Antioxidant properties of 2-hydroxyethyl methacrylate-based copolymers with incorporated sterically hindered amine

Article abstract of DOI:10.1021/acs.biomac.5b00599

A series of model linear copolymers of 2-hydroxyethyl methacrylate (HEMA) and a sterically hindered amine derivative [N-(2,2,6,6-tetramethyl-piperidin-4-yl)methacrylamide (HAS)] were synthesized and characterized. Scavenging activities of the copolymers against reactive oxygen species (peroxyl and hydroxyl radicals) and 2,2-diphenyl-1-picrylhydrazyl (DPPH) radicals were determined. It was found that copolymers with medium HAS content (3.5-4.0 mol %) were better scavengers than copolymers with lower and higher HAS content and also than polyHEMA and polyHAS homopolymers and the HAS monomer. Importantly, these copolymers compared favorably even to established low-molecular weight antioxidant standards (BHA and dexpanthenol). Monomer reactivity ratios were determined, and the microstructure of the copolymers was assessed. Subsequently, cross-linked copolymers in the powder and film forms with optimal HAS content were synthesized. Their scavenging activities against the three types of radicals were determined, revealing that these hydrogels are potent scavengers of reactive oxygen species.

Full text of DOI:10.1021/acs.biomac.5b00599

Article
pubs.acs.org/Biomac
Antioxidant Properties of 2‑Hydroxyethyl Methacrylate-Based
Copolymers with Incorporated Sterically Hindered Amine
L. Polakova,* V. Raus, L. Kostka, A. Braunova, J. Pilar, V. Lobaz, J. Panek, and Z. Sedlakova
́ ́ ́ ̌ ́ ́ ́
Institute of Macromolecular Chemistry, Academy of Sciences of the Czech Republic, Heyrovsky Sq. 2, 162 06 Prague 6, Czech
Republic
*
S Supporting Information
ABSTRACT: A series of model linear copolymers of 2-
hydroxyethyl methacrylate (HEMA) and a sterically hindered
amine derivative [N-(2,2,6,6-tetramethyl-piperidin-4-yl)-
methacrylamide (HAS)] were synthesized and characterized.
Scavenging activities of the copolymers against reactive oxygen
species (peroxyl and hydroxyl radicals) and 2,2-diphenyl-1-
picrylhydrazyl (DPPH) radicals were determined. It was found
that copolymers with medium HAS content (3.5−4.0 mol %)
were better scavengers than copolymers with lower and higher
HAS content and also than polyHEMA and polyHAS homopolymers and the HAS monomer. Importantly, these copolymers
compared favorably even to established low-molecular weight antioxidant standards (BHA and dexpanthenol). Monomer
reactivity ratios were determined, and the microstructure of the copolymers was assessed. Subsequently, cross-linked copolymers
in the powder and film forms with optimal HAS content were synthesized. Their scavenging activities against the three types of
radicals were determined, revealing that these hydrogels are potent scavengers of reactive oxygen species.
1
6
7
INTRODUCTION
systems, and carrier materials for wound healing. To
introduce specific properties into HEMA-containing polymeric
material, HEMA can be copolymerized with other vinylic
monomers, or hydroxyl groups in polyHEMA can be
chemically modified. For example, polyHEMA with antioxidant
properties has been recently prepared by HEMA polymer-
ization in the presence of natural antioxidant quercetin; the
synthesized polymers exhibited scavenging activities against 2,2-
■
Reactive oxygen species (ROS) such as peroxyl, hydroxyl, or
superoxide anion radicals play a crucial role in the wound
1
−4
healing process.
When a tissue is injured, enzymes in
phagocytic cells (neutrophils and macrophages) responsible for
the formation of ROS-type radicals are often activated. Under
certain circumstances, an undesirable cumulative increase in the
concentration of these radicals in the tissue can be observed. An
excessive concentration of ROS then causes oxidative damage
to healthy cells, resulting in the development of an
inflammatory reaction, and thereby impairing the wound
1
7
diphenyl-1-picrylhydrazyl (DPPH) and oxygen radicals.
Copolymers of HEMA with the antioxidant α-tocopherol
were successfully synthesized by Roman
the incorporation of the antioxidant into the polymer structure
effectively increased the oxidative stability of the polymer.
Recently, clinically relevant hybrid macromolecular antioxidants
18,19
́
et al.,
showing that
2
−5
healing process.
For more than six decades, hydrogels have been attracting
attention in many disciplines because of their convenient water
swellability, mechanical properties, and biocompatibility.
2
0,21
6
−8
bearing sterically hindered phenols have been synthesized.
Hindered amine stabilizers (HAS) rank among the synthetic
antioxidants capable of reacting with ROS. The reaction results
in the formation of stable and nontoxic nitroxides or
hydroxylamines. Similar behavior is expected from polymer-
izable derivatives of sterically hindered amines and their
copolymers with HEMA. Such copolymers have been
The shape-retentive polymeric network allows hydrogels to
preserve their viscoelastic properties despite a high water
9
content, which makes them attractive for certain medical
22
applications. Examples of hydrogel-based materials with a
potential use in the biomedicinal field include networks based
on poly(2-hydroxyethyl methacrylate) (polyHEMA), copoly-
10
23
́
synthesized by Kejlova et al., who proved that the cross-
mers of HEMA with N,N-dimethylaminopropyl methacryla-
1
1
12
linked polyHEMA with incorporated HAS monomer units is
highly biocompatible and exhibits good skin tolerance.
However, the antioxidant properties of the copolymer have
not been studied in detail.
mide, oligo(ethylene glycol) (meth)acrylate copolymers,
and copolymers of N,N-dimethylaminopropyl methacrylamide
1
3
14
with acrylic or itaconic acid.
