18] were both previously synthesized. Chol-T andChol-�-Gal were kept constant at 50 and 10 mole% in all liposomepreparations, respectively, whilst DOPE was adjusted to accom-modate changes in PEGylation as shown in Table 1. PEGylatedliposomes were formulated with 2 or 5 mole% of PEG. Lipid filmswere rehydrated in Hepes Buffered Saline (HBS, 20 mM HEPES,150 mM NaCl, pH 7.5) and vortexed then sonicated to clarity. Lipo-some preparations were stored at 4 ◦C.

The reporter gene (1 �g pCMV-luc) was mixed directly withdifferent amounts of appropriate cationic liposome across specificranges of mass (w/w) or N/P (+:−) ratios to produce the correspond-ing lipoplex. These mixtures were brought up to a constant volume(10 �L) with HBS, gently vortexed for 30 s and incubated for 30 minat room temperature for maturation of complexes.

2.4. Particle imaging

Liposome and lipoplex dispersions were diluted 1:5 (v/v) withHBS then deposited (2 �L) onto 200-mesh Lacey Formvar car-bon coated copper grids and contrasted 1:1 (v/v) with 4% uranylacetate negative stain. Samples were rapidly vitrified at −180 ◦Cthen viewed at −170 ◦C using a Cryogenic-TEM JEOL JEM-1010 elec-tron microscope (Jeol, Tokyo, Japan) operating at an acceleratingvoltage of 100 kV under low electron dose.

2.5. Particle size and zeta potential analysis

Size distribution (polydispersity index, PDI) and �-potentialmeasurements were carried out using a Malvern Nano-ZS Zeta-Sizer (Worcestershire, UK) with a He–Ne laser beam (633 nm, fixedscattering angle of 173◦) at 25 ◦C. Liposomes were allowed to equil-ibrate to room temperature prior to measurement and lipoplexpreparation. Briefly, 10 �L of liposome and pre-formed lipoplexdispersions were diluted in deionized water to 1 mL to obtainappropriate viscosities for measurement.

2.6. DNA binding, compaction and protection studies

For DNA binding studies, pre-formed lipoplexes were loadedinto 1% (w/v) agarose gels (containing 1 �g mL−1 EtBr), and elec-trophoresed at 50 V cm−1 for 90 min in Tris–acetate–EDTA (TAE)running buffer (36 mM Tris base, 30 mM NaH2PO4, 10 mM EDTAat pH 7.5). The location of plasmid DNA in the gels was analyzedunder UV300nm transillumination and images were captured usinga Vacutec Syngene G:Box BioImaging system (Cambridge, UK).

In the dye displacement assays (EtBr intercalation), EtBr (10 �L,100 �g mL−1) was added to HBS (250 �L) in 96-well FluorTrac flat-bottom black plates and a baseline relative fluorescence of 0%was established. This was followed by the introduction of pCMV-luc (3 �g) to the mixture and 100% fluorescence intensity of EtBrwas obtained at excitation and emission wavelengths of 520 nmand 600 nm, respectively using an automated SynergyMx ELX 800microplate reader (BioTek Instruments, Vermont, USA). Thereafter,cationic liposome dispersion was added stepwise (1 �L aliquots)to the EtBr/pCMV-luc solution followed by a 30 s mixing intervalbefore fluorescence intensities were recorded. This was continueduntil a plateau in fluorescence readings was reached.

For serum resistance assays, lipoplexes were assembled accord-ing to three weight ratios based on binding values obtained for gelretardation studies and lipoplexes were treated with 10% FCS and

ylated cationic liposomes for hepatocytic gene delivery, Colloids.07.010

incubated at 37 ◦C for 4 h. Digestion was terminated with EDTA(final 10 mM, pH 8) and lipoplexes disassociated using 0.5% SDS(w/v). Mixtures were incubated at 55 ◦C for 20 min followed byelectrophoresis at 50 V for 2 h as described above.

Please cite this article in press as: K. Naicker, et al., PEGylated galactosSurf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014

ARTICLE ING ModelCOLSUB-6525; No. of Pages 9

K. Naicker et al. / Colloids and Surfaces B

Tab

le

1C

har

acte

riza

tion

of

the

dif

fere

nt

cati

onic

lip

osom

e

and

lip

ople

x

susp

ensi

ons.

Cat

ion

ic

lip

osom

e

Com

pos

itio

n

(�m

oles

/500

�L)

Gel

reta

rdat

ion

lip

osom

e:D

NA

,w

/w

(N/P

)a

EtB

rin

terc

alat

ion

lip

osom

e:D

NA

,w

/w

(N/P

)a

Mea

n

par

ticl

e

size

and

zeta

pot

enti

als

(nm

)

n

=

3

Ch

ol-T

DO

PE

PEG

Ch

ol-�

-Gal

Lip

osom

e

Lip

ople

x

Size

(PD

I)�-

Pote

nti

al(m

V) ±

SDSi

ze

(PD

I)�-

Pote

nti

al(m

V) ±

SD

Ch

ol-T

Ch

ol-�

-Gal

1

0.80

–

0.20

13:1

(3.5

:1)

11:1

(3:1

)

183.

