WO2008080441A1 - Micropatterning of biomolecules - Google Patents

Abstract

The present invention concerns methods and compositions for detecting, monitoring, or measuring various intracellular processes and molecular interactions. In one embodiment, the present invention provides a method for assessing a molecular interaction comprising: (a) providing a cell comprising a labeled target molecule and a membrane moiety having an extracellular domain; (b) providing a patterned surface comprising a pattern of interaction areas, the interaction areas comprising immobilized ligands specific to the extracellular domain of the membrane bound moiety; (c) contacting the cell with the patterned surface; and (d) detecting a signal from the labeled target molecule, wherein a greater signal at the interaction areas of the patterned surface as compared to areas between the interaction areas indicates a molecular interaction between the labeled target molecule and the membrane bound moiety.

Description

Micropatterning of biomolecules

This application claims the benefit of the filing date of United States Provisional Patent Application No. 60/882,652 filed December 29, 2006, the disclosure of which is hereby incorporated herein by reference.

BACKGROUND OF THE INVENTION

Field of the Invention

The present invention relates generally to the field of molecular biology. More particularly, it concerns methods for the detection and qualitative and quantitative measurement of molecular interactions in cells.

Description of Related Art

Cellular signaling is mediated by sequential interactions of numerous receptor, effector and messenger proteins, which specify the particular signaling pathway.

Recent advances in bioinformatics yielded the theoretical framework for identifying and characterizing signaling pathways and their interconnections within the cellular signaling network, thereby laying the basis for systems biology. Large-scale measurements provide data on protein interactions, predominantly via strategies in which the two interaction partners act like bait and prey. The most straightforward strategy is based on immunoprecipitation of the fluorescently labeled prey. Alternatively, bait and prey can be linked to non-functional subunits of a complex; upon interaction, the two subunits are brought into contact and form the activated complex, e.g. a transcription factor in yeast 2 hybrid screens or a green fluorescent protein (GFP).

The major advantage of these approaches is their suitability for large scale screening. This advantage is, however, counterbalanced by the artificial character of the systems, which renders the systems susceptible to false positives and negatives. For example, with immunoprecipitation approaches the molecular interactions have to be stable enough to endure the preparation steps. In yeast 2 hybrid screens, molecular orientation and distance may hamper the formation of the readout complex. In particular, membrane proteins - the key players in early signaling - are difficult to analyze. Moreover, the local environment and protein content of yeast is different from mammalian cells, which may well influence protein interactions.

A different approach for sensing molecular proximity is based on energy transfer between a donor and an acceptor dye (Fόrster Resonant Energy Transfer, FRET). While this method has the advantage of being applicable to living cells, the interpretation of the results is severely complicated by the requirement of precise knowledge on the spectral properties of the dyes involved.

The ability to identify protein-protein interactions is important for applications such as characterizing signaling pathways and their interconnections within the cellular signaling network. Although various approaches for studying protein-protein interactions are known (e.g., the yeast 2 hybrid system, discussed above), they have their limitations. Accordingly, there is a need for new approaches to study intracellular processes and molecular interactions.

Recently, several research groups have introduced methods to arrange membrane proteins in specific patterns within the plasma membrane of living cells. Wu et al. (PNAS USA 101 (2004): 13798-803) used micrometer-size patterned lipid bilayers containing liganded lipids to control the location and size of receptor clusters on the cell membrane and enable direct visualization of structural reorganization of cellular components. Triggering the FCs receptor via the microstructured lipid bilayers allowed for studying the formation and composition of signaling complexes in the cell membrane (Wu et al., PNAS USA 101 (2004): 13798-803). US 2002/0160505 describes a method for immobilizing cells onto discrete areas with lipid membranes on a microarray for screening and modulation of living cell adhesion and growth on a solid substrate. US 6,559,474 describes a method and an apparatus for patterning of biological carrier materials in the nm-μra range. WO 01/88182 relates to the immobilization of cells on lipid bilayer membranes. US 2003/0143634 describes an intra-cellular 3-part hybrid system (referred to as "improved GFP assisted Readout of Interacting Proteins (iGRIP)") for detecting protein interactions.

SUMMARY OF THE INVENTION

It is an object of the invention to enable an efficient analysis of molecules, particularly of biomolecules.

In order to achieve the object defined above, a method for assessing a molecular interaction, a method for detecting a molecular interaction, a solid substrate comprising a patterned surface, a kit, and a measurement device according to the independent claims are provided.

According to an exemplary embodiment of the invention, a method for estimating the binding kinetics of molecular interactions in living cells may be provided. Particularly, a combination of micropatterned surfaces, TIRF (total internal reflection fluorescence) microscopy, and/or FRAP (fluorescence recovery after photobleaching) allows to estimate the binding kinetics of molecular interactions in living cells essentially without unspecific background. This may be particularly achieved by specifically designing the capture molecules. Taking such a measure may be important in order to provide interaction networks in a living cell. Thus, a quantitative measurement may be made possible which provides a significant improvement over conventional qualitative ("Yes/No" logic) methods. Furthermore, it may also become possible to measure very weak interactions. Beyond this, a study of the influence of different mutations and reagents on the cell may become possible.

By providing substrates having a patterned surface with interaction areas and non- interaction areas, high throughput screening may be made possible since different capture molecules may be immobilized on different interaction surfaces. By providing blocking molecules between the interacting surfaces, highly reliable signals may be compared and background signals may be eliminated or suppressed. Biochemical interactions occurring on such a substrate may be securely detected with a measurement device analyzing such a substrate using electromagnetic radiation sources and detectors. Particularly, the combination of such substrates with FRAP for analysis allows to properly detect quantitatively the interactions by acquiring recovery after a photobleaching pulse.

Embodiments of the present invention provide methods and compositions for monitoring various intracellular processes and molecular interactions, as well as for detecting or screening for interactions between two molecules (e.g., protein-protein interactions). In one embodiment, the present invention provides a method for assessing a molecular interaction comprising: (a) providing a cell comprising a labeled target molecule and a membrane moiety having an extracellular domain; (b) providing a patterned surface comprising a pattern of interaction areas, the interaction areas comprising immobilized ligands specific to the extracellular domain of the membrane bound moiety; (c) contacting the cell with the patterned surface; and (d) detecting a signal from the labeled target molecule, wherein a greater signal at the interaction areas of the patterned surface as compared to areas between the interaction areas indicates a molecular interaction between the labeled target molecule and the membrane bound moiety. Typically, the signal on the interaction areas is compared with a signal between the interaction areas. This allows a measurement of the difference between the specific signal on the interaction areas and non-specific binding or background noise. Of course, alternatively standardized reference values for the signal can be used for a quick evaluation of the interaction.

The membrane moiety or transmembrane moiety facilitates the measured interaction with the labeled target molecule and connects for visualization to the surface pattern. hi particular embodiments the membrane moiety is a biomolecule, such as a protein, nucleic acid, carbohydrate or lipid, hi certain aspects of the invention, the biomolecule is a protein such as a membrane protein. Any molecule, especially macromolecules, can be modified to comprise a membrane anchor to allow the localization of the membrane moiety on a cellular membrane. Transmembrane proteins have this feature inherently and have protein domains on both sides of the membrane. Accessibility from both sides of the membrane facilitates antibody binding and interactions of the bait molecule in the cytoplasm. This moiety is preferably a transmembrane protein but can also be selected from modified cytosolic proteins with an artificial transmembrane anchor.

In certain aspects of the invention, the labeled target molecule has a fluorescent label. The fluorescent label may be a fluorescent protein, such as a green fluorescent protein (GFP), yellow fluorescent protein (YFP), cyan fluorescent protein (CFP), or a red fluorescent protein (RFP), or their variants. Further examples of labels that may be used in embodiments of the present invention include chemical labels, especially optical labels such as chemical fluorophores, or radioactive labels. The labeled target molecule may be any biomolecule of interest including, for example, a protein, nucleic acid, carbohydrate, or lipid, hi certain aspects of the invention the labeled target molecule is a cytosolic protein. In other aspects of the invention, the labeled target molecule is a membrane bound protein. In certain aspects of the invention, the cell is transfected with an expression construct encoding a fusion protein comprising a target moiety and a label moiety.

The ligands immobilized on the interaction areas of the patterned surface may be any molecule that specifically binds to the extracellular domain of the membrane bound protein. The ligands may be, for example, proteins, carbohydrates, lipids, or nucleic acids. In certain embodiments, the ligands may be DARPins (Binz et al., Nat. Biotechnol. 22:575 (2004)), toxins, cytokines, chemokines, particles like LDL, HDL, extracellular signaling molecules like EGF, TNF, and aptamers that can bind specific membrane receptors with high affinity. In certain embodiments of the invention, the ligands are antibodies against the extracellular domain of the membrane bound protein or a signaling protein that binds the membrane bound protein. The antibody may be, for example, a monoclonal antibody, polyclonal antibody, or antibody fragment, such as Fab, Fab', F(ab)₂ and scFv (single chain variable fragment). The ligands may be immobilized on the surface by any method known to those in the art. In one embodiments of the invention, the ligands are immobilized on the surface using a linking moiety, such as a streptavidin and/or a biotin. In another embodiment the ligands are directly spotted onto the surface. Linking is in a further embodiment based on interaction between His-tagged proteins and a Ni-NTA functionalized surface, or immobilization of antibodies via Protein A or Protein G.

hi certain embodiments of the invention, assessing a molecular interaction further comprises photobleaching the labeled target molecule and observing the recovery from the photobleaching. Fluorescence recovery may be used to measure the kinetics of the interactions between two molecules. In some embodiments of the invention, assessing the molecular interaction further comprises contacting the cell with a ligand of the target molecule or a ligand of the transmembrane protein (which may or may not be the same as the ligand immobilized on the patterned surface) and observing a ligand-mediated change in interaction between the target molecule and the transmembrane protein.

In another embodiment, the present invention provides a method in which a target molecule of interest is displayed on the surface of a cell - for example, as a fusion with a membrane protein. The displayed target protein may then be screened for interactions against any other molecule of interest that is immobilized on the patterned array. As with the methods described above, an interaction may result in a redistribution of the membrane protein corresponding to the patterned surface. Accordingly, in one aspect, the present invention provides a method for detecting a molecular interaction comprising: (a) providing a cell comprising a labeled target molecule, the labeled target molecule comprising an extracellular target moiety, a membrane moiety, and a label moiety; (b) providing a patterned surface comprising a pattern of interaction areas, the interaction areas comprising immobilized ligand molecules; (c) contacting the cell with the patterned surface; and (d) detecting a signal from the labeled target molecule, wherein a greater signal at the interaction areas of the patterned surface as compared to areas between the interaction areas indicates a molecular interaction between the target moiety and the immobilized ligand molecules. Typically, the signal on the interaction areas is compared with a signal between the interaction areas. This allows a measurement of the difference between the specific signal on the interaction areas and non-specific binding or background noise. Of course, alternatively standardized reference values for the signal can be used for a quick evaluation of the interaction. In certain embodiments of the invention, the method may be used to assess or measure the interaction between two molecules known to interact with each other. In other embodiments of the invention, the method may be used to screen a plurality of molecules to identify those molecules that interact with each other.

The extracellular target moiety may typically be a protein, polypeptide, or peptide, although it may be another biomolecule such as a nucleic acid, carbohydrate, or lipid anchored to the membrane moiety. The membrane moiety may be a protein, such as a transmembrane protein, or a lipid. Where the extracellular target moiety and the membrane moiety are proteins, they may be an extracellular domain and a membrane domain of the same protein or they may be a fusion of two different proteins. A linker moiety may be placed between the extracellular domain and the membrane domain. The label moiety may be coupled to the extracellular target moiety or to the membrane moiety. In certain embodiments, the cell expresses the labeled target molecule. In certain embodiments of the invention, the cell is transfected with an expression construct encoding a fusion protein comprising an extracellular target moiety, a membrane moiety, and a label moiety.

