Analyst View Article Online
Published on 21 August 2014. Downloaded by Michigan Technological University on 04/09/2014 09:50:56.
PAPER
Cite this: DOI: 10.1039/c4an00921e
View Journal
Detection of c-reactive protein based on a magnetic immunoassay by using functional magnetic and fluorescent nanoparticles in microplates S. F. Yang,a B. Z. Gao,a H. Y. Tsai*bc and C. Bor Fuh*a We report the preparation and application of biofunctional nanoparticles to detect C-reactive protein (CRP) in magnetic microplates. A CRP model biomarker was used to test the proposed detection method. Biofunctional magnetic nanoparticles, CRP, and biofunctional fluorescent nanoparticles were used in a sandwich nanoparticle immunoassay. The CRP concentrations in the samples were deduced from the reference plot, using the fluorescence intensity of the sandwich nanoparticle immunoassay. When biofunctional nanoparticles were used to detect CRP, the detection limit was 1.0 ng ml 1 and the linear range was between 1.18 ng ml 1 and 11.8 mg ml 1. The results revealed that the method involving biofunctional nanoparticles exhibited a lower detection limit and a wider linear range than those of the enzyme-linked immunosorbent assay (ELISA) and most other methods. For CRP measurements of serum
Received 21st May 2014 Accepted 2nd August 2014
samples, the differences between this method and ELISA in CRP measurements of serum samples were less than 13%. The proposed method can reduce the analysis time to one-third that of ELISA. This
DOI: 10.1039/c4an00921e
method demonstrates the potential to replace ELISA for rapidly detecting biomarkers with a low
www.rsc.org/analyst
detection limit and a wide dynamic range.
1. Introduction The C-reactive protein (CRP) is an acute-phase protein, which has been widely used as a biomarker for diagnosing inammation, infection, and cardiovascular diseases.1 Elevated CRP levels are associated with many pathological states and cardiovascular diseases. Therefore, monitoring CRP levels is helpful in determining the progress and effective treatments of a disease. Currently, most CRP detection methods use the enzyme-linked immunosorbent assay (ELISA) method. However, ELISA is known to be very time consuming, thus a simple, rapid, and sensitive detection of CRP in clinical applications is necessary. A magnetic immunoassay that involves functional magnetic and uorescent nanoparticles in microplates has the potential to provide an alternative to improve the CRP detection. Functional nanoparticles have been used to enhance the performance of diagnosis, screening, and other measurement applications for highly sensitive and rapid immunoassay detection.2–4 Antibody labels on nanoparticles have several
advantages over those of ELISA. First, antibodies on nanoparticles have more available binding sites because of the high surface area, and are capable of effective reactions with analytes without limitations of planar xation. Second, magnetic nanoparticles can be easily collected in a washing cycle, by simply modifying a microplate with permanent magnets to generate a magnetic eld. Third, the amount of antibody on the nanoparticles is consistent in the same batch labels. Quantum dots are known to have several advantages as uorescent probes for various applications.5,6 These advantages include low photobleaching, high brightness, narrow emission bandwidths, and broad excitation spectra. Here, quantum dots were prepared and modied with silica and anti-CRP to form biofunctional nanoparticles for CRP detection. In this study, we prepared and applied biofunctional magnetic and uorescent nanoparticles using a magnetic sandwich immunoassay to detect CRP in microplates.
2. 2.1
a
Department of Applied Chemistry, National Chi Nan University, Puli, Nantou 545, Taiwan. E-mail: [email protected]; Fax: +886-49-2917-956; Tel: +886-49-2919-779
b
Department of Applied Chemistry, Chung Shan Medical University, Taichung, 402, Taiwan. E-mail: [email protected]; Tel: +886-4-24730022 ext. 12135
c
Clinical laboratory, Chung Shan Medical University Hospital, Taichung, 402, Taiwan
This journal is © The Royal Society of Chemistry 2014
Experimental Materials
Ferric chloride, ferrous chloride, and ammonium hydroxide were purchased from J.T. Baker (Philipsburg, New Jersey, USA). Selenium, hexadecylamine, stearic acid, and bis(trimethylsilyl) sulde were purchased from Acros (New Jersey, USA). Tetraethoxysilane, diethylzinc, 1,6-hexamethylenediamine, trioctylphosphine oxide
Analyst
View Article Online
Published on 21 August 2014. Downloaded by Michigan Technological University on 04/09/2014 09:50:56.
