Cytocompatible, Photoreversible, and Self-Healing Hydrogels for Regulating Bone Marrow Stromal Cell Differentiation

Article abstract of DOI:10.1002/mabi.201500457

Photo-crosslinking and self-healing have received considerable attention for the design of intelligent materials. A novel photostimulated, self-healing, and cytocompatible hydrogel system is reported. A coumarin methacrylate crosslinker is synthesized to modify the polyacrylamide-based hydrogels. With the [2+2] cyclo-addition of coumarin moieties, the hydrogels exhibit excellent self-healing capacity when they are exposed to light with wavelengths at 280 and 365 nm, respectively. To enhance cell compatibility, a poly (amidoamine) crosslinker is also synthesized. Variations in light exposure times and irradiation wavelengths are found to alter the self-healing property of the hydrogels. The hydrogels are shown to induce a regular cellular pattern. The hydrogels are used to regulate bone marrow stromal cells differentiation. The relative mRNA expressions are recorded to monitor the osteogenic differentiation of the cells. (Figure presented.).

Full text of DOI:10.1002/mabi.201500457

Macromolecular
Bioscience
Full Paper
Cytocompatible, Photoreversible, and Self-
Healing Hydrogels for Regulating Bone
Marrow Stromal Cell Differentiation
Lianlian Yu, Kaige Xu, Liangpeng Ge, Wenbing Wan, Ali Darabi,
Malcolm Xing,* Wen Zhong*
Photo-crosslinking and self-healing have received considerable attention for the design of
intelligent materials. A novel photostimulated, self-healing, and cytocompatible hydrogel
system is reported. A coumarin methacrylate crosslinker is synthesized to modify the poly-
acrylamide-based hydrogels. With the [2+2] cyclo-addition
of coumarin moieties, the hydrogels exhibit excellent self-
healing capacity when they are exposed to light with wave-
lengths at 280 and 365 nm, respectively. To enhance cell
compatibility, a poly (amidoamine) crosslinker is also syn-
thesized. Variations in light exposure times and irradiation
wavelengths are found to alter the self-healing property of
the hydrogels. The hydrogels are shown to induce a regular
cellular pattern. The hydrogels are used to regulate bone
marrow stromal cells differentiation. The relative mRNA
expressions are recorded to monitor the osteogenic differen-
tiation of the cells.
1. Introduction
biological process helps living organisms to restore their
integrity and prolong their lifespan.[1] Mimicking such
a natural healing feature, materials with the capacity of
self-repairing after being damaged are highly desirable in
biomedical applications. For example, if applied in tissue
engineering, a self-healing scaffold can self-repair the
cracks or damages that can occur during or after implanta-
tion, and consequently, minimize the needs of further pro-
cedures and enhance the efficacy of the treatment.
A variety of smart materials that have self-healing capac-
ities have been developed based on various strategies,[2–9]
including the photosensitive self-healing polymers.[10–12]
Coumarins are a class of molecules including hundreds
of derivatives. Their photosensitivity was discovered one
century ago.[13] Coumarin-containing polymers have been
well studied and widely applied in many fields, including
biochemicals, organic–inorganic hybrid materials, liquid
crystalline materials, electro-optical materials, and light
harvesting/energy transferring materials.[14–21] In addi-
tion, their photocleavage behavior and reversible photo-
crosslinking property have been widely investigated.[22–25]
The phenomenon of self-healing exists widely in living
organisms, from blood clotting to skin repair. This special
L. Yu, Prof. W. Zhong
Department of Biosystem Engineering
University of Manitoba
Winnipeg MB R3T 2N2, Canada
E-mail: wen.zhong@umanitoba.ca
K. Xu, L. Ge, W. Wan, A. Darabi, Prof. M. Xing
Department of Mechanical Engineering
University of Manitoba
Winnipeg MB R3T 2N2, Canada
E-mail: malcolm.xing@umanitoba.ca
Prof. M. Xing
Manitoba Institute of Child Health
Winnipeg MB R3T 2N2, Canada
Prof. M. Xing
Department of Biochemistry and Medical Genetics
University of Manitoba
Winnipeg MB R3T 2N2, Canada
Macromol. Biosci. 2016, DOI: 10.1002/mabi.201500457
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1
DOI: 10.1002/mabi.201500457
wileyonlinelibrary.com
Early View Publication; these are NOT the final page numbers, use DOI for citation !!

Macromolecular
Bioscience
L. Yu et al.
www.mbs-journal.de
Recently, there have also been studies on self-healing
materials containing coumarin compounds.[26] How-
ever, cell-compatibility of these materials has been a
concern due to the usage of such crosslinkers as formal-
dehyde.[27] It is still a great challenge to fabricate macro-
molecular materials with both superior self-healing and
cytocompatible performances. The [2+2] cycloaddition of
a couple of coumarin moieties forms a cyclobutane ring
under a UVA (long wave UV) irradiation (320–400 nm).