PolyHEMA-based polymers are prominent members of the
hydrogel family. Their good mechanical properties, desirable
swelling behavior in a water-based environment, and high
biocompatibility facilitate their use as biomedicinal materials,
Received: May 4, 2015
Revised: August 1, 2015
1
0
15
e.g., contact lenses, dental adhesives, drug delivery
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.biomac.5b00599
Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules
Article
This study is a part of a broader effort to develop and
characterize hydrogel-based wound dressing materials with
antioxidant properties. Here, a series of linear HEMA/HAS
copolymers with different HAS contents was synthesized (i) to
confirm and quantify the scavenging efficiency of the HEMA/
HAS copolymers against ROS and (ii) to determine the
optimal content of HAS units in the copolymers with regard to
their scavenging activity. Antioxidant properties of the
gradient-on is a convolution of the spectrum measured at the gradient-
off and the concentration profile of nitroxides inside the sample. To
record the concentration profile, the spectrum measured at the
gradient-off was deconvoluted out of the projection measured at the
gradient-on. Nitroxide concentrations were normalized with respect
to the spectrometer settings and to the mass of the swollen strip
calculated from the known swelling behavior of the film. Absolute
nitroxide concentrations were determined using a calibration
procedure based on a standard sample.
2
4
(
co)polymers were evaluated from DPPH, peroxyl, and
Synthesis of the HAS Monomer. The monomer was prepared by
N-acylation of 4-amino-2,2,6,6-tetramethylpiperidine by a mixed
anhydride formed via a reaction of methacrylic acid and ethyl
hydroxyl radical scavenging assays. Afterward, the optimal
content of HAS found in this model study was applied in the
synthesis of cross-linked copolymers in the application-friendly
forms of powder and film. Scavenging activities as well as
swelling properties of the cross-linked copolymers were then
evaluated.
25
chloroformate, according to a previously published procedure: yield
1
8
1
1
1
2%; mp 118.0−118.5 °C; H NMR (300.1 MHz, MeOD) δ 5.62 (s,
H, a), 5.33 (s, 1H, a′), 4.25 (m, 1H, c), 1.92 (s, 3H, b), 1.75 (dd, J =
2.6, 3.6 Hz, 2H, d′), 1.24 (s, 6H, e), 1.17 (d, 1H, d), 1.13 (s, 6H, e),
13
.09 (d, 1H, d) (hydrogen assignment according to Figure 1a);
C
EXPERIMENTAL SECTION
■
Materials. 2-Hydroxyethyl methacrylate (HEMA; Fluka) was
distilled prior to use. 2,2′-Azobis(2-methylpropionitrile) (AIBN;
Fluka) was recrystallized from methanol. Ethylene glycol dimethacry-
late (EGDMA, 98%; Sigma-Aldrich), Darocur 1173 (97%; Sigma-
Aldrich), poly(ethylene glycol) (PEG; Macrogolum 300, PENTA),
potassium dihydrogen phosphate (≥99.9%; Lach-Ner), disodium
hydrogen phosphate dodecahydrate (≥98%; Lach-Ner), triethylamine
(
TEA, p.a.; PENTA), 4-amino-2,2,6,6-tetramethylpiperidine (p.a.;
Merck), ethyl chloroformate (≥98%, Merck), 2,2′-azobis(2-methyl-
propionamidine) dihydrochloride (AAPH, ≥97%; Sigma-Aldrich), 2,2-
diphenyl-1-picrylhydrazyl (DPPH; Sigma-Aldrich), fluorescein (p.a.;
Fluka), a hydrogen peroxide solution [30% (w/w); Lach-Ner], sodium
Figure 1. Structure of the (a) HAS monomer and (b) HEMA/HAS
copolymers.
tungstate dihydrate (≥99%; Riedel-de Haen
̈
), ascorbic acid (≥99%;
NMR (150 MHz, MeOD) δ 170.7, 141.7, 120.0, 52.4, 45.1, 43.6, 34.3,
28.0, 18.9. Elemental analysis. Theoretical: H, 10.8%; C, 69.6%; N,
12.5%; O, 7.1%. Found: H, 10.8%; C, 69.5%; N, 12.3%; O, 7.4%.
Synthesis of Linear HEMA/HAS (Co)polymers. In a 100 mL
flask equipped with a magnetic stirrer, HAS (1.20 g, 5.3 mmol),
HEMA (5.0 g, 38.4 mmol), and AIBN (359 mg, 2.2 mmol, 5 mol %)
were dissolved in toluene (73 mL). The solution was bubbled with
nitrogen for 10 min and sealed. Polymerization was conducted at 70
°C. After 24 h, the liquids were evaporated, and the residuum was
reprecipitated twice from methanol to acetone and dried to a constant
weight. Other HEMA/HAS copolymers as well as homopolymers of
HEMA and HAS were prepared under similar conditions.
Synthesis of Cross-Linked HEMA/HAS Copolymers. The
cross-linked copolymer of HEMA/HAS in the form of a powder
(HHP) was synthesized by a slightly modified procedure. In a 100 mL
flask with a magnetic stirrer, HAS (0.31 g, 1.4 mmol), HEMA (5.0 g,
38.4 mmol), EGDMA (40 mg, 0.2 mmol, 0.5 mol %), and AIBN (330
mg, 2.0 mmol, 5 mol %) were dissolved in toluene (63 mL). The
reaction mixture was thoroughly bubbled with nitrogen and sealed.
Polymerization was conducted at 70 °C. After 24 h, the precipitate was
separated by filtration, washed several times with hot toluene, and
dried in vacuum to a constant weight. Conversion of 95%.
Fluka), dexpanthenol (pure Ph. Eur., Fluka Analytical), and a mixture
of 2-tert-butyl-4-hydroxyanisole and 3-tert-butyl-4-hydroxyanisole
(
BHA, 97%; Acros Organics) were used as received.
Methods. H nuclear magnetic resonance (NMR) and 13C NMR
1
spectra were recorded on the Bruker Avance DPX 300 spectrometer
300.1 MHz; MeOD as a solvent, 25 °C) and the Bruker Avance III
00 spectrometer (150 MHz; MeOD as a solvent, 25 °C), respectively.
(
6
Infrared (IR) spectra of the powdered samples in KBr pellets were
recorded on the Thermo Nicolet NEXUS 870 FTIR spectrometer.
