- Email: [email protected]
Novel in situ forming hydrogel based on xanthan and chitosan re-gelifying in liquids for local drug delivery
Accepted Manuscript Title: Novel in situ forming hydrogel based on xanthan and chitosan re-gelifying in liquids for local drug delivery Authors: Jinjian Huang, Youming Deng, Jianan Ren, Guopu Chen, Gefei Wang, Feng Wang, Xiuwen Wu PII: DOI: Reference:
S0144-8617(18)30025-0 https://doi.org/10.1016/j.carbpol.2018.01.025 CARP 13175
To appear in: Received date: Revised date: Accepted date:
29-4-2017 5-1-2018 9-1-2018
Please cite this article as: Huang, Jinjian., Deng, Youming., Ren, Jianan., Chen, Guopu., Wang, Gefei., Wang, Feng., & Wu, Xiuwen., Novel in situ forming hydrogel based on xanthan and chitosan re-gelifying in liquids for local drug delivery.Carbohydrate Polymers https://doi.org/10.1016/j.carbpol.2018.01.025 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Novel in situ forming hydrogel based on xanthan and chitosan re-gelifying in liquids for local drug delivery
Jinjian Huang1,3, Youming Deng1,2, Jianan Ren1#, Guopu Chen1, Gefei Wang1, Feng
IP T
Wang4, Xiuwen Wu1# Department of Surgery, Jinling Hospital, Nanjing, China
2
Medical School of Nanjing University, Nanjing, China
3
Medical School of Southeast University, Nanjing, China
4
Changgung Hospital, Tsinghua University, Beijing, China
M
A
N
U
SC R
1
TE D
First author: Jinjian Huang, Email: [email protected] Co-author: Youming Deng, Email: [email protected]
EP
Co-author: Jianan Ren, Email: [email protected]
CC
Co-author: Guopu Chen, Email: [email protected]
A
Co-author: Gefei Wang, Email: [email protected] Co-author: Feng Wang, Email: [email protected] Co-author: Xiuwen Wu, Email: [email protected]
1
#
Co-corresponding author
Department of Surgery, Jinling Hospital, 305 East Zhongshan Road, Nanjing, 210002, China, Mobile Phone: +08613605169808, Fax: +0862580860376,
IP T
Email: [email protected]
SC R
Highlights
1. Development of a novel injectable hydrogel based on Xan-CHO and NOCC. 2. Characterization of Xan-CHO/NOCC hydrogel properties including self-healing,
U
anti-enzymatic hydrolysis, biocompatibility as well as in vitro and in vivo
N
degradation,etc.
M
A
3. The advantage of Xan-CHO/NOCC hydrogel to prevent drug outburst in liquids over the commercial injectable fibrin gel, thus benefitting for wet wounds.
TE D
4. Potential application in regeneration of the abdominal wall defect, serving as a
EP
tissue scaffold.
CC
Abstract
Injectable hydrogels have been an attractive topic in biomaterials. However, during
A
gelation in vivo, they are easy to disperse due to tissue exudates, thus leading to failure of controlled drug release. To solve this problem, we present a novel polysaccharide-based injectable hydrogel via self-crosslinking of aldehyde-modified xanthan (Xan-CHO) and carboxymethyl-modified chitosan 2
(NOCC). The physical properties were optimized by adjusting the mass ratio of Xan-CHO and NOCC. Experiments revealed that this material exhibited the characteristics of self-healing, anti-enzymatic hydrolysis, biocompatibility and biodegradability. The releasing curve demonstrated stable release of BSA-FITC
IP T
10 hours after injection in liquids. After incorporation with a vascular endothelial growth factor, there was an interaction between this biomaterial and the host, which
SC R
accelerated the reconstruction of the abdominal wall in rats. Therefore, this injectable hydrogel, as a drug delivery system, can prevent drug outburst in a variety of settings
A
N
U
and function as a tissue scaffold.
Introduction
TE D
anti-enzymatic hydrolysis
M
Keywords: injectable hydrogel; self-healing; drug delivery; xanthan gum;
Hydrogels are three-dimensional, hydrophilic, polymeric networks capable of
EP
retaining large amounts of water or biological fluids and have attracted significant
CC
attention in biomedical applications such as drug delivery, tissue engineering and regenerative medicine (Kikuchi et al., 2016; Kim, Potta, Park & Song, 2017). From a
A
clinical perspective, injectable hydrogels are more suitable for medical applications because they can accommodate irregularly shaped defects and remain at the desired position by being implanted into tissues through minimally invasive techniques (Yu & Ding, 2008). Moreover, injectable hydrogels can deliver therapeutic drugs to the 3
target lesion. For example, Ma et al (Ma, Xu, Chen, Qin, Chi & Ye, 2018) found that after injection, fully crosslinked hydrogels could deliver a model protein (bovine serum albumin) in vitro in a sustainable manner. However, failure of controlled release occurs for injectable hydrogels under physiological conditions. Two reasons
IP T
account for the failure. First, injectable hydrogels are susceptible to tissue exudates before hydrogel formation. Second, the gelation process can be easily interfered if the
SC R
gelation time is too long. Therefore, stable and rapid gelating methods need to be further investigated.
U
Polysaccharides are ideal candidates to generate hydrogels because they are
N
biocompatible, biodegradable, and hydrophilic (Hamcerencu, Desbrieres, Popa &
A
Riess, 2012). Chitosan is a polysaccharide naturally derived from the chitin shells of
M
shrimp and other crustaceans (Illum, 1998). In recent studies, chitosan has shown an
TE D
advantage in the promotion of wound healing (Hu, Sun & Wu, 2013; Napavichayanun & Aramwit, 2017), but its poor solubility in physiological solvents greatly limits its
EP
further applications. Thus, carboxymethylation of chitosan is needed to achieve solubility. Xanthan is also a polysaccharide produced by the plant pathogenic
CC
bacterium Xanthomonas campestris (Bueno, Bentini, Catalani & Petri, 2013). A xanthan solution has a high-intrinsic viscosity owing to its high molecular weight,
A
interaction by hydrogen bonds and helicoidal structure. Blending xanthan with other polymers results in the formation of hydrogen bonds and electrostatic interactions to stabilize the compound (Li, Hou & Li, 2012). Based on this property of xanthan, a
4
re-gelifying strategy of controlled drug release in liquids is proposed. Specifically, after injection of aldehyde xanthan (Xan-CHO) solutions and carboxymethyl chitosan (NOCC) solutions, the resulting mixture can temporarily maintain a stable structure in liquids via physical interactions (mostly hydrogen bonds). Very soon after, the
IP T
chemical crosslinkage between the two reactants is formed via the Schiff’s base reaction. Through this way, the injected hydrogels are able to avoid the influence of
SC R
tissue exudates so that the loaded drug can realize its controlled release.
A damage or defect in the abdominal wall, resulting from a congenital deformity
U
or postnatal trauma, is a great challenge to clinicians (Azar et al., 2017;
N
Fleurke-Rozema, van de Kamp, Bakker, Pajkrt, Bilardo & Snijders, 2017). Mesh
A
repair and skin transplantation are two major management methods, but these
M
procedures involve significant postoperative morbidity (Gialamas, Balaphas, Assalino,
TE D
Morel, Toso & Staszewicz, 2017). Our previous study found that polypropylene mesh coated with chitosan/gelatin hydrogels could protect the gastrointestinal tract and
EP
promote the growth of granulation tissue for the abdominal wall (Deng et al., 2016), which created a favourable condition for subsequent skin transplantation.
CC
Angiogenesis was regarded as a key factor in soft tissue regeneration (Iijima et al., 2016). Therefore, a potent angiogenic factor (VEGF) was loaded by the novel
A
Xan-CHO/NOCC hydrogels to improve the effects of abdominal wall reconstruction. The synthesis, various physical properties and effects of the re-gelifying strategy on drug delivery in liquids, as well as the application of Xan-CHO/NOCC hydrogels in
5
the rat models were discussed carefully as followed. Experiment 2.1 Materials Xanthan gum (PubChem CID: 7107, pharmaceutical grade, Mw:~220000, viscosity of
IP T
1% aqueous solution at 20°C:1450-2000 mPa.s, Shanghai JiangLai CO. LTD),
SC R
chitosan (PubChem CID: 71853, deacetylation degree of 85%, Mw: 310000-375000, viscosity (1wt % in 1% acetic acid at 25 °C): 800-2000 cP, Sigma-Aldrich, USA), NaIO4 (PubChem CID: 23667635, Sigma-Aldrich, USA), monochloroacetic acid
U
(Shanghai Macklin Biochemical CO.LTD), BSA-FITC (BSA: 5.12 mg/mL, FITC:
N
45µg/mL, Beijing Solarbio Science & Technology CO. LTD), rhodamine B
M
A
(Sigma-Aldrich, USA), fresh duodenal juice (collected from the drainage fluid of a patient with a duodenal fistula, with the patient’s informed consent; the composition
TE D
of the drainage fluid included amylase, lipase, protease and bile acid), CCK-8 kit (Sigma-Aldrich, USA), Porcine Fibrin Gel (Hangzhou Puji CO. LTD), GFP
EP
gene-loaded lentivirus (Keygen, Nanjing), recombinant human vascular endothelial growth factor 121 (rhVEGF121, PeproTech CO.LTD, USA), CD31 monoclonal
CC
antibody (Keygen, Nanjing), α-SMA polyclonal antibody (Keygen, Nanjing),
A
TRITC-labelled goat-anti-mouse IgG (Keygen, Nanjing), FITC-labelled goat-anti-rabbit IgG (Keygen, Nanjing), VEGF receptor primer sequence (Forward: AAAGAGAGGGACTTTGGCCG, Reverse: GTCGCCACTTGACAAAACCC, designed by Primer 5 software), CD31 primer sequence (Forward: 6
AGTAGCATCCTGGTCAACATAACA, Reverse: ATACTGTGACAACACCGTCTCTTC, designed by Primer 5 software), double syringe (cross-sectional area ratio of Xan-CHO/NOCC: 2:1). All other chemicals and reagents were of the highest purity grade commercially available, and some
IP T
abbreviations are explained in Table S1.
SC R
2.2 Cell preparation
Fibroblasts (L929) were cultured in ENI.S medium. The GFP gene-loaded lentivirus
U
was added to the fibroblasts’ medium at a concentration of 10^8 TU/mL. The
N
polybrene was simultaneously added to enhance the transfection efficiency, followed
M
TE D
2.3 Synthesis of Xan-CHO
A
by co-culture for 72 hours.
A 0.6% (w/v) xanthan aqueous solution and 8% (w/v) NaIO4 aqueous solution were
EP
prepared well in advance. Aqueous xanthan solution (80 mL) was poured into a beaker and stirred after adding 2 mL NaIO4 aqueous solution dropwise (the molar
CC
ratio of xanthan to NaIO4 was 1: 1.5). Then, the mixed liquid was continuously stirred for 3 hours in the dark. The purpose of this step was to cleave carbon-carbon bonds of
A
the cis-diol groups and generate reactive aldehyde groups in xanthan units. Then,1 mL ethylene glycol was added to quench the unreacted NaIO4. The reaction was stirred for another 1 hour, and the solution was purified using dialysis bags (MWCO
7
12000-14000) against distilled water for 3 days (the water was renewed at least 5 times per day). Finally, the Xan-CHO product was freeze-dried in a freeze dryer and stored in a sealed plastic bag at 4
for further research. The degradation of xanthan
was observed using GPC measurements (Table S2), which may result from the
IP T
over-oxidation of NaIO4 (Vold & Christensen, 2005). The oxidation degree of Xan-CHO (proportion of oxidized xanthan repeating
SC R
units) was determined by measuring the aldehyde content via the hydroxylamine
hydrochloride titration method with some modifications (Yan et al., 2014a). Five
U
millilitres of the Xan-CHO aqueous solution (0.2% w/v) was dissolved in 15 mL
N
hydroxylamine hydrochloride solution (2.3% w/v). The pH1 was recorded by the pH
A
meter, and then the mixed solution was stirring for 24 hours. Then, the pH2 was
M
measured. The related reactions and calculation formula are as follows:
C1(H+)= 10-pH1 C2(H+)=10-pH2
TE D
Xan-(CHO)n+ nH2N-OH·HCl = Xan-(CH=N-OH)n + nH2O +nHCl
(1) (2) (3)
EP
ΔC = C2(H+) - C1(H+)
(4) (5)
CC
Oxidation degree (%) = 993×ΔC×20×10-3 / (2×W)
In formula (5), 993 is the molecular weight of Xan-CHO repeating units in g/mol, and
A
20 indicates the total volume of the reactive solution (15 mL+5 mL). W refers to the weight of Xan-CHO in grams. The oxidation degree of Xan-CHO was calculated to be 44.1% according to equations (1) ~ (5) above.
