Hydrophobically modified biomineralized polysaccharide alginate membrane for sustained smart drug delivery

International Journal of Biological Macromolecules 50 (2012) 747–753

Contents lists available at SciVerse ScienceDirect

International Journal of Biological Macromolecules journal homepage: www.elsevier.com/locate/ijbiomac

Hydrophobically modified biomineralized polysaccharide alginate membrane for sustained smart drug delivery Jun Shi ∗ , Zhengzheng Zhang, Wenyan Qi, Shaokui Cao ∗ School of Materials Science and Engineering, Zhengzhou University, Zhengzhou 450052, China

a r t i c l e

i n f o

Article history: Received 19 October 2011 Received in revised form 20 November 2011 Accepted 8 December 2011 Available online 17 December 2011 Keywords: Alginate Biomineralization Hydrophobically modified Multi-responsive Smart drug delivery

a b s t r a c t Hydrophobically modified biomineralized polysaccharide alginate membrane with smart drug release property using sodium palmitate as the hydrophobic component was prepared via a one-step method. The formation of CaHPO4 in the membrane was clearly identified through scanning electron microscopy (SEM), energy dispersive X-ray spectrometer (EDS), X-ray diffraction (XRD) and Fourier transform infrared (FT-IR) spectroscopy. Indomethacin release profiles of the modified alginate membrane were found to be pH- and thermo-responsive. The drug release of modified alginate membrane was around 60% within 12 h, while that of the alginate membrane was higher than 90%. These results indicate that the hydrophobic and biomineralized polysaccharide components can hinder the permeation of the encapsulated drug and reduce the drug release effectively. The resulting membrane can be used as “smart” materials for sustained dual-responsive drug delivery. © 2011 Elsevier B.V. All rights reserved.

1. Introduction There has been a great deal of research activity in the development of stimulus–responsive membranes, especially in those with pH, temperature or ionic strength sensitivity [1–6]. Alginate is a pH sensitive and biocompatible natural hydrogel material with relatively low cost [7–9]. Much research has also been done to associate biopolymers (such as alginate and chitosan) with thermo-sensitive macromolecules in an attempt to prepare the matrixes presenting a multiple and independent sensitivity to ambient changes [10,11]. Poly(N-isopropylacrylamide) (PNIPAAm) is one of the most widely studied thermo-sensitive polymers, which exhibits remarkable hydration–dehydration changes in aqueous solution in response to relatively small changes in temperature near its lower critical solution temperature (LCST) around 30 ◦ C [12]. It is of particular importance that PNIPAAm exhibits very low toxicity due to the excretion through glomerular filtration. Previous studies have shown biomineralized polysaccharide hybrid structure could improve the mechanical strength and sustained release behaviour of alginate beads [13–15]. Moreover, biomineralized polysaccharide hybrid materials are likely to be of generic importance in the design of biocompatible microcapsules because they often exhibit complementary properties, and have potential applications for cell growth, drug delivery and

∗ Corresponding authors. Tel.: +86 371 67763523; fax: +86 371 67763561. E-mail addresses: [email protected] (J. Shi), [email protected] (S. Cao). 0141-8130/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.ijbiomac.2011.12.003

the implications for tissue engineering [16–18]. On the other hand, hydrophobic surfactant can be introduced into alginatebased materials to decrease the swelling ratio and sustain the drug release [19]. Most of the hydrophobically modified alginate materials reported in the literatures were prepared via the chemical reactions between alginate and the hydrophobic component [20–22]. While, potential cytotoxicity of the residual chemicals should not be ignored for the biomedical applications. To the best of our knowledge, we have not found any report concerning the combination of biomineralized polysaccharide and hydrophobic components into alginate composite membranes to prepare smart polysaccharide materials. Herein, we describe the preparation of hydrophobically modified biomineralized polysaccharide alginate membranes with sustained smart drug release via a one-step method, in which the deposition of the porous alginate/chitosan polyelectrolyte around alginate membranes containing hydrophobic component is coupled with the controlled precipitation of calcium phosphate as illustrated in Scheme 1. PNIPAAm and sodium palmitate were employed as thermal-responsive and hydrophobic components, respectively. Sodium palmitate is a non-toxic fatty acid sodium salt with an appropriate alkyl chain length ( (CH2 )14 ), which can effectively inhibit water penetrating into the beads and decrease the water uptake. Biomineralized component (CaHPO4 ) formed between Ca2+ and HPO4 2− and the polyelectrolyte formed between positively charged chitosan and negatively charged alginate exist in the resulting membranes. The diffusion-controlled deposition of inorganic minerals and hydrophobic component within porous