Being exposed to an UVC (short wave UV) irradiation
(200–280 nm), on the other hand, the coumarin photo-
dimers are cleaved to release the original coumarin moi-
eties (Figure 6). Polymers containing coumarin structural
parts, therefore, exhibit fast photoresponsibility and
effective photoreversibility at different wavelengths of
UV light. Poly (amidoamine)s are a family of synthetic
polymers containing tertiary amino and amido groups
regularly arranged along their polymer chain.[28–30] They
are obtained by Michael-type polyaddition of primary or
secondary amines to bis-acrylamides. By using such func-
tionalized monomers as aminocarbohydrate derivatives,
PAAs can be easily modified to incorporate such side sub-
stituents as additional tertiary amino groups, carboxyl
groups, hydroxyl groups, and allyl groups.[31] However,
for materials used in biological application, biocompat-
ibility is another important issue. Despite of their very
promising properties, synthetic hydrogels are character-
ized by poor cell adhesion.[32] Efforts have been made to
improve biocompatibility and cell adhesion of hydro-
gels. For example, polycations such as poly(L-lysine) was
used to enhance the cytocompatibility of hydrogels.[33,34]
4-aminobutyl guanidine or agmatine was also introduced
to build a functional amphoteric repeating unit with a
structure similar to the tripeptidearginin-glycin-aspartic
acid (RGD), which presents in extra cellular matrix
(ECM).[31] PAAs hydrogels incorporating RGD-mimicking
units have been shown to have a superior cell adhesion
capability as compared to plain amphoteric PAAs.[35,36]
Based on the aforementioned, both cell-compatibility
and self-healing capacity might be achieved by taking
advantages of the photoresponsive behavior of coumarin
and RGD-mimicking units in the PAA, as it will be devel-
oped and discussed in this work. Most existing work
in this area focuses more on the reaction mechanisms
of the materials than on the biological applications of
coumarin. In addition, there has been no report on self-
healing macrohydrogels containing coumarin derivatives
with the feature of reversible photo-crosslinking and cell
attachment.
2. Experimental Section
2.1. Materials
2-Bromo-2-methylpropionyl
bromide,
acryloyl
chloride,
7-hydroxy-4-methylcoumarin, 2-bromoethanol, triethylamine,
potassium carbonate, ethanol, ether, anhydrous sodium
sulfate, dichloromethane, sodium chloride, DMSO, acryla-
mide, N,N′-methylene bis(acrylamide) (MBA), agmatine
sulfate salt, 1-(2-aminoethyl)-piperozine(AEPZ), N,N,N′,N′-
tetramethylethanediamine (TEMED), ammonium persulfate
(APS), and lithium hydroxide (LiOH) were all purchased from
Sigma-Aldrich without further purification. PBS (phosphate-
buffered saline) 1x powder, Hoechst 33342, BODIPY FL phalla-
cidin, To-Pro-3 iodide and Live-Cell staining kit were all purchased
from Invitrogen. The 3-(4,5-dimethyl thiazolyl-2)-2,5-diphe-
nyltetrazolium bromide (MTT) cell viability assay kits were from
Biotium Inc. (Hayward, CA). O.C.T. compounds were purchased
from V.W.R FITC fluorescence dye and RT-PCR kit was purchased
from Thermo Scientific. UV exposure power was 16 mw cm−2
.
2.2. Synthesis of Coumarin Methacrylate (CMA)
Crosslinker
The synthesis of coumarin methacrylate (CMA) crosslinker was
conducted by a two-step reaction. As showed in Figure 1. First,
a mixture of 7-hydroxy-4-methylcoumarin (5.0 g, 25.7 mmol),
2-bromoethanol (5.0 g, 40.0 mmol), and potassium carbonate
(3.0 g, 21.7 mmol) in 50 mL of ethanol was heated under reflux
for 20 h. The mixture was then cooled at rt, diluted with ether
and water. The mixture was separated, and the aqueous layer
was extracted with ether. The combined organic layer was dried
over magnesium sulfate, followed by the removal of the solvent.
Step 1
K2CO3, ethanol reflux, 20 h
OH
Br
HO
O
O
O
HO
O
O
Step 2
NEt3, CH2Cl2, 12 h, rt
Cl
O
HO
O
O
O
O
O
O
O
O
Figure 1. Schematic illustration of two-step synthesis of the CMA crosslinker.
Macromol. Biosci. 2016, DOI: 10.1002/mabi.201500457
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
2
www.MaterialsViews.com
Early View Publication; these are NOT the final page numbers, use DOI for citation !!

Macromolecular
Bioscience
Cytocompatible, Photoreversible, and Self-Healing Hydrogels for Regulating Bone Marrow Stromal Cell Differentiation
www.mbs-journal.de
The solid obtained, was pure enough to use in the next step
without further purification.
1H NMR (ppm) for PAA crosslinker: δ, 1.5 (H14 + H13, m, 4nH),
2.0-3.0(H2 + H5 + H7 + H9 + H10 + H11 + H12,m, 8nH), 3.17(H15, t,
2nH), 4.53 and 4.62(H1 + H6,s, (2n+2)H), 5.77 (ethylene, dd, 2H18),
6.21(ethylene, dd, 2H17), 6.225 (ethylene, dd, 2H16).
