Simultaneous Detection of High-SensitivityCardiac Troponin I and Myoglobin by

Modified Sandwich Lateral Flow Immunoassay:Proof of Principle

Jimin Zhu,1,2 Nengli Zou,2 Danian Zhu,1 Jin Wang,1 Qinghui Jin,2 Jianlong Zhao,2 and Hongju Mao2*

BACKGROUND: Although numerous lateral flow immu-noassays (LFIAs) have been developed and widelyused, inadequate analytical sensitivity and the lack ofmultiple protein detection applications have limitedtheir clinical utility. We developed a new LFIA devicefor the simultaneous detection of high-sensitivity car-diac troponin I (hs-cTnI) and myoglobin (Myo).

METHODS: We used a gold nanoparticle (AuNP) doublylabeled complex, in which biotinylated single-strandedDNA was used as a linkage to integrate 2 AuNPs andstreptavidin-labeled AuNP, as an amplifier to magnifyextremely low signals.

RESULTS: The detection limit of 1 ng/L achieved for hs-cTnI was 1000 times lower than that obtained in a con-ventional LFIA. The detection limit for simultaneouslymeasured Myo was 1 �g/L. The linear measurementranges for hs-cTnI and Myo were 1–10 000 ng/L and1–10 000 �g/L, respectively. We observed concordantresults between the LFIA and clinical assays in serafrom 12 patients with acute myocardial infarction (hs-cTnI r � 0.96; Myo r � 0.98). Assay imprecision was�11% for both hs-TnI and myo.

CONCLUSIONS: The described proof-of-principle LFIAmethod could be used as a point-of-care device inmultiple protein quantification and semiquantitativeanalysis.© 2011 American Association for Clinical Chemistry

Acute myocardial infarction (AMI)3 is a major cause ofdeath and disability worldwide. The early exclusion or

diagnosis of AMI allows efficient and cost-effective tri-age as well as the successful management of patientswith AMI. Serum high-sensitivity cardiac troponin I(hs-cTnI) and myoglobin (Myo) have been shown tobe biomarkers for early detection of AMI (1–3 ). Thesimultaneous detection of these 2 biomarkers may helpfacilitate a diagnosis of AMI within the first 90 minafter presentation. However, the rapid detection ofprotein markers, such as hs-cTnI, remains a challenge.

The simultaneous detection of multiple proteinspresent in the circulation in widely different concentra-tions is desirable (4–6). Although Western blotting (7),protein microarray (8), ELISA (9), flow cytometry(10, 11), and quantitative PCR (12, 13) provide relativelyspecific and sensitive detection, these techniques requiresophisticated laboratory facilities (14), highly trainedmedical technologists (15, 16), and time-consuming pro-cedures (17, 18). These problems can hamper the adop-tion of proteins analysis in point-of-care (POC) settings.

Lateral flow immunoassay (LFIA), known for its ap-plication to the commercially available pregnancy test(19), has the advantage of circumventing these difficultiesand inconveniences. LFIA not only provides a means forperforming the assay without the extensive handling ofspecimens, but also accelerates the analytical process to 15min in most cases. Thus, numerous LFIAs have been de-veloped and used in clinical medicine (20–22). Two maindrawbacks that limit the practical application of LFIAs inPOC settings: (a) limited applications of LFIA to multipleproteins at widely different concentrations and (b) themicrogram-per-liter detection limit range (18). For ex-ample, the nanogram-per-liter hs-cTnI concentration(23) necessary for the diagnosis of AMI cannot be de-tected in a conventional LFIA.

1 Department of Physiology and Pathophysiology, Shanghai Medical College ofFudan University, Shanghai, China; 2 State Key Laboratory of Transducer Tech-nology, Shanghai Institute of Microsystem and Information Technology, ChineseAcademy of Science, Shanghai, China.

* Address correspondence to this author at: State Key Laboratory of TransducerTechnology, Shanghai Institute of Microsystem and Information Technology,Chinese Academy of Science, 865 Changning Rd., Shanghai 200050, P.R. China.Fax �86-21-62511070-8714; e-mail hjmao@mail.sim.ac.cn.