Analyst
Paper
(TOPO), cadmium oxide, CRP, anti-CRP (capture: monoclonal antibody produced from mouse, detection: polyclonal antibody produced from goat), and phosphate-buffered saline (PBS) were purchased from Sigma-Aldrich (St. Louis, Missouri, USA). Permanent magnets (Nd-Fe-B, 6 mm in diameter and 13 mm in length) were purchased from Super Electronics (Taipei, Taiwan). Sera were obtained from Jackson ImmunoResearch (West Grove, California, USA). 2.2
Instruments
Atomic force microscopy (AFM) and transmission electron microscopy (TEM) were performed to observe the size and shape of the nanoparticles. A Fourier transform infrared spectrometer (Perkin Elmer RX-1, Norwalk, Connecticut, USA) was used to identify the functional groups of particles. The magnetization curves of the particles were studied using a superconducting quantum interference device (SQUID) magnetometer. A spectrometer (Varioskan, Thermo Electron, Waltham, Massachusetts, USA) with an excitation wavelength of 400 nm and an emission wavelength of 588 nm was used for uorescence detection. All uorescence intensities were measured in triplicate. 2.3
Preparation of biofunctional nanoparticles
Iron oxide and silica-modied iron oxide nanoparticles were prepared by chemical precipitation, as previously reported.3 In brief, 0.1 g of ferrous chloride and 0.27 g of ferric chloride in 50 ml of H2O were mixed with 4.5 ml of 28% (v/v) ammonium hydroxide to form iron oxide nanoparticles. Silica-modied magnetic nanoparticles (iron oxide@silica) were prepared by mixing 0.3 ml of a solution of iron oxide nanoparticles (0.5 g ml 1), 20 ml of tetraethoxysilane, and 0.5 ml of a 28% (v/v) ammonium hydroxide solution. Quantum dots (CdSe/ZnS) were prepared using previous methods with some modications.7–9 In brief, Solution A was
Fig. 1
Analyst
Fig. 2 (A) Infrared spectra of iron oxide, SiO2, and iron oxide @SiO2 (B) magnetization curves of magnetic nanoparticles after each modification step.
Schematic view of sandwich reaction steps in a magnetic immunoassay.
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 21 August 2014. Downloaded by Michigan Technological University on 04/09/2014 09:50:56.
Paper
prepared by dissolving 0.06 g of selenium in 2 ml of TOP. Solution B was prepared by mixing 4.83 g of TOPO, 0.569 g of stearic acid, 0.0642 g of CdO, and 2.41 g of hexadecylamine, followed by heating the mixture to 330 C under nitrogen. Solution A was added to Solution B and thoroughly mixed aer cooling down to 250 C. Solution C was prepared by mixing 0.1 ml of diethylzinc and 0.15 ml of bis(trimethylsilyl) sulde, and then adding this combination to the mixture of Solutions A and B. Quantum dots (CdSe/ZnS) were obtained from the nal mixture aer cooling to room temperature and washing with methanol 3 times. Silica-modied quantum dots [(CdSe/ZnS) @SiO2] were prepared by mixing 0.03 g of CdSe/ZnS nanoparticles in 21 ml of cyclohexane with 0.1 ml of tetraethoxysilane and 0.3 ml of a 28% (v/v) ammonium hydroxide solution. Biofunctional magnetic and uorescent nanoparticles were prepared by coupling 1 ml of 8.5 10 6 M anti-CRP with 30 mg of amine-modied nanoparticles through the coupling reaction of N-hydroxysuccinimide and 1-ethyl-3-(3-dimethyl aminopropyl) carbodiimide in an ice bath for 2 h.
Analyst
2.4
Magnetic sandwich immunoassay
Permanent magnets were placed under the bottom of a microplate for particle reconcentration aer washing. The magnetic eld strength of the magnets was 2.5 0.1 kG at the bottom well of the microplate. Fig. 1 shows the schematic view of the sandwich reaction steps in a magnetic immunoassay. One hundred microliters of anti-CRP labeled (biofunctional) magnetic nanoparticles (1.5 1011) in a PBS solution was placed into a microplate well. One hundred microliters of CRP with concentrations ranging from 10 6 M to 10 16 M were added into each well of a microplate to react for 10 min. The solutions were thoroughly mixed and washed with a PBS solution twice under a magnetic eld for 5 min to remove unreacted CRP, as depicted in the second step of Fig. 1. Subsequently, 100 ml of anti-CRP labeled (biofunctional) uorescent nanoparticles (9.0 1012) in a PBS solution was added to react for 10 min with the CRP on the immunocomplex (CRP/anti-CRP). The uorescence intensities of the resuspended sandwich nanoparticles were measured to determine the amount of CRP in the reaction
Fig. 3 (A) Electron micrographs of fluorescent nanoparticles of quantum dots (CdSe/ZnS) and silica-coated quantum dots [(CdSe/ZnS)@SiO2] (B) infrared spectra of quantum dots (CdSe/ZnS), SiO2, and (CdSe/ZnS)@SiO2. (C) Optimization of the number of biofunctional fluorescent nanoparticles for sandwich nanoparticle detection, using 10 6 M CRP and 1.5 1011 biofunctional magnetic nanoparticles.
This journal is © The Royal Society of Chemistry 2014
Analyst
View Article Online
Published on 21 August 2014. Downloaded by Michigan Technological University on 04/09/2014 09:50:56.
Analyst
aer mixing and washing to remove the unreacted biofunctional uorescent nanoparticles. A reference plot of uorescence intensity was constructed by plotting the measured uorescence intensity versus the various known concentrations of CRP added to the solution. The amounts of CRP in the samples were deduced from the reference plot using the measured uorescence intensity of the sample.
