Effects of Chitin Nanocrystal on Solution Behavior and Gelation of Chitosan/β-Glycerophosphate Thermosensitive Hydrogel

ULLAH Wajeeh ( )， GONG Feiei ()， JI Yali ()

State Key Laboratory for Modification of Chemical Fibers and Polymer Materials， College of Materials Science and Engineering， Donghua University， Shanghai 201620， China

Abstract: The chitosan/β-glycerophosphate (CS/β-GP)， a physical hydrogel system with thermosensitive and injectable features combined with biocompatibility and biodegradability， has great potentials as matrices for drug or cell encapsulation and delivery， or as in situ gel-forming materials for tissue repair. Here， the chitin nanocrystal (ChiNC) was introduced into the aforementioned system， and its effects on solution behavior and mechanical properties was investigated. The results showed the incorporation of ChiNC complicated sol-to-gel transition process; a higher loading ratio (20%) speeded up sol-to-gel transition rate， reduced the sol-to-gel transition temperature， while still maintained shear-thinning behavior or injectable feature.Moreover， the mechanical properties of gels were significantly enhanced by ChiNC， accompanied by decreased water uptake. The above mentioned behavior favored better applications as injectable tissue-repair implants.

Key words: chitosan; β-glycerophosphate; chitin nanocrystal; hydrogel; thermosensitivity; injectable

Introduction

Chitin nanocrystal (ChiNC， also called nanowhisker， or nanofibril)， is produced by acid hydrolysis of chitin via removing disordered or low-ordered regions and leaving highly crystalline residues[1-3]. It is an elongated rod-like particle with the size of 150-2 200 nm in length and 10-50 nm in width dependent on the chitin origins and hydrolytic conditions. Owing to its high modulus，i.e., about 150 GPa in longitudinal modulus and about 15 GPa in transverse modulus， the ChiNC as an emerging and novel nanofiller has been proved to exhibit great reinforcing effects on both synthetic and natural polymeric structures[4-5]. Furthermore， the ChiNC inherits chitin’s excellent biocompatibility and biodegradability and unique antibacterial activity， all of which make it one of the most promising nanofillers in biomedical material field[6-7]. The ChiNC was thought to only well disperse in aqueous solution， so many water-soluble or water-dispersible polymers，i.e. poly(vinyl alcohol)[8]， poly(ethylene glycol)[9]， alginate[10]， chitosan[11]， starch[12]， or waterborne polyurethane[13]were chosen as matrix polymer to prepare ChiNC reinforced nanocomposites in consideration of convenient processing via aqueous blending. Up to now， the ChiNC nanofiller has successfully applied in the fields of electrospun fiber mats[14]， wet spun fibers[15]， films[16]， cell 3D-scaffolds[17]， hydrogels[18]， aerogels[19]，etc. Almost all the research revealed that ChiNC nanofillers exerted positive influences on the mechanical and biological properties of the matrix materials.

Thermosensitive hydrogels have attracted much attention in the field of biomedicine recently[20-21]. The chitosan/β-glycerophosphate (CS/β-GP)， a physical thermosensitive hydrogel system， was prepared by neutralizing chitosan solutions with weak base， β-glycerophosphate (β-GP). This system remained liquid at or below room temperature and transformed into gel at body temperature， while keeping pH values within a physiologically acceptable range of 6.8-7.2[22-23]. The temperature-sensitive and injectable features combined with biocompatibility， biodegradability， and adhesive to human tissues， endow CS/β-GP with great potentials as matrices for drug or cell encapsulation and delivery， or as in situ gel-forming materials for tissue repair[24-25]. The molecular mechanism of gelation was thought to involve multiple interactions， including: (1) electrostatic repulsion between charged CS chains; (2) electrostatic attraction between oppositely charged CS and phosphate moiety of β-GP; (3) the hydrophobic or water-structuring character of the glycerol moiety of β-GP; (4) attractive hydrophobic and hydrogen bonding between CS chains[26]. If attractive interchain actions are predominant， the sol-to-gel transition occurs. It is generally assumed that hydrogen bonding interactions are not predominant at high temperature and thus hydrophobic effect is the main driving force for gelation of CS/β-GP[27]. Upon heating of CS/β-GP solution: (1) electrostatic repulsion forces between CS chains reduced; (2) structuring of free water by the glycerol of β-GP increased thus dehydrating CS chains; (3) electrostatic attraction between CS and β-GP decreased due to the transfer of protons from CS amine groups to the phosphate moiety of β-GP. All above are responsible for increasing interchain hydrophobic attraction of CS and thus producing sol-to-gel transition. Unfortunately， although the CS/β-GP hydrogel exhibited attractive gelation temperature and time for biomedical applications， its weak mechanical properties， resulting from lack of extra interactions except for hydrophobic and hydrogen bonding interactions， to some extent delayed its wide applications. Recent studies have been reported that its mechanical properties can be improved by blending with other materials while keeping thermosensitivity. For example， starch[28]， poly(vinyl alcohol)[29]， β-tricalcium phosphate[30]， or chopped silk fibers[31]were chosen to physically reinforce the hydrogel. In addition， chemical crosslinking was also adopted， such as poly(ethylene glycol) diacrylate[32]or thiolated chitosan[33]， which could provide additional chemical bonding to reinforce the system.

