European Polymer Journal 189 (2023) 111975

Contents lists available at ScienceDirect

European Polymer Journal

journal homepage: www.elsevier.com/locate/europolj

Dual composition Chondroitin Sulfate and gelatin biomimetic hydrogel

based on tyramine crosslinking for tissue regenerative medicine
Tien Thinh Nguyen a, b, 1, Le Hang Dang a, c, 1, Phuong Nguyen d, *, Truc Le-Buu Pham e,
Hai Khoa Le a, Minh-Ty Nguyen f, Tran Thi Yen Nhi g, Sijia Feng h, Jun Chen h, Ngoc
Quyen Tran a, c, *
a
Graduate University of Science and Technology, Vietnam Academy of Science and Technology, Hanoi 100000, Viet Nam
b
Faculty of Medicine-Pharmacy, Tra Vinh University, Tra Vinh City 87000, Viet Nam
c
Institute of Applied Materials and Science, Vietnam Academy of Science and Technology, Ho Chi Minh City 700000, Viet Nam
d
Faculty of Chemical Technology, HCMC University of Food Industry, Ho Chi Minh City 700000, Viet Nam
e
Department of Animal Biotechnology, Biotechnology Center of HCM City, Ho Chi Minh City 700000, Viet Nam
f
Institute of Chemical Technology VAST, 1A Thanh Loc 29 Str., Dist. 12, Ho Chi Minh City, Viet Nam
g
Institute of Applied Technology and Sustainable Development, Nguyen Tat Thanh University, Ho Chi Minh City 700000, Viet Nam
h
Sports Medicine Institute of Fudan University, Department of Sports Medicine, Huashan Hospital, Fudan University, Shanghai 200040, China

A R T I C L E I N F O A B S T R A C T

Keywords: Hydrogels with bone biomimetic microenvironment are of great importance for the development of bone-graft
Biomimetic hydrogel materials. Herein, a biomimetic hydrogel, which was based on biopolymer as an extracellular matrix (ECM)
Biomineralized hydrogel mimicking components with the incorporation of biphasic calcium phosphate (BCP) nanoparticles, was prepared
Chondroitin sulfate
via the enzymatic crosslinking. Chondroitin sulfate and gelatin were first modified to form tyramine (TA)
Gelatin
HRP enzyme
moieties on the natural polymers, which were ultilized as precursors for in-situ forming hydrogel (CDTA-GTA)
under physiological conditions. The amount of CDTA to GTA was optimized for gelation time as well as degraded
ability with in collagenase medium. Increasing the CDTA amount delayed the time for crosslinking process, but
the water holding capacity was increased. The subjection of BCP into CDTA-GTA hydrogel produced an injectable
system with suitable biomineralizated property. This hydrogel formed more rapidly in 90–100 s under 3 mM
H2O2 and 0.125 mg/mL HRP enzyme. BCP/CDTA-GTA hydrogel exhibited the porous structure (100–300 µm in
diameter) with the homogenous distribution of BCP nanoparticles on the wall of the interconnected networks,
suggesting the good environment for bone growth. Further, the degradation process of the obtained hydrogel
followed pseudo-first-order kinetics in the collagenase condition. After soaking in SBF for 28 days, SEM following
EDS technique confirmed the aggregation of the calcium and phosphorus on the surface of CDTA-GTA hydrogel
(Ca/P ratio ~ 1.34) which improved in case of BCP/ CDTA-GTA hydrogel (Ca/P ratios ~ 1.77). Through X-ray
diffraction, the formation of calcite crystals was only detected with BCP hydrogel. In vitro cytotoxicity assays
with hMSC cells exhibited non-toxicity of the biomimetic hydrogel. Interestingly, the remarkably enhancement
of the level intracellular calcium deposit as well as the activity of ALP-osteogenic related marker were observed
in MSC cells treated with biomimetic hydrogel, showing superior scaffold osteogenic potential. These results
indicate that enzymatic cross-linked CDTA-GTA gel with BCP nanoparticles would be a favorable option to
accelerate bone repair and thus should potentially be useful for osteochondral tissue engineering.

1. Introduction forming by water soluble polymers, have received much attention in

recent years [1,2,3,4]. Tunable physical, biological properties of poly­
Polymeric hydrogels, the porous three-dimensional (3D) network mer scaffold by the versatile precursor polymer materials make them

* Corresponding author.
E-mail addresses: [email protected] (T.T. Nguyen), [email protected] (L.H. Dang), [email protected] (P. Nguyen), [email protected].
vn (N.Q. Tran).
1
Equal contribution.

https://doi.org/10.1016/j.eurpolymj.2023.111975
Received 13 October 2022; Received in revised form 25 February 2023; Accepted 4 March 2023
Available online 7 March 2023
0014-3057/© 2023 Published by Elsevier Ltd.
T.T. Nguyen et al. European Polymer Journal 189 (2023) 111975

