BACKGROUND OF THE INVENTION

Assays used in biosensor systems can be roughly divided into two major classes; solid phase binding assays and homogeneous solution phase assays. Solid phase binding assays are far more common because of the advantages of anchoring the ligand- receptor interactions to a surface. Immobilizing the receptors simplifies many aspects of an assay, such as rinsing and exchange of reagents, and enables multiplexing by arraying different receptors across the surface. However, the solid support can also be a limitation, both because of fouling and the inefficient diffusion of target molecules to the two- dimensional surface. See Sheehan, P.E., Whitman, LJ. , 2005. Nano Lett. 5, 803-807. Although homogeneous assays are generally not limited by mass transport, they are more challenging to multiplex, both in terms of chemistry and detection. A common solution to multiplexing in homogeneous assay systems has been the use of microbeads with unique fluorescent signatures, with each ligand-receptor pair anchored to a different color bead.

Individual or multiplexed targets are then identified by fluorescence analysis; this technique has been used to detect a wide variety of DNA, protein, and small molecule targets (Anderson, et al., 2006. Anal. Chem. 78, 2279-2285.; Bandyopadhyay et al., 2007. J. Clin. Microbiol. 45, 2835-2840; Mahony et al., 2007 J. Clin. Microbiol. 45, 2965-2970; Sun et al., 2007 Acta Pharmacol. Sin. 28, 2011-2018). This approach has been very successful, but the required detection instrumentation has been relatively large, complex, and expensive.

Solid phase assays proliferate because they have the flexibility to be either rapid, inexpensive, and simple to execute — e.g., lateral flow immunoassays as implemented in home pregnancy tests; or massively multiplexed and capable of deciphering complex relationships — e.g., large-scale nucleic acid microarrays as applied to gene expression profiling (Epstein et al., 2002 Anal. Chim. Acta 469, 3-36; Michalet et al., 2003 Annu. Rev. Biophys. Biomol. Struct. 32, 161-182). Currently, the most sensitive multiplexed detection technology capable of both protein and nucleic acid detection is the bio-barcode assay (Hill and Mirkin, 2006, Nat. Protoc. 1, 324-336; Jaffrezic-Renault et al., 2007 Sensors 7, 589-614; Nam et al., 2005 Anal. Chem. 77, 6985-6988; Tang et al., 2007 J. Acq. Immune Def. Synd. 46, 231-237), with reported protein detection at attomolar concentrations (Georganopoulou et al., 2005 Proc. Natl. Acad. Sci. U. S. A. 102, 2273- 2276; Goluch et al., 2006 Lab Chip 6, 1293-1299; Nam et al., 2003 Science 301, 1884- 1886; Oh et al., 2006 Small 2, 103-108) and DNA at zeptomolar concentrations (Nam et al., 2004 J. Am. Chem. Soc. 126, 5932-5933). The excellent analytical sensitivity of this approach can be attributed to the signal amplification achieved by detecting thousands of oligonucleotide "barcodes" for each target molecule captured onto a microbead in solution (rather than detecting each target). Bio-barcode assays that use microarray-based "scanometric" detection of these oligos have achieved the stated level of sensitivity in 2-5 hr. Recently, barcode oligonucleotides have been fluorescently labeled and detected in solution, identifying a protein target (prostate specific antigen) in serum at 300 aM in only 90 min (Oh et al., 2006, Small 2, 103-108).

The fluid force discrimination (FFD) assay is an alternate approach to multiplexed detection that is simple, uses few reagents, is rapid (<25 min), and directly detects the biological target without amplification (Rife, et. al., U.S. Patent Application 11/424643; Mulvaney et al., 2007 Biosen. Bioelectron. 23, 191-200). FFD assays have been used to demonstrate multiplexed, femtomolar nucleic acid hybridization and immunoassays in a solid phase microarray format. In FFD assays, biomolecular targets are captured onto a microarray and then labeled with microbeads. Next, nonspecifically bound bead labels are preferentially removed by application of controlled, laminar fluidic forces. The density of beads remaining on each spot after FFD indicates the target identity and concentration. The use of microbead labels enhances the analytical sensitivity by enabling label detection at extremely low label densities («10 ⁴ /cm ^~2 ), down to single labels in a typical microarray spot (~100-μm diameter). Unlike approaches that rely on amplification, microbead assays are ideal for the analysis of samples in complex matrices because they do not suffer from the many forms of matrix interference that plague other labels (e.g., autofluoresence, electrochemical background). Finally, if the microbead labels are paramagnetic, FFD assays can be performed in a field-portable detection system incorporating magnetoresistive sensor array chips (Baselt et al., Biosens. Bioelectron. 13, 731-739, 1998; Edelstein et al., Biosen. Bioelectron. 14, 805-813, 2000; Miller et al., 2001; Mulvaney et al., 2007; Rife et al., Sens. Actuators, A 107, 209-21, 2003; Tamanaha et al., Biosensors and Bioelectronics 24 (2008) 1-13). BRIEF SUMMARY OF THE INVENTION

