Synthesis of Monodispersed Submillimeter-Sized Molecularly

Apr 8, 2015 - attention for the development of proteomics because they can be used as ..... S3a−f). When the Span 85 concentration increased, the av...
0 downloads 0 Views 5MB Size
Article pubs.acs.org/Langmuir

Synthesis of Monodispersed Submillimeter-Sized Molecularly Imprinted Particles Selective for Human Serum Albumin Using Inverse Suspension Polymerization in Water-in-Oil Emulsion Prepared Using Microfluidics Kyohei Takimoto, Eri Takano, Yukiya Kitayama, and Toshifumi Takeuchi* Graduate School of Engineering, Kobe University, 1-1, Rokkodai-cho, Nada-ku, Kobe 657-8501, Japan S Supporting Information *

ABSTRACT: We synthesized monodispersed submillimetersized (100 μm−1 mm) microgels by inverse suspension polymerization of water-soluble monomer species with a photoinitiator in water-in-oil (W/O) droplets formed by the microchannel. After fundamental investigations of the selection of suitable surfactants, surfactant concentration, and flow rate, we successfully prepared monodispersed submillimeter-sized W/O droplets. Because radical polymerization based on thermal initiation was not appropriated based on colloidal stability, we selected photoinitiation, which resulted in the successful synthesis of monodispersed submillimeter-sized microgels with sufficient colloidal stability. The microgel size was controlled by the flow rate of the oil phase, which maintained the monodispersity. In addition, the submillimeter-sized microgels exhibit high affinity and selective binding toward HSA utilizing molecular imprinting. We believe the monodispersed submillimeter-sized molecularly imprinted microgels can be used as affinity column packing materials without any biomolecules, such as antibodies, for sample pretreatment to remove unwanted proteins without a pump system.

1. INTRODUCTION Monodispersed particles/microgels that are more than a micrometer in size have been widely studied because these particles can be used in many research areas and industrial fields, such as medical diagnostics and a liquid crystal spacer.1−6 Recently, the particles have found use in life science including in the area of proteomics and protein purification as affinity chromatography column packing materials.7−10 In proteomics, the qualitative and quantitative analyses of small amounts of proteins have been performed by mass spectrometry, and a key process for successful proteomic studies is the removal of unwanted proteins by affinity chromatography.9 Monodispersed submillimeter-sized (100 μm−1 mm) particles/microgels with a high hydrophilicity, which is required for suppression of nonspecific adsorption, have attracted much attention for the development of proteomics because they can be used as affinity column packing materials for sample pretreatment to remove unwanted proteins without a pump system prior to mass spectrometry. Since the early study of affinity chromatography by Cuatrecasas et al.,11 agarose beads have been used as packing material for affinity chromatography. However, the process for preparing the affinity beads involves several challenges, such as the use of cyanogen bromide for antibody modification.12 Therefore, polymer-based affinity beads have been focused on due to their high flexibility in particles/microgels design.13,14 © XXXX American Chemical Society

Conventional methods of creating monodispersed particles that are more than a micrometer in size have involved heterogeneous polymerization, such as precipitation or dispersion polymerization.2,3 However, the particle size range of these methods is typically 1−10 μm. Suspension polymerization using a vortex is another potential process for the synthesis of polymer particles that are more than 100 μm in size. However, the monodispersity is quite low.15−17 A Sirasu porous glass (SPG) membrane technique enables us to prepare micrometer-sized monodispersed particles that limited to a size range of 1−10 μm.18−21 The similar size range limit also exists in the two-step swelling method proposed by Ugelstad et al.22 The dynamic swelling method proposed by Okubo et al. can prepare particles >10 μm with high monodispersity. However, a size of 100 μm has not been achieved.23 To date, a general synthetic method for monodispersed hydrophilic submillimeter-sized particles/microgels has not yet been established. A microchannel has the potential for use in the preparation of monodispersed submillimeter-sized droplets,4,24−30 and subsequent heterogeneous polymerization (i.e., suspension or inverse-suspension polymerizations) in the droplets enables the synthesis of monodispersed submillimeter-sized droplets. In Received: February 28, 2015 Revised: April 5, 2015

A

DOI: 10.1021/acs.langmuir.5b00769 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

