Quantitative Multianalyte Microarray Immunoassay Utilizing

Subscriber access provided by University Libraries, University of Memphis

Article

Quantitative multianalyte microarray immunoassay utilizing upconverting phosphor technology Henna Päkkilä, Minna Helena Ylihärsilä, Satu Lahtinen, Liisa Hattara, Niina Salminen, Riikka Arppe, Mika Lastusaari, Petri Saviranta, and Tero Soukka Anal. Chem., Just Accepted Manuscript • Publication Date (Web): 17 Sep 2012 Downloaded from http://pubs.acs.org on September 21, 2012

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Quantitative multianalyte microarray immunoassay utilizing upconverting phosphor technology Henna Päkkilä1*, Minna Ylihärsilä1,2*, Satu Lahtinen1, Liisa Hattara3, Niina Salminen1, Riikka Arppe1, Mika Lastusaari4,5, Petri Saviranta3, Tero Soukka1 1

2

Department of Biotechnology, University of Turku, Tykistökatu 6A, FI-20520 Turku, Finland,

Department of Virology, University of Turku, Kiinamyllynkatu 13, FI-20520 Turku, Finland, 3Medical Biotechnology Centre, VTT Technical Research Centre of Finland, Itäinen Pitkäkatu 4C, FI-20520 Turku, Finland, 4Department of Chemistry, University of Turku, Vatselankatu 2, FI-20014 Turku, Finland, 5Turku University Centre for Materials and Surfaces (MatSurf), Turku, Finland

*[email protected]; *[email protected]

Tel: +385-2-333-7081 or +358-2-333-8089. Fax: +358-2-333-8050.

A

quantitative

multianalyte

immunoassay

utilizing

luminescent

upconverting

single-crystal

nanoparticles as reporters on an antibody array-in-well platform was demonstrated. Upconverting nanoparticles are inorganic rare earth doped materials that have the unique feature converting low energy infrared radiation into higher energy visible light. Autofluorescence, commonly limiting the sensitivity of fluorescence-based assays, can be completely eliminated with photon upconversion technology because the phenomenon does not occur in biological materials. Biotinylated antibodies for

ACS Paragon Plus Environment

1


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 24

three analytes (prostate specific antigen, thyroid stimulating hormone and luteinizing hormone) were printed in an array format onto the bottom of streptavidin-coated microtiter wells. Analyte dilutions were added to the wells and the analytes were detected with antibody-coated upconverting nanoparticles. Binding of the upconverting nanoparticles was imaged with an anti-Stokes photoluminescence microwell imager and the standard curves for each analyte were quantified from the selected spot areas of the images. Single analyte and reference assays were also carried out to compare with the results of the multianalyte assay. Multiplexing did not have an effect on the assay performance. This study demonstrates the feasibility of upconverting single-crystal nanoparticles for imaging-based detection of quantitative multianalyte assays.


2

Page 3 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION Multianalyte immunoassay platforms enable simultaneous measurement of multiple protein biomarkers and they would be advantageous when several biomarkers need to be tested to ensure correct diagnosis. Multianalyte assays could enhance the diagnostic accuracy in the case of diseases for which there are no specific biomarkers and they could also be used for monitoring treatment responses (1, 2, 3). Single protein assays, such as sandwich enzyme-linked immunosorbent assay, measure only one analyte at a time, and may consume relatively large amount of specimen and reagents. Microarraybased immunoassays have capability to parallel protein measurements on the same specimen thereby reducing sample consumption and assay time. Therefore, multianalyte immunoassays could enable more effective and affordable diagnostics. Multianalyte assays can be divided into planar and bead-based suspension arrays based on the choice of support. In bead-based assays, detection occurs on microspheres. Each bead is coded and the code is used to distinguish different analyte detecting spheres from another. The codes utilized for the detection can be bar-codes (4, 5, 6, 7, 8), spectral difference of fluorophores (9, 10), fluorophore intensities (11, 12), mixtures of two or more fluorophores (13) and combinations of fluorophores and their intensities (14). Thus, multiplexing is accomplished by combining different bead sets with associated capture probes, such as antibodies, into one mix and incubating the mixture with each sample. In theory, over 10000 different codes can be generated (14) and bead-based suspension arrays enable detection of up to 100 biomarkers (15), but in practice the multiplexing degree of immunoassays has been much lower because the challenge is to find specific antibodies and assay conditions, where all different assays perform adequately. With Luminex’s xMAP coded microbead technology a multiplex assay for 25 biomarkers has been demonstrated (11). The drawback of bead-based assays is that they require dedicated instrumentation. The planar multianalyte assay format enables to perform several sandwich immunoassays in an array format either on a slide or a microtiter well. In the array-in-well format, each well of the 96-well microtiter plate contains an array of capture antibodies each recognizing a different analyte. The ACS Paragon Plus Environment

