WO2013137737A1 - Methods of creating and screening of dna encoded combinatorial libraries of chemical antibodies - Google Patents

Abstract

Method for the screening and identification of a binding region interacting with a target molecule comprising providing a library of binding regions, pooling the binding regions in pools wherein each binding region in a pool is connected with a pool-specific sequence identifier to yield a molecule, contacting the binding regions in the pools with a target molecule, identifying and selecting the pools that contain binding regions that interact with the target molecule using the pool specific identifier to thereby identify a subset of binding regions that form a successive library, generating the successive library and repeating the steps of pooling, contacting, identifying and selecting to generate further successive libraries of decreasing numbers of binding regions at least once until the individual binding region that interacts with the target molecule is identified.

Description

Title: Methods of creating and screening of DNA encoded combinatorial libraries of chemical antibodies

Technical Field

[01] The present invention relates to the generation of libraries of aptamers and the identification of aptamers using combinatorial libraries, pooling and high throughput sequencing.

Background Art

Aptamers:

[02] The term aptamer is derived from the Latin word "aptus"— which means fitting (Ellington and Szostak, 1990) and the Greek word "meros" meaning particle. Aptamers, also called chemical antibodies, are nucleic acids or peptide molecules that bind targets with an affinity and specificity that can rival antibody-antigen interactions. Nucleic acid aptamers are short single-stranded nucleic acid oligomers (ssDNA or RNA) with a specific and complex three-dimensional shape characterized by stems, loops, bulges, hairpins, pseudoknots, triplexes, or quadruplexes. Based on their three-dimensional structures, aptamers can well-fittingly bind to a wide variety of targets from single molecules to complex target mixtures or whole organisms.

[03] Binding of the aptamer to the target results from structure compatibility, stacking of aromatic rings, electrostatic and vanderWaals interactions, and hydrogen bondings, or from a combination of these effects (Hermann and Patel, 2000). The first aptamers developed consisted of unmodified RNA (Ellington and Szostak, 1990; Tuerk and Gold, 1990). Later on, single-stranded DNA aptamers for different targets were described (Ellington and Szostak, 1992) as well as aptamers containing chemically modified nucleotides (Green et al., 1995). Chemical modifications can introduce new features into the aptamers, improve their binding capabilities or enhance their stability (Gold et al., 1995). However due to the requirements of enzymatic amplification during the evolutionary selection process there are strict limitations which modifications are actually suitable (Keefe et al., Current Opinion in Chemical Biology 2008, 12:448-456; ).

[04] The main advantage of aptamers is their in vitro selection process, whereas the production of antibodies uses biological systems. To produce antibodies, the induction of an immune response is necessary. Therefore this procedure is limited to target proteins that do not have a similar structure or sequence to endogenous proteins and such process cannot be applied for toxic compound targets that would kill the animal. By isolating aptamers in vitro, an aptamer can be produced virtually for any target molecule, including inorganic and small organic molecules, peptides, proteins, carbohydrates, antibiotics, as well as complex targets like target mixtures or whole cells and organisms. Several publications summarized the various target molecules according to certain aspects (Famulok, 1999; Wilson and Szostak, 1999; Goeringer et al., 2003; Klussmann, 2006).

[05] Another complication for in vivo production of antibodies is that the antibodies can only work under physiological conditions. This restricts the range of application and function of antibodies. Aptamers on the other hand can be optimized for any conditions, they can also be manipulated to bind different region of the target in different conditions. Also, aptamers are more stable at high temperature and they can be regenerated easily after denaturation. Aptamers are in general more stable than antibodies, so they have a longer shelf life.

[06] The quality of aptamers is more consistent than antibodies because they are synthesized chemically and then purified. Also, through chemical modifications the kinetic parameters such as on or off rates can be changed for aptamers, and cannot be accomplished with antibodies. Finally, the labels on antibodies can cause them to loss their affinity to their target molecules. However, the position of these labelling molecules on aptamers can be easily changed to positions where binding is not affected.

[07] Chemical modification of natural aptamers is useful to protect against enzymatic degradation (Scherr et al., Bioorg.& Med. Chem. Lett. 7 (13) 1791-1796, 1997) an important requirement for a binding molecule to serve in a biological environment.

SELEX:

[08] Combinatorial chemistry is an important technology for industry as well as biotechnological and pharmaceutical research to discover new materials or molecules with desirable properties, new drugs, and catalysts. It is characterized by the synthesis and simultaneous screening of large libraries of related, but structurally distinct compounds to identify and isolate functional molecules.

[09] Nucleic acids are very attractive compounds for combinatorial chemistry, because they are able to fold into defined secondary and tertiary structures, and they can be amplified by PCR or in vitro transcription easily. Very complex libraries of random sequence oligonucleotides with about 10¹⁵ different molecules can be produced by chemical synthesis and screened in parallel for a particular functionality, such as high-affinity ligand-binding (aptamers) or catalytic activity (ribozymes, DNAzymes). In 1990, a new selection process was described, called "Systematic Evolution of Ligands by Exponential enrichment (SELEX)", Tuerk and Gold, 1990). Ellington and Szostak independently used a similar selection procedure to isolate - from a random sequence RNA library - RNA molecules, with the ability to fold into a stable three-dimensional structure, thus creating a specific binding site for small ligands. They named these selected, individual RNA sequences 'aptamers' (Ellington and Szostak, 1990). Two years later, the successful selection of single stranded DNA sequences from a chemically synthesized pool of random sequence DNA molecules could be shown (Ellington and Szostak, 1992). Since this early phase of the SELEX technology it became an important and widely used tool in molecular biological,

pharmaceutical, and medical research. Additionally, this technique often was modified (Stoltenburg et al., Biomolecular Engineering 24 (2007) 381-403; Wang et al., Chin J Anal Chem, 2009, 37(3), 454-460) to select aptamers for different applications.

[10] In the classic SELEX starting libraries have relatively long oligomers of DNA/ RNA sequences (80-120 nt) with central randomized regions (30-70 nt). These are sparsely sampled libraries with a probability of ~ 10^"4 that any particular sequence occurs in a typical

29

starting pool for a randomized 30mer, and of -10^" with randomized 70mers. This means that such SELEX experiments begin with single copies of those sequences that are present by random chance. Evolution occurs via the selective pressure of binding to the target followed by amplification of the survivors; selection and amplification are repeated in typically 5-20 rounds. "Winners" are found by cloning and sequencing, after which a minimal core binding sequence is sought by truncating segments of the parent aptamer that are not needed for the interaction with the target (WO201 1/109451).

[11] Despite the wide adoption of the SELEX procedure for the discovery of DNA/RNA aptamers, only a few hundred target-specific aptamers have been discovered to date using this method compared with the discovery of thousands of antibodies during the same period. This limited success may stem primarily from a significant number of drawbacks with the SELEX selection method itself. First, the chemical space of possible sequences in SELEX experiments (e.g., 10¹⁸ for a 30 nucleotide random stretch), is so large that direct synthesis and screening of all sequences is impossible, even given the high-throughput advancements made in DNA/RNA synthesizer instrumentation. Second, even when SELEX identifies nucleic acid sequences with extremely high affinity for the target, these sequences are generally relatively long (typically 80-120 monomer units in length), and often have complex internal structures (secondary structures). Such long folded molecules are often

disadvantageous for a variety of applications, where cost and ease of production are mandatory. Third, the SELEX methodology of repeated rounds of selection and amplification are cumbersome, time-consuming and expensive. Finally, once an RNA or DNA aptamer motif is identified there is the need of "post-SELEX" optimization (Keefe, Current Opinion in Chemical Biology 2008, 12:448-456) to overcome the susceptibility of 'natural' aptamers to degradation by nucleases. This quality is especially disadvantageous when developing RNA aptamers towards therapeutic applications, which requires injection into biological fluids. In order to circumvent this challenge, aptamers have been generated from libraries with oligonucleotides with various modified backbones. Specifically, modifications at the 2' position of the ribose sugar in pyrimidine nucleotides by substituting with amino (-NH₂) and fluoro (-F) groups have been shown to provide protection from RNA nucleases. However, as these modifications have to be compatible with enzymes in the SELEX process (Lauridsen et al., Chembiochem 13, 19-25, 2012). there is a clear limitation what modifications can be implemented to stabilize an Aptamer against enzymatic digestion.

