Analyst View Article Online
Published on 12 February 2014. Downloaded by UNIVERSIDAD DE BUENOS AIRES on 26/03/2018 08:43:48.
PAPER
View Journal | View Issue
Cite this: Analyst, 2014, 139, 2237
Surface plasmon resonance-based immunoassay for human fetuin A† S. K. Vashist,*ab E. M. Schneiderc and J. H. T. Luongd This article describes a highly-sensitive surface plasmon resonance (SPR)-based immunoassay (IA) for human fetuin A (HFA), a specific biomarker for atherosclerosis and hepatocellular carcinoma. The assay is based on a novel immobilization procedure that simply involves the dilution of an anti-HFA capture antibody (Ab) in 1% (v/v) 3-aminopropyltriethoxysilane (APTES), followed by its dispensing on a KOHtreated gold (Au)-coated SPR chip and incubation for 30 min. The developed SPR IA detected 0.3–20 ng mL1 of HFA with a limit of detection and sensitivity of 0.7 ng mL1 and 1 ng mL1, respectively. The highly-simplified Ab immobilization procedure is also 5-fold more rapid than conventional procedures. It leads to the leach-proof binding of the capture Ab, which means that the developed SPR IA is highly
Received 20th January 2014 Accepted 10th February 2014
cost-effective, as the Ab-bound SPR chip could be reused for many repeated HFA IAs after regeneration with 10 mM glycine–HCl, pH 2.0. The Ab-bound SPR chip, stored at 4 C, lost only 18% of its original
DOI: 10.1039/c4an00149d
activity after 4 months. For the detection of HFA spiked in diluted human whole blood and plasma, the
www.rsc.org/analyst
results obtained by the developed SPR IA agreed well with the commercial HFA sandwich ELISA.
Introduction HFA, also known as alpha2-Heremans-Schmid glycoprotein (AHSG), is a product of the liver and its concentration decreases during the acute phase reaction. It displays anti-inammatory properties, counteracting proinammatory cytokine production, and functions as an inhibitor of so tissue calcication.1 Additionally, HFA has been demonstrated to constitute a major component of mineralo-organic nanoparticles which are likely responsible for the vicious cycle of inammation and calcication.2 The decrease in serum HFA concentration occurs in acute alcoholic hepatitis, chronic autoimmune hepatitis, fatty liver, alcoholic and primary biliary cirrhosis, and hepatocellular cancer.3 Most importantly, low plasma concentration conditions of HFA appear for states of insulin resistance and metabolic syndrome.4 The latter is explained by a high binding affinity of HFA to insulin receptors that specically inhibit the tyrosine kinase activity of the insulin receptor.5 Moreover, it inhibits adiponectin and thus indirectly upregulates the a
HSG-IMIT - Institut f¨ ur Mikro - und Informationstechnik, Georges-Koehler Allee 103, 79110 Freiburg, Germany. E-mail: [email protected]
b
Laboratory for MEMS Applications, Department of Microsystems Engineering IMTEK, University of Freiburg, Georges-Koehler-Allee 103, 79110 Freiburg, Germany
c Sektion Experimentelle Anaesthesiologie, University Hospital Ulm, Albert Einstein Allee 23, 89081 Ulm, Germany d
Innovative Chromatography Group, Irish Separation Science Cluster (ISSC), Department of Chemistry and Analytical, Biological Chemistry Research Facility (ABCRF), University College Cork, Cork, Ireland † Electronic supplementary 10.1039/c4an00149d
information
(ESI)
This journal is © The Royal Society of Chemistry 2014
available:
See
DOI:
secretion of inammatory cytokines from macrophages such as tumor necrosis factor alpha (TNF-a), which increases cardiovascular disease and the risk of strokes. This is why HFA is one of the most promising biomarkers used in blood and serum to identify cardiovascular dysfunction at an early stage.6 Importantly, the intriguing relevance of HFA quantication has been extended to neuroinammatory diseases,7 particularly in multiple sclerosis, where it has been detected in demyelinated lesions and in grey matter.8 Therefore, the personalized monitoring of HFA in plasma and body uid is a promising approach for metabolic risk analysis in modern communities. The availability of rapid assays for HFA quantication appears to be highly desired in the acute states of organ (heart, liver, lung and kidney) diseases related to vascular calcication. There is an urgent need for a rapid immunoassay (IA) format to detect disease biomarkers and analytes because conventional IA formats such as enzyme-linked immunosorbent assays, chemiluminescent IAs and uorescent IAs take several hours to provide a result. During the last decade, SPR has signicantly reduced the IA duration to just a few minutes. Several prospective SPR IA formats have been developed using a wide range of antibody immobilization techniques.9–15 They utilize expensive commercially available SPR chips that are pre-functionalized with carboxymethyl dextran, streptavidin, nitriloacetic acid, long chain alkanethiol molecules or lipophilic groups.16 Fragment crystallizable (Fc)-binding proteins14,17,18 such as protein A, protein G or protein A/G, for oriented Ab immobilization have also been investigated. However, all of these techniques require a large number of process steps and take several hours to immobilize the capture Ab onto the SPR chip.
Analyst, 2014, 139, 2237–2242 | 2237
View Article Online
Published on 12 February 2014. Downloaded by UNIVERSIDAD DE BUENOS AIRES on 26/03/2018 08:43:48.
Analyst
The bioanalytical performance of a SPR-based immunoassay is critically dependent on the nature of antibody immobilization as well as the affinity, orientation and stability of the bound antibody.9,14,19 This article reveals a potential SPR IA format for HFA using a highly simplied Ab immobilization procedure that enables a leach-proof binding of the capture Ab on the Aucoated SPR chip in just 30 min. The developed SPR IA employs normal Au-coated SPR chips, which are 3-fold less expensive than pre-functionalized chips. It removes the multiple process steps and additional chemicals required in conventional immobilization techniques that are widely used for binding capture Ab on pre-functionalized SPR chips. Therefore, the Ab immobilization technique used for the developed SPR IA is highly-simplied, rapid and cost-effective. The analytical performance of the capture Ab-bound SPR chip in terms of sensitivity, dynamic range, total IA duration and increased costeffectiveness will be investigated and compared with our previously developed,14 and commercial carboxymethyl (CM5) dextran chip-based SPR IA procedures.1 The applicability of the assay format for HFA detection in diluted human whole blood and plasma is then compared with the commercial sandwich ELISA procedure. The reuse of the capture Ab-bound SPR chip is also demonstrated, together with its excellent storage stability.
