- Email: [email protected]
A facile route to the synthesis of spherical poly(acrylic acid) brushes via RAFT polymerization for high-capacity protein immobilization
Journal of Colloid and Interface Science 398 (2013) 82–87
Contents lists available at SciVerse ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
A facile route to the synthesis of spherical poly(acrylic acid) brushes via RAFT polymerization for high-capacity protein immobilization Zhenyuan Qu, Fenglin Hu, Kaimin Chen, Zongqiang Duan, Hongchen Gu, Hong Xu ⇑ School of Biomedical Engineering, Med-X Research Institute, Shanghai Jiao Tong University, Shanghai 200030, PR China
a r t i c l e
i n f o
Article history: Received 13 December 2012 Accepted 1 February 2013 Available online 11 February 2013 Keywords: Spherical brushes Poly(acrylic acid) (PAA) RAFT polymerization Protein immobilization
a b s t r a c t Spherical poly(acrylic acid) brushes were prepared via a facile RAFT polymerization route from silica nanoparticles (SiO2@PAAs). A silane functionalized RAFT chain transfer agent was designed and synthesized by a one-step reaction, and then immobilized onto silica nanoparticles (SiNPs) through its R group to afford RAFT polymerization. Key structural parameters and contents of carboxyl groups of SiO2@PAAs were thoroughly characterized by transmission electron microscopy, dynamic light scattering, gel permeation chromatography, thermogravimetric analysis and conductometric titration. The SiO2@PAAs exhibit excellent dispersity, tunable brush thicknesses (14.6–68.8 nm) and abundant carboxyl groups (0.82– 2.37 mmol/g). An ultra-high protein immobilization capacity (2600 lg streptavidin/mg SiO2@PAAs) was realized by virtue of its rich carboxyl groups and spherical brush structure, which opens up new possibilities for biomedical applications. Ó 2013 Elsevier Inc. All rights reserved.
1. Introduction Poly(acrylic acid) (PAA) brushes (either planar or spherical) have aroused great interest in the past decade due to its simple structure, abundant content of carboxyl groups and response to the environment stimuli [1]. These characteristics have resulted in wide applications of PAA brushes in biomedical fields including protein adsorption [2–4], immobilization of biomolecules [5–8], controlled drug release [9] and so on. Among them, spherical PAA brushes (SPAABs) in which PAA chains are attached to particles have emerged as an attractive structure [10]. On the one hand, the particles serve as ideal carriers to realize an effective separation of biomolecules or targeting to organism for both in vitro and in vivo studies. On the other hand, the PAA corona endows the particles with excellent dispersity while maintaining the above-mentioned general characteristics of PAA brushes. Various techniques have been attempted to fabricate the PAA brush layer. Guo et al. [10] firstly reported the synthesis of SPAABs via photoemulsion polymerization. The uncontrolled polymerization process, however, leads to an ill-defined structure [1,10]. A precise control over the brush structure is not available until the advent of living radical polymerization (LRP) at the end of the last century, which gradually becomes the predominant method to fabricate PAA brushes or brushes with PAA block. Among various LRP techniques, atom transfer radical polymerization (ATRP) has been widely adopted [5–8,11,12]. Nevertheless, as ATRP is not compati⇑ Corresponding author. Fax: +86 21 62932907. E-mail address: [email protected] (H. Xu). 0021-9797/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcis.2013.02.001
ble with monomers containing carboxyl groups, a two-step process is usually employed, i.e., polymerization of protected monomer and deprotection by hydrolysis [5,7,8,11] or pyrolysis [6,12]. Reversible addition–fragmentation chain transfer (RAFT) polymerization is another representative LRP technique that is very suitable for the synthesis of PAA brushes owing to its tolerance to the acrylic acid (AA) monomer. However, up to date, RAFT polymerization is much less adopted regarding the fabrication of PAA brushes in comparison with ATRP. Difficulties in the synthesis and/or immobilization of RAFT initiator [13], together with relatively low grafting density of the brushes, has been recognized as major factors resulting in the limitation of RAFT polymerization [14]. To conduct a surface RAFT polymerization, harsh reaction conditions (e.g. Grignard reaction) [15] or a multi-step derivation [9,14,16] is often necessary to immobilize the RAFT chain transfer agent (RAFT CTA), which can be tedious, complicated, and even demanding to some materialists without solid expertise in organic chemistry. It is, then, of significant importance to develop novel RAFT polymerization process with simplified synthetic route to the synthesis of SPAABs. Among various spherical solid carriers, silica nanoparticles (SiNPs) are widely used in biomedical fields due to its easy synthesis and modification, stable chemical properties and excellent biocompatibility [17]. SiNPs also serve as a robust and versatile shell for various nanoparticles [18,19] and can be easily extended to mesoporous silica [20,21], which will endow the material with more functionalities for application. SiNPs-based SPAABs, therefore, act as an important model material for both fundamental studies and biomedical applications.
Z. Qu et al. / Journal of Colloid and Interface Science 398 (2013) 82–87
In this paper we report a facile RAFT polymerization route to the synthesis of SPAABs from SiNPs with well-defined structure (denoted as SiO2@PAAs). A novel RAFT CTA with silane functional groups was synthesized by a one-step reaction. This RAFT CTA enables a simple immobilization via silanization and the subsequent RAFT polymerization from its R group to form SiO2@PAAs. Key parameters of SiO2@PAAs were determined and the control of RAFT process over brush thicknesses was performed. We further show the advantages of SiO2@PAAs in the improvement of protein immobilization capacity.
2. Materials and methods 2.1. Materials (4-(chloro)phenyl)trimethoxysilane was purchased from Gelest Inc., N-(3-dimethylamino-propyl)-N0 -ethylcarbodiimide (EDC) and ethanethiol were obtained from J&K chemical. N-hydroxysuccinimide (NHS) and bicinchoninic acid (BCA) reagent kit was obtained from Thermo Scientific. Streptavidin(SA) was obtained from Shanghai Yemin biotechnology Co., Ltd., biotinylated horseradish peroxidase (biotin-HRP) from Beijing biosynthesis biotechnology Co., Ltd., and biotin from Sangon biotech (Shanghai) Co., Ltd., 3,30 ,5,50 -Tetramethylbenzidine was obtained from Sigma. Acrylic acid (AA) was distilled under reduced pressure and stored in 20 °C prior to use. 2,20 -Azobisisobutyronitrile (AIBN) was recrystallized from ethanol. All the other reagents were the products of China National Medicines Group Shanghai Chemical Reagents Company and used without further purification.