excellent scaffold for tissue regeneration [2]. Among of polymeric hydrogel with BCP in the formation of apatite [33]. Therefore, in the
hydrogel, hydrogel based chemical crosslinking method provides higher present study, we focused our attention on the potency of enzymatic
stability and mechanical properties than the hydrogel with the physical cross-linked tyramine hybrid hydrogel containing gelatin and chon­
method [3, 4]. Therefore, chemical hydrogels are more suitable for bone droitin sulfate in case of template for mineralization.
regeneration. However, from the viewpoint of cyto-compatibility, there
are a couple of shortcomings due to the polymerization process, 2. Materials and method
including photo-initiators and irradiation, chemical cross-linker or
metal ions which are potentially detrimental to cell survival, causing 2.1. Materials
tissue damage and deactivating the incorporated proteins [5,6]. In
recent years, enzyme-mediated in situ forming hydrogel has become Chondroitine sulfate, gelatin type A from porcine skin (Bloom 300),
popular for the rational design of chemical crosslinking hydrogels tyramine (TA), 1Ethyl-3-(3 dimethylaminopropyl) carbodiimide (EDC)
because it involves the use of a mild cross-linking reaction, making it were obtained from Acros Organics (USA). Collagenase and horseradish
compatible with living cells, therapeutic proteins, and drugs [3,7,8]. peroxidase (HRP) enzymes (type VI, 298 purpurogallin unit/mg solid)
In this strategy, the use of peroxidases such as horseradish peroxi­ were purchased from Sigma-Aldrich (USA). Biphasic calcium phosphate
dase (HRP) is one of the most studied [7,9]. This enzyme catalyzes a (BCP) nanoparticles was synthesized at Institute of Applied Materials
variety of oxidative transformations using hydrogen peroxide forming [27]. All the solvents were supplied by Labscan (Thai Lan) and used
the radical moieties which are simultaneously formed the crosslinking without purification.
between these polymer chain, resulting the inject-ability for the ob­ For cell culture, human mesenchymal stem cells was come from
tained hydrogel [10]. These type hydrogel, can be used by minimally Lonza Bioscience (Basel, Switzerland). Culture media: Minimum
invasive treatment to fill irregular parts, showing great application Essential Media (MEM), glutamine, Penicillin-Streptomycin, trypsin-
prospects in bone tissue engineering [11]. EDTA and Phosphate-buffered saline (PBS, without calcium and mag­
In bone regeneration, great efforts have been made to improve the nesium) was originated from Gibco (Thermo Fisher Scientific – USA).
adhesion capacity of hydrogels [12,13]. Among all the polymer mate­ Other supplements such as Fetal Bovine Serum (FBS), sodium bicar­
rials examined thus far, chondroitin sulfate (CD), a glycosaminoglycan bonate was were obtained from Sigma. MTT kit, staining dyes (AO/PI/
that can be found in the non-collagenous ECM of human bone, has been Hoechst) and ALP activity colorimetric kit were provided by Abcam.
become the spotlight for cartilage tissue engineering [10]. CD is a
sulfated anionic polysaccharide of repeating units, (1,4)-linked glucos­ 2.2. Preparation of tyramine functionalized Chondroitine sulfate (CDTA)
amine and uronic acid residues and is known to bind growth factors and tyramine functionalized gelatin (GTA)
involved in bone healing process [14,15]. Because of the highly abun­
dant in carboxyl group and sulfonate groups on CD backbone, CD could CDTA and GTA derivatives were prepared from chondroitin sulfate
act as a materials having high apatite-forming ability in body environ­ and gelatin using carbodiimide coupling reagent, as illustrated in Fig. 1.
ment to show osteoconductivity [16]. In addition, it has been revealed For CDTA, chondroitin sulfate (2.0 g) was dissolved in 80 ml distilled
that glucosamine -rich unit in CD, facilitates cell adhesion [17] and (DI) water. TA (1.0 g, 7.3 mmol) was dissolved in 10 ml. pH of solution
promotes osteoblast proliferation [18]. By modifying with some was adjusted to 6 with HCl 5%. A solution of EDC (0.8 g, 4 mmol) and
reducing substrates of HRP such as tyramine, catechol or phloretic acid, NHS (0.5 mg) in 5 ml DI water was added to the chondroitin sulfate
various in situ forming CD hydrogels under catalyzed process of HRP solution and activated in 15 min. Then tyramine solution was drop-
have been introduced for bone scaffold [7,19]. However, the enzymatic wised in the activated chondroitin sulfate solution under stirring for
crosslinking of phenol functionalized CD in physiological buffers such as 24 h at room temperature. The solution was dialyzed against DI water
phosphate buffered saline (PBS) or cell growth media was found to be using membrane dialysis (Molecular Weight Cut-Off (MWCO) 12 – 14
much slower leading to the limitation for fabrication of injectable kDa) in pH 7.4 for 3 days and in continuous stirring condition at room
hydrogel based CD [7]. Faster gelation of material based CD can be temperature for 4–5 days. Finally, the solution after dialysis was freeze-
achieved by increasing the concentrations of HRP and/or H2O2. How­ dried to obtain a white solid product.
ever, high concentrations of HRP [20] and H2O2 [21] may cause po­ For GTA, the pH of the solution of gelatin (2 g) and TA (1.0 g, 7.3
tential immune responses in vivo and apoptotic or necrotic cytotoxicity. mmol) in DI water (30 ml) was adjusted to 6 by addition of EDC (0.50 g,
Increased the density of reducing substrate on CD backbone was also 2.5 mmol) under stirring for 24 h. Then, the solution was dialyzed
found to increase gelation speed [22], but viscosity of the hydrogels may against deionized water using membrane dialysis (MWCO 6000–8000)
raise mechanical properties concerns [10]. A simple way to fabricate for 3 days.
enzymatic crosslinking CD hydrogel with tunable gelation time is to
incorporate another polymeric component resulting in the hybrid 2.3. Hybrid hydrogel formation and gelation time
hydrogel [10,23]. Based on the previous reports with in situ phenol-
substituted polymers during HRP-mediated crosslinking [24,25,26] as The preparation conditions for gelatin or chitosan-based hydrogels
well as adhesive behavior of gelatin, here we proposed that hybrid were screened to determine a suitable setup for further works. Briefly,
hydrogels based CD can be supplemented with tyramine functional the solution of GTA (20 mg) in DI water (150 µl) was divided equally
gelatin to improve gelation rate and cytocompatibility as well as enzy­ into two vials. Enzyme HRP (40 µl of 0.05 mg/mL stock solution) and
matic degraded profile. H2O2 (40 µl of 0.05–0.4% wt/vol stock solution) were respectively
Along with suitable scaffold, many efforts have been dedicated to added into each tube. Then, two vials were mixed continuously to form
mimic the environment for bone regeneration. In the context of bio­ the gelatin-based hydrogels. In hydrogel, GTA polymer concentration
mimetic bone scaffolds, hydrogel − inorganic composite materials are was 10% wt/wt. A similar process was conducted for chondroitin-based
particularly promising material candidates [27,28]. Among of inorganic hydrogels’ preparation in which 40 mg CDTA was dissolved in 150 µl DI
materials, biphasic calcium phosphate (BCP), which consists of hy­ water with HRP (40 µl of 0.125 mg/mL stock solution) and H2O2 (40 µl
droxyapatite (HAp) and tricalcium phosphate (TCP) phases, is consid­ of 0.05–0.4% wt/vol stock solution). Polymers solution was obtained at
ered to be a more efficient ceramic material and an ideal bone substitute the final concentration of 10 % wt/wt. In situ GTA-CDTA hydrogels was
[29,30], due to its controllable degradability and favorable biological formed when mixing solution A (consist of GTA, CDTA and HRP) with
properties, such as having a high bioresorption rate. The incorporation solution B (contained GTA, CDTA and H2O2) at equal volume of the
of BCP in the hydrogel scaffold could approach a goal of biomimetic precursor polymer solution as demonstrated in the above experiments.
mineralization [31,32]. There have been previous studies on the role of The gelation time was determined by using the vial tilting method.

2
T.T. Nguyen et al. European Polymer Journal 189 (2023) 111975

Fig. 1. Synthetic scheme of CDTA (A) and GTA (B).