67

(0.5

25)

86.8

4

±

15.5

2 24

9.75

(0.4

07)

43.9

7

±

3.70

5C

hol

-T

Ch

ol-�

-Gal

2%PE

G1

0.76

0.04

0.20

10:1

(2.5

:1)

11:1

(2.8

:1)

71.4

8

(0.3

98)

28.2

5

± 2.

759

138.

26

(0.2

09)

27.9

8

±

8.08

7

Ch

ol-T

Ch

ol-�

-Gal

5%PE

G1

0.70

0.10

0.20

10:1

(2.3

:1)

20:1

(4.6

:1)

76.8

8

(0.2

00)

7.50

± 0.

694

168.

19

(0.1

55)

21.4

6

±

4.92

3

aTh

e

theo

reti

cal N

/P

rati

o

rep

rese

nts

the

char

ge

rati

o

of

cati

onic

lip

id

to

nu

cleo

tid

e

and

was

calc

ula

ted

assu

min

g

one

pos

itiv

e

char

ge

per

cyto

fect

in

mol

ecu

le

at

pH

7.5

and

one

neg

ativ

e

char

ge

per

DN

A

nu

cleo

tid

e

(ave

rage

mas

s

330

Da)

.

PRESS: Biointerfaces xxx (2014) xxx–xxx 3

2.7. Cytotoxicity studies

Cell viability in terms of mitochondrial integrity, and overallcellular metabolism was measured using the MTT [19] and AB[20] assays following literature procedures. Briefly, for both assaysHEK293 and HepG2 cells were seeded into 48-well plates at a den-sity of 2.8 × 104 cells per well and cultured overnight to achieve∼80% cell confluence. Cells bathed in 0.3 mL of complete culturemedium were then introduced to 10 �L of mature lipoplex [pre-pared at optimized N/P ratios (+:−) containing 1 �g pCMV-lucplasmid DNA] and incubated at 37 ◦C for 48 h. Next, cells were incu-bated for 4 h at 37 ◦C with fresh medium (0.2 mL) and 0.2 mL of MTTreagent (5 mg mL−1 in sterile PBS, pH 7.4). Thereafter, cells wererinsed (2× 300 �L PBS) and formazan crystals were solubilizedin 0.2 mL DMSO. Absorbance was recorded at 570 nm (detection�) and 630 nm (reference �) using a Vacutec MR-96A microplatephotometer (Vacutec, Hamburg, Germany).

In the AB assay however; 10× AB reagent was added directlyto cells in complete culture medium (30 �L/well) after the 48 hincubation period and plates were incubated for a further 4 h at37 ◦C in 5% CO2. Fluorescence measured in 96-well FluorTrac flat-bottom black plates using the GloMax®-Multi Detection System(Promega BioSystems, Sunnyvale, CA, USA) operated by Instinctsoftware at 570 nm (fluorescence excitation �) and 585 nm (flu-orescence emission �). Control groups were run simultaneouslywhich included untreated cells. The percentage of AB reduction wascalculated as follows: [(�570 Test − �585 Test)/(�570 Control − �585Control)] × 100.

2.8. Lectin-induced agglutination of galactosylatedliposome/lipoplexes

The protocol to investigate the accessibility of Chol-�-Galresidues on the surface of targeted liposomes and lipoplexes wasadapted from that of Managit and co-workers in 2005 [21]. Briefly,100 �L of liposome suspension (1 mg mL−1 total lipid in HBS, pH7.4) was directly incubated with 100 �L of RCA120 (1 mg mL−1 inPBS, pH 7.4), in 96-well flat-bottom microwell plates. After rapidmixing for 30 s, agglutination as induced by RCA120 was estimatedby the time-dependent increase in turbidity of the solution. Thus,absorbance was measured over 25 min (1 min intervals) using amicroplate reader at �max 450 nm. The absorbance of the control(liposome suspension without the addition of the lectin agglutinin)was subtracted from the absorbance of the aggregated liposomalsuspension. Reversibility of the agglutination was verified by thesubsequent addition of 100 �L (10 mg mL−1) free d-galactose fol-lowed by additional absorbance measurements.

Lipoplexes were first prepared at their optimal N/P ratios (+:−),matured, then diluted with HBS (pH 7.4) to afford a concentrationof 1 mg mL−1 total lipid. Thereafter, 100 �L of the diluted lipoplexsuspension was treated as described above.

2.9. In vitro transfection efficiency

HEK293 and HepG2 cells were plated (seeding density at2.6 × 104 cells per well) separately in 48-well plates. Twenty fourhours later, mature lipoplexes were introduced to cells as describedin Section 2.7. After the 48 h incubation, the growth medium wasdiscarded and cells washed with PBS (0.2 mL), lysed with 80 �Lof 1× reporter lysis buffer and plates were rocked for 15 min at30 rpm. Each well was gently scraped to detach remaining cells.Cell lysates were then centrifuged for 30 s at 12,000 rpm to pel-

ylated cationic liposomes for hepatocytic gene delivery, Colloids.07.010

let cellular debris. Harvested supernatants were assayed for fireflyluciferase activity according to the Promega Luciferase Assay proto-col (Promega Corp., Madison, USA). Light units were normalized bysoluble protein content in the cell extracts, determined by the BCA

ARTICLE IN PRESSG ModelCOLSUB-6525; No. of Pages 9

4 K. Naicker et al. / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

F low) aG parede

pIt

idwNcwOFr

2

Opgs

3

3

we(lpwotlllav

ig. 1. Transmission electron micrographs of liposomes (above) and lipoplexes (beal @ N/P 2.5; and 5% PEGylated Chol-T Chol-�-Gal @ N/P 2.3. Lipoplexes were prelectrophoretic mobility shift assays. Bar = 100 or 200 nm.

rotein assay at 562 nm, and were expressed as RLU mg−1 protein.n competition experiments, 300 �g of asialofetuin was introducedo cells 30 minutes prior to exposure to lipoplexes.

