WO2016080838A1

WO2016080838A1 - Self-assembled bivalent ligand complex (sablc) libraries and methods for screening such libraries

Info

Publication number: WO2016080838A1
Application number: PCT/NL2015/050815
Authority: WO
Inventors: Stefan Matysiak; Klaus Hellmuth; Marcus Jan Pieter DE JONG
Original assignee: Piculet Biosciences Technologies B.V.
Priority date: 2014-11-21
Filing date: 2015-11-20
Publication date: 2016-05-26
Also published as: NL2013857B1

Abstract

The present invention relates to identifying bispecific ligands binding to a target molecule, more particularly to self-assembled bivalent ligands comprising at least two separate ligands or two sub-libraries thereof, each ligand covalently attached via an optional spacer molecule to the ends of a nucleic acid sequence. The said ligands dimerise via the nucleic sequence and are subsequently covalently linked either directly or by incorporation of a third nucleic acid. The bivalent ligands (or monomeric or in some embodiments trivalent ligands) are provided as libraries and are screened for a specific binding interaction with a target molecule (e.g. a biological target). Bivalent ligands are identified by unique sequence identifier sequences that are comprised in the nucleic acid sequences.

Description

Title: Self-assembled bivalent ligand complex (SABLC) libraries and methods for screening such libraries

Technical Field: The present invention relates to the field of ligand libraries, and to methods for selecting and identifying ligands from libraries that bind to a specified target. In particular, the present invention relates to bispecific or bivalent ligands and methods of generating bispecific or bivalent ligand libraries and methods of selecting bispecific or bivalent ligands that bind to a specific target or two different targets.

Background art

Identification of specific binding molecules or ligands, especially organic molecules, is a real challenge in pharmaceutical sciences and general biology and chemistry. In order to find a specific binding molecule towards a target molecule currently millions of candidates have to be screened. Generation of large libraries with significant diversity to find suitable

candidates is rather costly. Identification of a specific binding motif from classic small molecule libraries with vast members is a very complex and time consuming task and requires significant investments in synthesis, screening, storage and handling.

Therefore alternatives to high throughput screening have been introduced recently including dynamic combinatorial chemistry, small-molecule microarrays, fragment-based lead discovery and DNA-encoded chemical libraries.

In particular, combining ligands with further functions has been the subject of development. For example, in US patent 5,573,905 (Lerner et a!.; Encoded combinatorial chemical libraries) an encoded combinatorial chemical library is provided which comprises a plurality of bifunctional molecules according to the formula A-B-C, where A is a polymeric chemical moiety. B is a linker molecule operatively linking A and C, consisting of a chain length of 1 to about 20 atoms and preferably comprising means for attachment to a solid support, and C codes for the identification of the polymer A and attaching the code C to the polymer A with a linker molecule B allows the polymer to be identified. The solution presented is limited to a specific type of a chemical moiety. Also, each individual synthesis has to be carried out for each individual member of a chemical library.

Another strategy is described in US patent application 2012053091 (Wagner, R.; Methods of Creating and Screening DNA-encoded Libraries). A method of DNA template assisted synthesis of a chemical compound is provided, where the final composition of the chemical compound is encoded in the attached nucleic acid sequence. In another strategy described in US20040014090 (Neri et al.; Encoded self-assembling chemical libraries (ESACHEL)) the combinatorial assembly of DNA encoded small molecule libraries is described.

Summary of the invention

The present invention relates to identifying bispecific or bivalent ligands binding to a target molecule, more particularly to self-assembled bispecific or bivalent ligands comprising at least two separate ligands or two sub-libraries thereof, each ligand covalently attached via an optional spacer molecule to the ends of a nucleic acid sequence. The said ligands dimerise via the nucleic sequence and are subsequently covalently linked. The bispecific or bivalent ligands (or monomeric ligands) are provided as libraries and are screened for a specific binding interaction with a target molecule (e.g. a biological target). In a first embodiment according to the invention such a self-assembled complex is characterized by (a) a first nucleic acid sequence, covalently attached to a first ligand, comprising an identifier sequence, coding for the first ligand and a first optional spacer, and at least one dimerisation sequence, capable of performing a specific assembly reaction with (b) a second

complementary nucleic acid sequence, covalently attached to a second ligand via an optional second spacer molecule, also comprising a identifier sequence, and (c) functional moieties at the end of each nucleic acid opposite to the corresponding end with the ligand, which allow a chemical or enzymatic ligation reaction. The two nucleic acid sequences of the bivalent ligand complex are covalently crosslinked (fixated) after combinatorial self-assembly to yield a stable single nucleic acid sequence, which forms a scaffold in hairpin form, presenting the two individual ligands at the tip of each end. The bispecific or bivalent ligands that are thus formed are single molecules that allow to orientate the ligands in the three- dimensional space while simultaneously encode or incorporate the information of the ligands and each of the spacer molecule used.