The process as above was also applied to prepare BCP-dispersed The media containing dye was discarded and replaced by fresh media.
GTA-CDTA hydrogels which contained 10 wt/wt% of the BCP. The The behavior of MSC cells treating with the extracted hydrogel was were
concentration of H2O2 was changed from 0.075 to 0.4 % wt/vol to observed under laser confocal microscopy (Andor, England) at 3 chan­
characterize the gelation time behaviors of the hydrogels and injectable nels: 350/461 (Hoestch), 490/525 (AO), 490 / 636 (PI).
nanocomposite hydrogels (BCP/GTA-CDTA). Morphology of the ob­
tained hydrogel was observed by Scanning Electron Microscopy (SEM)
2.5. In vitro mineralization and degradation
at the freeze-dried form. Concentration of HRP and H2O2 were adjusted
depending on weighted amount of the conjugated TA moieties in the
2.5.1. Collagenase-mediated degradation study
modified polymers. The enzyme-mediated hydrogels contained
The weight-losses of the hydrogels and nanocomposite hydrogel
approximately 10 wt/wt% of the polymer concentration.
immersed in phosphate buffer saline (PBS) solution pH 7.4 containing
collagenase (0.2 U/ml) at 37 ◦ C were monitored at different incubation
2.4. Cytotoxicity test times. Before being immersed in 1 ml of enzymatic solution, samples
with different mass ratios were accurately weighted (wi). At the deter­
Humam mesenchymal stem cells (MSCs, Lonza) were used for the in mined time, samples were removed from the incubation medium and the
vitro assay following the previous study [24]. MSCs were cultured in remaining weights were measured. The change in mass of hydrogel after
MEM supplemented with 10% FBS, 2 mM l-glutamine, and 1% pen­ soaking in compared to initial wet mass was used to determine water
icillin–streptomycin in a 5% CO2 incubator at 37 ◦ C. MSCs cells were holding capacity of hydrogel.
seeded in 96 well plate at the density of 1 × 104 cells per well. After 24 h
Wi − Wt
culture, the old media was withdraw and then replaced by the media % mass change = × 100%
Wi
containing the extracted BCP/CDTA-GTA hydrogel. At the determined
time, 24 h and 48 h, MTT assay (abcam) was applied to determine the The kinetic degradation of hydrogels in PBS buffer with the supple­
viability of cell through ELISA reader. For staining test, 1 × 105 MSC ment of collagenase was determined using the following formula:
cells were seeded in 12 well plates, and then treated with the extracted ( )
2.303 Wt
BCP/CDTA-GTA hydrogel. After 48 h, Hoestch 33,342 (Thermo Fisher, Rate constant (k) = ln (1)
t Wi
5 ppm in PBS buffer) was added into each wells and incubated for 15
mins in the incubator. Then, the media was removed and the wells were 0.693
washed with PBS 1x (Gibco) for 3 times before adding the new media. Half − life = (2)
k
The mixture dye in PBS (AO/PI) was added and incubated for 5 mins.

3
T.T. Nguyen et al. European Polymer Journal 189 (2023) 111975

incubated for 60 min. Detaining solution was added, the nodule for­
2.5.2. Biomineralization study
mation was examined by light microscope and well as the digital
Bio-mineralization is one of the most important elements to access
camera.
the potential of materials for bone regeneration. To identify bio-
mineralization of the resultant hydrogels (CDTA-GTA hydrogel and
2.5.3. In vitro osteogenic differentiation
BCP/CDTA-GTA hydrogel), two methods were involved in this study.
ALP activity was performed with ab83369. Briefly, MSC cells (2 ×
Mineralization study in SBF media: these materials were immersed in
105) culturing with samples (CDTA-GTA or BCP/CDTA-GTA) in 14 days
a SBF buffer solution (pH 7.4) in 4 weeks then collected and washed with
was harvested and washed with cold PBS without calcium and magne­
distilled water to remove soluble inorganic salts. After that, SEM, Energy
sium. 100 μl assay buffer was added following the homogenized pro­
Dispersive X-ray analysis (EDX) and X-ray diffraction (XRD) methods
cedure on the ice box. The centrifugation at 10.000 rpm was applied for
were applied to analyze the bio-mineralization ability of the hydrogel.
5 min. 100 μl each supernatant was transfer to 96-well plate. ALP assay
Mineralization study with MSC cells: The minimization potential of
was then applied following the guidance of Abcam.
these materials was identified via alizarin red staining techniques. MSC
cells (2 × 105) was mixture with hydrogel precursor in the eppendorf
2.6. Statistical analysis
before adding HRP enzyme and H2O2. 300ul gel solution was quickly
transferred to 48-well plate, and incubated in cell culture condition for
All quantitative data are presented as mean ± SD and the quantita­
30 min. Then, 300ul culture medium was added to maintain the culture
tive experiments were carried out at least in quadruplicate. Differences
system. The culture media containing osteogenic differentiation (Mes­
between the experimental and the control group were analyzed by one-
enCult™ Osteogenic Differentiation Kit) was replaced the older media
way analysis of variance, and *p < 0.05 was considered as statistically
after 2 days. The MSC culture without scaffold was used as the control.
significant. Statistical analysis and data processing were performed
At the designed time, the medium was removed and washed with PBS 1X
using GraphPad Prism 9.0 and Origin Pro. 2022b.
(Cas: 14190144). 1 ml 4% paraformaldehyde was applied for 1 h at
room temperature. 1 ml alizarin red (2%, pH 4.3) was used and

Fig. 2. Chemical structure of tyramine functional chondroitin sulfate (CDTA) analyzing via FTIR (A) and 1H NMR (B).

4
T.T. Nguyen et al. European Polymer Journal 189 (2023) 111975

3. Results on IR of CDTA at 1563 cm− 1, 1370 cm− 1 and 1158 cm− 1, indicating the
existence of phenyl group on the modified CD. The evidence of tyramine
3.1. The synthesis of gelatin-tyramine (GTA) and chondroitin – Tyramine substitution on CD was also observed by 1H NMR as shown in Fig. 2b.
(CDTA) Several class of proton assigning to native CD such as acetamine methyl
protons (H6 δ = 2.01 ppm), anomeric proton H1 (δ = 4.60 ppm), H2-H3
Two tyramine functional polymers, gelatin-tyramine (GTA) and (GaINAC, δ = 3.98 ppm), GlcUA H4 – GaINAC H5 (δ = 3.79 ppm), GlcUA
Chondroitin – tyramine (CDTA), were synthesized through carbodiimide H3 (δ = 3.57 ppm), similar to the signals seen in previous studied
coupling method as schematically represented in Fig. 1. The side-chain [34,35]. Along with the common chemical shift of native CD, the
carboxylic acid on chondroitin sulfate (CD) or gelatin backbone was spectrum of CDTA showed the the distinctive signals for both pairs of
activated with EDC-NHS chemistry for direct conjugation to primary aromatic ring protons at δ = 7.2 ppm and δ = 6.8 ppm, and signal proton
amines of tyramine (TA). The successful modification process was of aliphatic side chain at δ = 2.7 ppm, confirming the formation of
proved by 1H NMR, FT-IR, while the degree of substitution was esti­ tyramine functional CD. The degree of substitution of CD with tyramine
mated by UV–Vis. was identified by ultraviolet–visible (UV–Vis) at 275 nm using a cali­
The chemical structure of CDTA was presented in Fig. 2. The FT-IR bration curve of tyramine in Milli-Q water. The amount for tyramine on
spectrum of CDTA (Fig. 2A) showed two distinct absorption bands CD is 4.8 × 104 mol /mg. The degree of substitution of CD with tyramine
assigned to the vibration of hydroxyl group (–OH, 3608 cm− 1) and N–H was 22.2, which was similar as the result obtaining from 1H NMR (DS =
bending (3322 cm− 1), while both signals are overlapped in case of native 22.17).
CD, with a maximum at 3463 cm− 1. Below the IR region 3000 cm− 1, the For tyramine functional gelatin (GTA), the functional groups of
shift in the bending vibrations of the N–H (glycosaminoglycan amide I) native gelatin and GTA are similar; therefore, IR spectrum of GTA is
and and sulfate S = O absorption are observable on IR of CDTA. The IR closely resembled that of natural gelatin (Fig. 3A). However, the fre­
spectra of CD shows amide I at wavenumber of 1646 cm− 1, which is red quency shifts have been detected in spectrum of GTA. The observed shift
shift to 1658 cm− 1 in IR spectra of CDTA. The S = O asymmetric of the amide I band and amide II band towards lower wavenumbers,
stretching vibration shows blue shift from 1248 cm− 1 to 1232 cm− 1 after 1627 cm− 1 and 1520 cm− 1, respectively. The red shift of these amide
modifying with tyramine. In addition, the C–C aromatic stretch, –OH band is expected due to the participation of unsaturated carbon–carbon
bending in phenol group, and C-O stretching vibration is only observable in phenyl group of tyramine. 1H NMR spectrum of GTA expresses the