In GFP expression assays, HEK293 and HepG2 cells were seedednto 6-well plates at a density of 3.5 × 105 cells per well asescribed above. Twenty four hours later, cells were transfectedith lipoplexes containing pCMV-GFP (1 �g) prepared at optimal/P (+:−) ratios, as determined in luciferase assays, in 1.5 mL ofulture medium and incubated for 48 h at 37 ◦C. GFP expressionas detected using an inverted fluorescent microscope (CKX41,lympus, Japan) at 540 nm (excitation �) and 580 nm (emission �).luorescent images were acquired at a magnification of 20X andecorded using the TeamViewer software.

.10. Statistical analysis

All data are presented as mean ± standard deviation (SD, n = 3).ne-way analysis of variance (ANOVA) followed by Tukey–Kramerost hoc analysis was used for multiple comparisons betweenroups. Differences between experimental groups were consideredignificant when the P value was less than 0.05 (P < 0.05).

. Results and discussion

.1. Interaction of liposome with DNA and lipoplex formation

The supramolecular nanostructure of liposomes and lipoplexesere directly visualized by cryo-TEM. Micrographs described a het-

rogenous population of concentric spherical-shaped liposomesFig. 1). Minimal structural distortion was observed; howeveriposomes appeared well dispersed and did not aggregate underhysiological conditions. All liposome preparations were similarith the increasing mole% of PEGylation (0–5%) showing no obvi-

us effects on the liposomal morphology. This may be ascribedo steric repulsion and stabilization. Cryo-TEM characterization ofipoplexes revealed predominately large dense nebulous multi-

Please cite this article in press as: K. Naicker, et al., PEGylated galactosSurf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014

amellar aggregates (Fig. 1) with a few unilamellar structuresocated at the periphery of these large aggregative structures. Theggregates appeared to be composed of a varying number of indi-idual lipoplexes. Reduced levels of aggregation were observed

s follows: 0% PEGylated Chol-T Chol-�-Gal @ N/P 3.5; 2% PEGylated Chol-T Chol-�- at optimal binding lipid:pDNA or N/P (+:−) ratios (as indicated) established using

with PEGylated lipoplexes underscoring the steric stabilizationresulting in smaller liposome–DNA complexes compared to thenon-PEGylated lipoplex [22].

Dynamic light scattering (DLS) and �-potential measurementsof the liposomes and corresponding lipoplexes are summarized inTable 1. Of interest is the significant (P < 0.001) decrease in parti-cle size of liposomes (from 184 nm to 72 nm) and lipoplexes (from250 nm to 138 nm) upon the introduction of PEG, i.e. (0 → 2 mole%);however further increase in PEGylation, i.e. (2 → 5 mole%) lead toslight increases in liposome (from 72 nm to 77 nm) and lipoplex(from 138 nm to 168 nm) particle size. PEGylation improves col-loidal stability and resistance to aggregation which in turn leadsto smaller particle sizes resulting in improved biocompatibility.Polydispersity indices (PDI) improved as degree of PEGylationincreased, with 5% PEGylated liposomes displaying monodisper-sity (PDI = 0.20). Lipoplexes were larger than liposomes althoughPEGylation afforded significantly (P < 0.001) smaller lipoplexesthan non-PEGylated preparations (>200 nm). Furthermore, absenceof the steric stabilizer rendered colloidally unstable lipoplexes,evidenced by a visible increase in turbidity upon complexation andwider size distributions than the PEGylated preparations.

�-Potentials steadily declined with an increase in PEGyla-tion. This was expected with an increasing content of the PEGmoiety which contains a negatively charged phosphate group.Additionally, the PEG shell shifts the plane of shear around the par-ticle further away from their surface through which the positivecharge is increasingly neutralized by ions present in the dispersionmedium [23]. It is interesting to note, that PEG molecules assemblein a mushroom-like fashion on the surface of liposomes at concen-trations lower than 4 mole% [24], whereas higher concentrationsyield brush-like arrangements [25]. Positive surface charges main-tained by the lipoplexes can support electrostatic interactions andthus attachment to anionic cell membranes, as cells are known totake up positively charged liposomes more efficiently than neutralor negatively charged liposomes [26].

The electrostatic interaction between cationic liposomes and

ylated cationic liposomes for hepatocytic gene delivery, Colloids.07.010

plasmid DNA was examined by conventional gel retardation elec-trophoresis, at varying lipid:pDNA (+:−) N/P ratios (Table 1 andFig. 2A). All liposomes fully retarded plasmid DNA at N/P ratiosbetween 2.5:1 and 3.5:1 (+:−).