The methods of embodiments of the present invention may be used to detect changes in the interaction between a membrane moiety, e.g. a transmembrane protein, and a labeled moiety, e.g. a cytosolic protein, induced by interacting (or interfering) agents, e.g. interaction competitors or interaction regulators. Such interacting (or interfering) agents may be moieties that competitively bind the labeled moiety or agents that block the interaction between the membrane moiety and the labeled moiety. Such a method can be used for drug discovery. For example, the drug targets may be G-protein coupled receptors (GPCR). Substances are screened for modifying the interaction between the GPCR and the G protein. Using, for example, fluorescent G proteins as the labeled moiety and GPCRs as the membrane moiety, the release (or strengthening) of interaction can be easily observed. Due to its simplicity, the assay can be easily modified for high-throughput screening, in particular in conjunction with microfluidic devices.

The methods may also be used to identify regulators of a particular interaction. Many biomolecular interactions are indirect, and depend on adaptor proteins, or posttranslational modifications of the interaction partners. When using cells transfected with cDNA or RNA interference libraries, bait-prey interactions may be significantly modified in a subpopulation of cells, which can be harvested and analyzed. The up-regulated (for cDNA) or down-regulated (RNAi) genes correspond to candidates for regulators.

Therefore embodiments of the present invention provide the above methods with the addition of a potentially interfering or regulating agent, which modifies the interaction of the transmembrane moiety with the labeled target molecule, e.g. a competitive binder of the transmembrane moiety or the labeled target molecule, or an agent that modifies the binding strength of the transmembrane moiety to the labeled target molecule, such as a cofactor.

The immobilized ligand molecules in the interaction areas on the patterned surface may be proteins, nucleic acids, carbohydrates, or lipids. The ligand molecules may be immobilized on the patterned surface by a linking moiety such as, for example, a streptavidin and/or a biotin or spotted directly onto the surface.

The patterned surface may also comprise an inert, cell adhesive or blocking component. Such components may be blood proteins, or modified blood protein such as albumin, especially BSA (bovine serum albumin), or fibronectin. In certain embodiments of the invention, the inert, cell adhesive or blocking component is located between the interaction areas of the patterned surface, hi some embodiments of the invention a fluorescent molecule, distinguishable from a fluorescent molecule that may be used to label the target, is located between the interaction areas of the patterned surface. In one embodiment of the invention, the fluorescent molecule located between the interaction areas of the patterned surface is Cy5. In certain embodiments of the invention, the space between the interaction areas contains a blocking component, such as BSA, to inhibit nonspecific adsorption of proteins or other biomolecules.

The patterned surface may have a regular pattern or an irregular pattern of interaction areas. The size of the interaction areas and the spacing of the interaction areas may be such that when a cell is in contact with the patterned surface the cell will cover a plurality of interaction areas. The interaction areas may be essentially any shape including, for example, essentially square, rectangular, triangular, or circular. In certain embodiments of the invention, the interaction areas have a length and/or width of about 0.5 μm, 1 μm, 2 μm, 3 μm, 4 μm, 5 μm, 6 μm, 7 μm, 8 μm, 9 μm, or 10 μm, and a width of about 0.5 μm, 1 μm, 2 μm, 3 μm, 4 μm, 5 μm, 6 μm, 7 μm, 8 μm, 9 μm, or 10 μm, or any range of lengths and widths derivable therein. In some embodiments of the invention, the interaction areas are essentially circular and have a diameter of about 0.5 μm, 1 μm, 2 μm, 3 μm, 4 μm, 5 μm, 6 μm, 7 μm, 8 μm, 9 μm, or 10 μm, or any range therein. In certain embodiments of the invention, the spacing between the interaction areas is about 0.5 μm, 1 μm, 2 μm, 3 μm, 4 μm, 5 μm, 6 μm, 7 μm, 8 μm, 9 μm, or 10 μm, or any range therein. In certain embodiments of the invention, the patterned surface comprises 2, 3, 4, 5, 6, 7, 8, 9, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 1000, 10000, 100000, 10⁶, 10⁷, 10⁸, 10⁹ or 10¹⁰ or any range therein, interaction areas.

In certain embodiments, the surface is made from a light-transmitting material, such as glass, quartz, or transparent synthetic polymeric materials such as polyethylene, polypropylene and polyphenylene sulfide.

In certain embodiments of the invention, all of the interaction areas on the patterned surface contain the same prey molecule. In other embodiments of the invention, the interaction areas are grouped into regions with the interaction areas in one region containing prey molecules different from the prey molecules contained in the interaction areas of at least one other region on the patterned surface. In this way, interactions between two or more different sets of molecules may be detected on the same substrate, hi certain embodiments of the invention, the patterned surface comprises 2, 3, 4, 5, 6, 7, 8, 9, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 1000, 10000, 100000, 10⁶, 10⁷, 10⁸, 10⁹ or 10¹⁰ or any range therein, different regions of interaction areas.

In one embodiment, the present invention provides a solid substrate having a patterned surface. In certain embodiments, the invention provides a solid substrate comprising a patterned surface comprising: a pattern of interaction areas, the interaction areas comprising immobilized ligands specific to an extracellular domain of a transmembrane protein; and spaces between the interaction areas, the spaces comprising a blocking molecule. Optionally, the spaces may further comprise a fluorophore. However, in other embodiments, the spaces may be free of a fluorophore. The blocking molecule may be any molecule that blocks unspecific adsorption of proteins or other biomolecules. In certain embodiments of the invention, the blocking molecule is BSA. The fluorophore in the spaces between the interaction areas may be any fluorophore as long as it is distinguishable from the label used to label the target molecule to be assayed. In certain embodiments, the fluorophore is Cy5. As discussed above, the patterned surface may have a regular pattern or an irregular pattern of interaction areas, and may be essentially any shape including, for example, essentially square, rectangular, triangular, or circular. In some embodiments of the invention, the solid substrate having a patterned surface is provided in a kit. hi certain embodiments, the invention provides a kit for making such a solid substrate having a patterned surface.

In one embodiment, the present invention provides a measurement device (or an analysis device) comprising a carrier (such as a support) adapted for receiving (or accommodating) a solid substrate having the above mentioned features, an electromagnetic radiation source (comprising one or more electromagnetic radiation emitting units such as lasers) adapted for irradiating at least one of the interaction areas (particularly a specifically selected one of the interaction areas at a time) with primary electromagnetic radiation (which may have a wavelength appropriate for exciting fluorescence labels attached to the solid substrate), and an electromagnetic radiation detector adapted for detecting secondary electromagnetic radiation emitted/reflected from the at least one interaction area in response to the irradiation with the primary electromagnetic radiation (particularly for detecting a wavelength emitted from excited fluorescence labels).

In this context, the term "carrier" may particularly denote a component which is specifically designed to receive a solid substrate having the above-mentioned features. The term "electromagnetic radiation source" may particularly denote a device capable of generating electromagnetic radiation (particularly a continuous or pulsed beam of essentially monochromatic or polychromatic radiation).

The term "electromagnetic radiation" may particularly denote photons of an appropriate wavelength. Although embodiments of the invention are based on an optical irradiation with photons in a wavelength range of approximately 400 ran to approximately 800 nm, other embodiments of the invention may also use infrared, UV, or even X-rays.

The term "electromagnetic radiation detector" may denote any device capable of detecting electromagnetic radiation in a specific wavelength range in a qualitative or quantitative manner. Particularly, the electromagnetic radiation detector may be a charge coupled device (CCD), or may be a photodiode or an array of photodiodes.

A solid substrate having the above-mentioned features may be received by (that is to say may be mounted or assembled on) the carrier. A surface of the solid substrate bearing biologically active components may be exposed to light so as to be capable to perform an electromagnetic radiation based experiment.

The electromagnetic radiation source and the electromagnetic radiation detectors may be configured to form a fluorescence recovery after photobleaching (FRAP) arrangement. FRAP may be denoted as a live cell imaging technique used to study binding kinetics and/or the mobility of fluorescent molecules. A pulse of high intensity light may be used to photobleach a population of fluorophores in a target region. Recovery of fluorescence in the bleached region represents movement of fluorophores into that region. In other words, FRAP may be denoted as a technique used in cellular imaging where a fluorochrome attached to a molecule may be destroyed on purpose with an intense flash of light (by a laser). This may be performed in a well-defined area to study the repopulation of this area with peripheral molecules still fluorescent. This method may allow for a measurement of the speed of diffusion of molecules in living cells.

The electromagnetic radiation source and the electromagnetic radiation detector may be configured to form a total internal reflection fluorescence (TIRF) arrangement. A TIRF microscope may be denoted as a microscope with which a thin region of a specimen, for instance less than 200 nm, can be observed. A TIRF microscope may use evanescent waves to selectively illuminate and excite fluorophores in a restricted region of the specimen immediately adjacent to a solid-liquid surface. Evanescent waves may be generated when the incident light is totally reflected at the solid-liquid interface. The evanescent electromagnetic field may decay (for instance exponentially) from the interface, and thus penetrates to a depth of only approximately 100 nm into the sample medium. Thus, the TIRF enables the selective visualization of surface regions, particularly of an active biomolecule interaction region of the solid substrate according to an exemplary embodiment of the invention.

The measurement apparatus may be adapted to move the electromagnetic radiation source, the electromagnetic radiation detector, and the carrier relative to one another for scanning the interaction areas of the solid substrate (groupwise or individually). For instance, the solid substrate may comprise an array of interaction areas provided with different capture molecules. Thus, different portions of the solid substrate may be sensitive to different biomolecules allowing for high throughput screening applications, hi order to scan the individual interaction areas of the solid substrate, it is possible to keep the radiation source and radiation detector fixed, and to move the solid substrate on the carrier. Alternatively, it is also possible to move the optics (that is to say the electromagnetic radiation source and/or the electromagnetic radiation detector), and to maintain the solid substrate and the carrier spatially fixed. Particularly, the electromagnetic radiation source may comprise a first laser adapted for generating a pulsed laser beam having a first intensity and may comprise a second laser adapted for generating a pulsed or continuous laser beam having a second intensity, wherein the first intensity is larger than the second intensity. In other words, such a configuration may be used for FRAP, since the first laser photobleaches the sample, and the second laser uses a very low light amplitude to prevent any bleaching or saturation effects, thereby allowing to accurately detect recovery of the signal after photobleaching. A control unit may be provided for harmonizing or coordinating the function of the first laser and the second laser. In one embodiment, the sensing beam generated by the second laser may be a continuous beam or may be a pulsed beam.

The electromagnetic radiation source may comprise a laser adapted for generating a pulsed laser beam having a first intensity and may be adapted for generating a pulsed or continuous laser beam having a second intensity, wherein the first intensity is larger than the second intensity. In such an embodiment, both bleaching and sensing may be performed with a single laser which first bleaches and secondly detects recovery of the signal. The detection beam may be a continuous beam or may be a pulsed beam as well.

The electromagnetic radiation detector may comprise a charge coupled device (CCD). A charge coupled device (CCD) may be denoted as a sensor for recording images, comprising an integrated circuit containing an array of linked or coupled capacitors. Under the control of an electric circuit, each capacitor transfers its electric charge to one or other of its neighbours. According to an exemplary embodiment of the invention, CCDs may be used for high accuracy imaging of the fluorescence radiation generated by fluorescence particles close to the sensor surface of the solid substrate.

According to exemplary embodiments of the invention, micropatterning of membrane molecules may be performed. More particularly, embodiments of the invention enable live cell single molecule imaging and micropatterning of CD4 to reveal binding mechanisms to Lck.

