Hybrid macromolecular antioxidants based on hydrophilic polymers and sterically hindered phenols

Article abstract of DOI:10.1007/s11172-007-0117-x

A novel class of biologically active substances was created. These are hybrid macromolecular antioxidants (HMAO) based on hydrophilic polymers with chemically grafted sterically hindered phenols with different structural parameters. The antiradical activities of HMAO were assessed in reactions with 2,2-diphenyl-1-picrylhydrazyl and the corresponding sodium sulfonate in various solvents. The mechanism that explains the substantially enhanced activities of HMAO in water was proposed. The state of HMAO in solutions was studied by viscosimetry and photon correlation spectroscopy. HMAO were assayed in biological models.

Full text of DOI:10.1007/s11172-007-0117-x

Russian Chemical Bulletin, International Edition, Vol. 56, No. 4, pp. 781—790, April, 2007
781
Hybrid macromolecular antioxidants based on
hydrophilic polymers and sterically hindered phenols
D. V. Aref´ev,a I. S. Belostotskaya,b V. B. Vol´eva,b N. S. Domnina,a
ꢀ
N. L. Komissarova,bꢀ O. Yu. Sergeeva,a and R. S. Khrustalevac
aDepartment of Chemistry, St. Petersburg State University,
26 Universitetskii prosp., St. Peterhoff, 198504 St. Petersburg, Russian Federation.
Fax: +7 (812) 428 6939. Eꢀmail: domnin@id1042.spb.edu
bN. M. Emanuel´ Institute, Russian Academy of Sciences,
4 ul. Kosygina, 119991 Moscow, Russian Federation.
Fax: +7 (495) 137 4101. Eꢀmail: komissarova@polymer.chph.ras.ru
cV. A. Almazov Research Institute of Cardiology,
Ministry of Public Health and Social Development of the Russian Federation,
15 ul. Parkhomenko, 194156 St. Petersburg, Russian Federation.
Fax: +7 (812) 103 3976. Eꢀmail: crystal2000@front.ru
A novel class of biologically active substances was created. These are hybrid macromolecuꢀ
lar antioxidants (HMAO) based on hydrophilic polymers with chemically grafted sterically
hindered phenols with different structural parameters. The antiradical activities of HMAO
were assessed in reactions with 2,2ꢀdiphenylꢀ1ꢀpicrylhydrazyl and the corresponding sodium
sulfonate in various solvents. The mechanism that explains the substantially enhanced activiꢀ
ties of HMAO in water was proposed. The state of HMAO in solutions was studied by viscosimꢀ
etry and photon correlation spectroscopy. HMAO were assayed in biological models.
Key words: hybrid bioantioxidants, sterically hindered phenols, hydrophilic polymers, antiꢀ
radical activity, mechanism, aggregates, hydrodynamic properties, plasma substitutes.
Sterically hindered phenols (SHP) are the most repreꢀ
sentative and popular class of synthetic antioxidants used
to solve problems of both applied and fundamental charꢀ
acter. The firstꢀgeneration phenolic antioxidants (ionol,
simple substituted 2,6ꢀdi(tertꢀbutyl)phenols, phenozans)
have been widely employed as inhibitors of the thermoꢀ
oxidative degradation of polymers, oils, fats, fuels, etc., as
well as correctors of oxidant pathologies in living biologiꢀ
cal systems. However, a modern approach to creation of
bioantioxidants requires narrower specialization and comꢀ
bination of the antioxidant properties with the capability
for target delivery and structural interactions with the area
to be protected within a biosystem. This can be reached in
hybrid molecules whose separate fragments provide the
desired polyfunctionality.1 Examples of such hybrids inꢀ
clude "floating" SHP containing a charged onium group
(anchor) bound to a lipophilic longꢀchain alkyl substituꢀ
ent (float).2 Such a structure enables the antioxidant to
efficiently interact with the charged lipid bilayer of cell
membranes that need antioxidant protection for mainteꢀ
nance of a normal level of lipid peroxidation.
cally grafted to the polymer chain. This approach (called
biomimetic) to the design of biologically active comꢀ
pounds (BAC) was successfully applied in the 1950—1960s
for creation of a number of polymeric forms of BAC.
Both biopolymers (cellulose, starch, and chitosan) and
waterꢀsoluble synthetic polymers (polyvinylpyrrolidone,
polyacrylic acid, polyvinyl alcohol, etc.) were used as the
base polymers3 to develop highly efficient biocatalysts,
immunoactive drugs, polymeric derivatives of various lowꢀ
molecular bioregulators, biocides, etc.4—8
Such macromolecular systems have some advantages,
e.g., adjustable solubility, prolonged action, and the enꢀ
hanced stability and reduced toxicity of BAC. For inꢀ
stance, artificial biocatalysts are usually more resistant to
denaturing effects and can be used repeatedly.6
In the structure of polymeric hybrids, the nature of
linkage between the polymer and BAC is a substantial
element.5,9 An ether bond is stable in aqueous solutions
over a wide pH range (as well as in biological media). This
makes it possible to create compounds in which the atꢀ
tached BAC is active in the hybrid form, not as the liberꢀ
ated BAC molecule. This can be used to estimate how
much the biological activity of BAC has changed upon its
irreversible immobilization. When the molecule contains
a labile covalent (acetal, aldimine, or ester) bond linking
A novel class of hybrid macromolecular antioxidants
(HMAO) developed and studied over the last few years is
based on hydrophilic biopolymers and synthetic polymers
designed for medical purposes with SHP fragments chemiꢀ
Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 4, pp. 751—760, April, 2007.