3


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 24

detection of bound analyte is usually achieved using fluorescent-labelled detection antibodies (16, 17, 18), although electrochemical detection (19, 20), enzyme labels and enzymatic signal amplification (21, 22) and label-free detection methods, including surface plasmon resonance (23), have also been used. In the planar arrays, the signal from each spot is linked to a certain analyte based on the spot position. Therefore, detection of one fluorophore is enough and the instrumentation can be simpler than in suspension arrays. For planar arrays, measurement of spot intensities is often done by scanning the array area, but also imaging-based applications are now emerging (21, 24). The optical properties of the fluorescent reporters such as luminescence intensity and photostability are important factors affecting the sensitivity of the immunoassay detection. Thus, it is highly desirable to develop reporters enhancing the efficiency of the detection. Upconverting nanoparticles (UCNPs) are inorganic materials that are exceptionally photostable and have the unique feature of being capable of converting infrared radiation to visible light via sequential absorption and energy transfer of multiple infrared photons (25). UCNPs produce strong anti-Stokes photoluminescence that can be measured using a simple fluorescence readout. Due to this, the measurement can be done entirely free of autofluorescence and scattered infrared excitation light, enabling sensitive assays (26, 27, 28). Upconverting particles (size > 200 nm) were first used in bioassays to demonstrate their applicability on toxin detection as well as cell and tissue surface antigen detection (29, 30). Previously we have demonstrated the feasibility of upconverting particles (size > 110 nm) in qualitative oligonucleotide array with imaging detection (31). Larger upconverting particles (size 400 nm) have also been applied for quantitative nucleic acid microarrays using fluorescence scanning (32), but as a drawback the scanning-based detection requires complex and expensive instrumentation. UCNPs, however, enable also high sensitivity quantitative fluorescence imaging due to the low fluorescence background. The aim of this study was to develop a quantitative multianalyte sandwich immunoassay for three analytes utilizing nano-sized (30 nm) single-crystal UCNPs combined with a CCD sensor-based microarray imager for anti-Stokes photoluminescence.


4

Page 5 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


We demonstrate the possibility of printing multiple antibodies in the array format onto the bottom of microtiter well plate, and the performance of the array-based immunoassay was evaluated using a mix of secondary antibody-coated UCNPs, each specific to one of the measured analytes. To demonstrate the performance of this platform, prostate-specific antigen (PSA), thyroid-stimulating hormone (TSH), and luteinizing hormone (LH) were selected as model analytes. PSA is widely used as a serum biomarker to assist in the diagnosis of prostate cancer and to assess response to treatment and recurrence of cancer in patients (33). TSH is essential for the maintenance of normal thyroid function, and it is used in the diagnosis of primary and secondary hypo- and hyperthyroidism (34). LH levels are used to determine menopause, to pinpoint ovulation, and in diagnosis of pituitary disorders (35). Increased LH concentrations are measured during renal failure, cirrhosis, hyperthyroidism, and severe starvation, while low levels may indicate infertility in both males and females.


5


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 24

EXPERIMENTAL SECTION Materials. All assays were carried out in assay buffer (50 mM Tris base pH 7.75, 150 mM NaCl, 0.05% NaN3, 0.01% Tween-40, 0.05% bovine gammaglobulin, 20 µM diethylenetriaminepentaacetic acid, 0.5% BSA, 20 µg/mL cherry red, 0.2% polyvinyl alcohol). Tetra-methyl-orthosilicate and (N-(3trimethoxysilyl)propyl)ethylene diamine used for silanization of the UCNPs were purchased from Sigma-Aldrich (St. Louis, MO). Monoclonal anti-PSA antibody MabH117 (36, 37, 38), monoclonal recombinant anti-TSH antibody fragment Fab5404 (39), monoclonal anti-LH MabM21241 (Fitzgerald Industries International, Acton, MA) and polyclonal rabbit anti-mouse immunoglobulins (RaM) (DakoCytomation, Glostrup, Denmark) were used for printing of the individual spots in the array. Monoclonal antibodies used for UCNP coating were anti-PSA MabH50 (36, 37, 38), anti-TSH Mab5409 (Medix Biochemica, Kauniainen, Finland), anti-LH Mab8D10 (40) (Delfia LH kit, Wallac, Finland). PSA was produced as described earlier (36), human TSH was purchased from Immuno Diagnostic (Hämeenlinna, Finland) and human LH from Sigma-Aldrich. Biotinylation of capture antibodies. Capture antibodies were biotinylated for immobilization onto the streptavidin-coated microtiter 96-well plate surface. Briefly, antibody was biotinylated with a 30fold molar excess of isothiocyanate-derived biotin (41) in 50 mM carbonate buffer, pH 9.8 and the reaction was incubated for 4 h at room temperature. Biotinylated antibody was purified with Sephadex NAP-5 and NAP-10 columns (GE Healthcare Bio-Sciences AB, Uppsala, Sweden) using 50 mM TSA, pH 7.75. Synthesis, surface modification and conjugation of antibodies to UCNPs. The single-crystal NaYF4:Yb3+,Er3+ nanomaterials (crystallite size 25-28 nm) were prepared with a method reported earlier using the RCl3·6H2O: (YbCl3·6H2O: 99.99 % and ErCl3·6H2O: 99.995 %). The concentrations of all RCl3 methanol solutions were 0.2 mol/dm3 (ratios: xYb = 0.17 mol-%, xEr = 0.03 mol-%) (42). The phase and structural purity was confirmed with X-ray powder diffraction (Huber G670; Huber Diffraktionstechnik GmbH & Co. KG, Rimsting, Germany) and morphology with transmission electron ACS Paragon Plus Environment

