Enzymetically regulating the self-healing of protein hydrogels with high healing efficiency

Article abstract of DOI:10.1002/anie.201404531

Enzyme-mediated self-healing of dynamic covalent bond-driven protein hydrogels was realized by the synergy of two enzymes, glucose oxidase (GOX) and catalase (CAT). The reversible covalent attachment of glutaraldehyde to lysine residues of GOX, CAT, and bovine serum albumin (BSA) led to the formation and functionalization of the self-healing protein hydrogel system. The enzyme-mediated protein hydrogels exhibit excellent self-healing properties with 100 recovery. The self-healing process was reversible and effective with an external glucose stimulus at room temperature. Enzyme-mediated self-healing of dynamic covalent bond-driven protein hydrogels was realized by the synergy of two enzymes, glucose oxidase (GOX) and catalase (CAT). The dual-enzyme-mediated protein hydrogels exhibit excellent self-healing properties with 100 recovery. The self-healing process was reversible and effective with an external glucose stimulus at room temperature.

Full text of DOI:10.1002/anie.201404531

Angewandte
Chemie
DOI: 10.1002/anie.201404531
Self-Healing Materials Hot Paper
Enzymetically Regulating the Self-Healing of Protein Hydrogels with
High Healing Efficiency**
Yuzhou Gao, Quan Luo, Shanpeng Qiao, Liang Wang, Zeyuan Dong, Jiayun Xu, and
Junqiu Liu*
Abstract: Enzyme-mediated self-healing of dynamic covalent
bond-driven protein hydrogels was realized by the synergy of
two enzymes, glucose oxidase (GOX) and catalase (CAT). The
reversible covalent attachment of glutaraldehyde to lysine
residues of GOX, CAT, and bovine serum albumin (BSA) led
to the formation and functionalization of the self-healing
protein hydrogel system. The enzyme-mediated protein hydro-
gels exhibit excellent self-healing properties with 100%
recovery. The self-healing process was reversible and effective
with an external glucose stimulus at room temperature.
and mechanical strength of the materials is still a big
challenge for the self-healing systems.
To address this problem, self-healing materials based on
dynamic covalent bonds which employ the reversible but
relatively strong covalent bonds that control the structure of
the materials by the equilibrium of bond breaking and re-
[
6]
forming have been widely used. For most designed materi-
[7]
als, external stimuli are required to achieve healing. For
example, thermally reversible covalent bonds have been
introduced into polymers. Upon heating and cooling, they can
reversibly rupture and re-form to achieve self-healing
capacity. Recently, a transparent organic polymeric material
was shown to thermally repeatedly mend or re-mend itself
T
he ability to spontaneously heal injury and recover
functionality are the key features that increase the surviv-
[1]
[8]
ability and the lifetime of the organism. However, synthetic
materials usually fail after being damaged or fractured.
Inspired by nature, the demand for self-healing materials is
rapidly developing to offer a new strategy toward safer,
longer-lasting products and lower production costs. Over the
past few decades, three kinds of conceptual self-healing
systems, namely the capsule system, vascular system, and
based on Diels–Alder reactions. Furthermore, the redox
stimuli are also used to develop healing materials. In the
redox system, the disulfide bonds were cleaved and re-formed
[
9]
under the redox conditions. Acidity reversible covalent
bonds were also used to construct healing materials owing to
controllable properties and easy adjustment. The acylhydra-
[
2]
zone bonds and imine bonds were frequently used as
[3]
[10]
intrinsic system, have been reported. For the capsule system
and vascular system, self-healing materials mainly relied on
the encapsulated healing agents in the cavities. When the
capsules or vessels were damaged, the healing agents were
released to heal them through the formation of covalent
reversible covalent bonds.
These covalent bonds show
reversibility by breaking down the network and regenerating
the starting reagents by acid catalysis. These are relatively
simple stimulate processes; however, in living organisms, the
healing process should be complicated and hierarchically
controlled. Therefore, ways to simulate tissue repair and
construct enzyme-mediated self-healing systems is a very new
field that has strongly aroused our interest.
