Prodendronic polyamines from stable or labile methacrylates obtained by selective Michael addition onto asymmetric diacrylic compounds

Article abstract of DOI:10.1002/pola.27240

A new synthetic strategy for the preparation of methacrylic monomers and polymers carrying acyl β-amino groups is presented. The approach is based on the Michael addition of aliphatic amines onto asymmetric acrylic/methacrylic compounds, reacting the amine highly selectively with the acrylic unit while leaving the methacrylic moiety unreacted. The corresponding polymers are then obtained by conventional radical polymerization. The use of N,N,N,N- tetraethyldiethylenetriamine (TEDETA) as the secondary amine leads to TEDETA moieties supported on polymeric chains. The new aminopolymers are sensitive to pH and to temperature exhibiting a lower critical solution temperature of between 50 and 90 C. A further interesting feature of the new approach is that the stability toward hydrolysis of the side β-amino acyl compounds was found to be dependent on whether an acrylamide or an acrylate is employed as the acrylic group of the asymmetric starting material. The esters exhibit an enhanced sensitivity to hydrolysis, compared to standard aliphatic esters, and decompose releasing a derivative of the amine precursor, within hours or weeks, depending on the pH and temperature conditions. The use of the amides leads to stable polymers when the same experimental conditions are applied. The novel dendronic polyamines have been proven to interact with DNA and to transfect cells with efficiency close to that obtained with polyethyleneimine vectors used as positive controls.

Full text of DOI:10.1002/pola.27240

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
Prodendronic Polyamines from Stable or Labile Methacrylates Obtained
by Selective Michael Addition onto Asymmetric Diacrylic Compounds
1
1
2
1
Juan Alfonso Redondo, Rodrigo Navarro, Enrique Mart ꢀı nez-Campos, M oꢀ nica P eꢀ rez-Perrino,
3
2
1
1
1
Rodrigo Par ꢀı s, Jos eꢀ Luis L oꢀ pez-Lacomba, Carlos Elvira, Helmut Reinecke, Alberto Gallardo
1
Institute of Polymer Science and Technology, ICTP-CSIC. Juan de la Cierva 3, 28006 Madrid, Spain
2
Institute of Biofunctional Studies, Universidad Complutense de Madrid. Paseo de Juan XXIII 1, 28040 Madrid, Spain
3
Centro de Investigaci oꢀ n y Desarrollo, Repsol, Direcci oꢀ n de Tecnolog ꢀı a Qu ꢀı mica, Autov ꢀı a A-5, Km.18, 28931 M oꢀ stoles, Madrid,
Spain
Correspondence to: R. Navarro (E-mail: rodrigo23_nc@yahoo.es) or A. Gallardo (E-mail: gallardo@ictp.csic.es)
Received 6 February 2014; accepted 30 April 2014; published online 20 May 2014
DOI: 10.1002/pola.27240
ABSTRACT: A new synthetic strategy for the preparation of
methacrylic monomers and polymers carrying acyl b-amino
groups is presented. The approach is based on the Michael
addition of aliphatic amines onto asymmetric acrylic/metha-
crylic compounds, reacting the amine highly selectively with
the acrylic unit while leaving the methacrylic moiety unreacted.
The corresponding polymers are then obtained by conven-
metric starting material. The esters exhibit an enhanced sensi-
tivity to hydrolysis, compared to standard aliphatic esters, and
decompose releasing a derivative of the amine precursor,
within hours or weeks, depending on the pH and temperature
conditions. The use of the amides leads to stable polymers
when the same experimental conditions are applied. The novel
dendronic polyamines have been proven to interact with DNA
and to transfect cells with efficiency close to that obtained with
0
0
tional radical polymerization. The use of N,N,N ,N -tetraethyl-
diethylenetriamine (TEDETA) as the secondary amine leads to
TEDETA moieties supported on polymeric chains. The new
aminopolymers are sensitive to pH and to temperature exhibit-
polyethyleneimine vectors used as positive controls. V
2014
C
Wiley Periodicals, Inc. J. Polym. Sci., Part A: Polym. Chem.
2014, 52, 2297–2305
ing a lower critical solution temperature of between 50 and
ꢀ
90
C. A further interesting feature of the new approach is that
KEYWORDS: asymmetric diacrylic compounds; degradation;
drug delivery systems; functionalization of polymers; gene
delivery; hydrolysis; Michael addition; stimuli responsive
polymers
the stability toward hydrolysis of the side b-amino acyl com-
pounds was found to be dependent on whether an acrylamide
or an acrylate is employed as the acrylic group of the asym-
INTRODUCTION The Michael reaction is a powerful tool fre-
quently used in organic synthesis for the mild formation of
C-C or C-heteroatom bonds. Because of its high versatility and
remarkable selectivity, it has been employed in a large number
of processes including the total synthesis of natural prod-
tems leading to a versatile route to prepare new polymeriz-
able methacrylates (see Scheme 1).