8
2.4 Synthesis of NOCC NOCC was produced as described in the literature (Chen, Wu, Mi, Lin, Yu & Sung, 2004; Li et al., 2014a). Briefly, 10 g chitosan (CS) was suspended in 75 mL isopropyl
IP T
alcohol and stirred at room temperature. Then, 25 mL of 10 mol/L NaOH aqueous solution was equally divided into five portions and added to the stirred slurry
SC R
sequentially at 5-min intervals. The resulting slurry was stirred for 30 min. Then,
monochloroacetic acid (20 g) was added dropwise, followed by heating to 60 °C and
U
stirring at this temperature for 3 h. The reaction mixture was then filtered. The residue
N
(NOCC) was thoroughly washed for three cycles with 80% v/v methanol/water
A
mixture and two cycles with alcohol. The final product was obtained by
M
vacuum-drying and the substitution degree of NOCC was 85%. The degradation of
TE D
chitosan was also confirmed using GPC measurements (Table S2). This may have resulted from the hydrolytic action of the NaOH solution and monochloroacetic acid
EP
(Cai et al., 2014).
CC
2.5 Synthesis of Xan-CHO/NOCC hydrogel
A
The synthesis of Xan-CHO/NOCC hydrogel is displayed in Fig 1. The crosslinking of Xan-CHO and NOCC was attributed to the Schiff’s base reaction between the aldehyde groups of Xan-CHO and the amino groups of NOCC. Before the chemical reaction, the hydrogel bonds in the Xan-CHO aqueous solution could maintain a 9
stable three-dimensional network.
2.6 Hydrogel characterization 2.6.1 FTIR spectrometry. An FTIR spectrophotometer was used to obtain the FTIR
IP T
spectra of CS, NOCC, xanthan, Xan-CHO and the cross-linked hydrogel using a Nicolet-6700 spectrometer from Thermo Electron at room temperature in the wave
SC R
number range of 4000–500 cm-1 by the KBr pellet technique. The powders of each polymer were ground to dry KBr disk, and 32 scans at a resolution of 4 cm-1 were
U
used to record the spectra.
N
2.6.2 Gelation time. Different mass ratios (1%:0.16%, 1%:0.33%, 1%:0.66%, 1%:1%,
A
1%:2%) of Xan-CHO and NOCC were chosen to study the gelation time, as
M
determined by a vial tilting method (Ghobril & Grinstaff, 2015). Specifically, the
TE D
Xan-CHO and NOCC were mixed in a vial with stirring (25 ̊C, pH=7), and the time when no flow was observed after tilting the vial was recorded as the gelation time.
EP
2.6.3 Microstructure. The samples used in the experiment of gelation time were freeze-dried. Then, they were cut vertically and scanned by scanning electron
CC
microscopy (SEM, S-4800) at a voltage of 5kV. The pore size was analysed by randomly determining the pore diameters in three different visual fields.
A
2.6.4 Rheological property. (1) Storage modulus (G′) and loss modulus (G″) of the simple 1 mL Xan-CHO (1%) or a mixture of 1 mL Xan-CHO (1%) and 0.5 mL NOCC (0.16%, 0.33%, 0.66%, 1% and 2%, respectively) were measured with a
10
rheometer (Anton Paar, MCR302) using parallel plates at 37 C ̊ in oscillatory mode for 5 minutes (Gap size = 1 mm, strain = 1.0%). (2) The Xan-CHO/NOCC hydrogel (1%:0.33%) was tested under a 1.0% strain level, and the angular frequency (ω) was swept from 0.1 to 100 rad/s.
formula (Huang, Ren, Chen, Deng, Wang & Wu, 2017):
SC R
Swelling ratio (%) = (Mswollen gel – Mdried gel) / Mdried gel × 100
IP T
2.5.5 Swelling ratio. The hydrogel swelling ratio was determined by the following
Mdried gel and the Mswollen gel stand for the mass of dried hydrogel and swollen hydrogel
U
at equilibrium, respectively. Specifically, freeze-dried hydrogels were immersed in
N
PBS buffer. They were removed at each preset time, and the surface moisture was
A
immediately wiped off using tissue paper. The hydrogels were returned to the PBS
M
buffer after they were weighed. This process was repeated until equilibrium was
TE D
attained.
EP
2.7 Self-healing assay
The 1.5 mL Xan-CHO/NOCC hydrogel (1%:0.33%) was prepared with 0.1 mL of 1
CC
wt % rhodamine B in a round plastic mould and then cut into four pieces. We used another undyed hydrogel to recombine with the dyed pieces. The healing process of
A
the hydrogel was recorded by a smartphone. Additionally, a rheological recovery test (Huang et al., 2016) was performed on the 1.5 mL Xan-CHO/NOCC hydrogel (1%:0.33%) where the alternate step strain switched from a small strain (1%) to a
11
large strain (γ = 180%, 300%, 800%) in continuous step strain measurements at a fixed angular frequency (10 rad/s).
2.8 Local injection assay in the liquid environment
IP T
Hydrogels, 1.5 mL Xan-CHO/NOCC (1%:0.33%), containing 0.1 mL of 1wt % rhodamine B were directly injected into the PBS solution in plastic plates using a
U
minutes later, the images were recorded with a camera.
SC R
double syringe. The procedure was also performed with 1.5 mL of fibrin gel. Ten
N
2.9 Drug releasing assay
A
The drug-releasing law of Xan-CHO/NOCC hydrogel (1%:0.33%) was assessed using
M
BSA-FITC (Kim et al., 2009). The hydrogel containing 0.05 mL BSA-FITC was
TE D
injected into 20 mL PBS. At each time point (0.5 h, 1.5 h, 3 h, 6 h, 9 h, 12 h, 18 h, 24 h, 36 h, 48 h, 60 h, 72 h), 1 mL PBS (pH 6.8) was retrieved (T=37 ̊C) and the amounts
EP
of BSA-FITC were determined using a multi-functional microplate reader (HORIBA, model: FluoroMax-4) at an absorbance of 493 nm. Subsequently, the PBS was poured
CC
back. These procedures were also performed on the injectable fibrin gel. All experiments were repeated three times. The concentration of the released BSA-FITC
A
was calculated according to the following equation: FI = 3.644×108×C + 3.865×105 (Goodness of fit: r2 = 0.924) where FI refers to the fluorescence intensity and C represents the concentration of
12
BSA-FITC.
2.10 Anti-enzymatic hydrolysis assay Three blocks of Xan-CHO/NOCC hydrogels (1%:0.33%) and fibrin gels were well . The
IP T
prepared in a 6-well plate and then immersed in fresh duodenal juice at 37
duodenal juice was changed every 6 hours. After 6 h, 12 h, 24 h,48 h and 72 h, the
SC R
hydrogel blocks were removed, and the surface moisture was immediately wiped off using tissue paper. Then, the hydrogels’ weight was measured, and the hydrogel was
N
U
returned to the plate.
A
2.11 Cytocompatibility study
M
The biocompatibility of the hydrogel was evaluated by the extraction method. The
TE D
Xan-CHO/NOCC hydrogel (1%:0.33%) was prepared in a 24-well plate following extraction using DMEM with 10% FBS for 24 h at 37
. Then, sequential dilutions of
the stock solution were carried out to obtain different concentrations (100%, 50%,
EP
25%, 12.5%) of the leachate; 0.1 mL of 1.0×105/mL fibroblasts was incubated with
CC
the leachate (200μL) for another 48 h. Subsequently, 10μL CCK-8 was added to each well, followed by incubation at 37
for another 4 h. After the solution was
A
homogenized, the absorbance at 450 nm was measured. All experiments were performed in triplicate. For further assessment of the cytocompatibility, 0.1 mL of 1.0×104/mL
13
fluorescent fibroblasts was seeded to the leachate, the surface of Xan-CHO/NOCC hydrogel (1%:0.33%) and the space between the hydrogel and the 24-well plate. Subsequently, 0.8 mL DMEM medium was added to each well, and the samples were cultured for 48 h. The images were recorded with an inverted fluorescence
SC R
2.12 Degradation of Xan-CHO/NOCC hydrogel in vitro and in vivo
IP T
microscope at 24 h and 48 h.
The degradation of the Xan-CHO/NOCC hydrogel (1%:0.33%) was investigated via
U
immersion in PBS at pH 1.2, pH 7.4 and pH 12 for 11 days. Moreover, eighteen
N
BALB/c mice were employed to evaluate in vivo degradation by dorsal subcutaneous
A
injection of 0.4 mL Xan-CHO/NOCC hydrogels. At the scheduled time, the injection
M
sites of the three mice were exposed using surgical scissors to observe the hydrogel.
TE D
The skin above the hydrogel was carefully removed, fixed in 4% paraformaldehyde, embedded in paraffin, sectioned, and stained with H&E for further histological
EP
examination.
CC
2.13 Reconstruction of abdominal wall defect Eighteen adult male Sprague-Dawley rats (body weight: 200 g-220 g) were used for
A
this experiment. The animals were maintained at a temperature of 25°C and a relative humidity of 50–60% under natural light/dark cycles and allowed free access to food and water. The Xan-CHO and NOCC were dissolved in sterile water and then
14
underwent ultraviolet sterilization for 24 hours. All animal care and experimental protocols were approved by the Animal Investigation Ethics Committee of Jinling Hospital. All rats were fasted overnight and received general anaesthesia through
IP T
intraperitoneal injections of 10% chloral hydrate at 0.4 mL/100 g. These rats were then randomly divided into three groups with six rats per group. The abdomen was
SC R
prepared for aseptic surgery by sterilizing the operative area with iodophor three times, followed by placing sterile surgical drapes over the entire field. A 2×3 cm abdominal
U
wall defect was created using surgical scissors. Then, the rats received 1.5 mL PBS,
N
1.5 mL simple Xan-CHO/NOCC hydrogel and 1.5 mL VEGF(2μg)-loaded
A
Xan-CHO/NOCC hydrogel (Fig S1). Then, polypropylene meshes were applied for
M
temporal abdominal closure. No rats died during the entire treatment. One week later,
TE D
these animals were sacrificed, and fresh granulation tissues were excised and cut into two pieces. Promptly, one-half were immersed in 10% neutral formaldehyde for H&E
EP
and Masson trichrome staining, and the other half were stored at -80℃ for
A
CC
immunofluorescence staining and qPCR (Shin et al., 2013).
15
Results 3.1 FTIR analysis The preparation of Xan-CHO and NOCC was confirmed by FTIR, as shown in Fig S2. Both chitosan and NOCC showed the characteristic absorption bands of
IP T
polysaccharide at approximately 1340 cm-1 (C−C−H and O−C−H stretching) and 1100 cm-1 (C−O stretching). In comparison with chitosan, NOCC had a characteristic
SC R
absorption band at 1610 cm-1 and 1432 cm-1 due to the asymmetric and symmetric
stretching vibration of the COO- groups. In the xanthan and Xan-CHO, the typical
U
absorption bands of polysaccharide structures were also observed. Moreover, at the
N
1740 cm-1 was the C=O group stretching vibration, which was more obvious in
A
Xan-CHO than in xanthan. After Xan-CHO/NOCC gelation, the stretching vibration
M
of the C=O group (1740 cm-1) was diminished, while the absorption band of the C=N
TE D
group (1625 cm-1) was enlarged compared with that of Xan-CHO, suggesting the
EP
occurrence of a Schiff’s base reaction between two modified polysaccharides.
3.2 Characterization of Xan-CHO/NOCC hydrogel
CC
The morphology of the Xan-CHO/NOCC hydrogel that was prepared from various concentrations of NOCC is presented in Fig 2. The hydrogel with a mass ratio of
A
1%:0.33% (Xan-CHO/NOCC) exhibited a dense and solid structure, while an increase or decrease in the NOCC concentration led to an irregular and loose shape (Fig 2A). At a micro level, high-density and small-sized pores were obtained at 1%:0.33% of
16
Xan-CHO/NOCC, revealing that sufficient crosslinking between Xan-CHO and NOCC was achieved in this ratio. The rheological properties of Xan-CHO solution and Xan-CHO/NOCC hydrogel were recorded in Fig 3A. Interestingly, there was no focal point between the curves of
IP T
the storage modulus (G’) and the loss modulus (G’’). G’ for each sample is listed in Fig 3B. The hydrogel with a mass ratio of 1%:0.33% (Xan-CHO/NOCC) had a larger
SC R
G’ than the others. The gelation time was also recorded in Fig 3B, indicating that the Xan-CHO/NOCC hydrogel (1%:0.33%) exhibited the quickest gelation time. The
U
swelling ratio of the hydrogel (1%:0.33%) was also the highest among all samples
N
(Fig 5A). In the entire frequency range test, G′ was consistently greater than G″,
A
indicating that the hydrogel (1%:0.33%) was stable and behaved like a viscoelastic
M
solid (Fig S3).
TE D
Fig 4A demonstrated that the Xan-CHO/NOCC hydrogel had a self-healing property. It was clear that the red dye diffused to undyed hydrogel at 6 hours and
EP
distributed evenly in the recombined hydrogel at 24 hours. The hydrogel could be suspended under gravity, and no crack was observed on its surface (Fig 4B). In the
CC
rheological recovery test (Fig S4), G’ could almost completely return to the origin (strain=1%) after increasing the strain from 1% to 180%, 300% and 800%,
A
respectively.
