WO2003089900A2

WO2003089900A2 - Protein arrays and methods of production

Info

Publication number: WO2003089900A2
Application number: PCT/US2003/011826
Authority: WO
Inventors: Evgeny Vulfson; Christopher P. Marshall; Alexander Hoffmann
Original assignee: Avatar Medical, Llc
Priority date: 2002-04-15
Filing date: 2003-04-15
Publication date: 2003-10-30
Also published as: WO2003089900A3; AU2003225010A8; AU2003225010A1

Abstract

The present inventionprovides methods and compositions for orientation and covalent immobilization of protein on a surface, e,g., using light-mediated and/or oxidative reaction chemistry between functional groups on the surface and particular amino acid residue of the proteins. The present invention also provides protein arrays, such as protein arrays fabricated using light light mediated and oxidative chemistry. In particular, the invention provides methods for orientation and covalent immobilization of protein on a surface using light activated di-tyrosine cross-linking and arrays produced by light activated di-tyrosine cross-linking.

Description

PROTEIN ARRAYS AND METHODS OF PRODUCTION

1. FIELD OF THE INVENTION The present invention relates to methods and compositions for orientation and covalent immobilization of proteins on a surface, e.g., using light-mediated and/or oxidative reaction chemistry between functional groups on the surface and particular amino acid residues of the proteins. The present invention also relates to biosensors and protein arrays, such as biosensors and protein arrays fabricated using light-mediated and or oxidative chemistry.

2. BACKGROUND OF THE INVENTION

The ability of biological systems to recognize chemical substances is unparalleled. Using molecular bioreceptors honed by evolution, new means of chemical analysis have been developed utilizing the high selectivity of biological recognition systems. These systems, in combination with novel and improved transduction methods, have helped to promote the rapid development of bioanalysis with biosensors and biochips. Biosensors have a wide range of applications, primarily in the areas of biological/medical monitoring and environmental sensing.

Biosensors are defined as devices that consist of a bioreceptor and a transducer, whereby the interaction of the analyte with the bioreceptor is designed to produce an effect measured by the transducer, which converts the information into a measurable signal, such as an electrical pulse. In a typical biosensor bioreceptors are immobilized on or attached to a surface either to provide uniform coverage of said surface or in the form of spatially separated arrays. Arrays of bioreceptors and other biological molecules attached to a surface or solid support in such a spatially defined manner are generally referred to as biochips.

Bioreceptors bind the analyte of interest to the sensor for the measurement, and provide the specificity for biosensor technologies. These bioreceptors can take many forms and the different bioreceptors that have been used are as numerous as the different analytes that have been monitored using biosensors. Examples of bioreceptors include antibodies or antibody fragments, enzymes, and nucleic acids (see, e.g., Vo-Dinh et al., 2000, Fresenius J Anal Chem 366: 540-551). It is now widely acknowledged that future analytical and diagnostic methods will require large numbers of individual tests to be run in parallel, and controlled attachment of proteins to a surface is a crucial part of the fabrication of biochips (see, e.g., Vo-Dinli et al., 2000, Fresenius J Anal Chem 366: 540-551; Emili et al., 2000, Nature Biotech. 18: 393- 397). Biochips that contain proteins in precisely defined, spatially separated arrays therefore hold great promise, and the development of an immobilization technology that provides significant spatial control, and that improves upon the reproducibility in their fabrication is therefore a major technological challenge.

However, the development of biochip-based technologies, such as protem differential displays, has proven to be more difficult practically than the fabrication of analogous systems for the analysis of, for example, gene expression. This is firstly because the use of RT-PCR enables precise and facile amplification of mRNA, whereas no such method is available to amplify protein expression. Secondly, for functional assays, the proteins must be immobilized on the surface of a chip in such a way that they retain their activity and specificity. Impressive advances have been made with arrays of short polypeptides (see, e.g., Blawas et al., 1998, Biomaterials 19: 595-609) and in physical detection techniques such as surface plasmon resonance (SPR) and mass spectrometry (MALDI and SELDI) (see, e.g., Rich et al, 2000, Curr Opin. Biotechnol 11: 54-61; and Fung et al, 2001, Curr Opin. Biotechnol. 12: 65-69). However, the fabrication of arrays of large proteins and protein complexes remains a considerable challenge.

Currently, a variety of chemistries are available for covalent attachment of proteins to biochip and biosensor surfaces. For example, glass, coated with aldehyde containing silane reagent (MacBeath et al., 2000, Science 289: 1760-1763), or modified with aminopropyl trimethoxysilane (APTS) and activated with N-hydroxysuccinimide (Mendoza et al, 1999, Biotechniques 27: 778-788) or silicone treated with 3-glycidoxy propyltrimethoxysilane (Zhu et al, 2000, Nature Genet. 26: 283-289) has been used successfully for the immobilization of enzymes, antibodies and protein substrates. The common feature of all these techniques is that matrices coated/modified with reactive chemical groups do not allow specific orientation of the immobilized protein on the surface of the biochip, thereby impairing activity and leading to reduced and, in the worst case, ambiguous responses. Nevertheless, covalent attachment of proteins to surfaces remains the method of choice for large-scale analysis, as is evident from the fact that the large-scale proteomics studies referred to above all relied on covalent immobilization of proteins. Among the many advantages of using proteins covalently attached to surfaces, the following are

5 especially worth mentioning: covalently attached proteins do not leach, important particularly for continuous flow applications; covalently attached proteins can be dried on the surface of the chip, employing the beneficial effects of sugars and polymers for stabilization and refolding upon re-hydration; ligands can be specifically eluted thus allowing for further analysis and chip reuse (significant when biochips are linked to a

10 microprocessor).

Biochip technology allows simultaneous detection of multiple biotargets, and offers several advantages in size, performance, fabrication, analysis and production costs. A desired feature of the chemistry for fabrication of biochips is that it can be carried out with equal efficacy by chemical means or by irradiation in the presence of a photo-oxidant

15 system. The latter would be well suited for the generation of micropatterns using conventional lithographic masks and equipment used in integrated circuit manufacturing. This capability would offer significant advantages in terms of cost, scalability, minitiaturization, protein density and overall reproducibility and quality of the product. Bioreceptor orientation on the surface is a significant concern (see, e.g., Turkova,

20 1999, J. Chromatogr. B 722: 11-31). The major advantages of specifically and uniformly orienting proteins on biosensor and biochip surfaces lie in: (1) Increased ligandVanalyte binding capacity; and (2) Improved reproducibility and batch to batch consistency so that repeated biosensor/biochip calibration becomes unnecessary (uniform standard). Consequently, an increasing number of patents and publications address the use of highly

25 selective and high affinity protein-protein interactions as a means for protein attachment to surfaces. Specific orientation of proteins on the surface can be achieved, for example, by expressing them as fusions with appropriate binding domains and by chemical attachment of particular ligands. A number of such systems have been developed, with the classic poly-His tag and biotin/(strept)avidin pair-based devices being the most reliable.

30 The attachment of biotin either to the surface of biochips, or to the actual target, is now a well-established immobilization method (see, e.g. Burgener et. al, 2000, Bioconjugate Chem 11: 749-754; and Adamczyk et al, 2001, Bioconjugate Chem. 12: 139- 142). The expression of the target as a fusion with (strept)avidin is, on the first glance, simpler, as there are numerous protocols and commercially available reagents for covalent

35 modification of surfaces with biotin (see, e.g., Airenne et al, 1999, Biomolec. Eng. 16: 87- 92). However, in addition to forfeiting the above advantages to covalent attachment, these protocols also require a larger number of steps in the fabrication of the biosensors or biochips. For example, where the prote /peptide is biotinylated, it must be (1) synthesized/produced, (2) purified/analyzed, (3) biotinylated, (4) re-purified/re-analyzed and (5) attached to the surface. The covalent attachment of the protein to the surface of the chip is then implemented in additional step(s).

Oxidative cross-links between phenolic ring structures of proteins have been described both in vivo due to oxidative stress, aging and UV exposure (see, e.g., Stadman,

10 1992, Science 257: 1220-1224) and in vitro as a tool for the identification and mapping of polypeptides in multi-subunit protein complexes within biological systems (see, e.g., Brown et al, 1998, Biochemistry 37: 4397-4406; Fancy et al., 1999, Proc. Natl Acad. Sei. USA 96: 6020-6024). As shown in Figure 1, the covalent C-C bond between two tyrosyl-side chains is the result of an oxidative cross-link reaction between two tyrosines situated in a close

15 proximity to each other. These bonds are formed in the presence of a catalyst and oxidizing agent, and no atom is added. Although it is an oxidative reaction, no amino acids other than tyrosines have been shown to form cross-links under the conditions used, hi addition, there is a strict distance requirement between the tyrosine side-chains, with the bond forming only when the two are in very close proximity. Most di-tyrosine (DT) cross-linked proteins

20 examined so far have been shown to maintain activity (Malenic et al., 1994, Biochemistry 33: 13363-13376; Lardinois et al, 1999, J Biol Chem. 21 A: 35441-35448; Kanwar et al., 2000, Biochemistry 39: 14976-14983). The resulting C-C bond is stable under virtually any conditions. It has been shown to be stable in 20% beta-mercaptoethanol, 10% SDS at 100°C, and even under conditions of complete protein hydrolysis (6N HC1, 100°C). More

25 recently, it has been shown that the photolysis of ruthenium (II) tris-bypyridyl complex with visible light (-450 mn), in the presence of inorganic oxidant and proteins of interest, can lead to almost quantitative formation of DT-cross-linked products with irradiation time of less than 1 second (Fancy et al, 1999, Proc. Natl Acad. Sei. USA 96: 6020-6024). This chemistry has been used extensively for the formation of intra- and inter-molecular covalent

30 bonds within protein structures (see, for example, Malencik & Anderson, 1996, Biochemistry 35, 4375-4386 and references cited therein).

There is therefore a need for inexpensive, scalable and reproducible methods of attaching proteins to surfaces that enables (1) specific orientation of the target protein on the surface; and (2) covalent immobilization of the protein on the surface. In particular, there is

35 a need for methods that are compatible with current methods of fabrication of biochips and other biosensors for production of high density protein arrays and other novel or improved bioanalytical devices. Discussion or citation of a reference herein shall not be construed as an admission that such reference is prior art to the present invention.

