Homogeneous Assay for Whole Blood Folate Using Photon

Subscriber access provided by ERCIYES UNIV

Article

Homogeneous assay for whole blood folate using photon upconversion Riikka Arppe, Leena Mattsson, Krista Korpi, Sami Blom, Qi Wang, Terhi Talvikki Riuttamäki, and Tero Soukka Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/ac503691m • Publication Date (Web): 30 Dec 2014 Downloaded from http://pubs.acs.org on January 5, 2015

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 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

Analytical Chemistry

Homogeneous assay for whole blood folate using photon upconversion Riikka Arppe*, Leena Mattsson†, Krista Korpi††, Sami Blom†††, Qi Wang, Terhi Riuttamäki†† and Tero Soukka Department of Biochemistry / Biotechnology, University of Turku, Tykistökatu 6 A 6th floor, 20520 Turku, Finland ABSTRACT: Red blood cell folate is measured for folate deficiency diagnosis as it reflects the long-term folate level in tissues whereas serum folate only represents the dietary intake. Direct homogeneous assay from whole blood would be ideal but conventional fluorescence techniques in blood suffer from high background and strong absorption of light at ultraviolet and visible wavelengths. In this study a new photon upconversion-based homogeneous assay for whole blood folate is introduced based on resonance energy transfer from upconverting nanophosphor donor coated with folate binding protein to a near-infrared fluorescent acceptor dye conjugated to folate analog. The sensitized acceptor emission is measured at 740 nm upon 980 nm excitation. Thus, optically transparent wavelengths are utilized for both donor excitation and sensitized acceptor emission to minimize the sample absorption, and antiStokes detection completely eliminates the Stokes’-shifted autofluorescence. The IC50-value of the assay was 6.0 nM and the LOD was 1 nM. The measurable concentration range was two orders of magnitude between 1.0–100 nM, corresponding to 40–4000 nM folate in whole blood sample. Recoveries of added folic acid were 112–114 %. A good correlation was found when compared to a competitive heterogeneous assay based on the DELFIA-technology. The introduced assay provides a simple and fast method for whole blood folate measurement.

INTRODUCTION Unprocessed whole blood is a preferred sample matrix in point-of-care settings which require a simple and fast analysis. Furthermore, analytes bound by the red blood cells (e.g. folate) require whole blood specimen. Very sensitive fluorescence-based heterogeneous assays, i.e., solid-phase assays with washing steps, can be used for optically challenging sample matrix due to the separation steps which, however, complicate the instrumentation and increase the assay duration. Homogeneous assays, i.e. “mix and measure” type separation-free assays, are simple and fast to perform and thus suitable to point-of-care but the strong absorption of light by the whole blood limit the use of conventional fluorescent labels. Folic acid (vitamin B9) has a role in both amino acid and nucleic acid synthesis in cells1, and folic acid deficiency has been linked to an increased risk of vascular diseases such as coronary heart disease2, cancer3, and birth defects such as fetal neural tube defects4. The folic acid stored in red blood cells reflects the long-term folate level of tissues whereas folate in serum represents only the dietary intake. Thus red blood cell folic acid level should be measured for folate deficiency diagnosis5. The folates in red blood cells are in the form of polyglutamates whereas in the plasma they are only found as monoglutamates due to pteroylpoly-γ-glutamylcarboxypeptidase which deconjugates the polyglutamates.6,7 The red blood cells contain about 40 times the amount of folate compared to plasma.8 The whole blood folate comprises both red blood cell and plasma folates.

The available folate assays have been reviewed by Quinlivan et al.9 and they include microbiological assays, binding assays, chromatography assays and tandem mass spectrometry assays. The latter is currently considered as the candidate reference method for folate analysis being capable of differentiating between all different folate forms.10-12 The binding assays are mostly based on folate binding protein (FBP) instead of antibodies and they are usually competitive and heterogeneous. Commercial assays use radiolabels (QuantaPhase II, Bio-Rad Laboratories) or enzymes with chemiluminescence (ARCHITECT Folate, Abbott Laboratories; Advia Centaur Folate, Bayer Diagnostics; IMMULITE 2000, Siemens Medical Solutions Diagnostics) or electrochemiluminescence (Elecsys 2010, Roche Diagnostics) detection. The AutoDELFIA® Anemia assay (PerkinElmer Life Sciences, Wallac Oy) uses time-resolved fluorescence of a Eu3+-chelate. The published homogeneous binding assays for folate are based on fluorescence polarization and intensity13, chemiluminescence14, or enzyme labels with either bioluminescence15 or absorbance measurement16. However, from the published homogeneous assay concepts only in CEDIA® Red Blood Cell Folate (Roche Diagnostics/Boehringer Mannheim GmbH) assay whole blood is used as sample matrix. CEDIA® Red Blood Cell Folate assay is based on a color change generated by two fragments of β-galactosidase enzyme uniting in the presence of folate analyte.17,18 The strong absorption of ultraviolet and visible light by whole blood and autofluorescence background from biological material limit the use of conventional fluorescence-based labels in homogeneous whole blood assays. Upconverting

