Integration of Multiplexed Microfluidic Electrokinetic Concentrators

Feb 25, 2016 - concentrator, we could achieve a maximum concentration factor of ∼800 ... is that the fabrication of the microfluidic concentrator is...
0 downloads 0 Views 2MB Size
Subscriber access provided by MONASH UNIVERSITY

Article

Integration of Multiplexed Microfluidic Electrokinetic Concentrators with a Morpholino Microarray via Reversible Surface Bonding for Enhanced DNA Hybridization Diogo Martins, Xi Wei, Rastislav Levicky, and Yong-Ak Rafael Song Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.5b03875 • Publication Date (Web): 25 Feb 2016 Downloaded from http://pubs.acs.org on February 25, 2016

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 15

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

Analytical Chemistry

Integration of Multiplexed Microfluidic Electrokinetic Concentrators with a Morpholino Microarray via Reversible Surface Bonding for Enhanced DNA Hybridization Diogo Martins§, Xi Wei§,†, Rastislav Levicky†, and Yong-Ak Song§, † §

Division of Engineering, New York University Abu Dhabi, P.O. Box 129188, Abu Dhabi, United Arab Emirates Department of Chemical and Biomolecular Engineering, New York University Tandon School of Engineering, Brooklyn, New York 11201, United States



ABSTRACT: We describe a microfluidic concentration device to accelerate the surface hybridization reaction between DNA and morpholinos (MOs) for enhanced detection. The microfluidic concentrator comprises a single polydimethylsiloxane (PDMS) microchannel onto which an ion-selective layer of conductive polymer poly(3,4ethylenedioxythiophene)-poly(styrenesulfonate) (PEDOT:PSS) was directly printed, and then reversibly surface bonded onto a morpholino microarray for hybridization. Using this electrokinetic trapping concentrator, we could achieve a maximum concentration factor of ~800 for DNA and a limit of detection of 10 nM within 15 min. In terms of the detection speed, it enabled faster hybridization by around 10-fold when compared to conventional diffusionbased hybridization. A significant advantage of our approach is that the fabrication of the microfluidic concentrator is completely decoupled from the microarray; by eliminating the need to deposit an ion-selective layer on the microarray surface prior to device integration, interfacing between both modules, the PDMS chip for electrokinetic concentration and the substrate for DNA sensing, is easier and applicable to any microarray platform. Furthermore, this fabrication strategy facilitates a multiplexing of concentrators. We have demonstrated the proofof-concept for multiplexing by building a device with 5 parallel concentrators connected to a single inlet/outlet and applying it to parallel concentration and hybridization. Such device yielded similar concentration and hybridization efficiency compared to that of a single-channel device without adding any complexity to the fabrication and setup. These results demonstrate that our concentrator concept can be applied to the development of a highly multiplexed concentrator-enhanced microarray detection system for either genetic analysis or other diagnostic assays.

ACS Paragon Plus Environment

1

Analytical Chemistry

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

Page 2 of 15

INTRODUCTION Planar nucleic acid microarrays are a powerful 1 and indispensable tool in genomic research. These methods use highly multiplexed solid-phase hybridization to perform massively parallel screening 2 of DNA and RNA sequences in a single experiment. Despite the high-throughput supported by conventional microarray platforms, microarray experiments nevertheless require rather long 3 incubation times. Often, overnight hybridization (~ 20 h) is required to accumulate hybridized DNA on the microarray. An important challenge in this regard is realization of sufficient concentration of scarce or unamplified samples, so that adequate mass transport of the analyte DNA to the binding site can 4 be provided even from minute starting amounts. This constraint greatly limits the applicability of microarrays to point-of-care diagnostics where short detection times and simplified workflow are paramount. To overcome mass transport limitations, one possibility is to use microfluidics and electrokinetic phenomena to concentrate the biomolecules of interest into a smaller volume so that they can be 5 efficiently captured. There are many techniques available for sample preconcentration in microfluidic devices; the most reported and promising for 6–15 biosensing being isotachophoresis (ITP) and 16–19 electrokinetic trapping (ET) . Both techniques are particularly interesting since they are not limited to DNA, but can be used with any charged 20–22 and usually don’t require any flow molecules control elements or moving parts. In ITP, a special arrangement of buffers is required combined with a sequence of steps that use electric fields to extract molecules from the sample reservoir and concentrate them and then flow them through a microchannel for detection. Conversely, in ET, a single low ionic strength buffer with a DC voltage is needed and the plug remains approximately in the same position, with its concentration increasing continuously. ITP can, however, be performed using higher ionic strength buffers which is important to preserve the conformational integrity and function of most biomolecules, and to allow the right intermolecular interactions to form for biosensing purposes. Santiago’s group has pioneered on-chip ITPbased assays for rapid and high sensitivity nucleic 6–8,11,13 acid detection in free-solution. They reported the maximum increase of hybridization rate with ITP 6 (14,000 fold) for homogenous DNA/DNA reaction.

In addition, using a bead-based assay and ITP, they achieved a reduction in bead hybridization reaction 15 time (60-fold) while also improving their sensitivity. Acceleration of surface-based hybridization by ITP in a polydimethylsiloxane (PDMS) microchannel reversibly bonded to a glass slide has also been 10 The ITP-enhanced microarrays demonstrated. allowed hybridization at much higher rate (30 min versus 20 h) and with higher sensitivity (~ 8.2 fold) than conventional methods. Moreover, an ITPenhanced nucleic acid hybridization assay has been demonstrated by Karsenty et al. in just 3 min. with a 2 orders of magnitude improvement in limit of detection (LOD) using pre-labeled paramagnetic beads with desired capture probes as the biosensing 12 element. As for ET-based approaches, microfluidic preconcentration systems have been used to enhance both the reaction rate and detection 16,23–28 sensitivity of immunoassays as well as in the 29–31 standard enzymatic assays. Lee et al. reported concentration enhanced cell kinase assay in a micro/nanofluidic platform directly from cell lysates and demonstrated a 25-fold increase in reaction 29 velocity and 65-fold enhancement in sensitivity. Even though the device didn’t use unprocessed biological samples like blood, it was capable of handling a complex physiological sample. However, it required sample dilution prior to the concentration step. Acceleration of surface-based reactions using ET has been mostly targeted at surface-based immunoassays rather than genetic analysis since polymerase chain reaction (PCR) amplification can be used instead to concentrate nucleic acids. Nevertheless, PCR has its shortcomings that can be overcome with ET concentration including circumventing the need for identifying efficient PCR primers, avoiding artifacts that can arise from the enzymatic PCR reaction itself, and eliminating the associated materials and time costs. Therefore, ET biomolecular concentrators might be an interesting alternative to PCR in low-resource settings without trained personnel and equipment. The obstacle to using ET in DNA analysis is mismatch between the optimum buffer ionic strength for concentration using cation-selective membrane (~ 10 mM) and that favorable to DNA-DNA hybridization (0.1 - 1 M). However, by selecting morpholino (MO) as a synthetic DNA analogue capture probe with a neutral backbone, DNA hybridization can be readily performed in the low ionic strength buffers that are also optimally