Detection and quantification of protein interactions represent a cornerstone of proteomic research. Current screening platforms, however, rely on measuring protein interactions in highly artificial systems rendering the results difficult to confer on the in vivo situation. According to exemplary embodiments of the invention, a method is provided to identify and characterize interactions between a fluorescently labeled protein ("prey") and a membrane protein ("bait") in living mammalian cells. Cells transfected with a fluorescent protein fusion of the prey may be plated on micropatterned surfaces functionalized with specific antibodies to the exoplasmic domain of the bait. The technology may be applied for characterizing the interaction between CD4, the major co-receptor in T cell activation, and Lck, the protein tyrosine kinase essential for early T cell signaling. Equilibrium associations may be quantified by semi-automatic analysis of the Lck-micropatterns. Information on interaction dynamics may be obtained by combining the platform with photobleaching experiments and single molecule imaging, hi addition to the known zinc clasp, additional Lck domains may be identified contributing to CD4 binding. The membrane-anchor can be found to have a strong impact on the interaction: it mediates direct binding, and further stabilizes the interaction of other Lck domains. In total, membrane-anchorage may increase the interaction lifetime by two orders of magnitude. In addition, multiple membrane-distal domains can be found to modulate the binding affinity in a subtle way. The broad range of affinities contributed by different Lck domains render this molecule amenable to fine-tuning of its binding capabilities, thereby providing additional means for regulating Lck recruitment in cellular activation processes. With techniques according to exemplary embodiments of the invention, a general tool may be provided for characterizing protein interactions quantitatively - in particular its modulations and regulations in a live cell context. In this context, embodiments may use micropatterning tools in conjunction with photobleaching and single molecule imaging to characterize in detail the interaction between CD4, the major co-receptor in T cell activation, and Lck, the major Src family protein tyrosine kinase essential for early T cell signaling. Micropatterns of a specific monoclonal antibody to the bait protein CD4 may be formed on glass surfaces using microcontact printing, and may be found to yield specific CD4 redistribution in the plasma membrane of adherent cells. Protein interactions may be assayed via the concomitant redistribution of the fluorescent prey protein Lck.

It is contemplated that any method or composition described herein can be implemented with respect to any other method or composition described herein.

The use of the term "or" in the claims is used to mean "and/or" unless explicitly indicated to refer to alternatives only or the alternatives are mutually exclusive, although the disclosure supports a definition that refers to only alternatives and "and/or".

Throughout this application, the term "about" is used to indicate that a value includes the standard deviation of error for the device or method being employed to determine the value. Following long-standing patent law, the words "a" and "an," when used in conjunction with the word "comprising" in the claims or specification, denotes one or more, unless specifically noted.

Other objects, features and advantages of the present invention will become apparent from the following detailed description. It should be understood, however, that the detailed description and the specific examples, while indicating specific embodiments of the invention, are given by way of illustration only, since various changes and modifications within the spirit and scope of the invention will become apparent to those skilled in the art from this detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

The following drawings form part of the present specification and are included to further demonstrate certain aspects of the present invention. The invention may be better understood by reference to one or more of these drawings in combination with the detailed description of specific embodiments presented herein.

FIG. 1. Illustration of an assay/a solid substrate according to an exemplary embodiment of the invention. BSA-micropatterns were printed on functionalized glass coverslips using a microstructured PDMS-stamp. Interaction areas were filled with streptavidin and incubated with biotinylated antibody against a membrane protein. Cells expressing a fluorescence labeled cytosolic protein were incubated on the coverslips. Interactions between the antibody and the membrane protein result in a redistribution of the membrane protein in the cell membrane. Interactions between the membrane protein and the fluorescence labeled cytosolic protein were probed by measuring the degree of co-patterning. Specific protein-protein interactions between the membrane protein and the labeled cytosolic protein are characterized by enrichment of the fluorescence labeled protein at the interaction areas as illustrated in the cell on the left in FIG. 1. If no interaction occurs, the distribution of the fluorescence labeled protein will be homogenous as illustrated in the cell on the right in FIG. 1.

FIGs. 2 A to 2E. Micropatterning of the bait. T24 cells were transfected with CD4- YFP and plated on CD4 μ-biochips (FIG. 2A). The image was recorded in TIR configuration, showing a lxO.βmm² region of the chip. All transfected cells showed a redistribution of the bait CD4-YFP following the micropattern of the capture antibody. In the zoom (FIG. 2B) the 3 μm x 3 μm feature size of the micropattern is well resolved. FIG. 2C demonstrates the anticorrelation of CD4-YFP patterns with the BSA-Cy5 pattern recorded on the same area. As negative controls, T24 cells were transfected with CD4-YFP and plated on CD3 μ-biochips (FIG. 2D) or CD 147 μ- biochips (FIG. 2E); both result in a homogenous brightness over the plasma membrane.

FIGs. 3A to 3D. Interaction between the bait CD4 and the fluorescent prey Lck- YFP. T24 cells were cotransfected with unlabelled CD4 and Lck-YFP and plated on CD4 μ-biochips (FIG. 3A). Consistent with the expected stable association of Lck with CD4, Lck-YFP was found to closely follow the CD4 micropatterns. The specificity of the interaction was confirmed by plating cells transfected only with Lck-YFP on CD4 μ-biochips (FIG. 3B). Due to the lack of CD4 in the plasma membrane, no Lck-YFP micropatterns were detected. To test for the influence of the fluorescence tag, T24 cells were cotransfected with unlabelled CD4 and YFP-only (FIG. 3C); again, no redistribution of the fluorescence signal was observed. The interaction of Lck-YFP with a different membrane protein - endogenous CD 147 - was tested (FIG. 3D). T24 cells were transfected with Lck-YFP and plated on CD147 μ-biochips. Consistently, no redistribution of the fluorescence signal was observed.

FIGs. 4A to 4D. Zinc chelation releases CD4-Lck interaction. Cells cotransfected with unlabelled CD4 and Lck-YFP were plated on CD4 μ-biochips. Interaction of bait and prey was clearly visible (FIG. 4A) and released upon zinc chelation with 1 , 10-phenanthroline (FIG. 4B). Both TPEN and 1,10-phenanthroline yielded similar results. A statistical analysis of numerous cells is shown in density plots in FIG. 4C and FIG. 4D. For this representation, each feature of the micropattern was analyzed according to its fluorescence signal F and its contrast C. The 2D histograms represent 1609 (958) features obtained from 22 (20) cells. Upon chelation, a new population can be detected at high signal F and low contrast C. Interestingly, Lck-YFP micropatterns at the cell periphery characterized by a low fluorescence signal F typically sustain the chelation procedure. FIG. 5. Demonstration of the technique on a different cell line. HEK cells were cotransfected with unlabeled CD4 and lck-YFP and plated on CD4 μ-biochips. The results obtained on T24 cells could be qualitatively confirmed.

FIGs. 6A to 6D. Micropatterns of capture antibody on glass coverslips.

Micropatterns were prepared as described and incubated with Alexa555-labeled secondary antibody. The image shows a 270 μm x 190 μm detail of the coverslip scanned sequentially at 647 nm for excitation of Cy5-BSA (FIG. 6A) and at 514 run for excitation of Alexa555 (FIG. 6B). The overlay (FIG. 6C) demonstrates the efficient separation of the two reagents, yielding a high-contrast micropattern. For analysis of spots, an automatic gridding algorithm was applied to determine the grid- size and the rotation angle α of the image (FIG. 6D). The grid subdivides the total image into adjacent squares, each containing information about the feature brightness (F⁺) and the unspecific background (F^"). The background value F_bg was measured on biochip-regions containing no cells.

FIGs. 7A to 7D. Release of CD4-lck interaction (control for DMSO). Cells cotransfected with unlabeled CD4 and lck-YFP were plated on CD4 μ-biochips. Interaction of bait and prey was clearly visible (FIG. 7A). Treatment with mock solution (HBSS containing 0.5 % DMSO only) had no effect on the pattern (FIG. 7B). The 2D histograms in FIG. 7C and FIG. 7D represent 1235 (894) features obtained from 21 (21) cells.

FIGs. 8 A to 8D. Control for endurance of CD4 pattern upon zinc chelation. To confirm the endurance of the bait micropattern upon zinc chelation, CD4-YFP transfected cells were treated on CD4 μ-biochips with zinc chelator. The images of FIG. 8A and FIG. 8C of T24 cells before addition of the chelator, FIG. 8B and FIG. 8D show the same cell 10 minutes after addition of TPEN. No change in the bait micropattern was observed. The 2D histograms represent 1372 (767) features obtained from 25 (25) cells in FIG. 8B (FIG. 8D). FIG. 9. A measurement device according to an exemplary embodiment of the invention. FIG. 9 illustrates a combined TIRF and FRAP measurement.

FIG. 10. Sketch of all Lck-mutants used in the study. For clarity, the inserted fluorescent proteins were not shown in this plot. For wild-type Lck and Lck- C20/23 A, fusion constructs with C-terminal YFP were used, which contain an additional CFP-insertion at aal25. All other constructs were fused to mGFP at the C- terminus. Controls on mGFP -labeled wild-type Lck and Lck-C20/23 A revealed no difference compared to the presented constructs.

FIG. 11. Interaction of CD4 with various Lck-mutants. In each experiment, cells were cotransfected with CD4 and the respective Lck-mutant, and plated on CD4 μ- biochips. The left column shows typical images from single cells, the right column statistical analysis of many cells. For comparison, wild-type Lck was included in this plot (FIG. 1 IA: 1609 features obtained on 22 cells). The following constructs were analyzed: Lck-C20/23/A (FIG. 1 IB: 1643 features obtained on 16 cells), Lck-NIO (FIG. 11C: 1223 features obtained on 2 cells), a GPI-anchored mGFP (FIG. 1 ID: 524 features obtained on 12 cells), Lck-N65 (FIG. 1 IE: 1069 features obtained on 17 cells), Lck-ΔNIO (FIG. 1 IF: 1365 features obtained on 22 cells), Lck-ΔN65 (FIG. HG: 1351 features obtained on 16 cells), Lck-ΔN249 (FIG. HH: 1379 features obtained on 21 cells).

FIG. 12. Interaction kinetics between CD4 and Lck. Cells were cotransfected with CD4 and Lck- YFP (FIG. 12A) or the mutant ΔN10-mGFP (FIG. 12B), and plated on CD4 μ-biochips. Individual features of the micropattern were selected for the FRAP experiment. The illumination area was restricted to 20x5 μm² by imaging a field-stop. On the left, an image is shown before photobleaching, yielding the intensity- difference ΔF₀. Photobleaching reduced the fluorescence signal to zero, as apparent in the first recovery image obtained at 0.62s and 0.4s for wild-type and mutant, respectively. Both features recovered to the same shape, however, at dramatically different time-scales. The recovery of the signal ΔF(t) was analyzed by fitting with

^-£+ = (a - β) [l - exp(- t/τ)]+ β , yielding τ=186s (wild-type) and τ=3.9s (mutant)

ΔF₀

(FIG. 12C). Open and full circles relate to two different acquisition modes, which are described in the microscopy part.

FIG. 13. Mobility and association of single Lck-YFP molecules. T24 cells cotransfected with CD4 and Lck-YFP were plated on CD4 μ-biochips, and FRAP experiments were performed. At the onset of the recovery process (here after a recovery time of 2.4s), individual diffraction limited peaks were observed in the bleached area. To correlate the single molecule signals with the micropattern, FIG. 13A shows an overlay of the pre-bleach image and the recovery image. The brightness of one of the channels was reduced for better visibility. Individual peaks were tracked through subsequent images; a random selection of trajectories is shown in FIG. 13B. Individual trajectories were analyzed according to their effective diffusion constant D (FIG. 13C and FIG. 13D) and according to their brightness (FIG. 13E and FIG. 13F); data were split regarding to their localization with respect to the CD4 pattern: FIG. 13C and FIG. 13E show data observed within CD4 positive regions, FIG. 13D and FIG. 13F within CD4 negative regions. Lck-YFP diffusion is clearly heterogeneous, with a large fraction of free diffusion characterized by a diffusion constant D~lμm²/s. The tail towards D~0.01μm²/s indicates immobilization. Interestingly, the sites for immobilization do not correlate with the CD4 patterns, as both histograms yield significant data at low values of D. 28% (38%) of Lck trajectories were found to be characterized by D<0.1μm²/s in CD4 positive (negative) regions. Note that the decadic logarithm of D is plotted. The peak brightness shown in FIG. 13E and FIG. 13F is also broadly distributed. For comparison, the single molecule brightness observed for YFP coupled to CD 147 recorded using the same illumination parameter is shown. Taking a threshold of 300 counts for monomer brightness, ~55% (50%) of the observed signals represent dimers or larger associates.

