Carbohydrate Polymers 195 (2018) 63–70

Contents lists available at ScienceDirect

Carbohydrate Polymers
journal homepage: www.elsevier.com/locate/carbpol

Antibacterial and hemostatic hydrogel via nanocomposite from cellulose T

nanofibers
⁎ ⁎
Rui Liua, Lin Daia,b, , Chuanling Sia,b, , Zhaogang Zengc
a
Tianjin Key Laboratory of Pulp and Paper, Tianjin University of Science and Technology, Tianjin 300457, PR China
b
State Key Laboratory of Pulp and Paper Engineering, South China University of Technology, Guangzhou 510640, PR China
c
Xiangtan Sepiolite Technology Co., Ltd., Xiangtan 411100, PR China

A R T I C LE I N FO A B S T R A C T

Keywords: Bacterial infection and uncontrolled bleeding are the major challenges facing the wound treatment. In order to
Nanoellulose solve these problems, we have devised a green nanocomposite hydrogel by introducing the aminated silver
Cellulose nanofibers nanoparticles (Ag-NH2 NPs) and gelatin (G) to carboxylated cellulose nanofibers (CNF). Interpenetrating poly-
Hydrogel meric network (IPN) was formed by interaction of multicomponent, leading to the non-covalent (dynamic ionic
Nanoparticle
bridges) crosslinked hydrogel CNF/G/Ag. The produced hydrogel dressing with 0.5 mg/mL Ag-NH2 NPs (CNF/
Nanocomposite
G/Ag0.5) demonstrated stronger mechanical, self-recovery, antibacterial properties, satisfactory hemostatic
performance, and appropriate balance of fluids on the wound bed (2093.9 g/m2 per day). More importantly, the
wound healing model evaluation in vitro and in vivo of CNF/G/Ag0.5 showed an outstanding biocompatibility
(∼100% infected cell viability) and wound healing efficacy (∼90% healed and 83.3% survival after 14 days).
Our study paved a highly promising approach to improve the performance of cellulose-based hydrogel dressing
and would also be useful for developing ideal skin wound dressings by other green materials.

1. Introduction
The skin is the essential interface between the body and its en-
vironment which can heal itself when the wound is narrow and small.
However, the large wound of human skin is prone to infection and
difficult to heal. Therefore, to help wound healing, appropriate treat-
ments to the wound are necessary (Li et al., 2015). The conventional
wound dressing in clinical application is natural or synthetic bandages,
cotton wool, and gauzes, which may need long-term treatment, be in-
effective, or even adhere to desiccated wound surfaces (Radhakumary,
Antonty, & Sreenivasan, 2011). In order to overcome many of these
drawbacks, numerous wound dressing materials were being in-
vestigated (Çalamak, Erdoğdu, Özalp, & Ulubayram, 2014; Fan et al.,
2014; Ignjatović et al., 2016; Kang et al., 2017; Mei et al., 2017).
Among them, hydrogel-based wound dressing have been paid special
attentions due to that it can possess many essential properties, such as
providing a cooling sensation, a moist environment, allowing gaseous
exchange and wound exudate absorption (Cheng et al., 2017; Li et al.,
2017; Zhao et al., 2017). Unfortunately, up to now, there is almost no
clinically used hydrogel satisfying with all the requirements of the ideal
skin wound dressing.
The leading challenge facing the wound dressing is the uncontrolled
hemorrhage which is often the direct cause of the trauma deaths
(Gaston, Fraser, Xu, & Ta, 2018; Konieczynska et al., 2017). Many
clinical hemostatic formulations work with the use of hemostatically
active proteins such as fibrin, collagen, and thrombin, but may be dif-
ficult to store and use out of the hospital. Gelatin (G) is an interesting
candidate as a hemostatic based on its low prices, widely distribute,
hydrolysis into simple byproducts, and reliable hemostatic effect (Chen,
Guo et al., 2016; Lan et al., 2015; Mele, 2016). For the aspect of wound
contamination, the major reason is bacterial infection causing by Sta-
phylococcus aureus and Pseudomonas aeruginosa, which can easily enter
the body through the wounds, reach into deeper portions of the tissue,
furthermore, even lead to septicemia and death. In order to reduce the
abuse of broad-spectrum antibiotic, many antibiotics alternatives, such
as silver nanoparticles (Ag NPs) with low bacterial resistance and side
effects, have been widely used as topical antibacterial agents (Chen,
Zhang et al., 2016; Dhand et al., 2016; Nie et al., 2016). But the de-
velopment and application of nanomaterials has generated public de-
bate on the safety of nanotechnology and has been intervened by some
supervisory departments. This issue could potentially be overcame by
incorporation of nanoparticles into hydrogels resulting in decreased
risks to human health and the environment (Foss Hansen et al., 2016;
Thoniyot, Tan, Karim, Young, & Loh, 2015).

⁎
Corresponding authors. Present address: Tianjin University of Science and Technology, No. 29 at 13th Avenue, TEDA, Tianjin 300457, PR China.
E-mail addresses: dailin@tust.edu.cn, dailin1989@163.com (L. Dai), sichli@tust.edu.cn (C. Si).

https://doi.org/10.1016/j.carbpol.2018.04.085
Received 26 February 2018; Received in revised form 19 April 2018; Accepted 21 April 2018
Available online 22 April 2018
0144-8617/ © 2018 Elsevier Ltd. All rights reserved.
