Redox and Label-Free Array Detection of Protein ... - ACS Publications

May 9, 2014 - Label-free electroanalytical methods present, arguably, the most ... Citation data is made available by participants in Crossref's Cited...
0 downloads 0 Views 2MB Size
Article pubs.acs.org/ac

Redox and Label-Free Array Detection of Protein Markers in Human Serum Xiliang Luo,†,‡ Qiao Xu,‡ Tim James,§ and Jason J. Davis*,‡ †

Key Laboratory of Biochemical Analysis, Ministry of Education, College of Chemistry and Molecular Engineering, Qingdao University of Science and Technology, Qingdao 266042, China ‡ Department of Chemistry, University of Oxford, Oxford OX1 3QZ, U.K. § Department of Clinical Biochemistry, Oxford University Hospitals NHS Trust, Oxford OX3 9DU, U.K. S Supporting Information *

ABSTRACT: A substantial outstanding challenge in diagnostics and disease monitoring is an ability to rapidly and conveniently assay for protein biomarkers within complex biological media. Label-free electroanalytical methods present, arguably, the most promising and scalable means of achieving this but, as with all label-free assays, can struggle with response selectivity issues that arise from nonspecific surface interactions. Impedimetric methods are ultrasensitive and have been applied to the quantification of a wide range of proteins but have not previously been utilized in a multiplexed format capable of operation in complex analytical fluid. Herein, we present the use of thermally cross-linked poly(ethylene glycol) (PEG) polymer sensory array interfaces in the ultrasensitive quantification of two protein markers, insulin and C-reactive protein (CRP). This was achieved with detection limits of 171 ± 19 fM and 150 ± 10 pM, respectively. Significantly, the arrays not only enable the simultaneous, fast, nonamplified, and label-free detection of both markers without reagent addition but do so with little cross talk, even in human serum. A blind analysis of 17 real patient samples generated results in excellent agreement with those obtained through a clinically approved chemiluminescence assay.

D

Although these have become increasingly potent, the vast majority of reports are concerned with only the detection of single markers in buffered aqueous solution.8 Though a number of truly multiplexed examples have been presented,9 they have been based invariably on the utilization of multistep sandwich formats, necessitating the use of two antibodies (one of which is labeled in generating an amplified signal), and the assumption of a consistent presentation of two non overlapping epitopes on the target surface, for each target. A notable exception has been the work of the Lieber group,10 which demonstrated the simultaneous label-free electrical detection of three cancer markers within silicon nanowire sensor arrays. The demands associated with generating and scaling up highly reproducible nanowire arrays and their utility in high ionic strength solutions, are, though, significant. In recent years, we have developed a number of approaches to the ultrasensitive impedimetric assaying of specific markers in complex biological fluid through the use of a monolayer or polymer-film-based nonfouling chemistry and subsequent impedance or capacitative analyses.11 PEG is a nontoxic, hydrophilic biocompatible polymer that, when integrated with

uring the past two decades, the number of proposed and validated biomarkers has increased dramatically. To date, for example, there are accepted (though not necessarily individually specific, see below) markers for every common form of cancer as well as a range of cardiac diseases.1 An ability to assay such markers is of obvious profound value in early (and potentially lifesaving) diagnosis and in the monitoring of response to therapeutic intervention.1a,2 For many complex diseases, particularly those with a high innate heterogeneity, such as cancer, it has become increasingly clear that a single biomarker will, alone, often be insufficient to diagnose with sufficiently high specificity.3 A multiplexed assay of two or more biomarkers in the same biological sample is, therefore, of considerable value.4 It is, additionally, likely that future live and continuous reports of physiological status will invoke an assessment of a panel of markers, where relative protein levels will be of greater diagnostic utility than the absolute quantification of one marker alone.5 From a practical (time and cost) point of view, multiplexed assays are broadly valuable in facilitating rapid disease screening and minimizing patient stress.6 Among the various detection methods capable of supporting multiplexed assays,7 electrical approaches are of particular interest because of their innate combination of low cost, high sensitivity, and integration with microelectronic technologies. © 2014 American Chemical Society

Received: March 19, 2014 Accepted: May 9, 2014 Published: May 9, 2014 5553

dx.doi.org/10.1021/ac5010037 | Anal. Chem. 2014, 86, 5553−5558

Analytical Chemistry

Article

Figure 1. (A) Utilized microfabricated arrays (left) and their associated fluidic housing (right). PEGylated polymer modification is followed by a manual antibody functionalization of the electrodes (three in any array selective to one target, three to the second target). The volume of the measuring cell is 50 μL, with fluids injected/expelled through the luer lock inlets/outlets. (B) Schematic representation of the cross-linked functional PEG polymer electrode modification and subsequent antibody integration.