[12] A typical Post-SELEX Modification Processes involves several steps:

Minimization. The initial aptamer sequences isolated by SELEX are typically 70 to 80 nucleotides long. Aptamers of this length would be difficult and expensive using current manufacturing techniques, and production yields would be low. Therefore it is necessary to identify the active site or core of the aptamer and remove unnecessary nucleotides from the molecule. Successful minimization should yield aptamers to be between 20 and 40 nucleotides in length without compromising the affinity, specificity or functional activity of the aptamer for the target of interest.

Optimization. Once an aptamer of appropriate size is identified its affinity, functional activity and metabolic stability can be adjusted. By modifying the sequence and introduction of chemical modifications these variant aptamers have to be compared to each other and to the starting aptamer in order to determine which modifications improve affinity, functional activity or both.

Nuclease resistance. If not chemically altered, aptamers composed of unmodified nucleotides may be rapidly degraded, or metabolized, by enzymes which are naturally present in the blood and tissues. These enzymes, known as nucleases, bind to and metabolize the aptamer. While rapid drug clearance and a short duration of action are desirable for some clinical applications, a prolonged duration of action is necessary for other disease categories. Once the specific sites within an aptamer that are most susceptible to nuclease metabolism are identified, site-specific stabilizing substitutions into the aptamer can be introduced to achieve nuclease resistance.

PEGylation. Duration of action is often correlated to how long the aptamer remains in the body. Because aptamers are small in size, they may be naturally excreted before they have achieved their intended therapeutic effect. To slow the rate of excretion from the body, the size of the aptamer can be increased by attaching it to another molecule known as polyethylene glycol, or PEG, to create a larger molecule. This process is known as

PEGylation. Desired duration of action can be achieved by using different sizes, structures and attachment locations of PEG molecules. Once the aptamer is pegylated, it has to be tested again to determine whether the desired duration of action was achieved.

[13] Through this combination of SELEX and post-SELEX modification processes it is possible to design and select desired properties of an aptamer that address the proposed therapeutic indication.

[14] In summary the classic SELEX and Post-SELEX modification process appears to be rather lengthy and expensive. Therefore a low cost process, which allows the direct identification of already highly modified nucleic acid sequences with the required properties sequences out of a large library of binding candidates is highly desired.

[15] Such chemically modified nucleic acid sequences can be seen as a new form of antibodies. They can widely be used, including medical and pharmaceutical basic research, drug development, diagnosis, and therapy (Famulok et al., Chem. Rev. 107 (9) 3715).

Analytical and separation tools bearing chemical antibodies as molecular recognition and binding elements are another significant field of application.

Combinatorial chemistry

[16] Even though combinatorial chemistry (Lam et al. Chem. Rev. 1997, 97, 41 1-448 41 1) has been an essential part of early drug discovery for more than two decades, so far only one de novo combinatorial chemistry-synthesized chemical has been approved for clinical use by FDA. The analysis of poor success rate of the approach has been suggested to connect with the rather limited chemical space (Dobson, Nature Vol. 432, 16, 2004, Medina-Franco et al., Curr. Comp. -Aided Drug Design, Wikipedia) covered by products of combinatorial chemistry. When comparing the properties of compounds in combinatorial chemistry libraries to those of approved drugs and natural products, it was noted (Feher & Schmidt, J. Chem. Inf. Comput. Sci., 4, 218, 2003) that combinatorial chemistry libraries suffer particularly from the lack of chirality, as well as structure rigidity, both of which are widely regarded as drug-like properties. Even though natural product drug discovery has not probably been the most fashionable trend in pharmaceutical industry in recent times, a large proportion of new chemical entities still are nature-derived compounds, and thus, it has been suggested that effectiveness of combinatorial chemistry could be improved by enhancing the chemical diversity of screening libraries (Guido et al., Comb Chem High Throughput Screen. 2011). As chirality and rigidity are the two most important features distinguishing approved drugs and natural products from compounds in combinatorial chemistry libraries, these are the two issues emphasized in so-called diversity oriented libraries, i.e. compound collections that aim at coverage of the chemical space, instead of just huge numbers of compounds. DNA Encoded Libraries

[17] The burgeoning cost of drug discovery has led to the ongoing search for new methods of screening greater chemical space as inexpensively as possible to find molecules with greater potency and little to no toxicity. Combinatorial chemistry approaches in the 1980s were originally heralded as being methods to transcend the drug discovery paradigm, but largely failed due to insufficient library sizes and inadequate methods of de-convolution. Recently, the use of DNA-displayed combinatorial libraries of small molecules has created a new paradigm shift for the screening of therapeutic lead compounds.

[18] Major challenges in the use of DNA- displayed combinatorial approaches in drug discovery: (a) the synthesis of libraries of sufficient complexity and (b) the identification of molecules that are active in the screens used. In addition, Morgan et al. (U.S. Patent Application Publication No. 2007/0224607) states that the greater the degree of complexity of a library, i.e., the number of distinct structures present in the library, the greater the probability that the library contains molecules with the activity of interest. Thus, the chemistry employed in library synthesis must be capable of producing vast numbers of compounds within a reasonable time frame.

Summary of Invention

[19] There remains a need for economic and efficient methods for the identification of suitable aptamers for a target molecule by the screening of large numbers and converging in a rapid and efficient way to the identification of individual molecules.

[20] A major goal of this invention is to provide a method for the production of large libraries of nucleic acid derived chemical antibodies and a time and cost efficient screening process to identify high-affinity binders from these complex libraries against a large panel of potential target molecules.

[21] The present invention provides an iterative process to produce and identify new and optimized chemical antibodies (highly modified nucleic acid sequences) that have high binding affinity for a single or a plurality of targets using DNA encoded chemical antibody libraries to start with. The libraries, in preferred embodiments, comprises multiple sub- libraries (pools) of (highly degenerated) non-natural nucleic acid sequences (binding regions) covalently attached to a natural DNA sequence with a unique sequential code (sequence identifier) flanked by two universal primer regions (Fig1 a-b). Each bar-coded sub- library comprises a sufficiently large number of identical sequences such that they are more likely to be available to the application of acyclic identification methods that allow to avoid multiple cycles of evolutionary selection, which are the most time-consuming and costly steps in traditional SELEX method. Several rounds of selection, PCR amplification and decoding of the natural DNA barcode sequence with less and less degenerated libraries allow the identification of each particular binding region.

[22] The present inventors have found a method using the generation of series of pooled libraries of molecules comprising binding regions of interest and pool-specific sequence identifiers that can be screened against a target molecule. In the method, the pools of a library are screened, the positive pools are identified based on the sequence of the identifier. From the positive pools a further library is generated, preferably de novo, wherein the binding regions of the molecules of the positive pools are combined with a set of new pool-specific identifiers in a set of new pools in the further library. These pools, of which the binding regions are typically a subset of the binding regions of the previous library, are screened again against the target molecule(s). The process is repeated until a set of one or more individual molecules that contain binding regions is identified that can be used as aptamers and/or chemical antibodies. Detailed description of the invention

[23] Thus, in a first aspect the invention pertains to a method for the screening and identification of a molecule interacting or capable of interacting with a target molecule of interest.

