Article pubs.acs.org/ac
Localized Surface Plasmon Resonance Nanosensing of C‑Reactive Protein with Poly(2-methacryloyloxyethyl phosphorylcholine)Grafted Gold Nanoparticles Prepared by Surface-Initiated Atom Transfer Radical Polymerization 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: Highly sensitive and selective protein nanosensing based on localized surface plasmon resonance (LSPR) of gold nanoparticles (AuNPs) on which polymerized specific ligands were grafted as an artificial protein recognition layer for the target protein were demonstrated. As a model, optical nanosensing for Creactive protein (CRP), a known biomarker for chronic inflammation that predicts the risk of arteriosclerosis or heart attacks, was achieved by measuring the shift of LSPR spectra derived from the change of permittivity of poly(2-methacryloyloxyethyl phosphorylcholine)-grafted AuNPs (PMPC-g-AuNPs) upon interacting with CRP, in which the PMPC-g-AuNPs layer were grafted on AuNPs by surface-initiated atom transfer radical polymerization (ATRP). This nanosensing system was effective even for detecting CRP concentrations in a human serum solution diluted to 1% (w/w), at which point a limit of detection was ∼50 ng/mL and nonspecific adsorption of other proteins was negligible. The nanosensing system using specific ligand-grafted AuNPs has several strengths, such as low preparation cost, avoiding the need for expensive instruments, no necessary complex pretreatments, and high stability, because it does not contain biobased molecules. We believe this novel synthetic route for protein nanosensors, composed of AuNPs and a polymerized specific ligand utilizing surface-initiated controlled/living radical polymerization, will provide a foundation for the design and synthesis of nanosensors targeting various other biomarker proteins, paving the way for future advances in the field of biosensing. limits below 0.12 μg/mL.19−22 In our previous works, we have also demonstrated with success the highly sensitive detection of CRP by surface plasmon resonance (SPR) and reflectometric interference spectroscopy-based immunosensors.23,24 Localized surface plasmon resonance (LSPR)-based nanosensing has gained a significant reputation as a highly sensitive detection method. Gold nanoparticles (AuNPs) have great potential for use as optical nanosensing materials because they show localized surface plasmon resonance (LSPR) as specific wavelengths of visible light, whose extinction wavelength due to LSPR is shifted, depending on the permeability of the surrounding of the AuNPs.25−29 An LSPR-based nanosensing system has multiple advantages, such as avoiding the necessity of expensive instruments, ease-of-use, label-free detection, and short operation time. Although previous papers have reported successful sensing systems using antibodies (anti-CRP),23,24 these antibodies are both expensive and unstable, and artificial sensing materials provide a stable and attractive alternative. The use of polymerbased artificial recognition materials is especially attractive and
C
oronary heart disease is one of the leading cause of death in developed countries and has several underlying cause, one of which is atherosclerosis.1−3 Systemic inflammation plays a capital role in atherothrombtic inception and progression, during which cytokines produce acute phase reactants such as C-reactive protein (CRP) in the bodies of patients suffering from inflammation, making CRP levels a good indicator of inflammation, cancer, and tissue destruction.4−12 Conventionally, CRP detection for the clinical diagnosis of acute inflammation is carried out using one of several methods, such as immune nephelometry or latex aggregation; however, these methods cannot detect CRP levels below a threshold of 0.3 μg/mL and, therefore, are not sufficiently sensitive enough to detect cardiovascular risks. Recently, a highly sensitive CRP assay (hs-CRP), which can detect CRP quantitatively at concentrations under 0.15 μg/mL, has attracted much attention as powerful a predictor of future first stage coronary heart disease in presently healthy men and women3,13−17 as well as a diagnostic tool for inflammation detection in newborns, whose CRP concentrations are generally low compared with those in adults.18 To determine the presence of trace amount of CRP with high sensitivity, previous studies have reported the use of enzyme-linked immunosorbent assays and magnetic permeability assays, which were reported to have achieved detection © XXXX American Chemical Society
Received: April 2, 2014 Accepted: May 7, 2014
A
dx.doi.org/10.1021/ac501322x | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
and HAuCl4·2H2O (36 mg, 87.4 μmol) aqueous solution (water, 190 mg) was added to the ethanol solution. The water/ ethanol solution was then stirred vigorously, and 0.5 M NaBH4 aqueous solution (1.9 g) was slowly added and reacted for 15 h at room temperature. The obtained Br-functionalized AuNPs with undecyl linker (AuNPs-undecyl-Br) were washed by centrifugation with methanol/water (2:1 v/v) twice and with ethanol/hexane (5:1 v/v) three times. Finally, AuNPs-undecylBr was dispersed in ethanol. The Br-functionalized AuNPs with the ethyl linker (AuNPs-ethyl-Br) was also synthesized using bis[2-(2-bromoisobutyryloxy)-ethyl] disulfide by the same protocol as AuNPs-undecyl-Br. Surface-Initiated AGET ATRP of MPC with Br-AuNPs. MPC (236 mg, 800 μmol), CuBr2 (0.78 mg, 25 μmol), PMDETA (0.84 mg, 25 μmol) were dissolved in the AuNPsundecyl-Br dispersion in ethanol (2 g, solid content of AuNPsundecyl-Br: 0.019 wt %), in which DMSO was added as an internal standard for conversion measurement by 1H NMR. After three freeze−pump−thaw cycles, ascorbic acid (2.2 mg, 12.5 μmol) was dissolved in the ethanol solution (100 μL) and was added to the solution to reduce the copper species and start surface-initiated activator generated by electron transfer (AGET) ATRP. The polymerization was carried out at 40 °C in a water bath. After the polymerization, the obtained PMPCg-AuNPs were washed by centrifugation with ethanol twice and water three times. The washed PMPC-g-AuNPs were dispersed in 10 mM Tris−HCl buffer solution (pH 7.4) with 140 mM NaCl and 20 mM CaCl2. Surface-Initiated AGET ATRP of OEOMA with BrAuNPs. The poly(OEOMA) (POEOMA)-grafted AuNPs (POEOMA-g-AuNPs) were synthesized by surface-initiated AGET ATRP under a procedure similar to that of the protocol for the synthesis of PMPC-g-AuNPs. Synthesis of AuNPs by Citrate. HAuCl4·2H2O (60 mg, 151 μmol) was dissolved in water (450 mL), and the aqueous solution was refluxed. Sodium citrate (150 mg, 581 μmol) aqueous solution (15 mL) was added to the HAuCl4 aqueous solution, and the mixture was refluxed for 20 min. Calculation of the A/D Parameter. The A/D parameter is given by a ratio of the integral of the spectra of PMPC-gAuNPs in the range from 550 to 700 nm (A value) to that of the spectra in the range from 490 to 540 nm (D value). The Δ(A/D) value was given by (A/D) − (A/D)0, and a relative A/ D change against an initial A/D parameter was calculated by Δ(A/D)/(A/D)0, where (A/D)0 was an initial A/D parameter CRP Nanosensing with PMPC-g-AuNPs in 10 mM Tris−HCl buffer solution (pH 7.4) with 140 mM NaCl and 20 mM CaCl2. PMPC-g-AuNPs (solids content: 0.08 wt %) dispersed in 10 mM Tris−HCl buffer solution with 140 mM NaCl and 20 mM CaCl2 (1.5 mL) was diluted with the same volume of the buffer solution, which was used as a sample for absorbance measurement and stirred for 60 min at 25 °C. Various concentrations (10, 100, 1000 μg/mL) of CRP dissolved in the buffer (1.2−3.0 μL) were added stepwise to the AuNPs dispersed in the buffer, and the final concentrations of the CRP sample in each step were 0, 10, 50, 100, 500, 1000, and 3000 ng/mL. UV−vis spectra were measured after 30 min of incubation at 25 °C. To confirm the nonspecific binding to the PMPC-g-AuNPs, HSA was used as a reference protein. Each measurement was performed three times to check the repeatability. CRP Nanosensing with PMPC-g-AuNPs in Human Serum. PMPC-g-AuNPs (solids content: 0.08 wt %) dispersed
allows for important advances in the preparation of CRP sensing materials because unlike those based on antibodies and enzymes, polymer-based materials can be prepared easily and at a low cost. CRP is known to bind readily to phosphorylcholine groups in the presence of Ca2+ ions,30−38 so we selected 2methacryloyloxyethyl phosphorylcholine (MPC) for use as a polymerizable specific ligand toward CRP. Notably, Ishihara et al. clarified in a series of previous works that poly(MPC) (PMPC) can effectively suppress nonspecific binding with other proteins present with CRP.39−42 Therefore, PMPC should be a great candidate for use as an effective artificial receptor for selective CRP sensing. In this study, we demonstrated the synthesis and performance of a novel, highly sensitive CRP nanosensing material based on AuNPs grafted with poly(2-methacryloyloxyethyl phosphorylcholine) (PMPC) as an artificial CRP recognition layer (PMPC-g-AuNPs) prepared by surface-initiated atom transfer radical polymerization (ATRP).34,43−52 CRP sensing ability of obtained PMPC-g-AuNPs was demonstrated using UV−vis in a buffer solution. The nonspecific binding properties were also investigated with UV−vis using human serum albumin (HSA) as a reference protein. Finally, to confirm the effectiveness of highly sensitive detection of CRP using PMPCg-AuNPs, sensing experiments were carried out in a diluted human serum solution. Our proposed nanosensing system has several advantages compared with previously reported methods, such as avoiding the necessity of specific or complex pretreatment, the use of inexpensive instruments, and avoiding the need for expensive and unstable biobased molecules.