6

Page 7 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


microscopy (Tecnai 12 BioTwin TEM; FEI, Hillsboro, Oregon, USA) (Figures S1 and S2 in the Supporting Information). After the synthesis, the particles were silanized with slight modifications to the protocol described earlier (43). Briefly, 16 mg of UCNPs in cyclohexane were suspended in 250 µL of toluene. The mixture was heated at 80 ˚C until the cyclohexane had evaporated. Then, 10 mL of toluene, 25 µL tetra-methyl-orthosilicate and 50 µL of (N-(3-trimethoxysilyl)propyl)ethylene diamine were added and the reactants were mixed by bath sonication (Finnsonic m03; Finnsonic Oy, Lahti, Finland) for three minutes. Thereafter the reactants were mixed by vortexing and incubated at 70 ˚C for 40 min. The precipitate was collected by centrifuging (Sorvall evolution RC centrifuge) at 41500 g for 15 min and washed with toluene and ethanol. After the last washing, the precipitate was suspended in 600 µL of Milli-Q water. The silanized particles were carboxylated according to the protocol described earlier (42). The FT-IR analysis was carried out with the Nicolet Nexus 870 FT-IR equipment (Thermo Fisher Scientific, Waltham, MA). In the FT-IR spectra of the carboxylated UCNP sample, an infrared vibration band at 1716 cm-1, which is a stretching vibration from the carboxyl functional group, was observed (data not shown). For each analyte, a set of carboxylated UCNPs was conjugated with the corresponding antibody according to the procedure described earlier (44). Antibody concentration of 1 mg/mL was used for conjugation. Multianalyte array-in-well immunoassay. The biotinylated capture antibodies for three target analytes were printed to the white streptavidin-coated 96-well microtiter plates (Kaivogen, Turku, Finland) at the VTT Medical Biotechnology Centre (Turku, Finland). The biotinylated polyclonal RaM was included into the array as a positive control. The assay principle is illustrated in Figure 1. To generate standard curves, titrations with eight concentrations for LH (0-60 U/L), PSA (0-60 ng/mL) and TSH (0-200 mU/L) were made. The mixture of the standard dilutions of the three analytes was done in the assay buffer. A 10-µL volume of the standard dilution was added to the array well as triplicates together with 30 µL of the assay buffer containing antibody-coated UCNPs for TSH and PSA. Six replicates were used for the blank. The assay was incubated for 10 min at room temperature with ACS Paragon Plus Environment

7


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 24

shaking. After a 10-minute incubation, 10 µL of the antibody-coated UCNPs for LH was added and the array-in-well assay was incubated for 30 min at room temperature with shaking. The concentration of each UCNP was 20 µg/mL in the reaction. Anti-LH-UCNPs were added 10 min later than other UCNPs because without the sequential addition of the reporters, the anti-LH-UCNPs would have bound to the TSH-spots as well. This is because the tracer antibody for LH had affinity towards the alpha chain of the LH molecule which is identical in TSH and LH molecules. After the incubation, the wells were washed to remove the unbound reporters and the plate was dried. The assay was imaged with a microwell imager constructed for multianalyte detection based on a Plate Chameleon microplate reader (Hidex, Turku, Finland) (42). The parameters used in this study were: 2x binning, 1 s exposure per well and 7 W laser power. For multianalyte assay, also 0.2 and 5 s exposure times were used. The image analysis was performed with ImageJ software version 1.43n (http://rsbweb.nih.gov/ij/index.html). As a reference, array-based multianalyte assay was also measured with a modified Plate Chameleon fluorescence reader (Hidex Oy, Turku, Finland) equipped with 980 nm laser excitation and 535 nm emission bandpass filter (28). The excitation area of the laser beam of the plate reader was small (2.4 x 10-3 cm2) and thus each well was scanned with a 7 x 7 raster using 0.8 mm distance between the measurement points and a 2 s reading. From the raster, four measurement points for each analyte were selected to represent the spot. To verify potential cross-reactions and interference caused by the presence of reporters for other analytes in the multianalyte assay, single-analyte assays for PSA, TSH and LH were carried out in the array-in-wells in parallel with the multianalyte assay. Single-analyte assays were measured with the microwell imager.