[24] The method comprises the steps of providing a library of molecules, each molecule comprising a binding region and a sequence based identifier, organising the molecules in pools, screening the pools for interaction with a target molecule, identifying and selecting pools that interact with the target molecule (positive pools) using the sequence based identifier, providing a further library of molecules that contain the binding regions from the selected (positive) pools (the binding regions being a subset of the binding regions of the previous library), proving the binding regions in the further library with (different) sequence based identifiers, organising the molecules comprising the selected binding regions and pool specific identifiers in further pools and repeating the steps of selecting positive pools and providing further libraries from positive pools until convergence is achieved at the level of individual binding regions and the individual binding regions that interact with the target molecule are identified. Put differently, each library is a group of binding regions wherein each binding region is incorporated in a molecule together with a sequence based identifier.

[25] The method comprises providing a library of binding regions, wherein the binding regions in the library are pooled in pools and in the pool each binding region is covalently attached to a pool-specific nucleotide sequence identifier, contacting the pools with a target molecule, identifying and selecting pools that interact with the target molecule, providing a further library of binding regions from the selected pools and repeating the steps of contacting, identifying and selecting and providing further libraries, wherein the successive libraries contain a decreasing number of binding regions. Preferably the number of binding regions in the successive library is lower than in the previous library, so that the different set is a subset of the previous library. Preferably the successive libraries are (physically) generated (synthesized) de novo.

[26] The successive or further libraries of binding regions are combined with pool specific sequence identifiers and organised in pools. Preferably the distribution of the binding regions of the successive library over the pools is different compared to the previous library and/or the number of binding regions the pools is different, preferably smaller, compared to the previous library.

[27] In one embodiment a combination of microarray synthesis and combinatorial chemistry is applied for the synthesis of said libraries.

[28] In another embodiment high throughput solid phase synthesizer instruments (Cheng, J.-Y., Nucleic acids research 30, e93, 2002; Livesay, E. A. et al. A Scalable High- Throughput Chemical Synthesizer. Genome Research 12, 1950-1960, 2002) and combinatorial chemistry are be used.

[29] In the method of the invention, a target molecule is provided. The target molecule can be provided on a solid support, for instance being covalently attached to a bead or an array or in the form of a screening assay which may depend on characteristics of the target molecule. Affinity or size exclusion chromatography as well a gel electrophoresis can be applied with and without a reporter group by selection the elution section where the protein is located. The target molecule can also be provided in solution and the interaction between the molecules in the pools and the target molecule can be determined, for example using chromatography or electrophoretic techniques.

[30] The target molecule can be selected from the group consisting of proteins, polypeptides, nucleic acids, carbohydrates, lipids, polysaccharides, glycoproteins, hormones, receptors, antigens, antibodies, antibody mimics, viruses, pathogens, toxic substances, substrates, metabolites, transition state analogs, cofactors, inhibitors, drugs, small molecules, dyes, nutrients, pollutants, growth factors, cells, tissues, cell lysates or microorganisms, cell surface molecule, cell membrane protein and any component, fragment or portion thereof. A plurality of target molecules can be used, wherein preferably each target molecule carries a different label to distinguish it from other target molecules. A plurality of target molecules may be a set of structural homologues.

[31] The method comprises a step in which a library of molecules (binding regions) is provided. Such a library can vary in size from 10 to 1000, 10.000 or even 100.000 or 1.000.000 of molecules. Based on theoretical considerations it is possible to generate (combinatorial) libraries of natural and non-natural nucleic acid sequences comprising binding regions with up to 10 consecutive and separated degenerated positions (N¹⁰⁾. This corresponds to ca. 1 Mio (4¹⁰) different sequences if only 4 different monomers (N = 4) are used. All of these members in the library can be covalently attached to a particular DNA- barcode (sequence identifier). This DNA-barcode may be a natural DNA-barcode. This attachment can be provided by chain elongation during the (microarray) synthesis process or other combinatorial way of producing libraries of nucleic acid sequences that contain combined natural and non-natural nucleic acids. Either the synthesis is started with a particular coding region for each feature, followed by the potential binding sequence with the degenerated positions or it is done the other way round.

[32] Typically, a first library is provided that is commonly indicated as L1 , which library comprises molecules M (M1 , M2, M3, M4... Mn). Each molecule in the library comprises a binding region B and a sequence identifier ID. The molecules in the library can be divided or split into pools P (P1 , P2, P3, P4... Pm) for a library, indicated as P1-1 (the first pool from the first library), P1-2, P1-3 etc.. Preferably, each pool comprises a (different) selection of the molecules from the library. Each molecule in a pool preferably shares a pool-specific identifier, such that a first pool (P1) is distinguishable over a second pool (P2) by virtue of the pool specific identifier for pool P1 (ID1) over Pool P2 (ID2). Each molecule in a pool preferably differs in the binding region B. The same binding region B may be present in different pools, but with a different identifier ID. A more elaborate scheme is provided herein below for multiple libraries each comprising multiple pools each comprising multiple molecules.

[33] The pools from the library are contacted (screened) with the target molecule. Thus, the absence, presence or amount of interaction between the molecules contained in the pools with the target molecules is determined and the pools in which an interaction is determined are identified.

[34] The pools that contain molecules that have an interaction with the target molecule are identified. The identification typically comprises or consists of the identification of the identifier that is included in the molecule to identify the pools.

[35] The identified pools are then combined to create a successive or second library. In certain embodiments, the binding regions comprised in the identified pools of the first library are combined to create a second library. This can be done in silico. Thus a successive or second library that at least contains and preferably consists of the (binding regions of the molecules of the) pools that have been identified from the previous or first library is created. In certain embodiments, the successive or second library comprises or consists of a subset of the molecules (or binding regions) of the previous or first library. In certain embodiments, the successive or second library comprises a subset of the binding regions of the molecules of the pools of the first library. Thus, in a preferred embodiment, the successive or second library is composed of the binding regions of the molecules of the pools that have been identified during the screening of the pools of the previous or first library.

[36] The molecules of the successive or second library can be pooled, wherein each pool contains a different selection of the molecules of the successive or second library. Each pool in the successive or second library is distinguishable over the other pools in the successive or second library by the presence of a pool-specific identifier. Each molecule in a pool of the successive or second library contains the pool-specific identifier such that each pool of molecules of the successive or second library can be distinguished from the other pools of molecules in the successive or second library. The identifier sequence itself may the same (i.e. reused) or different from the identifiers used for the pools of the previous or first library.

[37] The pools from the successive or second library can be contacted with the target molecule and any interaction between the molecules in pools and the target molecules can be determined. The absence, presence or amount of interaction with the target molecule for the pools of the successive or second library can be determined. Depending on the interaction, pools can be selected for further screening, involving subsequent steps of library formation, pool formation and identification of molecules until in a final step, a set of individual molecules (or rather the binding regions) are screened for an interaction with the target molecule and the molecules (binding regions) that have an interaction can be individually identified. This means that in a final step, a library of individual molecules may be provided, wherein binding regions are not pooled and connected with a pool-specific identifier, but wherein each binding region is connected with a binding region-specific identifier.

[38] As used herein, the singular forms "a," "an" and "the" include plural referents unless the context clearly dictates otherwise. For example, a method for connecting a binding region with an identifier to yield "a" molecule , as used herein, includes a plurality of molecules (e.g. 10's, 100's, 1000's, 10's of thousands, 100's of thousands, millions, or more molecules).

[39] The interaction between the pools and the target molecule can be determined by determining the presence, absence or amount of a label. The label can refer to one or more reagents that can be used to detect interactions involving a target molecule and a binding region. A label (or detection moiety) is capable of being detected directly or indirectly. In general, any reporter molecule that is detectable can be a label. The interaction can also be determined by using chromatographic and/or electrophoretic techniques that do not use a label.