DESCRIPTION OF ILLUSTRATIVE EMBODIMENTS OF THE PRESENT INVENTION

A. GENERAL DESCRIPTION OF EMBODIMENTS OF THE PRESENT INVENTION

The present invention provides methods and compositions for detecting, analyzing, monitoring, or measuring various intracellular processes and molecular interactions. In addition, the present invention provides methods and compositions for detecting or screening for interactions between two molecules (e.g., protein-protein interactions).

In certain embodiments, the present invention may be used to assess the interaction between two membrane bound molecules or between a membrane bound molecule and a cytosolic molecule. Any molecule, especially macromolecules, can be modified to comprise a membrane anchor to allow the localization of the molecule on a cellular membrane. Transmembrane proteins have this feature inherently and have protein domains on both sides of the membrane. Accessibility from both sides of the membrane facilitates antibody binding and interactions of the bait molecule in the cytoplasm. For example, the interaction between a transmembrane protein and a cytosolic protein can be evaluated. The transmembrane protein can be referred to as the "bait" and the cytosolic protein can be referred to as the "prey." Micropatterns of a ligand, such as an antibody, specific to an extracellular domain of the bait are generated on a solid surface. When cells are incubated on the patterned surface a redistribution of the bait in the plasma membrane of the cells occurs. The prey protein is labeled, typically as a fluorescent protein fusion. Thus, protein interactions between the bait and the prey can be assayed via the redistribution of the fluorescently labeled prey corresponding to the patterned surface. This allows for the detection of the interaction between the bait and prey. In addition, the kinetics of the interactions can be measured by, for example, Fluorescence Recovery After Photobleaching (FRAP). In addition, the effect of other molecules on the interaction between the bait and prey can be assessed by administering the molecule to the cell and observing the effect on the distribution of the fluorescently labeled prey.

Embodiments of the present invention may be used to study the interaction between two molecules known to interact or suspected to interact; however, embodiments of the present invention may also be used to identify previously unknown interactions between molecules. Such interactions can be screened for by, for example, providing a cell comprising a labeled target molecule, the labeled target molecule comprising an extracellular target moiety, a membrane moiety, and a label moiety. In one embodiment, the labeled target molecule is a fusion protein comprising a fusion of a known transmembrane protein with a target moiety of interest to be displayed on the surface of the cell and a fluorescent protein. When using fusion constructs with a fluorescent protein, specificity is ensured inherently. Fusion with AGT (Alkyl guanine transferase) enables labeling with fluorescent 06- benzylguanine derivatives. A multi-histidine tag allows for labeling with Ni-NTA modified fluorophores. The cell is incubated on a patterned surface comprising a pattern of interaction areas, the interaction areas comprising immobilized ligand molecules against which the target moiety is to be screened for interaction. It may be advantageous that the immobilized ligand molecules do not interact with any endogenous moieties on the cell surface. Protein interactions between the target moiety and the ligand results in a redistribution of the labeled target molecule in the plasma membrane, which can be detected via the redistribution of the fluorescent label corresponding to the patterned surface. In this manner a single target moiety can be screened against any number of ligand molecules and/or a single ligand molecule can be screened against any number of target moieties. In addition to using transmembrane proteins to display the target moiety on the cell surface, the target moiety may also be coupled to the cell surface via lipids or carbohydrates. In addition, the ligands may be used to identify new interaction partners for the labeled target moiety, e.g. a cytosolic protein.

B. MICROPATTERNED SURFACES

Embodiments of the present invention employ substrates having micropatterned surfaces comprising interaction areas onto which ligands, such as antibodies, are immobilized. For fluorescent microscopy applications, the substrate may be made from a light-transmitting material, such as glass, quartz, or transparent synthetic polymeric materials such as polyethylene, polypropylene and polyphenylene sulfide. The use of such substrates is known in the art for micro- and nano-array chips and can be adapted for the method of embodiments of the present invention. The immobilized ligand may be bound to the surface via a linking moiety, such as streptavidin, avidin, and/or biotin. The non-covalent bond between streptavidin (or avidin) and biotin is one of the strongest in biological systems. This can be used to bind one of the binding partners to the surface and the other to the ligand (e.g. as biotinylation). The ligand can then be brought into contact with the surface and is subsequently bound to the surface.

The pattern of interaction areas may be a regular pattern or an irregular pattern. When the pattern is a regular pattern, this may facilitate automated manufacturing processes and the automated readout of the signals. The pattern can be essentially any shape including, for example, squares, rectangles, bars, circles, triangles, etc. The size of the pattern may be smaller than the size of the cells being assayed but still large enough for the resolution of the particular label to discern the discrete areas. In certain embodiments of the invention, the size and arrangement of the interaction areas is designed to allow 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 200, 300, 400, or 5000, or any range therein, interaction areas to fit into an area the size of the bound cell. In certain embodiments of the invention, the surface comprises an inert, cell adhesive or blocking component, in particular in the space between the interaction areas. Such components may be blood proteins, or modified blood protein such as albumin, especially BSA (bovine serum albumin), or fibronectin. In some embodiments of the invention a second label, distinguishable from the label that may be used to label the target, is located between the interaction areas of the patterned surface. In one embodiment of the invention, a fluorescent molecule, such as Cy5, is located between the interaction areas of the patterned surface.

In one embodiment the patterned surface is generated by microcontact printing using a poly(dimethylsiloxane) stamp to transfer Cy5-labeled bovine serum albumin (BSA) to an aldehyde-derivatized glass slide in the desired pattern. The BSA micropatterned glass slides are then incubated with streptavidin, and finally incubated with biotinylated ligand. BSA efficiently blocks nonspecific adsorption of both streptavidin and the ligand, thereby providing a well-defined micropattern specifically reactive to one cell surface target.

In the following, referring to FIG. 1, a solid substrate 100 according to an exemplary embodiment of the invention will be explained in detail.

The solid substrate 100 is shown in FIG. 1 on the left hand side in a first operation mode and on the right hand side in a second operation mode. The solid substrate 100 comprises a substrate 101, in the present embodiment a glass coverslip. On the surface of the substrate 101, a pattern of interaction areas 102 (which may also be denoted as active areas), and intermediate portions (which may also be denoted as inactive or passive portions) are provided. In each of the interaction areas 102, a streptavidin structure 104 is formed, whereas spaces 103 between the interaction areas 102 are free of streptavidin. On the streptavidin structures 104, biotinylated capture antibodies 105 are immobilized which are specific to an extracellular domain of a transmembrane protein which will be denoted in the following as baits 106.

The spaces 103 between the interaction areas 102 comprise BSA (bovine serum albumin) as blocking molecules blocking the deposition of baits 106. Optionally, the blocking molecules 106 may also comprise a fluorophore, that is to say a fluorescent active label or molecule. The baits 106 are configured to capture correspondingly configured preys 108 which also comprise fluorescence labels 109.

Thus, FIG. 1 shows an assay 100, wherein BSA micropatterns 107 are printed on the functionalized glass coverslip 101 using a microstructure PDMS stem. The interspaces 102 are filled with the streptavidin 104 and incubated with the biotinylated antibody 105 against the membrane protein 106. Interaction with the second fluorescence labelled protein 108 are probed by measuring the degree of co- patterning. Specific protein-protein interactions are characterized by enrichment of the fluorescence labelled protein (left-hand side of FIG. 1). When no interaction occurs, the distribution will be essentially homogeneous (right-hand side of FIG.

1).

C. MICROSCOPY AND LABEL DETECTION

According to embodiments of the present invention, interactions between molecules may be assayed via the redistribution of a label corresponding to the patterned surface. The label may be visualized by microscopy. For example, where the label is a fluorescent label, fluorescence microscopy may be employed. Typical components of a fluorescence microscope are the light source, the excitation filter, the dichroic mirror or beamsplitter, and the emission filter. The particular filters and the dichroic are chosen to match the spectral excitation and emission characteristics of the fluorophore label. After appropriate filtering, fluorescence may be imaged onto a CCD camera or other image capture device. Where two fluorophores are used, scans may be performed sequentially in two colors, with one scan at a wavelength selective for one fluorophore and the second scan at a wavelength selective for the other fluorophore. For example, using YFP and Cy5, a first scan at 514 nm would be made for selective excitation of YFP, and a second scan at 647 nm would be made for selective excitation of Cy5.

One advantage of embodiments of the present invention is that living cells can be used. The methods provide a mild treatment of the cell, which allows for in vivo monitoring of processes and interactions of biological pathways. One approach for monitoring these in vivo processes and interactions is FRAP (Fluorescence Recovery After Photobleaching). FRAP is an imaging technique in which a fluorescent label is destroyed on purpose with an intense light. Photobleaching is performed in a well-defined area to study the repopulation of this area with fluorescent molecules from areas that were not photobleached. This method allows the measurement of kinetic features of molecules in living cells, in particular the measurement of the lifetime of the interaction, e.g. the off-rate, which can reveal information about protein interaction partners, organelle continuity, and/or protein trafficking. In certain embodiments, a fluorescent protein, such as GFP, is used during FRAP.

Examples of fluorophores that may be used in the context of embodiments of the present invention include, but are not limited to the following: fluorescent proteins (e.g., GFP, EGFP, YFP, CFP, RFP), Alexa Fluor® dyes, AMCA, BODIPY® 630/650, BODIPY® 650/665, BODIPY®-FL, BODIPY®-R6G, BODIPY®-TMR, BODIPY®-TRX, Cascade Blue®, CyDyes™, including but not limited to Cy2™, Cy3™, and Cy5™, DNA intercalating dyes, 6-FAM™, Fluorescein, HEX™, 6- JOE, Oregon Green® 488, Oregon Green® 500, Oregon Green® 514, Pacific Blue™, REG, phycobilliproteins including, but not limited to, phycoerythrin and allophycocyanin, Rhodamine Green™, Rhodamine Red™, ROX™, TAMRA™, TET™, Tetramethylrhodamine, Texas Red®, Atto dyes, and quantum dots.

The image data may be analyzed using, for example, an automatic gridding algorithm, which determines the rotation of the image with respect to the scan direction, and the grid-size. Based on the grid, images may be segmented into squares containing the interaction areas. A central circle within each square containing an interaction area may be used for analysis of the corresponding image. Each square may then be characterized by the mean fluorescence intensity within the circle, Fhig_h, and the remaining part of the square, Fi_ow. In addition, the background signal of the glass surface may determined on a part of the surface containing no cells (F_bg).

In the following, referring to FIG. 9, a measurement apparatus 900 according to an exemplary embodiment of the invention will be explained.

The measurement apparatus 900 comprises a carrier or mount 901 on which a solid substrate 100 such as the one described referring to FIG. 1 is mounted or assembled.

A surface of the mount 901 may be configured (e.g. shaped, dimensioned, grooved, oriented, and/or positioned) specifically for securely mounting the solid substrate 100 on top of the mount 901 , at a specific position as to interact in a spatially correct manner with the optical detection components which will be defined in the following in more detail.

Furthermore, the measurement device 900 comprises a first laser 902 and a second laser 903 each for irradiating a dedicated one of the interaction areas 102 with primary electromagnetic radiation. More particularly, the first laser 902 is adapted to generate a beam (for instance a continuous or a pulsed beam) of a relative low intensity onto the interaction area 102. This is indicated in a graph 904. In contrast to this, the second laser 903 is adapted to generate intense and very short pulses (microseconds or less to milliseconds or more) onto the interaction area 102, as indicated schematically in a diagram 905. More particularly, the second laser beam 903 is adapted to photobleach an active portion of the interaction area 102. The recovery of the signal may then be measured based on a sampling signal of the kind as shown schematically in a diagram 904, which does not photobleach the system.

The first laser 902 is adapted to generate a first primary beam 906 which impinges on the interaction area 102. A first secondary electromagnetic radiation beam 907 may then be directed onto a charge coupled device (CCD) 908 serving as an electromagnetic radiation detector for detecting the first secondary electromagnetic radiation beam 907 from the interaction area 102 in response to the irradiation with the first primary electromagnetic radiation beam 906. An optional frequency- selective filter element 915 (for instance a band pass filter) is capable of absorbing specific wavelength ranges of electromagnetic radiation to suppress background or scattered radiation.