2.2. Synthesis of RAFT CTA The silane functionalized RAFT CTA, ethyl 4-(trimethoxysilyl)benzyl carbonotrithioate (1), was synthesized as follows. Ethanethiol (487 lL, 6.6 mmol) was charged into a stirred suspension of K3PO4 (1.02 g, 6.6 mmol) in anhydrous acetone (15 mL) and stirring for half an hour. CS2 (1.1 mL, 18 mmol) was added and the solution turned bright yellow. After stirring for another 10 min, (4-(chloromethyl)phenyl)-trimethoxysilane (1.43 mL, 6.6 mmol) was added and the mixture was then stirred at ambient temperature in nitrogen atmosphere for 13 h. The mixture was concentrated, diluted with dichloromethane and filtered off. After removing the solvent from the filtrate under reduced pressure the resulting yellow residue was purified by column chromatography on silica gel using a petroleum ether/ethyl acetate gradient to yield a bright yellow oil 1 (73%). 1 H NMR (400 MHz, CDCl3) d 7.60 (d, J = 7.9 Hz, 2H), 7.37 (d, J = 7.8 Hz, 2H), 4.62 (s, 2H), 3.61 (s, 9H), 3.38 (q, J = 7.4 Hz, 2H), 1.36 (t, J = 7.4 Hz, 3H). 13C NMR (CDCl3) d 223.37, 137.90, 135.10, 128.80, 128.06, 50.86, 41.13, 31.70, 13.04.m/z (EI) 348.1, 211.2. IR (cm1): 3070, 3016, 2967, 2941, 2871, 2840, 1602, 1454, 1399, 1191, 1124, 1082, 1030, 876, 809, 726. The free RAFT CTA, benzyl dodecyl carbonotrithioate (2), was prepared in a similar procedure to that of the silane functionalized RAFT CTA. Specifically, dodecanethiol (2.4 mL, 9.9 mmol) was charged to a stirred suspension of K3PO4H2O(2.64 g, 9.9 mmol) in acetone (30 mL) and stirring for half an hour. CS2 (1.65 mL, 27 mmol) was added and the solution turned bright yellow. After stirring for another 10 min, benzyl bromide (1.18 mL, 9.9 mmol) was added and the mixture was stirred overnight at ambient temperature. The mixture was concentrated and filtered off. After removing the solvent from the filtrate under reduced pressure, Compound (2) was crystallized in refrigeration and obtained as a yellow solid product with 92% isolated yield.
83
1
H NMR (400 MHz, CDCl3) d 7.37–7.26 (m, 5H), 4.62 (s, 2H), 3.38 (t, J = 7.4 Hz, 2H), 1.76–1.17 (m, 20H), 0.89 (t, J = 6.7 Hz, 3H). 13C NMR (CDCl3) d 224.06, 135.55, 129.85, 128.40, 128.07, 54.72, 41.40, 37.11, 31.99, 29.71, 29.63, 29.52, 29.42, 29.18, 28.99, 28.04, 22.77, 14.21. IR (cm1): 2954, 2917, 2852, 1494, 1464, 1454, 1063, 826, 880, 809, 698. 2.3. Synthesis of RAFT CTA functionalized SiNPs (SiO2@RAFT) SiNPs with a diameter of 80 nm (TEM image was shown in Fig. S1 in supporting information) were prepared by the Stöber procedure [22]. To modify SiNPs with RAFT CTA, compound 1 (412 lL) was added into 70 mL suspension of SiNPs in absolute ethanol (containing 3 g of SiNPs). The mixture was refluxed under nitrogen for 24 h. The product was washed with ethanol and obtained as yellow particles suspended in ethanol. The grafting density of RAFT CTA can be quantitatively determined by comparing the intensity of 308-nm absorption of SiO2@RAFT to the standard curve of silane functionalized RAFT CTA measured in ethanol [15] (e308 = 12,751 L mol1 cm1). 2.4. Synthesis of SiO2@PAAs SiO2@RAFT was transferred into N,N-dimethylformamide (DMF) by centrifugation and resuspension. Then, AA, AIBN and free RAFT CTA, 2, were added and the mixture was transferred into a schlenk tube. After three cycles of freeze–pump–thaw, the system was closed and placed into an oil bath thermostat at 70 °C. The mixture becomes sticky as the polymerization proceeds. The product was diluted by DMF and centrifuged to obtain SiO2@PAAs, which were then extensively washed by ethanol to fully remove untethered polymers and unreacted chemicals. In a typical run, the final concentration of AA, AIBN, free RAFT CTA and SiNPs were 4.5 M, 3.75 mM, 29.7 mM and 15 mg/mL, respectively. The polymerization was allowed to proceed in DMF for 3 h. Three SiO2@PAAs (denoted as SiO2@PAAs75, SiO2@PAAs150, SiO2@PAAs300) with various feed ratios of AA to free RAFT CTA (75, 150 and 300, respectively) were obtained. 2.5. Immobilization of SA on SiO2@PAAs SA was immobilized on the SiO2@PAAs by a two-step NHS/EDC coupling procedure. 1 mg of SiO2@PAAs300 was firstly activated by NHS (25 mg/mL) and EDC (50 mg/mL) in 0.4 mL of 10 mM phosphate buffer containing 0.05% (w/v) tween-20 (PB-T) for 15 min. After removing the excess reactants, SA was added to allow a 2 h incubation. The binding capacity of SA on SiO2@PAAs, corresponding to the difference of SA concentration in solution, was quantified by the determination of protein concentration before and after conjugation using BCA method. The SiO2@PAAs-SA complexes were blocked with 0.2% (w/v) BSA and 0.5% glycine overnight at 4 °C. Finally, the resulting SiO2@PAAs-SA complexes were stored in PB-T buffer containing 0.1% BSA. 2.6. Determination of binding capacity of SiO2@PAAs-SA complexes with biotin The binding capacity between SiO2@PAAs-SA complexes and biotin was measured by enzyme competitive inhibition method [23]. The SiO2@PAAs-SA complexes (40 lg) was mixed with a gradient amount of biotin for 10 min at 37 °C. After centrifugation and discarding the supernatant, 100 lL of diluted biotinylated horseradish peroxidase solution was added to the mixture for another 10 min incubation at 37 °C. The resulting mixture was washed five times with PB-T buffer and then incubated with 100 lL of substrate buffer (containing TMB) under shaken for 10 min at room temperature. The
84
Z. Qu et al. / Journal of Colloid and Interface Science 398 (2013) 82–87
reaction was ended by adding 25 lL of 2 M H2SO4 solution. After centrifugation, 100 lL of supernatant was transferred to a micro-well and measured at the wavelength of 450 nm by a multifunctional microplate reader. The saturation binding capacity of SiO2@PAAs-SA complexes with biotin corresponds to the point where the optical density (OD) drops to the lower plateau. 2.7. Characterization The monomer conversions were obtained by gravimetry. Number average molecular weight (Mn) and polydispersity index (PDI) of the polymers cleaved from the surface of SiNPs using hydrofluoric acid (HF) [15] were determined by gel permeation chromatography (GPC) in an equipment comprising a Waters 1525 binary HPLC pump, a 2707 auto sampler, and a 2414 refractive index detector. The columns are an Ultrahydrogel™ guard column (porosity 200 Å, particle size 6 lm, dimension 6 mm 40 mm), followed by three Ultrahydrogel™ columns (porosity 1000 Å, 250 Å, and 120 Å, particle size 12 lm for 1000 Å, and 6 lm for others, dimension 7.8 mm 300 mm). Phosphate buffer (0.1 M, pH = 7.2) with 0.3 M NaCl were used as the eluent, and calibration was done by sodium polyacrylate (PAANa) standards (purchased from American Polymer Standards Corporation) with a flow rate of 1.0 mL/min and injection volume of 50 lL at 35 °C. Thermogravimetric analysis (TGA) was performed on a Netzsch (TG 209 F1 Iris) thermogravimetric analyzer in air at a heating rate of 10 °C/min. The conductometric titration was done by a Mettler Toledo T50 autotitrator using 0.1 M NaOH as titrant. Transmission electron microscope (TEM) images were obtained by a JEM 2010 (JEOL, Japan) instrument operating at 200 kV. UV–Vis spectrum was recorded by a Unicam UV300 UV–Vis spectrometer. The quantification of SA concentration and the determination of free biotin binding capacity were carried out using a microplate protocol. The OD was read by a PerkinElmer 1420 Multilabel
Counter. The hydrodynamic diameter was determined by dynamic light scattering (DLS) using a Malvern Nano ZS particle size analyzer. 1H NMR (400 MHz) spectra were recorded on a Bruker 400 UltraShield spectrometer at 25 °C using CDCl3 as a solvent. Mass spectrometry (MS) was performed on a Pegasus 4D GCxGC-TOFMS mass spectrometer. Fourier transform infrared spectra were recorded using KBr pellet method on a Nicolet 6700 FT-IR spectrometer. 3. Results and discussion 3.1. Synthesis of SiO2@PAAs via a new RAFT agent with a siloxane end The synthetic route towards SiO2@PAAs is illustrated in Scheme 1 with the following steps. (I) A novel RAFT CTA with silane functional groups, 1, was synthesized in one step for the first time. (II) The silane functionalized RAFT CTA was immobilized onto the SiNPs by a silanization reaction. The reaction was done by refluxing in ethanol, which is a good dispersant for both SiNPs and SiO2@RAFT. (III) RAFT polymerization was conducted to form the PAA brushes. Free RAFT CTA, 2, was added in solution to tune the brush thicknesses of the brushes. Three SiO2@PAA samples with different feed ratios of AA to RAFT CTA were prepared. In the present work, we design a novel RAFT CTA with a silane functionalized benzyl groups as its R group and an ethyl group as its Z group. Using this agent, the immobilization of RAFT CTA by ‘‘R-group approach’’ is simplified as an efficient and robust silanization reaction. The Z group is selected to be a tiny ethyl group to minimize the steric hindrance in the immobilization of RAFT CTA. Different from previous works that involve a multi-step derivation [9,14,16] or reactions with stringent conditions [15] to form R-group immobilized RAFT CTA, the present route clearly simplifies the synthetic procedure with the use of milder reaction conditions. It has been recognized that the ‘‘R-group approach’’
Scheme 1. Synthetic route to SiO2@PAAs.
