Synthesis of a novel zwitterionic biodegradable poly (α,β-l -aspartic acid) derivative with some l -histidine side-residues and its resistance to non-specific protein adsorption

Colloids and Surfaces B: Biointerfaces 86 (2011) 237–241

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb

Synthesis of a novel zwitterionic biodegradable poly (␣,␤-l-aspartic acid) derivative with some l-histidine side-residues and its resistance to non-specific protein adsorption Xiaojuan Wang a , Guolin Wu a,∗ , Caicai Lu a , Yinong Wang a , Yunge Fan a , Hui Gao b , Jianbiao Ma b,∗ a b

The Key Laboratory of Functional Polymer Materials (Ministry of Eduction), Institute of Polymer Chemistry, Nankai University, Tianjin 300071, PR China School of Chemistry and Chemical Engineering, Tianjin University of Technology, Tianjin 300191, PR China

a r t i c l e

i n f o

Article history: Received 12 February 2011 Received in revised form 5 April 2011 Accepted 6 April 2011 Available online 13 April 2011 Keywords: Zwitterionic Polypeptide derivative Biodegradability Protein adsorption

a b s t r a c t A novel zwitterionic polypeptide derivative, denoted as His-PAsp/PAsp, was successfully synthesized by amidation of Poly (␣,␤-l-aspartic acid) with l-histidine methyl ester. Turbidity, zeta potential and 1 H NMR measurements were used to study the aggregation behaviors of His-PAsp/PAsp under different pH values. The modified polypeptide derivative composed of electro-negatively carboxylic and electro-positively imidazole residues randomly so as to bear opposite charges at pH values above or below the isoelectric point. When the zwitterionic polypeptide was coated on silicon wafer as a model substrate material, the absorption resistance of fibrinogen, a blood protein resulting in the blood coagulation cascade, on the coated surface was measured. It was found that the adsorption amount of fibrinogen on the polypeptidecoated surface depended on the dose of the polypeptide on silicon wafer. Obvious resistance of the fibrinogen adsorption on the polypeptide-coated surface was observed. Since its good biodegradability and superior anti-protein-fouling property, this pH-responsive zwitterionic polypeptide is a promising candidate for surface modification in many biomedical applications, including medical implants, drug delivery carriers, and biosensors. © 2011 Elsevier B.V. All rights reserved.

1. Introduction Biomaterials have been intensively studied due to their potential applications in biomedical fields, such as medical implants, drug delivery carriers, and biosensors. Protein adsorption is the first response from human body to foreign materials exposed to physical environment. However, most of the protein adsorption is unwanted and harmful for the biomaterials using in a living body. The protein adsorption may induce adverse bioresponses, including complement activation, coagulation and thrombosis, or other undesirable biophysical and biochemical processes [1–3]. For example, liposomes, vehicles, micelles and polymeric nanoparticles have been widely used as nanocarrier systems for targeted delivery [4,5], but the size increase and serious particle agglomeration leaded by adsorbing protein C3b, fibrinogen or immunoglobulins G, may resulting in rapid clearance from blood circulation and concentration in liver and spleen and severely compromising drug delivery to the target tissue [1,6–8]. The fibrinogen–cell interaction subsequently promotes thrombin, and coagulation cascade [9]. Actually, those reactions in biology mostly occur at interfaces [10,11], thus it