[40] Labels include, for example, (i) reporter molecules that can be detected directly by virtue of generating a signal,

(ii) specific binding pair members that can be detected indirectly by subsequent binding to a cognate that contains a reporter molecule,

(iii) mass tags detectable by mass spectrometry, and

(iv) oligonucleotide primers that can provide a template for amplification or ligation.

[41] The reporter molecule can be a catalyst, such as an enzyme, dye, fluorescent molecule, quantum dot, chemiluminescent molecule, coenzyme, enzyme substrate, radioactive group, a small organic molecule, amplifiable polynucleotide sequence, a particle such as latex or carbon particle, metal sol, crystallite, etc., which may or may not be further labelled with a dye, catalyst or other detectable group, a mass tag that alters the weight of the molecule to which it is conjugated for mass spectrometry purposes, and the like. The label can be selected from electromagnetic or electrochemical materials.

[42] The detection moiety can be detected by emission of a fluorescent signal, a chemiluminescent signal, or any other detectable signal that is dependent upon the identity of the moiety. In the case where the detectable moiety is an enzyme (for example, alkaline phosphatase), the signal can be generated in the presence of the enzyme substrate and any additional factors necessary for enzyme activity. In the case where the detectable moiety is an enzyme substrate, the signal can be generated in the presence of the enzyme and any additional factors necessary for enzyme activity. Suitable reagent configurations for attaching the detectable moiety to a target molecule include covalent attachment of the detectable moiety to the target molecule, non-covalent association of the detectable moiety with another labeling agent component that is covalently attached to the target molecule, and covalent attachment of the detectable moiety to a labelling agent component that is non-covalently associated with the target molecule.

[43] Selection of the molecules that interact can also be performed by chromatography or electrophoresis. Eluting the pools or subjecting the pools after being contacted with the target molecule in solution to electrophoretic processes may be used to also discriminate between molecules that interact with the target molecule and molecules that do not or to a lesser extent. In this way a preselection of molecules can be made at pool level.

[44] In certain embodiments, the 'contacting to determine an interaction' is under buffer conditions and/or stringency conditions that allow the molecules in the (pools of the) library to bind to the target molecule. Buffer conditions refer to the chemical nature of the buffer, pH, added salts, denaturants, detergents, molar ratio of target to molecules (aptamer candidates) and other parameters well known to those skilled in the art of modulating target interactions with nucleic acids. Stringency is used in reference to the conditions of temperature, ionic strength, and the presence of other compounds such as organic solvents or chaotropic reagents, under which the binding assays of the target molecule and the molecules in the pools are conducted.

[45] In certain embodiments, the identification of the pools comprises identifying the identifier for the pool. Preferably, the identifier in the molecule is amplifiable, i.e. can be amplified using amplification techniques such as PCR. To this end, the identifier may be accompanied by or flanked at one or both ends by a primer binding site to which primers can hybridize which serves as the starting point for amplification. An amplifiable identifier preferably contains only natural nucleotides.

[46] The identifier can be amplified using enzymes such as a polymerase and dNTPS. The identifier can be identified using sequencing, preferably high throughput sequencing such as Roche FLX or lllumina sequencing technology. High throughput sequencing in this context means the determination of thousand of nucleotide sequence simultaneously. The technology is well described in 'Next generation genome sequencing, Janitz ed. 2008, Wiley-VCH.

[47] The molecules in a library or in a pool are preferably synthetic molecules. The molecules are preferably synthesised simultaneously and/or in parallel, for instance on an array.

[48] Standard DNA synthesis uses microliters of reagents in tiny glass columns to make individual sequences. Microarrays miniaturize and parallelize the synthesis, producing thousands of sequences side by side; small volumes of reagents normally used for single reactions can wash over an entire slide of tens of thousands of nucleotides at once.

[49] Compared with traditional DNA synthesis techniques, microarrays offer a far less expensive source of oligonucleotides; long oligos (about 100 to 200 nucleotides) generally cost about $0.10 per nucleotide from commercial vendors, but microarray-based methods can be used to produce oligonucleotides for considerably less, about a million 60-mers for $600 in some cases, though prices can vary for many reasons (Nature Methods 8, 457-460, 201 1).

[50] In principle there a two basic methods to fabricate DNA microarrays either by printing pre-manufactured oligonucleotides with fine-pointed pins onto a solid support ("Spotting") or by in situ synthesis.

[51] In spotted microarrays, the probes are oligonucleotides, cDNA or small fragments of PCR products that correspond to mRNAs. A common approach utilizes an array of fine pins or needles controlled by a robotic arm that is dipped into wells containing DNA probes and then depositing each probe at designated locations on the array surface. The resulting "grid" of probes represents the nucleic acid profiles of the prepared probes and is ready to receive complementary cDNA or cRNA "targets" derived from experimental or clinical samples.

[52] The in-situ microarray synthesis process includes photolithography using pre-made masks (Affymetrix), dynamic micromirror devices (DMD) (Nimblegen, LCSciences,

MycroArray) or sequential laser deprotection (FlexGen), ink-jet printing (Agilent, Lausted C et al., Genome Biology 5 (8), 2004) or electrochemistry on microelectrode arrays

(CombiMatrix).

[53] Successful synthesis of up to 150mers using ink-jet technology has been reported (LeProust et al., Nucleic Acids Research, 2010, Vol. 38, No. 8. 2522-2540).

[54] Such library or pool according to the invention can be provided by a Flexarrayer synthesis system available via Flexgen B.V., Leiden the Netherlands, a solid phase DNA synthesiser that can produce hundreds of thousands of oligonucleotides in one run. Each library of molecules can be designed (in silico) together with the desired identifiers and synthesised economically and efficiently using such platforms. The (chemically) synthesised oligonucleotides can be provided in pools, either on a support or in solution or can be provided in solution and spotted on a support for further screening using known methods in the art. The libraries and pools of molecules can be synthesised independently from each other, i.e. each library and /or pool can be the result of a separate synthesising run.

[55] The pools that have been identified for their interaction with the target molecule (i.e. positive pools, regardless of whether the interaction was the absence, presence or amount of interaction) are selected and combined. The combining step typically means that the sequence information of the molecules is combined, the identifier sequence is stripped (in silico) and a new library is generated comprising the binding regions of the positive pools. These binding regions can be pooled again (in silico) and identifiers added to identify the pools. Subsequently the new library is generated by synthesising the corresponding oligonucleotides of the pools and or library. This method of generating libraries and the screening thereof differs from the prior art wherein large libraries of individual molecules are provided (synthesised/isolated) that are combined into pools, screened and from the pools the identity of the individual molecules is derived (via several steps). In the present invention the pools and libraries are preferably separately synthesised de novo. Thus in certain embodiments, the first, second and further libraries are independently chemically

synthesised oligonucleotide libraries. First, second and further libraries may be generated by only connecting binding regions and identifiers, which identifiers may be pool-specific identifiers or binding region-specific identifiers. Hence, the binding regions and identifiers may be provided once, and the generation of first, second and further libraries may be carried out by performing connecting steps de novo only. The binding regions and identifiers may in such a scenario be provided separately, and from these building blocks, the first, second and further libraries can be generated. Hence, in this scenario, the libraries are also independently generated.

[56] In certain embodiments and outlined herein before, the number of molecules (or unique binding regions) in a library of a lower order is less than that in a library of a higher order. Thus, the number of molecules in a lower order library Le is less than a higher order library Ld, with d, e being an integer and d<e and indicating the order of the library.

[57] In certain embodiment, in one pool, the molecules each have the same identifier. In certain embodiments, in one pool each molecule has a different binding region. The library however, may contain binding regions that are present in multiple pools. Pooling strategies such as 3D, row or column pooling strategies may be used to distribute binding regions over the library. Deconvolution techniques based on the sequence identifier may be used to deconvolute the data to identify the binding regions that interacted with the target molecule. Deconvolution is then based on the coincidence of the distribution of the binding regions over the pools and the pools that tested positive in the screen. Such methodologies can be advantageously, depending on the expected amount of positive screens, to more quickly converge to the individual binding regions.