In contrast to this, the second laser 903 generates a second primary beam 909 which also impinges on the interaction area 102. However, a second secondary beam 910 generated by irradiating the interaction area 102 with the second primary beam 909 may be directed to a wall 911 of an aperture element 912, but not on the CCD 908. Thus, with the array of FIG. 9, only the radiation originating from the first laser 902 can be detected by the CCD detector 908.

Summarizing, the second laser 903 functions to photobleach the sample in the irradiated interaction area 102. The first laser 902 functions to detect recovery of the signal after photobleaching the sample in the irradiated interaction area 102.

As an alternative to the described embodiment, it is also possible to provide a single common laser adapted for both bleaching and sensing. Such a laser may first bleach and afterwards detect recovery of the signal. The bleaching beam may be a pulsed beam. The detection beam may be a continuous beam or may also be a pulsed beam.

A control unit 913 is provided which controls cooperation of the various components, particularly operation of the first laser 902, the second laser 903, a motion mechanism for moving the carrier 901 (for instance a drive unit), and for receiving and evaluating data captured by the CCD 908. Furthermore, an input/output unit 914 is provided which may communicate with the CPU 913 to provide commands and/or to receive information. The input/output unit 914 may comprise an input unit such as a keypad, a button, a joystick, etc., via which a user may input information in the system 900. Furthermore, the input/output unit 914 may comprise an output unit, particularly a display unit such as an LCD display or a cathode ray tube for displaying information provided by the control unit 913.

With the measurement apparatus 900, a fluorescence recovery after photobleaching (FRAP) arrangement may be combined with a total internal reflection fluorescence (TIRF) arrangement.

TIRF microscopy may use the evanescent field generated by total reflection of the light beam 906 to excite fluorophores in a thin layer on the specimen surface, that is to say close to the surface of the interaction area 102. TIRF microscopy is based on a total reflection of the light beam 906 under certain optical and geometric conditions, for instance when specific angular properties are fulfilled. Thus, as a precisely defined evanescent field penetrates approximately 100 nm to 300 nm into the specimen, the TIRF method generates a highly accurate excitation within a thin layer above the coverslip 101 in the interaction area 102.

For applying the FRAP technology by the measurement apparatus 900, the second laser 903 generates the laser pulses 909 in order to destroy fluorescent molecules in the interaction area 102 by a very intensive laser pulse ("bleaching"). Subsequently, the fluorescence in the area 102 is detected in dependence of the time, with a light intensity of the continuous laser beam 906 of the first laser 902 which is small enough to prevent a further bleaching.

D. KITS

Any of the compositions described herein may be comprised in a kit. In a non- limiting example, a pre- fabricated solid substrate having a patterned surface may be comprised in a kit in suitable container means. The kit may also include one or more buffers, potential ligands, cells, and/or an expression vectors for recombinant proteins. The kit may also provide a nucleic acid encoding a fluorescent protein which can be fused to a target molecule. Other kits of embodiments of the invention may include components for making a solid surface having a patterned surface, and thus, may include, for example, a solid substrate, a PDMS stamp, BSA, biotin, avidin, streptavidin, a fluorophore, and/or an antibody.

The components of the kits may be packaged in aqueous media or in dried (e.g. lyophilized) form. The container means of the kits may generally include at least one vial, test tube, flask, bottle, syringe or other container means, into which a component may be placed, and possibly, suitably aliquoted. Where there is more than one component in the kit, the kit also will generally contain a second, third or other additional containers into which the additional components may be separately placed. However, various combinations of components may be comprised in a container. The kits of embodiments of the present invention also may include a means for containing the containers in close confinement for commercial sale. Such containers may include cardboard or injection or blow-molded plastic containers into which the desired containers are retained.

When the components of the kit are provided in one or more liquid solutions, the liquid solution may be an aqueous solution, with a sterile aqueous solution being particularly preferred. However, the components of the kit may be provided as dried powder(s). When reagents and/or components are provided as a dry powder, the powder can be reconstituted by the addition of a suitable solvent. It is envisioned that the solvent may also be provided in another container means.

A kit may also include instructions for employing the kit components as well the use of any other reagent not included in the kit. Instructions may include variations that can be implemented.

E. EXAMPLES

The following examples are included to demonstrate preferred embodiments of the invention. It should be appreciated by those of skill in the art that the techniques disclosed in the examples which follow represent techniques discovered by the inventors to function well in the practice of the invention, and thus can be considered to constitute exemplary modes for its practice. However, those of skill in the art should, in light of the present disclosure, appreciate that many changes can be made in the specific embodiments which are disclosed and still obtain a like or similar result without departing from the spirit and scope of the invention.

EXAMPLE 1

Materials and Methods

Reagents. The following buffers were used: phosphate buffered saline (PBS) containing 150 mM NaCl, 160 mM Na₂HPO₄, 40 mM NaH₂PO₄, pH adjusted to 7.4; PBSBT containing PBS, 0.1 % Tween 20, 0.2 % BSA.

The monoclonal antibodies to CD59, MEM-43/5, and to CD 147, MEM-6/4, were kindly provided by Vaclav Horejsi, Institute of Molecular Genetics, Prague, Czech Republic, and biotinylated by Standard protocols. Monoclonal biotinylated antibody to CD3 and CD4 was purchased from THP medical products, Vienna (CD3: UCHTI, #13-0038-80; CD4: #13-0049-80). A goat antimouse Cy3 (GE Health-care #PA43002) was used as secondary antibody at a concentration of 1 μg/ml in PBS. Fatty acid-free BSA (Sigma-Aldrich, Austria) was labeled at 2 mg/ml protein concentration with Cy5 monofunctional dye (GE Healthcare) by the procedure described by Gruber et al. (Bioconjug Chem 11 (2000): 161-166 & 696-704).

The following additional products were purchased from Sigma-Aldrich, Austria: Streptavidin (#S-0677); N,N,N';N'-tetrakis(2-pyridylmethyl)ethylenediamine (TPEN; #87641); 1 , 10-Phenanthroline (#131377).

Cell Culture and Transfection. ECV-304(T-24) cells (DSMZ # ACC 310) were maintained in monolayer cultures with RPMI 1640 medium (with L-glutamine, without phenolred; PAA-Laboratories, Linz, Austria) supplemented with 10% fetal calf serum (PAA-Laboratories), penicillin-streptomycin (100 units penicillin per ml, lOOμg streptomycin per ml, PAA-Laboratories) and 10 mM Hepes buffer (N-2- hydroxyethylpiperazine-N'2-ethane-sulfonic acid) and incubated at 37°C in a 5% CO₂ atmosphere. Confluent cells were harvested with trypsin/EDTA (PAA- Laboratories) and diluted to 30 mm petri-dishes at least 24 hours before transfection.

Human Embryonic Kidney cells (HEK-293; ATCC # CRL-1573) were maintained in monolayer cultures with DMEM High Glucose (PAA-Laboratories) supplemented with 10% fetal calf serum (PAA-Laboratories), 1% penicillin- streptomycin, and incubated at 37°C in a 5% CO₂ atmosphere. Confluent cells were harvested with trypsin/EDTA (PAA-Laboratories) and diluted to 30 mm petri- dishes at least 4 hours before transfection.

For transfection, unlabeled CD4 was cloned into the vector pEFIRES-P; all remaining constructs were cloned into the vector pBMN-Z (Gary Nolan Lab, Stanford). Cells were transfected with up to 4 μg DNA using the BioRad Transfectin Lipid Reagent #170-3350 according to the manufacturer's protocol. Cotransfections were performed using 2 μg DNA of each of the plasmids. Before the experiment, cells were removed from the Petri dish with a EDTA solution

(HOmM NaCl, 2mM KCl, 2.5mM Na₂HPO₄, ImM KH₂PO₄, 400 μM EDTA) and washed twice in medium. For microscopy, cells were placed in Hank's buffered salt solution (#H 15-008, PAA Laboratories, Austria). For chelation experiments, cells were incubated with 5mM TPEN or 1,10-Phenanthroline dissolved in HBSS containing 0.5% DMSO for 10 minutes at 37°C.

Fusion constructs. The GPI-GFP construct was constructed in the eukaryotic expression vector pJB20. It has an EcoRI site at the 5' end, a HindIII site at the 3' end and a Pstl site that separates the ecto and anchor domains. In addition it has a myc tag at the NH₂ terminus and contains an ER import signal. The GFP of this vector was mutated at amino acid position 206 from Alanine to Lysine to get a monomelic GFP variant using overlap extension PCR. The mutagenic forward primer 5'- CCTG AGTACCCAGTCCAAACTGAGCAAAGACCCCAACG-S' and the mutagenic reverse primer 5'- CGTTGGGGTCTTTGCTCAGTTTGGACTGGGTACTCAGG-S' were used in combination with the forward primer 5'-CGGTAGGCGTGTACGGTGGG-3' and the reverse primer 5'-GGCACTGGGGAGGGGTCACAGG-S' to generate the construct pJB20 GPI-mGFP). This plasmid was used as a template for PCR and amplified mGFP was fused to the different Lck-variants.

All Lck fusion constructs were prepared in the retroviral expression vector pBMN-Z using standard molecular biology methods. Monomelic GFP (mGFP) was PCR amplified from the pJB20 GPI-mGFP construct using the following oligos: mGFP- forward (5'-AAAGCATGCATGAGTAAAGGAGAAGA-S') and mGFP-reverse (5'-TGCGGCCGCTTATTATTTGTATAGTTCATCCAT-S'). Full-length Lck was PCR amplified from cDNA with the following oligos: Lck-forward (5'- AAAAGCTTATGGGCTGTGGCTGCA-3') and Lck-reverse (5'- TGCGGCCGCTTATTAGCATGCAGGCTGAGGCTGGTACTG-S'). Lck-C20/23 A was PCR amplified in two consecutive reactions using the following primers: Lck- C20/23A-forward-l (5'-GAAAACATCGATGTGGCCGAGAACGCCCATTAT-S') and Lck-reverse in the first reaction and Lck-C20/23A-forward-2 (5'- AAGCTTATGGGCTGTGGCTGCAGCTCACACCCGGAAGATGACTGGATGG AAAAC ATCGATGTGGCC-3') and Lck-reverse for 5 'elongation. Lck-ΔNIO was PCR amplified using the oligos Lck-ΔNlO-forward (5'- AAGCTTATGGACTGGATGGAAAACATCGATGTG-S ') and Lck-reverse. Lck- ΔN65 was PCR amplified using the primers Lck-ΔN65-forward (5'- AAGCTTATGGTT ATCGCTCTGCACAGCTATGAG-3') and Lck-reverse. For PCR amplification of Lck-ΔN249 Lck-ΔN249-forward (5'- AAGCTTATGAAGCTGGTGGAGCGGCTG-3') and Lck-reverse was used. To create mGFP-tagged constructs, Lck fragments and mGFP were cloned to the linearized pBMN-Z. For the minimal constructs Lck-N65-mGFP and Lck-N10- mGFP, mGFP was PCR amplified using mGFP-forward-1 (5'- AAACTCGAGATGGTGAGCAAGGGCGAG-3') and mGFP-reverse. Lck-N65 was PCR amplified using the following oligos: Lck-forward and Lck-N65 -reverse (5'- AGCGATAACCTCGAGCAGGTTGTCTTGCAGTGG-S'). The Lck-NIO fragment was built up by annealing Lck-NIO-forward (5'-

AGCTTATGGGCTGTGGCTGCAGCTCACACCCGGAAGATC-3') and Lck-N10- reverse (5'-TCGAGATCTTCCGGGTGTGAGCTGCAGCCACAGCCCATA-S'). The minimal constructs were constructed by ligating the Lck fragments and the mGFP to the linearized pBMN-Z.