ACS Paragon Plus Environment

2

Page 3 of 15

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

Analytical Chemistry

compatible with ET-enhanced MO-DNA hybridization 32 assays, as recently reported. This uniquely synergistic approach was enabled by the fact that the charge-neutral MOs are capable of efficiently capturing DNA analyte even at very low ionic 33–35 As a result, the hybridization rate of strengths. DNA target to MO probe molecules was increased by almost two orders of magnitude, enabling the detection time to be reduced from more than 24 h to 15 min. This enhancement in hybridization speed was achieved through a microprinting of poly(3,4ethylenedioxythiophene)-poly(styrenesulfonate) 36 (PEDOT:PSS) as a cation-selective membrane next to the MO capture probes on the same microarray substrate and subsequent sealing with a PDMS microchannel. A voltage difference applied across the channel initiated the formation of a DNA plug in front of the PEDOT:PSS membrane. The formation of a highly concentrated DNA plug resulted from the cation selectivity of PEDOT:PSS, which initiated the formation of an ion depletion region, in combination with an electroosmotic flow that transports biomolecules from the anodic 25,37 reservoir towards the membrane. In this way, the local concentration of DNA target was increased by three orders of magnitude directly at the site of the capture probes. Despite Nafion being the common choice of 16,19,23,25– material for cation-selective membranes, 27,29–31,37–41 the micro-flow patterning technique used to print it on a glass substrate, although simple, has 37 the following limitations: it is not reproducible; 27 leads to variable membrane thicknesses; and allows for “hidden” nanochannels along the membrane edges due to an imperfect PDMS sealing 37 that leads to leakage holes and dominates the 42 ionic transport, compromising the ionic selectivity of the membrane. To bypass such limitations, a different printing method and conductive polymer were used. Unlike Nafion resin, PEDOT:PSS dispersion in H2O has low viscosity [10-30 cP (20 °C)] and consequently can be directly deposited on a very confined area (with dimensions slightly smaller than the channel width) of the substrate 43,44 using an ultrasonically driven micropipette . In addition, PEDOT:PSS is compatible with cleanroom 45 processing (e.g. spin-coating, photolithography, and etching processes) and as such could potentially allow the simultaneous fabrication of a high-density array of concentrators on a substrate. Based on our previous work, a new design of PDMS/PEDOT:PSS microfluidic concentrator has been developed and applied to the DNA-MO hybridization assay. In this new concentrator, a cation-selective PEDOT:PSS solution is deposited directly onto a PDMS microchannel rather than on a

43

planar glass substrate using a microplotter . This approach is in contrast to most ion-selective membrane-based concentrators previously 23,25,37,46 published in which membranes were deposited on the glass substrate rather than on a PDMS microchannel. It enables on-chip electrokinetic concentration to be performed without any modification to the microarray surface. Therefore, direct interfacing of the concentrator chip to any microarray platform is facile and rapid. In addition, any irregularities in the size of the printed PEDOT:PSS layer may easily compromise the stability of the concentration plug due to incomplete sealing between the PEDOT:PSS membrane and the PDMS chip and resultant side leakages during concentration. By decoupling the electrokinetic concentrator from the microarray entirely, as proposed in this work, this sealing problem can be avoided so that the interfacing between the microfluidic concentrator and microarray is more robust and leakage-free. Lastly, since both the PDMS microchannnel fabrication and the PEDOT:PSS printing on PDMS chip are scalable, a high-density array of concentrators can be built and used in any surfacebased assay, conceptually similar to the device 23 proposed by Ko et al. This scalability is supported by high reproducibility of the ion-selective layer deposition through the spatial confinement the PDMS channel provides to the printed PEDOT:PSS solution. The current work thus highlights the potential of our microfluidic concentrator for highly multiplexed, concentrator-enhanced detection of nucleic acids in combination with MO microarrays.

EXPERIMENTAL SECTION Materials. The unlabeled MO probe sequence (5’ NH2-GTA GCT AAT GAT GTG GCA TCG GTT 3’) was purchased from Gene Tools LLC. The 24mer probes were modified with an amino group at the 5’ end to allow for surface attachment to aldehyde groups on Superaldehyde 2 microarray slides from Arrayit. Cyanine 5 (Cy5, λexc = 649 nm and λem = 670 nm) labeled complementary DNA targets (5’ CAA CCG ATG CCA CAT CAT TAG CTA C-Cy5 3’) were purchased from Integrated DNA Technologies. The 25 nucleotides long DNA targets were labeled at the 3’-end with Cy5 to enable fluorescence detection. All hybridization solutions were prepared in 0.1X PBS at pH 7.1 with no other additives. Conductive polymer PEDOT:PSS 2.2–2.6% in H2O (high conductivity grade) was obtained from SigmaAldrich. Phosphate buffer saline (PBS) solution at

ACS Paragon Plus Environment

3

Analytical Chemistry

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

0.1X pH 7.1 was prepared by diluting 100 times the Dulbecco's 10X PBS (composed of 26.7 mM KCl, 14.6 mM KH2PO4, 1379 mM NaCl, and 80.6 mM Na2HPO4-7H2O) purchased from Gibco. The pH after dilution with deionized (DI) water (18 MΩ.cm) was adjusted by adding 1 M NaOH The 0.1 M sodium phosphate buffer (PB) of pH 7.0 contained 39 mM H2NaO4P-2H2O and 61 mM HNa2O4P, and 0.1 M PB pH 8.0 contained 6 mM H2NaO4P-2H2O and 94 mM HNa2O4P. Sodium borohydride (NaBH4) and nonionic surfactant Tween 20 were purchased from Merck and Sigma-Aldrich, respectively. PDMS prepolymer set (Sylgard 184) was obtained from Dow Corning Inc.