1066ꢀ5285/07/5604ꢀ0781 © 2007 Springer Science+Business Media, Inc.

782
Russ.Chem.Bull., Int.Ed., Vol. 56, No. 4, April, 2007
Aref´ev et al.
BAC with the polymer, the active substance can be liberꢀ
ated from the hybrid at different rates depending on the
conditions (pH and temperature). It is the possibility of
gradual hydrolysis that makes polymeric derivatives of
plant growth regulators highly efficient. In contrast,
hybrids acting as immunomodulators are effective in the
polymer form only.5
Despite a sufficiently wide range of relevant publicaꢀ
tions and achievements in this field, so far many problems
associated with the effect of the structural parameters on
the activity of hybrid BAC are being solved empirically.
With HMAO as an example, we implement, probably for
the first time, a comprehensive study that includes develꢀ
opment of methods of synthesis, structural identification,
physicochemical characteristics and their dependence on
the medium, assay with biological models, and estimation
of the most promising ways of practical applications. Such
a study is possible because of a structural variety of HMAO
that differ in the nature and molecular weight (M ) of the
base polymer, the number of grafted SHP fragments, the
type of the covalent bond between SHP and the polymer
chain, and the nature and the length of the spacer beꢀ
tween the SHP core and the polymer.
70 000, and 200 000, hydroxyethylated starch (2) with
M = 200 000, and polyvinyl alcohol (3) with M = 10 000
were employed as macromolecular bases in the synthesis
of HMAO. These polymers are widely used in biology and
medicine (e.g., for preparation of colloidal plasma substiꢀ
tutes).10
For chemical modification, we used functionalized
2,6ꢀ and 2,4ꢀdi(tertꢀbutyl)phenols capable of interacting
with the hydroxy groups of the polymers: βꢀ[3,5ꢀdi(tertꢀ
butyl)ꢀ4ꢀhydroxyphenyl]propionic acid (phenozan
(4a)), 3,5ꢀdi(tertꢀbutyl)ꢀ4ꢀhydroxycinnamic acid,
4ꢀbromoꢀ or 4ꢀacetoxymethylꢀ2,6ꢀdi(tertꢀbutyl)phenol
(BP), 4ꢀ[3,5ꢀdi(tertꢀbutyl)ꢀ4ꢀhydroxybenzylidene]ꢀ5ꢀ
oxoꢀ2ꢀphenylꢀ4,5ꢀdihydrooxazole, 2,6ꢀdi(tertꢀbutyl)ꢀ4ꢀ
(2ꢀtosyloxy)ethoxymethylphenol, and 2,6ꢀdi(tertꢀbutyl)ꢀ
4ꢀ(4ꢀtosyloxy)butoxymethylphenol.
R = H (4a), K (4b)
Results and Discussion
Several series of HMAO with different structural paꢀ
rameters were synthesized. Hybrids of the first series
(HMAOꢀI) with ester bonds SHP—polymer were preꢀ
pared by condensation of the polymer with phenozan (4a)
in the presence of DCC and DMAP. By varying the ratio
and concentrations of the reagents and the esterification
time, we obtained hybrids based on dextran (1GꢀI),
hydroxyethylated starch (2GꢀI), and polyvinyl alcohol
(3GꢀI) with different degrees of substitution of the polyꢀ
mer units with SHP fragments (γ, mol.%). Hybrids 1GꢀI
were synthesized from dextrans with different molecular
masses.11—13 The γ values were determined by UV specꢀ
trophotometry (aromatic chromophore) and 1H NMR
spectroscopy that gave identical results.
Hydroxyꢀcontaining hydrophilic polymers such as
dextran (1) with M = 6000, 10 000, 18 000, 40 000,
We determined the solubilities of HMAO in water,
which depend on the polymer and γ. The maximum deꢀ
gree of replacement at which HMAO dissolved completely
(threshold water solubility) was 10 mol.% for 1GꢀI,
6 mol.% for 2GꢀI, and 2 mol.% for 3GꢀI. With a further
increase in γ, HMAO was dissolved in aqueous organic
solvents (water—ethanol and water—dioxane); at the highꢀ
est γ value obtained (for 1GꢀI with M = 40 000, γmax
=
47.7 mol.%), the polymeric product was dissolved in diꢀ
oxane, chloroform, and benzene. All HMAO were well
soluble in DMSO.