When amines are used as nucleophiles, there is an additional
issue related to this reaction: the addition to an acrylate
leads to a b-amino ester, which exhibits an enhanced sensi-
tivity to hydrolysis, transesterification, and so forth, when
compared to standard aliphatic ester such as the methacrylic
ester itself (also present in the structure, see Scheme 1). The
addition to an acrylamide, however, leads to a b-amino
amide group, which has been reported to be usually much
1
,2
ucts.
with electron poor olefins as (meth)acrylics. Focusing on
meth)acrylics as acceptors, alkyl methacrylates are relatively
It encompasses different couplings of nucleophiles
(
poor Michael acceptors, and require strong nucleophiles (such
as carbanions used in the Group-Transfer Polymerization
3
(
GTP) ). Alkyl acrylates or acrylamides, on the other hand, are
5
–8
more stable than the b-amino ester.
In other words, the
better acceptors and readily accept weaker nucleophiles such
4
as amines or thiols. Owing to these differences, in this work,
lability of the spacer can be predefined at some extent by
the appropriate choice of group X of the methacrylate in
Scheme 1 (X 5O highly degradable, X 5 NH comparatively
stable).
we have elaborated reaction conditions that allow for the
selective Michael addition of amines onto asymmetric
0
methacrylate-R-acrylate or methacrylate-R -acrylamide sys-
Additional Supporting Information may be found in the online version of this article.
2014 Wiley Periodicals, Inc.
V
C
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 2297–2305
2297

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
and freshly used. 2-Aminoethanol, TEDETA, diethylamine, PEI
25 kDa, hyperbranched) (Aldrich), and triethylamine (Schar-
(
lau) were used as received.
Synthesis
2-(Acryloylamino)-2-ethylmethacrylate
2
-(Acryloylamino)-2-ethylmethacrylate (AAEMA) was prepared
19
in a two-step synthesis as described previously. In the first
step, 20 mmol of 2-aminoethanol hydrochloride was dissolved
in 200 mL of acetonitrile. About 200 mg of hydroquinone as
inhibitor and 20 mmol of methacryloyl chloride were added,
and the mixture was refluxed for 2 h. The mixture was then
allowed to cool down to room temperature and the excess of
unreacted 2-aminoethanol hydrochloride was filtered off. The fil-
trate was dried under reduced pressure and used for the second
step without further purification. The obtained yield for this
SCHEME 1 Scheme of the selective Michael addiction pro-
posed and studied in this work. X5O or NH. In case of primary
1
amines, R 5 H.
step was 45%. Then, the residue was suspended in 300 mL of
ꢀ
chloroform and cooled to 0 C. Under a stream of dry N , the
2
In this work, we have applied this synthetic design to the
coupling of the dendronic N,N,N ,N -tetraethyldiethylenetri-
equimolar amount of acryloyl chloride and the double molar
amount of triethylamine were added. The reaction mixture was
allowed to warm up to room temperature. After 3 h, the organic
phase was washed three times with water, dried over anhydrous
0
0
amine (TEDETA) to two asymmetric diacrylic compounds
3
(
R 5 -CH
2
-CH
2
- in Scheme 1). Secondary amines as TEDETA
simplify the approach described before, as primary amines
must be considered as difunctional nucleophiles in this type
of reaction that may lead to double addition of the amine
and consequently to a mixture of products.
MgSO , and the solvent was stripped off. Purification of the raw
4
product was carried out by column chromatography using silica
gel and dichloromethane as the eluent. Yield 91%.
1
H NMR (CDCl , 300 MHz) d: 6.26 (d, 1H, J 5 15.0 Hz), 6.15–
3
The radical polymerization of these new methacrylates bear-
ing TEDETA will lead to a TEDETA moiety supported on a
polymeric chain. Supported TEDETA is very attractive in dif-
6
3
.06 (m, 2H), 5.66–5.57 (m, 2H), 4.26 (t, 2H, J 5 5.1 Hz),
.63 (q, 2H, J 5 5.1 Hz), 1.92 (s, 3H).
1
,9–12
ferent fields.
adsorb different proteins
cling of metallic catalysts as copper. Similar supported tria-
It has, for example, recently been used to
2
-(3-(Bis(2(diethylamino)ethyl)amino) propanamido)ethyl
1
3,14
or for the removal and recy-
methacrylate (TN)
1
5
For the synthesis of this monomer, 1.4 mmol of AAEMA were
dissolved in 1.4 mmol of TEDETA. To this solution, a catalytic
amount of acetic acid and hydroquinone were added. The reac-
1
6,17
mines have been proposed for CO capture.
The system
2
has furthermore been applied in atom transfer radical poly-
merization (ATRP) because TEDETA is a well-known ligand
in ATRP. In all these reports the TEDETA ligand is linked to
a preformed surface or polymer chain. In the present work,
we propose an alternative route which is based on a mono-
mer preparation followed by polymerization. The “bottom-
up” nature of this approach gives high flexibility to the
macromolecular design since the properties of the chains
determined by the TEDETA load may be easily modulated
just by choosing the most appropriate comonomers in the
right ratio.