Microcontact Printing. Aldehyde-derivatized glass coverslips were used. Poly(dimethylsiloxane) (PDMS) was generated from basic elastomer (GE Bayer Silicons, #RTV 615A) mixed with starter (GE Bayer Silicons, #RTV 615B) in a 10:1 ratio, and applied to a silicon master for 30 minutes at 80°C. The silicon master containing an array of squares with a feature size and a depth of 3 μm was generated by standard photolithography using a custom designed beam mask (Photronics MZD, Dresden, Germany). The PDMS stamp was peeled off the mask and stored at room temperature. Before the experiment, stamps were rinsed with 100 % ethanol and dd H₂O, dried with N₂, and incubated with 100 mg/ml Cy5- labeled bovine serum albumin (BSA) for 30 minutes at room temperature. Upon inking, stamps were washed extensively with PBS and dd H₂O, and dried with N₂. For Cy5-BSA transfer stamps were placed under their own weight onto the glass coverslips for 30 minutes. Upon removing the stamps, coverslips were sealed with adhesive silicon masks (Secure Seal, Schleicher & Schuell, Austria).

BSA micropatterned glass coverslips were incubated for 1 hour at room temperature with 50 μg/ml streptavidin dissolved in PBS, rinsed with PBS, and finally incubated with 10 μg/ml biotinylated monoclonal capture antibody in PBSBT. BSA efficiently blocks unspecific adsorption of both streptavidin and the antibody, thereby providing a well-defined 3 μm micropattern specifically reactive to one cell surface protein (FIGs. 1 A-IC).

Microscopy. The detection system was set up on an epifluorescence microscope (Axiovert 200m, Zeiss, Germany). Ar⁺ and Kr⁺-ion lasers (Innova, Coherent,

USA) were used for selective fluorescence excitation of YFP at 514 nm and Cy5 at 647 nm, respectively. Samples were illuminated in objective- type total internal reflection (TIR) configuration using a 10Ox oil immersion objective (NA=I.45, α- Fluar, Zeiss, Germany). After appropriate filtering using standard filter sets (Chroma Technology Corp., VT), fluorescence was imaged onto a back-illuminated CCD camera (SPEC 10:100B, Princeton Instruments, NJ). Readout was performed in time delay and integration (TDI) mode. For this, samples were shifted using a motorized xy-stage (Scan IM 120x100, Marzhauser, Germany) synchronized to the line-shift of the camera. The reader was equipped with an automated focus hold system operating during the scanning process: the back-reflected laser beam is imaged on a 2-segment photodiode, and the differential signal is used to control a z-piezo (PIFOC, Physik Instrumente, Germany) for fast refocusing. All scans were performed sequentially in two colors, with one scan at 514 run for selective excitation of YFP, the second scan at 647 nm for selective excitation of Cy5. Filters and dichroics were changed between the scans. All scans were recorded at room temperature.

FRAP (Fluorescence Recovery After Photobleaching) experiments were performed at 37°C (Moertelmaier, et al., Appl Phys Lett 87 (2005):263903). A Zeiss Axiovert 200 microscope was equipped with a 10Ox NA=I .45 Plan-Apo objective

(Olympus). Samples were illuminated in objective-type TIR configuration via the epiport using 514 nm light from an Ar+ Laser (Model 2020, Spectra Physics). A slit diaphragm with a width of 4μm in the object plane was used as field stop to confine the illumination area. Images were recorded using a back-illuminated liquid nitrogen cooled CCD camera (Micro Max 1300-PB, Roper Scientific). A bleaching time of 350 milliseconds (ms) was sufficient to completely destroy all fluorophores in the illuminated area. The recovery images were recorded at an illumination time of lms. To avoid additional photobleaching during the measurement of the recovery process, each point in the FRAP curve was measured separately. For this, pre-bleach images (characterized by the fluorescence signal FO) and recovery images (characterized by F(t)) were recorded, and F(t)/FO was plotted as a function oft. For a given time-lag t, 2-4 images were recorded and the average value of F(t) was used for the analysis.

Data Analysis. The BSA-Cy5 image was used for an automatic gridding algorithm, which determined the rotation of the image with respect to the scan direction, and the grid-size. Based on the grid, images were segmented into adjacent squares containing the BSA-Cy5-negative (and concomitantly capture antibody-positive) regions. A central circle with a diameter of 4 μm was used for analysis of the corresponding FP- image. Each square was characterized by the mean fluorescence intensity within the circle, F⁺, and the remaining part of the square, F^". hi addition, the background signal of the glass surface was determined on a part of the chip containing no cells (F_bg). For analysis of multiple cells, two-dimensional histograms were prepared for the

fluorescence signal F - F⁺ - F₁, against the signal contrast C - . To

F⁺ - F bug restrict the analysis to FP-expressing cells, only squares with a signal exceeding a preset threshold value were used for analysis.

FRAP data were analyzed by calculating AF = F⁺ - F^~ , normalized by the very first pre-bleach image recorded on one feature, ΔF₀. For single molecule analysis, images were analyzed using in-house algorithms implemented in MATLAB (Math Works). Individual diffraction limited signals were selected and fitted with a Gaussian profile, yielding the single molecule position r (t) and brightness. Single molecule

trajectories were reconstructed, and the effective diffusion constant D = Η ljag )r

At lag was calculated, with δr{t_lag )= r(t + t_lag ) - r(t) and t_lag = {t_lU + t_delay )i , n a number smaller than the maximum trajectory length.

EXAMPLE 2

Results

The Assay /Micropatterning of the Bait CD4 in Living T24 Cells. Micropatterns of membrane proteins in their native environment provide a novel tool to characterize protein-protein interactions. FIG. 1 shows the experimental strategy: a capture antibody 105 to the exoplasmic domain of a membrane protein 106 is assembled in a micropattern on the bio-chip surface, which leads to the redistribution of the membrane protein 106 (here referred to as "bait"). An interacting molecule 108 ("prey") will follow this redistribution; the resulting micropattern can be visualized by fluorescence labeling 109 of the prey 108. Excitation is performed in total internal reflection (TIR) configuration to discriminate plasma membrane proteins from residual intracellular fluorescence due to cytosolic proteins or auto fluorescence. Micropatterns of capture antibody 105 on glass coverslips 101 is shown in FIGS. 6A-6D. Micropatterns were prepared as described and incubated with Alexa555-labeled secondary antibody 105. The image shows the coverslip 101 scanned sequentially at 647 nm for excitation of BSA-Cy5 (FIG. 6A) and at 514 nm for excitation of Alexa555 (FIG. 6B). The over-lay (FIG. 6C) demonstrates the efficient separation of the two reagents, yielding a high-contrast micropattern. FIG. 6D illustrates gridding, and shows a zoomed region.

Micropatterning of a transmembrane protein in the plasma membrane of T24 cells. To demonstrate the redistribution of the bait, T24 human bladder carcinoma cells transfected with CD4-YFP were plated on coverslips micropatterned with CD4 capture antibody and grown for ~12 hours under standard cell culture conditions. During this time, cells adhered strongly enough to endure the subsequent washing steps. FIGs. 2A and 2B show the CD4 pattern on a T24 cell: strong CD4 enrichment and depletion correlated well with Cy5-BSA negative and positive regions on the coverslip, respectively. Similar results were obtained using CD147-YFP transfected cells and a CD 147 capture antibody (FIG. 2C).

To ensure that cells do not non-specifically react to the microstructured surface, CD4-YFP transfected cells were plated on μ-biochips functionalized with capture antibody to a protein not expressed in T24 cells (CD3): as expected, no concomitant CD4 redistribution was observed (FIG. 2D). Similarly, no CD4-YFP re-distribution was observed when cells were plated on u-biochips against CD 147, a membrane protein endogenously expressed in T24 cells (FIG. 2E).

Copatterning oflck-YFP with CD4. Stable CD4-lck association is believed to be the basis for lck recruitment to the immunological synapse, the critical site for initiation of T cell signaling. However, for CD4-lck interaction only indirect evidence via immunoprecipitation studies is available for CD4-Lck interaction. In particular, due to purification steps the contribution of weak binding sites with rapid turnover has not been addressable by conventional methods. Here, the interaction between CD4 and Lck was measured directly in living cells, by using unlabeled CD4 as bait and YFP-tagged lck as prey. Cotransfected cells showed clear enrichment of lck- YFP on the CD4 micropatterns (FIG. 3A). No lck- YFP pattern was observed when cells were transfected with lck- YFP only (FIG. 3B). Next, it was tested whether or not the fluorescent tag mediates the interaction: cells cotransfected with the bait CD4 and the prey YFP did not show any YFP redistribution (FIG. 3C). As a second negative control, endogenous CD147 was used as bait, for which no interaction with lck has been reported in the literature: no redistribution of the prey lck- YFP was observed (FIG. 3D). As further control, the so-called "zinc clasp" structure formed by a dicystein motif was addressed on each protein (Cys 420 and 422 on CD4 with Cys 20 and 23 on Lck). This complex can be disrupted by using zinc chelators (e.g. TPEN or 1,10-phenanthroline), as demonstrated in immunoprecipitation studies. Thus, it was tested whether disruption of the CD4-Lck interaction can be directly visualized with the assay.

Disruption of Zinc Chelation indicates novel Interaction Sites between CD4 and Lck. After addition of zinc chelators Lck- YFP did not preserve the clear micropatterns, indicating a significant decrease in CD4-Lck interaction affinity (FIG. 4). However, structures remained in particular in the periphery of the cells, which is in contrast to the negative controls showing no micropatterns over the whole cell surface (FIG. 3B to FIG. 3D). To exclude that chelation influences the bait micropatternper se cells transfected with CD4-YFP were plated on patterned CD4 capture antibody and were incubated with the chelator: no change in the micropattern was observed (FIG. 8A to FIG. 8D). Thus, micropatterns remaining after chelation seemed to be specifically caused through interaction of CD4 and Lck. To discriminate the different patterns quantitatively, an algorithm was developed to semi-automatically analyze the micropattems of a cohort of cells via their fluorescence brightness F and contrast C. Analysis of 20 cells by this method revealed a patterning of high contrast and variable brightness before addition of the chelator. Strikingly, zinc chelation revealed a new population of low contrast at high brightness in addition to the high contrast fraction that remained (FIG. 4B). Thus, the chelator treatment was not sufficient to fully disrupt the CD4-Lck interaction. This result confirms the essential role of zinc ions and the cysteine residues for the CD4- Lck interaction, but further indicates the existence of additional interaction sites.

Influencing the CD4-Lck interaction. It is believed that the interaction between CD4 and Lck is predominantly mediated by two cysteins on each protein (Cys 420 and 422 on CD4 with Cys 20 and 23 on lck) forming a complex including a zinc ion. This complex can be disrupted using zinc chelators (e.g. TPEN or 1,10- phenanthroline), as demonstrated in immunoprecipitation studies. It was therefore tested whether disruption of the CD4-lck interaction can be directly visualized with the present assay. Repeated imaging of cells cotransfected with CD4 as bait and lck- YFP as prey before and after addition of zinc chelators revealed clearly distinct pictures: while lck- YFP closely followed the bait before chelation (FIG. 4A), its pattern dissolved upon chelator treatment, resulting in a homogenous fluorescence distribution (FIG. 4B). Concomitantly, statistical analysis of 20 cells revealed a new population of high fluorescence intensity F and low contrast C. hi a control experiment, where cells were incubated with mock solution, no change in the prey pattern was observed (FIGs. 7A-7D). For testing whether chelation influences the bait micropattern, cells transfected with CD4-YFP were plated on patterned CD4 capture antibody, and incubated with the chelator: no change in the micropattern was observed (FIGs. 8A-8D). Both TPEN and 1,10-phenanthroline yielded similar results.

Contribution of the individual Lck domains to CD4 interaction. To analyze the contributions of Lck-C20 and C23 on the interaction with CD4, both cysteines were mutated to alanines (illustrated in FIG. 10) that was shown to abolish or diminish CD4-Lck interaction in co-immunoprecipitation studies. Cells cotransfected with Lck-C20/23A-YFP and CD4 were plated on CD4 μ-biochips. Compared to wild-type Lck-YFP, the induced micropatterns on CD4 μ-biochips showed dramatically reduced contrast in the central part of the cell (FIG. 1 IA and FIG. 1 IB); high contrast regions remained in the periphery. Concomitantly, statistical analysis of 16 cells revealed the emergence of a new population at high intensity and zero contrast, which was similar to the chelator experiment (FIG. 4). Though, the detection of high contrast regions in the cysteine double mutants ultimately evidenced additional mechanisms for CD4-Lck interaction.