To a solution of triethylamine (5.0 g, 6.89 mL, 49.4 mmol) and
7-(2-hydroxyethoxy)-4-methylcoumarin (5.0 g, 20.14 mmol) in
80 mL of dichloromethane, was added dropwise acryloyl chlo-
ride (5.0 g, 4.48 mL, 55.24 mmol). After being stirred for 12 h at
room temperature, water was added. The mixture was separated,
and the aqueous layer was extracted with dichloromethane. The
combined organic layer was washed with an aqueous solution
of NaCl, followed by drying over anhydrous sodium sulfate and
the removal of the solvent to give a solid. The crude product was
recrystallized from ethanol to afford colorless powdery crystals.
2.5. Synthesis of CMA Containing Hydrogel
Coumarin-containing hydrogel was prepared by copolymeriza-
tion of acrylamide (6.2 mmol), 7-(2-methacryloyloxyethoxy)-
4-methylcoumarin (0.31 mmol), PAA crosslinker (0.97 mmol)
using APS (0.20 mmol), and TEMED in 1.0 mL of DMSO at 65 °C.
The gel was purified by washing with large amounts of DMSO,
followed by exposure to water over several hours, then repeated
washing with water. To form the gel with cellular pattern, the
concentrations of components were listed as in Table 1.
2.3. Synthesis of the PAA Crosslinker
The agmatine containing poly(amidoamine) (PAA) crosslinker
was synthesized via Michael-addition polymerization as shown
in Figure 2. MBA (234 mg, 1.52 mmol), Agmatine (138 mg,
0.6 mmol), and AEPZ (52 mg, 0.4 mmol) were dissolved in water/
methanol (v/v = 5/1, total in 5 mL) while stirring. When the solu-
tion was clear, LiOH·H2O (25.2 mg, 0.6 mmol) was added to the
solution. The mixture was then gently stirred and allowed to react
at 45 °C in the dark for 72 h. After reaction, the solvent was evapo-
rated using a rotary-evaporator. The product was finally recovered
by lyophylization and stored in −20 °C refrigerator for future use.
2.6. Evaluation of Self-Healing Capacity of the Hydrogels
2.6.1. Sample Preparation
Hydrogels for self-healing evaluations were all prepared
between two glass cover slips (25 mm × 25 mm) by adding 250 μL
of gel precursor solution. After gelation under 60 °C for 20 min,
the cover slips were removed and then the gel was thoroughly
washed with sterilized PBS solution to remove chemical
residuals.
2.4. Characterization
2.6.2. Confocal Laser Scanning Microscope Images of the
Healing Hydrogels
In order to characterize the CMA crosslinker (Figure 3) and
PAA crosslinker (Figure 4), 1H NMR spectra was recorded on an
Advance 300 MHz spectrometer. 8 mg of crosslinker powders
were dissolved in 800 μL of NMR solvent. Chemical shifts (δ)
were reported in parts per million (ppm): 1H NMR (ppm, CDCl3)
The hydrogels precursor were prepared first and mixed evenly.
Then 200 μL precursor was dropped onto square glass cover slips
(length: 20 mm). After gelation under 65 °C, the slips with hydro-
gels on the surface were stained by FITC fluorescence dye. Then
a crack was made through the whole thickness of the hydrogel
by a blade. Fluorescence pictures before and after UV healing
process (λ = 365 nm) were recorded by confocal laser scanning
microscopy (CLSM) (Olympus FV1000, Japan).
δ for CMA crossliner: (above), 2.44 (CH3,s, 3H), 4.28 (CH2
O
aromatic,t, 2He), 4.55 (CH2 OCO,t, 2Hf), 6.10 (ethylene, dd, 1Hh),
6.28 (aromatic, s, 1Ha), 6.38 (ethylene, dd, 1Hg), 6.68 (ethylene,
dd, 1Hi), 7.14 (O C6H3 ,d, 1Hc), 7.17 (O C6H3 ,s, 1Hd),
7.64(O C6H3 ,d, 1Hb).
O
NH
NH2
O
O
HN
.
HO
S
OH
H2N
LiOH . H2O
N
H
N
H
N
H
N
NH2
O
AEPZ
Lithium
hydroxide
Agmatine
Agmatine sulfate salt
MBA
N,N-Methylene bis(acrylamide)
1-(2-Aminoethyl)-piperazine
°
Water/methanol, 45 C, 72 h
H
O
O
O
O
O
O
H
H
N
H
N
H
N
N
H
N
H
N
N
H
N
N
N
H
n
H
H
R2
H
n
R2: polymer chain
Hyperbranched polyamidoamine with acyl end groups
NH
HN
NH2
Figure 2. Schematic illustration of the preparation of the PAA crosslinker.
Macromol. Biosci. 2016, DOI: 10.1002/mabi.201500457
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3
www.MaterialsViews.com
Early View Publication; these are NOT the final page numbers, use DOI for citation !!

