A Paper-Based Multiplexed Resonance Energy Transfer Nucleic Acid Hybridization Assay Using a Single Form of Upconversion Nanoparticle as Donor and Three Quantum Dots as Acceptors

Samer Doughan, Uvaraj Uddayasankar, Aparna Peri, Ulrich J. Krull*

Chemical Sensors Group, Department of Chemical and Physical Sciences, University of Toronto Mississauga, 3359 Mississauga Road, Mississauga, ON L5L 1C6, Canada

*Author to whom correspondence should be addressedEmail: ulrich.krull@utoronto.caTel: 1-905-828-5437

Abstract

Monodisperse aqueous upconverting nanoparticles (UCNPs) were covalently immobilized on

aldehyde modified cellulose paper via reductive amination to evaluate the multiplexing capacity

of luminescence resonance energy transfer (LRET) between UCNPs and quantum dots (QDs).

This is the first account of a multiplexed bioassay strategy that demonstrates the principle of use

of a single form of UCNP as donor and three different color emitting QDs as acceptors to

concurrently determine three analytes. Broad absorbance profiles of green, orange and red QDs

that spanned from the first exciton absorption peak to the UV region were in overlap with a blue

emission band from UCNPs composed of NaYF4 that was doped with 30% Yb3+, 0.5% Tm3+,

allowing for LRET that was stimulated using 980 nm near-infrared radiation. The characteristic

narrow and well-defined emission peaks of UCNPs and QDs allowed for the collection of

luminescence from each nanoparticle using a band-pass optical filter and an epi-fluorescence

microscope. The LRET system was used for the concurrent detection of uidA, Stx1A and tetA

Abbreviations: Base Pair Mismatch (BPM), Cy3 (Indocarbocyanine), Fluorescence Resonance Energy Transfer (FRET), Fully complementary (FC), reduced L-glutathione (GSH), Luminescence Resonance Energy Transfer (LRET), Nanoparticle (NP), Oleic Acid (OA), o-phosphorylethanolamine (PEA), photomultiplier tube (PMT), tetramethylammonium hydroxide (TMAH), Upconverting Nanoparticle (UCNP). 1

gene fragments with selectivity even in serum samples, and reached limits of detection of 26

fmol, 56 fmol and 76 fmol, respectively.

Keywords:

Upconversion Nanoparticle, Quantum Dot, Luminescence Resonance Energy Transfer, Bioassay,

Multiplexed Detection, Paper.

1. Introduction

Upconversion nanoparticles (UCNPs) are lanthanide doped inorganic crystals with multiple

narrow and well-defined emission peaks. Upconversion is based on the sequential absorption of

two or more photons in the NIR or IR region of the electromagnetic spectrum followed by

emissions spanning the UV to NIR region. Excitation using NIR radiation minimizes

autofluorescence from biological samples and reduces optical background associated with scatter

from UV and visible excitation sources [1]. These properties have made UCNPs attractive for

use in bioassays for the detection of nucleic acids and proteins [1-4].

The tuneable narrow emission profiles governed by the electronic structure and concentration of

the lanthanide dopants allow for UCNPs to be used as multiplexing agents in bioassays. The use

of UCNPs as passive labels for multiplexing has been widely explored [5-10], and has primarily

been of interest owing to the opportunity for NIR excitation of optical processes. However, use

as LRET donors has been limited to the concurrent detection of two biomolecules [11-13]. A

further attribute suitable for analytical applications is that LRET methods offer access to

ratiometric methodology that provides for good precision [14]. Rantanen et al. reported a dual

parameter sandwich-based nucleic acid hybridization assay using two colors from one UCNP as

2

LRET donors and two molecular dyes as acceptors with a limit of detection of 28 fmol [12]. He

et al. demonstrated the first UCNP-based LRET assay on paper for the detection of matrix

metalloproteinase-2 (MMP-2) [15]. Recently, He et al. reported a portable UCNP-based paper

device for the detection of cocaine based on the quenching of UCNP luminescence by gold NPs

[16]. Zhou et al. demonstrated a two-plex assay on paper using dye-labelled nucleic acids [13]. A

similar study for a sandwich-based single-plex hybridization assay demonstrated limited

sensitivity and selectivity and had a limit of detection of 146 fmol [17]. Our research group has

reported a sandwich-based hybridization assay on paper using UCNPs as donors and QDs as

acceptors with a limit of detection of 13 fmol for the HPRT1 housekeeping gene fragment [18].

While QDs are used as LRET accepters herein, they are more typically used as donors in

fluorescence resonance energy transfer (FRET)-based bioassays. Multiplexed FRET-based

hybridization assays have been reported with a maximum of one QD donor and two molecular

dye acceptors. Three-plex detection was achieved only when two FRET channels between green

emitting QD donor and Cy3 and Rhodamine Red-X acceptors were used, while the third channel

was based on the direct excitation of Pacific Blue [19]. The work was extended for the

simultaneous detection of four different targets on optical fibres with the addition of red emitting

QD donors and Alexa Fluor 647 acceptor. Detection limits of 1-7 nM were achieved for dual

labelled nucleic acid reporters [20].

The FRET multiplexing capacity for one donor was limited by the small Stoke’s shift and broad

emission profiles of molecular dyes. Herein, QDs are used as LRET acceptors. In addition to

their characteristic photostability and high extinction coefficient compared to molecular dyes,

QDs have broad absorption profiles in the UV and blue region of the spectrum. This allows the

fluorescence of any color emitting QD to be sensitized by blue emitting UCNPs. Both UCNPs

3

and QDs have narrow and well defined emission bands, allowing for QD emission peaks to be

resolved in the visible region of the spectrum using only optical band pass filters. LRET has been

well characterized between UCNPs and QDs [21, 22], and in particular between lanthanides and

QDs [23, 24]. Forster Distance between UCNP and QDs was characterized to be 15 Å [22].