3. 3.1
Results and discussion
Paper
number of biofunctional uorescent nanoparticles from 1.0 1011 to 1.5 1013 were evaluated. Fig. 3C illustrates the optimization of the number of biofunctional uorescent nanoparticles for sandwich nanoparticle detection with the optimal number being 1.5 1011 biofunctional magnetic nanoparticles. The high number of biofunctional uorescent nanoparticles ensured the complete reaction with CRP on the affinity complex. The nonbinding nanoparticles were removed by washing to avoid interference on detection. The optimal number of biofunctional nanoparticles was used in the remaining magnetic immunoassay experiments.
Characterization of nanoparticles
The diameters of iron oxide and silica-coated iron oxide (iron oxide@silica) nanoparticles were approximately 5.5 nm and 11 nm, as estimated by AFM and TEM, respectively. Fig. 2A shows the infrared spectra of iron oxide, silica dioxide, and iron oxide@silica nanoparticles aer the surface modication steps were performed. The silica coating was obvious and provided sufficient protection for the iron oxide. Fig. 2B displays the magnetization curves of the magnetic nanoparticles aer each modication step of iron oxide. Minimizing the decrease of magnetization on each modication step is essential for magnetic applications. The saturated magnetization of biofunctional magnetic nanoparticles was reduced by 16% to approximately 46 emu g 1, which was favorable for the magnetic collection step. The diameters of the quantum dots (CdSe/ZnS) and silica-coated quantum dots [(CdSe/ZnS)@SiO2] were approximately 3 nm and 25 nm, respectively. Fig. 3A displays the electron micrograph of (CdSe/ZnS) and (CdSe/ZnS) @SiO2. Each silica matrix contained several quantum dots, providing a stable environment and easy biofunctionalization for uorescence detection. The uorescence intensities of (CdSe/ZnS)@SiO2 remained at 90% for 5 h under 150 W of UV irradiation, or for 56 days without UV irradiation. The maximum emission wavelength of the quantum dots shied from 581 nm to 588 nm when anti-CRP was conjugated with the quantum dots under a 400 nm excitation wavelength. Fig. 3B shows the infrared spectra of the quantum dots and the results of their surface modication. The amounts of anti-CRP on magnetic and uorescent nanoparticles were approximately 1.5 and 4, respectively. These values were calculated from the different amounts of anti-CRP before and aer the labels were divided by the number of particles. The number of anti-CRP was calculated from the molecular numbers of anti-CRP used for labeling subtracted the number of anti-CRP in suspension. The number of anti-CRP was equal to molar concentration (M) volume (L) 6.02 1023.
3.3
Magnetic sandwich immunoassay
A magnetic sandwich immunoassay that entails using biofunctional magnetic and uorescent nanoparticles was tested using a model biomarker, CRP. In this method, biofunctional (anti-CRP labeled) magnetic nanoparticles were placed into a well, the CRP samples were added into the well to react with the anti-CRP on the magnetic nanoparticles, and biofunctional (anti-CRP labeled) uorescent nanoparticles were then added to perform magnetic sandwich immunoassay, and identify the
3.2 Optimization of the number of biofunctional nanoparticles The maximum number of biofunctional magnetic nanoparticles that could be retained in a microplate under a magnetic eld was determined to be 1.5 1011 aer a number of particles from 1010 to 1013 were evaluated. The optimal number of biofunctional uorescent nanoparticles was determined to be 9.0 1012 for 1.5 1011 biofunctional magnetic nanoparticles, aer a Analyst
Fig. 4 (A) A reference plot of the fluorescence intensities of sandwich nanoparticles at various CRP concentrations added in diluted serum solutions. The numbers of biofunctional magnetic and fluorescent nanoparticles were approximately 1.5 1011 and 9.0 1012, respectively. (B) The linear range of reference plot.
This journal is © The Royal Society of Chemistry 2014
View Article Online
Paper
Published on 21 August 2014. Downloaded by Michigan Technological University on 04/09/2014 09:50:56.