748

J. Shi et al. / International Journal of Biological Macromolecules 50 (2012) 747–753

Scheme 1. Schematic illustration of hydrophobically modified alginate membrane.

organic polymeric matrices could prevent the permeation of the encapsulated drug and reduce the drug release effectively. The equilibrium swelling behaviour of the modified membranes as well as their controlled delivery performance was investigated as a function of pH and temperature. 2. Materials and methods 2.1. Materials N-isopropylacrylamide (NIPAAm, Tokyo Chemical Industry Co. Ltd., Japan), ammonium persulfate (APS, Shanghai Chemical Regent Co. Ltd., China), N,N,N ,N -tetramethyl ethylenediamine (TEMED, Shanghai Chemical Regent Co. Ltd., China), sodium alginate (1% aqueous solution with a viscosity of 20 cps at 20 ◦ C, M = 3.95 × 105 , mannuronic acid/guluronic acid (M/G) ratio of alginate = 0.78, Shanghai Chemical Regent Co. Ltd., China), chitosan (with a 92% degree of deacetylation and a viscosity of 55 cps, Shanghai Chemical Regent Co. Ltd., China) and indomethacin (Shanghai Houcheng Chemical, China) were used as received. Sodium palmitate was obtained by neutralization of palmitic acid (99% purity, Aldrich) with sodium hydroxide at 80 ◦ C. 2.2. Preparation of alginate and hydrophobically modified alginate membranes PNIPAAm was synthesized according to the literature procedures [14]. Homogeneous aqueous solutions composed of sodium alginate (3%, w/v), PNIPAAm (PNIPAAm:alginate = 1:3 (w/w)) and 20% (w/w) of indomethacin (relatively to the total weight of alginate and PNIPAAm) were prepared. The obtained solutions were sonicated in a sonication bath and left to stand until the trapped air bubbles were removed. Then the solution was spread on a PTFE plate (12 cm × 12 cm) and dried at room temperature. The dried membrane was immersed in 5% CaCl2 solution for 60 min. The membranes were washed, then dried in air overnight and vacuum dried at 40 ◦ C for 24 h. The preparation of the hydrophobically modified membranes is similar to the above procedure, except the addition of 0.45% (w/v) of sodium palmitate into the alginate solution.