Fig. 3. Chemical structure of tyramine functional gelatin (GTA) analyzing via FTIR (A) and 1H NMR (B).

5
T.T. Nguyen et al. European Polymer Journal 189 (2023) 111975

substitution of tyramine on gelatin backbone (Fig. 3B). The chemical tyramine functional polymer (10 wt%) and of HRP enzyme (0.125 mg/
shift of methyl proton of amino acid such as leucine, valine, isoleucine ml), the gelation rate of the enzymatically cross-linked polymer tyra­
(δ = 0.84 ppm), threonine (δ = 1.18 ppm), alanine (δ = 1.37 ppm), mine conjugate hydrogels could be controlled by changing H2O2 con­
arginine (δ = 1.64 ppm), and the aromatic proton of phenylalanine (δ = centration (Fig. 4). With lowest H2O2 concentration (0.075%, the final
7.31 ppm) occur on the spectrum of GTA. The two predominant peaks concentration was 3 mM), the gelation time was within 1–2 min,
for phenyl group of tyramine (δ = 7.23 ppm and δ = 6.79 ppm) along whereas the gelation time was reduced to 20–50 s when concentration of
with the proton signal of methylene (–CH2) on the side chain of tyramine H2O2 increased upto 0.4 mM. Along with H2O2 concentration, the ratio
(δ = 2.67 ppm and δ = 3.69 ppm) are assignable, implying successful between CDTA and GTA also have the strong impact to gelation time.
conjugation of TA to the gelatin backbone. In this one pot synthesis, the Despite the higher amount tyramine per mg in CDTA, the crosslinking
substitution was 6.99 × 10− 5 mol tyramine/mg for gelatin. The degree efficacy in 10 wt% CDTA hydrogel was lower than 10 wt% GTA hydrogel
of substitution of gelatin with tyramine was 45.25, which was two times (90%CI 7.475 to 32.53, p < 0.0001). The pure CDTA hydrogel was
higher than CDTA. formed within 102 ± 5 s with 0.075 mM H2O2. In the same manner, pure
GTA solution required 82 ± 8 s to transfer to solid stage. In the mixture
of GTA and CDTA, the gelation time was reduced if the concentration of
3.2. In situ gelation performance of biomimetic CDTA-GTA hydrogel GTA increased. The reduced gelation time from 94 ± 3 s to 81 ± 4 s
(90%CI 0.9408 to 25.06, p = 0.020) was observed when GTA increased
CDTA-GTA hydrogel were rapidly formed through the oxidative from 5% to 7%, suggesting the favorable crosslinking interactions to
coupling of tyramines under the help of horseradish peroxidase (HRP). GTA than that of CDTA. The longer time for hydrogel based CD has been
The mixture of CDTA solution and GTA solution with H2O2 was reported in the previous study [10,19]. This behavior could be explained
remained in the liquid stage and then then switched to solid stage by by the unfavorable interactions with the active site of the enzyme caused
adding HRP enzyme. For enzymatic crosslinking procedure, the gelation by the more acid groups in CS chains (e.g OSO3− and COO–) either
time is very importance. Control of gelation time possess several ad­ through steric hindrance or charge interactions [9,38].
vantages for drug delivery vehicles and injectable scaffold applications The biomimetic hydrogel was prepared with BCP nanoparticles
[36]. Fast gelation times are required for in situ local drug delivery to within the basic scaffold of CDTA-GTA hydrogel. The gelation was
prevent the diffusion of hydrogel precursor solution, drugs and growth observed after adding to the mixture CDTA/BCP and GTA/BCP with the
factors out of the target site or the surrounding hydrogel formation [37]. subjection of horseradish peroxidase crosslinking in the presence of
On the other hand, slower gelation times are desirable in injectable H2O2. The gelation time of BCP/CDTA-GTA was total depended on the
scaffold applications for tissue engineering to correctly fit the defect site concentration of H2O2 or the ratio between CDTA and GTA as similar to
[8]. In our system, it is necessary to use minimum H2O2 concentrations native CDTAGTA. The increase in gelation time of biomimetic hydrogel
since the remaining H2O2 can induce cytotoxicity effects [21]. During containing more CDTA polymer or lower concentration H2O2. The
the hydrogel formation, H2O2 changes to H2O and oxygen after incorporation of BCP nanoparticles also showed the impact to the
decomposition via the enzymatic reaction [10,23]. Therefore, to obtain gelation rate. Generally, the incorporation of BCP delays the gelation
high cell viabilities, the remaining H2O2 in sample need to be negligible. time of CDTA-GTA. Despite the increase of H2O2 concentration from
Therefore, in this study, optimizing H2O2 concentration as well as the 0.075 to 0.4 % wt/vol, the gelation time of BCP/ CDTA-GTA were longer
ratio CDTA/GTA had been determined. Under the constant amount of

Fig. 4. Gelation time of CDTA-GTA hydrogel with BCP (black bar) and without BCP (orange bar) as a functional CDTA/GTA weight ratio and H2O2 concentration
(0.075 to 0.4%, the final concentration of H2O2 was 3 mM to 15 mM), while the final concentration of polymer, HRP were kept at 10 wt% and 0.125 mg/mL,
respectively. Data were in form of mean ± SD (n = 4). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of
this article.)

