The influence of chitosan valence on the complexation and transfection of DNA The weaker the DNA-chitosan binding the higher the transfection efficiency
The DNA-chitosan polyplexes have attracted for some years now the attention of physical-chemists and biologists for their potential use in gene therapy, however, the correlation between the physicochemical properties of these polyplexes with their transfection efficiency remains still unclear. In a recent paper we demonstrated by means of DLS that the DNA-chitosan complexation is favored at acidic conditions considering that fewer amounts of chitosan were required to compact the DNA. As a second study, in the present work we analyze the influence of chitosan valence on the complexation and transfection of DNA. Three chitosans of different molecular weights (three different valences) are characterized as gene carriers at 25°C and pH 5 over a wide range of chitosan-Nitrogen to DNA-Phosphate molar ratios, N/P, by means of conductometry, electrophoretic mobility, isothermal titration calorimetry (ITC), transmission electron microscopy (TEM), atomic force microscopy (AFM), and β-galactosidase and luciferase expression assays.
Colloids and Surfaces B: Biointerfaces 82 (2011) 54–62
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces
journal homepage: www.elsevier.com/locate/colsurfb
The inﬂuence of chitosan valence on the complexation and transfection of DNA:
The weaker the DNA–chitosan binding the higher the transfection efﬁciency
, Pablo Taboada
, Florian Hartl
, Tobias Wagner
, Julio R. Rodríguez
Grupo de Nanomateriales y Materia Blanda, Departamento de Física de la Materia Condensada, Facultad de Física,
Universidad de Santiago de Compostela, E-15782 Santiago de Compostela, Spain
Grupo de Física de Coloides y Polímeros, Departamento de Física de la Materia Condensada, Facultad de Física,
Universidad de Santiago de Compostela, E-15782 Santiago de Compostela, Spain
Biontex Laboratories GmbH, D-82152 Martinsried/Planegg, Germany
Faculty of Precision Engineering and Micromechanics, Munich University of Applied Sciences, D-80335 Munich, Germany
Department of Mechanical and Process Engineering, Furtwangen University, D-78054 Villingen-Schwenningen, Germany
Received 4 May 2010
Received in revised form 14 July 2010
Accepted 6 August 2010
Available online 14 August 2010
The DNA–chitosan polyplexes have attracted for some years now the attention of physical-chemists and
biologists for their potential use in gene therapy, however, the correlation between the physicochemical
properties of these polyplexes with their transfection efﬁciency remains still unclear. In a recent paper
we demonstrated by means of DLS that the DNA–chitosan complexation is favored at acidic conditions
considering that fewer amounts of chitosan were required to compact the DNA. As a second study, in
the present work we analyze the inﬂuence of chitosan valence on the complexation and transfection
of DNA. Three chitosans of different molecular weights (three different valences) are characterized as
gene carriers at 25
C and pH 5 over a wide range of chitosan-Nitrogen to DNA-Phosphate molar ratios,
N/P, by means of conductometry, electrophoretic mobility, isothermal titration calorimetry (ITC), trans-
mission electron microscopy (TEM), atomic force microscopy (AFM), and ␤-galactosidase and luciferase
© 2010 Elsevier B.V. All rights reserved.
Looking at the compaction of DNA from a general perspective
might be helpful to understand the polyelectrolyte complexation
process per se, which is also important for the improvement of poly-
cation based gene delivery systems. DNA–polycation complexes
referred to as polyplexes are regarded as potential efﬁcient gene
delivery systems that may someday be used as a systematic means
to introduce the nucleic acid into living cells for therapeutic pur-
poses. A large body of literature has compared different formula-
tions of polyplexes with regard to their transfection efﬁciency. Most
commonly studied cationic polymers as gene carriers (vectors)
include chitosan, polyethylenimine, poly(l-lysine), poly(␤-amino
ester)s and poly(amidoamine) dendrimers [1–3] (and references
therein). Among them, chitosan is special for its biological prop-
erties such as biodegradability, biocompatibility, mucoadhesivity,
and permeability enhancer capacity [4,5]. Chitosan is a biodegrad-
Corresponding author. Tel.: +34 981563100; fax: +34 981520676.
E-mail addresses: manuel
firstname.lastname@example.org (M. Alatorre-Meda),
email@example.com (J.R. Rodríguez).
able polyaminosaccharide . Chemically it is a fully or partially
N-deacetylated derivative of chitin  extracted from crustacean
shells, and has been shown to be non-toxic in both animal  and
human trials . Chitosan has a positive charge and hydrophilic
character at acidic pHs . It is a continuum of primary aliphatic
amine that can be protonated by acids, the pK
of the chitosan
amine groups being around 6.3–6.5 . The presence of amino
groups in its backbone provides chitosan cationicity, and conse-
quently a capacity to form polyelectrolyte complexes with DNA.
Hybrid DNA–chitosan systems can be classiﬁed into two cate-
gories that differ in their mechanism of formation and morphology:
complexes and nanospheres. DNA–chitosan complexes are made
by simple mixing of solutions of DNA and of chitosan; whereas,
DNA–chitosan nanospheres are obtained by increasing the speed
of mixing and temperature upon blending of the two solutions
. Besides the mechanism of complexation, the chemical and bio-
chemical properties of the DNA–chitosan complexes also depend
on the chitosan cationic charge density and valence, the former
being inﬂuenced for the chitosan degree of acetylation and pH
of the medium, and the latter being proportional to the chitosan
molecular weight [9,10]. It is well known that the high charge
density of chitosan at pHs below its pK
results beneﬁcial for poly-
plex preparation, and also that its low charge density at pH 7.4
0927-7765/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
M. Alatorre-Meda et al. / Colloids and Surfaces B: Biointerfaces 82 (2011) 54–62 55
contributes to a low polyplex cytotoxicity and may facilitate the
intracellular release of DNA from the complex after its endocy-
totic cellular uptake . However, conﬂicting results have been
found for the chitosan-mediated DNA transfection [12–15]. Sev-
eral studies have promoted the use of high molecular weight (Mw)
chitosans [13,15,16], but others have found that lower Mw chi-
tosans were superior for gene transfer [12,14,17]. Similarly, some
derivatives including glycosylated [18–21], pegylated [22–24], and
trimethylated chitosans [22,25,26] have also been tried; neverthe-
less, the criteria and strategies for the design of efﬁcient chitosan
gene delivery systems remain inconclusive. On the other hand,
in what the physicochemical aspects is concerned, particularly at
acidic conditions, only a few papers have been published on the
complex formation. Early studies dealt with the size, -potential,
and morphology of the N/P ratios that produced maximal transfec-
tion (generally a ratio much higher than the isoneutrality point)
[15,27–29]; however, a complete physicochemical characteriza-
tion addressing the electrochemical, morphological, and energetic
properties of the polyplexes covering a wide range of chitosan-
Nitrogen to DNA-Phosphate molar ratios, N/P, is, to the best of our
knowledge, still missing. In a recently published work, our group
studied the inﬂuence of the molecular weight of chitosan and pH of
the media on the size and colloidal stability of DNA–chitosan poly-
plexes via dynamic light scattering (DLS) . By comparing the
effect of the three different pHs of 5, 6 and 6.5 on the DNA com-
paction, we concluded that the increased charge density of chitosan
resulting from a lowering of pH lead to greater binding afﬁnities
between chitosan and DNA for fewer chitosan was required to reach
the complexation. Consequently, and in order to further understand
the general aspects involved in the DNA–chitosan interactions, in
the present paper we study the inﬂuence of chitosan valence on
the complex formation at acidic conditions and on the subsequent
transfection of the complexes thereby formed. Three chitosans of
different molecular weights (three different valences) have been
studied as DNA carriers in terms of a wide range of N/P molar
ratios by means of conductometry, electrophoretic mobility, ITC,
TEM, AFM, and ␤-galactosidase and luciferase expression assays.