Device fabrication. The microfluidic device is composed of two parts: a MO microarray and a reversibly surface-bonded PDMS/PEDOT:PSS microfluidic concentrator [Figure 1 (a)]. To fabricate -1 the MO microarray, MO solution of 40 µmol L in 0.1M PB at pH 8.0 was printed using a fluid GIX microplotter II (Sonoplot Inc.) and a micro glass tip with a diameter of 50 µm (Sonoplot Inc.). The array 2 covered an area of 7×2 mm and the MO spots had a diameter of 50 µm at a spacing distance of 50 µm. After printing, slides were stored in a desiccator at 23 °C for 24 h. Next, unbound MO probes were 33 washed from the slide using the following protocol. First, we concomitantly deactivated unreacted aldehyde groups to hydroxyl groups and reduced the labile Schiff base bonds between probes and the slide by washing slides in 100 mL of freshly prepared NaBH4 solution (280 mg of NaBH4, 24mL of absolute ethanol, and 76 mL of PB 0.1M pH 7.0) for 5 min. The slides were then rinsed with DI water for 2 min. Next, a surfactant wash with 0.05% (w/v) Tween 20 solution for 5 min. was used to remove physisorbed probes, followed by a second 2 min. DI water rinse. All washing steps were performed under gentle mixing (60 rpm) on an orbital shaker (model PSU-10i, from Grantbio). Lastly, the slides were dried under a nitrogen stream. The microchannels (~17 µm high, 200 µm wide, and 1 cm long) were made in PDMS using methods 47 described elsewhere. After punching both the inlet and outlet holes (circular holes of 4 mm in diameter and square holes of 7 mm on each side for singleand multi-channel devices, respectively), the PDMS microchannel was reversibly sealed against a blank glass slide. Then, using a handheld corona treater (BD-20AC, Electro-Technic Products Inc.), a corona plasma (10 sec. in duration) was applied to the inner PDMS walls of the microchannel, rendering them 48 hydrophilic. A ~ 3 µm thick PEDOT:PSS membrane layer [Figure 1 c)-(d)] was then directly

Page 4 of 15

printed into a circular pattern at the center of the PDMS microchannel [Figure 1 (a), step 1]. By confining the hydrophilicity to the microchannel, a droplet of liquid PEDOT:PSS spread inside the circular pattern and remained confined without flowing over the edges of the circular pattern or into the microchannel due to surface tension [see Video S-1 in supporting information]. It is important that the conductive polymer conforms to the edges of the circular pattern that gives a shape to the ionselective layer. Otherwise, the reversible sealing around the PEDOT:PSS layer might be compromised due to the overflow of PEDOT:PSS, leading to leakage along the edges of the membrane during ion-polarization concentration (ICP). After printing of the PEDOT:PSS, the ion-selective resin layer was allowed to dry at room temperature for 10 min. Finally, the PDMS microfluidic concentrator was sealed against a MO microarray substrate by reversible bonding [Figure 1 (a), step 2]. The devices were stored at room temperature and under vacuum to ensure spontaneous filling (even a few days after the assembly) of aqueous solutions, thereby eliminating the need for any further plasma 49 treatment.

Chip operation. The microfluidic chip consisted either of a single PDMS microchannel or a set of parallel microchannels [Figure 1 (b)], with a PEDOT:PSS membrane printed on each channel separately, and reversibly sealed against an array of MO probes on a glass slide. To test the operation of the device in hybridization assays, the cathodic reservoir was loaded with 100 µL of 0.1X PBS buffer to fill the microchannel up to the edge of the opposite reservoir by capillarity; and 50 µL of DNA target solution was dispensed into the opposite (anodic) reservoir. After equilibrating the conductive polymer layer for 20 min, 75 V was applied along the microchannel through the Pt wires (0.5 mm in diameter, Goodfellow) in each reservoir, using a DC source meter (Keithley 2400). The source meter was controlled by a LabVIEW program (National Instruments). The voltage difference initiated the formation of a DNA plug in front of the PEDOT:PSS membrane [Figure 1 (a), step 3]. During a concentration period of 15 min., the DNA target molecules in the plug hybridized with the immobilized MO probes located at the bottom of the microchannel. At the end of the surface hybridization assay, the voltage was switched off and the plug was quickly washed away from the detection site by the pressure-driven flow

ACS Paragon Plus Environment

4

Page 5 of 15

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

Analytical Chemistry

created by the column height difference between the reservoirs.

Data acquisition and analysis. Concentration and hybridization experiments were monitored using an inverted epi-fluorescence microscope (Nikon TiEclipse) equipped with a mercury lamp (Nikon Intensilight C-HGFIE) and a digital camera (Andor Clara). Cy5 filter was used to observe fluorescence emission (λem=670 nm) of the DNA sample. The exposure time used for capture of the fluorescence signal was 2 s. During time-lapse imaging of hybridization experiments, an electronic shutter (Lambda SC, from Sutter Instrument) was used to minimize photobleaching and synchronized with the camera exposure so that images were captured in 5 s. intervals. Since highly concentrated plugs of fluorescent DNA often saturated the CCD camera during time-lapse acquisitions, a set of neutral density filters was used when required. Digital analysis of the fluorescent data acquired with Nikon Ti-Eclipse was performed using NIH ImageJ. We used fluorescent-tagged DNA molecules (without prior amplification via PCR) to quantify the local concentration enhancement of the DNA sample by measuring the increase of its fluorescence intensity after electrokinetic concentration and comparing this value to the fluorescence signal intensities of some known DNA concentrations measured for the calibration curve (such as 0.1 µM, 1 µM, 10 µM, and 100 µM). Using the fluorescent-labeled DNA, we can detect and quantify the hybridization results between DNA and MO in the same way by using the fluorescence

detection that is the most common detection method for commercial microarrays. Hybridization intensity of a MO–DNA spot was determined by measuring the average intensity of the pixels belonging to the spot and then subtracting the background signal given by the average intensity of the pixels around that same spot. Only those MO–DNA spots with uniform fluorescence signal intensity and located directly under the concentration plug were considered for quantitative analysis. The intensities measured from these spots were combined to provide averaged values per device. Final average intensities and respective standard deviations are representative of at least three independent measurements. To obtain a 3D profile of the DNA concentration plug, each plug was scanned through the entire height of the microchannel and its fluorescence intensity measured using an inverted confocal microscope (Olympus FLUOVIEW FV1000). The multi-focal depth z-stack images obtained were used to reconstruct the vertical concentration profile of the DNA plug during electrokinetic concentration. For all reported experiments, at least three independent measurements were taken under the same test condition to ensure reproducibility and statistical significance of the experiments. With the exception of MO-DNA hybridization experiments, all measurements were conducted with devices having deactivated superaldehyde glass slides without MO as substrates. The washing steps mentioned in the device fabrication section were performed on superaldehyde glass slides.