Z. Qu et al. / Journal of Colloid and Interface Science 398 (2013) 82–87
Fig. 1. UV–Vis specra of RAFT CTA, bare SiNPs and SiO2@RAFT measured in ethanol. The characteristic peak at 308 nm was observed for both RAFT CTA and SiO2@RAFT, which can be further utilized to determine the grafting density of RAFT CTA on SiNPs.
Table 1 Surface RAFT polymerization to fabricate SiO2@PAAs with different feed ratios of monomer to RAFT CTA. Sample
SiO2@PAAs75
SiO2@PAAs150
SiO2@PAAs300
[AA]/[RAFT] Conversion (%) Mn (kD) Mn(theo) (kD) PDI
75 80 4.5 4.7 1.13
150 84 10.2 9.4 1.15
300 77 18.4 17.0 1.48
resembles the ‘‘graft-from’’ approach. Thus a relatively high grafting density can be accomplished due to a reduced steric hindrance [14], which consequently contributes to a higher content of carboxyl groups of SiO2@PAAs for protein immobilization. The successful immobilization of RAFT CTA was demonstrated by UV–Vis spectrometry (Fig. 1), in which a characteristic peak centered at 308 nm was found in both RAFT CTA and SiO2@RAFT, while the same peak was absent in bare SiNPs. After silanization in ethanol, a good dispersant for both SiNPs and SiO2@RAFT, the SiO2@RAFT was obtained as yellow monodisperse particles (PDI = 0.040) with roughly identical hydrodynamic diameter to SiNPs. The excellent dispersity of SiO2@RAFT is not only important to obtain the resulting SiO2@PAAs with a well-defined structure,
85
Fig. 3. TGA curves of SiNPs, SiO2@RAFT, and SiO2@PAAs.
but also a prerequisite for the quantification of RAFT CTA grafting density using UV–Vis spectrometry. The grafting density of RAFT CTA was quantified to be 0.40 CTA/nm2, which is in accordance with the previously reported value [15]. The concentration of RAFT CTA is usually maintained to be much higher than that of radical initiators for a well-controlled RAFT polymerization. For surface RAFT polymerization, the addition of free RAFT CTA is a useful approach to help control the polymerization process when the concentration of surface-bound RAFT CTA is insufficient [16,24]. A rough estimation reveals that in our systems, surface-bound RAFT CTA (about only 0.3 mM with the particle concentration of 15 mg/mL) accounts for a very small proportion in total RAFT CTA due to a relatively low specific surface area of 80 nm-SiNPs made by stöber method. Hence the polymerization is mainly controlled by the free RAFT CTA added in solution. SiO2@PAAs with different molecular weights can be obtained by changing the amount of free RAFT CTA as PAAs grow synchronously on the surface and in solution. As shown in Table 1, the molecular weights of the grafted polymers are close to theoretical prediction, demonstrating a good control of RAFT process over molecular weights of PAA brushes. The low PDIs of the grafted polymers for SiO2@PAAs75 and SiO2@PAAs150 also indicate that the polymerizations are well controlled, generating uniform PAA brushes. On the other hand, the PDI for SiO2@PAAs300 increases to 1.48. The reason has not been clear yet.
Fig. 2. (a) Hydrodynamic diameters of SiNPs and SiO2@PAAs in deionized water determined by DLS. (b) Hydrodynamic diameters (based on cumulant method) of SiO2@PAAs in water as a function of pH.
86
Z. Qu et al. / Journal of Colloid and Interface Science 398 (2013) 82–87
Table 2 Structural parameters and the content of carboxyl groups of SiO2@PAAs with different feed ratios of monomer to RAFT CTA. Sample
SiO2@PAAs75
SiO2@PAAs150
SiO2@PAAs300
DSiNPsa L (nm)b Gr (g PAA/g SiO2)c r (nm2)d COOH (mmol/g)e
80 14.6 0.084 0.34 0.82
80 45.6 0.161 0.28 1.56
80 68.8 0.242 0.23 2.37
a Size of the spherical carrier (DSiNPs) was determined by TEM (see Fig. S1 in supporting information). b PAA brush thicknesses (L) = 12 ðdSiO2 @PAAs dSiNPs Þ where dSiO2 @PAAs and dSiNPs are the hydrodynamic diameters of fully-stretched SiO2@PAAs (at pH above 8) and SiNPs determined by dynamic light scattering (DLS), respectively. c Mass ratio of the grafted polymer to the SiNPs (Gr) estimated from thermogravimetric analysis (TGA). d Grafting density of PAA corona (r) was calculated in combination of Mn and Gr. The calculation method was described in Supporting information. e The titration background of SiO2@RAFT, which probably results from the residue silanol groups, was subtracted.
