Synthesis, characterization, and in vitro transfection activity of charge-reversal amphiphiles for DNA delivery

Article abstract of DOI:10.1021/bc1004526

A series of charge-reversal lipids were synthesized that possess varying chain lengths and end functionalities. These lipids were designed to bind and then release DNA based on a change in electrostatic interaction with DNA. Specifically, a cleavable ester linkage is located at the ends of the hydrocarbon chains. The DNA release from the amphiphile was tuned by altering the length and position of the ester linkage in the hydrophobic chains of the lipids through the preparation of five new amphiphiles. The amphiphiles and corresponding lipoplexes were characterized by DSC, TEM, and X-ray, as well as evaluated for DNA binding and DNA transfection. For one specific charge-reversal lipid, stable lipoplexes of approximately 550 nm were formed, and with this amphiphile, effective in vitro DNA transfection activities was observed.

Full text of DOI:10.1021/bc1004526

ARTICLE
pubs.acs.org/bc
Synthesis, Characterization, and In Vitro Transfection Activity of
Charge-Reversal Amphiphiles for DNA Delivery
Xiao-Xiang Zhang,z,‡ Carla A. H. Prata,z,‡ Jason A. Berlin, Thomas J. McIntosh,§ Philippe Barthelemy,† and
Mark W. Grinstaff*
Departments of Biomedical Engineering and Chemistry, Boston University, Boston, Massachusetts 02215, United States
†Inserm, U869, Bordeaux, F-33076 France
‡Universitꢀe de Bordeaux, F-33076, Bordeaux, France
§Department of Cell Biology, Duke University Medical Center, Durham, North Carolina 27710, United States
S Supporting Information
b
ABSTRACT: A series of charge-reversal lipids were synthesized that possess
varying chain lengths and end functionalities. These lipids were designed to
bind and then release DNA based on a change in electrostatic interaction with
DNA. Specifically, a cleavable ester linkage is located at the ends of the
hydrocarbon chains. The DNA release from the amphiphile was tuned by
altering the length and position of the ester linkage in the hydrophobic chains
of the lipids through the preparation of five new amphiphiles. The amphiphiles
and corresponding lipoplexes were characterized by DSC, TEM, and X-ray, as
well as evaluated for DNA binding and DNA transfection. For one specific
charge-reversal lipid, stable lipoplexes of approximately 550 nm were formed,
and with this amphiphile, effective in vitro DNA transfection activities was
observed.
’ INTRODUCTION
as PEG) has been placed between the cationic headgroup and
the hydrophobic chains23 to improve transfection efficiency.
Researchers have explored the use of glycosidic cationic head
groups in an attempt to lower toxicity through tighter interac-
tions with DNA24À26 or peptide head groups to mimic peptideÀ
DNA binding interactions.27À30 Recent reports from our labo-
ratory have also described the use of cationic lipids possessing
nucleic acid recognition components for nucleic acid delivery in
an effort to use more than one type of molecular interaction for
complexation.31À39
In addition to using these strategies to overcome the low
transfection efficiencies of synthetic vectors (estimated to be
0.1% to 1%), functional or stimuli-responsive vectors have also
been developed and evaluated. These cationic amphiphiles are
able to respond to a stimulus that changes their chemical or
physical properties, facilitating DNA delivery. Synthetic vectors
have been prepared that are responsive to pH,40À49 reducing
conditions,50À59 enzymes,43,60À62 or temperature.63,64 Many of
these functional vectors that degrade or change conformation
were designed based on the specific steps in the transfection
pathway. This pathway starts with cellular entry of the
DNAÀamphiphile complex by endocytosis, followed by endo-
somal escape, and finally nuclear targeting and DNA expression.
Gene therapy offers the potential to treat a variety of diseases
by delivering a functional copy of a missing or defective gene.
Current research in this field focuses on two major classes of
DNA carriers: viral1À5 and nonviral.6À16 Viral vectors transfect a
significantly greater percentage of genetic material than nonviral
systems, but suffer from a number of drawbacks, including toxicity
and induced immunological response.17 As a consequence, inter-
est in synthesizing and evaluating nonviral delivery systems,
including amphiphilic small molecule and polymeric compounds,
has increased over the last 10 years. Cationic lipids are one type of
amphiphile that can intracellularly transport DNA. Several lipo-
some formulations are commercially available with a few in
clinical trials.16,18 Three examples of such lipids (DOTAP,
DOTMA, and DMRIE) are shown in Figure 1.
Following the initial success of these lipids for DNA transfec-
tion, a number of different lipid forming complexes of varying
structure and composition have been investigated. Within this
context, a few representative examples are described below for
DNA delivery. The acyl chain length and linker bonds connect-
ing the cationic head to the acyl tail(s) have been varied to
enhance transfection efficiency and decrease cytotoxic effects
(e.g., DOTMA).19 Lipospermines, with polycationic headgroups,
have been investigated to increase the affinity for DNA.20
Fluorinated analogues of cationic lipids such as DOTMA have
been explored, and these amphiphilies exhibited higher transfec-
tion activities in vitro21 and in vivo.22 A hydrophilic spacer (such
Received: October 16, 2010
Revised:
March 2, 2011
Published: April 01, 2011
r
2011 American Chemical Society
690
dx.doi.org/10.1021/bc1004526 Bioconjugate Chem. 2011, 22, 690–699
|

Bioconjugate Chemistry
ARTICLE
Figure 2. Charge-reversal or charge- switchable effect of the
amphiphiles.
Figure 1. Chemical structures of several common cationic lipids.
’ RESULTS AND DISCUSSION
Our research effort is centered on the release of DNA from
the DNAÀamphiphile supramolecular complex following cel-
lular uptake, and as such, we have also focused on functional
synthetic vectors. Given that a strong electrostatic assembly is
formed between the cationic amphiphile and anionic DNA,
and that this assembly is unlikely to disassemble sponta-
neously, we are interested in altering the electrostatic interac-
tion to free the DNA.54,65,66 To this end, we have synthesized
and characterized new cationic amphiphiles that are capable of
undergoing an intracellular electrostatic transition from catio-
nic to anionic under the influence of esterases such as those
found in cell lysosomes/endosomes.67,68 This enables these
amphiphiles to release DNA from the supramolecular assembly
(Figure 2).