6
T.T. Nguyen et al. European Polymer Journal 189 (2023) 111975

than that of the native CDTA-GTA. The effect of BCP on gelation time electron microscope (SEM). Native CDTA-GTA composite hydrogel dis­
was strongly positive correlation to CDTA polymer compared to GTA played porous-like structures (Fig. 5a) with randomly oriented pores
polymer. Under the concentrations of 0.125 mg/mL HRP and 0.075% ranging in size from 50 to 350 μm and a mean pore size of ~150 μm as
H2O2, the solution of pure CDTA switched to solid stage after 102 ± 5 s calculated from ImageJ analysis (Fig. 5b). The addition of biphasic
and after 120 ± 6 s if 10% BCP was added (95%CI − 27.26 to − 8.744, p calcium phosphate (BCP) did not induce any change in the network of
< 0.0001). In case of pure GTA hydrogel, the gel without BCP was the native CDTA-GTA hydrogel. The BCP/CDTA-GTA hydrogel were
formed within 82 ± 8 s, which were quite related to BCP/GTA hydrogel highly porous with a 3D network of interconnected pore structures (4c)
(t = 84 ± 12 s, 90%CI − 18.15 to 14.15, p > 0.999). In term of composite as similar to CDTA-GTA hydrogel. With higher magnification, the sur­
hydrogel, the difference in gelation time between gel with BCP and face of the pore walls contained BCP granules with well–distribution
without BCP was 18 ± 0.13 s for 5 wt% CDTA (90%CI − 34.59 to (Fig. 3e-f). Despite the indistinguishable 3D network, the incorporation
− 1.414, p = 0.0229) and 4 ± 0.14 s for 3% CDTA (90% CI-21.10 to of BCP nanoparticles into hydrogel caused slight changes in the porous
9.098, p = 0.9648). This pattern was repeated to all tested H2O2 con­ microstructure of CDTAGTA hydrogel. The mean of pore size of CDTA-
centration range. GTA hydrogel with BCP shifted to 220 μm with the larger distribution
The structure of the resultant hydrogels were examined via scanning ranging from 50 to 400 μm (Fig. 5d), proposing the good scaffold for

Fig. 5. Scanning Electron Microscope images of (A) CDTA-GTA hydrogel and biomimetic CDTA-GTA hydrogel (C) along with their pore size distribution (B,D). The
distribution of BCP nanoparticles inside CDTA-GTA hydrogel network at the high magnification (x2000) of SEM image (E, D).

7
T.T. Nguyen et al. European Polymer Journal 189 (2023) 111975

bone regeneration [13,39]. BCP/CDTA-GTA hydrogel had a thicker wall groups and with oxygen atom in sulfonic group [40], resulting in the
than that of non BCP hydrogel. Furthermore, a secondary network was introduction of the second network in BCP/CDTA-GTA hydrogel. This
formed on the basic of CDTA-GTA hydrogel network after adding BCP hypothesis is consistence to gelation time data. Due to the dative bond
nanoparticles, suggesting the interactions between BCP particles and between CDTA and BCP nanoparticles, the interaction between tyra­
tyramine functional polymer chains. As compared to structure of CD, mine sides on the polymer backbone has a tendency to decrease leading
calcium ion in BCP nanoparticles might form the complex with CD to the delay in the gelation process in BCP hydrogel.
structure via the coordinative bond with oxygen atom in carboxyl

Fig. 6. The change in the mass of CDTA-GTA hydrogel (A) and BCP/ CDTAGTA hydrogel (B) after soaking in PBS buffer supplement with collagenase enzyme (16
mg.dm− 3) for 3, 6, 18, 42 and 90 h at room temperature, pH 7.4, [ion]total = 0.15 M, [polymer]total = 10wt/vol %. Data was presented as mean ± SD (n = 5).

8
T.T. Nguyen et al. European Polymer Journal 189 (2023) 111975

3.3. Kinetic degradation of biomimetic CDTA-GTA hydrogel was longer than that of non BCP composite hydrogel. Interestingly, the
dissociation of GTA hydrogel was remarkable reduced following the
The degradation profiles of CDTA-GTA hydrogel and BCP/ CDTA- addition of BCP nanoparticles. Only 74.52 ± 6.71% hydrogel loss after
GTA hydrogel were identified in the physiological buffer supplement 42 h.
of collagenase enzyme (16 mg/dm3). The analysis of the wet mass Further, the degradation profiles of different hydrogels were illus­
change of hydrogel showed that the ratio between CDTA and GTA trated on Fig. 7. The correlation curves of the natural log of remaining
polymers affected the resorption time of the resultant hydrogel (Fig. 6A). weight percent against incubation time in all groups were approximately
It was observed that sample with higher amount of CDTA displayed the linear with R2 value > 0.9, indicating that the degradation of the
higher in water holding capacity. The mass of pure CDTA hydrogel hydrogels followed the pseudo-first-order degradation kinetics. The
increased 19.99 ± 5.31% at first 3 h and reached the equilibrium after degraded rate constant for pure GTA hydrogel to the initial degradation
18 h. No change in the mass of pure CDTA hydrogel was detected until time period was found to be 0.037 h − 1 (t1/2 = 20.3 h) while rate
42 h (p = 0.9897). In the contrast, pure GTA hydrogel was not able to constants for cleavage of pure CDTA hydrogel was 17 times lower (k =
store water in the structure. The reduction in mass of hydrogel was taken 0.0021 mg.h-1. T1/2 = 187.3 h). The observed kinetics degradation for
after 3 h soaking and strongly dissociation in immersed collagenase CDTA (5%) - GTA (5%) hydrogel was at the rate constant of 0.0036 h-1
medium after 42 h with 93.13 ± 3.81% was recorded. The combination (t1/2 = 101.9 h) which was an order of magnitude larger than that
of CDTA and GTA showed the positive feature in the control the water observed for CDTA (3%)- GTA (7%) hydrogel. There was discernable
holding capacity of hydrogel. The addition of 3% CDTA in the GTA difference measured between non BCP hydrogel and BCP hydrogel. The
hydrogel increased the retained water in the network of GTA hydrogel. degraded rate of BCP/CDTA hydrogel was 0.0037 h-1 with the half-life
During 3–42 h, non-significance change in mass of hydrogel in respected of 330 h ~ 13.75 days- 1.8 times longer than non BCP hydrogel (p <
to initial mass was identified (all p > 0.05), suggesting the water mol­ 0.002). This tendency was well matched with nanocomposite CDTA-
ecules embedded within hydrogel structure. When the amount of CDTA GTA hydrogel. The degradation rate of BCP/CDTA-GTA hydrogel was
upto 5 wt%, the resultant hydrogel showed the increase in mass from 2 down to 75% or 50% in respected to non-BCP CDTA-GTA hydrogel.
to 8% during the first 18 h. The reduction in mass of hydrogel with 5%
CDTA at 90 h was 39.48 ± 11.94%, which was a half amount of hydrogel 3.4. In vitro cytotoxic of biomimetic CDTA-GTA hydrogel
with 3% CDTA at the same time (p = 0.0021). This phenomena was kept
in the biomimetic hydrogel. However, similar to gelation time, the The cytotoxicity activity of biomimetic CDTA-GTA hydrogel was
incorporation of BCP expressed their influence to water holding capacity accessed with MSC cells. In our research, various concentration of BCP/
(Fig. 6B). With addition of 10% BCP, the significant increase in the mass CDTA-GTA hydrogel (the final concentration in the culture well were
of pure CDTA hydrogel was detected after 6 h as compared to pure CDTA 10–1000 ppm) were cultured with hMSC cells for 24 h and 48 h. It was
(p < 0.0001). In case of nanocomposite hydrogel, water entered and showed that the viability of cell culturing with the low concentration of
remained on the network of the resultant hydrogel during 18 h, which BCP/CDTA-GTA hydrogel (under 500 ppm) was similar to the negative

Fig. 7. The first-order kinetic plot for the degradation rate of CDTA-GTA hydrogel (A) and biomimetic CDTA-GTA hydrogel (B) as a functional CDTA/GTA weight
ratio under physiological PBS buffer with collagenase enzyme (16 mg/dm3). The degradation kinetic parameters (C) were calculated from the fitting function by
using OriginLab software.