Disclosed is a semi-homogenous (SH) implementation of a fluidic force discrimination (FFD) assay using only two reagent mixtures and three assay steps that can be performed in as little as 10 min. Previously microbead labels and FFD have been combined to achieve multiplexed, femtomolar nucleic acid hybridization and immunoassays in a microarray format (Mulvaney et al., 2007. Biosen. Bioelectron. 23, 191-200). In SH FFD assays, the microbeads and any required intermediate receptors (e.g. secondary antibodies) are first mixed directly with a sample, allowing target analytes to be efficiently captured onto the beads. The target- loaded beads are then specifically captured onto a microarray surface, with nonspecifically bound beads removed by controlled, laminar fluidic forces. The remaining beads on each microarray capture spot are counted to determine the targets' identities and concentrations. SH target collection provides a 1000- fold improvement in the assay sensitivity, down to attomolar concentrations, as demonstrated by our detection of staphylococcal enterotoxin B (SEB) at 35 aM (1 fg/ml). SH assays are adaptable for extraction, preconcentration, and identification of analytes in complex sample matrices, including assays for SEB and ricin toxoid in serum and whole blood. Disclosed is a detailed model of the reaction kinetics that reveals how capturing the targets onto the beads in solution provides a significant kinetic advantage at low target concentrations where mass transport to a microarray surface is most limited.

The disclosed embodiment of FFD assays results in a 1000-fold improvement in analytical sensitivity — down to attomolar concentrations — using only two reagent mixtures and three assay steps that can be performed in as little as 10 min. In this semi- homogenous (SH) implementation of FFD, the microbeads and any required intermediate receptors (e.g. secondary antibodies) are mixed directly with the sample solution, allowing targets in solution to be efficiently captured onto the beads. The target- loaded beads are then specifically captured onto the microarray surface, with nonspecifically bound beads removed by FFD in buffer. The assay results combined with detailed modeling of the reaction kinetics confirms that capturing the targets onto the beads in solution (vs. on the microarray) provides a significant kinetic advantage at low target concentrations, where mass transport to a microarray surface is most limited. Furthermore, we demonstrate that SH assays are adaptable for extraction, preconcentration, and analysis of target in complex matrices, including assays for staphylococcal enterotoxin B (SEB) and ricin A chain (RCA) in serum and whole blood.

Also disclosed is a method for making a stable surface configured for attaching NeutrAvidin. A surface having primary amines the surface is reacted with succinimidyl A- formylbenzoate (SFB). NeutrAvidin is reacted with with succinimidyl A- hydrazinonicotinate acetone hydrazone (SANH). The SFB-reacted surface is incubated with the NA-SANH solution under acidic conditions to configure said surface with stable SFB-SANH. The stable surface is stored at 4° C and remains stable for at least 4 weeks.

The semi-homogenous binding assay method also includes an embodiment wherein a solution comprised of a target of interest is mixed with a micrometer-scale bead configured to bind specifically to the target or to an intervening receptor that binds specifically to the target and to the micrometer- scale bead. A capture surface configured to specifically bind the target is exposed to the target solution. A fluidic system is employed to remove non- specifically bound micrometer-scale particle labels from the capture surface, wherein said fluidic system applies a controlled and uniform laminar flow at the capture surface, said flow producing a Stokes force on said micron-scale bead of at least 1 pN. The Stokes force preferentially removes non- specifically bound micron sized labels from the capture surface.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a depiction of the modified surface chemistry;

FIG. 2 shows a sequential and a semi-homogenous FFD binding assay;

FIG. 3 shows the relative performances of sequential and SH FFD assays for the detection of SEB in buffer;