Unless otherwise stated, experimentation was carried out at pH 5
2.1.1. Cationic polymers
Chitosans (from crab shells) of low viscosity (≤200mPas,1%in
1% acetic acid, at 20
C), of middle viscosity (200–400 mPa s, 1% in
1% acetic acid, at 20
C), and of high viscosity (≥400mPas,1%in
1% acetic acid, at 20
C) were from Fluka. For simplicity, through-
out this work we refer to each chitosan in terms of its valence.
The chitosan valence was obtained by dividing the chitosan molec-
ular weight, Mw (previously obtained by our group as 111 ± 2,
266 ± 14, and 467 ± 18 kDa for chitosans of low, middle, and high
viscosity, respectively ) to the repeat unit-molecular weight
(161 g mol
). C(689), C(1652), and C(2901) are in consequence
used to identify the chitosans of low, middle, and high viscosity,
2.1.2. DNA and plasmids
Calf thymus DNA sodium salt with a reported molecular weight
of 10–15 million Da was from Sigma–Aldrich. Plasmids pCMV
Lac-Z (7 kbp, 1 mg mL
, suspended in water for injection, WFI),
and pCMV-Luc (6 kbp, 1 mg mL
, suspended in WFI) were from
PlasmidFactory. The purity of DNA and plasmids was tested by
UV–vis spectroscopy. The absorbance of the samples was mea-
sured at 260 nm (absorption band of nucleic acids) and at 280 nm
(absorption band of proteins). The absorbance ratio was found to
∼ 1.9 demonstrating that the samples were free of
proteins. (It has to be noted that samples presenting protein con-
tamination render values of A
< 1.8, see Ref. .)
2.1.3. Buffers for polyplex and nanospheres formation
Bis-Tris and EDTA (used in polyplex formation) as well as sodium
acetate and sodium sulfate (used in the nanosphere formation)
were from Sigma–Aldrich.
2.1.4. Transfection experiments
PRO transfection reagent (MEP) and HeLa cells
were from Biontex Laboratories GmbH, IZB/Planegg/Germany.
NaCl, KCl, and KH
from Roth, and Na
Applichem were used to prepare the phosphate buffered saline
buffer (PBS). Dulbecco’s Modiﬁed Eagle Medium (DMEM) and Fetal
Calf Serum (FCS) both from PAA laboratories were used as culture
medium and culture medium additive, respectively.
2.1.5. ˇ-Galactosidase assay
, KCl, MgSO
, and 2-␤-mercaptoethanol (used
to prepare the buffer A), Na
(used to prepare the stop solution),
and ONPG substrate were from Applichem. Tris and Igepal CA-630
(used to prepare the ONPG lysis buffer) were from Roth and from
2.1.6. Luciferase expression
, EDTA, and 1% Triton X-100 (used to prepare the lysis
buffer) were from Applichem. ATP sodium salt hydrate, MgCl
DDT (all from Applichem) and tricine (from SERVA) were used to
prepare the ATP solution. Tricine, MgCl
, DDT, and d-luciferin (from
P.J.K.) were employed to prepare the luciferin solution.
2.1.7. Protein concentration
BCA reagents (Reagent A and Reagent B) were from Thermo
All materials were used as received. Water puriﬁed in an
18 M cm MilliQ Plus water system was used throughout the work.
2.2. Sample preparation
It is difﬁcult to ascertain the exact amount of positive charges
contributed by chitosan, then the maximum possible positive
charge will be considered . Consequently, the repeat unit-
molecular weights used for the calculation of N/P were 161 and
330 g mol
for chitosan  and DNA , respectively.
2.2.1. Complex formation for physicochemical characterizations
DNA and chitosan stock solutions were prepared separately. The
DNA solutions of desired concentrations were, unless otherwise
stated, prepared in a 10 mM Bis-Tris–EDTA buffer, pH 5. Mean-
while, to prepare the chitosan stock solutions, chitosan was ﬁrstly
dissolved in 25 mM acetic acid at a concentration of 0.1% (w/v),
and then diluted to the desired concentrations with 10 mM Bis-
Tris–EDTA; once diluted the pH was re-adjusted to 5.
To form the polyplexes, both stock solutions were mixed, vor-
texed for 10 s, and left for incubation during 30 min at 25
A series of polyplexes with compositions of 1 ≤ (N/P) ≤ 20, was
prepared by adding chitosan solutions of desired concentrations
to a 0.061 mM DNA solution, both solutions being equal in vol-
ume. For conductometry experiments the Bis-Tris–EDTA buffer
and DNA concentrations were of 2 mM and 0.152 mM, respec-
tively. For the ITC experiments the DNA concentration in the cell
amounted to 0.152 mM, whereas the concentration of chitosan in
the injector amounted to 6.25 mM. For TEM, complexes with a
56 M. Alatorre-Meda et al. / Colloids and Surfaces B: Biointerfaces 82 (2011) 54–62
ratio N/P = 20 were prepared. For AFM, complexes with the same
composition as for TEM experiments were diluted 10-fold prior to
2.2.2. Transfection assays
DNA–chitosan polyplexes, DNA–chitosan nanoparticles, and a
DNA-MEP lipoplex were tested.
DNA–chitosan polyplexes were formed in the same way as for
the physicochemical characterization except for the fact that plas-
mids (1 mg mL
suspended in WFI), chitosan (0.1% (w/v) dissolved
in 25 mM acetic acid), and MEP were dissolved to the desired con-
centrations in 10 mM PBS (1× PBS), pH 7.4. A series of polyplexes
with compositions of 1 ≤ N/P ≤ 18 was prepared.
DNA–chitosan nanoparticles were formed following the pro-
tocol described by Mao et al. . Brieﬂy, plasmids (1 mg mL
suspended in WFI) and chitosans (powder) were dissolved in sep-
arate in 5 mM sodium sulfate and in 5 mM sodium acetate buffer,
pH 5.5, respectively. Both solutions were preheated to 50–55
at least 15 min. Equal volumes of both solutions were then quickly
mixed together and vortexed for 15–30 s. A series of nanoparticles
with compositions of 1 ≤ N/P ≤ 20 was prepared.
Finally, the DNA–MEP lipoplex, used as a positive blank, was
prepared in a ratio MEP to DNA of 4:1 (L:g). Once mixed, sam-
ples were incubated for 20 min at room temperature to reach the
The DNA concentration amounted to 0.091 mM (0.03 g L
in all cases.
Conductance data were collected with a SevenMulti
tivity meter from METTLER TOLEDO. The dip-type conductance cell
used was calibrated by measuring the conductivity of a series of
standard solutions of KCl at different concentrations. The cell con-
stant was determined to be equal to 0.139 cm
. To perform the
study, 20 mL of a reservoir solution (containing either DNA or pure
buffer) were placed in a conductivity cell that was continuously
stirred; then, aliquots of chitosan solutions were added. The instru-
ment measured the changes in the conductance of the solution after
chitosan injections; values of conductivity with deviations lower
than 0.4% were reported. DNA dilution after each chitosan injec-
tion was considered for the N/P calculations. The cell calibration and
measurements were carried out at 25.00 ± 0.01
C inside a Julabo
2.4. Electrophoretic mobility
The -potential of the samples was obtained with a Zetasizer
NanoSeries (Malvern) using folded capillary cells. The instrument
measured the electrophoretic mobility of the particles and con-
verted it to the -potential using the classical Smoluchowski
where ˛, ∈, , and denote the electrophoretic mobility, permit-
tivity of the media, -potential of the particles, and viscosity of the
media, respectively. All measurements were carried out at 25
The results are the average of ﬁve measurements.