These charge-reversal or charge-switchable amphiphiles
possess a positively charged ammonium headgroup that func-
tions to complex DNA and lipophilic acyl chains to promote
bilayer formation, similar to other cationic lipids like DOTAP
(Figure 1). However, these new amphiphiles also contain ester
functionalities near the terminal end of each acyl chain that,
upon hydrolysis, switch the overall charge of the amphiphile
from positive to negative. As an extension of our previous
communication54 in which we introduced the charge-reversal
amphiphile, in this manuscript we have synthesized and in-
vestigated nine amphiphiles possessing C10, C12, or C16 acyl
chain lengths terminally protected by benzyl, ethyl, butyl, or
acetyl groups (Figure 3). The optimal chain length and terminal
groups were determined based on amphiphileÀDNA binding
and DNA release. Standard ethidium bromide assays estab-
lished the DNA binding capability of each of the vectors over a
range of pH. The release of DNA from the DNAÀamphiphile
assembly was monitored also using an ethidium bromide assay.
To determine the thermal and structural properties of the lipid
and DNAÀlipid complexes, differential scanning calorimetery
(DSC), dynamic light scattering (DLS), transmission electronic
microscopy (TEM), and X-ray diffraction were performed.
Finally, the transfection of the amphiphile/DNA complexes
was evaluated in vitro in several cell lines.
To bind DNA, form lipoplexes, and then eventually release
DNA, each amphiphile requires three distinct structural compo-
nents: a cationic headgroup, hydrophobic chains, and terminal
ester linkages. As shown in Figure 3, amphiphile 1 has a cationic
headgroup to bind DNA, lipophilic acyl chains to form a bilayer,
and ester linkages at the end of the acyl chains for enzymatic
hydrolysis. We prepared the series of compounds 2À10, shown in
Figure 3, to systematically assess the role of each structural
component and to evaluate the effects of amphiphile structure
and reactivity (chain length, terminal end group, and ester or
amide terminal group) on DNA binding/release, supramolecular
assembly, and gene delivery efficacy. Specifically, amphiphile 1
possesses a cationic ammonium headgroup and two dodecanoic
acid chains protected with benzyl esters. Hydrolysis of the benzyl
esters of 1 affords compound 2, which is negatively charged at
neutral pH. Amphiphile 3 is an analogue of 1 that possesses
noncleavable amide linkages, whereas compound 4 lacks the long
acyl chains completely. Amphiphiles 6 and 7 are short and long
acyl chain analogues of 1, respectively, with the same terminal
benzyl ester group. Amphiphiles 8 and 9 possess terminal alkyl
ester linkages of two different chain lengths (ethyl and butyl ester,
respectively). Finally, amphiphile 10 contains a similar terminal
ester but upon hydrolysis affords two hydroxyl terminated acyl
chains and thus remains cationic atneutral pH inaqueous solution.
The synthetic route to the amphiphiles is shown in Scheme 1.
The preparation requires the use of functionalized fatty acids. The
monoprotected benzyl ester fatty acids were prepared from the
corresponding diacids (decanoic, dodecanoic, or hexanoic) with
benzyl formate in the presence of Dowex 50W-X2 in octane at
80 °C. The butyl ester fatty acid was prepared in a similar manner
using butyl formate. The ethyl ester fatty acids were obtained by
the addition of ethanol to a solution of dodecanedioyl dichloride
in tetrahydrofuran in the presence of triethylamine at 0 °C. The
benzyl amide analogue was prepared in a similar manner using
benzylamine. The acetylated fatty acid was prepared by acetyla-
tion of the hydroxydodecanoic acid using acetic anhydride in
pyridine. Next, the monofunctionalized fatty acid derivatives were
coupled to 3-dimethylaminopropanediol in the presence of DCC
and DMAP in dichloromethane. The reaction yields ranged from
80% to 95% for all steps. Finally, the amphiphiles were reacted
with methyl iodide in dichloromethane to quaternize the tertiary
amine in quantitative yield. Complete details of the synthesis for
each compound are found in the Supporting Information.
’ EXPERIMENTAL PROCEDURES
Please see the Supporting Information for complete experi-
mental details.
691
dx.doi.org/10.1021/bc1004526 |Bioconjugate Chem. 2011, 22, 690–699

Bioconjugate Chemistry
ARTICLE
Figure 3. Structures of the charge-reversal amphiphiles under investigation for gene delivery.
Scheme 1. Synthesis of the Charge-Reversal Amphiphiles
As the first step toward evaluating these amphiphiles for the
delivery of nucleic acids, we used an ethidium bromide displace-
ment assay to determine whether the amphiphiles bind DNA.69
This fluorescence quenching assay is an indirect method to
measure DNA binding by monitoring the displacement of the
ethidium bromide from DNA by the cationic amphiphile. This
assay is widely used to assess small molecule/DNA binding, and
it is assumed that the small molecule binder is responsible for
displacing ethidium bromide and causing fluorescence decrease.
As shown in Figure 4, the fluorescence intensity decreases for
amphiphiles 1, 3, 6À10, and DOTAP upon addition to the
ethidium bromideÀDNA solution, but does not decrease for
amphiphiles 2 and 4. Amphiphile 2 possesses an overall negative
charge and, consequently, does not bind DNA due to unfavor-
able electrostatic interactions. Compound 4 is a cationic amphi-
phile but does not bind DNA, indicating the importance of the
hydrophobic acyl chains for the formation of a strong interaction
with DNA.
significant effect on DNA binding, as all of the amphiphiles bind
DNA. The binding constants can be estimated and compared by
measuring the loss of EtBr fluorescence as a function of the vector
added. The concentration which produces 50% of the inhibition
of fluorescence (IC50) approximates the binding constant. The
apparent binding constants were calculated as follows: Kapp
=
(1.26 Â 10À6/ IC50) Â KEtBr with KEtBr = 107 MÀ1 as described
elsewhere.70 The binding constant for all the amphiphiles is on
the order of 106 MÀ1 and similar in magnitude to DOTAP (K =
106 MÀ1). These binding curves indicate that a ∼1:1 charge
complex is formed between DNA and most of these amphiphiles,
except for 7, which forms a ∼ 2:1 charge complex with DNA.
