ORIGINAL PAPER

Thermoresponsive hybrid hydrogel of oxidizednanocellulose using a polypeptide crosslinker

Jie Cheng • Minsung Park • Jinho Hyun

Received: 27 December 2013 / Accepted: 18 February 2014

� Springer Science+Business Media Dordrecht 2014

Abstract Thermoresponsive hybrid nanocellulose

hydrogels were prepared from a mixture of oxidized

nanocellulose and elastin-like polypeptide (ELP).

Positively charged ELP was used as a polymeric

crosslinker for conjugation with negatively charged

nanocellulose. Hydrogel formation was triggered by a

simple increase in temperature, and the hydrogel was

reversibly returned to the liquid phase by decreasing

temperature. Surface potential measurement con-

firmed the electrostatic properties of oxidized nano-

cellulose and ELP molecules. The surface morphology

of hydrogels was observed by atomic force micros-

copy and field emission-scanning electron micros-

copy. Conformational changes in the ELP/

nanocellulose hybrid were characterized by circular

dichroism. The ELP/nanocellulose hybrid hydrogel

was noncytotoxic and suitable for encapsulating cells,

indicating its potential for biomedical applications.

Keywords Bacterial cellulose � Polypeptide �Thermo-responsiveness � Hydrogel

Introduction

As the most abundant renewable resource on earth,

cellulose has potential in a wide range of applications

(Klemm et al. 2011; Eichhorn et al. 2005). Natural

cellulose and its derivatives have been widely studied

for the fabrication of hydrogels (Chang and Zhang

2011). Due to the strong interfibrillar forces, micro-

fibrillar cellulose obtained by mechanical disintegra-

tion of wood fibers forms a strong gel in water

suspension (Paakko et al. 2007). Nanocellulose-based

hydrogel can be fabricated from disintegrated cellu-

lose suspensions (Zhao et al. 2007; Ostlund et al. 2009;

Dong et al. 2013), cellulose derivatives (Marci et al.

2006; Fei et al. 2000), and cellulose-polymer blending

(Zhao et al. 2009; Chang et al. 2008).

Recently, stimuli responsive cellulose-based

hydrogels have been fabricated through chemical

modifications or incorporating of polymers (Amin

et al. 2012; Halake and Lee 2014). Chemical modi-

fications of cellulose can be exclusively performed on

the hydroxyl groups through esterification, etherifica-

tion and oxidation reactions. Interestingly, carboxy-

methyl cellulose can be pH-responsive (Tan et al.

2010; Ekici 2011), with pKa around 3–4. Hydroxy-

propylmethyl cellulose is a temperature-responsive

cellulose derivative exhibiting a sol–gel transition in

aqueous solution at 41 �C (Zhang et al. 2011; Bai et al.

2012; Li et al. 2001; Silva et al. 2008). However, the

gelation temperature is too high to encapsulate cells.

UV light or electron-beam irradiation was reported for

Electronic supplementary material The online version ofthis article (doi:10.1007/s10570-014-0208-4) contains supple-mentary material, which is available to authorized users.

J. Cheng � M. Park � J. Hyun (&)

Department of Biosystems and Biomaterials Science and

Engineering, Seoul National University, Seoul 151-921,

Korea

e-mail: jhyun@snu.ac.kr

123

Cellulose

DOI 10.1007/s10570-014-0208-4

the fabrication of responsive cellulose hydrogels, but

in situ encapsulation would be harmful to cells (Amin

et al. 2012; Halake and Lee 2014).

Bacterial cellulose (BC) synthesized by Acetobacter

xylimum possesses a series of novel properties, includ-

ing high porosity, high crystallinity, water absorbance,

excellent mechanical properties and biocompatibility.

These characteristics make BC a very desirable mate-

rial for bio-related applications. BC is a hydrogel

consisting of type I cellulose nanofibers with a layer-by-

layer structure. Recently, Isogai and coworkers devel-

oped a 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO)-

oxidation method to disintegrate cellulose into

TEMPO-oxidized cellulose nanofibers without losing

any of the original crystallinity of natural cellulose

(Fukuzumi et al. 2009; Isogai et al. 2011; Saito et al.

2007). Due to electrostatic repulsion and osmotic

effects, the TEMPO-oxidized cellulose nanofibers were

completely and individually dispersed in water (Isogai

et al. 2011).

