Rapid and sensitive cardiac troponin I immunoassay based on fluorescent europium(III)-chelate-dyed nanoparticles

Clinica Chimica Acta 414 (2012) 70–75

Contents lists available at SciVerse ScienceDirect

Clinica Chimica Acta journal homepage: www.elsevier.com/locate/clinchim

Rapid and sensitive cardiac troponin I immunoassay based on fluorescent europium(III)-chelate-dyed nanoparticles Marja-Leena Järvenpää a,⁎, Katri Kuningas a, Ilari Niemi a, Pirjo Hedberg b, Noora Ristiniemi a, Kim Pettersson a, Timo Lövgren a a b

Department of Biotechnology, University of Turku, Turku, Finland Laboratory of Oulu University Hospital, Department of Clinical Chemistry, University of Oulu, Oulu, Finland

a r t i c l e

i n f o

Article history: Received 23 December 2011 Received in revised form 26 August 2012 Accepted 27 August 2012 Available online 4 September 2012 Keywords: Troponin I Immunoassay Nanoparticle Time-resolved fluorometry

a b s t r a c t Background: Cardiac troponins are the preferred and recommended biomarkers of myocardial infarction. Unfortunately, most of the current commercial assays do not meet the guideline recommendations for sensitivity and low-end precision. Therefore, improvements in their analytical performance are still needed. Methods: Cardiac troponin I (cTnI) immunoassay was developed. The assay utilized a monoclonal antibody and a F(ab')2 antibody fragment immobilized onto the microtiter wells for capturing, and a monoclonal antibody covalently conjugated to fluorescent europium(III)-chelate-dyed nanoparticles for detecting. Following a 15-min incubation of the sample and nanoparticle-bioconjugates in the capture wells, cTnI was quantified directly from the washed well surface by time-resolved fluorometry. Results: The limits of detection and quantification were 0.0020 μg/l and 0.012 μg/l, respectively. The response was linear in the measured range of 0.003–9.6 μg/l. The within-run imprecisions were 9.8, 5.1, 7.7 and 5.4%, and the total imprecisions were 13.1, 10.4, 9.0 and 8.7% at cTnI levels of 0.007, 0.051, 0.52 and 2.62 μg/l, respectively. Plasma recoveries of added cTnI were 72–119%. Regression analysis with Innotrac Aio! 2nd generation cTnI assay yielded a slope (95% confidence intervals) of 1.197 (1.141 to 1.253) and y-intercept of 0.216 (−0.128 to 0.561) μg/l (Syx = 2.176 μg/l, n = 212, r = 0.945). Conclusions: The developed immunoassay based on europium(III)-chelate-dyed nanoparticle label allows rapid and sensitive measurement of cTnI. © 2012 Elsevier B.V. All rights reserved.

1. Introduction Cardiac troponins I and T (cTnI and cTnT, respectively) are the preferred and recommended biochemical markers of myocardial infarction (MI). In order to detect also minor cardiac muscle damage and to identify patients at risk for cardiovascular diseases, the current guidelines propose that troponin assays should be able to measure elevations above the 99th percentile concentration of a healthy population. Furthermore, at the 99th percentile, the coefficient of variation (CV) should be 10% or less. While cTnT has been predominantly measured by a single assay from Roche Diagnostics [1], several assays exist for cTnI. Unfortunately, most of the current commercial assays are not capable of meeting the suggested requirements [2,3], and

Abbreviations: cTnI, cardiac troponin I; cTnT, cardiac troponin T; CV, coefficient of variation; EDC, N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide; LoD, limit of detection; LoQ, limit of quantification; mAb, monoclonal antibody; MI, myocardial infarction; SA, streptavidin; Sulfo-NHS, N-hydroxysulfosuccinimide; TnC, troponin C. ⁎ Corresponding author at: Department of Biotechnology, University of Turku, Tykistökatu 6A, 6th floor, 20520 Turku, Finland. Tel.: +358 2 333 8572; fax: +358 2 333 8050. E-mail address: mljarv@utu.fi (M.-L. Järvenpää). 0009-8981/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.cca.2012.08.027