[58] The number of libraries may vary between two and 100. It depends largely on the number of molecules (binding regions) to be screened, the number of pools desired and the used of structured pooling techniques that allow for deconvolution based on the combined occurrence of the identifier. The number of pools in a library is typically between 2-100000. The number of pools can decrease or increase with an increasing order (generation) of the library. The number of binding motifs attached to a specific identifier sequence will usually decrease with an increasing order (generation) of the library. The final library usually is not pooled, but each molecule is individually labelled to identify the binding regions that have the interaction with the target molecule. For example, a first library may comprise 1 ,000- 10,000 binding regions per pool, the second library may comprise 50-200 binding regions per pool, and the final library may comprise molecules wherein single binding regions are connected to binding region-specific identifiers, i.e. each molecule is individually labelled. A pool can contain from 2-100000 molecules, typically 20-1000. The number of pools and molecules in a library and the number of molecules in a pool can vary, depending on the circumstances.

[59] The identifier, which is a unique sequence identifier between the pools of a library, is typically from 2-100 nucleotides in length, with a preference of 12-25 nucleotides usually being sufficient for most purposes.

[60] The molecules use in the present invention are typically oligonucleotides and preferably of a length from 10-250 nucleotides. The nucleotides in the oligonucleotide may be natural (A,C,T,G, U) or artificial (non-natural) nucleotides (PNA, LNA, UNA). The non- natural nucleotides can be modified in the nucleobase (such as 2,6-diaminopurine, isoG or isoC, pteridines, pyrene ) and/or in the carbohydrate (such as mirrorimage pyranose, 2'- methoxy, 2'-fluoro-substituted carbohydrates). In certain embodiments, the linkage between nucleotides are not natural/are synthetic, such as 2'-3', 3'-3', 2'-5, 5'-5', 2'2' linkages, phosphor-dithioate linkages, chiral linkage such as phosphor-thioates and phosphotriester-, alkylphosphonate internucleotide linkage. In another embodiment non-natural nucleotides, which allow "Click"-chemistry to covalently attach one or multiple different types of ligands such as small peptides or lipophilic steroid analogs are introduced into the binding region [Moses and Moorhouse, Chem. Soc. Rev. 2007, 1249-1262; Delft et al. Org. Lett. 2010, Vol.12 (23), 5486-5489] .

[61] The binding region is the part of the molecule that is designed to be tested for its affinity or binding capacity to the target molecule. In this context, a binding region or binding site is a certain stretch within a larger nucleic acid with a stable secondary structure to which specific other molecules and ions in this context collectively called target molecules form one or multiple non-covalent bonds (i.e. ionic bonds, hydrogen bonds, van der Waals interactions) across a certain contact surface area. Binding regions analogously exist on antibodies as specifically coded regions that bind antigens based upon their structure. The binding region can consist entirely of natural nucleotides, but it is preferred that the binding region contains non-natural nucleotides. Preferably the binding region has a length in the order of 5-150 nucleotides, more preferably in the order of 5-100, with higher preference from 5-60 nucleotides. Preferably the binding regions contains at least one non-natural nucleotide or at least one non-natural linkage. Typically and preferably the number of non- natural linkages and/or nucleotides varies, independently, from 5-20, more preferably from 7-15.

[62] The binding region of the molecules may be homodirectional; heterodirectional, preferably allowing intramolecular circularization by intramolecular base pairing; or flanked by complementary sequences and wherein the direction of 'the terminating sequence is switched' to allow intramolecular circularization by intramolecular base pairing. Examples hereof are provide in Figure 1. The binding region may be coupled to the identifier (DNA Barcode) covalently, either directly or via a spacer.

[63] The target molecule can be proteins, polypeptides, nucleic acids, carbohydrates, lipids, polysaccharides, glycoproteins, hormones, receptors, antigens, antibodies, antibody mimics, viruses, pathogens, toxic substances, substrates, metabolites, transition state analogs, cofactors, inhibitors, drugs, small molecules, dyes, nutrients, pollutants, growth factors, cells, tissues, or microorganisms and any fragment or portion of any of the foregoing. In one embodiment, a "target" refers to a cell surface molecule, such as a cell membrane protein.

[64] The term "small molecules" and analogous terms include, but are not limited to, peptides, peptidomimetics, amino acids, amino acid analogs, polynucleotides,

polynucleotide analogs, nucleotides, nucleotide analogs, other organic and inorganic compounds (i.e., including heteroorganic and organometallic compounds) having a molecular weight less than about 10,000 grams per mole. In some embodiments, the term refers to organic or inorganic compounds having a molecular weight less than about 5,000 grams per mole, less than about 1 ,000 grams per mole, less than about 500 grams per mole, less than about 100 grams per mole. Salts, esters, and other pharmaceutically acceptable forms of such compounds are also encompassed.

[65] Thus, in an exemplary embodiment, the invention relates to a method for the screening and identification of a molecule interacting with a target molecule comprising a) providing a target molecule T;

b) providing a first library of molecules L1 wherein each molecule comprises a first sequence identifier and a binding region, wherein the molecules are comprised in a plurality of first pools, each first pool comprising a different selection of the molecules from the first library, wherein each first pool is distinguished from other first pools in the first library by a unique (pool-specific) first sequence identifier (DNA barcode) and wherein each molecule in the first pool comprises the corresponding unique (pool-specific) first sequence identifier such that each first pool of molecules can be distinguished from other first pools of molecules in the first library by the unique first (pool-specific) sequence identifier;

c) contacting the molecules of the one or more first pools from the first library with the target molecule to thereby determine an interaction between the (binding region of the) molecules in the first pool and the target molecule;

d) identifying first pools from the first library that contain molecules that interact with the target molecule;

e) optionally providing a second library L2 (which is preferably a subset of molecules from L1J comprising the (binding regions of the) molecules from the identified first pools, each molecule further containing a second sequence identifier, wherein the molecules are comprised in a plurality of second pools, each second pool comprising a different selection of the molecules from the second library, wherein each second pool is distinguished from other second pools in the second library by a unique second sequence identifier and wherein each molecule in the second pool comprises the corresponding unique second sequence identifier such that each second pool of molecules can be distinguished from other second pools of molecules in the second library by the unique second sequence identifier; f) optionally, contacting one or more second pools from the second library with the target molecule to thereby determine an interaction between the molecules in the second pools and the target molecule;

g) optionally, identifying second pools from the second library that contain molecules that interact with the target molecule;

h) optionally, repeating steps f), g), and h) with third and further libraries, preferably comprising the molecules from the identified second and further pools, wherein the molecules are comprised in a plurality of third and further pools wherein each third and further pool is distinguished from other third and further pools in the third and further libraries by a unique third and further sequence identifier and wherein each molecule in the third and further pools pool comprises the corresponding unique third and further sequence identifier such that each third and further pool of molecules can be distinguished from other third and further pools of molecules in the third and further library by the unique third and further sequence identifier and identifying third and further pools that contain molecules that interact with the target molecule;

i) optionally, providing a final library comprising the molecules from the identified pool of step d), g) or h), wherein each molecule has a unique sequence identifier for each unique binding region;

j) optionally, contacting the final library with the target molecule;

k) identifying a molecule from the library that interacts with the target molecule.