1 ACS Paragon Plus Environment


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

nanophosphors (UCNPs), e.g. NaYF4: Yb3+, Er3+, are exciting new luminescent labels for their unique photophysical properties: they can convert low-energy infrared radiation into higher-energy visible emission light.19,20 The infrared radiation is absorbed by ytterbium acting as sensitizer and the excitation energy is transferred to erbium serving as activator. The photon upconversion is enabled by stacking up the energies of two or more sequentially absorbed infrared photons utilizing the ladder-like metastable excited energy levels of erbium resulting then in emission of one higher-energy photon. The antiStokes’ photoluminescence enables detection without autofluorescence which is Stokes’ shifted. Furthermore, the red emission (at 660 nm) of NaYF4: Yb3+, Er3+ nanophosphors upon 980 nm excitation render luminescence measurements possible in optically challenging sample matrices such as whole blood which strongly absorbs light with wavelengths shorter than 600 nm.21 Small molecular organic dyes can be used as acceptors in upconversion resonance energy transfer (UC-RET) based assays with UCNPs acting as the donors.22-27 The narrow emission peaks and large anti-Stokes’ shift of the UCNPs enable the measurement of the UC-RET without donor cross-talk at the measurement window of the sensitized acceptor emission. The near-infrared excitation of UCNPs at 980 nm excludes direct excitation of the acceptor. The first homogeneous whole blood assay based on upconverting nanoparticles was introduced in 2007 and it was a competitive UC-RET-based assay for estradiol with antiestradiol Fab-fragment-coated UCNPs as donors and estradiol–Alexa Fluor 680 (AF680) conjugate as the acceptor.21 The upconversion sensitized acceptor emission was measured at 740 nm. Since then, UCNPs have been used to detect pH and metal ions28, blood potassium29 and matrix metalloproteinase-2 (MMP-2)30 from blood. Here, we introduce the first fluorescence-based homogeneous assay for quantification of whole blood folic acid using photon upconversion. The silica-coated NaYF4: Yb3+, Er3+ UCNPs were conjugated with FBP and used as donors in UCRET to an Alexa Fluor 680 acceptor conjugated to folate analog. The folic acid in sample competed with the AF680-folate analog for binding to the FBP. The sensitized acceptor emission was measured at 740 nm upon 980 nm excitation of the donor. At this wavelength range both donor emission and whole blood absorption were minimal. The assay was tested with whole blood samples from healthy donors and compared to a competitive heterogeneous assay based on the DELFIA label technology.

MATERIALS AND METHODS REAGENTS AND MATERIALS Folate binding protein (FBP) was obtained from Scripps Laboratories (San Diego, CA), DELFIA Enhancement Solution (DES) and dithiothreitol (DTT) were from PerkinElmer Life and Analytical Sciences (Wallac Oy, Turku, Finland). N10-(trifluoroacetyl)pteroic acid and pteroyl-L-glutamic acid (98 %, folic acid) were obtained from Sigma-Aldrich (St. Louis, MO). Tween-85 was from Merck (Darmstadt, Germany), and bovine serum albumin (BSA) from Bioreba (Reinach, Switzerland). Alexa Fluor 680-NHS ester (AF680) was purchased from Molecular Probes (Eugene, OR). K3-EDTA Venoject tubes were from Terumo NV Europe (Leuven, Belgium), streptavidin coated microtiter plates were

Page 2 of 11

obtained from Kaivogen Oy (Turku, Finland) and black Half Area 96-well plates were from Corning Incorporated (Corning, NY). Folate measurement buffer (FMB: 50 mM borate pH 9.3, 9 g/L NaCl, 3.75 g/L BSA, 0.1 g/L Tween-40, 0.5 g/L NaN3) was used in the folic acid assays with BSA concentration adjusted to correspond the protein concentration in 40-fold diluted hemolysed blood sample (~0.375 % w/vol). Folic acid (pteroyl-L-glutamic acid) standard stock solution was prepared by dissolving folic acid to mQ-H2O until a saturated solution was achieved. The mixture was centrifuged for 10 min with 16 800 g (Eppendorf Centrifuge 5418, Hamburg, Germany) and the supernatant was transferred to a fresh tube. The concentration of the folic acid was calculated by measuring the absorbance of the supernatant at 282 nm (molar absorptivity 27 600 M-1cm-1)31 with NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific). N10-(trifluoroacetyl)pteroic acid was first amino-modified and then conjugated to either Alexa Fluor 680 or a 9-dentate Eu3+-chelate32. Detailed protocols for preparation of the amino-modified folate analog and the labeling reactions, as well as the biotinylation of FBP and conjugation of FBP with silica-coated, Ø 29×36 nm NaYF4: Yb3+, Er3+ UCNPs are described in the Supporting Information. WHOLE BLOOD SAMPLE PRETREATMENT