Received July 15, 2011; accepted October 3, 2011.Previously published online at DOI: 10.1373/clinchem.2011.1716943 Nonstandard abbreviations: AMI, acute myocardial infarction; hs-cTnI, high-

sensitivity cardiac troponin I; Myo, myoglobin; POC, point-of-care; LFIA, lateralflow immunoassay; AuNP, gold nanoparticle; ssDNA, single-stranded DNA;mAb, monoclonal antibody; CRP, C-reactive protein; BNP, brain natriureticpeptide; PEG, polyethylene glycol; UV-Vis, ultraviolet/visible.

Clinical Chemistry 57:121732–1738 (2011)

Point-of-Care Testing

1732

To overcome these 2 disadvantages of the conven-tional LFIA, we developed and evaluated a modifiedsandwich LFIA designed for simultaneous detection of2 proteins present in circulation at widely differentconcentrations. Our approach was to combine the de-tection of hs-cTnI and Myo on a single LFIA device.

Materials and Methods

ASSAY PRINCIPLE

Our new format used a gold nanoparticle (AuNP) dou-bly labeled complex, in which biotinylated single-stranded DNA (ssDNA) was used as a linkage to inte-grate 2 AuNPs, and streptavidin-labeled AuNP (41nm) was used as an amplifier of small-sized AuNP (13nm) to magnify extremely low signals. The slow migra-tion speed of a larger-sized AuNP (41 nm) with at-tached streptavidin ensured its binding, after the initialsandwich of hs-cTnI antigen and antibody was fixed inplace (Fig. 1), to the faster-moving AuNP (13 nm) thatwas coated with anti– hs-cTnI monoclonal antibody(mAb) and conjugated with biotinylated ssDNA.

MATERIALS

ssDNA 5� (SH)-TTTTT TTTTT GGCTT TCAGTTATAT GGATG ATGTG GTATT TTTTT TTT(Biotin)-3� was synthesized by TaKaRa. We purchasedanti– hs-cTnI mAb and standard sample of hs-cTnI

from HyTest; anti-Myo mAb and standard samples ofMyo and C-reactive protein (CRP) from Fitzgerald;standard sample of brain natriuretic peptide (BNP)from Tocris; BSA, polyvinylpyrrolidone, and goatanti–mouse IgG antibody from Sigma-Aldrich;streptavidin from New England Biolabs; hydrogen tet-rachloroaurate trihydrate (HAuCl4•3H2O) from Ac-ros; nitrocellulose membrane (HiFlow Plus HFB135)from Millipore; glass fiber (SB06), polyester fiber(VL68), and polyvinyl chloride from Gold-bio; andpolyethylene glycol (PEG) (Mr � 8000) and sucrosefrom Dingguo Biotechnology.

All other chemicals were of analytical-reagentgrade and used without further purification. Deionizedand autoclaved water (resistance �18 M�•cm) wasused throughout the experiments.

Storage buffer solution 1 consisted of 50 mmol/Lphosphate sodium buffer (pH 7.8) containing 5%polyvinylpyrrolidone (wt/vol), 1.25% sucrose (wt/vol),0.05% PEG8000 (wt/vol), 0.2% BSA (wt/vol), and0.05% Tween-20 (vol/vol). Storage buffer solution 2consisted of 10 mmol/L sodium borate buffer (pH 7.8)containing 0.05% PEG8000 (wt/vol) and 0.05%Tween-20 (vol/vol).

CHEMICAL SYNTHESIS OF AuNP

We prepared AuNP (13 nm in diameter) by trisodiumcitrate reduction of HAuCl4•3H2O, a method pio-

Fig. 1. Schematic diagram of the modified sandwich lateral flow immunoassay before (A) and after (B) assay.

Simultaneous Detection of 2 Proteins by LFIA

Clinical Chemistry 57:12 (2011) 1733

neered by Frens (24 ) and Turkevich et al. (25 ). Briefly,all glassware was cleaned with aqua regia (3 parts HCl,1 part HNO3), thoroughly rinsed with distilled water,and dried in an air drier before use. Aqueous solutionof 0.01% (wt/vol) chloroauric acid and 1% (wt/vol)sodium citrate were prepared. Using an oil-bath ther-mostat in a 2-neck round-bottomed flask with a mag-netic stirrer, 0.01% chloroauric acid solution (100 mL)was heated to boiling, followed by an addition of 3 mLof the 1% sodium citrate solution to the flask, withconstant stirring at a maximum speed to synthesize 13nm AuNP. The reaction time lasted 10 min, duringwhich the color of the solution changed to wine red.The AuNP solution was cooled to room temperaturebefore storage at 4 °C for future use.