Anonymized patient serum samples were collected by Clinical Biochemistry, John Radcliffe Hospital (Oxford, U.K.). Monoclonal anti-insulin antibody (mouse IgG1 isotype) was purchased from Santa Cruz Biotechnology, Inc., and goat antihuman CRP polyclonal antibody was purchased from AbD Serotec. Phosphate-buffered saline (PBS, 10 mM, pH 7.4) was prepared by dissolving PBS tablets (Sigma-Aldrich) in ultrapure water and filtering using a 0.22 μm membrane filter. All other chemicals were of analytical grade. Ultrapure water (18.2 MΩ/ cm) was obtained from an Ultrapure (Elga Water Flex3) system and used throughout. Electrochemical experiments were performed on an Autolab Potentiostat 12 equipped with an FRA2 module (Metrohm Autolab B.V.) using a three-electrode system. Microelectrode arrays with six gold working electrodes (200 μm in diameter) and a shared counter electrode (Figure 1A left) were designed in-house (Figure 1B right) and fabricated by Triteq Ltd., by standard printed circuit board (PCB) microfabrication techniques. A silver/silver chloride (Ag/AgCl, filled with 1.0 M KCl) reference electrode, located through the middle of the fluidic housing, was purchased from CH Instruments. The analytical solutions (50 μL) are injected through the inlet pipe of the fluidic housing by syringe and allowed to equilibrate, and then the cell is flushed with PBS prior to analysis in static PBS. The cells are rinsed at least three times with PBS prior to any repetitions. Sensor Surface Preparation. The preparation process is schematically summarized in Figure 1. Gold array microelectrodes were first rinsed with ethanol, followed by washing with water and electrochemical pretreatment in 0.5 M H2SO4 using cyclic voltammetry (CV) over the potential range from

solid surfaces, can be highly effective in reducing unwanted interactions with cells12 and proteins.11b Such interfaces operate through what is commonly described as “steric repulsion”.13 Although the precise mechanistic details have varied, it is clear that both unfavorable entropic change through (locally disordered) polymer compression and enthalpy change through enforced polymer desolvation are important.13c Thiolated PEG SAMs have been used productively to support specific protein detection in a variety of mixtures, most notably in serum.11c,13a Herein we have sought to utilize a commercial PEGylated monomer in generating, in a very simple manner, electrodeconfined cross-linked PEG polymer films. These were envisaged to present both very high levels of biocompatibility (important in maximizing antibody efficacy) and nonfouling characteristics at least as good as those associated with thiolated PEG SAMs. To the best of our knowledge, these films have not been utilized in any previous analytical platform. This capability was integrated within fabricated microelectrode arrays. Through the spatially controlled attachment of two different antibodies, the simultaneous and convenient assaying of insulin and C-reactive protein (CRP) from human patient serum samples can be achieved with clinically relevant reliability.



EXPERIMENTAL SECTION Materials and Apparatus. Human insulin, human Creactive protein (CRP), human blood serum, bovine serum albumin (BSA), 1-ethyl-3-(3-(dimethylamino)propyl) carbodiimide (EDC), and N-hydroxysuccinimde (NHS) were purchased from Sigma-Aldrich. In addition, 4-armed PEG-epoxide (molecular weight 2K) and PEG-amine (molecular weight 2K) precursors were purchased from Creative PEGWorks (U.S.A.). 5554

dx.doi.org/10.1021/ac5010037 | Anal. Chem. 2014, 86, 5553−5558

Analytical Chemistry

Article

Figure 2. Non-Faradaic impedance change of the antibody−polymer array interfaces after incubation with specified concentrations of insulin (A) and CRP (B) in PBS (10 mM, pH 7.4). Inset shows the corresponding calibration curve. Error bars represent the standard deviations of three measurements.

resulting thermo-polymerized film11 is sufficiently rich in accessible amine groups to facilitate convenient subsequent antibody attachment. Film generation was associated with an expected increase in baseline impedance (from 30 to 50 MOhm at 0.1 Hz, Figure S2, SI) and an air-dried, ellipsometer-derived, thickness of 7.15 ± 0.75 nm. In view of the volumetric swelling ratio of PEG polymers in PBS (∼6 at room temperature),14 these films are expected to be ∼40 nm thick when presented in PBS. Standard carbodiimide chemistry enables the subsequent covalent tethering of antibodies to the film. The so-prepared interfaces exhibit very high stability (