R. Liu et al.
Interpenetrating polymeric network (IPN) is a 3D network com-
posed of two or more different types of materials, which are partially or
fully interlaced on a molecular scale but not covalently bonded to each
other and cannot be separated unless chemical bonds are broken
(Dragan, 2014). Multicomponent hydrogels based on IPNs have shown
structural diversity, versatility, significant enhancement in the me-
chanical performance for tissue healing. Cellulose nanofibers (CNF), a
kind of green functional material, have been widely studied in the field
of food and biomedical field due to their nontoxic, biocompatible,
biodegradable, and environmental and economic sustainability
(Fernandes, Pires, Mano, & Reis, 2013; Liu et al., 2018; Saito, Kimura,
Nishiyama, & Isogai, 2007). TEMPO-oxidized cellulose nanofibers
(carboxylated CNF) have good hydrophilicity and water retention,
which can helpful to achieve good performance on blood absorption
and water vapor transmission. To the best of our knowledge, cellulose-
based multifunctional hydrogel dressing was rarely reported. It is well
accepted that cellulose and other biopolymers are promising building
blocks of sustainable materials with tailored characteristics (Fan et al.,
2017).
In this article, we presented a novel design for a CNF/G/Ag nano-
particles hydrogel as a wound dressing which did meet the controlling
of evaporative water loss, the stopping hemorrhage, and the good an-
tibacterial and biocompatible properties to promote wound healing.
The physical and chemical properties, bactericidal activities, safety as
well as the wound healing efficiency were comparatively investigated.
2. Experimental details
2.1. Reagents and materials
TEMPO-oxidized cellulose nanofibers (carboxylated CNF, surface
carboxylate density 1.1 mmol/g) were provided free by Dr. Yangbing
Wen (Tianjin University of Science and Technology, China). Polyvinyl
pyrrolidone (PVP, K15, MW = 10000), AgNO3, ethylene glycol (EG), 3-
aminopropyltriethoxysilane (APTES), gelatin (G) were bought form
bought from Sinopharm Chemical Reagent Co. Simulated body fluid
(SBF, pH 7.4, 142 mM Na+, 5 mM K+, 2.5 mM Ca2+, 148 mM Cl−,
4.2 mM HCO-3, 1 mM HPO2-4, and 5 mM SO2-4) was obtained from
Qingdao Jisskang Biotechnology Co., Ltd. Dulbecco’s Modified Eagle’s
Medium (DMEM, with high glucose, L-glutamine, sodium pyruvate),
fetal bovine serum (FBS), amphotericin B, phosphate-buffered saline
(PBS), streptomycin, and penicillin were all purchased from Invitrogen.
Cell-Counting Kit-8 (CCK-8) was received from the Dojindo
Laboratories (Beijing, China).
2.2. Animals and ethics
Adult Kunming mice (male, 18–22 g) used in the study were ob-
tained from Beijing HFK Biosciece Co., Ltd. (Beijing, China). The mice
are housed in the animal laboratory of Institute of Process Engineering
(Chinese Academy of Sciences, Beijing, China) in a controlled en-
vironment (22 ± 2 °C, 55 ± 5% relative humidity level, a 12 h light/
dark cycle). The animal experiments were in accordance with the
guidelines set by the National Institutes of Health (NIH Publication No.
85-23) and were approved by the Beijing Experimental Animal Ethics
Committee.
2.3. Preparation of aminated Ag NPs (Ag-NH2 NPs)
Ag-NH2 NPs were accomplished by the procedure reported in lit-
erature (Wang, Gao, Sun, Su, & Gao, 2016; Zhang et al., 2010). 1 g
AgNO3 completely dissolved in 25 g PVP/200 mL EG solution at room
temperature. Then, the mixture solution was heated to 120 °C at the
rate of 1 °C/min, and kept the reaction at 120 °C for 1 h. After natural
cooling, the solution was added was precipitated with acetone. The Ag
NPs were obtained by ultracentrifugation and continuous reacted in
64
Carbohydrate Polymers 195 (2018) 63–70
100 mL NaOH aqueous solution (0.5 mol/L) at 70 °C for 12 h. The re-
sulting product was ultracentrifugated, washed, redispersed in me-
thanol/water (9/1 v/v) solution, and followed by amination with
APTES at 70 °C for 12 h. Finally, the solids were collected, washed with
water (2 × 500 mL), and dried under vacuum at 30 °C to give Ag–NH2
NPs.
2.4. Preparation of CNF/G/Ag hydrogel dressings
Carboxylated CNF (200 mg) was completely dissolved in 14 mL
deionized water under room temperature and disposed by ultrasonic to
obtain the uniformly CNF hydrogel. Then 0.5 mL gelatin solution
(80 mg/mL) and different amounts of Ag-NH2 NPs solution were added
into the CNF hydrogel at the same time. The entire mixture kept under
vigorous stirring for 2 h at room temperature to form non-covalent
crosslinked hydrogels. Finally, the hydrogels were poured into a petri-
dish and freeze-dried overnight to obtain wound dressing and named as
CNF/G/Ag. The samples containing 0.2 mg/mL and 0.5 mg/mL Ag-NH2
NPs were abbreviated as CNF/G/Ag0.2 and CNF/G/Ag0.5, respectively.