Binding studies were performed using a VP-ITC titration
microcalorimeter from MicroCal Inc., Northampton, MA with a cell
volume of 1.355 mL at 25
C. Samples were degassed in a Ther-
moVac system (MicroCal) prior to use. The sample cell was ﬁlled
with the DNA solution and the reference cell with buffer solution
only. Chitosan solutions were introduced into the thermostated cell
by means of a syringe stirred at 250 rpm, which ensured rapid mix-
ing but did not cause foaming on solutions. Each titration consisted
of an initial 2 L injection (neglected in the analysis) followed by 55
subsequent 5 L injections programmed to occur at 400 s intervals,
sufﬁcient for the heat signal to return to the baseline. We present
the results of the ITC experiments in terms of the enthalpy change
per injection, Q, as a function of the N/P molar ratios. Heats of dilu-
tion from titrations of chitosan solutions into buffer only (without
DNA) were subtracted from the heats obtained from titrations of
the polymer solutions into the DNA solution to obtain the net bind-
ing heats. All experiments were carried out in duplicate and the
reproducibility was within ±3%.
Raw ITC data of polymers binding to DNA were processed as
described previously . Brieﬂy, the isotherms were ﬁtted by a
nonlinear least-squares analysis to a two-binding-site model. This
model employs the following ﬁtting equation that incorporates
Langmuir isotherm binding equilibria for two independent types
of association, where Q is the heat per injection, M is the macro-
molecule concentration (in this case DNA), V is the volume of the
cell, n and H are the stoichiometry and enthalpy of interactions,
respectively, and is the fraction of ligand bound to the macro-
Q = MV(n
where the sub-indices 1 and 2 stand for the two sets of sites.
One can solve for
using the equilibria equations for
binding constants K
, being X the concentration of ligand
and [X] the concentration of free ligand (the ligand in this case is
[X] = X − M(n
To achieve an accurate ﬁt of all six ﬂoating parameters to our
data, multiple attempts were performed starting from different ini-
tial parameters. The same six values were reached at the minimum
, regardless of the values of initialization.
TEM sample preparation and image acquisition procedures
followed in this work are described elsewhere . Brieﬂy, the sam-
ples were applied to carbon-coated copper grids, blotted, washed,
negatively stained with 2% (w/v) of phosphotungstic S5 acid, air
dried, and then examined with a Phillips CM-12 transmission elec-
tron microscope operating at an accelerating voltage of 120 kV.
Neither dilution nor addition of more than one drop to the grid
were needed. After preparation, the samples were left for incuba-
tion overnight prior to imaging. All experiments were carried out
at room temperature.
Tapping Mode AFM in air was performed with a Multimode
SPM (Nanoscope IIIa, Digital Instruments). The samples were
deposited on freshly cleaved mica, left for incubation during 2 min
and dried by capillarity immediately after. The experiments were
run using a J tube scanner (scan size: 10 m × 10 m, vertical
range: 5 m). Microfabricated crystal silicon probes with a spring
constant of 20–80 N/m and a resonant frequency of 281–319 kHz
(Veeco MPP-11100) were used as received. Z-scale accuracy was
checked once a day by means of a silicon grating (TGZ02 silicon grat-
ing, from Ultrasharp Cantilevers and Gratings) ensuring a nominal
M. Alatorre-Meda et al. / Colloids and Surfaces B: Biointerfaces 82 (2011) 54–62 57
height deviation lower than 2% at the highest scan size. To elimi-
nate imaging artifacts, the scan direction was varied to ensure a true
image. The images were obtained from at least ﬁve macroscopically
separated areas on each sample. All images were processed using
procedures for plane-ﬁt and ﬂatten in the WSxM 4.0 Develop 11.4
software  without any ﬁltering. All experiments were carried
out at room temperature.
2.8. Transfection efﬁciency
Transfection experiments were performed in triplicate. HeLa
cells with an optical conﬂuence of 80–90% were seeded into wells
) of a 48-well plate (200 L, 1 × 10
cells/well) and grown
at standard culture conditions (DMEM supplemented with 10% FCS
in an atmosphere of 10% CO
C) for 24 h. Afterwards, the com-
plexes formed as described before were added and incubated for
6 h at standard culture conditions. Then, the medium containing
the complexes was exchanged with fresh medium and cells were
incubated for further 42 h. Finally, the (i) ␤-galactosidase and (ii)
luciferase expressions as well as (iii) the total protein concentration
per well were determined as follows.
(i) ˇ-Galactosidase assay: The culture medium was discarded, and
the cells were washed with 1× PBS. The cells were then shaken
at room temperature (240 rpm, 15 min) in the presence of
120 L of lysis buffer (0.25 M Tris (pH 7.8), 0.6% Igepal CA-630).
Once the cells were lysed, 120 Lof1× PBS were added and
homogenized by pipetting up and down several times. Mean-
while an aliquot of 20 L was separated for the protein content
determination, the remaining 220 L of cell lysate were mixed
with 250 L of buffer A (100 mM KH
10 mM KCl, 1 mM MgSO
, and 50 mM 2-␤-mercaptoethanol).
Then, 60 L of ONPG substrate solution (10 mg mL
100 mM Na
O, pH 7.5) were added to the mixture
and incubated at room temperature until the solution became
yellow (8 min). After incubation, the reaction was terminated
by increasing the pH of the solution up to 11 (pH at which ␤-
galactosidase is inactivated) upon addition of 500 L of stop
solution (1 M Na
). Then, in order to get rid of any precip-
itate that could appear, the samples were cooled down at 4
for 15 min and warmed up at room temperature for several
minutes. Finally, the absorbance of the samples was measured
at 405 nm with an ELx800 Absorbance Microplate Reader from
Biotek. The analysis was carried out at room temperature in
(ii) Luciferase expression: The culture medium was discarded, and
the cells were washed with 1× PBS. In this case, the cells were
handled above a refrigerated gel pack to maintain the tem-
perature around 4
C. The cells were shaken (240 rpm, 30 min)
in the presence of 100 L of lysis buffer (100 mM Na
7.8), 2 mM EDTA (pH 8), 1% Triton X-100). In order to avoid
debris after lysation, the cell lysates were homogenized by
taking off and reﬁlling a constant volume of 80 L several
times. Then, like in the ␤-galactosidase assay, 20 L of cell
lysate were separated for the protein content determination.
The remaining 80 L were mixed with 80 L of ATP solution
(30 mM tricine, 3 mM ATP, 0.225 mM MgCl
, 1 mM DDT) and
shaken (320 rpm, 5 min). Finally, 40 L of each well were trans-
ferred to an opaque white plate and measured for 1 s for their
luminescence in the presence of luciferin solution (0.5 mM
d-luciferin, 30 mM tricine, 0.225 mM MgCl
, 1 mM DDT). The
analysis was carried out at non-sterile conditions. The lumines-
cence determination was performed with a MicroLumat Plus
Luminometer from Berthold Technologies.
(iii) Protein concentration: The protein concentrations of the cell
lysates were determined via the BCA assay. The 20 L aliquots
separated from the previous assays were mixed with 400 L
of BCA solution (98% Reagent A). Both samples, from the ␤-
galactosidase and luciferin assays, were incubated ﬁrstly at
the standard culture conditions (the former for 30 min and
the latter for 90 min), and secondly at room temperature (both
samples for 15 min). After incubation, the absorbance of the
samples was measured at 562 nm. The absorbance determina-
tion was carried out at non-sterile conditions with the same
instrument as for the ␤-galactosidase assay.