thus, improvements in their analytical sensitivity and low-end precision are still needed. One important issue to be considered when constructing a sensitive assay is the selection of the label and detection technology. Current commercial cardiac troponin assays are mainly enzyme immunoassays, where an immunocomplex formed between two or three anti-troponin antibodies is subsequently detected in most cases by fluorescence or chemiluminescence [2]. Recently, a new generation of high-sensitive research assays utilizing novel detection methods have emerged and can reliably detect cTnI concentrations below 0.001 μg/l. Erenna® System from Singulex (Alameda, CA) [4,5] is based on capillary flow single-molecule counting and Verigene® System from Nanosphere (Northbrook, IL) [6] on chemical signal enhancement of gold nanoparticles. Previously, highly fluorescent nanoparticle labels containing tens of thousands of europium(III)-chelates inside a polystyrene shell [7] have enabled very low detection limits for various analytes such as prostate-specific antigen [8], thyroid-stimulating hormone [9] and viral antigens [10,11]. The benefits of lanthanide-based time-resolved luminescence for developing sensitive immunoassays have been recently reviewed [12]. The main advantage of europium and other lanthanide chelates is their significantly long fluorescence lifetime

M.-L. Järvenpää et al. / Clinica Chimica Acta 414 (2012) 70–75

allowing lanthanide luminescence to be selectively discriminated from the background autofluorescence using temporal resolution. Therefore, higher sensitivity compared to conventional fluorescent labels is achieved. The potential for developing an ultrasensitive assay with lanthanide chelates packed inside a nanoscale shell stems from high specific activity of the particulate label. One binding event of an antibody conjugated to the nanoparticle surface results in an amplified signal intensity compared to a soluble antibody coupled with a small number of individual label moieties. In addition, high binding site density on the nanoparticle surface has been shown to enhance the inherent affinity of the original antibody [13]. In this report, we demonstrate that a sensitive and convenient immunoassay for cTnI can be constructed using europium(III)nanoparticle as a label. The developed assay is a one-step, two-site immunoassay utilizing a monoclonal antibody (mAb) and a F(ab')2 antibody fragment immobilized to microtiter well surface for capturing, and a mAb covalently conjugated to europium(III)-chelate-dyed nanoparticles for detecting. Following a 15-min incubation of sample and nanoparticles in the capture wells, cTnI is quantified directly from the washed well surface by time-resolved fluorometry. 2. Materials and methods 2.1. cTnI standards Human cardiac troponin (native, tissue-derived cTnI–cTnT–troponin C (TnC) complex) was purchased from HyTest, Ltd. (Turku, Finland). Standards were prepared by diluting the troponin complex in TSA buffer (50 mmol/l Tris–HCl, pH 7.75, 9 g/l NaCl, 0.5 g/l NaN3) supplemented with 75 g/l bovine serum albumin (BSA). Standards were divided into aliquots and stored at −20 °C so that a new set of standards was thawed for every assay. 2.2. Plasma samples Lithium heparin plasma from apparently healthy young volunteers (n = 17) were collected at the Department of Biotechnology, University of Turku (Turku, Finland), and used in this study with the consent of the blood donors. All specimens were tested for cardiac troponin specific autoantibodies according to the protocol previously described in detail [14,15]. Normal plasma pool was composed of the specimens that gave a negative result in autoantibody assay as well as had the measured cTnI concentration below the detection limit in the developed nanoparticle-based assay. Lithium heparin plasma from chest pain patients (n = 226) were randomly collected at the Laboratory of Oulu University Hospital (Oulu, Finland). For the anonymity of the patients, all identifying labels were removed at the hospital before the specimens were shipped frozen to the Department of Biotechnology, University of Turku. During the study, normal plasma specimens were stored at −20 °C and patient plasma specimens at −70 °C. All samples were thawed, mixed and centrifuged (1 min, 2000 g) to remove any particulate material before analysis with the developed assay. 2.3. Antibodies A monoclonal antibody (mAb) recognizing the epitope at amino acids (aa) 137–148 (8I7-mAb; International Point of Care, Toronto, Canada) was used as a detection antibody. Two antibodies, a mAb recognizing the epitope at aa 41–49 (19C7-mAb; HyTest Ltd.) and a F(ab')2 antibody fragment of a mAb recognizing the epitope at aa 190–196 in the C-terminal region of cTnI (9707-mAb; Medix Biochemica, Kauniainen, Finland), were used as capture antibodies. The F(ab')2 antibody fragment was produced from the parental 9707-mAb by enzymatic digestion with bromelain (ID-Diluent 1; Diamed, Switzerland) as described by Väisänen et al. [16].