Considering the ChiNC has excellent reinforcing ability and the similar origin as CS， herein， we tried to use the ChiNC as nanofiller to enhance the mechanical properties of CS/β-GP hydrogel， and investigated its effects on solution behavior， gelation process and mechanical properties.

1 Experimental

1.1 Materials

Chitin flakes (from shrimp shell) and β-GP disodium were purchased from Sigma-Aldrich Co. Ltd.， USA. Low molecular weight chitosan (Mw≈70 000 g/mol) with a high degree of deacetylation (DDA， about 91%) was provided by Qingdao Hecreat Biotech Co. Ltd.， China. All chemicals were used as received without further processing.

1.2 Preparation of ChiNC

The ChiNC was prepared according to our previous work[14]. Briefly， 3 mol/L hydrochloric acid (HCl) and chitin flakes were mixed together with the ratio of 30 mL/g， refluxed for 6 h under stirring， and then centrifuged to collect residue. The residue was treated again with 3 mol/L HCl twice as above， washed with deionized water for three times， dialyzed against deionized(DI) water for 3 d， followed by ultrasonic treatment in a sonicating bath (Whaledent Biosonic， USA) for 20 min and subsequent filtration to remove residual aggregates. Finally， the clear colloidal aqueous suspension was lyophilized to obtain light brown powders. The yield is about 55%. Before use， the ChiNC was redispersed in deionized water (50 g/L) by sonification.

1.3 Preparation of CS/β-GP/ChiNC solutions

CS was dissolved in 0.1 mol/L HCl solution， and to it a desired amount of 50 g/L ChiNC aqueous suspension was added. The mass ratios ofm(ChiNC):m(CS) were controlled at 0∶100， 5∶95， 10∶90， 15∶85， 20∶80 and 30∶70， respectively， while CS concentration was maintained at 18 g/L. Then， in an ice bath， 700 g/L β-GP in 0.1 mol/L HCl was added dropwise to the CS/ChiNC solutions under stirring， until a mass ratio of 3.6∶1 with CS. The CS/β-GP/ChiNC gels were prepared by heating the aqueous solutions in an oven at 37.0 ℃ for 2 h. According to ChiNC content， the gels were denoted as 0%， 5%， 10%， 15%， 20% and 30% ChiNC， respectively.

1.4 Morphology and dispersibility tests of ChiNCs

The morphology and dispersibility of ChiNCs were evaluated via transmission electron microscopy (TEM). A drop of diluted ChiNC suspension in H2O was cast onto a carbon coated copper grid， slowly evaporated at room temperature， and then observed on a JEM-1230 TEM (JEOL， Japan).

1.5 Rheological behavior measurements

Rheological measurements were performed on an ARES-RFS rotationalrheometer (TA Instruments, USA) using parallel plate geometry (50 mm in diameter， with a gap of 1.5 mm). The viscoelastic properties of the CS/β-GP/ChiNC systems were assessed by measuring storage modulus (G′) representing elastic behavior， and loss modulus (G″) reflecting viscous behavior. Freshly prepared degassed solutions of 5 mL were introduced between the plates and then covered with mineral oil in order to prevent evaporation during the tests. The values of the strain amplitude were verified in order to ensure that all measurements were performed within the linear viscoelastic region， so thatG′ andG″ were independent of the strain amplitude.