Herein, we use a paper-based assay to further evaluate the capacity for multiplexing using LRET

between UCNPs and QDs. The immobilization of UCNPs on paper offers a facile layer by layer

assembly of the assay without the need for tedious purification steps and avoids problems

associated with aggregation of nanoparticles in solution. The simultaneous detection of three

nucleic acid targets using one UCNP as donor and three different color emitting QDs as

acceptors allows for the controlled placement of QDs in close proximity to the UCNPs for

LRET. The large surface area associated with the three dimensional matrix of paper, in addition

to the contraction of wet paper upon drying that brings donors and acceptors in closer proximity,

can give up to a 10 fold enhancement in ratiometric signal by LRET[25].

This work presents the first account of LRET between a single form of UCNP and three

acceptors. The photoluminescence from three different color emitting QDs was concurrently

sensitized by blue emission of a single form of UCNP. Three independent optical channels for

green, orange and red emitting QDs were isolated using optical band pass filters. Examination of

the potential of the LRET system for higher-order multiplexing was done using a sandwich-

based hybridization assay for the concurrent detection of uidA, Stx1A and tetA gene fragments.

The uidA sequence is diagnostic of Escherichia coli [26], the Stx1A gene encodes the production

of Shiga toxins[27], and the tetA gene indicates resistance to tetracycline[28]. These markers

offer potential for the detection of E.coli, its pathogenicity and its resistance to an antibiotic.

4

2. Experimental

A full list of Materials and Instrumentation can be found in the Supporting Information.

2.1 Synthesis of NaYF4: 0.5% Tm3+, 30% Yb3+/NaYF4 Core/Shell UCNPs

Oleic acid (OA) capped NaYF4: 0.5% Tm3+, 30% Yb3+/NaYF4 core/shell UCNPs were

synthesized according to previous reports [29]. Core NaYF4: 0.5% Tm3+, 30% Yb3+ were

synthesized by first stirring 456.2 mg, 253.4 mg and 4.2mg of Y(CH3CO2)3.xH2O,

Yb(CH3CO2)3.4H2O, Tm(CH3CO2)3.xH2O, respectively, in 30 mL of octadecene and 12 mL of

OA at 115 °C under vacuum for 30 minutes. The clear solution was then cooled to 50 °C under

argon and a 20 mL solution of methanol containing 0.20 g of NaOH and 0.30 g of NH4F was

added. The cloudy solution was stirred for 30 minutes before the temperature was raised to 75 °C

to evaporate the methanol. Then the solution was rapidly heated to 300 °C and maintained at this

temperature for 1 hour while stirring. The solution was allowed to cool to room temperature and

the core UCNPs were collected in ethanol and were separated by centrifugation. The core

UCNPs were re-suspended in hexanes and recaptured with ethanol and centrifugation. The core

UCNPs were stored in hexanes overnight at 4 °C.

Core UCNPs were capped with a NaYF4 shell. 573.8 mg of Y(CH3CO2)3.xH2O was stirred in 30

mL of octadecene and 12 mL of OA at 115 °C under vacuum for 30 min. The clear solution was

allowed to cool to 80 °C under argon before the core UCNPs in hexanes were added. After the

hexanes evaporated, the temperature was lowered to 50 °C and a 20 mL solution of methanol

containing 0.14 g of NaOH and 0.26 g of NH4F was added. The reaction temperature was

increased to 75 °C to evaporate the methanol before it was rapidly increased to 300 °C and

maintained for an hour while stirring. The reaction was allowed to cool to room temperature and

5

the core/shell UCNPs were captured with ethanol and centrifugation. The core/shell UCNPs

were re-suspended in hexanes and recaptured with ethanol and centrifugation three times. The

OA capped core/shell UCNPs were stored in hexanes at 4 °C for subsequent modification.

2.2 Preparation of Water Soluble UCNPs

OA capped UCNPs were made water soluble via ligand exchange with o-

phosphorylethanolamine (PEA) according to previous reports [30]. 100 mg of OA capped

UCNPs in 2 mL of hexanes were added to a 10 mL solution of ethanol containing 400 mg of

PEA and 1 mL of tetramethylammonium hydroxide (TMAH). The solution was capped and

allowed to stir overnight at 70 °C. PEA capped UCNPs were collected by centrifugation at 3500

rpm. The NPs were washed three times by sonication in ethanol before the addition of hexanes,

followed by centrifugation to collect the UCNPs. The PEA capped UCNPs were passed through

a 0.22 µm polyethersulfone (PES) syringe filter to remove any aggregates, and collected solution

was stored at 4 °C with excess PEA.

2.3 Paper Modification for Immobilization of PEA-UCNPs and DNA Probes

To immobilize the UCNPs on paper, reaction zones with an inner diameter of 30 mm were

defined on Whatman™ 1 Chr chromatography paper using wax printing of a circular border. The

paper was placed in an oven at 120 °C for 2 min to melt the wax. Each zone was treated with a 5

µL oxidizing solution of 0.01 g mL-1 sodium(meta)periodate and 0.03 g mL-1 lithium chloride in

purified water. The paper was placed in an oven at 50 °C until dry. The procedure was repeated

before the paper was washed in purified water and placed in a desiccator to dry. Each spot of the

aldehyde modified paper was then treated with a 5 µL solution of 1.6 mg mL-1 PEA-UCNPs in

HEPES buffer (100mM, pH 7.2) containing 1 mM sodium cyanoborohydride. The solution was

allowed to incubate for 5 min before it was washed in HEPES buffer (100 mM, pH 7.2). The