[66] The above use of the notations of first, second, third and further serve to illustrate the concept of the invention. It is noted that the concept of the invention resides in the use of converging pools of molecules that are screened against a target molecule and that positive pools are identified, combined and used for the generation of a new library of pools, typically by synthesising the new library again. This is conceptually different from the methods in the art wherein the molecules in the library are individually synthesised, the individual molecules are pooled, the pools are screened, the positive pools are identified and a new library is formed from the individual molecules. In the present invention, the first library is designed in silico and the pools are designed in silico. The library is synthesised, i.e. the molecules in the library are synthesised simultaneous/in parallel, for instance on a carrier. Alternatively, for instance when the library is very large and exceeds the capacity of the synthesising platform, the library can be synthesised in parts, for instance pool by pool. Having screened the pools against the target molecule, the positive pools are identified by the identifier present in the molecules, typically by PCR and sequencing of the identifier. Based on the information obtained, the corresponding binding regions are identified in silico. From these binding regions, a new library is generated in silico, pooled (in silico) by combining the binding regions with identifiers. The resulting new combination of binding regions and identifiers then is synthesised de novo as outlined hereinabove, screened against the target molecule(s) after which the selection and design process can start again until the level of individual molecules is reached, which may be after two, there, four, five, ten or twenty repetitions.

Brief Description of Drawings

[67] Figure 1 Discloses two variants of the molecules that can be used in the libraries and pools of the present invention to screen for the possible interaction with a target molecule.

A: The molecule contains a binding region, which may comprise double stranded regions that can be coupled via a nucleotide spacer to an 8-25 bp identifier sequence (DNA barcode) flanked in this embodiment by two universal primer binding sites. The binding region can contain sections that are intramolecular complementary such that hairpins, bulges etc can be formed.

B: The molecule has the same structure as in Fig 1A, but comprises in the binding region a switch that reverse the direction of the oligonucleotide making up the binding region from 3'- to 5' or vice versa. Such conformation is described in applicant co-pending application WO 2102/102616A1 thus the switch has a heterodirectional design, i.e. the oligonucleotide comprises at least one intramolecular switch, a 3'-3' or 5'-5' covalent linkage, such that the polynucleotide comprises at least two polynucleotide segments in opposite direction.

Figure 2: AC_t of the 3 different identifier sequences before and after selection using one ID specific primer (5-7) and one universal primer (8) for qPCR. For each of the three DNA oligonucleotides (high (about 13), low (about 0.5) or no affinity (about 0) from left to right), the Δ C_t is shown. For each, from left to right, 1x, 2x, 5x, 10x wash buffer B at 25°C and 1x, 2x, 5x, 10x wash buffer B at 30°C

Schematic representation of the libraries, pools, molecules, identifiers and binding regions. L: Library;.

d: order of library;

P: Pool;

m, n, o: number of the pool

Pd.o: o-th pool of library d

M: Molecule

Md.o.1 : first molecule of pool o of library d;

x, y, z: the number of the molecule xn: the number x of molecules in pool n of library d.

ID : identifier

IDn: identifier for pool n

B: Binding region

B1.1.1 : Binding region of the first molecule M 1.1.1 in the first pool P1.1 in the first library L1 Binding regions B2 of L2 are a subset of binding region B1 of L1.

Identical binding regions B may occur in different pools in one library to create redundancy or to allow complex and deconvolutable pooling. For instance, B1.1.1 may be identical in its nucleotide and linkage composition to B1.3.3 in L1.

L P1.1 P1.2 P1.n

1 M1.1.10D1 , B1.1.1) M1.2.10D2, B1.2.1) M1.n.1 (IDn, Β1. Π.1)

M1.1.2(ID1 , B1.1.2) M1.2.2(1 D2, B1.2.2) M1.n.2(IDn, Β1. Π.2)

M1.1.30D1 , B1.1.3) M1.2.30D2, B1.2.3) M1. ri.30Dn, Β1. Π.3)

M1.1.40D1 , B1.1.4) M1.2.4(1 D2, B1.2.4) M1. ri.40Dn, B1.n.4)

M1.1.x1 (ID1 , B1.1 .X1 ) M1.2.x2(ID2, B1.2.X2) M1.n.xn(IDn, Bln.xn)

L P2.1 P2.2 P2.m

2 M2.1.10D1 , B2.1.1) M2.2.10D2, B2.2.1) M2m.1 (IDm, B2.m.1)

M2.1.20D1 , B2.1.2) M2.2.20D2, B2.2.2) M2m.2(IDm, B2.m.2)

M2.1.30D1 , B2.1.3) M2.2.3(ID2, B2.2.3) M2m.3(IDm, B2.m.3)

M2.1.40D1 , B2.1.4) M2.2.4(ID2, B2.2.4) M2m.4(IDm, B2.m.4)

M21.y1 (ID1 , B2.1.y1) M2.2.y2(ID2, B2.2.y2) M2.m.ym(IDm, B2.m.ym

L Pd.1 Pd.2 Pd.o

d

Md.1.1 (ID1 , Bd.1.1) Md.2.1 (ID2, Bd.2.1) Md.o.1 (IDo, Bd.o.1)

Md.1.2(ID1 , Bd.1.2) Md.2.2(ID2, Bd.2.2) Md.o.20Do, Bd.o.2)

Md.1.3(ID1 , Bd.1.3) Md.2.3(ID2, Bd.2.3) Md.o.3(IDo, Bd.o.3)

Md.1.4(ID1 , Bd.1.4) Md.2.4(ID2, Bd.2.4) Md.o.4(IDo, Bd.o.4)

Md.1.z1 (ID1 , Bd.1.z1) Md2.z2(ID2, Bd.2.z2) Md.o.zo(IDo, Bd.o.zo)

Examples

[68] For the in-situ synthesis of an oligonucleotide microarray a non porous, flat and planar, substrate with either hydroxyl- or primary amino-functions is used. Typically it is a glass surface with 5x10¹²- 2x10¹³ of such anchor groups per mm², which corresponds to 5x10⁶- 2x10⁷ anchor groups per um². The dimensions of a typical microarray are about 10- 30 mm x 10-30 mm, which are illuminated by up to 4.2 million individual pixels of light. The area where one specific sequence is generated is called a feature. Several individual pixels of light can be bundled to synthesize a particular sequence. A typical feature size is 5-100 micrometer in diameter. For a medium feature size of 30 um in diameter there ca. 3-10 x 10^! anchor-groups. During the beginning of the oligonucleotide synthesis only a fraction of these anchor groups are used, mostly due to steric hindrance. The diameter of a typical monomer is around 2.0-2.5 x 10^"9 m, similar to the diameter of a DNA double helix (Philips et al., NAR 2008 (36 (1) e7)). Based on these numbers and assuming the DNA helices are tightly packed cylinders on a totally flat surface there are ca. 3 x 10⁵ molecules per urn². However, hybridisation based analysis revealed a maximum loading of ~ 3 x 10¹² molecules per cm² or ~ 3x10⁴ molecules per urn². (Philips et al., NAR 36 (1) el, 2008). This corresponds to ~ 3 x 10⁷ molecules for a 30 urn diameter feature. With a typical stepwise coupling yield between 98.5% and 99.5% (~ 25-60 % total) such feature comprise around 1-2x10⁷ full length product for a 90mer nucleotide long sequence.

Length of Oligonucletide

Average Stepwise

Coupling Yield 100 90 80 70 60 50 40 30 20

97.5 0.080 0.102 0.132 0.170 0.219 0.282 0.363 0.468 0.603

98.0 0.133 0.162 0.199 0.243 0.298 0.364 0.446 0.545 0.668

98.5 0.221 0.257 0.298 0.347 0.404 0.470 0.546 0.635 0.739

99.0 0.366 0.405 0.448 0.495 0.547 0.605 0.669 0.740 0.818

99.5 0.606 0.637 0.670 0.704 0.740 0.778 0.818 0.860 0.905

100.0 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000

[69] Successful synthesis of up to 150mers using ink-jet technology has been reported (LeProust et ai, Nucleic Acids Research, 2010, Vol. 38, No. 8. 2522-2540).