removed. A homogeneous aqueous solution of chitosan (1%, v/v) in 1% acetic acid containing 5% of CaCl2 was used as a coagulation fluid, which was mixed for 2 h before use and the pH value was adjusted to 6.4 ± 0.2 by adding NaOH solution (1 M). Thereafter, the homogeneous alginate solution was spread on a PTFE plate (12 cm × 12 cm) and dried at room temperature. The dried membranes were immersed into the coagulation fluid for 60 min. The membranes were washed, then dried in air overnight and vacuum dried at 40 ◦ C for 24 h. The thicknesses of the resulting membranes were about 100 ␮m. The sample was labelled as alginate/sodium palmitate/biomineralized polysaccharide membranes (alginate/sodium palmitate/BP). 2.4. Characterization of membranes FT-IR spectra of the samples were recorded with a Bruker Tensor 27 FT-IR spectrometer in the range of 4000–500 cm−1 using KBr pellets. The morphology and composition of the prepared membranes were observed using scanning electron microscopy (SEM, FEI Quanta 200) at an accelerated voltage of 20 kV. Before SEM observation, the membranes were stabilized on aluminium stubs using adhesive and sputter coated with an approximate 100 A˚ layer of gold. The formation of biomineralized polysaccharide component was confirmed by means of energy dispersive X-ray spectrometer (EDS, EDAX). Thermogravimetric analysis (TGA) was carried out on a STA 409 PC/PG simultaneous thermal analyzer (Netzsch) at a heating rate of 10 ◦ C min−1 under Ar atmosphere. X-ray power diffraction (XRD) patterns were recorded on a Rigaku D/MAX 2550 ˚ with V X-ray diffractometer using Cu K␣ radiation ( = 1.54178 A) a graphite monochromator. 2.5. Swelling studies The swelling behaviour of the prepared membranes was studied in a phosphate buffer solution (PBS) with pH 7.4 and in an HCl solution with pH 2.1 at temperatures of 25 ◦ C and 37 ◦ C, respectively. At predetermined time intervals, the swollen membranes were weighted after wiped with soft paper tissue. The degree of swelling for each sample was calculated by using the following expression: Swelling ratio = (Ws − Wd )/Wd , where Ws and Wd are the weight of the swollen and the dried membranes, respectively.

2.3. Preparation of hydrophobically modified biomineralized polysaccharide membranes

2.6. Determination of drug encapsulation efficiency

Homogeneous aqueous solutions composed of sodium alginate (3%, w/v), PNIPAAm (PNIPAAm:alginate = 1:3 (w/w)), 0.45% (w/v) of sodium palmitate and 20% (w/w) of indomethacin (relatively to the total weight of alginate and PNIPAAm) and 50 mM of Na2 HPO4 were prepared. The obtained solutions were sonicated in a sonication bath and left to stand until the trapped air bubbles were

The prepared membranes (10 mg) were dispersed in 100 mL of PBS (pH 7.4, containing 5% (v/v) ethanol) under stirring during 24 h. The amount of free indomethacin was determined in the clear supernatant by UV spectrophotometry at 320 nm using the calibration curve. Such experiments allow the calculation of both the loading efficiency and the loading content. The loading efficiency

J. Shi et al. / International Journal of Biological Macromolecules 50 (2012) 747–753

749

Fig. 1. SEM micrographs of membranes. a, b and c refer to alginate, alginate/sodium palmitate and alginate/sodium palmitate/BP membranes, respectively. 1, 2 and 3 refer to the surface, the high magnification of the surface and the cross section of the membranes, respectively.

is defined as the weight percentage of loaded drug based on feed amount. 2.7. In vitro release studies The prepared membranes (10 mg) were suspended in 50 mL of PBS (pH 7.4) or HCl solution (pH 2.1). This dissolution medium was stirred at 50 rpm in a horizontal laboratory shaker and maintained at 25 and 37 ◦ C. The sample (2 mL) was periodically removed and the withdrawn sample was replaced by the same volume of fresh medium. The amount of released indomethacin was analysed with a UV spectrophotometer. All the tests, including measurement of the swelling ratio, determination of the drug content and in vitro drug release, were carried out in triplicate, and the average values were shown in this study. 3. Results and discussion Hydrophobically modified biomineralized polysaccharide alginate membranes with thermal- and pH-responsive property were prepared via a one-step method as described in Scheme 1. The deposition in porous alginate/chitosan polyelectrolyte complexes

around the resulting membrane is coupled with a controlled precipitation of calcium phosphate arising from counter-diffusion of ions across the polysaccharide interface at a suitable pH value [13,14,17]. The biomineralized polysaccharide layer hardens with time due to the deposition of calcium phosphate but remains permeable to Ca2+ ions, such that the membranes are also internally stabilized due to cross-linking of the alginate network. The hydrophobic and biomineralized polysaccharide component endow the resulting hybrid membranes with higher mechanical strength and sustained release properties, and does not ruin the thermal- and pH-sensitivity at the same time.