To allocate these binding sites, a set of Lck truncation mutants was generated (FIG. 10, FIG. 11). The role of the membrane-proximal part of Lck was addressed. FIG. 11C shows the behavior of the membrane anchor only (Lck-NIO-mGFP). The analysis revealed a dramatic broadening in contrast, which covers the full region between zero and one. The predominant subpopulation shifted to zero contrast, consistent with the presumption that essential binding sites were removed. However, a significant fraction at high contrast indicated that the membrane anchor itself was sufficient for partial targeting of Lck to CD4. Since direct protein interaction can be excluded for this mutant, co-recruitment of CD4 and Lck to lipid rafts provides the most plausible explanation for the observation. To further corroborate this interpretation, the interaction of CD4 with GPI-anchored mGFP was measured that is considered as raft-marker. Indeed, weak but significant co-patterning in peripheral regions of the cells were found (FIG. 1 ID). The attenuated interaction compared to Lck-NIO-mGFP was expected, since additional transbilayer coupling is required to target GPI-mGFP to CD4. Finally, different interaction characteristics were found at different locations in the plasma membrane, indicating an influence of the membrane organization for the interaction process. Next, the SH4 domain (Lck-N65-mGFP) was studied, which includes the acylation sites for membrane anchoring and the two cysteines 20 and 23. The data is similar to Lck wild type (FIG. 11 E) indicating that the main interaction sites with CD4 are contained in this part of the molecule. Detailed analysis of the data, however, revealed a broadening of the high contrast population towards lower contrast values, suggesting that also the missing part containing the SH2, SH3 and SHl (kinase) domains is involved in CD4 interaction.

To analyze the contribution of these Lck domains for CD4 interaction, the deletion mutants Lck-ΔNlO-mGFP (FIG. HF), Lck-ΔN65-mGFP (FIG. 1 IG) and Lck- ΔN249-mGFP were generated (FIG. 1 IH). In general, mutants lacking the membrane-anchor showed a lower contrast. However, significant brightness/intensity values indicated that also membrane-distal Lck-domains are able to interact directly or indirectly with CD4. Even for Lck-ΔN249-mGFP - which represents in essence the kinase domain only - weak interaction was found.

Time-course ofCD4-Lck interaction. Having established a method for comprehensive characterization of CD4-Lck equilibrium interaction, next the binding kinetics was analyzed. The micropatterning technique was combined with fluorescence recovery after photobleaching (FRAP) to study the interaction kinetics in a live cell context. In a typical FRAP experiment, the fluorescent ligand to an immobile receptor is photobleached by a strong laser pulse, and the fluorescence recovery due to exchange of bleached with unbleached ligand is recorded, hi general, various terms contribute to the observed recovery kinetics including the kinetic on- and off-rate, the mobility of the ligand, and the spatial distribution of binding sites; discrimination of the individual contributions is therefore a non-trivial.

However, in some situations approximations allow for a straightforward data interpretation. For example, if the interaction lifetime between a freely diffusing membrane constituent and an immobilized receptor is long compared to the diffusion time, the fluorescence recovery becomes a simple exponential function

1 - exp(- 11 τ) , with τ^"1 the reverse binding rate constant. For investigation of interactions between a surface-bound receptor and its soluble fluorescent ligand TIR- excitation can be employed, yielding the same mono-exponential functionality for the recovery curve. Yet, a difficulty remains whenever unspecific adsorption of the ligand to the membrane precedes specific binding. For discriminating the kinetics of specific from unspecific binding events, the presented micropatterned surfaces provide an advantageous addendum to FRAP, as they enable the relative recovery measure of ΔF = F⁺ - F^~ . Apparently, AF is independent of homogenous unspecific signal contributions.

First, the immobility of the bait CD4 within the micropattern was tested. Cells transfected with CD4-YFP were plated on CD4 μ-biochips. For the FRAP experiment, one spot of the micropattern with high contrast between F⁺ and F^" was selected. After photobleaching, no recovery of AF was observable within 4min. The high stability was expected, since the applied excess of CD4 antibody on the μ- biochip results in immediate rebinding, whenever a CD4 molecule may be released.

To study the time-course of interaction between CD4 and Lck-YFP, cells cotransfected with CD4 and Lck-YFP on CD4 micropatterns were plated. Areas of 20x5 μm² were photobleached, and the subsequent recovery of the respective areas was monitored (FIG. 12). No contrast was observed on images recorded immediately after photobleaching. On a minutes time-scale, an increase in the fluorescence signal could be observed; interestingly, in the observation time of up to 7 minutes, only few cells yielded recovery to ~100%, indicating Lck fractions which are even stronger associated.

At the onset of the recovery process (i.e. <ls after photobleaching), the first individual Lck-molecules could be followed entering the photobleached region (FIG. 13). The majority of these molecules was found to be mobile (-66%), with an average diffusion constant D-lμm /s. For further analysis, single molecule trajectories were discriminated according to their localization with respect to the CD4 micropatterns.

Occasional immobilization events did not correlate with CD4 positive regions, indicating additional CD4-independent interactions of Lck. Data obtained on CD4 positive and CD4 negative regions were compared, yielding similar statistics for the diffusion constant (compare FIG. 13C and FIG. 13D; p=0.11 in a Kolmogorov- Smirnov test). The observed free mobility is an apparent consequence of the stable interaction between CD4 and photobleached Lck, yielding hardly any vacant CD4 for the recovering fluorescent Lck-molecules. Importantly, the high mobility allows for straightforward interpretation of the recovery time as reverse rate-constant, since w² the diffusion time τ_D = — « 2s (w=l .5μm the size of the bleach spot) was found to

be negligible compared to the measured recovery time of hundreds of seconds.

Also the single molecule brightness was analyzed. Histograms obtained on CD4 positive and negative sites yielded similar distributions (FIG. 13E and FIG.13F; p=0.31), which show clear broadening towards large brightness values compared to the distribution of single YFP-molecules (p«10^"5). Apparently, Lck moves as small associates - preferentially monomers and dimers - in the plasma membrane.

For comparison, the interaction kinetics between CD4 and Lck-ΔNlO-mGFP was also measured. The data yield a completely different picture (FIG. 12). In regions with homogenous fluorescence signal (in particular at the cell center), photobleaching of the area was not possible due to immediate recovery of Lck-

ΔNlO-mGFP from the cytosol. hi regions with clearly visible contrast fundamental differences were found to the results on wild-type Lck: first, the fluorescence signal recovered much faster (seconds time-scale); second, on many cells the contrast could not be removed to zero, indicating an additional fraction with rapid turnover on time- scales much faster than hundred milliseconds.

For quantitative analysis, all data sets were fitted by the function

— — = (a - β) ■ [l - exp(- 11 r)] + β , which accounts for incomplete recovery (α) and

incomplete photobleaching (β). Analysis of 10 (12) different cells for wild type Lck (ΔN10) yield an average interaction lifetime of τ= 159s (τ=4.6s). The two orders of magnitude difference in lifetimes appears remarkable, since both Lck constructs contain the same protein backbone including all CD4 interaction sites. Apparently, Lck membrane-anchorage and raft targeting is of crucial relevance for stabilizing CD4-Lck interaction.

Discussion of the Results. Stable association between CD4 and Lck is currently regarded as central for early T cell signaling: the molecular link is suspected to enable recruitment of Lck to the TCR in an antigen-specific way for subsequent ITAM phosphorylation. To further characterize and quantify this interaction in the living cell, exemplary embodiments of the invention provide an easy-to-implement methodology using micropatterned surfaces. It is based on the arrangement of a membrane protein ("bait") in micropatterns; interaction with a second fluorescently labeled protein ("prey") is inferred from its assembly in the same micropatterns. The methodology allows also weak and less pertinent associations to be easily investigated. In addition, since the study is performed directly in the living cell, any impact of the cellular environment - e.g. the plasma membrane structure - on the interaction strength is preserved.

Rigorous controls have been performed to demonstrate the specificity of the assay. The bait (tested for CD4 and CD 147) was found to be reliably arranged in micropatterns perfectly complementary to the BSA-Cy5 patterns (FIG. 2C). The structures were formed at high contrast; no significant turnover was recorded in tens of minutes. There was no effect of the BSA patterns on the lateral distribution of the bait, as tested by plating CD4-YFP transfected cells on CD 147 or CD3 μ-biochips (FIG. 2D and FIG. 2E); in these cases, no CD4 micropattems were observable.

In order to record selectively membrane-associated prey, TIR fluorescence microscopy can be employed for imaging. Due to excitation of only a small volume next to the biochip surface, signal arising from cytosolic or organelle-anchored prey as well as cellular autofluorescence can thereby be efficiently suppressed. As a consequence, however, distance variations of the cellular plasma membrane will give rise to concomitant variations in the brightness. It is therefore critical for the assay that the plasma membrane lies flat on the surface. This was confirmed in multiple experiments. First, cells cotransfected with CD4 and YFP were plated on CD4 μ- biochips; YFP did not show brightness variations anti correlated with the BSA-Cy5 pattern (FIG. 3C). Second, cells cotransfected with CD4 and the fluorescent Lck- mutants NlO, ΔN10, ΔN65, ΔN249 or C20/23A were plated on CD4 μ-biochips, yielding large areas with homogenous fluorescence (FIG. 11). Third, single molecule signals were measured in cells cotransfected with Lck-YFP and CD4 plated on CD4 μ-biochips; single Lck-YFP molecules could be traced in the cellular plasma membrane both at CD4-positive and -negative regions, yielding the same focal plane and fluorescence brightness (FIG. 13).

Lck was found to interact strongly with CD4, the interaction was zinc-dependent, and can be significantly attenuated by zinc chelation (FIG. 4) or point-mutation of the two involved cysteines on Lck (FIG. HB). However, Lck contains 4 characteristic domains, which a priori may contribute to its interaction with CD4. To allocate the essential binding sites, a set of Lck truncation mutants was studied (see FIG. 10). The mutant Lck-N65-mGFP showed micropattems with only slightly reduced contrast compared to Lck wild-type, indicating that the membrane-anchor in conjunction with the SH4 domain is sufficient for CD4 interaction. In contrast, additional truncation of the SH4 domain (Lck-NIO-mGFP) dramatically reduced the interaction; similar results were obtained on the mutant Lck-C20/23A-YFP.

Interestingly, both Lck-NIO-mGFP and Lck-C20/23 A-YFP still show significant interaction, in particular in the periphery of the cell. Apparently, in the first case this interaction can only be mediated by the membrane anchor itself. It is known that Lck and CD4 are recruited to lipid rafts, based on extraction studies using mild detergents. Here, in a live cell context, the interaction of CD4 with Lck-NIO-mGFP were found, which can be regarded as a marker of rafts in the cytosolic leaflet of the plasma membrane, and with GPI-mGFP, a marker of lipid rafts in the exoplasmic leaflet. This interaction was restricted to the periphery of the cells, consistent with recent reports indicating lamellipodia-targeting of lipid rafts. Since the degree of raft association was reported to be regulated by reversible palmitoylation of both CD4 and Lck, the observed spatial modulation of raft-association may be caused by heterogeneity in the degree of palmitoylation. Variations in the lipid composition within the plasma membrane may further add to this effect.