Table 1
Analyst Comparisons of linear range and detection limit for CRP detection
Literature
Linear range
Detection limit
Reference
J. Chromatogr. A, 2013, 1315, 188–194 Anal. Chim. Acta, 2010, 662, 182–192 Biosens. Bioelectron., 2011, 26, 3072–3076 Sens. Actuators, B, 2013, 88, 1277–1283 Biosens. Bioelectron., 2013, 39, 94–98 This study
10–150 mg ml 1 50–450 ng ml 1 10 ng ml 1–10 mg ml 1 20 ng ml 1–30 mg ml 1 59 ng ml 1–5.9 mg ml 1 1.18 ng ml 1–11.8 mg ml
9.2 mg ml 1 50 ng ml 1 10 ng ml 1 20 ng ml 1 21 ng ml 1 1 ng ml 1
10 11 12 13 14
CRP samples in the immunoassay complex, as shown in Fig. 1. A magnetic eld was applied to reconcentrate nanoparticles aer unreacted species were removed with each washing. All of these reactions were highly effective under nearly homogeneous conditions. The uorescence intensities of the sandwich nanoparticles were correlated with the CRP concentrations in the samples aer calibration. Fig. 4 displays the reference plot of uorescence intensities, determined from the sandwich nanoparticles added to the serum solutions at various CRP concentrations (10 6 to 10 16 M). The selectivity of the CRP was tested by replacing CRP with a blank and three concentrations (10 8, 10 9, and 10 10 M) of IgG. The averaged uorescence intensities were all below 21, indicating the favorable selectivity of the proposed method. The uorescence intensities were linear from 10 7 M (11.8 mg ml 1) to 10 11 M (1.18 ng ml 1) for CRP detection. The linear range of the proposed method was wider than those of ELISA and most other detection methods.10–14 Table 1 summarizes the comparison of the linear range and detection limit for CRP detection obtained in this study with those in the literature. The detection limit of the proposed method was 1.0 ng ml 1 based on a signalto-noise ratio of 3. This detection limit was 7 times lower than that of ELISA. The proposed method that involves using a magnetic immunoassay based on a sandwich nanoparticle detection reduced the analysis time to one-third that of ELISA. Serum samples spiked with 3 CRP concentrations (3.0 10 8, 3.0 10 9, and 3.0 10 10 M) were tested for sandwich nanoparticle detection in a magnetic immunoassay, and their measured concentrations differed from the spiked concentrations by 6.4%, 5.7%, and 8.5%, respectively. Serum samples spiked with 3 other CRP concentrations (7.0, 21, and 60 ng
1
ml 1) were also used to compare ELISA and the proposed method. The CRP concentrations determined by ELISA were 7.5, 23, and 57 ng ml 1 with 5.5% of averaged RSDs, whereas those determined by the proposed method were 6.7, 20, and 64 ng ml 1 with 5.8% of averaged RSDs. Fig. 5 shows the easy comparison of CRP measurements using ELISA and this study with standard deviation. All of these measurements differed from the spiked concentrations by less than 9%. The differences between ELISA and the proposed method were also less than 13%. The proposed method can be easily performed by simple modications of the present ELISA instrument for allowing automation with high throughputs. The proposed method provides a favorable alternative for ELISA with a lowerer detection limit, greater linear range, and shorter analysis time than those of ELISA.
4. Conclusions The proposed method has several advantages over ELISA for CRP detection, including the analysis time reduced by onethird, a lower detection limit, and a greater dynamic range. The proposed method provides an effective alternative to detect protein and other biochemical compounds.
Acknowledgements This work was supported by the Ministry of Science and Technology of Taiwan (Grant MOST-99-2113-M-260-003). The authors would like to thank Z. K. Chen and Mr B. C. Zhou for their assistances.
References
Fig. 5
Comparison of CRP measurements using ELISA and this study.
This journal is © The Royal Society of Chemistry 2014
1 J. E. Volanakis, Mol. Immunol., 2001, 38, 189–197. 2 K. Saha, S. S. Agasti, C. Kim, X. Li and V. M. Rotello, Chem. Rev., 2012, 112, 2739–2779. 3 C. S. Li, S. H. Chen, H. Y. Tsai and C. B. Fuh, Sens. Actuators, B, 2012, 166–167, 819–823. 4 S. T. Selvan, T. T. Y. Tan, D. K. Yi and N. R. Jana, Langmuir, 2009, 26, 11631–11641. 5 W. R. Algar, K. Susumu, J. B. Delehanty and I. L. Medintz, Anal. Chem., 2011, 83, 8826–8837. 6 A. M. Smith and S. Nie, Acc. Chem. Res., 2009, 43, 190–200. 7 S. K. Panda, S. G. Hickey, C. Waurisch and A. Eychmuller, J. Mater. Chem., 2011, 21, 11550–11555.
Analyst
View Article Online
Published on 21 August 2014. Downloaded by Michigan Technological University on 04/09/2014 09:50:56.
Analyst
8 M. J. Murcia, D. L. Shaw, H. Woodruff, C. A. Naumann, B. A. Young and E. C. Long, Chem. Mater., 2006, 18, 2219– 2225. 9 P. Reiss, J. Bleuse and A. Pron, Nano Lett., 2002, 2, 781–784. 10 Y.-J. Lin, J.-Y. Yang, T.-Y. Shu, T.-Y. Lin, Y.-Y. Chen, M.-Y. Su, W.-J. Li and M.-Y. Liu, J. Chromatogr. A, 2013, 1315, 188–194. 11 V. Raj and K. Sreenivasan, Anal. Chim. Acta, 2010, 662, 186– 192.
Analyst
Paper
12 C.-S. Chen, K.-N. Chang, Y.-H. Chen, C.-K. Lee, B. Y.-J. Lee and A. S.-Y. Lee, Biosens. Bioelectron., 2011, 26, 3072–3076. 13 P. Ding, R. Liu, S. Liu, X. Mao, R. Hu and G. Li, Sens. Actuators, B, 2013, 188, 1277–1283. 14 T. Bryan, X. Luo, P. R. Bueno and J. J. Davis, Biosens. Bioelectron., 2013, 39, 94–98.