3. SUMMARY OF THE INVENTION

The present invention provides methods for immobilizing polypeptides on a surface. In one embodiment, the present invention provides a method for immobilizing a polypeptide comprising at least one tyrosine residue on a support comprising at least one substituted phenolic group by cross-linking the polypeptide at a tyrosine residue to a substituted phenolic group on the support. In preferred embodiments, the cross-linking is carried out by a method comprising irradiating the support in the presence of the polypeptide with light of a wavelength in the range of 190-750 nm. In a preferred embodiment, the method further comprises a step of introducing the tyrosine residue or residues into the polypeptide. The solid support can be a planar surface. The solid support can also be a bead, e.g., a porous bead. In some embodiments, the substituted phenolic group is selected from the group consisting of o-alkyl, o-amino, o-alkoxy-, o-dialkylamino, j alkyl,^-amino, j9-alkoxy-, -dialkylamino phenolic groups.

In another embodiment, the present invention provides a method for immobilizing a polypeptide covalently attached to a molecular moiety having at least a given affinity for binding to a support and comprising at least one first functional moiety on a support comprising at least one second functional moiety which forms one or more covalent bonds with the first functional moiety under suitable conditions. The method comprises cross- linking the polypeptide to the second functional moiety on the support via the first functional moiety by a method comprising irradiating the support in the presence of the polypeptide with light of a wavelength in the range of 190-750 nm. In a specific embodiment, the invention provides a method for immobilizing a polypeptide on a support comprising at least one surface functional moiety, comprising (a) selecting a molecular moiety having at least a given affinity for binding to said support and comprising at least one first functional moiety, wherein said first functional moiety forms one or more covalent bonds with said surface functional moiety under suitable conditions; (b) attaching covalently said molecular moiety to said polypeptide; and (c) cross-linking said polypeptide at said first functional moiety to said surface functional moiety on said support by a method comprising irradiating said support in the presence of said polypeptide with light of a wavelength in the range of 190-750 nm. Preferably, the molecular moiety is a peptide. In a preferred embodiment, the peptide comprises 5-20 amino acids. In another preferred embodiment, the peptide is selected from a random peptide library. In a more preferred embodiment, the first functional moiety in the peptide is a tyrosine residue and the second functional moiety is a substituted phenolic groups, e.g., a substituted phenolic group selected from the group consisting of o-alkyl, o-amino, ø-alkoxy-, o-dialkylamino, -alkyl, _/3-amino, ^-alkoxy-, />-dialkylamino phenolic groups. The solid support can be a planar surface. The solid support can also be a bead, e.g., a porous bead. In one embodiment, the molecular moiety comprises a plurality of histidines. Preferably, the molecular moiety is a peptide of at least 8, 10 or 12 amino acids in length. Preferably, at least 20%, 30%>, 40%> or 50% of the amino acids in the molecular moiety are His, Pro or Ser, or a mixture thereof. In a preferred embodiment, the molecular moiety comprises a motif arranged in the sequence HIHHH, or any one of numerous variations, such as HHTHHH, HLHHTH, HHTHAH, HHHSH, HfflHI, HHHKHH. Such a histidine-rich motif is preferably at either the N- or C- terminus of the polypeptide, and can be in either orientation. In another preferred embodiment, the molecular moiety further comprises an additional 5-6 amino acids, of which one or more amino acids can form hydrogen bonds, such as serine and arginine. In one embodiment, the molecular moiety is selected from the 12-mer peptides as described by SEQ ID NO:l through SEQ ID NO:28. Preferably, the molecular moiety also comprises one or more tyrosine residues, hi a preferred embodiment, the affinity tag is selected from the group consisting of TTYSRHDHfflHH, HHTHYSPHRGTP, HHffllYKFTESV, and IHQHHFPLWPYP. In a preferred embodiment, polypeptides comprising such molecular moiety are bound to a phenol-modified surface.

The invention also provides methods for producing an array of immobilized polypeptides. h one embodiment, the invention provides a method for producing an array comprising a support and a plurality of different polypeptides, each attached to the support at a different predetermined location. Each of the plurality of different polypeptides comprises at least one tyrosine residue, and the support comprises a plurality of substituted phenolic groups. The method comprises cross-linking each of the plurality of different polypeptides to one of the plurality of substituted phenolic groups via a tyrosine residue at a predetermined location on the support. In preferred embodiments, the cross-linking is carried out by a method comprising irradiating the support in the presence of one or more of the plurality of different polypeptides with light of a wavelength in the range of 190-750 nm. In a preferred embodiment, the method further comprises a step of introducing the tyrosme residue or residues into one or more of the plurality of different polypeptides. The solid support is preferably a planar surface. In some embodiments, the substituted phenolic group is selected from the group consisting of o-alkyl, o-amino, o-alkoxy-, o-dialkylamino, /?-alkyl, >-amino, p-akoxy-, ^»-dialkylamino phenolic groups.

In another embodiment, the invention provides a method for producing an array comprising a support comprising a plurality of second functional moieties which forms one or more covalent bonds with a first functional moiety under suitable conditions and a plurality of different polypeptides, each of which is covalently attached to a molecular moiety having at least a given affinity for binding to said support and comprising at least one first functional moiety. Each of the plurality of different polypeptides is attached to the support at a different predetermined location. The method comprises cross-linking each of the polypeptides to one of the plurality of the second functional moieties on the support via the first functional moiety by a method comprising irradiating the support in the presence of the polypeptide with light of a wavelength in the range of 190-750 nm. In a specific embodiment, the invention provides a method for producing an array of immobilized polypeptides, said array comprising a support and a plurality of different polypeptides, said support comprises a plurality of surface functional moieties, each said different polypeptide being attached to said support at a different predetermined location, said method comprising (a) selecting one or more molecular moieties having at least a given affinity for binding to said support and comprising at least one first functional moiety, wherein said first functional moiety forms one or more covalent bonds with said surface functional moiety under suitable conditions; (b) attaching covalently one of said one or more molecular moieties to each of said plurality of polypeptides; and (c) cross-linking each said polypeptide at said first functional moiety to said surface functional moiety at a predetermined location on said support by a method comprising irradiating said support in the presence of each said polypeptide with light of a wavelength in the range of 190-750 nm. Preferably, the molecular moiety is a peptide. In a preferred embodiment, the peptide comprises 5-20 amino acids. In another preferred embodiment, the peptide is selected from a random peptide library. In a more preferred embodiment, the first functional moiety in the peptide is a tyrosme residue and the second or surface functional moiety is a substituted phenolic group, e.g., a substituted phenolic group selected from the group consisting of o- alkyl, o-amino, o-alkoxy-, o-dialkylamino, 7-alkyl, /?-amino, -alkoxy-, j^dialkylamino phenolic groups. The solid support can be a planar surface. In one embodiment, the molecular moiety comprises a plurality of histidines. Preferably, the molecular moiety is a peptide of at least 8, 10 or 12 amino acids in length. Preferably, at least 20%), 30%>, 40% or 50% of the amino acids in the molecular moiety are His, Pro or Ser, or a mixture thereof. In a preferred embodiment, the molecular moiety comprises a motif arranged in the sequence HIHHH, or any one of numerous variations, such as HHTHHH, HLHHTH, HHTHAH, HHHSH, HfflHI, HHHKHH. Such a histidine-rich motif is preferably at either the N- or C- terminus of the polypeptide, and can be in either orientation, hi another preferred embodiment, the molecular moiety further comprises an additional 5-6 amino acids, of which one or more amino acids can form hydrogen bonds, such as serine and arginine. In one embodiment, the molecular moiety is selected from the 12-mer peptides as described by SEQ ID NO:l through SEQ LD NO:28. Preferably, the molecular moiety also comprises one or more tyrosine residues. In a preferred embodiment, the affinity tag is selected from the group consisting of TTYSRHDHfflHH, HHTHYSPHRGTP, HfflHIYKFTESV, and ffiQHHFPLWPYP. In a preferred embodiment, polypeptides comprising such molecular moiety are bound to a phenol-modified surface.

The invention also provides arrays of immobilized polypeptides. In one embodiment, the invention provides an array comprising a support comprising a plurality of substituted phenolic groups and a plurality of different polypeptides, each of which comprising at least one tyrosine residue. On such an array, each of the polypeptides is cross-linked at the tyrosine residue to a substituted phenolic group on the support at a predetermined location.

In another embodiment, the invention provides an array comprising a support comprising a plurality of substituted phenolic groups and a plurality of different polypeptides, each of which is covalently attached to a predetermined location on the support via a molecular moiety having at least a given affinity for binding to the support and comprising at least one first functional moiety which is covalently attached to one of the substituted phenolic groups on the support. In one embodiment, the molecular moiety comprises a plurality of histidines. Preferably, the molecular moiety is a peptide of at least 8, 10 or 12 amino acids in length. Preferably, at least 20%>, 30%>, 40%> or 50%o of the amino acids in the molecular moiety are His, Pro or Ser, or a mixture thereof. In a preferred embodiment, the molecular moiety comprises a motif arranged in the sequence HIHHH, or any one of numerous variations, such as HHTHHH, HLHHTH, HHTHAH, HHHSH, HHEHI, HHHKHH. Such a histidine-rich motif is preferably at either the N- or C-terminus of the polypeptide, and can be in either orientation. In another preferred embodiment, the molecular moiety further comprises an additional 5-6 amino acids, of which one or more amino acids can form hydrogen bonds, such as serine and arginine. In one embodiment, the molecular moiety is selected from the 12-mer peptides as described by SEQ ID NO:l through SEQ ID NO:28. Preferably, the molecular moiety also comprises one or more tyrosine residues, hi a preferred embodiment, the affinity tag is selected from the group consisting of TTYSRHDHIHHH, HHTHYSPHRGTP, HHIHiYKFTESV, and IHQHHFPLWPYP. h a preferred embodiment, polypeptides comprising such molecular moiety are bound to a phenol-modified surface.

The solid support of the arrays of the invention is preferably a planar surface. In some embodiments, the substituted phenolic groups on the solid support are selected from the group consisting of o-alkyl, o-amino, o-alkoxy-, o-dialkylammo,^-alkyl, >-amino,^>- alkoxy-, / dialkylamino phenolic groups.

Preferably, the arrays produced by any of the methods of the invention comprise in the range of 100 to 1,000 different polypeptides per 1 cm², in the range of 1,000 to 10,000 different polypeptides per 1 cm², in the range of 10,000 to 50,000 different polypeptides per 1 cm², in the range of 50,000 to 100,000 different polypeptides per 1 cm², or in the range of 100,000 to 100,000 different polypeptides per 1 cm². Preferably, the arrays comprise at least 100; 1,000; 10,000; 50,000; 100,000; 500,000; 10⁶; 10⁷; 10⁸; 10⁹; or 10¹⁰ different polypeptides per 1 cm . In some preferred embodiments, the arrays produced by any of the methods of the invention comprise polypeptides comprising at least 5, 10, 25, 50, 100, 150, 200, 500 or 1,000 amino acids.