The second series of macromolecular antioxidants with
ether bonds SHP—polymer (HMAOꢀII) was synthesized
by reactions of polymers with 4ꢀbromoꢀ or 4ꢀacetoxyꢀ
methylꢀ2,6ꢀdi(tertꢀbutyl)phenol. The resulting polymeric
products contained different numbers of chemically bound

Hybrid macromolecular antioxidants
Russ.Chem.Bull., Int.Ed., Vol. 56, No. 4, April, 2007
783
SHP fragments and, consequently, were differently soluble
in water.
backbone chain of the polymer (dextrans (Dn) 1 with
M = 1500, 6000, 10 000, 18 000, 40 000, 70 000,
and 200 000).
To estimate the antioxidant activity of the HMAO
obtained, we studied the kinetics of their reactions with
the free radical 2,2ꢀdiphenylꢀ1ꢀpicrylhydrazyl (5a•)
(Scheme 2). A correlation between the observed antiradiꢀ
cal and antioxidant activities has been confirmed14,15 with
reactions of lowꢀmolecular SHP with 5a• as examples.
Therefore, reactions of HMAO with 5a• can serve as
A variant of HMAOꢀII are compounds obtained from
4ꢀbromomethylꢀ2,6ꢀdi(tertꢀbutyl)phenol and ethylene
and butylene glycols. The synthesis of compounds of this
series involved isolation of intermediate toluenesulfonates,
which react with the hydroxy groups of the polymer
(Scheme 1).
This method of synthesis of HMAO allowed us to vary
the length of the spacer between the SHP core and the

784
Russ.Chem.Bull., Int.Ed., Vol. 56, No. 4, April, 2007
Aref´ev et al.
Table 1. Rate constants of the reactions of 5a• with the antioxiꢀ
dants in various solvents
Antioxidant
K/L (mol s)–1
Chloroform
Benzene
Dioxane
Phenozan
1GꢀI
0.188 0.009 0.142 0.008 0.056 0.004
0.153 0.008 0.083 0.005 0.135 0.007
(γ = 47.7%,
M = 40 000)
The reactions of polymeric antioxidants with 5a• were
carried out at 20 °C, showing the pseudofirst order with
respect to the radical. The rate constants of the reactions
with 5a• were measured in both organic solvents (benꢀ
zene, chloroform, and dioxane) and dioxane—water mixꢀ
tures with different ratios of the components.
The antiradical effect with respect to 5a• is usually
tested in nonpolar solvents such as benzene, hexane, carꢀ
bon tetrachloride, and dioxane.5,16
a test for quantitative estimation of their antioxidant propꢀ
erties.
We found that the rate constants of the reactions of
lowꢀmolecular SHP with 5a• in chloroform and benzene
are higher than those for HMAO, while in dioxane
(Table 1) and in aqueous dioxane (Fig. 1), the pattern was
opposite.
Scheme 1
With an increase in the water content in aqueous diꢀ
oxane (see Fig. 1), the rate constants of the reactions of
both lowꢀmolecular SHP and 1GꢀI with 5a• increase.
Interestingly, the higher the water content of the solvent
the larger the difference between the rate constants of the
reactions of 5a• with both antioxidants.
Thus, the composition of the medium strongly influꢀ
ences the rate of the reactions of the antioxidants with
5a•, 1GꢀI being much more sensitive to variations in its
composition. For instance, at the highest water content of
aqueous dioxane, 1GꢀI have the highest rate constants of
the reactions with 5a• compared to lowꢀmolecular SHP.
K/L (mol s)–1
Scheme 2
2
12
10
8
6
4
1
2
0
20
40
60
C (mol.%)
Fig. 1. Plots of the rate constants of the reactions of 5a• with
lowꢀmolecular SHP (1) and 1GꢀI (γ = 10.4%, M = 40 000) (2)
vs. the composition of the mixed solvent (C is the water content
in aqueous dioxane).

Hybrid macromolecular antioxidants
Russ.Chem.Bull., Int.Ed., Vol. 56, No. 4, April, 2007
785
K/L (mol s)–1
In connection with this, we found it expedient to study
the antiradical activity in water and other solvents (dioxꢀ
ane and aqueous dioxane) and establish a correlation beꢀ
tween the activity and the medium polarity. To this end,
we synthesized sodium 2,2ꢀdiphenylꢀ1ꢀpicrylhydrazylsulꢀ
fonate (5b•), which is a waterꢀsoluble analog of 5a•.16
8
6
1
3
2
5
4
4b
4
2
2
4
6
8
10
12
14 γ (%)
Fig. 2. Plots of the rate constants of the reactions of 5b• with
1GꢀII (1—4) and 1GꢀI (5) in aqueous dioxane (1 : 1 v/v) vs. the
degree of substitution γ of glucose units in dextrans with
M = 10 000 (1), 40 000 (2), 70 000 (3), 200 000 (4), and
40 000 (5).