Number and weight averages of molecular weights of the prepared
linear (co)polymer were determined by size-exclusion chromatography
(
SEC) on a Shimadzu high-performance liquid chromatography
system equipped with a TSK 4000 SW column (Tosoh Bioscience)
and UV, RI, and multiangle light-scattering (MALLS) detectors. A
mixture (80/20) of methanol and acetate buffer (0.3 M, pH 6.5) was
used as an eluent at a flow rate of 0.5 mL/min.
A PerkinElmer Lambda 20 UV−vis spectrometer with a halogen
lamp and JASCO spectrofluorometer (model FP-6 200) using a water-
cooled Peltier thermostated single-cell holder with a stirrer (model
ETC-272T, at 37 °C) were employed to determine scavenging
activities of the synthesized polymers. The fluorescence intensity was
recorded at λex = 485 nm and λem = 519 nm.
The HEMA/HAS copolymer in the form of a thin, transparent film
(HHF) was prepared by photoinitiated free radical polymerization. In
a typical experiment, a stock polymerization solution was prepared by
mixing HEMA (10.67 mL, 88 mmol), HAS (0.72 g, 3.21 mmol),
EGDMA (86.7 μL, 0.458 mmol), Darocur 1173 (228 μL, 1.49 mmol),
and water (8.2 mL) under argon. Then, 8 mL of the stock solution was
sampled to a round flat Teflon mold with a diameter of 11 cm, and the
mold was placed in a polymerization reactor having a form of a box
equipped with five 4 W UV-A lamps (Narva LT 4W/073 BLB,
emission maximum at 365 nm; distance from the sample, ∼10 cm).
The polymerization mixture was irradiated for 15 min under a constant
argon flow through the reactor, after which the prepared film was
removed from the mold and immersed in distilled water for 12 days
with a daily water exchange. The film was then dried in vacuum at 40
°C to a constant weight.
Concentration profiles of nitroxide radicals in the oxidized film
HHF) were determined by an electron spin resonance imaging
ESRI) technique using a Bruker ELEXSYS E 540 X-band
(
(
spectrometer equipped with a pair of figure 8-shaped Lewis gradient
coils that can produce a vertical magnetic field gradient (Gmax ∼ 320
G/cm) perpendicular to the external magnetic field. ESR spectra
without a magnetic field gradient and projections at a magnetic field
gradient of 280 G/cm were measured at 298 K at a microwave power
output of 6 mW with 100 kHz magnetic modulation. A 4 mm × 4 mm
rectangle was cut out of the oxidized film and divided into four 1 mm
wide strips. Each strip was wiped dry and placed into a quartz tube
positioned vertically in the cavity of the spectrometer, parallel to the
direction of the magnetic field gradient. Concentration profiles of
nitroxides in the samples along the direction of the magnetic field
gradient were determined by application of a suitable deconvolution
procedure to one-dimensional ESRI data. Projection measured at the
Determination of Antioxidant Properties of the (Co)-
2
6−28
29
30,31
polymers. The peroxyl,
hydroxyl, and DPPH radical
B
DOI: 10.1021/acs.biomac.5b00599
Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules
Article
Table 1. Main Characteristics of the Synthesized Linear Polymers
a
b
Mw × 10−4
c
c
d
d
d
sample
FHAS
f HAS
M /Mn
w
conversion (%)
nHAS
nHEMA
nHAS + nHEMA
polyHEMA
HH1
0.0
1.8
0.0
1.5
31.4
27.3
27.6
28.7
30.2
36.7
23.2
2.08
2.20
2.32
2.41
2.40
2.35
2.09
95
98
95
97
89
75
0
2413
2007
1914
1962
1798
1575
2413
2060
2034
2103
2101
2297
1034
53
120
HH2
4.0
3.5
HH3
6.2
4.0
141
HH4
20.0
40.0
100.0
8.9
303
HH5
21.0
100.0
722
polyHAS
51
1034
0
a
b
c
d
Mole percent of HAS in the feed. Mole percent of incorporated HAS, from elemental analysis. From SEC-MALLS. Number of monomeric units
calculated from M considering the copolymer composition (weight percent).
w
assays were performed according to published literature protocols. The
detailed experimental procedures can be found in the Supporting
Information.
Determination of the Degree of HEMA/HAS Film Swelling.
The film was immersed in a great excess of distilled water. At
predetermined time intervals, it was removed and water uptake was
assessed gravimetrically. The swelling experiment was repeated in
triplicate. The degree of swelling was calculated using the following
increased slightly with an increasing HAS unit content, which
explains the somewhat higher DP of the HH5 copolymer.
Dispersity values were greater than 2, which is typical for
32
polymers prepared by free radical polymerization.
To assess the monomeric sequence patterns in the HEMA/
HAS copolymers, we determined the monomer reactivity ratios
33
at low conversions. For the calculation, Finemann−Ross and
3
4
formula:
Kelen−Tu
̈
do
̈
s
methods were employed, giving very similar
values: r
= 2.24 and rHAS = 0.11, respectively, and rHEMA
=
HEMA
degree of swelling (%) = [(m − m )/m ] × 100
(1)
t
0
0
2
.21 and rHAS = 0.10, respectively. The r
r
product gives
HEMA HAS
where m and m are weights of the swollen film at time t and of the
dry film, respectively.
a value of 0.22, suggesting that the process had predominantly
t
0
35
the character of alternating radical copolymerization. This
finding indicates that the HAS units should be relatively
uniformly distributed in the copolymer chains. From the
monomer reactivity ratios, we calculated the conditional
probabilities of monomeric dyad formation during copoly-
Copolymer Oxidation for ESR/ESRI Experiments. Typically, a
piece of the HHF sample (263 mg) was placed into a 15 mL vial
containing Na WO ·2H O (4.8 mg) dissolved in water (2 mL). After
2
4
2
addition of 30% H O (10 mL), the reaction mixture was kept for 5
2
2
days without being stirred at room temperature. Then, the swollen and
slightly yellow membrane was washed several times with water. The
HH2 linear copolymer was oxidized in the same way.