The invention further provides a surface comprising a plurality of substituted phenolic groups and a surface comprising a plurality of substituted phenolic groups, each cross-linked to a tyrosine residue of a polypeptide. The substituted phenolic groups on the surface of the invention can be selected from the group consisting of o-alkyl, o-amino, o- alkoxy-, o-dialkylamino, )-alkyl, -amino,j)-alkoxy-,/?-dialkylamino phenolic groups.

4. BRIEF DESCRIPTION OF FIGURES Figure 1 shows the chemistry and reaction mechanism of DT cross-linking (clockwise).

Figure 2 shows a schematic illustration of the chemistry used for surface modification and subsequent attachment of protein ligands through an oligopeptide tag.

Figure 3 shows a schematic illustration of the chemical process for derivatization of beads for cross-linking with proteins.

Figure 4 shows a schematic illustration of the chemistry used for the introduction of phenolic residues on the surface of the chip. Figure 5 shows binding of peptide 1 (TTY-Sequence 1 hybrid) to phenol-modified glass. The curve fitting of experimental points using GraphPad Prism (software) gave the best fit for Kd=4.45 x 10^"8.

5. DETAILED DESCRIPTION OF THE INVENTION The invention provides methods and compositions for the orientation and covalent immobilization of proteins on surfaces and for production of protein arrays. In particular, the invention provides methods that combine photochemistry and/or oxidative cross-linking chemistry with microfabrication methodologies for production of high density protein arrays. In the invention, a solid support, e.g., a glass surface, containing immobilized functional groups, e.g., functional phenolic groups, is prepared. Proteins or polypeptides to be immobilized in a particular orientation are fused to a peptide which has affinity to solid support and contains one or more amino acid residues, e.g., tyrosine residues, which can form covalent bond(s) with the functional groups on the surface of solid support. Preferably, the amino acid residues of the peptide form covalent bonds with the functional groups on the solid support by a light-activated reaction. Fusion proteins are then covalently cross-linked to the solid support to produce a high density array of one or more proteins, each immobilized at a particular location on the solid support.

5.1. PREPARATION OF SOLID SUPPORT

The invention provides methods for preparing a solid support bearing functional groups, e.g., phenolic functional groups. Such solid supports can be used for covalent attachment of proteins or oligopeptides via a reaction, e.g., a light-activated reaction, between suitable residues, e.g., tyrosine residues, on the proteins and the immobilized functional residues, e.g., phenolic residues, on the supports.

As used herein, "solid support" refers to any solid material that comprises at least one surface to which functional groups, e.g., phenolic groups, can be covalently attached.

Any suitable material can be used as the solid support, including but not limited to latex, polystyrene, polypropylene, polyacrylamide, glass, silicon including modified silicon, other oxides and modified oxides, and modified metal surfaces such as gold and silver. In a preferred embodiment, the solid support is a planar solid surface. In another preferred embodiment, the solid support is a solid surface containing wells. In still another preferred embodiment, the solid support is a bead. In still another preferred embodiment, the bead is a porous bead. In a preferred embodiment the solid support is a glass surface, e.g., surface of a microscope slide. In another preferred embodiment, the solid support is an oxidized silicon. Oxidized silicon can be prepared using a standard procedure for producing an insulating layer of silicon dioxide on the surface of a wafer, e.g., thermal oxidation. The glass or oxidized silicon surface is then cleaned and activated, e.g., to expose hydroxy groups, using a standard procedure known in the art. In one embodiment, the glass or oxidized silicon surface is hydrolyzed under strong basic conditions, rinsed, dried and then treated with piranha solution (70%> sulphuric acid, 30%> hydrogen peroxide), washed and dried again. The activated surface is then silanized using a standard method known in the art (see, e.g., Durfor et αZ, 1994, Langmuir 10: 148-152).

The surfaces can then be functionalized with immobilized functional groups, e.g., phenolic groups. In a preferred embodiment, surfaces containing phenolic groups are prepared by a method comprising using commercially available Benzooxasilepin dimethyl ester to introduce 2-phenolic groups (which are as reactive as 4-phenols) onto the surface. The method requires only a minimum number of steps as schematically illustrated in Figure 2. This treatment produces a modified glass surface with covalently attached phenolic residues. In another preferred embodiment, the widely-used and highly efficient aminopropyl trimethoxysilane-based method is employed (Durfor et al, 1994, Langmuir 10: 148-152; Mendoza et al, 1999, Biotechniques 27:778-788; Cunliffe et al, 1999, Biotech. Lett 22:141-145; and Cunliffe et al, 2000, Appl Environ. Microbiol. 65:4995-5002). An ordinary skilled person in the art would be able to recognize that a number of other methods, techniques and reagents can be used to introduce the functional groups of the present invention to the surface.

The surface density of functional phenolic groups is an important parameter for optimal coupling of proteins. Too high a degree of modification could prove detrimental for attaching proteins, as proteins are known to adhere non-specifically to highly hydrophobic interfaces. On the other hand, too low a degree of modification would be undesirable from a functional standpoint, and may impair the efficiency of cross-linking. The desired surface density of functional phenolic groups can be determined by an ordinary skilled person in the art by routine procedures. For example, a range of surfaces with varying density of functional phenolic groups are prepared. The densities are then correlated with the efficiency of cross-linking. The surface with the optimal attachment of proteins, e.g., via di-tyrosine photo cross-linking, can then be determined. The surface density of functional phenolic groups depends in part on the density of surface hydroxy groups. In one embodiment, the density of surface hydroxy groups, i.e., the degree of hydrolysis, is adjusted by changing the time or temperature of the piranha treatment (Cunliffe et al, 1999, Biotech. Lett 22:141-145). The surface density of functional phenolic groups also depends on the concentration of Benzooxasilepin dimethyl ester used and the duration of treatment. In one embodiment, the concentration is adjusted such that a desired surface density of functional phenolic groups is obtained. An ordinary skilled person in the art will be able to determine the appropriate combination of the density of surface hydroxy groups and the concentration of Benzooxasilepin dimethyl ester once the desired surface density of functional phenolic groups is determined.

Various standard methods can be used to monitor the average surface density of phenolic groups. In a preferred embodiment, a standard fluorescent assay, e.g., derivatization with 4-(N-phthalimidinyl) benzenesulfonyl chloride to give fluorescent sulfonyl esters (Tsuruta et al. 1996, Anal. Biochem. 243:86-91) is used to routinely monitor the average surface density of phenolic groups. Preferably, the density of functional groups, e.g., phenolic groups, on the support is in the range of 1 per 10 to 100 A². In a preferred embodiment, the support disclosed herein contains phenolic groups at a desired density on surface. Those skilled persons in the art would be able to recognize that other functionality such as for example, various substituted phenolic groups, e.g., those selected from the group consisting of o-alkyl, o-amino, o-alkoxy-, o-dialkylamino, p-alkyl, p-ax no, /^-alkoxy-, j?-dialkylamino phenolic groups can all be used successfully to practice the present invention. In addition, any other substituted phenolic groups that can react with amino acid residues of a protein, such as tyrosine residues, when immobilized on a solid support can also been used. Preferably, the substituted phenolic groups react with amino acid residue of the protem in a light activated manner.

In another preferred embodiment, porous beads are prepared for immobilization of proteins via DT cross-linking. High surface area resin beads, which are the most suitable for the immobilization of polypeptides via the DT cross-linking technology, are made industrially by copolymerization of styrene and divinylbenzene in aqueous suspensions (Arshady, 1991, J. Chromatography 586: 181-219; VivaldoLima et al, 1997, bid. Eng. Chem. Res. 36:939-965, each of which is incorporated by reference herein in its entirety). Examples of such polymers include but not limited to Amberlite™, Dowex™ and Merrifield resins, hi order to make these resins compatible with the DT cross-linking technology the introduction of suitable functionality, e.g., phenolic groups or substituted phenolic groups, is required. This is achieved by replacing some of the styrene in the monomer feed by acetoxy-styrene (phenolic groups are effective inhibitors of free radical polymerization). The acetoxy-styrene monomer is commercially available and inexpensive. The subsequent treatment of the resin with base yields the desired immobilization materials. The overall process is schematically illustrated in FIG. 3.

The great advantage of the approach is that phenolic groups, which are necessary for cross-linking, can also be introduced into weak ion-exchange resins of the type mentioned above to achieve initial non-covalent adsorption of the proteins to the surface. This methodology is versatile, inexpensive and well suited for applications where non-covalent attachment of protein to surface is insufficient. It should also be stressed that these resins are widely used in a whole host of large-scale industrial applications and their surface area, pore size and mechanical characteristics have been extensively optimized for such use. The polymers are prepared using conventional protocols of suspension polymerization on 10-50g scale (Arshady, 1991, J. Chromatography 586: 181-219; VivaldoLima et al, 1997, Ind. Eng. Chem. Res. 36:939-965) and characterized with regard to surface area and pore size using the N₂ adsorption isotherm method and electron microscopy. If necessary, the content of 4-acetoxy styrene in the resins can be optimized to maximize the efficiency of the cross-linking and protein loading, without compromising the overall structure, porosity and the mechanical strength of the material. The attachment of proteins to the resin is accomplished as described in Example 1 for fusion scFvs.

5.2. PREPARATION OF PROTEINS

The invention provides methods for producing polypeptide linked with an affinity tag. As used herein, an "affinity tag" refers to a molecular moiety which is linked to a polypeptide or protein and which attaches to a surface with an affinity above a given threshold affinity level. In a preferred embodiment, the threshold affinity level is determined by a particular wash condition, e.g., the pH and/or ionic strength of the solution used and the duration and/or temperature of the wash. The affinity tags which remain bound to a surface after being subject to a particular wash is said to attach to the surface with an affinity above the threshold defined by the wash condition. An ordinary skilled person in the art will be able to choose an appropriate wash condition for a desired affinity level. Preferably, the affimty of an affinity tag is higher than the poplypeptide or protein which it links to. The attachment of such an affinity tag to a surface enhances the orientation of the linked protein on the surface. Preferably, the affinity tag attaches to a surface comprising functional groups with an affinity above a given threshold affinity level. More preferably, the affinity tag attaches to a surface comprising functional phenolic groups with an affimty above a given threshold affinity level. In preferred embodiments, the affinity tag attaches to the surface by non-covalent attachment, e.g., physical adsorption. Preferably, the affinity tag is a peptide comprising 1-25 amino acids. More preferably, the affinity tag comprises one or more tyrosine residues. In a preferred embodiment, a peptide affinity tag attaches to a surface with an affinity higher than at least 50%>, 60%>, 70%, 80%>, 90%, 95%o, 99%o, or 99.9% of other peptides having the same number of amino acid residues. In an another preferred embodiment, a peptide affinity tag attaches to a surface with an affinity higher than at least 50%, 60%, 70%, 80%, 90%, 95%, 99%, or 99.9% of other peptides in a random peptide library. In some embodiments, the affinity tag is linked with the polypeptide directly, e.g., by a peptide bond. In some other embodiments, the affinity tag is linked with the polypeptide via a linker peptide. The affinity tag or affinity peptide of the present invention can be linked to either the N- or C- terminus of the polypeptide or protein to be immobilized. The affinity tag or affinity peptide can also be inserted at an appropriate location in the polypeptide or protein such that the immobilized polypeptide or protein oriented in a desired orientation.