2.5. Characterization
After coating with 5 nm of Pt/Pd, the morphology of samples was
observed with a Hitachi S-4800 field-emission scanning electron mi-
croscope (FE-SEM). The structure and phase composition of samples
were characterized by Shimadzu XRD-6100 wide-angle X-ray diffrac-
tion (WAXD) using Ni-filtered CuKα radiation (40 kV, 30 mA) with 4°/
min scanning rate at room temperature. Diffraction intensity was
measured in a range of 2θ = 5–70°. The changes in the functional
groups were studied by Varian 670-IR fourier transform infrared spec-
troscopy (FTIR) at room temperature in the spectral range of
4000–400 cm−1. Rheological characterization of the hydrogels was
measured and analyzed using a TA Instruments AR2000 rheometer with
a 25 mm diameter parallel plate. Samples were placed between the
37 °C pre-heated plates of the rheometer. The self-recovery properties of
CNF/G/Ag0.5 were detected by dynamic strain amplitude cyclic test
(γ = 1% for 120 s and γ = 80% for 60 s) at 37 °C.
2.6. Blood clotting assay
The fresh whole blood samples (1 mL) were collected from mice,
and heparin was added immediately. The whole blood clotting assay
was according to Ong, Wu, Moochhala, Tan, and Lu (2008). The hy-
drogel dressings (0.5 cm square) were pre-warmed to 37 °C and fully
immerse into the blood solution, and then 10 mL of 0.2 M CaCl2 solu-
tion added. All the samples were placed at 37 °C with 30 rpm shaken for
15 min. Free erythrocytes were hemolyzed with 20 mL of deionized
water, and the absorbance of the resulting hemoglobin solution was
measured at were detected at 541 nm using an Infinite M200 microplate
spectrophotometer from three independent tests.
2.7. Platelet adhesion
The fresh whole blood samples (1 mL) were centrifuged at 200 rpm
at 4 °C for 15 min. The pellets were further centrifuged at 1500g at 4 °C
for 10 min, and then removed and resuspended in a 2.5 mM CaCl2/
1.0 mM MgCl2.soultion. The platelet adhesion tests were adapted from
Lih, Jung, Joung, Ahn, and Han (2016). The hydrogel dressing samples
(0.5 cm square) were immerse into 2 mL platelet suspension, incubated
at 37 °C for 1 h, washed twice to remove the nonadherent and loosely
attached platelets with PBS buffer. Adhered platelets on the samples
were lysed using 1 mL of 2% (v/v) Triton X-100 at 37 °C for 15 min.
Then, 100 μL adhered platelet solution was seeded in 96-well plate with
adding 100 μL lactate dehydrogenase activity assay kit and detected at
450 nm.
R. Liu et al.
Fig. 1. Schematic illustration of the prepared aminated silver nanoparticles (Ag-NH2 NPs), CNF/G/Ag hydrogel dressing and mechanisms of crosslinking reactions of
multicomponent.
Fig. 2. FE-SEM micrographs of CNF, Ag-NH2 NPs, and the interactions of different component in CNF/G/Ag0.5.
2.8. Thrombin generation
The hydrogel dressing samples (0.5 cm square) were incubated with
1 mL of whole blood at 37 °C. 30 min later, sodium citrate (20 mL,
0.633 M) was added to stop thrombin generation. The level of throm-
bin–antithrombin complex was tested by ELISA method.
2.9. Equilibrium simulated body fluid and blood absorption
The absorption capacities of hydrogel dressings were determined by
swelling the simulated body fluid (SBF) and citrated whole blood at
37 °C. A known weight (∼100 mg, Wdry) of freeze-dried dressing was
placed in the media for the enough period of time. And then the wet
weight until equilibrium swelling (Wswollen) was detected by removing
freely draining liquid and calculated by the following formula:
(Wswollen − Wdry )
Equilibrium fluid uptake (g/g) =
Wdry
2.10. Water vapor transmission rate
The hydrogel dressings with a diameter of 45 mm were fastened on
the mouth of cylindrical plastic cups (40 mm diameter) containing
Carbohydrate Polymers 195 (2018) 63–70
15 mL water and placed in an incubator at 37 °C and 35% relative hu-
midity. The whole system was weighed at set intervals of time. The
water vapor transmission rate (WVTR) was calculated by the following
formula,
WVloss
WVTR =
AT (2)
where WVloss is the weight of water vapor loss after time T, A is the test
area of the sample in square meter. Experiments were done in triplicate.
2.11. Antibacterial measurement
The evaluation of the antibacterial activity of the hydrogel dressings
was determined by two methods: the optical density and disk diffusion.
Quantitative analyses were performed in triplicate for S. aureus (ATCC
29213) and P. aeruginosa (ATCC 15692). Microbial inocula were pre-
(1) pared by subculturing bacteria in 20 mL of Tryptic Soy Broth at 37 °C
overnight and then diluted to approximately 105–106 CFU/mL. The
microorganisms were treated with hydrogel dressings (0.5 cm square)
at 37 °C. The absorbance of the media at 600 nm was measured at
different time. The agar disk diffusion method were according to the
standardized method by Moosdeen, Williams, and Secker (1988). CNF,
CNF/G/Ag0.2, and CNF/G/Ag0.5 were added to individual inoculated
65
R. Liu et al.
Fig. 3. FTIR (a) and WAXD (b) spectra of the hydrogels.
6
agar plates (10 CFU/mL), respectively, which were then incubated at
37 °C overnight. Three replicate experiments were performed. Zones of
inhibition were presented as the diameter of the area of no bacterial
growth minus the diameter of the sample itself.