Whole blood samples were taken from healthy volunteers into K3-EDTA Venoject tubes and the hematocrits were measured. The blood samples were stored at –20°C. For the assay, the whole blood samples were slowly thawed in room temperature and then mixed for 15 min in rotation before the pretreatment. First, the red blood cells were hemolysed by diluting the sample 20-fold in freshly prepared 0.5 % (w/v) ascorbic acid and mixing for 90 min at room temperature33. Thereafter, 30 µl of a mixture of 1 M NaOH and 11 mM DTT was added into 300 µl of lysed blood, and proteins were allowed to denature at pH ~11 for 15 min at room temperature in rotation. Finally, the pH was neutralized by adding 270 µl of FMB with 1 mM DTT and incubating for 15 min at room temperature in rotation. The hemolysate samples were spiked with folic acid diluted into FMB so that the final folate concentrations were 0, 5 and 10 nM in the homogeneous assay, and 0, 2 and 3 nM in the heterogeneous assay. By taking into account the dilution factors (1:20-dilution in lysing and 1:2-dilution in denaturation, i.e., 1:40 total dilution) during the pretreatment, the calculated spike amounts in the original whole blood sample would be 200 and 400 nM in the homogeneous assay, and 80 and 120 nM in the heterogeneous assay, respectively. The samples were spiked in duplicate for the homogeneous assay and in triplicate for the heterogeneous assay. The standard calibrators of folic acid in FMB were also treated the same way as the samples after lysing, i.e., into 300 µl of folic acid standard dilutions (0–1200 nM in heterogeneous assay, 0–2400 nM in homogeneous assay) in FMB, 30 µl of mixture of 1 M NaOH and 11 mM DTT were added and incubated for 15 min after which the pH was neutralized by addition of 270 µl of FMB with 1 mM DTT. HETEROGENEOUS REFERENCE ASSAY FOR WHOLE BLOOD FOLIC ACID The competitive heterogeneous reference assay for folic acid was performed in yellow streptavidin-coated microtitration


Page 3 of 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


wells and all reagents were diluted in FMB. First, 20 ng of bio-FBP was bound on a prewashed streptavidin-coated well in 50 µl by mixing the plate for 30 min in room temperature protected from light. The plate was washed twice and 30 µl of FMB was added to the wells rapidly after which triplicates of the pretreated folic acid standards or whole blood samples in 1:40-dilution spiked with folic acid to 0, 2 and 3 nM concentrations (corresponding 0, 80 and 120 nM folic acid in the whole blood sample before pretreatment) were added in 50 µl and incubated for 1 h. Thereafter, 50 nM Eu3+-labeled folate analog was added in 20 µl (total reaction volume 100 µl) and incubated for 1 h. The plate was washed for four times, 200 µl of DES was added and incubated for 30 min, and the Eu3+fluorescence was measured at 615 nm with Victor 1420 Multilabel Counter from the solution using the standard program for europium. HOMOGENEOUS UC-RET-BASED ASSAY FOR WHOLE BLOOD FOLIC ACID The homogeneous UC-RET-based assays were performed at room temperature in a total volume of 80 µl in black Half Area 96-well plates and all reagents were diluted in FMB. All reagent concentrations are given as the added concentrations instead of the final concentrations in total assay volume. The optimization of the homogeneous UC-RET-based assay for folic acid is described in the Supporting Information. The pretreated folic acid standard dilutions (0–2400 nM) in six replicates or whole blood samples in 1:40-dilution spiked in duplicate with folic acid to 0, 5 and 10 nM concentrations (corresponding to 0, 200 and 400 nM folic acid in the whole blood sample before pretreatment) in 40 µl volume were mixed in three replicates with 20 µg/ml of FBP-UCNP in 20 µl volume for 30 min at room temperature, protected from light in slow rotation to prevent blood cell sedimentation. Then, 12 nM AF680-labeled folate analog was added in 20 µl volume and the sensitized emission of the AF680 acceptor was measured at 740 nm with modified Plate Chameleon fluorometer (Hidex Oy, Finland) equipped with a 980 nm laser diode as an excitation source34 after 30 min of incubation at room temperature in slow rotation, protected from light.

RESULTS AND DISCUSSION WHOLE BLOOD SAMPLE PRETREATMENT In deoxygenated venous blood the folate polyglutamates in the red blood cells are bound by deoxyhemoglobin.35 Although the binding affinity is low, the high number of hemoglobin leads to the folate binding.36 The red blood cells were hemolysed to release the folate to assayable format by diluting the blood sample 20-fold into 0.5 % (w/vol) ascorbic acid causing the red blood cells to lyse osmotically. Ascorbic acid acted as an antioxidant protecting the reduced folates from oxidizing.36 At the same time, the ascorbic acid decreased the pH of the hemolysate to 4–5 allowing for the deconjugation of the erythrocyte folate polyglutamates to monoglutamates by endogenous human plasma pteroylpoly-γ-glutamylcarboxypeptidase.37 At this point, the polyglutamate forms might still be bound by deoxyhemoglobin with low affinity, but because most of the polyglutamate residues extrude the hemoglobin molecule38, they are available for the plasma conjugase.36 The removal of glutamate residues decreases the binding affinity of deoxyhemoglobin to folate dissociating it to