Experimental Materials 1-Ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC), Nhydroxysulfosuccinimide, 2-(N-morpholino)ethane sulfonic acid (MES, pH 4.7) and bovine serum albumin (BSA) were purchased from Thermo Scientic. 3-Aminopropyltriethoxysilane (APTES, purity 98%, w/v), Tween 20, H2O2 (30%, v/v) and H2SO4 (97.5%, v/v) were obtained from Sigma-Aldrich. A human fetuin A kit with all of the necessary components was obtained from R&D Systems, USA. Buffers and solutions were prepared with 18 MU Milli-Q ultrapure water (UPW) ltered through a 2 mm lter. SPR was performed using a BIAcore 3000 from GE Healthcare, Uppsala, Sweden. The SIA kit (containing SPR Au chips), carboxymethyl dextran (CMD)-functionalized Au chips, ethanolamine hydrochloride (1 M, pH 8.5), HBS-EP (0.01 M 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) pH 7.4, 0.15 M NaCl, 3 mM EDTA, 0.005% v/v surfactant P20), and glycine–HCl (10 mM, pH 2.0) were purchased from GE Healthcare. The SPR Au chip was assembled as provided by the manufacturer. HBS-EP was used as the running buffer for BIAcore and all sample dilutions were made in the running buffer. The conventional sandwich ELISA procedure was followed by the manufacturer's guidelines without any modication. Developed Ab immobilization procedure The assay format involves the sequential dilution of anti-HFA Ab in 1% (v/v) APTES (in the ratio of 1 : 1 v/v), dispensing on a KOH-treated Au SPR chip, incubation for 30 min, and washing with HEPES-buffered saline (HBS) (Fig. 1). In brief, the Au chip was cleaned by treatment with 90 mL of 1% (w/v) KOH for 5 min
2238 | Analyst, 2014, 139, 2237–2242
Paper
followed by extensive washing with UPW. The capture anti-HFA Ab (200 mg mL1 in HBS) was mixed with 1% APTES (in the ratio of 1 : 1 v/v). Thereaer, 90 mL of this capture anti-HFA Ab solution, with a nal Ab concentration of 100 mg mL1 in 0.5% APTES, was dispensed onto the Au chip and incubated for 30 min at room temperature (RT) in a fume hood. The anti-HFA Abbound Au SPR chip was washed extensively with HBS. It was then docked into the BIAcore 3000 and primed. Non-specic binding sites on the Ab-bound chip were then blocked by injecting 20 mL of 1% (w/v) BSA at 10 mL min1. Detailed information of the immobilization procedures is provided in the ESI.† HFA detection HFA (50 mL, 0.3 ng mL1) was passed through three detecting ow cells at 10 mL min1 with the fourth ow cell serving as a reference. The resultant changes in the SPR response units (RUs) were recorded for all ow cells. The Ab-bound SPR chip was regenerated aer each HFA IA by treatment with 20 mL of 10 mM glycine–HCl, pH 2.0 at 10 mL min1. This procedure was repeated for the consecutive detection of other HFA concentrations being employed in the IA: 0.6, 1.2, 2.5, 5.0, 10.0 and 20.0 ng mL1. Subsequently, the RU values of the reference ow cell were subtracted from the RU values obtained for captured HFA in other ow cells. The dilution of HFA was made in BSApreblocked glass vials that were prepared by incubation with 1% (w/v) BSA for 30 min. This minimizes the sample loss due to non-specic adsorption on sample tube surfaces and/or adverse effects due to altered immunogenicity.20 The SPR-based HFA detection curves were plotted with SigmaPlot, version 11.2, using a four-parameter logistic t. The EC50, R2 and Hill slope values were determined from the report generated by the soware during the standard curve analysis based on the fourparameter logistic function. The analytical sensitivity, limit of detection (LOD), and the intra- and inter-day variability were determined by the standard procedures as specied in our previous reports.21–24
Results and discussion The developed SPR IA enables highly sensitive detection of HFA in the range of 0.3–20 ng mL1 (R2 ¼ 0.999) with a LOD and analytical sensitivity of 0.7 ng mL1 and 1 ng mL1, respectively (Fig. 2A and B). The intra- and inter-assay variability were determined by ve assay repeats (in triplicate) performed on a single day and ve consecutive days, respectively. The intra- and inter-assay variability for various HFA concentrations in the IA were in the range of 1.2–5.8 and 2.1–8.6, respectively. The maximum half-effective concentration (EC50) of the SPR IA was 3.8 ng mL1 and this high sensitivity was mainly due to the higher Ab immobilization density provided by the Ab immobilization procedure. The developed Ab immobilization procedure results in a higher number of Ab molecules immobilized on the SPR chip per unit area, thereby increasing the availability of antiHFA Ab to bind to HFA. The Ab immobilization and HFA detection exhibited several advantages over conventional SPR IA
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 12 February 2014. Downloaded by UNIVERSIDAD DE BUENOS AIRES on 26/03/2018 08:43:48.
Paper
Analyst
Schematic of the developed surface plasmon resonance (SPR)-based immunoassay (IA) procedure for human fetuin A (HFA). It involves the generation of hydroxyl groups on a gold surface by treatment with 1% KOH followed by binding to the alkoxy groups of APTES in a solution comprising anti-HFA antibodies in 1% APTES. The anti-HFA antibody binds to the APTES by ionic reaction. Thereafter, the antibody-bound chip was blocked with 1% BSA and used for the detection of HFA.