Fig. 4. Rheological properties of the samples at 37 °C. The values of the storage moduli (G′, solid) and the loss moduli (G″, open) on frequency (a) and strain sweep
(b). Viscosity (c) and photographs (d) of hydrogel at different Ag-NH2 NPs loadings. Dynamic strain amplitude cyclic test (γ = 1% for 120 s and γ = 80% for 60 s) of
CNF/G/Ag0.5 (G′ solid lines, G″ dash lines) (e).
66
Carbohydrate Polymers 195 (2018) 63–70
2.12. Cellular viability by CCK-8 assay
A normal neonatal human dermal fibroblasts (NHDF) model was
used to mimic infected wounds. NHDF were cultured in DMEM with
high glucose, L-glutamine, sodium pyruvate, 10% FBS, 1% penicillin–-
streptomycin, and 0.25 mg/mL amphotericin B. Cells were added into
48-well plates at a density of 5 × 103 cells/well and incubated at 37 °C
in a humidified incubator with 5% CO2 overnight. 10 μL of S. aureus or
P. aeruginosa (5 × 103 CFU/mL) suspension was added to the surface of
NHDF monolayer. Then the hydrogel dressing with a diameter of 1 mm
was placed on top of the contaminated NHDF. After cultured 24 h, the
cytotoxicity was measured by CCK-8 assay and detected by Tecan M200
multimode reader at 450 nm (Dai, Liu, Hu, Zou, & Si, 2017; Unger,
Wittmar, & Kissel, 2007).
2.13. In vivo wound healing
Eighteen adult Kunming mice (male, 18–22 g) were randomly di-
vided and evenly into three groups to evaluate the wound healing
characteristics. On the first day, one 8 cm thickness skin wound was
prepared on the dorsum of each mouse, and covered with the dressing
samples tightly and completely. During treatment, mice were mon-
itored for wound sizes and body weights every other day. The reduction
in wound size was defined by the following formula,
S0 − St
Wound size(%) = × 100%
S0 (3)
where S0 and St represent the wound area at first day and at a time
interval, respectively.
2.14. Statistical analysis
All the experiments were replayed at least three times. The data
were presented as the means ± standard deviation (SD). T-test or
ANOVA was performed in statistical evaluation. A p-value < 0.05 was
considered statistically significant.
R. Liu et al.
Fig. 5. The blood clotting results as measured by absorbance of haemoglobin (a), platelet adhesion (b), thrombin generation as evaluated from thrombin–anti-
thrombin complex (c), fluid uptake of different samples (d), and water vapour transmission loss from hydrogel dressings (d).
3. Results and discussion
3.1. Structural evaluation
Our products consisted of four components: water, carboxylated
cellulose nanofibers (CNF), gelatin (G), and aminated silver nano-
particles (Ag-NH2 NPs). The samples containing 0.2 mg/mL and
0.5 mg/mL Ag-NH2 NPs were abbreviated as CNF/G/Ag0.2 and CNF/G/
Ag0.5, respectively. Carboxylated CNF displayed mostly negative car-
boxyl groups’ functionality that entangled with one another and dis-
persed uniformly by attractive interactions of negative-charged surface
carboxylic groups with Ag-NH2 NPs. But it is considered here that there
was not only one interaction, at the same time, gelatin with amine and
carboxyl groups also can involve in the attraction each other and co-
operate in the construction of the structure, resulting in the formation
of hydrogel (CNF/G/Ag) with IPN structure (Fig. 1). The theoretical
reasoning described above was supported by the experimental ob-
servations from field-emission scanning electron microscope (FE-SEM)
of CNF, Ag-NH2 NPs (∼20 nm), and CNF/G/Ag hydrogel (Fig. 2). The
microstructures of attractive interactions between CNF, Ag-NH2 NPs,
and gelatin in CNF/G/Ag0.5 were performed clearly from high power
images.
The FTIR analysis was also confirmed the occurrence of the inter-
actions (Fig. 3a). The FTIR spectrum of CNF/G showed typical char-
acteristic peaks of carboxylated CNF and G and agrees with literature.
For carboxylated CNF, the characteristic peaks at 3600–3200 cm−1
(OeH streching vibrations), 2890 cm−1 (overlapping symmetric and
asymmetric CeH streching), 1570–1600 cm−1 (carboxylate groups),
1313 and 1047 cm−1 (CeO and CeOeC of polysaccharide structure of
cellulose) were visible (Chen et al., 2011). Gelatin (G) exhibited char-
acteristic absorption bands at 3290 cm−1 (due to OeH streching vi-
bration of water molecules and N−H streching), 2967 cm−1 (NeH
steching) (Hosseini, Rezaei, Zandi, & Ghavi, 2013). The main absorp-
tion bands for peptide bond of gelatin at 1646 cm−1 (amide I, C]O and
67
Carbohydrate Polymers 195 (2018) 63–70
CeN stretching) was disappeared, confirming that interaction between
carboxyl groups (eCOOH) of CNF and amino groups (eNH2) of gelatin
occurred. The small peak at 1558 cm−1 belongs to C]O and CeN
stretching vibration (assigned to amino group II). This could be due to
C]O is involved in the reaction with CNF. The decreased in the in-
tensity at 1234 cm−1 associated with the cooperation of NeH from
amino group II in the reaction. Compared with CNF/G, CNF/G/Ag0.2
demonstrated less difference due to the little amount of Ag-NH2 NPs
addition. FTIR spectrum of CNF/G/Ag0.5 showed the increased in the
intensity at 1646 cm−1, 1558 cm−1, and 1234 cm−1 associated with the
cooperation of excess Ag-NH2 NPs in the hydrogel (Naseri, Deepa,
Mathew, Oksman, & Girandon, 2016).