3.3 Local injection and drug releasing assay
17
Compared with injectable fibrin gel, the Xan-CHO/NOCC hydrogel could maintain a stable and dense ‘S’ shape after 10 minutes of local injection into PBS (Fig 4C). The red dye was scattered in the fibrin gel, but in the Xan-CHO/NOCC hydrogel, the dye was retained. BSA-FITC releasing assay indicated that the Xan-CHO/NOCC hydrogel
IP T
could achieve a steady and continuous drug release for ten hours. The releasing ratio of BSA-FITC was over 80% after 20 hours (Fig 4D). However, the initial drug
SC R
outburst could not be avoided for the fibrin gel. Moreover, the fibrin gel would
disappear after immersion in digestive juice for 12 hours, but the Xan-CHO/NOCC
U
hydrogel was resistant to enzymatic hydrolysis, with more than 70% of its original
M
3.4 Cytocompatibility
A
N
weight after immersion for 72 hours (Fig 5C).
TE D
Different concentrations of the leachates were employed to culture fluorescent fibroblasts. Fig 6a-h indicates that the spindle-shaped fibroblasts could spread well in
EP
the medium after 24 hours and 48 hours regardless of leachate’s proportions. As seen in Fig 6l, the CCK-8 assay further confirmed that the leachate would not influence
CC
cell proliferation regardless of its percentage. Fig 6i shows the spherical shape of fibroblasts on the hydrogel surface, while elongated and spread morphologies were
A
observed when the fibroblasts were cultured between the hydrogel and plate surface (Fig 6j). The spherical cells were transferred to fresh medium for 24 hours, and the well-spread shapes were observed again (Fig 6k).
18
3.5 Degradation in vitro and in vivo The degradation rates in vitrowere similar at pH 7.4 and pH 12 but not pH 1.2 (Fig 5B). In the long term, the acid environment could accelerate the degradation of
IP T
Xan-CHO/NOCC hydrogels (Fig S5). In vivo, the weight of the hydrogel was reduced after injection, particularly on the first day (Fig 7A, B). After degradation for 14 days,
SC R
the remaining weight was nearly 30% that of the initial weight. Histological
examinations (Fig 7C) indicated that the number of neutrophils increased significantly
U
on the 7th day with the degradation of the hydrogel. On the 14th day, the
N
inflammatory cells were reduced. Moreover, the layer of subcutaneous soft tissue on
A
the 14th day was obviously thicker than before. The mice showed no adverse
TE D
M
reactions or symptoms of toxicity during the entire process.
3.6 Reconstruction of abdominal wall defects
EP
H&E and Masson staining of the regenerative granulation tissues revealed that hydrogel+VEGF was effective to generate thicker abdominal walls than other
CC
interventions (Fig 8A, C). CD31 and a-SMA are biomarkers of vascular endothelial
A
cells and capillary formation, respectively (Rufaihah et al., 2017; Thaunat, 2015).The results of immunofluorescence staining confirmed that hydrogel+VEGF can promote maturation of blood vessels compared with PBS or simple hydrogel (Fig S6A). The transcription activities of VEGF receptors (Fig S6B) and CD31 (Fig S6C) were 19
significantly enhanced in the hydrogel+VEGF group.
Discussion In this study, we developed a novel polysaccharide-derived injectable
IP T
(Xan-CHO/NOCC) hydrogel based on the “re-gelifying” strategy. The xanthan was
SC R
elaborately chosen because it could exhibit a stable structure supported by hydrogen bonds before chemical gelation to realize controlled drug release. Xanthan and
chitosan were modified to generate the hydrogel based on a Schiff’s base reaction. To
U
homogenize the hydrogel, the concentration of Xan-CHO solution was decreased to 1%
N
(w/v) to ensure full crosslinking of Xan-CHO and NOCC. The gelation formulation
M
A
was adjusted to 1%:0.33% (Xan-CHO/NOCC) based on the following requirements: 1) appropriate gelation time; 2) relatively high storage modulus, which aimed to reach a
TE D
steady state at the early stage and provide the target lesion with a solid scaffold (Wang, Zhang, Tsang, Wan & Wu, 2017).
EP
Remarkably, when Xan-CHO/NOCC was 1%:0.33%, the ratio of -CHO groups in Xan-CHO to -NH2 groups in NOCC was approximately 3:4 with the quickest gelation
CC
time. A change in the ratio would prolong the gelation time. Yan et al (Yan et al.,
A
2014b) also studied the influence of the –CHO/-NH2 ratio on the gelation time of the Schiff’s base reaction, but in their study, the quickest gelation time (approximately 10 seconds) was achieved at a –CHO/-NH2 ratio of 1:3. Moreover, our findings suggested that at the optimized ratio of 1%:0.33%, the hydrogel showed a rigid 20
appearance with the largest G’ at the macro level and the highest pore density at the micro level, which would be ascribed to the high crosslinking degree of Xan-CHO and NOCC. Therefore, the hydrogel exerted the largest swelling ratio to absorb large amounts of liquids at 1%:0.33%.
IP T
The G’ and G’’ curves of the Xan-CHO solution were shown to lack the focal point, which indicated that physical interactions (hydrogen bonds) existed between
SC R
the Xan-CHO molecules. After crosslinking with the NOCC, the hydrogel displayed a self-healing character based on the dynamic equilibrium between the Schiff’s base
U
linkage and the aldehyde and amine reactants. Zhang et al (Zhang, Tao, Li & Wei,
N
2011) also reported a hydrogel made of chitosan and DF-PEG through the Schiff’s
A
base reaction and found its repeated self-healing property.
M
The Xan-CHO/NOCC hydrogel was well tolerated at different pH values and
TE D
with various digestive enzymes in the human duodenal juice. Recently, several reports focused on the pH-sensitive hydrogels as the oral administrative drug carriers (Abdel,
EP
Radwan & Ali, 2016; Shi et al., 2016). Although the smart biomaterials presented a lower drug release rate at pH 1.2 and a higher drug release rate at pH 6.8, these
CC
studies ignored the effects of digestive enzymes on the degradation of the hydrogels, thus, leading to a bias between in vitro and in vivo experiments. Koetting et al
A
(Koetting, Guido, Gupta, Zhang & Peppas, 2016) produced hydrogel microparticles as protein delivery, which suggested that this pH-sensitive material completely released the drug within one hour in the presence of 10 mg/mL pancreatin. Therefore, these
21
hydrogels inevitably had a drug outburst effect in the intestinal tract. The Schiff’s base reaction was unstable in an acid environment; however, the Xan-CHO/NOCC hydrogel exhibited an anti-acid property of 10% degradation rate within four hours (gastric emptying time), which led to only a minor amount of the drug being liberated
IP T
into the stomach. The reason might be the protective effect of the Xan-CHO’s side chains and a high degree of polymerization. Moreover, Xan-CHO/NOCC hydrogels
SC R
showed anti-enzymatic hydrolysis, probably because their precursors (xanthan,
chitosan) are low-digestible dietary fibres (Koh, De Vadder, Kovatcheva-Datchary &
U
Backhed, 2016; Tan, Wei, Zhao, Xu & Peng, 2017). Hence, this material exerted its
N
potential ability to control drug release in the gut for oral administration or behave as
A
a plugging material for the management of gastrointestinal fistulas (Ren, Wu, Hong &
M
Li, 2012).
TE D
Additionally, the Xan-CHO/NOCC hydrogel could be gelated in situ in a fluid environment. Compared with fibrin gel, the Xan-CHO/NOCC hydrogel did not show
EP
drug outburst when applied in liquids owing to the temporary protective roles of the Xan-CHO physical interactions. Thus, the hydrogel had the advantage of delivering
CC
drugs to exudative tissues. The drug release rate was not in accordance with the degradation rate of the hydrogel. This could be attributed to the determinant factors of
A
drugs, including hydrophilia, molecular weight, and concentration gradient (Shin et al., 2013b; Peret & Murphy, 2008). The Xan-CHO/NOCC hydrogel was compatible with the host cells in our
22
experiment. The fibroblasts could survive when they were covered with the hydrogel, which revealed that the oxygen and nutritional supplements could be freely exchanged through this pore-contained biomaterial (Murphy & O'Brien, 2010). The cells cultured on the hydrogel surface were alive but in a spherical shape. This was in
IP T
accordance with the reports by Motealleh et al(Motealleh & Kehr, 2017) that the mechanical stiffness less than 5 kPa would lead to the circular morphology of cells.
SC R
During biodegradation in vivo, we did not observe the multinuclear giant cells of severe foreign body reactions. Additionally, the subcutaneous connective tissue
U
became thicker on the 14th day, perhaps because the hydrogel could supply a matrix
N
for cell growth (Li et al., 2014b).
A
In the abdominal wall defect model, we found that the simple Xan-CHO/NOCC
M
hydrogel was effective in tissue regeneration. The potential reasons might be the
TE D
glucosamine units of chitosan that initiate fibroblast proliferation and collagen production as well as angiogenesis (Napavichayanun & Aramwit, 2017). The
EP
angiogenic effects associated with VEGF have been widely reported (Poels, Abou-Ghannam, Decamps, Leyman, Rieux & Wyns, 2016; Yin et al., 2016). In this
CC
study, we applied the hydrogel to carry VEGF and confirmed the interactions between the granulation tissues and the hydrogel+VEGF. As seen in Fig 8B, several chemical
A
groups such as imine, amine and nitro could be formed between the hydrogel and the surrounding tissues, and furthermore, the transcriptional level of VEGF receptors was elevated. Therefore, the growth of granulation tissues was enhanced by the materials
23
and theoretically created an excellent condition for the subsequent skin transplantation (Archer, Billmire, Falcone & Warner, 2006). As a preliminary research, the limitations were noteworthy. First, the rate of degradation in vivo fell behind the healing process of wounds, so it will require more
IP T
work to realize the synchronization. Second, we focused only on the blood supply of the regenerative abdominal wall but ignored the inflammatory reactions in the
SC R
granulation tissues. Finally, the subsequent effects, such as organ function, associated complications, and the prognosis of skin graft, were not studied. Therefore, further
U
studies are necessary to assess these consequences.
N
Conclusions
A
In this study, we developed a novel injectable hydrogel from Xan-CHO and NOCC.
M
This hydrogel showed various properties, e.g., porosity, self-healing, anti-enzymatic
TE D
hydrolysis, cytocompatibility and promotion of tissue healing. Moreover, this biomaterial presented the advantage of attaining controlled drug release, especially in
EP
tissues with much excretion or exudation such as digestive tracts and open wounds. The hydrogel loaded with VEGF resulted in accelerated angiogenesis and regenerated
CC
the abdominal wall tissue. Altogether, these results demonstrated that this Xan-CHO/NOCC hydrogel could be applied as a drug delivery system in various
A
settings.
24
Declaration of Conflicting Interests We declare no potential conflicts of interest concerning the research, authorship,
SC R
IP T
and/or publication of this article.
Funding
The study was supported by the National Natural Science Foundation of China (Grant
N
U
no. 81571881) and Key Project of Jiangsu Social Development (Grant no.
A
BE2016752).
M
Acknowledgements
We thank Tingting Xu and Yichang Pan at Nanjing Tech University for instructions
A
CC
EP
TE D
regarding Fourier transform infrared spectroscopy and scanning electron microscopy.
25
References: Abdel, G. A., Radwan, R. R., & Ali, H. E. (2016). Radiation Synthesis of Poly(Starch/Acrylic acid) pH Sensitive Hydrogel for Rutin Controlled Release. International Journal of Biological Macromolecules,
IP T
92, 957-964.
Archer, L. P., Billmire, D. A., Falcone, R. J., & Warner, B. W. (2006). Reconstruction of an acquired
SC R
abdominal wall defect in a neonate using acellular human dermis. Plastic and Reconstructive Surgery, 118(7), 163e-166e.
U
Azar, F. K., Crawford, T. C., Poruk, K. E., Farrow, N., Cornell, P., Nadra, O., Azoury, S. C., Soares, K.
N
C., Cooney, C. M., & Eckhauser, F. E. (2017). Ventral hernia repair in patients with abdominal loss of
A
domain: an observational study of one institution's experience. Hernia, 21(2), 245-252.
M
Bueno, V. B., Bentini, R., Catalani, L. H., & Petri, D. F. (2013). Synthesis and swelling behavior of
TE D
xanthan-based hydrogels. Carbohydr Polym, 92(2), 1091-1099.
Cai, J., Yang, Y., Man, J., Huang, J., Wang, Z., Zhang, C., Gu, M., Liu, Q., & Wei, C. (2014). Structural
EP
and functional properties of alkali-treated high-amylose rice starch. Food Chemistry, 145, 245-253.
CC
Chen, S. C., Wu, Y. C., Mi, F. L., Lin, Y. H., Yu, L. C., & Sung, H. W. (2004). A novel pH-sensitive hydrogel composed of N,O-carboxymethyl chitosan and alginate cross-linked by genipin for protein
A
drug delivery. Journal of Controlled Release, 96(2), 285-300.
Deng, Y., Ren, J., Chen, G., Li, G., Guo, K., Hu, Q., Wu, X., Wang, G., Gu, G., & Li, J. (2016). Evaluation of polypropylene mesh coated with biological hydrogels for temporary closure of open abdomen. Journal of Biomaterials Applications, 31(2), 302-314.
26
Fleurke-Rozema, H., van de Kamp, K., Bakker, M., Pajkrt, E., Bilardo, C., & Snijders, R. (2017). Prevalence, timing of diagnosis and pregnancy outcome of abdominal wall defects after the introduction of a national prenatal screening program. Prenat Diagn.
adhesives for wound closure: a tutorial. Chemical Society Reviews, 44(7), 1820-1835.