We next determined the effect of pH on DNA binding for
amphiphiles 1 and 2. We chose two pH conditions near phys-
iological pH (pH 7 and 8) and a lower pH of 5.5 that mimics the
conditions of the early endosome. Figure 5 shows the fluores-
cence intensity as a function of amphiphile/DNA charge ratio for
amphiphiles 1, 2, and 5 (DOTAP) at the three different pH
conditions. The fluorescence intensity decreases as a function of
charge ratio for both amphiphile 1 and DOTAP across all three
The different terminal ester groups (ethyl, butyl, benzyl,
acetyl) of the cationic amphiphiles do not appear to have a
692
dx.doi.org/10.1021/bc1004526 |Bioconjugate Chem. 2011, 22, 690–699

Bioconjugate Chemistry
ARTICLE
pHs. The binding curves look similar, with a 1:1 charge complex
being formed between the amphiphile and DNA. The fluores-
cence intensity is constant and does not decrease for amphiphile
2, indicating that 2 does not displace EtBr from DNA. This result
is consistent with 2 being anionic over this pH range and as that
the pKa for the terminal carboxylic acids is below 5.5. Overall,
these pH conditions do not play a significant role in DNA binding
of the amphiphiles.
To evaluate the release of the DNA from the amphiphi-
leÀDNA complex, the amphiphile/EtBr/DNA solution was
incubated with a porcine liver esterase at pH 7.4 (1000 units/
mL; 37 °C; 100 mM NaCl, 100 mM Tris buffer). For these
release studies, the fluorescence intensities as a function of time
for the various amphiphiles (including DOTAP) are shown in
Figures 6 and 7. The increase in fluorescence from EtBr results
from reintercalation of EtBr in the DNA; once the DNA is “free”
from the amphiphile complex, ethidium bromide can bind with
the DNA. As shown in Figure 6, amphiphile 1 no longer binds
strongly to the DNA as a consequence of hydrolysis of its ter-
minal benzyl esters. This result is consistent with the binding
studies conducted with 2, the hydrolyzed product of 1 prepared
independently. However, in the presence of the enzyme, no
increase in fluorescence over time is observed with amphiphile 3
or 5 (DOTAP), which possesses a terminal amide linkage or
methyl group, respectively. A slightly slower increase in fluores-
cence is observed with compound 10, which upon hydrolysis
affords a neutral hydroxyl group.
To confirm that enzymatic hydrolysis cleaved the benzyl ester
bond, liposomes formed from amphiphile 1 were incubated in
PBS at 37 °C in the presence of porcine liver esterase, and the
reaction products were examined by LC-MS. The results show
that, within 30 min, ∼58% of 1 had converted to the hydrolyzed
product 2, and 1 was completely hydrolyzed by 2.5 h.
When the terminal ester is changed from a benzyl to an ethyl
group as in compound 8, similar EtBrÀDNA binding curves to
those of compound 1 are obtained (Figure 7). The increases in
fluorescence indicate that both benzyl and ethyl terminal esters
are readily hydrolyzed. Likewise, similar results are obtained
when the acyl chain length is shortened from C12 to C10 (1 vs 6)
while maintaining the terminal benzyl esters. However, upon
extending the acyl chain length to C16 from C12 (1 vs 7), no
increase in fluorescence is observed. Our interpretation of this
Figure 4. (top) Fluorescence intensity as a function of amphiphile/
DNA charge ratio. (bottom) The fluorescence intensity as a function of
amphiphile/DNA charge ratio for amphiphiles 1, 6À10, and DOTAP.
N = 3 Avg ( SD.
Figure 5. Fluorescence intensity as a function of amphiphile/DNA charge ratio in pH solutions of 5.5, 7.0 and 8.0. N = 3 Avg ( SD.
693
dx.doi.org/10.1021/bc1004526 |Bioconjugate Chem. 2011, 22, 690–699

Bioconjugate Chemistry
ARTICLE
Table 1. DSC Results for the Amphiphile Assemblies
compound
Tm (°C)
ΔH (kcal/mol)
1
52.55
46.32
75.45
78.44
76.14
50.77
2.03
3.18
1.57
8.47
3.78
0.102
6
7
8
9
10
Figure 6. Ethidium bromide intercalation assay showing the fluores-
cence intensity as a function of time in the presence of a porcine liver
esterase (1000 units/mL). N = 3 Avg ( SD.
Figure 8. Transfection efficiency of the different amphiphiles in CHO
cells. N = 3 Avg ( SD.
Figure 7. Ethidium bromide intercalation assay showing the fluores-
cence intensity as a function of time in the presence of a porcine liver
esterase. N = 3 Avg ( SD.
The complexes were added to CHO cells after 15 min. The
amount of DNA used was the same as was used in the naked
DNA control (no amphiphile) and positive control experi-
ments (DOTAP and Transfast). The negative control was
amphiphile 1 without DNA. After incubation at 37 °C and 5%
CO2 for 2 h, the solution containing the lipoplexes was
removed and fresh growth medium with serum was added.
Transfection efficiencies were assessed after 48 h using the
β-galactosidase enzyme assay in conjunction with a standard
curve. The efficacy of each transfection was calculated as β-gal
activity normalized to total protein. We first performed a
screen to identify the optimal amphiphile 1/DNA ratio for
greatest transfection examining ratios of 1:1, 2:1, 5:1, 10:1,
15:1, and 45:1. We also varied the time of incubation of the
lipoplex with the cells from 30 min to 3 h, as well as performed
analyses at the 24 and 48 h time points. On the basis of these
experiments, the highest transfection efficiency was at a 5:1
ratio, an exposure time of 2 h, and analysis after 48 h. With
these optimal conditions, we then performed a series of
transfection experiments with the charge-reversal amphiphiles,
as summarized in Figure 8. Amphiphile 1 was the most effective
vector for transfection. The other compounds (2À4 and
6À10) showed minimal transfection activity, comparable to
the negative control and naked DNA. The transfection efficacy
of amphiphile 1 was significantly better than that of DOTAP
(p < 0.05) and comparable to that of Transfast (p > 0.05). In
the literature, it has also been shown that the addition of
neutral lipids (i.e., helper lipids) such as dioleylphosphatidy-
lethanolamine (DOPE) to a cationic lipid formulation can
result is that the long chain analogue forms a more stable bilayer
structure, reducing accessibility of the ester bonds to the enzyme,
and therefore inhibiting hydrolysis. The phase-transition tem-
perature (Tm) for 7 is 23 °C higher than for 1, as shown in
Table 1, which is consistent with this interpretation. Upon
changing the benzyl to an ethyl ester but maintaining the chain
length of C12, we observe a similar rate of hydrolysis (6 vs 8).