6

paper was allowed to dry in a desiccator. A 3 µL aliquot of a 2 mM aqueous solution of NHS-

PEG4-biotin was applied to each spot and allowed to dry. The paper was then washed in purified

water and allowed to dry in a desiccator. Subsequently, each spot was treated with a 3 µL aliquot

of a 30 µM avidin solution in HEPES buffer (100 mM, pH 7.2). The paper was washed in borate

buffer (50 mM, pH 9.25). The dried spots were then treated with a 3 µL of solution containing 15

µM of each of uidA, Stx1A and tetA biotinylated oligonucleotide probes in borate buffer (50

mM, pH 9.25), followed by 3 µL aliquots of a 20 µM solution of unlabelled oligonucleotide, in

order to minimize the potential for fouling of the surface by adsorption of oligonucleotides in

hybridization experiments.

2.4 Preparation of HexahistidineFunctionalized Oligonucleotides and mPEG

Reporter nucleic acid sequences were functionalized with hexahistidine according to previous

reports [31, 32]. Thiol modified nucleic acid sequences were incubated with 500 molar excess of

dithiothreitol (DTT) in 1x PBS buffer for 1 hour to reduce the disulfide moieties. Excess DTT

was extracted four times with ethyl acetate. The isolated nucleic acid was mixed with 20 molar

equivalence of maleimide functionalized hexahistidine (6-Maleimidohexanoic acid –

G(Aib)GHHHHHH) dissolved in dimethyl sulfoxide (DMSO) and was allowed to shake

overnight. Excess peptide was removed using a NAP-5 desalting column. The oligonucleotide

was quantified by UV-vis spectroscopy, and was stored at -20 °C.

Thiol functionalized polyethyleneglycol (PEG)-methyl ether (MW: 6000 g mol-1) was incubated

with 10 molar excess of tris (2-carboxyethyl) phosphine TCEP for 2 hours to reduce any

disulfides. TCEP was removed using a NAP-5 desalting column and PEG was incubated with 20

molar equivalents of maleimide functionalized hexahistidine overnight. Excess peptide was

removed using a NAP-5 desalting column.

7

2.5 Preparation of QD-Reporter Conjugates

Alkyl Trilite™ QDs with emission maxima at 525 nm, 575 nm and 620 nm were obtained from

Crystalplex (Pittsburgh, USA). The QDs were rendered water soluble via ligand exchange with

reduced L-glutathione (GSH)[33]. Briefly, 75 µL of a 10 µM solution of alkyl QDs was

dispersed in 2 mL of chloroform and was added drop-wise to a 600 µL solution of TMAH

containing 0.2 g of GSH. The solution was briefly placed on a vortex mixer and then allowed to

stand overnight. The QDs were extracted from the chloroform solution three times into a total of

300 µL of borate buffer (50 mM, pH 9.25) and were precipitated with 20 µL of 3M NaCl and

300 µL of 95% ethanol by centrifugation at 8000 rpm for 5 minutes. The QDs were resuspended

in 300 µL of pH 9.25 borate buffer (50 mM, 100mM NaCl), and then were precipitated with an

equal volume of 95% ethanol and centrifugation at 8000 rpm two more times. The GSH coated

QDs were dissolved in pH 9.25 borate buffer (50 mM, 100 mM NaCl) and stored at 4 °C.

The 525 nm, 575 nm and 620 nm GSH coated QDs were incubated with a 5 fold excess of

hexahistidine modified uidA, Stx1A and tetA probes, respectively, in pH 9.25 borate buffer (50

mM, 100 mM NaCl) for one hour. The QDs were captured with ethanol and then centrifugation a

total of three times to remove unbound nucleic acids. The QDs were then incubated in 15 fold

excess hexahistidine modified PEG for one hour. The PEG stabilized DNA conjugated QDs were

stored at 4 °C.

2.6 Hybridization Assay on Paper Substrates

For the detection of individual oligonucleotide targets, 3 µL solutions of the target with

concentrations ranging from 0.01 µM to 2 µM in pH 9.25 borate buffer (50 mM, 100 mM NaCl)

were applied to the interaction zones on paper defined by the wax borders. For the simultaneous

detection of all three oligonucleotide targets, 3 µL solutions containing equimolar amounts of

8

each target with concentrations ranging from 0.01 µM to 2 µM in pH 9.25 borate buffer (50 mM,

100 mM NaCl) were applied to the zones. 3 µL aliquots of a solution containing 0.5 µM of each

of the QD-DNA reporter conjugates in borate buffer (pH 9.25, 50 mM, 100 mM NaCl) were then

applied and allowed to incubate for 5 minutes. The paper was washed in pH 9.25 borate buffer

(50 mM, 100 mM NaCl) with 0.1% v/v Tween for 2 minutes. The paper was allowed to dry in a

desiccator before it was imaged using an epi-fluorescence microscope. Filters of 420-490, 535-

545, 565-585 and 615-625 nm band pass were used for the collection of luminescence from

UCNPs, 525 QDs, 575 QDs and 620 QDs, respectively. The luminescence signal was based on

integrated intensity from the area of each spot.

Selectivity of the assay was investigated. Simultaneous detection of the three oligonucleotide

targets in 10% goat serum was indicative of response in a complicated sample matrix. Evaluation

of the potential for discrimination of targets that were differentiated by single nucleotide

polymorphism involved comparison of the assay response from a fully complementary (FC)

uidA target to one base pair mismatched (1 BPM) uidA target in the presence of FC Stx1A and

tetA targets. Solution containing 2 µM of each oligonucleotide target was spotted into the

interaction zones and the papers were washed in pH 9.25 borate buffer (50 mM, 100 mM NaCl)

with either no formamide or 5% v/v formamide. The papers washed in 5% formamide were

subsequently washed in pH 9.25 borate buffer before drying. A 3 µL aliquot of a solution

containing 0.5 µM of each of the QD-DNA reporter conjugates in the pH 9.25 borate buffer was

then applied to each interaction zone and allowed to incubate for 5 minutes. The papers were

then washed for 2 minutes in pH 9.25 borate buffer that contained 0.1% v/v Tween before being

dried and imaged.