As a polymeric crosslinker of nanocellulose-based

hydrogels, positively charged elastin-like polypeptide

(ELP) was adopted. ELP, an artificial smart polypep-

tide, has successfully been prepared by a number of

different strategies. ELPs are biopolymers of the

pentapeptide repeat Val-Pro-Gly-Xaa-Gly derived

from human tropoelastin, where the ‘‘guest residue’’

Xaa can be any of the natural amino acids except for

Pro. Their biocompatibility, biodegradability, non-

immunogenicity, stimuli-responsive characteristics

and thermo-responsive characteristics make ELPs a

very promising candidate for drug delivery and other

biomedical applications (Nettles et al. 2010; Bidwell

and Raucher 2010; Simnick et al. 2007; Cheng et al.

2013). ELPs are soluble in aqueous solution below the

inverse transition temperature (Tt); once the temper-

ature increases above Tt, ELPs undergo a sharp phase

transition accompanied by aggregation.

Herein, a thermoresponsive polysaccharide hydro-

gel based on nanofibrous cellulose and ELP was

prepared. When TEMPO-oxidized bacterial cellulose

(TOBC) and ELP were mixed together at temperatures

lower than the Tt of ELPs, positively charged ELPs

spontaneously adsorbed onto the surface of negatively

charged TOBC due to electrostatic interactions. ELPs

folded and coacervated when the temperature

increased above Tt, causing TOBCs to rope together

and form a porous hybrid hydrogel of ELP and TOBC.

The thermoresponsive polysaccharide hydrogel of

TOBC and ELP was unique because of its noncyto-

toxicity and reversible sol–gel transition. Compared

with hydrogels bonded covalently using smaller

crosslinkers, the polysaccharide hydrogel of TOBC

and ELP was resuspended by simply decreasing the

temperature. This is a promising material for applica-

tions in biotechnology and life science fields that

require properties such as cell encapsulation, tissue

engineering and controlled drug release.

Materials and experimental methods

Preparation of TOBC

BC was prepared by fermenting a single Glu-

conacetobacter xylinus (KCCM 40216) colony in

mannitol culture medium. BC was harvested and

purified by soaking in 1 wt% sodium hydroxide with

gentle agitation for 24 h at room temperature. Subse-

quently, BC was thoroughly washed with running

distilled water and then freeze-dried.

TOBC was prepared following the method reported

by the Isogai group (Isogai et al. 2011). BC was

oxidized by TEMPO. To obtain TEMPO-oxidized BC,

the never-dried BC pellicle (0.45 g of cellulose

content) was cut into small pieces and then suspended

in 500 ml distilled water containing 20 mg TEMPO

and 0.5 g NaBr. Then, 15 ml NaClO solution was

added to the BC suspension to start the oxidation. The

solution was maintained at pH 10, and the mixture was

vigorously agitated with a magnetic stirring bar for

2 days. At the end of the reaction, oxidation was

quenched by adding ethanol to the suspension. The

products were washed three times via centrifugation

and then freeze-dried for further use.

Synthesis of ELP

The ELP gene [(VPGVG)14(VPGKG)]8[VPGVG]40

was oligomerized by recursive directional ligation and

transformed into E. coli strain BLR (DE3) (Novagen,

USA) for expression. The ELP was purified by inverse

transition cycling and the collected ELP was freeze-

dried (Meyer and Chilkoti 2002). Detailed information

regarding ELP synthesis is described in supplementary

information.

Cellulose

123

Sol–gel transition

TOBC/ELP solution (2 wt%, TOBC:ELP = 1:1) was

prepared at 4 �C and then transferred to an incubator at

37 �C to trigger sol–gel phase transition. For compar-

ison, pure TOBC and ELP solutions with a concen-

tration of 1 wt% were also prepared at 4 �C and

transferred to an incubator at 37 �C. After these

solutions were heated to 37 �C and equilibrated for

30 min, the sol–gel transition was determined by vial

inversion, which is a visual inspection of the sol–gel

phase transition.

Characterization of TOBC/ELP

To determine the Tt of ELP and TOBC/ELP solution,

OD350 was monitored by a UV–visible spectropho-

tometer equipped with a water circulating temperature

controller (Mecasys, South Korea). The samples were

heated at a rate of 1 �C/min.

To further confirm hydrogel formation, dynamic

rheological analysis was performed to determine the

change in sample viscoelasticity as a function of

temperature. Rheological measurements of ELP

(2 wt%), TOBC (2 wt%) and TOBC/ELP (2 wt% for

a 1:1 TOBC–ELP mixture) were performed using the

Advanced Rheometric Expansion System (Rheomet-

ric Scientific, UK). Measurements were carried out by

increasing temperature in the range of 25–40 �C at an

angular frequency of 1.0 rad/s and a shear stress of

2.0 Pa.

Secondary ELP structure was determined by circu-

lar dichroism (CD) (Applied photophysics, UK) at

different temperatures. ELP (0.5 mg/ml) and TOBC/

ELP (1 mg/ml for a 1:1 TOBC-ELP mixture) solutions

were prepared for CD analysis.