71

2.4. Other reagents N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide (EDC) and N-hydroxysulfosuccinimide (sulfo-NHS) were obtained from Fluka (Buchs, Switzerland). Biotin isothiocyanate was synthesized at the Department of Biotechnology, University of Turku. BSA was purchased from Bioreba (Nyon, Switzerland), bovine gamma globulin (BGG; G-5009) from Sigma-Aldrich (St. Louis, MO) and native and denatured mouse IgG from Meridian Life Science (Saco, ME). Casein and heparin were purchased from Calbiochem (La Jolla, CA). Tween 85 was a product of Merck (Darmstadt, Germany). Low-fluorescent, normal capacity streptavidin (SA)‐coated 12-well microtiter strips and Kaivogen buffer and wash solutions were obtained from Kaivogen Oy (Turku, Finland). All other reagents used were of analytical grade.

2.5. Nanoparticle-bioconjugates Highly fluorescent europium(III)-chelate-dyed Fluoro-Max™ polystyrene nanoparticles (107 nm in diameter) were purchased from Seradyn, Inc. (Indianapolis, IN). Fluorescent properties of these nanoparticles have been described previously [7]. An 8I7-mAb was covalently coupled to activated carboxyl groups on the nanoparticle surface using a procedure from Valanne et al. [10] with some minor modifications. The carboxyl groups were activated in 10 mmol/l phosphate buffer (pH 7.0) by applying EDC and sulfo-NHS to final concentrations of 0.75 mmol/l and 10 mmol/l, respectively. The coupling reaction was performed in 900 μl of 10 mmol/l phosphate buffer (pH 8.0) containing 0.33% (w/v) of activated nanoparticles (corresponds to 4.6 × 1012 nanoparticles), 0.5 g/l of 8I7-mAb and 5.8 g/l of NaCl. Finally, the nanoparticle-bioconjugates were stored in 10 mmol/l tris buffer (pH 8.5) supplemented with 0.5 g/l NaN3 and 0.1 g/l Tween 85. Before the first instance of use, the nanoparticle-bioconjugates were mixed thoroughly, sonicated and cautiously centrifuged at 350 g for 5 min to separate noncolloidal aggregates from the monodisperse suspension.

2.6. Capture antibody wells For the preparation of antibody-coated microtiter wells, 19C7-mAb was biotinylated with a 10-fold and 9707-F(ab')2 with a 30-fold molar excess of biotin-isothiocyanate [17] using a procedure described earlier [18]. Thereafter, 100 ng of the biotinylated 19C7-mAb and 66 ng of the biotinylated 9707-F(ab')2 in 50 μl of Kaivogen buffer solution were incubated in SA-coated microtiter wells for 1 h at room temperature without shaking. The wells were washed twice with Kaivogen wash solution and used immediately in the assay.