plasma.36 The predominant form of folate in human serum is N-5-methyltetrahydrofolate.6 Next, the endogenous folate binders of blood were denatured by increasing the pH to 11 by adding NaOH and dithiothreitol (DTT) which also acts as a strong reductant stabilizing the reduced folate monoglutamate forms at higher pH.39 At last, the hemolysate was neutralized by addition of folate measurement buffer (FMB) with DTT to adjust the pH to ~9.3 where the FBP has the same affinity towards both 5methyltetrahydrofolate and folic acid.40 Pteroyl-L-glutamic acid, i.e. folic acid, which is the fully oxygenated form of folate, was used as the standard calibrator in the assay. Because the whole blood folates were kept at reduced forms, also the folic acid standard calibrators were subjected to the pretreatment protocol to bring them to reduced form. WHOLE BLOOD FOLIC ACID ASSAY PRINCIPLES The two competitive folic acid assay principles are illustrated in Figure 1. In the heterogeneous reference assay the biotinylated FBP (bio-FBP) is first bound onto the surface of streptavidin coated microtiter wells and the folic acid in the sample competes with the 9d-Eu3+-chelate-labeled folate analog for binding to the FBP. Unbound reagents are removed by washing steps and the fluorescence of the Eu3+-label is detected at 615 nm using time-resolved measurement (Figure 1B). In the homogeneous UC-RET-based assay the FBP is conjugated onto the surface of the silica-coated UCNPs and the folic acid in the sample competes with the AF680-labeled folate analog for binding to the FBP (Figure 1A). When the AF680-labeled folate analog is in close proximity with the UCNP-donor, i.e., bound to the FBPs on the surface, upconversion resonance energy transfer (UC-RET) can occur between the donor and acceptor resulting in sensitized acceptor emission. The spectral principle of the assay is illustrated in Figure 2. The 1:20-diluted whole blood in 0.5 % ascorbic acid absorbs light strongly below 600 nm but at the assay dilution (1:80) the absorbance is reduced. The overlap of the excitation spectrum of the AF680 dye and the 660 nm emission peak of the NaYF4: Yb3+, Er3+ UCNP upon excitation at 980 nm was exploited for the resonance energy transfer. The UC-RETsensitized emission of the AF680 acceptor was then measured at 740±20 nm. At this wavelength region the UCNP-donor has an emission minimum and the whole blood absorption is low. Thus the background originating from blood and from donor crosstalk was minimized. Further, the background caused by Stokes’ shifted autofluorescence is totally avoided spectrally because of the anti-Stokes nature of the upconversion photoluminescence.



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 11

Figure 1. Principles of the two competitive assays for whole blood folic acid. A) In the homogeneous UC-RET-based assay the folic acid in the sample competes with AF680-labeled folate analog for binding to the FBP conjugated onto the surface of the UCNP. Upon 980-nm excitation, the upconversion energy is transferred to the bound AF680 via UC-RET and the sensitized emission is measured at 740 nm. B) In the heterogeneous reference assay the biotinylated FBP is first bound onto the streptavidin coated wells and the folic acid in the sample competes with 9d-Eu3+-chelate labeled folate analog for binding to the FBP. All excess unbound reagents are removed by washing steps. The fluorescence of the Eu3+ at 615 nm is detected upon 340-nm excitation using time-resolved measurement. With increasing folic acid concentration the UC-RET or fluorescence of 9d-Eu3+ decreases. UC-RET, upconversion resonance energy transfer; AF680, Alexa Fluor 680; FBP, folate binding protein; UCNP, upconverting nanophosphor; 9d-Eu3+, 9-dentate Eu3+-chelate.

Figure 2. The spectral principle of the UC-RET-based assay. The absorbance of 1:20-diluted whole blood in 0.5 % ascorbic acid (solid red line) and 1:80-diluted pretreated whole blood (assay dilution) in folate measurement buffer (dotted red line) are presented together with the upconversion luminescence spectrum of the UCNP donor (solid green line), and excitation (dotted black line) and emission spectrum (solid black line) of the AF680 acceptor. The sensitized emission of the acceptor is measured at 740±20 nm using a band-pass emission filter and continuous laser excitation at 980 nm. Autofluorescence (grey filled area) is avoided due to the anti-Stokes measurement. The upconversion luminescence and acceptor fluorescence spectra have been normalized to the same intensity levels. a.u., arbitrary unit.