In aqueous dioxane, radical 5b• proved to be less acꢀ
tive than 5a•, regardless of the antioxidant (Table 2).
As in the reactions with 5a•, a mixture of chemiꢀ
cally nonbound dextran and phenoxan (potassium
βꢀ[3,5ꢀdi(tertꢀbutyl)ꢀ4ꢀhydroxyphenyl]propionate (4b))
reacted with 5b• at the same rate constant as did the
reference lowꢀmolecular antioxidant 4b. Thus, the comꢀ
parison of the antiradical effects on both the free radicals
showed the correctness of using 5b• in such investigaꢀ
tions.
The influences of the molecular mass of the starting
dextran and the degree of substitution γ of the glycoside
units with the SHP fragments on the antiradical activity
of the resulting hybrid compound were studied with 5b•
in aqueous dioxane (1 : 1) (Fig. 2). It can be seen that the
rate constants for 1GꢀI are virtually γꢀindependent, varyꢀ
ing within the experimental error (3.5 0.2 L (mol s)–1).
Another pattern was observed for 1GꢀII. Up to γ = 12%,
these hybrid compounds were more active than 4b. Howꢀ
ever, with an increase in γ, they became as active as or
even less active than compound 4b. An increase in the
molecular weight of HMAO from 10 000 to 200 000 lowꢀ
ered the rate constants by half.
length of the spacer. The highest rate constant was obꢀ
tained for 1GꢀI with an ester bond. One can infer from
the presented data that it is the hybrid structure of the
polymeric antioxidant that determines the observed subꢀ
stantially higher antiradical effects of HMAO with respect
to both free radicals in water and an aqueous organic
solvent.
It is known that the nature of the medium usually has
a weak influence on the rates of radical reactions. Howꢀ
ever, such an influence was revealed in our study of the
antiradical activity of HMAO in media with different poꢀ
larities. This is evident from the plot of the rate constant
vs. the composition of the solvent (Fig. 3).
With a decrease in the water content (the minimum
water content is 30%, which is associated with the soluꢀ
bility of 1G), the rate constants diminished, being close
for all the antioxidants at equal water : dioxane ratios.
Such a pronounced dependence on the medium comꢀ
position contradicts the generally accepted radical mechaꢀ
nism of hydrogen transfer from the SHP molecule to
compound 5b• (see Ref. 16) (Scheme 3, pathway a).5
In water alone, HMAO was more active than lowꢀ
molecular antioxidant 4b (see Fig. 1, Table 3).
It is worth noting that the difference between the
rate constants for 1G is determined by the nature and
Table 3. Rate constants of the reactions of
HMAO and phenoxan (4b) with 5b• in water
Table 2. Rate constants of the reactions of the antioxidants with
the free radicals in aqueous dioxane (1 : 1)
Antioxidant
K/L (mol s)–1
Antioxidant
Solubility
in water
K/L (mol s)–1
1GꢀI
1GꢀII
1GꢀIII
1GꢀIV
2GꢀI
2GꢀII
3GꢀI
3GꢀII
4b
1100 50
270 13
356 18
28.5 1.5
544 18
226 10
5a•
5b•
Dextran (М = 40000)
Phenoxan (4b)
Phenozan (4a)
Phenoxan—dextran (1 : 10)
Phenozan—dextran (1 : 10)
1GꢀI (γ = 10.0 mol.%)
+
+
—
+
—
+
0
0
10.0 0.5 3.0 0.2
3.4 0.2 1.2 0.05
10.0 0.5 3.0 0.2
3.4 0.2 1.2 0.05
12.5 0.8 3.5 0.2
95
70
5
3
35.7 1.2

786
Russ.Chem.Bull., Int.Ed., Vol. 56, No. 4, April, 2007
Aref´ev et al.
K/L (mol s)–1
logK
1.6
2
1000
1.2
0.8
0.4
800
600
400
200
4
3
1
0.5
1.0
1.5
2.0
2.5
3.0 Y (rel. unit)
40
60
80
C (vol.%)
Fig. 4. Correlation between the logarithms of the rate constants
logK of the reaction of 5b• with phenoxan 4b and the ionizing
capacity of aqueous dioxane.
Fig. 3. Plots of the rate constants of the reactions of 5b• with
phenoxan 4b (1) and various dextranꢀbased HMAO
(M = 40 000) (2—4) vs. the composition of the mixed solvent:
(2) 1GꢀI (γ = 9.1%), (3) 1GꢀII (γ = 9.1%), and (4) 1GꢀIII
(γ = 9.1%).
pose, we investigated the kinetics of the reaction of 5b•
with phenoxan (4b) in mixed solvents with known ionizꢀ
ing capacities (Y ) and plotted the logarithm of its rate
constant versus Y; the correlation coefficient was 0.992
(Fig. 4). The resulting linear plot indicates that the reacꢀ
tion kinetics obeys the Grünwald—Winstein equation and,
consequently, confirms the radicalꢀcation mechanism of
the reaction.