2.7. Nanoparticle-based immunoassay The nanoparticle-based cTnI-immunoassay (Fig. 1) was performed in freshly prepared capture antibody wells coated with 19C7-mAb and 9707-F(ab')2. The assay buffer (37.5 mmol/l Tris–HCl, pH 7.75, 30 g/l NaCl, 0.4 g/l NaN3, 0.6 g/l BGG, 25 g/l BSA, 50 g/l D-trehalose, 0.8 g/l native mouse IgG, 0.05 g/l denatured mouse IgG, 2 g/l casein, and 37.5 IU/ml heparin) was modified from von Lode et al. [19]. In the basic assay protocol, 3.75 × 10 8 nanoparticle-bioconjugates in 40 μl of assay buffer and 10 μl of standard or sample were applied to the capture antibody wells and incubated for 15 min at +36 °C (650 rpm). Thereafter, the wells were washed thoroughly with Kaivogen wash solution. Time-resolved fluorescence from the surface-bound nanoparticlebioconjugates was measured at 615 nm with Victor™ 1420 Multilabel Counter (Perkin-Elmer Life and Analytical Sciences, Wallac Oy, Turku, Finland).

72

M.-L. Järvenpää et al. / Clinica Chimica Acta 414 (2012) 70–75

Fig. 1. Principle of a one-step, two-site nanoparticle-based cTnI immunoassay. Two biotinylated capture antibodies – a monoclonal antibody and an enzymatically digested F(ab')2 antibody fragment – are immobilized in streptavidin (SA)‐coated microtiter wells. Sample is applied simultaneously with europium(III)-chelate-dyed nanoparticles covalently coupled to several detection antibodies. After 15-min incubation, the wells are washed and europium fluorescence of the bound nanoparticle-bioconjugates is measured from the surface in a time-resolved mode (340 nm, excitation wavelength; 615 nm, emission wavelength). Dashed arrows represent four different binding events that may take place in the assay well. (Picture not to scale).

2.8. Method comparison Fresh plasma specimens from 226 chest pain patients were initially analyzed at the Laboratory of Oulu University Hospital with Innotrac Aio! 2nd generation cTnI assay (Innotrac Diagnostics Oy, Turku, Finland) [20,21]. Following the shipment to the Department of Biotechnology, University of Turku, the samples were measured with the developed nanoparticle-based assay. Deming regression parameters were calculated with GraphPad Prism 4 statistical software (GraphPad Software Inc., La Jolla, CA). 3. Results 3.1. Kinetics Fluorescence signal of the developed cTnI-immunoassay increased linearly until 30 min. Standards with low (0.01 μg/l) and high (1 μg/l) cTnI concentration levels gave comparable kinetics. The kinetic pattern of the endogenous cTnI was determined with a normal plasma pool spiked with a patient sample of high troponin, and was confirmed to be similar to that of the HyTest's standard material. Approximately 35–50% of the maximum signal was reached in 15 min, which was chosen as the incubation time for the current assay. Kinetic curves are represented in the Supplemental data Fig. A1.

samples were 0.007, 0.051, 0.52, and 2.62 μg/l. The within-run imprecisions were 9.8%, 5.1%, 7.7%, and 5.4%, and the total imprecisions were 13.1%, 10.4%, 9.0%, and 8.7%, respectively. 3.3. Linearity and functional detection limit Linearity on dilution was studied with serial dilutions of five patient samples whose initial cTnI concentrations were 0.31, 0.47, 0.62, 2.0 and 9.6 μg/l. The specimens were diluted 3- to 243-fold in a normal plasma pool and measured in six replicates. Linearity was assessed with linear regression analysis by plotting the observed cTnI concentration against the dilution factor (Fig. 3A). The assay response was linear (r = 0.998–0.999) throughout the measured cTnI concentration range (0.003–9.6 μg/l). In order to estimate the limit of quantification (LoQ, defined as the lowest cTnI concentration measured with CV of 10%), the imprecision profiles of the dilutions whose observed cTnI concentrations were below 0.05 μg/l, were studied in more detail (Fig. 3B). The LoQ was 0.012 μg/l.