FIG. 4 the semi-homogeneous FFD detection signal for 35 pM and 35 fM SEB in buffer as a function of the homogeneous mixing time (1-20 min);

FIG. 5 shows a comparison of SH FFD assays for SEB in buffer and in serum;

FIG. 6 shows the detection of 150 pM (10 ng/ml) RCA in whole blood with SH FFD assays. DETAILED DESCRIPTION OF THE INVENTION Each assay was performed in an acrylic flow cell mounted on a microscope slide. The flow cell was 2.8 mm long x 2.2 mm wide x 100 μm high and had a tapered entrance and exit to insure uniform, laminar flow of reagents across the capture spots, as discussed in detail in (Mulvaney et al., 2007; Tamanaha et al., J. Micromechan. Microeng. 12, 347-347, 2002). In particular, the flow cell design creates a very uniform fluid velocity across the middle of the channel where the assay occurs, ensuring consistent spot-to-spot application of fluidic shear forces. The flow rates were controlled with a peristaltic pump (Instech Laboratories, Inc.).

The surface chemistry used here was a variation of that described previously

(Mulvaney et al., 2007), with a different method used to couple NeutrAvidin (NA) to the substrate. Standard microscope slides coated with silicon nitride were purchased from Lance Goddard Films. As-received slides were dipped in 1% HF solution, rinsed with copious amounts of H ₂ O, and dried in a N ₂ stream. Slides were then immediately plasma cleaned in -70% humidified air for 50 min at 135 W, producing a combination of hydroxyl and primary amine groups on the surface (Cole, et. al., U.S. Patent Application 11/424643; Stine et al., Langmuir 23, 4400-4404, 2007).

The modified surface chemistry disclose herein is shown in FIG. 1. NA, succinimidyl 4-hydrazinonicotinate acetone hydrazone (SANH), and succinimidyl A- formylbenzoate (SFB) were purchased from Pierce Biotechnolgies, Inc. Prior to the plasma treatment of the slides, NA at 1 mg/ml was reacted with SANH according to manufacture recommendation and then dialyzed for 24 hr to remove unreacted SANH. The primary amine groups created on the slide surface after plasma treatment were reacted with 50 mM SFB in DMSO (Aldrich) for 2 hr. Next, the SFB solution was aspirated and the slides were incubated with the NA-SANH solution for 2 hr under acidic conditions, coupling the SFB and SANH via a condensation reaction. Finally, slides were rinse with copious amounts of H ₂ O and stored at 4 ⁰ C until used. Surfaces prepared this way had the advantage of being stable under storage conditions for at least 4 weeks at 4 ⁰ C. This presents a major advantage over previously disclosed surface chemistries. The assay buffer was comprised of one liter of 1 x phosphate buffered saline was mixed with 50 g of dehydrated skim milk (Carnation), 1 ml of Tween 20 (Aldrich), and 1 ml of 1 % thimersol (Aldrich).

Sequential FFD assays were performed as detailed in Mulvaney et al., 2007. A NA functionalized microscope slide was spotted with biotinylated sheep anti-SEB (Toxin Technologies, Inc.) and goat anti-ricin (Tetracore) capture antibodies (1 μg/ml) and sealed in the microfluidic cell. The surface was then challenged for 5 min by stop flow with SEB or RCA at the stated concentration. Next, a rabbit anti-SEB (Toxin Technologies, Inc) or rabbit anti-ricin (Tetracore) secondary antibody solution (1 μg/ml) was introduced for 5 min by stop flow. Sheep anti-rabbit beads (Invitrogen) were then introduced and allowed to settle to the surface (3 min). Finally, FFD was applied with buffer to remove nonspecifically bound beads (3-5 min) and the remaining beads counted. From target introduction through bead counting, the total assay time was<25 min.

Semi-homogeneous FFD assays were also performed. Target solutions (1 ml) at the stated concentration were prepared in either buffer or canine serum (Innovated

Research, Inc) as indicated. For each experiment, 100 μl of secondary antibody solution (1 μg/ml) and 100 μl of sheep anti-rabbit beads were added to the target solution. The solution was mixed on a 360° rotator for 5 min. The target-loaded beads were then used to challenge the capture probe array, with a 3 min settling time. FFD was performed in buffer to remove nonspecifically bound beads (3-5 min). (Note that force discrimination was completed in <1 min, but additional time was required to flush all beads from the tubing used in this generation instrument.) From the start of the mixing step through bead counting, the total assay time was <20 min.