Fluorescence emission/excitation spectra were

recorded by FP-8300 spectrofluorometer (Jasco,

Japan). To determine the maximum excitation spec-

trum, ELP solution with a concentration of 0.25 mg/ml

was scanned in the range of 200–500 nm, and emission

spectra were recorded in the range of 220–750 nm.

Sample emissions were recorded in the range of

280–450 nm with the excitation wavelength set to

270 nm at given temperatures.

Zeta-potentials of ELP, TOBC and TOBC/ELP

complexes were measured by dynamic light scattering

(Zetasizer Nano, Malvern, UK) to determine the

electrophoretic mobility of samples. All samples were

suspended in deionized water, and measurements were

carried out at 20 �C with folded capillary cells.

Sample surface morphologies were imaged by

atomic force microscopy (AFM, Park Science) with a

non-contact mode (tip model, spring constant 42 N/m)

in air. The pure TOBC and TOBC/ELP solutions were

casted on a mica surface and dried at room temperature

prior to AFM imaging.

The inner morphologies of the TOBC solution, the

TOBC/ELP solution and the TOBC/ELP hydrogel

were observed by imaging the cross-section of the

samples by scanning electron microscope (SEM).

SEM images were acquired with a SUPRA 55VP field-

emission SEM (Carl Zeiss, Germany). After samples

were equilibrated at given temperatures, they were

dipped into liquid nitrogen and freeze-dried. Freeze-

dried samples were coated by gold sputter prior to

SEM observation.

Cell viability and proliferation

NIH-3T3 fibroblast cells were used for MTT assay.

The cells were maintained in Dulbecco’s minimal

essential medium (DMEM) supplemented with 10 %

fetal bovine serum, penicillin (100 U/ml), streptomy-

cin (100 lg/ml) and HEPES (7.5 mM). Briefly, 200 ll

of NIH-3T3 fibroblast cells were seeded on a 96-well

plate at a density of 4 9 104 cells/well. They were

incubated (37 �C, 5 % CO2) overnight to allow the

cells to attach to the surface of the wells. The medium

was sequentially removed and the wells were washed

with PBS. Then, 100 ll of fresh medium was added to

each well followed by the addition of 100 ll of

sterilized TOBC/ELP aqueous complex at a concen-

tration of 0.02–5 mg/ml. This resulted in final con-

centrations of 0.01, 0.1, 0.5, 1.0, and 2.5 mg/ml. As a

control, 100 ll of sterilized DW was added to the well

and mixed with 100 ll of fresh medium. Cells were

incubated for another 48 h in 5 % CO2 at 37 �C. After

incubation, 20 ll of MTT solution (5 mg/ml in PBS)

was added to each well and incubated for another 4 h.

Next, the medium containing the MTT solution was

aspirated. The crystallized product was dissolved in

100 ll of DMSO, and absorbance was measured at

570 nm using a plate reader. Cell viability was

calculated by dividing the optical density (OD) value

of the experimental sample with the OD value of the

control sample.

Cellulose

123

Cell encapsulation with TOBC/ELP

The TOBC/ELP aqueous complex for cell encapsula-

tion was sterilized by autoclave. NIH-3T3 fibroblast

cells were seeded into the TOBC/ELP aqueous

complex at 20 �C with a final concentration of

2 9 106 cells/ml. The cell-containing TOBC/ELP

complex was then transferred to a 37 �C humidified

environmental incubator and incubated for 30 min to

trigger the sol–gel transition. After cell encapsulation,

the TOBC/ELP hydrogel was rinsed with pre-warmed

DMEM medium several times to replace the remain-

ing water inside the hydrogel. Encapsulated cells were

cultured in a 5 % CO2 humidified incubator at 37 �C

for 2, 5 or 7 days until examination.

A live/dead viability fluorescence assay was per-

formed by staining cells with a live/dead viability/

cytotoxicity kit for mammalian cells (Invitrogen,

USA). The combined live/dead assay reagents were

added to cover hydrogel containing cells and then

incubated for 30 min at room temperature in the dark.

Cell viability was observed by fluorescence micros-

copy (BX51, Olympus, Japan).