̈
̈
merization (rHEMA and rHAS obtained by the Kelen−Tudos
35
method were used for the calculation). The results, listed in
Table 2, show that the probability of HAS−HAS dyad
RESULTS AND DISCUSSION
■
Table 2. Conditional Probabilities of Dyad Formation
during the HEMA/HAS Copolymerization
Linear Polymers. PolyHEMA and polyHAS homopol-
ymers and linear HEMA/HAS copolymers with different
contents of HAS were synthesized by free radical polymer-
ization. The structures of the HAS monomer and the resulting
HEMA/HAS copolymer are shown in Figure 1. The purity of
the prepared copolymers was verified by NMR spectroscopy; in
particular, the absence of unreacted monomers was checked.
Because of overlapping signals of HAS and HEMA, the content
b
b
b
b
pHEMA,HEMA
pHEMA,HAS
pHAS,HEMA
pHAS,HAS
(%)
a
copolymer f HAS
(%)
(%)
(%)
HH1
HH2
HH3
HH4
HH5
1.5
99.3
98.1
97.1
89.8
76.7
0.7
1.9
99.8
99.6
99.3
97.6
93.7
0.2
0.4
0.7
2.4
6.3
3.5
4.0
2.9
8.9
10.2
23.3
1
of HAS units in copolymers could not be determined by H
NMR spectroscopy and was calculated from elemental analysis
instead.
21.0
a
b
Mole percent of HAS in the copolymer. Conditional probability of
forming the respective dyad in the copolymer chain.
As can be seen from Table 1, the amount of HAS units in the
copolymers increased steadily from 1.5 to 21.0 mol %, in
accordance with the increasing HAS:HEMA ratio in the
polymerization feed. The incorporation of HAS units into the
copolymer was less efficient for the feeds richer in HAS,
probably because of the lower reactivity of the methacrylamide-
type comonomer. This was also reflected in monomer
conversions that were noticeably lower for HAS-rich feeds. If
formation increased with increasing HAS unit content in the
copolymers. This means that whereas in copolymers HH1−
HH3 almost all the HAS units are isolated by HEMA units, in
the HAS-rich copolymers (HH4 and HH5), an appreciable
amount of HAS−HAS dyads can be expected. Further, we
calculated the mole fractions of different lengths of HEMA
we focus on the M of the copolymers obtained from SEC−
w
MALLS, the values were in the range of approximately
sequences [(HEMA) ]. For the extreme cases of copolymers
n
2
70000−370000 and increased toward the samples with higher
HH1 and HH5, these data are visualized in Figure 2. As clearly
shown, the distribution of sequence lengths is vastly different,
with the HH5 copolymer having a much higher fraction of
shorter sequences. To better illustrate this difference, the mole
fraction values for n = 1−10 and n > 10 are listed in Table S1.
As follows from the data, while the HH5 copolymer contains
only ∼7% of longer (n > 10) HEMA sequences, for the HH1
copolymer the share of such longer sequences increases to
>93%. In Figure 3, we attempt to schematically depict the
HAS contents. This is caused by the higher molar weight of the
HAS monomeric unit because the degrees of polymerization
(
DPs), calculated from the M and copolymer composition,
w
were very similar for all the copolymers (see the last column of
Table 1). The similarity of DP can be ascribed to the way in
which the polymerization was conducted, i.e., the products
precipitated out of the toluene solution during polymerization.
We observed that the solubility of copolymers in toluene
C
DOI: 10.1021/acs.biomac.5b00599
Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules
Article
The efficiency of copolymers in radical scavenging was tested
in three model assays employing different types of radicals: two
ROS (peroxyl and hydroxyl radicals) and DPPH radicals. From
these three types of radicals, the OH radicals are the most
reactive, followed by ROO radicals and finally DPPH.
The ROO radicals were generated in vitro via a standard
procedure utilizing thermal decomposition of AAPH at 37 °C.
•
The scavenging activity against ROO was then measured by
monitoring the fluorescence intensity decrease resulting from
•
26−28
the fluorescein oxidation by ROO .
Figure 4a presents the
plots of scavenging activities of the HAS monomer and of the
prepared (co)polymers against their concentration in the
reaction mixture. This plot illustrates the performance of the
studied materials as a whole by taking the same amount of the
sample regardless of the HAS unit content.
As can be seen, neat polyHEMA exhibited practically no
scavenging activity against peroxyl radicals; a similar result was
17
observed by Cirillo et al. The HAS monomer and polyHAS
showed similar, relatively good scavenging activity, indicating
that the antioxidant activity of HAS was retained after
homopolymerization. Scavenging activities of HEMA/HAS
copolymers were significantly dependent on the copolymer
composition. The HH1 copolymer (1.5 mol % of HAS units)
exhibited poor activity at all the used concentrations; the
remaining copolymers fared considerably better, even though
their performance was noticeably concentration-dependent.
Whereas at the highest sample concentration (2.8 mg/mL)
almost the same activity value (∼80%) was observed for all the
copolymers, at lower concentrations HH2 (3.5 mol % of HAS
units) performed much better than the other samples. Note
that the overall performance of the HAS monomer and
polyHAS was comparable to that of the HH3 copolymer.
The reasons for the observed behavior are not immediately
clear; most probably, several different factors play a role. The
higher scavenging activity of some of the HEMA/HAS
copolymers compared to those of the HAS monomer and
polyHAS indicates that the presence of HEMA units is
essential. Quite obviously, a certain amount of HAS must be
present to achieve the antioxidant activity. However, the data
Figure 2. Sequence length distribution for HEMA units in copolymers
HH1 and HH5 (n = 1−35).
sequence-related differences between HH1 and HH5 copoly-
mers.
Figure 3. Visualization of main differences in monomer (HEMA,
green; HAS, red) sequence structure between copolymers HH1 (left)
and HH5 (right).