Fig. 1
formats based on our previously developed covalent and CM5 dextran-based Ab immobilization procedures (Fig. S1†) (Table 1). The experimental variability in the results was circumvented as all of the IAs were performed using the same assay components and experimental conditions. The Ab immobilization density, for the various immobilization procedures, was in the decreasing order of developed > CM5 dextran > previously developed (Table 1). In contrast, the number of molecules of HFA detected was in the decreasing order of developed > previously developed > CM5 dextran (Table 1). As demonstrated in our previous report,14 an increase in Ab immobilization density does not always correspond to higher analyte detection as the Ab may not be bound in a functionally active orientation. However, the developed Ab immobilization procedure is more efficient and highly sensitive as it has both the highest Ab immobilization density and the highest number of HFA molecules detected. To our knowledge, the developed procedure is the most sensitive SPR IA procedure reported to date for HFA detection. It was further employed to detect 0.3–20 ng mL1 of HFA spiked in diluted human whole blood and plasma (Fig. 2A). The determination of HFA by the developed SPR IA correlated well with the commercial HFA sandwich ELISA (Table 2), in the range of 1.2–20 ng mL1, thereby demonstrating high analytical precision. The desired range was selected based on the analytical sensitivity of the developed IA (1 ng mL1) as it has a higher precision in the determination of HFA above this concentration. Consequently, it can be reliably employed in healthcare,
This journal is © The Royal Society of Chemistry 2014
industrial and bioanalytical settings for the detection of HFA and other disease biomarkers/analytes. Our previously developed covalent Ab immobilization procedure14 involves many steps that take 2.5 h and bind Abs by their carboxyl terminal groups, located far away from the antigen-binding sites that are present near their amino terminal regions. For comparison, the CM5 dextran-based Ab immobilization procedure binds Ab by their amino terminal groups located near their antigen-binding sites. Therefore, our previously developed procedure was better than that of CM5 dextran. But the developed Ab immobilization procedure only takes 30 min, i.e. 5-fold less time than our previously developed and CM5 dextran chip-based Ab immobilization procedures. It only involves the binding of Ab to the APTES in solution, which is more efficient than the binding of Ab to the APTES-functionalized SPR chip's surface, used in our previous procedure. Therefore, it increases the likelihood of ionic binding of the carboxyl groups on the Abs to the amino groups on APTES molecules, which simultaneously bind to the hydroxyl groups on the Au-coated SPR chip surface. The developed Ab immobilization procedure is highly simplied and cost-effective, as it only employs a single step compared to the multiple steps required and the additional chemicals used in conventional Ab immobilization procedures. The leach-proof binding of the anti-HFA capture Abs to the Au SPR chip was also analyzed by performing repeated IAs on the same anti-HFA Ab-bound SPR chip. The anti-HFA Ab-bound
Analyst, 2014, 139, 2237–2242 | 2239
View Article Online
Published on 12 February 2014. Downloaded by UNIVERSIDAD DE BUENOS AIRES on 26/03/2018 08:43:48.
Analyst
Paper
Developed SPR-based HFA IA. (A) Detection of HFA spiked in various sample matrices i.e. HEPES-buffered saline (HBS, 10 mM, pH 7.4), diluted human whole blood and diluted human plasma. (B) Sensorgram for the detection of various HFA concentrations in HBS. (C) Multiple consecutive IAs for the detection of 5 ng mL1 HFA on the same anti-HFA antibody-bound SPR chip by regenerating it after each IA using glycine–HCl (10 mM, pH 2.0). (D) Stability of the anti-HFA antibody-bound SPR chip stored at 4 C, determined by the detection of 5 ng mL1 HFA. All experiments were done in triplicate with the error bars representing the standard deviation. Fig. 2
Determination of the molecular densities of the immobilized anti-HFA antibody and the detected amount of HFA with different SPR immunoassay formats, based on various antibody immobilization procedures
Table 1
Immobilization of anti-HFA antibody Antibody immobilization procedure
DRU
a
HFAd
Mass densityb Molecular densityc (ng cm2) (molecules per cm2) DRUa
Developed 1826.0 9.9 182.6 1.0 Covalent (previously developed) 1536.0 10.4 153.6 1.0 Covalent CM5 dextran 1728.0 12.4 172.8 1.2
7.3 1011 6.1 1011 6.9 1011
Mass densityb Molecular densityc EC50 (ng cm2) (molecules per cm2) (ng mL1)
165.0 11 16.5 1.1 149.0 9.6 14.9 1.0 138.0 10.4 13.8 1.0
(2.4 0.2) 1011 (2.1 0.1) 1011 (1.9 0.1) 1011
3.8 3.8 4.1
DRU: change in resonance units (RU) caused by binding. b Calculated using the commonly used conversion factor i.e. 1000 RU ¼ 100 ng cm2 .25–29 Calculated by [mass density (ng cm2)/molecular weight (in ng)]. Molecular weight of anti-HFA antibody and HFA were 150 kDa and 43.5 kDa, respectively. In order to calculate molecular weight in SI units, the conversion factor 1 kDa ¼ 1000 Da ¼ 1000 g was used. The molecular weight of anti-HFA antibody and HFA are 24.9 1011 ng and 7.0 1011 ng, respectively. d Calculations were performed for the detection of 5 ng mL1 of HFA i.e. the concentration just above the EC50. a c
2240 | Analyst, 2014, 139, 2237–2242
This journal is © The Royal Society of Chemistry 2014
View Article Online
Paper
Analyst
Table 2 Determination of spiked HFA concentrations in diluted human whole blood and plasma by the developed SPR immunoassay and the commercial sandwich ELISA. The experiments were performed in triplicate. The results are presented as mean S.E.
Published on 12 February 2014. Downloaded by UNIVERSIDAD DE BUENOS AIRES on 26/03/2018 08:43:48.
Detected HFA conc. 1
Sample matrix
Spiked HFA conc. (ng mL )
Developed SPR immunoassay
Diluted human whole blood
20 10 5 2.5 1.2 20 10 5 2.5 1.2
20.3 10.2 5.2 2.4 1.1 20.1 10.3 4.9 2.5 1.3
Diluted human plasma
SPR chip was regenerated aer each analysis by treatment with glycine–HCl (10 mM, pH 2.0) and reused for 35 consecutive HFA IAs to detect 5 ng mL1 HFA. There was no decrease in the HFA detection response in consecutive IAs (Fig. 2C), showing the leach-proof covalent binding of Ab to the SPR chip. The storage stability of the anti-HFA Ab-bound SPR chip was also determined by storing it at 4 C and using it aer every two weeks for the detection of 5 ng mL1 HFA. There was no decrease in the HFA detection response of the Ab-bound SPR chip for 2 months and only an 18% decrease aer 4 months (Fig. 2D). The developed SPR IA procedure is ideal for mass-use applications in bioanalytical sciences, where Ab-prebound SPR chips are routinely employed for rapid analyte detection. Being highly cost-effective, simple and rapid, means that it enables the enduser to prepare the Ab-bound SPR chip in just 30 min before its intended use. Therefore, it can lead to tremendous cost-savings and enhanced analytical performance in comparison to conventional SPR IA procedures.