ACS Paragon Plus Environment

5

Analytical Chemistry

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

Page 6 of 15

Figure 1. Fabrication of a multiplexed microfluidic concentrator chip in PDMS. (a) 3D schematic diagram of the fabrication of the microfluidic concentrator by PEDOT:PSS printing on each PDMS microchannel (step 1), subsequent integration with a MO microarray by reversible sealing (step 2), and electrokinetically enhanced hybridization by applying a constant voltage between the “DNA sample” and “Buffer” reservoirs (step 3). b) Optical image of a multiplexed PDMS device consisting of 5 parallel microchannels (200 µm wide, 17 µm high, and 10 mm long), all connected to a single inlet and outlet, and a PEDOT:PSS layer deposited in the center of each channel. c) SEM image of the circular center area of a microchannel, showing a PEDOT:PSS membrane (in false color) contained in a circular pattern of the microchannel. (d) Profile of the PEDOT:PSS membrane along the centerline of the microchannel between point A and B. The printed membranes were typically ~2.8 µm thick and had a diameter of ~400 µm.

ACS Paragon Plus Environment

6

Page 7 of 15

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

Analytical Chemistry

RESULTS AND DISCUSSION Single-channel preconcentration. Using a single channel concentrator device, we conducted experiments with a fluorescently labeled DNA sample to characterize its performance for various starting DNA concentrations under a constant voltage of 75 V. The devices for these experiments used deactivated superaldehyde glass slides, without a printed MO array. When the DC voltage was applied to the channel, the DNA molecules in the anodic reservoir were driven by electroosmotic flow (EOF) towards the PEDOT:PSS membrane. Since PEDOT:PSS is ion-selective, an ion-depletion zone was formed next to the membrane; as a result, the DNA molecules started accumulating in front of the membrane. As shown in Figure 2 (a), the DNA plug remained stable during the first 15 min. of concentration, displaying only a slight increase of the ion depletion zone with time. Due to a continuous supply and accumulation of DNA molecules, the size of the plug kept increasing from 0 min. to 15 min.. Figure 2 (b) depicts an increase in fluorescence signal intensity of the DNA plug as a measure of the DNA concentration obtained from three different initial DNA concentrations ranging from 1 to 100 nM. The data show that in all cases the concentration increased rapidly by more than two orders of magnitude in the first minute and then stabilized after about 5 min. The DNA concentration reached a stable value of ~0.3 µM (300-fold increase), ~4 µM (i.e. 400-fold increase), and 80 µM (800-fold increase) for 1 nM, 10 nM, and 100 nM, respectively. As a comparison, the concentration factor was ~2times less than the one obtained with the PEDOT:PSS layer printed on the substrate (termed as PEDOT:PSS-glass device), especially at 1 and 10 nM, which indicated that the concentrator was less effective at low initial DNA concentrations when the ion-selective membrane was printed on the PDMS channel directly (termed as PEDOT:PSSPDMS device). The supplied electrical current was monitored during all experiments. A stable value of ~3.6 µA was recorded, with the only exception occurring during the first few minutes when the current dropped close to ~ 1 µA.

Figure 2. Characterization of the single-channel concentration experiment. (a) Time-lapse fluorescence images of the single-channel concentration experiment showing the DNA plug at different times adjacent to the membrane. (b) Typical fluorescence intensity curves of DNA plugs for different initial concentrations (1, 10, 100 nM), at a constant voltage of 75 V. Fluorescence signal intensity data obtained with devices with PEDOT:PSS-PDMS and PEDOT:PSS-glass are plotted for direct comparison. Dotted lines represent reference fluorescence signal intensities from the following DNA concentrations: 0.1, 1, 10 and 100 µM.

ACS Paragon Plus Environment

7

Analytical Chemistry

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

Profiling the concentration distribution in zdirection. Given the possibility that a DNA concentration plug may not be uniformly distributed along the channel depth, along the z direction, we investigated its three-dimensional concentration profile with confocal microscopy. If the plug does not extend across the depth of the channel fully to the glass slide, it would not be as effective in enhancing surface hybridization, As shown schematically in Figure 3 (a), the entire height of the microchannel was scanned while electrokinetic concentration was in progress, and the fluorescence intensity was measured to determine DNA concentration as a function of the z-direction. The results show that the concentration distribution was not uniform along the z-direction, as illustrated in Figure 3 (b) (see also the 3D image of the DNA concentration plug in Figure S1 and corresponding Video S-2 in supporting information). As measured quantitatively in Figure 3 (c), the fluorescence signal intensity of the DNA plug was maximal near the glass surface and decreased almost linearly as the focal plane moved upwards to the top of the channel, where the fluorescence signal intensity value decreased to about one-fourth of the maximal value. A similar experiment has been reported 37 previously. A PDMS microchannel with similar dimensions was sealed against a glass substrate where a cation-selective Nafion membrane with the same thickness as the PEDOT:PSS layer had been deposited by a micro-flow patterning technique, and used to concentrate FITC. As opposed to our concentration results with DNA, that study found the FITC molecules to concentrate better at the top of the PDMS microchannel rather than at the bottom. Moreover, repeating our confocal microscopy measurement for a PEDOT:PSS-glass concentrator found that the DNA plug concentration profile remained unchanged irrespective of the membrane position (see Figure S-2 and Video S-3 in supporting information). Interestingly, both cases exhibited an almost identical spatial distribution of the fluorescence signal intensities in the z-direction. We hypothesize that the reason for this concentration distribution could be linked to the different electroosmotic mobilities of glass and PDMS. The electroosmostic speed νeo near the double layer region for a homogeneous surface is given by:      