9

.Table 1: Oligonucleotide sequences used in hybridization assays. H6 = hexahistidine

Target (QD) Sequence Name Sequence

uidA (525 nm)

Probe 5’ – Biotin – AGTCTTACTTCCATG – 3’

Target 3’ – TCAGAATGAAGGTACTAAAGAAATTGATAC – 5’

1 BPM Target 3’ – TCAGAATCAAGGTACTAAAGAAATTGATAC – 5’

Reporter 5’ – ATTTCTTTAACTATG – H6 – 3’

Stx1A (575 nm)

Probe 5’ – Biotin – GTCACCAGACAATGT – 3’

Target 3’ – CAGTGGTCTGTTACATTGGCGACAACATGG – 5’

Reporter 5’ – AACCGCTGTTGTACC – H6 – 3’

tetA (620 nm)

Probe 5’ – Biotin – GAAGAAGACCGCCAT – 3’

Target 3’ – CTTCTTCTGGCGGTAGTCCCGCCGCTGCTG – 5’

Reporter 5’ – CAGGGCGGCGACGAC – H6 – 3’

Blocking Strand 5’ – ACACACACACACACACACACACACA – 3’

10

3. Results and Discussion

3.1 UCNP Synthesis, Ligand Exchange and Immobilization

11

Figure 1: The strategy for the concurrent detection of uidA, Stx1A and tetA gene fragments. A paper substrate

was wax printed to prepare circular reaction zones with an inner diameter of 3 mm. Probe oligonucleotides were

conjugated to immobilized UCNPs in the reaction zones to capture the target sequences. In a sandwich assay

format, unhybridized portions of the target sequences captured reporter oligonucleotides that were conjugated to

QDs, resulting in localization of the QDs near the UCNPs for LRET.

NaYF4: 0.5% Tm3+, 30% Yb3+/NaYF4 core/shell UCNPs were synthesized using the seeded

growth method [29]. An inert shell was added to minimize quenching of dopant ions near the

surface of the core. The TEM image in Figure S1 shows monodisperse OA capped core UCNPs

that are hexagonal in shape, with an average diameter of 20.5 ± 5.6 nm based on 181

nanoparticles. Dynamic Light Scattering results shown in Figures S2 andS3 indicate a 2.3 ± 0.9

nm increase in diameter upon the addition an inert NaYF4 shell. The OA capped UCNPs were

made water soluble via ligand exchange with PEA. The TEM image in Figure S4 shows

monodisperse PEA-UCNPs. The phosphate groups of PEA coordinate strongly to the UCNP

surface leaving amine groups available for conjugation [30, 34].

The PEA capped UCNPs were covalently immobilized on aldehyde functionalized paper via

reductive amination with sodium cyanoborohydride in situ. We have previously reported the

reproducibility of this immobilization technique, which had a standard deviation based on

luminescence measurements of about 8% for 5 interaction zones [18]. Biotin-PEG4-NHS was

used to decorate the immobilized UCNPs for subsequent modification with avidin. The avidin

served to capture the biotinylated uidA, Stx1A and tetA probes. While the distance separation

between a donor NP and an acceptor is typically measured from the center of the NP, it is

measured from the surface of UCNPs due to the presence of lanthanides throughout the host

lattice. This minimizes the decrease in the LRET efficiency as the use of avidin increases the

separation distance between the donor and acceptor. The paper was then treated with a

concentrated solution of non-complementary DNA to minimize any subsequent non-specific

adsorption of oligonucleotides onto the paper and the UCNPs.

3.2 Characterization of QD-DNA Conjugates

12

Previous studies have shown that the hybridization kinetics of reporter nucleic acids that are

conjugated to QDs with target nucleic acids are independent of the number of immobilized

reporters on the QDs[18]. Herein, we incubate the QDs with only 5-fold molar excess of reporter

DNA. Hexahistidine functionalized uidA, Stx1A and tetA reporter DNA sequences were

conjugated to 525 nm, 575 nm and 620 nm GSH coated QDs, respectively. The QDs were then

incubated with excess hexahistidine-modified PEG to prevent non-specific adsorption of QDs on

the paper [18].

The gel image in Figure S6 compares the electrophoretic mobility of GSH QDs, PEG coated

QDs and DNA conjugated PEG stabilized QDs for 525, 575 and 620 nm QDs. The high

electrophoretic mobility of the GSH QDs is attributed to the carboxylate anions present on the

surface of the QDs. Neutral PEG molecules provide no mechanism for migration of QDs, while

the PEG stabilized QDs that are conjugated to reporter oligonucleotides show a slight

electrophoretic mobility due to negative charges associated with the phosphate backbone of

nucleic acids.

3.3 Enhancement in LRET Ratio in Dry Paper

Noor et. al reported up to a 10 fold enhancement in FRET ratio between green QD donors and

Cy3 acceptors in dry paper compared to hydrated paper [25]. The enhancement was attributed to

the contraction of wet paper upon drying that brings donors and acceptors in closer proximity.

Herein, dried paper offered a maximum 12-fold enhancement in LRET ratio between blue

emitting UCNP donors and green emitting QD acceptors, and an order of magnitude

improvement in the LOD for the single-plex nucleic acid hybridization assay (Figure 2). An

13

LOD of 12 fmol was obtained for dried paper compared to 140 fmol for hydrated paper. All

subsequent papers were dried before imaging.