Conclusions We have developed a rapid and highly sensitive SPR IA procedure for the detection of HFA. It employs a highly simplied Ab immobilization procedure for the leach-proof binding of Ab to a Au SPR chip in just 30 min, i.e., 5-fold more rapid than the conventional SPR IA formats. It exhibits an increased detection range, a lower limit of detection, a better analytical sensitivity and higher cost-effectiveness than conventional SPR IA formats. The Ab-bound SPR chip can be used reproducibly for many consecutive IAs and has good storage stability of up to 4 months when stored at 4 C. The developed SPR IA detects HFA spiked in diluted human whole blood and plasma, in agreement with the commercial HFA sandwich ELISA. Therefore, it can be employed for the detection of various disease biomarkers and analytes in healthcare, industrial and bioanalytical settings.
Notes and references 1 H. Wang and A. E. Sama, Curr. Mol. Med., 2012, 12, 625–633. This journal is © The Royal Society of Chemistry 2014
0.8 0.3 0.2 0.1 0.1 0.6 0.4 0.3 0.2 0.1
Commercial sandwich ELISA 19.8 1.1 9.9 0.4 5.3 0.3 2.5 0.2 1.3 0.2 20 0.4 10.1 0.5 4.8 0.4 2.7 0.3 1.2 0.1
2 C. Y. Wu, J. Martel, W. Y. Cheng, C. C. He, D. M. Ojcius and J. D. Young, J. Biol. Chem., 2013, 288, 30571–30584. 3 L. Kalabay, L. Jakab, Z. Prohaszka, G. Fust, Z. Benko, L. Telegdy, Z. Lorincz, P. Zavodszky, P. Arnaud and B. Fekete, Eur. J. Gastroenterol. Hepatol., 2002, 14, 389–394. 4 B. Huddam, A. Azak, G. Kocak, N. Bayraktar and S. Sezer, J. Clin. Lab. Anal., 2013, 27, 317–322. 5 A. S. Goustin and A. B. Abou-Samra, Cell Signal., 2011, 23, 980–990. 6 N. Stefan and H. U. Haring, Nat. Rev. Endocrinol., 2013, 9, 144–152. 7 K. Mori, M. Emoto and M. Inaba, Recent Pat. Endocr., Metab. Immune Drug Discovery, 2011, 5, 124–146. 8 V. K. Harris, N. Donelan, Q. J. Yan, K. Clark, A. Touray, M. Rammal and S. A. Sadiq, Mult. Scler. J., 2013, 19, 1462– 1472. 9 G. Bergstr¨ om and C.-F. Mandenius, Sens. Actuators, B, 2011, 158, 265–270. 10 F. Clow, J. D. Fraser and T. Pro, Biotechnol. Lett., 2008, 30, 1603–1607. 11 M. Conti, G. Falini and B. Samor`ı, Angew. Chem., 2000, 112, 221–224. 12 D. J. O'Shannessy, M. Brigham-Burke and K. Peck, Anal. Biochem., 1992, 205, 132–136. 13 J. Svitel, H. Boukari, D. Van Ryk, R. C. Willson and P. Schuck, Biophys. J., 2007, 92, 1742–1758. 14 S. K. Vashist, C. K. Dixit, B. D. MacCraith and R. O'Kennedy, Analyst, 2011, 136, 4431–4436. 15 S. A. Walper, P. Lee, E. R. Goldman and G. P. Anderson, J. Immunol. Methods, 2013, 388, 68–77. 16 GE Healthcare Bio-Sciences AB, Biacore Sensor Surface Handbook BR-1005-71 Edition AB, 2008, https://www.biacore.com/lifesciences/products/Consumables/guide/, accessed 19th Mar., 2014. 17 E. de Juan-Franco, A. Caruz, J. R. Pedrajas and L. M. Lechuga, Analyst, 2013, 138, 2023–2031. 18 T. H. Ha, S. O. Jung, J. M. Lee, K. Y. Lee, Y. Lee, J. S. Park and B. H. Chung, Anal. Chem., 2006, 79, 546–556. 19 S. K. Vashist, Diagnostics, 2012, 2, 23–33. Analyst, 2014, 139, 2237–2242 | 2241
View Article Online
Published on 12 February 2014. Downloaded by UNIVERSIDAD DE BUENOS AIRES on 26/03/2018 08:43:48.
Analyst
20 C. K. Dixit, S. K. Vashist, B. D. MacCraith and R. O'Kennedy, Analyst, 2011, 136, 1406–1411. 21 C. K. Dixit, S. K. Vashist, B. D. MacCraith and R. O'Kennedy, Nat. Protoc., 2011, 6, 439–445. 22 C. K. Dixit, S. K. Vashist, F. T. O'Neill, B. O'Reilly, B. D. MacCraith and R. O'Kennedy, Anal. Chem., 2010, 82, 7049–7052. 23 S. K. Vashist, Biosens. Bioelectron., 2013, 40, 297–302. 24 S. K. Vashist, Anal. Biochem., 2014, 446, 96–101. 25 C. Huang, T. Stakenborg, Y. Cheng, F. Colle, T. Steylaerts, K. Jans, P. Van Dorpe and L. Lagae, Biosens. Bioelectron., 2011, 26, 3121–3126.
2242 | Analyst, 2014, 139, 2237–2242
Paper
26 S. Peeters, T. Stakenborg, G. Reekmans, W. Laureyn, L. Lagae, A. Van Aerschot and M. Van Ranst, Biosens. Bioelectron., 2008, 24, 72–77. 27 U. J¨ onsson, L. F¨ agerstam, B. Ivarsson, B. Johnsson, R. Karlsson, K. Lundh, S. L¨ of˚ as, B. Persson, H. Roos and I. R¨ onnberg, BioTechniques, 1991, 11, 620–627. 28 Y. L. Bunimovich, Y. S. Shin, W.-S. Yeo, M. Amori, G. Kwong and J. R. Heath, J. Am. Chem. Soc., 2006, 128, 16323– 16331. 29 W. Pos, J. T. Crawley, R. Fijnheer, J. Voorberg, D. A. Lane and B. M. Luken, Blood, 2010, 115, 1640–1649.