ꢀ
tion was heated up to 100 C and allowed to stir over 6 h.
The resulting residue was purified by column chromatography
using alumina gel, Brockmann II, using dichloromethane/meth-
anol (14:1) as the eluent. The final product was obtained with
a yield close to 80% (Supporting Information Fig. S1).
1
H NMR (CDCl , 300 MHz) d: 6.13 (s, 1H), 5.52 (s, 1H), 4.19
3
(t, 2H, J 5 6 Hz), 3.51 (q, 2H, J 5 6 Hz), 2.71–2.65 (m, 2H),
2
0
.59–2.47 (m, 16H), 2.36–2.32 (m, 2H), 1.93 (s, 3H), 1.02–
.96 (m, 12H).
In the last part of the present article, the novel TEDETA car-
rying polymers were tested for possible use in gene therapy.
This study is of interest due to the similarity of the triamine
dendron to an inner block of hyperbranched polyethylenei-
mine (PEI) a standard polyamine used with good success in
2
-(Acryloyloxy)ethyl methacrylate (AOEMA)
This asymmetric diacrylic crosslinker was obtained by esteri-
fication of HEMA with acryloyl chloride. Details of the syn-
thesis have been reported elsewhere.
2
0–23
1
8
1
these applications.
H NMR (CDCl , 300 MHz) d: 6.43 (d, 1H, J 5 9 Hz), 6.20–
3
6
4
.05 (m, 2H), 5.86 (d, 1H, J 5 12 Hz), 5.59 (s, 1H), 4.39 (s,
H), 1.94 (s, 3H)
EXPERIMENTAL
Materials
2-((3-(Bis(2-(diethylamino)ethyl)amino)
propanoyl)oxy)ethyl methacrylate (TO)
0
2,2 -Azobisisobutyronitrile (AIBN, Merck) was recrystallized
twice from ethanol. Methacryloyl chloride, acryloyl chloride and
-hydroxyethyl methacrylate (HEMA, Aldrich) were distilled
To a solution of 2-(acryloyloxy)ethyl methacrylate (1.4
mmol) in chloroform (5 mL) a catalytic amount of acetic
2
2
298
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 2297–2305

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
acid and hydroquinone as radical inhibitor were added.
Then, an equivalent of TEDETA was added dropwise. The
resulting mixture was kept at 40 C and stirred over 48 h.
were performed in triplicate and nontreated cells and cells
treated with naked DNA were used as negative controls.
ꢀ
Cellular viability was tested after the transfection measure-
ments with the calcein AM assay (Molecular Probes, Eugene,
OR). This fluorescent dye can be transported through the cel-
lular membrane into living cells, making it a cell viability
marker.
The final product was used without further purification. The
final product was obtained with a yield close to 99% accord-
ing to H NMR characterization (Supporting Information
1
Fig. S2).
1
H NMR (CDCl
3
, 300 MHz) d: 6.14 (s, 1H), 5.59 (s, 1H), 4.28
After the transfection measurement, the remaining media
was removed from each well, and 100 lL of calcein solution
(t, 2H, J 5 4.5 Hz), 3.86 (t, 2H, J 5 4.5 Hz), 2.82 (t, 2H, J 5 7.2
Hz), 2.57–2,43 (m, 18H), 1.95 (s, 3H), 1.04–0.98 (m, 12H).
(
3
4 lg/mL) was added. Cells were incubated for 30 min at
ꢀ
7 C in a humidified atmosphere of 5% CO . Finally, fluores-
2
General Polymerization Procedure
cence was measured at kex 495 nm and kem 515 nm, using
a BioTek Synergy HT (Biotek Instruments) plate reader. Data
were normalized against data obtained from nontreated
cells.
The homo and copolymerizations were carried out by con-
ꢀ
ventional free radical polymerization at 60 C, using AIBN
(
2
2
[I] 5 1.5 3 10 mol/L) as the radical initiator and a mono-
mer concentration of 1M using DMF as the solvent. All solu-
tions were carefully degassed through dried nitrogen flow to
remove all the oxygen traces. After 24 h, the obtained poly-
mers were extensively washed using dialysis in distilled
water and finally the water was removed by a freeze-drying
process. The dialysis membranes Spectra/PorVR (Spectrum
Methods
Thin-layer chromatography (TLC) was performed on alumi-
num sheets 60 F254 Merck silica gel and compounds were
visualized by irradiation with UV light and/or by treatment
with a solution of Ce MoO in water, a solution of Ninhydrin
Laboratories) present a molecular weight cut off of 3.5 kDa.
2
4
Polyplex Formation and Characterization
in n-BuOH/EtOH or H SO4 (5%) in EtOH followed by heat-
2
Polymer/DNA complexes (polyplexes) were prepared at
room temperature in Milli-Q water 1 h before use. Various
nitrogen/phosphate (N/P) mol ratios were made from 1/1
up to 8/1, with a final DNA concentration of 1 lg/100 lL of
water.
ing. Flash chromatography was performed using thick walled
columns, employing silica gel (Merck 60: 0.040–0.063 nm)
or neutral activated aluminum oxide, Brockmann II (Aldrich).