[
4]
bond. For intrinsic self-healing materials, self-repair occurs
by the inherent reversibility of chemical bonds or physical
interactions between the damaged regions. The easy prepa-
ration and modification of the intrinsic self-healing materials
as well as their stimuli-responsive self-healing properties led
to them attracting much more attention. Up to now, various
non-covalent interactions, such as hydrogen bonds, p–p
stacking interactions, host–guest interactions, and ionic inter-
actions have been used for the development of stimuli-
responsive healable materials, such as self-healing films and
Herein, we report a new strategy for the construction of
self-healing protein hydrogels based on imine bond that is
regulated by two synergetic enzymes: glucose oxidase (GOX)
and catalase (CAT). As shown in Scheme 1a, the reversible
covalent attachment of glutaraldehyde to lysine residues of
GOX, CAT, and bovine serum albumin (BSA) is suitable for
the formation and functionalization of the self-healing protein
hydrogel system. The BSA scaffold supports the hydrogel
system and the GOX as a catalytic center plays a key role to
adjust the pH of the system by the addition of extra traces of
glucose. First, glucose is oxidized to gluconolactone by GOX
catalysis. Then gluconolactone hydrolyzed to gluconic acid to
decrease the pH value of the hydrogel system. The H O
[5]
rubbers. However, how to enhance the structural stability
[*] Dr. Y. Z. Gao, Prof. Q. Luo, S. P. Qiao, L. Wang, Prof. Z. Y. Dong,
J. Y. Xu, Prof. J. Q. Liu
State Key laboratory of Supramolecular Structure and Materials
College of Chemistry, Jilin University
2
2
generated from the catalytic process will be decomposed into
H O and O by the enzyme CAT to avoid the imine bonds
2
2
2699 Qianjin Road, Changchun 130012 (China)
being oxidized. The generated O is re-utilized by GOX and
2
E-mail: junqiuliu@jlu.edu.cn
can accelerate the whole catalytic reaction (Scheme 1b). With
the change of pH, the imine bonds provide the opportunity to
heal the protein hydrogel (Scheme 1c).
[
**] We acknowledge financial support from the Natural Science
Foundation of China (21234004, 21221063, 21004028), the 111
project (B06009), and the Chang Jiang Scholars Program of China.
We first explored the impact of pH in the protein hydrogel
systems. Mechanical properties of protein systems under
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201404531.
Angew. Chem. Int. Ed. 2014, 53, 9343 –9346
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9343

.
Angewandte
Communications
Scheme 1. The enzyme-mediated self-healing of the protein hydrogel
system. a) Protein hydrogel system; b) the mechanism of enzyme-
regulated catalytic reactions; c) the procedure of self-healing in the
protein hydrogel system.
Figure 1. a) Storage modulus (G’) and loss modulus (G’’) versus
angular frequency (w) of the protein hydrogel (protein concentraction
1
2 wt%) with different glutaraldehyde concentractions at 258C.
b) Storage modulus (G’) and loss modulus (G’’) vs angular frequency
w) of the protein hydrogel (protein concentraction 12 wt%) with
(
different conditions were studied by dynamic rheological
measurements. The storage moduli (G’) and the loss moduli
different pH values at 258C. Solid symbols represent G’, and open
symbols represent G’’.
(G’’) were presented at different concentration of glutaralde-
hyde and different pH values (Figure 1). In Figure 1a, three
pairs of G’ and G’’ curves were included with different
glutaraldehyde concentrations with a constant protein con-
centration (12 wt%), and 20 min was required to form the
hydrogels. With the increase of glutaraldehyde content, G’
and G’’ were significantly raised. When the glutaraldehyde
However, at low pH of 3.5, no hydrogelation was observed
even after 24 h of standing still (Supporting Information,
Figure S1 f). These results suggest that the pH of the solution
plays an important effect during imine formation. On one
+
hand, the free H facilitates this reaction. Therefore, from
concentration reached to 0.5 wt%, the G’ value (average 1.5 ꢀ
pH 8 to 5 the reaction rate was promoted with the increase of
3
+
+
1
0 Pa) is much larger than G’’ (average 100 Pa), indicating
H concentration. On the other hand, when the H concen-
tration is too high, the amino group would be protonated, and
the nucleophilic attack occurs too slowly to inhibit the
formation of the imine bond, leading to the reaction
inefficiency. This is consistent with that observation where
no hydrogelation took place at pH 3.5.
