WO2013122689A1

WO2013122689A1 - Devices and methods for biomolecule detection

Info

Publication number: WO2013122689A1
Application number: PCT/US2013/020261
Authority: WO
Inventors: Wayne Frasch
Original assignee: Arizona Board Of Regents, A Body Corporate Of The State Of Arizona, Acting For And On Behalf Of Arizona State University
Priority date: 2012-01-06
Filing date: 2013-01-04
Publication date: 2013-08-22

Abstract

The present invention provides molecular detection devices and methods for using the devices to detect biomolecules.

Description

Devices and Methods for Biomolecule Detection

Cross-reference

This application claims priority to U.S. Provisional Patent Application Serial No. 61/583719 filed January 6, 2012, incorporated by reference herein in its entirety.

Statement of Government Rights

This work was supported under grant number N66001-03-C-8022ARZS from the Defense Advanced Research Projects Agency (DARPA) and FA9550-05-1-0424 from the Air Force Office of Scientific Research (AFOSR). The U.S. government has certain rights in the invention.

Background

The production of small amounts of protein by just a few specific cells can have profound effects on the normal development or pathology of an entire organism. A single neuron can affect an entire neural circuit culminating in motor action², or affect behavioral responses in rats. In vivo transplantation of individual adult stem cells can give rise to an entire prostate, or a functioning mammary gland. Small numbers of cells can be identified by their DNA signature by first amplifying the DNA copy number via techniques like PCR. However, detecting protein targets remains a much greater challenge due to the inability to amplify the number of copies of target molecules. An average mammalian cell has been determined to contain only about 100,000 copies of each of 1,000 proteins. Unfortunately, the protein from about 5000 cells is required to detect a protein analyte using a typical Western blot assay. Thus, improved detection devices and methods are needed.

Summary of the Invention

In a first aspect, the present invention provides devices, comprising

(a) a substrate;

(b) a plurality of first components, wherein each first component comprises

(i) a first capture ligand, wherein the first capture ligand binds to a first portion of a biomolecule target; and (ii) a first molecular motor attached to the first capture ligand, wherein the first molecular motor is attached to the substrate; and

(c) a plurality of second components, wherein each second component comprises

(i) a first detection ligand, wherein the first detection ligand binds to a second portion of the biomolecule target; and

(ii) a first nanorod attached to the first detection ligand.

In one embodiment, the first molecular motor comprises an Fi-ATPase; in this embodiment, the Fi-ATPase may be attached to a first member of a first binding pair, the first member of the first binding pair may be bound to a second member of the first binding pair, and and the second member of the first binding pair may be attached to the first capture ligand. In another embodiment, the second member of the first binding pair may be bound to a first binding molecule, and the first binding molecule may be bound to the first capture ligand. In any of these embodiments, the first nanorod may be attached to a second binding molecule, and the second binding molecule may be bound to the first detection ligand.

In any of these embodiments or combinations thereof, the biomolecule may comprise a polypeptide; the first capture ligand may comprise an antibody selective for a first site on the polypeptide, and the first detection ligand may comprise an antibody selective for a second site on the polypeptide. In exemplary embodiments, the polypeptide is an Stx2 toxin or an HSP70 polypeptide. In another exemplary embodiment, the first capture ligand may comprise an antibody selective for a Stx2B subunit of the Stx2 toxin, and the first detection ligand may comprise an antibody selective for a Stx2A subunit of the Stx2 toxin.

In any of these embodiments or combinations thereof, the device may further comprise

d) a plurality of third components, wherein each third component comprises i. a second capture ligand, wherein the second capture ligand binds to a nucleic acid of interest; and

ii. a second molecular motor attached to the second capture ligand,

wherein the second molecular motor is attached to the substrate; and e) a plurality of fourth components, wherein each fourth component comprises iii. a second detection ligand, wherein the second detection ligand binds to the nucleic acid of interest; and

iv. a second nanorod attached to the second detection ligand, wherein the second nanorod is distinguishable from the first nanorod. In one embodiment, the second molecular motor may comprise an Fi-ATPase. In another embodiment, the Fi-ATPase may be attached to a first member of a binding pair, the second capture ligand may comprise a second member of the binding pair, and the Fi-ATPase may be attached to the second capture ligand through binding of the first member of the binding pair to the second member of the binding pair. In another embodiment, the first member of the binding pair may comprise avidin, and the second member of the binding pair may comprise biotin. In a further embodiment, the second capture ligand may further comprise a first member of a second binding pair, and the second detection ligand may comprise a second member of the second binding pair; in one embodiment, the second member of the second binding pair may comprise an antibody selective for the first member of the second binding pair. In a still further embodiment, the second nanorod may be bound to a third binding molecule, and the third binding molecule may be bound to the second detection ligand.

In a further embodiment of any other embodiment or combination of embodiments, the second capture ligand may comprise a plurality of oligonucleotides that are perfectly complementary to the nucleic acid of interest, wherein upon hybridization to the target nucleic acid, the target-specific nucleic acid probes form a series of target-specific nucleic acid probes directly adjacent to one another

In a second aspect, the present invention provides kits, comprising

(a) a plurality of first components, wherein each first component comprises

(i) a first capture ligand, wherein the first capture ligand binds to a first subunit of a multi-subunit biomolecule target; and

(ii) a first molecular motor attached to the first capture ligand, wherein the first molecular motor is capable of attachment to a substrate; and

(b) a plurality of second components, wherein each second component comprises

(i) a first detection ligand, wherein the first detection ligand binds to a second subunit of the multi-subunit biomolecule target; and

(ii) a first nanorod attached to the first detection ligand.

In a third aspect, the present invention provides methods for biomolecule detection, comprising

(a) contacting the detection device according to any embodiment or combination of embodiments of the first aspect of the invention with a sample under conditions whereby the first capture ligand and the first detection ligand bind to different sited on the biomolecule target, if it is present in the sample; (b) inducing movement of the molecular motor; and

(c) detecting movement of the molecular motor, wherein the movement of the molecular serves to detect the biomolecule target in the sample.

In one embodiment, the test sample may be a bodily fluid sample. In another embodiment, the biomolecule target may be a polypeptide. In a further embodiment, the method may comprise multiplex detection of a polypeptide target and a nucleic acid of interest in the test sample, wherein the detection device comprises

(c) a plurality of third components, wherein each third component comprises

(i) a second capture ligand, wherein the second capture ligand binds to a nucleic acid of interest; and

(ii) a second molecular motor attached to the second capture ligand, wherein the second molecular motor is attached to the substrate; and

(d) a plurality of fourth components, wherein each fourth component comprises

(i) a second detection ligand, wherein the second detection ligand binds to the nucleic acid of interest; and

(ii) a second nanorod attached to the second detection ligand, wherein the second nanorod is distinguishable from the first nanorod.

In a fourth aspect, the present invention provides physical computer-readable storage medium containing instructions executable by a processor that, when executed, cause the processor to perform certain functions on a detection device for carrying out the detection methods of the present invention, wherein the functions include one or more of receiving a signal representing light scattered from the nanorods on the surface of the support as recited in any of the preceding claims, stitching of signals obtained from multiple consecutive fields, and detecting movement of the molecular motor based on the signals so obtained.

Description of the Figures

Figure 1. (A) Stx2 structure. (B) Gold nanorod component for Stx2 detection coated with Protein G (light blue) with bound monoclonal antiStx2A (dark blue). (C) Microscope slide component for Stx2 detection including FiATPase molecular motor containing the biotinylated γ subunit rotor (brown), Avidin (green), biotinylated-Protein G (blue), and monoclonal antiStx2B (red). (D) Stx2 target-dependent nanodevice assembly: (i) application of sample containing Stx2 target to microscope slide component (C); (ii) addition of gold nanorod component (B); (iii) buffer wash. (E) Nanorod rotation distinguishes Stx2 target- dependent nanodevice assembly from nonspecific nanorod binding. Protein structures based on pdblDs 1R4P, 1E79, 1IGT, 1 AVE, 2ZWO.

Figure 2. (A) Field of view using dark field microscopy of nanorods bound to the surface of a microscope slide as the result of Stx2 toxin-dependent nanodevice assembly. (B) Dark-field microscopy image of a 3 mm diameter area of a slide assembled from consecutive, stitched fields of view to observe the area exposed to a ΙΟμΙ_^ sample that was applied to the slide in the presence of a silicone mask that created 3 mm diameter wells on the slide. Area encompassed by 10 and 30 consecutive, stitched fields of view (csfov) is indicated.