•
Figure 4. Dependence of ROO scavenging activity on (a) sample concentration and (b) HAS unit concentration. Average values (±SD) from four
independent experiments are presented.
D
DOI: 10.1021/acs.biomac.5b00599
Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules
Article
•
Figure 5. Dependence of HO scavenging activity on (a) sample concentration and (b) HAS unit concentration. Average values (±SD) from three
independent experiments are presented.
suggest that an overly high HAS content can be detrimental.
Instead, a certain optimal HAS content probably exists, and
from our samples, HH2 appears to be closest to this optimal
composition. These effects are somehow better visualized in
Figure 4b, where the scavenging activities are plotted against
the actual HAS unit concentration in the tested samples. These
concentrations were obtained by multiplying the original
sample concentrations by the respective weight ratios of HAS
units in the copolymers. In this way, the individual points are
normalized for the HAS quantity, and the qualitative differences
among samples become more obvious. Considering the trends
shown in Figure 4b, it is clear that the HAS units
copolymerized with HEMA are much better scavengers than
the HAS monomer and homopolymer. Once again, copolymer
HH2 acts as the strongest scavenger, achieving very high
scavenging activity of almost 90% at a low HAS unit
concentration (0.165 mg/mL).
Figure 5a shows how scavenging activities of the tested
samples depend on sample concentration in the reaction
mixture. As in literature reports, neat polyHEMA exhibited a
very poor ability to scavenge hydroxyl radicals. In contrast to
the behavior observed for peroxyl radicals, the HAS monomer
and homopolymer performed vastly differently here.
Whereas polyHAS showed quite negligible scavenging
activity, the HAS monomer was identified as the best scavenger.
It is noteworthy that the HAS monomer molecules contain a
double bond that can potentially react with hydroxyl radicals,
which could artificially inflate the scavenging activity. Such
behavior could be enhanced by the high reactivity of OH
radicals and also by their high instantaneous concentration
17
(
compared to the ROO radical assay). Scavenging activity
values of HEMA/HAS copolymers were only slightly depend-
ent on the HAS content with an optimum at 3.5−4.0 mol %.
The “qualitative” performance of the HAS units in different
samples can be evaluated using the plots in Figure 5b. Besides
the already discussed HAS monomer behavior, the trends are
similar as for the ROO radicals with the exception of copolymer
HH1, for which the HAS units showed much higher scavenging
activity even at the very low HAS concentration.
To compare the antioxidant behavior of the HEMA/HAS
copolymers to that of established antioxidant compounds, we
determined the half-maximal inhibitory concentration (IC )
50
values of the HAS units in copolymer HH2 and of the BHA
standard. The obtained values of 12.5 ± 0.3 and 5.40 ± 0.14
μg/mL, respectively, show that the tested copolymer compares
favorably to the standard.
As previously, the scavenging efficiency of copolymer HH2
was compared to a suitable standard in terms of the IC values.
5
0
Here, dexpanthenol was used as a standard instead of BHA
because of solubility reasons. Also in this case, the obtained
values were very close to each other (0.25 ± 0.01 and 0.29 ±
The second ROS type studied in terms of scavenging activity
was the hydroxyl radical. Hydroxyl radicals are the most
reactive species of the ROS family, and their overproduction
during an inflammatory process might cause considerable
0
.01 mg/mL, respectively).
DPPH is a commercially available stable nitrogen radical that
2
−5
oxidative damage to the affected tissue.
In this study, the
is widely used to assess the in vitro scavenging activity of both
31,36
hydroxyl radicals were generated in vitro by redox reactions of a
transition metal ion with hydrogen peroxide. The formed free
radicals are able to oxidize fluorescein, resulting in a decrease in
the fluorescence intensity at 519 nm. Typically, Fe(II) is
employed for the reaction with peroxide (Fenton reaction);
however, the oxidized form of HAS (nitroxide) has been shown
to oxidize Fe(II) and thus disrupt the system of redox
low-molecular weight compounds and polymers.
DPPH method consists of the direct reaction of an antioxidant
hindered amine in our case) with the DPPH radical, resulting
The
(
in a decrease in absorbance at 527 nm in the UV−vis spectrum.
In contrast to the peroxyl and hydroxyl assays, the DPPH
radicals are not generated continuously; instead, an arbitrary
amount of radicals is introduced into the reaction mixture at the
beginning of the reaction. Advantageously, this arrangement
allows us to set up an experiment with a predetermined
HAS:DPPH molar ratio. Figure 6 shows the relative decrease in
3
3
reactions. For this reason, an established alternative system
utilizing sodium tungstate to decompose the hydrogen peroxide
2
9
was used here.
E
DOI: 10.1021/acs.biomac.5b00599
Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules
Article
Figure 7. Time-dependent relative decrease in absorbance at 527 nm
of a DPPH solution in the presence of the tested samples.
Figure 6. Time-dependent relative decrease in absorbance at 527 nm
of a DPPH solution in the presence of the tested samples. The initial
HAS:DPPH molar ratio was 10:1. Copolymer HH4 was not included
because of its problematic solubility in the used medium.
Let us now discuss the rather unexpected scavenging
behavior of the tested samples in the three performed assays.
The study of monomeric sequences presented above indicates
that the structure of the prepared HEMA/HAS copolymers
varies significantly with their changing composition. It is very
likely that the monomeric unit sequence patterns are somehow
responsible for the differences in the scavenging capability of
individual copolymers, but the exact chain of events leading to
the observed results is at present unknown. When considering
the possible reasons, we should note that the respective
homopolymers show different solubilities in water, with
polyHAS being quite soluble and polyHEMA practically
insoluble. Consequently, it can be reasonably expected that
the studied copolymers will exhibit different conformational
behavior in water-rich environments used here for scavenging
assays. For instance, the copolymers can differ in coil
absorbance at 527 nm that depends on the reaction time for the
analyzed samples; the molar ratio of HAS units to DPPH was
1
0:1. The neat polyHEMA showed no relevant scavenging
17
ability, which is in accordance with literature reports.