3. Results and discussion
The polyplex formation is a process dominated by electrostat-
ics, therefore, a careful electrochemical study of the polyplexes in
solution is strongly recommended to better understand the inter-
actions. The next two sections detail the electrochemical analysis
that was done at two levels: (i) the bulk, by means of conductometry
experiments (Section 3.1), and (ii) the surface charge of the poly-
plexes formed, by means of electrophoretic mobility experiments
Upon mixing, oppositely charged compounds interact electro-
statically and form complexes in a process that is accompanied
by a release of counterions [38,39]. The tracking of this release by
means of conductometry can be employed as a tool to assess the
conditions at which the interactions take place as recently demon-
strated for the DNA compaction mediated by surfactants  and
liposomes . In the present work we measured the change in
conductivity provoked by the addition of chitosan solutions to
both, DNA and pure buffer solutions. Unlike the physicochemi-
cal experiments detailed in following sections, for the conductivity
experiments here presented the DNA and chitosan solutions were
prepared in 2 mM Bis-Tris–EDTA buffer, pH 5. This lower concen-
tration was chosen based on preliminary results showing that the
higher ionic strength of 10 mM of the buffer masked the changes
in conductivity related to the DNA–chitosan interactions. Also the
concentrations of the DNA and chitosan solutions were tunned. The
DNA concentration at which the other physicochemical tests were
run (0.061 mM, see next sections) turned out to be too low to pro-
voke a detectable change in the conductivity outcome, whereas a
concentration of 0.303 mM led to polyplex precipitation during the
polycation addition. Consequently, on assumption that the elec-
trostatic interactions in the DNA compaction is DNA-concentration
independent , the DNA concentration chosen for this experi-
ment was of 0.152 mM.
Fig. 1 presents a representative plot of the electrical conductiv-
ity, , as a function of the polycation concentration, recorded for
C(2901). The ﬁlled and empty squares stand for the DNA and pure
buffer reservoir solutions, respectively. was calculated from the
, and corrected for the conductiv-
ity of the buffer solution,
, according to =
. As can be
seen, in the buffer solution alone the conductivity increased linearly
with the chitosan concentration (empty squares) indicating that
no aggregation took place under the whole range of the polycation
concentration. In the DNA solution (ﬁlled squares) by contrast, the
conductivity grew linearly, however, with a clear change in slope at
the chitosan concentration of 0.104 mM, corresponding to the N/P
ratio of 0.82. A similar inﬂection in a conductivity plot during DNA
compaction upon addition of a cationic vector has been observed
previously [41,42]. The authors suggested that the increase in con-
ductivity related to the counterion release from the polycation
injected is accentuated by the release of counterions resulting from
the complexation process (in our case Na
from DNA and CH
58 M. Alatorre-Meda et al. / Colloids and Surfaces B: Biointerfaces 82 (2011) 54–62
Fig. 1. Electrical conductivity, , vs. C(2901) concentration. Filled and empty squares
stand for the addition of chitosan to a DNA and to a pure buffer solution, respectively.
(Note the difference in scales.)
from chitosan) thereby justifying a higher slope in the conductivity
plot below the inﬂection point. On the other hand once the inﬂec-
tion occurred, the lower slope was attributed to the fact that only
the counterions coming from the chitosan dissociation now con-
tribute to the conductivity of the solution. This change in slope of
the conductivity plot can in consequence be considered as the point
from which DNA is compacted [41,42].
Compared to the DNA compaction ratio we determined by
2 (chitosan/DNA mass ratio = 1 in Ref. ), the lower
ratio of N/P = 0.82 here suggested can be ascribed to the differ-
ence in ionic strength of the media used in both experiments
and to the fact that contrary to DLS experiments in the con-
ductivity experiment the complex formation is run at constant
stirring. Similar results of fewer quantities of cationic vector
needed to compact DNA as ionic strength decreases have been
observed somewhere else. For instance, the compaction of calf thy-
mus DNA by cationic liposomes constituted by a 1:1 mixture of
a cationic lipid, 1,2-distearoyl-3-(trimethylammonium) propane
chloride (DSTAP), and a zwitterionic lipid, 1,2-dioleoyl-sn-glycero-
3-phosphatidylethanolamine (DOPE) showed isoneutrality at a
liposome to DNA mass ratio (L/D) = 3.6 when dissolved in water;
whereas when dissolved in 40 mM HEPES the isoneutrality point
was determined as (L/D) = 5.6 .
Concerning the other two chitosans, C(689) and C(1652),
although with slight differences in the conductivity values, they
revealed the inﬂection point at exactly the same N/P ratio as com-
pared to C(2901) (plots not shown).
3.2. Electrophoretic mobility
With the interaction of DNA with a cationic vector, the negative
charge of DNA is expected to diminish or even to shift to posi-
tive; this phenomenon facilitates the DNA approach and uptake
through the cell membrane during transfection [43–46]. For many
polyplexes the cross-over from a negative to a positive -potential
occurs at or very close to the isoneutrality point (N/P)ϕ .(N/P)ϕ
is deﬁned as the point at which the N/P ratio of the polyplex equals
1, that is, the ratio where the negative charges of DNA are stoichio-
metrically neutralized by the positive charges of the polycation .
In the present study, the characterization was done in the range
1 ≤ N/P ≤ 20 for all polyplexes.
The -potential of the polyplexes is plotted as a function of
N/P in Fig. 2. It can be seen that at N/P = 1 all polyplexes, in par-
ticular those formed with C(1652) and C(2901), reveal a lower
-potential as compared to the rest of compositions. The cationic
Fig. 2. -potential of DNA–chitosan polyplexes vs. N/P. DNA–C(689) (squares),
DNA–C(1652) (circles), and DNA–C(2901) (triangles) are plotted. The dotted line
stands for the DNA solution. Results are the average of 5 runs. Uncertainty bars
represent the standard deviation.
vector-mediated DNA coil to globule transition demonstrated by
other authors [48–51] in conjunction with the base line-absent
DLS correlation functions we obtained for these systems at ratios
N/P ≤ 1 , may provide an explanation to this feature. Apparently,
larger amounts of chitosan are needed to reach a complete DNA
compaction and in consequence populations with varying extents
of DNA compaction are expected to be present in the bulk. On
the other hand, at ratios higher than (N/P)ϕ, all polyplexes reach
a plateau around 16 mV regardless of chitosan Mw, which is in
good agreement with other systems including polyaminoacids 
and cationic polymers in either linear conﬁgurations or dendrimer
structures . This positive -potential of the polyplexes suggests
that DNA compaction is completely achieved with chitosan chains
probably pointing to the outer part of the polyplexes as inferred by
other authors .
The capability of DNA–polycation complexes to avoid prema-
ture dissociation from media and to promote the release of genetic
materials from the endosome once inside the cell are parameters
strongly related to the binding afﬁnity between the DNA and the
polycation in question . Isothermal titration calorimetry was
performed to investigate the DNA binding afﬁnity of chitosan.
ITC is an extremely sensitive technique in which the summation
of several heat effects determines the shape of the binding isotherm
. In fact, the dilution of both the polycation and DNA solu-
tions, condensation and/or aggregation of the resultant polyplexes,
coupled protonation effects, and possible conformational changes
upon binding are contributions that commonly affect the ﬁnal
results . To obtain accurate thermodynamic binding parame-
ters, the nonbinding heat contributions from the polymer dilution
were subtracted before applying the curve-ﬁtting algorithms.
Fig. 3 shows the enthalpy per injection as a function of the N/P
molar ratios for chitosans upon binding with DNA after the sub-
traction of polymer dilution effects.