However, 9, which possesses an n-butyl ester as opposed to an
ethyl ester (9 vs 1), shows a slower rate of fluorescence increase
during the assay. In this amphiphile, the hydrolyzable ester is
buried further in the hydrophobic chain, leading to a slower rate
of hydrolysis. The enzymatic mechanism is still unknown,
although we suspect that, during the flip-flop of the lipids, the
enzyme hydrolyzes the terminal ester bonds.
Overall, the fluorescence data indicate that release of DNA
from the supramolecular assembly does not occur for amphi-
philes lacking a terminal hydrolyzable ester linkage, and those
linkages near the headgroup or buried deeper in the chain are less
accessible to enzymatic hydrolysis. The data also show the
difference in DNA release from an amphiphilic complex that
generates a resulting anionic amphilphile product versus a
neutral one (1 vs 10).
Next, transfection experiments using the reporter gene,
β-galactosidase (β-gal, pVax-LacZ1, Invitrogen), were performed
with Chinese hamster ovarian (CHO) cells. The reporter gene was
first mixed with the cationic lipid, at a defined amphiphile/DNA
ratio, in potassium phosphate buffer (PBS) at room temperature.
694
dx.doi.org/10.1021/bc1004526 |Bioconjugate Chem. 2011, 22, 690–699

Bioconjugate Chemistry
ARTICLE
Figure 10. Cytotoxicity of the different lipids in CHO cells. N = 3
zAvg ( SD.
Figure 9. Photograph of β-galactosidase transfected cells using amphi-
phile 1 at low and high magnification. Dark blue stain represents B-gal
expression.
Figure 11. Transfection efficiency of amphiphile 1 with HEK293 and
K562 cell lines. N = 3 Avg ( SD. First data set with 1 is at 15:1 and the
second data set is at 45:1 charge ratio.
improve DNA transfection efficiency.19 Consequently, the
above transfection experiments were repeated using the helper
lipid, DOPE, and the charge-reversal amphiphiles at a 1:1 ratio.
No increase in transfection efficiency was observed for these
conditions (data not shown).
Micrographs of the β-galactosidase transfected CHO cells
using amphiphile 1 are shown in Figure 9. This transfection
experiment was conducted in a similar manner except it was done
on a glass slide that had been previously modified with collagen
to anchor the cells. The cells are expressing the protein β-gal
(dark blue stain), and there appear to be few patches containing
no cells, indicating minimal toxicity under these experimental
conditions. This lack of cytotoxicity was quantitatively deter-
mined in our next set of experiments.
Cytotoxicity experiments were performed with CHO cells
using both a formazan-based proliferation assay and a total
protein assay. The cells were seeded onto a 96-multiwell micro-
liter plate with an approximate density of 1 Â 104 cells per well.
The compounds were added to the cells 24 h later. After 24 h, cell
proliferation was determined and expressed as a percentage of
nontreated cells as shown in Figure 10. None of the amphiphiles
showed a significant cytotoxicity, with values similar to the
negative control (i.e., nontreated cells).
With these encouraging results, we then evaluated the trans-
fection activity of amphiphile 1 in four other cell lines. Transfec-
tion experiments using human embryonic kidney (HEK293),
erythroleukemic (K562), mouse fibroblast (NIH 3T3), and
human liver carcinoma (HepG2) cell lines were performed with
compound 1 (see Figure 11). Once again, we varied amphiphile
1/DNA ratio, and only at high amphiphile/DNA ratio did we
observe transfection (HEK 15:1; K562 45:1). Amphiphile 1
transfected DNA with similar efficiency to Transfast in the
HEK293 and K562, but we did not observe significant transfec-
tion in the NIH 3T3 or HepG2 lines (data not shown).
With the above transfection activities recorded, we embarked
on a series of physiochemical experiments to elucidate the key
structural and physical properties of amphiphile 1 likely respon-
sible for transfection activity. Given the polar headgroup and
long hydrophobic acyl chains present in amphiphile 1, it is likely
to form bilayer structures in aqueous solution. To test this idea,
differential scanning calorimeter (DSC), dynamic light scattering
(DLS), transmission electron microscopy (TEM), and X-ray
diffraction (XRD) experiments were performed. The samples
were prepared for these studies by introducing a chloroformic
695
dx.doi.org/10.1021/bc1004526 |Bioconjugate Chem. 2011, 22, 690–699

Bioconjugate Chemistry
ARTICLE
solution of amphiphiles into a pear-shaped flask. The solution
was evaporated under vacuum, leaving a thin film deposited onto
the flask wall. A total of 1 mL Tris buffer (100 mM Tris, 100 mM
NaCl, pH 7.4) was then added, and the film was peeled off by
vortexing. After 20 extrusions though a 50 nm polycarbonate
membrane (Avanti Polar Lipids), the milky aqueous suspension
turned into a homogeneous solution. The lipoplexes were
prepared in an analogous manner by addition of the DNA to
the previously prepared liposome solution.
DSC traces of the aqueous hydrated amphiphiles (1, 6À10)
show for each lipid a single phase-transition temperature (Tm).
The Tm and transition enthalpy (ΔH) vary as the chain lengthens
(C10, C12, and C16; 6 vs 1 vs 7, respectively) when the end
groups remain the same. For example, the Tm increases from
46 °C for 6 to 75 °C for 7, indicating that amphiphile 7 forms a
more stable structure than 1 or 6. This trend with increasing
chain length and higher Tm is true for Tm, but not for enthalpy.
The enthalpy is smaller for 7 than for 6 or 1. These results are
consistent with reports on other lipid families (e.g., phosphati-
dylcholine and Avanti polar technical data). When the chain
lengths are kept constant but the end group is changed from a
benzyl to an alkyl ester on the amphiphile (e.g., 1 vs 8), the Tm
and enthalpy increase by about 25 °C. On the other hand,
reversing the directionality of the ester linkage (10 vs 8) leads to a
decrease in the Tm, possibly reflecting the greater ease that water
molecules can hydrate and destabilize the terminal chain end
group in 8 relative to 10. In all, the DSC data showed that these
charge-reversal lipids form organized structures with melting
temperatures higher than 0 °C, the melting temperature of
DOTAP, a common amphiphile used for gene transfection.
The mean diameter of the supramolecular structures formed
by the different amphiphiles and by the amphiphile/DNA
complexes was measured by dynamic light scattering (DLS).