Figure 2: Response curves for the detection of uidA target sequence in hydrated paper ( ■ ) and

dry paper ( ♦ ). For hydrated paper, y = 0.13x - 0.01, R2 = 0.99, Dynamic range (DR) was: 140

fmol – 3 pmol; For dry paper, y = 0.33x + 0.02, R2= 0.99, DR: 12 fmol – 1.5 pmol. The

secondary vertical axis shows the enhancement in LRET ratio( o ) for dried paper compared to

hydrated paper for different quantities of uidA targets.

3.4 Selection of QDs and Optical Channels for Multiplexing

The LRET-based hybridization assay used one donor and three acceptors. The spectral overlap

between the emission of blue emitting UCNPs and the absorbance of different color emitting

QDs is shown in Figure 3. Four independent optical channels were created using band pass filters

in an epi-fluorescence microscope, and were used to collect the luminescence from the UCNPs,

and the green, orange and red QDs. Filters with band pass of 420-490, 535-545, 565-585 and

615-625 nm were used to define the blue, green, orange and red optical channels, respectively.

Figure 4 shows an overlay of the normalized transmittance profiles of all the bandpass filters and

the normalized emission profiles of the QDs and UCNP.

14

Figure 3: Spectral overlap between the emission of NaYF4: 0.5% Tm3+, 30% Yb3+/NaYF4 core/shell UCNPs and absorbance of green, orange and red emitting QDs.

Figure 4: An overlay of UCNP emission (full blue) and green (full green), orange (full orange) and red (full red) QD emission. The dotted lines of respective colours show the transmittance of the band-pass filters used to collect luminescence from each NP.

15

The photomultiplier tube (PMT) gain was adjusted to improve contrast of the different optical

channels. For example, the gain in the orange channel was reduced to minimize background

signals from the green QDs, as seen in Figure S7. Figure 5 shows the LRET ratios obtained for

the independent detection of each target sequence with concentrations ranging from 30 fmol to 6

pmol. The LRET ratio is defined as the ratio of the signal of one colour QD to that of the UCNP

in the same spot. A definitive response curve in an optical channel was observed only when the

16

Figure 5: Response curves from the (a) green, (b) orange and (c) red optical channels for the

independent detection of uidA (i), Stx1A (ii) and tetA (iii) target sequence. For the selective

hybridization reactions: a(i) had a linear response described by y = 0.54x + 0.04, R2 = 0.99, DR: 10

fmol – 1.5 pmol; the b(ii) response was y = 0.27x + 0.04, R2= 0.99, DR: 50 fmol – 1.5 pmol; the

c(iii) response was y = 0.14x + 0.03, R2 = 0.99, DR: 65 fmol – 1.5 pmol

overhang of the DNA target was complementary to the reporter nucleic acids on the

corresponding QD. The reporter labelled QDs that did not correspond to the target were not

retained and the signals obtained in the respective optical channels were within the noise. The

limit of detection of uidA, Stx1A, and tetA gene fragments were 10 fmol, 50 fmol and 65 fmol,

respectively, calculated as three standard deviations above the LRET ratio in the presence of

non-complementary target and QD reporters. We note that the three target gene fragments used

to demonstrate the principles of operation of the detection strategy are relatively short

oligonucleotides. The question as to whether longer oligonucleotide targets can be quantitatively

determined by resonance energy transfer using hybridization of probes on nanoparticles was

previously investigated by our team, and functionality with sequences of lengths that are relevant

to PCR products (~150mer lengths) was achieved using quantum dots [35].

3.5 Multiplexed Hybridization Assays

Calibration curves were obtained for the simultaneous detection of equimolar amounts of uidA,

Stx1A and tetA targets. We previously reported the use of 2 µM QD-reporter conjugates for the

detection of HPRT1 target and demonstrated that the response curve is governed by the QD

concentration [18]. The limit of linearity was constrained by the concentration of QDs. Since our

previously reported single-plex assay showed an increase in LRET ratio with an increasing

concentration of QD reporter solution up to 2 µM, each QD reporter concentration herein was

kept at 0.5 µM, with a total concentration of 1.5 µM to ensure ample space for the QD reporter to

hybridize to target strands on the surface of UCNPs. The response curves obtained for each

target in the multiplexed assay were compared to their respective response curves in the single-

plex assay. Figure 6 shows a slight increase in the LRET ratio in the multiplexing format

compared to single-plex detection. The increase in the LRET response is due to the decrease in

17

UCNP luminescence in the presence of more QD acceptors in the multiplex format, as shown in

Figure S8. Limits of detection for uidA, Stx1A and tetA gene fragments were calculated to be 26

fmol, 62 fmol and 78 fmol, respectively. To avoid the change in LRET response in the multiplex

format, large UCNPs can be used to provide constant background reference emission as there

will always be a substantial amount of signal in the core of the UCNPs that cannot participate in

LRET. Figure 7 compares the kinetics of hybridization for each QD reporter in the single-plex

and the three-plex format showing no change in kinetics within one standard deviation. By

immobilizing a large number of DNA probes and maintaining a low QD concentration, no

change in kinetics was observed. Only 0.5 µM of each colour QD was used herein, whereas 8-10

µM of reporter is typically used in bioassays that rely on organic fluorophores as acceptors [17,

33].

18

19

Figure 6: Response curves for the independent (•) and mixture (▪) detection of (a) uidA, (b)

Stx1A, and (c) tetA in green, orange and red optical channels, respectively.