IP T
Ghobril, C., & Grinstaff, M. W. (2015a). The chemistry and engineering of polymeric hydrogel
SC R
Gialamas, E., Balaphas, A., Assalino, M., Morel, P., Toso, C., & Staszewicz, W. (2017). Complex abdominal wall defect repair using biologic porcine matrix. Minerva Chirurgica.
U
Hamcerencu, M., Desbrieres, J., Popa, M., & Riess, G. (2012). Original stimuli-sensitive polysaccharide
N
derivatives/N-isopropylacrylamide hydrogels. Role of polysaccharide backbone. Carbohydr Polym,
A
89(2), 438-447.
M
Hu, L., Sun, Y., & Wu, Y. (2013). Advances in chitosan-based drug delivery vehicles. Nanoscale, 5(8),
TE D
3103-3111.
Huang, J., Ren, J., Chen, G., Deng, Y., Wang, G., & Wu, X. (2017). Evaluation of the Xanthan-Based
EP
Film Incorporated with Silver Nanoparticles for Potential Application in the Nonhealing Infectious
CC
Wound. Journal of Nanomaterials, 2017, 1-10.
Huang, W., Wang, Y., Chen, Y., Zhao, Y., Zhang, Q., Zheng, X., Chen, L., & Zhang, L. (2016). Strong
A
and Rapidly Self-Healing Hydrogels: Potential Hemostatic Materials. Advanced Healthcare Materials, 5(21), 2813-2822.
Iijima, M., Aubin, H., Steinbrink, M., Schiffer, F., Assmann, A., Weisel, R. D., Matsui, Y., Li, R. K., Lichtenberg, A., & Akhyari, P. (2016). Bio-active coating of decellularized vascular grafts with a
27
temperature-sensitive VEGF-conjugated hydrogel accelerates autologous endothelialization in vivo. J Tissue Eng Regen Med.
Illum, L. (1998). Chitosan and its use as a pharmaceutical excipient. Pharm Res, 15(9), 1326-1331.
IP T
Kikuchi, I. S., Galante, R. S., Dua, K., Malipeddi, V. R., Awasthi, R., Ghisleni, D. D., & de Jesus, A. P. T. (2016). Hydrogel based drug delivery systems: A review with special emphasis on challenges
SC R
associated with decontamination of hydrogels and biomaterials. Current Drug Delivery.
Kim, B. S., Oh, J. M., Kim, K. S., Seo, K. S., Cho, J. S., Khang, G., Lee, H. B., Park, K., & Kim, M. S.
U
(2009). BSA-FITC-loaded microcapsules for in vivo delivery. Biomaterials, 30(5), 902-909.
N
Kim, Y. M., Potta, T., Park, K. H., & Song, S. C. (2017). Temperature responsive chemical crosslinkable
A
UV pretreated hydrogel for application to injectable tissue regeneration system via differentiations of
M
encapsulated hMSCs. Biomaterials, 112, 248-256.
TE D
Koetting, M. C., Guido, J. F., Gupta, M., Zhang, A., & Peppas, N. A. (2016). pH-responsive and enzymatically-responsive hydrogel microparticles for the oral delivery of therapeutic proteins: Effects of
EP
protein size, crosslinking density, and hydrogel degradation on protein delivery. Journal of Controlled
CC
Release, 221, 18-25.
Koh, A., De Vadder, F., Kovatcheva-Datchary, P., & Backhed, F. (2016). From Dietary Fiber to Host
A
Physiology: Short-Chain Fatty Acids as Key Bacterial Metabolites. Cell, 165(6), 1332-1345.
Li, H., Hou, W., & Li, X. (2012). Interaction between xanthan gum and cationic cellulose JR400 in aqueous solution. Carbohydr Polym, 89(1), 24-30.
28
Li, L., Wang, N., Jin, X., Deng, R., Nie, S., Sun, L., Wu, Q., Wei, Y., & Gong, C. (2014a). Biodegradable and injectable in situ cross-linking chitosan-hyaluronic acid based hydrogels for postoperative adhesion prevention. Biomaterials, 35(12), 3903-3917.
Li, L., Wang, N., Jin, X., Deng, R., Nie, S., Sun, L., Wu, Q., Wei, Y., & Gong, C. (2014b). Biodegradable
IP T
and injectable in situ cross-linking chitosan-hyaluronic acid based hydrogels for postoperative adhesion
SC R
prevention. Biomaterials, 35(12), 3903-3917.
Ma, X., Xu, T., Chen, W., Qin, H., Chi, B., & Ye, Z. (2018). Injectable hydrogels based on the hyaluronic
U
acid and poly (gamma-glutamic acid) for controlled protein delivery. Carbohydr Polym, 179, 100-109.
A
Engineering. Advanced Healthcare Materials, 6(1).
N
Motealleh, A., & Kehr, N. S. (2017). Nanocomposite Hydrogels and Their Applications in Tissue
M
Murphy, C. M., & O'Brien, F. J. (2010). Understanding the effect of mean pore size on cell activity in
TE D
collagen-glycosaminoglycan scaffolds. Cell Adh Migr, 4(3), 377-381.
Napavichayanun, S., & Aramwit, P. (2017). Effect of animal products and extracts on wound healing
EP
promotion in topical applications: a review. J Biomater Sci Polym Ed, 1-27.
CC
Peret, B. J., & Murphy, W. L. (2008). Controllable Soluble Protein Concentration Gradients in Hydrogel Networks. Advanced Functional Materials, 18(21), 3410-3417.
A
Poels, J., Abou-Ghannam, G., Decamps, A., Leyman, M., Rieux, A., & Wyns, C. (2016). Transplantation of testicular tissue in alginate hydrogel loaded with VEGF nanoparticles improves spermatogonial recovery. Journal of Controlled Release, 234, 79-89.
29
Ren, J., Wu, X., Hong, Z., & Li, J. (2012). Management of a patient with colocutaneous fistula by autologous platelet-rich fibrin glue. Am Surg, 78(2), 86-87.
Rufaihah, A. J., Johari, N. A., Vaibavi, S. R., Plotkin, M., Di Thien, D. T., Kofidis, T., & Seliktar, D. (2017). Dual delivery of VEGF and ANG-1 in ischemic hearts using an injectable hydrogel. Acta
IP T
Biomaterialia, 48, 58-67.
SC R
Shi, Y., Wang, X., Deng, X., Tian, R., Zhang, Y., Shang, Q., & Chen, N. (2016). Release profile of
insulin from pH-sensitive hydrogel and its hypoglycemic effect by oral administration. J Biomater Sci
U
Polym Ed, 27(1), 86-96.
N
Shin, S. H., Lee, J., Lim, K. S., Rhim, T., Lee, S. K., Kim, Y. H., & Lee, K. Y. (2013a). Sequential
A
delivery of TAT-HSP27 and VEGF using microsphere/hydrogel hybrid systems for therapeutic
M
angiogenesis. Journal of Controlled Release, 166(1), 38-45.
TE D
Shin, S. H., Lee, J., Lim, K. S., Rhim, T., Lee, S. K., Kim, Y. H., & Lee, K. Y. (2013b). Sequential delivery of TAT-HSP27 and VEGF using microsphere/hydrogel hybrid systems for therapeutic
EP
angiogenesis. Journal of Controlled Release, 166(1), 38-45.
Tan, C., Wei, H., Zhao, X., Xu, C., & Peng, J. (2017). Effects of dietary fibers with high water-binding
CC
capacity and swelling capacity on gastrointestinal functions, food intake and body weight in male rats.
A
Food & Nutrition Research, 61(1), 1308118.
Thaunat, O. (2015). Soothing touch of CD31 protects endothelium during cellular immune responses. Proc Natl Acad Sci U S A, 112(43), 13133-13134.
30
Vold, I. M., & Christensen, B. E. (2005). Periodate oxidation of chitosans with different chemical compositions. Carbohydr Res, 340(4), 679-684.
Wang, J., Zhang, F., Tsang, W. P., Wan, C., & Wu, C. (2017). Fabrication of injectable high strength
IP T
hydrogel based on 4-arm star PEG for cartilage tissue engineering. Biomaterials, 120, 11-21.
Yan, S., Wang, T., Feng, L., Zhu, J., Zhang, K., Chen, X., Cui, L., & Yin, J. (2014a). Injectable in situ
SC R
self-cross-linking hydrogels based on poly(L-glutamic acid) and alginate for cartilage tissue engineering. Biomacromolecules, 15(12), 4495-4508.
U
Yan, S., Wang, T., Feng, L., Zhu, J., Zhang, K., Chen, X., Cui, L., & Yin, J. (2014b). Injectable in situ
N
self-cross-linking hydrogels based on poly(L-glutamic acid) and alginate for cartilage tissue engineering.
A
Biomacromolecules, 15(12), 4495-4508.
M
Yin, N., Han, Y., Xu, H., Gao, Y., Yi, T., Yao, J., Dong, L., Cheng, D., & Chen, Z. (2016). VEGF-conjugated alginate hydrogel prompt angiogenesis and improve pancreatic islet engraftment and
TE D
function in type 1 diabetes. Mater Sci Eng C Mater Biol Appl, 59, 958-964.
Yu, L., & Ding, J. (2008). Injectable hydrogels as unique biomedical materials. Chemical Society
EP
Reviews, 37(8), 1473-1481.
CC
Zhang, Y., Tao, L., Li, S., & Wei, Y. (2011). Synthesis of multiresponsive and dynamic chitosan-based
A
hydrogels for controlled release of bioactive molecules. Biomacromolecules, 12(8), 2894-2901.
31
IP T SC R
Fig 1. Synthesis of Xan-CHO/NOCC hydrogels. (a) Xanthan, (b) Xan-CHO, (c)
N
U
chitosan, d) NOCC. Before injection, the drug (VEGF) is carried by Xan-CHO
A
solutions, where hydrogen bonds exist to maintain a stable structure. After injection,
M
the physical interactions are replaced with chemical interactions via the Schiff’s base
A
CC
EP
TE D
reaction, leading to formation of the novel Xan-CHO/NOCC hydrogels.
Fig 2. (A) The appearance of Xan-CHO/NOCC hydrogel, (B) the microstructure of Xan-CHO/NOCC hydrogel. (a) 1%: 0.16%, (b) 1%:0.33%, (c) 1%:0.66%, (d) 1%:1%,
32
(e) 1%:2%, (f) pore sizes at different mass ratios of Xan-CHO and NOCC. (Bar =100
U
SC R
IP T
µm).
N
Fig 3. (A) The results of the rheological property, strain=1.0%, [mass ratio of
A
Xan-CHO/NOCC: (a) 1%:0%, (b) 1%:0.16%, (c) 1%:0.33%, (d) 1%:0.66%, (e)
M
1%:1%, (f) 1%:2%], (B) the gelation time and G’ of each hydrogel with varied mass
EP
TE D
ratios.
CC
Fig 4. (A) Self-healing process of Xan-CHO/NOCC hydrogel, (B) lifting up the Xan-CHO/NOCC hydrogel after self-healing, (C) local injection of Xan-CHO/NOCC
A
hydrogel and fibrin gel in PBS (red arrow: Xan-CHO/NOCC hydrogel, green arrow: fibrin gel), (D) cumulative releasing ratio of BSA-FITC after local injection in PBS.
33
IP T
Fig 5. (A) Swelling ratio of Xan-CHO/NOCC hydrogel with varied mass ratios, (B) degradation of Xan-CHO/NOCC hydrogel (1%:0.33%) in acid, neutral and alkaline
SC R
buffers within 24 hours, (C) anti-enzymatic hydrolysis of Xan-CHO/NOCC hydrogel
M
A
N
U
and fibrin gel (red arrow: Xan-CHO/NOCC hydrogel, green arrow: fibrin gel).
TE D
Fig 6. Cytocompatibility of Xan-CHO/NOCC hydrogel. (a)-(d) Co-culture of fibroblasts in different proportions of leachate (12.5%, 25%, 50% and 100%,
EP
respectively) for 24 hours, (e)-(h) co-culture of fibroblasts in different proportions of leachate (12.5%, 25%, 50% and 100%, respectively) for 48 hours, 100×, (i) culture of
CC
fibroblasts on the surface of hydrogel for 48 hours, 100×, (j) culture of fibroblasts
A
between the hydrogel and the surface of the plastic plate for 48 hours, 100×, (k) culture of the spherical cells in (i) on the plastic plate for 24 hours, 100×, (l) CCK-8 assay detecting cell-proliferation by culturing fibroblasts in different proportions of leachate for 24 hours and 48 hours. 34
IP T SC R
U
Fig 7. Degradation of Xan-CHO/NOCC hydrogel in vivo. (A) The appearance of the
N
hydrogel and the surrounding tissues at the scheduled time (black arrow: remained
A
hydrogels), (B) remaining mass of hydrogel after in vivo degradation, (C) H&E
CC
EP
TE D
M
staining of superficial skin, 10×.