8

Page 9 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1. Multianalyte array-in-well immunoassay principle. Biotinylated antibodies (Bio-Ab) were printed on streptavidin-coated (SA) microtiter wells. The captured analytes and control spots were detected with antibody-coated UCNPs. The array was imaged using a microwell imager and the standard curves for each analyte were formed from the images by quantifying the mean intensities of the spots with ImageJ software.

Reference assay. The assays for TSH, PSA and LH were performed also in regular white streptavidin-coated 96-well microtiter plates (Kaivogen, Turku, Finland) to compare the assay performance of the array-in-well assay with regular whole well assay. Biotinylated capture antibodies (2 ng/µL) were first incubated onto the well surface for 40 min, at room temperature with shaking. After the incubation, the wells were washed and analyte dilutions were made in the assay buffer. The analyte concentrations were the same that were used in the multianalyte assay (LH 0-60 U/L, PSA 0-60 ng/mL and TSH 0-200 mU/L). In each assay, a 10 µL of analyte dilution was added to the well as three replicates (six replicates of the blanks) together with 40 µL of the corresponding antibody-coated UCNPs (20 µg/mL in reaction). The assay was incubated for 40 min at room temperature with shaking. The wells were washed and the plate was dried. The photoluminescence intensities of the surface-bound UCNPs were measured with the plate reader. The central area of each well was scanned with a 3 x 3 raster with 1 mm distance between the measurement points to get a representative signal from the wells. The average of the nine measurement points covering the whole well bottom, was calculated. The assay was also imaged with a microwell imager, as described earlier. ACS Paragon Plus Environment

9


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 24

Time dependence of the signal in array-in-well format. Time dependence of the one-step array-inwell assay was studied in buffer using three PSA concentrations (0, 1.6 and 12 ng/mL) all as three replicates. After incubation of PSA dilutions and anti-PSA H50-coated UCNPs (20 µg/mL in reaction), the wells were washed and photoluminescence signals were measured with the microwell imager. The signal was measured at eight different time points after 5-120 min of incubation. Effect of sample matrix on the PSA assay in array-in-well format. The effect of sample matrix on the performance of the array-in-well format was studied with PSA assay. Blood sample was collected from a female volunteer and used in the assay with her informed consent. The plasma was stored at -20 °C and the PSA level in the sample was measured by a published PSA assay (45) before the use in the array-in-well assay. The plasma was spiked with PSA to the same standard concentrations as for the multianalyte assay and the array-in-well assay was performed as described earlier. Array-in-well data analysis. Array wells containing all the reagents except the sample were used as blanks. For each spot, a round shape area of 12064 pixels was evaluated. The mean intensities were determined by integrating over the entire spot area. Specific spot intensity was determined as the mean intensity of the pixels within the spot minus the mean intensity of the pixels within the blank. For each specific spot, signal-to-background and signal-to-noise ratios (mean specific spot intensity divided by mean intensity or standard deviation of the blank background, respectively) were obtained.

RESULTS AND DISCUSSION Time dependence of the signal. The signal level of the array-in-well assay depended on the incubation time. The time dependence was studied using blank and two different PSA concentrations. The saturated signal levels for both antigen concentrations were obtained at 60 min (Figure S3 in the Supporting Information) but the background signal increased over time. Thus the signal-to-background ratios (data not shown) started to peak already after 40 min of incubation and therefore it was chosen for all assay formats.


10

Page 11 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Assay performance. In this study, quantitative response was obtained with all three analytes in all assay formats (Figure 2). Linear fitting was used to create the standard curve. The lower limits of detection (LOD) were calculated from the standard curves as the analyte concentrations corresponding to a signal of three standard deviations (SD) of the blanks. The LODs obtained from the multianalyte assays were comparable both to the LODs from the single analyte assays performed in the array wells and to the reference assays measured with the plate reader (Table 1). Thus, the presence of reporters for other analytes did not significantly affect on the assay performance. UCNPs for only one analyte were included in the single analyte assays, thus the total UCNP concentration (20 µg/mL in reaction) was only one third compared with the UCNP concentration (60 µg/mL) in the multianalyte assay. The smaller reporter concentration had to bind both to the analyte and to the control spot, which explains the lower signal levels observed in the single analyte assays. Dynamic ranges of the assays in the multianalyte format were approximately two orders of magnitude and the signal-to-background ratios e.g. for the TSH calibration standard points were from 1 to 28 and the signal-to-noise ratios from 2 to 197. The obtained LODs and dynamic ranges were in clinically relevant ranges although, for the measurement of TSH, the LOD should have been over one order of magnitude lower to be suitable for measurement of hyperthyroidism. However, TSH is known to be a very demanding analyte requiring ultrahigh assay sensitivities and a very wide dynamic range. The sensitivity of the PSA assay was in the same range as obtained in previously published array-in-well multianalyte assay (46). However, comparison to previously published multianalyte assay results is difficult because different model analytes, detection technologies and antibodies, which are known to limit the microarray performance, have been used (47). Achievement of adequate analytical sensitivity and performance is not a simple task and many of the published papers concerning novel multianalyte assay technologies lack the adequate assay sensitivities (48, 49). Still, there are also examples of sandwich multianalyte assays reaching very good assay sensitivities and linear ranges of over five orders of magnitude. Very low detection limits in fg/mL range for tumor markers has been reached using chemiluminescence imaging, ACS Paragon Plus Environment