| |



(1)



where µeo is the electroosmotic mobility, E electric field strength, ε the dielectric constant of solution, ζ the zeta potential, η the viscosity, V applied voltage, and L the distance between

the the the the

Page 8 of 15

50

electrodes. Glass is more negatively charged (i.e. higher zeta potential) than native PDMS, and so has -4 2 a higher electroosmotic mobility (µeo= 4×10 cm /Vs -4 2 for glass substrate and µeo= 1×10 cm /Vs for native 51 PDMS); therefore, when a voltage difference is applied along the microchannel the fluid velocity profile is not uniform, but rather is higher near the glass substrate than next to the PDMS surface. The approximately four times higher electroosmotic flow velocity near the glass surface means that more DNA molecules per unit time reach the ion depletion region at the bottom of the channel compared to the top. As a result, DNA molecules accumulate more rapidly at the bottom of the channel adjacent to the ion depletion region. The four times difference in νeo between the top and bottom surfaces could explain the one-fourth drop in fluorescence intensity from the bottom to the top of the channel seen in Figure 3 (c).

Figure 3. Confocal measurement of the DNA concentration plug along the channel depth in the zdirection. (a) Schematic side view of an electrokinetic DNA concentration experiment. (b) Reconstruction of the DNA plug profile from the confocal images of fluorescence intensities of Cy5-

ACS Paragon Plus Environment

8

Page 9 of 15

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

Analytical Chemistry

labeled DNA in 0.1X PBS after 15 min. of concentration at a voltage of 75 V (note: for clarity, the z-axis is magnified by 10× relative to the y-axis). c) Fluorescence intensities of the DNA plug along the z-direction [dashed-line connecting A and B in (b)] for 100 nM DNA. The fluorescence signal intensity was highest at the bottom of the microchannel and decreased linearly to about onefourth of that value at the top of the PDMS channel. Black dotted lines indicate reference DNA fluorescence signal intensities.

Surface hybridization of DNA to MO arrays under electrokinetic trapping. This section compares the hybridization efficiency of the current concentrator-enhanced DNA-MO biosensor with the 32 one reported in our previous study, in which the PEDOT:PSS membrane was printed on the glass substrate (PEDOT:PSS-glass). Figure 4 shows that, as for the previous concentrator, the presence of MO probes on the slide did not affect the electrokinetic sample concentration as a similar (~500-800)-fold increase in concentration was obtained for the deactivated superaldehyde glass slides without MO probes. However, the confocal microscopy study for 100 nM DNA showed a lower fluorescence signal intensity at the bottom of the microchannel in the case of the MO printed substrate (Figure S-3 and Video S-4 in supporting information) compared to the simply deactivated superaldehyde substrate in Figure S-2. This result seemed to suggest that the electrically neutral MO probes might affect the electrophoretic mobility of the substrate, thus the concentration speed, as discussed previously. Using a starting concentration of 100 nM DNA target and a 15 min. concentration time, the fluorescence signal achieved with both types of concentrators was comparable. In the case of lower concentrations, 1 nM and 10 nM, however, the hybridization signal achieved using the PEDOT:PSS-PDMS devices was lower than the signal obtained with the PEDOT:PSS-glass devices. The hybridization signal for C0=10 nM was ~40% lower, and the signal for C0=1 nM was below the noise level σ of the fluorescence background (σ=10 A.U.). Consequently, a LOD (LOD=3×σ) of 10 nM was determined for the hybridization assay using PEDOT:PSS-PDMS devices, whereas 1 nM had been achieved previously with PEDOT:PSS-glass devices. These results reflect the data in Figure 2, where the fluorescence signal was about 50% lower in the case of PEDOT:PSS-PDMS devices at the lower DNA concentrations. The origin of the difference in fluorescence intensity between the PEDOT:PSS-PDMS and PEDOT:PSS-glass

devices, especially at the lower DNA concentrations between 1 nM and 10 nM, was also investigated using confocal microscopy (see Figure S-4 in supporting information for 10 nM DNA). Confocal microscopy measurements of the DNA concentration along the z-axis found that the DNA concentration at the MO array varied between the two device designs, thus offering an explanation for the different hybridization signals. In the case of PEDOT:PSS3 glass device, the signal intensity was 1.6×10 A. U. 2 for 10 nM, whereas 7.5×10 A. U. was achieved with the PEDOT:PSS-PDMS device. To estimate the increase in sensitivity via ET after 15 min. of concentration, the standard hybridization signal was measured at 15 min. for different initial DNA concentrations (Figure 4). Compared to standard hybridization, ET-enhanced hybridization shows the following increase in sensitivity: ~ 2.9-fold (PEDOT:PSS-glass) and 2.1fold (PEDOT:PSS-PDMS) at 100 nM; 5.3-fold (PEDOT:PSS-glass) and 2.1-fold (PEDOT:PSSPDMS) in sensitivity at 10 nM. At 1 nM, the signals from both standard and concentrator-enhanced hybridization were below the noise level, and consequently did not allow for a quantitative comparison. Instead of fluorescence detection that in general requires labeling of the purified DNA sample (with an exception provided by “fluorescence unquenching” 52–55 , other highly sensitive and label-free methods) detection methods (although generally more 56 technically challenging) based on electrochemical, 57 optical (e.g. surface plasmon resonance and 58 optical ring resonator ), and mass-based (e.g. 59,60 quartz crystal microbalance and 61 microcantilevers ) modalities, could also be integrated with our concentrator and potentially yield a significantly lower LOD (ideally in the aM-fM range, the concentration naturally occurring in body fluids). Since pH can have a profound effect on biomolecular interaction, the pH change induced by 62,63 ICP at the preconcentration region should be carefully considered so that the optimum pH for hybridization is achieved at the reaction site Using a pH-sensitive fluorescein dye, a pH drop by ~0.5 pH unit has been estimated in the DNA concentration plug (see section S6 in supporting information).