NMR ( H NMR) spectra were recorded on 300 MHz (Inova
300 or Bruker 300) and 400 MHz (Inova 400 or Mercury
1
4
00) spectrometers, using CDCl or D O as solvents at room
3 2
Transfection and Cell Viability Assays
temperature. Chemical shift values are reported in parts per
million (d) relative to tetramethylsilane (TMS). Coupling con-
stants (J values) are reported in Hertz (Hz), and spin multi-
plicities are indicated by the following symbols: s (singlet), d
(doublet), t (triplet), q (quartet), m (multiplet).
For this study, mouse Swiss 3T3 embryonic fibroblasts (CCL-92,
ATCC) were used. Cells were routinely grown in Dulbecco’s
modified eagle medium (Gibco), supplemented with 10% fetal
bovine serum and 1% of penicillin/streptomycin. Cells were
subcultured every 3–4 days using trypsin/ethylenediaminotetr-
ꢀ
The degradation experiments of the polymer poly(T ) were
O
acetic acid (EDTA) and maintained at 37 C in a humidified 5%
carried out at different pHs (5, 7, 8, and 9) and monitored in
situ by dynamic NMR ARRAY. These experiments were car-
ried out in a Varian 400. To perform quantitative experi-
ments, the following conditions were used: a pulse sequence
CO atmosphere.
2
Transfection experiments were performed using the lucifer-
ase gene contained in the pCMV-GLuc (New England Biolabs)
plasmid as a reporter gene. The polyplexes were evaluated
using serum-free medium as solvent. PEI/DNA formulations
prepared at the same N/P mol ratios were used as positive
ꢀ
of 7 ls equivalent to a 90 tip angle and a 120-s delay time
were applied in order to allow for the total relaxation of the
protons and to process the individual data. The spinning
rate of the samples was 7 Hz. The sample temperature was
2
controls. About 10,000 cells/cm were seeded in 96-well
ꢀ
plates 24 h before the transfection. In a standard transfec-
tion experiment, cells were incubated with the desired
amounts of polyplexes (100 lL dispersion with 1 lg DNA
maintained at 37 C using the heat controller of the NMR
equipment. The experimental procedure is described in more
2
4
detail elsewhere. The isolated signals at 2.63 and 2.40
ppm, which have been assigned to the CH -COO- of the poly-
per well) for 4 h in a humidified 5% CO atmosphere. Then,
2
2
the medium was removed, and fresh culture medium was
added, leaving cells in culture for 2 additional days. After
this time, the medium was collected and analyzed for the
production of luciferase protein, using the BioLuxVR Gaussia
Luciferase Assay Kit (New England Biolabs) according to the
supplier’s instructions. The luciferase expression was quanti-
fied by measuring the luminescence with a plate reader Bio-
Tek Synergy HT (Biotek Instruments). All measurements
mer and to the CH2-COOH of the free TEDETA residue, were
used to determine the hydrolysis ratio by comparing their
areas, according to these equations:
A2:6352HTO
A2:4052Hh
(1)
(2)
Therefore, the hydrolysis ratio f is given by
h
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 2297–2305
2299

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
Hh
fh5
(3)
HTO1Hh
Control stability experiments of the polymer poly(T ) were
N
carried out under the same experimental conditions as
ꢀ
before (37 C and the same buffers), carrying out one NMR
spectra every 24 h over a 1-week period. No degradation
was observed in any of the experiments.
The LCST of the polymers was determined measuring the
optical transmittance in aqueous buffer solutions at different
pHs, at 500 nm wavelength as a function of the temperature.
The copolymer concentrations were 2 mg/mL. The analyses
were performed in a UV–visible spectrometer Cary 3 BIO-
Varian. The temperature was gradually increased from 25 to
SCHEME 2 Michael addition of TEDETA over the two asym-
metric crosslinkers AOEMA and AAEMA. The best conditions
for each reaction are indicated below: TEDETA over AOEMA: in
ꢀ
ꢀ
ꢀ
chloroform, 40 C (48 h). Quantitative reaction; TEDETA over
9
0
C with a heating rate of 1 C/min. As the T
N
system
ꢀ
ꢀ
AAEMA: mass reaction, 100 C (6 h). 85% of conversion.
gave very similar results at 1 and 5 C/min, the selected
heating rate for poly-T
ꢀ
O
was 5 C/min, trying to minimize
RESULTS AND DISCUSSION
the influence of the degradation process on LCST. The LCST
was estimated as the temperature at the inflection point of
the curve of absorbance against temperature.