11


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 24

cardiac troponin I can be detected in sub picomole per litre levels and drugs of abuse in sub nanogram per litre levels using magnetic nanoparticles and optical detection (21, 50). In this study, the non-specific binding of the reporter was limiting the assay sensitivity. In addition, the surface chemistry used for biofunctionalization of the UCNPs and the assay conditions, such as buffer, may affect colloidal stability of UCNPs during the assay. This can increase the background signal originating from non-specific binding of the UCNPs. The non-specific binding cannot be completely eliminated but it is a inherent characteristic of antibodies that can be minimized by careful optimization of antibody combinations and assay conditions (51). By carefully choosing the antibodies used in the assay and optimizing the coating protocol of the UCNPs for each of the antibodies, the nonspecific binding could be further reduced. The performed multianalyte assay was stored at 4 °C for two months. The average signals after storage period were 96%, 99%, and 100% of the original signals for TSH, PSA, and LH, respectively. These results indicate excellent stability of the reporter signal and imply that assays could be archived for later read-out.


12

Page 13 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2. Calibration curves for (a) TSH, (b) PSA and (c) LH in different assay formats.


13


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 24

Table 1. Analytical performance of the different assays. TSH

Analyte

PSA

LH

LODa

R2

LODb

R2

LODc

R2

Assay type

Measurement device

Multianalyte

Microwell imager

0.64

0.936

0.17

0.985

0.45

0.998

Single analyte in array

Microwell imager

0.53

0.997

0.15

0.986

0.36

0.973

Reference assay

Microwell imager

4.5

0.990

1.0

0.989

0.55

0.997

Reference assay

Plate reader

0.31

0.998

0.23

0.996

0.031

0.998

a

Limits of detection units are mU/L, bng/mL, c U/L. R2: The correlation coefficient of the linear fitting.

Plate reader and microwell imager as detection instruments. The CCD camera of the imager is not as sensitive for light detection as the photomultiplier tube (PMT) in the plate reader. In the reference assays the LODs were approximately one order of magnitude lower with the plate reader compared with the microwell imager. In this study, the same analyte-specific UCNP concentration was used for all the assay formats but the reference assays could have been more sensitive than the multianalyte assay if the UCNP concentration had been optimized for the reference assay. The dynamic range in the CCD-based microwell imager is also limited because the maximum value for a particular pixel is 65535. This explains the narrower dynamic range in CCD-based assays compared to the assays carried out via PMT-equipped plate reader and for this reason the dynamic range for LH and PSA did not cover the whole clinically relevant range. The limited dynamic range can be compensated to some extent by measuring the array-wells with different exposure times (Figure 3). With the longer 5 s exposure time in multianalyte assay format, lower LOD for TSH (0.38 mU/L) was obtained but the drawback was that the CCD rapidly reached the saturation charge level. With shorter 0.2 s exposure time, the smallest analyte concentrations did not deviate from the background but higher ACS Paragon Plus Environment

14

Page 15 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


analyte concentrations could be measured without saturation of the CCD. By combining different measurement times, a standard curve covering wider analyte concentration range can be obtained. Extending the dynamic range by combining different exposure times has been demonstrated earlier for imaging (52). To our knowledge, the combination of exposure times has not been used for array-based immunoassays before, because long exposure times have resulted in increase of the background signal (24, 53). UCNPs enable the use of different exposure times because of the low background and high photostability inherent to UCNP technology. Another factor limiting the dynamic range could be the spot itself, which has lower binding capacity compared to the whole wells used in the reference assay. The scanning measurement with the plate reader did not increase the linear range of the multianalyte assays (Figure S4 in the Supporting Information). This means that the wide dynamic range of the reference assays measured with the plate reader does not originate solely from different measurement technology but binding capacity of the solid phase also affects to the dynamic range. The plate reader was not optimal for scanning of the arrays as the resolution was not adequate. Thus the LODs for the multianalyte array-in-well assays were different compared to the results obtained with the microwell imager. The LODs for PSA and LH (0.05 ng/mL and 0.20 U/L, respectively) were lower with the plate reader implying that it is more sensitive but the LOD for TSH (1.7 mU/L) was significantly higher than with the microwell imager. This was because of the high non-specific binding of the label to the area around TSH spot. The advantage of imaging is that it is easier to discover inhomogeneities in the array or in the spot compared to low resolution scanning.