Compound 1 formed structures with a mean diameter of 110 nm,
and the diameter increased to 562 nm in the presence of DNA.
Amphiphile 6 formed structures with the same mean diameter
(110 nm) and in presence of DNA formed structures with a mean
diameter of 680 nm. Compounds 8 and 9 formed aggregates of 1
μm and larger in diameter as measured by DLS. Likewise,
compounds 2, 3, 7, and 10 afforded very large aggregates and
precipitated products.
At 22 °C, X-ray diffraction patterns of amphiphile 1 in the
absence of DNA gave three orders of a lamellar repeat period of
5.22 ( 0.03 nm with a sharp wide-angle spacing of 0.46 (
0.01 nm. These data indicate that compound 1 formed gel-phase
multibilayers at 22 °C, consistent with the DSC data (Table 1).
In the presence of DNA, fully hydrated amphiphile 1 gave a
lamellar repeat period of 5.31 ( 0.14 nm with a sharp wide-angle
spacing of 0.46 ( 0.01 nm. Thus, the presence of the DNA did
not significantly change d, indicating that the DNA was not
intercalated between the lipid layers. Typically, an increase of 1 to
1.4 nm in d is observed with formation of a DNAÀlipid sandwich
structure.71 These XRD data suggest a model where the DNA is
entrapped at the surface or at the interface between the multi-
lamellar vesicles (MLVs) in solution. This model is consistent
with the increase in size observed by DLS, which could be caused
by DNA bridging between these large MLVs. We were able to
obtain a TEM image of amphiphile 1 in the presence, but not the
absence, of DNA. The TEM micrograph showed multilamellar
structures with an average vesicle diameter of 200 nm
(Figure 12). Repeated attempts to obtain X-ray data on 6 in
the presence and absence of DNA were unsuccessful.
Figure 12. TEM micrographs of amphiphile 1 in aqueous solution,
negative staining with ammonium molybdate 1% in water.
Figure 13. Flow cytometry histogram for CHO cells transfected with
rhodamine-labeled DNA using either 1 (blue) or 6 (red). Negative
control (black).
Of the amphiphiles, compound 6 has a melting temperature
slightly lower than 1 and forms similarly sized structure by DLS
as 1, but does not give high transfection activity. We subsequently
examined if the difference in activity between 1 and 6 was related
to the amount of lipoplex taken up by the CHO cells. Cells were
transfected with rhodamine-labeled DNA using either 1 or 6, and
the resulting cellular rhodamine fluorescence signal was recorded
by flow cytometry. As shown in Figure 13, a significant amount of
fluorescence was observed when DNA was delivered using 1,
while the fluorescence signal was almost the same as that of the
negative control when delivered using 6. Cell population analysis
revealed that 27% of the cells were rhodamine-positive when
transfected with 1, compared to only 0.5% cells when transfected
with 6. These data suggest that the difference in cellular uptake
may be responsible for the difference in the transfection activity
between 1 and 6. The underlying mechanism for the difference in
cellular uptake is unknown and currently under investigation, but
6 does form a relatively weaker supramolecular structure than 1
as evident by the lower Tm.
Altogether, the results suggest a number of key findings
responsible for the higher transfecton activity observed with
amphiphile 1 compared to compounds 2À10. First, compounds
2 and 4 do not bind DNA, while amphiphiles 1, 3, and 5À10 do
bind DNA, at roughly the same magnitude. Amphiphiles 1, 6, 8,
696
dx.doi.org/10.1021/bc1004526 |Bioconjugate Chem. 2011, 22, 690–699

Bioconjugate Chemistry
ARTICLE
and 10 release DNA relatively rapidly in the presence of an
esterase compared to the amphiphile 9. Amphiphiles 3, 5
(DOTAP), and 7 do not release DNA in the presence of an
esterase. Amphiphiles 3, 7, 8, 9, and 10 formed aggregates of 1
μm or larger in diameter. Amphiphiles 1 and 6 both form
similarly sized nanometer lipoplexes; however, more lipoplexes
formed with amphiphile 1 are taken up by CHO than 6.
In summary, in vitro gene delivery using charge reversal or
charge switchable amphiphiles is a new approach and, like other
functional delivery systems, provides a means to increase transfec-
tion activity, as well as potentially new reagents to study the
transfection pathway. These amphiphiles were designed to trans-
form from a cationic species that binds DNA to an anionic species
that loses affinity to DNA after cleavage by an esterase. This
charge-reversal effect would release DNA from the supramolecular
complex. To evaluate the role of key amphiphile structural
components, we synthesized in good yield a series of amphiphiles
to determine the physicochemical properties and transfection
efficiency of these amphiphiles. This series of amphiphiles in-
cluded those that had C10 to C16 chains and a benzyl, n-butyl, or
ethyl ester terminal group, as well as amphiphiles possessing very
short alkyl chains and chains terminated with a carboxylate. All of
these amphiphiles bind DNA and form amphiphile/DNA assem-
blies except for the compounds lackinglonghydrophobicchains or
those terminated with a carboxylate. The DNA release from the
complexes and the dissociation rate of the complexes is controlled
by the chemical structure of the lipid, which plays an important
role in the stability and organization of the resulting supramole-
cular assembly. For example, the enzymatic hydrolysis rate of the
ester linkage was dependent on hydrophobicity of the amphiphile,
as the rate of hydrolysis was faster for the ethyl than the n-butyl
analogue. The XRD data showed that amphiphile 1 forms bilayer
organizations in the presence and absence of DNA, and DLS and
TEM data showed that this amphiphile formed vesicles and
lipoplexes with DNA on the order of 550 nm. The other
amphiphiles organized themselves in larger aggregates. DSC data
showedthat the meltingtemperatures werebetween 45 and 80 °C,
with those amphiphiles possessing the longer alkyl chains having
higher melting temperatures.
The transfection experiments showed that the highest DNA
transfection activity was observed for amphiphile 1. This activity
was observed in three different cell lines—Chinese hamster
ovarian (CHO), human embryonic kidney (HEK293), and
erythroleukemic (K562) cells. All other amphiphiles (except
DOTAP and Transfast) showed minimal transfection activity,
comparable to that of the negative control. Further studies will
investigate the mechanism of cellular uptake and DNA delivery to
the nucleus. This concept of charge reversal or charge switching
has been successfully extended to additional cationic lipid
structures,66 zwitteronic lipids,65 ferrocene-containing lipids,55,56
and polymers41,72 for DNA transfection, suggesting that this is an
important, and potentially general, approach. Our data show that
charge-reversal amphiphile 1 has all of the characteristics neces-
sary for DNA delivery: it binds DNA, releases DNA upon charge
reversal, forms well-ordered bilayers, forms lipoplexes of nan-
ometer size, is taken up by cells, and efficiently transfects several
cell types.