This journal is © The Royal Society of Chemistry 2014
Published on 12 February 2014. Downloaded by UNIVERSIDAD DE BUENOS AIRES on 26/03/2018 08:43:48.
PAPER
View Journal | View Issue
Cite this: Analyst, 2014, 139, 2237
Surface plasmon resonance-based immunoassay for human fetuin A† S. K. Vashist,*ab E. M. Schneiderc and J. H. T. Luongd This article describes a highly-sensitive surface plasmon resonance (SPR)-based immunoassay (IA) for human fetuin A (HFA), a specific biomarker for atherosclerosis and hepatocellular carcinoma. The assay is based on a novel immobilization procedure that simply involves the dilution of an anti-HFA capture antibody (Ab) in 1% (v/v) 3-aminopropyltriethoxysilane (APTES), followed by its dispensing on a KOHtreated gold (Au)-coated SPR chip and incubation for 30 min. The developed SPR IA detected 0.3–20 ng mL1 of HFA with a limit of detection and sensitivity of 0.7 ng mL1 and 1 ng mL1, respectively. The highly-simplified Ab immobilization procedure is also 5-fold more rapid than conventional procedures. It leads to the leach-proof binding of the capture Ab, which means that the developed SPR IA is highly
Received 20th January 2014 Accepted 10th February 2014
cost-effective, as the Ab-bound SPR chip could be reused for many repeated HFA IAs after regeneration with 10 mM glycine–HCl, pH 2.0. The Ab-bound SPR chip, stored at 4 C, lost only 18% of its original
DOI: 10.1039/c4an00149d
activity after 4 months. For the detection of HFA spiked in diluted human whole blood and plasma, the
www.rsc.org/analyst
results obtained by the developed SPR IA agreed well with the commercial HFA sandwich ELISA.
Introduction HFA, also known as alpha2-Heremans-Schmid glycoprotein (AHSG), is a product of the liver and its concentration decreases during the acute phase reaction. It displays anti-inammatory properties, counteracting proinammatory cytokine production, and functions as an inhibitor of so tissue calcication.1 Additionally, HFA has been demonstrated to constitute a major component of mineralo-organic nanoparticles which are likely responsible for the vicious cycle of inammation and calcication.2 The decrease in serum HFA concentration occurs in acute alcoholic hepatitis, chronic autoimmune hepatitis, fatty liver, alcoholic and primary biliary cirrhosis, and hepatocellular cancer.3 Most importantly, low plasma concentration conditions of HFA appear for states of insulin resistance and metabolic syndrome.4 The latter is explained by a high binding affinity of HFA to insulin receptors that specically inhibit the tyrosine kinase activity of the insulin receptor.5 Moreover, it inhibits adiponectin and thus indirectly upregulates the a
HSG-IMIT - Institut f¨ ur Mikro - und Informationstechnik, Georges-Koehler Allee 103, 79110 Freiburg, Germany. E-mail: [email protected]
b
Laboratory for MEMS Applications, Department of Microsystems Engineering IMTEK, University of Freiburg, Georges-Koehler-Allee 103, 79110 Freiburg, Germany
c Sektion Experimentelle Anaesthesiologie, University Hospital Ulm, Albert Einstein Allee 23, 89081 Ulm, Germany d
Innovative Chromatography Group, Irish Separation Science Cluster (ISSC), Department of Chemistry and Analytical, Biological Chemistry Research Facility (ABCRF), University College Cork, Cork, Ireland † Electronic supplementary 10.1039/c4an00149d
information
(ESI)
This journal is © The Royal Society of Chemistry 2014
available:
See
DOI:
secretion of inammatory cytokines from macrophages such as tumor necrosis factor alpha (TNF-a), which increases cardiovascular disease and the risk of strokes. This is why HFA is one of the most promising biomarkers used in blood and serum to identify cardiovascular dysfunction at an early stage.6 Importantly, the intriguing relevance of HFA quantication has been extended to neuroinammatory diseases,7 particularly in multiple sclerosis, where it has been detected in demyelinated lesions and in grey matter.8 Therefore, the personalized monitoring of HFA in plasma and body uid is a promising approach for metabolic risk analysis in modern communities. The availability of rapid assays for HFA quantication appears to be highly desired in the acute states of organ (heart, liver, lung and kidney) diseases related to vascular calcication. There is an urgent need for a rapid immunoassay (IA) format to detect disease biomarkers and analytes because conventional IA formats such as enzyme-linked immunosorbent assays, chemiluminescent IAs and uorescent IAs take several hours to provide a result. During the last decade, SPR has signicantly reduced the IA duration to just a few minutes. Several prospective SPR IA formats have been developed using a wide range of antibody immobilization techniques.9–15 They utilize expensive commercially available SPR chips that are pre-functionalized with carboxymethyl dextran, streptavidin, nitriloacetic acid, long chain alkanethiol molecules or lipophilic groups.16 Fragment crystallizable (Fc)-binding proteins14,17,18 such as protein A, protein G or protein A/G, for oriented Ab immobilization have also been investigated. However, all of these techniques require a large number of process steps and take several hours to immobilize the capture Ab onto the SPR chip.
Analyst, 2014, 139, 2237–2242 | 2237
View Article Online
Published on 12 February 2014. Downloaded by UNIVERSIDAD DE BUENOS AIRES on 26/03/2018 08:43:48.
Analyst
The bioanalytical performance of a SPR-based immunoassay is critically dependent on the nature of antibody immobilization as well as the affinity, orientation and stability of the bound antibody.9,14,19 This article reveals a potential SPR IA format for HFA using a highly simplied Ab immobilization procedure that enables a leach-proof binding of the capture Ab on the Aucoated SPR chip in just 30 min. The developed SPR IA employs normal Au-coated SPR chips, which are 3-fold less expensive than pre-functionalized chips. It removes the multiple process steps and additional chemicals required in conventional immobilization techniques that are widely used for binding capture Ab on pre-functionalized SPR chips. Therefore, the Ab immobilization technique used for the developed SPR IA is highly-simplied, rapid and cost-effective. The analytical performance of the capture Ab-bound SPR chip in terms of sensitivity, dynamic range, total IA duration and increased costeffectiveness will be investigated and compared with our previously developed,14 and commercial carboxymethyl (CM5) dextran chip-based SPR IA procedures.1 The applicability of the assay format for HFA detection in diluted human whole blood and plasma is then compared with the commercial sandwich ELISA procedure. The reuse of the capture Ab-bound SPR chip is also demonstrated, together with its excellent storage stability.