Page 5 of 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


WHOLE BLOOD FOLIC ACID ASSAYS The developed homogeneous UC-RET-based assay was compared with a heterogeneous reference assay based on the established DELFIA-technology which has been used e.g. in the AutoDELFIA® Anemia assay by PerkinElmer Life Sciences (Wallac Oy). The sensitivity and the measurable concentration range of the homogeneous UC-RET-based assay can be changed by adjusting the amount of binding sites i.e. the concentration of the UCNP-FBP in the assay. For the homogeneous whole blood folic acid assay 20 µg/ml UCNP-FBP was chosen to bring the standard curve to similar measurable concentration region with the heterogeneous reference assay for assay comparison. The typical sigmoidal standard curves of both homogeneous UC-RET-based assay and heterogeneous reference assay for the whole blood folic acid are depicted in Figure 3. The sigmoidal Logistic fit function y=A2+(A1–A2)/(1+(x/x0)p) (where A1 is the initial value, A2 is the final value, x0 is the center value and p is the power) of Origin v8 (OriginLab Corporation, Northampton, MA) was used to create the standard curves. The IC50-values, i.e. the folate concentration x0 in which the signal response is half of the maximum, were 6.0 and 2.0 nM, respectively, which correspond to 240 and 80 nM whole blood folate concentrations (when multiplied by the dilution factor of 40), respectively. The lower limits of detection (LOD) were calculated from the standard curves as the mean of the sensitized emission of the zero calibrator subtracted by three times the standard deviation. The LODs for the homogeneous and heterogeneous assays were 1.0 and 0.4 nM (corresponding to 40 and 16 nM whole blood folate concentrations), respectively. The concentration coefficients of variations were below 20 % at folic acid concentration range of 3.2–100 nM and 0.5–50 nM, respectively, corresponding to 128–4000 nM and 20–2000 nM whole blood folic acid concentrations, respectively. Thus, the measurable concentration range of both assays was two orders of magnitude. According to Snow41, the lower limit of the reference range for serum folate is about 6.8 nM and for red blood cells 340–567 nM, whereas, according to ARCHITECT Folate Assay (Abbott Diagnostics, Chicago, IL), the normal folic acid reference range in red blood cells is 285–1475 nM and in serum 7–46.5 nM. The assay sensitivity is not as important requirement as a wide dynamic range in measuring whole blood folate. Thus, the homogeneous whole blood folic acid assay met the requirements for folate deficiency analysis with the wide dynamic range. Since the resonance energy transfer is inversely distance dependent process to the sixth power, only the acceptors which are bound to the UCNP-surface via FBP can participate in the UC-RET. Further, only the erbium ions within the UCNP that are in the required distance can transfer their energy to the acceptors. The previously presented UC-RET-based whole blood assay for estradiol was performed on ~300 nm sized upconverting phosphor.21 Thus, only the erbium ions located near the particle surface were able to transfer their energy to the acceptors and the core of the particle volume only produced background by nonproximity-based reabsorptive energy transfer. In this study, the upconverting nanophosphors were 29×36 nm in diameter and highly monodispersed and uniform as can be seen from the transmission electron microscopy image (Figure S1, Supporting Information). Most erbium ions

are within the required distance adding to the value of the obtained UC-RET signals.

Figure 3. The sigmoidal standard curve (open symbols) and the concentration coefficient of variation percentage (CV%) profile (filled symbols) of the a) homogeneous UC-RET-based assay (R2 = 0.992) and b) heterogeneous reference assay for folate (R2 = 0.995). The UC-RET was measured from six replicate wells and the time-resolved fluorescence from three replicate wells. The error bars represent the standard deviation of the measured replicas. The dotted line indicates the limit of detection (LOD) defined as the mean of the sensitized emission of zero calibrator – three times the standard deviation. The LODs were a) 1.0 and b) 0.4 nM. The IC50 values for the assays were a) 6.0 and b) 2.0 nM. UC-RET, upconversion resonance energy transfer; cts, counts; TR-fluorescence, time-resolved fluorescence.

The folic acid content and the analytical recovery (i.e. trueness) were measured from the same whole blood sample (hematocrit 49 %) with both assays using the obtained standard curves to study the correlation and reliability of the developed homogeneous UC-RET based assay. The whole blood sample were spiked with two folic acid concentrations and the



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 recoveries were calculated by subtracting the endogenous folate amount (0-spiked) from the found amount and dividing it by the spiked amount. A recovery of 100 % would mean that all added folic acid is found by the assay and there is no interfering matrix effect present. For the heterogeneous assay, two replicate blood hemolyses were done, and from the other hemolysate two replicate pretreatments were performed. For the homogeneous assay, one hemolysis and two replicate pretreatments were performed. The mean endogeneous whole blood folic acid concentration of the 0-spiked sample was 148.0 nM and 86.5 nM according to the homogeneous and the heterogeneous assays, respectively (Table 1). The hematocrit corrected folic acid amounts were 302 nM/ml of packed cells and 176 nM/ml of packed cells, respectively. The analytical recoveries from spiked blood samples were 112–114 % and 126–131 %, respectively (Table 1). Table 1. Recovery of folic acid (FA) from whole blood sample hemolysed and pretreated in replicates Homogeneous UC-RET-based assay (n = 2) Spiked FA (nM) WBFA (nM)a,b Recovery %b,c 0 148.0 (±25.4) 200

373.6 (±6.6)

112.8 (±3.3)