AuNP (41 nm in diameter) solution was also pre-pared by using the same method except that only 1 mLof the 1% sodium citrate solution was added to the flaskfor the reaction with the 0.01% chloroauric acid solu-tion (100 mL).

The successful formation of 13-nm and 41-nmAuNPs was qualitatively characterized by transmissionelectron microscopy and ultraviolet/visible (UV-Vis)spectroscopy (Fig. 2).

PREPARATION OF AuNP CONJUGATES

We prepared AuNP labeled with anti– hs-cTnI mAbaccording to our previously reported procedure (26 ),but with a few modifications. We adjusted the pH valueof the AuNPs (13 nm) to 8.5 with 0.2 mol/L K2CO3

dropwise and added 9 �L anti– hs-cTnI mAb (1 g/L) to

the 1 mL AuNPs. After a 30-min incubation at roomtemperature, the AuNPs modified with anti– hs-cTnImAb reacted with ssDNA at a final concentration of 1�mol/L for 16 h. We added PEG-8000 to a final con-centration of 0.5% to stabilize the AuNPs. We broughtthe solution to a concentration of 0.1 mol/L NaCl byadding phosphate buffer solution (pH 7.4, 10 mmol/L). The excess antibody and ssDNA molecules wereremoved by centrifugation at 8603g for 50 min at 4 °C,and 200 �L of storage buffer solution 1 was added tothe AuNP conjugate to be resuspended. We repeatedthe centrifugation and suspension process twice. Theprecipitate was suspended in 200 �L storage buffer so-lution 1 and stored at 4 °C for future use.

We prepared anti-Myo mAb–labeled AuNP fol-lowing the reported method (26 ), with a few modifica-tions. We adjusted the AuNP solution to pH 8.5 with0.2 mol/L K2CO3 and added 8 �L anti-Myo mAb (1g/L) to the 1 mL AuNP solution dropwise. The mixturewas incubated for 30 min at room temperature, fol-lowed by addition of 100 �L of 10% BSA solution toblock the residual surface of the AuNP particles. Theobtained solution was centrifuged at 15 294g for 25min at 4 °C, after which the supernatant was discardedand 1 mL of 1% BSA solution was added to the AuNPconjugate to be resuspended. We repeated the centrif-ugation and suspension process twice, with the precip-itate resuspended in 200 �L storage buffer solution 1and stored at 4 °C for future use.

We prepared streptavidin-labeled AuNP followingthe reported method (26 ), with a few modifications.

Fig. 2. Photograph of 13-nm (A) and 41-nm (B) AuNPs shown at the transmission electron microscopy image (scalebars, 50 nm), together with UV-Vis spectra of 13-nm AuNP (black) and 41-nm AuNP (red) (C).

1734 Clinical Chemistry 57:12 (2011)

We used AuNP (41 nm) for the conjugation of strepta-vidin. We adjusted the AuNP solution to pH 9.3 with0.2 mol/L K2CO3 and added 10 �L streptavidin (1 g/L)to the 1 mL AuNP solution dropwise. The mixture wasincubated for 30 min at room temperature, followed byaddition of 100 �L of 1% PEG-8000 solution to blockthe residual surface of the AuNP particles. The ob-tained solution was centrifuged at 15 294g for 25 min at4 °C, after which the supernatant was discarded and 1mL storage buffer solution 2 was added to the AuNPconjugate to be resuspended. We repeated the centrif-ugation and suspension process twice, with the precip-itate resuspended in 200 �L storage buffer solution 2and stored at 4 °C for future use.

PRETREATMENT OF SAMPLE PAD, CONJUGATE PAD, AND

NITROCELLULOSE MEMBRANE

The sample pad, made from glass fiber, was saturatedwith a buffer (pH 7.4) containing 20 mmol/L sodiumborate, 1% (wt/vol) sucrose, 1% (wt/vol) BSA, 0.5%(vol/vol) Tween-20, and 0.05% (wt/vol) NaN3 in waterfor 1 h before drying in an air drier at 50 °C for 2 h andstored in a dry state for future use.