Interestingly, the HAS monomer exhibited a very poor ability
to scavenge DPPH radicals; a similar result was also observed
for polyHAS.
Conversely, the HEMA/HAS copolymers showed consid-
erably better scavenging properties. Again, the content of HAS
units in the copolymer played a crucial role in HAS reactivity.
Whereas copolymers HH1 and HH5 (1.5 and 21.0 mol % of
HAS units, respectively) acted as relatively weak scavengers,
copolymers HH2 and HH3 (3.5 and 4.0 mol % of HAS units,
respectively) showed a strong ability to quench DPPH radical,
which was illustrated by halving of the sample absorbance
within only 15 min. Copolymer HH2 was identified as the
strongest DPPH scavenger. Clearly, HAS units need to be
copolymerized with HEMA to achieve this effect as is illustrated
by the low scavenging efficiency of the sample where a simple
polyHEMA homopolymer mixture with the HAS monomer was
used (see Figure 6). Similar behavior has been previously
38
dimensions and the related coil density, and also in their
tendency to aggregate or form various self-assembled structures
3
9,40
(e.g., unimolecular micelles,
flower-type core−shell nano-
41
structures, etc.). Clearly, these factors could account for the
different observed qualitative behavior of HAS units; i.e., they
could influence the accessibility of HAS units to the radical
species.
The solution behavior of the copolymers can be rather
complex, because of the complexity of the used solvents that,
besides water, also contain other components (PEG, DMF, and
salts). In addition, the solvent composition had to be different
for each of the assays presented here. In addition, the three
types of radicals studied here differed not only in their reactivity
but also in their bulkiness. This aspect is important mainly for
the substantially bulkier DPPH radical. The local steric
hindrance at the amine site, hampering the hydrogen radical
transfer between HAS and DPPH, then could be the reason for
the weak observed polyHAS and especially HAS monomer
scavenging ability in the DPPH assay. Such an assumption
necessarily points to an important role of HEMA units that not
only might play a role of a “dummy” comonomer but also could
take part in the actual scavenging mechanism. The nature of
this interaction is presently unknown to us though.
37
observed for an analogous DPPH-based system.
In Figure 7, the scavenging performance of the HH2
copolymer at lower HAS:DPPH ratios (5:1 and 1:1) is
compared to that at the 10:1 ratio used above. Interestingly,
HH2 is an efficient DPPH quencher even at low loadings. On
the other hand, even a 10-fold increase in polyHAS
concentration (100:1 HAS:DPPH) did not lead to a substantial
increase in scavenging efficiency, which was still much lower
than for HH2 at the lowest loading (1:1 HAS:DPPH). Finally,
it appears that the specific HEMA unit structure is vital for its
positive role in the scavenging mechanism because when
HEMA was replaced with N-(2-hydroxypropyl) methacryla-
mide (HPMA) in the HH2 copolymer, the scavenging
efficiency was significantly suppressed (Figure 7; the procedure
for synthesis of the HPMA/HAS copolymer can be found in
the Supporting Information).
Cross-Linked Copolymers. Evaluation of scavenging
activities of the linear HEMA/HAS copolymers confirmed
that (i) they show considerable antioxidant qualities and (ii)
the optimal molar composition (from the ones tested) of the
F
DOI: 10.1021/acs.biomac.5b00599
Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules
Article
copolymer is 96.5:3.5. Hence, this ratio was further applied in
the synthesis of cross-linked HEMA/HAS copolymers. Cross-
linking of HEMA-based polymers leads to hydrogel materials
that are mechanically stable and swellable, and thus suitable for
the intended wound dressing applications. In this study, we
considered two basic forms of the copolymer, a fine powder
(denoted HHP) and a thin film (HHF), yielding a gel and a
flexible foil, respectively, after swelling in a suitable solvent.
Scavenging activities of the copolymers against peroxyl,
hydroxyl, and DPPH radicals were determined. Compared to
linear copolymers, the situation with cross-linked copolymers is
more complicated. It can be expected that diffusion of low-
molecular weight species into the copolymer network will play
an important role in the radical scavenging process. It is clear
that limitations of this kind will be more prominent for the film
because of its lower specific surface. For this reason, a swelling
experiment with HHF was first conducted. As follows from
Figure 8, the film reached the swelling equilibrium in water
after ∼8 h.
•
Figure 9. Dependence of ROO scavenging activity of cross-linked
samples HHP and HHF on sample concentration and reaction time.
Average values (±SD) from three independent experiments are
presented.
of radical formation and higher reactivity of the radicals in this
system. This would also influence the time dependence of the
scavenging activity. As previously, HHP showed scavenging
activity superior to that of HHF, presumably because of the
diffusion effects.
Figure 8. Time dependence of the degree of swelling for the film
sample (HHF).
Figure 11 shows the plot of the relative decrease of
absorbance at 527 nm against reaction time for the cross-
linked samples reacting with DPPH radicals. Besides the HHP
and HHF samples, the data for the HH2 linear copolymer are
shown for comparison. The powder sample achieved almost the
same result as the linear, non-cross-linked sample, albeit over a
longer time. On the other hand, the film sample exhibited very
low activity against DPPH radicals in the time frame of the
experiment. This observation can be related to the increased
bulkiness of the DPPH radicals, probably accentuating the
diffusion effects discussed previously.
For the intended wound dressing application, the scavenger
(HAS) should be uniformly distributed in the copolymeric
structure. This can be reasonably expected for the copolymer in
the form of a fine powder (gel after swelling); to show that this
is true also for the film, we utilized the ESRI method. The ESR
technique is a standard method used for the detection of
nitroxide radicals produced by oxidation of the sterically
hindered amine with oxygen radicals. For the analysis, the
copolymers were oxidized by a large excess of hydrogen
peroxide for 5 days; our trial experiments showed that these
conditions are sufficient for the quantitative HAS oxidation.