The selection of affinity tags or affinity peptides suitable for practicing the present invention can be accomplished by a number of methods well known in the art. h a preferred embodiment the affinity peptides are selected using conventional phage display methodology (Smith, 1985, Science 228 :1315-1317). This methodology has been used successfully to identify small peptides with high affinity to a number of proteins and non- peptide molecules (McGregor, 1996, Molecular Biotechnol. 6:155-162; Cesareni et al, 1999, Combinatorial Chemistry & High Throughput Screening 2:1-17). In the method, random oligopeptides are displayed as fusions to coat proteins of filamentous phages. Biopanning is used to select for phages with affinity for the target, e.g., the surface of a solid support. Subsequent amplification of the bound phages allows for an iterative purification of phage carrying high affinity sequences. In a more preferred embodiment, the selection is carried out in two steps. In the first step oligopeptides with high affinity to the surface of the solid support are obtained. A second step is then followed to allow the selection of sequences suitable for use in the light-mediated cross-linking from the oligopeptides obtained in the first step.

It is preferable to use one of the commercially available filamentous fungi, for example, Ml 3 with protein/polypeptide library fused to the phage coat protein. Any random library of short peptides, e.g., peptides containing 7-12 amino acids, and library of longer peptides and proteins, such as antibodies, which can be used to preferentially enrich particular amino acids, e.g., tyrosine residues, for oxidative cross-linking can be used successfully. The library preferably contains about 10 -10 different sequences for adequate selection. More preferably, libraries having larger diversities are used. In one embodiment, phages are first allowed to bind to the surface of the present invention. The unbound phages are then washed away. The ones that remain bound, i.e., the specifically bound phages, are then eluted. Such phages are used to infect E.coli for titration and amplification. Following 3-10 cycles of biopanning, preferably 4-6 cycles, the enrichment is maximal, and clones with the desired characteristics are isolated and sequenced. The peptides obtained can be used directly in construction of fusion proteins. In a preferred embodiment, the peptides obtained are subject to additional protein processing such as adding further amino acids, e.g., tyrosine, to the peptides to enhance the yield of immobilized protein. Any methods for introducing tyrosine residues into a polypeptide can be used (see, e.g., PCT publication WO 01/29247, published on April 26, 2001; and U.S. patent application 09/837,235, filed on.April 18, 2001, each of which is incorporated by reference herein in its entirety). In another preferred embodiment, dimerization or multimerization of the obtained peptides are earned out to enhance their binding affinities to the surface. An ordinary skilled person in the art will be able to recognize that both addition of residues and multimerization can also be carried out. An ordinary skilled person in the art will also be able to recognize that any other protein modification techniques known in the art which may lead to improved affinity and efficiency of cross-linking or both can also be used. Although the use of a phage display is the prefened embodiment of the present invention, an ordinary skilled person in the art will be able to recognize that any other library of peptides and/or selection procedure can also be used in conjunction with the present invention. For example, chemically synthesized peptide libraries, cell displays, and protein-nucleic acid fusions can also be used in the present invention.

Once obtained, an affimty tag or affinity peptide of the present invention is preferably fused to either the N- or C- tenninus of a polypeptide or protein to be immobilized using conventional molecular biological techniques. The affinity tag or affinity peptide can also be inserted at an appropriate location in a polypeptide or protein such that the immobilized polypeptide or protein oriented in a desired orientation. In a prefened embodiment, the affinity tag is inserted in a particular surface domain, i.e., a domain which is exposed to the environment, of a polypeptide or protein such that the polypeptide or protein is bound to the surface via the particular surface domain and the

5 affinity tag. It will be apparent to an ordinary skilled person in the art that such a domain can comprise hydrophilic residues. In such an embodiment, the polypeptide or protem is oriented such that domains other than the particular surface domain linked to the affinity tag can have normal or close to normal activities. A large number of techniques and host systems for production of fusion proteins have been described in the literature over the last

10 20 years and any person skilled in the art can find a suitable protocol for constructing and expressing the fusion proteins. Preferably, the affinity tag is fused to one end of a polypeptide. More preferably, the affinity tag is fused to one subunit of multi-subunit polypeptide. The polypeptide to be immobilized can contain as little as 5 amino acids. The polypeptide to be immobilized can also be as large as 1,000 kilodaltons (KD) in cases of

15 complex multi-subunit proteins or even bigger in cases of protein complexes or assemblies, e.g., ribosomes. In a particular embodiment, the polypeptide to be immobilized is linked to a cell. Such an embodiment is useful for immobilizing cells on a solid support, h prefened embodiments, polypeptides or proteins of molecular weight from 30KD to 300KD are used as this is a typical range of the molecular weight of the majority of bioreceptors such as

20 antibodies and enzymes.

hi one embodiment, the affinity tag of the invention comprises a plurality of histidines. Preferably, the affinity tag is at least 8, 10 or 12 amino acids in length. Preferably, at least 20%>, 30%>, 40%) or 50% of the amino acids in the affinity tag are His, Pro or Ser, or a mixture thereof. In a prefened embodiment, the affinity tag of the invention

25 comprises a motif arranged in the sequence HIHHH, or any one of numerous variations, such as HHTHHH, HLHHTH, HHTHAH, HHHSH, HHTHI, HHHKHH. Such a histidine- rich motif is preferably at either the N- or C-terminus of the polypeptide, and can be in either orientation. In another prefened embodiment, the affinity tag further comprises an additional 5-6 amino acids, of which one or more amino acids can form hydrogen bonds,

30 such as serine and arginine. In one embodiment, the affinity tag is selected from the 12-mer peptides as described by SEQ ID NO:l through SEQ ID NO:28. Preferably, the affinity tag also comprises one or more tyrosine residues. In a prefened embodiment, the affinity tag is selected from the group consisting of TTYSRHDHIHHH, HHTHYSPHRGTP, HHffllYKFTESV, and IHQHHFPLWPYP. The invention thus also provides protein arrays

35 comprising a plurality of different polypeptides, each of which comprises an affinity tag described above. 5.3. PRODUCTION OF PROTEIN ARRAYS The present invention provides methods for producing high density anays of oriented and covalently immobilized proteins. The methods of the invention comprise cross-linking polypeptides or proteins described in Section 5.2. with a surface described in Section 5.1. Preferably, each different fusion protein is immobilized to a different predetermined location on the surface. The cross-linking can be carried out using any one of the standard methods known in the art. Preferably, a light activated cross-linking, e.g., light activated di-tyrosine cross-linking, is used (see, e.g., PCT publication WO 01/29247, published on April 26, 2001; and U.S. patent application 09/837,235, filed on April 18, 2001, each of which is incorporated by reference herein in its entirety). hi one embodiment of the invention, the orientation of immobilized polypeptide is achieved by employing an affinity tag fused to the polypeptide. The affinity tag attaches to the surface with an affinity above a given threshold affinity level. After allowing the fusion polypeptide contacting the surface, the surface is washed with a suitable solution to remove any fusion polypeptides which attach to the surface with an affinity below the threshold affinity level as defined by the wash condition. The reaction between a functional moiety in the affinity tag and a functional moiety on the surface is then activated such that fusion polypeptides which remain on the surface are cross-linked to the surface. By such a procedure, the immobilized polypeptides are linked to the surface via the affinity tag, and therefore, are oriented. Such a procedure can be used in conjunction with any of the methods described infra for producing anays of immobilized and oriented polypeptides. In some embodiments of the invention, a photolithographic method is used to produce high density anays. In one embodiment, a region on a suitable surface is illuminated using a suitable light source in the presence of the polypeptide or protein to be immobilized at the location under reaction conditions such that cross-linking of the polypeptide to the surface occurs. The procedure is then carried out to cross-link another polypeptide at a different region. By sequentially applying the procedure with a different polypeptide and at a different region, a plurality of different polypeptides are each immobilized at a different region on the surface. Any one of the known photolithographic methods can be adapted for use in the present invention. In the invention, light having a wavelength or wavelengths in the range of 190-750 nm can be used. In some embodiments, cross-linking is activated by light directly. In some other embodiments, the cross-linking is achieved by the catalysis of a light activated catalyst. An ordinary skilled person in the art will be able to choose the appropriate wavelength or wavelengths once the cross-linking chemistry is selected. It will also be apparent to one skilled in the art that in such photolithographic methods, the size of each polypeptide spot, i.e., the size of each surface region that contains a particular immobilized polypeptides, and the density of such spots are determined by, e.g., the mask feature size and the wavelength of light used. In a preferred embodiment, a di-tyrosine photoactivated cross-linking chemistry is used, in which the photo-oxidation is carried out using 150 W Xe arc lamp with a 280 nm longpass cutoff filter and ruthenium (II) in the presence of tris-bypyridyl and ammonium persulphate as catalyst and oxidant respectively (Fancy et. al., 1999, Proc. Natl. Acad. Sei. USA 96: 6020-6024, which is incorporated by reference herein in its entirety).

In a prefened embodiment of a photolithographic method, a mask contains a hole of a chosen size, e.g., 500 x 500 nm, is used, hi a prefened embodiment, the surface is illuminated via the mask by an 150 W Xe arc lamp with a longpass cutoff filter at 280 nm. The hole in the mask is scanned to each different predetermined location for each different polypeptide so that the polypeptide is immobilized at the location, h another prefened embodiment, a beam of light of a desired wavelength or range of wavelengths is used. The beam is focused to a chosen spot size on the surface. The surface is scanned such that a different predetermined location is illuminated when a different polypeptide is added so that the polypeptide is immobilized at the location.