3.2. Calibration curve and assay imprecision A typical calibration curve is represented in Fig. 2 together with an intra-assay precision profile. The calibration curve was linear up to 10 μg/l (r = 0.999), and no high-dose hook effect was observed even with 1000 μg/l. The CVs of the fluorescence signals between the replicas (n = 4) were 1.6–12.4%. The limit of blank (LoB) and limit of detection (LoD) were determined according to Clinical Laboratory Standards Institute (CLSI) Guideline EP17-A. LoB was 0.0012 μg/l defined as the mean of zero calibrator (n = 20) + 1.645 times the standard deviation (SD). LoD was 0.0020 μg/l defined as the LoB + 1.645 times the SD of a sample containing low concentration of troponin (n = 20). Within-run and total assay imprecisions were assessed according to CLSI Guideline EP5-A. Ternary troponin complex was spiked to normal plasma pool and the samples were assayed in duplicate twice a day for 10 days. The measured mean concentrations of cTnI in the

Fig. 2. Calibration curve (solid symbols) and precision profile (open symbols). The time-resolved fluorescence was measured from four replicate reaction wells. The dotted line indicates the limit of detection (LoD=0.0020 μg/l), defined as the concentration of cTnI required to give a signal equal to the limit of blank (LoB=0.0012 μg/l)+1.645 times the SD of a sample containing low concentration of troponin (n=20).

M.-L. Järvenpää et al. / Clinica Chimica Acta 414 (2012) 70–75

73

Fig. 4. Method comparison. Correlation between the developed nanoparticle-based cTnI assay and Innotrac Aio! 2nd generation cTnI assay (n = 212). Dashed line represents the line of identity.

Fig. 3. (A), Linearity and (B), functional detection limit of the nanoparticle-based cTnI assay. Five clinical plasma samples containing different amounts of cTnI were diluted 3–243-fold with a pool of plasma from apparently healthy donors. A; linearity of dilutions. B; within-run imprecision profiles (closed symbols) for the low end concentrations (open symbols). The thick solid line in B indicates the estimated functional detection limit defined as the lowest concentration of cTnI measured with a CV of 10% (0.012 μg/l).

the specimens whose cTnI concentrations exceeded the corresponding LoQs (0.012 μg/l for the developed assay and 0.040 μg/l for the reference assay [21], respectively) were included in the comparison. In nine specimens, the cTnI concentration was below the LoQs by both assays, and in five specimens below the LoQ of the reference assay but above that of the developed assay. None of the specimens whose cTnI concentration was below the LoQ with the developed assay had detectable concentrations with Innotrac Aio! assay either. Deming regression analysis between the developed and the reference assay yielded a slope (95% confidence intervals) of 1.197 (1.141 to 1.253) and y-intercept of 0.216 (−0.128 to 0.561)μg/l (Sy|x =2.176 μg/l, n= 212, r=0.945). Bland–Altman plot of the relative differences between the methods is shown in Fig. 5. The mean relative difference (95% limits of agreement) was 5.5% (−88.1% to 99.2%).

3.4. Analytical recovery In the recovery study, 17 normal plasma specimens were spiked to 0.5 μg/l cTnI with the troponin complex and measured in triplicate. Plasma backgrounds were determined without the addition of troponin and used to correct for any signal increment caused by endogenous cTnI or nonspecific binding in the specimens. The backgrounds, expressed as cTnI concentrations, were below the LoD of the developed assay (b 0.0020 μg/l) in 11 specimens. The highest cTnI background within the normal plasma specimens was 0.017 μg/l. Analytical recoveries were expressed as the measured concentration (assessed by using the background-subtracted signals) divided by the expected concentration. The cTnI recovery in plasma specimens tested negative for circulating autoantibodies against cardiac troponins (n= 15) varied between 72 and 119% (median 93%). The recoveries in two specimens tested positive for the autoantibodies were 54% and 60%. 3.5. Method comparison Comparison of the results between the developed assay and Innotrac Aio! 2nd generation cTnI assay are shown in Fig. 4. Only

Fig. 5. Bland–Altman analysis of agreement between the developed nanoparticle-based assay and Innotrac Aio! 2nd generation reference assay. The mean difference (solid line) was 5.5% with 95% limits of agreement (dotted lines) of −88.1% to 99.2%.