The wide-angle X-ray diffraction patterns were illustrated in Fig. 3b.
Due to the amorphous nature of gelatin showed a dispersive broad peak
around 20°. The diffraction peaks at 14.2°, 16.9°, 22.4°, and 34.4°
(marked with *) were attributed to cellulose (Klemm, Heublein, Fink, &
Bohn, 2005). The sharp peaks at 37.8, 44.0, and 64.3 in corresponding
to lattice planes 111, 200, and 220 of highly crystalline silver, respec-
tively (Sun & Xia, 2002), which confirmed the incorporation of Ag-NH2
NPs into hydrogel.
3.2. Viscoelastic behavior
One of the most salient-featured by natural materials is the colloidal
reinforcement bound together by dynamic bonds on molecular level,
allowing dissociation and reformation (Wegst, Bai, Saiz, Tomsia, &
Ritchie, 2014). The efficient stress transfer by multicomponent inter-
action was studied in more details. All these samples show a single
plateau region in the dynamic moduli, and the storage moduli (G′)
values have a substantial elastic response and dominate over the loss
moduli (G″) values in the whole range of frequencies (Fig. 4a). The
enhancement of mechanical properties was easy controlled by the ad-
dition of Ag-NH2 NPs, and gelatin. CNF/G/Ag0.5 exhibited greater G′
value and viscosity which proved the interaction role of Ag-NH2 NPs. As
R. Liu et al.
Fig. 6. Antibacterial activity (a, b) and zones of inhibition (c) for S. aureus and P. aeruginosa up to 15 h. Illustration (d) and cell viability (e) of the wound infection
model by dressing treatment.
shown in Fig. 4b, the strain sweep measurements showed a significant
decrease in G′ and G″ with the increase of applied strain. Furthermore,
as the shear rate increased, the viscosity of all the samples decreased
(Fig. 4c). The collapse of the dynamical interaction and shear thinning
phenomenon were just because of the non-covalent interactions of CNF,
Ag-NH2 NPs, and gelatin in the hydrogel. Interesting to note mechanical
properties of CNF/G/Ag0.5 can be recovered after the 80% strain within
a short time (∼20 s) (Fig. 4d). This means that even an unavoidable
deformation of wound dressing treatment, such as joint movement, the
reformation of IPN structure between multicomponent was able to fast
recovery at room temperature (Yang, Zhang, Ma, & Xu, 2015).
3.3. In vitro blood test
In order to evaluate blood clotting performance, the samples were
completely immersed in the whole blood for 15 min. CNF/G/Ag0.2 and
CNF/G/Ag0.5 led to significantly lower absorbance values of free he-
moglobin than CNF (Fig. 5a) which represented a better blood clotting
capacity of CNF/G/Ag. Blood clotting capacity of CNF/G/Ag0.5 was
slightly lower than CNF/G/Ag0.2 which could be due to unfulfilled
potential of higher crosslinking degree within a short time. Platelet has
a major role in hemostasis when blood contact with foreign surfaces.
The amount of thrombin–antithrombin complex can reflect the degree
of thrombin formulation over a period of time. As shown in Fig. 5b and
c, considerably more platelets and thrombin–antithrombin complex
were detected on CNF/G/Ag hydrogels and in the blood samples after
incubated with CNF/G/Ag, respectively, compared to CNF. Overall, the
68
Carbohydrate Polymers 195 (2018) 63–70
results of the blood test in vitro (blood clotting, platelet adhesion, and
thrombin generation) revealed that the CNF/G/Ag have good hemo-
static properties, which may be benefited from the multicomponent and
IPN structure of the hydrogels. Each component has different me-
chanism of coagulation. Uncrossinked and protonated amine groups of
Ag-NH2 NPs and gelatin could attracte negatively-charged residues on
red blood cell membranes, and thus displayed strong hemagglutination.
Gelatin can promote the erythropoiesis, and increase the number of
platelets and white blood cells to blockade bleeding (Broderick et al.,
2005). In addition, the hydrogel formulations adhered to the surface of
the wound and led to platelet aggregation.
3.4. Fluid uptake and water vapour transmission
The absorption efficiency of the samples was evaluated by in-
cubating for 3 h in simulated body fluid (SBF) and whole blood
(Fig. 5d). The SBF absorption of CNF/G/Ag was just a little higher than
CNF. Noteworthy here was the blood absorption efficiency, CNF/G/
Ag0.5 showed about 2-fold higher than CNF. The significantly improved
absorption may be due to specific attraction of blood proteins and other
blood components. The good blood clotting performance of CNF/G/
Ag0.5 should contribute to the blood absorption and blockade bleeding.
It has been reported that the rate of evaporative water loss was
204 g/m2 per day for normal skin and 279 g/m2 per day for injured
skin, respectively (Lamke, Nilsson, & Reithner, 1977). An ideal water
vapor transmission rate (WVTR) of wound dressing should be
2000–2500 g/m2 per day to achieve the appropriate surface moisture
R. Liu et al. Carbohydrate Polymers 195 (2018) 63–70

Fig. 7. Photographs (a), wound size change (b), weight change (c), and survival curves (d) of mice by different dressing treatment.