In some other embodiments of the invention, a spraying, spotting or printing method is used to produce the high density anays. Any one of the known methods can be used, including but not limited to pipette spotting, pin spotting, inkjet spotting, and electrospraying (see, e.g., U.S. Patent Nos. 6,063,339; 6,350,609; 6,323,043; 6,319,674, each of which is incorporated by reference herein in its entirety). In such a method, each polypeptide or protein to be immobilized on the surface is delivered to the surface sequentially using, e.g., an inkjet printing method, under reaction conditions such that cross- linking of the polypeptide to the surface can occur. In such methods, the size of polypeptide spots, i.e., the surface regions containing immobilized polypeptides, and the density of spots are determined by the particular method and are known to one skilled in the art based on the method used. These methods can be used in conjunction with any suitable cross-linking chemistries. In a preferred embodiment, a di-tyrosine cross-linking chemistry is used, in which the reaction is initiated by the addition of H₂O₂ (0.1-lmM) and 0.1 mM hemin catalyst (see, e.g., Campbell, Kodadek & Brown, 1996, Bioorganic and Medicinal Chemistry, 6, 1301-1307, which is incorporated by reference herein in its entirety) or the addition of H O (0.1-lmM) and 1 mg/mL horse radish peroxidase (see, e.g., Malencik & Anderson, 1996, Biochemistry 35, 4375-4386, which is incorporated by reference herein in its entirety); and the incubation is continued for 5 and 30 minutes for hemin and horse radish peroxidase, respectively.

The spotting or printing methods can also be used in conjunction with light mediated cross-linking. In such an embodiment, a suitable surface is illuminated using a suitable light source. Each polypeptide or protem to be immobilized on the surface is then delivered to the surface sequentially using, e.g., an inkjet, under reaction conditions that allow photoactivated cross-linking.

5.4. PROTEIN ARRAYS PRODUCED The invention also provides anays produced by the methods described, supra, hi one embodiment, the invention provides anays comprising a support which comprises a plurality of particular functional moieties and a plurality of different polypeptides each of which is covalently attached to an affinity tag molecule of the invention (see, Section 5.2.) having at least a given affinity to said support and comprising at least a functional moiety which forms one or more covalent bonds with the particular functional moiety on the surface. Preferably, the support is a planar support.

In a preferred embodiment, the invention provides arrays comprising a support which comprises a plurality of substituted phenolic groups and a plurality of different polypeptides, in which each of the different polypeptides comprises at least one tyrosine residue and is cross-linked via the tyrosine residue to one or more phenolic groups on the support at a predetermined location. Preferably, the support is a planar support. Preferably, the substituted phenolic group groups are selected from the following: o-alkyl, o-amino, o- alkoxy-, o-dialkylamino^-alkyl, j9-amino, ^-alkoxy-, -dialkylamino phenolic groups.

Preferably, the anays produced by any of the methods of the invention comprise in the range of 100 to 1,000 different polypeptides per 1 cm², in the range of 1,000 to 10,000 different polypeptides per 1 cm², in the range of 10,000 to 50,000 different polypeptides per 1 cm , in the range of 50,000 to 100,000 different polypeptides per 1 cm , or in the range of 100,000 to 100,000 different polypeptides per 1 cm². Preferably, the arrays comprise at least 100; 1,000; 10,000; 50,000; 100,000; 500,000; 10⁶; 10⁷; 10⁸; 10⁹; or 10¹⁰ different polypeptides per 1 cm². Preferably, each of the polypeptides on the anay comprises at least 5, 10, 25, 50,

100, 150, 200, 500 or 1,000 amino acids. In some other prefened embodiments, the anays produced by any of the methods of the invention comprise polypeptides having molecular weights in the range of 30KD to 300KD. In some embodiments, the arrays produced by any of the methods of the invention comprise polypeptides having molecular weights as high as 1,000 KD.

Preferably, each spot or region of polypeptide on the array contains the polypeptide with a high purity. In a prefened embodiment, each spot is at least 10%>, 20%, 40%, 50%>, 60%, 70%), 80%o, 90%), 95%ι, or 99%> pure of the respective polypeptides immobilized in the spot with the desired orientation. As used herein, a percentage of purity of a given polypeptide with a particular conformation in a spot refers to the percentage of molecules of the polypeptide with the particular conformation in the spot in all polypeptide molecules in the spot. hi one embodiment, the protein anays comprise a plurality of different polypeptides, each of which comprises an affinity tag comprising a plurality of histidines. Preferably, the affinity tag is at least 8, 10 or 12 amino acids in length. Preferably, at least 20%>, 30%>, 40% or 50%) of the amino acids in the affinity tag are His, Pro or Ser, or a mixture thereof. In a preferred embodiment, the affinity tag comprises a motif ananged in the sequence HIHHH, or any one of numerous variations, such as HHTHHH, HLHHTH, HHTHAH, HHHSH, HHLHI, HHHKHH. Such a histidine-rich motif is preferably at either the N- or C-terminus of the polypeptide, and can be in either orientation, h another prefened embodiment, the affimty tag further comprises an additional 5-6 amino acids, of which one or more amino acids can form hydrogen bonds, such as serine and arginine. In one embodiment, the affinity tag is selected from the 12-mer peptides as described by SEQ ID NO:l through SEQ ID NO:28. Preferably, the affinity tag also comprises one or more tyrosine residues. In a prefened embodiment, the affinity tag is selected from the group consisting of TTYSRHDHfflHH, HHTHYSPHRGTP, FfflffllYKFTESV, and IHQHHFPLWPYP. In a preferred embodiment, polypeptides comprising such affinity tags are bound to a phenol- modified surface. 6. EXAMPLES

The following examples are presented by way of illustration of the present invention, and are not intended to limit the present invention in any way.

6.1. EXAMPLE 1 Preparation of immobilization surface:

Modified glass surface is prepared by hydrolyzing commercially available glass beads (Sigma- Aldrich) or 13 mm microscope glass slides are hydrolyzed by immersion in sodium hydroxide solution (aqueous, 5M) for 1 h and then washed thoroughly with deionized water. They are then soaked in the piranha solution (70% sulphuric acid, 30% hydrogen peroxide), washed and dried again. The modified glass surfaces are prepared immediately prior to silylation by drying them in a hot oven for 30 min. The silylation is carried out in methanol-acetic acid containing 4% water and 1%> Benzooxasilepin dimethyl ester as described by Durfor et al, 1994 (Langmuir 10: 148-152) for lh with occasional gentle shaking. The resultant phenolic group containing surface is washed with methanol 10 and cured in a hot oven for 30 min.

Primary oligopeptide selection:

A commercially available, random 10-amino acid peptide library which is constructed as a fusion with the major coat protein P8 is used. The diversity of the library is 20⁶ = 6.4xl0⁶. The phage library is incubated at 10¹² per ml at room temperature for 2hrs in

15 the presence of glass beads or microscope slides prepared as described in detail below. After washing (using pH and/or ionic strength in different micro-tubes), the bound phages are eluted using standard procedures (lOOmM HC1 at RT). The eluted phages are neutralized, and used to infect E.coli for titration and amplification. The resulting phage is purified from E.coli lysates by centrifugation, and titered spectrophotometrically (OD268 of

20 1.0 = 5 x 10¹² phages/ml). Following 3 or 4 cycles of biopanning, enrichment is maximal, and clones are sequenced to determine the residues of each of the diverse set of sequences.

Secondary oligopeptide selection:

Based on the sequences selected by the above-described method, a phage display

7 ^J5 library containing oligopeptides with appropriate invariant residues identified during primary selection and additional tyrosines (N-terminal, following the protein signal sequence and poly-Glycine spacer) is constructed by standard molecular biological methods. The vector pGP-FlOO (Paschke et al, 2001, Biotechniques 30: 720-725) is used for this selection step, as it contains following the sequence of the peptide a recognition

^JV sequence for Tev protease. This system allows recovery of the phage even if the peptide is covalently bound to the surface. The biopanning procedure is done as described above with the cross-linking reaction triggered after the last wash. Unreacted phage is removed with lOOmM HC1, and covalently bound phages are recovered by Tev protease digestion, as described in Paschke et al, 2001, Biotechniques 30: 720-725. 3 or 4 reiterations of this

³⁵ procedure typically yields a high affinity tag that can be expressed as a fusion with proteins (such as scFv's) and can be immobilized covalently on the selected surface. Confirmation of polypeptide affinity to surface: The identified and selected oligopeptides are synthesized using conventional solid- state chemistry with the t-Boc or F-moc chemistry. This experiment confirms that the selected oligopeptide sequences possess high affinity to the surface. The oligopeptides are prepared in a fluorescently labeled form, i.e., "capped" with fluoresceine, i.e., modified with fluorescein-5-isothiocyanate, to facilitate the analysis of the covalent reaction 0 efficiency described below. The oligopeptides are purified by a preparative reverse phase HPLC and the structure of the purified products confirmed by Mass Spectrometry and amino acid analysis. The affinity of the oligopeptides to the modified surface is then detennined fluorometrically using conventional Scatchard analysis.

Preparation of fusion protein: 5

The identified oligopetide tags are added to the C-terminus of scFv proteins by standard molecular biological methods. An additional sequence encoding the His(6) is sometimes added to the the 3' end of the encoding DNA to facilitate the purification of the fusion proteins. The scFvs is cloned into the secretion vector pAK400 (Krebber et al. 1997, ^υ J. Immunol Methods 201 : 35-55) directing high periplasmic protein expression, expressed in Escherichia coli SB536 (Bass et al, 1996, J. Bacteriol. 178:1154-1161), and purified over a Ni ⁺-nitrilotriacetic acid column (Qiagen).

Immobilization of fusion protein: 5

The fusion scFvs are covalently immobilized on the modified glass surface prepared as described above by reacting tyrosine residues in the polypeptide tag with 2-phenolic groups on the surface. Several sets of experimental conditions are used. In one experiment, the reaction is initiated by adding H₂O₂ (0.1-lmM) and 0.1 mM hemin catalyst sequentially to the reaction mixture containing the surface and the polypeptides and incubating for 5 0 minutes (Campbell, Kodadek & Brown, 1996, Bioorganic and Medicinal Chemistry, 6, 1301-1307). In another experiment, the same procedure as above is carried out except that 1 mg/mL horse radish peroxidase is used as the catalyst in place of hemin (Malencik & Anderson, 1996, Biochemistry 35, 4375-4386) and the cross-linking reaction is carried out for 30 minutes. In another experiment, photo-oxidation is carried out using 150 W Xe arc 5 lamp with a 280 nm longpass cutoff filter and ruthenium (II) tris-bypyridyl and ammonium persulphate as catalyst and oxidant respectively (Fancy & Kodadek, 1999, Proc. Natl. Acad. Sei. USA 96: 6020-6024). In the latter case, only the modified glass slides are used and the light is applied through a mask containing a hole of 500 x 500 nm for a period of 1 min. After exposure the mask is removed and the glass slide is thoroughly washed with an aqueous buffer containing 10 mM benzoic acid to remove non-covalently bound proteins. The procedure is repeated with the mask positioned at different locations of the surface to generate a pattern of regions containing immobilized scFvs. The pattern is revealed by treating the surface containing array of covalently immobilized scFv fragments with

10 commercially available fluorescently labeled antibodies (Sigma- Aldrich).