■
EXPERIMENTAL PART Materials. HAuCl4, NaBH4, NaCl, sodium citrate, CuBr2, Lascorbic acid, and human serum albumin (HSA) were purchased from Nacalai Tesque (Kyoto, Japan). Bis[2-(2bromoisobutyryloxy)-ethyl] disulfide, bis[2-(2-bromoisobutyryloxy)-undecyl] disulfide, N,N,N′,N″,N″ pentamethyldienediethylenetriamine (PMDETA), were purchased from SigmaAldrich (USA). Methanol, ethanol, hexane, DMSO, 2,2′azobis(isobutyronitril) (AIBN) were purchased from Wako Pure Chemical Industries (Osaka, Japan). Deionized water used was obtained from a Millipore Milli-Q purification system. MPC was kindly supplied by Prof. Kazuhiko Ishihara at the University of Tokyo. Oligo(ethylene oxide) methyl ether methacrylate (OEOMA) (degree of polymerization of ethylene oxide ∼ 8) was purchased from NOF Corporation (Tokyo, Japan). Characterization. 1H NMR spectra were measured using 300 MHz FT-NMR apparatus (JNM-LA300 FT NMR system, JEOL Ltd., Tokyo, Japan). FT-IR measurements were carried out by KBr method using (Varian 660 KU-IR, Agilent Inc., California, USA). XPS measurements were carried out using a PHI X-tool (ULVAC PHI, Inc., Kanagawa, Japan), where the takeoff angle was 45° and X-ray condition was 20 kV and 98 W with Al Kα. The particle sizes and morphologies were observed using a JEM-1230 electron microscope (JEOL Ltd., Tokyo, Japan). The particle size and size distribution were obtained using a dynamic light-scattering system (Zetasizer Nano ZS, Malvern Instruments Ltd., U.K.). UV−vis spectral measurements were performed using a V-560 spectrophotometer (JASCO Ltd., Japan). Preparation of Br-Functionalized AuNPs. A typical procedure is as follows: Bis[2-(2-bromoisobutyryloxy)undecyl]disulfide (15.3 mg, 21.7 μmol) was dissolved in ethanol (7.7 g), B
dx.doi.org/10.1021/ac501322x | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
in 10 mM Tris−HCl buffer solution with 140 mM NaCl and 20 mM CaCl2 (3.0 mL) was added to the same volume of 2 wt % human serum buffer solution. The dispersion was used as a sample for absorbance measurement and stirred for 60 min to maintain the temperature at 25 °C. Various concentrations (10, 100, 1000 μg/mL) of CRP dissolved in the buffer (1.2−3.0 μL) were added stepwise to the AuNPs dispersed in the 100-timesdiluted human serum, and the final concentrations of the CRP sample in each step were 0, 10, 50, 100, 500, 1000, and 3000 ng/mL. UV−vis spectra were measured after 30 min of incubation at 25 °C. To confirm the nonspecific binding to the PMPC-g-AuNPs, HSA was used as a reference protein. Each measurement was performed three times to check the repeatability. Determination of Binding Constant. The binding constant of PMPC-g-AuNPs toward CRP was determined by curve-fitting using DeltaGraph 5.4.5v. The fitting equation is shown below, which was generally used for determination of the binding constant of formation of the 1:1 complex, in which we hypothesized the binding between PMPC-g-AuNP and CRP occurred as one-to-one binding. Y = [(1 + KG + KH ) − D × 2KG
Figure 2. UV−vis spectra of AuNPs-undecyl-Br (a) and AuNPs-ethylBr (b) dispersed in ethanol.
obtained for AuNPs-undecyl-Br and AuNPs-ethyl-Br dispersions, respectively. The XPS measurements for S 2p and Br 3d were carried out to investigate whether the AuNPs chemically bonded with the disulfide derivatives (bis[2-(2′-bromoisobutyryloxy)undecyl]disulfide or bis[2-(2′-bromoisobutyryloxy)ethyl]disulfide) (Supporting Information (SI) Figure S1). The S 2p peak, which was not observed with AuNPs prepared with citric acid (AuNPs-citrate) as a reference material, appeared clearly with both AuNPs-undecyl-Br and AuNPs-ethyl-Br. Br 3d peaks also appeared with both AuNPs-undecyl-Br and AuNPs-ethyl-Br but were absent with AuNPs-citrate. These results indicate that AuNPs-undecyl-Br and AuNPs-ethyl-Br were chemically conjugated with bis[2-(2′-bromoisobutyryloxy)undecyl]disulfide and bis[2-(2′-bromoisobutyryloxy)ethyl]disulfide, respectively, as stabilizing agents and ATRP initiators. The particle sizes of AuNPs-undecyl-Br and AuNPs-ethyl-Br were measured by dynamic light scattering (DLS), as shown in Figure 3. Despite the stabilization of both AuNPs by disulfide derivatives, the particle sizes of the obtained AuNPs were clearly different: the z-average diameter (dz) of AuNPs-undecylBr was ∼10 nm, whereas the dz value of AuNPs-ethyl-Br was ∼340 nm. To investigate the morphologies of AuNPs-undecylBr and AuNPs-ethyl-Br, their TEM imaging was carried out. In TEM images of AuNPs-undecyl-Br, the AuNPs existed individually, and no aggregation was observed (SI Figure S2). The particle size of AuNPs-undecyl-Br measured from the TEM image was smaller than that obtained by DLS (∼10 nm). This is because the particle size obtained by DLS is the hydrodynamic particle size containing the stabilizing 2-(2bromoisobutyryloxy)ethyl thiol layer, whereas the particles used in TEM imaging were absent this layer. TEM images of AuNPsethyl-Br revealed large AuNPs (>1 μm) formed by the coagulation of small AuNPs. These results indicated that AuNPs-ethyl-Br initially formed as small particles stabilized by bis[2-(2′-bromoisobutyryloxy)ethyl]disulfide, but the particles had low stability, resulting in significant coagulation. From these results, we selected bis[2-(2′-bromoisobutyryloxy)undecyl]disulfide as the ATRP-initiator and stabilizer of AuNPs. Surface-Initiated AGET ATRP for the Preparation of PMPC-g-AuNPs. To prepare the PMPC-g-AuNPs, we carried out surface-initiated AGET ATRP of MPC in ethanol at 40 °C (Scheme 1). It is well-known that the solvent significantly affects the polymerization rate and controls the molecular weight distribution in ATRP systems.53,54 The ATRP polymerization rate of MPC was accelerated, but the control/livingness worsened as the polarity of the solvent increased.52,55 We selected ethanol as our solvent in this work because it offered
(1 + KG + KH ) − 4K 2HG ]
where Y is ΔA/D/(A/D)0, K is the affinity constant, H is the concentration of PMPC-g-AuNPs, G is the CRP concentration, and D is the maximum ΔA/D/(A/D)0.
■
RESULTS AND DISCUSSION Synthesis of Bromo-Functionalized AuNPs (AuNPsBr). Br-functionalized AuNPs were synthesized by reduction of HAuCl4 in the presence of bromoisobutyryloxydisulfide derivatives with different alkyl chain lengths (bis[2-(2bromoisobutyryloxy)undecyl]disulfide and bis[2-(2bromoisobutyryloxy)ethyl]disulfide) in an ethanol/water solution (Figure 1). When a NaBH4 aqueous solution was added
Figure 1. Structures of disulfide derivatives used in this study: (a) bis[2-(2′-bromoisobutyryloxy)undecyl]disulfide and (b) bis[2-(2′bromoisobutyryloxy)ethyl]disulfide.
to the HAuCl4 solution, dark purple particles immediately precipitated. The obtained precipitants were washed with methanol/water (2:1 v/v) and ethanol/hexane (5:1 v/v) solutions, followed by centrifugation. Figure 2 shows UV−vis spectra of the obtained AuNPs-Br dispersions in ethanol using AuNPs-undecyl-Br or AuNPs-ethyl-Br, respectively. The AuNPs-undecyl-Br dispersion showed LSPR activity with an extinction peak at 520 nm, but an extinction peak derived from LSPR did not appear for the AuNPs-ethyl-Br dispersion. The spectra were in accordance with the red and black colors C
dx.doi.org/10.1021/ac501322x | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
Figure 3. Particle size distribution of AuNPs-undecyl-Br (a) and AuNPs-ethyl-Br (b) in ethanol measured by DLS.