This journal is © The Royal Society of Chemistry 2014
Published on 21 August 2014. Downloaded by Michigan Technological University on 04/09/2014 09:50:56.
PAPER
Cite this: DOI: 10.1039/c4an00921e
View Journal
Detection of c-reactive protein based on a magnetic immunoassay by using functional magnetic and fluorescent nanoparticles in microplates S. F. Yang,a B. Z. Gao,a H. Y. Tsai*bc and C. Bor Fuh*a We report the preparation and application of biofunctional nanoparticles to detect C-reactive protein (CRP) in magnetic microplates. A CRP model biomarker was used to test the proposed detection method. Biofunctional magnetic nanoparticles, CRP, and biofunctional fluorescent nanoparticles were used in a sandwich nanoparticle immunoassay. The CRP concentrations in the samples were deduced from the reference plot, using the fluorescence intensity of the sandwich nanoparticle immunoassay. When biofunctional nanoparticles were used to detect CRP, the detection limit was 1.0 ng ml 1 and the linear range was between 1.18 ng ml 1 and 11.8 mg ml 1. The results revealed that the method involving biofunctional nanoparticles exhibited a lower detection limit and a wider linear range than those of the enzyme-linked immunosorbent assay (ELISA) and most other methods. For CRP measurements of serum
Received 21st May 2014 Accepted 2nd August 2014
samples, the differences between this method and ELISA in CRP measurements of serum samples were less than 13%. The proposed method can reduce the analysis time to one-third that of ELISA. This
DOI: 10.1039/c4an00921e
method demonstrates the potential to replace ELISA for rapidly detecting biomarkers with a low
www.rsc.org/analyst
detection limit and a wide dynamic range.
1. Introduction The C-reactive protein (CRP) is an acute-phase protein, which has been widely used as a biomarker for diagnosing inammation, infection, and cardiovascular diseases.1 Elevated CRP levels are associated with many pathological states and cardiovascular diseases. Therefore, monitoring CRP levels is helpful in determining the progress and effective treatments of a disease. Currently, most CRP detection methods use the enzyme-linked immunosorbent assay (ELISA) method. However, ELISA is known to be very time consuming, thus a simple, rapid, and sensitive detection of CRP in clinical applications is necessary. A magnetic immunoassay that involves functional magnetic and uorescent nanoparticles in microplates has the potential to provide an alternative to improve the CRP detection. Functional nanoparticles have been used to enhance the performance of diagnosis, screening, and other measurement applications for highly sensitive and rapid immunoassay detection.2–4 Antibody labels on nanoparticles have several
advantages over those of ELISA. First, antibodies on nanoparticles have more available binding sites because of the high surface area, and are capable of effective reactions with analytes without limitations of planar xation. Second, magnetic nanoparticles can be easily collected in a washing cycle, by simply modifying a microplate with permanent magnets to generate a magnetic eld. Third, the amount of antibody on the nanoparticles is consistent in the same batch labels. Quantum dots are known to have several advantages as uorescent probes for various applications.5,6 These advantages include low photobleaching, high brightness, narrow emission bandwidths, and broad excitation spectra. Here, quantum dots were prepared and modied with silica and anti-CRP to form biofunctional nanoparticles for CRP detection. In this study, we prepared and applied biofunctional magnetic and uorescent nanoparticles using a magnetic sandwich immunoassay to detect CRP in microplates.
2. 2.1
a
Department of Applied Chemistry, National Chi Nan University, Puli, Nantou 545, Taiwan. E-mail: [email protected]; Fax: +886-49-2917-956; Tel: +886-49-2919-779
b
Department of Applied Chemistry, Chung Shan Medical University, Taichung, 402, Taiwan. E-mail: [email protected]; Tel: +886-4-24730022 ext. 12135
c
Clinical laboratory, Chung Shan Medical University Hospital, Taichung, 402, Taiwan
This journal is © The Royal Society of Chemistry 2014
Experimental Materials
Ferric chloride, ferrous chloride, and ammonium hydroxide were purchased from J.T. Baker (Philipsburg, New Jersey, USA). Selenium, hexadecylamine, stearic acid, and bis(trimethylsilyl) sulde were purchased from Acros (New Jersey, USA). Tetraethoxysilane, diethylzinc, 1,6-hexamethylenediamine, trioctylphosphine oxide
Analyst
View Article Online
Published on 21 August 2014. Downloaded by Michigan Technological University on 04/09/2014 09:50:56.