The optimization of the reaction conditions of the Michael
addition was carried out using N,N-diethylamine as a model
for an aliphatic secondary amine; once the optimum parame-
ters were found, the reaction conditions were successfully
transferred to the TEDETA addition. As asymmetric diacrylic
crosslinkers, compounds AOEMA and AAEMA were selected
The molecular weight of the polymers was determined by
size exclusion chromatography (SEC) in a Perkin Elmer chro-
matography system with Styragel (300 3 7.8 mm, 5 lm
nominal particle size) water columns. THF with 4% Triethyl-
amine was used as the eluent. The measurements were car-
(
Scheme 2). Both exhibit an ethylene spacer between the
methacrylic and the acrylic unit. AOEMA and AAEMA are not
commercial products but their syntheses have been reported
ꢀ
ried out at 35 C at a flow rate of 1.0 mL/min using a RI
detector. The polymer concentration was 3 mg/mL. The cali-
bration was performed with monodispersed PS standards in
the range of 2.0 and 9000 kDa.
19–23
previously.
In both cases, the amine addition over AOEMA and AAEMA
turned indeed out to be selective toward the acrylate or
acrylamide although with some differences as it is indicated
in Scheme 2. Basically, the addition over the acrylate is much
more favored than the reaction with the acrylamide and
The dissociation constant (pKa) of the poly-T derivative was
N
determined by acid–base titration of a 5-mg polymer solu-
tion in 5-mL aqueous 0.15 M NaCl. An aqueous 0.2 M NaOH
solution was used to carry out the titration using small vol-
umes in order to avoid the modification of the ionic strength.
Before starting the titration experiment, a small volume of a
ꢀ
under gentle conditions such as 40 C complete selectivity is
reached. Besides, the reaction is quantitative and there is no
need for extra purification. On the other hand, the best
1M HCl solution was added to ensure the ionization of the
results for the addition over the acrylamide were obtained at
amine groups. The changes of pH were measured with a
Schott (Schott Instruments) pH meter.
ꢀ
1
00 C without any solvent. In this case, a conversion of
approximately 85% was reached. Selectivity was not com-
plete and some addition over the methacrylate is produced.
A purification step, using alumina deactivated with water to
level Brockmann II, was required.
To assess the polyplex formation via migration properties, a
0.9% (w/v) agarose gel was used, applying 100 V for 30
min. Naked k-Hind III DNA (used as control) and polyplex
samples were mixed with a sucrose loading dye and loaded
into the agarose gel. Afterward, UV light was used to visual-
ize the bands, marked with ethidium bromide (EtBr).
Both, TN and T , exhibited a good polymerizability under
O
conventional radical polymerization conditions. Homopoly-
mers poly-TN and poly-T have been prepared and character-
O
ized. A schematic representation of the polymerization is
shown in Scheme 3.
Polymers and polyplexes were sized using dynamic light
scattering (DLS) (NanoZS Malvern Instruments), and the sur-
face charge was analyzed using a Zetamaster system (Mal-
vern Instruments). Polyplexes were made up in distilled
water to a final volume of 1 mL with 10 lg of DNA and
added to a clear disposable zeta cell. Polymers were meas-
Figure 1 shows the proton NMR spectra of both homopoly-
mers, which are very similar. The two intense signals cen-
tered at 0.95 and 2.58 ppm correspond to the populated
CH -CH -N and -CH -N of the TEDETA unit, respectively. The
3
2
2
ured at a concentration of 2 mg/mL at pH 7 and 9. Three
b-CH3 and b-CH2 signals of the backbone are centered at
around 0.75 and 1.80 ppm, in agreement with other aliphatic
polymethacrylates. The main difference can be found in the -
ꢀ
readings were obtained for each measurement, at 25
C
allowing 1 min of equilibration time. ALV-Correlator Control
Software was used for data.
CH - signals found at lower field (assigned to two of the
2
2
300
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 2297–2305

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
TABLE 1 The Molecular Weights and Polydispersities of Pre-
pared Polymers Were Obtained by SEC Chromatography
ꢀ
LCST ( C)
M
w
M
n
M /M
w n
pH 7.4 pH 8 pH 9 pH 10
Poly-T
N
O
117,000 52,000 2.3
153,000 73,000 2.1
88
69
68
56
56
42
51
–
Poly-T
was not analyzed due to its hydrolyzability. Titration of poly-
TN shows a main pKa of around 8.5. It should be taken into
account that, due to the polymeric form of the compounds
and due to the dendronic structure of the cationizable side
chain, every amine is surrounded by a particular environ-
ment and, therefore, no single discrete pKa should be
expected but a certain interval around this value of 8.5.
X
SCHEME 3 Radical polymerization of T , being X 5 NH or O.
other CH2 of the spacer), which in poly-TO are overlapped
while in poly-T they are not. Integrals are in agreement
with the structure.
N
Molecular weights were characterized by GPC using THF and
% triethylamine as the eluent. The molecular weights and
polydispersities of these polymers are quoted in Table 1.
4
In addition to the pH sensitivity related to the protonation of
the tertiary amines, the polymers have also shown thermo-
sensitivity (lower critical solution temperature, LCST) in
aqueous solution. Polymers became nonsoluble above a criti-
cal temperature (the LCST), which was strongly dependent
Poly-T had a lower molecular weight than poly-T .