Scheme 1. Surface-Initiated ATRP of MPC with AuNPsundecyl-Br As Initiator Seed Particles
an appropriate control/livingness in the ATRP of MPC while allowing the AuNPs to maintain high stability without coagulating both before and after polymerization. This is because ethanol is a good solvent for both bis[2-(2bromoisobutyryloxy)undecyl]disulfide and PMPC, which leads to steric repulsion between the individual AuNPs. The time−conversion plot of the polymerization obtained from 1H NMR analysis is shown in Figure S3 in the Supporting Information. Conversion increased with polymerization time, which indicates that the Br group in AuNPs-undecyl-Br worked as the initiating group. The first-order plot kept linearity during the polymerization with passing through the origin, indicating that the propagating radical concentration were constant (SI Figure S3(b)). Furthermore, the particle size distribution of PMPC-gAuNPs at various polymerization times clearly shifted toward larger particle sizes as polymerization progressed (Figure 4), and the dz values also clearly increased (Figure 5). In the
Figure 5. z-Average particle size (da) values of PMPC-g-AuNPs prepared by surface-initiated AGET ATRP of MPC at 40 °C in ethanol at different polymerization times.
assumption of “fully stretched” PMPC chains,56,57 the degrees of polymerizations calculated from both particle sizes and conversions were approximately 70, 125, 250, and 345 for 12, 24, 45, and 70 h polymerizations, respectively. These results indicate that the polymerization proceeded with livingness, and the PMPC layer expanded as the polymerization proceeded. In the FT-IR spectra, peaks at 950, 1100, and 1250 cm−1 in PMPC-g-AuNPs were derived from the P−O, PO, and N−H bonds, respectively, in the phosphorylcholine moiety (SI Figure S4). Peaks at 800, 1750, and 3000 cm−1 were derived from C− Br, COO, and C−C (alkyl) bonds, respectively. From these spectra, it is clear that the PMPC-g-AuNPs prepared by surface-
Figure 4. Particle size distributions of PMPC-g-AuNPs prepared by surface-initiated AGET ATRP of MPC for 0, 12, 24, 45, and 70 h polymerizations in ethanol at 40 °C. D
dx.doi.org/10.1021/ac501322x | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
Figure 6. XPS spectra of N 1s (a) and P 2p (b) of PMPC-g-AuNPs.
initiated AGET ATRP contain both AuNPs-undecyl-Br and PMPC. These peaks were obtained with all PMPC-g-AuNPs prepared at different polymerization times. Moreover, XPS measurements revealed that P 2p and N 1s peaks did not appear in the AuNPs-undecyl-Br seed particles, but they did appear in PMPC-g-AuNPs, as shown in Figure 6. These results indicate that a PMPC layer successfully formed on the AuNPs after surface-initiated AGET ATRP. CRP Nanosensing Based on LSPR Property of PMPCg-AuNPs. In this study, CRP nanosensing was carried out utilizing the A/D parameter (Δ(A/D)/(A/D)0), where A and D stand for the aggregated state (550−700 nm) and the dispersed state (490−540 nm). Use of the A/D parameter is an effective technique for nanosensing systems of high sensitivity because it is based on the transformation of a peak-top shiftsensing system to peak-area ratio-metric sensing.58−60 Δ(A/ D)/(A/D)0 values provide an index of the electromagnetic field change around PMPC-g-AuNPs due to adsorption of CRP. This value (Δ(A/D)/(A/D)0) is defined as (A/D) − (A/D)0, where (A/D)0 and (A/D) are the values before and after the addition of various concentrations of CRP. When the AuNPs are adsorbed with target proteins and aggregate, the Δ(A/D)/ (A/D)0 value increases as a result of the change in the index of the electromagnetic field around the particles. The stability of PMPC-g-AuNPs prepared with different polymerization times were investigated in a 10 mM Tris−HCl buffer solution (pH 7.4) containing 140 mM NaCl and 20 mM CaCl2 by measuring the UV−vis spectra at 25 °C. The Δ(A/ D)/(A/D)0 values for PMPC-g-AuNPs at various stabilized times were constant (SI Figure S5), indicating that the PMPCg-AuNPs were highly stable in the buffer solution and the particles were suitable for CRP nanosensing (Scheme 2). We carried out CRP nanosensing based on the LSPR properties of PMPC-g-AuNPs prepared at 45 h polymerization in a 10 mM Tris−HCl buffer solution containing 140 mM NaCl and 20 mM CaCl2. CaCl2 was added because CRP characteristically shows calcium-dependent binding to phosphorylcholine,61 a condition chosen according to previously reported conditions.62 CRP nanosensing based on LSPR properties was carried out by adding various concentrations of CRP to PMPC-g-AuNPs in 10 mM Tris−HCl buffer solution (pH 7.4) containing 140 mM NaCl and 20 mM CaCl2. The LSPR spectra gradually changed with the addition of CRP (Figure 7a,b). The Δ(A/D)/(A/D)0 values clearly increased with the addition of CRP to the PMPC-g-AuNPs buffer solution. A limit of detection for CRP with this system was
Scheme 2. CRP Nanosensing Based on LSPR Property of PMPC-g-AuNPs by UV−Vis
calculated from the Δ(A/D)/(A/D)0 plot to be ∼50 ng/mL,63 which is comparable to detection limits achieved using antibody-based CRP sensors.24 The affinity constant estimated by curve-fitting (DeltaGraph 5.4.5v) was 4.98 × 108 M−1. To determine the effectiveness of PMPC as a recognition layer, we compared it with CRP sensing using POEOMA-gAuNPs that were prepared by surface-initiated AGET ATRP using polymerization conditions similar to those for PMPC-gAuNPs. The average particle size of the POEOMA-g-AuNPs was ∼175 nm after 48 h polymerization (∼11% conversion), as shown in SI Figure S7. The Δ(A/D)/(A/D)0 values changed slightly (SI Figure S8), but the variation was very small, even after the addition of 3000 ng/mL of CRP, which indicates that the POEOMA layer did not work as an effective recognition layer. Comparatively, the PMPC layer was determined to work well as a specific ligand for CRP recognition and could be quite important for future sensitive CRP detection. To apply optical nanosensing materials in actual clinical diagnoses, it is very important to suppress the nonspecific binding of unwanted proteins. We carried out a HSA binding test using PMPC-g-AuNPs in the same manner as CRP. No change was observed on the A/D parameter (Figure 7), and the values remained unchanged, even at concentrations of 100 μg/ mL. These results suggest PMPC-g-AuNPs avoid nonspecific binding while working as highly specific nanosensors for CRP based on LSPR properties. Notably, CRP nanosensing with this system was successfully carried out using only a UV−vis spectrometer, absent any special pretreatments and expensive equipment. E
dx.doi.org/10.1021/ac501322x | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
Figure 7. PMPC-g-AuNPs nanosensing responses for CRP (a,c (red)) and HAS (b,c (blue)) in 10 mM Tris−HCl buffer solution containing 140 mM NaCl and 20 mM CaCl2.
M−1 which was comparable to that in a buffer solution. These results indicate that our CRP nanosensing system worked effectively, even in diluted human serum. From these results, we have successfully demonstrated the highly sensitive and selective CRP detection ability of the present PMPC-g-AuNPs.