Analyst
Paper
(TOPO), cadmium oxide, CRP, anti-CRP (capture: monoclonal antibody produced from mouse, detection: polyclonal antibody produced from goat), and phosphate-buffered saline (PBS) were purchased from Sigma-Aldrich (St. Louis, Missouri, USA). Permanent magnets (Nd-Fe-B, 6 mm in diameter and 13 mm in length) were purchased from Super Electronics (Taipei, Taiwan). Sera were obtained from Jackson ImmunoResearch (West Grove, California, USA). 2.2
Instruments
Atomic force microscopy (AFM) and transmission electron microscopy (TEM) were performed to observe the size and shape of the nanoparticles. A Fourier transform infrared spectrometer (Perkin Elmer RX-1, Norwalk, Connecticut, USA) was used to identify the functional groups of particles. The magnetization curves of the particles were studied using a superconducting quantum interference device (SQUID) magnetometer. A spectrometer (Varioskan, Thermo Electron, Waltham, Massachusetts, USA) with an excitation wavelength of 400 nm and an emission wavelength of 588 nm was used for uorescence detection. All uorescence intensities were measured in triplicate. 2.3
Preparation of biofunctional nanoparticles
Iron oxide and silica-modied iron oxide nanoparticles were prepared by chemical precipitation, as previously reported.3 In brief, 0.1 g of ferrous chloride and 0.27 g of ferric chloride in 50 ml of H2O were mixed with 4.5 ml of 28% (v/v) ammonium hydroxide to form iron oxide nanoparticles. Silica-modied magnetic nanoparticles (iron oxide@silica) were prepared by mixing 0.3 ml of a solution of iron oxide nanoparticles (0.5 g ml 1), 20 ml of tetraethoxysilane, and 0.5 ml of a 28% (v/v) ammonium hydroxide solution. Quantum dots (CdSe/ZnS) were prepared using previous methods with some modications.7–9 In brief, Solution A was
Fig. 1
Analyst
Fig. 2 (A) Infrared spectra of iron oxide, SiO2, and iron oxide @SiO2 (B) magnetization curves of magnetic nanoparticles after each modification step.
Schematic view of sandwich reaction steps in a magnetic immunoassay.
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 21 August 2014. Downloaded by Michigan Technological University on 04/09/2014 09:50:56.
Paper
prepared by dissolving 0.06 g of selenium in 2 ml of TOP. Solution B was prepared by mixing 4.83 g of TOPO, 0.569 g of stearic acid, 0.0642 g of CdO, and 2.41 g of hexadecylamine, followed by heating the mixture to 330 C under nitrogen. Solution A was added to Solution B and thoroughly mixed aer cooling down to 250 C. Solution C was prepared by mixing 0.1 ml of diethylzinc and 0.15 ml of bis(trimethylsilyl) sulde, and then adding this combination to the mixture of Solutions A and B. Quantum dots (CdSe/ZnS) were obtained from the nal mixture aer cooling to room temperature and washing with methanol 3 times. Silica-modied quantum dots [(CdSe/ZnS) @SiO2] were prepared by mixing 0.03 g of CdSe/ZnS nanoparticles in 21 ml of cyclohexane with 0.1 ml of tetraethoxysilane and 0.3 ml of a 28% (v/v) ammonium hydroxide solution. Biofunctional magnetic and uorescent nanoparticles were prepared by coupling 1 ml of 8.5 10 6 M anti-CRP with 30 mg of amine-modied nanoparticles through the coupling reaction of N-hydroxysuccinimide and 1-ethyl-3-(3-dimethyl aminopropyl) carbodiimide in an ice bath for 2 h.
Analyst
2.4
Magnetic sandwich immunoassay
Permanent magnets were placed under the bottom of a microplate for particle reconcentration aer washing. The magnetic eld strength of the magnets was 2.5 0.1 kG at the bottom well of the microplate. Fig. 1 shows the schematic view of the sandwich reaction steps in a magnetic immunoassay. One hundred microliters of anti-CRP labeled (biofunctional) magnetic nanoparticles (1.5 1011) in a PBS solution was placed into a microplate well. One hundred microliters of CRP with concentrations ranging from 10 6 M to 10 16 M were added into each well of a microplate to react for 10 min. The solutions were thoroughly mixed and washed with a PBS solution twice under a magnetic eld for 5 min to remove unreacted CRP, as depicted in the second step of Fig. 1. Subsequently, 100 ml of anti-CRP labeled (biofunctional) uorescent nanoparticles (9.0 1012) in a PBS solution was added to react for 10 min with the CRP on the immunocomplex (CRP/anti-CRP). The uorescence intensities of the resuspended sandwich nanoparticles were measured to determine the amount of CRP in the reaction
Fig. 3 (A) Electron micrographs of fluorescent nanoparticles of quantum dots (CdSe/ZnS) and silica-coated quantum dots [(CdSe/ZnS)@SiO2] (B) infrared spectra of quantum dots (CdSe/ZnS), SiO2, and (CdSe/ZnS)@SiO2. (C) Optimization of the number of biofunctional fluorescent nanoparticles for sandwich nanoparticle detection, using 10 6 M CRP and 1.5 1011 biofunctional magnetic nanoparticles.
This journal is © The Royal Society of Chemistry 2014
Analyst
View Article Online
Published on 21 August 2014. Downloaded by Michigan Technological University on 04/09/2014 09:50:56.
Analyst
aer mixing and washing to remove the unreacted biofunctional uorescent nanoparticles. A reference plot of uorescence intensity was constructed by plotting the measured uorescence intensity versus the various known concentrations of CRP added to the solution. The amounts of CRP in the samples were deduced from the reference plot using the measured uorescence intensity of the sample.