3.2. Characterization of SiO2@PAAs with controlled brush thicknesses The regular structure makes spherical brushes an ideal model material for the study concerning the relationship between its structure and performance. The key structural parameters including the size of core, brush thicknesses and the grafting density of the brushes can be tuned and explicitly determined by a variety of methods [25]. Fig. 2a shows the hydrodynamic diameters of SiO2@PAAs and the bare SiNPs. All the three SiO2@PAAs have an increased hydrodynamic diameter compared with bare SiNPs, indicating the formation of PAA corona, whose thicknesses increase with the increase of feed ratios of monomer to RAFT CTA. Fig. 2b shows the change of hydrodynamic diameters of SiO2@PAAs as a function of pH. The PAA corona collapses at a pH around its pKa and reaches a fully-stretched conformation in slightly basic conditions, which is in conformity with the typical behavior of PAA
brushes in aqueous solution [26]. It is also worth mentioning that the monodispersity of all SiO2@PAAs are maintained at pH > 3 (PDI < 0.08) due to both steric and electrostatic stabilization effect of PAA corona to nanoparticles. The TGA curves of the materials are shown in Fig. 3. An additional weight loss was observed for SiO2@PAAs in comparison with SiO2@RAFT and the weight loss also increases with the increase of feed ratios. In combination of the results obtained by DLS, TGA and GPC, together with TEM, the structural parameters of core size, brush thicknesses and grafting density of the brushes can be determined, which are summarized in Table 2. It can be seen that as the feed ratios of monomer to RAFT CTA increase, a higher amount of polymers can be grafted to SiNPs, and SiO2@PAAs with larger brush thicknesses and lower density will be obtained. The PAA grafting density for SiO2@PAAs and the aforementioned grafting density of RAFT CTA for SiO2@RAFT suggest that a proportion of about 57–85% surface-bound RAFT CTA kept active in the process of RAFT polymerization. The results of conductometric titration (Table 2) further confirm that the brushes have abundant contents of carboxyl groups. The results are in well agreement with TGA and the contents of carboxyl groups of SiO2@PAAs are at least one order of magnitude higher than that achievable by conventional monolayer functionalization method [27]. 3.3. High-capacity immobilization of streptavidin (SA) The SPAABs with an exceptionally rich content of carboxyl groups provide abundant sites for protein immobilization. Here we made a preliminary demonstration on the ultra-high protein binding capacity of SiO2@PAAs obtained in this work. SA is chosen as the model protein as the specific recognition between SA and biotin (and also biotinylated biomolecules) has found wide applications in biotechnology [28]. SA was immobilized onto SiO2@PAAs300 by means of NHS/EDC coupling chemistry. And the resulting SiO2@PAAs-SA complex
Fig. 4. (a) Schematic illustration of the protein immobilization process and the binding of biotin to SA. (b) Amount of SA found (lg/mg) on particles and in dependence of total SA supplied. (c) Binding capacity of biotin to SiO2@PAAs-SA complexes determined by enzyme competitive inhibition method.
Z. Qu et al. / Journal of Colloid and Interface Science 398 (2013) 82–87
has the ability to further capture biotin or biotinylated biomolecules. The process is schematically illustrated in Fig. 4a. Fig. 4b depicts the binding capacity of SA to SiO2@PAAs300 with increasing amounts of SA added. A saturation point was observed, indicating that SiO2@PAAs300 has a binding capacity of ca. 2600 lg SA/mg. This value is about 60-fold higher than that achieved by conventional monolayer functionalization method [27]. It is also higher than that has been accomplished by Cullen et al. [6] with planar PAA brushes, in which a 16-fold improvement are reported utilizing the same conjugation process. At the same time, the binding efficiency of SA to per unit carboxyl group is also increased by two times in comparison with the monolayer modification structure [27]. The considerably rich carboxyl content of the materials brought about by the PAA brushes clearly accounts for the high binding capacity. And it is also probable that the structure of the spherical brush, with an increasing accessibility in its outer layers for the bulky protein molecules, contributes to the enhanced protein binding efficiency. Detailed investigation on the binding mechanism is in progress. The biological activity of the surface-bound SA was tested by the enzyme competitive inhibition method (Fig. 4c) [23]. The binding capacity of biotin was about 91 nmol/mg, corresponding to ca. 2 biotins per SA, which is in consistency with the previously reported value [27]. It demonstrates that the biological activity of SA keeps intact after conjugation.
4. Conclusion In summary, a facile and efficient route to the synthesis of SiO2@PAAs via RAFT polymerization was developed. The RAFT process provides a good control over molecular weights and thicknesses of the PAA brushes. The resulting SiO2@PAAs systems have a tunable brush thicknesses of 14.6–68.8 nm, PAA grafting density of 0.23–0.34 nm2 with a rich carboxyl group content of 0.82–2.37 mmol/g. The protein binding capacity to the brushes was substantially improved to 2600 lg SA/mg SiO2@PAAs owing to its high carboxyl content and the spherical brush structure. The present synthetic route also provides versatile method to tune the core size and grafting density of spherical brushes, as the former can be tuned by stöber process and the latter by adjusting the grafting density of RAFT CTA via silanization reaction. The relationship between the structure of SiO2@PAAs and their performance in protein immobilization will be the focus of future studies.
87
Acknowledgments This work was supported by NSF of China (21075082), 863 High Tech Program (2012AA020103), Shanghai Nano program (11nm0505600), the Science and Technology Commission of the Xuhui District (RCT 201007), SJTU funding (YG2012ZD03) and China Postdoctoral Science Foundation (2012M511088).
Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jcis.2013.02.001. References [1] H. Mori, A.H.E. Müller, Prog. Polym. Sci. 28 (2003) 1403. [2] A. Wittemann, B. Haupt, M. Ballauff, Phys. Chem. Chem. Phys. 5 (2003) 1671. [3] S. Rosenfeldt, A. Wittemann, M. Ballauff, E. Breininger, J. Bolze, N. Dingenouts, Phys. Rev. E 70 (2004) 061403. [4] O. Hollmann, T. Gutberlet, C. Czeslik, Langmuir 23 (2006) 1347. [5] J. Dai, Z. Bao, L. Sun, S.U. Hong, G.L. Baker, M.L. Bruening, Langmuir 22 (2006) 4274. [6] S.P. Cullen, X. Liu, I.C. Mandel, F.J. Himpsel, P. Gopalan, Langmuir 24 (2007) 913. [7] M. Rutnakornpituk, N. Puangsin, P. Theamdee, B. Rutnakornpituk, U. Wichai, Polymer 52 (2011) 987. [8] F. Audouin, R. Larragy, M. Fox, B. O’Connor, A. Heise, Biomacromolecules 13 (2012) 3787. [9] C.-Y. Hong, X. Li, C.-Y. Pan, J. Mater. Chem. 19 (2009) 5155. [10] X. Guo, A. Weiss, M. Ballauff, Macromolecules 32 (1999) 6043. [11] D. Li, X. Sheng, B. Zhao, J. Am. Chem. Soc. 127 (2005) 6248. [12] N.D. Treat, N. Ayres, S.G. Boyes, W.J. Brittain, Macromolecules 39 (2005) 26. [13] C. Li, J. Han, C.Y. Ryu, B.C. Benicewicz, Macromolecules 39 (2006) 3175. [14] R. Ranjan, W.J. Brittain, Macromol. Rapid. Commun. 29 (2008) 1104. [15] C. Li, B.C. Benicewicz, Macromolecules 38 (2005) 5929. [16] Y. Tsujii, M. Ejaz, K. Sato, A. Goto, T. Fukuda, Macromolecules 34 (2001) 8872. [17] Y. Piao, A. Burns, J. Kim, U. Wiesner, T. Hyeon, Adv. Funct. Mater. 18 (2008) 3745. [18] R. Ghosh Chaudhuri, S. Paria, Chem. Rev. 112 (2011) 2373. [19] A. Guerrero-Martínez, J. Pérez-Juste, L.M. Liz-Marzán, Adv. Mater. 22 (2010) 1182. [20] J. Zhang, X. Li, J.M. Rosenholm, H.-C. Gu, J. Colloid Interface Sci. 361 (2011) 16. [21] Z.X. Li, J.C. Barnes, A. Bosoy, J.F. Stoddart, J.I. Zink, Chem. Soc. Rev. 41 (2012) 2590. [22] W. Stöber, A. Fink, E. Bohn, J. Colloid Interface Sci. 26 (1968) 62. [23] Z. Gan, R.R. Marquardt, J. Biochem. Bioph. Methods 39 (1999) 1. [24] Y. Zhao, S. Perrier, Macromolecules 40 (2007) 9116. [25] M. Ballauff, Prog. Polym. Sci. 32 (2007) 1135. [26] X. Guo, M. Ballauff, Langmuir 16 (2000) 8719. [27] T. Schiestel, H. Brunner, G.E.M. Tovar, J. Nanosci. Nanotechno. 4 (2004) 504. [28] S.-C. Huang, M.D. Stump, R. Weiss, K.D. Caldwell, Anal. Biochem. 237 (1996) 115.