According to the results of the HSA sensing experiments, PMPC-g-AuNPs did not bind to HSA, even at a concentration of 100 μg/mL HSA, a concentration similar to the clinical sample concentration of 1% human serum. We also carried out CRP nanosensing in a 1% human serum solution. The Δ(A/ D)/(A/D)0 values, which were calculated from the UV−vis spectra obtained at various concentrations of CRP (SI Figure S9), increased gradually with the addition of CRP to a dispersion of PMPC-g-AuNPs in 1% human serum, as shown in Figure 8. The affinity constant was estimated to be 2.96 × 108
■
CONCLUSIONS We demonstrated highly sensitive and selective CRP nanosensing utilizing PMPC-g-AuNPs dispersed in a 10 mM Tris− HCl buffer solution containing 140 mM NaCl and 20 mM CaCl2. PMPC-g-AuNPs were synthesized by surface-initiated AGET ATRP of MPC with AuNPs-undecyl-Br as initiator seed particles, and the polymerization proceeded with livingness, as confirmed by DLS measurements. The LSPR spectra shifted when CRP was added to the PMPC-g-AuNPs dispersion, whereas no shift was observed when HSA was added, indicating the selective binding of PMPC-g-AuNPs toward CRP. More importantly, we successfully demonstrated CRP detection, even in 1% (w/w) human serum solution. The detection limit was ∼50 ng/mL, which is comparable to our previously reported value using biobased recognition materials. This level of sensitivity is sufficient for use in clinical diagnostics as hsCRP. It is expected that the optimization of the chain length or chain density of PMPC enables us to detect CRP more sensitively; therefore, investigation of such effects on the CRP sensing is ongoing in our laboratory. We believe this novel, highly sensitive nanosensing material (PMPC-g-AuNPs) would significantly contribute to the early
Figure 8. CRP response of PMPC-g-AuNPs in a 1% human serum diluted by 10 mM Tris−HCl buffer solution containing 140 mM NaCl and 20 mM CaCl2. F
dx.doi.org/10.1021/ac501322x | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
(25) Haes, A. J.; Zou, S. L.; Schatz, G. C.; Van Duyne, R. P. J. Phys. Chem. B 2004, 108, 6961. (26) Willets, K. A.; Van Duyne, R. P. Annu. Rev. Phys. Chem. 2007, 58, 267. (27) Haes, A. J.; Hall, W. P.; Chang, L.; Klein, W. L.; Van Duyne, R. P. Nano Lett. 2004, 4, 1029. (28) Zhao, J.; Zhang, X. Y.; Yonzon, C. R.; Haes, A. J.; Van Duyne, R. P. Nanomedicine 2006, 1, 219. (29) Mayer, K. M. H.; J, H. Chem. Rev. 2011, 111, 3828. (30) Kaplan, M. H.; Volanaki, Je J. Immunol. 1974, 112, 2135. (31) Volanaki, Je; Kaplan, M. H. Proc. Soc. Exp. Biol. Med. 1971, 136, 612. (32) Padlan, E. A.; Davies, D. R.; Rudikoff, S.; Potter, M. Immunochemistry 1976, 13, 945. (33) Volanakis, J. E.; Wirtz, K. W. A. Nature 1979, 281, 155. (34) Morinaga, T.; Ohkura, M.; Ohno, K.; Tsujii, Y.; Fukuda, T. Macromolecules 2007, 40, 1159. (35) Matyjaszewski, K.; Tsarevsky, N. V. Nat. Chem. 2009, 1, 276. (36) Siegwart, D. J.; Oh, J. K.; Matyjaszewski, K. Prog. Polym. Sci. 2012, 37, 18. (37) Lutz, J. F.; Ouchi, M.; Liu, D. R.; Sawamoto, M. Science 2013, 341, 628. (38) Thompson, D.; Pepys, M. B.; Wood, S. P. Structure Fold. & Des. 1999, 7, 169. (39) Ishihara, K.; Nomura, H.; Mihara, T.; Kurita, K.; Iwasaki, Y.; Nakabayashi, N. J. Biomed. Mater. Res. 1998, 39, 323. (40) Ishihara, K.; Ziats, N. P.; Tierney, B. P.; Nakabayashi, N.; Anderson, J. M. J. Biomed. Mater. Res. 1991, 25, 1397. (41) Feng, W.; Zhu, S. P.; Ishihara, K.; Brash, J. L. Biointerphases 2006, 1, 50. (42) Goda, T.; Kjall, P.; Ishihara, K.; Richter-Dahlfors, A.; Miyahara, Y. Adv. Healthcare Mater. 2014, DOI: 10.1002/adhm.201300625. (43) Kato, M.; Kamigaito, M.; Sawamoto, M.; Higashimura, T. Macromolecules 1995, 28, 1721. (44) Kamigaito, M.; Ando, T.; Sawamoto, M. Chem. Rev. 2001, 101, 3689. (45) Matyjaszewski, K.; Gnanou, Y.; Leibler, L. Macromolecular Engineering; Wiley-VCH, Weinheim 2007. (46) Kitayama, Y.; Kagawa, Y.; Minami, H.; Okubo, M. Langmuir 2010, 26, 7029. (47) Tanaka, T.; Okayama, M.; Kitayama, Y.; Kagawa, Y.; Okubo, M. Langmuir 2010, 26, 7843. (48) Zetterlund, P. B.; Kagawa, Y.; Okubo, M. Chem. Rev. 2008, 108, 3747. (49) Qiu, J.; Charleux, B.; Matyjaszewski, K. Prog. Polym. Sci. 2001, 26, 2083. (50) Monteiro, M. J.; Cunningham, M. F. Macromolecules 2012, 45, 4939. (51) Tsujii, Y.; Ohno, K.; Yamamoto, S.; Goto, A.; Fukuda, T. In Surface-Initiated Polymerization I; Jordan, R., Ed.; Springer: Berlin, New York, 2006; Vol. 197, p 1. (52) Sugihara, S.; Sugihara, K.; Armes, S. P.; Ahmad, H.; Lewis, A. L. Macromolecules 2010, 43, 6321. (53) Horn, M.; Matyjaszewski, K. Macromolecules 2013, 46, 3350. (54) Braunecker, W. A.; Tsarevsky, N. V.; Gennaro, A.; Matyjaszewski, K. Macromolecules 2009, 42, 6348. (55) Ma, Y. H.; Tang, Y. Q.; Billingham, N. C.; Armes, S. P.; Lewis, A. L.; Lloyd, A. W.; Salvage, J. P. Macromolecules 2003, 36, 3475. (56) Yamamoto, S.; Ejaz, M.; Tsujii, Y.; Matsumoto, M.; Fukuda, T. Macromolecules 2000, 33, 5602. (57) Yamamoto, S.; Ejaz, M.; Tsujii, Y.; Fukuda, T. Macromolecules 2000, 33, 5608. (58) de la Rica, R.; Fratila, R. M.; Szarpak, A.; Huskens, J.; Velders, A. H. Angew. Chem., Int. Ed. 2011, 50, 5703. (59) de la Rica, R.; Velders, A. H. Small 2011, 7, 66. (60) Uchida, A.; Kitayama, Y.; Takano, E.; Ooya, T.; Takeuchi, T. RSC Adv. 2013, 3, 25306. (61) Sui, S. F.; Sun, Y. T.; Mi, L. Z. Biophys. J. 1999, 76, 333.
clinical diagnosis of inflammation, coronary heart disease, and cancer. From the results of our study, we see great potential in combining CLRP as the construction method for artificial receptors and AuNPs (or other noble metal nanoparticles) as signaling materials to create various organic/inorganic hybrid nanomaterials for sensing, imaging, and detection. This strategy opens a new door for further advancement in medical diagnosis and preventive medicine.
■
ASSOCIATED CONTENT
S Supporting Information *
Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.”
■
AUTHOR INFORMATION
Corresponding Author
*E-mail: [email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS The authors gratefully thank to Prof. Kazuhiko Ishihara and Dr. Kyoko Fukazawa at the University of Tokyo for kindly supplying MPC.
■
REFERENCES
(1) Quershi, A.; Gurbuz, Y.; Kang, W. P.; Davidson, J. L. Biosens. Bioelectron. 2009, 25, 877. (2) Ridker, P. M.; Wilson, P. W. F.; Grundy, S. M. Circulation 2004, 109, 2818. (3) Rifai, N.; Ridker, P. M. Clin. Chem. 2001, 47, 403. (4) Kuby, J. Cytokine Secretion and Biological Activity of TH1 and TH2 Subsets; WH Freeman & Co.: New York, 1994. (5) Tietz, N. W. C-Reactive Protein; WB Saunders Co.: Philadelphia, 1987. (6) Oliveira, E. B.; Gotschlich, E. C.; Liu, T. Y. J. Biol. Chem. 1979, 254, 489. (7) Pepys, M. B.; Hirschfield, G. M. J. Clin. Invest. 2003, 111, 1805. (8) Volanakis, J. E. Mol. Immunol. 2001, 38, 189. (9) Pepys, M. B.; Baltz, M. L. Adv. Immunol. 1983, 34, 141. (10) Conen, D.; Ridker, P. M. Biomark. Med. 2007, 1, 229. (11) Ridker, P. M. Clin. Chem. 2009, 55, 209. (12) Ridker, P. M. J. Am. Coll. Cardiol. 2007, 49, 2129. (13) Casas, J. P.; Shah, T.; Hingorani, A. D.; Danesh, J.; Pepys, M. B. J. Intern. Med. 2008, 264, 295. (14) Hansson, G. K. N. Engl. J. Med. 2005, 352, 1685. (15) Bassuk, S. S.; Rifai, N.; Ridker, P. M. Curr. Probl. Cardiol. 2004, 29, 439. (16) Haverkate, F.; Thompson, S. G.; Pyke, S. D. M.; Gallimore, J. R.; Pepys, M. B. Lancet 1997, 349, 462. (17) Kuller, L. H.; Tracy, R. P.; Shaten, J.; Meilahn, E. N. Am. J. Epidemiol. 1996, 144, 537. (18) Kitahashi, S.; Tatsumi, N.; Tagawa, S.; Matsui, T.; Higashihata, M.; Shintaku, H.; Tomoda, S.; Tsuda, I. J. Autom. Chem. 1998, 20, 195. (19) Kriz, K.; Ibraimi, F.; Lu, M.; Hansson, L. O.; Kriz, D. Anal. Chem. 2005, 77, 5920. (20) Meyer, M. H. F.; Hartmann, M.; Keusgen, M. Biosens. Bioelectron. 2006, 21, 1987. (21) Tsai, H. Y.; Hsu, C. F.; Chiu, I. W.; Fuh, C. B. Anal. Chem. 2007, 79, 8416. (22) Macy, E. M.; Hayes, T. E.; Tracy, R. P. Clin. Chem. 1997, 43, 52. (23) Choi, H. W.; Sakata, Y.; Kurihara, Y.; Ooya, T.; Takeuchi, T. Anal. Chim. Acta 2012, 728, 64. (24) Murata, A.; Ooya, T.; Takeuchi, T. Biosens. Bioelectron. 2013, 43, 45. G
dx.doi.org/10.1021/ac501322x | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
(62) Rowe, I. F.; Soutar, A. K.; Trayner, I. M.; Baltz, M. L.; Debeer, F. C.; Walker, L.; Bowyer, D.; Herbert, J.; Feinstein, A.; Pepys, M. B. J. Exp. Med. 1984, 159, 604. (63) Kirchmer, C. J. In Action in Analytical Chemistry; Currie, L. A., Ed.; American Chemical Society: Washington, DC, 1987; Vol. 361, p 78.