ACS Paragon Plus Environment

9

Analytical Chemistry

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

Page 10 of 15

of DNA fragments to the substrate was negligible. Therefore, no additional surface passivation step prior to the concentration experiment was required. These features, together with the ease of integration with microarray substrates, make the present method a promising approach to lab-on-chip diagnostics. Currently, the multiplexed concentrator device is limited in addressing specific probes in a microarray. One of the potential ways to apply the concentrators to specific MO probes on a microarray slide is to match the footprint size of the PDMS concentrator chip to that of a microarray slide so we can simply align them to each other along the outer edges. The location of the concentrators within the PDMS chip itself can be then determined during the mask design to address specific MO probes on the microarray slide.

Figure 4. Comparison of DNA concentration and electrokinetically enhanced hybridization intensities, after 15 min. of concentration of complementary 32 DNA target, for the PEDOT:PSS-glass and the PEDOT:PSS-PDMS device configurations. As a reference signal, the standard hybridization signal without electrokinetic concentration is shown. For C0 = 100 nM, the concentration enhancement as well as the hybridization results are independent of the membrane position. However, PEDOT:PSS-PDMS devices yielded lower hybridization signals compared to the PEDOT:PSS-glass devices for initial DNA concentrations of C0 = 1 nM and C0 = 10 nM. The preconcentration data suggest that the difference in hybridization signals resulted from a difference in preconcentration performance at the lower 1 nM and 10 nM initial DNA concentrations.

Multiplexing concentration-enhanced MODNA hybridization assays. To demonstrate potential for multiplexing, the number of concentrators per chip was increased to five, as shown in Figure 5 (a). This concentrator array was then integrated with a MO microarray simply by reversible bonding, and used to perform parallel hybridization assays [Figure 5 (b-d)]. The concentrated sample plugs and ion depletion regions in the five parallel channels are shown in Figure 5 (b). The locations of the concentration plugs remained stable with minimal variations across the multiple channels (see Video S-5 in supporting information) due to the robust fabrication process for the ion-selective membrane, as described earlier. During the concentration step, non-specific binding

Figure 5. A multiplexed microfluidic concentrator for scaling up electrokinetically enhanced MO-DNA hybridization. (a) Photograph of the multiconcentrator device. The multiple parallel channels are connected to the same inlet and outlet. Therefore, only a single sample and buffer loading reservoir, and a common pair of electrodes, were needed. (b) Fluorescent image showing the simultaneous preconcentration of a DNA sample next to PEDOT:PSS membranes in the five channels. (c) Zoomed-in image of a plug of Cy5labeled 100 nM DNA in one of the five channels, above an array of MO spots. (d) Hybridized spots after a 15 min. electrokinetic concentration step. The difference in signal intensity between the

ACS Paragon Plus Environment

10

Page 11 of 15

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

Analytical Chemistry

concentrated and non-concentrated zones is clearly visible. Figure 6 presents a quantitative summary of the concentration-enhanced hybridization experiments for 10 nM and 100 nM DNA samples in the 5channel devices. For both initial DNA concentrations tested, the fluorescence intensity of the plugs and DNA-MO hybridization were uniform across the microchannels and showed similar concentration factors compared to the single microchannel device. Unlike microflow-based patterning method which may generate non-uniformities in the membrane that would cause the parallel concentrators to exhibit 37 variations in concentration factors, our improved fabrication method provided a high chip-to-chip and channel-to-channel reproducibility (see Table S-1 in supporting information). To quantify the variability between the channels and the improvement in hybridization rate, the time needed for hybridization without electrokinetic concentration to achieve an equal signal to that after a 15 min. preconcentration were compared. For this comparison, the microchannel was filled with 100 µL of the target DNA solution in each reservoir and the fluorescence signal of MO-DNA hybridization was measured in real-time. For the two DNA concentrations tested, a 9-fold electrokinetic enhancement in hybridization rate was found for 100 nM DNA target (135 min.), while an 11-fold enhancement was found at 10 nM (165 min.).

Figure 6. Channel-to-channel comparison of the fluorescence signal intensity of the DNA plug (squares) under simultaneous electrokinetic concentration for 15 min. and the corresponding hybridization result (triangles).

SUMMARY AND CONCLUDING REMARKS We have demonstrated sample preconcentration and accelerated surface hybridization of DNA on a MO microarray using an electrokinetic concentrator consisting of a cation-selective PEDOT:PSS membrane layer printed directly onto a PDMS microchannel. The direct printing onto the PDMS channel allowed decoupling of the concentrator from the substrate where the capture probes are immobilized. This concentrator can be simply and reversibly bonded to an existing microarray and used for accelerated surface hybridization with minimal alignment efforts. It produced concentration factors of 300- to 800-fold after 15 min., starting with DNA concentrations from 1 to 100 nM. The threedimensional concentration profile of the plug showed that the DNA concentration was highest at the surface of the glass substrate where MO probes were immobilized and decreased almost linearly with distance from the slide. This non-uniform distribution, advantageous for surface hybridization applications, was attributed to the different EOF velocities between the bottom and the top of the microchannel caused by the different materials used for the top (PDMS) and the bottom (glass). Finally, a five-channel concentration device was used to demonstrate robustness of fabrication as well as simultaneous DNA hybridization on-chip. Our results show that, in comparison with standard surface hybridization, ET has the potential to speed up surface hybridization (~10-fold) while also improving detection sensitivity (~2-fold). Compared to our previously reported PEDOT:PSSglass device, the current device with the PEDOTPSS printed directly on the PDMS channel yielded lower concentration factors (~50% less) and, as a consequence, lower hybridization signals after the same 15 min. assay duration. This translated to a higher detection limit of 10 nM target compared to the previously reported limit at 1 nM. However, the sensitivity could likely be further enhanced by optimizing the PEDOT:PSS membrane geometry and properties such as its thickness and electrical conductivity, or, as previously mentioned, by using highly sensitive label-free detection methods. Presently, the multiple-channel concentrator device can only address a very limited fraction of probes on the microarray. By decreasing the gap between adjacent microchannels and the MOs spot size (down to 5 µm), we could increase both the

ACS Paragon Plus Environment

11

Analytical Chemistry

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

concentrator and microarray density, thereby minimizing the amount of unused MO probes and significantly increasing the throughput of the assay. The work proposed here lays the foundation for future development of fast and sensitive, highly multiplexed concentrator-enhanced MO microarrays. Such devices could be used to detect genes expressed at very low levels, as in eukaryotic gene expression. In addition, our fabrication and integration scheme is readily transferable to other high-throughout diagnostics such as immunoassays for detecting low abundance biomarkers. Furthermore, we believe that the combination of a high-density concentrator chip with a hand-held 64,65 portable microarray reader could hold great promise in point-of-care settings. Hence, the proposed technology has the potential to markedly enhance the detection speed and sensitivity of various biosensors for future clinical and point-ofcare diagnostics.