ARTICLE IN PRESSG ModelCOLSUB-6525; No. of Pages 9

K. Naicker et al. / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx 5

Fig. 2. Gel retardation (A) and EtBr intercalation displacement (B) studies of lipoplexes composed of: Chol-T Chol-�-Gal 0% PEG (left, �); Chol-T Chol-�-Gal 2% PEG (middle, �);and Chol-T Chol-�-Gal 5% PEG (right, �). For the gel retardation studies, samples from left to right are as follows: lane 1: 1 �g pCMV-luc; lanes 2–8: respective liposome/pCMV-luc (1 �g) complexes prepared at different liposome mass (�g) as indicated. White arrows indicate the end-point or point of electroneutrality. EtBr intercalation displacementw tofectu

tinpsTldt[tllDaotto

here percentage relative fluorescence (Fr) was plotted as a function of nmoles of cyntil a plateau in fluorescence was achieved.

PEGylation is one of the most popular approaches for prolonginghe serum half-life of lipoplexes. Lipoplex transfection efficiencys largely inhibited due to serum proteins that associate with theanocomplex [27]. Reports have detailed the existence of a richrotein layer, called a protein corona, which is associated with theurface of lipoplexes that come into contact with serum [28,29].he protein corona can have several unfavourable effects on theipoplex such as, aggregation, surface charge neutralization andestabilization of the lipoplex structure thus exposing the DNAo intrinsic enzymatic degradation or even rapid opsonization30]. For that reason, FCS-catalyzed digestion coupled with elec-rophoretic analysis was used to gauge the protective abilities ofipoplexes towards DNA in the presence of serum (Fig. S1). Alliposomes tested showed substantial protection of their plasmidNA cargo in the presence of 10% (v/v) FCS. Plasmid DNA associ-ted with liposomes remained comparatively stable in the presence

Please cite this article in press as: K. Naicker, et al., PEGylated galactosSurf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014

f serum for all tested lipid:pDNA weight ratios (w/w) comparedo the fully digested control (FCS-treated DNA). Lipoplexes par-ially protected the DNA with some nicking in evidence after 4 hf incubation. From the banding patterns, Chol-T Chol-�-Gal 5%

in. Cationic liposome was systematically added to an EtBr/pCMV-luc (3 �g) solution

PEG lipoplexes provided least protection as evidenced by faintbands and low molecular weight breakdown products. As men-tioned earlier, above a 4 mole% of PEG grafted onto the surfaceof liposomes results in brush-like arrangements of the polymer.The extension length (L) of the polymer chain from the surfaceof the liposome extends to about 35 A and 50 A for the “mush-room” and “brush” configurations, respectively [24,25]. Therefore,it can be assumed that the larger protrusion profile of the 5% PEGy-lated lipoplex (compared to the 2% PEGylated lipoplex) may haveaffected the complete compaction of pDNA allowing partial degra-dation by serum nucleases. The reduced compaction of pCMV-lucDNA by Chol-T Chol-�-Gal 5% PEG was further corroborated by theEtBr intercalation studies (refer below). This obviously was not dis-advantageous as the reduced compaction of pDNA by this lipoplexpossibly conferred better release of the DNA during the process oflipofection bringing about better levels of reporter gene transfec-

ylated cationic liposomes for hepatocytic gene delivery, Colloids.07.010

tion (as discussed in Section 3.4).Fluorescence quenching assays were conducted to examine

liposome–DNA binding and condensation. The points of inflectionat which the liposome maximally displaced the intercalated EtBr

ARTICLE IN PRESSG ModelCOLSUB-6525; No. of Pages 9

6 K. Naicker et al. / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

F B). Lipa e suspT e (10 m

aaac

3

dtiecTNb

ig. 3. RCA120 induced agglutination of galactosylated liposomes (A) and lipoplexes (t t = 1 min followed by continuous readings for 25 min. Non-galactosylated liposomhe reversibility of the agglutination was verified by the addition of free d-galactos

nd condensed the pDNA are indicated in Table 1 and are in generalgreement with end-point N/P ratios obtained by gel retardationnalyses. The DNA in Chol-T Chol-�-Gal 5% PEG lipoplexes was leastompacted as evidenced by reduced dye displacement (Fig. 2B).

.2. Cytotoxicity analysis

It is recommended that more than one cell viability assay withifferent chemistries and detection mechanisms should be usedo test the tolerability and toxicity of any delivery system uponnteraction with cells. Here, the MTT and AB reduction assays weremployed to quantify the levels of in vitro cytotoxicity the Chol-T

Please cite this article in press as: K. Naicker, et al., PEGylated galactosSurf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014

ationic lipoplexes had induced in HEK293 and HepG2 cell lines.he MTT assay (Fig. S2A, C and E) shows the higher the lipoplex/P ratio (+:−) the higher the cytotoxicity observed, as indicatedy the lower cell viabilities, with maximum reduction in cell

osomes and lipoplexes (1 mg mL−1 total lipid) were treated with RCA120 (1 mg mL−1)ension (Chol-T) without the addition of the lectin agglutinin was used as a control.g mL−1) at the 20th min.