Deviations from this protocol are noted in the text. Specifically, for time- dependent SH assays, the mixing time on a 360° rotator was varied from 1-20 min as indicated. For SH assays via extraction, RCA was spiked into whole blood (Innovated Research, Inc.) and 100 μl of rabbit anti-ricin solution (1 μg/ml) and 100 μl of sheep anti- rabbit beads were added. The solution was mixed on a 360° rotator for 5 min. The whole blood matrix was then exchanged for buffer while magnetically retaining the beads. The target-loaded beads were then resuspended in buffer and introduced to the microarray as usual. The density of beads remaining on a capture spot after FFD is proportional to the analyte concentration. All data reported herein were acquired with optical bead counting and are reported as the number of beads in a 200-μm diameter spot, with error bars representing the standard deviation of at least three measurements. For optical counting, an image of the capture spot is collected utilizing a CCD camera and optical microscope. A custom LabVIEW-based application then defines a 200-μm diameter circle in the center of the capture spot. Using National Instrument Vision software, the number of beads within that circle is determined by setting a binary threshold in the bright field image to delineate the beads, calculating the area covered by beads, and then dividing by the average area of a single bead. Analyzing the images in this manner allows for accurate bead counts even when some beads are too close (or touching) for the software to resolve them individually. The total image analysis time is currently <2 min per spot, dominated by the time required to manually locate and focus on each capture spot. With automated image capture, under development, we expect the total image analysis time to be <30 s per spot. Those skilled in the art would understand that other means may be used to count the beads.

Both the sequential and SH FFD assay schemes are depicted in FIG. 2. The sequential assay shown is typical for labeled, solid phase assays (e.g., ELISAs) where an immobilized capture probe is exposed to a series of solutions to create a biomolecular "sandwich." In the example depicted, the target is captured first, a secondary antibody (2° Ab) that recognizes a second epitope on the target follows, and then an antibody- conjugated microbead that recognizes the Fc portion of the 2° Ab completes the sandwich. Note that using a 2° Ab facilitates multiplexing: if all 2° Ab are from the same source (i.e., rabbit anti-target), a single type of Ab-conjugated microbead (i.e., sheep anti-rabbit) can label all assays. Target identities and concentrations in the sample may then be determined from the density of bead labels at each capture spot. In contrast, in a SH assay, the biomolecular sandwich is partially formed by simultaneously mixing the target, the 2° Ab, and the microbeads in a single solution. The sandwich is completed in a second step by exposing the capture probe microarray to the target-loaded beads. Finally, in both assays the nonspecifically attached beads are preferentially removed with FFD and the remaining beads at each capture spot counted.

The relative performances of sequential and SH FFD assays for the detection of SEB in buffer are compared in FIG. 3. For the sequential assays, the target and 2° Ab were separately introduced in stop flow with a 5 min incubation time for each. Detection was reproducible over 6 orders of magnitude, from 35 fM (1 pg/ml) to 35 nM (1 μg/ml). (At 3.5 fM, the signal is comparable to the background.) For the SH assays, the target, 2° Ab, and microbeads were combined in a 1 ml solution for 5 min. Note that this step results in a net time savings because the 2° Ab labeling was performed concurrently with target capture. The SH FFD assays resulted in greater bead capture at all target concentrations. The SH FFD assay is more sensitive by three orders of magnitude, with SEB easily detected at 35 aM (1 fg/ml).

The improvement in sensitivity in a SH assay is attributed to the greater efficiency of target capture onto a bead by homogeneous mixing as compared to laminar flow delivery to a microarray surface. In sequential assays, diffusion of the target and 2° Ab to the capture probes on the microarray surface is known to be a limiting step (Sheehan and Whitman, 2005). SH assays are more efficient because of the homogeneous sampling of target molecules and 2° Ab by the microbead labels, with each target molecule in the 1 ml SH mixture having numerous opportunities to bind to a 2° Ab and be captured onto a bead. In a sequential assay, the flow cell volume is ~2 μl, such that — on a simple per volume basis — SH assays have a 500-fold sampling advantage. Although sequential assays could compensate for this sampling advantage by flowing target solution over the capture array, reasonable flow rates would add significant time to the assay; for example, flowing 1 ml of solution at 5 μl/min would require 200 min. At the lowest concentrations, where sequential assays become limited by diffusion over these time scales (minutes), the advantage of SH assays becomes most pronounced. Whereas the resulting difference in final bead density between the sequential and SH SEB assay at 35 nM is only 6%, at 35 fM it grows to 36%.