This material is available free of charge via the Internet at http://
pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*Corresponding author. E-mail: mgrin@bu.edu.
Author Contributions
zContributed equally to the work.
’ ACKNOWLEDGMENT
This work was supported by the Army Research Office (PB)
and Grants GM27278 (TJM) and EB005658 (MWG) from the
National Institutes of Health. We thank Dr. Dan Luo and Erin
King for helpful discussions and technical advice at Cornell
University. We thank Mr. Yuxing for help with the initial studies.
We also thank Drs. Jun-lin Guan and Zara Melkoumian for
providing the cell lines, and Dr. Stephen Lee (ARO) for support
and discussions.
’ REFERENCES
(1) Deregowski, P., and Canalis, E. (2008) Gene delivery by retro-
viruses. Methods Mol. Biol. 455, 157–162.
(2) Flotte, T. R. (2004) Gene therapy progress and prospects:
Recombinant adeno-associated virus (rAAV) vectors. Gene Ther.
11, 805–810.
(3) Lundstrom, K. (2004) Gene therapy applications of viral vectors.
Technol. Cancer Res. Treatment 3, 467–477.
(4) Flotte, T. R. (2007) Gene therapy: The first two decades and the
current state-of-the-art. J. Cell. Physiol. 213, 301–305.
(5) Ozawa, K. (2007) Gene therapy using AAV vectors. Drug
Delivery Syst. 22, 643–650.
(6) Mintzer, M. A., and Simanek, E. E. (2009) Nonviral vectors for
gene delivery. Chem. Rev. 109, 259–302.
(7) Ewert, K. K., Ahmad, A., Bouxsein, N. F., Evans, H. M., and
Safinya, C. R. (2008) Non-viral gene delivery with cationic liposome-
DNA complexes. Methods Mol. Biol. 433, 159–175.
(8) Wolff, J. A., and Rozema, D. B. (2007) Breaking the bonds: non-
viral vectors become chemically dynamic. Mol. Ther. 16, 8–15.
(9) Wong, S. Y., Pelet, J. M., and Putnam, D. (2007) Polymer
systems for gene delivery-past, present, and future. Prog. Polym. Sci.
32, 799–837.
(10) Dang, J. M., and Leong, K. W. (2006) Natural polymers for
gene delivery and tissue engineering. Adv. Drug Delivery Rev.
58, 487–499.
(11) Christine, C. C., and Huang, L. (2005) Recent advances in non-
viral gene delivery. Adv. Genetics 53, 3–18.
(12) Reineke, T. M., and Grinstaff, M. W. (2005) Designer materials
for nucleic acid delivery. Mat. Res. Soc. Bull. 30, 635–639.
(13) Luo, D., and Saltzman, W. M. (2000) Synthetic DNA delivery
systems. Nat. Biotechnol. 18, 33–37.
(14) Kabanov, A. V., Felgner, P. L., and Seymour, L. W. (1998) Self-
assembling complexes for gene delivery: From laboratory to clinical trial,
John Willey and Sons, New York.
(15) Behr, J. P. (1994) Gene transfer with synthetic cationic
amphiphiles: prospects for gene therapy. Bioconjugate Chem. 5, 382–389.
(16) Miller, A. D. (1998) Cationic liposomes for gene therapy.
Angew. Chem. Int. 37, 1768–1785.
(17) Marshall, E. (2000) Clinical research - FDA halts all gene
therapy trials at Penn. Science 287, 565–567.
(18) Felgner, P. L., Gadek, T. R., Holm, M., Roman, R., Chan, H. W.,
Wenz, M., Northrop, J. P., Ringgold, G. M., and Danielsen, M. (1987)
Lipofectin: A highly efficient, lipid mediated DNA-transfection proce-
dure. Proc. Natl. Acad. Sci. U.S.A. 84, 7413–7417.
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental procedures,
b
fluorescence studies, DNA transfection, and cytotoxicity data.
697
dx.doi.org/10.1021/bc1004526 |Bioconjugate Chem. 2011, 22, 690–699

Bioconjugate Chemistry
ARTICLE
(19) Felgner, J. H., Kumar, R., Sridhar, C. N., Wheeler, C. J., Tsai, Y. J.,
Border, R., Ramsey, P., Martin, M., and Felgner, P. L. (1994) Enhanced
gene delivery and mechanism studies with a novel series of cationic lipid
formulations. J. Biol. Chem. 269, 2550–2561.
(39) Camplo, M., Khiati, S., Ceballos, C., Prata, C. A. H., Giorgio, S.,
Marsal, P., Barthꢀelꢀemy, P., and Grinstaff, M. W. (2010) Cationic
nucleoside lipids derived from universal bases: a rational approach for
siRNA transfection. Bioconjugate Chem. 21, 1062–1069.
(20) Behr, J. P., Demeneix, B., Loeffler, J. P., and Perez-Mutul, J.
(1989) Efficient gene transfer into mammalian primary endocrine cells
with lipopolyamine coated DNA. Proc. Natl. Acad. Sci. U.S.A.
86, 6982–6986.
(21) Gaucheron, J., Santaella, C., and Vierling, P. (2002) Transfec-
tion with fluorinated lipoplexes based on fluorinated analogues of
DOTMA, DMRIE and DPPES. Biochim. Biophys. Acta À Biomembr.
1564, 349–358.
(22) Boussif, O., Gaucheron, J., Boulanger, C., Santaella, C., Kolbe,
H. V. J., and Vierling, P. (2001) Enhanced in vitro and in vivo cationic
lipid-mediated gene delivery with a fluorinated glycerophosphoethano-
lamine helper lipid. J.Gene Med. 3, 109–114.
(23) Schulze, U., Schmidt, H., and Safinya, C. R. (1999) Synthesis of
novel cationic poly(ethylene glycol) containing lipids. Bioconjugate
Chem. 10, 548–552.
(24) Bessodes, M., Dubertret, C., Jaslin, G., and Scherman, D.