74

M.-L. Järvenpää et al. / Clinica Chimica Acta 414 (2012) 70–75

4. Discussion The ability to detect minor elevations of cardiac troponins in circulation is important for rapid diagnosis of MI and identification of patients at risk for cardiovascular diseases. Unfortunately, most of the current commercial assays do not meet the guideline recommendations for sensitivity and low-end precision, and therefore, improvements in their analytical performance are still needed. In this report, polystyrene nanoparticles doped with tens of thousands of fluorescent europium(III)-chelates [7] were exploited as labels to construct a one-step, two-site immunoassay for cTnI (Fig. 1). The assay utilized two biotinylated antibodies – a mAb and a F(ab')2 antibody fragment – immobilized onto SA-coated microtiter wells, and a mAb covalently conjugated to the nanoparticles. The assay antibodies were selected based on their specificity for cTnI (no cross-reactivity for cTnT or TnC), and furthermore, their configuration was designed to circumvent the inhibiting effect of the circulating autoantibodies that are mainly bound to the central part of cTnI [22]. Thus, only one capture antibody recognized the epitope at the stable mid-fragment of cTnI (aa 41–49), whereas the other capture antibody and the detection antibody had their recognition sites outside the central part in the C-terminal region (after aa 110). After a 15-min incubation of the sample and nanoparticle-bioconjugates in the reaction wells coated with the capture antibodies, the amount of cTnI was quantified from the washed surface by measuring the fluorescence of the bound nanoparticles in a time-resolved mode. In the developed assay, multiple antibodies were coupled to the surface of a highly fluorescent nanoparticle. This approach increases the binding area of the label surface, and thus, enhances its interactions with the target antigen. However, high density of the detection antibody may also render the assay more vulnerable to the matrix-related analytical interferences as compared to the immunoassays employing a soluble antibody labeled with a few small-sized reporter molecules. Thus, in order to reduce a risk for interference by heterophilic antibodies shown to display the highest specificity for a murine IgG1 antibody subclass [23], the capture antibody representing that particular subclass was employed as an enzymatically digested F(ab')2 fragment of the parental antibody. Replacing the F(ab')2 antibody fragment with a corresponding site-specifically biotinylated recombinant Fab has also been studied during the assay development. Based on our unpublished data, the use of fragments instead of an intact antibody helps to reduce nonspecific interferences from heterophilic antibodies significantly. In the previous studies, site-specific biotinylation of Fab fragments has provided further improvements by enabling better capture orientation and enhanced capacity of the binding surface [24,25]. The LoD of the developed cTnI assay (0.0020 μg/l) was below the detection limits reported for the commercial cTnI assays currently available on the automated platforms [2]. In order to provide an estimate for LoQ (CV of 10% at 0.012 μg/l), the within-run imprecision results from the dilution linearity test were utilized. A more reliable determination of LoQ would require more extensive between-run imprecision data at low cTnI level. However, since the assay is presently not performed with an automatic device, its particular sources of variability (especially timing), would most probably have an adverse effect on the between-run imprecision and therefore it would not reflect the potential performance of our test concept. The within-run imprecision of the assay has been very reassuring throughout the study even at low cTnI concentrations (see Fig. 2 and 20-day precision data). If transferred to an automated platform, a more robust study for the repeatability and LoQ of the assay would be necessary. In the method comparison study, the developed assay showed a good correlation to Innotrac Aio! 2nd generation cTnI-assay. Each plasma specimen, whose cTnI result was above the respective LoD with the Aio! reference assay, had detectable concentrations also with the nanoparticle-based assay. Furthermore, five specimens were found to have cTnI concentration above the respective LoD