∗ Corresponding authors. Tel.: +86 22 23507746; fax: +86 22 23502749. E-mail addresses: [email protected] (G. Wu), [email protected] (J. Ma). 0927-7765/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfb.2011.04.010

is very necessary for surface modification to inhibit or prevent protein adsorption [12]. The non-cytotoxicity of the materials is also required for biomedical applications [13–15]. Poly(ethylene glycol), PEG, is the most preferable biologically inert polymer material with the low-fouling property of proteins [16]. Biomaterials coated with PEG have been shown to escape clearance by the mononuclear phagocytic system with a prolonged circulation time in the blood-stream [17]. However, the susceptibility of PEG to oxidation damage compromised its utility for long-term applications [18]. Its repulsive properties are diminished above 35 ◦ C [19]. In addition, studies shown that the PEG surfaces limit the interaction between carriers and the target tissues, compromising effective cellular uptake of the loaded cargo [20,21]. Recently, zwitterionic polymers are known to be effective in reducing protein adsorption [22–25]. The ultra-low fouling property is achieved by a uniformly mixture of balanced charges at the molecular scale on the modified surface. It is a very challenging to achieve biocompatibility and biodegradability, which are primary requirements for in vivo applications. There are only a few works reported to circumvent this problem. Jiang had synthesized biodegradable ultra-low fouling natural peptides via ring opening polymerization, which exhibit high resistance to nonspecific protein adsorption [26]. Yeo reported a zwitterionic chitosan synthesized by amidation of chitosan with succinic anhydride. This

238

X. Wang et al. / Colloids and Surfaces B: Biointerfaces 86 (2011) 237–241

chitosan derivative served as a coating material to prevent protein adsorption to cationic surfaces, which can be removed in a pH-responsive manner [27]. Poly(amino acids) have been extensively studied as carriers for drug delivery, due to their unique regular secondary structures, rigid backbone, biocompatibility, biodegradability and various side groups for further functionalization. The self-assembled morphology and the aggregate size of the zwitterionic poly(amino acids) may be controlled by changing the environmental pH. Self-assembly properties of block and random ampholytic polypeptides under different conditions have been investigated by several groups [28,29]. Most of the polypeptides reported so far were prepared by ring-opening polymerization of amino acid Ncarboxyanhydrides (NCAs). However, the NCA route is of a high cost because the side reactive groups of the amino acids need to be protected prior to polymerization and deprotected after the polymer formation under harsh conditions to give the functional side groups [30]. Poly (␣,␤-l-aspartic acid), PAsp, can be synthesized by thermal polycondensation of l-aspartic acid (l-Asp) with low cost. It would be a good candidate for preparing zwitterionic polypeptide derivatives. A biodegradable and zwitterionic derivative of poly (␣,␤-laspartic acid) was recently synthesized in our laboratory via partial conjugation with l-histidine methyl ester (His-OME) onto its side carboxylic groups. This zwitterionic derivative of poly (␣,␤-laspartic acid) has an isoelectric point (IEP) owing to both weak basic and weak acid groups as the side chains. At pH values near the isoelectric point, the polymer is expected to be electrically neutral and mostly insoluble in an aqueous medium. Its zwitterionic properties and ability to serve as an anti-biofouling material were then investigated.

acidic, alkaline and deionized water subsequently. The final product, denoted as His-PAsp/PAsp, was obtained through lyophilization. Its chemical structure and histidine-grafting ratio were confirmed by proton nuclear magnetic resonance (1 H NMR). 2.3. Nuclear magnetic resonance (NMR) spectroscopy 1 H NMR spectra were recorded on a Varian UNITY-plus 400 spectrometer using D2 O as the solvent.

2.4. Buffering capacity measurement The buffering capacity of the polymer was evaluated using the acid–base titration method. The polymer solution at 1 mg/mL of concentration was prepared in 10 mM NaCl solution. The initial pH of His-PAsp/PAsp solution was adjusted to pH 11 with 1 M NaOH. The titration curve was obtained by monitoring the pH decrease of the solution upon addition of 0.1 M HCl stepwise. The pH values were measured with a Sartorrious PB-10 pH meter. 2.5. Turbidity measurements Turbidity was measured at 380 nm using a Shimadzu2405-UV UV-vis spectrophotometer. All measurements were carried out in a quartz cuvette (1 cm width). 2.6. Zeta potential measurements

2. Materials and methods

The zeta potentials were measured using Nano-ZS90 (Malvern) with He–Ne laser at a wavelength of 633 nm.