15


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 24

Figure 3. Effect of exposure times on signal levels in TSH in multianalyte assay in array wells.

PSA assay in array-in-well format using plasma. To demonstrate the array-in-well immunoassay applicability for clinically relevant samples, PSA was used as a model analyte in the single analyte assay. The measured total PSA level in the female plasma sample was 0.06 ng/mL, which was below the detection limit of the array-in-well assay (Table 1). The standard curve obtained in lithium heparin plasma was similar than in buffer although the signal levels were lower (Figure S5 in the Supporting Information). Thus the detection limit in the plasma was six times worse as the detection limit obtained in buffer. Formation of non-immunoreactive complexes of PSA to PSA binding proteins present in the sample is the principal reason for compromised performance in plasma (54). Cross-reactivity. In multianalyte assays, high background and low sensitivity are frequent problems caused by the non-specific binding of antibodies (11). Therefore, highly specific antibodies are important for avoiding cross-reactivity between antibodies and other proteins in the specimen. The cross-reactivity leads to reduced specificity, sensitivity, and false quantification. In the single analyte array-in-wells, the analytes bound only to the corresponding spots and no cross-reactions were detected (Figure 4). The dilution series from the array-based single analyte assays are shown in Figure S6 in the ACS Paragon Plus Environment

16

Page 17 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Supporting Information. RaM spots that were used as positive control were clearly visible in all array wells confirming the functionality of the array performance.

Figure 4. (a) A fluorescence image of the multianalyte array-in-well immunoassay containing: TSH (200 mU/L), PSA (60 ng/mL), and LH (60 U/L). Fluorescence images of the singe-analyte array-in-well immunoassay results for analytes (b) TSH (200 mU/L), (c) PSA (60 ng/mL), and (d) LH (60 U/L).

CONCLUSIONS In this study, we have developed a quantitative multianalyte immunoassay utilizing UCNP technology and simple anti-Stokes photoluminescence imaging instrumentation for the measurement of the assay. The imaging was possible because of the low fluorescence background which is inherent advantage of upconversion technology. Also, no optimization of assay materials for fluorescence background was needed with UCNPs. The multianalyte array-in-well immunoassay was carried out in 96-well microtiter plates, which offers a high sample throughput as simultaneous detection of 96 samples could be carried out in one plate. Compared to slide-based arrays, array-in-well format enables the use of conventional


17


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 24

instrumentation, such as plate washers, to perform assay steps. The multianalyte immunoassay included the detection of PSA, TSH and LH. The obtained multianalyte assay sensitivity for PSA and LH were in the clinically relevant range and it was demonstrated that the linear range of the assays could be extended by combining measurements using different exposure times. However, the assay sensitivity for TSH needs to be improved for the measurement of hyperthyroidism. The multianalyte immunoassay platform showed potential for developing customized quantitative assays for protein biomarkers in plasma samples for the need of clinical diagnostics. The technology could be applied to e.g. cardiac marker multianalyte assay, drug discovery, and other biomarker panels where more effective diagnosis is needed.

ACKNOWLEDGMENTS H.P. and M.Y. contributed equally to this work. The authors thank Jorma Hölsä and Minnea Tuomisto for synthesizing the upconverting nanoparticles used in this study and Pirjo Laaksonen for drawing blood samples. This study was supported by Tekes, the Finnish Funding Agency for Technology and Innovation, Academy of Finland (Grant number 140758) and DIA-NET, The Graduate School of Advanced Diagnostic Technologies and Applications.

Supporting Information Available: This material is available free of charge via the Internet at http://pubs.acs.org.

REFERENCES (1)

Martins, T. B.; Litwin, C. M.; Hill, H. R. Am.J.Clin.Pathol. 2008, 129, 34-41.

(2)

Ramasamy, I. Clin.Chim.Acta. 2011, 412, 1279-1296.


18

Page 19 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(3)


Cohen, M. S.; Shaw, G. M.; McMichael, A. J.; Haynes, B. F. N.Engl.J.Med. 2011, 364, 1943-

1954. (4)

Appleyard, D. C.; Chapin, S. C.; Doyle, P. S. Anal.Chem. 2011, 83, 193-199.