Page 12 of 15

through the NYUAD Research Enhancement Fund 2013. We thank Dr. James Weston of NYUAD for the SEM pictures. The device fabrication was conducted in the microfabrication core facility of NYUAD.

REFERENCES

(1)

Trevino, V.; Falciani, F.; Barrera-saldaña, H. A. Mol. Med. 2007, 13 (9-10), 527–541.

(2)

Gresham, D.; Dunham, M. J.; Botstein, D. Nat. Rev. Genet. 2008, 9 (4), 291–302.

(3)

Cheung, V. G.; Morley, M.; Aguilar, F.; Massimi, a; Kucherlapati, R.; Childs, G. Nat. Genet. 1999, 21 (1 Suppl), 15–19.

(4)

Gadgil, C.; Yeckel, A.; Derby, J. J.; Hu, W. S. J. Biotechnol. 2004, 114 (1-2), 31–45.

ASSOCIATED CONTENT

(5)

Napoli, M.; Eijkel, J. C. T.; Pennathur, S. Lab Chip 2010, 10 (8), 957–985.

Supporting Information

(6)

Bercovici, M.; Han, C. M.; Liao, J. C.; Santiago, J. G. Proc. Natl. Acad. Sci. U. S. A. 2012, 109 (28), 11127–11132.

(7)

Eid, C.; Garcia-Schwarz, G.; Santiago, J. G. Analyst 2013, 138 (11), 3117–3120.

(8)

Garcia-Schwarz, G.; Santiago, J. G. Anal. Chem. 2012, 84 (15), 6366–6369.

(9)

Schoch, R. B.; Ronaghi, M.; Santiago, J. G. Lab Chip 2009, 9 (15), 2145–2152.

(10)

Han, C. M.; Katilius, E.; Santiago, J. G. Lab Chip 2014, 14 (16), 2958–2967.

(11)

Bercovici, M.; Kaigala, G. V; Mach, K. E.; Han, C. M.; Liao, J. C.; Santiago, J. G. Anal. Chem. 2011, 83 (11), 4110–4117.

(12)

Karsenty, M.; Rubin, S.; Bercovici, M. Anal. Chem. 2014, 86 (6), 3028–3036.

(13)

Bahga, S. S.; Han, C. M.; Santiago, J. G. Analyst 2013, 138 (1), 87–90.

(14)

Persat, A.; Marshall, L. a; Santiago, J. G. Anal. Chem. 2009, 81 (22), 9507–9511.

(15)

Shintaku, H.; Palko, J. W.; Sanders, G. M.; Santiago, J. G. Angew. Chemie Int. Ed. 2014, 53 (50), 13813–13816.

(16)

Liu, V.; Song, Y.-A.; Han, J. Lab Chip 2010, 10 (11), 1485–1490.

Supporting information provides: a video showing the automatic PEDOT:PSS printing process on PDMS channels, 3D confocal videos of the DNA plug during concentration experiments, confocal data of the vertical concentration profile of the plug in microfluidic concentrators with PEDOT:PSS at the top and bottom of the microchannel, estimation of the pH shift during ET concentration, a video showing simultaneous DNA concentration in 5 parallel channels, and a measurement of the channel-to-channel and device-to-device variation.

AUTHOR INFORMATION Corresponding Author E-mail: [email protected]. Phone: +971-2-6284781. Fax: +971-2-659-0794.

Notes The authors declare no competing financial interest.

ACKNOWLEDGMENTS We gratefully acknowledge the financial support from the New York University Abu Dhabi (NYUAD)

ACS Paragon Plus Environment

12

Page 13 of 15

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

Analytical Chemistry

(17)

Choi, E.; Kwon, K.; Kim, D.; Park, J. Lab Chip 2015, 15 (2), 512–523.

(18)

Kim, P.; Kim, S. J.; Han, J.; Suh, K. Y. Nano Lett. 2010, 10 (1), 16–23.

(19)

Kwak, R.; Kim, S. J.; Han, J. Anal. Chem. 2011, 83 (19), 7348–7355.

(20)

Smart Mater. Struct. 2013, 22 (8), 085026. (37)

Kim, M.; Jia, M.; Kim, T. Analyst 2013, 138 (5), 1370–1378.

(38)

Song, H.; Wang, Y.; Garson, C.; Pant, K. Microfluid. Nanofluidics 2014, 17 (4), 693– 699.

Syed, a; Mangano, L.; Mao, P.; Han, J.; Song, Y. Lab Chip 2014, 14 (23), 4455–4460.

(39)

Jeon, H.; Lee, H.; Kang, K. H.; Lim, G. Sci. Rep. 2013, 3, 3483.

(21)

Moghadam, B. Y.; Connelly, K. T.; Posner, J. D. Anal. Chem. 2015, 87 (2), 1009–1017.

(40)

(22)

Moghadam, B. Y.; Connelly, K. T.; Posner, J. D. Anal. Chem. 2014, 86 (12), 5829–5837.

Kim, S. J.; Ko, S. H.; Kwak, R.; Posner, J. D.; Kang, K. H.; Han, J. Nanoscale 2012, 4 (23), 7406–7410.

(41)

Kwak, R.; Kim, S. J.; Han, J. 2010, No. October, 887–889.

(42)

Kim, B.; Heo, J.; Kwon, H. J.; Cho, S. J.; Han, J.; Kim, S. J.; Lim, G. ACS Nano 2013, 7 (1), 740–747.

(43)

Ko, S. H.; Song, Y.-A.; Kim, S. J.; Kim, M.; Han, J.; Kang, K. H. Lab Chip 2012, 12 (21), 4472–4482.

Larson, B. J.; Gillmor, S. D.; Lagally, M. G. Rev. Sci. Instrum. 2004, 75 (4), 832.

(44)

Cheow, L. F.; Ko, S. H.; Kim, S. J.; Kang, K. H.; Han, J. Anal. Chem. 2010, 82 (8), 3383– 3388.