2.1. Materials

2.7. Protein adsorption measurements

l-Aspartic acid (purity > 99.5%) was obtained from Shanghai Chemical Co., 1-ethyl-3-(3-dimethyllaminopropyl) carbodiimide (EDC, purity > 99.5%), l-histidine methyl ester dihydrochloride (His-OMe, purity > 98%), Aminopropyltriethoxysilane (APTES, >99%) were purchased from Alfa Aesar. Triethylamine (TEA, Tianjin Chemical Co.) was purified by distillation before use. silicon wafers were obtained from Beijing Top Vendor Science & Technology Co. Ltd., for protein adsorption experiments. The wafers were cleaned and treated as described [31]. The wafers slides were submerged in Piranha solution (70/30, v/v, sulfuric acid/hydrogen peroxide) for 20 min and subsequently rinsing them with water. After this process was repeated, the slides were sonicated in 50% (v/v) isopropanol/water for 20 min and then washed with water. The slides were heated to 60 ◦ C in RCA solution (5:1:1, v/v/v, water/hydrogen peroxide/ammonia solution) for 20 min. Finally, they were washed with water and dried under a stream of nitrogen. Fibrinogen was purchased from Lianxing biotechnology and used without any treatment. All solvents were redistilled before use.

The silica wafers were pre-coated with the polymer HisPAsp/PAsp before protein adsorption experiments. Rectangular silicon wafer slides (1.1 mm × 1.1 mm) were treated with a mixture containing 2 mL of ethanol, 0.4 mL of 3-aminopropyl triethoxysilane (APTES) and 0.1 mL of (25–28%) ammonia solution for 2 h. The slides were then rinsed with ethanol and water, dried under a stream of nitrogen to afford the amine-functionalized silica wafers with a electro-positive surface, named as NH2 -SW [31]. The NH2 -SW slides were incubated in His-PAsp/PAsp solutions (0.2–4 mg/mL, 10 mM NaCl and pH 5.0) for 5 h at room temperature, followed by water washes (3× 10 min). His-PAsp/PAsp was coated on the NH2 -SW surface via electrostatic interaction. The polymercoated slides were incubated with fibrinogen solution (0.2 mg/mL) in a phosphate buffer(10 mM phosphate, 150 mM NaCl, pH = 7.4) for 3 h on a shaker at 37 ◦ C. The protein concentration was measured by the Bradford method [32]. The amounts of adsorbed protein were determined by measuring the fibrinogen concentration before and after the adsorption process. The adsorption capacity of fibrinogen on the polymer-coated surface per area (q, ␮g/cm2 ) was calculated using the following Eq. (1) [33]:

2.2. Polymer synthesis q= Poly (␣,␤-l-aspartic acid), PAsp, was synthesized by hydrolysis of poly (succinimide) (PSI, M␩ = 29,700 g/mol, PDI = 1.46) using a previously reported method [30]. The zwitterionic polymer was prepared by direct conjugation of l-histidine methyl ester dihydrochloride, which was used in 10 equiv excess to the amount of carboxyl groups of PAsp. The reaction was activated by EDC. The solution was dialyzed against

(Ci − Cf)V S

(1)

where Ci and Cf are the initial protein concentration and the protein concentration in the supernatant after the adsorption, respectively. V is the total volume of the solution and S is the total surface area of the silicon wafers. The reported data were mean values of triplicate samples. The APTES-modified slides (NH2 -SW) without coating polymer were used as the control.

X. Wang et al. / Colloids and Surfaces B: Biointerfaces 86 (2011) 237–241

239

Scheme 1. The synthesis route of His-PAsp/PAsp.