Contents lists available at SciVerse ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
A facile route to the synthesis of spherical poly(acrylic acid) brushes via RAFT polymerization for high-capacity protein immobilization Zhenyuan Qu, Fenglin Hu, Kaimin Chen, Zongqiang Duan, Hongchen Gu, Hong Xu ⇑ School of Biomedical Engineering, Med-X Research Institute, Shanghai Jiao Tong University, Shanghai 200030, PR China
a r t i c l e
i n f o
Article history: Received 13 December 2012 Accepted 1 February 2013 Available online 11 February 2013 Keywords: Spherical brushes Poly(acrylic acid) (PAA) RAFT polymerization Protein immobilization
a b s t r a c t Spherical poly(acrylic acid) brushes were prepared via a facile RAFT polymerization route from silica nanoparticles (SiO2@PAAs). A silane functionalized RAFT chain transfer agent was designed and synthesized by a one-step reaction, and then immobilized onto silica nanoparticles (SiNPs) through its R group to afford RAFT polymerization. Key structural parameters and contents of carboxyl groups of SiO2@PAAs were thoroughly characterized by transmission electron microscopy, dynamic light scattering, gel permeation chromatography, thermogravimetric analysis and conductometric titration. The SiO2@PAAs exhibit excellent dispersity, tunable brush thicknesses (14.6–68.8 nm) and abundant carboxyl groups (0.82– 2.37 mmol/g). An ultra-high protein immobilization capacity (2600 lg streptavidin/mg SiO2@PAAs) was realized by virtue of its rich carboxyl groups and spherical brush structure, which opens up new possibilities for biomedical applications. Ó 2013 Elsevier Inc. All rights reserved.
1. Introduction Poly(acrylic acid) (PAA) brushes (either planar or spherical) have aroused great interest in the past decade due to its simple structure, abundant content of carboxyl groups and response to the environment stimuli [1]. These characteristics have resulted in wide applications of PAA brushes in biomedical fields including protein adsorption [2–4], immobilization of biomolecules [5–8], controlled drug release [9] and so on. Among them, spherical PAA brushes (SPAABs) in which PAA chains are attached to particles have emerged as an attractive structure [10]. On the one hand, the particles serve as ideal carriers to realize an effective separation of biomolecules or targeting to organism for both in vitro and in vivo studies. On the other hand, the PAA corona endows the particles with excellent dispersity while maintaining the above-mentioned general characteristics of PAA brushes. Various techniques have been attempted to fabricate the PAA brush layer. Guo et al. [10] firstly reported the synthesis of SPAABs via photoemulsion polymerization. The uncontrolled polymerization process, however, leads to an ill-defined structure [1,10]. A precise control over the brush structure is not available until the advent of living radical polymerization (LRP) at the end of the last century, which gradually becomes the predominant method to fabricate PAA brushes or brushes with PAA block. Among various LRP techniques, atom transfer radical polymerization (ATRP) has been widely adopted [5–8,11,12]. Nevertheless, as ATRP is not compati⇑ Corresponding author. Fax: +86 21 62932907. E-mail address: [email protected] (H. Xu). 0021-9797/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcis.2013.02.001
ble with monomers containing carboxyl groups, a two-step process is usually employed, i.e., polymerization of protected monomer and deprotection by hydrolysis [5,7,8,11] or pyrolysis [6,12]. Reversible addition–fragmentation chain transfer (RAFT) polymerization is another representative LRP technique that is very suitable for the synthesis of PAA brushes owing to its tolerance to the acrylic acid (AA) monomer. However, up to date, RAFT polymerization is much less adopted regarding the fabrication of PAA brushes in comparison with ATRP. Difficulties in the synthesis and/or immobilization of RAFT initiator [13], together with relatively low grafting density of the brushes, has been recognized as major factors resulting in the limitation of RAFT polymerization [14]. To conduct a surface RAFT polymerization, harsh reaction conditions (e.g. Grignard reaction) [15] or a multi-step derivation [9,14,16] is often necessary to immobilize the RAFT chain transfer agent (RAFT CTA), which can be tedious, complicated, and even demanding to some materialists without solid expertise in organic chemistry. It is, then, of significant importance to develop novel RAFT polymerization process with simplified synthetic route to the synthesis of SPAABs. Among various spherical solid carriers, silica nanoparticles (SiNPs) are widely used in biomedical fields due to its easy synthesis and modification, stable chemical properties and excellent biocompatibility [17]. SiNPs also serve as a robust and versatile shell for various nanoparticles [18,19] and can be easily extended to mesoporous silica [20,21], which will endow the material with more functionalities for application. SiNPs-based SPAABs, therefore, act as an important model material for both fundamental studies and biomedical applications.
Z. Qu et al. / Journal of Colloid and Interface Science 398 (2013) 82–87
In this paper we report a facile RAFT polymerization route to the synthesis of SPAABs from SiNPs with well-defined structure (denoted as SiO2@PAAs). A novel RAFT CTA with silane functional groups was synthesized by a one-step reaction. This RAFT CTA enables a simple immobilization via silanization and the subsequent RAFT polymerization from its R group to form SiO2@PAAs. Key parameters of SiO2@PAAs were determined and the control of RAFT process over brush thicknesses was performed. We further show the advantages of SiO2@PAAs in the improvement of protein immobilization capacity.
2. Materials and methods 2.1. Materials (4-(chloro)phenyl)trimethoxysilane was purchased from Gelest Inc., N-(3-dimethylamino-propyl)-N0 -ethylcarbodiimide (EDC) and ethanethiol were obtained from J&K chemical. N-hydroxysuccinimide (NHS) and bicinchoninic acid (BCA) reagent kit was obtained from Thermo Scientific. Streptavidin(SA) was obtained from Shanghai Yemin biotechnology Co., Ltd., biotinylated horseradish peroxidase (biotin-HRP) from Beijing biosynthesis biotechnology Co., Ltd., and biotin from Sangon biotech (Shanghai) Co., Ltd., 3,30 ,5,50 -Tetramethylbenzidine was obtained from Sigma. Acrylic acid (AA) was distilled under reduced pressure and stored in 20 °C prior to use. 2,20 -Azobisisobutyronitrile (AIBN) was recrystallized from ethanol. All the other reagents were the products of China National Medicines Group Shanghai Chemical Reagents Company and used without further purification.
2.2. Synthesis of RAFT CTA The silane functionalized RAFT CTA, ethyl 4-(trimethoxysilyl)benzyl carbonotrithioate (1), was synthesized as follows. Ethanethiol (487 lL, 6.6 mmol) was charged into a stirred suspension of K3PO4 (1.02 g, 6.6 mmol) in anhydrous acetone (15 mL) and stirring for half an hour. CS2 (1.1 mL, 18 mmol) was added and the solution turned bright yellow. After stirring for another 10 min, (4-(chloromethyl)phenyl)-trimethoxysilane (1.43 mL, 6.6 mmol) was added and the mixture was then stirred at ambient temperature in nitrogen atmosphere for 13 h. The mixture was concentrated, diluted with dichloromethane and filtered off. After removing the solvent from the filtrate under reduced pressure the resulting yellow residue was purified by column chromatography on silica gel using a petroleum ether/ethyl acetate gradient to yield a bright yellow oil 1 (73%). 1 H NMR (400 MHz, CDCl3) d 7.60 (d, J = 7.9 Hz, 2H), 7.37 (d, J = 7.8 Hz, 2H), 4.62 (s, 2H), 3.61 (s, 9H), 3.38 (q, J = 7.4 Hz, 2H), 1.36 (t, J = 7.4 Hz, 3H). 13C NMR (CDCl3) d 223.37, 137.90, 135.10, 128.80, 128.06, 50.86, 41.13, 31.70, 13.04.m/z (EI) 348.1, 211.2. IR (cm1): 3070, 3016, 2967, 2941, 2871, 2840, 1602, 1454, 1399, 1191, 1124, 1082, 1030, 876, 809, 726. The free RAFT CTA, benzyl dodecyl carbonotrithioate (2), was prepared in a similar procedure to that of the silane functionalized RAFT CTA. Specifically, dodecanethiol (2.4 mL, 9.9 mmol) was charged to a stirred suspension of K3PO4H2O(2.64 g, 9.9 mmol) in acetone (30 mL) and stirring for half an hour. CS2 (1.65 mL, 27 mmol) was added and the solution turned bright yellow. After stirring for another 10 min, benzyl bromide (1.18 mL, 9.9 mmol) was added and the mixture was stirred overnight at ambient temperature. The mixture was concentrated and filtered off. After removing the solvent from the filtrate under reduced pressure, Compound (2) was crystallized in refrigeration and obtained as a yellow solid product with 92% isolated yield.