3.6 Hybridization Assays in a Complex Matrix:

The sensitivity of the triplex assay in 10% goat serum was minimally impacted by this more

complex matrix (Figure 8). Assays in 10% serum achieved limits of detection of 52, 56 and 76

fmol for uidA, Stx1A and tetA targets, respectively. This compares to limits of detection of 26,

62 and 78 fmol, respectively, in buffer. The differences in the limits of detection are within a

factor of 2 or less, and this similarity in performance is attributed to the use of NHS-PEG4-biotin

and random DNA as antifouling agents to passivate the surface of the immobilized UCNPs. The

selectivity of the assay was evaluated in 10% goat serum (Figure 9), and contrast ratios of the

20

Figure 7: Kinetic curves for the hybridization of (a) green, (b) orange and (c) red QD reporters, collected

for independent targets(•) and for targets in mixture (▪). Figure (d) shows an overlay of the kinetic curves

for all three reporters in concurrent detection, green QD (•), orange QD (◊) and red QD (▪).

LRET ratios for the FC uidA target to 1 BPM uidA target in the presence of FC Stx1A and tetA

targets were 3.5:1 and 40:1 with 0% and 5% formamide in the wash buffer, respectively. There

was no difference in the detection of FC Stx1A or tetA targets in the presence of FC or 1 BPM

uidA target. This selectivity is superior to a previously reported two-plex assay using UCNPs as

donors and molecular dyes as acceptor in a direct hybridization assay, where a contrast ratio of

only 2.16 was achieved with 20% formamide [13]. In a similar single-plex sandwhich format

hybridization assay, a contrast ratio of only 1.81 was achieved with 20% formamide in buffer.

The improvement in selectivity is attributed to the decoration of the QDs with PEG to minimize

non-specific adsorption on the surface of the paper[18].

Figure 8: Response curves for the simultaneous detection of (a) uidA, y = 0.33x + 0.02, R2 =

0.99, DR: 52 fmol – 3 pmol (b) Stx1A, y = 0.17x + 0.02, R2 = 0.99, DR: 56 fmol – 3 pmol and

(c) tetA, y = 0.08x + 0.02, R2 = 0.98, DR: 76 fmol – 3 pmol, in 10% goat serum.

21

Figure 9: Assays demonstrating the selectivity response.(i) 6 pmol FC uidA (dark) and 1 BPM

uidA (light);(ii) 6 pmol FC Stx1A,and (iii) 6 pmol FC tetA, in (a) 0% and (b) 5% formamide.

22

4. Conclusion:

We have demonstrated a LRET-based assay for the concurrent determination of three nucleic

acid targets using a single form of UCNP as donor and QDs as acceptors. All three of the

different color QDs had broad absorption profiles in the blue region of the spectrum and their

fluorescence was excited by blue emitting UCNPs. Both UCNPs and QDs have narrow and well

defined emission profiles, allowing for QD emission peaks to be resolved for multiplexing.

Independent optical channels were established using band pass filters to collect fluorescence

from green, orange and red emitting QDs, as well as the blue emission band of the UCNPs. The

assays had limits of detection of 26 fmol, 62 fmol and 78 fmol in buffer for the concurrent

detection of uidA, Stx1A and tetA targets, respectively. The limits of detection were comparable

in 10% goat serum. A contrast ratio derived from the signal for fully complementary FC uidA in

comparison to 1 BPM target was 40:1 when stringency was controlled using formamide,

demonstrating the selectivity of the assay. The triplexed assay demonstrated an order of

magnitude improvement in LOD and superior selectivity compared to the duplex detection of

DNA using UCNPs as donors and dyes as acceptors. This was achieved by coating the QDs with

PEG to annihilate non-specific adsorption on paper.

Supporting Information. Reagents, instrumentation, characterization of upconverting

nanoparticles and quantum dot-DNA conjugates, and additional results.

23

Acknowledgements:

We thank Dr. Sreekumari Nair for obtaining TEM images. We gratefully acknowledge the

Natural Sciences and Engineering Research Council of Canada (NSERC) for financial support of

this research. S.D. and U. U are thankful to NSERC for graduate fellowships.

References:

[1] M.V. DaCosta, S. Doughan, Y. Han, U.J. Krull, Lanthanide upconversion nanoparticles and applications in bioassays and bioimaging: A review, Anal. Chim. Acta, 832 (2014) 1-33.[2] M. Wang, G. Abbineni, A. Clevenger, C.B. Mao, S.K. Xu, Upconversion nanoparticles: synthesis, surface modification and biological applications, Nanomed. Nanotech. Biol. Med., 7 (2011) 710-729.[3] A. Sedlmeier, H.H. Gorris, Surface modification and characterization of photon-upconverting nanoparticles for bioanalytical applications, Chem. Soc. Rev., 44 (2015) 1526-1560.[4] H.H. Gorris, O.S. Wolfbeis, Photon-Upconverting Nanoparticles for Optical Encoding and Multiplexing of Cells, Biomolecules, and Microspheres, Angew. Chem. Int. Ed., 52 (2013) 3584-3600.[5] N. Duan, S.J. Wu, C.Q. Zhu, X.Y. Ma, Z.P. Wang, Y. Yu, Y. Jiang, Dual-color upconversion fluorescence and aptamer-functionalized magnetic nanoparticles-based bioassay for the simultaneous detection of Salmonella Typhimurium and Staphylococcus aureus, Anal. Chim. Acta, 723 (2012) 1-6.[6] Q.Q. Dou, N.M. Idris, Y. Zhang, Sandwich-structured upconversion nanoparticles with tunable color for multiplexed cell labeling, Biomaterials, 34 (2013) 1722-1731.[7] S.J. Wu, N. Duan, X.Y. Ma, Y. Xia, Y. Yu, Z.P. Wang, H.X. Wang, Simultaneous detection of enterovirus 71 and coxsackievirus A16 using dual-colour upconversion luminescent nanoparticles as labels, Chem. Comm., 48 (2012) 4866-4868.[8] M.Z. Zhang, W.J. Chen, X. Chen, Y.J. Zhang, X.J. Lin, Z.Y. Wu, M.F. Li, Multiplex Immunoassays of Plant Viruses Based on Functionalized Upconversion Nanoparticles Coupled with Immunomagnetic Separation, J. Nanomater., (2013) 317437.[9] O. Ehlert, R. Thomann, M. Darbandi, T. Nann, A four-color colloidal multiplexing nanoparticle system, ACS Nano, 2 (2008) 120-124.[10] S.J. Wu, N. Duan, Z. Shi, C.C. Fang, Z.P. Wang, Simultaneous Aptasensor for Multiplex Pathogenic Bacteria Detection Based on Multicolor Upconversion Nanoparticles Labels, Anal. Chem., 86 (2014) 3100-3107.[11] S.J. Wu, N. Duan, Z. Shi, C.C. Fang, Z.P. Wang, Dual fluorescence resonance energy transfer assay between tunable upconversion nanoparticles and controlled gold nanoparticles for the simultaneous detection of Pb2+ and Hg2+, Talanta, 128 (2014) 327-336.