Fig 8. (A) H&E and Masson staining of regenerative abdominal wall tissues (green
A
arrows: native tissues, red framework: regenerative tissues) (bar=1 mm), (B) schematic diagram of tissue growth promoting effects through hydrogel-to-host interactions, (C) thickness of regenerative abdominal wall tissues. *P<0.05, **P<0.01.
35
S0144-8617(18)30025-0 https://doi.org/10.1016/j.carbpol.2018.01.025 CARP 13175
To appear in: Received date: Revised date: Accepted date:
29-4-2017 5-1-2018 9-1-2018
Please cite this article as: Huang, Jinjian., Deng, Youming., Ren, Jianan., Chen, Guopu., Wang, Gefei., Wang, Feng., & Wu, Xiuwen., Novel in situ forming hydrogel based on xanthan and chitosan re-gelifying in liquids for local drug delivery.Carbohydrate Polymers https://doi.org/10.1016/j.carbpol.2018.01.025 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Novel in situ forming hydrogel based on xanthan and chitosan re-gelifying in liquids for local drug delivery
Jinjian Huang1,3, Youming Deng1,2, Jianan Ren1#, Guopu Chen1, Gefei Wang1, Feng
IP T
Wang4, Xiuwen Wu1# Department of Surgery, Jinling Hospital, Nanjing, China
2
Medical School of Nanjing University, Nanjing, China
3
Medical School of Southeast University, Nanjing, China
4
Changgung Hospital, Tsinghua University, Beijing, China
M
A
N
U
SC R
1
TE D
First author: Jinjian Huang, Email: [email protected] Co-author: Youming Deng, Email: [email protected]
EP
Co-author: Jianan Ren, Email: [email protected]
CC
Co-author: Guopu Chen, Email: [email protected]
A
Co-author: Gefei Wang, Email: [email protected] Co-author: Feng Wang, Email: [email protected] Co-author: Xiuwen Wu, Email: [email protected]
1
#
Co-corresponding author
Department of Surgery, Jinling Hospital, 305 East Zhongshan Road, Nanjing, 210002, China, Mobile Phone: +08613605169808, Fax: +0862580860376,
IP T
Email: [email protected]
SC R
Highlights
1. Development of a novel injectable hydrogel based on Xan-CHO and NOCC. 2. Characterization of Xan-CHO/NOCC hydrogel properties including self-healing,
U
anti-enzymatic hydrolysis, biocompatibility as well as in vitro and in vivo
N
degradation,etc.
M
A
3. The advantage of Xan-CHO/NOCC hydrogel to prevent drug outburst in liquids over the commercial injectable fibrin gel, thus benefitting for wet wounds.
TE D
4. Potential application in regeneration of the abdominal wall defect, serving as a
EP
tissue scaffold.
CC
Abstract
Injectable hydrogels have been an attractive topic in biomaterials. However, during
A
gelation in vivo, they are easy to disperse due to tissue exudates, thus leading to failure of controlled drug release. To solve this problem, we present a novel polysaccharide-based injectable hydrogel via self-crosslinking of aldehyde-modified xanthan (Xan-CHO) and carboxymethyl-modified chitosan 2
(NOCC). The physical properties were optimized by adjusting the mass ratio of Xan-CHO and NOCC. Experiments revealed that this material exhibited the characteristics of self-healing, anti-enzymatic hydrolysis, biocompatibility and biodegradability. The releasing curve demonstrated stable release of BSA-FITC
IP T
10 hours after injection in liquids. After incorporation with a vascular endothelial growth factor, there was an interaction between this biomaterial and the host, which
SC R
accelerated the reconstruction of the abdominal wall in rats. Therefore, this injectable hydrogel, as a drug delivery system, can prevent drug outburst in a variety of settings
A
N
U
and function as a tissue scaffold.
Introduction
TE D
anti-enzymatic hydrolysis
M
Keywords: injectable hydrogel; self-healing; drug delivery; xanthan gum;
Hydrogels are three-dimensional, hydrophilic, polymeric networks capable of
EP
retaining large amounts of water or biological fluids and have attracted significant
CC
attention in biomedical applications such as drug delivery, tissue engineering and regenerative medicine (Kikuchi et al., 2016; Kim, Potta, Park & Song, 2017). From a
A
clinical perspective, injectable hydrogels are more suitable for medical applications because they can accommodate irregularly shaped defects and remain at the desired position by being implanted into tissues through minimally invasive techniques (Yu & Ding, 2008). Moreover, injectable hydrogels can deliver therapeutic drugs to the 3
target lesion. For example, Ma et al (Ma, Xu, Chen, Qin, Chi & Ye, 2018) found that after injection, fully crosslinked hydrogels could deliver a model protein (bovine serum albumin) in vitro in a sustainable manner. However, failure of controlled release occurs for injectable hydrogels under physiological conditions. Two reasons
IP T
account for the failure. First, injectable hydrogels are susceptible to tissue exudates before hydrogel formation. Second, the gelation process can be easily interfered if the
SC R
gelation time is too long. Therefore, stable and rapid gelating methods need to be further investigated.
U
Polysaccharides are ideal candidates to generate hydrogels because they are
N
biocompatible, biodegradable, and hydrophilic (Hamcerencu, Desbrieres, Popa &
A
Riess, 2012). Chitosan is a polysaccharide naturally derived from the chitin shells of
M
shrimp and other crustaceans (Illum, 1998). In recent studies, chitosan has shown an
TE D
advantage in the promotion of wound healing (Hu, Sun & Wu, 2013; Napavichayanun & Aramwit, 2017), but its poor solubility in physiological solvents greatly limits its
EP
further applications. Thus, carboxymethylation of chitosan is needed to achieve solubility. Xanthan is also a polysaccharide produced by the plant pathogenic
CC
bacterium Xanthomonas campestris (Bueno, Bentini, Catalani & Petri, 2013). A xanthan solution has a high-intrinsic viscosity owing to its high molecular weight,
A
interaction by hydrogen bonds and helicoidal structure. Blending xanthan with other polymers results in the formation of hydrogen bonds and electrostatic interactions to stabilize the compound (Li, Hou & Li, 2012). Based on this property of xanthan, a
4
re-gelifying strategy of controlled drug release in liquids is proposed. Specifically, after injection of aldehyde xanthan (Xan-CHO) solutions and carboxymethyl chitosan (NOCC) solutions, the resulting mixture can temporarily maintain a stable structure in liquids via physical interactions (mostly hydrogen bonds). Very soon after, the
IP T
chemical crosslinkage between the two reactants is formed via the Schiff’s base reaction. Through this way, the injected hydrogels are able to avoid the influence of
SC R
tissue exudates so that the loaded drug can realize its controlled release.
A damage or defect in the abdominal wall, resulting from a congenital deformity
U
or postnatal trauma, is a great challenge to clinicians (Azar et al., 2017;
N
Fleurke-Rozema, van de Kamp, Bakker, Pajkrt, Bilardo & Snijders, 2017). Mesh
A
repair and skin transplantation are two major management methods, but these
M
procedures involve significant postoperative morbidity (Gialamas, Balaphas, Assalino,
TE D
Morel, Toso & Staszewicz, 2017). Our previous study found that polypropylene mesh coated with chitosan/gelatin hydrogels could protect the gastrointestinal tract and
EP
promote the growth of granulation tissue for the abdominal wall (Deng et al., 2016), which created a favourable condition for subsequent skin transplantation.
CC
Angiogenesis was regarded as a key factor in soft tissue regeneration (Iijima et al., 2016). Therefore, a potent angiogenic factor (VEGF) was loaded by the novel
A
Xan-CHO/NOCC hydrogels to improve the effects of abdominal wall reconstruction. The synthesis, various physical properties and effects of the re-gelifying strategy on drug delivery in liquids, as well as the application of Xan-CHO/NOCC hydrogels in
5
the rat models were discussed carefully as followed. Experiment 2.1 Materials Xanthan gum (PubChem CID: 7107, pharmaceutical grade, Mw:~220000, viscosity of
IP T
1% aqueous solution at 20°C:1450-2000 mPa.s, Shanghai JiangLai CO. LTD),
SC R
chitosan (PubChem CID: 71853, deacetylation degree of 85%, Mw: 310000-375000, viscosity (1wt % in 1% acetic acid at 25 °C): 800-2000 cP, Sigma-Aldrich, USA), NaIO4 (PubChem CID: 23667635, Sigma-Aldrich, USA), monochloroacetic acid
U
(Shanghai Macklin Biochemical CO.LTD), BSA-FITC (BSA: 5.12 mg/mL, FITC:
N
45µg/mL, Beijing Solarbio Science & Technology CO. LTD), rhodamine B
M
A
(Sigma-Aldrich, USA), fresh duodenal juice (collected from the drainage fluid of a patient with a duodenal fistula, with the patient’s informed consent; the composition
TE D
of the drainage fluid included amylase, lipase, protease and bile acid), CCK-8 kit (Sigma-Aldrich, USA), Porcine Fibrin Gel (Hangzhou Puji CO. LTD), GFP
EP
gene-loaded lentivirus (Keygen, Nanjing), recombinant human vascular endothelial growth factor 121 (rhVEGF121, PeproTech CO.LTD, USA), CD31 monoclonal
CC
antibody (Keygen, Nanjing), α-SMA polyclonal antibody (Keygen, Nanjing),
A
TRITC-labelled goat-anti-mouse IgG (Keygen, Nanjing), FITC-labelled goat-anti-rabbit IgG (Keygen, Nanjing), VEGF receptor primer sequence (Forward: AAAGAGAGGGACTTTGGCCG, Reverse: GTCGCCACTTGACAAAACCC, designed by Primer 5 software), CD31 primer sequence (Forward: 6
AGTAGCATCCTGGTCAACATAACA, Reverse: ATACTGTGACAACACCGTCTCTTC, designed by Primer 5 software), double syringe (cross-sectional area ratio of Xan-CHO/NOCC: 2:1). All other chemicals and reagents were of the highest purity grade commercially available, and some
IP T
abbreviations are explained in Table S1.
SC R
2.2 Cell preparation
Fibroblasts (L929) were cultured in ENI.S medium. The GFP gene-loaded lentivirus
U
was added to the fibroblasts’ medium at a concentration of 10^8 TU/mL. The
N
polybrene was simultaneously added to enhance the transfection efficiency, followed
M
TE D
2.3 Synthesis of Xan-CHO
A
by co-culture for 72 hours.
A 0.6% (w/v) xanthan aqueous solution and 8% (w/v) NaIO4 aqueous solution were
EP
prepared well in advance. Aqueous xanthan solution (80 mL) was poured into a beaker and stirred after adding 2 mL NaIO4 aqueous solution dropwise (the molar
CC
ratio of xanthan to NaIO4 was 1: 1.5). Then, the mixed liquid was continuously stirred for 3 hours in the dark. The purpose of this step was to cleave carbon-carbon bonds of
A
the cis-diol groups and generate reactive aldehyde groups in xanthan units. Then,1 mL ethylene glycol was added to quench the unreacted NaIO4. The reaction was stirred for another 1 hour, and the solution was purified using dialysis bags (MWCO
7
12000-14000) against distilled water for 3 days (the water was renewed at least 5 times per day). Finally, the Xan-CHO product was freeze-dried in a freeze dryer and stored in a sealed plastic bag at 4
for further research. The degradation of xanthan
was observed using GPC measurements (Table S2), which may result from the
IP T
over-oxidation of NaIO4 (Vold & Christensen, 2005). The oxidation degree of Xan-CHO (proportion of oxidized xanthan repeating
SC R
units) was determined by measuring the aldehyde content via the hydroxylamine
hydrochloride titration method with some modifications (Yan et al., 2014a). Five
U
millilitres of the Xan-CHO aqueous solution (0.2% w/v) was dissolved in 15 mL
N
hydroxylamine hydrochloride solution (2.3% w/v). The pH1 was recorded by the pH
A
meter, and then the mixed solution was stirring for 24 hours. Then, the pH2 was
M
measured. The related reactions and calculation formula are as follows:
C1(H+)= 10-pH1 C2(H+)=10-pH2
TE D
Xan-(CHO)n+ nH2N-OH·HCl = Xan-(CH=N-OH)n + nH2O +nHCl
(1) (2) (3)
EP
ΔC = C2(H+) - C1(H+)
(4) (5)
CC
Oxidation degree (%) = 993×ΔC×20×10-3 / (2×W)
In formula (5), 993 is the molecular weight of Xan-CHO repeating units in g/mol, and
A
20 indicates the total volume of the reactive solution (15 mL+5 mL). W refers to the weight of Xan-CHO in grams. The oxidation degree of Xan-CHO was calculated to be 44.1% according to equations (1) ~ (5) above.
8
2.4 Synthesis of NOCC NOCC was produced as described in the literature (Chen, Wu, Mi, Lin, Yu & Sung, 2004; Li et al., 2014a). Briefly, 10 g chitosan (CS) was suspended in 75 mL isopropyl
IP T
alcohol and stirred at room temperature. Then, 25 mL of 10 mol/L NaOH aqueous solution was equally divided into five portions and added to the stirred slurry
SC R
sequentially at 5-min intervals. The resulting slurry was stirred for 30 min. Then,
monochloroacetic acid (20 g) was added dropwise, followed by heating to 60 °C and
U
stirring at this temperature for 3 h. The reaction mixture was then filtered. The residue
N
(NOCC) was thoroughly washed for three cycles with 80% v/v methanol/water
A
mixture and two cycles with alcohol. The final product was obtained by
M
vacuum-drying and the substitution degree of NOCC was 85%. The degradation of
TE D
chitosan was also confirmed using GPC measurements (Table S2). This may have resulted from the hydrolytic action of the NaOH solution and monochloroacetic acid
EP
(Cai et al., 2014).