The enhanced antiradical activity of HMAO comꢀ
pared to the lowꢀmolecular antioxidant in aqueous media
may be due to the solvent shell of the macromolecule.
The properties of water in the solvent shell of hydrophilic
polymers differ from those of ordinary (bulk) water in
increased ionizing capacity because of different orientaꢀ
tions and mobility of water molecules and breaking
of hydrogen bonds between them.18,19 This water deꢀ
structurization reduces the energy required to break hydroꢀ
gen bonds in the solvation of radical cations.
One can assume that the mechanism of this reaction
in aqueous media involves the intermediate formation of
a radical cation (Scheme 3, pathway b).
It is known that the kinetics of the process involving
radical cations is described by the Grünwald—Winstein
equation:17
logK/K0 = m*Y,
where Y is a measure of the ionizing capacity of the solꢀ
vent, m is a factor that reflects the sensitivity of the reacꢀ
tion to a change in the ionizing capacity of the solvent,
K is the reaction rate constant, and K0 is the reaction rate
constant in a standard solvent (80% ethanol).
This criterion was used to confirm the radicalꢀcation
mechanism of the reactions under study. For this purꢀ
Scheme 3

Hybrid macromolecular antioxidants
Russ.Chem.Bull., Int.Ed., Vol. 56, No. 4, April, 2007
787
K/L (mol s)–1
Intrinsic viscosity is related to the speed of motion of
macromolecules in a solvent and depends on the mass
and shape of a polymer species in motion. Using photon
correlation spectroscopy, one can determine the hydroꢀ
dynamic radius of a lightꢀscattering species (Rh). A comꢀ
bination of these two methods allows a qualitative deꢀ
scription of the behavior of macromolecules in liquids.
Dextranꢀbased HMAO are studied best, because these
compounds have a wider solubility range. The data from
both methods were obtained for aqueous solutions and
the aprotic solvent DMSO. In DMSO, in which hydroꢀ
phobic interactions are weakened, the intrinsic viscosity η
changes only slightly compared to the starting dextran, is
1000
800
600
400
50000
100000
150000
200000 М
virtually independent of γ, and equals 0.21—0.23 dL g–1
.
Fig. 5. Plot of the rate constants of the reaction of 1GꢀI
(γ = 9 mol.%) with 5b• in water (T = 20 °C) vs. the molecular
mass of dextran.
In water, the intrinsic viscosities of the same products
monotonically decrease from 0.13 dL g–1 for γ = 4%
to 0.09 for γ = 9.4% (for the starting dextran, η =
0.16 dL g–1). Such a shape of the dependence is explained
by compactification of the macromolecules of modified
dextran, which becomes more hydrophobic as the numꢀ
ber of SHP fragments in the polymer chain increases.
This is supported by PCS data on dynamic light scatꢀ
tering. The spectrum of dextran shows two peaks due to
species with very different sizes (Fig. 6). The leftꢀhand
peak corresponds to molecular sizes, while the rightꢀhand
peak corresponds to macromolecular aggregates. The same
pattern was observed in DMSO, except that the aggreꢀ
gates are smaller than those in water.
Modification of dextran with SHP fragments dramatiꢀ
cally changes the size distribution of species, lowers the
polydispersity index, and gives rise to a pronounced peak
that is narrower than the peak of the starting dextran. The
species that correspond to this peak are much larger, for
all HMAO, than their molecules. This suggests aggregaꢀ
tion of modified dextran in aqueous solutions into species
with Rh = 0.23—0.43 µm. In DMSO, the pattern is qualiꢀ
tatively the same, but the hydrodynamic radii are lower
(Rh = 0.09—0.16 µm). Apparently, this accounts for the
aforementioned fact that the antiradical activity of HMAO
in DMSO and other organic solvents differs only slightly
from that of the lowꢀmolecular analog SHP, being strongly
dependent on the length of the spacer in aqueous soluꢀ
tions.
The hydrodynamic properties make HMAO well
biocompatible; because of this, we estimated its antioxiꢀ
dant activity with several biological models. The range of
optimum HMAO concentrations for biosystems was deꢀ
termined in the model of the hypotonic hemolysis of
erythrocytes (Fig. 7).
The structural relation of HMAO to hydrophilic polyꢀ
mers used in medical practice as components of plasma
substitutes prompted investigations of the properties of
HMAO in the model of hemorrhagic shock (acute masꢀ
sive blood loss). Medicobiological tests were carried out
at the V. A. Almazov Research Institute of Cardiology of
In aqueous solutions, SHP fragments covalently bound
to the polymer are most likely surrounded by solvent shells
in which they seem to interact with the radical. The higher
ionizing capacity of the solvent shell of the macromolꢀ
ecule favors better stabilization of the radicalꢀion transiꢀ
tion state; this explains why hybrid compounds are highly
active in water but become less active when the water
content of the mixed solvent decreases.