Supported on calorimetric measurements it is well accepted
that polyelectrolyte complex formation and coacervation are
mainly entropically driven through the release of condensed coun-
terions via the ion-exchange process in which an endothermic
signal recorded during complex formation is typically present
[39,54]. By contrast, in a result most commonly observed in the
formation of protein–ligand complexes, it can be observed from
Fig. 3 that for all experiments the injection of chitosan appears as
a markedly negative signal at the beginning of the binding process,
M. Alatorre-Meda et al. / Colloids and Surfaces B: Biointerfaces 82 (2011) 54–62 59
Fig. 3. Integrated heats of interaction of the titration of chitosan to DNA vs. N/P.
Chitosans C(689) (squares), C(1652) (circles), and C(2901) (triangles) were titrated.
Solid lines represent the two site model ﬁtting to the experimental data. The heat
contributions from chitosan dilution were subtracted.
that is, complexation is exothermic. A similar result was previ-
ously found by Bharadwaj et al. for the poly(styrene sulfonate)
(PSS)–poly(allylamine hydrochloride) (PAH) complexation .
The authors observed a decrease in the quantity of heat released on
successive injections of titrant; this decrease was interpreted as an
indicative of the progressive neutralization of charges in the reser-
voir molecule. The zone of the thermogram in which a plateau in
the heat released was reached was attributed to the complete DNA
In this context, comparing the onset of the plateau yielded for
the three chitosans during the DNA compaction it is clear that for
C(2901) N/P ∼ 1, whereas for C(689) and C(1652) N/P ∼ 0.5. This
result is consistent with those we observed by DLS  and -
potential (Section 3.2) indicating that the use of higher valence
chitosans, even at low pHs, apparently demands larger amounts
of cationic polymer for DNA compaction. This is also supported by
the 4–5-fold smaller enthalpic contribution rendered by C(2901)
compared to its lower valence homologues.
3.3.1. Thermodynamic parameter determination
Table 1 summarizes the enthalpy, entropy, binding constant,
and the stoichiometry of the DNA–chitosan interaction derived on
the basis of the two-binding-site model.
As can be observed from Table 1, the DNA binding constants
obtained for all chitosans are on the order of 10
for the ﬁrst and second class of binding sites, respectively.
These results are in good agreement with previously reported val-
ues for other systems including cationic polymers [35,56,57] and
proteins [58,59]. On the other hand, the decreasing values of the
binding constants with the chitosan valence is an indicative that
chitosan chain may undergo steric restrictions as Mw (valence)
increases, restrictions that in turn apparently hamper the inter-
polyelectrolyte interactions [9,10,30].
Concerning the enthalpy, it is well known that it results from
a combination of electrostatics, conformational changes (espe-
cially for second binding sites), and hydrogen bonding interactions;
therefore, H cannot be strictly related to any one contribution.
However, and despite the experimental evidence demonstrating
that the binding enthalpy H was negative, the DNA–chitosan
complexation was proven to be entropically driven. This result is
in good agreement with other electrostatic polyelectrolyte associ-
ations promoted by the release of counterions and solvent upon
3.4. TEM and AFM
Single molecule experiments can detect, localize, and ana-
lyze individual aggregates of a heterogeneous population, thereby
revealing events that would otherwise be hidden. In this context
TEM and AFM are frequently used in parallel for the visual charac-
terization of biological molecules [42,62–64].
Fig. 4 presents typical TEM (A) and AFM (B) images obtained for
the DNA–C(689) polyplexes at N/P = 20. Polyplexes formed with the
other chitosans at the same ratios are not shown due to the similar
morphology they exhibited with respect to the one presented.
It has been reported previously that the DNA compaction with
chitosans might result in a blend of structures: toroids, rods, and
globules, with the relative amounts of the different structures
apparently depending on the actual chitosan, charge ratio, and
solution properties like pH and ionic strength [9,10]. What we
observe from Fig. 4A and B is a heterogeneous population of poly-
plexes with particle sizes ranging from 250 to 500 nm in good
agreement with a previous DLS characterization . Both images
depict polyplexes with a brush-like conformation where glob-
ules/aggregates comprise a dense core that is surrounded by a
“hairy” shell of polymer chains. As demonstrated by Maurstad et al.
, this globular conformation is characteristic of the DNA com-
plexation with high molecular weight chitosans (Mw > 100 kDa,
from Ref. ); by contrast, complexes formed with lower molecular
weight chitosans adopt toroid- and rod-like conformations. Similar
brush-like structures were also obtained for the DNA complexation
with transferrin-poly(l-lysine) conjugates. In this case, the com-
plex morphology was found to depend on the conjugate to DNA
ratio . Carnerup et al. suggest that the signiﬁcant morphological
rearrangement undergone by DNA when it is condensed with low
molecular weight polycations is because of the low charge density
of the polycation in question . In order for toroidal aggregates to
form, the electrostatic attraction has to be moderate; that is, a bal-
ance between mobility and high afﬁnity binding of the DNA to the
polymer has to exist. In such a system, the condensed DNA chains
will be able to arrange into a toroid. On the contrary, if the charge
density of the polymer is too high (as expected for chitosan at pH
5), the DNA chains will entangle with the polymer ones, forming
globular aggregates .
Thermodynamic parameters of the DNA–chitosan binding process. Sub-indices 1 and 2 next to each parameter stand for the two types of sites.
C(689) 29.9 ± 0.75 0.27 ± 0.03 −2.176 ± 0.23 0.022
C(1652) 29.0 ± 1.23 0.27 ± 0.05 −2.712 ± 0.12 0.020
C(2901) 4.82 ± 0.08 0.73 ± 0.01 −0.598 ± 0.07 0.024
C(689) 0.095 ± 0.01 0.86 ± 0.02 −1.304 ± 0.16 0.014
C(1652) 0.011 ± 0.02 0.75 ± 0.04 −1.357 ± 0.12 0.014
C(2901) 0.009 ± 0.01 0.62 ± 0.07 −0.529 ± 0.02 0.012
60 M. Alatorre-Meda et al. / Colloids and Surfaces B: Biointerfaces 82 (2011) 54–62
Fig. 4. TEM (A) and height AFM (B) images of DNA–C(689) polyplexes, N/P = 20. The bar next to (B) represents the Z-scale in nm.
Very importantly, the morphological structure of our system
depicted by TEM and AFM in conjunction with the markedly posi-
tive -potentials obtained for the polyplexes at this high N/P ratio
(Section 3.2) appear to be in line with the core–shell structure pro-
posed for polycation-excessive DNA complexes . This model
states that DNA is condensed in the inner part of the polyplex
by the binding of short segments of a large number of poly-
cation chains, whereas the remaining segments of these same
chains are expected to be free in the outer part of the poly-
plexes giving raise to markedly positive polyplex surface charges
3.5. Transfection efﬁciency
Transfection efﬁciency was evaluated by the expression of the ␤-
galactosidase and luciferase reporter enzymes. In order to avoid any
kind of contamination related to the commercial polymer samples,
two protocols prior to the polyplex formation were followed. On the
one hand polymers were sterile ﬁltered (ﬁlter pore size of 0.22 m,
calcium acetate ﬁlter), on the other hand polymer solutions were
added with 0.2% wt of sodium azide (a biocide) and were left to
incubate for at least 12 h. The (i) ␤-galactosidase and (ii) luciferase
assays are described below:
(i) ␤-galactosidase assay
The potential of chitosans C(689) and C(2901) as DNA car-
riers towards HeLa cells was evaluated. The polyplexes were
formed at ratios in the range 1 ≤ N/P ≤ 18. In addition, a MEP-
DNA lipoplex (4:1 L:g) and naked DNA were measured as
positive and negative controls, respectively. The speciﬁc activ-
ity, reported as the transfection efﬁciency, was calculated by
dividing the sample absorbance by the product of the total pro-
tein mass obtained from the BCA assay (in mg) and the ONPG
incubation time (in s).