Fibroblast cells seeded and spread in a hydrogel

were fixed with 2 % aqueous glutardaldehyde solution

for 30 min followed by washing with DI water and

freeze-drying. Cell morphology in hydrogel was

observed by FE-SEM (Carl Zeiss, Germany).

Results and discussion

Gel formation of TOBE/ELP

Pure ELP, pure TOBC and TOBC/ELP mixture

solutions in vials all showed fluid-like states when

the vials were inverted at 20 �C. The turbidity of pure

ELP solution abruptly increased at 37 �C due to the

hydrophobic aggregation of ELP molecules, but this

mixture was still a fluid-like ELP suspension. Con-

versely, the temperature increase enabled hydrogel

formation from the TOBC/ELP mixture, as shown in

Fig. 1.

To determine the critical temperature that induces

hydrogel formation, changes in the turbidity of TOBC/

ELP mixtures were measured with thermo-regulated

UV/Vis spectrometry at 350 nm as a function of

temperature. The phase transition of TOBC/ELP

complexes started at *26 �C, a much lower temper-

ature (Fig. 2b) than the pure ELP solution (Fig. 2a).

a

Heating(37°C)

Cooling(20°C)

b c

Fig. 1 Temperature-triggered sol–gel transition of the TOBC/

ELP complex. a TOBC only, b ELP only and c TOBC/ELP

5

4

3

2

1

0

OD

(35

0 nm

)

50454035302520

Temperature (°C)

TOBC/ELP

ELP

Fig. 2 Turbidity variations of ELP and TOBC/ELP solutions as

a function of temperature (CD: 350 nm, heating: 1 �C/min)

Cellulose

123

ELPs were adsorbed onto the surface of TOBCs due to

electrostatic interactions, leading to a very high local

concentration of ELPs on the surface of TOBC.

Rheological analysis of TOBE/ELP hydrogel

Dynamic rheology is one of the most direct and

reliable way to determine the sol–gel transition of

polysaccharide hydrogels (Zhang et al. 2008; Brun-

Graeppi et al. 2010). The rheological properties of the

cellulose suspensions (Karppinen et al. 2011) and the

aggregation of the cellulose fibers (Myllytie et al.

2009) have been studied by mixing different polymers

with cellulose suspension.

Hydrogel formation was confirmed by dynamic

rheological analysis, which determined the change in

sample viscoelasticity as a function of temperature.

Measurements were carried out at angular frequency

of 1.0 rad/s and shear stress of 2.0 Pa. For both TOBC/

ELP complex and pure TOBC solutions, the elastic

modulus (G0) was always higher than the viscous

modulus (G00) throughout the whole temperature

range, as shown in Fig. 3a, b, indicating that elasticity

was dominant at both low and high temperatures

(Brun-Graeppi et al. 2010). In other words, either the

TOBC/ELP complex or the pure TOBC solution was

not a free-flowing fluid but a viscous semi-solid or

semi-transparent swollen gel at temperatures lower

than the Tt of ELP (Karppinen et al. 2011; Paakko

et al. 2007). This is possibly due to the high

concentration and relatively large size of TOBCs.

For the TOBC/ELP complex, G0 increased dramati-

cally as temperature increased. This is due to the

transition of semitransparent swollen gel to nontrans-

parent shrunken/solid gel. The decline of both G0 and

G00 at temperatures above the Tt of ELP was due to

phase separation, leading to the formation of a

biopolymer-rich phase and water-rich phase. The

sharp enhancement of G0 at around 30 �C was caused

by the folding and coacervation of ELPs adsorbed onto

the surface of TOBC, which acted as a physical

crosslinker between TOBC chains. Meanwhile, G0 and

G00 of TOBC/ELP complex were significantly

decreased after the transition compared with pure

ELP solution (Fig. 3a, c). It resulted from relatively

stiff cellulose nanofibers to hinder aggregation of ELP

molecules adsorbed onto the surface (Karppinen et al.

2011). The sharp enhancements of both G0 and G00

were not observed for pure TOBC solution as

temperature increased (Fig. 3b). However, a sharp

enhancement in G0 was observed for pure ELP

solution as temperature increased. This indicates that

colloidal ELP particles precipitated when the

24 26 28 30 32 34 36 38 40 42

0

5000

10000

15000

20000

25000

G',G

''(d

yn/c

m2 )

G' G''

ELP (20mg/ml)

24 26 28 30 32 34 36 38 40 420

50

100

150

200

250

300

G',G

''(d

yn/c

m2 )

Temperature (°C)

G' G''

TOBC+ELP=1;1 (20mg/ml)

0

50

100

150

200

250

300

G',G

'' (d

yn/c

m2 )

G' G''

TOBC (20mg/ml)

24 26 28 30 32 34 36 38 40 42

Temperature (°C)

Temperature (°C)

a

b

c

Fig. 3 Oscillatory temperature ramp measurements to confirm

hydrogel formation. a TOBC/ELP complex, b pure TOBC

solution and c pure ELP solution. Heating rate: 2 �C/min.