viability being observed at the highest N/P ratios for all liposomes,irrespective of degree of PEGylation. This shows that the lipoplexeshave a dose-dependent cytotoxic effect on both cell lines. This is inagreement with the notion that toxicity is closely associated withN/P ratios between cationic lipids and anionic oligonucleotides, aswell as, the dose of lipoplex administered [31]. It has been pro-posed that lipoplexes at higher N/P ratios are more toxic to cells,perhaps due to free liposomes which compete with lipoplexes forbinding and uptake by cells [32]. Complexes were however welltolerated by both cell lines, with cell viability remaining above65%. The non-PEGylated lipoplexes were less toxic (Chol-T Chol-�-Gal: 11% in HEK293 and 13% in HepG2) than their PEGylated

ylated cationic liposomes for hepatocytic gene delivery, Colloids.07.010

counterparts (Chol-T Chol-�-Gal 2% PEG: 19% in HEK293 and 24%in HepG2) and (Chol-T Chol-�-Gal 5% PEG: 24% in HEK293 and 35%in HepG2). This is in agreement with the findings of an earlier studywith untargeted lipoplexes [33].

ARTICLE IN PRESSG ModelCOLSUB-6525; No. of Pages 9

K. Naicker et al. / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx 7

Fig. 4. In vitro transfection studies in HEK293 (�) and HepG2 (�) cell lines. Competitive inhibition of targeted transfection in HepG2 cells is also shown (�). Lipoplexes(containing 1 �g of pCMV-luc DNA) were prepared at various N/P ratios (+:−) as indicated: (A) non-PEGylated Chol-T Chol-�-Gal; (B) Chol-T Chol-�-Gal 2% PEG; and (C) Chol-T Chol-�-Gal 5% PEG. TurboFectTM-pCMV-luc DNA complexes were prepared according to the manufacturer instructions and used a positive control. Control 1: untreated cells;Control 2: cells exposed to 1 �g of pCMV-luc DNA only. Normalized light units were expressed as RLU mg−1 protein. Each data point was averaged over three replicates (n = 3)a one-wi

i(llcntMeiTi

3l

lmart(rmt

nd results are presented as means ± SD. Statistical analysis was performed using

ndicate a significant difference.

Fig. S2(B, D and F) illustrates the percentage reduction of ABn HEK293 and HepG2 cells as a function of lipoplex N/P ratio+:−) after 48 h exposure. As with the MTT reduction assay, theipoplexes displayed a dose-dependent cytotoxic effect on both cellines. However, the AB assay revealed more marked decreases inell viability with an increase in N/P ratios than the MTT assay. It isoteworthy that as AB is not only reduced by mitochondrial reduc-ases but also by cytochromes, whereas tetrazolium salts (such as

TT) are not. Apparently, cells that have detached and are consid-red non-viable continue to transform MTT into formazan resultingn an overestimation of cell viability as deduced by the MTT assay.his may explain the discrepancy in cytotoxic response observedn both cell lines in the two assay systems.

.3. Targeting ligand accessibility of galactosylatediposome/lipoplexes

Fig. 3 illustrates the RCA120 induced agglutination of galactosy-ated liposomes (A) and lipoplexes (B), respectively. Results show a

arked difference in the binding ability of the non-galactosylatednd galactosylated liposomes to RCA120, with untargeted liposomesemaining unaffected (Fig. 3A). The ligand–lectin agglutina-ion was rapidly reversed by challenge with excess d-galactose

Please cite this article in press as: K. Naicker, et al., PEGylated galactosSurf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014

10 mg mL−1) [34]. Results obtained with targeted lipoplexes mir-ored those obtained with liposomes confirming that the galactooieties remain prominently displayed after liposome complexa-

ion with DNA.

ay ANOVA followed by Tukey–Kramer post hoc test. * (P < 0.05) and *** (P < 0.001)

3.4. In vitro transfection efficiency

The ability of the lipoplexes to transfect HEK293 and HepG2cells with the reporter gene pCMV-luc was examined using theluciferase reporter gene assay. Fig. 4 shows the transfection effi-ciencies of Chol-T Chol-�-Gal (A), Chol-T Chol-�-Gal 2% PEG (B), andChol-T Chol-�-Gal 5% PEG (C) at predetermined N/P ratios (+:−),while TF was used according to manufacturer instructions for com-parison. Both controls showed negligible bioluminescence outputindicating poor transfection of the reporter gene when not associ-ated with a gene carrier. Higher transfection activity was achievedin the ASGP-R-positive HepG2 cell line than in HEK293 cells sug-gesting that lipoplexes actively targeted towards the ASGP-R areable to significantly improve receptor-mediated transfection levels.The highest transfection levels achieved by the lipoplexes in HepG2cells were: Chol-T Chol-�-Gal (N/P 4) 6.02 × 106 RLU mg−1, Chol-TChol-�-Gal 2% PEG (N/P 2.5) 9.78 × 106 RLU mg−1, and Chol-T Chol-�-Gal 5% PEG (N/P 2.3) 1.06 × 107 RLU mg−1. Chol-T Chol-�-Gal 5%PEG proved to be the best transfecting lipoplex with expressionlevels comparable to that of TF (Fig. 4C). Interestingly, Chol-T Chol-�-Gal 5% PEG also elicited the highest level of cell death and didnot afford the same level of protection to DNA as the other twoformulations.