Experimental Materials 1-Ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC), Nhydroxysulfosuccinimide, 2-(N-morpholino)ethane sulfonic acid (MES, pH 4.7) and bovine serum albumin (BSA) were purchased from Thermo Scientic. 3-Aminopropyltriethoxysilane (APTES, purity 98%, w/v), Tween 20, H2O2 (30%, v/v) and H2SO4 (97.5%, v/v) were obtained from Sigma-Aldrich. A human fetuin A kit with all of the necessary components was obtained from R&D Systems, USA. Buffers and solutions were prepared with 18 MU Milli-Q ultrapure water (UPW) ltered through a 2 mm lter. SPR was performed using a BIAcore 3000 from GE Healthcare, Uppsala, Sweden. The SIA kit (containing SPR Au chips), carboxymethyl dextran (CMD)-functionalized Au chips, ethanolamine hydrochloride (1 M, pH 8.5), HBS-EP (0.01 M 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) pH 7.4, 0.15 M NaCl, 3 mM EDTA, 0.005% v/v surfactant P20), and glycine–HCl (10 mM, pH 2.0) were purchased from GE Healthcare. The SPR Au chip was assembled as provided by the manufacturer. HBS-EP was used as the running buffer for BIAcore and all sample dilutions were made in the running buffer. The conventional sandwich ELISA procedure was followed by the manufacturer's guidelines without any modication. Developed Ab immobilization procedure The assay format involves the sequential dilution of anti-HFA Ab in 1% (v/v) APTES (in the ratio of 1 : 1 v/v), dispensing on a KOH-treated Au SPR chip, incubation for 30 min, and washing with HEPES-buffered saline (HBS) (Fig. 1). In brief, the Au chip was cleaned by treatment with 90 mL of 1% (w/v) KOH for 5 min
2238 | Analyst, 2014, 139, 2237–2242
Paper
followed by extensive washing with UPW. The capture anti-HFA Ab (200 mg mL1 in HBS) was mixed with 1% APTES (in the ratio of 1 : 1 v/v). Thereaer, 90 mL of this capture anti-HFA Ab solution, with a nal Ab concentration of 100 mg mL1 in 0.5% APTES, was dispensed onto the Au chip and incubated for 30 min at room temperature (RT) in a fume hood. The anti-HFA Abbound Au SPR chip was washed extensively with HBS. It was then docked into the BIAcore 3000 and primed. Non-specic binding sites on the Ab-bound chip were then blocked by injecting 20 mL of 1% (w/v) BSA at 10 mL min1. Detailed information of the immobilization procedures is provided in the ESI.† HFA detection HFA (50 mL, 0.3 ng mL1) was passed through three detecting ow cells at 10 mL min1 with the fourth ow cell serving as a reference. The resultant changes in the SPR response units (RUs) were recorded for all ow cells. The Ab-bound SPR chip was regenerated aer each HFA IA by treatment with 20 mL of 10 mM glycine–HCl, pH 2.0 at 10 mL min1. This procedure was repeated for the consecutive detection of other HFA concentrations being employed in the IA: 0.6, 1.2, 2.5, 5.0, 10.0 and 20.0 ng mL1. Subsequently, the RU values of the reference ow cell were subtracted from the RU values obtained for captured HFA in other ow cells. The dilution of HFA was made in BSApreblocked glass vials that were prepared by incubation with 1% (w/v) BSA for 30 min. This minimizes the sample loss due to non-specic adsorption on sample tube surfaces and/or adverse effects due to altered immunogenicity.20 The SPR-based HFA detection curves were plotted with SigmaPlot, version 11.2, using a four-parameter logistic t. The EC50, R2 and Hill slope values were determined from the report generated by the soware during the standard curve analysis based on the fourparameter logistic function. The analytical sensitivity, limit of detection (LOD), and the intra- and inter-day variability were determined by the standard procedures as specied in our previous reports.21–24
Results and discussion The developed SPR IA enables highly sensitive detection of HFA in the range of 0.3–20 ng mL1 (R2 ¼ 0.999) with a LOD and analytical sensitivity of 0.7 ng mL1 and 1 ng mL1, respectively (Fig. 2A and B). The intra- and inter-assay variability were determined by ve assay repeats (in triplicate) performed on a single day and ve consecutive days, respectively. The intra- and inter-assay variability for various HFA concentrations in the IA were in the range of 1.2–5.8 and 2.1–8.6, respectively. The maximum half-effective concentration (EC50) of the SPR IA was 3.8 ng mL1 and this high sensitivity was mainly due to the higher Ab immobilization density provided by the Ab immobilization procedure. The developed Ab immobilization procedure results in a higher number of Ab molecules immobilized on the SPR chip per unit area, thereby increasing the availability of antiHFA Ab to bind to HFA. The Ab immobilization and HFA detection exhibited several advantages over conventional SPR IA
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 12 February 2014. Downloaded by UNIVERSIDAD DE BUENOS AIRES on 26/03/2018 08:43:48.
Paper
Analyst
Schematic of the developed surface plasmon resonance (SPR)-based immunoassay (IA) procedure for human fetuin A (HFA). It involves the generation of hydroxyl groups on a gold surface by treatment with 1% KOH followed by binding to the alkoxy groups of APTES in a solution comprising anti-HFA antibodies in 1% APTES. The anti-HFA antibody binds to the APTES by ionic reaction. Thereafter, the antibody-bound chip was blocked with 1% BSA and used for the detection of HFA.