N
O
Dual pH and Temperature Sensitivity
ꢀ
Polymers exhibit pH sensitivity related to the cationizable
tertiary amine groups of the dendronic TEDETA blocks. Poly-
TN was titrated in water against a 0.2-M NaOH solution as
on the pH and ranged between 40 and 90 C (see Table 1
and Fig. 3). Below the LCST, the polymer–water interactions
are favored but, above the LCST, the polymer–polymer inter-
actions dominate and the macromolecule becomes hydropho-
bic leading to phase separation. The measured LCST values
are pH dependent, and they have been observed only above
or in the vicinities of the pKa, that is, when at least part of
indicated in the Experimental section (see Fig. 2). Poly-T
O
X
the T units are neutral (being X 5 NH or O). As an example,
Figure 3 shows the turbidimetric study of the poly-TN at pH
ꢀ
9
. This polymer, at this pH, is soluble below 56 C and non-
soluble above this transition temperature. It should be noted
ꢀ
that a high-temperature ramp rate of 5 C/min was chosen
to minimize the influence of the hydrolysis of poly-T on the
O
O
measurement. The LCST of poly-T at pH 10 was not
observed because of its fast degradation at this pH.
Poly-TN exhibits slightly higher LCST values than poly-TO at
all pH values tested. This double sensitivity has been previ-
ously described for structurally related polymers such as
1
FIGURE 1 H NMR spectra of homopolymers poly-T
N
and
poly-T
O
.
N
FIGURE 2 Volumetric titration of poly-T .
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 2297–2305
2301

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
this region, the hydrolysis produces the progressive reduc-
tion of the -CH - signal centered at 2.63 ppm, while two
2
other -CH - peaks at 2.40 and 2.78 ppm are increasing corre-
2
latively. It should be noted that differences between the
spectrum of poly-TO in Figure 1 and the spectrum in Figure
4
(Time 0) are due to the different solvents used: while the
first one was obtained in CDCl , the second spectrum was
3
recorded in D O that produces the shift of some -CH - sig-
2
2
nals, probably due to the protonation of the amines.
FIGURE 3 LCST values of both polymers versus the pH. The
N
turbidimetry analysis of poly-T at pH 9 is shown as example.
2
2
5
6
poly-(N,N-diethylaminoethyl methacrylate), poly-DEAEM,
poly-(N,N-dimethylaminoethyl methacrylate), poly-DMAEM,
2
7
or poly-(ethylpyrrolidine methacrylate), poly-EPyM, which
also have aliphatic amines in their structure. Poly-DEAEM,
the polymer structurally closest to the TX structures, has a
ꢀ
25–29
pKa of 7.5 and LCST values of around 40 C at pH 7.
DLS analysis of poly-T solutions at pH 7 and 9, that is,
N
below and above the pKa, and at different temperatures,
exhibits temperature-dependent aggregation at pH 9 and a
nonaggregative behavior at pH 7 (Supporting Information
Fig. S3). At pH 9, the closer the temperature to LCST value
(
at temperatures lower than the LCST) the larger the size of
the aggregate which is in agreement with a correlative
increase in polymer–polymer interactions in this predomi-
nantly neutral state. At pH 7, where charged units are pre-
dominant, the Coulombic repulsion seems to prevent
aggregation.
Hydrolytic Behavior of Poly-T
O
Scheme included in Figure 4 shows the two types of esters
found in a T unit, the b-aminoester (1) and the methacrylic
O
that links the side chain to the macromolecular backbone
(
2). The anomalous enhanced sensitivity to hydrolysis of
Ester 1 compared to standard esters such as 2 has been
indeed reported to be related to the position of the amine in
5
the b position. Based on this sensitivity, self-inmolative poly
(
b-amino)-esters have been prepared and evaluated in gene
5
,6
therapy applications. In this work, the hydrolytic behavior
of poly-TO was studied at 37 C using different buffer solu-
ꢀ
tions by proton NMR in situ. Control experiments carried out
with poly-TN (see Experimental section) did not show any
degradation of this polymer after 7 days under the same
experimental conditions (37 C and the same buffers). The
degradation of poly-TO could be monitored in situ by analyz-
1
FIGURE
4
H
NMR spectra of three representative times,
ꢀ
0 (up), 24 (medium), and 48 h (bottom) during the NMR moni-
toring of the hydrolysis of poly-TO at pH 7. A scheme indicat-
ing the hydrolysis of a representative unit has been included.
ing the region between 2.2 and 2.8 ppm (see Fig. 4). Inside
2
302
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 2297–2305

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
h
FIGURE 5 Hydrolysis rate, f , versus the time at different pHs
obtained by in situ NMR monitoring as explained in the text.
The signals at 2.63 and 2.40 ppm, assigned to the -CH -COO-
2
of the polymer and to the -CH -COO- of free TEDETA deriva-
2
tive, respectively, have been used to determine the hydrolysis
rate which has been depicted in Figure 5 at different pHs. A
more detailed explanation of this calculation can be found in
the Experimental section.
As can be expected, there is a clear dependence of the
hydrolytic sensitivity to the environmental pH. The half-life
times of the kinetics change from hours at basic pH to days
at pH 7 and months at pH 5, due to the well-known sensitiv-
ity of the ester group to the basic media and to the higher
ester activation by the b-amino group in its neutral state.