3.1. Characterization of membranes Three kinds of membranes were prepared in this study: alginate/sodium palmitate/BP, alginate/sodium palmitate and alginate membranes. The drug loading efficiency of the alginate/sodium palmitate/BP membranes (75.6%) was higher than that of the alginate/sodium palmitate (69.2%) and alginate membranes (62.1%). The possible reason to this phenomenon is that sodium palmitate and biomineralized inorganic component decrease the membrane hydrophilicity, which leads to a relatively low loss of indomethacin

750

J. Shi et al. / International Journal of Biological Macromolecules 50 (2012) 747–753

Fig. 2. EDS spectra of alginate membrane (A), alginate/sodium palmitate membrane (B), alginate/sodium palmitate/BP membrane (C) and TG curves of membranes (D).

during the membrane preparation process. Moreover, the electrostatic interaction between chitosan and alginate also form a barrier layer to reduce the loss of entrapped drug during the preparation process. Surface morphology of the studied membranes was observed with SEM (Fig. 1). High magnification SEM images showed a progressive change in the texture of the outer space with the introduction of the hydrophobic and biomineralized polysaccharide composition [23]. The morphology of modified membrane was rougher than that of the alginate ones. The hydrophobic alkyl chains ( (CH2 )14 ) of sodium palmitate occupy a particular space in the hydrogel networks (see Scheme 1), which leads to a lower crosslinking density and rough structure. Moreover, the diffusion-controlled deposition of inorganic minerals within alginate matrices also increases the roughness of the modified alginate composite membranes. A spot of indomethacin crystal could be observed in the alginate membrane from Fig. 1, while no indomethacin crystal could be found in alginate/sodium palmitate/BP and alginate/sodium palmitate membranes. This phenomenon may be attributed to the hydrophobic interaction between sodium palmitate and indomethacin. The electrostatic interaction between chitosan and alginate also brings a barrier layer to entrap the drug within the hybrid membrane. The presence of Ca and P elements within the outer biomineralized polysaccharide shell was confirmed by EDS analyses (Fig. 2). A signal for P element was observed for alginate/sodium palmitate/BP membranes (Fig. 2C), which could not be found from the alginate

(Fig. 2A) and alginate/sodium palmitate ones (Fig. 2B). The higher contents of P, Ca and O elements indicate that the outer shell of hydrophobically modified biomineralized polysaccharide alginate membrane is mainly composed of CaHPO4 . EDS analyses indicate that calcium ion could diffuse and react with the phosphate leading to CaHPO4 formation surrounding the membranes. Thermogravimetric analyses (Fig. 2D) could also be employed to demonstrated the existence of CaHPO4 component in the hybrid membrane. It can be found that the thermal stability of hydrophobically modified biomineralized polysaccharide alginate membrane was better than that of alginate and hydrophobically modified alginate membranes. As illustrated in Fig. 2D, 45% of hydrophobically modified biomineralized polysaccharide alginate membranes were not decomposed at 800 ◦ C, indicating the existence of CaHPO4 . Fig. 3A presents the FT-IR spectra of the prepared membranes. After the introduction of sodium palmitate, strong absorptions at 2856 and 2919 cm−1 appeared as a result of the introduction of hydrophobic methyl and methylene group. Additionally, three strong absorptions at around 1033 cm−1 , 601 cm−1 and 566 cm−1 assignable to the P O bonds could be observed for alginate/sodium palmitate/BP membranes, suggesting the formation of CaHPO4 within the membrane [13,14]. Fig. 3B shows the XRD patterns of the prepared membranes. For alginate/sodium palmitate/BP membranes, two distinct typical reflections of 211 and 222 centred at 2 of 32 and 45 corresponding to the HAP crystals could be observed, which could not be found in the results of the other two membranes.

J. Shi et al. / International Journal of Biological Macromolecules 50 (2012) 747–753

751

A

a

2930

Transmittance

2974 2856

2919

b c

2856

566

2919

601 1033

4000

3500

3000

2500

2000

1500

1000

500

-1

Wavenumbers (cm )

B

Intensity (a.u.)