Cheun, H.; Rugheimer, P. P.; Larson, B. J.; Gopalan, P.; Lagally, M. G.; Winokur, M. J. J. Appl. Phys. 2006, 100 (7), 073510.

(45)

Lee, J. H.; Song, Y.-A.; Han, J. Lab Chip 2008, 8 (4), 596–601.

Charlot, B.; Sassine, G.; Garraud, A.; Sorli, B.; Giani, A.; Combette, P. Microsyst. Technol. 2012, 19 (6), 895–903.

(46)

Wang, Y.-C.; Han, J. Lab Chip 2008, 8 (3), 392–394.

Kim, S. J.; Song, Y.-A.; Han, J. Chem. Soc. Rev. 2010, 39 (3), 912–922.

(47)

Lee, J. H.; Cosgrove, B. D.; Lauffenburger, D. a; Han, J. J. Am. Chem. Soc. 2009, 131 (30), 10340–10341.

Mcdonald, J. C.; Duffy, D. C.; Anderson, J. R.; Chiu, D. T. Electrophoresis 2000, 21, 27– 40.

(48)

Lee, J. H.; Song, Y.; Tannenbaum, S. R.; Han, J. Lab Chip 2008, 80 (9), 6592–6598.

Haubert, K.; Drier, T.; Beebe, D. Lab Chip 2006, 6 (12), 1548–1549.

(49)

Sarkar, A.; Han, J. Lab Chip 2011, 11 (15), 2569–2576.

Tang, T.; Li, G.; Jia, C.; Gao, K.; Zhao, J. Biomicrofluidics 2014, 8 (2), 026501.

(50)

Martins, D.; Levicky, R.; Song, Y.-A. Biosens. Bioelectron. 2015, 72, 87–94.

Kirby, B. J.; Hasselbrink, E. Electrophoresis 2004, 25 (2), 203–213.

(51)

Ren, X.; Bachman, M.; Sims, C.; Li, G. P.; Allbritton, N. J. Chromatogr. B Biomed. Sci. Appl. 2001, 762 (2), 117–125.

(52)

Yang, R.; Jin, J.; Chen, Y.; Shao, N.; Kang, H.; Xiao, Z.; Tang, Z.; Wu, Y.; Zhu, Z.; Tan, W.; Uni, V.; Gaines, V. J. Am. Chem. Soc. 2008, 130 (16), 8351–8358.

(53)

Du, H.; Disney, M. D.; Miller, B. L.; Krauss, T. D. J. Am. Chem. Soc. 2003, 125 (14), 4012– 4013.

(23)

(24) (25)

(26)

(27) (28) (29)

(30) (31) (32)

Ko, S. H.; Kim, S. J.; Cheow, L. F.; Li, L. D.; Kang, K. H.; Han, J. Lab Chip 2011, 11 (7), 1351–1358. Lee, J. H.; Han, J. Microfluid. Nanofluidics 2010, 9 (4), 973–979.

(33)

Qiao, W.; Kalachikov, S.; Liu, Y.; Levicky, R. Anal. Biochem. 2013, 434 (2), 207–214.

(34)

Tercero, N.; Wang, K.; Gong, P.; Levicky, R. J. Am. Chem. Soc. 2009, 131 (13), 4953– 4961.

(35)

Tercero, N.; Wang, K.; Levicky, R. Langmuir 2010, 26 (17), 14351–14358.

(36)

Kim, S.-S.; Jeon, J.-H.; Kee, C.-D.; Oh, I.-K.

ACS Paragon Plus Environment

F.

13

Analytical Chemistry

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

(54)

Steemers, F. J.; Ferguson, J. a; Walt, D. R. Nat. Biotechnol. 2000, 18, 91–94.

(55)

Fang, X.; Liu, X.; Schuster, S.; Tan, W. J. Am. Chem. Soc. 1999, 121 (12), 2921–2922.

(56)

Drummond, T. G.; Drummond, T. G.; Hill, M. G.; Hill, M. G.; Barton, J. K.; Barton, J. K. Nat. Biotechnol. 2003, 21 (10), 1192–1199.

(57)

Zagorodko, O.; Spadavecchia, J.; Serrano, A. Y.; Larroulet, I.; Pesquera, A. Anal. Chem. 2014, 86 (22), 11211–11216.

(58)

Suter, J. D.; White, I. M.; Zhu, H.; Shi, H.; Caldwell, C. W.; Fan, X. Biosens. Bioelectron. 2008, 23 (7), 1003–1009.

(59)

Feng, K.; Li, J.; Jiang, J.-H.; Shen, G.-L.; Yu, R.-Q. Biosens. Bioelectron. 2007, 22 (8), 1651–1657.

(60)

Okahata, Y.; Kawase, M.; Niikura, K.; Ohtake, F.; Furusawa, H.; Ebara, Y. Anal. Chem. 1998, 70 (7), 1288–1296.

(61)

McKendry, R.; Zhang, J.; Arntz, Y.; Strunz, T.; Hegner, M.; Lang, H. P.; Baller, M. K.; Certa, U.; Meyer, E.; Güntherodt, H.-J.; Gerber, C. Proc. Natl. Acad. Sci. U. S. A. 2002, 99 (15), 9783–9788.

(62)

Andersen, M. B.; Rogers, D. M.; Mai, J.; Schudel, B.; Hatch, A. V; Rempe, S. B.; Mani, A. Langmuir 2014, 30 (26), 7902–7912.

(63)

Mai, J. Y.; Miller, H.; Hatch, A. V. ACS Nano 2012, 6 (11), 10206–10215.

(64)

Thompson, D. L.; Coleman, M. A.; Lane, S. M.; Albala, J.; Wachsmann-Hogiu, S. Handheld Portable Microarray Reader for Biodetection. US 2011/0188782 A1, 2011.

(65)

Thompson, D. L.; Pearson, F.; Thomas, C.; Rao, R.; Matthews, D.; Albala, J. S.; Wachsmann-Hogiu, S.; Coleman, M. a. Sensors 2009, 9 (4), 2524–2537.

ACS Paragon Plus Environment

Page 14 of 15

14

Page 15 of 15

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

Analytical Chemistry

for TOC only

ACS Paragon Plus Environment

15