9
T.T. Nguyen et al. European Polymer Journal 189 (2023) 111975

control at both determined time points (all p > 0.2) (Fig. 8a). A decrease medium, the remained hydrogel was frozen-dried and then evaluated
in cell viability was found for the higher tested concentrations (p < the mineralization process. The phase of the CDTA-GTA hydrogel and
0.001). With the highest tested concentration (1000 ppm), the viability BCP/CDTA-GTA hydrogel after immersion were analyzed by XRD
was 84.13 ± 4.1% after 24 h culture and 84.65 ± 4.74% after 48 h pattern and shown in Fig. 9a. No diffraction peak in the XRD curve of
culture. The MSC cells cultured with 1000 ppm were staining with live/ native CDTA-GTA hydrogel at day 0 nor at day 28. In contrast, CDTA-
dead assay in combination with Hoechst. As shown in Fig. 8b, the dis­ GTA hydrogel with BCP revealed HAp and TCP peaks on its XRD
tribution of MSC nuclei treating with BCP/CDTAGTA hydrogel was pattern. The peaks at 2θ = 25.8◦ (0 0 2), 28.35◦ (1 0 2), 31.78◦ (2 1 1),
indistinguishable to the control. In addition, the intensity of red signal 41.72◦ (1 3 0), and 46.96o (2 2 2) agree well with the diffraction peaks of
corresponding to dead cell was very low in control and treated group. HA crystals (JCPDS No. 9–0432). The peaks of β-TCP appear at 2θ =
The bright field and green field (due to AO dye) revealed that MSC cells 27.14◦ (1 0 1), 28.74◦ (2 1 4), 29.45.2◦ (2 2 2), 30.23◦ (0210), 34.76◦
grew in a confluent monolayer, with a fibroblastic-like polygonal (2 2 0) (JCPDS No. 9–0169). Along with the crystal peaks of BCP nano­
morphology (Fig. 8b). Along with green signal, the presentation of the particles, XRD pattern of BCP/CDTA-GTA hydrogel at day 28 displayed
gel fragments were detected, which may cause the bias result in MTT new peaks at 2θ = 23.20◦ (0 1 2), 28.52◦ (1 0 4), 36.03◦ (1 1 0), 39.48◦
assay. Therefore, the MTT result and live/dead assay were inconsistent. (1 1 3), 43.24◦ (2 0 2), 47.62◦ (0 1 8), 48.61◦ (1 1 6), confirming the for­
In other words, this biomimetic hydrogel were considered non- mation of calcite. Under SEM analysis, the calcite crystals were aggre­
cytotoxic. gated on the surface of BCP/CDTA-GTA hydrogel (Fig. 9B1), while this
phenomena did not appear in case of native CDTA-GTA hydrogel
3.5. Mineralization behavior of biomimetic CDTA-GTA hydrogel (Fig. 9B2). Further, the Energy Dispersive Spectrometer (EDS) spectrum
displayed the element of both hydrogel types, mainly Ca, P and O,
From the degradation kinetics of difference CDTA-GTA hydrogel and consistent with the chemical composition of apatite layer. This
its BCP hydrogel, the hydrogel from CDTA (5%) and GTA (5%) were confirmed that CDTA-GTA hydrogel has ability of calcification. The
used for further study. Before soaking and after 28 days soaking in SBF evidence of the calcium phosphate formation on the native CDTA-GTA

Fig. 8. (A) The viability of MSC cell after 24 h and 48 h incubated with different concentration of biomimetic CDTA-GTA hydrogel extract in PBS buffer in respected
to non-treated cell. (B) Propidium iodide, Hoechst, acridine orange staining of MSC cell treated with 1000 ppm extracted hydrogel for 48 h; control was PBS buffer.
(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

10
T.T. Nguyen et al. European Polymer Journal 189 (2023) 111975

Fig. 9. (A) XRD pattern of CDTA-GTA hydrogel and its biomimetic hydrogel before and after soaking 28 days in SBF solution. (B) SEM image and (C) energy
dispersive spectroscopy (EDS) analysis of hydrogel sample and biomimetic hydrogel at day 28th.

gels without BCP after soaking in SBF indicates that the carboxyl groups intracellular calcium deposits than the non-treated group after 14 days
on CD and in the gel structure may act as a site to make heterogeneous incubation, which further increase with the addition of BCP nano­
nucleation of calcium phosphate at amorphous phase. With the addition particles. Similar to alizarin red staining, alkaline phosphatase (ALP) –
of BCP, the intensity of the peaks for these elements was increased the marker for early osteogenesis, was more significantly active in case
(Fig. 9C2). The Ca/P ratios were 1.34 for native CDTA-GTA hydrogel of BCP/CDTA-GTA compared to other groups (Fig. 10B). The level ALP
and 1.77 for BCP/ CDTA-GTA hydrogel after 28 days soaking. The activity of biomimetic hydrogel was improved 5 times relative to control
resulting Ca/P ratios in biomimetic hydrogel was higher than the ratio in the osteogenic culture media and 2.3 times to bare CDTA-GTA
value of BCP nanoparticles (theory Ca/P ratio in BCP was 1.57), hydrogel. The obtained data suggested that a biomimetic hydrogel
providing the evidence of the aggregation of Ca and P from the SBF on BCP/CDTA-GTA could trigger osteogenic differentiation.
the surface of biomimetic hydrogel [27]. In addition, together with the
signal of apatite nucleation, another element such as Na or S appeared 4. Conclusion
on the elemental map of BCP/ CDTA-GTA. The deposition of sodium on
BCP hydrogel indicates the exchanged process of calcium ion inside the In this study, injectable hydrogel based on the hybrid polymer sys­
hydrogel to sodium ion in SBF medium. The escape of calcium from BCP tem was prepared via enzymatic crosslinking. The naturally occurring
and existence of Ca2+ in SBF medium could be interacted to carboxyl glycosaminoglycan, Chondroitin sulfate, and the derivate collagen,
groups in the gels helps increase in the degree of super-saturation with gelation were first modified with tyramine. The combination of CDTA
respect to the hydroxyapatite to proceed the further crystallization [41]. and GTA greatly improved the holding water capacity of hydrogel,
Thus, the Ca/P ratios in BCP/ CDTA-GTA hydrogel was higher than which is favorable for good nutrient transportation for cell culture. The
expected. combination of afforded CDTA and GTA hydrogels controlled gelation as
Further, to provide more evidence for biomimetic hydrogel BCP/ well as degraded profile. It is found that incorporation of BCP NPs,
CDTA-GTA on bone regeneration, the in vitro osteogenic differentia­ leaded to significant change in gelation time, swelling and kinetic
tion potential of BCP/CDTA-GTA was evaluated by MSC cells. For this, degradation of the nano-composite hydrogel. The nano-composite
MSC cells was laden in hydrogel network before incubating with the hydrogel could be a great injectable scaffold for minimally invasive
osteogenic differentiation media. After 14 days, the mineralized nodule bone regeneration due to the suitable gelation time (~1min), degraded
formations were occurred at all tested samples (Fig. 10A). The treated rate (0.0068 h-1) with the half-live of 192.5 h., the extract of the selected
groups with CDTA-GTA hydrogel scaffolds displayed the higher level hydrogel was considered as nontoxic to MSC cells, confirming via MTT