Fig. 5 shows the transfection efﬁciency of the polyplexes and
controls as well as the protein content of the wells after lysa-
tion. Two features are observed from Fig. 5. On the one hand it is
clear that polyplexes within the whole range of ratios rendered
levels of ␤-galactosidase expression slightly higher than that of
the negative control (DNA without polymer) with transfection
efﬁciency being increased with chitosan valence. This result is
somehow logical taking into account the lower binding afﬁni-
ties depicted by ITC assays for the DNA–C(2901) complex (see
Table 1); that is, the DNA release from this complex in the inte-
rior of the cell is expected to be favored. On the other hand,
compared to that of the DNA–MEP lipoplex, the transfection
efﬁciency of the polyplexes is considerably lower.
Speculating that the low transfer rate of the polyplexes (com-
pared to that of the lipoplex) might be related to cytotoxicity
effects, the protein concentration was determined via the BCA
assay. From Fig. 5 (right hand side axis) a comparable level of
protein content for all the formulations including the blanks
was observed. Even if not totally conclusive regarding a measure
of the cytotoxicity, this result shows that the cells prolifer-
ated approximately in the same way; consequently, the poor
transfection efﬁciency shown by the DNA–chitosan polyplexes
cannot be ascribed to cytotoxicity.
Finally, suspecting that the transfection results of the poly-
plexes might have been brought about by the ﬁltering of the
polymers, another protocol was proposed. Instead of being ster-
ile ﬁltered, the chitosan solutions were added with 0.2% wt
sodium azide prior to the complexation. The results of this
experiment showed that transfection efﬁciencies and protein
contents were comparable to those obtained with ﬁltered sam-
ples (data not shown), hence, it could be concluded that the
results did not depend on the polymer treatment.
In the light of the transfection efﬁciency depicted by the
polyplexes, a second kind of experiment was implemented.
Fig. 5. Transfection efﬁciency of complexes (columns) and protein content in wells
after lysation (squares) vs. N/P. Uncertainty bars represent the standard deviation
of 3 experiments.
M. Alatorre-Meda et al. / Colloids and Surfaces B: Biointerfaces 82 (2011) 54–62 61
DNA–chitosan nanospheres were prepared following the pro-
tocol described by Mao et al. . The transfection efﬁciency
results for C(689) are also plotted in Fig. 5. Transfection experi-
ments were not run with C(2901) for this chitosan could not be
dissolved in sodium acetate.
Compared to the polyplexes, no transfection improvement
was observed, on the contrary, nanospheres in general showed
a decreasing in gene expression yielding results only compa-
rable to the naked DNA administration. It has to be noted that
the same steps regarding chitosan treatment prior to complex-
ation were attended, and similarly to polyplexes, no effect was
(ii) Luciferase expression assay
The luciferase assay was conducted in order to conﬁrm the
results obtained by the ␤-galactosidase assay. The transfection
efﬁciency for all cases was similar to that observed in the ␤-
galactosidase method. In general, naked DNA and complexes
regardless of the N/P ratio and structure yielded a luminescence
three orders of magnitude lower than that of the DNA–MEP
lipoplex. Concerning the protein content determined by the
BCA assay, the polyplexes and nanospheres rendered protein
contents slightly higher than that of the DNA–MEP lipoplex
(data not shown). In consequence, the transfection efﬁciency
of polyplexes was conﬁrmed to be low compared to that of the
In general, the low capacity of DNA to escape from endosomes
is regarded as one of the major limitations for the transfection
efﬁciency of polyplexes . This feature has been ascribed to a
number of factors. While some authors support that an excess of
polycation, even in the presence of chloroquine, limits the pro-
tein expression due to in vitro cytotoxicity , other authors
afﬁrm that the low tolerance to dissociation observed by DNA
from polyplexes is presumably due to the bulky form in solution
they adopt .
In our case the low transfection efﬁciency of the
DNA–chitosan polyplexes compared to that of the lipoplex is
most likely related to the colloidal properties they exhibited,
that is, the morphological structure depicted by TEM and
AFM in conjunction with the markedly positive -potentials
obtained that suggest a core–shell structure with chitosan
occupying the outer part of the complex. In consequence, the
possibility for the enzymes ␤-galactosidase and luciferase
(attached to DNA) to be reached by the agents needed for their
expression (ONPG for the former and ATP, Mg
, and O
latter) is hampered.
Finally, as mentioned before, the increment of transfec-
tion efﬁciency observed with chitosan valence seems to be
explained by the fact that the DNA–chitosan binding strength
lowers as chitosan valence increases.
In general, all polyplexes exhibited good and similar colloidal
properties over the whole range of N/P ratios studied irrespective of
chitosan valence; nevertheless, the DNA–chitosan binding afﬁnity
and the transfection efﬁciency revealed to be valence-dependent.
On the one hand the binding constants obtained by ITC, K, were
found to follow a decreasing trend with chitosan valence; on the
other hand, although considerably lower compared to that of the
DNA–MEP lipoplex tested as a blank, the transfection efﬁciency
of the polyplexes increased with chitosan valence as depicted by
the ␤-galactosidase and luciferase expression assays. These results
demonstrate that the transfection efﬁciency is hampered by the
DNA–polycation binding strength.
The authors are grateful to the Department of Physical Chem-
istry 1, Center for Chemistry and Chemical Engineering, Lund
University, to Dr. Ulf Olsson in particular for lending AFM facilities
and for his fruitful discussions.
The authors also thank Dr. Alfredo González-Perez (University
of Southern Denmark) and Luis Pegado (Lund University) for their
Pablo Taboada thanks Xunta de Galicia for ﬁnancial support by
project INCITE 09206020PR.
Julio R. Rodriguez thanks the Dirección Xeral de I+D+I of the
Xunta de Galicia and the European Regional Development Fund
(INCITE07PXI206076ES) for the ﬁnancial supports.
Manuel Alatorre-Meda is supported by the Programme Alban,
the European Union Programme of High Level Scholarships for Latin
America, scholarship No. (E06D101860MX).
 S.C. De Smedt, J. Demeester, W.E. Hennink, Cationic polymer based gene deliv-
ery systems, Pharm. Res. 17 (2000) 113–126.
 G. Borchard, Chitosans for gene delivery, Adv. Drug Deliv. Rev. 52 (2001)
 H. Jiang, P. Taranekar, J.R. Reynolds, K.S. Schanze, Conjugated polyelectrolytes:
synthesis, photophysics, and applications, Angew. Chem. Int. Ed. 48 (2009)
 L. Illum, Chitosan and its use as a pharmaceutical excipient, Pharm. Res. 15
 M. Kumar, R.A.A. Muzzarelli, C. Muzzarelli, H. Sashiwa, A.J. Domb, Chi-
tosan chemistry and pharmaceutical perspectives, Chem. Rev. 104 (2004)
 H. Onishi, Y. Machida, Biodegradation and distribution of water-soluble chi-
tosan in mice, Biomaterials 20 (1999) 175–182.
 S.B. Rao, C.P. Sharma, Use of chitosan as a biomaterial: studies on its safety and
hemostatic potential, J. Biomed. Mater. Res. 34 (1997) 21–28.
 T.J. Aspden, J.D.T. Mason, N.S. Jones, J. Lowe, O. Skaugrud, L. Illum, Chitosan as
a nasal delivery system: the effect of chitosan solutions on in vitro and in vivo
mucociliary transport rates in human turbinates and volunteers, J. Pharm. Sci.
86 (1997) 509–513.
 S. Danielsen, K.M. Varum, B.T. Stokke, Structural analysis of chitosan medi-
ated DNA condensation by AFM: inﬂuence of chitosan molecular parameters,
Biomacromolecules 5 (2004) 928–936.