211

222

c b a 10

20

30

40

50

60

70

80

2 (degree) Fig. 3. FT-IR spectra (A) and XRD curves of membranes (B). a, b and c refer to alginate, alginate/sodium palmitate and alginate/sodium palmitate/BP membranes, respectively.

3.2. Swelling study Fig. 4A shows the swelling behaviour of the prepared membranes at different pH values and temperatures. The swelling ratio decreased progressively with the introduction of sodium palmitate and biomineralized polysaccharide components. The interaction between sodium palmitate and biomineralized polysaccharide component endows the alginate/sodium palmitate/BP membrane with lower swelling ratio compared with other two membranes. The long alkyl chain ( (CH2 )14 ) of sodium palmitate inhibits water from penetrating into the membrane, which leads to the decrease in swelling ratio [20,21]. On the other hand, the biomineralized polysaccharide outer shell can prevent the water to penetrate into the membranes and then decrease the swelling ratio [24]. It also can be observed from Fig. 4A that the swelling ratio at 37 ◦ C was higher than that at 25 ◦ C for all the membranes. There is a competition between the two procedures in the studied system: one is the diffusion affected with temperature, the other is the precipitation of PNIPAAm when temperature above its LCST [25]. The swelling behaviour of the studied membranes will be dominated by the superior procedure. A high temperature can help the water to penetrate into the membranes, which contributes to the higher swelling ratio. At pH 2.1, most of the carboxylic groups in alginate are in the form of COOH, as its pKa is about 3.2. The hydrogen bonds between COOH in alginate and CONH in

Fig. 4. Changes of temperature- and pH-dependence in equilibrium swelling ratio for the studied membranes (A) and the sustained drug release profiles of the membranes (B).

PNIPAAm lead the polymer–polymer interactions predominating over the polymer–water interactions. As a result, the swelling ratio of the studied membranes is very low at pH 2.1. As the pH value changed to 7.4, the carboxylic groups become ionized and a small quantity of H+ in water acts as a bridge within alginate molecules, resulting in the increase of the swelling ratio [26]. It also should be pointed that calcium alginate can be rapidly changed to alginic acid that is insoluble and less swelling in acidic acid. 3.3. Drug release study 3.3.1. Sustained release Fig. 4B shows the indomethacin release profiles of the studied membranes at 37 ◦ C and pH 7.4. The drug release was around 60% within 12 h for the alginate/sodium palmitate/BP membrane, while the values for the alginate and alginate/sodium palmitate membranes were 94% and 67%, respectively. The drug release results indicated that the in situ mineralization process and the introduction of sodium palmitate were successful in decreasing membrane permeability and then the drug release [17,27]. A Student’s t-test analysis was conducted to support this point, the p value obtained in the comparison between the data of alginate/sodium palmitate/BP and alginate membranes was 0.0015 for the referred time period. The difference between these data was statistically significant (greater than 95% confidence) for the considered time period. It must be noted that the drug release of alginate/sodium palmitate/BP membranes was lower than that of the biomineralized polysaccharide alginate membranes [15]. The results demonstrated

752

J. Shi et al. / International Journal of Biological Macromolecules 50 (2012) 747–753

A

100

Drug release (%)

80

60

40

o

pH7.4, 37 C o pH2.1, 37 C

20

0 0

2

4

6

8

10

12

Time (h)