Next, the activity of the GOX in the hydrogel system was
tested. The mechanism of the enzymatic catalysis is shown in
the Supporting Information, Figure S2. To test the catalytic
activity, hydrogel systems with different concentration of
glutaraldehyde were employed in the catalytic system (Sup-
porting Information, Figure S3). It can be seen from the
experimental results that the activity of the GOX was
gradually reduced with the increase of the glutaraldehyde
content, but it still maintains a certain degree of activity under
the hydrogel state. The gluconic acid can still be generated by
the glucose/glucose oxidase system to regulate the pH of the
system. At the same time, we also tested the relative GOX
activity versus different pH values at 258C (Supporting
the formation of strong chemical gel with covalent cross-
linked network. In Figure 1b, with a constant protein
concentration (12 wt%) and glutaraldehyde concentration
(
0.5 wt%), after 40 min to form the hydrogels, four pairs of G’
and G’’ curves were included with different pH values (pH 5,
, 7, and 8). These data showed that G’ and G’’ were
6
significantly decreased with the increase of pH, suggesting the
slow linkage reaction of amine with aldehyde, the extent of
cross-linking decreasing, and also the reducing mechanical
strength of the hydrogel.
The formation process of protein hydrogels with different
pH values (the numbers 1 to 6 on the little vials corresponding
to pH 8, 7, 6, 5, 4, and 3.5, respectively) was also investigated.
Hydrogels were formed in 20 min at pH 5 (Supporting
Information, Figure S1b) and 40 min at pH 4 (Supporting
Information, Figure S1c) after allowing the solution to stand.
The formation time of protein hydrogel was increased with
the increase of pH (Supporting Information, Figure S1d,e).
9
344 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 9343 –9346

Angewandte
Chemie
Information, Figure S4). The results show that the GOX
remains in relatively stable catalytic activity with the change
of pH from 3.5 to 7.5. The pH values versus time with
different glutaraldehyde concentrations in the protein hydro-
gel system was also explored (Supporting Information, Fig-
ure S5).
Hydrogel systems with or without GOX were prepared to
investigate whether the addition of GOX would promote
gelation. While the protein hydrogels were formed only in the
presence of 0.01 wt% GOX (Supporting Information, Fig-
ure S6a), we found that the pH value decreased from 6.5 to
scars had almost disappeared. Even when manually stretching
the sample, no rupture occurred. A similar self-healing test
was performed on the sample without GOX or CAT; the
control gel could not be coalesced and can be easily separated
into two pieces by manually stretching the sample. When the
coalesced gel was cut into separated pieces along the scar
again and then put them together closely as before, the self-
healing process occurred once again, proving that this process
could be repeated.
Then the self-healing property of the gel in scars recovery
was tested (Figure 3). It can be seen that when the gels were
broken, after self-healing for 5 h, a very significant difference
between the samples was observed in the present of GOX and
CAT and without GOX nor CAT. After 5 h, the samples with
GOX and CAT were almost completely repaired, but the
samples without GOX nor CAT were repaired at a very low
level.
5
.5 and the solution color gradually turned deep yellow-
brown during the formation process of protein hydrogel. In
this case, the formed hydrogel was breakable with low
elasticity. This is probably due to the oxidation of a part of
imine bond caused by H O , resulting in the damage of the net
2
2
structure of protein hydrogel. To demonstrate this conjecture,
.05 wt% CATwas added to the above system to catalyze the
decomposition of H O . As expected, the hydrogel color
0
2
2
turned light (Supporting Information, Figure S6b) after the
addition of CAT to clear out H O and thus to prevent the
2
2
oxidation of the imine bond. And we also test the activity of
CAT in the hydrogel system with different concentrations of
glutaraldehyde and different pH (Supporting Information,
Figures S7 and S8, respectively). The result indicated that
CAT could keep high activity in the protein hydrogel system.