 G. Maurstad, S. Danielsen, B.T. Stokke, The inﬂuence of charge density of
chitosan in the compaction of the polyanions DNA and xanthan, Biomacro-
molecules 8 (2007) 1124–1130.
 S.P. Strand, S. Lelu, N.K. Reitan, C. de Lange Davies, P. Artursson, K.M. Varum,
Molecular design of chitosan gene delivery systems with an optimized balance
between polyplex stability and polyplex unpacking, Biomaterials 31 (2009)
 M. Koping-Hoggard, K.M. Varum, M. Issa, S. Danielsen, B.E. Christensen, B.T.
Stokke, P. Artursson, Improved chitosan-mediated gene delivery based on eas-
ily dissociated chitosan polyplexes of highly deﬁned chitosan oligomers, Gene
Ther. 11 (2004) 1441–1452.
 M. Huang, C.W. Fong, E. Khor, L.Y. Lim, Transfection efﬁciency of chitosan vec-
tors: effect of polymer molecular weight and degree of deacetylation, J. Control.
Release 106 (2005) 391–406.
 M. Lavertu, S. Methot, N. Tran-Khanh, M.D. Buschmann, High efﬁciency
gene transfer using chitosan/DNA nanoparticles with speciﬁc combinations
of molecular weight and degree of deacetylation, Biomaterials 27 (2006)
 F.C. MacLaughlin, R.J. Mumper, J.J. Wang, J.M. Tagliaferri, I. Gill, M. Hinchcliffe,
A.P. Rolland, Chitosan and depolymerized chitosan oligomers as condensing
carriers for in vivo plasmid delivery, J. Control. Release 56 (1998) 259–272.
 T. Kiang, H. Wen, H.W. Lim, K.W. Leong, The effect of the degree of chitosan
deacetylation on the efﬁciency of gene transfection, Biomaterials 25 (2004)
 M. Koping-Hoggard, Y.S. Mel’nikova, K.M. Varum, B. Lindman, P. Artursson,
Relationship between the physical shape and the efﬁciency of oligomeric chi-
tosan as a gene delivery system in vitro and in vivo, J. Gene Med. 5 (2003)
 M. Hashimoto, M. Morimoto, H. Saimoto, Y. Shigemasa, T. Sato, Lactosylated
chitosan for DNA delivery into hepatocytes: the effect of lactosylation on
the physicochemical properties and intracellular trafﬁcking of pDNA/chitosan
complexes, Bioconjug. Chem. 17 (2006) 309–316.
 M. Hashimoto, M. Morimoto, H. Saimoto, Y. Shigemasa, H. Yanagie, M. Eriguchi,
T. Sato, Gene transfer by DNA/mannosylated chitosan complexes into mouse
peritoneal macrophages, Biotechnol. Lett. 28 (2006) 815–821.
 M.M. Issa, M. Koping-Hoggard, K. Tommeraas, K.M. Varum, B.E. Christensen,
S.P. Strand, P. Artursson, Targeted gene delivery with trisaccharide-substituted
62 M. Alatorre-Meda et al. / Colloids and Surfaces B: Biointerfaces 82 (2011) 54–62
chitosan oligomers in vitro and after lung administration in vivo, J. Control.
Release 115 (2006) 103–112.
 S.P. Strand, M.M. Issa, B.E. Christensen, K.M. Varum, P. Artursson, Tailor-
ing of chitosans for gene delivery: novel self-branched glycosylated chitosan
oligomers with improved functional properties, Biomacromolecules 9 (2008)
 O. Germershaus, S.R. Mao, J. Sitterberg, U. Bakowsky, T. Kissel, Gene delivery
using chitosan, trimethyl chitosan or polyethylenglycol-graft-trimethyl chi-
tosan block copolymers: establishment of structure–activity relationships in
vitro, J. Control. Release 125 (2008) 145–154.
 X. Jiang, H. Dai, K.W. Leong, S.H. Goh, H.Q. Mao, Y.Y. Yang, Chitosan-g-PEG/DNA
complexes deliver gene to the rat liver via intrabiliary and intraportal infusions,
J. Gene Med. 8 (2006) 477–487.
 Y.Q. Zhang, J.J. Chen, Y.D. Zhang, Y.F. Pan, J.F. Zhao, L.F. Ren, M.M. Liao, Z.Y.
Hu, L. Kong, J.W. Wang, A novel PEGylation of chitosan nanoparticles for gene
delivery, Biotechnol. Appl. Biochem. 46 (2007) 197–204.
 T. Kean, S. Roth, M. Thanou, Trimethylated chitosans as non-viral gene delivery
vectors: cytotoxicity and transfection efﬁciency, J. Control. Release 103 (2005)
 S.R. Mao, X.T. Shuai, F. Unger, M. Wittmar, X.L. Xie, T. Kissel, Synthesis, char-
acterization and cytotoxicity of poly(ethylene glycol)-graft-trimethyl chitosan
block copolymers, Biomaterials 26 (2005) 6343–6356.
 H.Q. Mao, K. Roy, V.L. Troung-Le, K.A. Janes, K.Y. Lin, Y. Wang, J.T. August, K.W.
Leong, Chitosan–DNA nanoparticles as gene carriers: synthesis, characteriza-
tion and transfection efﬁciency, J. Control. Release 70 (2001) 399–421.
 K. Roy, H.Q. Mao, S.K. Huang, K.W. Leong, Oral gene delivery with chitosan–DNA
nanoparticles generates immunologic protection in a murine model of peanut
allergy, Nat. Med. 5 (1999) 387–391.
 Z.R. Cui, R.J. Mumper, Chitosan-based nanoparticles for topical genetic immu-
nization, J. Control. Release 75 (2001) 409–419.
 M. Alatorre-Meda, P. Taboada, J. Sabin, B. Krajewska, L.M. Varela, J.R. Rodriguez,
DNA–chitosan complexation: a dynamic light scattering study, Colloids Surf. A:
Physicochem. Eng. Aspects 339 (2009) 145–152.
 W. Saenger, Principles of Nucleic Acid Structure, Springer-Verlag, New York,
 P.L. Felgner, Y. Barenholz, J.P. Behr, S.H. Cheng, P. Cullis, L. Huang, J.A. Jessee, L.
Seymour, F. Szoka, A.R. Thierry, E. Wagner, G. Wu, Nomenclature for synthetic
gene delivery systems, Hum. Gene Ther. 8 (1997) 511–512.
 P. Erbacher, S.M. Zou, T. Bettinger, A.M. Steffan, J.S. Remy, Chitosan-based
vector/DNA complexes for gene delivery: biophysical characteristics and trans-
fection ability, Pharm. Res. 15 (1998) 1332–1339.
 J.P. Camp, A.T. Capitano, Size-dependent mobile surface charge model of cell
electrophoresis, Biophys. Chem. 113 (2005) 115–122.
 L.E. Prevette, T.E. Kodger, T.M. Reineke, M.L. Lynch, Deciphering the role of
hydrogen bonding in enhancing pDNA–polycation interactions, Langmuir 23
 P. Taboada, S. Barbosa, E. Castro, V. Mosquera, Amyloid ﬁbril formation and
other aggregate species formed by human serum albumin association, J. Phys.
Chem. B 110 (2006) 20733–20736.
 I. Horcas, R. Fernandez, J.M. Gomez-Rodriguez, J. Colchero, J. Gomez-Herrero,
A.M. Baro, WSXM: a software for scanning probe microscopy and a tool for
nanotechnology, Rev. Sci. Instrum. 78 (2007).