400

606.6d

114.6d

Heterogeneous reference assay (n = 3) Spiked FA (nM) WBFA (nM)a,b

Page 6 of 11

ies, especially in the heterogeneous reference assay (Table 1). This might be due to the spiked folate species being folic acid, i.e., pteroyl-L-glutamic acid, whereas the predominant endogenous folate form in serum is N-5-methyltetrahydrofolate. Even though the pH in the assay was set to ~9.3 where FBP has the same binding affinity towards both folate species, there might have been differences in local environments of the two assays. In fact, the bio-FBP bound on well surface was found to be sensitive for drying in the heterogeneous assay: if the bio-FBP wells dried before the addition of sample, the binding affinity suffered and the obtained signal levels at the end of the 96-well microtiter plate (which was pipetted the last) were ~51 % from the signal levels at the beginning of the plate (data not shown). This decrease in binding affinity was permanent. However, this was solved by adding 30 µl of FMB to the wells quickly with a 12-channel pipet after washing the excess bioFBP away to prevent the drying before the addition of standard calibrators or samples. Also Wilson et al.44 have noticed that uptake of atmospheric CO2 by one of the ARCHITECH folate assay reagent affected the binding affinity of FBP to pteroylglutamic acid probably due to a change of pH. Whole blood folate measurements have shown a wide interlaboratory and inter-method variability due to difficulties in standardization.42 However, Thorpe et al.42 introduced a reference material for whole blood folate which was accepted as an International Standard. The introduced proof-of-concept assay could be further tested for correlation with the International Standard to investigate the clinical value of the assay.

Recovery %b,c

CONCLUSIONS

0

86.5 (±6.6)

80

191.6 (±6.5)

131.4 (±8.1)

120

237.2 (±13.3)

125.6 (±11.1)

a

WBFA, Whole blood folic acid, not hematocrit corrected; Mean (Standard deviation); c([WBFA in spiked sample] – [WBFA without spike]) / [Spiked FA] * 100; dn = 1. b

The inter-method variability (calculated from the folic acid concentrations of 0-spiked sample replicates) was 32 % (n = 5), which is similar to that reported (30–41 %) for an interlaboratory comparison study f or whole blood folate with an International Standard42, and also compared to the variability (35.7 %) reported in an international round robin comparison study43. The source of the assay variability between the homogeneous UC-RET-based assay and the heterogeneous reference assay must be in the analysis method itself as the assay standards were prepared from the same folate species (pteroylL-glutamic acid, folic acid), and the assayed sample and the sample pretreatment protocol were the same for both assays. The signals obtained from the spiked samples fell upon the upper part (i.e., lower standard concentrations) and in the middle of the standard curve with the homogeneous UC-RET assay, but upon the lower part (i.e., the higher standard concentrations) with the heterogeneous assay. The dynamic range in the standard curve of the heterogeneous assay was slightly wider than the dynamic range of homogeneous assay. The replicate whole blood pretreatments were repeatable, as the CVs in the homogeneous assay were in between of 1.6–17.1 % and in the heterogeneous assay 3.4–7.6 %. The spiked whole blood samples gave falsely high folate concentrations as can be seen from the high analytical recover-

In this study, a new competitive homogeneous assay for quantification of whole blood folate was demonstrated with superior characteristics compared to the conventional folate assays. The assay was based on UC-RET between a FBPconjugated UCNP donor and an AF680 acceptor labeled folate analog which enabled the use of near-infrared and far-red wavelengths for both excitation and the measurement of the sensitized acceptor emission. At this wavelength region the whole blood sample matrix is relatively transparent. Thus, the assay is performed free from background caused by autofluorescence or direct excitation of the acceptor and without attenuation of upconversion photoluminescence by the whole blood absorption. The developed homogeneous UC-RET-based assay is simple, fast and sensitive, and it can be used with a large range of sample folate concentrations (which is the case with whole blood folate). The “mix and measure” type assay without a need for separation steps also simplifies the instrument requirements for assay automation. The homogeneous UC-RET-based whole blood folate assay was compared with a competitive heterogeneous assay based on the DELFIAtechnology and a fairly low inter-method variability was found. The sample pretreatment is the most time-consuming step taking about 2.5 h while the homogeneous assay itself is performed in one hour. In comparison, the performance of the heterogeneous reference assay takes up 4 h even excluding the sample pretreatment. The simplicity and the unique capabilities of homogeneous UC-RET-based assays provide a sophisticated new tool for rapid clinical diagnostics also in optically challenging sample materials.


Page 7 of 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


ASSOCIATED CONTENT Supporting Information Preparation of assay reagents: bio-FBP, amine-modified folate analog, labeling of folate analog with Eu3+-chelate and Alexa Fluor 680 and preparation of UCNP-bioconjugate. Optimization of the homogeneous UC-RET-based folic acid assay. This material is available free of charge via the Internet at http://pubs.acs.org.

AUTHOR INFORMATION Corresponding Author * Riikka Arppe, Department of Biochemistry / Biotechnology, University of Turku, Tykistökatu 6A, 20520 Turku, Finland, Tel.: (+358) 2-333-8089, Fax: (+358) 2-333-8050, E-mail: [email protected]

Present Addresses †

AIT Austrian Institute of Technology GmbH, Konrad-LorenzStrasse 24, 3430 Tulln, Austria †† Kaivogen Oy, Tykistökatu 4D, 2nd Floor, 20520 Turku, Finland ††† Institute of Molecular Medicine Finland FIMM, Biomedicum Helsinki 2U, Tukholmankatu 8, 00290 Helsinki, Finland

Author Contributions All authors have given approval to the final version of the manuscript.