Figure 9 shows how scavenging activities of HHP and HHF
against ROO radicals changed with sample concentration and
with reaction time. Note that for analysis of HHP and HHF,
different polymer concentrations were used so as to stay in a
concentration window in which the time development of
fluorescence intensity can be observed. As follows from Figure
9
, high scavenging activity (>70%) was achieved for both
copolymer forms in a relatively short time (10 min). As
expected, scavenging activities increased with an increase in
polymer concentration. Further, the instantaneous scavenging
activity decreased with time as a result of the scavenging site
saturation. This decrease is more prominent for HHF because
of the diffusion effects mentioned above. These effects are
probably responsible also for the generally higher scavenging
activities of HHP compared to those of HHF (cf. the data for a
concentration of 12.6 mg/mL).
The scavenging activity against OH radicals is depicted in
Figure 10. The concentration and time dependence trends are,
in general, the same as for ROO radicals. The somewhat lower
values of scavenging activity can be ascribed to different kinetics
G
DOI: 10.1021/acs.biomac.5b00599
Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules
Article
Figure 12. ESR spectrum of the oxidized linear HH2 copolymer in (a)
aqueous solution and (b) the oxidized HHF film.
slow motional ESR spectrum was obtained for the cross-linked
film sample (Figure 12b). The anisotropy of nitroxide magnetic
properties is responsible for this difference. In the first case, the
anisotropy is averaged by the fast rotation of the nitroxide
moiety in solution, while in the second case, the anisotropy
develops because the rotational mobility of the nitroxide is
restricted by its incorporation into the polymer matrix.
Nitroxide concentration profiles were determined by deconvo-
lution of the ESR spectrum of the film (Figure 13). It can be
•
Figure 10. Dependence of HO scavenging activity of cross-linked
samples HHP and HHF on sample concentration and reaction time.
Average values (±SD) from three independent experiments are
presented.
Figure 13. Concentration profiles of the nitroxide radicals present in
the oxidized HHF film as determined by the ESRI method. The
concentration is expressed in nanomoles per millimeter per milligram.
Four equivalent concentration profiles representing average values
over a strip width are drawn for each strip.
clearly seen that there is only a slight variation in the absolute
values of the nitroxide concentration, confirming that the
scavenger is evenly spread in the polymer material. Note that
the broad small peaks visible at edges of the profiles are due to
the imperfection of the deconvolution procedure that is not
able to represent steep change in concentration at the edges of
the strip correctly.
CONCLUSIONS
■
HEMA/HAS copolymers with variable contents of the sterically
hindered amine were synthesized. In peroxyl, hydroxyl, and
DPPH radical scavenging assays, some of these copolymers
showed excellent scavenging activities, comparable to those of
established low-molecular weight antioxidant standards. From
the copolymers tested, the sample with 3.5 mol % HAS showed
consistently the best antioxidant performance. The reason for
this behavior, although presently not well-understood, can be
related to vastly different microstructures of the copolymers, as
determined from the monomer reactivity ratios. The
Figure 11. Time-dependent decrease in absorbance at 527 nm of
DPPH solutions in the presence of copolymer HH2, HHP, or HHF.
Average values (±SD) from three independent experiments are
presented. The HAS:DPPH molar ratio was 10:1.
Figure 12 shows (a) the ESR spectrum of the oxidized HH2
copolymer in comparison with (b) the ESR spectrum of the
oxidized HHF film strip. The linear copolymer in solution
showed a motionally narrowed ESR spectrum composed of
three narrow lines (Figure 12a). On the other hand, a broader
H
DOI: 10.1021/acs.biomac.5b00599
Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules
Article
subsequently synthesized cross-linked samples (powder and
film) with the optimal 3.5 mol % HAS exhibited, in the swollen
state, scavenging properties comparable to that of their linear
analogue in solution. As expected, diffusion effects played a role
in this case, which was reflected in the superior performance of
the powder form compared to the film. The biocompatibility
and antioxidant properties make the HEMA/HAS copolymers
promising materials for wound dressing applications.
(16) Ferreira, L.; Vidal, M. M.; Gil, M. H. Int. J. Pharm. 2000, 194,
69−180.
1
(
17) Curcio, M.; Cirillo, G.; Parisi, O. I.; Iemma, F.; Spizzirri, U. G.;
Altimari, I.; Picci, N.; Puoci, F. J. Funct. Biomater. 2011, 2, 1−17.
18) Ortiz, C.; Vazquez, B.; San Roman, J. Polymer 1998, 39, 4107−
114.
19) Ortiz, C.; Vazquez, B.; San Roman, J. J. Biomed. Mater. Res.
999, 45, 184−191.
20) Filippov, S. K.; Porsch, B.; Sergeeva, O.; Yu; Olifirenko, A. S.;
Lesnichin, S. B.; Domnina, N. S.; Komarova, E. A.; Lezov, A. V.;
Garamus, V. M.; Walterova, Z.; Stepanek, P. Eur. Phys. J. E: Soft Matter
Biol. Phys. 2011, 34, 123−132.
21) Sergeeva, O.; Vlasov, P. S.; Domnina, N. S.; Bogomolova, A.;
(
́
́
4
(
́
́
1
(
ASSOCIATED CONTENT
Supporting Information
̌
̌ ́
■
*
S
(
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.bio-
mac.5b00599.
̌
́
̌ ́
Konarev, P. V.; Svergun, D. I.; Walterova, Z.; Horsky, J.; Step ek, P.;
an
Filippov, S. K. RSC Adv. 2014, 4, 41763−41771.
(
22) Krishna, M. C.; DeGraff, W.; Hankovszky, O. H.; Sar, C. P.;
́
NMR and FTIR characterization of the HAS monomer,
poly(HEMA), and a HEMA/HAS copolymer; a table
with sequence length distributions for HEMA units in
copolymers; the synthetic procedure of the HPMA/HAS
copolymer; and experimental procedures for determi-
nation of antioxidant properties of linear and cross-linked
polymers (peroxyl, hydroxyl, and DPPH radical scaveng-
ing assays) (PDF)
Kal
1998, 41, 3477−3492.
23) Kejlova, K.; Labsky,
005, 19, 957−962.