The influence of the solvent shell on the reaction of
HMAO with 5b• can be traced in the series of HMAO
with different molecular masses and different spacer
lengths.
The highest rate constants for the reaction with 5b•
were obtained for 1GꢀI (see Table 3). The rate constants
for 1GꢀII and 1GꢀIII were lower, while 1GꢀIV with the
longest spacer is as active as lowꢀmolecular SHP 4b. This
can be explained by the escape of the SHP core with this
spacer from the solvent shell.
The influence of the solvent shell on the reaction of
HMAO with 5b• is responsible for the appearance of a
maximum on the plot of the rate constant vs. the molecuꢀ
lar mass of the base polymer (Fig. 5). According to the
literature data,18,19 an increase in M of hydrophilic polyꢀ
mers enlarges the solvent shell and hence the reaction
of 5b• with HMAO should be accelerated. However, above
a certain M value, interactions of radicals with phenol
fragments are sterically hindered, which results in the
lowering of the rate constants.
Therefore, the nature of the spacer is one of the most
significant factors that influence the antiradical activity
of HMAO. Interpretation of this fact was substantially
contributed by a study of the hydrodynamic properties of
HMAO by viscosimetry and photon correlation spectroꢀ
scopy (PCS)*.
* The investigations were carried out by S. K. Filippov and
supervised by A. V. Lezov at the Department of Physics of the
St. Petersburg State University.

788
Russ.Chem.Bull., Int.Ed., Vol. 56, No. 4, April, 2007
Aref´ev et al.
r(R)
0.16
modern medicine. Currently available plasma substitutes
do not protect an organism that has experienced blood
loss from postꢀtransfusional complications (ischemia of
organs and tissues, edemas, and encephalopathy) associꢀ
ated with an avalanche increase of oxidative processes
after the restoration of bloodstream (oxygen paradox).
The use of HMAO in plasma substitutes may be a solution
to this problem. Laboratory tests in rats showed that in
contrast to nonmodified plasma substitutes, compensaꢀ
tion of blood loss with HMAOꢀcontaining ones (2•10–6
mol L–1 with γ taken into account) stabilizes the hemodyꢀ
namic parameters 12—15 min after the beginning of infuꢀ
sion and keeps them constant throughout the observation
time (~1 h). Infusion of required amounts of Reopolyꢀ
glukin caused only a shortꢀtime hypertensive effect
(Fig. 8), with edemas of tissues in 72% of rats.
a
2
0.12
0.08
0.04
0
1
10–6 10–4 10–2 100 102 104
Rh/µm
r(R)
A series of special experiments with massive blood loss
preceded by infusion of HMAOꢀcontaining plasma subꢀ
stitutes revealed a 52% increase in the rat survival comꢀ
pared to a reference group. Thus, by using HMAOꢀconꢀ
taining plasma substitutes for compensation of blood loss,
one can restore and maintain the hemodynamic paramꢀ
eters in animals at nearly the initial level, suppress free
radical oxidation, and, consequently, prevent damage due
to perfusion upon restoration of bloodstream.
The use of HMAO in plasma substitutes is an imporꢀ
tant but not unique line of their practical application. By
varying the structural parameters, one can adapt HMAO
to solution of various problems in biology, medicine, and
agriculture. The latest data on the immunostabilizing and
cytoprotective properties of HMAO21 and on possible use
of HMAO in formulations for lowꢀtemperature preservaꢀ
tion of genetic materials provide convincing evidence for
the necessity of further investigations of this class of
bioantioxidants.
b
2
0.04
0.02
0
1
10–4
10–2
100
102
104 Rh/µm
Fig. 6. Distribution of aggregates over the hydrodynamic radii
for dextran (1) and dextranꢀbased HMAO (γ = 8.2%) (2) in
water (a) and DMSO (b) (C = 0.6 wt.%), θ = 90°.
the Ministry of Public Health and Social Development
of the RF.
Thus, we created a novel class of phenolic antioxiꢀ
dants based on hydrophilic polymers with SHP fragments
chemically grafted to the polymer chain and studied the
Development of new means and methods for transfuꢀ
sional therapy belong to the most topical problems of
N (%)
pA/Torr
a
70
60
50
40
30
20
10
128.000
42.500
12.30
25.00
37.30
t/min
pA/Torr
128.000
42.500
b
–3 –4 –5 –6 –7 –8 –9 –10 –11 –12
x
12.30
25.00
37.30
t/min
Fig. 7. Determination of the optimum concentration of HMAO
(CHMAO/mol L–1) from the osmotic resistance of erythrocytes
(upon blood loss); CHMAO = (1•10x),where x = –3÷–12 and
N (%) is the number of lysated erythrocytes.