83
1
H NMR (400 MHz, CDCl3) d 7.37–7.26 (m, 5H), 4.62 (s, 2H), 3.38 (t, J = 7.4 Hz, 2H), 1.76–1.17 (m, 20H), 0.89 (t, J = 6.7 Hz, 3H). 13C NMR (CDCl3) d 224.06, 135.55, 129.85, 128.40, 128.07, 54.72, 41.40, 37.11, 31.99, 29.71, 29.63, 29.52, 29.42, 29.18, 28.99, 28.04, 22.77, 14.21. IR (cm1): 2954, 2917, 2852, 1494, 1464, 1454, 1063, 826, 880, 809, 698. 2.3. Synthesis of RAFT CTA functionalized SiNPs (SiO2@RAFT) SiNPs with a diameter of 80 nm (TEM image was shown in Fig. S1 in supporting information) were prepared by the Stöber procedure [22]. To modify SiNPs with RAFT CTA, compound 1 (412 lL) was added into 70 mL suspension of SiNPs in absolute ethanol (containing 3 g of SiNPs). The mixture was refluxed under nitrogen for 24 h. The product was washed with ethanol and obtained as yellow particles suspended in ethanol. The grafting density of RAFT CTA can be quantitatively determined by comparing the intensity of 308-nm absorption of SiO2@RAFT to the standard curve of silane functionalized RAFT CTA measured in ethanol [15] (e308 = 12,751 L mol1 cm1). 2.4. Synthesis of SiO2@PAAs SiO2@RAFT was transferred into N,N-dimethylformamide (DMF) by centrifugation and resuspension. Then, AA, AIBN and free RAFT CTA, 2, were added and the mixture was transferred into a schlenk tube. After three cycles of freeze–pump–thaw, the system was closed and placed into an oil bath thermostat at 70 °C. The mixture becomes sticky as the polymerization proceeds. The product was diluted by DMF and centrifuged to obtain SiO2@PAAs, which were then extensively washed by ethanol to fully remove untethered polymers and unreacted chemicals. In a typical run, the final concentration of AA, AIBN, free RAFT CTA and SiNPs were 4.5 M, 3.75 mM, 29.7 mM and 15 mg/mL, respectively. The polymerization was allowed to proceed in DMF for 3 h. Three SiO2@PAAs (denoted as SiO2@PAAs75, SiO2@PAAs150, SiO2@PAAs300) with various feed ratios of AA to free RAFT CTA (75, 150 and 300, respectively) were obtained. 2.5. Immobilization of SA on SiO2@PAAs SA was immobilized on the SiO2@PAAs by a two-step NHS/EDC coupling procedure. 1 mg of SiO2@PAAs300 was firstly activated by NHS (25 mg/mL) and EDC (50 mg/mL) in 0.4 mL of 10 mM phosphate buffer containing 0.05% (w/v) tween-20 (PB-T) for 15 min. After removing the excess reactants, SA was added to allow a 2 h incubation. The binding capacity of SA on SiO2@PAAs, corresponding to the difference of SA concentration in solution, was quantified by the determination of protein concentration before and after conjugation using BCA method. The SiO2@PAAs-SA complexes were blocked with 0.2% (w/v) BSA and 0.5% glycine overnight at 4 °C. Finally, the resulting SiO2@PAAs-SA complexes were stored in PB-T buffer containing 0.1% BSA. 2.6. Determination of binding capacity of SiO2@PAAs-SA complexes with biotin The binding capacity between SiO2@PAAs-SA complexes and biotin was measured by enzyme competitive inhibition method [23]. The SiO2@PAAs-SA complexes (40 lg) was mixed with a gradient amount of biotin for 10 min at 37 °C. After centrifugation and discarding the supernatant, 100 lL of diluted biotinylated horseradish peroxidase solution was added to the mixture for another 10 min incubation at 37 °C. The resulting mixture was washed five times with PB-T buffer and then incubated with 100 lL of substrate buffer (containing TMB) under shaken for 10 min at room temperature. The
84
Z. Qu et al. / Journal of Colloid and Interface Science 398 (2013) 82–87
reaction was ended by adding 25 lL of 2 M H2SO4 solution. After centrifugation, 100 lL of supernatant was transferred to a micro-well and measured at the wavelength of 450 nm by a multifunctional microplate reader. The saturation binding capacity of SiO2@PAAs-SA complexes with biotin corresponds to the point where the optical density (OD) drops to the lower plateau. 2.7. Characterization The monomer conversions were obtained by gravimetry. Number average molecular weight (Mn) and polydispersity index (PDI) of the polymers cleaved from the surface of SiNPs using hydrofluoric acid (HF) [15] were determined by gel permeation chromatography (GPC) in an equipment comprising a Waters 1525 binary HPLC pump, a 2707 auto sampler, and a 2414 refractive index detector. The columns are an Ultrahydrogel™ guard column (porosity 200 Å, particle size 6 lm, dimension 6 mm 40 mm), followed by three Ultrahydrogel™ columns (porosity 1000 Å, 250 Å, and 120 Å, particle size 12 lm for 1000 Å, and 6 lm for others, dimension 7.8 mm 300 mm). Phosphate buffer (0.1 M, pH = 7.2) with 0.3 M NaCl were used as the eluent, and calibration was done by sodium polyacrylate (PAANa) standards (purchased from American Polymer Standards Corporation) with a flow rate of 1.0 mL/min and injection volume of 50 lL at 35 °C. Thermogravimetric analysis (TGA) was performed on a Netzsch (TG 209 F1 Iris) thermogravimetric analyzer in air at a heating rate of 10 °C/min. The conductometric titration was done by a Mettler Toledo T50 autotitrator using 0.1 M NaOH as titrant. Transmission electron microscope (TEM) images were obtained by a JEM 2010 (JEOL, Japan) instrument operating at 200 kV. UV–Vis spectrum was recorded by a Unicam UV300 UV–Vis spectrometer. The quantification of SA concentration and the determination of free biotin binding capacity were carried out using a microplate protocol. The OD was read by a PerkinElmer 1420 Multilabel
Counter. The hydrodynamic diameter was determined by dynamic light scattering (DLS) using a Malvern Nano ZS particle size analyzer. 1H NMR (400 MHz) spectra were recorded on a Bruker 400 UltraShield spectrometer at 25 °C using CDCl3 as a solvent. Mass spectrometry (MS) was performed on a Pegasus 4D GCxGC-TOFMS mass spectrometer. Fourier transform infrared spectra were recorded using KBr pellet method on a Nicolet 6700 FT-IR spectrometer. 3. Results and discussion 3.1. Synthesis of SiO2@PAAs via a new RAFT agent with a siloxane end The synthetic route towards SiO2@PAAs is illustrated in Scheme 1 with the following steps. (I) A novel RAFT CTA with silane functional groups, 1, was synthesized in one step for the first time. (II) The silane functionalized RAFT CTA was immobilized onto the SiNPs by a silanization reaction. The reaction was done by refluxing in ethanol, which is a good dispersant for both SiNPs and SiO2@RAFT. (III) RAFT polymerization was conducted to form the PAA brushes. Free RAFT CTA, 2, was added in solution to tune the brush thicknesses of the brushes. Three SiO2@PAA samples with different feed ratios of AA to RAFT CTA were prepared. In the present work, we design a novel RAFT CTA with a silane functionalized benzyl groups as its R group and an ethyl group as its Z group. Using this agent, the immobilization of RAFT CTA by ‘‘R-group approach’’ is simplified as an efficient and robust silanization reaction. The Z group is selected to be a tiny ethyl group to minimize the steric hindrance in the immobilization of RAFT CTA. Different from previous works that involve a multi-step derivation [9,14,16] or reactions with stringent conditions [15] to form R-group immobilized RAFT CTA, the present route clearly simplifies the synthetic procedure with the use of milder reaction conditions. It has been recognized that the ‘‘R-group approach’’
Scheme 1. Synthetic route to SiO2@PAAs.