FIGURE 7 (a) f-Potential and (b) average size measurements of
the polyplexes of PEI (used as control) and poly-T
N
measured
at different N/P ratios in distilled water.
mV. At 1/1 N/P ratios, the sample showed neutral
f-potential values, higher than the f-potential of the com-
plexed PEI, both being much higher than free DNA (around
Polyplex Formation and Characterization
Poly-TN was evaluated as a gene vector in preliminary in
vitro assays. Poly-TO was not studied due to its fast hydroly-
sis rate at pH 7 that causes the polyplex to prematurely
release the charged moieties responsible for the interaction
with DNA. The self-assembly of the new polymer Poly-TN
and pDNA was examined by agarose gel electrophoresis
2
70 mV). This higher f-potential of the polyplexes formed
with poly-T at the lower N/P ratios, compared with PEI,
indicates that poly-T interacts more efficiently than PEI. The
N
N
particle size measurements [Fig. 7(b)] showed that, at 2/1
N/P ratios or higher, these polyplexes had a mean average
size of around 150 nm, similar to the PEI derivatives.
Although in this PEI case, there is a trend toward smaller
diameters while increasing the N/P ratio. At the 1/1 ratio,
the polyplex formed with poly-TN exhibited the highest size,
around 300 nm. This may be related to the neutral
f-potential shown by this sample that favors aggregation.
(
Fig. 6). The DNA was condensed by the polymer at all the
N/P ratios prepared.
f-Potential measurements [Fig. 7(a)] were positive at 2/1
and higher N/P ratios, with values of approximately 140
Cell viability after transfection was evaluated by a calcein
assay. Figure 8(b) represents the percentage of relative via-
bility of cells in the presence of the homopolymer poly-T in
N
comparison with the positive control PEI. The data were nor-
malized with respect to those of the nontreated cells (100%
viability). Both polymers showed no significant toxicity (via-
bility higher than 80%) after transfection at low N/P ratios.
Nevertheless, at ratio 8/1, viability was significantly reduced
in the case of PEI, while the poly-T
cytocompatibility.
N
showed still good
FIGURE 6 Gel electrophoretic mobility of plasmid DNA in the
polyplexes with different N/P ratios. M 5 k-Hind III DNA.
The transfection efficiency of the polymer complexes was
studied using mouse Swiss-3T3 fibroblasts, and the results
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 2297–2305
2303

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
and to temperature, exhibiting LCST values, in a pH interval
between 7 and 9, above physiological temperature. This sta-
ble polymer has been shown to be comparably efficient in
terms of gene vector capability to commercial hyper-
branched PEI analogs in in vitro cell studies.
ACKNOWLEDGMENTS
The authors thank C. Garc ꢀı a for her help with the GPC measure-
ments. The authors affiliated with the Institute of Polymer Sci-
ence and Technology gratefully acknowledge support from Grant
MAT- 2010–20001. J. A. Redondo and R. Navarro gratefully
acknowledgment funding from the Spanish Research Council
(
CSIC) and European Social Fund (ESF) through the JAE pro-
gram. The authors affiliated with the Insititute of Biofunctional
Studies gratefully acknowledge the support of the Ministerio de
Econom ꢀı a y Competitividad (MINECO) (CONSOLIDER project
CSD2009-00088), Ministerio de Ciencia e Innovaci oꢀ n (MICINN)
(
SAF2011–27863), and Seventh Framework Programme of the
European Commission (EuroNanomed ERA-NET initiative,
REBONE project; PI10/02985FIS).
REFERENCES AND NOTES
1
M. E. Honigfort, W. J. Brittain, Macromolecules 2003, 36,
3
111–3114.
2
E. Juaristi, V. Soloshonok, Enantioselective Synthesis of
FIGURE 8 (a) Transfection efficiency and (b) Cell viability of
Beta-Amino Acids, John Wiley & Sons, Hoboken, New Jersey,
2
N
poly-T polyplexes 48 h after transfection on mouse Swiss-3T3
005.
fibroblasts, in comparison with the positive control PEI.
3
P. Teyssi eꢀ , J. Baran, P. Dubois, R. J eꢀ rome, J.-S. Wang, J. Yu,
obtained are shown in the Figure 8(a). Polyplexes prepared at
different polymer/plasmid DNA (N/P) molar ratios ranging
from 1/1 to 8/1 were used. Polyplexes were incubated with
the cells for 4 h in the absence of serum. Nontransfected cells
and cells treated with naked DNA were used as negative con-
trols. Polyplexes formed with hyperbranched PEI were used
Y. Yu, T. Zundel, Macromol. Symp. 1998, 132, 303–307.
4 B. D. Mather, K. Viswanathan, K. M. Miller, T. E. Long, Prog.
Polym. Sci. 2006, 31, 487–531.
5
Y.-b. Lim, Y. H. Choi, J.-s. Park, J. Am. Chem. Soc. 1999, 121,
5
633–5639.