Sample concentration: 2 wt%

Cellulose

123

temperature increased, leading to a gel-like film on the

detecting plate surface.

Morphological analysis of TOBC/ELP hydrogel

After the temperature change of TOBC solution,

TOBC/ELP solution and TOBC/ELP hydrogel, the

samples were immediately dipped into liquid nitrogen

and then freeze-dried to keep the morphological

differences at the temperatures. As shown in FE-

SEM images, TOBC fibers were closely packed in

pure TOBC (Fig. 4a, a1). The density of TOBC fibers

stored in TOBC/ELP solution at 20 �C was lower than

the density in pure TOBC solution (Fig. 4b, b1).

ca b

c1a1 b1

c2a2 b2

TOBC/ELPTOBCd

200nm

200nm

Fig. 4 FE-SEM images of freeze-dried a TOBC, b TOBC/ELP solution (swollen gel) and c TOBC/ELP shrunken hydrogel. d AFM

images of TOBC and TOBC/ELP complex

Cellulose

123

TOBC, ELP and water were evenly mixed together to

form a single aqueous phase system. The phase

separation phenomenon occurred as the hydrogel

was formed at 37 �C, leading to two aqueous phases

inside the hydrogel, the biopolymer (ELP and TOBC)-

rich phase and water-rich phase. The water-rich phase

formed pores after freeze-drying due to the removal of

water from the region (Fig. 4c, c1).

ELPs were adsorbed onto the surface of TOBCs by

electrostatic interactions (Fig. 4b2), just as with pre-

viously reported protein/polysaccharide systems (de

Kruif et al. 2004; Schmitt et al. 1998; Turgeon et al.

2007). When the temperature increased above Tt, ELP

transitioned from hydrophilic to hydrophobic states by

exposing its hydrophobic side groups, driving ELPs to

fold and coacervate by hydrophobic interactions.

Therefore, TOBCs were held together when temper-

atures increased above Tt due to the coacervation of

ELPs adsorbed onto their surface. In addition, strong

hydrophobic interactions between ELPs led to the

formation of TOBC/ELP hybrid hydrogel by cross-

linking TOBCs tightly (Fig. 4c2). Due to high water

content in the water-rich phase, pores were left by

water sublimation after lyophilization, as shown in

Fig. 4c, c1.

AFM height images also verified the morphological

differences between pure TOBC and TOBC/ELP

complex (Fig. 4d). The root mean square roughness

(Rq) of a TOBC/ELP layer was *11 nm. It showed a

smoother surface compared with a TOBC layer

(Rq * 13 nm). TOBC nanofibers became thicker

due to ELP molecules adsorbed onto the surface.

The obvious difference in morphology was another

strong evidence that positively-charged ELPs were

spontaneously adsorbed onto the surface of nega-

tively-charged TOBC nanofibers through the electro-

static interaction.

Mechanism of TOBC/ELP hydrogel formation

Zeta potential is widely used for quantification of the

magnitude of the electrical charge at the double layer.

Although zeta potential is not equal to the electric

surface potential in the double layer, zeta potential is

often the only available path for characterization of

double-layer properties. From the experimentally-

determined electrophoretic mobility, pure ELP

showed a positive surface potential (or surface charge)

in distilled water at pH 6 due to its lysine residues

(Fig. 5). Conversely, TOBC had a very strongly

negative surface potential at pH 6 as a result of

TEMPO-oxidation. When TOBC and ELP were

mixed together, this strongly negative surface poten-

tial became weakened dramatically.

When the positively charged ELPs were deproto-

nated at high pH, the TOBC/ELP mixture lost its

ability to form a hydrogel. The pKa of the a-amino

group at the ELP N-terminus and the e-amino group in

the ELP lysine are 8.9 and 10.5, respectively. The

medium was adjusted to pH 12 to eliminate positive

surface charges of ELP derived from primary amines.

As shown in Fig. 6, a free-flowing milky suspension

was formed instead of a shrunken hydrogel when the

temperature increased to 37 �C. This was similar to

the behavior of pure ELP solution, shown in Fig. 1.