Z. Qu et al. / Journal of Colloid and Interface Science 398 (2013) 82–87
Fig. 1. UV–Vis specra of RAFT CTA, bare SiNPs and SiO2@RAFT measured in ethanol. The characteristic peak at 308 nm was observed for both RAFT CTA and SiO2@RAFT, which can be further utilized to determine the grafting density of RAFT CTA on SiNPs.
Table 1 Surface RAFT polymerization to fabricate SiO2@PAAs with different feed ratios of monomer to RAFT CTA. Sample
SiO2@PAAs75
SiO2@PAAs150
SiO2@PAAs300
[AA]/[RAFT] Conversion (%) Mn (kD) Mn(theo) (kD) PDI
75 80 4.5 4.7 1.13
150 84 10.2 9.4 1.15
300 77 18.4 17.0 1.48
resembles the ‘‘graft-from’’ approach. Thus a relatively high grafting density can be accomplished due to a reduced steric hindrance [14], which consequently contributes to a higher content of carboxyl groups of SiO2@PAAs for protein immobilization. The successful immobilization of RAFT CTA was demonstrated by UV–Vis spectrometry (Fig. 1), in which a characteristic peak centered at 308 nm was found in both RAFT CTA and SiO2@RAFT, while the same peak was absent in bare SiNPs. After silanization in ethanol, a good dispersant for both SiNPs and SiO2@RAFT, the SiO2@RAFT was obtained as yellow monodisperse particles (PDI = 0.040) with roughly identical hydrodynamic diameter to SiNPs. The excellent dispersity of SiO2@RAFT is not only important to obtain the resulting SiO2@PAAs with a well-defined structure,
85
Fig. 3. TGA curves of SiNPs, SiO2@RAFT, and SiO2@PAAs.
but also a prerequisite for the quantification of RAFT CTA grafting density using UV–Vis spectrometry. The grafting density of RAFT CTA was quantified to be 0.40 CTA/nm2, which is in accordance with the previously reported value [15]. The concentration of RAFT CTA is usually maintained to be much higher than that of radical initiators for a well-controlled RAFT polymerization. For surface RAFT polymerization, the addition of free RAFT CTA is a useful approach to help control the polymerization process when the concentration of surface-bound RAFT CTA is insufficient [16,24]. A rough estimation reveals that in our systems, surface-bound RAFT CTA (about only 0.3 mM with the particle concentration of 15 mg/mL) accounts for a very small proportion in total RAFT CTA due to a relatively low specific surface area of 80 nm-SiNPs made by stöber method. Hence the polymerization is mainly controlled by the free RAFT CTA added in solution. SiO2@PAAs with different molecular weights can be obtained by changing the amount of free RAFT CTA as PAAs grow synchronously on the surface and in solution. As shown in Table 1, the molecular weights of the grafted polymers are close to theoretical prediction, demonstrating a good control of RAFT process over molecular weights of PAA brushes. The low PDIs of the grafted polymers for SiO2@PAAs75 and SiO2@PAAs150 also indicate that the polymerizations are well controlled, generating uniform PAA brushes. On the other hand, the PDI for SiO2@PAAs300 increases to 1.48. The reason has not been clear yet.
Fig. 2. (a) Hydrodynamic diameters of SiNPs and SiO2@PAAs in deionized water determined by DLS. (b) Hydrodynamic diameters (based on cumulant method) of SiO2@PAAs in water as a function of pH.
86
Z. Qu et al. / Journal of Colloid and Interface Science 398 (2013) 82–87
Table 2 Structural parameters and the content of carboxyl groups of SiO2@PAAs with different feed ratios of monomer to RAFT CTA. Sample
SiO2@PAAs75
SiO2@PAAs150
SiO2@PAAs300
DSiNPsa L (nm)b Gr (g PAA/g SiO2)c r (nm2)d COOH (mmol/g)e
80 14.6 0.084 0.34 0.82
80 45.6 0.161 0.28 1.56
80 68.8 0.242 0.23 2.37
a Size of the spherical carrier (DSiNPs) was determined by TEM (see Fig. S1 in supporting information). b PAA brush thicknesses (L) = 12 ðdSiO2 @PAAs dSiNPs Þ where dSiO2 @PAAs and dSiNPs are the hydrodynamic diameters of fully-stretched SiO2@PAAs (at pH above 8) and SiNPs determined by dynamic light scattering (DLS), respectively. c Mass ratio of the grafted polymer to the SiNPs (Gr) estimated from thermogravimetric analysis (TGA). d Grafting density of PAA corona (r) was calculated in combination of Mn and Gr. The calculation method was described in Supporting information. e The titration background of SiO2@RAFT, which probably results from the residue silanol groups, was subtracted.