CC
2.5 Synthesis of Xan-CHO/NOCC hydrogel
A
The synthesis of Xan-CHO/NOCC hydrogel is displayed in Fig 1. The crosslinking of Xan-CHO and NOCC was attributed to the Schiff’s base reaction between the aldehyde groups of Xan-CHO and the amino groups of NOCC. Before the chemical reaction, the hydrogel bonds in the Xan-CHO aqueous solution could maintain a 9
stable three-dimensional network.
2.6 Hydrogel characterization 2.6.1 FTIR spectrometry. An FTIR spectrophotometer was used to obtain the FTIR
IP T
spectra of CS, NOCC, xanthan, Xan-CHO and the cross-linked hydrogel using a Nicolet-6700 spectrometer from Thermo Electron at room temperature in the wave
SC R
number range of 4000–500 cm-1 by the KBr pellet technique. The powders of each polymer were ground to dry KBr disk, and 32 scans at a resolution of 4 cm-1 were
U
used to record the spectra.
N
2.6.2 Gelation time. Different mass ratios (1%:0.16%, 1%:0.33%, 1%:0.66%, 1%:1%,
A
1%:2%) of Xan-CHO and NOCC were chosen to study the gelation time, as
M
determined by a vial tilting method (Ghobril & Grinstaff, 2015). Specifically, the
TE D
Xan-CHO and NOCC were mixed in a vial with stirring (25 ̊C, pH=7), and the time when no flow was observed after tilting the vial was recorded as the gelation time.
EP
2.6.3 Microstructure. The samples used in the experiment of gelation time were freeze-dried. Then, they were cut vertically and scanned by scanning electron
CC
microscopy (SEM, S-4800) at a voltage of 5kV. The pore size was analysed by randomly determining the pore diameters in three different visual fields.
A
2.6.4 Rheological property. (1) Storage modulus (G′) and loss modulus (G″) of the simple 1 mL Xan-CHO (1%) or a mixture of 1 mL Xan-CHO (1%) and 0.5 mL NOCC (0.16%, 0.33%, 0.66%, 1% and 2%, respectively) were measured with a
10
rheometer (Anton Paar, MCR302) using parallel plates at 37 C ̊ in oscillatory mode for 5 minutes (Gap size = 1 mm, strain = 1.0%). (2) The Xan-CHO/NOCC hydrogel (1%:0.33%) was tested under a 1.0% strain level, and the angular frequency (ω) was swept from 0.1 to 100 rad/s.
formula (Huang, Ren, Chen, Deng, Wang & Wu, 2017):
SC R
Swelling ratio (%) = (Mswollen gel – Mdried gel) / Mdried gel × 100
IP T
2.5.5 Swelling ratio. The hydrogel swelling ratio was determined by the following
Mdried gel and the Mswollen gel stand for the mass of dried hydrogel and swollen hydrogel
U
at equilibrium, respectively. Specifically, freeze-dried hydrogels were immersed in
N
PBS buffer. They were removed at each preset time, and the surface moisture was
A
immediately wiped off using tissue paper. The hydrogels were returned to the PBS
M
buffer after they were weighed. This process was repeated until equilibrium was
TE D
attained.
EP
2.7 Self-healing assay
The 1.5 mL Xan-CHO/NOCC hydrogel (1%:0.33%) was prepared with 0.1 mL of 1
CC
wt % rhodamine B in a round plastic mould and then cut into four pieces. We used another undyed hydrogel to recombine with the dyed pieces. The healing process of
A
the hydrogel was recorded by a smartphone. Additionally, a rheological recovery test (Huang et al., 2016) was performed on the 1.5 mL Xan-CHO/NOCC hydrogel (1%:0.33%) where the alternate step strain switched from a small strain (1%) to a
11
large strain (γ = 180%, 300%, 800%) in continuous step strain measurements at a fixed angular frequency (10 rad/s).
2.8 Local injection assay in the liquid environment
IP T
Hydrogels, 1.5 mL Xan-CHO/NOCC (1%:0.33%), containing 0.1 mL of 1wt % rhodamine B were directly injected into the PBS solution in plastic plates using a
U
minutes later, the images were recorded with a camera.
SC R
double syringe. The procedure was also performed with 1.5 mL of fibrin gel. Ten
N
2.9 Drug releasing assay
A
The drug-releasing law of Xan-CHO/NOCC hydrogel (1%:0.33%) was assessed using
M
BSA-FITC (Kim et al., 2009). The hydrogel containing 0.05 mL BSA-FITC was
TE D
injected into 20 mL PBS. At each time point (0.5 h, 1.5 h, 3 h, 6 h, 9 h, 12 h, 18 h, 24 h, 36 h, 48 h, 60 h, 72 h), 1 mL PBS (pH 6.8) was retrieved (T=37 ̊C) and the amounts
EP
of BSA-FITC were determined using a multi-functional microplate reader (HORIBA, model: FluoroMax-4) at an absorbance of 493 nm. Subsequently, the PBS was poured
CC
back. These procedures were also performed on the injectable fibrin gel. All experiments were repeated three times. The concentration of the released BSA-FITC
A
was calculated according to the following equation: FI = 3.644×108×C + 3.865×105 (Goodness of fit: r2 = 0.924) where FI refers to the fluorescence intensity and C represents the concentration of
12
BSA-FITC.
2.10 Anti-enzymatic hydrolysis assay Three blocks of Xan-CHO/NOCC hydrogels (1%:0.33%) and fibrin gels were well . The
IP T
prepared in a 6-well plate and then immersed in fresh duodenal juice at 37
duodenal juice was changed every 6 hours. After 6 h, 12 h, 24 h,48 h and 72 h, the
SC R
hydrogel blocks were removed, and the surface moisture was immediately wiped off using tissue paper. Then, the hydrogels’ weight was measured, and the hydrogel was
N
U
returned to the plate.
A
2.11 Cytocompatibility study
M
The biocompatibility of the hydrogel was evaluated by the extraction method. The
TE D
Xan-CHO/NOCC hydrogel (1%:0.33%) was prepared in a 24-well plate following extraction using DMEM with 10% FBS for 24 h at 37
. Then, sequential dilutions of
the stock solution were carried out to obtain different concentrations (100%, 50%,
EP
25%, 12.5%) of the leachate; 0.1 mL of 1.0×105/mL fibroblasts was incubated with
CC
the leachate (200μL) for another 48 h. Subsequently, 10μL CCK-8 was added to each well, followed by incubation at 37
for another 4 h. After the solution was
A
homogenized, the absorbance at 450 nm was measured. All experiments were performed in triplicate. For further assessment of the cytocompatibility, 0.1 mL of 1.0×104/mL
13
fluorescent fibroblasts was seeded to the leachate, the surface of Xan-CHO/NOCC hydrogel (1%:0.33%) and the space between the hydrogel and the 24-well plate. Subsequently, 0.8 mL DMEM medium was added to each well, and the samples were cultured for 48 h. The images were recorded with an inverted fluorescence
SC R
2.12 Degradation of Xan-CHO/NOCC hydrogel in vitro and in vivo
IP T
microscope at 24 h and 48 h.
The degradation of the Xan-CHO/NOCC hydrogel (1%:0.33%) was investigated via
U
immersion in PBS at pH 1.2, pH 7.4 and pH 12 for 11 days. Moreover, eighteen
N
BALB/c mice were employed to evaluate in vivo degradation by dorsal subcutaneous
A
injection of 0.4 mL Xan-CHO/NOCC hydrogels. At the scheduled time, the injection
M
sites of the three mice were exposed using surgical scissors to observe the hydrogel.
TE D
The skin above the hydrogel was carefully removed, fixed in 4% paraformaldehyde, embedded in paraffin, sectioned, and stained with H&E for further histological
EP
examination.
CC
2.13 Reconstruction of abdominal wall defect Eighteen adult male Sprague-Dawley rats (body weight: 200 g-220 g) were used for
A
this experiment. The animals were maintained at a temperature of 25°C and a relative humidity of 50–60% under natural light/dark cycles and allowed free access to food and water. The Xan-CHO and NOCC were dissolved in sterile water and then
14
underwent ultraviolet sterilization for 24 hours. All animal care and experimental protocols were approved by the Animal Investigation Ethics Committee of Jinling Hospital. All rats were fasted overnight and received general anaesthesia through
IP T
intraperitoneal injections of 10% chloral hydrate at 0.4 mL/100 g. These rats were then randomly divided into three groups with six rats per group. The abdomen was
SC R
prepared for aseptic surgery by sterilizing the operative area with iodophor three times, followed by placing sterile surgical drapes over the entire field. A 2×3 cm abdominal
U
wall defect was created using surgical scissors. Then, the rats received 1.5 mL PBS,
N
1.5 mL simple Xan-CHO/NOCC hydrogel and 1.5 mL VEGF(2μg)-loaded
A
Xan-CHO/NOCC hydrogel (Fig S1). Then, polypropylene meshes were applied for
M
temporal abdominal closure. No rats died during the entire treatment. One week later,
TE D
these animals were sacrificed, and fresh granulation tissues were excised and cut into two pieces. Promptly, one-half were immersed in 10% neutral formaldehyde for H&E
EP
and Masson trichrome staining, and the other half were stored at -80℃ for
A
CC
immunofluorescence staining and qPCR (Shin et al., 2013).
15
Results 3.1 FTIR analysis The preparation of Xan-CHO and NOCC was confirmed by FTIR, as shown in Fig S2. Both chitosan and NOCC showed the characteristic absorption bands of
IP T
polysaccharide at approximately 1340 cm-1 (C−C−H and O−C−H stretching) and 1100 cm-1 (C−O stretching). In comparison with chitosan, NOCC had a characteristic
SC R
absorption band at 1610 cm-1 and 1432 cm-1 due to the asymmetric and symmetric
stretching vibration of the COO- groups. In the xanthan and Xan-CHO, the typical
U
absorption bands of polysaccharide structures were also observed. Moreover, at the
N
1740 cm-1 was the C=O group stretching vibration, which was more obvious in
A
Xan-CHO than in xanthan. After Xan-CHO/NOCC gelation, the stretching vibration
M
of the C=O group (1740 cm-1) was diminished, while the absorption band of the C=N
TE D
group (1625 cm-1) was enlarged compared with that of Xan-CHO, suggesting the
EP
occurrence of a Schiff’s base reaction between two modified polysaccharides.
3.2 Characterization of Xan-CHO/NOCC hydrogel
CC
The morphology of the Xan-CHO/NOCC hydrogel that was prepared from various concentrations of NOCC is presented in Fig 2. The hydrogel with a mass ratio of
A
1%:0.33% (Xan-CHO/NOCC) exhibited a dense and solid structure, while an increase or decrease in the NOCC concentration led to an irregular and loose shape (Fig 2A). At a micro level, high-density and small-sized pores were obtained at 1%:0.33% of
16
Xan-CHO/NOCC, revealing that sufficient crosslinking between Xan-CHO and NOCC was achieved in this ratio. The rheological properties of Xan-CHO solution and Xan-CHO/NOCC hydrogel were recorded in Fig 3A. Interestingly, there was no focal point between the curves of
IP T
the storage modulus (G’) and the loss modulus (G’’). G’ for each sample is listed in Fig 3B. The hydrogel with a mass ratio of 1%:0.33% (Xan-CHO/NOCC) had a larger
SC R
G’ than the others. The gelation time was also recorded in Fig 3B, indicating that the Xan-CHO/NOCC hydrogel (1%:0.33%) exhibited the quickest gelation time. The
U
swelling ratio of the hydrogel (1%:0.33%) was also the highest among all samples
N
(Fig 5A). In the entire frequency range test, G′ was consistently greater than G″,
A
indicating that the hydrogel (1%:0.33%) was stable and behaved like a viscoelastic
M
solid (Fig S3).
TE D
Fig 4A demonstrated that the Xan-CHO/NOCC hydrogel had a self-healing property. It was clear that the red dye diffused to undyed hydrogel at 6 hours and
EP
distributed evenly in the recombined hydrogel at 24 hours. The hydrogel could be suspended under gravity, and no crack was observed on its surface (Fig 4B). In the
CC
rheological recovery test (Fig S4), G’ could almost completely return to the origin (strain=1%) after increasing the strain from 1% to 180%, 300% and 800%,
A
respectively.
3.3 Local injection and drug releasing assay
17
Compared with injectable fibrin gel, the Xan-CHO/NOCC hydrogel could maintain a stable and dense ‘S’ shape after 10 minutes of local injection into PBS (Fig 4C). The red dye was scattered in the fibrin gel, but in the Xan-CHO/NOCC hydrogel, the dye was retained. BSA-FITC releasing assay indicated that the Xan-CHO/NOCC hydrogel
IP T
could achieve a steady and continuous drug release for ten hours. The releasing ratio of BSA-FITC was over 80% after 20 hours (Fig 4D). However, the initial drug
SC R
outburst could not be avoided for the fibrin gel. Moreover, the fibrin gel would
disappear after immersion in digestive juice for 12 hours, but the Xan-CHO/NOCC
U
hydrogel was resistant to enzymatic hydrolysis, with more than 70% of its original
M
3.4 Cytocompatibility
A
N
weight after immersion for 72 hours (Fig 5C).