24) Pilar, J.; Kríz,
009, 50, 4543−4551.
25) Rho, H. S.; Baek, H. S.; Kim, D. H.; Chang, I. S. Bull. Korean
́
̈
ai, T.; Jeko, J.; Russo, A.; Mitchell, J. B.; Hideg, K. J. Med. Chem.
(
́
́
́ ́
J.; Jírova, D.; Bendova, H. Toxicol. In Vitro
2
(
̌
̌ ́
̌ ̌ ́
J.; Meissner, B.; Pradny, M.; Kadlec, P. Polymer
2
(
Chem. Soc. 2006, 27, 584−586.
(
(
26) Cook, A.; le, A. J. Phys. Chem. 2006, 10, 44−49.
27) Werber, J.; Wang, Y. J.; Milligan, M.; Li, X.; Ji, J. A. J. Pharm. Sci.
2
(
011, 100, 3307−3315.
28) Huang, D.; Ou, B.; Hampsch-Woodill, M.; Flanagan, J. A.; Prior,
R. L. J. Agric. Food Chem. 2002, 50, 4437−4444.
29) Onoda, M.; Uchiyama, T.; Mawatari, K. I.; Kaneko, K.;
AUTHOR INFORMATION
Corresponding Author
E-mail: polakova@imc.cas.cz.
■
*
(
Notes
Nakagomi, K. Anal. Sci. 2006, 22, 815−817.
The authors declare no competing financial interest.
(30) Papariello, G. J.; Janish, M. A. M. Anal. Chem. 1965, 37, 899−
9
02.
(
31) Soares, A. A.; de Souza, C. G. M.; Daniel, F. M.; Ferrari, G. P.;
da Costa, S. M. G.; Peralta, R. M. Food Chem. 2009, 112, 775−781.
32) Gilbert, R. G.; Hess, M.; Jenkins, A. D.; Jones, R. G.; Kratochvíl,
ACKNOWLEDGMENTS
■
This work was supported by the Ministry of Education, Youth
and Sports of the Czech Republic (Grants LH14292 and
EE2.3.30.0029) and by BIOCEV (CZ.1.05/1.1.00/02.0109),
Biotechnology and Biomedicine Centre of Academy of Sciences
and Charles University from the European Regional Develop-
ment Fund.
(
P.; Stepto, R. F. T. Pure Appl. Chem. 2009, 81, 351−353.
(33) Fineman, M.; Ross, S. D. J. Polym. Sci. 1950, 5, 259−262.
̈
̈
(34) Kelen, T.; Tudos, F. J. Macromol. Sci., Chem. 1975, 9, 1−27.
(35) Odian, G. Principles of polymerization, 4th ed.; John Wiley &
Sons, Inc.: New York, 2004.
(
36) Mitchell, J. B.; Samuni, A.; Krishna, M. C.; DeGraff, W. G.; Ahn,
M. S.; Samuni, U.; Russo, A. Biochemistry 1990, 29, 2802−2807.
37) Arefiev, D. V.; Domnina, N. S.; Komarova, E. A.; Bilibin, A.Yu.
Eur. Polym. J. 2000, 36, 857−860.
38) Wang, X.; Qiu, X.; Wu, C. Macromolecules 1998, 31, 2972−
976.
REFERENCES
■
(
(
(
(
(
1) Gordillo, G. M.; Sen, C. K. Am. J. Surg. 2003, 186, 259−263.
2) Khodr, B.; Khalil, Z. Free Radical Biol. Med. 2001, 30, 1−8.
3) Apel, K.; Hirt, H. Annu. Rev. Plant Biol. 2004, 55, 373−399.
4) Schreml, S.; Szeimies, R. M.; Prantl, L.; Karrer, S.; Landthaler, M.;
(
2
(
(
(
39) Joanny, J. F.; Leibler, L. J. Phys. (Paris) 1990, 51, 545−557.
40) Khokholov, A. R. J. Phys. A: Math. Gen. 1980, 13, 979−987.
41) de Gennes, P. G. Isr. J. Chem. 1995, 35, 33−35.
Babilas, P. Br. J. Dermatol. 2010, 163, 257−268.
̈
(
5) Senel, O.; Cetinkale, O.; Ozbay, G.; Ahci
Plast. Surg. 1997, 35, 443−558.
6) Aurand, E. R.; Wagner, J.; Lanning, C.; Bjugstad, K. B. J. Funct.
̧ oglu, F.; Bulan, R. Ann.
(
Biomater. 2012, 3, 839−863.
(
(
7) Lee, K. Y.; Mooney, D. J. Chem. Rev. 2001, 101, 1869−1879.
8) Van Vlierberghe, S.; Dubruel, P.; Schacht, E. Biomacromolecules
2
(
011, 12, 1387−1408.
9) Sun, T. L.; Kurokawa, T.; Kuroda, S.; Ihsan, A. B.; Akasaki, T.;
Sato, K.; Haque, M. A.; Nakajima, T.; Gong, J. P. Nat. Mater. 2013, 12,
9
(
(
32−937.
10) Seidel, J. M.; Malmonge, S. M. Mater. Res. 2000, 3, 79−83.
11) Mishra, R. K.; Ramasamy, K.; Majeed, A. B. A. J. Appl. Polym. Sci.
2
012, 126, E98−E107.
12) Zhang, D.; Macias, C.; Ortiz, C. Macromolecules 2005, 38,
530−2534.
13) Das, A.; Mehndiratta, M.; Chattopadhyay, P.; Ray, A. R. J. Appl.
Polym. Sci. 2010, 115, 393−403.
14) Mishra, R. K.; Ray, A. R. J. Appl. Polym. Sci. 2011, 119, 3199−
206.
15) Michelsen, V. B.; Moe, G.; Strøm, M. B.; Jensen, E.; Lygre, H.
Dent. Mater. 2008, 24, 724−731.
(
2
(
(
3
(
I
DOI: 10.1021/acs.biomac.5b00599
Biomacromolecules XXXX, XXX, XXX−XXX