Fig. 1
formats based on our previously developed covalent and CM5 dextran-based Ab immobilization procedures (Fig. S1†) (Table 1). The experimental variability in the results was circumvented as all of the IAs were performed using the same assay components and experimental conditions. The Ab immobilization density, for the various immobilization procedures, was in the decreasing order of developed > CM5 dextran > previously developed (Table 1). In contrast, the number of molecules of HFA detected was in the decreasing order of developed > previously developed > CM5 dextran (Table 1). As demonstrated in our previous report,14 an increase in Ab immobilization density does not always correspond to higher analyte detection as the Ab may not be bound in a functionally active orientation. However, the developed Ab immobilization procedure is more efficient and highly sensitive as it has both the highest Ab immobilization density and the highest number of HFA molecules detected. To our knowledge, the developed procedure is the most sensitive SPR IA procedure reported to date for HFA detection. It was further employed to detect 0.3–20 ng mL1 of HFA spiked in diluted human whole blood and plasma (Fig. 2A). The determination of HFA by the developed SPR IA correlated well with the commercial HFA sandwich ELISA (Table 2), in the range of 1.2–20 ng mL1, thereby demonstrating high analytical precision. The desired range was selected based on the analytical sensitivity of the developed IA (1 ng mL1) as it has a higher precision in the determination of HFA above this concentration. Consequently, it can be reliably employed in healthcare,
This journal is © The Royal Society of Chemistry 2014
industrial and bioanalytical settings for the detection of HFA and other disease biomarkers/analytes. Our previously developed covalent Ab immobilization procedure14 involves many steps that take 2.5 h and bind Abs by their carboxyl terminal groups, located far away from the antigen-binding sites that are present near their amino terminal regions. For comparison, the CM5 dextran-based Ab immobilization procedure binds Ab by their amino terminal groups located near their antigen-binding sites. Therefore, our previously developed procedure was better than that of CM5 dextran. But the developed Ab immobilization procedure only takes 30 min, i.e. 5-fold less time than our previously developed and CM5 dextran chip-based Ab immobilization procedures. It only involves the binding of Ab to the APTES in solution, which is more efficient than the binding of Ab to the APTES-functionalized SPR chip's surface, used in our previous procedure. Therefore, it increases the likelihood of ionic binding of the carboxyl groups on the Abs to the amino groups on APTES molecules, which simultaneously bind to the hydroxyl groups on the Au-coated SPR chip surface. The developed Ab immobilization procedure is highly simplied and cost-effective, as it only employs a single step compared to the multiple steps required and the additional chemicals used in conventional Ab immobilization procedures. The leach-proof binding of the anti-HFA capture Abs to the Au SPR chip was also analyzed by performing repeated IAs on the same anti-HFA Ab-bound SPR chip. The anti-HFA Ab-bound
Analyst, 2014, 139, 2237–2242 | 2239
View Article Online
Published on 12 February 2014. Downloaded by UNIVERSIDAD DE BUENOS AIRES on 26/03/2018 08:43:48.
Analyst
Paper
Developed SPR-based HFA IA. (A) Detection of HFA spiked in various sample matrices i.e. HEPES-buffered saline (HBS, 10 mM, pH 7.4), diluted human whole blood and diluted human plasma. (B) Sensorgram for the detection of various HFA concentrations in HBS. (C) Multiple consecutive IAs for the detection of 5 ng mL1 HFA on the same anti-HFA antibody-bound SPR chip by regenerating it after each IA using glycine–HCl (10 mM, pH 2.0). (D) Stability of the anti-HFA antibody-bound SPR chip stored at 4 C, determined by the detection of 5 ng mL1 HFA. All experiments were done in triplicate with the error bars representing the standard deviation. Fig. 2
Determination of the molecular densities of the immobilized anti-HFA antibody and the detected amount of HFA with different SPR immunoassay formats, based on various antibody immobilization procedures
Table 1
Immobilization of anti-HFA antibody Antibody immobilization procedure
DRU
a
HFAd
Mass densityb Molecular densityc (ng cm2) (molecules per cm2) DRUa
Developed 1826.0 9.9 182.6 1.0 Covalent (previously developed) 1536.0 10.4 153.6 1.0 Covalent CM5 dextran 1728.0 12.4 172.8 1.2
7.3 1011 6.1 1011 6.9 1011
Mass densityb Molecular densityc EC50 (ng cm2) (molecules per cm2) (ng mL1)
165.0 11 16.5 1.1 149.0 9.6 14.9 1.0 138.0 10.4 13.8 1.0
(2.4 0.2) 1011 (2.1 0.1) 1011 (1.9 0.1) 1011
3.8 3.8 4.1
DRU: change in resonance units (RU) caused by binding. b Calculated using the commonly used conversion factor i.e. 1000 RU ¼ 100 ng cm2 .25–29 Calculated by [mass density (ng cm2)/molecular weight (in ng)]. Molecular weight of anti-HFA antibody and HFA were 150 kDa and 43.5 kDa, respectively. In order to calculate molecular weight in SI units, the conversion factor 1 kDa ¼ 1000 Da ¼ 1000 g was used. The molecular weight of anti-HFA antibody and HFA are 24.9 1011 ng and 7.0 1011 ng, respectively. d Calculations were performed for the detection of 5 ng mL1 of HFA i.e. the concentration just above the EC50. a c
2240 | Analyst, 2014, 139, 2237–2242
This journal is © The Royal Society of Chemistry 2014
View Article Online
Paper
Analyst
Table 2 Determination of spiked HFA concentrations in diluted human whole blood and plasma by the developed SPR immunoassay and the commercial sandwich ELISA. The experiments were performed in triplicate. The results are presented as mean S.E.
Published on 12 February 2014. Downloaded by UNIVERSIDAD DE BUENOS AIRES on 26/03/2018 08:43:48.
Detected HFA conc. 1
Sample matrix
Spiked HFA conc. (ng mL )
Developed SPR immunoassay
Diluted human whole blood
20 10 5 2.5 1.2 20 10 5 2.5 1.2
20.3 10.2 5.2 2.4 1.1 20.1 10.3 4.9 2.5 1.3
Diluted human plasma
SPR chip was regenerated aer each analysis by treatment with glycine–HCl (10 mM, pH 2.0) and reused for 35 consecutive HFA IAs to detect 5 ng mL1 HFA. There was no decrease in the HFA detection response in consecutive IAs (Fig. 2C), showing the leach-proof covalent binding of Ab to the SPR chip. The storage stability of the anti-HFA Ab-bound SPR chip was also determined by storing it at 4 C and using it aer every two weeks for the detection of 5 ng mL1 HFA. There was no decrease in the HFA detection response of the Ab-bound SPR chip for 2 months and only an 18% decrease aer 4 months (Fig. 2D). The developed SPR IA procedure is ideal for mass-use applications in bioanalytical sciences, where Ab-prebound SPR chips are routinely employed for rapid analyte detection. Being highly cost-effective, simple and rapid, means that it enables the enduser to prepare the Ab-bound SPR chip in just 30 min before its intended use. Therefore, it can lead to tremendous cost-savings and enhanced analytical performance in comparison to conventional SPR IA procedures.