24

[12] T. Rantanen, M.L. Jarvenpaa, J. Vuojola, R. Arppe, K. Kuningas, T. Soukka, Upconverting phosphors in a dual-parameter LRET-based hybridization assay, Analyst, 134 (2009) 1713-1716.[13] F. Zhou, U.J. Krull, Spectrally Matched Duplexed Nucleic Acid Bioassay Using Two-Colors from a Single Form of Upconversion Nanoparticle, Anal. Chem., 86 (2014) 10932-10939.[14] M.O. Noor, E. Petryayeva, A.J. Tavares, U. Uddayasankar, W.R. Algar, U.J. Krull, Building from the "Ground" Up: Developing interfacial chemistry for solid-phase nucleic acid hybridization assays based on quantum dots and fluorescence resonance energy transfer, Coord. Chem. Rev., 263 (2014) 25-52.[15] M.Y. He, Z.H. Liu, Paper-Based Microfluidic Device with Upconversion Fluorescence Assay, Anal. Chem., 85 (2013) 11691-11694.[16] M.Y. He, Z. Li, Y.Y. Ge, Z.H. Liu, Portable Upconversion Nanoparticles-Based Paper Device for Field Testing of Drug Abuse, Anal. Chem., 88 (2016) 1530-1534.[17] F. Zhou, M.O. Noor, U.J. Krull, A Paper-Based Sandwich Format Hybridization Assay for Unlabeled Nucleic Acid Detection Using Upconversion Nanoparticles as Energy Donors in Luminescence Resonance Energy Transfer, Nanomaterials, 5 (2015) 1556-1570.[18] S. Doughan, U. Uddayasankar, U.J. Krull, A paper-based resonance energy transfer nucleic acid hybridization assay using upconversion nanoparticles as donors and quantum dots as acceptors, Anal. Chim. Acta, 878 (2015) 1-8.[19] W.R. Algar, U.J. Krull, Developing Mixed Films of Immobilized Oligonucleotides and Quantum Dots for the Multiplexed Detection of Nucleic Acid Hybridization Using a Combination of Fluorescence Resonance Energy Transfer and Direct Excitation of Fluorescence, Langmuir, 26 (2010) 6041-6047.[20] W.R. Algar, U.J. Krull, FRET-Based Solid-Phase Three-Color and Four Color Hybridization Assays Using Mixed Films and Quantum Dots and Oligonucleotides, Mater. Res. Soc. Symp. P., (2010) 45-55.[21] L. Mattsson, K.D. Wegner, N. Hildebrandt, T. Soukka, Upconverting nanoparticle to quantum dot FRET for homogeneous double-nano biosensors, RSC Adv., 5 (2015) 13270-13277.[22] A. Bednarkiewicz, M. Nyk, M. Samoc, W. Strek, Up-conversion FRET from Er3+/Yb3+:NaYF4 Nanophosphor to CdSe Quantum Dots, J. Phys. Chem. C, 114 (2010) 17535-17541.[23] L.J. Charbonniere, N. Hildebrandt, R.F. Ziessel, H.G. Loehmannsroeben, Lanthanides to quantum dots resonance energy transfer in time-resolved fluoro-immunoassays and luminescence microscopy, J. Am. Chem. Soc., 128 (2006) 12800-12809.[24] D. Geissler, S. Linden, K. Liermann, K.D. Wegner, L.J. Charbonniere, N. Hildebrandt, Lanthanides and Quantum Dots as Forster Resonance Energy Transfer Agents for Diagnostics and Cellular Imaging, Inorg. Chem., 53 (2014) 1824-1838.[25] M.O. Noor, U.J. Krull, Camera-Based Ratiometric Fluorescence Transduction of Nucleic Acid Hybridization with Reagentless Signal Amplification on a Paper-Based Platform Using Immobilized Quantum Dots as Donors, Anal. Chem., 86 (2014) 10331-10339.[26] J.L. Alonso, K. Soriano, I. Amoros, M.A. Ferrus, Quantitative determination of E-coli and fecal coliforms in water using a chromogenic medium, J. Environ. Sci. Health A Tox. Hazard Subst. Environ. Eng., 33 (1998) 1229-1248.[27] M.D. Gray, K.A. Lampel, N.A. Strockbine, R.E. Fernandez, A.R. Melton-Celsa, A.T. Maurelli, Clinical Isolates of Shiga Toxin 1a-Producing Shigella flexneri with an Epidemiological Link to Recent Travel to Hispaniola, Emerg. Infect. Dis., 20 (2014) 1669-1677.