Macromolecular
Bioscience
L. Yu et al.
www.mbs-journal.de
Hb
Hd
Hc
O
Ha
O
Hg
Me
e
Hh
O
O
f
Hi
O
Hb Hd+Hc
Ha
Hi
Hh
Hg
Hf
He
PPM
8.0
7.0
6.0
5.0
4.0
3.0
2.0
1.0
Figure 3. 1H NMR spectrum of the CMA crosslinker.
H
O
O
O
O
18
7
8
4
1'
1
3
9
O
O
H
18
3
1
11
H
N
N
H
N
N
N
N
17
6
2
5
H
H
H
N
H
12
N
R2
N
N
H
H
m
16
17
n
2
H
10
13
15
H
16
14
NH
HN
NH2
2-12
′
1
1
16
13,14
15
17
1.0
7.0
6.0
5.0
4.0
3.0
2.0
Figure 4. 1H NMR spectrum of the PAA crosslinker.
intimate contact without pressure. Samples were then irradi-
ated under UV light at the wavelength of 365 nm for 10, 20, 30
and 60 min, respectively. After the self-healing process by irra-
diation, the connected bars were fixed onto the INSTRON mate-
rials testing machine as showed in Figure 5. The test speed was
set at 0.1 mm min−1. Stress–strain curves were recorded by the
2.6.3. Tensile Tests
To further evaluate the healing capacity of the hydrogels, hydro-
gels containing CMA crosslinker were synthesized. Specimen
of 20 mm in length, 4 mm in width and 2 mm in thickness
was firstly cut by a razor. Samples were kept in alignment and
Table 1. Concentrations of components.
Whole concentration
Name
AM
Concentration [mg mL−1
]
Mole
1.47315 mmol
0.0152 mmol
0.0304 mmol
5.7 mg
Volume [μL]
523.6
1.5 mmol mL−1 (97:1:2) Total 1 mL
200
50
30
200
–
CMA
87.25
Crosslinker
APS
405.3
28.5
TEMED
2.9 mg
3.74
Macromol. Biosci. 2016, DOI: 10.1002/mabi.201500457
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
4
www.MaterialsViews.com
Early View Publication; these are NOT the final page numbers, use DOI for citation !!

Macromolecular
Bioscience
Cytocompatible, Photoreversible, and Self-Healing Hydrogels for Regulating Bone Marrow Stromal Cell Differentiation
www.mbs-journal.de
Figure 5. Images of sample test setting.
INSTRON software. All measurements were carried out at con-
stant temperature and humidity room. The sample size of each
group is 3 (n = 3). The original hydrogel samples without cut
were also tested as a control to show the healing efficiency of the
hydrogels.
errors, all swelling ratio results should be obtained from tripli-
cate samples.
2.8. In Vitro Biocompatibility Test
2.8.1. Cell Culture and Growth on Hydrogel
2.6.4. Photoreversible Test
The hydrogels precursor for cell culture were prepared first and
mixed evenly. Then 100 μL precursor was dropped onto glass
cover slips (diameter: 10 mm). After gelation under 37 °C, the
glass cover slips were put into PBS solution for 2 h, in order to
remove chemical residuals. Then the glass cover slips were taken
out and irradiated under UV light in a biosafety hood for 30 min.
Finally, the gels coated glass cover slips were placed in cell cul-
ture dishes (35 mm × 10 mm). Mouse bone marrow stromal cells
(BMSCs from ATCC, USA) were cultured with Dulbecco’s Modified
Eagle’s Medium (DMEM, GIBCO) supplemented with 10% fetal
bovine serum (FBS, GIBCO), 1.0 × 105 Ul−1 penicillin (Sigma) and
100 mg mL−1 streptomycin (Sigma) at 37 °C in 5% CO2. Live/dead
staining was carried out to evaluate the cytotoxicity of hydro-
gels, according to a protocol from Invitrogen Inc., Canada. BMSC
cultures were performed using hydrogels and stained at 24 h, 3 d,
and 7 d. Images were taken by CLSM.
In order to verify the photoreversible property of hydrogels, UV
light (λ < 280 nm) at another wavelength was applied to the
mechanically bisected and then healed samples. Specifically,
sample hydrogels 20 mm in length, 4 mm in width, and 2 mm
in thickness were cut into two halves and subsequently allowed
to heal for 30 min. The healed samples were further exposed to
the UV light at the wavelength of 260 nm for 10 and 20 min,
respectively. Their tensile stress and breaking elongation were
again recorded on an INSTRON tester, following the procedures
described in the previous section.
2.7. Swelling Test
Hydrogels were prepared in a specific size in 24-well plate as
described before. Then weighed and immersed in vials which
containing PBS buffer (pH = 7.4) at room temperature. At
designed time intervals, the hydrogels were taken out from solu-
tion, and weighed after wiping off any visible surface moisture.