Figure 3. Nanodevice assembly -based detection of purified Stx2 Toxin protein. (A) Detection specificity based on the nanorods observed in 10 csfov after addition of ΙΟμΙ_^ of: 3.14 pmol of purified Stx2 protein (+Stx2); 3.14 pmol of purified Stxl protein (+Stxl); or water. Monoclonal anti-Stx2B capture antibody was immobilized as in Figure 1C (+CA), except where indicated (-CA). (B) Purified Stx2 protein detection sensitivity via nanodevice assembly expressed as nanorods/10 csfov after subtraction of an average of 290 nanorods/10 csfov nonspecifically bound nanorods.

Figure 4. (A) Sensitivity of rotation-based detection of purified Stx2 protein. Data are expressed as the number of rotating gold nanorods observed in 30 csfov of the microscope after subtraction of an average of 7.5 nanorods per 30 csfov that showed some variation in red light intensity as a function of time in the absence of Stx2. (B) Intensities of red light scattered from a nanorod viewed by dark field microscopy through a polarizing filter when the rotary position of the nanorod is perpendicular and parallel to the plane of polarization.

Figure 5. (A) Stx2 detection in cell supernatants from 10⁹ cfu of E. coli strains EDL933 and MG1655 after 10-fold dilution into water or saliva, measured as nanorods per 30csfov from ΙΟμΙ of sample. (B) Nanodevice assembly-dependent detection sensitivity of Stx2 measured as the percent increase in nanorods observed in cell supernatant from equivalent cfu of the EDL933 versus the MG1655 strain in 10 csfov. (C) Rotation-dependent detection sensitivity of Stx2 in EDL933 cell supernatant measured in 30csfov.

Figure 6. Self-assembling nanodevices for multiplexed detection of Stx2 protein and stxl gene DNA in one sample well. (A) Slides coated with inter-dispersed Fi molecules containing antiStx2B or exposed Avidin were exposed to samples followed by red and green protein G-coated nanospheres functionalized with antiStx2A and antiFITC, respectively, stxl DNA-specific nanodevice assembly occurred via 5 '-FITC, 3 '-biotin DNA 40mer bridges generated from stxl target by LXR. Detection measured as nanospheres in lOcsfov from cell supernatant samples containing: (B) EDL933 with lOpmole of stxl DNA; (C) EDL933 without stxl DNA; (D) MG1633 with lOpmole of stxl DNA; and (E) MG1633 without stxl DNA.

Figure 7. Detection of the cancer biomarker protein HSP70 using monoclonal capture and detection antibodies to the analyte for attachment to the molecular motor and the nanorod, respectively.

Detailed Description of the Invention

All of these embodiments can be combined with any other embodiment, unless the context clearly dictates otherwise.

In a first aspect, the present invention provides a device, comprising

(a) a substrate;

(b) a plurality of first components, wherein each first component comprises

(i) a first capture ligand, wherein the first capture ligand binds to a first portion of a biomolecule target; and

(ii) a first molecular motor attached to the first capture ligand, wherein the first molecular motor is attached to the substrate; and

(c) a plurality of second components, wherein each second component comprises

(ii) a first nanorod attached to the first detection ligand.

The devices of this aspect of the invention can be used, for example, in detection assays to detect low levels of non-nucleic acid biomolecules in a test sample. Detection of such non-nucleic acid targets is more challenging than nucleic acid targets due to the inability to amplify the number of non-nucleic acid biomolecules.

As used herein, "biomolecule" is any biomolecule except a nucleic acid. In various non-limiting embodiments, the biomolecule comprises polypeptides, polysaccharides, lipids, metabolites, isoprenoids, etc. The biomolecule may be a single subunit molecule, so long as the capture and detection ligands bind to different and non-overlapping domains of the biomolecule. In another embodiment, the biomolecule is a multi-subunit biomolecule; in one embodiment, the capture ligand and detection ligands bind to different subunits of the multi- subunit biomolecule. As used herein "attached" means attached in any way, whether directly (for example, via covalent binding) or indirectly (for example, via a linker).

As used herein, the "capture ligand" is anything that is capable of binding to a first site on the biomolecule target of interest (directly or indirectly) and also to the molecular motor (directly or indirectly). As used herein, the "detection ligand" is anything that is capable of binding to a second site on the biomolecule target (directly or indirectly) and to the nanorod (directly or indirectly. The first capture ligand can bind to the molecular motor and the first detection ligand can bind to the nanorod either directly (for example by a covalent bond) or indirectly through another molecule. In one embodiment, the first capture ligand (or detection ligand) binds directly to the biomolecule target; in an exemplary embodiment, the first capture ligand may comprise an antibody binding to a first site on the biomolecule and the detection ligand may comprise an antibody binding to a second site on the biomolecule. In another embodiment, the first capture ligand (or detection ligand) binds directly to the biomolecule target and to an affinity target, wherein the affinity target is bound to the molecular motor (or the nanorod). Together, the capture ligand (or detection ligand) and affinity target make up a binding pair. Either member of a binding pair can be used as a capture ligand (or detection ligand), and either member can be used as an affinity target. An affinity target includes both separate molecules and portions of molecules, such as an epitope of a protein that interacts specifically with a capture ligand (or detection ligand). Antibodies, either member of a receptor/ligand pair, and other molecules with specific binding affinities can be used as affinity tags. Binding an affinity tag to biomolecule target thus permits an indirect linkage between the biomolecule and the molecular motor or the nanorod.

A non-limiting example of a binding pair is biotin/avidin. Other non-limiting binding pair examples include Protein G and an antibody specific for the biomolecule of interest, and other antigen/antibody pairs. Epitope tags, such as a his-tag, and antibodies directed against the epitope tag (or fragments thereof) are further examples of binding pairs for use with the methods of the present invention. Those of skill in the art will understand that certain embodiments listed herein as indirect binding of the affinity tag and the molecular motor or detection ligand can also be used for direct binding embodiments. For example, where the affinity tag is an epitope tag as described above, the detection ligand can be a labeled antibody against the epitope tag. Many further such examples will be readily apparent to those of skill in the art.

The first capture ligand and the first detection ligand may be the same or different as is most suitable for their ultimate attachment to the specific molecular motor and the nanorod employed. In one embodiment, the biomolecule target is a polypeptide, and the first detection ligand comprises an antibody that binds to a first site on the polypeptide, and the detection ligand binds to a second site on the polypeptide. In another embodiment, the molecular motor (such as an Fi-ATPase) is attached to a first member of a first binding pair, wherein the first member of the first binding pair is bound to a second member of the first binding pair, and wherein the second member of the first binding pair is attached to the first capture ligand. In a further embodiment, the second member of the first binding pair is bound to a first binding molecule, and wherein the first binding molecule is bound to the first capture ligand. Any suitable first binding molecule can be used. In one embodiment, the first binding molecule provides specificity for a given capture ligand and/or second member of the first binding pair. In another embodiment, the first binding molecule is a general binding molecule, including but not limited to protein G, protein A, and avidin, permitting multiple reuse of the device. In a further embodiment of any of these embodiments, the first nanorod is attached to a second binding molecule, and wherein the second binding molecule is bound to the first detection ligand. In one embodiment, the second binding molecule provides specificity for a given detection ligand and/or nanorod. In another embodiment, the second binding molecule is a general binding molecule, including but not limited to protein G, protein A, and avidin, permitting multiple reuse of the device.

In one exemplary embodiment where the biomolecule target is a polypeptide, biotinylated-Protein G is allowed to bind to the avidin attached to the Fi-ATPase on the substrate. Protein G binds tightly to a wide variety of antibodies in a manner that does not interfere with the antigen binding sites of the antibodies. The biotinylated Protein G is then immobilized to a capture antibody specific for the intended target protein. The nanorods (such as gold nanorods) are coated with non-biotinylated Protein G to which a detection antibody is attached. The detection antibody is specific to a different domain of the target protein from the capture antibody. The use of Protein G increases the versatility of the device because the same substrates and nanorods can be adapted to detect any protein target for which two non-overlapping antibodies (such as monoclonal antibodies) can be obtained.