Fig. 8. Dynamics of the blood pressure level (PA) on compensaꢀ
tion of the blood loss with Reopolyglukin (a) and dextranꢀbased
HMAO (b).

Hybrid macromolecular antioxidants
Russ.Chem.Bull., Int.Ed., Vol. 56, No. 4, April, 2007
789
Model of the hypotonic hemolysis of erythrocytes. Hemolysis
was carried out at 20 °C in a spectrophotometer cell containing
0.5% NaCl (2.5 mL) or a solution of HMAO (2.5 mL) in
0.5% NaCl. A suspension of erythrocytes (0.02 mL, ~150•107
cells/mL) in saline was added with a micropipette. The kinetics
of hypotonic hemolysis (yield of hemoglobin) was estimated
from the diminished turbidity of the solution at λ = 800 nm. The
efficiency of HMAO was calculated as a ratio of the degree of
hemolysis with and without the antioxidant.
Model of hemorrhagic shock. Tests were carried out with
male rats of the Sprague Dawley line (420—470 g) narcotized
with Nembutal (intraperitoneal injection, 25 mg kg–1). Blood
pressure was measured directly in the femoral artery with a
Baxter transducer and analyzed with the KardioPlus program.
The heart rate was estimated from the intersystolic interval; the
other femoral artery was used for bloodletting (specific blood
volume 30 mL kg–1, rate 0.8—1.0 mL (kg min)–1). To comꢀ
pensate blood losses, the plasma substitutes Reopolyglukin
and Hemodez ("Biokhimik", Saransk), 0.9% NaCl, the
Locke—Ringer solution, and Infukoll (SerumꢀWerk Bernburg
AG, Germany) were injected into rats of a reference group.
Blood losses in rats of a test group were compensated with soluꢀ
tions of HMAO.
structural factors of the high antiradical activity of HMAO
in aqueous media. Using the kinetic method, we demonꢀ
strated that the mechanism of redox processes with HMAO
in water involves the intermediate formation of radical
ions. We considered an important role of supramolecular
structures resulting from HMAO aggregation in aqueous
media. Based on the PCS data on the structural features
of solutions of HMAO, we explained why HMAO is difꢀ
ferently active in organic and aqueous media. The bioꢀ
compatibility and high antioxidant activity of HMAO was
illustrated with living biological models. We outlined the
main trends of practical application of HMAO.
Experimental
Commercial hydrophilic polymers with different molecular
weights were used; their weights were additionally refined by visꢀ
cosimetry on an Ubbelohde viscosimeter in water and DMSO
at 25 °C. Polymers were purified by dialysis against water folꢀ
lowed by freezeꢀdrying.
Functionalized SHP that can etherify and esterify the OH
groups of the polymer were used for the synthesis of HMAO.
Typical procedures for the synthesis, isolation, and purification
of HMAO are described in Ref. 12. The number of SHP fragꢀ
ments grafted to the polymer was determined by spectrophoꢀ
tometry from the absorption of the aromatic chromophore
(λmax = 275 nm, EtOH — water (1 : 1)) and expressed in terms of
γ (mol.%) defined as the fraction of substituted units in the total
number of the monomer units in the polymer. Phenozan was
used as a reference compound. The reproducibility of the results
was repeatedly confirmed by performing syntheses under identiꢀ
cal conditions. The absence of lowꢀmolecular SHP in the reacꢀ
tion products was confirmed by gel permeation chromatography
(Sephadex LHꢀ20, EtOH—water (1 : 1)). The localization of
SHP in the glucose unit of dextran followed from previous
data22,23 and was not studied further.
The kinetics of the reaction of HMAO with 5b• was moniꢀ
tored with an SFꢀ56 spectrophotometer (LOMO) by recording
changes in the optical density of 5b• at 520 nm (absorption
peak). The molar extinction coefficient in water was 10 200 100
L (mol cm)–1. The initial concentration of 5b• was 2•10–5
mol L–1; the initial concentration of HMAO was 4•10–4 mol L–1
(with γ taken into account). The reaction was stopped when the
conversion of 5b• reached 20%. In a control experiment, the
concentration of the individual radical in aqueous dioxane reꢀ
mained unchanged upon 1ꢀh irradiation in the cell at λ = 520 nm.
Dynamic light scattering experiments were carried out by
PCS at 25, 38.8, and 50 °C on a Photo Cor Complex setup fitted
with an automatic goniometer, a PhotoCorꢀFC realꢀtime
correlator, and a He—Ne laser (λ = 632.8 nm, ouput power
25 mW) as a light source. Scattering angles were 20—90°. The
concentration of the polymer in solution was 0.6 wt.%. The
signal acquisition time was 200—300 s. Autocorrelation funcꢀ
tions were processed by the regulation method with the DyneLS
program.
References
1. E. B. Burlakova, in Khimicheskaya i biologicheskaya kinetika.
Novye gorizonty [Chemical and Biological Kinetics. New Scope],
Khimiya, Moscow, 2005, 2, 10 (in Russian).