As noted earlier, PEGylation was employed to induce a size sta-

ylated cationic liposomes for hepatocytic gene delivery, Colloids.07.010

bilizing effect which led to significantly smaller liposomes beingformed, as well as, providing stability to the lipoplex in the pres-ence of serum. It is possible that non-PEGylated lipoplexes attractproteins from serum thus creating a protein corona, which may

ARTICLE ING ModelCOLSUB-6525; No. of Pages 9

8 K. Naicker et al. / Colloids and Surfaces B

Fig. 5. GFP expression in HEK293 (ASGP-R negative) and HepG2 (ASGP-R positive)cells. Cells were transfected with targeted cationic liposome–DNA lipoplexes (con-taining 1 �g of pCMV-GFP) prepared at N/P ratios (+:−) that achieved maximalluciferase activity in the presence of serum and GFP expression was examined 48 hpost-transfection. GFP micrographs for both cell lines are as follows: Control: cellstreated with 1 �g of pCMV-GFP only; 0% PEGylated Chol-T Chol-�-Gal (N/P 4); 2%PEGylated Chol-T Chol-�-Gal (N/P 2.5); and 5% PEGylated Chol-T Chol-�-Gal (N/P2

tlTctl

Htrtf(ifacliCsld

[3] B. Wang, J. Zhou, S. Cui, B. Yang, Y. Zhao, B. Zhao, Y. Duan, S. Zhang, Cationicliposomes as carriers for gene delivery: physicochemical characterization andmechanism of cell transfection, Afr. J. Biotechnol. 11 (2012) 2763–2773.

.3). Bar = 500 �m.

hen interfere with ligand–lectin interaction. In addition, the PEGipid used, DSPE-PEG2000, is linked to DSPE via a carbamate bond.his carbamate linkage is sensitive to the low pH of the endosomalompartment and it is this lability that may have also contributedo the increase in transfection levels observed with the PEGylatedipoplexes as opposed to the non-PEGylated complexes.

There was no statistical difference in the luciferase activity inepG2 cells transfected with Chol-T Chol-�-Gal lipoplexes when

he N/P ratio increased from 3 to 3.5 (Fig. 4A). However, when theatio was further increased from 3.5 to 4 so did the level of transfec-ion (P < 0.05). An increase in Chol-T Chol-�-Gal 2% PEG N/P ratiorom 2 to 2.5 was paralleled by an increase in luciferase activityP < 0.05) (Fig. 4B). However, further increases in N/P ratio resultedn a decrease in transfection activity, albeit non-significantly. Thisurther increase in N/P ratio probably resulted in the presence ofn excess of “empty” liposomes that may have competed withomplete lipoplexes for cell internalization thus resulting in lowerevels of transfection [35]. Although fluctuations in luciferase activ-ty was seen in HepG2 cells exposed to varying N/P ratios ofhol-T Chol-�-Gal 5% PEG lipoplexes, these differences were con-idered not significant. These findings corroborate the fact that the

Please cite this article in press as: K. Naicker, et al., PEGylated galactosSurf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014

ipid/DNA ratio is an important parameter to be considered wheneveloping lipid complexes.

PRESS: Biointerfaces xxx (2014) xxx–xxx

The specificity of ASGP-R mediated gene transfection was ver-ified by the competitive inhibition assay. Luciferase activity ofthe targeted lipoplexes decreased considerably (>83%; P < 0.001)compared to the unchallenged targeted transfection complexes.This confirms that AF molecules successfully inhibited the inter-nalization of the galactose-containing complexes. The decline intransfection levels with the challenged lipoplexes was consistentregardless of degree of PEGylation. This shows that PEGylation upto 5 mole% does not significantly interfere with lipoplex uptake andgene transfer mechanisms in vitro.

To directly visualize the transfected cells expressing pCMV-GFPreporter gene, lipoplex-mediated GFP expressions in HEK293 andHepG2 cells, were observed by an inverted fluorescent microscope.Fig. 5 illustrates the micrographs of GFP expression from trans-fected HEK293 and HepG2 cells at 48 h. Cells incubated with nakedpCMV-GFP (Fig. 5, Control) were barely transfected resulting in lit-tle or no fluorescence, validating that unaided DNA was incapableof successfully transfecting either of the two tested cell lines. Inagreement with luciferase gene expression results, green fluores-cence intensity was greater in HepG2 cells than HEK293. There wasalso higher GFP expression in both cell lines as PEGylation wasintroduced (0%–2%) and then increased (2%–5%).

4. Conclusion

Significant advantages conferred by PEGylated complexes overnon-PEGylated complexes included, (1) smaller and colloidallystable particles, (2) good lipoplex forming abilities due to com-paction and condensation properties and substantial protectionof the DNA cargo, (3) good cell tolerability and, (4) enhancedin vitro transfection of HepG2 cells largely through ASGP-R medi-ated recognition and uptake. This confirms that steric protectionof cationic lipoplexes is necessary for physiological stability andimproved transfection efficiency. The targeted lipoplexes bear-ing the galactosylated moiety were efficiently recognized by theASGP-Rs on the target cell resulting in higher transfection activitieslargely through receptor-mediated uptake of the lipoplexes. Theselipoplexes prove to be prospective second-generation cationic genedelivery vectors with potential application in hepatoma gene ther-apy. Future studies include evaluating the size-dependent uptakeof our lipoplexes by Kupffer cells (macrophages in the liver) andpotential in vivo delivery.