The first molecular motor can be any biological or synthetic molecule capable of induced translational or rotational movements that are capable of detection. In a preferred embodiment, the molecular motor comprises a biomolecular motor. Non-limiting examples of such biomolecular motors comprise Fi-ATPase, actomyosin, ciliary axonemes, bacteria flagellar motors, kinesin/microtubules, and nucleic acid helicases and polymerases. In a preferred embodiment, the molecular motor comprises an Fi-ATPase. In a further preferred embodiment, the Fi-ATPase is attached to the substrate. In a further embodiment, the first capture ligand attaches to a moiety on the moving component of the Fi-ATPase. By way of example, the rotating subunits on the Fi-ATPase include the γ and ε subunits, while the α, β, and δ subunits do not rotate. Thus, where the Fi-ATPase is used as the molecular motor, it is preferred that the first capture ligand binds (directly or indirectly) to the γ and/or ε subunits, while the molecular motor is attached via a functional moiety, such as a his-tag, to a substrate through the α, β, and δ subunits. In a preferred embodiment, the Fi-ATPase rotary biomolecular motor is a complex of α₃β₃γδε, α₃β₃γδ, or a_{3 3}y. subunits. This complex provides optimal binding of the Fi-ATPase to the solid support and the γ subunit to the biomolecule. In another preferred embodiment, a linker (including but not limited to a 6-his-tag) is added to the base of the stator of the Fi-ATPase to provide for binding of the Fi-ATPase to the support (such as a nickel-NTA-coated substrate) in an orientation such that the rotary axle of Fi faces away from the surface. Details of Fi-ATPase assembly, subunit composition, and inducement of Fi-ATPase rotation are well known to those of skill in the art; see, for example, US 20030215844; Yoshida et al, Journal of Biological Chemistry, 252:3480-3485 (1977); Du et al, Journal of Biological Chemistry, 276: 1 1517-11523 (2001); Bald et al, Journal of Biological Chemistry, 275: 12757-12762 (2000); Kato-Yamada et al, Journal of Biological Chemistry, 273 : 19375-19377 (1998), Kato et al, Journal of Biological Chemistry, 272:24906-24912 (1997); Tucker et al, Journal of Biological Chemistry, 279:47415-8 (2004); Tucker et al, Eur. J. Biochem. 268:2179-86 (2001), and Du et al, Journal of Biological Chemistry, 276: 11517-23 (2001). US 20090035751 and US20100015616, incorporated by reference herein in their entirety, provides non-limiting examples of Fi- ATPase subunits that can be used in the methods and compositions of the present invention, and other details on molecular motor design, each of which can be used in the present invention.

Any suitable nanorod can be used in the devices of the invention. In a preferred embodiment, the nanorod is an elemental metal nanorods, including but not limited to gold, silver, aluminum, platinum, copper, zinc, and nickel nanorod. The nanorod may be any length or shape (such as rod-shaped) suitable for an intended purpose (ie: 1 nm-999 nm). In various embodiments, the nanorod is between 10 nm-500 nm, 20 nm-400 nm, 30 nm-300 nm, 40 nm -250 nm, 50 nm-200 nm, etc., in length.

In a preferred embodiment, ATPase substrate density is such that spacing between Fi molecules is the minimal distance where one nanorod is incapable of binding to two Fi molecules to form a bridge. This maintains a 1 : 1 molecule ratio between the nanorods observed and the target molecules present. The distance will vary depending on the size of the nanorods used and the size of the target. For example, when nanorods that are about 75 nm long are used, the Fi molecules are preferably be separated on the substrate by 150 nm or more. In another embodiment, nanorods 35 nm long are used, the Fi molecules are preferably separated on the substrate by 70 nm or more. Those of skill in the art will recognize, based on the teachings herein, that many such permutations are possible.

The molecular motors (such as Fi-ATPases) can be present on the substrate in any total number suitable for a given application, depending on the size of the substrate. In a further embodiment, the molecular motors each are bound to between 1 and 5 first capture ligands. In a preferred embodiment, the first capture ligand comprises an antibody and the molecular motor binds to between 1 and 5, 1-4, 1-3, 1-2, or 1 antibody(ies); most preferably to 1 antibody when the molecular motor comprises an Fi-ATPase, given the relative size of the antibody to the Fi-ATPase.

The nanorod is bound to any number of detection ligands suitable for an intended use. In one embodiment, the surface of the nanorod is coated with detection ligands, and thus the number of detection ligands depends on the size of the nanorod used. It is preferred to have the nanorods bound to as many detection ligands (such as antibodies) as possible, to increase the chances it will bind to the biomolecule target.

As used herein, a "substrate" comprises a solid surface, with molecular motors or nanorods attached to said surface. In one embodiment, the substrates comprise a plurality of molecular motors linked to different capture ligands that are coupled to a surface of a substrate in different, known locations. For example, there are several silane derivatives to attach a variety of functional groups to a glass surface. The term "solid surface" as used herein refers to a material having a rigid or semi-rigid surface. Such materials will preferably take the form of chips, plates, slides, cover slips, small beads, small magnetic beads, pellets, disks or other convenient forms, although other forms may be used. Such solid surfaces can be coated in any way that improves desired binding to its surface and/or minimizes nonspecific binding to its surface. In a preferred embodiment, nickel-nitrilotriacetic acid (Ni- NTA) affinity resin (Sigma-Aldrich product #P661 1) is used. In a further embodiment, acetylated BSA can be added to reduce non-specific binding. The substrate can be of any size suitable for an intended use and detection technique to be used.

In a preferred embodiment, the biomolecule target comprises a polypeptide. The device may target any polypeptide capable of binding (directly or indirectly) at a first site to a detection ligand and at a second site to a capture ligand. Thus, the polypeptide may be a single subunit protein with distinct binding sites for the capture ligand and the detection ligand, or may be a multi-subunit protein where a binding site (direct or indirect) for the capture ligand may be, for example, on a first subunit, and a binding site (direct or indirect) for the detection ligand may be, for example, on a second subunit. In one non-limiting embodiment, the multi-subunit protein is the Stx2 Shiga toxin produced from E. coli 0157:H7. Shiga toxin-producing E. coli (STEC) strains are the leading cause of bacterial enteric infections in the USA from which the pathogenic 0157:H7 strain is the most widely recognized. The common trait among STEC strains is the production of the cytotoxins known as the Stxl and Stx2 Shiga toxins that have a subunit composition of A,Bs_. In this embodiment, it is further preferred that the first capture ligand comprises an antibody selective for a Stx2B subunit (SEQ ID NO:2) of the Stx2 toxin, and the first detection ligand comprises an antibody selective for a Stx2A subunit (SEQ ID NO: 1) of the Stx2 toxin. In another non-limiting embodiment, the polypeptide is HSP70 (SEQ ID NO:4).

The devices of the invention are suitable for multiplex detection, using different devices on the substrate that are designed to detect different biomolecule targets. Similarly, the devices can be designed to detect both a polypeptide target of interest, and a nucleic acid of interest in a sample. Devices for this purpose are as described above, but further comprising

(c) a plurality of third components, wherein each third component comprises

(d) a plurality of fourth components, wherein each fourth component comprises

Thus, in this embodiment, the substrate further comprises second population of molecular motors (such as Fi-ATPases) that are bound (directly or indirectly) to a second capture ligand that binds a nucleic acid of interest (where the polypeptide target is bound, directly or indirectly, by the first capture ligand on the first population of devices on the substrate). Similarly, the second nanorod is attached (directly or indirectly) to a second detection ligand which binds to a nucleic acid target of interest. In this embodiment, the nucleic acid of interest can be any suitable nucleic acid, and may be DNA, RNA, single stranded, double stranded, etc. In one embodiment, the nucleic acid of interest is a nucleic acid encoding all or a part of the polypeptide target being detected. Exemplary nucleic acids are those encoding all or part of the Stx2A subunit or Stx2B subunit (SEQ ID NO:3) and HSP70 (SE ID NO:5).

In one preferred embodiment, the second capture ligand and the second detection ligand comprise first and second target-specific nucleic acids, wherein the first and second target-specific nucleic acids each comprise sequences complementary to the nucleic acid of interest; wherein the first target specific nucleic acid is bound to a first affinity tag and the second target-specific nucleic acid is bound to a second affinity tag, wherein the first affinity tag binds to the molecular motor, and wherein the second affinity tag binds to the nanorod.

In another embodiment, the second capture ligand can be the same as the first capture ligand, for example, the case in which two nucleotide target sequences have the same sequence at one end (e.g. 5') but differ at the other end (e.g. 3 '). In this embodiment, two different detection ligands would be used.

In one embodiment, the second capture ligand comprises a plurality of

oligonucleotides that are perfectly complementary to the nucleic acid of interest, wherein upon hybridization to the target nucleic acid, the target-specific nucleic acid probes form a series of target-specific nucleic acid probes directly adjacent to one another. In another preferred embodiment, the first and second target-specific nucleic acids are capable of hybridizing to the target nucleic acid such that upon hybridization to the target nucleic acid the first and second target-specific nucleic acids are directly adjacent to each other. See, for example, US 20090035751 and US20100015616, incorporated by reference herein in their entirety. In another embodiment, the second capture ligand comprises a plurality of oligonucleotides that are perfectly complementary to the nucleic acid of interest, and further provide a specific binding site for a protein.