with the developed assay although the result was negative with the reference assay. High relative differences between the methods with some specimens may be partly explained by different time frames and conditions of the sample analysis (fresh vs. frozen specimens). During the assay optimization phase, some of the patient samples have undergone several freeze/thaw-cycles that may have affected the measurable cTnI concentrations with the research assay. Additionally in the research assay, the antibody against the epitope at aa residues 190–196 (clone 9707) is an enzymatically fragmented F(ab')2 instead of an intact antibody utilized in the reference assay (clone MF4), which is likely to decrease nonspecific interactions in the samples measured with the research assay. The major difference between the research and reference assay is the labeling reagent (europium-chelate-dyed nanoparticle vs. europium-chelate). Compared to a tracer antibody coupled to small-sized label moieties, a large and rigid nanoparticle suffers more significantly from steric hindrances that may prevent it binding to an analyte. This can be seen in particular with the samples containing circulating autoantibodies against cardiac troponins. Despite the careful selection of the capture and detection antibodies, slightly lower analytical recoveries were obtained with the current assay from the two samples tested positive for troponin autoantibodies. The lower binding efficiency in these samples is likely to be explained by strong sterical effects on the binding of the nanoparticle-bioconjugate to the cTnI subunit partially obscured by autoantibodies. Patient samples were not tested for troponin specific autoantibodies. Troponin assays should not only be very sensitive but also provide the results readily to facilitate the rapid identification of the patients requiring critical care. Thus, the total assay time should be reduced without sacrificing the analytical sensitivity. The use of a highly fluorescent nanoparticle label with high specific activity allowed the developed immunoassay to be performed in a kinetic format. Thus, the signal was measured when approximately half of the steady-state was reached. This may increase the between-run imprecision because even a small variation in incubation time between different assay runs may affect the measured value. Such kinetic errors could be avoided by exact timing in an automated analyzer. Dry chemistry format has been previously demonstrated and suggested to be applicable with europium(III)-nanoparticles [9] providing a good basis for an automated and rapid ultrasensitive cTnI assay. Automation would reduce assay steps and overall turn-around time, and also decrease the betweenassay variation by avoiding kinetic errors. From the present study, it can be concluded that europium(III)chelate-dyed nanoparticles and time-resolved fluorometry can be utilized to construct a rapid and sensitive immunoassay for cTnI. Furthermore, the developed assay principle could challenge the current commercial cTnI assays in analytical performance if optimized on an automated platform. Previously, a new generation of research assays [4,6] has emerged that can reliably detect at least 10-fold lower cTnI concentrations than the current generation assays. However, as those novel highly sensitive methods are rather laborious and have increased turn-around times, there will be a need for technological developments before they can be utilized in everyday clinical practice. Instead, the assay principle demonstrated in this report represents a sophisticated method that allows convenient detection of low-end cTnI concentrations. Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.cca.2012.08.027.

References [1] Giannitsis E, Kurz K, Hallermayer K, Jarausch J, Jaffe AS, Katus HA. Analytical validation of a high-sensitivity cardiac troponin T assay. Clin Chem 2010;56:254-61. [2] Tate JR. Troponin revisited 2008: assay performance. Clin Chem Lab Med 2008;46: 1489-500. [3] Jaffe AS, Ordonez-Llanos J. High sensitivity troponin in chest pain and acute coronary syndromes. A step forward? Rev Esp Cardiol 2010;63:763-9.