(5)

Wang, J.; Ahmad, H.; Ma, C.; Shi, Q.; Vermesh, O.; Vermesh, U.; Heath, J. Lab.Chip. 2010, 10,

3157-3162. (6)

Derveaux, S.; Stubbe, B. G.; Roelant, C.; Leblans, M.; De Geest, B. G.; Demeester, J.; De

Smedt, S. C. Anal.Chem. 2008, 80, 85-94. (7)

Nicewarner-Pena, S. R.; Freeman, R. G.; Reiss, B. D.; He, L.; Pena, D. J.; Walton, I. D.;

Cromer, R.; Keating, C. D.; Natan, M. J. Science. 2001, 294, 137-141. (8)

Pregibon, D. C.; Toner, M.; Doyle, P. S. Science. 2007, 315, 1393-1396.

(9)

Wang, L.; Yang, C.; Tan, W. Nano Lett. 2005, 5, 37-43.

(10) Klostranec, J. M.; Xiang, Q.; Farcas, G. A.; Lee, J. A.; Rhee, A.; Lafferty, E. I.; Perrault, S. D.; Kain, K. C.; Chan, W. C. Nano Lett. 2007, 7, 2812-2818. (11) Skogstrand, K.; Thorsen, P.; Norgaard-Pedersen, B.; Schendel, D. E.; Sorensen, L. C.; Hougaard, D. M. Clin.Chem. 2005, 51, 1854-1866. (12) Cook, E. B.; Stahl, J. L.; Lowe, L.; Chen, R.; Morgan, E.; Wilson, J.; Varro, R.; Chan, A.; Graziano, F. M.; Barney, N. P. J.Immunol.Methods. 2001, 254, 109-118. (13) Gazagne, A.; Claret, E.; Wijdenes, J.; Yssel, H.; Bousquet, F.; Levy, E.; Vielh, P.; Scotte, F.; Goupil, T. L.; Fridman, W. H.; Tartour, E. J.Immunol.Methods. 2003, 283, 91-98. (14) Han, M.; Gao, X.; Su, J. Z.; Nie, S. Nat.Biotechnol. 2001, 19, 631-635. (15) Vignali, D. A. J.Immunol.Methods. 2000, 243, 243-255.


19


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 24

(16) Peng, C.; Li, Z.; Zhu, Y.; Chen, W.; Yuan, Y.; Liu, L.; Li, Q.; Xu, D.; Qiao, R.; Wang, L.; Zhu, S.; Jin, Z.; Xu, C. Biosens.Bioelectron. 2009, 24, 3657-3662. (17) King, E. M.; Vailes, L. D.; Tsay, A.; Satinover, S. M.; Chapman, M. D. J.Allergy Clin.Immunol. 2007, 120, 1126-1131. (18) Grossmann, K.; Roggenbuck, D.; Schroder, C.; Conrad, K.; Schierack, P.; Sack, U. Cytometry A. 2011, 79, 118-125. (19) Wu, J.; Yan, F.; Tang, J.; Zhai, C.; Ju, H. Clin.Chem. 2007, 53, 1495-1502. (20) Lai, G.; Yan, F.; Wu, J.; Leng, C.; Ju, H. Anal.Chem. 2011, 83, 2726-2732. (21) Zong, C.; Wu, J.; Wang, C.; Ju, H.; Yan, F. Anal.Chem. 2012, 84, 2410-2415. (22) Chan, S. M.; Ermann, J.; Su, L.; Fathman, C. G.; Utz, P. J. Nat.Med. 2004, 10, 1390-1396. (23) Rebe Raz, S.; Liu, H.; Norde, W.; Bremer, M. G. Anal.Chem. 2010, 82, 8485-8491. (24) Ahmad, F.; Seyrig, G.; Tourlousse, D. M.; Stedtfeld, R. D.; Tiedje, J. M.; Hashsham, S. A. Biomed.Microdevices. 2011, 13, 929-937. (25) Auzel, F. Chem.Rev. 2004, 104, 139-173. (26) Kuningas, K.; Ukonaho, T.; Päkkilä, H.; Rantanen, T.; Rosenberg, J.; Lövgren, T.; Soukka, T. Anal.Chem. 2006, 78, 4690-4696. (27) Ukonaho, T.; Rantanen, T.; Jämsen, L.; Kuningas, K.; Päkkilä, H.; Lövgren, T.; Soukka, T. Anal.Chim.Acta. 2007, 596, 106-115. (28) Soukka, T.; Kuningas, K.; Rantanen, T.; Haaslahti, V.; Lövgren, T. J.Fluoresc. 2005, 15, 513528.