H
dx.doi.org/10.1021/ac501322x | Anal. Chem. XXXX, XXX, XXX−XXX
Localized Surface Plasmon Resonance Nanosensing of C‑Reactive Protein with Poly(2-methacryloyloxyethyl phosphorylcholine)Grafted Gold Nanoparticles Prepared by Surface-Initiated Atom Transfer Radical Polymerization 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: Highly sensitive and selective protein nanosensing based on localized surface plasmon resonance (LSPR) of gold nanoparticles (AuNPs) on which polymerized specific ligands were grafted as an artificial protein recognition layer for the target protein were demonstrated. As a model, optical nanosensing for Creactive protein (CRP), a known biomarker for chronic inflammation that predicts the risk of arteriosclerosis or heart attacks, was achieved by measuring the shift of LSPR spectra derived from the change of permittivity of poly(2-methacryloyloxyethyl phosphorylcholine)-grafted AuNPs (PMPC-g-AuNPs) upon interacting with CRP, in which the PMPC-g-AuNPs layer were grafted on AuNPs by surface-initiated atom transfer radical polymerization (ATRP). This nanosensing system was effective even for detecting CRP concentrations in a human serum solution diluted to 1% (w/w), at which point a limit of detection was ∼50 ng/mL and nonspecific adsorption of other proteins was negligible. The nanosensing system using specific ligand-grafted AuNPs has several strengths, such as low preparation cost, avoiding the need for expensive instruments, no necessary complex pretreatments, and high stability, because it does not contain biobased molecules. We believe this novel synthetic route for protein nanosensors, composed of AuNPs and a polymerized specific ligand utilizing surface-initiated controlled/living radical polymerization, will provide a foundation for the design and synthesis of nanosensors targeting various other biomarker proteins, paving the way for future advances in the field of biosensing. limits below 0.12 μg/mL.19−22 In our previous works, we have also demonstrated with success the highly sensitive detection of CRP by surface plasmon resonance (SPR) and reflectometric interference spectroscopy-based immunosensors.23,24 Localized surface plasmon resonance (LSPR)-based nanosensing has gained a significant reputation as a highly sensitive detection method. Gold nanoparticles (AuNPs) have great potential for use as optical nanosensing materials because they show localized surface plasmon resonance (LSPR) as specific wavelengths of visible light, whose extinction wavelength due to LSPR is shifted, depending on the permeability of the surrounding of the AuNPs.25−29 An LSPR-based nanosensing system has multiple advantages, such as avoiding the necessity of expensive instruments, ease-of-use, label-free detection, and short operation time. Although previous papers have reported successful sensing systems using antibodies (anti-CRP),23,24 these antibodies are both expensive and unstable, and artificial sensing materials provide a stable and attractive alternative. The use of polymerbased artificial recognition materials is especially attractive and
C
oronary heart disease is one of the leading cause of death in developed countries and has several underlying cause, one of which is atherosclerosis.1−3 Systemic inflammation plays a capital role in atherothrombtic inception and progression, during which cytokines produce acute phase reactants such as C-reactive protein (CRP) in the bodies of patients suffering from inflammation, making CRP levels a good indicator of inflammation, cancer, and tissue destruction.4−12 Conventionally, CRP detection for the clinical diagnosis of acute inflammation is carried out using one of several methods, such as immune nephelometry or latex aggregation; however, these methods cannot detect CRP levels below a threshold of 0.3 μg/mL and, therefore, are not sufficiently sensitive enough to detect cardiovascular risks. Recently, a highly sensitive CRP assay (hs-CRP), which can detect CRP quantitatively at concentrations under 0.15 μg/mL, has attracted much attention as powerful a predictor of future first stage coronary heart disease in presently healthy men and women3,13−17 as well as a diagnostic tool for inflammation detection in newborns, whose CRP concentrations are generally low compared with those in adults.18 To determine the presence of trace amount of CRP with high sensitivity, previous studies have reported the use of enzyme-linked immunosorbent assays and magnetic permeability assays, which were reported to have achieved detection © XXXX American Chemical Society
Received: April 2, 2014 Accepted: May 7, 2014
A
dx.doi.org/10.1021/ac501322x | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
and HAuCl4·2H2O (36 mg, 87.4 μmol) aqueous solution (water, 190 mg) was added to the ethanol solution. The water/ ethanol solution was then stirred vigorously, and 0.5 M NaBH4 aqueous solution (1.9 g) was slowly added and reacted for 15 h at room temperature. The obtained Br-functionalized AuNPs with undecyl linker (AuNPs-undecyl-Br) were washed by centrifugation with methanol/water (2:1 v/v) twice and with ethanol/hexane (5:1 v/v) three times. Finally, AuNPs-undecylBr was dispersed in ethanol. The Br-functionalized AuNPs with the ethyl linker (AuNPs-ethyl-Br) was also synthesized using bis[2-(2-bromoisobutyryloxy)-ethyl] disulfide by the same protocol as AuNPs-undecyl-Br. Surface-Initiated AGET ATRP of MPC with Br-AuNPs. MPC (236 mg, 800 μmol), CuBr2 (0.78 mg, 25 μmol), PMDETA (0.84 mg, 25 μmol) were dissolved in the AuNPsundecyl-Br dispersion in ethanol (2 g, solid content of AuNPsundecyl-Br: 0.019 wt %), in which DMSO was added as an internal standard for conversion measurement by 1H NMR. After three freeze−pump−thaw cycles, ascorbic acid (2.2 mg, 12.5 μmol) was dissolved in the ethanol solution (100 μL) and was added to the solution to reduce the copper species and start surface-initiated activator generated by electron transfer (AGET) ATRP. The polymerization was carried out at 40 °C in a water bath. After the polymerization, the obtained PMPCg-AuNPs were washed by centrifugation with ethanol twice and water three times. The washed PMPC-g-AuNPs were dispersed in 10 mM Tris−HCl buffer solution (pH 7.4) with 140 mM NaCl and 20 mM CaCl2. Surface-Initiated AGET ATRP of OEOMA with BrAuNPs. The poly(OEOMA) (POEOMA)-grafted AuNPs (POEOMA-g-AuNPs) were synthesized by surface-initiated AGET ATRP under a procedure similar to that of the protocol for the synthesis of PMPC-g-AuNPs. Synthesis of AuNPs by Citrate. HAuCl4·2H2O (60 mg, 151 μmol) was dissolved in water (450 mL), and the aqueous solution was refluxed. Sodium citrate (150 mg, 581 μmol) aqueous solution (15 mL) was added to the HAuCl4 aqueous solution, and the mixture was refluxed for 20 min. Calculation of the A/D Parameter. The A/D parameter is given by a ratio of the integral of the spectra of PMPC-gAuNPs in the range from 550 to 700 nm (A value) to that of the spectra in the range from 490 to 540 nm (D value). The Δ(A/D) value was given by (A/D) − (A/D)0, and a relative A/ D change against an initial A/D parameter was calculated by Δ(A/D)/(A/D)0, where (A/D)0 was an initial A/D parameter CRP Nanosensing with PMPC-g-AuNPs in 10 mM Tris−HCl buffer solution (pH 7.4) with 140 mM NaCl and 20 mM CaCl2. PMPC-g-AuNPs (solids content: 0.08 wt %) dispersed in 10 mM Tris−HCl buffer solution with 140 mM NaCl and 20 mM CaCl2 (1.5 mL) was diluted with the same volume of the buffer solution, which was used as a sample for absorbance measurement and stirred for 60 min at 25 °C. Various concentrations (10, 100, 1000 μg/mL) of CRP dissolved in the buffer (1.2−3.0 μL) were added stepwise to the AuNPs dispersed in the buffer, and the final concentrations of the CRP sample in each step were 0, 10, 50, 100, 500, 1000, and 3000 ng/mL. UV−vis spectra were measured after 30 min of incubation at 25 °C. To confirm the nonspecific binding to the PMPC-g-AuNPs, HSA was used as a reference protein. Each measurement was performed three times to check the repeatability. CRP Nanosensing with PMPC-g-AuNPs in Human Serum. PMPC-g-AuNPs (solids content: 0.08 wt %) dispersed
allows for important advances in the preparation of CRP sensing materials because unlike those based on antibodies and enzymes, polymer-based materials can be prepared easily and at a low cost. CRP is known to bind readily to phosphorylcholine groups in the presence of Ca2+ ions,30−38 so we selected 2methacryloyloxyethyl phosphorylcholine (MPC) for use as a polymerizable specific ligand toward CRP. Notably, Ishihara et al. clarified in a series of previous works that poly(MPC) (PMPC) can effectively suppress nonspecific binding with other proteins present with CRP.39−42 Therefore, PMPC should be a great candidate for use as an effective artificial receptor for selective CRP sensing. In this study, we demonstrated the synthesis and performance of a novel, highly sensitive CRP nanosensing material based on AuNPs grafted with poly(2-methacryloyloxyethyl phosphorylcholine) (PMPC) as an artificial CRP recognition layer (PMPC-g-AuNPs) prepared by surface-initiated atom transfer radical polymerization (ATRP).34,43−52 CRP sensing ability of obtained PMPC-g-AuNPs was demonstrated using UV−vis in a buffer solution. The nonspecific binding properties were also investigated with UV−vis using human serum albumin (HSA) as a reference protein. Finally, to confirm the effectiveness of highly sensitive detection of CRP using PMPCg-AuNPs, sensing experiments were carried out in a diluted human serum solution. Our proposed nanosensing system has several advantages compared with previously reported methods, such as avoiding the necessity of specific or complex pretreatment, the use of inexpensive instruments, and avoiding the need for expensive and unstable biobased molecules.