The percentage amount of buffer absorbed can be calculated
using the following formula
2.8.2. Quantitative Real-Time PCR Analysis
100 mL of BMSCs in medium at a density of 2 × 106 cells per mL was
seeded onto the hydrogels for 30 min. Then, 2 mL medium were
added to the disk. After 2 d, cell-encapsulating hydrogels were
transferred to FBS-supplemented DMEM containing a combina-
tion of osteogenic chemical supplements: 50 mg mL−1 L-ascorbic
acid 1-phosphate (Sigma), 10 × 10−3 M b-glycerophosphate (Sigma)
and 100 × 10−6 M dexamethasone (Sigma). Gels with cells were
incubated for various time points. The osteogenic medium was
(1)
Water (%) = (Wt-W0)/W0*100
where Wt is the weight of hydrogels at weighing time and W0
is the weight of the initial weight hydrogel. In order to reduce
Macromol. Biosci. 2016, DOI: 10.1002/mabi.201500457
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
5
www.MaterialsViews.com
Early View Publication; these are NOT the final page numbers, use DOI for citation !!

Macromolecular
Bioscience
L. Yu et al.
www.mbs-journal.de
changed for every 2 d. At predetermined time intervals, the media
were aspirated with the unattached cells and the wells were
washed with DPBS (Dulbecco's phosphate-buffered saline). Then,
cells on hydrogels were treated with liquid nitrogen and smashed.
In order to validate the gene expression of osteogenic differentia-
tion in all samples, total RNA isolation and cDNA synthesis were
conducted using TRIzol and Oligo dT (Thermo Scientific, USA),
according to the standard procedures. Then quantitative real-time
PCR (qPCR) was performed by SYBER Green assays (Applied Bio-
systems, USA). Amplification conditions were as follows: hold at
95 °C for 10 min, followed by 40 cycles at 15 s at 95 °C and 1 min
in 60 °C. Thermal cycling and fluorescence detection were done
using the StepOnePlus real-time PCR System (Applied Biosystems,
USA). The mRNA expression levels were determined relative to
the GAPDH by the ΔCt method. The five gene premiers were (form
5′–3′): ALP-F: CTC CAA AAG CTC AAC ACC AAT G, ALP-R: ATT TGT
CCA TCT CCA GCC G;BSP-F: CCA CAC TTT CCA CAC TCT CG, BSP-R:
CGT CGC TTT CCT TCA CTT TTG; COL I–F: AAC AGT CGC TTC ACC
TAC AG, COL I-R: AAT GTC CAA GGG AGC CAC; OPN-F: CTA CGA
CCA TGA GAT TGG CAG, OPN-R: CAT GTG GCT ATA GGA TCT GGG;
GAPDH-F: AGG TCG GTG TGA ACG GAT TTG, GAPDH-R: TGT AGA
CCA TGT AGT TGA GGT CA.
To evaluate the intrinsic healing behavior of cou-
marin-contained gels, the effect of coumarin on the
healing process was further explored. A crack was made
on the gel by a razor. The bisected samples were put in
contact at their cut cross sections and exposed to 365 nm
UV light irradiation for 30 min to allow self-healing. Con-
focal microscopy images (Figure 7) exhibited that the
autonomic fusion of cut surface occurred after the UV
treatment for 30 min, suggesting a good self-healing con-
dition of the hydrogel.
Stress–strain tests were conducted to further investi-
gate the mechanical properties of the healed hydrogels
as compared to the original hydrogels. Samples were
first cut into two halves, and then the two halves were
brought into contact (Figure 8a). The bisected samples
were exposed to 365 nm UV light irradiation for 10, 20,
30, and 60 min, respectively. Tensile tests were con-
ducted to quantify the capacity of the damaged gels to
restore their strength. For instance, the tensile stress of
the gel which was irradiated for 60 min (200.2 24.7kPa)
is nearly five times as that of the gel which was only
irradiated for 10 min (44.9 17.2 kPa), indicating that
the introduction of coumarin groups improves the self-
healing performance of gels. Moreover, increasing the
irradiation time gives rise to a remarkable improve-
ment in the tensile modulus as well as in the breaking
elongation. Gels which were irradiated for 10, 20, 30,
and 60 min, showed the elongations at break at 21.6
7.2%, 43.4 5.2%, 67.7 10.4%, and 96.2 9.5%, respec-
tively (Figure 8b). It can also be seen from the chats that
60 min of healing allows the cut samples restore to 88.6%
of the original tensile modulus (Figure 8d), suggesting
an excellent healing efficiency of the CMA-contained
hydrogels.
To further confirm the photoreversibility of the hydro-
gels. The samples were cut and subsequently expose to
365 nm UV irradiation to allow self-healing for 30 min.
The self-repaired samples were exposed to 254 nm UV
irradiation for 10 and 20 min, respectively. Figure 9 shows
reduced gel strengths with the increase in exposure time,
which may be caused by the photocleavage of the formed
cyclobutane rings under 254 nm UV irradiation.