ELPs were adsorbed onto the TOBC surface through

electrostatic interactions at pH 6 to form a polypep-

tide–polysaccharide complex (Turgeon et al. 2007).

-140

-120

-100

-80

-60

-40

-20

0

20

po

ten

tial

(m

V)

TOBC

TOBC/ELP

ELP

Fig. 5 n-Potential of ELP, TOBC and TOBC/ELP at 20 �C

Heating(37°C)

Cooling(20°C)

Fig. 6 Effect of pH on the formation of TOBC/ELP hydrogels.

The TOBC/ELP solution was adjusted to pH 12 to eliminate

positive charges on the ELP molecules. This resulted in the

inability to form a hydrogel despite increased temperature

Cellulose

123

Although phase separation is a very common

phenomenon in biopolymer mixtures, many research-

ers have demonstrated that this is not the case for

mixtures consisting of protein/polypeptide and

charged polysaccharide (Antonov et al. 1999;

Tolstoguzov 1991; Antonov and Wolf 2005). To

confirm whether ELP chains were adsorbed onto the

surface of TOBC backbones, the secondary ELP

structure was determined through CD and fluores-

cence measurements in the presence and absence of

TOBC. Pure ELP had a disordered secondary structure

with a negative peak around 198 nm at 10 �C (black

curve in Fig. 7). However, the intensity of this

negative peak was enhanced when pure ELP was

mixed with TOBC, indicating that the random-coil-

based disordered structure of ELP increased dramat-

ically (blue curve in Fig. 7). Altered secondary

structure in the presence of TOBC was consistent

with previous reports by other groups that reported

that the extent of protein secondary structures

increased with interactions between protein and poly-

saccharide (Turgeon et al. 2007). ELP molecules

transitioned into a more ordered state with a positive

peak around 212 nm when the temperature increased

to 50 �C (red curve in Fig. 7), indicating a beta-turn

structure (Cho et al. 2009; Nuhn and Klok 2008; Urry

1993). The more interesting result was that ELPs were

more disordered in the presence of TOBC despite the

Fig. 7 CD spectra of TOBC/ELP and ELP obtained at different

temperatures in water. ELP concentration was 0.5 mg/ml in

both the TOBC/ELP and pure ELP solutions. (Color figure

online)

0

20

40

60

80

100

120

Cel

l via

bili

ty (

%)

100

c d

a

5

Concentration of TOBC/ELP (mg/ml)

0.01 0.1 0.5 1.0 2.5

b

100

Fig. 8 a MTT results of the

TOBC/ELP solution. b FE-

SEM micrographs of cells

encapsulated in a TOBC/

ELP hydrogel. Fluorescence

microscopic images of

fibroblast cells encapsulated

in a TOBC/ELP hydrogel

after c 1 day and d 7 days of

incubation (cells were live/

dead stained.)

Cellulose

123

high temperature. This result was in the opposite

direction of pure ELP (green curve in Fig. 7). This

extraordinary phenomenon was attributed to the strong

electrostatic interaction between ELP and TOBC.

Biocompatibility of TOBC/ELP complex

MTT assay was used to assess the biocompatibility of

TOBC/ELP complexes and the proliferation of NIH-

3T3 fibroblast cells (Fig. 8a). Cell viability (%) was

calculated as the ratio of experimental sample OD

value to the control OD value. High fibroblast

viability was observed regardless of TOBC/ELP

concentration, as shown in Fig. 8a, indicating that

the TOBC/ELP complex was not cytotoxic. The

morphology of cells inside the hydrogel was observed

by FE-SEM. Cells bound to the hydrogel properly and

spread on the surface, as shown in Fig. 8b. The

viability of encapsulated cells was investigated by

live/dead staining. As shown in Fig. 8c, d, the number

of green spots inside the hydrogel increased after

7 days of incubation, indicating that most encapsu-

lated cells remained alive and continued to proliferate

inside the hydrogel.

Conclusions

A thermoresponsive hybrid hydrogel was prepared by

mixing negatively charged TOBC nanofibers and

positively charged ELP molecules containing lysine

residues. Positively charged ELPs were bound to the

surface of negatively charged TOBC by electrostatic

interactions to form a single phase. Increased temper-

ature triggered ELP folding and aggregation, leading

to the formation of a TOBC/ELP hybrid hydrogel.

Conformational analyses based on CD supported the

occurrence of electrostatic interactions between

TOBC and ELP. A cytotoxicity test showed that the

TOBC/ELP hydrogel was suitable for encapsulating

cells, indicating its potential for biomedical

applications.