ACKNOWLEDGMENT This study was supported by Tekes, the Finnish Funding Agency for Technology and Innovation (Grant number 40202/12), and the National Doctoral Programme of Advanced Diagnostic Technologies and Applications (DIA-NET). The authors would like to thank Emilia Palo for synthesizing the UCNPs. This work made use of the Laboratory of Electron Microscopy premises at University of Turku.

REFERENCES (1) Appling, D. R. FASEB J. 1991, 5, 2645–2651. (2) Robinson, K.; Arheart, K.; Refsum, H.; Brattström, L.; Boers, G.; Ueland, P.; Rubba, P.; Palma-Reis, R.; Meleady, R.; Daly, L.; Witteman, J.; Graham, I. Circulation 1998, 97, 437–443. (3) Choi, S.-W.; Mason, J. B. J. Nutr. 2000, 130, 129–132. (4) Botto, L. D.; Moore, C. A.; Khoury, M. J.; Erickson, J. D. N. Engl. J. Med. 1999, 341, 1509–1519. (5) Amos, R. J.; Dawson, D. W.; Fish, D. I.; Leeming, R. J.; Linnell, J. C. Clin. Lab. Haematol. 1994, 16, 101–115. (6) Herbert, V.; Larrabee, A. R.; Buchanan, J. M. J. Clin. Invest. 1962, 41, 1134–1138. (7) Perry, J.; Chanarin, I. Br. J. Haematol. 1977, 35, 397–402. (8) Hoffbrand, A. V.; Newcombe, F. A.; Mollin, D. L. J. Clin. Pathol. 1966, 19, 17–28. (9) Quinlivan, E. P.; Hanson, A. D.; Gregory, J. F. Anal. Biochem. 2006, 348, 163–184. (10) Nelson, B. C.; Pfeiffer, C. M.; Margolis, S. A.; Nelson, C. P. Anal. Biochem. 2003, 313, 117–127. (11) Nelson, B. C.; Pfeiffer, C. M.; Margolis, S. A.; Nelson, C. P. Anal. Biochem. 2004, 325, 41–51. (12) Pfeiffer, C. M.; Fazili, Z.; McCoy, L.; Zhang, M.; Gunter, E. W. Clin. Chem. 2004, 50, 423–432. (13) Martin, H.; Comeskey, D.; Simpson, R. M.; Laing, W. A.; McGhie, T. K. Anal. Biochem. 2010, 402, 137–145. (14) Adamczyk, M.; Fino, J. R.; Mattingly, P. G.; Moore, J. A.; Pan, Y. Bioorg. Med. Chem. Lett. 2004, 14, 2313–2317. (15) Huang, W.; Feltus, A.; Witkowski, A.; Daunert, S. Anal. Chem. 1996, 68, 1646–1650.