3. Results and discussion 3.1. Synthesis and structural characterization l-Histidine-conjugated poly (␣,␤-l-aspartic acid) was synthesized by hydrolysis of polysuccinimide (PSI), followed by partial amidation of the carboxyl groups with methyl l-histidinate in the presence of EDC as a coupling reagent at room temperature. The synthetic route was showed in Scheme 1. In the 1 H NMR spectrum of His-PAsp/PAsp as shown in Fig. 1, new peaks appeared at 7.2 and 8.6 ppm attributed to the imidazole group of histidine, indicating the successful conjugation of histidine. A comparison of the peak integration corresponding to the methine proton of methyl l-histidinate (3.2 ppm) and the chain methine proton (4.5 ppm) of the PAsp backbone leads to the grafting ratio of 0.39. Because the amidation of soluble poly (␣,␤-l-aspartic acid) was performed on aqueous solution where the side groups on the polymeric chain were free to react, the resultant His-PAsp/PAsp might be considered as a random copolymer of aspartic acid and asparagine with a residue of l-histidine methyl ester. 3.2. Aqueous solution behavior of the zwitterionic His-PAsp/PAsp 3.2.1. Buffering capacity of the His-PAsp/PAsp The coexistence of both weak acidic and basic units along the polymer chain results in the possibility to form zwitterinic fragments at certain pH range. The pH titration profile of the His-

Fig. 1. The 1 H NMR spectrum of His-PAsp/PAsp in D2 O.

PAsp/PAsp solution was shown in Fig. 2. During the titration, the pH value of His-PAsp/PAsp was decreased gradually. In the range of pH = 9–4, His-PAsp/PAsp has a good buffing capacity ascribed to the combination of carboxylic acid of poly(aspartic acid) and imidazole group of histidine, whose pKa values is 4.01 and 6.04, respectively. It is noted that the zwitterionic polymer His-PAsp/PAsp was precipitated from solution near the isoelectric point (pI) owing to electrostatic interactions between the oppositely charged segments, which is a characteristic behavior of polyampholytes. It is observed two transition points corresponding to the beginning of the ionization of the imidazole group of histidine and the end of the protonation of the carboxylic group of the PAsp segment. 3.2.2. Phase-transition The pH-induced solubility transition of His-PAsp/PAsp was characterized by turbidity measurements. The initial solution was prepared by dissolving the polymer in deionized water, and then the pH of solution was adjusted by adding 0.1 M HCl or 0.1 M NaOH. The dependence of transmittance tendency as a function of pH at  = 380 nm was given in Fig. 3. Results indicated that the pH of the solution played an important role in the polyelectrolyte aggregation. At different pH values, the ionization degree and solubility of side groups are different, which may induce aggregation or disaggregation of the copolymer. At very low pH (pH = 2), the transmittance was 95% so that the polymer His-PAsp/PAsp could be considered as a polycation as it possessed the uncharged carboxylic groups (–COOH) and electro-positively protonated imidazole groups. The repulsive forces between the charged imidazole groups lead to a chain

Fig. 2. The pH titration curves of His-PAsp/PAsp (concentration of 1 mg/mL, 25 ◦ C).

240

X. Wang et al. / Colloids and Surfaces B: Biointerfaces 86 (2011) 237–241

Fig. 3. Transmittance of His-PAsp/PAsp aqueous solution at different pHs.