■
EXPERIMENTAL PART Materials. HAuCl4, NaBH4, NaCl, sodium citrate, CuBr2, Lascorbic acid, and human serum albumin (HSA) were purchased from Nacalai Tesque (Kyoto, Japan). Bis[2-(2bromoisobutyryloxy)-ethyl] disulfide, bis[2-(2-bromoisobutyryloxy)-undecyl] disulfide, N,N,N′,N″,N″ pentamethyldienediethylenetriamine (PMDETA), were purchased from SigmaAldrich (USA). Methanol, ethanol, hexane, DMSO, 2,2′azobis(isobutyronitril) (AIBN) were purchased from Wako Pure Chemical Industries (Osaka, Japan). Deionized water used was obtained from a Millipore Milli-Q purification system. MPC was kindly supplied by Prof. Kazuhiko Ishihara at the University of Tokyo. Oligo(ethylene oxide) methyl ether methacrylate (OEOMA) (degree of polymerization of ethylene oxide ∼ 8) was purchased from NOF Corporation (Tokyo, Japan). Characterization. 1H NMR spectra were measured using 300 MHz FT-NMR apparatus (JNM-LA300 FT NMR system, JEOL Ltd., Tokyo, Japan). FT-IR measurements were carried out by KBr method using (Varian 660 KU-IR, Agilent Inc., California, USA). XPS measurements were carried out using a PHI X-tool (ULVAC PHI, Inc., Kanagawa, Japan), where the takeoff angle was 45° and X-ray condition was 20 kV and 98 W with Al Kα. The particle sizes and morphologies were observed using a JEM-1230 electron microscope (JEOL Ltd., Tokyo, Japan). The particle size and size distribution were obtained using a dynamic light-scattering system (Zetasizer Nano ZS, Malvern Instruments Ltd., U.K.). UV−vis spectral measurements were performed using a V-560 spectrophotometer (JASCO Ltd., Japan). Preparation of Br-Functionalized AuNPs. A typical procedure is as follows: Bis[2-(2-bromoisobutyryloxy)undecyl]disulfide (15.3 mg, 21.7 μmol) was dissolved in ethanol (7.7 g), B
dx.doi.org/10.1021/ac501322x | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
in 10 mM Tris−HCl buffer solution with 140 mM NaCl and 20 mM CaCl2 (3.0 mL) was added to the same volume of 2 wt % human serum buffer solution. The dispersion was used as a sample for absorbance measurement and stirred for 60 min to maintain the temperature at 25 °C. Various concentrations (10, 100, 1000 μg/mL) of CRP dissolved in the buffer (1.2−3.0 μL) were added stepwise to the AuNPs dispersed in the 100-timesdiluted human serum, and the final concentrations of the CRP sample in each step were 0, 10, 50, 100, 500, 1000, and 3000 ng/mL. UV−vis spectra were measured after 30 min of incubation at 25 °C. To confirm the nonspecific binding to the PMPC-g-AuNPs, HSA was used as a reference protein. Each measurement was performed three times to check the repeatability. Determination of Binding Constant. The binding constant of PMPC-g-AuNPs toward CRP was determined by curve-fitting using DeltaGraph 5.4.5v. The fitting equation is shown below, which was generally used for determination of the binding constant of formation of the 1:1 complex, in which we hypothesized the binding between PMPC-g-AuNP and CRP occurred as one-to-one binding. Y = [(1 + KG + KH ) − D × 2KG
Figure 2. UV−vis spectra of AuNPs-undecyl-Br (a) and AuNPs-ethylBr (b) dispersed in ethanol.
obtained for AuNPs-undecyl-Br and AuNPs-ethyl-Br dispersions, respectively. The XPS measurements for S 2p and Br 3d were carried out to investigate whether the AuNPs chemically bonded with the disulfide derivatives (bis[2-(2′-bromoisobutyryloxy)undecyl]disulfide or bis[2-(2′-bromoisobutyryloxy)ethyl]disulfide) (Supporting Information (SI) Figure S1). The S 2p peak, which was not observed with AuNPs prepared with citric acid (AuNPs-citrate) as a reference material, appeared clearly with both AuNPs-undecyl-Br and AuNPs-ethyl-Br. Br 3d peaks also appeared with both AuNPs-undecyl-Br and AuNPs-ethyl-Br but were absent with AuNPs-citrate. These results indicate that AuNPs-undecyl-Br and AuNPs-ethyl-Br were chemically conjugated with bis[2-(2′-bromoisobutyryloxy)undecyl]disulfide and bis[2-(2′-bromoisobutyryloxy)ethyl]disulfide, respectively, as stabilizing agents and ATRP initiators. The particle sizes of AuNPs-undecyl-Br and AuNPs-ethyl-Br were measured by dynamic light scattering (DLS), as shown in Figure 3. Despite the stabilization of both AuNPs by disulfide derivatives, the particle sizes of the obtained AuNPs were clearly different: the z-average diameter (dz) of AuNPs-undecylBr was ∼10 nm, whereas the dz value of AuNPs-ethyl-Br was ∼340 nm. To investigate the morphologies of AuNPs-undecylBr and AuNPs-ethyl-Br, their TEM imaging was carried out. In TEM images of AuNPs-undecyl-Br, the AuNPs existed individually, and no aggregation was observed (SI Figure S2). The particle size of AuNPs-undecyl-Br measured from the TEM image was smaller than that obtained by DLS (∼10 nm). This is because the particle size obtained by DLS is the hydrodynamic particle size containing the stabilizing 2-(2bromoisobutyryloxy)ethyl thiol layer, whereas the particles used in TEM imaging were absent this layer. TEM images of AuNPsethyl-Br revealed large AuNPs (>1 μm) formed by the coagulation of small AuNPs. These results indicated that AuNPs-ethyl-Br initially formed as small particles stabilized by bis[2-(2′-bromoisobutyryloxy)ethyl]disulfide, but the particles had low stability, resulting in significant coagulation. From these results, we selected bis[2-(2′-bromoisobutyryloxy)undecyl]disulfide as the ATRP-initiator and stabilizer of AuNPs. Surface-Initiated AGET ATRP for the Preparation of PMPC-g-AuNPs. To prepare the PMPC-g-AuNPs, we carried out surface-initiated AGET ATRP of MPC in ethanol at 40 °C (Scheme 1). It is well-known that the solvent significantly affects the polymerization rate and controls the molecular weight distribution in ATRP systems.53,54 The ATRP polymerization rate of MPC was accelerated, but the control/livingness worsened as the polarity of the solvent increased.52,55 We selected ethanol as our solvent in this work because it offered
(1 + KG + KH ) − 4K 2HG ]
where Y is ΔA/D/(A/D)0, K is the affinity constant, H is the concentration of PMPC-g-AuNPs, G is the CRP concentration, and D is the maximum ΔA/D/(A/D)0.
■
RESULTS AND DISCUSSION Synthesis of Bromo-Functionalized AuNPs (AuNPsBr). Br-functionalized AuNPs were synthesized by reduction of HAuCl4 in the presence of bromoisobutyryloxydisulfide derivatives with different alkyl chain lengths (bis[2-(2bromoisobutyryloxy)undecyl]disulfide and bis[2-(2bromoisobutyryloxy)ethyl]disulfide) in an ethanol/water solution (Figure 1). When a NaBH4 aqueous solution was added
Figure 1. Structures of disulfide derivatives used in this study: (a) bis[2-(2′-bromoisobutyryloxy)undecyl]disulfide and (b) bis[2-(2′bromoisobutyryloxy)ethyl]disulfide.
to the HAuCl4 solution, dark purple particles immediately precipitated. The obtained precipitants were washed with methanol/water (2:1 v/v) and ethanol/hexane (5:1 v/v) solutions, followed by centrifugation. Figure 2 shows UV−vis spectra of the obtained AuNPs-Br dispersions in ethanol using AuNPs-undecyl-Br or AuNPs-ethyl-Br, respectively. The AuNPs-undecyl-Br dispersion showed LSPR activity with an extinction peak at 520 nm, but an extinction peak derived from LSPR did not appear for the AuNPs-ethyl-Br dispersion. The spectra were in accordance with the red and black colors C
dx.doi.org/10.1021/ac501322x | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
Figure 3. Particle size distribution of AuNPs-undecyl-Br (a) and AuNPs-ethyl-Br (b) in ethanol measured by DLS.