Confirmation of uniformity of binding:

The binding characteristics of the model scFv fragments are determined according to an adapted protocol published by Schwesinger et al, 2000, Proc. Natl. Acad. Sei. USA

15 97:9972-9977. The association kinetics for the scFv fragments is measured in solution, on a Ni²⁺-nitrilotriacetic acid column, and on the immobilization matrix before and after covalent immobilization. Off-rates are measured by competitive dissociation assay (Boder et al, 1997, Nature Biotechnol 15:553-557). The uniformity of the binding constants obtained suggests that antibody is bound to the surface in the same, i.e., oriented,

²⁰ conformation.

6.2. EXAMPLE 2: SELECTION OF A TAG THAT FORMS A LIGHT-INDUCED, COVALENT BOND WITH PHENOL GROUPS ON A MODIFIED SILICON MATRIX The following example illustrates certain variations of the methods of the invention _jc immobilizing oriented proteins on a matrix toward the fabrication of protein based biosensors and biochips, by means of covalent attachment of proteins to a surface in a uniform orientation. This example is presented by way of illustration and not by way of limitation to the scope of the invention.

Biosensors have a wide range of applications, primarily in the areas of

30 biological/medical monitoring and environmental sensing. It is now widely acknowledged that future diagnostic methods will require large numbers of individual tests to be run in parallel, and controlled attachment of proteins to the surface is a crucial part of the fabrication of biochips (Vo-Dinh & Cullum, 2000; Emili & Cagney, 2000).

Application of the instant investion provides an accurate, inexpensive means of

35 immobilizing proteins with the combined benefits of oriented and covalent attachment of proteins to the surface of biosensors and biochips.

In the following section, methods of immobilizing a protein oriented on a modified silicon surface by light induced, covalent attachment of a tyrosine containing tag with high affinity for the matrix is described in detail.

In this particular embodiment of the technology a light-activated oxidative protein cross-link methodology is used that introduces very stable covalent C-C bonds between tyrosyl side-chains in close proximity to each other (Fancy, DA & Kodadek, T, 1999). Due to the reaction specificity for tyrosyl side-chains, and the distance dependence of bond formation (Brown, KC et al., 1998), cross-linked proteins maintain their activities and specificities (Malenic DA & Anderson SR, 1994; Lardinois OM. et al, 1999; Kanwar R & Balasubramanian D, 2000).

The formation of di-tyrosine bonds er se is a phenomenon well established in the literature. The results described in this illustrate the use of this chemistry in the fabrication of protein-based biosensors and biochips for a variety of applications.

In the set of experiments described herein, the goal was specifically to:

(1) Develop a matrix to which proteins are immobilized through the fonnation of covalent bonds between tyrosyl side-chains of polypeptides upon light activation; (2) Identify/isolate a polypeptide tag that (a) has high affinity for the matrix, and (b) contains one or more tyrosyl side-chains that can form a covalent bond with the matrix.

(3) Characterize the conditions required for efficient and reproducible cross-linking.

Generation of an immobilization matrix

This task was aimed at preparing surfaces bearing phenol functionality to effect the covalent attachment of oligopeptide tags via the reaction between tyrosine(s) and phenolic residues on the surface of the chip. This entailed the covalent introduction of phenols on the surface of the glass biochip.

Silylation of glass surfaces is a well-established chemistry with numerous published methods and readily available reagents.

As the first step, glass was hydrolyzed under strong basic conditions, using 5M NaOH (lhr, room temperature). After extensive rinses with distilled water, the cover slips were cleaned with piranha solution (70%> sulphuric acid, 30%> hydrogen peroxide) for 30 min., washed and dried. The glass was prepared immediately prior to the silylation reaction. The experiments were carried out using fine glass microscope cover slips 25x25 mm (Fisher Scientific; catalogue No 12-548C). 3-Aminopropyltriethoxysilane (APTES) was used to introduce amino groups on the surface, that were then reacted with 3-(4-Hydroxyphenoxy) propionic acid N-hydroxy succinimide ester to yield the desired phenols on the surface of the chip, as schematically illustrated in Figure 4.

The silylation reaction was canied in aqueous methanol (95%> MetOH, 4%> H O, ImM CH₃COOH and 1.0% APTES) for 30 min. at room temperature as described by

Cunliffe et al (2000). The modified glass was extensively washed with MetOH, cured in hot oven over a period of several hours and stored under dry conditions prior to further use.

The introduction of phenol groups was achieved by reacting the amino glass with 3-(4-

Hydroxyphenoxy) propionic acid N-hydroxy succinimide ester. Tins reaction was carried 5 out for 3 hrs. at room temperature using 50 mM reagent in 0.2 M NaHCO₃ at pH 8.5. Also, at a later stage of the investigation an alternative silalation protocol involving the modification of glass in dry toluene at 2.0% APTES was used as a control (Falsey et al,

2001). The amino glass obtained by this method had very similar properties to that r, modified in methanol (see below).

The concentration of amino groups on the surface, and the degree of their modification, was determined in a binding assay of fluorescein to APTES-modified glass as described by Tiller et al (2002). In brief, the modified glass slides were immersed in an aqueous solution of fluorescein sodium salt (1%> w/v) and incubated for 5 min. at room 5 temperature with stirring. The glass was then washed with distilled water and the bound fluorescein was eluted with a solution of cetyltrimethylammonium bromide (0.1%). The eluate was diluted with 200mM phosphate buffer, pH 8.0, and the concentration of fluorescein was determined using a standard calibration curve on a Hitachi F3010 Fluorescence Spectrometer (excitation at 490 nm, emission at 525 nm). Typically, the NH -group content was determined as 8.0-9.0x10^"10 mol/cm^"2 prior to the modification with 3-(4-Hydroxyphenoxy) propionic acid N-hydroxy succinimide ester. This value dropped to 1.2-1.5xl0^"10 mol/cm^"2 after the introduction of phenols, providing evidence for 85%o modification. The achieved density of NH₂-groups on the surface of one per 30-35A² falls well into the range reported in the literature (see, for example, Akkoyun and Bilitewski, 2002, and references cited above).

Thus, the phenol-modified surfaces were prepared for the next phase of experimentation using the protocols available in the literature. The surface density of phenolic residues on the surface was calculated as 1 per about 40-45 A² on the basis of the number available amino groups and the extent of their modification (85 >) with 3-(4- Hydroxyphenoxy) propionic acid N-hydroxy succinimide ester.

Isolation of high affinity cross-linkable polypeptide tag

, _π The objective of this set of experiments was to identify short oligopeptide tags with high affinity to the phenol-bearing surface. These oligopeptides should contain at least one tyrosine residue per molecule to enable the covalent cross-linking to the surface phenol of the tagged protem after its non-covalent orientation on the surface of the biochip.

To achieve this goal the phage display selection system, where oligopeptides libraries

15 are displayed on the surface of the filamentous phage M13 as fusioned with the major coat protein P8, has been employed. These libraries are produced commercially by New England Biolabs (MA) with heptamers and dodecamers (7 and 12 amino acid residues, respectively) available from the company. Initially, both libraries have been used in the biopanning experiments described below.

20 Glass slides were placed in 8 well cell culture microplates (Fisher Scientific) and washed 5 times with 2.0 ml of 50 TBS buffer (50 mM Tris, 150mM NaCl, pH 7.5). Phage libraries (10¹⁰ cpu) were added to a well with a single glass panel containing 1.2 ml of TBS buffer and 0.1 % Tween 20 (TBST), and the plates were incubated for 60 min at room temperature with shaking. The plates were then washed with TBST (10x2.0 ml) and the phage was eluted with 1.0 M Glycine, pH 2.2. These elution conditions are recommended by the manufacturer for elution in cases where high affinity ligands are unavailable for competitive elution. The resulting eluent was neutralized with Tris and the phage was amplified in E. coli, purified by precipitation with PEG and titred, as described in the

30 manual.

Typically 6 rounds of biopanning were carried for each set of experimental conditions. In each case, the phage was plated on agarose after the 3^rd, 4^th, 5^th and 6^th rounds, and individual colonies were picked and amplified. The phage DNA was recovered from the amplified clones using standard procedures and sequenced. Overall in excess of 300 clones

35 from numerous experiments were sequenced during this phase of the investigation. The following observations were made: (1) No significant differences were observed with the APTES glass modified using the two protocols described above (modification in methane vs. toluene, see page three);

(2) No consensus sequences were obtained with the library of heptamers;

(3) Similar sequences were obtained when the 12-mer library was eluted with glycine, pH 2.2 and 50 mM 3-Hydroxyphenol propionic acid in TBS; (4) Most sequences obtained after the 3^rd and 4^th round of biopanning (typically no consensus at this stage) contained an extraordinary high proportion of His, Pro, and Ser amino acid residues;

(5) The consensus sequences were repeatedly obtained with the 12-mer library and, in one experiments the consensus collapsed to a single.

Table 1: Consensus sequences obtained in rounds 5 and 6 of biopanning experiments (12 mers)

Underlined sequences were observed in multiple copies in several independent experiments. Sequences in bold/italics were selected for synthesis, see text.

Examination of the sequences obtained revealed a motif with an extraordinary number of histidines arranged in the sequence HIHHH, with numerous variations, such as ₅ HHTHHH, HLHHTH, HHTHAH, HHHSH, HHIHI, HHHKHH, and others. Furthermore, they revealed that the histidine-rich motif is equally effective at both the N- and C-terminus of the peptide - in either orientation. The remaining 5-6 amino acids were found to be predominantly hydrogen bonders, such as serine and arginine. The sequences contained a surprisingly low frequency of tyrosine, calculated as 40%> of what would be expected for 10 random incorporation. It was believed that the phenols, that are strong hydrogen bonders, in the context of the residual hydroxyls and silica - the underlying structural feature of the glass - drove the selection toward structured hydrophilic peptides with a high propensity to form hydrogen bonds.

To confirm the validity of this observation, we carried out a number of additional

15 control experiments. Biopanning experiments were repeated in the presence of ImM

EDTA to eliminate the possibility of some participation of metal ions in the strong binding observed (note that metal binding poly-His tags constitute one of the most popular affinity purification systems for recombinant proteins on a laboratory scale). The silylation reaction _jn. was carried out under different conditions and the resulting glass was used for the modification and biopanning. Finally, biopanning was repeated using glass slides cleaned with piranha solution, hi all cases, the histidine-rich motif was obtained after the 5^th and 6^th rounds of biopanning.

The high affinity of the resulting peptides to the phenol-modified surface and the

25 well defined histidine-rich domain linked to a cluster of mainly hydrogen bonding amino acids, enabled introduction of tyrosines at the stage of solid state peptide synthesis (see below).

Preliminary binding experiments carried out at this stage confirmed that these peptides have high affinity to the surface, and in conclusion, the key objective of selecting

30 polypeptide sequences with high affinity to the phenol-modified surface was achieved.