2. V. B. Vol´eva, I. S. Belostotskaya, N. L. Komissarova, and
G. A. Nikiforov, in Khimicheskaya i biologicheskaya kinetika.
Novye gorizonty [Chemical and Biological Kinetics. New Scope],
Khimiya, Moscow, 2005, 1, 306 (in Russian).
3. A. F. Nikolaev and G. I. Okhrimenko, Vodorastvorimye
polimery [WaterꢀSoluble Polymers], Khimiya, Leningrad,
1979, 144 pp. (in Russian).
4. Polimery meditsinskogo naznacheniya [Polymers for Medical
Applications], INKhS Akad. Nauk SSSR, Moscow, 204 pp.
(in Russian).
5. V. V. Korshak and M. I. Shtilman, Polimery v protsessakh
immobilizatsii i modifikatsii prirodnykh soedinenii [Polymers
in Immobilization and Modification of Natural Compounds],
Nauka, Moscow, 1984, 261 pp. (in Russian).
6. N. A. Plate and A. E. Vasil´ev, Fiziologicheski aktivnye
polimery [Physiologically Active Polymers], Khimiya, Mosꢀ
cow, 1986, 296 pp. (in Russian).
7. V. P. Torchilin, Immobilizovannye fermenty v meditsine [Imꢀ
mobilized Enzymes in Medicine], Znanie, Moscow, 1986,
28 pp. (in Russian).
8. G. E. Afinogenov and E. F. Panarin, Antimikrobnye polimery
[Antimicrobial Polymers], Gippokrat, St. Petersburg, 1993,
264 pp. (in Russian).
9. M. I. Shtilman, Immobilization on Polymers, VSP, Utrecht—
Tokyo, 1993, p. 479.
10. Spravochnik VIDAL´. Lekarstvennye preparaty v Rossii
[VIDAL Handbook. Drugs in Russia], AstraPharmService,
Moscow, 2002, 1488 (in Russian).
The properties of HMAO were studied with biological modꢀ
els of the hypotonic hemolysis of erythrocytes and hemorrhagic
shock.
11. D. V. Aref´ev, N. S. Domnina, and E. A. Komarova,
Zh. Prikl. Khim., 1999, 72, 670 [Russ. J. Appl. Chem., 1959,
72 (Engl. Transll.)].

790
Russ.Chem.Bull., Int.Ed., Vol. 56, No. 4, April, 2007
Aref´ev et al.
12. D. V. Arefiev, N. S. Domnina, E. A. Komarova, and A. Yu.
Bilibin, Eur. Polym. J., 2000, 36, p. 857.
18. M. Yasushi, T, Noriaki, and K. Hiromi, J. Phys. Chem.,
1993, 97, 13903.
13. N. S. Domnina, D. V. Arefiev, E. A. Komarova, and A. Yu.
Bilibin, Macromol. Symp. "Degradability, Renewability
and RecyclingꢀKey Functions for Future Materials", 1999,
144, 339.
19. T. Terada, Y. Maeda, and H. Kitano, J. Phys. Chem., 1993,
97, 3619.
20. P. G. de Gennes, in Scaling Concepts in Polymer Physics,
Cornell University Press, Ithaca, 1979.
14. I. G. Arzamanova, R. M. Logvinenko, and Ya. A. Gurevich,
Zh. Fiz. Khim., 1973, 47, 707 [J. Phys. Chem. USSR, 1973, 47
(Engl. Transl.)].
15. I. G. Arzamanova, M. I. Naiman, and Ya. A. Gurevich,
Zh. Fiz. Khim., 1979, 53, 1007 [J. Phys. Chem. USSR, 1979,
53 (Engl. Transl.)].
16. E. G. Rozantsev and E. G. Sholle, Organicheskaya khimiya
svobodnykh radikalov [The Organic Chemistry of Free Radiꢀ
cals], Khimiya, Moscow, 1979, 344 pp. (in Russian).
17. A. S. Dneprovskii and T. I. Temnikova, Teoreticheskie osnovy
organicheskoi khimii [Theoretical Foundations of Organic
Chemistry], Khimiya, Leningrad, 1979, 520 pp. (in Russian).
21. N. S. Domnina, R. S. Khrustaleva, V. B. Vol´eva,
I. S. Belostotskaya, and N. L. Komissarova, Tez.
VII Mezhdunarodnoi konferentsii "Bioantioksidant" [Abstrs
VII Int. Conf. "Bioantioxidant"], 2006, Moscow, 30 pp. (in
Russian).
22. M. SanchezꢀChaves and F. Arranz, Polymer, 1997, 38,
No. 10, 2501.
23. M. SanchezꢀChaves, J. L. Rodrigues, and F. Arranz,
Macromol. Chem. Phys., 1997, 198, 3465.
Received February 1, 2007;
in revised form March 30, 2007