25

[28] W. Hillen, C. Berens, Mechanisms Underlying Expression Of Tn10 Encoded Tetracycline Resistance, Annu. Rev. Microbiol., 48 (1994) 345-369.[29] H.S. Qian, Y. Zhang, Synthesis of Hexagonal-Phase Core-Shell NaYF4 Nanocrystals with Tunable Upconversion Fluorescence, Langmuir, 24 (2008) 12123-12125.[30] S. Doughan, Y. Han, U. Uddayasankar, U.J. Krull, Solid-Phase Covalent Immobilization of Upconverting Nanoparticles for Biosensing by Luminescence Resonance Energy Transfer, ACS Appl. Mater. Interfaces, 6 (2014) 14061-14068.[31] I.L. Medintz, L. Berti, T. Pons, A.F. Grimes, D.S. English, A. Alessandrini, P. Facci, H. Mattoussi, A reactive peptidic linker for self-assembling hybrid quantum dot-DNA bioconjugates, Nano Lett., 7 (2007) 1741-1748.[32] U. Uddayasankar, Z.F. Zhang, R.T. Shergill, C.C. Gradinaru, U.J. Krull, Isolation of Monovalent Quantum Dot-Nucleic Acid Conjugates Using Magnetic Beads, Bioconjug. Chem., 25 (2014) 1342-1350.[33] M.O. Noor, U.J. Krull, Paper-Based Solid-Phase Multiplexed Nucleic Acid Hybridization Assay with Tunable Dynamic Range Using Immobilized Quantum Dots As Donors in Fluorescence Resonance Energy Transfer, Anal. Chem., 85 (2013) 7502-7511.[34] J.C. Boyer, M.P. Manseau, J.I. Murray, F. van Veggel, Surface Modification of Upconverting NaYF4 Nanoparticles with PEG-Phosphate Ligands for NIR (800 nm) Biolabeling within the Biological Window, Langmuir, 26 (2010) 1157-1164.[35] M.O. Noor, D. Hrovat, M. Moazami-Goudarzi, G.S. Espie and U.J. Krull, Ratiometric fluorescence transduction by hybridization after isothermal amplification for determination of zeptomole quantities of oligonucleotide biomarkers with a paper-based platform and camera-based detection, Anal. Chim. Acta, 885 (2015) 156–165.

26

Figure Captions:

Figure 1: The strategy for the concurrent detection of uidA, Stx1A and tetA gene fragments. A

paper substrate was wax printed to prepare circular reaction zones with an inner diameter of 3

mm. Probe oligonucleotides were conjugated to immobilized UCNPs in the reaction zones to

capture the target sequences. In a sandwich assay format, unhybridized portions of the target

sequences captured reporter oligonucleotides that were conjugated to QDs, resulting in

localization of the QDs near the UCNPs for LRET.

Figure 2: Response curves for the detection of uidA target sequence in hydrated paper ( ■ ) and

dry paper ( ♦ ). For hydrated paper, y = 0.13x - 0.01, R2 = 0.99, Dynamic range (DR) was: 140

fmol – 3 pmol; For dry paper, y = 0.33x + 0.02, R2= 0.99, DR: 12 fmol – 1.5 pmol. The

secondary vertical axis shows the enhancement in LRET ratio( o ) for dried paper compared to

hydrated paper for different quantities of uidA targets.

Figure 3: Spectral overlap between the emission of NaYF4: 0.5% Tm3+, 30% Yb3+/NaYF4

core/shell UCNPs and absorbance of green, orange and red emitting QDs.

Figure 4: An overlay of UCNP emission (full blue) and green (full green), orange (full orange)

and red (full red) QD emission. The dotted lines of respective colours show the transmittance of

the band-pass filters used to collect luminescence from each NP.

Figure 5: Response curves from the (a) green, (b) orange and (c) red optical channels for the

independent detection of uidA (i), Stx1A (ii) and tetA (iii) target sequence. For the selective

hybridization reactions:a(i) had a linear response described by y = 0.54x + 0.04, R2 = 0.99, DR:

10 fmol – 1.5 pmol; the b(ii)response was y = 0.27x + 0.04, R2= 0.99, DR: 50 fmol – 1.5 pmol;

the c(iii) response was y = 0.14x + 0.03, R2 = 0.99, DR: 65 fmol – 1.5 pmol

Figure 6: Response curves for the independent (•) and mixture (▪) detection of (a) uidA, (b)

Stx1A, and (c) tetA in green, orange and red optical channels, respectively.

27

Figure 7: Kinetic curves for the hybridization of (a) green, (b) orange and (c) red QD reporters,

collected for independent targets(•) and for targets in mixture (▪). Figure (d) shows an overlay of

the kinetic curves for all three reporters in concurrent detection, green QD (•), orange QD (◊) and

red QD (▪).

Figure 8: Response curves for the simultaneous detection of (a) uidA, y = 0.33x + 0.02, R2 =

0.99, DR: 52 fmol – 3 pmol (b) Stx1A, y = 0.17x + 0.02, R2 = 0.99, DR: 56 fmol – 3 pmol and

(c) tetA, y = 0.08x + 0.02, R2 = 0.98, DR: 76 fmol – 3 pmol, in 10% goat serum.

Figure 9: Assays demonstrating the selectivity response. (i) 6 pmol FC uidA (dark) and 1 BPM

uidA (light); (ii) 6 pmol FC Stx1A,and (iii) 6 pmol FC tetA, in (a) 0% and (b) 5% formamide.

28