and controllable rate of water loss from the wound (Balakrishnan,
Mohanty, Umashankar, & Jayakrishnan, 2005). WVTR of CNF, CNF/G/
Ag0.2, and CNF/G/Ag0.5 was 3431.5, 2372.6, and 2093.9 g/m2 per day
(calculated from Fig. 5e), respectively. The high WVTR of CNF would
result in rapid dehydration and dressing adhesion to desiccated wound
surfaces. CNF/G/Ag samples, by contrast, have appropriate values. The
permeation of water vapor through hydrogel dressing should pass ad-
sorption and diffusion steps. The moderate crosslinked hydrogel could
increase water absorption efficiency and decrease the water vapor
diffusion, as consequence decreased the WVTR to keep an appropriate
balance of fluids on the wound bed.
3.5. Antibacterial activity and cytotoxicity
S. aureus and P. aeruginosa were treated with different samples, and
bacterial growth was recorded by measuring two methods: the optical
density of the liquid culture medium (Fig. 6a and b) and the agar disk
diffusion (Fig. 6c). For the agar disk diffusion method, antibacterial
efficiency was rated “good” (zone of inhibition > 1 mm), “fairly good”
(zone of inhibition ≤1 mm), “sufficient” (just not growth up on the
sample), “limited” (amount of growth on the sample < 50%) or “poor”
(amount of growth on the sample ≥50%) (Pollini, Russo, Licciulli,
Sannino, & Maffezzoli, 2009). Since the well-known bioactivity, Ag NPs
have been widely used as coatings for medical devices to prevent in-
fections. The treatment of two most common bacteria with CNF/G/Ag
inhibited their growth and CNF/G/Ag0.5 showed the better perfor-
mance, while the treatment with CNF did not. In order to investigate
the practical effect of CNF/G/Ag hydrogel dressing, a wound infection
model was established to simulate the therapeutic process (Fig. 6d). The
cell viability was tested after incubating normal neonatal human
dermal fibroblasts (NHDF) and the infected NHDF treated with different
hydrogel dressings for 24 h. All the dressings could accelerate the
growth of cells. It is worth noting that CNF/G/Ag0.5 presented highest
satisfactory infected cell viability (> 100%), which may benefit from
the growth-promoting effect of gelatin and also proved that the CNF/G/
Ag hydrogel dressing held the ability of antibacterial infection and
accelerate cell proliferation.
3.6. In vivo wound healing
All the above results gave us great confidence to evaluate the real
wound healing efficacy of CNF/G/Ag hydrogel dressing in vivo. Fig. 7a
and b shows the optical images and tracing detection of minor wounds
(0.8 cm) treated with CNF, CNF/G/Ag0.2 and CNF/G/Ag0.5 hydrogel
dressing for 1–14 days. The CNF group had little effect on the wound
size during the treatment. As expected, CNF/G/Ag groups had much
declined size of the wound than that of CNF, specifically that wound
has been healed ∼90% after treated with CNF/G/Ag0.5 for 14 days
(Fig. 7b). On the other hand, during the first week of all the treatments,
bacterial infection of the wound was reflected by weight loss to a cer-
tain degree (Fig. 7c). However, the treatment with CNF/G/Ag0.5 hy-
drogel dressing displayed a clear advantage in survival rate (83.3%)
compared with other groups (Fig. 7d). Such nice properties and wound
healing efficacy of CNF/G/Ag0.5 could benefit from the synergy of
multicomponent.
4. Conclusion
In conclusion, a new nanocomposite hydrogel dressing has been
prepared based on cellulose nanofibers (CNF) by crosslinking with ge-
latin (G), and aminated silver nanoparticles (Ag-NH2 NPs).
Incorporation of multicomponent effectively improved in the

69
R. Liu et al.
performance of mechanical, self-recovery, hemostatic (gelation), anti-
bacterial properties (Ag-NH2 NPs), and fluid balance on the wound bed.
Although the hemostatic effect of CNF/G/Ag0.5 hydrogel dressing de-
creased slightly with the increase of Ag-NH2 NPs content, it still reached
a satisfactory level. The wound healing evaluation in vitro and in vivo
demonstrated that the CNF/G/Ag0.5 hydrogel dressing exhibited good
biocompatibility, outstanding efficacy of antibacterial and wound
healing over a short period. In brief a few words, this concept and
approach of nanocomposite hydrogel with multicomponent would
promote the development of cellulose-based, nanocomposite, and other
green materials for potential wound dressing applications.
Conflict of interest
The authors declare no competing financial interest.
Acknowledgements
The authors wish to thank Dr. Yangbing Wen for preparing the
TEMPO-oxidized cellulose nanofibers, and the financial support of this
research by the National Natural Science Foundation of China
(21706193), Young Elite Scientists Sponsorship Program by Tianjin
(TJSQNTJ-2017-19), Natural Science Foundation of Tianjin
(17JCQNJC05200) and the State Key Laboratory of Pulp and Paper
Engineering (201768, 201822).
References
Balakrishnan, B., Mohanty, M., Umashankar, P. R., & Jayakrishnan, A. (2005). Evaluation
of an in situ forming hydrogel wound dressing based on oxidized alginate and gelatin.
Biomaterials, 26(32), 6335–6342.
Broderick, E. P., O'Halloran, D. M., Rochev, Y. A., Griffin, M., Collighan, R. J., & Pandit, A.
S. (2005). Enzymatic stabilization of gelatin-based scaffolds. Journal of Biomedical
Materials Research Part B: Applied Biomaterials, 72B(1), 37–42.
Çalamak, S., Erdoğdu, C., Özalp, M., & Ulubayram, K. (2014). Silk fibroin based anti-
bacterial bionanotextiles as wound dressing materials. Materials Science and
Engineering: C, 43(Suppl. C), 11–20.