6
D. M. Lynn, R. Langer, J. Am. Chem. Soc. 2000, 122, 10761–
1
0768.
as positive controls. The homopolymer poly-T
fection levels close to that of the positive PEI control at the
/1 ratio, while its transfection efficiency decreased dramati-
N
showed trans-
7
A. Metters, J. Hubbell, Biomacromolecules 2005, 6, 290–301.
8
P. Ferruti, J. Polym. Sci. Part A: Polym. Chem. 2013, 51,
4
2
319–2353.
cally at 8/1 ratio, where PEI showed higher transfection. It
should be noted that PEI at this 8/1 ratio showed a lower
9
D. Fournier, S. Pascual, V. Montembault, L. Fontaine, J.
Polym. Sci. Part A: Polym. Chem. 2006, 44, 5316–5328.
0 B. Grignard, C. J eꢀ r o^ me, C. Calberg, R. J eꢀ r o^ me, C.
Detrembleur, Eur. Polym. J. 2008, 44, 861–871.
cell viability than poly-T . In other words, and focusing on the
N
1
N
4/1 ratio, poly-T has shown to be comparable to PEI as a
gene vector on these in vitro studies. Besides, poly-TN seems
to be more cytocompatible than PEI.
1
1 Y. Shen, S. Zhu, Macromolecules 2001, 34, 8603–8609.
1
2 Y. Shen, S. Zhu, R. Pelton, Macromolecules 2001, 34, 5812–
5
818.
CONCLUSIONS
13 G. Bayramoglu, B. Karagoz, B. Altintas, M. Y. Arica, N.
Bicak, Bioprocess Biosyst. Eng. 2011, 34, 735–746.
The proposed selective Michael Addition of amines over
asymmetric methacrylate/acrylic (acrylate or acrylamide)
compounds has been successfully applied to the preparation
of two novel methacrylates bearing dendronic TEDETA moi-
eties. The addition over the acrylate leads to a labile ester
because of the b position of the amine group. This ester sen-
sitivity renders it to be hydrolysable at neutral or basic pHs.
1
4 G. Bayramoglu, B. Karagoz, M. Yilmaz, N. Bicak, M. Y. Arica,
Bioresour. Technol. 2011, 102, 3653–3661.
1
5 Y. Shen, H. Tang, S. Ding, Prog. Polym. Sci. 2004, 29, 1053–1078.
1
2
6 S. Satyapal, T. Filburn, J. Trela, J. Strange, Energy Fuels
001, 15, 250–255.
1
2
7 R. Serna-Guerrero, Y. Belmabkhout, A. Sayari, Adsorption
010, 16, 567–575.
1
The hydrolysis behavior has been studied by H NMR in situ
1
8 R. Kircheis, L. Wightman, E. Wagner, Adv. Drug Delivery
with the t1/2 of the ester ranging from months at pH 5 to
hours at pH 9. The polymers display double sensitivity
behavior to pH (related to the protonation of the amines)
Rev. 2001, 53, 341–358.
1
9 D. Demirgoz, R. Navarro, M. Perez, H. Reinecke, A. Gallardo,
€
ꢀ
J. Appl. Polym. Sci. 2010, 115, 896–900.
2
304
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 2297–2305

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
2
0 X. Ma, J. Tang, Y. Shen, M. Fan, H. Tang, M. Radosz, J. Am.
Chem. Soc. 2009, 131, 14795–14803.
1 H. Weickmann, M. Gurr, O. Meincke, R. Thomann, R.
M u€ lhaupt, Adv. Funct. Mater. 2010, 20, 1778–1786.
25 A. Schmalz, M. Hanisch, H. Schmalz, A. H. M u€ ller, Polymer
2010, 51, 1213–1217.
2
26 S. Guo, Y. Qiao, W. Wang, H. He, L. Deng, J. Xing, J. Xu,
X.-J. Liang, A. Dong, J. Mater. Chem. 2010, 20, 6935–6941.
2
2
2 Z. Xu, B. He, J. Shen, W. Yang, M. Yin. Chem. Commun.
013, 49, 3646–3648.
27 N. Gonz aꢀ lez, C. Elvira, J. S. Rom aꢀ n, Macromolecules 2005,
38, 9298–9303.
2
3 G. Y. N. Chan, D. H. Solomon, M. G. Looney, S.
28 S. Liu, J. V. Weaver, Y. Tang, N. C. Billingham, S. P. Armes,
K. Tribe, Macromolecules 2002, 35, 6121–6131.
Veluayitham, Aust. J. Chem. 1998, 51, 31–36.
4 M. R. Aguilar, A. Gallardo, M. d. M. Fern aꢀ ndez, J. S. Rom aꢀ n,
2
29 Y. Shen, X. Ma, B. Zhang, Z. Zhou, Q. Sun, E. Jin, M. Sui, J.
Tang, J. Wang, M. Fan, Chem. Eur. J. 2011, 17, 5319–5326.
Macromolecules 2002, 35, 2036–2041.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 2297–2305
2305