TE D
Different concentrations of the leachates were employed to culture fluorescent fibroblasts. Fig 6a-h indicates that the spindle-shaped fibroblasts could spread well in
EP
the medium after 24 hours and 48 hours regardless of leachate’s proportions. As seen in Fig 6l, the CCK-8 assay further confirmed that the leachate would not influence
CC
cell proliferation regardless of its percentage. Fig 6i shows the spherical shape of fibroblasts on the hydrogel surface, while elongated and spread morphologies were
A
observed when the fibroblasts were cultured between the hydrogel and plate surface (Fig 6j). The spherical cells were transferred to fresh medium for 24 hours, and the well-spread shapes were observed again (Fig 6k).
18
3.5 Degradation in vitro and in vivo The degradation rates in vitrowere similar at pH 7.4 and pH 12 but not pH 1.2 (Fig 5B). In the long term, the acid environment could accelerate the degradation of
IP T
Xan-CHO/NOCC hydrogels (Fig S5). In vivo, the weight of the hydrogel was reduced after injection, particularly on the first day (Fig 7A, B). After degradation for 14 days,
SC R
the remaining weight was nearly 30% that of the initial weight. Histological
examinations (Fig 7C) indicated that the number of neutrophils increased significantly
U
on the 7th day with the degradation of the hydrogel. On the 14th day, the
N
inflammatory cells were reduced. Moreover, the layer of subcutaneous soft tissue on
A
the 14th day was obviously thicker than before. The mice showed no adverse
TE D
M
reactions or symptoms of toxicity during the entire process.
3.6 Reconstruction of abdominal wall defects
EP
H&E and Masson staining of the regenerative granulation tissues revealed that hydrogel+VEGF was effective to generate thicker abdominal walls than other
CC
interventions (Fig 8A, C). CD31 and a-SMA are biomarkers of vascular endothelial
A
cells and capillary formation, respectively (Rufaihah et al., 2017; Thaunat, 2015).The results of immunofluorescence staining confirmed that hydrogel+VEGF can promote maturation of blood vessels compared with PBS or simple hydrogel (Fig S6A). The transcription activities of VEGF receptors (Fig S6B) and CD31 (Fig S6C) were 19
significantly enhanced in the hydrogel+VEGF group.
Discussion In this study, we developed a novel polysaccharide-derived injectable
IP T
(Xan-CHO/NOCC) hydrogel based on the “re-gelifying” strategy. The xanthan was
SC R
elaborately chosen because it could exhibit a stable structure supported by hydrogen bonds before chemical gelation to realize controlled drug release. Xanthan and
chitosan were modified to generate the hydrogel based on a Schiff’s base reaction. To
U
homogenize the hydrogel, the concentration of Xan-CHO solution was decreased to 1%
N
(w/v) to ensure full crosslinking of Xan-CHO and NOCC. The gelation formulation
M
A
was adjusted to 1%:0.33% (Xan-CHO/NOCC) based on the following requirements: 1) appropriate gelation time; 2) relatively high storage modulus, which aimed to reach a
TE D
steady state at the early stage and provide the target lesion with a solid scaffold (Wang, Zhang, Tsang, Wan & Wu, 2017).
EP
Remarkably, when Xan-CHO/NOCC was 1%:0.33%, the ratio of -CHO groups in Xan-CHO to -NH2 groups in NOCC was approximately 3:4 with the quickest gelation
CC
time. A change in the ratio would prolong the gelation time. Yan et al (Yan et al.,
A
2014b) also studied the influence of the –CHO/-NH2 ratio on the gelation time of the Schiff’s base reaction, but in their study, the quickest gelation time (approximately 10 seconds) was achieved at a –CHO/-NH2 ratio of 1:3. Moreover, our findings suggested that at the optimized ratio of 1%:0.33%, the hydrogel showed a rigid 20
appearance with the largest G’ at the macro level and the highest pore density at the micro level, which would be ascribed to the high crosslinking degree of Xan-CHO and NOCC. Therefore, the hydrogel exerted the largest swelling ratio to absorb large amounts of liquids at 1%:0.33%.
IP T
The G’ and G’’ curves of the Xan-CHO solution were shown to lack the focal point, which indicated that physical interactions (hydrogen bonds) existed between
SC R
the Xan-CHO molecules. After crosslinking with the NOCC, the hydrogel displayed a self-healing character based on the dynamic equilibrium between the Schiff’s base
U
linkage and the aldehyde and amine reactants. Zhang et al (Zhang, Tao, Li & Wei,
N
2011) also reported a hydrogel made of chitosan and DF-PEG through the Schiff’s
A
base reaction and found its repeated self-healing property.
M
The Xan-CHO/NOCC hydrogel was well tolerated at different pH values and
TE D
with various digestive enzymes in the human duodenal juice. Recently, several reports focused on the pH-sensitive hydrogels as the oral administrative drug carriers (Abdel,
EP
Radwan & Ali, 2016; Shi et al., 2016). Although the smart biomaterials presented a lower drug release rate at pH 1.2 and a higher drug release rate at pH 6.8, these
CC
studies ignored the effects of digestive enzymes on the degradation of the hydrogels, thus, leading to a bias between in vitro and in vivo experiments. Koetting et al
A
(Koetting, Guido, Gupta, Zhang & Peppas, 2016) produced hydrogel microparticles as protein delivery, which suggested that this pH-sensitive material completely released the drug within one hour in the presence of 10 mg/mL pancreatin. Therefore, these
21
hydrogels inevitably had a drug outburst effect in the intestinal tract. The Schiff’s base reaction was unstable in an acid environment; however, the Xan-CHO/NOCC hydrogel exhibited an anti-acid property of 10% degradation rate within four hours (gastric emptying time), which led to only a minor amount of the drug being liberated
IP T
into the stomach. The reason might be the protective effect of the Xan-CHO’s side chains and a high degree of polymerization. Moreover, Xan-CHO/NOCC hydrogels
SC R
showed anti-enzymatic hydrolysis, probably because their precursors (xanthan,
chitosan) are low-digestible dietary fibres (Koh, De Vadder, Kovatcheva-Datchary &
U
Backhed, 2016; Tan, Wei, Zhao, Xu & Peng, 2017). Hence, this material exerted its
N
potential ability to control drug release in the gut for oral administration or behave as
A
a plugging material for the management of gastrointestinal fistulas (Ren, Wu, Hong &
M
Li, 2012).
TE D
Additionally, the Xan-CHO/NOCC hydrogel could be gelated in situ in a fluid environment. Compared with fibrin gel, the Xan-CHO/NOCC hydrogel did not show
EP
drug outburst when applied in liquids owing to the temporary protective roles of the Xan-CHO physical interactions. Thus, the hydrogel had the advantage of delivering
CC
drugs to exudative tissues. The drug release rate was not in accordance with the degradation rate of the hydrogel. This could be attributed to the determinant factors of
A
drugs, including hydrophilia, molecular weight, and concentration gradient (Shin et al., 2013b; Peret & Murphy, 2008). The Xan-CHO/NOCC hydrogel was compatible with the host cells in our
22
experiment. The fibroblasts could survive when they were covered with the hydrogel, which revealed that the oxygen and nutritional supplements could be freely exchanged through this pore-contained biomaterial (Murphy & O'Brien, 2010). The cells cultured on the hydrogel surface were alive but in a spherical shape. This was in
IP T
accordance with the reports by Motealleh et al(Motealleh & Kehr, 2017) that the mechanical stiffness less than 5 kPa would lead to the circular morphology of cells.
SC R
During biodegradation in vivo, we did not observe the multinuclear giant cells of severe foreign body reactions. Additionally, the subcutaneous connective tissue
U
became thicker on the 14th day, perhaps because the hydrogel could supply a matrix
N
for cell growth (Li et al., 2014b).
A
In the abdominal wall defect model, we found that the simple Xan-CHO/NOCC
M
hydrogel was effective in tissue regeneration. The potential reasons might be the
TE D
glucosamine units of chitosan that initiate fibroblast proliferation and collagen production as well as angiogenesis (Napavichayanun & Aramwit, 2017). The
EP
angiogenic effects associated with VEGF have been widely reported (Poels, Abou-Ghannam, Decamps, Leyman, Rieux & Wyns, 2016; Yin et al., 2016). In this
CC
study, we applied the hydrogel to carry VEGF and confirmed the interactions between the granulation tissues and the hydrogel+VEGF. As seen in Fig 8B, several chemical
A
groups such as imine, amine and nitro could be formed between the hydrogel and the surrounding tissues, and furthermore, the transcriptional level of VEGF receptors was elevated. Therefore, the growth of granulation tissues was enhanced by the materials
23
and theoretically created an excellent condition for the subsequent skin transplantation (Archer, Billmire, Falcone & Warner, 2006). As a preliminary research, the limitations were noteworthy. First, the rate of degradation in vivo fell behind the healing process of wounds, so it will require more
IP T
work to realize the synchronization. Second, we focused only on the blood supply of the regenerative abdominal wall but ignored the inflammatory reactions in the
SC R
granulation tissues. Finally, the subsequent effects, such as organ function, associated complications, and the prognosis of skin graft, were not studied. Therefore, further
U
studies are necessary to assess these consequences.
N
Conclusions
A
In this study, we developed a novel injectable hydrogel from Xan-CHO and NOCC.
M
This hydrogel showed various properties, e.g., porosity, self-healing, anti-enzymatic
TE D
hydrolysis, cytocompatibility and promotion of tissue healing. Moreover, this biomaterial presented the advantage of attaining controlled drug release, especially in
EP
tissues with much excretion or exudation such as digestive tracts and open wounds. The hydrogel loaded with VEGF resulted in accelerated angiogenesis and regenerated
CC
the abdominal wall tissue. Altogether, these results demonstrated that this Xan-CHO/NOCC hydrogel could be applied as a drug delivery system in various
A
settings.
24
Declaration of Conflicting Interests We declare no potential conflicts of interest concerning the research, authorship,
SC R
IP T
and/or publication of this article.
Funding
The study was supported by the National Natural Science Foundation of China (Grant
N
U
no. 81571881) and Key Project of Jiangsu Social Development (Grant no.
A
BE2016752).
M
Acknowledgements
We thank Tingting Xu and Yichang Pan at Nanjing Tech University for instructions
A
CC
EP
TE D
regarding Fourier transform infrared spectroscopy and scanning electron microscopy.
25
References: Abdel, G. A., Radwan, R. R., & Ali, H. E. (2016). Radiation Synthesis of Poly(Starch/Acrylic acid) pH Sensitive Hydrogel for Rutin Controlled Release. International Journal of Biological Macromolecules,
IP T
92, 957-964.
Archer, L. P., Billmire, D. A., Falcone, R. J., & Warner, B. W. (2006). Reconstruction of an acquired
SC R
abdominal wall defect in a neonate using acellular human dermis. Plastic and Reconstructive Surgery, 118(7), 163e-166e.
U
Azar, F. K., Crawford, T. C., Poruk, K. E., Farrow, N., Cornell, P., Nadra, O., Azoury, S. C., Soares, K.
N
C., Cooney, C. M., & Eckhauser, F. E. (2017). Ventral hernia repair in patients with abdominal loss of
A
domain: an observational study of one institution's experience. Hernia, 21(2), 245-252.
M
Bueno, V. B., Bentini, R., Catalani, L. H., & Petri, D. F. (2013). Synthesis and swelling behavior of
TE D
xanthan-based hydrogels. Carbohydr Polym, 92(2), 1091-1099.
Cai, J., Yang, Y., Man, J., Huang, J., Wang, Z., Zhang, C., Gu, M., Liu, Q., & Wei, C. (2014). Structural
EP
and functional properties of alkali-treated high-amylose rice starch. Food Chemistry, 145, 245-253.
CC
Chen, S. C., Wu, Y. C., Mi, F. L., Lin, Y. H., Yu, L. C., & Sung, H. W. (2004). A novel pH-sensitive hydrogel composed of N,O-carboxymethyl chitosan and alginate cross-linked by genipin for protein
A
drug delivery. Journal of Controlled Release, 96(2), 285-300.
Deng, Y., Ren, J., Chen, G., Li, G., Guo, K., Hu, Q., Wu, X., Wang, G., Gu, G., & Li, J. (2016). Evaluation of polypropylene mesh coated with biological hydrogels for temporary closure of open abdomen. Journal of Biomaterials Applications, 31(2), 302-314.
26
Fleurke-Rozema, H., van de Kamp, K., Bakker, M., Pajkrt, E., Bilardo, C., & Snijders, R. (2017). Prevalence, timing of diagnosis and pregnancy outcome of abdominal wall defects after the introduction of a national prenatal screening program. Prenat Diagn.
adhesives for wound closure: a tutorial. Chemical Society Reviews, 44(7), 1820-1835.