Fig. 10. In vitro mineralization assay for MSC culture with CDTA-GTA and biomimetic hydrogel BCP/CDTA-GTA at days 14. A) The alizarin red staining with light
microscopy image and their quantitative result of osteogenic efficacy via ALP assay. The control sample was the MSC with pure completed cell culture media. Data
was presented as mean ± SD (n = 3). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

11
T.T. Nguyen et al. European Polymer Journal 189 (2023) 111975

test and live/dead staining assay. Notably, this type of hydrogel formed [10] F. Chen, S. Yu, B. Liu, Y. Ni, C. Yu, Y. Su, X. Zhu, X. Yu, Y. Zhou, D. Yan, Sci. Rep. 6
(2016) 1–12, https://doi.org/10.1038/srep20014.
bone-like apatite layer after 28 days soaking in SBF, detecting by XRD,
[11] J. Meyer, L.E. Meyer, S.J. Kara, Eng. Life Sci. 22 (2022) 165–177, https://doi.org/
SEM/EDS techniques as well as alizarin red staining method. Further­ 10.1002/elsc.202100087.
more, the study with MSC cells through the ALP activity proposed that [12] X. Bai, M. Gao, S. Syed, J. Zhuang, X. Xu, X.-Q. Zhang, Bioact. Mater. 3 (2018)
biomimetic hydrogel promote osteogenic potential. These results indi­ 401–417, https://doi.org/10.1016/j.bioactmat.2018.05.006.
[13] L. Polo-Corrales, M. Latorre-Esteves, J.E. Ramirez-Vick, J. Nanosci. Nanotechnol.
cate that the as-prepared injectable biomimetic hydrogels based on 14 (2014) 15–56, https://doi.org/10.1166/jnn.2014.9127.
hybrids of two type tyramine functional polymers are very promising for [14] H. Wibowo, P. Widiyanti, S. Asmiragani, J. Basic Clin. Physiol. Pharmacol. 32
the development of scaffolds with osteoconduction and bioresorbability. (2021) 631–635, https://doi.org/10.1515/jbcpp-2020-0406.
[15] M. Moss, G.O. Kruger, D.C. Reynolds, Oral Surg. Oral Med. Oral Pathol. 20 (1965)
795–801, https://doi.org/10.1016/0030-4220(65)90143-X.
CRediT authorship contribution statement [16] L. Zhou, L. Fan, F.-M. Zhang, Y. Jiang, M. Cai, C. Dai, Y.-A. Luo, L.-J. Tu, Z.-
N. Zhou, X.-J. Li, Bioact. Mater. 6 (2021) 890–904, https://doi.org/10.1016/j.
bioactmat.2020.09.012.
Tien Thinh Nguyen: Conceptualization, Methodology, Software, [17] T. Miyazaki, S. Miyauchi, A. Tawada, T. Anada, O.J. Suzuki, Dent. Mater. J. 29
Data curation, Writing – original draft, Visualization, Investigation, (2010) 403–410, https://doi.org/10.4012/dmj.2009-132.
Software, Validation, Writing – review & editing. Le Hang Dang: [18] C. Lv, L. Wang, X. Zhu, W. Lin, X. Chen, Z. Huang, L. Huang, S.J.B. Yang,
Pharmacotherapy, Biomed Pharmacother. 99 (2018) 271–277, https://doi.org/
Conceptualization, Methodology, Software, Data curation, Writing – 10.1016/j.biopha.2018.01.066.
original draft, Writing – review & editing. Phuong Nguyen: Data [19] Y. Zhang, H. Chen, T. Zhang, Y. Zan, T. Ni, Y. Cao, J. Wang, M. Liu, R.J. Pei, Mater
curation, Writing – original draft, Supervision. Truc Le-Buu Pham: Sci Eng C Mater Biol Appl. 96 (2019) 841–849, https://doi.org/10.1016/j.
msec.2018.12.014.
Visualization, Investigation. Hai Khoa Le: Visualization, Investigation. [20] F. Lee, K.H. Bae, M.J.B.M. Kurisawa, Biomed. Mater. 11 (2015), 014101, https://
Minh-Ty Nguyen: Visualization, Investigation. Tran Thi Yen Nhi: . doi.org/10.1088/1748-6041/11/1/014101.
Sijia Feng: Software, Validation. Jun Chen: Software, Validation. Ngoc [21] Y. Saito, K. Nishio, Y. Ogawa, J. Kimata, T. Kinumi, Y. Yoshida, N. Noguchi, E. Niki,
Free Radic. Res. 40 (2006) 619–630, https://doi.org/10.1080/
Quyen Tran: Conceptualization, Methodology, Software, Visualization,
10715760600632552.
Investigation, Supervision, Writing – review & editing. [22] J.K. Sahoo, J. Choi, O. Hasturk, I. Laubach, M.L. Descoteaux, S. Mosurkal, B. Wang,
N. Zhang, D.L. Kaplan, Biomater. Sci. 8 (2020) 4176–4185, https://doi.org/
10.1039/D0BM00512F.
Declaration of Competing Interest [23] O. Hasturk, K.E. Jordan, J. Choi, D.L.J.B. Kaplan, Biomater. Sci. 232 (2020),
119720, https://doi.org/10.1016/j.biomaterials.2019.119720.
The authors declare that they have no known competing financial [24] T.P. Ton, V.T. Nguyen, P. Doan, D.T. Nguyen, T.P. Nguyen, C.K. Huynh, T.C.
Q. Ngo, N.Q.J. Tran, New J. Chem. 45 (2021) 18327–18336, https://doi.org/
interests or personal relationships that could have appeared to influence 10.1039/D1NJ01426A.
the work reported in this paper. [25] T.D. Van, N.Q. Tran, D.H. Nguyen, C.K. Nguyen, D.L. Tran, P.T. Nguyen,