Shorter assay times are a significant advantage of SH assays. Moreover, the assay time may be varied to balance speed versus sensitivity. This tradeoff is illustrated in FIG. 4 for two concentrations of SEB in buffer and SH mixing times varied from 1 to 20 min. FIG. 4 shows the semi-homogeneous FFD detection signal for 35 pM (□) and 35 fM (O) SEB in buffer as a function of the homogeneous mixing time (1-20 min). The 35 fM data has been fit to the model described in the text, yielding the rate constants shown. For both concentrations, the signal is well above background with ~1 min of mixing, reaches about 50% of the maximum within 5 min, and is approaching saturation for 35 fM at 20 min. Thus, depending on the end user's need, 1 min of homogeneous mixing (<10 min total assay time) may be sufficient, or longer mixing times may be used to achieve additional sensitivity.

A distinct advantage of FFD assays is that the microbeads are physical labels and therefore compatible with complex sample matrices. As previously reported, FFD assays have been successfully performed in bacterial growth medium, milk, saliva, feces, urine, plasma, serum, and blood diluted 10-fold (Mulvaney et al., 2007). The only matrix requirement is that the chemical composition does not prevent biomolecular recognition. SH assays maintain this exceptional compatibility with complex matrices as demonstrated by the preliminary dose response curve for the detection of SEB in serum. FIG. 5 shows a comparison of SH FFD assays for SEB in buffer (same data as FIG. 3) and in serum. The background represents the average bead count over a negative capture spot (goat anti-ricin). It is well known that complex matrices typically reduce analytical sensitivity (Maraldo and Mutharasan, Anal. Chem. 79, 7636-7643, 2007; Morgan et al., J. Immunol. Methods 217, 51-60, 1998), and similar effects were observed here. The added components present in serum, mostly proteins, add viscosity, steric hindrance, and nonspecific competition for the ligands and receptors under investigation. The effects of this competition should be more pronounced with decreasing relative target concentrations, consistent with the steeper slope of the dose-response observed in serum versus buffer. At 35 aM SEB our detection signal is just above background; we estimate a statistically significant limit of detection under these conditions (5 min mix) to be about 3.5 fM (100 fg/ml).

SH assays are not only compatible with complex matrices, but when combined with magnetic microbeads also provide a capability for sample extraction and preconcentration. Paramagnetic microbeads were used so that the assays can be incorporated into a compact detection system based on magnetoelectronic sensor chips. Traditionally, magnetic microbeads are used to homogeneously capture and magnetically extract targets of interest from complex samples for subsequent analysis. Magnetic extraction has been shown to be effective even from whole blood, one of the most complicated matrices to assay. The abundance of proteins, lipids, and cellular material found in whole blood both foul surfaces and create steric barriers that hinder target-probe binding (Tian et al., Anal. Biochem. 283, 175-191, 2000). A 10-fold dilution of blood with buffer is sufficient to enable detection with a sequential FFD assay (Mulvaney et al., 2007). Simple dilution is also effective with SH assays; however, whole blood can be analyzed without dilution by using the magnetic beads for target extraction and then resuspending the target-loaded beads in buffer. FIG. 6 shows the detection of 150 pM (10 ng/ml) RCA in whole blood with SH FFD assays. The optical micrographs (250 μm x 250 μm) show bead capture within goat anti-ricin and negative capture probe (sheep anti-SEB) spots. One SH assay was performed completely in blood; in the other, the target- loaded beads were extracted from the sample via magnetic retention and then resuspended in buffer before challenging the microarray. Without extraction and resuspension, the bead signal in the SH FFD assay for 150 pM RCA in blood was barely above the small nonspecific background on a SEB capture spot. In contrast, a strong signal was observed when the beads were extracted and resuspended in buffer before challenging the microarray. Extraction is also an effective method of preconcentration from large volumes, successfully detecting both SEB and RCA from volumes as large as 10 ml after resuspension into <1 ml volumes.