M.-L. Järvenpää et al. / Clinica Chimica Acta 414 (2012) 70–75 [4] Wu AH, Fukushima N, Puskas R, Todd J, Goix P. Development and preliminary clinical validation of a high sensitivity assay for cardiac troponin using a capillary flow (single molecule) fluorescence detector. Clin Chem 2006;52:2157-9. [5] Todd J, Freese B, Lu A, et al. Ultrasensitive flow-based immunoassays using single-molecule counting. Clin Chem 2007;53:1990-5. [6] Wilson SR, Sabatine MS, Braunwald E, Sloan S, Murphy SA, Morrow DA. Detection of myocardial injury in patients with unstable angina using a novel nanoparticle cardiac troponin I assay: observations from the PROTECT-TIMI 30 Trial. Am Heart J 2009;158:386-91. [7] Härmä H, Soukka T, Lövgren T. Europium nanoparticles and time-resolved fluorescence for ultrasensitive detection of prostate-specific antigen. Clin Chem 2001;47: 561-8. [8] Soukka T, Antonen K, Härmä H, Pelkkikangas A, Huhtinen P, Lövgren T. Highly sensitive immunoassay of free prostate-specific antigen in serum using europium(III) nanoparticle label technology. Clin Chim Acta 2003;328:45-58. [9] Pelkkikangas A, Jaakohuhta S, Lövgren T, Härmä H. Simple, rapid, and sensitive thyroid-stimulating hormone immunoassay using europium(III) nanoparticle label. Anal Chim Acta 2004;517:169-76. [10] Valanne A, Huopalahti S, Soukka T, Vainionpää R, Lövgren T, Härmä H. A sensitive adenovirus immunoassay as a model for using nanoparticle label technology in virus diagnostics. J Clin Virol 2005;33:217-23. [11] Valanne A, Huopalahti S, Vainionpää R, Lövgren T, Härmä H. Rapid and sensitive HBsAg immunoassay based on fluorescent nanoparticle labels and timeresolved detection. J Virol Methods 2005;129:83-90. [12] Hagan AK, Zuchner T. Lanthanide-based time-resolved luminescence immunoassays. Anal Bioanal Chem 2011;400:2847-64. [13] Soukka T, Härmä H, Paukkunen J, Lövgren T. Utilization of kinetically enhanced monovalent binding affinity by immunoassays based on multivalent nanoparticleantibody bioconjugates. Anal Chem 2001;73:2254-60. [14] Eriksson S, Halenius H, Pulkki K, Hellman J, Pettersson K. Negative interference in cardiac troponin I immunoassays by circulating troponin autoantibodies. Clin Chem 2005;51:839-47. [15] Pettersson K, Eriksson S, Wittfooth S, Engström E, Nieminen M, Sinisalo J. Autoantibodies to cardiac troponin associate with higher initial concentrations and

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23] [24] [25]

75

longer release of troponin I in acute coronary syndrome patients. Clin Chem 2009;55:938-45. Väisänen V, Peltola MT, Lilja H, Nurmi M, Pettersson K. Intact free prostatespecific antigen and free and total human glandular kallikrein 2. Elimination of assay interference by enzymatic digestion of antibodies to F(ab')2 fragments. Anal Chem 2006;78:7809-15. Mukkala VM, Hänninen E, Hemmilä I. Synthesis of two biotin isothiocyanate derivatives. Proceedings of the 8th European Symposium on Organic Chemistry; 1993. p. 422. Eriksson S, Junikka M, Laitinen P, Majamaa-Voltti K, Alfthan H, Pettersson K. Negative interference in cardiac troponin I immunoassays from a frequently occurring serum and plasma component. Clin Chem 2003;49:1095-104. von Lode P, Rosenberg J, Pettersson K, Takalo H. A europium chelate for quantitative point-of-care immunoassays using direct surface measurement. Anal Chem 2003;75:3193-201. Pagani F, Stefini F, Panteghini M. Innotrac aio! Second-generation cardiac troponin I assay: imprecision profile and other key characteristics for clinical use. Clin Chem 2004;50:1271-2. Hedberg P, Valkama J, Suvanto E. Evaluation of innotrac aio! Second-generation cardiac troponin I assay: the main characteristics for routine clinical use. J Autom Methods Manag Chem 2006;2006:39325. Eriksson S, Junikka M, Pettersson K. An interfering component in cardiac troponin I immunoassays—its nature and inhibiting effect on the binding of antibodies against different epitopes. Clin Biochem 2004;37:472-80. Bjerner J, Olsen KH, Bormer OP, Nustad K. Human heterophilic antibodies display specificity for murine IgG subclasses. Clin Biochem 2005;38:465-72. Ylikotila J, Hellström JL, Eriksson S, et al. Utilization of recombinant Fab fragments in a cTnI immunoassay conducted in spot wells. Clin Biochem 2006;39:843-50. Brockmann E, Vehniäinen M, Pettersson K. Use of high-capacity surface with oriented recombinant antibody fragments in a 5-min immunoassay for thyroid-stimulating hormone. Anal Biochem 2010;396:242-9.