stretching, and make the polymer dissolved molecularly in aqueous solution. A sharp decrease of transmittance was observed in the range of pH 2–5. Because the pKa of PAsp is 4.01, a significant amount of –COOH groups should be ionized at pH 4. The formation of large aggregates at pH 4 should be leaded by the electrostatic interactions between the electro-positive imidazole segment and electro-negative carboxylic segment. If the pH values were high over 5, the polymer His-PAsp/PAsp was again soluble and the transmittance was close to 95%. Although the imidazole groups were deprotonated and rendered hydrophobic, the repulsion between the ionized carboxylic groups could also make the polymer soluble and keep the aqueous solution transparent. As mentioned above, His-PAsp/PAsp showed the typical characteristic expected for a polyampholyte: the solution was almost transparent at either low or high pH, and had a minimum transmittance at a special pH, which can be defined as the isoelectric point (pI). At this pH, electrostatic attraction lead to relatively tight aggregations of the polymer and made it dispersed as an emulsion. Because of the precipitation, it is difficult to clearly observe the pI by the titration measurements. The pI of the polyampholyte could be estimated as the midpoint of the precipitation range between pH 3.5 and 4. 3.3. Electrophoretic mobility and the determination of isoelectric point Isoelectric point (pI) is an important characteristic of the aqueous solution of polyampholytes. To detect the electrophoretic mobility and locate the pI, the -potential was monitored under the same condition as those in the turbidity measurements on His-PAsp/PAsp solution at various pH values. At a pH much lower than the pI, the solution has a net positive charge. At the pI, the net charge is zero. And it turns to a net negative charge at a pH values much higher than the pI. As shown in Fig. 4 the zeta potential () decreased from +18 mV (pH = 2.0) to −44 mV (pH = 8) as the pH increased. These zeta potential measurements provided the direct information on the charge density of the solution and electrostatic interactions among charged segments. Theoretically, a polymer solution would be unstable and aggregate when the || < 30 mV [34]. There were enough protonated imidazole and deprotonated carboxylic groups coexisted, where the || was closed to zero near the pI. Phase transition was induced by their electrostatic attraction. The pI determined by the electrophoretic mobility of the polymer in aqueous solution was 3.6 close to the theoretical

Fig. 4. The -potential of His-PAsp/PAsp aqueous solution at different pH (no salt added).

value 3.9 calculated by the formula (2) [35].

   

IEP = pKa − log

R 2

−(1 − R) + R



1−R R

2   +

4 R

 pKa −pKb

× 10

(2)

where R is the acid/base molar ratio and pKa and pKb are dissociation constants for the negative and positive charges, respectively. In this case, pKa = 4.01, pKb = 6.04, R = 61/39. 3.4. Protein adsorption The most important proteins promote cell adhesion on biomaterials include fibrinogen, von-Willebrand-factor, collagen and vitronectin. Fibrinogen is abundant in plasma and well studied in the context of blood compatibility [3]. Thus fibrinogen as a model protein is often used to investigate cell-adhesion and blood compatibility. It is well known that protein adsorption is strongly influenced by the surface properties of biomaterials. Recently, surface modification via adsorption of polyampholytes has evoked great interest and resulted in a completely different behavior compared with the unmodified substrates. The zwitterionic polymer was firstly pre-coated on the positive silica surface for the purpose of evaluating its protein-resistant characteristics. These APTES-treated silicon wafers served as water-insoluble platforms carrying positive charges at pH 5, when His-PAsp/PAsp bore negative charges. The polymer His-PAsp/PAsp could be coated on the wafers due to the electrostatic attraction. The fibrinogen adsorption measurements were carried out in 10 mM PBS buffer (150 mM NaCl, pH 7.4) to maintain a constant pH value during the adsorption process. The amount of fibrinogen adsorbed to each sample was quantified and compared with those on the NH2 -SWs without polymer coating. In compared with the unmodified surface, the protein amount adsorbed on the His-PAsp/PAsp coated surfaces dropped dramatically. It is the zwitterionic polypeptide to reduce the protein adsorption in a dose-dependent manner. As shown in Fig. 5, the amine-functionalized silica wafer had the most serious fibrinogen adsorption probably due to abundantly positive charges of the NH2 -SW substrate, while the His-PAsp/PAsp coated wafer exhibited very low adsorption of fibrinogen. It was suggested that the fibrinogen adsorption depended on the surface charge. Under the adsorption condition, His-PAsp/PAsp acted as a polyanion, thus this pre-adsorbed polymer switched the net charge of adsorption surface at pH 7.4 from electro-positivity to electro-negativity. Meanwhile, fibrinogen with a pI of 5.5 is negatively charged at pH 7.4. The repulsively electrostatic interac-