(2000) Synthesis and biological properties of new glycosidic cationic
lipids for DNA delivery. Bioorg. Med. Chem. Lett. 10, 1393–1395.
(25) Herscovici, J., Egron, M. J., Quenot, A., Leclercq, F., Leforestier, N.,
Mignet, N., Wetzer, B., and Scherman, D. (2001) Synthesis of new
cationic lipids from unsaturated glycoside scaffold. Org. Lett. 3
1893–1896.
(26) Arigon, J., Prata, C. A. H., Grinstaff, M. W., and Barthꢀelꢀemy, P.
(2005) Nucleic acid complexing glycosyl nucleoside-based amphiphile.
Bioconjugate Chem. 16, 864–872.
(27) Prata, C. A. H., Zhang, X., Luo, D., McIntosh, T. J., Barthꢀelꢀemy, P.,
and Grinstaff, M. W. (2008) Lipophilic peptides for gene delivery.
Bioconjugate Chem. 19, 418–420.
(28) Chan, E., Amon, M., Marano, R. J., Wimmer, N., Kearns, P. S.,
Manolios, N., Rakoczy, P. E., and Toth, I. (2007) Novel cationic
lipophilic peptides for oligodeoxynucleotide delivery. Bioorg. Med. Chem.
15, 4091–4097.
(29) Montier, T., Benvegnu, T., Jaffres, P. A., Yaouanc, J. J., and
Lehn, P. (2008) Progress in cationic lipid-mediated gene transfection: a
series of bio-inspired lipids as an example. Curr. Gene Ther. 8, 296–312.
(30) Obata, Y., Suzuki, D., and Takeoka, S. (2008) Evaluation of
cationic assemblies constructed with amino acid based lipids for plasmid
DNA delivery. Bioconjugate Chem. 19, 1055–1063.
(31) Gissot, A., Camplo, M., Grinstaff, M. W., and Barthꢀelꢀemy, P.
(2008) Nucleoside, nucleotide and oligonucleotide based amphiphiles:
a successful marriage of nucleic acids with lipids. Org. Biomol. Chem.
6, 1324–1333.
(32) Moreau, L., Barthꢀelꢀemy, P., Maataoui, M. E., and Grinstaff,
M. W. (2004) Supramolecular assemblies of nucleoside-based amphi-
philes. J. Am. Chem. Soc. 126, 7533–7539.
(33) Moreau, L., Ziarelli, F., Grinstaff, M. W., and Barthꢀelꢀemy, P.
(2006) Self-assembled microspheres from F-block elements and nu-
cleoamphiphiles. Chem. Commun. 15, 1661–1663.
(40) Boomer, J. A., and Thompson, D. H. (1999) Synthesis of acid-
labile diplasmenyl lipids for drug and gene delivery applications. Chem.
Phys. Lipids 99, 145–153.
(41) Liu, X., Yang, J. W., and Lynn, D. M. (2008) Addition of
‘charge-shifting’ side chains to linear poly(ethyleneimine) enhances cell
transfection efficiency. Biomacromolecules 9, 2063–2071.
(42) Bell, P. C., Bergsma, M., Dolbnya, I. P., Bras, W., Stuart, M. C.,
Rowan, A. E., Feiters, M. C., and Engberts, J. B. F. N. (2003)
Transfection mediated by genimi Surfactants: Engineered escape from
the endosomal compartment. J. Am. Chem. Soc. 125, 1551–1558.
(43) Lynn, D. M., and Langer, R. (2000) Degradable poly(beta-
amino esters): Synthesis, characterization, and self-assembly with plas-
mid DNA. J. Am. Chem. Soc. 122, 10761–10768.
(44) Pack, D. W., Putnam, D., and Langer, R. (2000) Design of
imidazole-containing endosomolytic biopolymers for gene delivery.
Biotechnol. Bioeng. 67, 217–223.
(45) Jones, R. A., Cheung, C. Y., Black, F. E., Zia, J. K., Stayton, P. S.,
Hoffman, A. S., and Wilson, M. R. (2003) Poly(2-alkylacrylic acid)
polymers deliver molecules to the cytosol by pH-sensitive disruption of
endosomal vesicles. Biochem. J. 372, 65–75.
(46) Zhu, J., Munn, R. J., and Nantz, M. H. (2000) Self-cleaving
ortho ester lipids: A new class of pH-vulnerable amphiphiles. J. Am.
Chem. Soc. 122, 2645–2646.
(47) Jewell, C. M., Zhang, J., Fredin, N. J., Wolff, M. R., Hacker, T. A.,
and D.M., L. (2006) Release of plasmid DNA from intravascular stents
coated with ultrathin multilayered polyelectrolyte films. Biomacromole-
cules 7, 2483–2491.
(48) Legendre, J. Y., and Szoka, F. C. J. (1992) Delivery of plasmid
DNA into mammalian cell lines using pH sensitive liposomes: Compar-
ison with cationic liposomes. Pharm. Res. 10, 1235–1242.
(49) Midoux, P., Pichon, C., Yaouanc, J. J., and Jaffrꢁes, P. A. (2009)
Chemical vectors for gene delivery: a current review on polymers,
peptides and lipids containing histidine or imidazole as nucleic acids
carriers. Br. J. Pharmacol. 157, 166–178.
(50) McGregor, C., Perrin, C., Monck, M., Camilleri, P., and Kirby,
A. J. (2001) Rational apporaches to the design of cationic genimi
surfactants for gene delivery. J. Am. Chem. Soc. 123, 6215–6220.
(51) Gosselin, M. A., Guo, W., and Lee, R. J. (2002) Incorporation of
reversibly cross-linked polyplexes into LPDII vectors for gene delivery.
Bioconjugate Chem. 13, 1044–1053.
(52) Tang, F., Wang, W., and Hughes, J. A. (1999) Cationic
liposomes containing disuldide bonds in delivery of plasmid DNA. J.
Liposome Res. 9, 331–347.
(53) Byk, G., Wetzer, B., Frederic, M., Dubertret, C., Pitard, B.,
Jaslin, G., and Scherman, D. (2000) Reduction-sensitive lipopolya-
mines as a novel nonviral gene delivery system for modulated release of
DNA with improved transgene expression. J. Med. Chem.
43, 4377–4387.
(34) Moreau, l., Campins, N., Grinstaff, M. W., and Barthꢀelꢀemy, P.