 G.S. Manning, Molecular theory of polyelectrolyte solutions with applica-
tions to electrostatic properties of polynucleotides, Q. Rev. Biophys. 11 (1978)
 D. Matulis, I. Rouzina, V.A. Bloomﬁeld, Thermodynamics of DNA binding and
condensation: isothermal titration calorimetry and electrostatic mechanism, J.
Mol. Biol. 296 (2000) 1053–1063.
 R.S. Dias, L.M. Magno, A.J.M. Valente, D. Das, P.K. Das, S. Maiti, M.G. Miguel, B.
Lindman, Interaction between DNA and cationic surfactants: effect of DNA con-
formation and surfactant headgroup, J. Phys. Chem. B 112 (2008) 14446–14452.
 A. Rodriguez-Pulido, E. Aicart, O. Llorca, E. Junquera, Compaction process of
calf thymus DNA by mixed cationic-zwitterionic liposomes: a physicochemical
study, J. Phys. Chem. B 112 (2008) 2187–2197.
 M. Alatorre-Meda, P. Taboada, B. Krajewska, M. Willemeit, A. Deml, R. Klösel,
J.R. Rodríguez, DNA–poly(diallyldimethylammonium chloride) complexation
and transfection efﬁciency, J. Phys. Chem. B 114 (2010) 9356–9366.
 P.L. Felgner, Y.J. Tsai, L. Sukhu, C.J. Wheeler, M. Manthorpe, J. Marshall, S.H.
Cheng, Improved cationic lipid formulations for in vivo gene therapy, DNA
Vaccines 772 (1995) 126–139.
 R.S. Dias, B. Lindman, M.G. Miguel, DNA interaction with catanionic vesicles, J.
Phys. Chem. B 106 (2002) 12600–12607.
 D.D. Lasic, N.S. Templeton, Liposomes in gene therapy, Adv. Drug Deliv. Rev. 20
 J.O. Radler, I. Koltover, T. Salditt, C.R. Saﬁnya, Structure of DNA–cationic lipo-
some complexes: DNA intercalation in multilamellar membranes in distinct
interhelical packing regimes, Science 275 (1997) 810–814.
 M.X. Tang, F.C. Szoka, The inﬂuence of polymer structure on the interactions of
cationic polymers with DNA and morphology of the resulting complexes, Gene
Ther. 4 (1997) 823–832.
 R.S. Dias, J. Innerlohinger, O. Glatter, M.G. Miguel, B. Lindman, Coil-globule
transition of DNA molecules induced by cationic surfactants: a dynamic light
scattering study, J. Phys. Chem. B 109 (2005) 10458–10463.
 K. Minagawa, Y. Matsuzawa, K. Yoshikawa, A.R. Khokhlov, M. Doi, Direct obser-
vation of the coil-globule transition in DNA-molecules, Biopolymers 34 (1994)
 Y. Yamasaki, K. Yoshikawa, Higher order structure of DNA controlled by the
redox state of Fe
, J. Am. Chem. Soc. 119 (1997) 10573–10578.
 K. Yoshikawa, M. Takahashi, V.V. Vasilevskaya, A.R. Khokhlov, Large discrete
transition in a single DNA molecule appears continuous in the ensemble, Phys.
Rev. Lett. 76 (1996) 3029–3031.
 C.W. Pouton, P. Lucas, B.J. Thomas, A.N. Uduehi, D.A. Milroy, S.H. Moss,
Polycation–DNA complexes for gene delivery: a comparison of the biophar-
maceutical properties of cationic polypeptides and cationic lipids, J. Control.
Release 53 (1998) 289–299.
 T. Wiseman, S. Williston, J.F. Brandts, L.N. Lin, Rapid measurement of binding
constants and heats of binding using a new titration calorimeter, Anal. Biochem.
179 (1989) 131–137.
 C.G. de Kruif, F. Weinbreck, R. de Vries, Complex coacervation of proteins and
anionic polysaccharides, Curr. Opin. Coll. Int. Sci. 9 (2004) 340–349.
 S. Bharadwaj, R. Montazeri, D.T. Haynie, Direct determination of the ther-
modynamics of polyelectrolyte complexation and implications thereof for
electrostatic layer-by-layer assembly of multilayer ﬁlms, Langmuir 22 (2006)
 C.K. Nisha, S.V. Manorama, M. Ganguli, S. Maiti, J.N. Kizhakkedathu, Com-
plexes of poly(ethylene glycol)-based cationic random copolymer and calf
thymus DNA: a complete biophysical characterization, Langmuir 20 (2004)
 Y.L. Zhou, Y.Z. Li, The interaction of poly(ethylenimine) with nucleic acids and
its use in determination of nucleic acids based on light scattering, Spectrochim.
Acta A: Mol. Biomol. Spectrosc. 60 (2004) 377–384.
 S. Milev, H.R. Bosshard, I. Jelesarov, Enthalpic and entropic effects of salt
and polyol osmolytes on site-speciﬁc protein–DNA association: the integrase
Tn916–DNA complex, Biochemistry 44 (2005) 285–293.
 L.E. Engler, K.K. Welch, L. JenJacobson, Speciﬁc binding by EcoRV endonuclease
to its DNA recognition site GATATC, J. Mol. Biol. 269 (1997) 82–101.
 S. Srinivasachari, Y.M. Liu, L.E. Prevette, T.M. Reineke, Effects of trehalose click
polymer length on pDNA complex stability and delivery efﬁcacy, Biomaterials
28 (2007) 2885–2898.
 L.E. Prevette, M.L. Lynch, T.M. Reineke, Amide spacing inﬂuences pDNA binding
of poly(amidoamine)s, Biomacromolecules 11 (2010) 326–332.
 H. Arakawa, K. Umemura, A. Ikai, Protein images obtained by STM, AFM and
TEM, Nature 358 (1992) 171.
 A.C. Lin, M.C. Goh, Investigating the ultrastructure of ﬁbrous long spacing col-
lagen by parallel atomic force and transmission electron microscopy, Proteins:
Struct. Funct. Genet. 49 (2002) 378–384.
 M. Alatorre-Meda, A. Gonzalez-Perez, J.R. Rodriguez, DNA-METAFECTENE
PRO complexation: a physical chemistry study, Phys. Chem. Chem. Phys. 12
 E. Wagner, M. Cotten, R. Foisner, M.L. Birnstiel, Transferrin polycation dna
complexes—the effect of polycations on the structure of the complex and dna
delivery to cells, Proc. Natl. Acad. Sci. U.S.A. 88 (1991) 4255–4259.
 A.M. Carnerup, M.L. Ainalem, V. Alfredsson, T. Nylander, Watching DNA con-
densation induced by poly(amido amine) dendrimers with time-resolved
cryo-TEM, Langmuir 25 (2009) 12466–12470.
 A.V. Kabanov, V.A. Kabanov, Interpolyelectrolyte and block ionomer complexes
for gene delivery: physicochemical aspects, Adv. Drug Deliv. Rev. 30 (1998)
 P. Midoux, C. Pichon, J.J. Yaouanc, P.A. Jaffres, Chemical vectors for gene deliv-
ery: a current review on polymers, peptides and lipids containing histidine or
imidazole as nucleic acids carriers, Br. J. Pharmacol. 157 (2009) 166–178.
 D. Fischer, H. Dautzenberg, K. Kunath, T. Kissel, Poly(diallyldimethylammonium
chlorides) and their N-methyl-N-vinylacetamide copolymer-based DNA-
polyplexes: role of molecular weight and charge density in complex formation,
stability, and in vitro activity, Int. J. Pharm. 280 (2004) 253–269.
 V.A. Izumrudov, M.V. Zhiryakova, S.E. Kudaibergenov, Controllable stability of
DNA-containing polyelectrolyte complexes in water-salt solutions, Biopoly-
mers 52 (1999) 94–108.