X. Wang et al. / Colloids and Surfaces B: Biointerfaces 86 (2011) 237–241

241

polypeptide derivative as a surface modifying material possesses protein repulsion ability with biocompatibility and biodegradability so as to make it applicable in the field of biomedical sciences. Acknowledgments This work was funded by the National Key Basic Research Development Program of China (973 Program) (No. 2009CB626612), the Natural Science Foundation of Tianjin (Nos. 09JCYBJC03400, 10JCYBJC26800), the Specialized Research Fund for the Doctoral Program of Higher Education (No. 20090031120012). References

Fig. 5. The adsorption amount of fibrinogen on the amine-functionalized and HisPAsp/PAsp -coated silica wafers.

tion between precoated silicon wafer and the negative fibrinogen resisted its adsorption. It demonstrated an important influence of the precoated polymer on resistance of protein adsorption. The electrostatic attraction between positive APTES-modified silicon wafers and the negative fibrinogen made the NH2 -SW adsorbing a lot of fibrinogen from its solution. Beside the electrostatic interaction, the hydrophilic effect of fibrinogen adsorption on the substrate should be inhibited also with the hydrophilic coating of His-PAsp/PAsp. It is well known that PAsp is well soluble in water, so the polymer His-PAsp/PAsp pre-coated on the wafer should be highly hydrated. Such a hydrophilic surface would decrease the adsorption amount of fibrinogen. The pI or the degree of histidine functionalization plays a crucial role in their antifouling abilities to fibrinogen as a model of serum proteins. The highest substitution degree of l-histidine sideresidues was 39% as we synthesized in the given conditions. The l-histidine-modified poly (␣,␤-l-aspartic acid) with 39% of substitution degree was used as the sample in this study. Samples with lower DS around 10% did not show the typical characteristic behavior of the polyampholytes. The reason should be that there were too few basic groups along the polymer to balance the acid groups, so the polymer with low substitution degree acted like a polyanion. 4. Conclusion Non-specific protein adsorption often induces biological incompatibility. The anti-biofouling behavior of a material surface through electrostatic interactions was affected by the surface charge density. Fibrinogen adsorption on wafer surfaces was inhibited effectively by His-PAsp/PAsp polymer coating. Such a result implied that a zwitterionic polypeptide coating would be an alternative strategy to circumvent this problem. The zwitterionic