(2006) A fluorocarbon nucleoamphiphile for the construction of
actinide loaded microspheres. Tetrahedron Lett. 47, 7117–7120.
(35) Barthꢀelꢀemy, P., Prata, C. A. H., Filocamo, S. F., Immoos, C. E.,
Maynor, B. W., Hashmi, S. A. N., Lee, S. J., and Grinstaff, M. W. (2005)
Supramolecular assemblies of DNA with neutral nucleoside amphi-
philes. Chem. Commun. 1261–1263.
(36) Campins, N., Dieudonnꢀe, P., Grinstaff, M. W., and Barthꢀelꢀemy,
P. (2007) Nanostructured assemblies from nucleotide based amphi-
philes. New. J. Chem. 31, 1928–1934.
(54) Prata, C. A. H., Zhao, Y., Barthꢀelꢀemy, B., Li, Y., Luo, D.,
McIntosh, T. J., Lee, S. J., and Grinstaff, M. W. (2004) Charge-reversal
amphiphiles for gene delivery. J. Am. Chem. Soc. 126, 12744–12745.
(55) Abbott, N. L., Jewell, C. M., Hays, M. E., Kondo, Y., and Lynn,
D. M. (2005) Ferrocene-containing cationic lipids: Influence of redox
state on cell transfection. J. Am. Chem. Soc. 127, 11576–11577.
(56) Jewell, C. M., Hays, M. E., Kondo, Y., Abbott, N. L., and Lynn,
D. M. (2008) Chemical activation of lipoplexes formed from DNA and a
redox-active, ferrocene-containing cationic lipid. Bioconjugate Chem.
19, 2120–2128.
(37) Chabaud, P., Camplo, M., Payet, D., Serin, G., Moreau, L.,
Barthꢀelꢀemy, P., and Grinstaff, M. W. (2006) Cationic nucleoside lipids
for gene delivery. Bioconjugate Chem. 17, 466–472.
(38) Moreau, L., Barthꢀelꢀemy, P., Li, Y., Luo, D., Prata, C. A. H., and
Grinstaff, M. W. (2005) Nucleoside phosphocholine amphiphile for in
vitro DNA transfection. Mol. BioSystems 1, 260–264.
(57) Fuchs, S., Buethe, D., Khanna, A., Yadava, P., and Hughes, J. A.
(2004) Sulfhydryl based cationic surfactants and the impact of polyan-
ions on disulfide bond formation: Implications for gene transfer vectors.
J. Drug Targeting 12, 347–353.
(58) Jewell, C. M., Hays, M. E., Kondo, Y., Abbott, N. L., and Lynn,
D. M. (2006) Ferrocene-containing cationic lipids for the delivery of
698
dx.doi.org/10.1021/bc1004526 |Bioconjugate Chem. 2011, 22, 690–699

Bioconjugate Chemistry
ARTICLE
DNA: Oxidation state determines transfection activity. J. Controlled
Release 112, 129–138.
(59) Miyata, K., Kakizawa, Y., Nishiyama, N., Harada, A., Yamasaki, Y.,
Koyama, H., and Kataoka, K. (2004) Block catiomer polyplexes with
regulated densities of charge and disulfide cross-linking directed to enhance
gene expression. J. Am. Chem. Soc. 126, 2355–2361.
(60) Wang, J., Guo, X., Xu, Y., Barron, L., and Szoka, F. C. (1998)
Synthesis and characterization of long chain alkyl acyl carnitine esters.
Potentially biodegradable cationic lipids for use in gene delivery. J. Med.
Chem. 41, 2207–2215.
(61) Pak, C. C., Ali, S., Janoff, A. S., and Meers, P. (1998) Triggerable
liposomal fusion by enzyme cleavage of a novel peptide-lipid conjugate.
Biochem. Biophys. Acta 1372, 13–27.
(62) Wang, J., Mao, H. Q., and Leong, K. W. (2001) A novel
biodegradable gene carrier based on polyphosphoester. J. Am. Chem.
Soc. 123, 9480–9481.
(63) Takeda, N., Nakamura, E., Yokoyama, M., and Teuro, O.
(2004) Temperature-responsive polymeric carriers incorporating hy-
drophobic monomers for effective transfection in small doses. J. Con-
trolled Release 95, 343–355.
(64) Kurisawa, M., Yokoyama, M., and Okano, T. (2000) Gene
expression control by thermo-responsive polymeric gene carriers
J. Controlled Release 69, 127–137.
(65) Prata, C. A. H., Li, Y., Luo, D., McIntosh, T. J., Barthꢀelꢀemy, P.,
and Grinstaff, M. W. (2008) A new helper phospholipids for gene
delivery. Chem. Commun. 1566–1568.
(66) Zhang, X. X., Prata, C. A. H., McIntosh, T. J., Barthꢀelꢀemy, P.,
and Grinstaff, M. W. (2010) The effect of charge-reversal amphiphile
spacer composition on DNA and siRNA delivery. Bioconjugate Chem.
21, 988–993.
(67) Mellman, I., Fuchs, R., and Helenius, A. (1986) Acidification of
the endocytic and exocytic pathways. Annu. Rev. Biochem. 55, 663–700.
(68) Lavis, L. D., Chao, T. Y., and Raines, R. T. (2006) Fluorogenic
label for biomolecular imaging. ACS Chem. Biol. 1, 252–260.
(69) Geall, A. J., and Blasgbrough, I. S. (2000) Rapid and sensitive
ethidium bromide fluorescence quenching assay of polyamine conju-
gate-DNA interactions for the analysis of lipoplex formation in gene
therapy. J. Pharm. Biomed. Anal. 22, 849–859.
(70) Cain, B. F., Baguley, B. C., and Demmy, M. W. (1978) Potential
antitumor agents. 28. Deoxyribonucleic acid polyintercalating agents. J.
Med. Chem. 21, 658–668.
(71) Koltover, I., Salditt, T., Radler, J. O., and Safinya, C. R. (1998)
An inverted hexagonal phase of cationic liposome-DNA complexes
related to DNA release and delivery. Science 281, 78–81.
(72) Liu, X. H., Yang, J. W., Miller, A. D., Nack, E. A., and Lynn, D. M.
(2005) Charge-shifting cationic polymers that promote self-assembly
and self-disassembly with DNA. Macromolecules 38, 7907–7914.
699
dx.doi.org/10.1021/bc1004526 |Bioconjugate Chem. 2011, 22, 690–699