J. Electron. Mater. 45 (2016) 2415–2422, https://doi.org/10.1007/s11664-016-
4354-3.
Data availability [26] S. Xu, Q. Li, H. Pan, Q. Dai, Q. Feng, C. Yu, X. Zhang, Z. Liang, H. Dong, X. Cao,
ACS Biomater Sci. Eng. 6 (2020) 6896–6905, https://doi.org/10.1021/
Data will be made available on request. acsbiomaterials.0c01183.
[27] T.P. Nguyen, B.H.P. Doan, D.V. Dang, C.K. Nguyen, N.Q. Tran, Adv. Nat. Sci.
Nanosci. Nanotechnol. 5 (2014), 015012, https://doi.org/10.1088/2043-6262/5/
Acknowledgments 1/015012.
[28] K. Flegeau, O. Gauthier, G. Rethore, F. Autrusseau, A. Schaefer, J. Lesoeur,
J. Veziers, A. Brésin, H. Gautier, P.J.B.S. Weiss, Biomater. Sci. 9 (2021)
We would like to thank Dr. Minh-Dung Truong (Biotechnology 5640–5651, https://doi.org/10.1039/D1BM00403D.
center of Ho Chi Minh City) for osteogenic differentiation medium. [29] S.E. Lobo, T.L.J.M. Arinzeh, Acta Biomater. 3 (2010) 815–826, https://doi.org/
10.1016/j.actbio.2017.01.076.
[30] R.D. Ventura, A.R. Padalhin, B.T.J.M. Lee, Design, Mater. Des 192 (2020), 108653,
Funding https://doi.org/10.1007/s11664-016-4354-3.
[31] S. Yao, Y. Xu, Y. Zhou, C. Shao, Z. Liu, B. Jin, R. Zhao, H. Cao, H. Pan, R.J. Tang,
ACS Appl Bio Mater. 2 (2019) 4408–4417, https://doi.org/10.1021/
This work was supported by Vietnam National Foundation for Science acsabm.9b00270.
and Technology Development (NAFOSTED) under grant number 104.02- [32] A. Ayala-Ham, J. Lopez-Gutierrez, M. Bermúdez, M. Aguilar-Medina, J.
2017.60. I. Sarmiento-Sánchez, C. López-Camarillo, G. Sanchez-Schmitz, R.J. Ramos-Payan,
Front. Mater. (2021) 294, https://doi.org/10.3389/fmats.2021.708945.
[33] L. Nie, Y. Deng, P. Li, R. Hou, A. Shavandi, S.J.A.O. Yang, ACS Omega 5 (2020)
References 10948–10957, https://doi.org/10.1021/acsomega.0c00727.
[34] B. Diehl, NMR applications for polymer characterisation, NMR spectroscopy in
[1] L.H. Dang, P. Doan, T.T.Y. Nhi, D.T. Nguyen, B.T. Nguyen, T.P. Nguyen, N.Q. Tran, pharmaceutical analysis, Elsevier, 2008, pp. 157-180.
Int. J. Biol. Macromol. 185 (2021) 592–603, https://doi.org/10.1016/j. [35] S. Sakai, H. Akiyama, Y. Sato, Y. Yoshioka, R.J. Linhardt, Y. Goda, T. Maitani, T.
ijbiomac.2021.06.183. J. Toida, J. Biol. Chem. 281 (2006) 19872–19880, https://doi.org/10.1074/jbc.
[2] M. Filippi, G. Born, M. Chaaban, A.J.F.i.b. Scherberich, biotechnology, Front. M509058200.
Bioeng. Biotechnol. 8 (2020) 474. https://doi.org/10.3389/fbioe.2020.00474. [36] R. Dimatteo, N.J. Darling, T.J. Segura, Adv. Drug Deliv. Rev. 127 (2018) 167–184,
[3] Y. Kobayashi, K. Kohara, Y. Kiuchi, H. Onoda, O. Shoji, H.J.C.C. Yamaguchi, https://doi.org/10.1016/j.addr.2018.03.007.
ChemComm 56 (2020) 6723–6726, https://doi.org/10.1039/D0CC01332C. [37] V.T. Nguyen, T.P. Le, T.P. Ton, D.T. Nguyen, N.N. Dang, B.T. Nguyen, V. Van Van,
[4] L.H. Dang, T.H. Nguyen, H.L.B. Tran, V.N. Doan, N.Q. Tran, J. Healthc. Eng. 2018 T.H. Nguyen, N.Q. Tran, Int. J. Biol. Macromol. 177 (2021) 360–369, https://doi.
(2018) 5754890, https://doi.org/10.1155/2018/5754890. org/10.1016/j.ijbiomac.2021.02.147.
[5] X. Cheng, J. Liu, L. Wang, R. Wang, Z. Liu, R.J.R.A. Zhuo, RSC Adv. 6 (2016) [38] R. Jin, L.S.M. Teixeira, P.J. Dijkstra, C.A. van Blitterswijk, M. Karperien, J.J. Feijen,
101334–101346, https://doi.org/10.1039/C6RA18479K. J. Control. Release 152 (2011) 186–195, https://doi.org/10.1016/j.
[6] J. Qi, Y. Yan, B. Cheng, L. Deng, Z. Shao, Z. Sun, X. Li, ACS Appl. Mater. Interfaces jconrel.2011.01.031.
10 (2018) 6180–6189, https://doi.org/10.1021/acsami.7b18535. [39] Z. Yang, L. Xie, B. Zhang, G. Zhang, F. Huo, C. Zhou, X. Liang, Y. Fan, W. Tian, Y.
[7] X. Zhang, Z. Ma, Y. Ke, Y. Xia, X. Xu, J. Liu, Y. Gong, Q. Shi, J.J.M.A. Yin, Adv. J. Tan, Front. Bioeng. Biotechnol. 10 (2022), https://doi.org/10.3389/
Mater. 2 (2021) 5150–5159, https://doi.org/10.1039/D1MA00137J. fbioe.2022.854693.
[8] P.J. Kondiah, Y.E. Choonara, P.P. Kondiah, T. Marimuthu, P. Kumar, L.C. Du Toit, [40] Y. Liu, S. Wang, D. Sun, Y. Liu, Y. Wang, C. Liu, H. Wu, Y. Lv, Y. Ren, X.J. Guo, Sci.
V.J.M. Pillay, Molecules 21 (2016) 1580, https://doi.org/10.3390/ Rep. 6 (2016) 1–13, https://doi.org/10.1038/srep29858.
molecules21111580. [41] P. Zhu, Y. Masuda, K.J.B. Koumoto, Biomaterials 25 (2004) 3915–3921, https://
[9] N.Q. Tran, Y.K. Joung, E. Lih, K.D. Park, Biomacromolecules 12 (2011) 2872–2880. doi.org/10.1016/j.biomaterials.2003.10.022.

12