In this embodiment, the second nanorod is distinguishable from the first nanorod. Any means for distinguishing the nanorods can be used. For example, as is well known to those of skill in the art, nanorods can be made in a variety of colors for use in multiplexing target analysis in a single well. Since the light scattered from the nanorods is diffraction- limited, each nanorod will give rise to a round spot of light with a diameter that is about half the wavelength of the light responsible for its color, and thus their movement can be distinguished, as described in more detail below. All embodiments of the molecular motors, capture ligands, detection ligands, etc. as disclosed above can be used in these embodiments as well. Thus, in one embodiment, the second molecular motor comprises an Fi-ATPase. In a further embodiment, the Fi-ATPase is attached to a first member of a binding pair, the second capture ligand comprises a second member of the binding pair, and the Fi-ATPase is attached to the second capture ligand through binding of the first member of the binding pair to the second member of the binding pair. In another embodiment, the first member of the binding pair comprises avidin, and the second member of the binding pair comprises biotin. In another embodiment, the second capture ligand further comprises a first member of a second binding pair, and wherein the second detection ligand comprises a second member of the second binding pair. In a further embodiment, the second member of the second binding pair comprises an antibody selective for the first member of the second binding pair. In a still further embodiment, wherein the second nanorod is bound to a third binding molecule, the third binding molecule is bound to the second detection ligand.

In a second aspect, the present invention provides kits comprising

(a) a plurality of first components, wherein each first component comprises

(ii) a first capture ligand, wherein the first capture ligand binds to a first site on a biomolecule target; and

(b) a plurality of second components, wherein each second component comprises

(i) a first detection ligand, wherein the first detection ligand binds to a second site on the biomolecule target; and

(ii) a first nanorod attached to the first detection ligand.

The kits of this second aspect of the invention can be used, for example, to prepare the devices of the first aspect of the invention. All embodiments and combinations of embodiments of the capture ligands, detection ligands, nanorods, and molecular motors disclosed herein are suitable for use in the kits of the invention. Similarly, all definitions of these components and the biomolecules as used above apply throughout the application.

(a) contacting the detection device according to any embodiment or combination of embodiments of the invention with a sample under conditions whereby the first capture ligand and the first detection ligand bind to different sited on the biomolecule target, if it is present in the sample;

(b) inducing movement of the molecular motor; and

The methods of this aspect of the invention can be used, for example, to detect low levels of non-nucleic acid biomolecules in a test sample. Detection of such non-nucleic acid targets is more challenging than nucleic acid targets due to the inability to amplify the number of non-nucleic acid biomolecules. In a preferred embodiment, the biomolecule is a polypeptide. In one such embodiment, the polypeptide is a protein (including but not limited to a multi-domain protein) and the detection ligand and capture ligand bind to different sites on (or different domains of) the protein, as described above. All embodiments and combinations of embodiments of capture ligands, detection ligands, molecular motors, affinity tags, substrates, etc., as described herein, can be used in the methods of the invention. In a preferred embodiment, the molecular motor comprises an Fi-ATPase, including but not limited to all embodiments and combinations of embodiments of Fi-ATPases described herein.

Any suitable test sample may be used, including but not limited to a body fluid sample (saliva, urine, whole blood, plasma, tears, sputum, vaginal secretions, breast milk, etc.), water samples (including but not limited to water samples from ponds, streams, lakes, oceans, seas, wastewater, reservoirs, drinking water, water distribution pipeline, etc.) as well as any other types of liquid, such as beverage samples, liquid medicine samples, as well as any other liquid borne samples. A test sample does not need to initially be a liquid, but instead can be derived from a non-liquid sample (ie: solid or gas) that is subsequently placed in a liquid prior to detection. In one non-limiting example, a food sample can be swabbed and the swab can subsequently be inserted into a liquid (ie: water, buffer, etc.) to generate a liquid test sample from a solid. Those of skill in the art will understand how to prepare other such non-liquid samples for use as a test sample based on the teachings herein. The non- liquid test samples can be any sample of interest, including but not limited to, food samples, environmental samples (from, e.g., medical centers such as linens, medical devices, etc.); pharmaceutical facilities (from, e.g., manufacturing or processing lines); food production facilities; livestock facilities; solid waste samples and diagnostic samples. Analysis of large sample volumes typically will be carried out using a substrate with a large surface area, while applications that analyze a small volume of liquid will typically utilize a substrate with a smaller surface area.

In one exemplary and non-limiting embodiment where the biomolecule target is a polypeptide, the first components comprise biotinylated-Protein G bound to avidin attached to an Fi-ATPase molecular motor, where the Fi-ATPase is attached to the substrate. Protein G binds tightly to a wide variety of antibodies in a manner that does not interfere with the antigen binding sites of the antibodies. The biotinylated Protein G is also bound to a capture antibody specific for the intended target polypeptide. The nanorods (such as gold nanorods) are also coated with non-biotinylated Protein G to which a detection antibody is attached. The detection antibody is specific to a different domain of the target protein from the capture antibody. The use of Protein G increases the versatility of the device because the same substrates and nanorods can be adapted to detect any protein target for which two non- overlapping antibodies (such as monoclonal antibodies) can be obtained.

Inducing movement of the molecular motor is done by standard methods in the art for a given molecular motor. For example, the movement of Fi-ATPases is induced by adding ATP using standard techniques (Noji, FL, Yasuda, R., Yoshida, M. and Kinosita, K. (1997) Nature 386, 299-302). Suitable concentrations of ATP for use in the methods of the invention range from 1 uM to 2 mM; preferably between 200 uM and 1 mM The rate of rotation of the Fi-ATPase can be controlled by the ATP concentration used. For example, some detection methods are capable of detecting greater rates of rotation than others, and thus the specific concentration of ATP used will depend in part on the detection technique to be employed.

Those of skill in the art are able to determine how to induce movement of other known molecular motors using similar published protocols. The only motion that will be detected will result from molecular motors that are connected to the nanorod through the bridge formed by the binding of the first and second components to the target biomolecule. Thus, observation of this motion will identify the presence of the target biomolecule.

Detecting movement of the molecular motor through nanorod can be accomplished by any suitable means. In one embodiment, direct visualization of the movement is used. In a preferred embodiment, elemental metal nanorods capable of visual observation by microscope are used. Other means of observation include, but are not limited to dark field microscopy and the methods disclosed in US 20090035751 and US20100015616, incorporated by reference herein in their entirety.

In a preferred embodiment, metal (such as gold) nanorods are used with visible light (400-700 nm wavelength range) to detect the rotation, to provide improved detection capability (See, for example, WO 2004/053501, US 20090035751 and US20100015616). The light scattered from the nanorods is polarized with the longer and shorter wavelengths scattered from the long and short axes, respectively, of the rod. When viewed through a polarizing filter, the intensity of scattered light depends on the angle of the rod relative to the direction of the filter. The light scattered from the long and short axes of the rod is observed to have a maximum value when those axes are parallel to the direction of the filter and a minimum when perpendicular to the filter. For example, if the long and short wavelengths of scattered light are red and green, respectively, the intensity of the red will be maximum when the green is minimum. Thus, rotation of a metal nanorod viewed through a polarizing filter will appear to blink red and green. In this embodiment, monitoring the oscillation of intensity of both the red and green light as the nanorod rotates provides independent confirmation that the rod is rotating. In a further preferred embodiment, the oscillation of intensity of light of only one wavelength is measured, which further improves signal to noise ratios. In these embodiments, measurements can be made using both wavelengths (using, for example, a beamsplitter or a color camera) or just one wavelength of light (using, for example, a green or red filter). Some digital cameras are limited with regard to the frame rate (speed of data collection) at which the camera is still sensitive enough to measure the intensity oscillations from the rotating nanorods. Single photon counters can be used to make the oscillation measurement. The pin hole acts as a camera obscura and the oscillation of only one rod at a time can be measured; it is capable of much greater frame rates at much higher signal to noise. Digital cameras can collect oscillation data on many nanorods at once, while the speed and sensitivity of the camera only needs to be sufficient to capture the rate of rotation of the rod. The preferred oscillation rate is one that is easily measured with the detection device used to make the measurement. In these embodiments employing elemental metal nanorods, dark field microscopy is the preferred detection method, because only the light scattered off the nanorods is observed, further improving signal to noise ratios. In another embodiment, detection is performed using light field microscopy.

The methods of the invention are suitable for multiplex detection, using different devices on the substrate that are used to detect different biomolecule targets. Similarly, the methods can be used to detect both a polypeptide target of interest, and a nucleic acid target of interest. Methods for this purpose are as described above, but where the detection device comprises

(c) a plurality of third components, wherein each third component comprises (i) a second capture ligand, wherein the second capture ligand binds to a nucleic acid of interest; and

(d) a plurality of fourth components, wherein each fourth component comprises

Thus, in this embodiment the methods further comprise detecting movement of the second molecular motor, which serves to detect the nucleic acid of interest. All other embodiments of the detection device for multiplex analysis as disclosed above can be used in these methods. In one preferred embodiment, the second capture ligand and the second detection ligand comprise first and second target-specific nucleic acids, wherein the first and second target-specific nucleic acids each comprise sequences complementary to the nucleic acid of interest (such as, but not limited to, as a nucleic acid encoding at least a portion of the polypeptide); wherein the first target specific nucleic acid is bound to a first affinity tag and the second target-specific nucleic acid is bound to a second affinity tag, wherein the first affinity tag binds to the molecular motor, and wherein the second affinity tag binds to the nanorod.