Conclusions We have developed a rapid and highly sensitive SPR IA procedure for the detection of HFA. It employs a highly simplied Ab immobilization procedure for the leach-proof binding of Ab to a Au SPR chip in just 30 min, i.e., 5-fold more rapid than the conventional SPR IA formats. It exhibits an increased detection range, a lower limit of detection, a better analytical sensitivity and higher cost-effectiveness than conventional SPR IA formats. The Ab-bound SPR chip can be used reproducibly for many consecutive IAs and has good storage stability of up to 4 months when stored at 4 C. The developed SPR IA detects HFA spiked in diluted human whole blood and plasma, in agreement with the commercial HFA sandwich ELISA. Therefore, it can be employed for the detection of various disease biomarkers and analytes in healthcare, industrial and bioanalytical settings.
Notes and references 1 H. Wang and A. E. Sama, Curr. Mol. Med., 2012, 12, 625–633. This journal is © The Royal Society of Chemistry 2014
0.8 0.3 0.2 0.1 0.1 0.6 0.4 0.3 0.2 0.1
Commercial sandwich ELISA 19.8 1.1 9.9 0.4 5.3 0.3 2.5 0.2 1.3 0.2 20 0.4 10.1 0.5 4.8 0.4 2.7 0.3 1.2 0.1
2 C. Y. Wu, J. Martel, W. Y. Cheng, C. C. He, D. M. Ojcius and J. D. Young, J. Biol. Chem., 2013, 288, 30571–30584. 3 L. Kalabay, L. Jakab, Z. Prohaszka, G. Fust, Z. Benko, L. Telegdy, Z. Lorincz, P. Zavodszky, P. Arnaud and B. Fekete, Eur. J. Gastroenterol. Hepatol., 2002, 14, 389–394. 4 B. Huddam, A. Azak, G. Kocak, N. Bayraktar and S. Sezer, J. Clin. Lab. Anal., 2013, 27, 317–322. 5 A. S. Goustin and A. B. Abou-Samra, Cell Signal., 2011, 23, 980–990. 6 N. Stefan and H. U. Haring, Nat. Rev. Endocrinol., 2013, 9, 144–152. 7 K. Mori, M. Emoto and M. Inaba, Recent Pat. Endocr., Metab. Immune Drug Discovery, 2011, 5, 124–146. 8 V. K. Harris, N. Donelan, Q. J. Yan, K. Clark, A. Touray, M. Rammal and S. A. Sadiq, Mult. Scler. J., 2013, 19, 1462– 1472. 9 G. Bergstr¨ om and C.-F. Mandenius, Sens. Actuators, B, 2011, 158, 265–270. 10 F. Clow, J. D. Fraser and T. Pro, Biotechnol. Lett., 2008, 30, 1603–1607. 11 M. Conti, G. Falini and B. Samor`ı, Angew. Chem., 2000, 112, 221–224. 12 D. J. O'Shannessy, M. Brigham-Burke and K. Peck, Anal. Biochem., 1992, 205, 132–136. 13 J. Svitel, H. Boukari, D. Van Ryk, R. C. Willson and P. Schuck, Biophys. J., 2007, 92, 1742–1758. 14 S. K. Vashist, C. K. Dixit, B. D. MacCraith and R. O'Kennedy, Analyst, 2011, 136, 4431–4436. 15 S. A. Walper, P. Lee, E. R. Goldman and G. P. Anderson, J. Immunol. Methods, 2013, 388, 68–77. 16 GE Healthcare Bio-Sciences AB, Biacore Sensor Surface Handbook BR-1005-71 Edition AB, 2008, https://www.biacore.com/lifesciences/products/Consumables/guide/, accessed 19th Mar., 2014. 17 E. de Juan-Franco, A. Caruz, J. R. Pedrajas and L. M. Lechuga, Analyst, 2013, 138, 2023–2031. 18 T. H. Ha, S. O. Jung, J. M. Lee, K. Y. Lee, Y. Lee, J. S. Park and B. H. Chung, Anal. Chem., 2006, 79, 546–556. 19 S. K. Vashist, Diagnostics, 2012, 2, 23–33. Analyst, 2014, 139, 2237–2242 | 2241
View Article Online
Published on 12 February 2014. Downloaded by UNIVERSIDAD DE BUENOS AIRES on 26/03/2018 08:43:48.
Analyst
20 C. K. Dixit, S. K. Vashist, B. D. MacCraith and R. O'Kennedy, Analyst, 2011, 136, 1406–1411. 21 C. K. Dixit, S. K. Vashist, B. D. MacCraith and R. O'Kennedy, Nat. Protoc., 2011, 6, 439–445. 22 C. K. Dixit, S. K. Vashist, F. T. O'Neill, B. O'Reilly, B. D. MacCraith and R. O'Kennedy, Anal. Chem., 2010, 82, 7049–7052. 23 S. K. Vashist, Biosens. Bioelectron., 2013, 40, 297–302. 24 S. K. Vashist, Anal. Biochem., 2014, 446, 96–101. 25 C. Huang, T. Stakenborg, Y. Cheng, F. Colle, T. Steylaerts, K. Jans, P. Van Dorpe and L. Lagae, Biosens. Bioelectron., 2011, 26, 3121–3126.
2242 | Analyst, 2014, 139, 2237–2242
Paper
26 S. Peeters, T. Stakenborg, G. Reekmans, W. Laureyn, L. Lagae, A. Van Aerschot and M. Van Ranst, Biosens. Bioelectron., 2008, 24, 72–77. 27 U. J¨ onsson, L. F¨ agerstam, B. Ivarsson, B. Johnsson, R. Karlsson, K. Lundh, S. L¨ of˚ as, B. Persson, H. Roos and I. R¨ onnberg, BioTechniques, 1991, 11, 620–627. 28 Y. L. Bunimovich, Y. S. Shin, W.-S. Yeo, M. Amori, G. Kwong and J. R. Heath, J. Am. Chem. Soc., 2006, 128, 16323– 16331. 29 W. Pos, J. T. Crawley, R. Fijnheer, J. Voorberg, D. A. Lane and B. M. Luken, Blood, 2010, 115, 1640–1649.
This journal is © The Royal Society of Chemistry 2014