Acknowledgments This research was supported by the Basic

Science Research Program through the National Research

Foundation of Korea (NRF), funded by the Ministry of

Science, ICT & Future Planning (Grant Number 2013023612).

We also acknowledge support from the Research Institute for

Agriculture and Life Sciences.

References

Amin MCIM, Ahmad N, Halib N, Ahmad I (2012) Synthesis

and characterization of thermo- and pH-responsive bacte-

rial cellulose/acrylic acid hydrogels for drug delivery.

Carbohydr Polym 88(2):465–473

Antonov YA, Wolf BA (2005) Calorimetric and structural

investigation of the interaction between bovine serum

albumin and high molecular weight dextran in water.

Biomacromolecules 6(6):2980–2989

Antonov YA, Lefebvre J, Doublier JL (1999) On the one-phase

state of aqueous protein uncharged polymer systems:

casein–guar gum system. J Appl Polym Sci 71(3):471–

482

Bai YY, Zhang Z, Zhang AP, Chen L, He CL, Zhuang XL, Chen

XS (2012) Novel thermo- and pH-responsive hydroxy-

propyl cellulose- and poly (L-glutamic acid)-based mi-

crogels for oral insulin controlled release. Carbohydr

Polym 89(4):1207–1214

Bidwell GL, Raucher D (2010) Cell penetrating elastin-like

polypeptides for therapeutic peptide delivery. Adv Drug

Deliv Rev 62(15):1486–1496

Brun-Graeppi AKAS, Richard C, Bessodes M, Scherman D,

Narita T, Ducouret G, Merten OW (2010) Study on the sol–

gel transition of xyloglucan hydrogels. Carbohydr Polym

80(2):555–562

Chang CY, Zhang LN (2011) Cellulose-based hydrogels: pres-

ent status and application prospects. Carbohydr Polym

84(1):40–53

Chang CY, Lue A, Zhang L (2008) Effects of crosslinking

methods on structure and properties of cellulose/PVA

hydrogels. Macromol Chem Phys 209(12):1266–1273

Cheng J, Na K, Kim HS, Lee CK, Hyun J (2013) ‘‘Smart’’

microspheres for self-renewal of embryonic stem cells.

Macromol Res 21(2):134–136

Cho Y, Sagle LB, Iimura S, Zhang YJ, Kherb J, Chilkoti A,

Scholtz JM, Cremer PS (2009) Hydrogen bonding of beta-

turn structure is stabilized in D2O. J Am Chem Soc

131(42):15188–15193

de Kruif CG, Weinbreck F, de Vries R (2004) Complex coac-

ervation of proteins and anionic polysaccharides. Curr

Opin Colloid Interface Sci 9(5):340–349

Dong H, Snyder JF, Williams KS, Andzelm JW (2013) Cation-

induced hydrogels of cellulose nanofibrils with tunable

moduli. Biomacromolecules 14(9):3338–3345

Eichhorn SJ, Young RJ, Davies GR (2005) Modeling crystal and

molecular deformation in regenerated cellulose fibers.

Biomacromolecules 6(1):507–513

Ekici S (2011) Intelligent poly (N-isopropylacrylamide)-car-

boxymethyl cellulose full interpenetrating polymeric net-

works for protein adsorption studies. J Mater Sci

46(9):2843–2850

Fei B, Wach RA, Mitomo H, Yoshii F, Kume T (2000) Hydrogel

of biodegradable cellulose derivatives. I. Radiation-induced

crosslinking of CMC. J Appl Polym Sci 78(2):278–283

Fukuzumi H, Saito T, Wata T, Kumamoto Y, Isogai A (2009)

Transparent and high gas barrier films of cellulose nanof-

ibers prepared by TEMPO-mediated oxidation. Biomac-

romolecules 10(1):162–165

Cellulose

123

Halake KS, Lee J (2014) Superporous thermo-responsive

hydrogels by combination of cellulose fibers and aligned

micropores. Carbohydr Polym (in press)

Isogai A, Saito T, Fukuzumi H (2011) TEMPO-oxidized cel-

lulose nanofibers. Nanoscale 3(1):71–85

Karppinen A, Vesterinen AH, Saarinen T, Pietikainen P, Sep-

pala J (2011) Effect of cationic polymethacrylates on the

rheology and flocculation of microfibrillated cellulose.