B

100

4. Conclusions

80

Drug release (%)

alginate membrane, the drug release at 37 ◦ C was higher than that at 25 ◦ C. Student’s t-test analysis showed that the difference between the release of alginate/sodium palmitate/BP membrane at 37 and 25 ◦ C was statistically different (p value was 0.0499, greater than 95% confidence). The plausible explanation is that the effective crosslinking density of the calcium–alginate network would be reduced by the precipitation of PNIPAAm at higher temperature, which would fill the coil space or porosity of the membranes and then accelerate the drug release [14]. Therefore, the precipitation of PNIPAAm in the gel matrix plays a critical role in squeezing out the entrapped drug molecules from the gel membranes at 37 ◦ C. Moreover, the squeezing of PNIPAAm at 37 ◦ C can also break the balance of the semi-IPN network and then accelerate the disruption of the membranes. The pH- and thermo-responsive properties indicate that alginate/sodium palmitate/BP membranes could prevent the permeation of the encapsulated drug and reduce the drug release effectively, and preserve the stimuli–responsive properties after the formation of CaHPO4 at the same time.

60

40 o

37 C, pH7.4 o 25 C, pH7.4

20

0 0

2

4

6

8

10

12

Time (h) Fig. 5. pH-dependent release profiles of indomethacin at 37 ◦ C measured at pH 2.1 and 7.4 (A) and temperature-dependent release profiles of indomethacin at pH 7.4 measured at 25 ◦ C and 37 ◦ C (B) from alginate/sodium palmitate/BP membranes.

that the interaction between sodium palmitate and biomineralized polysaccharide component could endow the alginate/sodium palmitate/BP membrane with satisfied sustained drug release property. The drug release properties are in line with the results of surface morphology and swelling ratio of the alginate/sodium palmitate/BP membranes. 3.3.2. Temperature/pH-sensitivity Fig. 5A presents the drug release behaviours at 37 ◦ C for alginate/sodium palmitate/BP membrane at pH 2.1 and pH 7.4. It was clear that the drug release was 60% after 12 h at pH 7.4, whereas the value was less than 5% at pH 2.1 with the same treatment. The relatively low release at pH 2.1 should be related to the low swelling ratio of the membrane in acidic conditions as shown in Fig. 4A [28]. Therefore, a significant pH-dependent response can be observed for hydrophobically modified composite membranes. It should be pointed that the solubility of indomethacin in acidic conditions is much lower than in neutral solutions. Hence, besides the response of alginate to pH, clearly detected in swelling, the release profile may also be dependent on pH value due to the difference in indomethacin solubility. Fig. 5B shows the drug release profiles of alginate/sodium palmitate/BP membrane in PBS (pH 7.4) at 25 and 37 ◦ C, respectively. The selected thermal-responsive section in our study is PNIPAAm, whose LCST is around 30 ◦ C. Therefore, we chose 25 ◦ C, which is lower than the LCST, and 37 ◦ C, which is higher than the LCST, as the temperature conditions in our study. A temperature dependent response could be observed for the hydrophobically modified