(16) Bachas, L. G.; Meyerhoff, M. E. Anal. Chem. 1986, 58, 956– 961. (17) Engel, W. D.; Khanna, P. L. J. Immunol. Methods 1992, 150, 99–102. (18) Khanna, P. L.; Dworschack, R. T.; Manning, W. B.; Harris, J. D. Clin. Chim. Acta 1989, 185, 231–239. (19) Haase, M., Schäfer, H. Angew. Chem. Int. Ed. 2011, 50, 5808– 5829. (20) Riuttamäki, T., Soukka, T. Biorev. 2014, 1, 155–204. (21) Kuningas, K.; Päkkilä, H.; Ukonaho, T.; Rantanen, T.; Lövgren, T.; Soukka, T. Clin. Chem. 2007, 53, 145–146. (22) Kuningas, K.; Rantanen, T.; Ukonaho, T.; Lövgren, T.; Soukka, T. Anal. Chem. 2005, 77, 7348–7355. (23) Kuningas, K.; Ukonaho, T.; Päkkilä, H.; Rantanen, T.; Rosenberg, J.; Lövgren, T.; Soukka, T. Anal. Chem. 2006, 78, 4690– 4696. (24) Rantanen, T.; Järvenpää, M. L.; Vuojola, J.; Arppe, R.; Kuningas, K.; Soukka, T. The Analyst 2009, 134, 1713–1716. (25) Riuttamäki, T.; Hyppänen, I.; Kankare, J.; Soukka, T. The J. Phys. Chem. C 2011, 115, 17736–17742. (26) Arppe, R., Näreoja, T., Nylund, S., Mattsson, L., Koho, S., Rosenholm, J.M., Soukka, T., Schäferling, M. Nanoscale 2014, 6, 6837–6843. (27) Liu, C., Chang, L., Wang, H., Bai, J., Ren, W., Li, Z. Anal. Chem. 2014, 86, 6095–6102. (28) Xie, L.; Qin, Y.; Chen, H.-Y. Anal. Chem. 2012, 84, 1969– 1974. (29) Xie, L.; Qin, Y.; Chen, H.-Y. Anal. Chem. 2013, 85, 2617– 2622. (30) Wang, Y.; Shen, P.; Li, C.; Wang, Y.; Liu, Z. Anal. Chem. 2012, 84, 1466–1473. (31) Póo-Prieto, R.; Haytowitz, D. B.; Holden, J. M.; Rogers, G.; Choumenkovitch, S. F.; Jacques, P. F.; Selhub, J. J. Nutr. 2006, 136, 3079–3083. (32) von Lode, P.; Rosenberg, J.; Pettersson, K.; Takalo, H. Anal. Chem. 2003, 75, 3193–3201. (33) Hansen, S. I.; Holm, J.; Nexo, E. Clin. Chem. 1987, 33, 1360– 1363. (34) Soukka, T.; Kuningas, K.; Rantanen, T.; Haaslahti, V.; Lövgren, T. J. Fluoresc. 2005, 15, 513–528. (35) Benesch, R. E.; Benesch, R.; Kwong, S.; Baugh, C. M. Proc. Natl. Acad. Sci. USA 1983, 80, 6202–6205. (36) Wright, A. J.; Finglas, P. M.; Southon, S. Clin. Chem. 1998, 44, 1886–1891. (37) Lakshmaiah, N.; Ramasastri, B. V. Int. J. Vitam. Nutr. Res. 1975, 45, 183–193. (38) Arnone, A.; Rogers, P. H.; Benesch, R. E.; Benesch, R.; Kwong, S. J. Biol. Chem. 1986, 261, 5853–5857. (39) Lucock, M. D.; Green, M.; Hartley, R.; Levene, M. I. Food Chem. 1993, 47, 79–86. (40) Givas, J. K.; Gutcho, S. Clin. Chem. 1975, 21, 427–428. (41) Snow, C. F. Arch. Intern. Med. 1999, 159, 1289–1298. (42) Thorpe, S. J.; Sands, D.; Heath, A. B.; Hamilton, M. S.; Blackmore, S.; Barrowcliffe, T. Clin. Chem. Lab. Med. 2004, 42, 533–539. (43) Gunter, E. W.; Bowman, B. A.; Caudill, S. P.; Twite, D. B.; Adams, M. J.; Sampson, E. J. Clin. Chem. 1996, 42, 1689–1694. (44) Wilson, D. H.; Williams, G.; Herrmann, R.; Wiesner, D.; Brookhart, P. Clin. Chem. 2005, 51, 684–687.



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 11


Page 9 of 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


Principles of the two competitive assays for whole blood folic acid. A) In the homogeneous UC-RET-based assay the folic acid in the sample competes with AF680-labeled folate analog for binding to the FBP conjugated onto the surface of the UCNP. Upon 980-nm excitation, the upconversion energy is transferred to the bound AF680 via UC-RET and the sensitized emission is measured at 740 nm. B) In the heterogeneous reference assay the biotinylated FBP is first bound onto the streptavidin coated wells and the folic acid in the sample competes with 9d-Eu3+-chelate labeled folate analog for binding to the FBP. All excess unbound reagents are removed by washing steps. The fluorescence of the Eu3+ at 615 nm is detected upon 340-nm excitation using time-resolved measurement. With increasing folic acid con-centration the UC-RET or fluorescence of 9d-Eu3+ decreases. UC-RET, upconversion resonance energy transfer; AF680, Alexa Fluor 680; FBP, folate binding protein; UCNP, upconverting nanophosphor; 9d-Eu3+, 9-dentate Eu3+-chelate. 177x86mm (300 x 300 DPI)

ACS Paragon Plus Environment


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

The spectral principle of the UC-RET-based assay. The absorbance of 1:20-diluted whole blood in 0.5 % ascorbic acid (solid red line) and 1:80-diluted pretreated whole blood (assay dilution) in folate measurement buffer (dotted red line) are presented together with the upconversion luminescence spectrum of the UCNP donor (solid green line), and excitation (dotted black line) and emission spectrum (solid black line) of the AF680 acceptor. The sensitized emission of the acceptor is measured at 740±20 nm using a band-pass emission filter and continuous laser excitation at 980 nm. Autofluorescence (grey filled area) is avoided due to the anti-Stokes measurement. The upconversion luminescence and acceptor fluorescence spectra have been normalized to the same intensity levels. a.u., arbitrary unit. 84x63mm (300 x 300 DPI)


Page 10 of 11

Page 11 of 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


The standard curve (open symbols) and the concentration coefficient of variation percentage (CV%) profile (filled symbols) of the a) homogeneous UC-RET-based assay and b) heterogeneous reference assay for folate. The UC-RET was measured from six replicate wells and the time-resolved fluorescence from three replicate wells. The error bars represent the standard deviation of the measured replicas. The dotted line indicates the limit of detection (LOD) defined as the mean of the sensitized emission of zero calibrator – three times the standard deviation. The LODs were a) 1.0 and b) 0.4 nM. The IC50 values for the assays were a) 6.0 and b) 2.0 nM. UC-RET, upconversion resonance energy transfer; cts, counts; TR-fluorescence, time-resolved fluorescence. 148x259mm (600 x 600 DPI)


Homogeneous Assay for Whole Blood Folate Using Photon

Recommend Documents