Hydrogels are necessary for their high water content
and low interfacial tension with the surrounding biolog-
ical environment.[35] Understanding the role of gel prop-
erties on ECM formation is important for numerous bio-
medical applications.[24] Swelling property is important to
evaluate the scaffold’s biocompatibility. Figure 10 shows
that the swelling ratio of both gels with and without CMA
increased gradually with time. Hydrogels with CMA have
a slightly lower swelling ratio as compared to the gels
without CMA, which may be resulted from the changes in
the crosslinking density.[15,35]
3. Results and Discussions
In this work, the synthesis of a CMA crosslinker was con-
ducted by a two-step reaction. In order to introduce an
agmatine group into the N,N′-methylene bis(acrylamide),
the poly(amidoamine) (PAA) crosslinker was synthesized
via Michael-addition. NMR was conducted to verify the
successful synthesis of both crosslinkers (Figures 1–4).
Chemical structure of gels and mechanism of the self-
healing process were showed in Figure 6. To endow poly-
acrylamide with photochemical reactivity, coumarin
methacrylate plays dual-roles of both side groups and
crosslinkers. It means that double bond groups have
to be present on the backbone of the polymer to con-
nect with PAA crosslinker and acrylamide. As a result,
multi-functional homopolymer of polyacrylamide that
contains coumarin groups and agmatine groups were
synthesized.
Generally, damaged gels that contain coumarin groups
showed significant self-healing feature upon the irra-
diation of 365 nm UV light. For the control group (gels
without coumarin groups), however, the two halves
cannot bind each other. With the treatment of UV light
(365 nm), the results reflect the contribution of π–π*
transitions of the conjugated benzene nucleus and
pyrone nucleus in 4-methylcoumarin chromophores.[37]
Because the double bonds of 4-methylcoumarin gradu-
ally dimerize to form cyclobutane rings under 365 nm UV
illumination, the conjugation between double bonds and
phenyl groups was destructed (Figure 6). As a result, the
two halves can connect together.
Macromol. Biosci. 2016, DOI: 10.1002/mabi.201500457
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
6
www.MaterialsViews.com
Early View Publication; these are NOT the final page numbers, use DOI for citation !!

Macromolecular
Bioscience
Cytocompatible, Photoreversible, and Self-Healing Hydrogels for Regulating Bone Marrow Stromal Cell Differentiation
www.mbs-journal.de
Figure 6. a) The schematic illustration of the self-healing process and b) chemical structure of the polyacrylamide gels and the photo-
dimerization and photocleavage of coumarin side groups.
BMSCs were seeded onto the hydro-
gels. MTT assay was used to evaluate
the cell viability for 24 h. Viability of
cells seeded on tissue culture plates
(TCPS) was normalized to 100%. As
the control group, hydrogels without
cells were also evaluated. As shown
Figure 7. Fluorescence confocal microscopy images of healing process.
in Figure 11a, hydrogels with the PAA
Macromol. Biosci. 2016, DOI: 10.1002/mabi.201500457
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
7
www.MaterialsViews.com
Early View Publication; these are NOT the final page numbers, use DOI for citation !!

Macromolecular
Bioscience
L. Yu et al.
www.mbs-journal.de
Figure 8. a) Images of sample preparation with a cut, b) breaking elongations, c) tensile stress, and d) tensile modulus of self-healed samples
versus different UV irradiation time (365 nm) with 16 mw cm−2, as compared to original hydrogels (n = 3).
crosslinker had high cell viability after 24 h culture. The
guanidine groups from agmatine combined with the
amido groups from MBA may have significantly pro-
moted cell attachment.[36] Live/dead assay was used to
evaluate the cell viability for 72 h. Hydrogels with and
without the PAA crosslinker were evaluated for the cell
proliferation and cell attachment. After 24 h culture,
most cells were found to attach to the gels. After 72 h
culture, some dead cells were found but most of them
were still alive (as shown in Figure 11). The number of
BMSCs increased significantly in the hydrogels that
contain the PAA crosslinker. However, in the control
group, a significantly lower cell density was observed,
which may be due to the poor adhesion to the untreated
hydrogel. The results suggest that the PAA crosslinker
may improve the cell attachment to the hydrogels, there-
fore, endowing the self-healing gels wider applications.
More interestingly, a pattern of cell distributions was
observed after the growth of the cells on the gels for
24 h. Confocal laser microscopy images demonstrate that
the cells tend to align in a certain direction as show in
Figure 12, when CMA and PAA crosslinker reacted with
a proper ratio (Table 1), although more experiments are
needed to determine the underlying mechanisms.
1400
gel with CMA
gel without CMA
1200
1000
800
600
400
200
0
1
3
24
Time(hrs)
72
Figure 9. Stress–strain curves at different healing time with
Figure 10. Swelling ratio of hydrogels with and without CMA
(n = 3).
reverse conditions: 254 nm.
Macromol. Biosci. 2016, DOI: 10.1002/mabi.201500457
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
8
www.MaterialsViews.com
Early View Publication; these are NOT the final page numbers, use DOI for citation !!