The conjugate pad, made from polyester fiber, wasimmersed in a solution containing 5% (wt/vol) sucroseand 0.05% (wt/vol) NaN3 in water for 1 h, and thendried at 50 °C for 2 h. Anti– hs-cTnI mAb–AuNP com-plex was then mixed together with anti-Myo mAb–AuNP complex in a 1:1 ratio. When 3 �L/strip ofAuNP-labeled antibody probe complex was placedonto the polyester fiber to be used as the first conjugatepad, 2 �L/strip of AuNP-labeled streptavidin probewas applied to another polyester fiber as the secondconjugate pad. Both pads were dried for 30 min at37 °C and stored in a dry state for future use.

The nitrocellulose membrane was treated with abuffer (pH 7.4) containing 10 mmol/L PBS solution,3% (wt/vol) BSA, and 0.5% Tween-20 for 2 h, dried for2 h at 50 °C, and stored in a dry state for future use.

ASSEMBLY AND ANALYSIS OF THE LFIA

The strip comprised 5 main elements: a sample pad, 2conjugate pads, a nitrocellulose membrane, and an ab-sorbent pad. The strip was positioned in such a waythat the ends of the elements overlapped, ensuring acontinuous flow by capillary action of the developingsolution from the sample pad to the absorbent pad(Fig. 1A). Afterward, 1 g/L anti– hs-cTnI capture anti-body (test line 1, a), anti-Myo capture antibody (testline 2, b), and goat anti–mouse IgG antibody (controlline, c) were each immobilized on the nitrocellulosemembrane by a dispenser system. The whole devicewas subsequently dried overnight at 37 °C and storedin a dry state until use.

We measured hs-cTnI and Myo in sera of 12 pa-tients with AMI by use of the LFIA strips and comparedthe results to values obtained by a Hitachi Modular7600 Chemistry Analyzer.

To obtain the quantitative results, after 10 min ofcolor development, the optical images of the dipstickswere captured using a digital scanner (Epson 1640SU)and imported into Quantity One software to be con-verted automatically to grayscale. The intensities ofeach signal were then quantified using the Peak Densityoption in Band Attributes.

Results and Discussion

ASSAY PROCEDURE

In this study, AuNPs of 2 different diameters in con-nection with biotin–streptavidin linkage were used to-gether for the development of an LFIA device to detect2 model protein analytes, hs-cTnI and Myo, through amodified sandwich strategy (Fig. 1). In the second con-jugate pad, part of the AuNPs (13 nm) was coated withanti– hs-cTnI mAb and then with biotinylated ssDNA,whereas another part of the AuNPs (13 nm) was treatedwith anti-Myo mAb only. In addition, the streptavidin-coated AuNPs (41 nm) were immobilized on the firstconjugate pad as an intensifier, designed to bind spe-cifically with the AuNPs (13 nm) labeled with anti– hs-cTnI mAb in the second conjugate pad to enlarge ex-tremely low signals (down to 1 ng/L).

First, we added the solution containing hs-cTnI andMyo onto the sample pad. Next, the solution migrated viacapillary action and rehydrated the AuNP–anti–hs-cTnImAb and AuNP–anti-Myo mAb conjugates, and hencethe binding between hs-cTnI and AuNP–anti–hs-cTnImAb and between Myo and AuNP–anti-Myo mAb oc-curred via interaction between the antigen and corre-sponding detection antibody. When the complexesreached test lines 1 and 2, they were captured, respectively,with the capture anti–hs-cTnI mAb and capture anti-Myo mAb immobilized on the distinct test line via theinteraction between the antigen and corresponding cap-ture mAb. Meanwhile, the streptavidin-coated AuNPcontinued to migrate along the strip at a relatively slowspeed because of its larger size. When streptavidin-coatedAuNP reached test line 1, the binding between biotinyl-ated ssDNA–coated AuNP (13 nm) and streptavidin-coated AuNP (41 nm) occurred via the interaction be-tween biotin and streptavidin. Thus, 2 characteristic redbands were observed for the accumulation of AuNPs inthe 2 test lines (Fig. 1B, a and b). The capillary actioncaused the liquid sample to migrate further. Once the so-lution passed through the control line, the excess conju-gates were captured on the control line via the conjuga-tion between the goat antimouse IgG and either of the 2superfluous AuNP-coated mAbs, producing a third red

Simultaneous Detection of 2 Proteins by LFIA

Clinical Chemistry 57:12 (2011) 1735

band in the control line, which confirmed the exact per-formance of the strip (Fig. 1B, c). In the absence of Myoand hs-cTnI, only the red band was observed in the con-trol line rather than in the 2 test lines (Fig. 1B, a and b).