Chen, W., Yu, H., Liu, Y., Chen, P., Zhang, M., & Hai, Y. (2011). Individualization of
cellulose nanofibers from wood using high-intensity ultrasonication combined with
chemical pretreatments. Carbohydrate Polymers, 83(4), 1804–1811.
Chen, C., Zhang, T., Dai, B., Zhang, H., Chen, X., Yang, J., et al. (2016a). Rapid fabrication
of composite hydrogel microfibers for weavable and sustainable antibacterial appli-
cations. ACS Sustainable Chemistry & Engineering, 4(12), 6534–6542.
Chen, H., Guo, L., Wicks, J., Ling, C., Zhao, X., Yan, Y., et al. (2016b). Quickly promoting
angiogenesis by using a DFO-loaded photo-crosslinked gelatin hydrogel for diabetic
skin regeneration. Journal of Materials Chemistry B, 4(21), 3770–3781.
Cheng, F., Liu, C., Wei, X., Yan, T., Li, H., He, J., et al. (2017). Preparation and char-
acterization of 2,2,6,6-Tetramethylpiperidine-1-oxyl (TEMPO)-oxidized cellulose
nanocrystal/alginate biodegradable composite dressing for hemostasis applications.
ACS Sustainable Chemistry & Engineering, 5(5), 3819–3828.
Dai, L., Liu, R., Hu, L. Q., Zou, Z. F., & Si, C. L. (2017). Lignin nanoparticle as a novel
green carrier for the efficient delivery of resveratrol. ACS Sustainable Chemistry &
Engineering, 5(9), 8241–8249.
Dhand, V., Soumya, L., Bharadwaj, S., Chakra, S., Bhatt, D., & Sreedhar, B. (2016). Green
synthesis of silver nanoparticles using Coffea arabica seed extract and its antibacterial
activity. Materials Science and Engineering: C, 58(Suppl. C), 36–43.
Dragan, E. S. (2014). Design and applications of interpenetrating polymer network hy-
drogels. A review. Chemical Engineering Journal, 243, 572–590.
Fan, Z., Liu, B., Wang, J., Zhang, S., Lin, Q., Gong, P., et al. (2014). A novel wound
dressing based on Ag/graphene polymer hydrogel: Effectively kill bacteria and ac-
celerate wound healing. Advanced Functional Materials, 24(25), 3933–3943.
Fan, J., Li, T., Ren, Y., Qian, X., Wang, Q., Shen, J., et al. (2017). Interaction between two
oppositely charged starches in an aqueous medium containing suspended mineral
particles as a basis for the generation of cellulose-compatible composites. Industrial
Crops & Products, 97, 417–424.
Fernandes, E. M., Pires, R. A., Mano, J. F., & Reis, R. L. (2013). Bionanocomposites from
lignocellulosic resources: Properties, applications and future trends for their use in
the biomedical field. Progress in Polymer Science, 38(10), 1415–1441.
Foss Hansen, S., Heggelund, L. R., Revilla Besora, P., Mackevica, A., Boldrin, A., & Baun,
A. (2016). Nanoproducts—What is actually available to European consumers?
Environmental Science: Nano, 3(1), 169–180.
Gaston, E., Fraser, J. F., Xu, Z. P., & Ta, H. T. (2018). Nano- and micro-materials in the
treatment of internal bleeding and uncontrolled haemorrhage. Nanomedicine:
Nanotechnology, Biology and Medicine, 14(2), 507–519.
Hosseini, S. F., Rezaei, M., Zandi, M., & Ghavi, F. F. (2013). Preparation and functional
properties of fish gelatin-chitosan blend edible films. Food Chemistry, 136(3),
70
Carbohydrate Polymers 195 (2018) 63–70
1490–1495.
Ignjatović, N., Wu, V., Ajduković, Z., Mihajilov-Krstev, T., Uskoković, V., & Uskoković, D.
(2016). Chitosan-PLGA polymer blends as coatings for hydroxyapatite nanoparticles
and their effect on antimicrobial properties, osteoconductivity and regeneration of
osseous tissues. Materials Science and Engineering: C, 60, 357–364.
Kang, S.-W., Choi, J.-J., Kim, H.-N., Seo, J., Park, S.-J., Kim, E.-Y., et al. (2017). EGF-
loaded hyaluronic acid based microparticles as effective carriers in a wound model.
Particle & Particle Systems Characterization, 34(5).
Klemm, D., Heublein, B., Fink, H.-P., & Bohn, A. (2005). Cellulose: Fascinating biopo-
lymer and sustainable raw material. Angewandte Chemie International Edition, 44(22),
3358–3393.
Konieczynska, M. D., Villa-Camacho, J. C., Ghobril, C., Perez-Viloria, M., Blessing, W. A.,
Nazarian, A., et al. (2017). A hydrogel sealant for the treatment of severe hepatic and
aortic trauma with a dissolution feature for post-emergent care. Materials Horizons,
4(2), 222–227.
Lamke, L. O., Nilsson, G. E., & Reithner, H. L. (1977). The evaporative water loss from
burns and the water-vapour permeability of grafts and artificial membranes used in
the treatment of burns. Burns, 3(3), 159–165.
Lan, G., Lu, B., Wang, T., Wang, L., Chen, J., Yu, K., et al. (2015). Chitosan/gelatin
composite sponge is an absorbable surgical hemostatic agent. Colloids and Surfaces B:
Biointerfaces, 136(Suppl. C), 1026–1034.