Acknowledgments

This work was financially supported by the National ResearchFoundation (NRF) (Grant No. SFH2009072900005161) of SouthAfrica. We also thank Dr. James-Wesley Smith for cryo-TEM analy-sis.

Appendix A. Supplementary data

Supplementary data associated with this article can befound, in the online version, at http://dx.doi.org/10.1016/j.colsurfb.2014.07.010.

References

[1] D.L. Elbert, J.A. Hubbell, Surface treatments of polymers for biocompatibility,Annu. Rev. Mater. Res. 26 (1996) 365–394.

[2] J.-H. Lee, H.B. Lee, J.D. Andrade, Blood compatibility of polyethylene oxide sur-faces, Prog. Polym. Sci. 20 (1995) 1043–1079.

ylated cationic liposomes for hepatocytic gene delivery, Colloids.07.010

[4] H. Nordenstedt, D.L. White, H.B. El-Serag, The changing pattern of epidemiologyin hepatocellular carcinoma, Dig. Liver Dis. 42 (2012) S206–S214.

ING ModelC

aces B

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

potential for liver targeting, Carbohydr. Res. 367 (2013) 41–47.

ARTICLEOLSUB-6525; No. of Pages 9

K. Naicker et al. / Colloids and Surf

[5] C. Plank, K. Zatloukal, M. Cotten, K. Mechtler, E. Wagner, Gene transferinto hepatocytes using asialoglycoprotein receptor mediated endocytosis ofDNA complexed with an artificial tetra-antennary galactose ligand, Bioconjug.Chem. 3 (1992) 533–539.

[6] P.C.N. Rensen, L.A.J.M. Sliedrgt, M. Ferns, E. Kieviet, S.M.W. van Rossenberg, S.H.van Leeuwen, T.J.C. van Berkel, E.A.L. Biessen, Determination of the upper sizelimit for uptake and processing of ligands by the asialoglycoprotein receptoron hepatocytes in vitro and in vivo, J. Biol. Chem. 276 (2001) 37577–37584.

[7] X.-Q. Zhang, X.-L. Wang, P.-C. Zhang, Z.-L. Liu, R.-X. Zhuo, H.-Q. Mao, K.W.Leong, Galactosylated ternary DNA/polyphosphoramidate nanoparticles medi-ate high gene transfection efficiency in hepatocytes, J. Control. Release 102(2005) 749–763.

[8] J.E. Zijderhand-Bleekemolen, A.L. Schwartz, J.W. Slot, G.J. Strous, H.J. Geuze,Ligand- and weak base-induced redistribution of asialoglycoprotein receptorsin hepatoma cells, J. Cell Biol. 104 (1987) 1647–1654.

[9] C. Bies, C.M. Lehr, J.F. Woodley, Lectin-mediated drug targeting: history andapplications, Adv. Drug Deliv. Rev. 56 (2004) 425–435.

10] R.B. Dodd, K. Drickamer, Lectin-like proteins in model organisms: implicationsfor evolution of carbohydrate-binding activity, Glycobiology 11 (2001) 71–79.

11] M. Monsigny, P. Midoux, R. Mayer, A.C. Roche, Glycotargeting: influence of thesugar moiety on both the uptake and the intracellular trafficking of nucleic acidcarried by glycosylated polymers, Biosci. Rep. 19 (1999) 125–132.

12] S. Kawakami, S. Fumoto, M. Nishikawa, F. Yamashita, M. Nishida, In vivo genedelivery to the liver using novel galactosylated cationic liposomes, Pharm. Res.17 (2000) 306–313.

13] L.A.J.M. Sliedrgt, P.C.N. Resen, E.T. Rump, P.J. van Santbrink, M.K. Bijsterbosch,A.R. Valentijn, G.A. van der Marel, J.H. van Boom, T.J. van Berkel, E.A. Biessen,Design and synthesis of novel amphiphilic dendritic galactosides for selectivetargeting of liposomes to the hepatic asialoglycoprotein receptor, J. Med. Chem.42 (1999) 609–618.

14] J. Gaucheron, C. Santaella, P. Vierling, In vitro gene transfer with a novel galac-tosylated spermine bolaamphiphile, Bioconjug. Chem. 12 (2001) 569–575.

15] B. Frisch, M. Carrie, C. Largeau, F. Mathey, C. Masson, F. Schuber, D. Scher-man, V. Escriou, A new triantennary galactose-targeted PEGylated gene carrier,characterization of its complex with DNA, and transfection of hepatoma cells,Bioconjug. Chem. 15 (2004) 754–764.

16] X. Gao, L. Huang, A novel cationic liposome reagent for efficient transfection ofmammalian cells, Biochem. Biophys. Res. Commun. 179 (1991) 280–285.