Light-mediated Cross-linking of the Polypeptide Tags to the Phenol-Modified Glass Surface

35 The main goal of this last part of the experimental program was the preparation of labeled polypeptides and the characterization of the conditions required for their efficient and reproducible cross-linking to the surface of the chip. Given that in preliminary experiments, the oligopeptides showed strong binding to the surface, the diversity of the sequences shown in Table 1 was judged to be sufficient for the design and synthesis of a meaningful matrix of labeled tag candidates. Sequence 1, (KSLSRHDHIHHH), 11 (HHTHYSPHRGTP), 19 (HHffllYKFTESV) and 26 (IHQHHFPLWPYP) were selected for the following reasons.

Sequence 1 was obtained the most frequently in biopanning experiments. Although 0 it does not contain a tyrosine residue, the degree of variation in the amino acid compositions of related sequences obtained allowed us to assume that a minor variation would have no dramatic effect on the binding properties, provided that the key motif is preserved. Because the tripeptide sequence TTY was frequently obtained in earlier rounds of biopanning (e.g. the 3^rd), it was used to replace the N-terminal sequence KSL. ⁵ Sequences 11 and 19 (HHTHYSPHRGTP, HHffllYKFTESV) both contain a tyrosine residue in the middle of the sequence.

Sequence 26 (IHQHHFPLWPYP) contains a tyrosine residue near the C-terminus. Thus, the designed TTY-containing (Sequence 1 hybrid: TTYSRHDHHTHH (SEQ _n ID NO:28)), together with sequences 11, 19, and 26, together provided a good initial matrix, also because in this selection tyrosines are placed on both sides of, and at various distances from, the conserved histidine-rich motif.

The oligopeptides in the Ml 3 phage library are linked to the coat protein through a spacer comprising three glycine residues. These were preserved in the oligopeptides where 5 the labeled amino acid was introduced at the C-terminus. The peptides were synthesized using the conventional Merrifield methods, and HPLC purified after cleavage from the resin and deprotection. The sequences were confirmed by MS using a Ciphergen ProteinChip® system, as well as by amino acid sequencing, in some cases. Three types of labeling approaches were attempted: Fluorescein and biotin. 0 To quantitate binding of the peptides, direct fluorometnc assays were employed for the two former labels, and binding of fluorescein-labeled avidin to biotin was assayed for the latter. Also, a competitive binding assay was implemented whereby inhibition of adsorption of fluorescently labeled protem to phenol-modified glass was determined by pre- treatment of the glass with the peptide of interest. In particular, this assay was used in determining the degree of the peptide cross-linking to the surface (see below)

Binding experiments were carried out in the same eight-well plates used for the biopanning experiments. The glass was incubated with oligopeptides in 1.2 ml of TBS, pH 7.5 for 1 hr. at room temperature with shaking. The panels were then taken out of the wells, small amounts of residual buffer was wiped off with filter paper. The peptide was eluted by placing the glass in a 0.25% Tween 20 TBS, pH 7.5 buffer for 30 min, as it was shown in control experiments that this amount of Tween 20 prevented/disrupted peptide binding. Fluorescence in the eluent was quantified using the excitation wavelength of 490 and 545 and emission at 525 and 577 for the fluorescein- and rhodamin-labels, respectively. The following observations were made:

(1) All the peptides showed appreciable binding to the phenol-modified glass when added to the incubation media at a concentration of 10^" -10^" M.

(2) The sensitivity of Rhodamine-based assay proved to be insufficient for accurate analysis in this concentration range.

(3) In some cases, the amount of bound peptide kept increasing with increasing concentration of the peptide in solution, not showing well-defined saturation, and suggesting the formation of multi-layers on the surface of the chip. This was not surprising given the propensity of the oligopeptides to hydrogen bond and should not be a problem when the tag is covalently attached to target proteins.

Figure 5 depicts a typical binding curve obtained with TTYSRHDHIHHH. In this case, peptide binding clearly shows saturation. Computer fitting of the experimental points using commercially available software resulted in a calculated Kd of 4.45 +/- 0.52 x 10^"8, well within a targeted 10 nM range., and thererfore no optimization of binding parameters was undertaken. Binding could, however, still probably be enhanced further (while still suppressing or eliminating aggregation), for example, by adjusting the pH and salt composition of the media. Thus, a high affinity peptide tag was identified. Subsequently, cross-linking of the target oligopeptides to the surface was investigated. Fluorescine labeled material prepared for the binding studies proved to be unusable in the investigation of cross-linking, as fluorescein was found to loose sensitivity upon cross-linking.

Therefore, biotin-labeled peptides were used for these experiments. The peptide was cross-linked to the phenol-modified glass in eight- well plates, essentially according to the procedure described by Fancy & Kodadek (1999). Ruthenium (II) tris-bypyridyl complex (125 μM, Aldrich) was used as a catalyst in the presence of 2.5 mM ammonium persulfate under inadiation for 3 min. The reaction was quenched; glass washed with water, 70%> ethanol, DNF & methanol, dried, and subsequently analyzed, either by the above displacement assay, or by using direct binding of flurescein-labeled avidin to the surface. As is evident from Table 2 below, both methods yielded comparable results with the degree of cross-linking in line with the 80%> target value. Expectedly, some background fluorescence was observed in the absence of the reagents.

Table 2: Percent crosslinking of TTYSRHDHLHHH to the phenol-modified glass.

Crosslinking Conditions Ru , hv H₂Q₂, MnTPPS w/o peptide 7% 7% w/o oxidant 31% 11% w/o catalyst/light 33% nd Full set of reagents 67% 79%

Thus, also a high degree of cross-linking of the high affinity tag was achieved.

References:

Akkoyun, A, Bilitewski, U (2002) Optimisation of glass surfaces for optical immunosensors. Biosensors Bioelectronics 17: 655-664.

Brown, K.C., Yu, Z., Burlingame, A.L., Craik, C.S. (1998) Determining protein-protein interactions by oxidative cross-linking of glycine-glycine-histidine fusion protein. Biochemistry 37: 4397-4406. Cunliffe, D., Smart, CA, Alexander, C. Vulfson, E.N. (1999) Bacterial adhesion at synthetic surfaces. Appl Env. Microb. 65: 4995-5002.

Fancy, D.A., Kodadek, T. (1999) Chemistry for the analysis of protein-protein interactions: Rapid and efficient cross-linking triggered by long wavelength light. Proc. Natl. Acad. Sei. USA 96: 6020-6024.

Kanwar, R., Balasubramanian, D. (2000) Structural studies of some di-tyrosine cross- linked globular proteins. Biochemistry 39: 14976-14983.

Lardinois, O.M., Medzihradszky, K.F., Ortiz de Montellano, P.R. (1999). Spin trapping and protem cross-linking of the lactoperoxidase protein radical. J. Biol. Chem. 274: 35441-35448.

Malenic, D.A., Andersson, S.R. (1994) Dityrosine formation in calmodulin: conditions for intennolecular cross-linking. Biochemistry ' 33: 13363-13376. Tiller, J.C, Liao, C J., Lewis, K., Klibanov, A.M. (2002). Designing surfaces that kill bacteria on contact. Proc. Natl. Acad. Sei. USA 98:5981-5985.

Falsey, J.R., Renil, M., Park, S., Li, S., Lam, K.S. (2001) Peptide and small molecule array for high throughput cell adhesion and functional assays: Bioconjugate Chem. 12: 346-353.

7. REFERENCES CITED All references cited herein are incorporated herein by reference in their entirety and for all purposes to the same extent as if each individual publication or patent or patent application was specifically and individually indicated to be incorporated by reference in its entirety for all purposes.

Many modifications and variations of the present invention can be made without departing from its spirit and scope, as will be apparent to those skilled in the art. The specific embodiments described herein are offered by way of example only, and the invention is to be limited only by the terms of the appended claims along with the full scope of equivalents to which such claims are entitled.

Claims

WHAT IS CLAIMED IS:

1. A method for immobilizing a polypeptide on a support, wherein said polypeptide comprises at least one tyrosine residue, and wherein said support comprises at least one substituted phenolic group, said method comprising cross-linking said polypeptide at said tyrosine residue to said substituted phenolic group on said support.

2. The method of claim 1, further comprising a step of introducing said at least one tyrosine residue into said polypeptide.

3. The method of claim 1, wherein said substituted phenolic group is selected from the group consisting of o-alkyl, o-amino, o-alkoxy-, o-dialkylamino, ?-alkyl,p-amino, -alkoxy-, -dialkylamino phenolic groups.

4. The method of claim 1, wherein said cross-linking is ca ied out by a method comprising inadiating said support in the presence of said polypeptide with light of a wavelength in the range of 190-750 mn.

5. The method of claim 1, 2, 3 or 4, wherein said support is a planar surface.

6. The method of claim 1, 2, 3 or 4, wherein said support is a bead.

7. A method for producing an array of immobilized polypeptides, said array comprising a support and a plurality of different polypeptides, wherein each said polypeptide comprises at least one tyrosine residue, wherein said support comprises a plurality of substituted phenolic groups, and wherein each said different polypeptide is attached to said support at a different predetermined location, said method comprising cross-linking each said polypeptide at said tyrosine residue to one of said plurality of substituted phenolic groups at said predetermined location on said support.

8. The method of claim 7, further comprising a step of introducing said at least one tyrosine residue into one or more of said plurality of different polypeptides.

9. The method of claim 7, wherein said substituted phenolic group is selected from the group consisting of o-alkyl, o-amino, o-alkoxy-, o-dialkylamino, -alkyl, j3-amino, -alkoxy-, j9-dialkylamino phenolic groups.

10. The method of claim 6, wherein said cross-linking is carried out by a method comprising inadiating said support in the presence of said polypeptide with light of a wavelength in the range of 190-750 nm.

11. The method of claim 7, 8, 9, or 10, wherein said support is a planar surface.

12. The method of claim 7, wherein said anay comprises in the range of 100 to 1,000 different polypeptides per 1 cm .

13. The method of claim 7, wherein said anay comprises in the range of 1,000 to

10,000 different polypeptides per 1 cm².

14. The method of claim 7, wherein said array comprises in the range of 10,000 to

50,000 different polypeptides per 1 cm².