Rheological characterizations of the CS/β-GP/ChiNC aqueous solutions consisted of different experiments as follows. Dynamic frequency sweep tests were performed at 22 ℃.G′ andG″ were measured over angular frequencies (ω) increasing through logarithmic steps from 0.1 to 100.0 rad/s. Gelation time of the thermogelling systems was assessed by time sweep tests， at constant angular frequency of 6.28 rad/s and constant temperature of 37 ℃. Temperature sweep tests were carried out at a constant frequency of 6.28 rad/s， with a temperature increase from 20 ℃ to 45 ℃ at a rate of 0.8 ℃/min. For all experiments， the sol-to-gel transition temperature (Ts/g) or time point (ts/g) correspond to the intersection of the curves ofG′ andG″.

1.6 Mechanical property measurements

The mechanical behaviors of gels were examined on a HY-941 universal material testing machine (Hengyu Instrument Co.， China) under compression mode at room temperature. The cylinder samples (15 mm in diameter， 10 mm in thickness) were compressed at a crosshead speed of 5 mm/min and a force of 100 N until 60% strain. The compression modulus (Young’s) was calculated from the initial slope of the stress-strain curve. The results of compression modulus and stress were the average of 5 specimens.

1.7 Morphology observation

The gels were frozen at -40 ℃ for over 4 h， and lyophilized at -50 ℃. Then the dried porous samples were cryo-fractured in liquid nitrogen， sputter coated with platinum， and examined via a field emission scanning electron microscopy (FESEM， SU8010， Hitachi Ltd.， Japan).

1.8 Water uptake tests

The water uptake of gels was examined in DI water. The freeze-dried gels of the known mass (Wa) were incubated in DI water for 24， 48 and 72 h， respectively， then removed and gently wiped with filter paper to remove excess water on the surface， and then weighed (Wb). The water uptake (S) was calculated as follows.

S/%=[(Wb-Wa)/Wa]×100.

(1)

In this study，three individual experiments were performed and data were averaged.

2 Results and Discussion

The gelation process of CS/β-GP system was dependent on the pH of solution， the molecule weight and deacetylation degree of CS， the concentrations of CS and β-GP[23]. Here， to explore the effects of ChiNC on CS/β-GP system， we kept the above variables constant and only discussed the variation of ChiNC content. Figure 1 shows the morphology and distribution of ChiNCs in H2O. The ChiNCs were individually distributed， and the estimated average length (L) and width (d) were around 300 nm and 20 nm， respectively， hence the aspect ratio(L/d) was around 15.

2.1 Rheological analysis

The time sweep tests were performed at 37 ℃， with the ChiNC content varying from 0% to 30% for exploring its influence on gelation time. The corresponding results are depicted in Fig. 2 and also summarized in Table 1. As expected， the progression ofG′ andG″ showed a clear sol-to-gel transition whenG′ became higher thanG″. As compared to the pure CS/β-GP system， the incorporation of ChiNC into CS/β-GP gave rise to strong differences in gelation time. Moreover， the gelation time was complicated by the addition amount of ChiNC. When the ChiNC content was 5%， the gelation time sharply increased from 344 s to 716 s, then further increased to 858 s at 10% ChiNC content, and followed by a little decrease to 807 s at 15% ChiNC content. Clearly， the gelation was retarded by the loading of ChiNC， when the amount of ChiNC ranged from 5% to 15%. While， as the ChiNC content was up to 20%， the gelation time sharply reduced to 244 s， shorter than that of pure CS/β-GP system, and then further reduced to 189 s at 30% ChiNC content. Generally， the sol-to-gel transition of CS/β-GP upon heating was ascribed to the increasing interchain hydrophobic attraction of CS[25]， while the presence of ChiNC would produce additional hydrogen-bond attraction between CS and ChiNC[34]， which would prevent interchain hydrophobic attraction of CS and thus delay gelation process. However， the gelation was accelerated when the added amount of ChiNC was not less than 20%， possibly due to the reason that the ChiNC can form a three-dimensional percolation structure via hydrogen-bonding interaction at a higher concentration[35]and thus construct a second network structure to produce gelation in this combined system.

Fig.2 Loss modulus (G′) and storage modulus (G″) of CS/ β-GP/ChiNC sol solutions as a function of time

The temperature sweep tests were conducted at a given rate. The corresponding rheological curves are shown in Fig. 3 and specific data are summarized in Table 1. The changes of gelation temperature with ChiNC content exhibited the same trend as that in the time sweep tests. Firstly， it increased from 32.7oC in the pure CS/β-GP system to 37.3oC at 10% ChiNC content; then began to continuously decrease to 28.3oC at 30% ChiNC content. These results disclose the clear influence of the loading ratio of ChiNC on the gelation temperature. And they confirmed our aforementioned hypothesis that at a lower concentration the ChiNC can form hydrogen-bond with CS against interchain hydrophobic attraction of CS; at a higher concentration the ChiNC can form a percolation structure and thus construct a second network to promote gelation at a low temperature.