Scheme 1. Surface-Initiated ATRP of MPC with AuNPsundecyl-Br As Initiator Seed Particles
an appropriate control/livingness in the ATRP of MPC while allowing the AuNPs to maintain high stability without coagulating both before and after polymerization. This is because ethanol is a good solvent for both bis[2-(2bromoisobutyryloxy)undecyl]disulfide and PMPC, which leads to steric repulsion between the individual AuNPs. The time−conversion plot of the polymerization obtained from 1H NMR analysis is shown in Figure S3 in the Supporting Information. Conversion increased with polymerization time, which indicates that the Br group in AuNPs-undecyl-Br worked as the initiating group. The first-order plot kept linearity during the polymerization with passing through the origin, indicating that the propagating radical concentration were constant (SI Figure S3(b)). Furthermore, the particle size distribution of PMPC-gAuNPs at various polymerization times clearly shifted toward larger particle sizes as polymerization progressed (Figure 4), and the dz values also clearly increased (Figure 5). In the
Figure 5. z-Average particle size (da) values of PMPC-g-AuNPs prepared by surface-initiated AGET ATRP of MPC at 40 °C in ethanol at different polymerization times.
assumption of “fully stretched” PMPC chains,56,57 the degrees of polymerizations calculated from both particle sizes and conversions were approximately 70, 125, 250, and 345 for 12, 24, 45, and 70 h polymerizations, respectively. These results indicate that the polymerization proceeded with livingness, and the PMPC layer expanded as the polymerization proceeded. In the FT-IR spectra, peaks at 950, 1100, and 1250 cm−1 in PMPC-g-AuNPs were derived from the P−O, PO, and N−H bonds, respectively, in the phosphorylcholine moiety (SI Figure S4). Peaks at 800, 1750, and 3000 cm−1 were derived from C− Br, COO, and C−C (alkyl) bonds, respectively. From these spectra, it is clear that the PMPC-g-AuNPs prepared by surface-
Figure 4. Particle size distributions of PMPC-g-AuNPs prepared by surface-initiated AGET ATRP of MPC for 0, 12, 24, 45, and 70 h polymerizations in ethanol at 40 °C. D
dx.doi.org/10.1021/ac501322x | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
Figure 6. XPS spectra of N 1s (a) and P 2p (b) of PMPC-g-AuNPs.
initiated AGET ATRP contain both AuNPs-undecyl-Br and PMPC. These peaks were obtained with all PMPC-g-AuNPs prepared at different polymerization times. Moreover, XPS measurements revealed that P 2p and N 1s peaks did not appear in the AuNPs-undecyl-Br seed particles, but they did appear in PMPC-g-AuNPs, as shown in Figure 6. These results indicate that a PMPC layer successfully formed on the AuNPs after surface-initiated AGET ATRP. CRP Nanosensing Based on LSPR Property of PMPCg-AuNPs. In this study, CRP nanosensing was carried out utilizing the A/D parameter (Δ(A/D)/(A/D)0), where A and D stand for the aggregated state (550−700 nm) and the dispersed state (490−540 nm). Use of the A/D parameter is an effective technique for nanosensing systems of high sensitivity because it is based on the transformation of a peak-top shiftsensing system to peak-area ratio-metric sensing.58−60 Δ(A/ D)/(A/D)0 values provide an index of the electromagnetic field change around PMPC-g-AuNPs due to adsorption of CRP. This value (Δ(A/D)/(A/D)0) is defined as (A/D) − (A/D)0, where (A/D)0 and (A/D) are the values before and after the addition of various concentrations of CRP. When the AuNPs are adsorbed with target proteins and aggregate, the Δ(A/D)/ (A/D)0 value increases as a result of the change in the index of the electromagnetic field around the particles. The stability of PMPC-g-AuNPs prepared with different polymerization times were investigated in a 10 mM Tris−HCl buffer solution (pH 7.4) containing 140 mM NaCl and 20 mM CaCl2 by measuring the UV−vis spectra at 25 °C. The Δ(A/ D)/(A/D)0 values for PMPC-g-AuNPs at various stabilized times were constant (SI Figure S5), indicating that the PMPCg-AuNPs were highly stable in the buffer solution and the particles were suitable for CRP nanosensing (Scheme 2). We carried out CRP nanosensing based on the LSPR properties of PMPC-g-AuNPs prepared at 45 h polymerization in a 10 mM Tris−HCl buffer solution containing 140 mM NaCl and 20 mM CaCl2. CaCl2 was added because CRP characteristically shows calcium-dependent binding to phosphorylcholine,61 a condition chosen according to previously reported conditions.62 CRP nanosensing based on LSPR properties was carried out by adding various concentrations of CRP to PMPC-g-AuNPs in 10 mM Tris−HCl buffer solution (pH 7.4) containing 140 mM NaCl and 20 mM CaCl2. The LSPR spectra gradually changed with the addition of CRP (Figure 7a,b). The Δ(A/D)/(A/D)0 values clearly increased with the addition of CRP to the PMPC-g-AuNPs buffer solution. A limit of detection for CRP with this system was
Scheme 2. CRP Nanosensing Based on LSPR Property of PMPC-g-AuNPs by UV−Vis
calculated from the Δ(A/D)/(A/D)0 plot to be ∼50 ng/mL,63 which is comparable to detection limits achieved using antibody-based CRP sensors.24 The affinity constant estimated by curve-fitting (DeltaGraph 5.4.5v) was 4.98 × 108 M−1. To determine the effectiveness of PMPC as a recognition layer, we compared it with CRP sensing using POEOMA-gAuNPs that were prepared by surface-initiated AGET ATRP using polymerization conditions similar to those for PMPC-gAuNPs. The average particle size of the POEOMA-g-AuNPs was ∼175 nm after 48 h polymerization (∼11% conversion), as shown in SI Figure S7. The Δ(A/D)/(A/D)0 values changed slightly (SI Figure S8), but the variation was very small, even after the addition of 3000 ng/mL of CRP, which indicates that the POEOMA layer did not work as an effective recognition layer. Comparatively, the PMPC layer was determined to work well as a specific ligand for CRP recognition and could be quite important for future sensitive CRP detection. To apply optical nanosensing materials in actual clinical diagnoses, it is very important to suppress the nonspecific binding of unwanted proteins. We carried out a HSA binding test using PMPC-g-AuNPs in the same manner as CRP. No change was observed on the A/D parameter (Figure 7), and the values remained unchanged, even at concentrations of 100 μg/ mL. These results suggest PMPC-g-AuNPs avoid nonspecific binding while working as highly specific nanosensors for CRP based on LSPR properties. Notably, CRP nanosensing with this system was successfully carried out using only a UV−vis spectrometer, absent any special pretreatments and expensive equipment. E
dx.doi.org/10.1021/ac501322x | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
Figure 7. PMPC-g-AuNPs nanosensing responses for CRP (a,c (red)) and HAS (b,c (blue)) in 10 mM Tris−HCl buffer solution containing 140 mM NaCl and 20 mM CaCl2.
M−1 which was comparable to that in a buffer solution. These results indicate that our CRP nanosensing system worked effectively, even in diluted human serum. From these results, we have successfully demonstrated the highly sensitive and selective CRP detection ability of the present PMPC-g-AuNPs.