17] M. Singh, N. Kisoon, M. Ariatti, Receptor-mediated gene delivery to HepG2 cellsby tertiary assemblies containing cationic liposomes and cationized asialooro-somucoid, Drug Deliv. 8 (2001) 29–34.

18] M. Singh, C.B. Rogers, M. Ariatti, Targeting of glycosylated lipoplexes in HepG2cells: anomeric and C-4 epimeric preference of the asialoglycoprotein receptor,S. Afr. J. Sci. 103 (2007) 204–210.

Please cite this article in press as: K. Naicker, et al., PEGylated galactosSurf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014

19] J. van Meerloo, G.J.L. Kaspers, J. Cloos, Cell sensitivity assays: the MTT assay,Cancer Cell Cult.: Methods Mol. Biol. 731 (2011) 237–245.

20] S.M. Chowdhury, G. Lalwani, K. Zhang, J.Y. Yang, K. Neville, B. Sitharaman,Cell specific cytotoxicity and uptake of graphene nanoribbons, Biomaterials34 (2013) 283–293.

[

PRESS: Biointerfaces xxx (2014) xxx–xxx 9

21] C. Managit, S. Kawakami, F. Yamashita, M. Hashida, Effect of galactose densityon asialoglycoprotein receptor-mediated uptake of galactosylated liposomes,J. Pharm. Sci. 94 (2005) 2266–2275.

22] L. Xu, M.F. Wempe, T.J. Anchordoquy, The effect of cholesterol domains onPEGylated liposomal gene delivery in vitro, Therapeut. Deliv. 2 (2011) 451–460.

23] W.L.J. Hinrichs, F.A. Mancenido, N.N. Sanders, K. Braeckmans, S.C. de Smedt,J. Demeester, H.W. Frijlink, The choice of a suitable oligosaccharide to pre-vent aggregation of PEGylated nanoparticles during freeze-thawing and freezedrying, Int. J. Pharm. 311 (2006) 237–244.

24] O. Garbuzenko, Y. Barenholz, A. Priev, Effect of grafted PEG on liposome size andon compressibility and packing of lipid bilayer, Chem. Phys. Lipids 135 (2005)117–129.

25] T. Gjetting, N.S. Arildsen, C.L. Christensen, T.T. Poulsen, J.A. Roth, V.N. Handlos,H.S. Poulsen, In vitro and in vivo effects of polyethylene glycol (PEG)-modifiedlipid in DOTAP/cholesterol-mediated gene transfection, Int. J. Nanomed. 5(2010) 371–383.

26] S. Resina, P. Prevot, A.R. Thierry, Physicochemical characteristics of lipoplexesinfluence cell uptake mechanisms and transfection efficacy, PLoS ONE 4 (2009)6058–6069.

27] D. Simberg, S. Weisman, Y. Talmon, A. Faerman, T. Shoshani, Y. Barenholz, Therole of organ vascularization and lipoplex-serum initial contact in intravenousmurine lipofection, J. Biol. Chem. 278 (2003) 39858–39865.

28] G. Caracciolo, L. Callipo, S. Candeloro De Sanctis, C. Cavaliere, D. Pozzi, A. Laganà,Surface adsorption of protein corona controls the cell internalization mecha-nism of DC-Chol–DOPE/DNA lipoplexes in serum, Biochim. Biophys. Acta 1798(2010) 536–543.

29] M. Lundqvist, J. Stigler, G. Elia, I. Lynch, T. Cedervall, K.A. Dawson, Nanopar-ticle size and surface properties determine the protein corona with possibleimplications for biological impacts, Proc. Natl. Acad. Sci. U. S. A. 105 (2008)14265–14270.

30] S. Audouy, G. Molema, L. De Leij, D. Hoekstra, Serum as a modulator of lipoplex-mediated gene transfection: dependence of amphiphile, cell type and complexstability, J. Gene Med. 2 (2000) 465–476.

31] C.R. Dass, E.G. Saravolac, Y. Li, L.Q. Sun, Cellular uptake, distribution, and sta-bility of 10–23 deoxyribozymes, Antisense Nucleic Acid Drug Dev. 12 (2002)289–299.

32] Y. Xu, S.W. Hui, P. Frederik, F.C. Szoka Jr., Physicochemical characterization andpurification of cationic lipoplexes, Biophys. J. 77 (1999) 341–353.

33] M. Singh, J. Borain, N. Noor-Mahomed, M. Ariatti, The effect of pegylation onthe transfection activity of two homologous cationic cholesteryl cytofectins,Afr. J. Biotechnol. 10 (2011) 1400–1407.

34] S.M. Shah, P.O. Pathak, A.S. Jain, C.R. Barhate, M.S. Nagarsenker, Synthesis,characterization, and in vitro evaluation of palmitoylated arabinogalactan with

ylated cationic liposomes for hepatocytic gene delivery, Colloids.07.010

35] T.A. Balbino, A.A.M. Gasperini, C.L.P. Oliveira, A.R. Azzoni, L.P. Cavalcanti, L.G.de La Torre, Correlation of the physicochemical and structural properties ofpDNA/cationic liposome complexes with their in vitro transfection, Langmuir28 (2012) 11535–11545.