OPTIMIZATION OF THE STREPTAVIDIN/ANTIBODY-AuNP

CONJUGATES

The gold aggregation test, which was designed to detectsalt-induced AuNP aggregation and determine the op-timal streptavidin concentration and pH value for thecoupling of streptavidin with AuNP, was carried out bymixing 50 �L AuNP with a series of volumes of K2CO3

and then with 10 �L streptavidin solutions at differentconcentrations. After incubation for 15 min, 5 �L of10% NaCl solution was added. Subsequently, thecolor change and UV-Vis absorption spectrum of theAuNP suspension were analyzed with a UV-Visspectrophotometer.

The AuNPs showed no aggregation in the streptavi-din at a concentration of�8 mg/L (Fig. 3). To stabilize theAuNP, we chose approximately 120% of the minimumunaggregated amount of streptavidin as its working con-centration. Moreover, the AuNPs indicated no aggrega-tion in the streptavidin with pH 9.0–9.3 (data not shown).Therefore, in our modified sandwich LFIA test, the finaloptimal streptavidin concentration was chosen to be 10mg/L, which was higher than the critical concentration ofAuNP aggregation. Meanwhile, the pH value for the con-jugation of streptavidin to AuNP was chosen to be 9.0.Anti-Myo mAb–labeled AuNP and anti–hs-cTnI mAb–labeled AuNP were also tested using the same method tovalidate the optimal concentration and pH value (datanot shown). In our approach, the final optimal anti-MyomAb concentration for the conjugation of anti-Myo mAbto AuNP (pH 8.5) was chosen to be 8 mg/L, and the final

optimal anti–hs-cTnI mAb concentration for the conju-gation of anti–hs-cTnI mAb to AuNP (pH 8.5) was cho-sen to be 9 mg/L.

VISUAL ASSESSMENT OF ANALYTICAL SENSITIVITY OF LFIA

It is essential for each analyte from the solution to ac-curately conjugate only with the specific capture anti-body, which is immobilized on the corresponding testline of the strip. Cross-reactivity, if present, would gen-erate nonspecific signals at various test lines, leading tofalse-positive results. To investigate whether there wasany cross-reactivity of the immobilized capture anti-body, 3 standard solutions, 2 containing each of thetested analytes and the third containing combined CRPand BNP solutions, were subjected to the testing. Theresults demonstrated the specificity of the assay (Fig.4A), suggesting that hs-cTnI (Fig. 4A-1) and Myo (Fig.4A-2) were captured only by the corresponding testline and not the other, and that 50 �g/L CRP and 50�g/L BNP (Fig. 4A-3) produced no observable bindingin either test line. Thus, it is reasonable to expect thatthere is no detectable cross-reactivity at the positionthat corresponds to the missing analytes.

To investigate whether the LFIA could provide si-multaneous and semiquantitative detection of Myoand hs-cTnI, the assay was performed using the stan-dard samples with a series of known concentrations ofhs-cTnI and Myo. In these experiments, 1 of the ana-lyte concentrations was kept constant and the otheranalyte concentration was adjusted to validate the dy-namic range for each analyte when measured in ourmodified sandwich format.

The hs-cTnI assay performance was carried outwith the Myo concentration held constant at 10 �g/L(Fig. 4B). Consequently, a distinct red line was ob-served in test line 1, and the color intensity of the hs-cTnI line increased gradually, corresponding to in-creasing hs-cTnI concentrations of 1, 10, 100, 1000,and 10 000 ng/L hs-cTnI. In this study, the detection ofhs-cTnI in the modified sandwich LFIA was about1000-fold higher compared with that of the conven-tional LFIA format, which was mainly attributed to theapplication of biotin-streptavidin linkage and the ag-gregation of 2 different diameters of AuNPs in the testline. The conventional format detecting only hs-cTnIwithout any signal amplification seems to show lowerdetectability. The dose–response curve spanned ordersof magnitude beyond the suggested clinically relevantranges (Fig. 4D). The error bars (Fig. 4D) representSDs, calculated from triplicate experiments performedon different LFIAs. In comparison with the conven-tional format, our new approach produced no differ-ences in Myo detectability, in which Myo producedgradual concentrations of 1, 10, 100, 1000, and 10 000

Fig. 3. Photographs of corresponding UV spectra ofAuNP conjugate solutions to find optimal streptavidinconcentration.