20

Page 21 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(29) Wright, W. H.; Mufti, N. A.; Tagg, N. T.; Webb, R. R.; Schneider, L. V. Proc. SPIE Int. Soc. Opt. Eng. 1997, 2985, 248-255. (30) Zijlmans, H. J.; Bonnet, J.; Burton, J.; Kardos, K.; Vail, T.; Niedbala, R. S.; Tanke, H. J. Anal.Biochem. 1999, 267, 30-36. (31) Ylihärsilä, M.; Valta, T.; Hattara, L.; Karp, M.; Harju, E.; Hölsä, J.; Saviranta, P.; Waris, M.; Soukka T. Anal. Chem. 2011, 83, 1456-1461. (32) van De Rijke, F.; Zijlmans, H.; Li, S.; Vail, T.; Raap, A. K.; Niedbala, R. S.; Tanke, H. J. Nat.Biotechnol. 2001, 19, 273-276. (33) Thompson, I. M.; Ankerst, D. P. CMAJ. 2007, 176, 1853-1858. (34) Baskin, H. J.; Cobin, R. H.; Duick, D. S.; Gharib, H.; Guttler, R. B.; Kaplan, M. M.; Segal, R. L. Endocr.Pract. 2002, 8, 457-469. (35) Tsutsumi, R.; Webster, N. J. Endocr.J. 2009, 56, 729-737. (36) Rajakoski, K.; Piironen, T.; Pettersson, K.; Lövgren, J.; Karp, M. Prostate Cancer.Prostatic Dis. 1997, 1, 16-20. (37) Piironen, T.; Villoutreix, B. O.; Becker, C.; Hollingsworth, K.; Vihinen, M.; Bridon, D.; Qiu, X.; Rapp, J.; Dowell, B.; Lövgren, T.; Pettersson, K.; Lilja, H. Protein Sci. 1998, 7, 259-269. (38) Piironen, T.; Nurmi, M.; Irjala, K.; Heinonen, O.; Lilja, H.; Lövgren, T.; Pettersson, K. Clin.Chem. 2001, 47, 703-711. (39) Ylikotila, J.; Välimaa, L.; Vehniäinen, M.; Takalo, H.; Lövgren, T.; Pettersson, K. J.Immunol.Methods. 2005, 306, 104-114. (40) Nilsson, C.; Seppälä, M.; Pettersson, K. J.Endocrinol. 2001, 168, 107-116.


21


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(41) Qin,

Q.;

Christiansen,

M.;

Lövgren,

T.;

Norgaard-Pedersen,

Page 22 of 24

B.;

Pettersson,

K.

J.Immunol.Methods. 1997, 205, 169-175. (42) Ylihärsilä, M.; Harju, E.; Arppe, R.; Hattara, L.; Hölsä, J.; Saviranta, P.; Soukka, T.; Waris, M. Clin.Microbiol.Infect. 2012, Accepted Article, DOI: 10.1111/j.1469-0691.2012.03926.x. (43) Jana, N. R.; Earhart, C.; Ying, J. Y. Chemistry of Materials. 2007, 19, 5074-5082. (44) Kuningas, K.; Rantanen, T.; Ukonaho, T.; Lövgren, T.; Soukka, T. Anal.Chem. 2005, 77, 73487355. (45) Väisänen, V.; Peltola, M. T.; Lilja, H.; Nurmi, M.; Pettersson, K. Anal.Chem. 2006, 78, 78097815. (46) Wiese, R.; Belosludtsev, Y.; Powdrill, T.; Thompson, P.; Hogan, M. Clin.Chem. 2001, 47, 14511457. (47) Nielsen, U. B.; Geierstanger, B. H. J.Immunol.Methods. 2004, 290, 107-120. (48) Copeland, S.; Siddiqui, J.; Remick, D. J.Immunol.Methods. 2004, 284, 99-106. (49) Park, J.; Sunkara, V.; Kim, T. H.; Hwang, H.; Cho, Y. K. Anal.Chem. 2012, 84, 2133-2140. (50) Bruls, D. M.; Evers, T. H.; Kahlman, J. A.; van Lankvelt, P. J.; Ovsyanko, M.; Pelssers, E. G.; Schleipen, J. J.; de Theije, F. K.; Verschuren, C. A.; van der Wijk, T.; van Zon, J. B.; Dittmer, W. U.; Immink, A. H.; Nieuwenhuis, J. H.; Prins, M. W. Lab.Chip. 2009, 9, 3504-3510. (51) Ellington, A. A.; Kullo, I. J.; Bailey, K. R.; Klee, G. G. Clin.Chem. 2010, 56, 186-193. (52) Brauers, J.; Schulte, N.; Bell, A.; Aach, T. In Multispectral high dynamic range imaging; IS\&T/SPIE Electronic Imaging: San Jose, 2008, 680704:1-680704:12.


22

Page 23 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(53) Momeni, N.; Ramanathan, K.; Larsson, P.; Danielsson, B.; Bengmark, S.; Khayyami, M. Anal.Chim.Acta. 1999, 387, 21-27. (54) Wu, P.; Zhu, L.; Stenman, U. H.; Leinonen, J. Clin.Chem. 2004, 50, 125-129.


23


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 24

For TOC only:


24

Quantitative Multianalyte Microarray Immunoassay Utilizing

Recommend Documents