According to the results of the HSA sensing experiments, PMPC-g-AuNPs did not bind to HSA, even at a concentration of 100 μg/mL HSA, a concentration similar to the clinical sample concentration of 1% human serum. We also carried out CRP nanosensing in a 1% human serum solution. The Δ(A/ D)/(A/D)0 values, which were calculated from the UV−vis spectra obtained at various concentrations of CRP (SI Figure S9), increased gradually with the addition of CRP to a dispersion of PMPC-g-AuNPs in 1% human serum, as shown in Figure 8. The affinity constant was estimated to be 2.96 × 108
■
CONCLUSIONS We demonstrated highly sensitive and selective CRP nanosensing utilizing PMPC-g-AuNPs dispersed in a 10 mM Tris− HCl buffer solution containing 140 mM NaCl and 20 mM CaCl2. PMPC-g-AuNPs were synthesized by surface-initiated AGET ATRP of MPC with AuNPs-undecyl-Br as initiator seed particles, and the polymerization proceeded with livingness, as confirmed by DLS measurements. The LSPR spectra shifted when CRP was added to the PMPC-g-AuNPs dispersion, whereas no shift was observed when HSA was added, indicating the selective binding of PMPC-g-AuNPs toward CRP. More importantly, we successfully demonstrated CRP detection, even in 1% (w/w) human serum solution. The detection limit was ∼50 ng/mL, which is comparable to our previously reported value using biobased recognition materials. This level of sensitivity is sufficient for use in clinical diagnostics as hsCRP. It is expected that the optimization of the chain length or chain density of PMPC enables us to detect CRP more sensitively; therefore, investigation of such effects on the CRP sensing is ongoing in our laboratory. We believe this novel, highly sensitive nanosensing material (PMPC-g-AuNPs) would significantly contribute to the early
Figure 8. CRP response of PMPC-g-AuNPs in a 1% human serum diluted by 10 mM Tris−HCl buffer solution containing 140 mM NaCl and 20 mM CaCl2. F
dx.doi.org/10.1021/ac501322x | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
(25) Haes, A. J.; Zou, S. L.; Schatz, G. C.; Van Duyne, R. P. J. Phys. Chem. B 2004, 108, 6961. (26) Willets, K. A.; Van Duyne, R. P. Annu. Rev. Phys. Chem. 2007, 58, 267. (27) Haes, A. J.; Hall, W. P.; Chang, L.; Klein, W. L.; Van Duyne, R. P. Nano Lett. 2004, 4, 1029. (28) Zhao, J.; Zhang, X. Y.; Yonzon, C. R.; Haes, A. J.; Van Duyne, R. P. Nanomedicine 2006, 1, 219. (29) Mayer, K. M. H.; J, H. Chem. Rev. 2011, 111, 3828. (30) Kaplan, M. H.; Volanaki, Je J. Immunol. 1974, 112, 2135. (31) Volanaki, Je; Kaplan, M. H. Proc. Soc. Exp. Biol. Med. 1971, 136, 612. (32) Padlan, E. A.; Davies, D. R.; Rudikoff, S.; Potter, M. Immunochemistry 1976, 13, 945. (33) Volanakis, J. E.; Wirtz, K. W. A. Nature 1979, 281, 155. (34) Morinaga, T.; Ohkura, M.; Ohno, K.; Tsujii, Y.; Fukuda, T. Macromolecules 2007, 40, 1159. (35) Matyjaszewski, K.; Tsarevsky, N. V. Nat. Chem. 2009, 1, 276. (36) Siegwart, D. J.; Oh, J. K.; Matyjaszewski, K. Prog. Polym. Sci. 2012, 37, 18. (37) Lutz, J. F.; Ouchi, M.; Liu, D. R.; Sawamoto, M. Science 2013, 341, 628. (38) Thompson, D.; Pepys, M. B.; Wood, S. P. Structure Fold. & Des. 1999, 7, 169. (39) Ishihara, K.; Nomura, H.; Mihara, T.; Kurita, K.; Iwasaki, Y.; Nakabayashi, N. J. Biomed. Mater. Res. 1998, 39, 323. (40) Ishihara, K.; Ziats, N. P.; Tierney, B. P.; Nakabayashi, N.; Anderson, J. M. J. Biomed. Mater. Res. 1991, 25, 1397. (41) Feng, W.; Zhu, S. P.; Ishihara, K.; Brash, J. L. Biointerphases 2006, 1, 50. (42) Goda, T.; Kjall, P.; Ishihara, K.; Richter-Dahlfors, A.; Miyahara, Y. Adv. Healthcare Mater. 2014, DOI: 10.1002/adhm.201300625. (43) Kato, M.; Kamigaito, M.; Sawamoto, M.; Higashimura, T. Macromolecules 1995, 28, 1721. (44) Kamigaito, M.; Ando, T.; Sawamoto, M. Chem. Rev. 2001, 101, 3689. (45) Matyjaszewski, K.; Gnanou, Y.; Leibler, L. Macromolecular Engineering; Wiley-VCH, Weinheim 2007. (46) Kitayama, Y.; Kagawa, Y.; Minami, H.; Okubo, M. Langmuir 2010, 26, 7029. (47) Tanaka, T.; Okayama, M.; Kitayama, Y.; Kagawa, Y.; Okubo, M. Langmuir 2010, 26, 7843. (48) Zetterlund, P. B.; Kagawa, Y.; Okubo, M. Chem. Rev. 2008, 108, 3747. (49) Qiu, J.; Charleux, B.; Matyjaszewski, K. Prog. Polym. Sci. 2001, 26, 2083. (50) Monteiro, M. J.; Cunningham, M. F. Macromolecules 2012, 45, 4939. (51) Tsujii, Y.; Ohno, K.; Yamamoto, S.; Goto, A.; Fukuda, T. In Surface-Initiated Polymerization I; Jordan, R., Ed.; Springer: Berlin, New York, 2006; Vol. 197, p 1. (52) Sugihara, S.; Sugihara, K.; Armes, S. P.; Ahmad, H.; Lewis, A. L. Macromolecules 2010, 43, 6321. (53) Horn, M.; Matyjaszewski, K. Macromolecules 2013, 46, 3350. (54) Braunecker, W. A.; Tsarevsky, N. V.; Gennaro, A.; Matyjaszewski, K. Macromolecules 2009, 42, 6348. (55) Ma, Y. H.; Tang, Y. Q.; Billingham, N. C.; Armes, S. P.; Lewis, A. L.; Lloyd, A. W.; Salvage, J. P. Macromolecules 2003, 36, 3475. (56) Yamamoto, S.; Ejaz, M.; Tsujii, Y.; Matsumoto, M.; Fukuda, T. Macromolecules 2000, 33, 5602. (57) Yamamoto, S.; Ejaz, M.; Tsujii, Y.; Fukuda, T. Macromolecules 2000, 33, 5608. (58) de la Rica, R.; Fratila, R. M.; Szarpak, A.; Huskens, J.; Velders, A. H. Angew. Chem., Int. Ed. 2011, 50, 5703. (59) de la Rica, R.; Velders, A. H. Small 2011, 7, 66. (60) Uchida, A.; Kitayama, Y.; Takano, E.; Ooya, T.; Takeuchi, T. RSC Adv. 2013, 3, 25306. (61) Sui, S. F.; Sun, Y. T.; Mi, L. Z. Biophys. J. 1999, 76, 333.
clinical diagnosis of inflammation, coronary heart disease, and cancer. From the results of our study, we see great potential in combining CLRP as the construction method for artificial receptors and AuNPs (or other noble metal nanoparticles) as signaling materials to create various organic/inorganic hybrid nanomaterials for sensing, imaging, and detection. This strategy opens a new door for further advancement in medical diagnosis and preventive medicine.
■
ASSOCIATED CONTENT
S Supporting Information *
Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.”
■
AUTHOR INFORMATION
Corresponding Author
*E-mail: [email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS The authors gratefully thank to Prof. Kazuhiko Ishihara and Dr. Kyoko Fukazawa at the University of Tokyo for kindly supplying MPC.
■
REFERENCES
(1) Quershi, A.; Gurbuz, Y.; Kang, W. P.; Davidson, J. L. Biosens. Bioelectron. 2009, 25, 877. (2) Ridker, P. M.; Wilson, P. W. F.; Grundy, S. M. Circulation 2004, 109, 2818. (3) Rifai, N.; Ridker, P. M. Clin. Chem. 2001, 47, 403. (4) Kuby, J. Cytokine Secretion and Biological Activity of TH1 and TH2 Subsets; WH Freeman & Co.: New York, 1994. (5) Tietz, N. W. C-Reactive Protein; WB Saunders Co.: Philadelphia, 1987. (6) Oliveira, E. B.; Gotschlich, E. C.; Liu, T. Y. J. Biol. Chem. 1979, 254, 489. (7) Pepys, M. B.; Hirschfield, G. M. J. Clin. Invest. 2003, 111, 1805. (8) Volanakis, J. E. Mol. Immunol. 2001, 38, 189. (9) Pepys, M. B.; Baltz, M. L. Adv. Immunol. 1983, 34, 141. (10) Conen, D.; Ridker, P. M. Biomark. Med. 2007, 1, 229. (11) Ridker, P. M. Clin. Chem. 2009, 55, 209. (12) Ridker, P. M. J. Am. Coll. Cardiol. 2007, 49, 2129. (13) Casas, J. P.; Shah, T.; Hingorani, A. D.; Danesh, J.; Pepys, M. B. J. Intern. Med. 2008, 264, 295. (14) Hansson, G. K. N. Engl. J. Med. 2005, 352, 1685. (15) Bassuk, S. S.; Rifai, N.; Ridker, P. M. Curr. Probl. Cardiol. 2004, 29, 439. (16) Haverkate, F.; Thompson, S. G.; Pyke, S. D. M.; Gallimore, J. R.; Pepys, M. B. Lancet 1997, 349, 462. (17) Kuller, L. H.; Tracy, R. P.; Shaten, J.; Meilahn, E. N. Am. J. Epidemiol. 1996, 144, 537. (18) Kitahashi, S.; Tatsumi, N.; Tagawa, S.; Matsui, T.; Higashihata, M.; Shintaku, H.; Tomoda, S.; Tsuda, I. J. Autom. Chem. 1998, 20, 195. (19) Kriz, K.; Ibraimi, F.; Lu, M.; Hansson, L. O.; Kriz, D. Anal. Chem. 2005, 77, 5920. (20) Meyer, M. H. F.; Hartmann, M.; Keusgen, M. Biosens. Bioelectron. 2006, 21, 1987. (21) Tsai, H. Y.; Hsu, C. F.; Chiu, I. W.; Fuh, C. B. Anal. Chem. 2007, 79, 8416. (22) Macy, E. M.; Hayes, T. E.; Tracy, R. P. Clin. Chem. 1997, 43, 52. (23) Choi, H. W.; Sakata, Y.; Kurihara, Y.; Ooya, T.; Takeuchi, T. Anal. Chim. Acta 2012, 728, 64. (24) Murata, A.; Ooya, T.; Takeuchi, T. Biosens. Bioelectron. 2013, 43, 45. G
dx.doi.org/10.1021/ac501322x | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
(62) Rowe, I. F.; Soutar, A. K.; Trayner, I. M.; Baltz, M. L.; Debeer, F. C.; Walker, L.; Bowyer, D.; Herbert, J.; Feinstein, A.; Pepys, M. B. J. Exp. Med. 1984, 159, 604. (63) Kirchmer, C. J. In Action in Analytical Chemistry; Currie, L. A., Ed.; American Chemical Society: Washington, DC, 1987; Vol. 361, p 78.
H
dx.doi.org/10.1021/ac501322x | Anal. Chem. XXXX, XXX, XXX−XXX