Macromolecular
Bioscience
Cytocompatible, Photoreversible, and Self-Healing Hydrogels for Regulating Bone Marrow Stromal Cell Differentiation
www.mbs-journal.de
Figure 13. Real time-qPCR of osteogenic gene expression levels
of mouse BMSCs cultured in vitro. Total RNA was prepared from
BMSCs grown on hydrogel for 2 and 7 d. BSP, ALP, COL, OPN gene
expressions were quantified using real time-qPCR methods,
GAPDH was used as an internal control. Data values are expressed
as mean SE (n = 3).‘‘*’’means p < 0.05 versus the groups on day 2.
or differentiated stem cells.[38] On day 2, the expression of
BSP gene was found to increase and the same trend was
observed on day 7 (Figure 13a). Collagen (COL) is a critical
marker in the late stage of bone differentiation.[39] The
expression level of COL was also found to increase with
time. Osteopontin (OPN) is another important human
gene which is related to osteogenesis.[40] It plays a role
in organic linking component of the bone. As shown in
Figure 13, the expression profiling of OPN was relatively
lower than that of both BSP and COL. But the expression
profiling of OPN still have a positive trend after 7 d during
the osteogenetic differentiation. Alkaline phosphatase
(ALP) is an enzyme that exists in all tissue in human body,
which can remove the phosphate group from nucleotides
or proteins. ALP may be a marker for bone metabolism: if
a high level of ALP was observed, active bone tissues were
generally believed to have formed.[41] The ALP expres-
sion was evaluated when BMSCs were cultured on hydro-
gels. The results show that the hydrogels had a high ALP
expression level throughout one week (Figure 13).
Figure 11. a) Cell viability on hydrogels after 24 h culture based
on MTT assay (absorbance is normalized to 1 for the TCPS con-
trol sample) (n = 3). Live/dead assay of BMSCs cultured on b,d)
hydrogels without PAA crosslinker, c,e) hydrogels containing PAA
after 24 and 72 h. (Green for live cells and red for dead cells, scale:
100 μm.)
To evaluate the effects of the hydrogels on the osteo-
genic differentiation of BMSCs, a real time-qPCR was
conducted on the BMSc-seeded hydrogels (Figure 13).
Their responses to the hydrogels were recorded using the
osteogenic related mRNAs for ALP, BSP, COL, and OPN.
As shown in Figure 13, osteogenic gene expressions in
BMSCs increased significantly after 7 d culture.
Bone sialoprotein (BSP) is an important protenforosteo-
genetic differentiation. It can be expressed in osteoblasts
3. Conclusions
In summary, we demonstrate a novel macrohydrogels
with self-healing capability and biocompatibility. Cou-
marin derivatives are successively introduced into the
polyacrylamide-based hydrogels. The reversible photo-
dimerization and photocleavage reactivity of coumarin
have been imparted to the polymer. The resultant func-
tional gels show noticeable self-healing property after
healing under 365 nm UV irradiation, ascribed to the
intramolecular [2+2] photocyclo-addition reactions of cou-
marins. The self-healing gels also possess high mechanical
Figure 12. Confocal microscopy images of BMSC cells grown on
the gel after 24 h. (Scale: 200 μm.)
Macromol. Biosci. 2016, DOI: 10.1002/mabi.201500457
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
9
www.MaterialsViews.com
Early View Publication; these are NOT the final page numbers, use DOI for citation !!

Macromolecular
Bioscience
L. Yu et al.
www.mbs-journal.de
[17] M. P. Brun, L. Bischoff, C. Garbay, Angew. Chem. Int. Ed. 2004,
43, 3432.
[18] L. Zhao, D. A. Loy, K. J. Shea, J. Am. Chem. Soc. 2006, 128,
14250.
[19] P. O. Jackson, M. O’Neill, W. L. Duffy, P. Hindmarsh,
S. M. Kelly, G. J. Owen, Chem. Mater. 2001, 13, 694.
[20] M. Chen, K. P. Ghiggino, A. Launikonis, A. W. Mau,
E. Rizzardo, W. H. Sasse, S. H. Thang, G. J. Wilson, J. Mater.
Chem. 2003, 13, 2696.
strength after healing for 60 min with stress intensity
of 2 × 105 Pa and the elongation at break over 96%. More
interestingly, cell attachment property was significantly
improved by adding the PAA crosslinker. The development
of methods and materials in this work may provide a new
insight into the fabrication of novel biocompatible self-
healing materials for extensive applications.
[21] C. Kim, A. Trajkovska, J. U. Wallace, S. H. Chen,
Macromolecules 2006, 39, 3817.
[22] Q. Fu, L. Cheng, Y. Zhang, W. Shi, Polymer 2008, 49, 4981.
[23] P. P. Wells, H. Morrison, J. Am. Chem. Soc. 1975, 97, 154.
[24] S. R. Trenor, A. R. Shultz, B. J. Love, T. E. Long, Chem. Rev.
2004, 104, 3059.
Acknowledgements: The authors would wish to thank NSERC
(Natural Sciences and Engineering Research Council of Canada)
for funding the present research.
Received: December 14, 2015; Revised: February 2, 2016;
Published online: ; DOI: 10.1002/mabi.201500457
[25] B. Ren, D. Zhao, S. Liu, X. Liu, Z. Tong, Macromolecules 2007,
40, 4501.
[26] B. Kiskan, Y. Yagci, J. Polym. Sci., Part A: Polym. Chem. 2014,
52, 2911.