The self-healing property of the cross-linked protein
hydrogel was investigated under a closed environment to
minimize water evaporation at room temperature. At first,
the self-healing property was demonstrated by connecting
two pieces of protein hydrogel into one piece (Figure 2). The
Figure 2. Photographs of the self-healing behavior of the protein hydro-
gel at room temperature: original, cut, mend, and stretched states,
respectively.
Figure 3. Photographs of the self-healing behavior of the protein hydro-
gel at room temperature; I–IV are the photographs of protein hydrogel
with dual enzyme, I) original, II) after damage, III) after standing for
gel was prepared in pure water with the protein concentration
(
12 wt%, GOX, CAT, and BSA for 0.1%, 0.5%, 11.4%,
respectively), and the glutaraldehyde concentration was
30 min, IV) after standing for 5 h, V–VIII are photographs of protein
0
7
.5 wt%, and then the pH of the solution was adjusted to
.0. The control gel was prepared under the same conditions
hydrogel without dual enzyme, V) original, VI) after damage, VII) after
standing for 30 min, and VIII) after standing for 5 h.
only without GOX nor CAT. To allow the reaction reach the
equilibrium, the mixture reacted for 20 h. As depicted in
Figure 2, the gel samples with good visibility were prepared
and cut with a blade into two pieces which were then brought
together with a little pressure to ensure the surfaces were fully
in contact. The contact samples with addition of glucose
Finally, the tensile tests were performed to quantitatively
evaluate self-healing properties of the original hydrogels and
repaired samples with GOX/CATand without GOX nor CAT
(Figure 4). The healing process was performed as described
before when connecting the two pieces of protein hydrogel
into one piece. Figure 4a shows typical stress-strain curves for
(
0.2 mg) were kept at room temperature. After 5 h, self-
healing of the contacted sample could be observed and the
Angew. Chem. Int. Ed. 2014, 53, 9343 –9346
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
9345

.
Angewandte
Communications
healing properties, which not only can be visually seen, but
can also be studied by mechanical test, showing 100%
recovery. The self-healing process was reversible and effective
with an external glucose stimulus at room temperature.
Moreover, the pH of this system was moderately adjusted by
two enzymes: glucose oxidase and catalase. The study of the
self-healing protein hydrogels by enzymatic manipulation of
dynamic covalent bonds will bring light to the properties of
dynamic and smart materials and will stimulate the develop-
ment of the biocompatible materials that would have
potential applications in biological repair smart materials.
Received: April 21, 2014
Revised: May 26, 2014
Published online: July 7, 2014
Keywords: dynamic covalent bonds · enzyme mediation ·
.
glucose oxidase · protein hydrogels · self-healing
[
1] Y. L. Chen, A. M. Kushner, G. A. Williams, Z. B. Guan, Nat.
Chem. 2012, 4, 467 – 472.
[
2] a) R. P. Wool, Soft Matter 2008, 4, 400 – 418; b) S. D. Bergman, F.
Wudl, J. Mater. Chem. 2008, 18, 41 – 62; c) M. D. Hager, P. Greil,
C. Leyens, S. van der Zwaag, U. S. Schubert, Adv. Mater. 2010,
22, 5424 – 5430; d) S. Burattini, B. W. Greenland, D. Chappell,
H. M. Colquhoun, W. Hayes, Chem. Soc. Rev. 2010, 39, 1973 –
1
7
985; e) A. B. South, L. A. Lyon, Angew. Chem. 2010, 122, 779 –
83; Angew. Chem. Int. Ed. 2010, 49, 767 – 771.
[
[
3] B. J. Blaiszik, S. L. B. Kramar, S. C. Olugebefola, J. S. Moore,
Figure 4. Typical stress–strain (stress–stretch ratio) curves of protein
hydrogels. a) Self-healing with dual enzymes. I): Original; II)–V): the
protein hydrogels were healed for different times. b) Self-healing with
dual enzymes and without enzymes. The samples were cut into
completely separate pieces using a razor blade and the cut faces were
gently brought together with glucose at room temperature for different
times.
N. R. Sottos, S. R. White, Annu. Rev. Mater. Res. 2010, 40, 179 –
2
11.