15. The method of claim 7, wherein said anay comprises more than 50,000 different polypeptides per 1 cm².

16. The method of claim 7, wherein said array comprises a polypeptide consisting of at least 5 residues.

17. The method of claim 16, wherein said array comprises a polypeptide consisting of at least 10 residues.

18. The method of claim 17, wherein said anay comprises a polypeptide consisting of at least 25 residues.

19. The method of claim 18, wherein said anay comprises a polypeptide consisting of at least 50 residues.

20. The method of claim 19, wherein said anay comprises a polypeptide consisting of at least 100 residues.

21. The method of claim 20, wherein said anay comprises a polypeptide consisting of at least 150 residues.

22. The method of claim 21, wherein said anay comprises a polypeptide consisting of at least 200 residues.

23. An array of immobilized polypeptides, said anay comprising a support and a plurality of different polypeptides, wherein each said polypeptide comprises at least one tyrosine residue, and wherein said support comprises a plurality of substituted phenolic groups, and wherein each said polypeptide is cross-linked at said tyrosine residue to one of said plurality of substituted phenolic groups at a predetermined location on said support.

24. The array of claim 24, wherein said substituted phenolic group is selected from the group consisting of o-alkyl, o-amino, o-alkoxy-, o-dialkylamino, 7-alkyl, p-a ino, p- alkoxy-, -dialkylamino phenolic groups.

25. The array of claim 23 or 24, wherein said support is a planar surface.

26. The anay of claim 23, wherein said array comprises in the range of 100 to 1,000 different polypeptides per 1 cm².

27. The array of claim 23, wherein said array comprises in the range of 1,000 to 10,000 different polypeptides per 1 cm².

28. The array of claim 23, wherein said array comprises in the range of 10,000 to 50,000 different polypeptides per 1 cm .

29. The anay of claim 23, wherein said array comprises more than 50,000 different polypeptides per 1 cm .

/Λ 30. The anay of claim 23, wherein said anay comprises a polypeptide consisting of at least 5 residues.

31. The anay of claim 23, wherein said anay comprises a polypeptide consisting of at least 10 residues.

5 32. The array of claim 31 , wherein said anay comprises a polypeptide consisting of at least 25 residues.

33. The array of claim 32, wherein said anay comprises a polypeptide consisting of at least 50 residues.

34. The anay of claim 33, wherein said anay comprises a polypeptide consisting of at least 100 residues.

35. The array of claim 34, wherein said array comprises a polypeptide consisting of at least 150 residues.

5 36. The array of claim 35, wherein said array comprises a polypeptide consisting of at least 200 residues.

37. A surface comprising a plurality of substituted phenolic groups.

38. A surface comprising a plurality of substituted phenolic groups, each cross-linked to a tyrosine residue of a polypeptide.

39. The surface of claim 37 or 38, wherein said substituted phenolic group is selected from the group consisting of o-alkyl, o-amino, o-alkoxy-, o-dialkylamino, ?-alkyl, /?-amino, _jP-alkoxy-, -dialkylamino phenolic groups.

40. A method for immobilizing a polypeptide on a support, wherein said polypeptide is covalently attached to a molecular moiety having at least a given affinity for binding to said support and comprising at least one first functional moiety, and wherein said support comprises at least one second functional moiety, wherein said second functional moiety forms one or more covalent bonds with said first functional moiety under suitable conditions, said method comprising cross-linking said polypeptide at said first functional moiety to said second functional moiety on said support, wherein said cross-linking is carried out by a method comprising irradiating said support in the presence of said polypeptide with light of a wavelength in the range of 190-750 mn.

41. The method of claim 40, wherein said molecular moiety is a peptide.

42. The method of claim 40 or 41, wherein said support is a planar surface.

43. The method of claim 40 or 41, wherein said support is a bead.

44. The method of claim 41, wherein said peptide comprises 5-20 amino acids.

45. The method of claim 41, wherein said first functional moiety is a tyrosine residue and said second functional moiety is a substituted phenolic group.

46. A method for producing an anay of immobilized polypeptides, said array comprising a support and a plurality of different polypeptides, each said different polypeptide being attached to said support at a different predetermined location, wherein each said polypeptide is covalently attached to a molecular moiety having at least a given affinity for binding to said support and comprising at least one first functional moiety, and wherein said support comprises a plurality of second functional moieties, wherein said second functional moieties forms one or more covalent bonds with said first functional moiety under suitable conditions, said method comprising cross-linking each said polypeptide at said first functional moiety to one of said plurality of said second functional moieties on said support, wherein said cross-linking is carried out by a method comprising inadiating said support in the presence of said polypeptide with light of a wavelength in the range of 190-750 nm.

47. The method of claim 46, wherein said molecular moiety is a peptide.

5

48. The method of claim 46 or 47, wherein said support is a planar surface.

49. The method of claim 47, wherein said peptide comprises 5-20 amino acids.

50. The method of claim 47 or 49, wherein said first functional moiety is a tyrosine 10 residue and said second functional moiety is a substituted phenolic group.

51. The method of claim 46, wherein said anay comprises in the range of 100 to 1,000 different polypeptides per 1 cm .

52. The method of claim 46, wherein said anay comprises in the range of 1,000 to 15 10,000 different polypeptides per 1 cm².

53. The method of claim 46, wherein said array comprises in the range of 10,000 to 50,000 different polypeptides per 1 cm².

54. The method of claim 46, wherein said anay comprises more than 50,000 different polypeptides per 1 cm².

55. The method of claim 46, wherein said anay comprises a polypeptide consisting of at least 5 residues.

56. The method of claim 55, wherein said array comprises a polypeptide consisting of at least 10 residues.

57. The method of claim 56, wherein said anay comprises a polypeptide consisting of at least 25 residues.

30 58. The method of claim 57, wherein said anay comprises a polypeptide consisting of at least 50 residues.

59. The method of claim 58, wherein said anay comprises a polypeptide consisting of at least 100 residues.

35 60. The method of claim 59, wherein said array comprises a polypeptide consisting of at least 150 residues.

61. The method of claim 60, wherein said anay comprises a polypeptide consisting of at least 200 residues.

62. An array of immobilized polypeptides, said anay comprising a support and a plurality of different polypeptides, wherein said support comprises a plurality of phenolic groups, and wherein each said polypeptide is covalently attached to a predetermined location on said support via a molecular moiety having at least a given affinity for binding to said support and comprising at least one first functional moiety, said first functional moiety being covalently attached to one of said phenolic groups on said support.

63. The array of claim 62, wherein said molecular moiety is a peptide.

64. The array of claim 62 or 63, wherein said support is a planar surface.

65. The method of claim 63, wherein said peptide comprises 5-20 amino acids.

66. The method of claim 63 or 65, wherein said first functional moiety is a tyrosine residue.

67. The array of claim 62, wherein said anay comprises in the range of 100 to 1,000 different polypeptides per 1 cm².

68. The array of claim 62, wherein said anay comprises in the range of 1,000 to

10,000 different polypeptides per 1 cm².

69. The array of claim 62, wherein said array comprises in the range of 10,000 to 50,000 different polypeptides per 1 cm².

70. The anay of claim 62, wherein said anay comprises more than 50,000 different polypeptides per 1 cm².

71. The anay of claim 62, wherein said anay comprises a polypeptide consisting of at least 5 residues.

72. The anay of claim 71, wherein said anay comprises a polypeptide consisting of at least 10 residues.

73. The anay of claim 72, wherein said anay comprises a polypeptide consisting of at least 25 residues.

74. The anay of claim 73, wherein said anay comprises a polypeptide consisting of at least 50 residues.

75. The anay of claim 74, wherein said array comprises a polypeptide consisting of at least 100 residues.

76. The array of claim 75, wherein said anay comprises a polypeptide consisting of at least 150 residues.

77. The array of claim 76, wherein said array comprises a polypeptide consisting of at least 200 residues.

78. A method for immobilizing a polypeptide on a support comprising at least one surface functional moiety, comprising

(a) selecting a molecular moiety having at least a given affinity for binding to said support and comprising at least one first functional moiety, wherein said first functional moiety forms one or more covalent bonds with said surface functional moiety under suitable conditions;

(b) attaching covalently said molecular moiety to said polypeptide; and

(c) cross-linking said polypeptide at said first functional moiety to said surface functional moiety on said support by a method comprising inadiating said support in the presence of said polypeptide with light of a wavelength in the range of 190-750 nm.

79. The method of claim 78, wherein said molecular moiety is a peptide.

80. The method of claim 79, wherein said peptide is selected from a random peptide library.

81. The method of claim 78, 79, or 80, wherein said support is a planar surface.

82. The method of claim 78, 79, or 80, wherein said support is a bead.

83. The method of claim 79 or 80, wherein said peptide comprises 5-20 amino acids.

84. The method of claim 79 or 80, wherein said first functional moiety is a tyrosine residue and said surface functional moiety is a substituted phenolic group.

85. A method for producing an anay of immobilized polypeptides, said anay comprising a support and a plurality of different polypeptides, said support comprises a plurality of surface functional moieties, each said different polypeptide being attached to said support at a different predetermined location, said method comprising (a) selecting one or more molecular moieties having at least a given affinity for binding to said support and comprising at least one first functional moiety, wherein said first functional moiety forms one or more covalent bonds with said surface functional moiety under suitable conditions;

(b) attaching covalently one of said one or more molecular moieties to each of said ^•plurality of polypeptides; and (c) cross-linking each said polypeptide at said first functional moiety to said surface functional moiety at a predetermined location on said support by a method comprising inadiating said support in the presence of each said polypeptide with light of a wavelength in the range of 190-750 nm.

86. The method of claim 85, wherein said one or more molecular moieties are peptides.

87. The method of claim 86, wherein said peptides are selected from a random peptide library.

88. The method of claim 85, 86, or 87, wherein said support is a planar surface.

89. The method of claim 85, wherein said peptide comprises 5-20 amino acids.

90. The method of claim 86 or 87, wherein said first functional moiety is a tyrosine residue and said surface functional moiety is a substituted phenolic group.

91. The method of claim 85, wherein said array comprises in the range of 100 to

1,000 different polypeptides per 1 cm².

92. The method of claim 85, wherein said array comprises in the range of 1,000 to 10,000 different polypeptides per 1 cm².

93. The method of claim 85, wherein said anay comprises in the range of 10,000 to

50,000 different polypeptides per 1 cm².

94. The method of claim 85, wherein said anay comprises more than 50,000 different polypeptides per 1 cm².

95. The method of claim 85, wherein said anay comprises a polypeptide consisting of at least 5 residues.

96. The method of claim 95, wherein said array comprises a polypeptide consisting of at least 10 residues.

97. The method of claim 96, wherein said array comprises a polypeptide consisting of at least 25 residues.

98. The method of claim 97, wherein said array comprises a polypeptide consisting of 0 ^υ at least 50 residues.

99. The method of claim 98, wherein said array comprises a polypeptide consisting of at least 100 residues.

100. The method of claim 99, wherein said anay comprises a polypeptide consisting of at least 150 residues.

101. The method of claim 100, wherein said anay comprises a polypeptide consisting of at least 200 residues.

0

5

0

5