1736 Clinical Chemistry 57:12 (2011)

�g/L, each maintaining a constant hs-cTnI concentra-tion of 10 ng/L (Fig. 4E).

To document whether the LFIA can work using se-rum specimens, hs-TnI and Myo were assessed in serumsamples from 12 patients with AMI. In addition, the 2analytes were measured using a standard laboratory in-strument (Hitachi Modular 7600 chemistry analyzer).The serum samples had hs-cTnI and Myo concentrationsin the range of 9.2–13 900 ng/L and 5.9–2771 �g/L, re-spectively. The results showed that the modified LFIA hadconcordant trends reflecting the concentrations of hs-cTnI (r � 0.96) and Myo (r � 0.98). Finally, we tested 2different serum samples with high (1038 ng/L) and low(16 ng/L) concentrations of hs-cTnI 10 times to evaluateimprecision. The concentrations of Myo in these 2 sam-ples were 25.6 �g/L and 132 �g/L. The imprecision was�11% for the 2 analytes at both concentrations.

In summary, in this proof of principle, our modifiedsandwich LFIA method may be an attractive approach forrapid POC detection of multiple protein analytes. In ad-dition, our approach can provide a higher detectability forthe analytes, at extremely low concentrations, that are dif-ficult to detect via the conventional LFIA.

Author Contributions: All authors confirmed they have contributed tothe intellectual content of this paper and have met the following 3 re-quirements: (a) significant contributions to the conception and design,acquisition of data, or analysis and interpretation of data; (b) draftingor revising the article for intellectual content; and (c) final approval ofthe published article.

Authors’ Disclosures or Potential Conflicts of Interest: Upon man-uscript submission, all authors completed the Disclosures of PotentialConflict of Interest form. Potential conflicts of interest:

Employment or Leadership: None declared.Consultant or Advisory Role: J. Wang, Fudan University.Stock Ownership: None declared.Honoraria: None declared.Research Funding: The National Natural Science Foundation(31000791) and the Science and Technology Commission of Shang-hai Municipality (0952nm05700 and 1052nm061000).Expert Testimony: None declared.

Role of Sponsor: The funding organizations played no direct role indesign of study, choice of enrolled patients, review and interpretationof data, and preparation and approval of manuscript.

Acknowledgments: We are grateful to Wade W. Han, MD, for acritical reading of the manuscript and helpful discussions.

Fig. 4. (A), Specificity of the capture lines of the LFIA.

(B), LFIA assessing the detection of varying hs-cTnI concentrations with constant Myo. (C), LFIA assessing the detection ofvarying Myo concentrations with constant hs-cTnI. (D), Dose–response curves of varying hs-cTnI concentrations with constantMyo. (E), Dose–response curves of varying Myo concentrations with constant hs-cTnI. AU, arbitrary units.

Simultaneous Detection of 2 Proteins by LFIA

Clinical Chemistry 57:12 (2011) 1737

References

1. McCord J, Nowak RM, McCullough PA, Fore-back C, Borzak S, Tokarski G, et al. Ninety-minute exclusion of acute myocardial infarctionby use of quantitative point-of-care testing ofmyoglobin and troponin I. Circulation 2001;104:1483– 8.

2. Plebani M, Zaninotto M. Diagnostic strategies inmyocardial infarction using myoglobin measure-ment. Eur Heart J 1998;19:N12–N5.

3. Plebani M, Zaninotto M. Diagnostic strategiesusing myoglobin measurement in myocardial in-farction. Clin Chim Acta 1998;272:69–77.

4. Abbott A. A post-genomic challenge: learning toread patterns of protein synthesis. Nature 1999;402:715–20.

5. Humphery-Smith I, Blackstock W. Proteomeanalysis: genomics via the output rather than theinput code. J Protein Chem 1997;16:537–44.

6. Drabovich AP, Okhonin V, Berezovski M, KrylovSN. Smart aptamers facilitate multi-probe affinityanalysis of proteins with ultra-wide dynamicrange of measured concentrations. J Am ChemSoc 2007;129:7260–1.