In another embodiment, the second capture ligand comprises a plurality of oligonucleotides that are perfectly complementary to the nucleic acid of interest, wherein upon hybridization to the target nucleic acid, the target-specific nucleic acid probes form a series of target-specific nucleic acid probes directly adjacent to one another. In another preferred embodiment, the first and second target-specific nucleic acids are capable of hybridizing to the target nucleic acid such that upon hybridization to the target nucleic acid the first and second target-specific nucleic acids are directly adjacent to each other. See, for example, US 20090035751 and US20100015616, incorporated by reference herein in their entirety.

In these embodiments, the second nanorod is distinguishable from the first nanorod. Any means for distinguishing the nanorods can be used. For example, as is well known to those of skill in the art, nanorods can be made in a variety of colors for use in multiplexing target analysis in a single well. Since the light scattered from the nanorods is diffraction- limited, each nanorod will give rise to a round spot of light with a diameter that is about half the wavelength of the light responsible for its color, and thus their movement can be distinguished. In the various embodiments of the methods of the invention, the support may contain any number/density of detection devices as desired for a given application, as described above.

In one non-limiting exemplary detection means, dark field microscopy is used to acquire the data, such that only the light scattered from the nanorods on the surface of the slide is recorded by the CCD camera. A motorized microscope stage driven by computer software enables multiple consecutive fields of view to be stitched together to eliminate overlap. This increases the accuracy and sensitivity of the detection. Any suitable number of such "stitched fields" can be used. In one embodiment, 10-30 consecutive stitched fields of view are analyzed.

In one non-limiting embodiment, multiplex detection of the Stx2 protein toxin and the DNA sequence to detect the stxl gene DNA sequence in the same well are shown in the examples that follow. The support surface in each well contains Fi-ATPase that was modified in one of two ways. One field of Fi-ATPase molecules contained an exposed avidin for DNA detection, and the second superimposed field of Fi-ATPase contained an exposed capture antibody for Stx2 protein detection. Samples of supernatant from is. coli 0157:H7 were incubated in the wells followed by the addition of a solution containing nanorods. The red and green nanorods added to each well were coated with detection antibodies for the Stx2 protein and for FITC, respectively. When the stxl gene target was present, it was converted to 3'biotinylated, 5' FITC-labeled DNA for nanodevice assembly thereby enabling it to assemble only with green nanorods. The data shown below demonstrates that specific protein and DNA targets can be detected and distinguished from each other when probed simultaneously in the same well.

In a fourth aspect, the present invention provides a physical computer-readable storage medium contains instructions executable by a processor that, when executed, cause the processor to perform certain functions on a detection device for carrying out the detection methods of the present invention. The functions include, but are not limited to, one or more of receiving a signal representing light scattered from the nanorods on the surface of the support, stitching of signals obtained from multiple consecutive fields, and detecting movement of the molecular motor based on the signals so obtained. Example 1. Single-molecule protein detection by molecular motor-driven nanodevices Abstract

Specific proteins made in small amounts from a few cells can have dramatic and rapid physiological effects on an entire organism. For example, Stx2 toxin produced by infection from as few as ten E. coli 0157:H7 cells may be lethal if not treated within 24 hours.

However, protein analyte detection is a challenge because, unlike DNA that can be amplified by PCR, the number of protein molecules cannot be increased. Here we present a molecular motor-powered nanodevice that detects single molecules of the Stx2 protein Shiga toxin from E. coli 0157:H7 in ~25 minutes in complex samples. Nanodevice assembly of a microscopy- visible gold nanorod and the Fi-ATPase motor is target mediated. Inducing Fi -dependent nanorod rotation discriminates analyte-dependent nanodevices from background, dramatically improving the LOD. Use of Protein G provided target flexibility with a common detection platform that enabled simultaneous detection of Stx2 protein and the stxl DNA sequence.

Introduction

The production of small amounts of protein by just a few specific cells can have profound effects on the normal development or pathology of an entire organism¹. A single neuron can affect an entire neural circuit culminating in motor action², or affect behavioral responses in rats³. In vivo transplantation of individual adult stem cells can give rise to an entire prostate⁴, or a functioning mammary gland⁵. Small numbers of cells can be identified by their DNA signature by first amplifying the DNA copy number via techniques like PCR. However, detecting protein targets remains a much greater challenge due to the inability to amplify the number of copies of target molecules. An average mammalian cell has been determined to contain only about 100,000 copies of each of 1,000 proteins⁶. Unfortunately, the protein from about 5000 cells is required to detect a protein analyte using a typical Western blot assay⁷.

Due to the secretion of small amounts of Shiga toxin⁸'⁹, the most important pathogen infection by enterohemorragic strains of E. coli like 0157:H7 require only 10-1000 cells for infection¹⁰, which occurs most often by the consumption of contaminated food. Children and elderly often develop hemolytic uremic syndrome (HUS), a serious life-threatening complication for which no effective therapy exists. Although Shiga toxins Stxl and Stx2 have a subunit composition of A1B5 (Figure 1A) and 60% sequence homology, Stx2 most often associated with HUS.

The effects of a lethal challenge of 0157:H7 can be ameliorated in mice by anti-Stx2, but only if treatment occurs within 24 hours of infection¹¹. Unfortunately, the symptoms develop in 2-5 days, and there is no chance for antibodies to be effective once significant amounts of Stx2 enter the bloodstream and cause irreversible tissue damage¹²'¹³. Thus, the timely diagnosis of the presence of small amounts of Stx2 toxin is imperative.

We now report the design and implementation of nanodevices capable of detecting single molecules of Stx2 shiga toxin protein from E. coli 0157:H7, as well as the multiplex detection of the Stx2 protein and the stxl gene DNA sequence in the same well at the same time. We can detect the toxin secreted into crude cell supernatant in ~25 min from as few as 10 cfu, each of which has secreted -340 toxin molecules. This suggests that ingestion by as few as -3400 toxin molecules is enough to elicit significant illness. Since nanodevices are assembled using Protein G that bind most antibodies, the detection system is easily adapted to detect a wide variety of proteins for different applications.

Nanodevice Design and Assembly

The nanodevices were designed to undergo analyte-dependent assembly of two components that can be prepared in advance and stored until use. To detect Stx2 protein, the first component was a preparation of 75nm X 35nm gold nanorods coated with non- biotinylated Protein G to which monoclonal anti-Stx2A (the detection antibody) was immobilized (Figure IB). The second component was comprised of a molecule of anti-Stx2B monoclonal antibody (the capture antibody), which was attached to a molecule of the Fi- ATPase molecular motor. Directed-assembly of the capture antibody and Fi on a microscope slide with the specific orientation shown in Figure 1C was facilitated by several factors. Six histidines inserted at the N-terminus of the Fi-a subunit (a 6-his tag) enabled attachment of Fi to a nickel-NTA coated microscope slide so that the γ subunit rotor faced away from the surface. A cysteine created by mutagenesis on the Fi-γ subunit was modified to contain a biotin that enabled the binding of avidin, which in turn served as the docking site for biotinylated Protein G. The capture antibody for the target, anti-Stx2B, was immobilized to the molecular motor via the Fi-bound Protein G. In this manner, a field of motors each containing a molecule of capture antibody was randomly dispersed over the surface of the slide as shown in the top slide of Figure ID. The Stx2 toxin-dependent assembly of the nanodevices on the microscope slide was accomplished by the sequential addition of the sample to be analyzed for the presence of target followed by addition of the functionalized gold nanorods (Figure ID). After each addition, the slide was flushed with buffer to minimize the number of target molecules and nanorods nonspecifically bound to the surface of the slide. Although the motors were randomly dispersed on the slide surface, the average spacing between Fi molecules was optimized to minimize the incidence in which a nanorod bridged between two immobilized target molecules. Consequently, the detection platform was designed such that number of nanorods observed by dark field microscopy reported the number of Stx2 toxin molecules present. When the number of assembled nanodevices, as measured by the number of nanorods visible by microscopy, is used as a measure of analyte detection, nonspecific binding of nanorods to the surface sets the lower limit of detection. However, this detection limit may be circumvented by adding ATP to fuel rotation catalyzed by the Fi-ATPase molecular motor, since nanorod rotation reports correctly assembled nanodevices that contain a molecule of the analyte (Figure IE).

Figure 2 A shows a field of view of the microscope following analyte-dependent nanodevice assembly. By using dark field microscopy, the CCD camera recorded only the light scattered from the nanorods on the surface of the slide. Nanorods are too small to discern their shape or to easily resolve the small spatial movements that result from rotation. The length of a gold nanorod is a major determinant of the wavelength of light scattered from it (14) such that each -75x35 nm nanorod shown in Figure 2A appears as a discrete round spot with a diameter that is about half the wavelength of the scattered light. The spots arising from single nanorods were easily distinguished from nanoscale-inclusions in the glass surface that appear as irregularly-shaped white spots.

To analyze multiple samples on the same microscope slide, a removable silicone mask perforated with 3mm diameter holes was applied to the surface of the slide. This created an array of wells that enabled analysis of ΙΟμΙ samples. A motorized microscope stage driven by computer software enabled multiple consecutive fields of view to be stitched together to eliminate overlap and avoid counting the same nanorod twice. The area of the slide exposed to sample in each well created by the silicone mask was approximately equivalent to 900 stitched fields of view (Figure 2B).

Performance of Nanodevice-Based Stx2 Protein Detection. The ability of the nanodevice to discriminate between purified Stxl and Stx2 toxins that have a -60% amino acid sequence homology is shown in Figure 3A. In the absence of the Stx2 target, the average number of nonspecifically bound nanorods observed in 10 stitched consecutive fields of view (csfov) was -400. This was comparable to the number of nanorods observed after a 10 μΐ sample containing 3.14 nM of purified Stxl toxin was examined. Addition of the same amount of purified Stx2 target resulted in more than a 5 -fold increase in the number of nanorods observed. Thus, the nanodevice is capable of distinguishing between the Stx2 target and a closely related Stxl toxin. It is noteworthy that omission of the capture antibody from the slides decreased nonspecific binding of nanorods to the surface to about 100 (-10 nanorods per field of view).

Figure 3B shows the number of nanorods observed in 10 csfov as a function of the concentration of purified Stx2 target protein after subtraction of nanorods bound in the absence of target. The number of assembled nanodevices increased in the presence of increasing Stx2 over a range of more than 7 orders of magnitude of analyte concentration. The slides used in the experiments of Figure 3B were prepared so that each field of view had an average of -250 immobilized Fi-ATPase molecules containing bound capture antibody. This provided an average spacing between each capture antibody that minimized the occurrence of a nanorod becoming bound to two immobilized Stx2 molecules such that there was a 1 : 1 ratio between observed nanorods and Stx2 molecules. When the amount of Stx2 exceeded 3 pmol, the increase in the assembled nanodevices approached an apparent maximum of -2500 in 10 csfov. Thus, in the presence of such high amounts of toxin, the majority of available Fi-ATPase molecules had captured a Stx2 molecule within the 3 mm diameter surface area of the microscope slide to which the analyte had been exposed. The larger errors in the measurements at the highest concentrations were then the result of the variability in the number of Fi molecules per field of view. The lower limit of detection when samples were examined in this manner was -1 fmol of Stx2 in the ΙΟμΙ of sample analyzed.

The rotation-based detection limit of the system using purified Stx2 protein is shown in Figure 4A. The observation of a rotating nanorod confirms that the nanorod is bound to the molecular motor via a molecule of the analyte. As shown in Figure 4B, the intensity of the red light scattered from a nanorod viewed through a polarizing filter is maximal and minimal when the long and short axes of the nanorod are aligned with the polarizer, respectively^15"17. Consequently, a rotating nanorod powered by the Fi motor is easily observed through a polarizer by a pronounced periodic oscillation in intensity of scattered light. Upon analysis of 30 csfov, an average of 7 nanorods were observed to rotate when the 3 mm diameter surface was exposed to as few as 13.5 zmol (8130 molecules) of Stx2 target in a 10 μΐ sample. Thus, analysis of rotating nanorods increased the detection sensitivity by about 5 orders of magnitude. Since 30 csfov is about 3.3 % of the total surface area exposed to the sample, then of the 8130 target molecules added, a total of about 210 target molecules (-2.5%) are estimated to have been incorporated into rotating nanodevices on the entire surface.

Detection of Stx2 Protein Target in Complex Samples.

Detecting Stx2 toxin in complex samples of either cell supernatant or saliva provides a measure of the ability of the diagnostic test to work under conditions that more closely represent those obtained from patients. Such real-world samples contain a variety of bacteria, glycoproteins, enzymes, and antibodies that may interfere with target detection. A

comparison of the effect of cell supernatant and saliva on the nonspecific binding of nanorods to the surface is shown in Figure 5A. Samples of crude cell supernatant from the EDL933 toxin-producing strain of E. coli and the MG1655 non-toxin producing strain were diluted ten fold into either water or saliva prior to analysis for the presence of the Stx2 toxin by counting the number of nanorods observed in 30 csfov. Prior to analysis, protein G was added to the samples to bind to the large abundance of antibodies present in saliva. When this step was omitted, the antibodies present in saliva increased nonspecific binding of nanorods to the surface in the saliva-containing samples, which interfered with analyte detection.

In the absence of saliva, the amount of Stx2 toxin in the cell supernatant produced from the EDL933 strain was enough to increase the number of nanorods observed by almost an order of magnitude above that observed in the MG1655 (Stx2 control) sample (Figure 5A). After the pretreatment with Protein G, samples containing saliva and cell supernatant from the toxin-producing strain (EDL933) were clearly distinguished from the saliva samples with the MG1655 supernatant even though the presence of saliva decreased the signal-to- noise of Stx2 detection relative to that observed for the same amount of cell supernatant analyzed in the absence of saliva.

Figure 5B shows the number of nanorods observed in 10 csfov as a function of the concentration of cell forming units of the Stx2-forming EDL933 strain in crude supernatant after subtraction of nanorods bound in the absence of target. The amount of Stx2 produced by the cells measured by the assembly of nanodevices resulted in a lower detection limit of <10⁸ cfu EDL933/ml (10⁶ cfu in the 10 μΐ sample). As shown in Figure 5C, this detection limit was improved by 4 orders of magnitude to a limit of -1000 cfu EDL933/ml (10 cfu total per sample) by measuring the number of rotating nanorods in 30 csfov. The number of rotating nanorods observed was about half the number observed in the presence of 13.5 zmole of purified Stx2. On this basis, it is estimated that about 350 Stx2 molecules were produced per cfu.

Multiplex Detection of Stx2 toxin and the stxl gene DNA sequence.

To detect Stx2 protein toxin and the stxl DNA gene sequence simultaneously in the same well, the slide surface was coated with randomly inter-dispersed fields of Fi molecules that contained an exposed capture antibody for Stx2 protein detection, or an exposed avidin for stxl gene detection (Figure 6A). For stxl gene detection, the ligation exonuclease reaction (LXR) was used to make a 5' FITC, 3' biotinylated DNA 40mer, a "DNA bridge", from each copy of the stxl gene present in the sample as described previously¹⁴ that enabled nanodevice assembly. The bridges were made by probing samples of either EDL933 or MG1655 supernatant, the latter containing the stxl target DNA fragment, with 5' FITC- labeled 20mers and 3 ' biotinylated, 5' phosphorylated 20mers complimentary to adjacent sequences specific for stxl. After the hybridized probes were ligated, and the original gene sequence was eliminated by exonuclease, exposure of the sample to the slide allowed the assembly of the Stx2 protein and the stxl -specific DNA bridges to assemble with their respective modified Fi molecules on the surface.

Figure 6B shows the simultaneous multiplex detection of protein and DNA for the Stx2 toxin and stxl gene after addition of a mixture of red and green gold nanoparticles that had been functionalized with detection anti-Stx2A and anti-FITC, respectively. These data show that specific protein and DNA targets can be detected and distinguished from each other when probed simultaneously in the same well using this method. The amounts of nonspecific nanorod binding for the Stx2 protein were lower than those for the stxl gene, which increased to some extent in samples that contained only Stx2 protein. This is likely to be the result of an inherent difference in the affinities for the surface of the respective antibodies used to functionalize the gold.

Conclusion

We have constructed a molecular motor-powered nanodevice capable of detecting single molecules of the Stx2 protein Shiga toxin and the stxl gene DNA sequence of E. coli 0157:H7. A key component is the Fi-ATPase molecular motor that can be induced to rotate a gold nanorod that is visible by microscopy. Since the limit of detection is dominated by nanorods nonspecifically bound to the microscope slide, rotation provides the ability to distinguish nanorods that are a component of the analyte-dependent assembly of the nanodevice, which enables an increase in sensitivity by 4 orders of magnitude. The single molecule detection presented here provides a dramatic improvement in sensitivity compared to a typical ELISA assay⁷. Based on the capture efficiency of the nanodevices for a particular analyte, the 1 : 1 stoichiometry of rotating nanorods to analyte molecules detected provides a quantitative measure of the number of analyte molecules present when compared to fluorophore-based detection systems. Further improvements in sensitivity and precision are likely to be possible by increasing the number of nanodevices per field of view as well as the number of fields of view examined.

Shiga toxin E. coli strains are particularly hardy and tolerate many different environmental conditions. As a result of this hardiness, the infectious dose for E. coli 0157:H7 is between 10-1000 cfu¹⁰. The sensor platform presented here can detect the toxin produced from such small number of bacteria, and is easily adapted to detect both protein and DNA targets. The assay also requires only two simple pipetting/washing steps prior to analysis by microscopy such that analysis can be completed in about 25 minutes, which is imperative for diagnosis of the Stx2 producing E. coli strain in a time scale necessary to ameliorate a lethal challenge. The ease of substituting antibodies in the nanodevices provided by protein G, allows the system to be adapted to detect a wide variety of proteins for different applications. Consequently, this detection platform should have wide applicability wherever the detection of small changes of cellular proteins is important like in neurobiology or stem cell research^1"5.

METHODS SUMMARY

Stx2 producing EDL933 (ATCC 43895) Escherichia coli 0157:H7 and E. coli K-12 strain MG1655 (ATCC 700926) strains (American Type Culture Collection, Manassas, VA) were grown in Luria-Bertani broth at 37°C with shaking for 18hrs. Cells were pelleted by centrifugation at 10,000xg for 2 min and supernatants were filtered through 0.22 μιη pore membranes (Millipore). Nanodevice assembly resulted from sequential 5 minute incubations on a Ni²⁺-NTA-coated coverslip (Xenopore, Inc) of: (1) ^g of 6xHis-tagged, biotinylated Fi- ATPase in 50mM Tris-HCl, pH 8.0, lOmM KC1 (Fi buffer); (2) SuperBlock (Pierce); (3) 10μg of NeutrAvidin (Invitrogen); (4) 2μg of biotinlyated protein G (Pierce); (5) O. ^g of monoclonal AntiStx2B (sc-65470, Santa Cruz Biotechnology); (6) ΙΟμΙ of sample to be analyzed; and (7) gold nanorods functionalized with monoclonal antiStx2A (sc-58080, Santa Cruz Biotechnology) via Protein G. An Fi buffer wash followed each step. Saliva was freshly isolated using the Salivette saliva collection system (Starstedt). Prior to addition of EDL933 or MG1655 supernatant, protein G was added to a final concentration of O. lmg/mL and incubated at 25°C for 30 minutes before diluting cell supernatant ten-fold. Gold nanoparticles ( anoPartz) were washed and resuspended in ImM CTAB were incubated in 0.2mg/mL Protein G (Pierce) at 25°C for 1 hour with rocking before incubating in 0.9% BSAc for 5 minutes. Antibody (anti-Stx2A or anti-FITC) was added to a final concentration of 8^g/mL and incubated at 25°C for lhr with rocking before diluting 10-fold with Fi buffer, stxl DNA- specific nanodevice assembly occurred via 5'-FITC, 3 '-biotin DNA 40mer bridges generated from stxl target by the Ligation-Exonucleation¹⁴ with some modifications. Gold

nanoparticles were visualized by microscopy¹⁴ with a Zeiss Axiovert 200M MAT dark- field, inverted microscope using the Mosaicx plug-in for the Zeiss Axiovision software to obtain consecutive stitched fields of view, and a Zeiss Axiocam Hsc camera to collect digital images.

Ruakhin, SS, Romanova, EV, Nemes, P & Sweedler, JV. Profiling metabolites and peptides in single cells. Nature Methods 8, S20-29 (2011).

DOI: 10.1038/NMETH.1549

Brecht, M, Schneider, M, Sakmann, B & Margrie, TW. Whisker movements evoked by stimulation of single pyramidal cells in rate motor cortex. Nature 427, 704-710 (2004).

Houweling, AR & Brecht, M. Behavioural report of single neuron stimulation in somatosensory cortex. Nature 451, 65-68 (2008).

Leong, KG, Wang, BE, Johnson, L & Gao, WQ. Generation of a prostate from a single adult stem cell. Nature 456, 804-808 (2008).

Shackleton, M et al. Generation of a functional mammary gland from a single stem cell. Nature 439, 84-88 (2006).

Schmid, A, Hortmann, H, Dittrich, PS, and Blank, LM. Chemical and biological single cell analysis. Curr. Opin. Biotechnol. 21, 12-20 (2010).

Schulte, R, Talamas J, Doucet C & Hetzer MW. Single bead affinity detection (SI BAD) for the analysis of protein-protein interactions. PLoS ONE 3, e2061 (2008). doi: 10.1371/journal.pone.0002061

Donnenberg, MS. Pathogenic strategies of enteric bacteria. Nature 430, 250-256 (2000).

Kaper, JB, Natero, JP & Mobley, HL. Pathogenic Escherichia coli. Nat. Rev.

Microbiol 21, 123-140 (2004).

Teuneis, PFM, Takumi, K, Shinagawa, K. Dose response for infection by Escherichia coli 0157:H7 from outbreak data. Risk Analysis 24, 401-407 (2004).

Yamaguami, et al. Efficacy of postinfection treatment with anti-Shiga toxin (Stx2) humanized monoclonal antibody TMA-15 in mice lethally challenged with Stx- producing Escherichia coli. J. Infect. Diseases 184, 738-742 (2001).

Bonifacino, JS & Rojas, R. Retrograde transport from endosomes to the trans-Golgi network. Nat. Rev. Cell Biol. 7, 568-579 (2006).

teLoo DM, et al. Binding and transfer of verocytotoxin by polymorphonuclear leukocytes in hemolytic uremic syndrome. Blood 95, 3396-3402 (2000).

York, J, Spetzler, D, Xiong, F & Frasch, WD. Single-molecule detection of DNA via sequence-specific links between Fi-ATPase motors and gold nanorod sensors. Lab Chip 4, 15-19 (20008). 15. Sonnichsen, C & Alivisatos, AP. Gold nanorods as novel nonbleaching Plasmon- based orientation sensors for polarized single-particle microscopy. Nano Lett. 5, 301- 304 (2005).

16. Spetzler, D, York, J, Daniel, D, Fromme, R, Lowry, D & Frasch, WD. Microsecond time scale rotation measurements of single Fi-ATPase molecules. Biochemistry 45, 31 17-3124 (2006).

17. Ishmukhametov, R, Hornung, T, Spetzler, D & Frasch, WD. Direct Observation of stepped proteolipid ring rotation in E. coli F₀Fi-ATP synthase. EMBO J 29, 3911- 3923 (2010).

METHODS

Bacterial strains and Culture filtrates. The high Stx2 producing EDL933 (ATCC 43895) Escherichia coli 0157:H7 strain was obtained from the American Type Culture Collection (ATCC, Manassas, VA). E. coli K-12 strain MG1655 (ATCC 700926) was used as a negative control. Bacterial strains were grown in Luria-Bertani broth at 37°C with shaking for 18hrs. Cells were pelleted by centrifugation at 10,000xg for 2 min in a Galaxy 16DH centrifuge (VWR) and the supernatants were filtered through 0.22 μιη pore membrane filters (Millipore).

Purification of Stx2. The stx2AB genes were PCR amplified with Herculase II polymerase (Stratagene) from an overnight culture of EDL933 using Ι μΙ, of 100-fold diluted overnight culture in a 50 μΐ, reaction volume. The stx2 genes were amplified using the Stx2F primer, which incorporates an Nco I restriction site and the Stx2R primer with aXho I restriction site for cloning into the pET24d(+) expression vector (Novagen).

The pET24d(+)::stx2AB plasmid was freshly transformed into BL21(DE3) cells (Novagen). An overnight culture was diluted 1 : 100 in lOOmL of fresh LB + Kanamycin and grown to an OD₆oo = 0.3. Expression of Stx2 in E. coli was induced with ImM isopropyl β- D(-)thiogalactopyranoside (Sigma) for 3 hrs at 37° with shaking. Stx2-thrombin-His was extracted with SoluLyse™ (Genlantis) containing 50 μg/mL Polymyxin B sulfate (Sigma), 5 μg/mL lysozyme (Sigma), and 250 U Benzonase (Novagen). The soluble fraction was combined with 1 part 2x binding buffer (50mM Tris-HCl pH 7.5, 300mM NaCl, 2mM imidazole, 20% glycerol), run over a Ni-NTA column (Qiagen), washed twice with wash buffer (25mM Tris-HCl pH 7.5, 150mM NaCl, lOmM imidazole, 10% glycerol), and eluted with elution buffer (25mM Tris-HCl pH 7.5, 150mM NaCl, 500mM imidazole, and 10% glycerol). Buffer exchange and concentration was performed using a Nanosep 3K Omega concentrator (Pall Corp). The 6xHis tag was removed using the Thrombin CleanCleave kit (Sigma) according to the manufacturer's instructions.