Sensitivity, selectivity, and speed are the primary metrics by which bioassay performance is measured. Polymerase chain reaction (PCR) and other asymmetric amplification schemes enable the detection of <10 copies of DNA in typical sample volumes of <100 μl (Decaro et al., 2006; Hoffmann et al., 2006; Mackay, 2004). However, these amplification schemes add complexity and sources of error to the analysis, especially when analyzing samples in complex matrices or outside a controlled laboratory setting. The bio-barcode assay, as discussed above, addresses many of these shortcomings by using magnetic extraction combined with internal amplification to achieve PCR-like sensitivity in complex matrices. Similar to the present SH FFD assay, the bio-barcode approach combines homogeneous collection of target molecules onto microbeads with subsequent microarray detection of the identifying barcode labels. Because of the simplicity and efficiency of the labels in FFD assays, however, SH FFD assays are able to achieve comparable attomolar sensitivities in a notably simpler and faster format, using only two reagent mixtures (target + 2° Ab + bead solution and buffer) and three assay steps.

Homogeneous capture of target and secondary label onto the microbeads enhances the assay sensitivity in two ways: primarily through improved ligand-receptor capture; and, to a lesser extent, through assisted delivery of on-bead target to the microarray surface. In sequential FFD assays, performance is limited by diffusion of target molecules to the microarray surface, which can be enhanced somewhat by flow. The accumulation rate of target species on the surface depends on flow cell dimensions, sensor size, the concentration of the target, and the physical properties of the sample solution (viscosity, salinity, etc.). Even if a microarray system is designed with these principles in mind, overcoming the diffusion limit requires assisted target delivery (Morozov et al., J. Am. Chem. Soc. 129, 12628, 2007). Many of these factors have previously been discussed at length as they relate to DNA capture at and delivery to disk-, hemispherical-, or hemicylindrical- shaped sensors (Sheehan and Whitman, 2005). The kinetics of DNA hybridization, however, are simplified by the negligible dissociation rate of double-stranded DNA. In contrast, immunocomplexes are more sensitive to association and dissociation rates, so each binding interaction must be considered.

In addition to improved target capture, the homogeneous collection step also improves assay efficiency via bead-assisted target delivery. In SH FFD assays, gravity is used to enhance delivery of the target to the microarray surface. When injected into the flow cell, the relatively dense magnetic microbeads will quickly settle from solution onto the microarray. In a 100-μm high flow cell, a bead at the top of the flow cell will reach the microarray surface in <1 min. The three-dimensional search for a binding site is then reduced to a two-dimensional search as the bead continues to sample the surface through Brownian motion. The reduction of dimensionality due to bead settling is partially responsible for the greater sensitivity of SH assays. It is also possible for beads to be more rapidly pulled to the surface using an applied magnetic field, a technique that works well for nanometer-diameter beads (de Boer et al., Biosen. Bioelectron. 22, 2366-2370, 2007;

Graham et al., Sens. Actuators, B 107, 936-944, 2005; Graham et al., Trends Biotechnol. 22, 455-462 ,2004), but which adds complexity and increased instances of agglomeration with microbeads. When a magnetic field is applied for this purpose many of the magnetized microbeads cluster prior to reaching the surface, an undesirable effect when using FFD.

It is a major challenge to create assays that are sensitive and specific yet highly multiplexed, while keeping the approach rapid, simple, and inexpensive. Microarrays are excellent strategies for multiplexing, but are susceptible to surface fouling and are limited by mass transport and the reaction rate of target species with the microarray capture probes. In contrast, homogeneous assays enable very efficient target labeling, but multiplexing greatly increases the complexity of the chemistry and detection. With semi-homogeneous fluidic force discrimination assays we combine the capture efficiency of homogeneous mixing, the ease of multiplexing with microarrays, and the sensitivity and specificity enhancements provided by FFD. Detailed analysis of the assay kinetics shows that by blending these strategies SH FFD assays provide a 1000-fold advantage in target capture and a 100-fold advantage in reaction rate by delivering target-laden microbeads to the microarray for discrimination and detection. This powerful combination enables unamplified, attomolar protein detection in complex sample matrices using only two reagent mixtures and three assay steps that can be performed in as little as 10 min, with options for extraction and additional preconcentration from large volumes.