3.2. Characterization of SiO2@PAAs with controlled brush thicknesses The regular structure makes spherical brushes an ideal model material for the study concerning the relationship between its structure and performance. The key structural parameters including the size of core, brush thicknesses and the grafting density of the brushes can be tuned and explicitly determined by a variety of methods [25]. Fig. 2a shows the hydrodynamic diameters of SiO2@PAAs and the bare SiNPs. All the three SiO2@PAAs have an increased hydrodynamic diameter compared with bare SiNPs, indicating the formation of PAA corona, whose thicknesses increase with the increase of feed ratios of monomer to RAFT CTA. Fig. 2b shows the change of hydrodynamic diameters of SiO2@PAAs as a function of pH. The PAA corona collapses at a pH around its pKa and reaches a fully-stretched conformation in slightly basic conditions, which is in conformity with the typical behavior of PAA
brushes in aqueous solution [26]. It is also worth mentioning that the monodispersity of all SiO2@PAAs are maintained at pH > 3 (PDI < 0.08) due to both steric and electrostatic stabilization effect of PAA corona to nanoparticles. The TGA curves of the materials are shown in Fig. 3. An additional weight loss was observed for SiO2@PAAs in comparison with SiO2@RAFT and the weight loss also increases with the increase of feed ratios. In combination of the results obtained by DLS, TGA and GPC, together with TEM, the structural parameters of core size, brush thicknesses and grafting density of the brushes can be determined, which are summarized in Table 2. It can be seen that as the feed ratios of monomer to RAFT CTA increase, a higher amount of polymers can be grafted to SiNPs, and SiO2@PAAs with larger brush thicknesses and lower density will be obtained. The PAA grafting density for SiO2@PAAs and the aforementioned grafting density of RAFT CTA for SiO2@RAFT suggest that a proportion of about 57–85% surface-bound RAFT CTA kept active in the process of RAFT polymerization. The results of conductometric titration (Table 2) further confirm that the brushes have abundant contents of carboxyl groups. The results are in well agreement with TGA and the contents of carboxyl groups of SiO2@PAAs are at least one order of magnitude higher than that achievable by conventional monolayer functionalization method [27]. 3.3. High-capacity immobilization of streptavidin (SA) The SPAABs with an exceptionally rich content of carboxyl groups provide abundant sites for protein immobilization. Here we made a preliminary demonstration on the ultra-high protein binding capacity of SiO2@PAAs obtained in this work. SA is chosen as the model protein as the specific recognition between SA and biotin (and also biotinylated biomolecules) has found wide applications in biotechnology [28]. SA was immobilized onto SiO2@PAAs300 by means of NHS/EDC coupling chemistry. And the resulting SiO2@PAAs-SA complex
Fig. 4. (a) Schematic illustration of the protein immobilization process and the binding of biotin to SA. (b) Amount of SA found (lg/mg) on particles and in dependence of total SA supplied. (c) Binding capacity of biotin to SiO2@PAAs-SA complexes determined by enzyme competitive inhibition method.
Z. Qu et al. / Journal of Colloid and Interface Science 398 (2013) 82–87
has the ability to further capture biotin or biotinylated biomolecules. The process is schematically illustrated in Fig. 4a. Fig. 4b depicts the binding capacity of SA to SiO2@PAAs300 with increasing amounts of SA added. A saturation point was observed, indicating that SiO2@PAAs300 has a binding capacity of ca. 2600 lg SA/mg. This value is about 60-fold higher than that achieved by conventional monolayer functionalization method [27]. It is also higher than that has been accomplished by Cullen et al. [6] with planar PAA brushes, in which a 16-fold improvement are reported utilizing the same conjugation process. At the same time, the binding efficiency of SA to per unit carboxyl group is also increased by two times in comparison with the monolayer modification structure [27]. The considerably rich carboxyl content of the materials brought about by the PAA brushes clearly accounts for the high binding capacity. And it is also probable that the structure of the spherical brush, with an increasing accessibility in its outer layers for the bulky protein molecules, contributes to the enhanced protein binding efficiency. Detailed investigation on the binding mechanism is in progress. The biological activity of the surface-bound SA was tested by the enzyme competitive inhibition method (Fig. 4c) [23]. The binding capacity of biotin was about 91 nmol/mg, corresponding to ca. 2 biotins per SA, which is in consistency with the previously reported value [27]. It demonstrates that the biological activity of SA keeps intact after conjugation.
4. Conclusion In summary, a facile and efficient route to the synthesis of SiO2@PAAs via RAFT polymerization was developed. The RAFT process provides a good control over molecular weights and thicknesses of the PAA brushes. The resulting SiO2@PAAs systems have a tunable brush thicknesses of 14.6–68.8 nm, PAA grafting density of 0.23–0.34 nm2 with a rich carboxyl group content of 0.82–2.37 mmol/g. The protein binding capacity to the brushes was substantially improved to 2600 lg SA/mg SiO2@PAAs owing to its high carboxyl content and the spherical brush structure. The present synthetic route also provides versatile method to tune the core size and grafting density of spherical brushes, as the former can be tuned by stöber process and the latter by adjusting the grafting density of RAFT CTA via silanization reaction. The relationship between the structure of SiO2@PAAs and their performance in protein immobilization will be the focus of future studies.
87
Acknowledgments This work was supported by NSF of China (21075082), 863 High Tech Program (2012AA020103), Shanghai Nano program (11nm0505600), the Science and Technology Commission of the Xuhui District (RCT 201007), SJTU funding (YG2012ZD03) and China Postdoctoral Science Foundation (2012M511088).
Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jcis.2013.02.001. References [1] H. Mori, A.H.E. Müller, Prog. Polym. Sci. 28 (2003) 1403. [2] A. Wittemann, B. Haupt, M. Ballauff, Phys. Chem. Chem. Phys. 5 (2003) 1671. [3] S. Rosenfeldt, A. Wittemann, M. Ballauff, E. Breininger, J. Bolze, N. Dingenouts, Phys. Rev. E 70 (2004) 061403. [4] O. Hollmann, T. Gutberlet, C. Czeslik, Langmuir 23 (2006) 1347. [5] J. Dai, Z. Bao, L. Sun, S.U. Hong, G.L. Baker, M.L. Bruening, Langmuir 22 (2006) 4274. [6] S.P. Cullen, X. Liu, I.C. Mandel, F.J. Himpsel, P. Gopalan, Langmuir 24 (2007) 913. [7] M. Rutnakornpituk, N. Puangsin, P. Theamdee, B. Rutnakornpituk, U. Wichai, Polymer 52 (2011) 987. [8] F. Audouin, R. Larragy, M. Fox, B. O’Connor, A. Heise, Biomacromolecules 13 (2012) 3787. [9] C.-Y. Hong, X. Li, C.-Y. Pan, J. Mater. Chem. 19 (2009) 5155. [10] X. Guo, A. Weiss, M. Ballauff, Macromolecules 32 (1999) 6043. [11] D. Li, X. Sheng, B. Zhao, J. Am. Chem. Soc. 127 (2005) 6248. [12] N.D. Treat, N. Ayres, S.G. Boyes, W.J. Brittain, Macromolecules 39 (2005) 26. [13] C. Li, J. Han, C.Y. Ryu, B.C. Benicewicz, Macromolecules 39 (2006) 3175. [14] R. Ranjan, W.J. Brittain, Macromol. Rapid. Commun. 29 (2008) 1104. [15] C. Li, B.C. Benicewicz, Macromolecules 38 (2005) 5929. [16] Y. Tsujii, M. Ejaz, K. Sato, A. Goto, T. Fukuda, Macromolecules 34 (2001) 8872. [17] Y. Piao, A. Burns, J. Kim, U. Wiesner, T. Hyeon, Adv. Funct. Mater. 18 (2008) 3745. [18] R. Ghosh Chaudhuri, S. Paria, Chem. Rev. 112 (2011) 2373. [19] A. Guerrero-Martínez, J. Pérez-Juste, L.M. Liz-Marzán, Adv. Mater. 22 (2010) 1182. [20] J. Zhang, X. Li, J.M. Rosenholm, H.-C. Gu, J. Colloid Interface Sci. 361 (2011) 16. [21] Z.X. Li, J.C. Barnes, A. Bosoy, J.F. Stoddart, J.I. Zink, Chem. Soc. Rev. 41 (2012) 2590. [22] W. Stöber, A. Fink, E. Bohn, J. Colloid Interface Sci. 26 (1968) 62. [23] Z. Gan, R.R. Marquardt, J. Biochem. Bioph. Methods 39 (1999) 1. [24] Y. Zhao, S. Perrier, Macromolecules 40 (2007) 9116. [25] M. Ballauff, Prog. Polym. Sci. 32 (2007) 1135. [26] X. Guo, M. Ballauff, Langmuir 16 (2000) 8719. [27] T. Schiestel, H. Brunner, G.E.M. Tovar, J. Nanosci. Nanotechno. 4 (2004) 504. [28] S.-C. Huang, M.D. Stump, R. Weiss, K.D. Caldwell, Anal. Biochem. 237 (1996) 115.