[1] P. Aggarwal, J.B. Hall, C.B. McLel, M.A. Dobrovolskaia, S.E. McNeil, Adv. Drug Deliv. Rev. 61 (2009) 428. [2] D.G. Castner, B.D. Ratner, Surf. Sci. 500 (2002) 28. [3] C. Sperling, M. Fischer, M.F. Maitz, C. Werner, Biomaterials 30 (2009) 4447. [4] Y. Zhao, M. Haney, N. Klyachko, S. Li, S. Booth, S. Higginbotham, J. Jones, M. Zimmerman, R. Mosley, A. Kabanov, Nanomedicine 6 (2011) 25. [5] M. Garcia-Fuentes, D. Torres, M. Alonso, Int. J. Pharm. 296 (2005) 122. [6] B. Thierry, L. Zimmer, S. McNiven, K. Finnie, C. Barbeˇı, H.J. Griesser, Langmuir 24 (2008) 8143. [7] C. Rodriguez-Emmenegger, A. Jäger, E. Jäger, P. Stepanek, A.B. Alles, S.S. Guterres, A.R. Pohlmann, E. Brynda, Colloids Surf. B: Biointerfaces 83 (2011) 376. [8] E.C. Cho, L. Au, Q. Zhang, Y. Xia, Small 6 (2010) 517. [9] C. Geer, I. Rus, S. Lord, M. Schoenfisch, Acta Biomater. 3 (2007) 663. [10] I. Lynch, A. Salvati, K.A. Dawson, Nat. Nano 4 (2009) 546. [11] A.E. Nel, L. Madler, D. Velegol, T. Xia, E.M.V. Hoek, P. Somasundaran, F. Klaessig, V. Castranova, M. Thompson, Nat. Mater. 8 (2009) 543. [12] C.M. Magin, S.P. Cooper, A.B. Brennan, Mater. Today 13 (2010) 36. [13] Q.A. Wei, B.J. Li, N. Yi, B.H. Su, Z.H. Yin, F.L. Zhang, J. Li, C.S. Zhao, J. Biomed. Mater. Res. A 96A (2011) 38. [14] Y.Q. Zou, P.Y.J. Yeh, N.A.A. Rossi, D.E. Brooks, J.N. Kizhakkedathu, Biomacromolecules 11 (2010) 284. [15] S. Burkert, E. Bittrich, M. Kuntzsch, M. Muller, K.J. Eichhorn, C. Bellmann, P. Uhlmann, M. Stamm, Langmuir 26 (2010) 1786. [16] R. Gref, M. Lück, P. Quellec, M. Marchand, E. Dellacherie, S. Harnisch, T. Blunk, R.H. Müller, Colloids Surf. B: Biointerfaces 18 (2000) 301. [17] D.E. Owens Iii, N.A. Peppas, Int. J. Pharm. 307 (2006) 93. [18] Y.Y. Luk, M. Kato, M. Mrksich, Langmuir 16 (2000) 9604. [19] N.V. Efremova, S.R. Sheth, D.E. Leckband, Langmuir 17 (2001) 7628. [20] B. Romberg, W. Hennink, G. Storm, Pharm. Res. 25 (2008) 55. [21] H. Hatakeyama, H. Akita, K. Kogure, M. Oishi, Y. Nagasaki, Y. Kihira, M. Ueno, H. Kobayashi, H. Kikuchi, H. Harashima, Gene Ther. 14 (2006) 68. [22] J. Ladd, Z. Zhang, S. Chen, J.C. Hower, S. Jiang, Biomacromolecules 9 (2008) 1357. [23] Z.G. Estephan, J.A. Jaber, J.B. Schlenoff, Langmuir 26 (2010) 16884. [24] Q. Jin, J.P. Xu, J. Ji, J.C. Shen, Chem. Commun. (2008) 3058. [25] G. Cheng, G.Z. Li, H. Xue, S.F. Chen, J.D. Bryers, S.Y. Jiang, Biomaterials 30 (2009) 5234. [26] S.F. Chen, Z.Q. Cao, S.Y. Jiang, Biomaterials 30 (2009) 5892. [27] P.S. Xu, G. Bajaj, T. Shugg, W.G. Van Alstine, Y. Yeo, Biomacromolecules 11 (2010) 2352. [28] J. Rodríguez-Hernández, S. Lecommandoux, J. Am. Chem. Soc. 127 (2005) 2026. [29] J. Li, T. Wang, D. Wu, X. Zhang, J. Yan, S. Du, Y. Guo, J. Wang, A. Zhang, Biomacromolecules 9 (2008) 2670. [30] Y. Wang, Y. Wang, G. Wu, Y. Fan, J. Ma, Colloids Surf. B: Biointerfaces 68 (2009) 13. [31] C.J. Ochs, G.K. Such, B. Stadler, F. Caruso, Biomacromolecules 9 (2008) 3389. [32] M.M. Bradford, Anal. Biochem. 72 (1976) 248. [33] J.N. Zheng, H.G. Xie, W.T. Yu, X.D. Liu, W.Y. Xie, J. Zhu, X.J. Ma, Langmuir 26 (2010) 17156. [34] B. Peng, Y. Hao, H. Kang, X. Han, C. Peng, H. Liu, Carbohydr. Res. 345 (2010) 101. [35] C.S. Patrickios, W.R. Hertler, N.L. Abbott, T.A. Hatton, Macromolecules 27 (1994) 930.