4] a) S. H. Cho, S. R. White, P. V. Braun, Adv. Mater. 2009, 21, 645 –
49; b) S. R. White, N. R. Sottos, P. H. Geubelle, J. S. Moore,
M. R. Kessler, S. R. Sriram, E. N. Brown, S. Viswanathan, Nature
001, 409, 794 – 798; c) K. S. Toohey, N. R. Sottos, J. A. Lewis,
6
2
J. S. Moore, S. R. White, Nat. Mater. 2007, 6, 581 – 585; d) C. J.
Hansen, W. Wu, K. S. Toohey, N. R. Sottos, S. R. White, J. A.
Lewis, Adv. Mater. 2009, 21, 4143 – 4147.
the original self-healed samples with GOX/CAT and without
GOX nor CAT after healing for different time. An important
result is that the healed sample with GOX and CAT did not
break at the contacted surface under the tensile testing. This
observation indicated that the break stress and strain of the
sample with added glucose can reach 100% of the original
strength and elongation abilities, but the sample without
GOX nor CAT only achieves 50% of the healing efficiency
[5] a) S. Burattini, B. W. Greenland, D. H. Merino, W. Weng, J.
Seppala, H. M. Colquhoun, W. Hayes, M. E. Mackay, I. W.
Hamley, S. J. Rowan, J. Am. Chem. Soc. 2010, 132, 12051 – 12058;
b) P. Cordier, F. Tournilhac, C. Souli-Ziakovic, L. Leibler, Nature
2008, 451, 977 – 980; c) P. Mukhopadhyay, N. Fujita, A. Takada,
T. Kishida, M. Shirakawa, S. Shinkai, Angew. Chem. 2010, 122,
6482 – 6486; Angew. Chem. Int. Ed. 2010, 49, 6338 – 6342.
[6] S. J. Rowan, S. J. Cantrill, G. R. L. Cousins, J. K. M. Sanders, J. F.
Stoddart, Angew. Chem. 2002, 114, 938 – 993; Angew. Chem. Int.
Ed. 2002, 41, 898 – 952.
(
Figure 4b). The mechanism of this self-healing property is
[
7] a) M. Burnworth, L. Tang, J. R. Kumpfer, A. J. Duncan, F. L.
related with the equilibrium of the imine bond formation, the
reaction can be promoted forward by dual enzymes synergetic
catalysis. The stretch experiments demonstrated that the
GOX protein hydrogel without CAT exhibited a slightly
higher tensile strength than that in the presence of CAT, but
its self-healing ability is lower, only 75% recovery (Support-
ing Information, Figure S9). This is mainly due to its relatively
lower reversibility resulted from the oxidation of a part of the
imine bond without CAT.
In conclusion, we have prepared the healable dual-
enzyme-mediated protein hydrogels and demonstrated that
the self-healing nature of dynamic imine bond cross-linked
chemical gels. The protein hydrogels exhibit excellent self-
Beyer, G. L. Fiore, S. J. Rowan, C. Weder, Nature 2011, 472, 334 –
338; b) H. M. Klukovich, Z. S. Kean, S. T. Iacono, S. L. Craig, J.
Am. Chem. Soc. 2011, 133, 17882 – 17888; c) G. H. Deng, F. Y. Li,
H. X. Yu, F. Y. Liu, C. Y. Liu, W. X. Sun, H. F. Jiang, Y. M. Chen,
ACS Macro Lett. 2012, 1, 275 – 279.
[
8] a) X. Chen, M. A. Dam, K. Ono, A. K. Mal, H. Shen, S. R. Nutt,
K. Sheran, F. Wudl, Science 2002, 295, 1698 – 1702; b) X. Chen, F.
Wudl, A. K. Mal, H. Shen, S. R. Nutt, Macromolecules 2003, 36,
1802 – 1807.
[
9] M. Nakahata, Y. Takashima, H. Yamaguchi, A. Harada, Nat.
Commun. 2011, 2, 511 – 516.
[10] G. H. Deng, C. M. Tang, F. Y. Li, H. F. Jiang, Y. M. Chen,
Macromolecules 2010, 43, 1191 – 1194.
9
346 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 9343 –9346