[70] Combinatorial libraries of natural and non-natural nucleic acid sequences are generated comprising binding regions with up to 10 consecutive and separated degenerated positions (N10). This corresponds to ca. 1 Mio (4¹⁰) different sequences if only 4 different monomers (N = 4) are used. All of these members in the library are covalently attached to a particular natural DNA-barcode. This attachment simply happens by chain elongation during the microarray synthesis process. Either the synthesis is started with a particular coding region for each feature, followed by the potential binding sequence with the degenerated positions or it is done the other way round.

[71] With 10 Mio molecules per feature in total there are still- 10 molecules of each individual sequence if there are 4 monomers and 10 degenerated positions (4¹⁰). To identify if there is a good binder within these ~1 Mio library the binding affinity itself, the exact conditions of the selection process as well as "even" amplification of the barcodes during the PCR are relevant. Literature references (Kupakuwana et al., PLOS 2011 , 6 (5), e19395) stipulate that relative enrichment of 10000 and more is feasible. Thus, a library size of several thousand individual sequences per barcode is sufficient. [72] For instance, with a degeneration of N⁷=16384 and 100.000 barcodes the total size of the 1 st generation library is already ~ 16384x10⁵=16x10⁸ different molecules. These are numbers which reach the required chemical space (Kupakuwana et al., PLoS ONE 6 (5) e19395, 201 1) to find at least one good binder. The separation of the natural DNA barcode, which can enzymatically be amplified and the non-natural binding sequence liberates from the SELEX intrinsic requirement to work with monomers which can be processed during PCR.

Example 1 : Illustrative example for the general principle.

[73] For a first library of potential aptamers, 100.000 potential binding regions are designed in silico and distributed over 1000 pools. 1000 pool-specific identifiers are designed in silico and combined in silico with the binding regions. The pools of the library are synthesized on a Flexarrayer (Flexgen, Leiden the Netherlands) and cleaved off by chemical means. The pools are contacted with a labelled target molecule under stringent conditions. The pools (100) that express an interaction with the target molecule are identified. The identifiers associated with the 100 pools are amplified using universal primers in a PCR reaction The pools are identified based on the sequence of the identifier by Roche FLX sequencing. The binding regions in the identified pools are identified (in silico). The binding regions from the identified pools are distributed over 10 pools and combined with 10 pool- specific identifiers (in silico) to form a second library (of 10 pools). The second library is synthesised on a Flexarrayer. The pools form the second library are contacted with the labelled target molecule under stringent conditions. The pools (2) that express an interaction with the target molecule are identified. The identifiers associated with the 2 pools are amplified using universal primers in a PCR reaction The pools are identified based on the sequence of the identifier by Roche FLX sequencing. The binding regions in the identified pools are identified (in silico). The binding regions (24) from the identified pools are individually distributed and combined with 24 pool-specific identifiers (in silico) to form a third library (of 24 individual molecules). The molecules are synthesised and the interaction of the 24 molecules with the target molecule is determined at an individual level. The binding regions that interact with the target molecule are identified. The binding regions are re- synthesised and assayed against the target molecule.

Example 2

To test the concept a representative model library of molecules was prepared. 3 DNA oligonucleotides (1-3) of 90 bases length comprising different binding motifs with good, low and no binding affinity against streptavidin (Bing, T., et al. Bioorganic & medicinal chemistry 18, 1798-805 (2010)), each covalently attached to an identifier sequence (ID) were used. These 3 different molecules were mixed in a 1 : 1 : 1 : 100000 ratio with oligonucleotide^), which represents a complex library of 4⁷= 16384 identifier sequences each of them attached to a pool of 4⁶ = 4096 different binding motifs. This corresponds to total of -67 million molecules.

List of oligomer sequences used for the proof-of-concept experiment.

Sequence 1-4 comprises a binding motif (underlined) covalently attached to a unique identifier sequence (cursive) flanked by two primer binding sites. 5,6 and 7 are ID specific primer sequences which are used with sequence 8 for qPCR to quantify the abundance of the identifier sequences (ID1 ,2,3) after selection. The oligomers were purchased from Biomers.net GmbH in Ulm (Germany), synthesized after standard protocols, HPLC purified and carefully quantified via UV absorption (NanoDrop). N: at this position all 4 (A,C,G,T) monomers are possible. Prior selection the test library, in PBS pH 7.4, 1 m gCI₂, 0.01 % Tween 20 (buffer A), was heated to 95°C for 3 min, then put on ice for 5 min and then stored at room temperature (RT) for 5 min. For selection ca. 25 microL (ca. 0.25 mg) Streptavidin coated magnetic beads Dynabeads® MyOne™ Streptavidin T1 (Life Technologies) were incubated with 00 pmoi of test library in buffer a for 30-50 min at RT and mixing. The beads are washed once with 250 microL of buffer A for 3 min at 25 or 30°C and 1 , 2, 5 and 10 times with wash buffer B (PBS pH 7.4, 1 mM gCi₂, 0.05% Tween 20). Quantitative PGR (annealing temperature 48°C) was performed directly after resuspension in 20 microL EB buffer (Quiagen) using iQ SYBR Green Super ix (BioRad).

Results:

Depending on the washing conditions a AC_t (before and after selection) of up to 15 was determined for the particular identifier sequence (ID) attached to a good binding motif, whilst a weak binding motif's ID was hardly enriched compared to a non-binding motif's ID. (Graph 1). This corresponds to a enrichment factor of about 10000-30000 of an ID attached to a significant good binding motif.

Based in this experiment it is possible to allow at least 100-1000 pool members per identifier sequence and still detect via the corresponding IDs significant enrichment of a high-binding motif over a non-binding motif .This may be even more likely as members of such a pool can normally all be related to each other and therefore may contribute as "family" members more or less to the overall binding affinity of a pool and consequently to abundance of the corresponding attached ID sequence after the selection step. Citation List

Patent Literature

[74] WO201 1/109451

[75] US2007/0224607

[76] WO 2012/102616A1

Non Patent Literature

[77] Hermann and Patel, 2000 Science 4 February 2000: Vol. 287 no. 5454 pp. 820- 825.

[78] Green LS, Jellinek D, Bell C, Beebe LA, Feistner BD, Gill SC, Jucker FM, Janjic N

Chemistry & Biology 2:683695, 1995.

[79] Gold L, Polisky B, Uhlenbeck O, Yarus M (1995) Diversity of oligonucleotide

functions. Ann. Rev. Biochem 64:763-797. [80] Keefe et al., Current Opinion in Chemical Biology 2008, 12:448-456.

[81] Wilson DW, Szostak JW. In vitro Selection of Functional Nucleic Acids. Ann. Rev.

Biochem., 1999; 68:61 1-648.

[82] Goeringer et al. , 2003.

[83] Klussmann, 2006.

[84] Scherr ef a/. , Bioorg.& Med. Chem. Lett. 7 (13) 1791-1796, 1997.

[85] Tuerk C, Gold L (1990) Systematic evolution of ligands by exponential enrichment:

RNA ligands to bacteriophage T4 DNA polymerase. Science 249:505-510

[86] Ellington and Szostak, 1990.

[87] Ellington AE, Szostak JW. Selection in vitro of single-stranded DNA molecules that fold into specific ligand binding structures. Nature 1992; 355:850-852.

[88] Stoltenburg et al., Biomolecular Engineering 24 (2007) 381-403.

[89] Wang et al., Chin J Anal Chem, 2009, 37(3), 454-460.

[90] Lauridsen, L. H., Rothnagel, J. a & Veedu, R. N. Chembiochem 13, 19-25 (2012).

[91] Famulok ef al., Chem. Rev. 1999 107 (9) 3715.