[27] J. Ling, M. Z. Rong, M. Q. Zhang, J. Mater. Chem. 2011, 21,
18373.
Keywords: crosslinker; cytocompatible; photoreversible; self-
healing hydrogel
[28] P. Ferruti, Ion-Chelating Polymers (Medical Applications), CRC
Press Inc., Boca Raton, FL 1996, p. 5/3334.
[29] P. Ferruti, M. A. Marchisio, R. Duncan, Macromol. Rapid
Commun. 2002, 23, 332.
[30] P. Ferruti, S. Manzoni, S. C. Richardson, R. Duncan,
N. G. Pattrick, R. Mendichi, M. Casolaro, Macromolecules
2000, 33, 7793.
[31] J. Franchini, E. Ranucci, P. Ferruti, M. Rossi, R. Cavalli,
Biomacromolecules 2006, 7, 1215.
[32] B. V. Slaughter, S. S. Khurshid, O. Z. Fisher,
A. Khademhosseini, N. A. Peppas, Adv. Mater. 2009, 21, 3307.
[33] L. Cai, J. Lu, V. Sheen, S. F. Wang, Biomacromolecules 2012,
13, 1663.
[34] L. Cai, J. Lu, V. Sheen, S. F. Wang, Biomacromolecules 2012,
13, 342.
[35] E. Jacchetti, E. Emilitri, S. Rodighiero, M. Indrieri,
A. Gianfelice, C. Lenardi, A. Podestà, E. Ranucci, P. Ferruti,
P. Milani, J. Nanobiotechnol. 2008, 6, 14.
[36] J. Shi, J. Ouyang, Q. Li, L. Wang, J. Wu, W. Zhong, M. M. Xing,
J. Mater. Chem. 2012, 22, 23952.
[1] C. Yu, C. F. Wang, S. Chen, Adv. Funct. Mater. 2013, 24,
1235.
[2] H. Ceylan, M. Urel, T. S. Erkal, A. B. Tekinay, A. Dana,
M. O. Guler, Adv. Funct. Mater. 2013, 23, 2100.
[3] M. J. Barthel, T. Rudolph, A. Teichler, R. M. Paulus, J. Vitz,
S. Hoeppener, M. D. Hager, F. H. Schacher, U. S. Schubert,
Adv. Funct. Mater. 2013, 23, 4921.
[4] D. J. Lipomi, M. Vosgueritchian, B. C. Tee, S. L. Hellstrom,
J. A. Lee, C. H. Fox, Z. Bao, Nat. Nanotechnol. 2011, 6, 788.
[5] T. S. Coope, U. F. Mayer, D. F. Wass, R. S. Trask, I. P. Bond, Adv.
Funct. Mater. 2011, 21, 4624.
[6] M. Burnworth, L. Tang, J. R. Kumpfer, A. J. Duncan, F. L. Beyer,
G. L. Fiore, S. J. Rowan, C. Weder, Nature 2011, 472, 334.
[7] J. H. Park, P. V. Braun, Adv. Mater. 2010, 22, 496.
[8] B. Ghosh, M. W. Urban, Science 2009, 323, 1458.
[9] S. R. White, N. Sottos, P. Geubelle, J. Moore, M. R. Kessler,
S. Sriram, E. Brown, S. Viswanathan, Nature 2001, 409, 794.
[10] S. D. Bergman, F. Wudl, J. Mater. Chem. 2008, 18, 41.
[11] Y. Yuan, T. Yin, M. Rong, M. Zhang, Express Polym. Lett. 2008,
2, 238.
[12] Y. Yang, M. W. Urban, Chem. Soc. Rev. 2013, 42, 7446.
[13] G. Ciamician, P. Silber, Ber. Dtsch. Chem. Ges. 1902, 35, 1992.
[14] C. Graf, W. Schärtl, N. Hugenberg, Adv. Mater. 2000, 12, 1353.
[15] N. K. Mal, M. Fujiwara, Y. Tanaka, Nature 2003, 421, 350.
[16] K. Hara, T. Sato, R. Katoh, A. Furube, Y. Ohga, A. Shinpo,
S. Suga, K. Sayama, H. Sugihara, H. Arakawa, J. Phys. Chem. B
2003, 107, 597.
[37] J. Ling, M. Z. Rong, M. Q. Zhang, Polymer 2012, 53, 2691.
[38] R. H. Kim, H. S. Shapiro, J. J. Li, J. L. Wrana, J. Sodek, Matrix
Biol. 1994, 14, 31.
[39] C. Chun, H. J. Lim, K.-Y. Hong, K.-H. Park, S.-C. Song,
Biomaterials 2009, 30, 6295.
[40] K. X. Wang, D. T. Denhardt, Cytokine Growth Factor Rev.
2008, 19, 333.
[41] R. Aschaffenburg, J. Mullen, J. Dairy Res. 1949, 16, 58.
Macromol. Biosci. 2016, DOI: 10.1002/mabi.201500457
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
10
www.MaterialsViews.com
Early View Publication; these are NOT the final page numbers, use DOI for citation !!