3. 3.1
Results and discussion
Paper
number of biofunctional uorescent nanoparticles from 1.0 1011 to 1.5 1013 were evaluated. Fig. 3C illustrates the optimization of the number of biofunctional uorescent nanoparticles for sandwich nanoparticle detection with the optimal number being 1.5 1011 biofunctional magnetic nanoparticles. The high number of biofunctional uorescent nanoparticles ensured the complete reaction with CRP on the affinity complex. The nonbinding nanoparticles were removed by washing to avoid interference on detection. The optimal number of biofunctional nanoparticles was used in the remaining magnetic immunoassay experiments.
Characterization of nanoparticles
The diameters of iron oxide and silica-coated iron oxide (iron oxide@silica) nanoparticles were approximately 5.5 nm and 11 nm, as estimated by AFM and TEM, respectively. Fig. 2A shows the infrared spectra of iron oxide, silica dioxide, and iron oxide@silica nanoparticles aer the surface modication steps were performed. The silica coating was obvious and provided sufficient protection for the iron oxide. Fig. 2B displays the magnetization curves of the magnetic nanoparticles aer each modication step of iron oxide. Minimizing the decrease of magnetization on each modication step is essential for magnetic applications. The saturated magnetization of biofunctional magnetic nanoparticles was reduced by 16% to approximately 46 emu g 1, which was favorable for the magnetic collection step. The diameters of the quantum dots (CdSe/ZnS) and silica-coated quantum dots [(CdSe/ZnS)@SiO2] were approximately 3 nm and 25 nm, respectively. Fig. 3A displays the electron micrograph of (CdSe/ZnS) and (CdSe/ZnS) @SiO2. Each silica matrix contained several quantum dots, providing a stable environment and easy biofunctionalization for uorescence detection. The uorescence intensities of (CdSe/ZnS)@SiO2 remained at 90% for 5 h under 150 W of UV irradiation, or for 56 days without UV irradiation. The maximum emission wavelength of the quantum dots shied from 581 nm to 588 nm when anti-CRP was conjugated with the quantum dots under a 400 nm excitation wavelength. Fig. 3B shows the infrared spectra of the quantum dots and the results of their surface modication. The amounts of anti-CRP on magnetic and uorescent nanoparticles were approximately 1.5 and 4, respectively. These values were calculated from the different amounts of anti-CRP before and aer the labels were divided by the number of particles. The number of anti-CRP was calculated from the molecular numbers of anti-CRP used for labeling subtracted the number of anti-CRP in suspension. The number of anti-CRP was equal to molar concentration (M) volume (L) 6.02 1023.
3.3
Magnetic sandwich immunoassay
A magnetic sandwich immunoassay that entails using biofunctional magnetic and uorescent nanoparticles was tested using a model biomarker, CRP. In this method, biofunctional (anti-CRP labeled) magnetic nanoparticles were placed into a well, the CRP samples were added into the well to react with the anti-CRP on the magnetic nanoparticles, and biofunctional (anti-CRP labeled) uorescent nanoparticles were then added to perform magnetic sandwich immunoassay, and identify the
3.2 Optimization of the number of biofunctional nanoparticles The maximum number of biofunctional magnetic nanoparticles that could be retained in a microplate under a magnetic eld was determined to be 1.5 1011 aer a number of particles from 1010 to 1013 were evaluated. The optimal number of biofunctional uorescent nanoparticles was determined to be 9.0 1012 for 1.5 1011 biofunctional magnetic nanoparticles, aer a Analyst
Fig. 4 (A) A reference plot of the fluorescence intensities of sandwich nanoparticles at various CRP concentrations added in diluted serum solutions. The numbers of biofunctional magnetic and fluorescent nanoparticles were approximately 1.5 1011 and 9.0 1012, respectively. (B) The linear range of reference plot.
This journal is © The Royal Society of Chemistry 2014
View Article Online
Paper
Published on 21 August 2014. Downloaded by Michigan Technological University on 04/09/2014 09:50:56.