[92] Renner et al., Future Med Chem. 201 1 Apr;3(6):751-66. "Recent trends and

observations in the design of high-quality screening collections".

[93] B Buller et a;., Bioconjug Chem. 2010 Sep 15;21 (9): 1571 -80."Drug discovery with

DNA-encoded chemical libraries".

[94] Scheuermann et ai, Chembiochem. 2010 May 3; 1 1 (7):931-7. "DNA-encoded chemical libraries: a tool for drug discovery and for chemical biology".

[95] Clark MA., Curr Opin Chem Biol. 2010 Jun; 14(3):396-403. Epub 2010 Mar26.

"Selecting chemicals: the emerging utility of DNA-encoded libraries".

[96] Melkko ef al., Drug Discov Today. 2007 Jun; 12(1 1-12):465-71. Epub 2007 Apr26.

"Lead discovery by DNA-encoded chemical libraries".

[97] Scheuermann et al., J Biotechnol. 2006 Dec 1 ; 126(4): 568-81. Epub 2006 Jun 9.

"DNA-encoded chemical libraries".

[98] Nature Methods 8, 457-460, 201 1.

[99] Heera Krishna and Marvin H. Caruthers; J. Am. Chem. Soc, 201 1 , 133 (25), pp

9844-9854 "Solid-Phase Synthesis, Thermal Denaturation Studies, Nuclease

Resistance, and Cellular Uptake of (Oligodeoxyribonucleoside)methylborane

Phosphine-DNA Chimeras".

[100] Philips et al., NAR 2008 (36 (1) e7.

[101] Kupakuwana et al., PLOS 201 1 , 6 (5), e19395. Douglas J. Dellinger, Christina M. Yamada, Marvin H. Caruthers Current Protocols in Nucleic Acid Chemistry Unit 2.4 "Oligodeoxyribonucleotide Analogs

Functionalized with Phosphonoacetate and Thiophosphonoacetate Diesters". Markus Schweitzer & Joachim W. Engels; Nucleosides and Nucleotides Volume 17, Issue 1-3, 1998 "M ethyl phosphonate Modified DNA Hairpin Loops".

LeProust et al., Nucleic Acids Research, 2010, Vol. 38, No. 8. 2522-2540

Cheng, J.-Y., Chen, H.-H., Kao, Y.-S., Kao, W.-C. & Peck, K. High throughput parallel synthesis of oligonucleotides with 1536 channel synthesizer. Nucleic acids research 30, e93, 2002.

Livesay, E. A. et al. A Scalable High-Throughput Chemical Synthesizer. Genome Research 12, 1950-1960, 2002.

Bing, T., Yang, X., Mei, H., Cao, Z. & Shangguan, D. Conservative secondary structure motif of streptavidin-binding aptamers generated by different laboratories. Bioorganic & medicinal chemistry 18, 1798-805 (2010).

Claims

Method for the screening and identification of a binding region interacting with a target molecule comprising providing a library of binding regions, pooling the binding regions in pools wherein each binding region in a pool is connected with a pool- specific sequence identifier to yield a molecule, contacting the binding regions in the pools with a target molecule, identifying and selecting the pools that contain binding regions that interact with the target molecule using the pool specific identifier to thereby identify a subset of binding regions that form a successive library, generating the successive library and repeating the steps of pooling, contacting, identifying and selecting to generate further successive libraries of decreasing numbers of binding regions at least once until the individual binding region that interacts with the target molecule is identified.

Method according to claim 1 , wherein determining the interaction between the target molecule and the (binding regions of) the molecules in the pool is by determining the presence, absence or amount of a label and/or by chromatography or

electrophoresis.

Method according to claims 2, wherein the label is a fluorescent dye, a fluorescence quencher, an energy-transfer pair, a quantum dot, or a chemiluminescent precursor Method according to claims 1-3, wherein the 'contacting to determine an interaction' is under buffer conditions and/or stringency conditions that allow the molecules in the library to bind or interact with the target molecule.

Method according to claims 1-4, wherein the molecules are ranked based on affinity with the target molecule under the applied buffer conditions and/or stringency conditions.

Method according to claims 1-5, wherein identifying the pool comprises determining the identifier for the pool.

Method according to claims 1-6, wherein the identifier in a molecule is a natural DNA sequence comprised of natural nucleotides

Method according to claims 1-7, wherein the identifier is amplifiable, preferably using enzyme(s).

Method according to claims 1-8, wherein the identifier is flanked by two primer binding sites.

Method according to claims 1-9, wherein the identifier is amplified using primers for the primer binding sites.

Method according to claims 1-10, wherein identifying the identifier comprises sequencing, preferably high throughput sequencing. Method according to claims 1-1 1 , wherein the molecules in a library are synthesized simultaneously/in parallel/in one system.

Method according to claims 1-12, wherein the molecules in a library are provided by synthesis in parallel/ on a solid support/by in situ synthesis/by spotting.

Method according to claims 1-13, wherein the library is a chemically synthesised oligonucleotide library.

Method according to claims 1-14, wherein the first, second and further libraries are independently chemically synthesised oligonucleotide libraries.

Method according to claims 1-15, wherein the number of molecules in a lower order library Lm is less than a higher order (successive) library Ln, with n, m being an integer and m>n and indicating the order of the library.

Method according to claims 1-16, wherein the molecules of a lower order library are a subset from a higher order library

Method according to claims 1-17, wherein the number of libraries is from 2 to 100. Method according to claims 1-18, wherein the pools in libraries are from 2-1000000. Method according to claims 1-19, wherein a pool contains from 1 to 100000 molecules.

Method according to claims 1-20, wherein the unique sequence identifier is from 2- 100 nucleotides.

Method according to claims 1-21 , wherein the binding region is homodirectional; heterodirectional, preferably allowing intramolecular circularization by intramolecular base pairing; or flanked by complementary sequences and wherein the direction of the terminating sequence is reversed to allow intramolecular circularization by intramolecular base pairing.

Method according to claims 1-22,wherein the binding region is covalently linked to the identifier (DNA barcode).

Method according to claims 1-23,wherein the binding region is covalently linked via a spacer to the identifier (DNA barcode).

Method according to claims 1-24, wherein the target molecule is selected from the group consisting of proteins, polypeptides, nucleic acids, carbohydrates, lipids, polysaccharides, glycoproteins, hormones, receptors, antigens, antibodies, antibody mimics, viruses, pathogens, toxic substances, substrates, metabolites, transition state analogs, cofactors, inhibitors, drugs, small molecules, dyes, metal ions, nutrients, pollutants, growth factors, cells, tissues, cell lysates or microorganisms, cell surface molecule, cell membrane protein and any component, fragment or portion thereof.

Method according to claims 1-25, wherein a plurality of target molecules are used.

27. Method according to claims 1-26, wherein each of the plurality of target molecules contains a different label.

28. Method according to claims 1-27, wherein the plurality of target molecules contain structural analogs.

29. Method according to claims 1-28, wherein the label is detected by detection of

fluorescence.

30. Method according to claims 1-29, wherein the binding region comprises natural (A, C, T,G, U) and/or artificial (non-natural) nucleotides (PNA, LNA, UNA).

31. Method according to claims 1-30, wherein the non-natural nucleotides are modified in the nucleobase and/or in the carbohydrate.

32. Method according to claims 1-31 , wherein the linkage between nucleotides are not natural/are synthetic,.

33. Method according to claims 1-32, wherein a binding region contains form one to ten non-natural nucleotides.

34. Method according to claims 1-33, wherein the pooling in a library is a random pooling, or a non-random (ordered) pooling, such as row, column, plate pooling,

multidimensional pooling such as 3D, 4D, 5D, 6D and multidimensional pooling.

35. Method according to claims 1-34, wherein deconvolution of the ordered pooling in the library identifies the molecule that interacts with the target molecule.