7. Westermeier R, Marouga R. Protein detectionmethods in proteomics research. Bioscience Rep2005;25:19–32.

8. Rusling JF, Kumar CV, Gutkind JS, Patel V. Mea-surement of biomarker proteins for point-of-careearly detection and monitoring of cancer. Analyst2010;135:2496–511.

9. Imafuku Y, Omenn GS, Hanash S. Proteomicsapproaches to identify tumor antigen directedautoantibodies as cancer biomarkers. Dis Markers2004;20:149–53.

10. Schleis SE, LeBlanc AK, Neilsen NR, LeBlanc CJ.Flow cytometric evaluation of multidrug resis-tance proteins on grossly normal canine nodal

lymphocyte membranes. Am J Vet Res 2008;69:1310–5.

11. Baran J, Kowalczyk D, Ozog M, Zembala M.Three-color flow cytometry detection of intracel-lular cytokines in peripheral blood mononuclearcells: comparative analysis of phorbol myristateacetate-ionomycin and phytohemagglutinin stim-ulation. Clin Diagn Lab Immunol 2001;8:303–13.

12. Swartzman E, Shannon M, Lieu P, Chen SM,Mooney C, Wei E, et al. Expanding applications ofprotein analysis using proximity ligation andqPCR. Methods 2010;50:S23–S6.

13. Reuter T, Gilroyed BH, Alexander TW, Mitchell G,Balachandran A, Czub S, McAllister TA. Prionprotein detection via direct immuno-quantitativereal-time PCR. J Microbiol Meth 2009;78:307–11.

14. Hilton S. Near patient testing in general-practice:a review. Br J Gen Pract 1990;40:32–6.

15. Patti G, Nusca A, Mangiacapra F, Gatto L,D’Ambrosio A, Di Sciascio G. Point-of-care mea-surement of clopidogrel responsiveness predictsclinical outcome in patients undergoing percuta-neous coronary intervention results of theARMYDA-PRO (Antiplatelet Therapy for Reduc-tion of Myocardial Damage during Angioplasty–Platelet Reactivity Predicts Outcome) study. J AmColl Cardiol 2008;52:1128–33.

16. Yager P, Domingo GJ, Gerdes J. Point-of-carediagnostics for global health. Ann Rev BiomedEng 2008;10:107–44.

17. Chan CP, Cheung YC, Renneberg R, Seydack M.New trends in immunoassays. Adv Biochem EngBiotechnol 2008;109:123–54.

18. Gordon J, Michel G. Analytical sensitivity limitsfor lateral flow immunoassays. Clin Chem 2008;54:1250–1.

19. Osikowicz G, Beggs M, Brookhart P, Caplan D,

Ching SF, Eck P, et al. One-step chromatographicimmunoassay for qualitative determination ofchoriogonadotropin in urine. Clin Chem 1990;36:1586.

20. Agoritsas K, Mack K, Bonsu BK, Goodman D,Salamon D, Marcon MJ. Evaluation of the QuidelQuickVue test for detection of influenza A and Bviruses in the pediatric emergency medicine set-ting by use of three specimen collection methods.J Clin Microbiol 2006;44:2638–41.

21. Harma H, Soukka T, Lovgren T. Europium nano-particles and time-resolved fluorescence for ultra-sensitive detection of prostate specific antigen.Clin Chem 2001;47:561–8.

22. Cazacu AC, Demmler GJ, Neuman MA, ForbesBA, Chung S, Greer J, et al. Comparison of a newlateral-flow chromatographic membrane immu-noassay to viral culture for rapid detection anddifferentiation of influenza A and B viruses inrespiratory specimens. J Clin Microbiol 2004;42:3661–4.

23. Eggers KM, Jaffe AS, Lind L, Venge P, Lindahl B.Value of cardiac troponin I cutoff concentrationsbelow the 99th percentile for clinical decision-making. Clin Chem 2009;55:85–92.

24. Frens G. Controlled nucleation for regulation ofparticle-size in monodisperse gold suspensions.Nat Phys Sci 1973;241:20–2.

25. Turkevich J, Stevenson PC, Hillier J. A study of thenucleation and growth processes in the synthesisof colloidal gold. Discuss Faraday Soc 1951:55–75.

26. Liu MY, Jia CP, Huang YY, Lou XH, Yao SH, JinQH, et al. Highly sensitive protein detection usingenzyme-labeled gold nanoparticle probes. Ana-lyst 2010;135:327–31.

1738 Clinical Chemistry 57:12 (2011)