Li, Y., Wang, S., Huang, R., Huang, Z., Hu, B., Zheng, W., et al. (2015). Evaluation of the
effect of the structure of bacterial cellulose on full thickness skin wound repair on a
microfluidic chip. Biomacromolecules, 16(3), 780–789.
Li, Y., Han, Y., Wang, X., Peng, J., Xu, Y., & Chang, J. (2017). Multifunctional hydrogels
prepared by dual ion cross-linking for chronic wound healing. ACS Appl Mater
Interfaces, 9(19), 16054–16062.
Lih, E., Jung, J. W., Joung, Y. K., Ahn, D. J., & Han, D. K. (2016). Synergistic effect of anti-
platelet and anti-inflammation of drug-coated Co-Cr substrates for prevention of in-
itial in-stent restenosis. Colloids and Surfaces B: Biointerfaces, 140(Suppl. C), 353–360.
Liu, Y., Sui, Y., Liu, C., Liu, C., Wu, M., Li, B., et al. (2018). A physically crosslinked
polydopamine/nanocellulose hydrogel as potential versatile vehicles for drug de-
livery and wound healing. Carbohydrate Polymers, 188, 27–36.
Mei, L., Fan, R., Li, X., Wang, Y., Han, B., Gu, Y., et al. (2017). Nanofibers for improving
the wound repair process: The combination of a grafted chitosan and an antioxidant
agent. Polymer Chemistry, 8(10), 1664–1671.
Mele, E. (2016). Electrospinning of natural polymers for advanced wound care: Towards
responsive and adaptive dressings. Journal of Materials Chemistry B, 4(28),
4801–4812.
Moosdeen, F., Williams, J. D., & Secker, A. (1988). Standardization of inoculum size for
disc susceptibility testing: A preliminary report of a spectrophotometric method.
Journal of Antimicrobial Chemotherapy, 21(4), 439–443.
Naseri, N., Deepa, B., Mathew, A. P., Oksman, K., & Girandon, L. (2016). Nanocellulose-
based interpenetrating polymer network (IPN) hydrogels for cartilage applications.
Biomacromolecules, 17(11), 3714–3723.
Nie, C., Cheng, C., Peng, Z., Ma, L., He, C., Xia, Y., et al. (2016). Mussel-inspired coatings
on Ag nanoparticle-conjugated carbon nanotubes: Bactericidal activity and mammal
cell toxicity. Journal of Materials Chemistry B, 4(16), 2749–2756.
Ong, S.-Y., Wu, J., Moochhala, S. M., Tan, M.-H., & Lu, J. (2008). Development of a
chitosan-based wound dressing with improved hemostatic and antimicrobial prop-
erties. Biomaterials, 29(32), 4323–4332.
Pollini, M., Russo, M., Licciulli, A., Sannino, A., & Maffezzoli, A. (2009). Characterization
of antibacterial silver coated yarns. Journal of Materials Science: Materials in Medicine,
20(11), 2361.
Radhakumary, C., Antonty, M., & Sreenivasan, K. (2011). Drug loaded thermoresponsive
and cytocompatible chitosan based hydrogel as a potential wound dressing.
Carbohydrate Polymers, 83(2), 705–713.
Saito, T., Kimura, S., Nishiyama, Y., & Isogai, A. (2007). Cellulose nanofibers prepared by
TEMPO-mediated oxidation of native cellulose. Biomacromolecules, 8(8), 2485–2491.
Sun, Y., & Xia, Y. (2002). Shape-controlled synthesis of gold and silver nanoparticles.
Science, 298(5601), 2176–2179.
Thoniyot, P., Tan, M. J., Karim, A. A., Young, D. J., & Loh, X. J. (2015). Nanoparticle-
hydrogel composites: Concept, design, and applications of these promising, multi-
functional materials. Advanced Science (Weinh), 2(1–2), 1400010.
Unger, F., Wittmar, M., & Kissel, T. (2007). Branched polyesters based on poly[vinyl-3-
(dialkylamino)alkylcarbamate-co-vinyl acetate-co-vinyl alcohol]-graft-poly(d,l-lac-
tide-co-glycolide): Effects of polymer structure on cytotoxicity. Biomaterials, 28(9),
1610–1619.
Wang, J., Gao, X., Sun, H., Su, B., & Gao, C. (2016). Monodispersed graphene quantum
dots encapsulated Ag nanoparticles for surface-enhanced Raman scattering. Materials
Letters, 162(Suppl. C), 142–145.
Wegst, U. G. K., Bai, H., Saiz, E., Tomsia, A. P., & Ritchie, R. O. (2014). Bioinspired
structural materials. Nature Materials, 14, 23.
Yang, J., Zhang, X., Ma, M., & Xu, F. (2015). Modulation of assembly and dynamics in
colloidal hydrogels via ionic bridge from cellulose nanofibrils and poly(ethylene
glycol). ACS Macro Letters, 4(8), 829–833.
Zhang, F., Braun, G. B., Shi, Y., Zhang, Y., Sun, X., Reich, N. O., et al. (2010). Fabrication
of Ag@SiO2@Y2O3:Er nanostructures for bioimaging: Tuning of the upconversion
fluorescence with silver nanoparticles. Journal of the American Chemical Society,
132(9), 2850–2851.
Zhao, X., Wu, H., Guo, B., Dong, R., Qiu, Y., & Ma, P. X. (2017). Antibacterial anti-oxidant
electroactive injectable hydrogel as self-healing wound dressing with hemostasis and
adhesiveness for cutaneous wound healing. Biomaterials, 122(Suppl. C), 34–47.