Membrane anchorage was found to be dispensable for targeting Lck to CD4, as all observed deletion mutants Lck-ΔNlO-mGFP, Lck-ΔN65-mGFP and Lck-ΔN249- mGFP yielded significant co-association with CD4. This finding appears plausible for Lck-ΔNlO-mGFP, which still includes the SH4 domain and therefore might bind to CD4 by forming the zinc clasp structure; however, this type of binding is ruled out for Lck-ΔN65-mGFP and Lck-ΔN249-mGFP. Surprisingly, the CD4 affinities to Lck-ΔN65-mGFP and to Lck-ΔNlO-mGFP were hardly distinguishable. This is remarkable in the context of the observed dramatically attenuated interaction of CD4 with Lck-NIO-mGFP compared to Lck-N65-mGFP. Apparently, Lck-binding to CD4 via its SH4 domain is context-specific and only efficient for membrane-anchored Lck. It is believed to be likely that Lck membrane-anchorage restricts the degrees of freedom for Lck-orientation relative to CD4, thereby facilitating their interaction. Similar to the membrane-anchored mutants, also the Lck-constructs lacking the membrane-anchor showed co-association with CD4 only in the peripheral regions of the cells.

Application of FRAP revealed more detailed insights into the interaction behavior of Lck with CD4. While in equilibrium, Lck-ΔNIO micropatterns observed in peripheral regions were similar to those observed for wild-type Lck, the measured interaction lifetime was dramatically different. First, a significant fraction of up to 20% recovered instantaneously (below the time-resolution of the FRAP-system of -100ms). Analysis was performed on AF, which is sensitive to contrast. Therefore, the observed instantaneous recovery cannot be attributed to unspecific adsorption of cytosolic Lck to the plasma membrane, but clearly represents a specific interaction with CD4. Second, the characteristic time-constant of the recovery shifted from 159s (wild-type Lck) to much faster kinetics of 4.6s (Lck-ΔNIO). Together, the data reveal a dramatically prolonged CD4 interaction of membrane- anchored Lck compared to its soluble counterpart. This finding indicates that Lck membrane-targeting enhances its affinity to CD4 not (only) by facilitating the complex formation with CD4, but in particular by reducing the dissociation rate of the complex. The presence of recovery kinetics on various different time-scales confirms the contribution of distinct binding sites for the total interaction process.

At the onset of the recovery process, single Lck- YFP molecules were observed to move into the photobleached region. The major fraction was found to diffuse freely with a diffusion constant D~lμm²/s. A significant fraction of immobile Lck (34% on average) was found; the sites of immobilizations did not correlate with CD4 micropatterns. Frequent observations of single molecules stopping or being released indicate that this interaction is transient on a subseconds time-scale. The moving entities observed were identified as Lck-monomers and oligomers.

Thus, by live cell experiments multiple CD4 interaction sites on Lck were found. This adds new aspects to the currently favored hypothesis of quasi irreversible interaction between Lck and CD4, which is presumed for correct signal initiation in T cells. Lck-mediated ITAM phosphorylation represents the first discernable and decisive phenotype that characterizes antigen recognition by a T cell. Its tight regulation is therefore crucial for correct signal transduction and processing.

Generalizing the approach. It was tested whether different human cell lines are equally suited for the novel method. For this, HEK cells were cotransfected with unlabeled CD4 and lck- YFP. Similar to T24 cells, clear micropatterns of lck- YFP were observed (FIG. 5).

Conclusion. In summary, a new method was developed to study interactions between membrane proteins directly in living cells. The interaction between CD4 and YFP-labeled lck was recorded as specific enrichment of lck- YFP on CD4 positive features, which were imposed onto the cellular plasma membrane by micropatterned surfaces. Lck- YFP micropatterns were found to dissolve upon Zn⁺- chelation, as expected for lck-CD4 interaction. Furthermore μ-biochips were utilized to study the interaction lifetime between lck and CD4. Interestingly, rapid turnover accompanied the known quasi irreversible interaction. Release of lck from the plasma membrane has so far not been analyzed and might represent an additional mechanism for downstream signal propagation.

* * * * * * * * * * * * * * *

All of the compositions and methods disclosed and claimed herein can be made and executed without undue experimentation in light of the present disclosure. While the compositions and methods of this invention have been described in terms of preferred embodiments, it will be apparent to those of skill in the art that variations may be applied to the compositions and methods and in the steps or in the sequence of steps of the methods described herein without departing from the concept, spirit and scope of the invention. More specifically, it will be apparent that certain agents which are both chemically and physiologically related may be substituted for the agents described herein while the same or similar results would be achieved. All such similar substitutes and modifications apparent to those skilled in the art are deemed to be within the spirit, scope and concept of the invention as defined by the appended claims.

REFERENCES

The following references, to the extent that they provide exemplary procedural or other details supplementary to those set forth herein, are specifically incorporated herein by reference.

US 6,559,474

US 2003/0143634 US 2002/0160505 WO 01/88182 Binz et al., Nat. Biotechnol. 22:575 (2004) Gruber et al., Bioconjug Chem 1 1 : 161-166 & 696-704 (2000) Moertelmaier et al., Appl Phys Lett 87:263903 (2005) Wu et al., PNAS USA 101 : 13798-803 (2004)

Claims

C L A I M S

1. A method for assessing a molecular interaction comprising:

(a) providing a cell comprising a labeled target molecule and a transmembrane moiety;

(b) providing a patterned surface comprising a pattern of interaction areas, the interaction areas comprising immobilized ligands specific to an extracellular domain of the transmembrane moiety;

(c) contacting the cell with the patterned surface; and (d) detecting a signal from the labeled target molecule, wherein a greater signal over the interaction areas of the patterned surface as compared to areas between the interaction areas indicates a molecular interaction between the labeled target molecule and the transmembrane moiety.

2. The method of claim 1, wherein the transmembrane moiety is a transmembrane protein.

3. The method of claim 1 or 2, wherein the labeled target molecule is labeled with a fluorescent protein.

4. The method of any one of claims 1 to 3, wherein the fluorescent protein is a green fluorescent protein, yellow fluorescent protein, or a red fluorescent protein.

5. The method of any one of claims 1 to 4, wherein the labeled target molecule is a protein, nucleic acid, carbohydrate, or lipid.

6. The method of any one of claims 1 to 5, wherein the labeled target molecule is a cytosolic protein.

7. The method of any one of claims 1 to 6, wherein the labeled target molecule is a membrane bound protein.

8. The method of any one of claims 1 to 7, wherein the ligands are immobilized on the patterned surface by a linking moiety.

9. The method of claim 8, wherein the linking moiety is a streptavidin and/or a biotin.

10. The method of any one of claims 1 to 9, wherein the interaction areas are about 3 μm by 3 μm.

11. The method of any one of claims 1 to 10, wherein the distance between interaction areas is about 3 μm.

12. The method of any one of claims 1 to 11, further comprising photobleaching the labeled target molecule and observing the recovery from the photobleaching.

13. The method of any one of claims 1 to 12, further comprising contacting the cell with a ligand and observing a ligand-mediated change in interaction between the target molecule and the transmembrane protein.

14. The method of any one of claims 1 to 13, further comprising: adding a potentially interfering or regulating agent; detecting a signal from the labeled target molecule in the presence of the potentially interfering or regulating agent; and comparing the signal from the labeled target molecule in the presence of the potentially interfering or regulating agent with the signal from the labeled target molecule in the absence of the potentially interfering or regulating agent.

15. The method of claim 14, wherein the agent is a competitive binder of the transmembrane moiety or the labeled target molecule.

16. The method of claim 14 or 15, wherein the agent modifies the binding strength of the transmembrane moiety to the labeled target molecule.

17. A method for detecting a molecular interaction comprising:

(a) providing a cell comprising a labeled target molecule, the labeled target molecule comprising an extracellular target moiety, a membrane moiety, and a label moiety;

(b) providing a patterned surface comprising a pattern of interaction areas, the interaction areas comprising immobilized ligand molecules;

(c) contacting the cell with the patterned surface; and

(d) detecting a signal from the labeled target molecule, wherein a greater signal over the interaction areas of the patterned surface as compared to areas between the interaction areas indicates a molecular interaction between the target moiety and the ligand molecules.

18. The method of claim 17, wherein the label moiety is a fluorescent protein.

19. The method of claim 18, wherein the fluorescent protein is a green fluorescent protein, yellow fluorescent protein, or a red fluorescent protein.

20. The method of any one of claims 17 to 19, wherein the extracellular target moiety is a protein, nucleic acid, carbohydrate, or lipid.

21. The method of any one of claims 17 to 20, wherein the membrane moiety is a protein or lipid.

22. The method of any one of claims 17 to 21, wherein the extracellular target moiety and the membrane moiety are proteins.

23. The method of claim 22, wherein the extracellular target moiety and the membrane moiety are an extracellular domain and a membrane domain of the same protein.

24. The method of claim 22, wherein the extracellular target moiety and the membrane moiety are from different proteins.

25. The method of any one of claims 17 to 24, wherein the label moiety is coupled to the extracellular target moiety.

26. The method of any one of claims 17 to 25, wherein the label moiety is coupled to the membrane moiety.

27. The method of any one of claims 17 to 26, wherein the immobilized ligand molecules are proteins, nucleic acids, carbohydrates, or lipids.

28. The method of any one of claims 17 to 27, wherein the immobilized ligand molecules are immobilized on the patterned surface by a linking moiety.

29. The method of claim 28, wherein the linking moiety is a streptavidin and/or a biotin.

30. The method of any one of claims 17 to 29, wherein the patterned surface comprises an inert or blocking component.

31. The method of any one of claims 17 to 30, wherein the inert or blocking component is between the interaction areas of the patterned surface.

32. The method of any one of claims 17 to 31, wherein the patterned surface has a regular pattern.

33. The method of any one of claims 17 to 32, wherein the interaction areas are about 3μm by 3 μm.

34. The method of any one of claims 17 to 33, wherein the distance between interaction areas is about 3μm.

35. A solid substrate comprising a patterned surface, the patterned surface comprising:

(a) a pattern of interaction areas, the interaction areas comprising immobilized ligands specific to an extracellular domain of a transmembrane protein; and (b) spaces between the interaction areas, the spaces comprising a blocking molecule.

36. The solid substrate of claim 35, wherein the blocking molecule is BSA.

37. The solid substrate of claim 35 or 36, wherein the spaces comprise a fluorophore.

38. The solid substrate of claim 37, wherein the fluorophore is Cy5.

39. The solid substrate of any one of claims 35 to 38, wherein the interaction areas are about 3 μm by 3 μm squares.

40. The solid substrate of any one of claims 35 to 39, wherein the solid substrate is glass.

41. A kit comprising the solid substrate of any one of claims 35 to 40.

42. A measurement device, the measurement apparatus comprising: a carrier adapted for receiving a solid substrate of any one of claims 35 to 40; an electromagnetic radiation source adapted for irradiating at least one of the interaction areas with primary electromagnetic radiation; an electromagnetic radiation detector adapted for detecting secondary electromagnetic radiation received from the irradiated at least one interaction area in response to the irradiation with the primary electromagnetic radiation.

43. The measurement apparatus of claim 42, comprising a solid substrate of any one of claims 35 to 40 received by the carrier.

44. The measurement apparatus of claim 42 or 43, wherein the electromagnetic radiation source and the electromagnetic radiation detector are configured to form a Fluorescence Recovery After Photobleaching arrangement.

45. The measurement apparatus of any one of claims 42 to 44, wherein the electromagnetic radiation source and the electromagnetic radiation detector are configured to form a Total Internal Reflection Fluorescence arrangement.

46. The measurement apparatus of any one of claims 42 to 45, adapted to move the electromagnetic radiation source, the electromagnetic radiation detector, and the carrier relative to one another for scanning the interaction areas of the solid substrate.

47. The measurement apparatus of any one of claims 42 to 46, wherein the electromagnetic radiation source comprises a first laser adapted for generating a pulsed laser beam having a first intensity and comprises a second laser adapted for generating a pulsed or continuous laser beam having a second intensity, wherein the first intensity is larger than the second intensity.

48. The measurement apparatus of any one of claims 42 to 46, wherein the electromagnetic radiation source comprises a laser adapted for generating a pulsed laser beam having a first intensity and adapted for generating a pulsed or continuous laser beam having a second intensity, wherein the first intensity is larger than the second intensity.

49. The measurement apparatus of any one of claims 42 to 48, wherein the electromagnetic radiation detector comprises a charge coupled device.