2-2. Droplet Formation by the Y-Junction Microfluidic Device Using Various Flow Rates of the Oil Phases. The droplets were prepared using a Y-junction microfluidic device (KC-MY-SUS, YMC Co., Ltd., Kyoto, Japan) with a flow channel that was 500 μm wide and 100 μm deep. To investigate the relationship between the droplet size and the flow rate of the oil and aqueous phases, the flow rate of the oil phase was varied from 150 to 750 μL/ min using a syringe pump (YSP-101, YMC Co., Ltd., Kyoto, Japan) with a 10 mL gastight glass syringe (Hamilton Co., USA). The aqueous phase containing AAm (2.01 mol/L), MBAA (0.46 mol/L), HEMA (3.53 mol/L), Irgacure 2959 (0.23 mol/L), pyrrolidyl acrylate (PyA) (0.027 mol/L), and HSA (0.69 mmol/L) dissolved in 10 mM phosphate buffer (pH 7.4) was maintained at a fixed flow rate of 3 μL/ min using a syringe pump (Econoflo Syringe Pumps, Harvard Apparatus, USA) with a 1 mL gastight glass syringe (Hamilton Co., USA). The size of the droplets was measured using a stereoscopic microscope (Nikon Digital Sight DS-L1, Nikon Corp., Tokyo, Japan). 2-3. Preparation of HSA-Imprinted Microgels. For the preparation of monodispersed submillimeter-sized HSA-imprinted microgels, the aqueous and oil phases with the same content as that in the W/O droplets formation experiments flowed into the Y-junction microfluidic device using syringe pumps at an oil phase flow rate between 150 and 750 μL/min. The obtained W/O droplets were collected in a Petri dishes and photopolymerized for 24 h at 25 °C using a 365 nm LED lamp (YMC-P-0049, YMC Co., Ltd., Kyoto, Japan). After polymerization, the HSA-imprinted microgels were washed with MeOH and hexane. The morphology of the microgels was observed using a scanning electron microscope (VE-9800, Keyence Corp., Osaka, Japan). To remove the HSA template, the microgels were washed three times for 1 h at room temperature (rt) with each of the following solutions: a 0.5% SDS aqueous solution, a 0.3% SDS aqueous solution containing 200 mM sodium chloride, and a 10 mM TCEP aqueous solution. In addition, the microgels were washed three times with a 0.1% SDS aqueous solution containing 10 mM TCEP for 1 h at 70 °C. To remove the SDS, the microgels were washed more than 10 times with pure water. The removal rate of the template HSA was determined by the HSA in the washing solution using UV−vis spectroscopy (V-560, JASCO Ltd., Tokyo, Japan). The volume change ratio from the W/O droplets to the microgels was calculated from eq 1

most of the previously reported studies, the droplets made in a microchannel were the oil-in-water (O/W) type, and the resulting particles were also highly hydrophobic. For the application of these particles as separating agents for biological molecules, such as proteins, hydrophilic submillimeter-sized polymer particles are necessary for suppressing nonspecific adsorption. However, there are few studies on the production of water-in-oil (W/O) type droplets using a microchannel,31,32 and hydrophilic monodispersed submillimeter-sized polymer particles have been rarely reported.33 Herein, we report the synthesis of monodispersed hydrophilic submillimeter-sized microgels by polymerizing W/O type droplets containing monomer species using a microchannel. In addition, we focused on hydrophilic submillimeter-sized monodispersed microgels for use as affinity microgels for specific proteins without any antibodies utilizing a molecularly imprinting strategy. The molecular imprinting is a type of template polymerization where the polymerization is carried out in the presence of the target protein (template) and the functional monomers interacting with the target protein. The subsequent removal of the template results in the formation of specific binding cavities in the polymer matrix.34−44 Herein, we demonstrate the preparation of monodispersed submillimetersized artificial microgels for human serum albumin (HSA) removal. The developed strategy for the submillimeter-sized artificial microgels has several important advantages, such as high versatility of target proteins, wide functionality, ease of synthesis, high stability, and long shelf life compared with conventional affinity beads using antibodies, which allows for the use of these novel artificial affinity microgels without any antibodies. Scheme 1. Schematic Diagram of the Synthesis of Monodispersed Submillimeter-Sized Microgels Using a YShaped Microchannel and Photopolymerization

volume change ratio =

Vd − Vp Vd

(1)