Table 1
Analyst Comparisons of linear range and detection limit for CRP detection
Literature
Linear range
Detection limit
Reference
J. Chromatogr. A, 2013, 1315, 188–194 Anal. Chim. Acta, 2010, 662, 182–192 Biosens. Bioelectron., 2011, 26, 3072–3076 Sens. Actuators, B, 2013, 88, 1277–1283 Biosens. Bioelectron., 2013, 39, 94–98 This study
10–150 mg ml 1 50–450 ng ml 1 10 ng ml 1–10 mg ml 1 20 ng ml 1–30 mg ml 1 59 ng ml 1–5.9 mg ml 1 1.18 ng ml 1–11.8 mg ml
9.2 mg ml 1 50 ng ml 1 10 ng ml 1 20 ng ml 1 21 ng ml 1 1 ng ml 1
10 11 12 13 14
CRP samples in the immunoassay complex, as shown in Fig. 1. A magnetic eld was applied to reconcentrate nanoparticles aer unreacted species were removed with each washing. All of these reactions were highly effective under nearly homogeneous conditions. The uorescence intensities of the sandwich nanoparticles were correlated with the CRP concentrations in the samples aer calibration. Fig. 4 displays the reference plot of uorescence intensities, determined from the sandwich nanoparticles added to the serum solutions at various CRP concentrations (10 6 to 10 16 M). The selectivity of the CRP was tested by replacing CRP with a blank and three concentrations (10 8, 10 9, and 10 10 M) of IgG. The averaged uorescence intensities were all below 21, indicating the favorable selectivity of the proposed method. The uorescence intensities were linear from 10 7 M (11.8 mg ml 1) to 10 11 M (1.18 ng ml 1) for CRP detection. The linear range of the proposed method was wider than those of ELISA and most other detection methods.10–14 Table 1 summarizes the comparison of the linear range and detection limit for CRP detection obtained in this study with those in the literature. The detection limit of the proposed method was 1.0 ng ml 1 based on a signalto-noise ratio of 3. This detection limit was 7 times lower than that of ELISA. The proposed method that involves using a magnetic immunoassay based on a sandwich nanoparticle detection reduced the analysis time to one-third that of ELISA. Serum samples spiked with 3 CRP concentrations (3.0 10 8, 3.0 10 9, and 3.0 10 10 M) were tested for sandwich nanoparticle detection in a magnetic immunoassay, and their measured concentrations differed from the spiked concentrations by 6.4%, 5.7%, and 8.5%, respectively. Serum samples spiked with 3 other CRP concentrations (7.0, 21, and 60 ng
1
ml 1) were also used to compare ELISA and the proposed method. The CRP concentrations determined by ELISA were 7.5, 23, and 57 ng ml 1 with 5.5% of averaged RSDs, whereas those determined by the proposed method were 6.7, 20, and 64 ng ml 1 with 5.8% of averaged RSDs. Fig. 5 shows the easy comparison of CRP measurements using ELISA and this study with standard deviation. All of these measurements differed from the spiked concentrations by less than 9%. The differences between ELISA and the proposed method were also less than 13%. The proposed method can be easily performed by simple modications of the present ELISA instrument for allowing automation with high throughputs. The proposed method provides a favorable alternative for ELISA with a lowerer detection limit, greater linear range, and shorter analysis time than those of ELISA.
4. Conclusions The proposed method has several advantages over ELISA for CRP detection, including the analysis time reduced by onethird, a lower detection limit, and a greater dynamic range. The proposed method provides an effective alternative to detect protein and other biochemical compounds.
Acknowledgements This work was supported by the Ministry of Science and Technology of Taiwan (Grant MOST-99-2113-M-260-003). The authors would like to thank Z. K. Chen and Mr B. C. Zhou for their assistances.
References
Fig. 5
Comparison of CRP measurements using ELISA and this study.
This journal is © The Royal Society of Chemistry 2014
1 J. E. Volanakis, Mol. Immunol., 2001, 38, 189–197. 2 K. Saha, S. S. Agasti, C. Kim, X. Li and V. M. Rotello, Chem. Rev., 2012, 112, 2739–2779. 3 C. S. Li, S. H. Chen, H. Y. Tsai and C. B. Fuh, Sens. Actuators, B, 2012, 166–167, 819–823. 4 S. T. Selvan, T. T. Y. Tan, D. K. Yi and N. R. Jana, Langmuir, 2009, 26, 11631–11641. 5 W. R. Algar, K. Susumu, J. B. Delehanty and I. L. Medintz, Anal. Chem., 2011, 83, 8826–8837. 6 A. M. Smith and S. Nie, Acc. Chem. Res., 2009, 43, 190–200. 7 S. K. Panda, S. G. Hickey, C. Waurisch and A. Eychmuller, J. Mater. Chem., 2011, 21, 11550–11555.
Analyst
View Article Online
Published on 21 August 2014. Downloaded by Michigan Technological University on 04/09/2014 09:50:56.
Analyst
8 M. J. Murcia, D. L. Shaw, H. Woodruff, C. A. Naumann, B. A. Young and E. C. Long, Chem. Mater., 2006, 18, 2219– 2225. 9 P. Reiss, J. Bleuse and A. Pron, Nano Lett., 2002, 2, 781–784. 10 Y.-J. Lin, J.-Y. Yang, T.-Y. Shu, T.-Y. Lin, Y.-Y. Chen, M.-Y. Su, W.-J. Li and M.-Y. Liu, J. Chromatogr. A, 2013, 1315, 188–194. 11 V. Raj and K. Sreenivasan, Anal. Chim. Acta, 2010, 662, 186– 192.
Analyst
Paper
12 C.-S. Chen, K.-N. Chang, Y.-H. Chen, C.-K. Lee, B. Y.-J. Lee and A. S.-Y. Lee, Biosens. Bioelectron., 2011, 26, 3072–3076. 13 P. Ding, R. Liu, S. Liu, X. Mao, R. Hu and G. Li, Sens. Actuators, B, 2013, 188, 1277–1283. 14 T. Bryan, X. Luo, P. R. Bueno and J. J. Davis, Biosens. Bioelectron., 2013, 39, 94–98.
This journal is © The Royal Society of Chemistry 2014