Cellulose 18(6):1381–1390

Klemm D, Kramer F, Moritz S, Lindstrom T, Ankerfors M, Gray

D, Dorris A (2011) Nanocelluloses: a new family of nature-

based materials. Angew Chem Int Ed 50(24):5438–5466

Li L, Thangamathesvaran PM, Yue CY, Tam KC, Hu X, Lam

YC (2001) Gel network structure of methylcellulose in

water. Langmuir 17(26):8062–8068

Marci G, Mele G, Palmisano L, Pulito P, Sannino A (2006)

Environmentally sustainable production of cellulose-based

superabsorbent hydrogels. Green Chem 8(5):439–444

Meyer DE, Chilkoti A (2002) Genetically encoded synthesis of

protein-based polymers with precisely specified molecular

weight and sequence by recursive directional ligation:

examples from the elastin-like polypeptide system. Bio-

macromolecules 3(2):357–367

Myllytie P, Holappa S, Paltakari J, Laine J (2009) Effect of

polymers on aggregation of cellulose fibrils and its impli-

cation on strength development in wet paper web. Nord

Pulp Pap Res J 24(2):125–134

Nettles DL, Chilkoti A, Setton LA (2010) Applications of

elastin-like polypeptides in tissue engineering. Adv Drug

Deliv Rev 62(15):1479–1485

Nuhn H, Klok HA (2008) Secondary structure formation and

LCST behavior of short elastin-like peptides. Biomacro-

molecules 9(10):2755–2763

Ostlund A, Lundberg D, Nordstierna L, Holmberg K, Nyden M

(2009) Dissolution and gelation of cellulose in TBAF/

DMSO solutions: the roles of fluoride ions and water.

Biomacromolecules 10(9):2401–2407

Paakko M, Ankerfors M, Kosonen H, Nykanen A, Ahola S,

Osterberg M, Ruokolainen J, Laine J, Larsson PT, Ikkala

O, Lindstrom T (2007) Enzymatic hydrolysis combined

with mechanical shearing and high-pressure homogeniza-

tion for nanoscale cellulose fibrils and strong gels. Bio-

macromolecules 8(6):1934–1941

Saito T, Kimura S, Nishiyama Y, Isogai A (2007) Cellulose

nanofibers prepared by TEMPO-mediated oxidation of

native cellulose. Biomacromolecules 8(8):2485–2491

Schmitt C, Sanchez C, Desobry-Banon S, Hardy J (1998)

Structure and technofunctional properties of protein–

polysaccharide complexes: a review. Crit Rev Food Sci

Nutr 38(8):689–753

Silva SMC, Pinto FV, Antunes FE, Miguel MG, Sousa JJS, Pais

A (2008) Aggregation and gelation in hydroxypropylm-

ethyl cellulose aqueous solutions. J Colloid Interface Sci

327(2):333–340

Simnick AJ, Lim DW, Chow D, Chilkoti A (2007) Biomedical

and biotechnological applications of elastin-like polypep-

tides. Polym Rev 47(1):121–154

Tan JJ, Liu RG, Wang W, Liu WY, Tian Y, Wu M, Huang Y

(2010) Controllable aggregation and reversible pH sensi-

tivity of AuNPs regulated by carboxymethyl cellulose.

Langmuir 26(3):2093–2098

Tolstoguzov VB (1991) Functional properties of food proteins

and role of protein–polysaccharide interaction. Food Hy-

drocoll 4(6):429–468

Turgeon SL, Schmitt C, Sanchez C (2007) Protein–polysac-

charide complexes and coacervates. Curr Opin Colloid

Interface Sci 12(4–5):166–178

Urry DW (1993) Molecular machines: how motion and other

functions of living organisms can result from reversible

chemical-changes. Angew Chem Int Ed Engl

32(6):819–841

Zhang YY, Xu XJ, Zhang L (2008) Dynamic viscoelastic

behavior of triple helical lentinan in water: effect of tem-

perature. Carbohydr Polym 73(1):26–34

Zhang Z, Chen L, Zhao CW, Bai YY, Deng MX, Shan HL,

Zhuang XL, Chen XS, Jing XB (2011) Thermo- and pH-

responsive HPC-g-AA/AA hydrogels for controlled drug

delivery applications. Polymer 52(3):676–682

Zhao HB, Kwak JH, Wang Y, Franz JA, White JM, Holladay JE

(2007) Interactions between cellulose and N-methylmor-

pholine-N-oxide. Carbohydr Polym 67(1):97–103

Zhao Q, Qian JW, An QF, Gao CJ, Gui ZL, Jin HT (2009)

Synthesis and characterization of soluble chitosan/sodium

carboxymethyl cellulose polyelectrolyte complexes and

the pervaporation dehydration of their homogeneous

membranes. J Membr Sci 333(1–2):68–78

Cellulose

123