where Vd is the droplet volume and Vp is the particle volume. 2-4. Preparation of Nonimprinted Microgels. For the preparation of nonimprinted microgels, the aqueous phase and the oil phase flowed into the Y-junction microfluidic device using syringe pumps at flow rates of 3 and 150 μL/min, respectively. The obtained W/O droplets were collected in a Petri dishes and photopolymerized for 24 h at 25 °C using a 365 nm LED lamp (YMC-P-0049, YMC Co., Ltd., Kyoto, Japan). After polymerization, the polymer particles were washed with MeOH and hexane. 2-5. Adsorption Experiments. HSA-imprinted microgels (10 mg) that were prepared using oil and aqueous phase flow rates of 150 and 3 μL/min, respectively, were incubated in a 10 mM phosphate buffer (pH 7.4, 2 mL). HSA (5 μM) dissolved in 10 mM phosphate buffer (pH 7.4, 10 μL) was added to the microgel dispersion and incubated with the HSA-imprinted microgels for 30 min. After 30 min of incubation, the absorbance of the supernatant (Aafter) was measured using UV−vis spectroscopy, and the amount of HSA adsorbed was calculated using the absorbance of the supernatant after 0 min of incubation (Abefore). For nonimprinted microgels, adsorption experiments were also performed using a similar protocol. 2-6. Evaluation of the Selectively of HSA-Imprinted Microgels. HSA-imprinted microgels (10 mg) that were prepared using oil and aqueous flow rates of 150 and 3 μL/min, respectively, were incubated in a 10 mM phosphate buffer (pH 7.4, 2 mL). 5 μM BSA chymotrypsin cytochrome c in 10 mM phosphate buffer (pH 7.4, 10 μL) was prepared, and these solutions were separately incubated for 30

2. EXPERIMENTAL PART 2-1. Materials. 2-Hydroxyethyl methacrylate (HEMA) and sodium chloride were purchased from Nacalai Tesque Co. (Kyoto, Japan). Acrylamide (AAm), N,N′-methylenebis(acrylamide) (MBAA), albumin from human serum (HSA), disodium hydrogen phosphate, and sodium dodecyl sulfate (SDS) were purchased from Wako Pure Chemical Industries (Osaka, Japan). Tris(2-carboxyethyl)phosphine hydrochloride (TCEP), 2-hydroxy-4′-(2-hydroxyethoxy)-2-methylpropiophenone (Irgacure 2959), sorbitan monolaurate (Span 20), and sorbitan monopalmitate (Span 40) were purchased from Tokyo Chemical Industries, Co. (Tokyo, Japan). Mineral oil and sorbitan trioleate (Span 85) were purchased from Sigma-Aldrich Japan (Tokyo, Japan). Sodium dihydrogen phosphate dihydrate was purchased from Katayama Chemical Industries, Co. (Osaka, Japan). B

DOI: 10.1021/acs.langmuir.5b00769 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 1. Structures of Span 20 (a), Span 40 (b), and Span 85 (c).

Figure 2. Stereoscopic microscope images of the W/O droplets prepared using the microchannel with various oil flow rates. The oil phase flow rates (μL/min) were as follows: (a) 150, (b) 250, (c) 375, (d) 400, (e) 500, and (f) 750. min. After 30 min of incubation, the absorbance of the supernatant (Aafter) was measured using UV−vis spectroscopy, and the amount of HSA adsorbed was calculated using the absorbance of the supernatant after 0 min of incubation (Abefore).

study, PyA was determined to be an effective functioned monomer for HSA imprinting.45 All of the surfactants did not dissolve in the aqueous phase, indicating that they can be used as surfactants for the W/O emulsion. In addition, Span 40 did not dissolve in the mineral oil at rt. To investigate the solubilized temperature of Span 40 in mineral oil, the transmittance of Span 40 (2 vol %) in mineral oil was measured from 25 to 45 °C, and the solubilized temperature existed between 35 and 40 °C (Figure S1). This result indicates that Span 40 is not appropriate for W/O droplet preparation at rt. However, the other surfactants dissolved in the mineral oil at rt. The W/O droplets prepared without surfactants were not stable. When Span 20 was used as a surfactant, the obtained droplets were unstable with high polydispersity (coefficient of variation (CV): 49%). However, monodispersed aqueous droplets with high stability (CV: 3.0%) were obtained using Span 85 (Figure S2). The steric repulsion due to the long alkyl chains of Span 85 worked effectively between the formed droplets resulting in stable aqueous droplets.

3. RESULTS AND DISCUSSION 3-1. Selection of a Suitable Surfactant for Monodispersed Submillimeter-Sized W/O Droplets. The selection of a suitable surfactant is critical for the preparation of stable monodispersed submillimeter-sized microgels. To select a suitable surfactant, aqueous droplets with dissolved water-soluble monomer species including AAm (comonomer), HEMA (comonomer), MBAA (cross-linker), and PyA (functional monomer) with HSA (molecular weight: 66 500 Da, pI = 4.8) were prepared with a mineral oil phase that contained 2 vol % Span 85 (HLB: 1.8), Span 40 (HLB: 6.7), or Span 20 (HLB: 8.6) as surfactants using a Y-type microchannel, where one channel was used for the mineral oil and the other channel was used for the aqueous phase at rt (Figure 1). In our previous C

DOI: 10.1021/acs.langmuir.5b00769 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 3. Average sizes (a) and CV (b) of droplet (circles) or microgel (cube) prepared using a microchannel at various oil phase flow rates.