IP T
Ghobril, C., & Grinstaff, M. W. (2015a). The chemistry and engineering of polymeric hydrogel
SC R
Gialamas, E., Balaphas, A., Assalino, M., Morel, P., Toso, C., & Staszewicz, W. (2017). Complex abdominal wall defect repair using biologic porcine matrix. Minerva Chirurgica.
U
Hamcerencu, M., Desbrieres, J., Popa, M., & Riess, G. (2012). Original stimuli-sensitive polysaccharide
N
derivatives/N-isopropylacrylamide hydrogels. Role of polysaccharide backbone. Carbohydr Polym,
A
89(2), 438-447.
M
Hu, L., Sun, Y., & Wu, Y. (2013). Advances in chitosan-based drug delivery vehicles. Nanoscale, 5(8),
TE D
3103-3111.
Huang, J., Ren, J., Chen, G., Deng, Y., Wang, G., & Wu, X. (2017). Evaluation of the Xanthan-Based
EP
Film Incorporated with Silver Nanoparticles for Potential Application in the Nonhealing Infectious
CC
Wound. Journal of Nanomaterials, 2017, 1-10.
Huang, W., Wang, Y., Chen, Y., Zhao, Y., Zhang, Q., Zheng, X., Chen, L., & Zhang, L. (2016). Strong
A
and Rapidly Self-Healing Hydrogels: Potential Hemostatic Materials. Advanced Healthcare Materials, 5(21), 2813-2822.
Iijima, M., Aubin, H., Steinbrink, M., Schiffer, F., Assmann, A., Weisel, R. D., Matsui, Y., Li, R. K., Lichtenberg, A., & Akhyari, P. (2016). Bio-active coating of decellularized vascular grafts with a
27
temperature-sensitive VEGF-conjugated hydrogel accelerates autologous endothelialization in vivo. J Tissue Eng Regen Med.
Illum, L. (1998). Chitosan and its use as a pharmaceutical excipient. Pharm Res, 15(9), 1326-1331.
IP T
Kikuchi, I. S., Galante, R. S., Dua, K., Malipeddi, V. R., Awasthi, R., Ghisleni, D. D., & de Jesus, A. P. T. (2016). Hydrogel based drug delivery systems: A review with special emphasis on challenges
SC R
associated with decontamination of hydrogels and biomaterials. Current Drug Delivery.
Kim, B. S., Oh, J. M., Kim, K. S., Seo, K. S., Cho, J. S., Khang, G., Lee, H. B., Park, K., & Kim, M. S.
U
(2009). BSA-FITC-loaded microcapsules for in vivo delivery. Biomaterials, 30(5), 902-909.
N
Kim, Y. M., Potta, T., Park, K. H., & Song, S. C. (2017). Temperature responsive chemical crosslinkable
A
UV pretreated hydrogel for application to injectable tissue regeneration system via differentiations of
M
encapsulated hMSCs. Biomaterials, 112, 248-256.
TE D
Koetting, M. C., Guido, J. F., Gupta, M., Zhang, A., & Peppas, N. A. (2016). pH-responsive and enzymatically-responsive hydrogel microparticles for the oral delivery of therapeutic proteins: Effects of
EP
protein size, crosslinking density, and hydrogel degradation on protein delivery. Journal of Controlled
CC
Release, 221, 18-25.
Koh, A., De Vadder, F., Kovatcheva-Datchary, P., & Backhed, F. (2016). From Dietary Fiber to Host
A
Physiology: Short-Chain Fatty Acids as Key Bacterial Metabolites. Cell, 165(6), 1332-1345.
Li, H., Hou, W., & Li, X. (2012). Interaction between xanthan gum and cationic cellulose JR400 in aqueous solution. Carbohydr Polym, 89(1), 24-30.
28
Li, L., Wang, N., Jin, X., Deng, R., Nie, S., Sun, L., Wu, Q., Wei, Y., & Gong, C. (2014a). Biodegradable and injectable in situ cross-linking chitosan-hyaluronic acid based hydrogels for postoperative adhesion prevention. Biomaterials, 35(12), 3903-3917.
Li, L., Wang, N., Jin, X., Deng, R., Nie, S., Sun, L., Wu, Q., Wei, Y., & Gong, C. (2014b). Biodegradable
IP T
and injectable in situ cross-linking chitosan-hyaluronic acid based hydrogels for postoperative adhesion
SC R
prevention. Biomaterials, 35(12), 3903-3917.
Ma, X., Xu, T., Chen, W., Qin, H., Chi, B., & Ye, Z. (2018). Injectable hydrogels based on the hyaluronic
U
acid and poly (gamma-glutamic acid) for controlled protein delivery. Carbohydr Polym, 179, 100-109.
A
Engineering. Advanced Healthcare Materials, 6(1).
N
Motealleh, A., & Kehr, N. S. (2017). Nanocomposite Hydrogels and Their Applications in Tissue
M
Murphy, C. M., & O'Brien, F. J. (2010). Understanding the effect of mean pore size on cell activity in
TE D
collagen-glycosaminoglycan scaffolds. Cell Adh Migr, 4(3), 377-381.
Napavichayanun, S., & Aramwit, P. (2017). Effect of animal products and extracts on wound healing
EP
promotion in topical applications: a review. J Biomater Sci Polym Ed, 1-27.
CC
Peret, B. J., & Murphy, W. L. (2008). Controllable Soluble Protein Concentration Gradients in Hydrogel Networks. Advanced Functional Materials, 18(21), 3410-3417.
A
Poels, J., Abou-Ghannam, G., Decamps, A., Leyman, M., Rieux, A., & Wyns, C. (2016). Transplantation of testicular tissue in alginate hydrogel loaded with VEGF nanoparticles improves spermatogonial recovery. Journal of Controlled Release, 234, 79-89.
29
Ren, J., Wu, X., Hong, Z., & Li, J. (2012). Management of a patient with colocutaneous fistula by autologous platelet-rich fibrin glue. Am Surg, 78(2), 86-87.
Rufaihah, A. J., Johari, N. A., Vaibavi, S. R., Plotkin, M., Di Thien, D. T., Kofidis, T., & Seliktar, D. (2017). Dual delivery of VEGF and ANG-1 in ischemic hearts using an injectable hydrogel. Acta
IP T
Biomaterialia, 48, 58-67.
SC R
Shi, Y., Wang, X., Deng, X., Tian, R., Zhang, Y., Shang, Q., & Chen, N. (2016). Release profile of
insulin from pH-sensitive hydrogel and its hypoglycemic effect by oral administration. J Biomater Sci
U
Polym Ed, 27(1), 86-96.
N
Shin, S. H., Lee, J., Lim, K. S., Rhim, T., Lee, S. K., Kim, Y. H., & Lee, K. Y. (2013a). Sequential
A
delivery of TAT-HSP27 and VEGF using microsphere/hydrogel hybrid systems for therapeutic
M
angiogenesis. Journal of Controlled Release, 166(1), 38-45.
TE D
Shin, S. H., Lee, J., Lim, K. S., Rhim, T., Lee, S. K., Kim, Y. H., & Lee, K. Y. (2013b). Sequential delivery of TAT-HSP27 and VEGF using microsphere/hydrogel hybrid systems for therapeutic
EP
angiogenesis. Journal of Controlled Release, 166(1), 38-45.
Tan, C., Wei, H., Zhao, X., Xu, C., & Peng, J. (2017). Effects of dietary fibers with high water-binding
CC
capacity and swelling capacity on gastrointestinal functions, food intake and body weight in male rats.
A
Food & Nutrition Research, 61(1), 1308118.
Thaunat, O. (2015). Soothing touch of CD31 protects endothelium during cellular immune responses. Proc Natl Acad Sci U S A, 112(43), 13133-13134.
30
Vold, I. M., & Christensen, B. E. (2005). Periodate oxidation of chitosans with different chemical compositions. Carbohydr Res, 340(4), 679-684.
Wang, J., Zhang, F., Tsang, W. P., Wan, C., & Wu, C. (2017). Fabrication of injectable high strength
IP T
hydrogel based on 4-arm star PEG for cartilage tissue engineering. Biomaterials, 120, 11-21.
Yan, S., Wang, T., Feng, L., Zhu, J., Zhang, K., Chen, X., Cui, L., & Yin, J. (2014a). Injectable in situ
SC R
self-cross-linking hydrogels based on poly(L-glutamic acid) and alginate for cartilage tissue engineering. Biomacromolecules, 15(12), 4495-4508.
U
Yan, S., Wang, T., Feng, L., Zhu, J., Zhang, K., Chen, X., Cui, L., & Yin, J. (2014b). Injectable in situ
N
self-cross-linking hydrogels based on poly(L-glutamic acid) and alginate for cartilage tissue engineering.
A
Biomacromolecules, 15(12), 4495-4508.
M
Yin, N., Han, Y., Xu, H., Gao, Y., Yi, T., Yao, J., Dong, L., Cheng, D., & Chen, Z. (2016). VEGF-conjugated alginate hydrogel prompt angiogenesis and improve pancreatic islet engraftment and
TE D
function in type 1 diabetes. Mater Sci Eng C Mater Biol Appl, 59, 958-964.
Yu, L., & Ding, J. (2008). Injectable hydrogels as unique biomedical materials. Chemical Society
EP
Reviews, 37(8), 1473-1481.
CC
Zhang, Y., Tao, L., Li, S., & Wei, Y. (2011). Synthesis of multiresponsive and dynamic chitosan-based
A
hydrogels for controlled release of bioactive molecules. Biomacromolecules, 12(8), 2894-2901.
31
IP T SC R
Fig 1. Synthesis of Xan-CHO/NOCC hydrogels. (a) Xanthan, (b) Xan-CHO, (c)
N
U
chitosan, d) NOCC. Before injection, the drug (VEGF) is carried by Xan-CHO
A
solutions, where hydrogen bonds exist to maintain a stable structure. After injection,
M
the physical interactions are replaced with chemical interactions via the Schiff’s base
A
CC
EP
TE D
reaction, leading to formation of the novel Xan-CHO/NOCC hydrogels.
Fig 2. (A) The appearance of Xan-CHO/NOCC hydrogel, (B) the microstructure of Xan-CHO/NOCC hydrogel. (a) 1%: 0.16%, (b) 1%:0.33%, (c) 1%:0.66%, (d) 1%:1%,
32
(e) 1%:2%, (f) pore sizes at different mass ratios of Xan-CHO and NOCC. (Bar =100
U
SC R
IP T
µm).
N
Fig 3. (A) The results of the rheological property, strain=1.0%, [mass ratio of
A
Xan-CHO/NOCC: (a) 1%:0%, (b) 1%:0.16%, (c) 1%:0.33%, (d) 1%:0.66%, (e)
M
1%:1%, (f) 1%:2%], (B) the gelation time and G’ of each hydrogel with varied mass
EP
TE D
ratios.
CC
Fig 4. (A) Self-healing process of Xan-CHO/NOCC hydrogel, (B) lifting up the Xan-CHO/NOCC hydrogel after self-healing, (C) local injection of Xan-CHO/NOCC
A
hydrogel and fibrin gel in PBS (red arrow: Xan-CHO/NOCC hydrogel, green arrow: fibrin gel), (D) cumulative releasing ratio of BSA-FITC after local injection in PBS.
33
IP T
Fig 5. (A) Swelling ratio of Xan-CHO/NOCC hydrogel with varied mass ratios, (B) degradation of Xan-CHO/NOCC hydrogel (1%:0.33%) in acid, neutral and alkaline
SC R
buffers within 24 hours, (C) anti-enzymatic hydrolysis of Xan-CHO/NOCC hydrogel
M
A
N
U
and fibrin gel (red arrow: Xan-CHO/NOCC hydrogel, green arrow: fibrin gel).
TE D
Fig 6. Cytocompatibility of Xan-CHO/NOCC hydrogel. (a)-(d) Co-culture of fibroblasts in different proportions of leachate (12.5%, 25%, 50% and 100%,
EP
respectively) for 24 hours, (e)-(h) co-culture of fibroblasts in different proportions of leachate (12.5%, 25%, 50% and 100%, respectively) for 48 hours, 100×, (i) culture of
CC
fibroblasts on the surface of hydrogel for 48 hours, 100×, (j) culture of fibroblasts
A
between the hydrogel and the surface of the plastic plate for 48 hours, 100×, (k) culture of the spherical cells in (i) on the plastic plate for 24 hours, 100×, (l) CCK-8 assay detecting cell-proliferation by culturing fibroblasts in different proportions of leachate for 24 hours and 48 hours. 34
IP T SC R
U
Fig 7. Degradation of Xan-CHO/NOCC hydrogel in vivo. (A) The appearance of the
N
hydrogel and the surrounding tissues at the scheduled time (black arrow: remained
A
hydrogels), (B) remaining mass of hydrogel after in vivo degradation, (C) H&E
CC
EP
TE D
M
staining of superficial skin, 10×.
Fig 8. (A) H&E and Masson staining of regenerative abdominal wall tissues (green
A
arrows: native tissues, red framework: regenerative tissues) (bar=1 mm), (B) schematic diagram of tissue growth promoting effects through hydrogel-to-host interactions, (C) thickness of regenerative abdominal wall tissues. *P<0.05, **P<0.01.
35