Table 1 Summery of gelation temperature and time of CS/β-GP/ChiNC solutions

Fig.3 Loss modulus (G′) and storage modulus (G″) of CS/β-GP/ ChiNC sol solution as a function of temperature

The dynamic frequency sweep tests of CS/β-GP/ChiNC solutions at room temperature (22 ℃)，i.e. the commonly used operation temperature， are shown in Fig. 4. Since this temperature was below the sol-to-gel transition temperature， the systems should have displayed viscoelastic liquid behavior at a relatively low angular frequency range，i.e.G″>G′. However， the incorporation of ChiNC made the systems behaved elastic character， that is theG′ was independent of angular frequency and higher thanG″ over the initial frequency range. Moreover， theG″ increased with angular frequency resulting in a crossover followed byG″ exceedingG′， In addition， the crossover was shifted to higher frequencies at higher ChiNC concentration. Thus， the presence of ChiNC gave rise to an extra gel-like structure by the formation of ChiNC percolation network and reinforced the CS/β-GP system. The storage modulusG′ was extracted at a constant frequency (ω=1 rad/s) and expressed as a function of ChiNC concentration， which is depicted in Fig. 5. Clearly， upon increasing ChiNC content， theG′ increased， while theG″ basically unchanged， confirming a gel-like structure. Although that， the systems containing ChiNC behaved a shear-thinning feature， which endowed the composite gels with a good fluidity when injecting to fulfill the application requirement(Fig. 6).

Fig.4 Loss modulus (G′) and storage modulus (G″) of CS-β-GP/ChiNC sol solutions as a function of shear frequency at room temperature

Fig.5 Effects of ChiNC content on G′and G″ of CS/β-GP gels at shear frequency of 1 rad/s at room temperature

Fig.6 Complex viscosity of CS/β-GP/ChiNC solutions as a function of shear frequency at room temperature

2.2 Mechanical properties of CS/β-GP/ChiNC gels

The mechanical properties of CS/β-GP/ChiNC composite gels were evaluated by compression tests as depicted in Fig. 7. The stress-strain curves for gels with varied ChiNC content (Fig. 7(a)) and corresponding data (Fig. 7(b)) clearly showed that the compression moduli and stresses highly improved upon increasing ChiNC ratio， which was in agreement with other related reports[17]. The highest modulus ((172.07±36.38) kPa) and stress ((50.79±2.01) kPa) appeared at 30% ChiNC content， nearly 10 times and 5 times increase， respectively， as compared to the pure CS/β-GP gel.

2.3 Pore shape and structure of CS/β-GP/ChiNC gels

The cross-section morphologies of freeze-dried gels are shown in Fig. 8. All the samples exhibited laminated structure， and no obvious differences between ChiNC loaded and unloaded gels were observed.

Fig.7 Mechanical properties of CS/β-GP/ChiNC gels: (a) compression stress-strain curves; (b) compression Young’s modulus and stress at 60% strain

Fig.8 SEM images of CS-β-GP/ChiNC gels in dry state: (a) 0% ChiNC; (b) 10% ChiNC; (c) 20% ChiNC; (d) 30% ChiNC (scale bar = 500 μm)

2.4 Water uptake property of CS/β-GP/ChiNC gels

Another prominent result came from the water uptake as depicted in Fig. 9. The introduction of ChiNC significantly decreased water uptake， which was very important forinvivoapplication， considering a high swelling of implant would exert compression to nerve or vessel around.

Fig.9 Water uptake of CS/β-GP/ChiNC gels (*p<0.05)

3 Conclusions

The CS/β-GP temperature-sensitive sol-to-gel system was highly affected by introducing ChiNC. At a higher loading ratio， the ChiNC speeded up the sol-to-gel transition rate and reduced the sol-to-gel transition temperature， while， exhibited a distinct shear-thinning behavior. The above mentioned rules favored the application as injectable tissue-repair implants. Most importantly， the weak mechanical properties of CS/β-GP hydrogel were significantly improved by loading ChiNC to a certain degree， accompanied by a positive result of decreased water uptake. Thus， ChiNC is a promising nanofiller candidate for improving the properties of CS/β-GP gel.