Figure 4. SEM images of the obtained microgels prepared by inverse suspension polymerization in W/O droplets prepared using a microchannel with various flow rates. Oil phase flow rate (μL/min): (a) 150, (b) 250, (c) 375, (d) 400, (e) 500, and (f) 750.

3-2. Preparation of Monodispersed SubmillimeterSized W/O Droplets. In suspension polymerization, the microgel size and its polydispersity after polymerization primarily depended on these characteristics of the W/O droplets, which dissolved the monomer species. Therefore, the key to the preparation of monodispersed submillimetersized microgels is to prepare the monodispersed W/O droplets with the target microgel size, where a larger droplet size compared with the target particle size is typically necessary because the volume will shrink due to an increase in the density from monomer to polymer. To investigate the optimum surfactant concentration of the monodispersed submillimeter-sized droplets, W/O droplets containing monomers and HSA were prepared with various concentrations of Span 85 (0.5−10 vol %), and the average droplet size and polydispersity were measured. In all of the experiments using this surfactant concentration range, the formed droplets were highly stable without coagulation (Figure S3a−f). When the Span 85 concentration increased, the average droplet size decreased from ca. 300 μm to ca. 100 μm. However, the polydispersity substantially increased when the

surfactant concentration increased higher than 2 vol % (Figure S4). In the microchannel device, the aqueous droplets were formed by the difference in the aqueous flow and the high flow rate mineral oil at the Y-junction point, where the formed droplet size depended on the oil/water interfacial tension because the critical aqueous phase volume for droplet formation decreased as the interfacial tension decreased when both flow rates were same. Therefore, the droplet size decreased as the surfactant concentration decreased. However, at a high surfactant concentration, the oil/water interfacial tension became too small, which appeared to lead to unexpected breaking of the formed aqueous droplets. The high CV value at a high surfactant concentration may be due to this reason. The effect of the oil flow rate on the formed droplet size was investigated using a fixed aqueous phase flow rate with 2 vol % of Span 85 (Figure 2a−f). The average size of the formed aqueous droplets decreased from ca. 450 μm to ca. 180 μm as the flow rate of the oil phase increased from 150 to 750 μL/min (Figure 3). In this series of experiments, the Reynolds number was 0.180 in the aqueous phase and 0.110−0.549 for the oil D

DOI: 10.1021/acs.langmuir.5b00769 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

S6d−g). The difference in these particle morphologies may be the reason for the different volume change ratio. We will investigate how the morphology difference affected the binding kinetics in the near future. 3-4. Functionality of Monodispersed SubmillimeterSized HSA-Imprinted Microgels. We selected 150 and 3 μL/min as the flow rates of the oil phase and aqueous phase, respectively, for preparation of the monodispersed submillimeter-sized HSA-imprinted microgels. To remove the template HSA from the obtained microgels, the microgels were washed using three types of washing solutions, as follows: SDS, TCEP, and sodium chloride aqueous solutions. SDS acted as a denaturing agent for HSA, and TCEP acted as a reducing agent for the disulfide linkage in HSA. After washing with these washing solutions, the total removing efficiency was 75%, which was determined by UV−vis spectroscopy. This efficiency indicated that the HSA recognition cavities are formed in the submillimeter-sized microgels (HSA-imprinted microgels). The obtained submillimeter-sized HSA-imprinted microgels were incubated for 30 min in a 10 mM phosphate buffer (pH 7.4, 2 mL) with 5 μM HSA. After incubation, the amount of HSA adsorption by the HSA-imprinted microgels was measured based on the concentration of remaining HSA in the supernatant after incubation using UV−vis spectroscopy, where the linearity between the absorbance and HSA concentration was confirmed (Figure S7). Therefore, the total adsorption amounts by the submillimeter-sized HSA-imprinted microgels increased as the HSA concentration increased (n = 3) (Figure 5).

phase (the aqueous and oil phase viscosities were 0.001 and 0.066 Pa·s, respectively). Therefore, both flows were laminar flows. The sufficiently low polydispersity (CV: