Devices and methods for measuring nanometer level binding reactions

Abstract

Systems and methods for measuring nanometer level binding reactions are described herein. In some embodiments, the disclosed subject matter is directed to a device comprising (a) a substrate having a surface and (b) an ordered array of posts, pits, or patches over the surface, wherein the posts, pits, or patches are capable of binding a protein or small molecule ligand, and wherein the pitch between adjacent posts, pits, or patches is less than about 100 nm. In some other embodiments, the disclosed subject matter is also directed to methods for identifying the presence of an analyte in a fluid and to methods for measuring relative binding specificity or affinity between an analyte in a fluid and the posts, pits, or patches, using the device of the disclosed subject matter.

Description

This application claims the benefit of the filing date of International Application Serial No. PCT/US2004/034987, filed Oct. 15, 2004, and U.S. patent application Ser. No. 11/404,716, filed Apr. 13, 2006, both of which claim the benefit of provisional application U.S. Ser. No. 60/511,799, filed Oct. 15, 2003, which are hereby incorporated by reference into the subject application in their entireties. This application also claims the benefit of the filing date of provisional application U.S. Ser. No. 60/837,701, filed Aug. 15, 2006, which is hereby incorporated by reference into the subject application in its entirety.

All patents, patent applications and publications cited herein are hereby incorporated by reference in their entirety. The disclosures of these publications in their entireties are hereby incorporated by reference into this application in order to more fully describe the state of the art as known to those skilled therein as of the date of the disclosed subject matter described and claimed herein.

Copyright Statement: This patent disclosure may contain material that is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by anyone of the patent document or the patent disclosure as it appears in the U.S. Patent and Trademark Office patent file or records, but otherwise reserves any and all copyright rights.

TECHNOLOGY AREA

Devices and methods for measuring nanometer level binding reactions are provided.

BACKGROUND OF THE APPLICATION

Many proteins have been studied extensively at the single molecule level. However, in the cell those proteins form into larger complexes or modules wherein the spacing of components on a nanometer scale is critical. New technologies in patterning now enable us to systematically measure the dependence of interactions on nanometer level patterns and to then exploit that spatial dependence in sensing and nanofabrication of materials through directed self-assembly. As an example, the signals from extracellular matrices affect normal and cancerous cell growth and there is evidence that the spacing of the matrix molecules makes a critical difference in that signal (Jiang, G. et al. (2003) Nature, 424:334-37). These mechanisms must be studied at a scale that matches the size and/or spacing of features of specific protein or subcellular protein complexes, which are generally at the nanometer level.

There is a great need to measure the binding of complex protein assemblies with spatially ordered ligands. Thus, there is a great need for a device that tests the binding of an analyte to specific spatial arrays of ligands.

SUMMARY OF THE APPLICATION

In some embodiments, the disclosed subject matter is directed to a device for measuring nanometer level binding reaction. The device can include (a) a surface and (b) an ordered array of posts, pits, or patches on the surface, wherein the posts, pits, or patches are capable of binding a protein or small molecule ligand, and wherein the pitch between adjacent posts, pits, or patches is less than about 100 nm.

In some other embodiments, the disclosed subject matter is also directed to methods for measuring nanometer level binding reactions. Methods of the disclosed subject matter can include (a) providing a device having a surface, the surface comprising an ordered array of posts, pits, or patches, wherein the pitch between adjacent posts, pits, or patches is less than about 100 nm, wherein each post, pit, or patch is coated with ligand, (b) contacting the surface of the device with a sample, and (c) determining whether or not an analyte from the sample interacts or binds to a ligand-coated post, pit, or patch, thereby identifying the presence of the analyte in the fluid sample. The method can also include isolating the analyte from the sample.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 depicts fibronectin and a slip bond formed between a single trimer of fibronectin and a cellular cytoskeleton.

FIG. 2 depicts a proposed process flow for the fabrication of a device according to some embodiments of the disclosed subject matter.

FIG. 3 depicts features with nanometer scale dimensions patterned by electron beam lithography.

FIG. 4 shows arrays of dots of hydrogen silsequioxane, a negative tone electron beam resist.

FIG. 5 depicts an electrode pair for the study of transport in individual molecules fabricated by direct-write e-beam lithography.

FIG. 6 is a graphic representation of resist thickness as a function of applied dose for a 3:1 isopropanol:water mixture using various molecular weights of poly(methyl)methacrylate.

FIG. 7 shows surface roughness of PMMA as a function of molecular weight.

FIG. 8 illustrates the reduction of dot size by thermal treatment.

FIG. 9 gives an example of high resolution placement accuracy e-beam lithography.

FIG. 10 depicts an exemplary device according to some embodiments of the disclosed subject matter.

FIG. 11 depicts micrographs of a prototype chip with patterned arrays according to some embodiments of the disclosed subject matter.

FIG. 12 depicts prototype dot pair arrays according to some embodiments of the disclosed subject matter.

FIG. 13 shows a prototype dot array according to some embodiments of the disclosed subject matter.

FIG. 14 depicts atomic force microscopy of prototype arrays according to some embodiments of the disclosed subject matter.

FIG. 15 shows a system for fabricating exemplary devices according to some embodiments of the disclosed subject matter.

FIG. 16 shows lithographic ways of patterning according to some embodiments of the disclosed subject matter.

DETAILED DESCRIPTION OF THE APPLICATION

The present application enables the formation of protein arrays on the scale of nanometers as in a cell and the application of this technology to measure and exploit the spatial dependence of interactions in biological and nanofabrication areas. There is a need to control the spatial parameter in molecular interactions in a defined way.

The disclosed subject matter provides a device that facilitates binding of an analyte to a set of posts, pits, or patches having a spacing that facilitates the binding. Binding to a single post, pit, or patch is dependent upon the chemistry and steric nature of the analyte and the protein attached to the post, pit, or patch.

The device can include (a) a surface and (b) an ordered array of posts, pits, or patches on the surface, wherein the posts, pits, or patches are capable of binding a protein or small molecule ligand, and wherein the pitch between adjacent posts, pits, or patches is less than about 100 nm.

The disclosed subject matter is also directed to methods for measuring nanometer level binding reaction. Methods of the disclosed subject matter can include (a) providing a device having a surface, the surface comprising an ordered array of posts, pits, or patches, wherein the pitch between adjacent posts, pits, or patches is less than about 100 nm, wherein each post, pit, or patch is coated with ligand, (b) contacting the surface of the device with a sample, and (c) determining whether or not an analyte from the sample interacts or binds to a ligand-coated post, pit, or patch, thereby identifying the presence of the analyte in the fluid sample. The method can also include isolating the analyte from the sample.

The present application is further directed to methods for measuring relative binding specificity or affinity between an analyte and a ligand. Methods of the disclosed subject matter can include (a) providing a device having a surface, wherein the surface comprises an ordered array of posts, pits, or patches on the surface having a ligand, wherein the pitch between adjacent posts, pits, or patches is less than about 100 nm, (b) contacting the surface of the device with one or more analytes, (c) determining whether or not an analyte interacts or binds to the ligand, and (d) determining the binding specificity or affinity between the analyte and the ligand.

The disclosed subject matter is also directed to methods for crystallizing a protein. Methods of the disclosed subject matter can include (a) providing a device having a surface, wherein the surface comprises an ordered array of posts, pits, or patches over the surface, wherein the pitch between adjacent posts, pits, or patches is less than about 100 nm, and is identical throughout the array, and wherein the post, pit, or patch is capable of binding a protein, and (b) contacting the surface of the device with a protein to be crystallized, wherein at least one post, pit, or patch functions as a seed crystal or nucleus for crystallization, thereby crystallizing the protein.

The disclosed subject matter is directed to methods for making a device. Methods of the disclosed subject matter can include (a) designing an array pattern, (b) writing the array pattern onto a substrate, (c) forming a post, pit, or patch on the substrate, and optionally (d) shrinking or enlarging the post, pit, or patch, thereby making the device.

The disclosed subject matter is also directed to kits. Kits of the disclosed subject matter can test for the presence or absence of an analyte in a sample, determine a subject's risk for developing a disease, or monitor the status of a disease in a subject. Kits of the disclosed subject matter can include a device that specifically binds to an analyte in an amount effective to detect the analyte in the sample. The kit can contain a device having a surface and an ordered array of posts, pits, or patches over the surface, wherein the posts, pits, or patches are capable of binding a ligand and wherein the pitch between adjacent posts, pits, or patches comprises less than about 100 nm. Further, the interaction between the device and the analyte is detectable. The kit can include one or more reagents for detecting amounts of one or more analytes bound to the device. In another embodiment, the kit can further include one or more reagents for detecting amounts of the one or more analytes bound to the device.

The disclosed subject matter is further directed to methods for detecting a protein isomer in a mixture. Methods of the disclosed subject matter can include (a) providing a device having a surface; the surface comprising an ordered array of posts, pits, or patches over the surface, wherein the pitch between adjacent posts, pits, or patches is less than about 100 nm, wherein the post, pit, or patch is coated with ligand; (b) contacting the surface of the device with a mixture; and (c) determining whether or not a protein isomer in the mixture interacts or binds to a ligand-coated post, pit, or patch, thereby detecting the protein isomer in the mixture.

The disclosed subject matter is directed to methods for detecting a microorganism in a sample. Methods of the disclosed subject matter can include (a) providing a device having a surface; the surface comprising an ordered array of posts, pits, or patches over the surface, wherein the pitch between adjacent posts, pits, or patches is less than about 100 nm, wherein the post, pit, or patch is coated with ligand; (b) contacting the surface of the device with a sample; and (c) determining whether or not a microorganism in the sample interacts or binds to a ligand-coated post, pit, or patch, thereby detecting the microorganism in the fluid sample.

In one embodiment, the device comprises a gradient of pitch values. For example, the gradient of pitch values can be from about 5 nm to about 100 nm. In one embodiment, the pitch is less than about 50 nm. In other embodiments, the pitch is less than about 40 nm, less than about 35 nm, less than about 30 nm, less than about 25 nm, less than about 20 nm, less than about 15 nm, less than about 10 nm, and in another embodiment, less than about 5 nm, less than about 3 nm in diameter, and in another embodiment, less than about 2 nm in diameter.

In another embodiment, the pitch is homogenous. For example, the ordered array of pairs of posts, pits, or patches can have a pitch that is less than about 50 nm. In other embodiments, the pitch is less than about 40 nm, less than about 35 nm, less than about 30 nm, less than about 25 nm, less than about 20 nm, less than about 15 nm, less than about 10 nm, and in another embodiment, less than about 5 nm, less than about 3 nm in diameter, and in another embodiment, less than about 2 nm in diameter.

In another embodiment, each post, pit, or patch is less than about 20 nm in diameter. In other embodiments, each post, pit, or patch is less than about 10 nm in diameter, less than about 5 nm in diameter, less than about 3 nm in diameter, and in another embodiment, less than about 2 nm in diameter.

In one embodiment, each post, pit, or patch independently comprises a material selected from the group consisting of metal, semiconductor, organic insulator, inorganic insulator, biocompatible material, or a combination thereof. For example, each post, pit, or patch independently comprises a material selected from gold, nickel, palladium, polyethyleneglycol, oxides of aluminum, hafnium, silicon, tantalum, titanium, zinc, or zirconium, nitrides of aluminum, hafnium, silicon, tantalum, titanium, zinc, or zirconium or carbides of aluminum, hafnium, silicon, tantalum, titanium, zinc, or zirconium or combinations thereof. In another embodiment, each post, pit, or patch includes a material which does not quench fluorescence, such as titanium oxide. In yet another embodiment, each post, pit, or patch includes a material that provides small grain structure and a smooth surface, such as a gold-palladium alloy.

In another embodiment, each post, pit, or patch has affixed thereto exactly one protein molecule. In another embodiment, each post, pit, or patch displays more than one protein molecule. In another embodiment, the protein molecule comprises at least a portion of a full-length dynein heavy chain, a tubulin, a kinesin, a myosin, a cytoplasmic domain of an integrin, an actin, an extracellular matrix protein, a fibronectin, a collagen, a laminin, a DNA solenoid, a DNA, a histone-DNA complex, an RNA, an RNA-protein complex, a bacterial coat protein, an antibody, a lectin, an avidin, or any combination thereof. In one embodiment, the protein molecule comprises a full-length dynein heavy chain, a dynein motor domain, or an N-terminal portion of a dynein motor domain.

In one embodiment, the relative binding specificity or affinity is a function of pitch. For example, pitch can be increased from a small spacing to a large spacing to determine the maximum separation that allows binding.

In one embodiment of the disclosed subject matter, the subject is a mammal. The mammal can be a human or a primate. The subject can be a human patient, or an animal that exhibits symptoms of a human immune disease and is therefore an animal model of a human disease, such as a murine transgenic disease model or a primate disease model or a model of human disease established in a SCID mouse reconstituted with the human immune system. The mammal can be, but is not limited to, a human, a primate, a rat, a dog, a cat, a swine. In another aspect of the disclosed subject matter, the subject is a murine subject, a bovine subject, a primate subject, an equine subject, a swine subject, or a canine subject.

In one embodiment, the sample is a blood sample or a serum sample.

In one embodiment, the analyte is labeled with a detectable marker. In one embodiment, the detectable marker is selected from the group consisting of a fluorescent marker, a radioactive marker, an enzymatic marker, a colorimetric marker, a chemiluminescent marker or a combination thereof.

In one embodiment, optical means, electrical means, mechanical means, or a combination thereof can be utilized to detect the presence of a species on a post, pit, or patch. In one embodiment, total internal reflection fluorescence (TIRF) microscopy, ellipsometry, or phase microscopy, atomic force microscopy (AFM), fluorescence resonance energy transfer (FRET) microscopy, fluorescence microscopy, two-photon microscopy, electrical conduction, or a combination thereof can be utilized.

In one embodiment, designing the array pattern comprises using a computer-aided design or an algorithmic design system. In another embodiment, the algorithmic design system allows for the systematic variation of post, pit, or patch configuration and spacing.

In another embodiment, writing the array pattern comprises using a high resolution electron beam lithography system. In one embodiment, writing the array pattern comprises patterning a mask.

In one embodiment, the substrate is coated with a resist. In one embodiment, the resist is a positive resist. In another embodiment, the resist is a negative resist.

In another embodiment, forming the post, pit, or patch comprises a technique selected from the group consisting of liftoff, electroplating, reactive ion etching, ion milling, controlled wet etching, or a combination thereof.

In one embodiment, the method of making the device further comprises forming microfluidic channels. In one embodiment, the method further comprises binding a ligand to a post, pit, or patch on the device.

Patches can be advantageous in some situations by providing a simple fabrication method. For example, patterns can be stamped using a master template (e.g., microcontact printing). Pits can be advantageous in some situations by providing regions where the portions that bind to the surface can be shielded from observation if desired.

Definitions

The term “post” is used herein to mean a support rising vertically from a surface.

The term “pit” is used herein to mean a structure or void penetrating vertically into a surface.

The term “patch” is used herein to mean a region on or near a surface having at least one chemical difference from its surrounding.

The term “surface” is used herein to mean the outer part of any material.

The term “pitch” is used herein to mean the distance between center points of adjacent posts, pits, or patches.

The term “gradient” is used herein to mean a change that can be abrupt or gradual.

The term “biocompatible” is used herein to mean being compatible with living tissue by virtue of a lack of toxicity or ability to cause immunological response.

The term “TIRF” is used herein to mean total internal reflection fluorescence.

The term “FRET” is used herein to mean fluorescence resonance energy transfer.

The term “AFM” is used herein to mean atomic force microscopy.

The term “PMMA” is used herein to mean poly(methyl)methacrylate.

The term “resist” is used herein to mean a radiation sensitive layer.

The term “CAD” is used herein to mean computer aided design.

The term “unit cell” as used herein refers to a set of posts, pits, or patches having a specific geometric configuration, which can include a pair of posts, pits, or patches separated by a given spacing; or a small number of posts, pits, or patches separated by a given spacing with fixed angles between them, so that they can be arranged as, for example, squares, rectangles, or general trapezoids.

The term “AAA” is used herein to mean ATPases associated with cellular activity.

The term “HC” is used herein to mean heavy chain.

The term “IPA” is used herein to mean isopropanol.

The term “about” is used herein to mean approximately, roughly, around, or in the region of. When the term “about” is used in conjunction with a numerical range, it modifies that range by extending the boundaries above and below the numerical values set forth. In general, the term “about” is used herein to modify a numerical value above and below the stated value by a variance of 20 percent up or down (higher or lower).

As used herein, the word “or” means any one member of a particular list and also includes any combination of members of that list.

Fabrication of Nanoarray Devices

The disclosed subject matter provides a coherent strategy for the preparation of nanoscale patterns for the study of biological molecules. The methods demonstrate precision and control at the nanometer regime.

This work takes advantage of metal and semiconductor processing on scales that match the size and/or spacing of features of specific protein or subcellular protein complexes, some of which have dimensions of only a few nanometers. In order to create these structures, multiple approaches can be utilized: electron beam lithography, nanoimprinting lithography, contact printing, and directed self-assembly.

For some applications described herein, hierarchical arrays of diverse materials to which specific proteins can bind, are patterned on a transparent substrate. The arrays consist of unit cells comprised of small numbers (about 2-4) of metal dots separated from one another by distances of about 10-200 nm and in some cases arranged in specific spatial configurations according to certain crystallographic space groups. Each array contains a sufficient number of posts within an area of about 1-5 micrometers so as to be easily detected by TIRF microscopy. Permutations on the dot spacing and/or the precise spatial configuration of the dots in each unit cell form the next level of hierarchy in the arrays, which cover areas of several micrometers to several millimeters. Such hierarchical arrays cannot be created exclusively by chemical self-assembly techniques (G. M. Whitesides, J. P. Matthias, and C. T. Seto, (1991) Science, 254, 1312), which are better suited to applications requiring vast numbers of identical elements, but with little or no long range order and no simple way of varying the intra-unit cell spacing. Instead, more conventional “top down” lithographic techniques are used (C. R. K. Marian and D. M. Tennant (2003), J. Vac. Sci. and Technology, A21, S207), which have the flexibility to form patterns comprising many different unit cell variations.

The required dimensions (both the feature size and spacing) of such bioarrays are near the current limits of lithographic patterning techniques. As an example, ultrahigh resolution electron beam lithography, with precise exposure dose control for critical features, and the use of small radius of gyration resist materials with low temperature processes that enhance resolution and maintain contrast extends the limits of these techniques. The patterns are transferred by liftoff and/or ion beam etching to metals for which the binding chemistry of various biological molecules is known (A. Ulman (1996), Chem. Rev., 96, 1533 (Washington, D.C.)). For some applications, arrays comprising diverse chemical species can be of interest. For such applications, multiple lithographic steps can be used with ultrahigh placement accuracy to pattern arrays of dissimilar metals (for example, gold dots followed by nickel dots could create a pattern for binding sulfhydryl and poly-histidine tags, respectively). As another example, patterning can be achieved by scanning probe lithography, including dip-pen, x-ray, or extreme UV lithography, and contact printing (C. R. K. Marian and D. M. Tennant (2003), J. Vac. Sci. and Technology, A21, S207).

A proposed process flow for the fabrication of hierarchical bioarrays with molecular spacings is shown in FIG. 2. First, arrays can be designed using either a generic CAD package or an algorithmic design system, such as CADENCE, MENTOR GRAPHICS, DESIGN CAD, and the like, allowing for the systematic variation of unit cell configuration and spacing. The design data can be used as input to a high resolution electron beam lithography (e-beam) system, which writes each pattern onto a transparent substrate coated with a radiation sensitive layer (resist). E-beam lithography can be used to pattern a mask, from which the array patterns are subsequently transferred by nanoimprint lithography, described below. Depending upon the details of the pattern transfer process, either a positive or negative tone resist is used. Pattern transfer can then take place via an additive process, such as liftoff or electroplating, or via a subtractive process, such as reactive ion etching, ion milling or controlled wet etching, depending upon the selected materials. Once metal dots have been formed on the substrate, subsequent processing depends upon the precise application. For example, for the investigation of protein binding, microfluidic channels can be formed, and ligands can be bound to the appropriate dots on the array. After washing and blocking the rest of the surface to prevent non-specific interactions, the binding molecules tagged with an appropriate fluorophore can be added through the flow channel, incubated to allow binding, and washed out to measure release. A linker molecule can be introduced which adheres to the metal by chemical bonding. For protein crystallization, the arrays can be used with or without further processing.

Generally, it is difficult to form features smaller than about 10 nm and arrays with center-to-center spacings less than about 50 nm using top-down patterning (i.e., conventional lithography) (Broers, A. N. (1988) IBM J. Research and Dev't, 32(4):502-513). However, as can be seen in FIG. 3, as early as 1981, features with nm-scale dimensions have been patterned by electron beam lithography (Isaacson, M. and A. Murray (1981) J. Vacuum Sci. Tech., 19(4): 1117-1120), and arrays with pitches of about 10 nm have been realized (Langheinrich, W. and H. Beneking (1993) Jap. J. App. Physics Part 1—Reg. Papers Short Notes Rev. Papers, 32(12B):6218-6223). FIG. 3(a) shows sub-5 nm lines patterned in NaCl film by electron beam-induced vaporization as in Isaacson and Murray, 1981. FIG. 3(b) shows sub-10 nm pitch lines patterned in Al-doped LiF, also as in Isaacson and Murray, 1981. In each case, exceedingly high doses were required, and pattern transfer was problematic. Until recently, the key challenges in accessing the sub-10 nm regime have been in identifying practical lithographic materials with sensitivities that allow for manageable exposure times and are amenable to direct pattern transfer. Such materials have recently been discovered (Namatsu, H. et al. (1998) J. Vacuum Sci. Tech., 16(6):3315-3321; Rooks, M. J. and A. Aviram (1999) J. Vacuum Sci. Tech. B. 17(6):3394-3397; Yasin, S. et al. (2001) App. Physics Lett., 78(18):2760-2762), and new approaches to processing (Yasin, S. et al. (2001) J. Vacuum Sci. Tech. B, 19(1):311-313) are also extending the resolution of commonly used electron beam resists, such as PMMA. The e-beam system is an FEI scanning electron microscope fitted with Nabity e-beam lithography control system. Lithography can be done at energies between 1-30 keV, and the system has already demonstrated sub-10 nm patterning capability, although low probe current limits practical pattern density and coverage for large arrays. The Leica VB-6HR is generally operated at 100 keV. This system has sufficient probe current to pattern arrays spanning many mm in a reasonable exposure time, and it has also shown sub-10 nm patterning capability, and sub 5-nm patterning capabilities are being pursued. FIG. 4 shows arrays of dots of hydrogen silsequioxane (HSQ), a negative tone electron beam resist. FIG. 4(a) shows ultrahigh resolution patterning of 6 nm HSQ dots on 100 nm pitch, while FIG. 4(b) shows a 25 nm pitch n×n array. Relatively isolated features as small as 6 nm have been patterned, and dense arrays, with center-to-center spacing down to 25 nm have already been achieved. Such high resolution is possible because of the cage-like structure of the HSQ molecule (Siew, Y. K. et al. (2000) J. Electrochem. Soc., 147(1):335-339) and its small radius of gyration (Namatsu, H. et al. (1998) J. Vacuum Sci. Tech. B, 16(1):69-76). Reducing the molecular weight of the material is likely to result in further improvements in “raw resolution.” HSQ can be easily converted to SiO₂ by thermal treatment or by exposure to an oxygen plasma, so that careful etching in dilute hydrogen fluoride (HF) can be used to reduce the dimensions of the dots even further. The dots can be used as a mask to transfer the pattern into an underlying film of metal or other material to which the selected proteins can bind, by ion milling or reactive ion etching.

In addition to HSQ, resolution enhancing processes in the positive tone resist PMMA can be performed. As mentioned above, Yasin et al. (Yasin, S. et al. (2001) App. Physics Lett., 78(18):2760-2762) have demonstrated about 5 nm patterning capability by use of a new resist development process using two nonsolvents, isopropanol and water, combined with ultrasonic agitation. This new process offers superior contrast and resolution when compared to the conventional developer dilute methyl-isobutyl-ketone (in IPA). It appears that EPA and water combine to form a co-solvent which is able to more efficiently penetrate the polymer matrix with minimum resist swelling (which normally leads to a loss of resolution). The ultrasonic agitation induces microstreaming, yielding an efficient mechanism for providing fresh developer to the polymer and rapid removal of dissolved material. Following Rooks et al., (Rooks, M. J. et al. (2002) J. Vacuum Sci. Tech. B, 20(6):2937-2941), this process was improved by reducing the developer temperature, resulting in increased contrast relative to room temperature development (as well as conventional developers) (Rooks, M. J. et al. (2002) J. Vacuum Sci. Tech. B. 20(6):2937-2941). This process is employed for the fabrication of electrodes for the study of molecular electronics, an example of which is shown in FIG. 5, with sub-5 nm spacing. FIG. 5 shows a an electrode pair for the study of transport in individual molecules fabricated by direct-write e-beam lithography and low temperature ultrasonic development using an IPA:H₂O mixture. We are extending both the resolution and reliability of this process by the use of low molecular weight PMMA. Initial experiments (Greci (2003) private communication) indicate that low temperature development using an IPA/water mixture with ultrasonic agitation maintains the resist contrast even as the molecular weight is reduced, as shown in FIG. 6. FIG. 6 shows the remaining resist thickness as a function of applied dose for a fixed development time, for four different molecular weights of PMMA. As can be seen in FIG. 6, contrast does not degrade with a reduction in molecular weight when for used for ultrasonically assisted low temperature using an IPA:H₂O mixture. This is strikingly different from what is found using conventional developers (i.e., MIBK solutions), where both contrast and resolution suffer when low Mw is used (Dobisz, E. A. et al. (2000) J. Vacuum Sci. Tech. B, 18(1): 107-111). In addition, scanning probe analysis of partially developed resist shows that the low molecular weight has significantly lower surface roughness than the higher molecular weight material when processed using low temperature ultrasonic development, as shown in FIG. 7. FIG. 7 shows atomic force micrographs of partially developed PMMA of different molecular weights. Reduced surface roughness has been shown to correlate with improved resolution (Namatsu, H. et al. (1998) J. Vacuum Sci. Tech. B. 16(6):3315-3321; Yasin, S. et al. (2001) J. Vacuum Sci. Tech. B. 19(1):311-313), and preliminary experiments indicate that improved resolution and linewidth control are indeed achieved with low molecular weight PMMA.

When a positive tone resist, such as PMMA, is used for the fabrication of bioarrays, pattern transfer is achieved by liftoff or by electro- and electroless plating. The electron beam evaporation system can include sub-nanometer thickness control and liquid nitrogen-cooled stage, offering the possibility of ultrafine grain metal deposition. When a negative tone resist, such as HSQ, is used, the pattern transfer is done by etching—reactive ion etching, ion milling, or carefully controlled wet chemical etching, depending on the materials involved.

Once dots with sub-10 nm have been formed, further dimensional reduction can be achieved as illustrated by the thermal treatment depicted in FIG. 8. After lithography and pattern transfer, each metal dot approximates a cylindrical disc sitting on the substrate. Thermal treatment causes the dots to minimize their surface area, forming spheroids. Reduction of the dot thickness to a fraction of its diameter results in a significant reduction in the final dot size. For example, a dot with a diameter of 6 nm and a thickness of 0.5 nm is reduced to about a 3 nm sphere. For this approach, precise control over film thickness and grain structure is also important. Table I tabulates the reduction in dimensions depicted in FIG. 8.

TABLE 1
Dimensional Reduction by Thermal Treatment
dc (nm)	h (nm)	ds (nm)
10	1	3.9
10	0.5	3.0
8	1	3.3
8	0.5	2.6

For some applications, it is desirable to form patterns of closely spaced posts of different materials, such as gold and nickel-NTA. In this case, two levels of lithography is required, with level-to-level overlay of about 10 nm. Such placement accuracy is achievable with e-beam lithography using careful alignment strategies (Guillorn, M. A. et al. (2000) J. Vacuum Sci. Tech. B. 18(3):1177-1181). As an example, FIG. 9 shows a set of four interdigitated metal lines in which successive pairs of alternative lines were patterned in separate lithographic exposures and metal depositions (Wind, S. J. et al. (2003) J. Vacuum Sci. Tech. B, accepted for publication). Through high-resolution placement accuracy using e-beam lithography, levels A and B depicted in FIG. 9 were patterned and processed separately. The enlargement on the right in FIG. 9 shows better than 5 nm overlay.

FIG. 10 depicts an exemplary device fabricated by the above process.

FIG. 11 depicts an optical electron micrograph (FIG. 11(a)) and a scanning electron micrograph (FIG. 11(b)) of a prototype chip with patterned arrays according to the some embodiments of the disclosed subject matter. The micrographs are each about 300 microns square. The circular flames around each of the arrays are visible in FIG. 11(a).

FIG. 12 depicts prototype dot pair arrays fabricated by the above process.

FIG. 13 shows a prototype dot array fabricated by the above process, and demonstrates a linear array of dot strings having specific spacings.

FIG. 14 depicts atomic force microscopy of gold-patterned prototype arrays fabricated by the above process.

Nanoimprint lithography can also be utilized. FIG. 15 depicts a thermal system and a photocurable system for mass-producing a device using nanoimprint lithography. As shown, nanoimprint lithography includes imprinting and pattern transferring. Imprinting can include pressing a mold onto a substrate having a layer of resist material using heat or radiation and removing the mold. Pattern transfer can include reactive ion etching to obtain a desired device in accordance with some embodiments of the disclosed subject matter. Other lithographic alternatives for patterning include PMMA, platinum, and direct patterning of self-assembled monolayers, as depicted in FIG. 16.

NANOARRAY DEVICE AND USES THEREOF

The disclosed subject matter can be utilized in a number of different applications. For example, nanoarray devices of the disclosed subject matter can be utilized as sensor devices (e.g., to detect hazardous materials or disease markers), as tools for biological mechanism studies (e.g., tools to study binding specificity or affinity), crystallization templates (e.g., crystallization of proteins, viruses, and the like), separation devices (e.g., separation based on different structural isomers, position isomers, functional group isomers, stereoisomers, and the like), and the like. For example, the disclosed subject matter can be utilized in bioinformatics, tissue engineering, portable screening devices, and the like.

In one embodiment of the disclosed subject matter is provided a device comprising (a) a substrate having a surface and (b) an ordered array of posts, pits, or patches over the surface, wherein the posts, pits, or patches are capable of binding a ligand, and wherein the pitch between adjacent posts, pits, or patches is less than about 100 nm. The device according to some embodiments of the disclosed subject matter is a flexible device that provides many new capabilities for measuring not only the spatial dependence of binding to one ligand, but also the spatial dependence of binding to several different ligands that are patterned on the surface. Using high throughput techniques, including multiple stamping techniques, several different chemicals could be patterned on a surface; for example, gold dots followed by nickel dots can create a pattern for binding sulfhydryl and poly-histidine tags, respectively. Stamping techniques include nanoimprint lithography and contact printing.

In another embodiment, the disclosed subject matter provides a method to rapidly measure highly specific interactions that depend upon the molecular level spacing of components in large complexes. By changing the distance between 2-5 nm posts, pits, or patches on the nanometer scale, it is possible to create spatial arrays of ligands that match the spacing and distribution of binding sites in a binding complex. Binding constants can be measured directly using a microfluidics flow system and either TIRF microscopy or ellipsometric measurements to monitor the rates of binding and release of the material at known concentrations.

In a typical measurement, an array of 2-5 nm posts, pits, or patches are created on the template (C. R. K. Marian and D. M. Tennant (2003), J. Vac. Sci. and Technology, A21, S207). The template is then used to create a pattern of dots on a small area (light microscope resolution limits the minimum size for reliable measurements to 2-5 μm²) of a glass coverslip. Tens to hundreds of different patterns are imprinted on adjacent regions of the coverslip to fill the 10,000 μm² viewing area of the microscope objective (typically a 60×, 1.45 n.a. objective capable of through objective TIRF). Once imprinted with the designated patterns, the imprinted coverslip is assembled with a microfluidics channel and the ligands are bound to the appropriate dots on the array. After washing and blocking the rest of the surface to prevent non-specific interactions, the binding molecules tagged with an appropriate fluorophore are added through the flow channel, incubated to allow binding, and washed out to measure release. Continuous monitoring by TIRF microscopy enables the rates of binding and release to be measured simultaneously for all different ligand configurations. Regions with the optimal spacing are then correlated with the structure of the binding complex, where known.

In diagnostic applications, configurations are selected that give the greatest discrimination in binding between different complexes or bacterial species. The rate-limiting step in these measurements can be the off rate, since binding affinities are expected to be very high for the multiple sites. For some applications where multiple measurements are to be made, the device is washed with mild denaturants to remove the bound complexes.

Stamping or imprinting enables the mass production of the arrays and the TIRF microscopy is automated to perform the measurements on a series of samples. Standard arrays are constructed for general research applications.

Molecular specificity with randomly oriented ligands on surfaces confers a range of binding affinities whereas this technique should provide critical spatial ordering on the molecular scale that would show a high degree of specificity. Many cellular functions rely upon this spatial order to confer specificity and there are an increasing number of nanodevices that will have a similar spatial order on the nanometer level. Measuring spatial dependent binding is critical for defining both cellular and artificial complexes.

The device according to some embodiments of the disclosed subject matter utilizes the spatial dependence of binding to provide further specificity of binding. Several different spacings are placed on the surface in adjacent areas in the microscope field so that relative binding specificity is measured simultaneously. A microfluidics flow channel over the device enables measurements to be made with as little as 1 microliter of material. The amount of binding is determined by total internal reflection fluorescence microscopy and on and off rates from regions with different spacings enables the proper spacing to be determined. For larger scale purifications and analyses, the optimal spacing is used to create an area large enough for a standard ellipsometry measurement, thus enabling purification and other types of analyses that need larger amounts of material, and expanding the range of uses of the spatial binding techniques.

Specificity of binding is much greater with proper spacing of multiple sites because physical chemical calculations indicate that the binding affinity increases to at least the product of the affinity constants. In addition, other spacings on adjacent regions of the same device provide controls for non-specific interactions. Because the volume of material needed for a measurement is very small (1 microliter), it is possible to test extremely small samples.

In Vitro Motor and Motility Analysis

Motor proteins such as kinesin and cytoplasmic dynein have been studied extensively at the single molecule level. However, there are many questions about the structure of these proteins and the mechanism of motility that cannot be addressed without being able to alter the array of proteins that the motors move upon. For dynein, there are also questions about the in vitro synthesized motor domain function that can be addressed by arraying it on the surface. In terms of the microtubules, different microtubule arrays have been formed but there is no current mechanism to systematically alter the array of tubulin subunits and then to assay for binding and/or motility. Using the device and methods of the disclosed subject matter, the effect of tubulin dimer spacing on motor binding are determined using GFP-tubulin that is bound to a specific array of anti-GFP antibodies. A parallel analysis of myosin binding to arrayed actin monomers is also performed. By engineering tubulin dimers, dimers that can be oriented on a stamped array are developed. The ordered arrays provide one means of carrying out an in vitro motility assay and determining the effect of array characteristics on motor function. This is an area of motor function that has not been addressed because it has not been possible with earlier technologies.

While much is known of the molecular mechanism of other classes of motor proteins, insight into the mechanism of action of the dyneins remains limited because of their large size. The dyneins are a family of microtubule based motor proteins responsible for ciliary and flagellar motility, and play diverse roles in cell division, organelle transport, and cell movement. The dynein motor domain is unrelated to that of other cellular motors, but is a relatively divergent member of the class of ATPases associated with cellular activity. Each dynein heavy chain subunit contains six AAA modules arrayed in a ring, from which two projections emerge. One is referred to as the stalk, and binds microtubules to its distal tip about 10 nm from the edge of the AAA ring. How ATP hydrolysis by the AAA units is converted into movement is poorly understood. Kinesin is much better understood as a motor. The microtubule binding characteristics of kinesin have been extensively studied and provide a nice comparison for the dynein binding characteristics.

Ordered nanoarrays provide an improved and in many cases unique tool to address a variety of important questions regarding dynein mechanochemistry. Cytoplasmic dynein, a two-headed motor protein, was found to act processively in its interaction with microtubules (Wang (1995) Biophys. J., 69:2011-23). Whether the individual motor domains are capable of sustained force production or processive behavior remains to be determined. A single-headed form of flagellar dynein was subsequently reported to produce force processively (Sakakibara (1999) Nature, 400:586-590). It is uncertain whether this feature is a special evolutionary adaptation of this molecule. The dynein molecules were randomly adsorbed to coverslips using a simple adsorption method performed at low protein concentrations. Although the distribution of particles was confirmed by TIRF microscopy using a fluorescent ATP analogue, aggregates of dynein are not readily detected. Furthermore, this method detects fully active dynein, but so-called “dead-heads,” which fail to bind ATP but bind microtubules strongly, would be missed.

The rat cytoplasmic dynein motor domain has been expressed using baculovirus infection of insect cells. Full-length dynein heavy chain has also been expressed and purified. Although it had some tendency to aggregate, full-length dynein HC was found to have limited motility activity (Mazumdar (1996) Proc. Natl. Acad. Sci. USA, 93:6552-6). A construct of about 350 kDa corresponding to the complete motor domain, and lacking the projecting microtubule-binding region, referred to as the stalk (Gee, (1997) Nature, 390:636-9) has also been produced. Each of these constructs has a hexahistidine and an epitope tag, which is used for purification and/or linkage to the nanoarray supports. Both constructs act as unique species by sedimentation and sizing chromatography, and have high levels of ATPase activity. The motor domain construct binds microtubules efficiently, which are released using ATP. These properties are strongly consistent with mechanochemical activity.

Additional experiments include testing for microtubule activation of the ATPase activity, a further sign of motor activity. Microtubule gliding assays, in which the motor domains are adsorbed at high density to coverslips, are performed as a more direct test for force production. The motor domains are then attached to Ni²⁺ posts. The spacing between posts is varied to test whether the motor domains act alone or cooperatively. If spacing of less than about 12 nm (i.e., the diameter of the motor domain) is required for full microtubule gliding activity, this suggests that two motor domains must act in concert. If the spacing must be even shorter than this distance, it suggests that the motor domains interact through the flat top or bottom surfaces of the AAA rings. A compact morphology for the two cytoplasmic dynein motor domains has been observed in one study (Amos (1989) J. Cell Sci., 93:19-28), though no direct evidence for such an interaction between dynein motor domains has been forthcoming. If the motor domains are functional at greater spacing, this will provide evidence that the individual domains are functional. In order to ensure no more than one motor domain per microtubule, short microtubules (about 1 μm) are applied to nanoarrays bearing dynein motor domains at even greater spacing.

If the motor domain fails at any spacing to support microtubule gliding activity, this means that the peptide tag being used can be located at a suboptimal site. Versions of the motor domain with tags at each end are then tested. Different modes of attachment are also tested, for example using antibody against the epitope tag to increase the flexibility and length of the link to the coverslip.

If motility is observed using both motor domain constructs, it is of considerable interest to compare the step size obtained in each case. Based on single particle image averaging of dynein electron micrographs, it is proposed that the power stroke primarily involves a shift in the interaction between the stem and the AAA ring, and secondarily between the microtubule-binding stalk and the AAA ring. By measuring the step size for each construct (Gelles (1988) Nature 331:450-3; Wang (1995) Biophys J., 69:2011-23), the disclosed subject matter provides methods for testing directly which portion of the motor domain makes the more significant contribution, and, in turn, gain valuable new insight into how the motor domain functions.

Numerous fragments of the cytoplasmic dynein heavy chain using both baculovirus infection of insect cells and transformation of bacterial cells for recombinant protein production have been produced to analyze dynein function. Of particular interest is the full motor domain construct, which contains all of the elements thought to be required for force production. This construct is being produced in milligram quantities with either a hexahistidine or FLAG epitope tag. At least two types of experimental set-ups are possible using ordered nanoarrays. First, the dynein motor domain is attached to Ni²⁺ bearing dots through the hexahistidine tag synthesized as part of the full motor domain construct. Microtubules are applied to the array in the presence of ATP, and microtubule gliding motility is evaluated and quantified as a function of dot spacing. These experiments provide new insight into whether dynein motor domains function by a cooperative mechanism and how the optimal spacing between motor domains compares with the spacing between tubulin subunits in the microtubule lattice. The nanoarrays of dynein are also tested for their ability to seed crystallization, as a first step toward determining the structure of the motor domain at atomic resolution.

To test for the important spacing in the substrate array, tubulin and tubulin fragments are linked to the dots within the nanoarray. Initially, tubulin dimers are linked by hexahistidine tags to Ni-NTA dots, giving a random orientation of the dimers. A similar experiment is performed with hexahistidine actin to look for myosin binding. To look for more detailed aspects of the binding and the motility, an oriented array of tubulin dimers is developed. Hexahistidine is used for one subunit and cysteines used for the other subunit. Cysteine reacts with gold and hexahistidine reacts with Ni-NTA dots, respectively. The gold and Ni-NTA posts are successively imprinted on the substrate with a directed 5 nm displacement.

For measurement of binding and possibly mobility, the dynein motor domain is bound to latex beads or native dimers will be purified. The beads or dynein dimers are applied to the nanoarrays and tested for binding or dynein-mediated movement. These experiments determine the spacing of the tubulin binding sites, step size inherent in the dynein crossbridge cycle, and whether the motor protein can accommodate to an imperfect lattice and still produce force.

Analysis of Integrin and Actin Interactions as a Function of Spacing

There are many large cytoskeletal proteins with multiple binding sites that are spaced by 20-100 nm (see Djinovic-Carugo, K. et al. (2002) FEBS Lett., 513:119-23; Goldmann, W. H. et al. (1996) J. Muscle Res. Cell Motil., 17:1-5; Liu, S. et al. (1997) Eur. J. Biochem., 243:430-6). These proteins have critical functions in cells and tissues that depend upon their specific binding to other components. We are exploring the spatial dependence of their binding interactions by measuring their relative interaction with different arrays of their binding partners. Theoretical analyses suggest that the proper spacing can increase binding avidity by orders of magnitude. Nanofabricated arrays now make it possible to measure the exact spatial dependence of the binding interactions. As one example, we are exploring the spatial dependence of integrin and actin arrays on their interactions with a variety of binding partners both in vivo and in vitro.

There is considerable evidence that the spacing between liganded integrins strongly affects binding to cytoplasmic proteins such as talin. For example, the binding of a fibronectin trimer causes specific attachment of talin1 to the cytoplasmic tail of integrin avb3. Talin1 is an anti-parallel dimer with both actin and integrin binding sites that has an overall length of about 50 nm whereas the fibronectin-integrin binding sites on the trimer are 40-70 nm apart. One explanation for the specific binding of the trimer to the actin cytoskeleton is that the spacing of the liganded integrins matches the spacing of the talin1 integrin binding sites. To test for this, arrays of cytoplasmic domains of integrins with different spacings are created. If a particular spacing binds GFP-talin1 more avidly than other spacings, a determination is made as to whether this corresponds to the full length of the talin1 dimer or to some other parameter of the molecule.

More specifically, recent studies indicate that the spacing of integrins and of actin is critical for the specific binding of many proteins (Calderwood, D. A. and M. H. Ginsberg (2003) Nat. Cell Biol., 5:694-97). For example, a single trimer of fibronectin (FIG. 1(a)) forms a slip bond with the cytoskeleton (FIG. 1(b)), whereas randomly spaced monomers of fibronectin do not (Jiang, G. et al. (2003) Nature, 424:334-37). A critical aspect of forming slip bonds is the selective binding of talin1. Talin1 is an anti-parallel dimer with an overall length of 56 nm (Winkler, J. et al. (1997) Eur. J. Biochem., 243:430-36). Since talin has an integrin binding site at the N-terminal end and an actin binding site at the C-terminus, an optimal spacing for beta integrin binding at both sites is theoretically about 50 nm. In the fibronectin trimer, the spacing between the pairs of RGD binding sequences is maximally about 60 nm (Coussen, F. et al. (2002) J. Cell Sci., 115:2581-90). The match between the spacing of the binding sites on the outside of the cell and those on the inside is very good and could be an important factor in creating the specific binding complexes. Indeed, the trimer rapidly forms a slip bond to the cytoskeleton that is broken at a force of 2 pNewton whereas monomer-coated beads bind more slowly and don't show preferential breaking at 2 pN. It is believed that the spacing of the monomeric integrins on beads varied and was often much greater than can be spanned by talin1. The disclosed subject matter enables placing fibronectins or integrins at specific spacings on a surface. The optimal spacing can be determined by in vitro and in vivo testing.

Cells bind to extracellular matrix-coated glass differently than to the same matrices in three dimensions, as a result of the spatial organization of matrix subunits. We are analyzing cell spreading on different arrays of fibronectin with defined spacings in the range of 20-150 nm. Initially, we will prepare substrates with 30 micron squares with a given array of fibronectin to allow cells to bind to that array. A small spacing is used and increased to determine the maximum separation that allows binding. Pairs of gold dots (2-5 nm in diameter) are centered in 150×300 nm areas and the spacing between the dots is varied from 25 to 150 nm. Other arrays with rows of dots spaced by 20-150 nm are formed at a row-to-row spacing of 150 nm. Stamping or imprinting technology produces more closely spaced rows (giving square arrays from 20×20 nm). The order in the arrays produces order in the cell spreading process. Cell spreading analysis enables us to quantitatively analyze the effect of order on spreading. The spreading process of talin1 mutant cell lines and other cell lines are also investigated.

There are many actin binding and integrin binding proteins that have an anti-parallel dimer structure (see Dhermy, D. (1991) Biol. Cell., 71:249-54; Goldmann, W. H. et al. (1996) J. Muscle Res. Cell Motil., 17:1-5; Liu, S. et al. (2000) Eur. J. Biochem., 243:430-6). The length of these molecules is somewhat variable because they are often composed of repeated domains such as the spectrin domains that have flexible linkages between the domains. Theoretically, alpha actinin can span about 60 nm whereas the larger talin molecule can span over 100 nm and is reported to have 4 binding sites for beta 1 and 3 integrin cytoplasmic tails. Since it is believed that talin causes a separation of the alpha and beta tails, it is likely that the isolated beta cytoplasmic domains will be capable of binding talin. For in vitro experiments, the device has smaller areas (1.5×3 μm) with the same spacings of the gold posts. This gives 100 binding pairs over that area, which can be viewed readily in TIRF.

The array of posts on a clean glass surface is prepared, (Chemiasvskaya et al., (2005), J. Vac. Sci. and Technol. B, 23:2972) and then the open areas of glass are reacted with PEG-sylanizing reagent to prevent non-specific absorption of the protein. To ensure the proper orientation of the bound proteins, the beta cytoplasmic tails are expressed in bacteria with a construct that places biotin on the amino terminus, which faces the membrane. Avidin is bound to the gold dots and then the biotin fragments will be added. Initially, the sites are saturated with biotin to examine the effect of spacing. Addition of a cysteine in the N-terminal region of the peptide enables us to fluorescently tag the fragment and then assay the density of bound cytoplasmic fragments in regions with single 5 nm gold dots spaced by over 0.5 micron from other dots. Images of single fluorophores are quantified to determine the average number of fluorophores per dot, using the intensity and the bleaching characteristics, since bleaching of one of two fluorophores cuts the intensity in half, while single fluorophores blink out, as readily apparent to one of ordinary skill in the art. Similarly, binding regions are manifest as a function of incremental changes in fluorescence level.

EXAMPLES

Example 1

As shown in FIG. 2, a CAD software, such as CADENCE from SYNOPSIS or DESIGNCAD from IMSI, is utilized to form a pattern that can be utilized to form an array of posts. First, a thin layer of PMMA is deposited on a silicon wafer using spin casting. The pattern formed using the CAD software is transferred onto the PMMA layer utilizing electron beam lithography, such as a scanning electron microscope (FEI XL 30 SIRION) equipped with a NABITY NPGS pattern generator or a LEICA VB6-HR. After the pattern is transferred, the PMMA layer has regions of holes of approximately 5 to 20 nm. A thin layer of titanium (ranging from about 0.5 to 30 nms) is deposited on the surface of the wafer. Lift-off of the PMMA layer provides posts of titanium standing on the surface of the silicon wafer. Size reduction of the titanium posts is carried out under a non-oxidizing environment at about 400° C. for about 1 hour in argon or nitrogen gas. The titanium posts are oxidized to form posts of titanium dioxide at about 300° C. in air for a few minutes. The titanium dioxide posts can have advantages over gold posts because titanium dioxide does not significantly contribute to fluorescence quenching when observing fluorescent molecules bound on the titanium dioxide posts.

Example 2

As shown in FIG. 2, a CAD software, such as CADENCE from SYNOPSIS or DESIGNCAD from IMSI, is utilized to form a pattern that can be utilized to form an array of pits. A thin layer of nickel is deposited on the silicon wafer surface followed by a thin layer of gold-palladium alloy (ranging from about 0.5 to 30 nms). The gold-palladium alloy is utilized because the grains of the gold-palladium alloy are significantly smaller than the gold and provide a smoother surface. A layer of PMMA (approximately 20 to 60 nm thick) is deposited on top of the gold-palladium alloy layer using spin casting. The pattern obtained using the CAD software is transferred onto the PMMA layer utilizing electron beam lithography, such as a scanning electron microscope (FEI XL 30 SIRION) equipped with a NABITY NPGS pattern generator or a LEICA VB6-HR. After pattern transfer, the PMMA layer has regions of holes of approximately 5 to 20 nm. Reactive ion etching or ion milling is carried out to form pits of gold-palladium on the gold-palladium layer to form pits in the gold-palladium layer. Any PMMA material is optionally removed using a solvent (e.g., acetone) or by heating above the depolymerization temperature of PMMA (e.g., 200° C.).

Example 3

As shown in FIG. 2, a CAD, such as CADENCE from SYNOPSIS or DESIGNCAD from IMSI, is utilized to form a pattern that can be utilized to form an array of pits. To carry out nanoimprint lithography, a master template is formed as follows. On a silicon wafer, PMMA layer is deposited and the pattern obtained from the CAD software is transferred onto the PMMA layer using electron beam lithography. Then, reactive ion etching or ion milling is carried to form posts of silicon. On a different silicon wafer, a layer of titanium is deposited and oxidized to form a layer of titanium dioxide. On top of the titanium dioxide layer, a layer of polystyrene is deposited using spin casting, where the thickness of the polystyrene layer is smaller or same as the height of the silicon posts described above. The silicon wafer having the polystyrene layer is heated above the glass transition temperature of the polystyrene (e.g., 180° C.) and the wafer having silicon posts is pressed onto the polystyrene layer and cooled down to room temperature. Once the polystyrene has vitrified, the wafer having silicon posts is removed leaving a pattern of holes in the polystyrene layer. Then, ion milling is carried out to form pits in the titanium dioxide layer. Any remaining polystyrene is removed using solvents (e.g., toluene).

Example 4

As shown in FIG. 2, a CAD software, such as CADENCE from SYNOPSIS or DESIGNCAD from IMSI, is utilized to form a pattern that can be utilized to form an array of patches. A thin layer of nickel is deposited on the silicon wafer surface followed by a thin layer of gold (ranging from about 0.5 to 30 nms). A layer of PMMA (about 20 to 60 nm thick) is deposited on top of the gold layer using spin casting. The pattern obtained using the CAD software is transferred onto the PMMA layer utilizing electron beam lithography, such as a scanning electron microscope (FEI XL 30 SIRION) equipped with a NABITY NPGS pattern generator or a LEICA VB6-HR. After pattern transfer, the PMMA layer has regions of holes, exposing small areas of gold of about 5-20 nm in size. A solution or vapor of thiol-functionalized molecules is brought in contact with the exposed gold regions to form a monolayer of molecules. The PMMA layer is removed using a solvent (e.g., acetone) producing patches of molecules on the gold surface.

While the disclosed subject matter has been described in detail with reference to some embodiments thereof, it will be understood that the disclosed subject matter is not limited to these embodiments. Indeed, modifications and variations are within the spirit and scope of that which is described and claimed.

Claims

1. A device for measuring nanometer level binding reactions, the device comprising:

(a) a surface; and

(b) an ordered array of posts, pits, or patches on the surface, wherein the posts, pits, or patches are capable of binding a protein or small molecule ligand, wherein the pitch between adjacent posts, pits, or patches is less than about 100 nm, and wherein each post consists essentially of titanium dioxide or gold-palladium alloy.

2. The device of claim 1, wherein the device comprises a gradual change of pitch that varies from about 5 nm to about 100 nm.

3. The device of claim 1, wherein the post, pit, or patch has affixed thereto at least one protein molecule.

4. The device of claim 3, wherein one of the post, pit, or patch has affixed thereto exactly one protein molecule.

5. The device of claim 1, wherein the protein molecule comprises at least a portion of a dynein heavy chain, a tubulin, a kinesin, a myosin, a cytoplasmic domain of an integrin, an actin, an extracellular matrix protein, a fibronectin, a collagen, a laminin, a DNA solenoid, a DNA, a histone-DNA complex, an RNA, an RNA-protein complexes, a bacterial coat proteins, an antibody, a lectin, an avidin, or any combination thereof.

6. The device of claim 1, wherein the protein molecule comprises a full-length dynein heavy chain.

7. The device of claim 1, wherein the protein molecule comprises a dynein motor domain.

8. The device of claim 7, wherein the protein molecule comprises an N-terminal portion of the dynein motor domain.

9. A method for measuring a nanometer level binding reaction, the method comprising

(a) providing a device comprising a surface; the surface comprising an ordered array of posts, pits, or patches, wherein the pitch between adjacent posts, pits, or patches is less than about 100 nm, wherein each post, pit, or patch is coated with ligand, and wherein each post consists essentially of titanium dioxide or gold-palladium alloy;

(b) contacting the surface of the device with a sample; and

(c) determining whether or not an analyte from the sample interacts or binds to a ligand-coated post, pit, or patch to measure the nanometer level binding reaction.

10. The method of claim 9, further comprising isolating the analyte from the sample.

11. The method of claim 9, wherein the device comprises a gradual change of pitch that varies from about 5 nm to about 100 nm.

12. The method of claim 9, wherein the post, pit, or patch has affixed thereto at least one protein molecule.

13. The method of claim 12, wherein one of the post, pit, or patch has affixed thereto exactly one protein molecule.

14. The method of claim 12, wherein the protein molecule comprises at least a portion of a dynein heavy chain, a tubulin, a kinesin, a myosin, a cytoplasmic domain of an integrin, an actin, an extracellular matrix protein, a fibronectin, a collagen, a laminin, a DNA solenoids, a DNA, histone-DNA complex, an RNA, an RNA-protein complex, a bacterial coat protein, an antibody, a lectin, an avidin, or any combination thereof.

15. The method of claim 12, wherein the protein molecule comprises a full-length dynein heavy chain.

16. The method of claim 12, wherein the protein molecule comprises a dynein motor domain.

17. The method of claim 16, wherein the protein molecule comprises an N-terminal portion of the dynein motor domain.