Hydrophobically modified biomineralized polysaccharide alginate membranes with sustained smart drug release property were prepared via in situ biomineralization procedure. SEM, EDS, FT-IR, TG and XRD results demonstrated the formation of CaHPO4 in the resulting membranes. Swelling ratio and drug release behaviours of the composite membrane demonstrated that sodium palmitate and biomineralized polysaccharide component could prevent the permeation of the encapsulated drug and reduce the drug release effectively, and did not ruin the thermal- and pH-sensitivity at the same time. These results suggest that the resulting hybrid alginate membranes have the potential to be used as “smart” polysaccharide material for sustained multi-responsive drug delivery. Acknowledgement This work was financially supported by the National Natural Science Foundation of China (Project 20874090). References [1] L. Grossin, D. Cortial, B. Saulnier, O. Félix, A. Chassepot, G. Decher, P. Netter, P. Schaaf, P. Gillet, D. Mainard, J.C. Voegel, N. Benkirane-Jessel, Adv. Mater. 21 (2009) 650–655. [2] H. Mjahed, C. Porcel, B. Senger, A. Chassepot, P. Netter, P. Gillet, G. Decher, J.C. Voegel, P. Schaaf, N. Benkirane-Jesselab, F. Boulmedais, Soft Matter 4 (2008) 1422–1429. [3] C.Y. Wang, S.Q. Ye, L. Dai, X.X. Liu, Z. Tong, Biomacromolecules 8 (2007) 1739–1744. [4] H.J. Dai, X.F. Li, Y.L. Long, J.J. Wu, S.M. Liang, X.L. Zhang, N. Zhao, J. Xu, Soft Matter 5 (2009) 1987–1989. [5] Z. Cao, B.Y. Du, T.Y. Chen, H.T. Li, J.T. Xu, Z.Q. Fan, Langmuir 24 (2008) 5543–5551. [6] J.B. Qu, L.Y. Chu, M. Yang, R. Xie, L. Hu, W.M. Chen, Adv. Funct. Mater. 16 (2006) 1865–1872. [7] M.R. Abidian, D.C. Martin, Adv. Funct. Mater. 19 (2009) 573–585. [8] S.M. Jay, W.M. Saltzman, J. Control. Release 134 (2009) 26–34. [9] A.K. Nayak, D. Pal, Int. J. Biol. Macromol. 49 (2011) 784–793. [10] M. George, T.E. Abraham, J. Control. Release 114 (2006) 1–14. [11] G. Fundueanu, M. Constantin, P. Ascenzi, Biomaterials 29 (2008) 2767–2775. [12] J. Shi, N.M. Alves, J.F. Mano, Adv. Funct. Mater. 17 (2007) 3312–3318. [13] J. Shi, L.H. Liu, X.P. Liu, X.M. Sun, S.K. Cao, Polym. Adv. Technol. 19 (2008) 1467–1473. [14] J. Shi, L.H. Liu, X.M. Sun, S.K. Cao, J.F. Mano, Macromol. Biosci. 8 (2008) 260–267. [15] J. Shi, X.P. Liu, Y.J. Shang, S.K. Cao, J. Membr. Sci. 352 (2010) 262–270. [16] G.M. Luz, J.F. Mano, Compos. Sci. Technol. 70 (2010) 1777–1788. [17] D.W. Green, I. Leveque, D. Walsh, D. Howard, X. Yang, K. Partridge, S. Mann, R.O.C. Oreffo, Adv. Funct. Mater 15 (2005) 917–923. [18] D. Baskar, R. Balu, T.S.S. Kumar, Int. J. Biol. Macromol. 49 (2011) 385–389. [19] M. Leonard, M.R. De Boissesona, P. Hubert, F. Dalencon, E. Dellacherie, J. Control. Release 98 (2004) 395–405. [20] B.L. Yao, C.H. Ni, C. Xiong, C.P. Zhu, B. Huang, Bioprocess Biosyst. Eng. 33 (2010) 457–463.

J. Shi et al. / International Journal of Biological Macromolecules 50 (2012) 747–753 [21] V. Burckbuchler, A.L. Kjoniksen, C. Galant, R. Lund, C. Amiel, K.D. Knudsen, B. Nystrom, Biomacromolecules 7 (2006) 1871–1878. [22] B. Nystrom, A.L. Kjoniksen, N. Beheshti, K.Z. Zhu, K.D. Knudsen, Soft Matter 5 (2009) 1328–1339. [23] I. Leveque, K.H. Rhodes, S. Mann, J. Mater. Chem. 12 (2002) 2178–2180. [24] J. Li, Z.Y. Jiang, H. Wu, L.H. Long, Y.J. Jiang, L. Zhang, Compos. Sci. Technol. 69 (2009) 539–544.

[25] [26] [27] [28]

753

E.S. Gil, S.M. Hudson, Prog. Polym. Sci. 29 (2004) 1173–1222. J. Shi, Z.Z. Zhang, G.F. Li, S.K. Cao, J. Mater. Chem. 21 (2011) 16028–16034. J.H. Liu, L. Li, Y.Y. Cai, Eur. Polym. J. 42 (2006) 1767–1774. A.K. Anal, D. Bhopatkar, S. Tokura, H. Tamura, W.F. Stevens, Drug Dev. Ind. Pharm. 29 (2003) 713–724.