Stx2 Detection. Monoclonal antibodies (anti-Stx2B (sc-65470) and Anti-Stx2A (sc- 58080)) to Stx2 of E. coli 0157:H7 were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Assembly of the nanodevice proceeded with the sequential addition of nanodevice components. First, ^g of 6x his-tagged, biotinylated Fi-ATPase was attached to the surface of Ni²⁺-coated coverslips (Xenopore, Inc) for 5 min and then washed with Fi buffer (50mM Tris-HCl, pH 8.0, lOmM KC1). The surface was blocked by treatment with SuperBlock (Pierce) for 5 minutes. The bound Fi-ATPase was avidinated by the addition of 10μg of NeutrAvidin (Invitrogen) onto the Fi-coated slide. Subsequently, 2μg of biotinlyated protein G (Pierce) was linked to the Fi-ATPase via the attached Avidin, then 0.1 μg of Anti- Stx2B monoclonal antibody was added to attach the antibody to the above complex through the attached protein G. The bacterial supernatant/lysate was added followed by the addition of anti-Stx2A functionalized gold nanorods. All incubations were performed for 5 minutes at 25°C.

Stx2 Detection in saliva. Saliva was freshly isolated using the Salivette saliva collection system (Starstedt). Prior to addition of EDL933 or MG1655 supernatant, protein G was added to a final concentration of O. lmg/mL and incubated at room temperature for 30 minutes to bind endogenous IgA and IgG. Cell supernatant was diluted ten-fold in the saliva, which was used for Stx2 detection.

Antibody Functionalized Gold Nanoparticles. The gold nanoparticles were obtained from NanoPartz (Salt Lake City, UT). The gold nanoparticles were washed once with ImM cetyltrimethylammonium bromide (CTAB) and then resuspended in ImM CTAB at half the original volume. Protein G (Pierce) was added to a final concentration of 0.2mg/mL and incubated at 25°C for 1 hour with rocking. BSAc was then added to a final concentration of 0.9% and incubated for 5 minutes at 25°C. Antibody (anti-Stx2A or anti-FITC) was then added to a final concentration of 8^g/mL and incubated at 25°C for lhr with rocking. The resulting functionalized gold was then diluted 10-fold with Fi buffer (50mM Tris-HCl, pH 8.0, lOmM KC1).

LXR. The Ligation-Exonucleation Reaction was used to generate DNA bridges from the stxl target DNA sequence that enabled nanodevice assembly as per York et al.¹⁴ with the following modifications. Two 20mer DNA probes with sequences that were complimentary to adjacent sequences specific for the stxl gene were prepared in advance. One probe contained a 5'-FITC (5'-s/xi-FiTC) while the other contained a 3'-biotin and a 5 '-phosphate (3 '-s/ 7-Biotin). To generate the 5'-FITC, 3'-biotin DNA ligation product that served as the DNA bridge, 50 pmol of each probe was added to the ligation reaction in the presence of the sample containing the stxl DNA target, lx of T4 DNA ligase buffer, and 5U T4 DNA ligase (Invitrogen). To remove non-ligated probes, the ligation was treated with 10U of Φ-29 DNA polymerase for 1 hr at 30°C.

Dark field microscopy and image analysis. Gold nanoparticles were visualized in a manner similar to that reported by York et al. ^u using a Zeiss Axiovert 200M MAT dark- field, inverted microscope. The consecutive stitched fields of view were obtained using the Mosaicx plug-in for the Zeiss Axiovision software, and the digital images of each field of view were recorded using a Zeiss Axiocam Hsc camera.

Example 2.

A further study was performed as in Example 1, but with devices designed for detection of HSP70. All conditions and techniques were as in Example 1 unless noted differently. Detection of the cancer biomarker protein HSP70 was performed using monoclonal capture and detection antibodies (Luminex kit manufactured by EMD/Millipore) to the analyte for attachment to the molecular motor and the nanorod, respectively. For analysis, 0.01 mL of the sample containing the concentration of HSP70 indicated in Figure 7 was applied to the microscope slide and 30 fields of view (FOV) of the 900 total FOV exposed to sample were examined for the presence or rotating nanorods. At 0.001 pg HSP70/ml the 0.01 ml of sample examined contained a total of 86 molecules of HSP70 analyte.

Claims

We claim

1. A device, comprising

(a) a substrate;

(b) a plurality of first components, wherein each first component comprises

(c) a plurality of second components, wherein each second component comprises

(ii) a first nanorod attached to the first detection ligand.

2. The device of claim 1, wherein the first molecular motor comprises an Fi-ATPase.

3. The device of claim 2, wherein the Fi-ATPase is attached to a first member of a first binding pair, wherein the first member of the first binding pair is bound to a second member of the first binding pair, and wherein the second member of the first binding pair is attached to the first capture ligand.

4. The device of claim 3, wherein the second member of the first binding pair is bound to a first binding molecule, and wherein the first binding molecule is bound to the first capture ligand.

5. The device of any one of claims 1-4, wherein the first nanorod is attached to a second binding molecule, and wherein the second binding molecule is bound to the first detection ligand.

6. The device of any one of claims 1-5, wherein the biomolecule comprises a polypeptide.

7. The device of claim 6, wherein the first capture ligand comprises an antibody selective for a first site on the polypeptide, and the first detection ligand comprises an antibody selective for a second site on the polypeptide.

8. The device of claim 6 or 7, wherein the polypeptide is the Stx2 toxin.

9. The device of claim 8, wherein the first capture ligand comprises an antibody selective for a Stx2B subunit of the Stx2 toxin, and the first detection ligand comprises an antibody selective for a Stx2A subunit of the Stx2 toxin.

10. The device of any one of claims 6-9, further comprising:

b) a plurality of third components, wherein each third component comprises i. a second capture ligand, wherein the second capture ligand binds to a nucleic acid of interest; and

ii. a second molecular motor attached to the second capture ligand,

wherein the second molecular motor is attached to the substrate;

c) a plurality of fourth components, wherein each fourth component comprises i. a second detection ligand, wherein the second detection ligand binds to the nucleic acid of interest; and

ii. a second nanorod attached to the second detection ligand, wherein the second nanorod is distinguishable from the first nanorod.

11. The device of claim 10, wherein the second molecular motor comprises an Fi- ATPase.

12. The device of claim 11 , wherein the Fi-ATPase is attached to a first member of a binding pair, and wherein the second capture ligand comprises a second member of the binding pair, and wherein the Fi-ATPase is attached to the second capture ligand through binding of the first member of the binding pair to the second member of the binding pair.

13. The device of claim 12, wherein the first member of the binding pair comprises avidin, and the second member of the binding pair comprises biotin.

14. The device of claim 13, wherein the second capture ligand further comprises a first member of a second binding pair, and wherein the second detection ligand comprises a second member of the second binding pair.

15. The device of claim 14, wherein the second member of the second binding pair comprises an antibody selective for the first member of the second binding pair.

16. The device of any one of claims 10-15, wherein the second nanorod is bound to a third binding molecule, and wherein the third binding molecule is bound to the second detection ligand.

17. The device of any one of claims 10-16, wherein the second capture ligand comprises a plurality of oligonucleotides that are perfectly complementary to the nucleic acid of interest, wherein upon hybridization to the target nucleic acid, the target-specific nucleic acid probes form a series of target-specific nucleic acid probes directly adjacent to one another

18. A kit, comprising

(a) a plurality of first components, wherein each first component comprises

(i) a first capture ligand, wherein the first capture ligand binds to a first subunit of a multi-subunit biomolecule target; and (ii) a first molecular motor attached to the first capture ligand, wherein the first molecular motor is capable of attachment to a substrate; and

(b) a plurality of second components, wherein each second component comprises

(ii) a first nanorod attached to the first detection ligand.

19. A method for biomolecule detection, comprising

(a) contacting the detection device according to any one of claims 1-17 with a sample under conditions whereby the first capture ligand and the first detection ligand bind to different sited on the biomolecule target, if it is present in the sample;

(b) inducing movement of the molecular motor; and

20. The method of claim 19, wherein the molecular motor comprises Fi-ATPase.

21. The method of claim 19 or 20, wherein the test sample is a bodily fluid sample.

22. The method of any one of claims 19-21 , wherein the biomolecule target is a polypeptide.

23. The method of any one of claims 19-22, wherein the method comprises multiplex detection of a polypeptide target and a nucleic acid of interest in the test sample, wherein the detection device comprises

(c) a plurality of third components, wherein each third component comprises

(d) a plurality of fourth components, wherein each fourth component comprises

24. A physical computer-readable storage medium contains instructions executable by a processor that, when executed, cause the processor to perform certain functions on a detection device for carrying out the detection methods of the present invention, wherein the functions include one or more of receiving a signal representing light scattered from the nanorods on the surface of the support as recited in any of the preceding claims, stitching of signals obtained from multiple consecutive fields, and detecting movement of the molecular motor based on the signals so obtained.