As the bispecific or bivalent ligand complex constitutes a single molecule, it is ensured that bispecific or bivalent molecules are maintained and do not allow for

rearrangement of two nucleic arms. The bispecific or bivalent ligands of the current invention is also referred to as Self-Assembled Bivalent Ligand Complexes (SABLCs). A library of bispecific or bivalent ligands is made from combinations of two nucleic acid arms, each arm having a different ligand and spacer molecule (which spacer molecule may vary in size, i.e. it may also be absent and thus have no size). The said nucleic acid arms comprise sequence identifiers that allow to identify the ligand and spacer molecule. In the process of identifying a bispecific ligand, or a specific combination of ligands, as the information about the ligands and spacer molecule is contained in the said single molecule, it is ensured that all information about the bispecific or bivalent ligand complex is retained. Hence, the current invention provides for unique and highly efficient method to provide for bispecific or bivalent ligand libraries, that allow a time and cost efficient screening process and allow for a very efficient and fast deconvolution method that allows to highly reliably identify bispecific or bivalent ligands that bind to a specific target or several specific targets. The bivalent ligand libraries of the current invention, i.e. the SABLC libraries, provide for a significant improvement over the prior art as no method in the prior art provide for scaffolds that present different ligands and that allow flexibility with regard to combining a multitude of ligands while at the same time allow to identify each bispecific or bivalent ligand with regard to scaffold and ligands in a highly reliable fashion. Figures

Figure 1. A bivalent ligand: Self-assembled bivalent ligand complex (SABLC)

A self-assembled bivalent ligand complex, binding to a target molecule, comprises binding region (with the ligands), and optional spacer molecule, (spacer region) and an assembly region. The assembly region comprises one linear mostly self-complementary nucleic acid sequence, which forms a hairpin structure with a bulge in the middle. The assembly region functions first as a molecular scaffold and secondly encodes all necessary information about the binding and spacer region in form of two separate nucleotide sequences, opposite to each other in the bulge. Figure 2. Design of the nucleic acid arm, i.e. a monovalent ligand, also referred to as an individual Monovalent Precursor Molecule (MPM). One arm, a "left arm", that constitutes a bivalent ligand complex is shown. A SABLC comprises a combination of two of such precursor molecules (a "left" and a "right" arm). A specific or a plurality of ligands A (w) are covalently attached via an optional spacer molecule S1 (x) to a Dimerisation Sequence (DS1), which is adjacent to an Identifier Sequence IS (w, x), coding for the spacer molecule S1 (x) and the Ligand A(w). Towards the 5 ' terminus of the nucleic acid arm is a second universal Dimerisation Sequence (DS2). The DS1 in part or in whole may comprise a Primer Binding Site (PBS1). In some embodiments a short Extension of the Primer Binding Site (ExPBSI)) between PBS 1 and the Identifier Sequence (IS) is introduced, which allows strand specific enzymatic amplification of the "DNA Core" or "Assembly" region. Both, DS 1 and 2, are used to assemble and to stabilize the final self-assembled bivalent ligand complex (SABLC) via Watson-Crick base pairing. At the terminus is a functional group "F1" which, after combinatorial hybridisation to another "arm" with complementary universal sequences and a terminal group "F2" allows site specific covalent crosslinking either chemically or enzymatically. Figure 3. A Self-assembled bivalent ligand complex (SABLC) comprises a Binding Region, a Spacer Region and an Assembly Region. The later one comprises a first

Dimerisation Region DR1 to scaffold the ligands to form a SABLC specific Binding Region. Below the Bulge Region BR, which contains all information about the Binding and the Spacer Region, is a second Dimerisation Region DR2 which can be used to initiate chemical or enzymatic ligation or crosslinking between the functional groups in the Crosslinking Region CR. Parts of DR1 together with parts of BR can be used to allow annealing of sequence specific primer sequences to the left and right "arm" of the Assembly Region for later enzymatic amplification reaction to the left and right "arm" of the Assembly Region.

Figure 4. Schematic of how enzymatic ligation may be employed to site specifically crosslink ("fixate") one member "w" of MPM Sub-library A with another member "y" of MPM sub-library B. A(w) = ligand "w" of sub-library A, B(y) = ligand "Y" of sub-library B; S1 (x) stands for an optional first spacer molecule "x". S3(z) for an optional spacer molecule "z". DS1 stands for 1 st Dimerisation Sequence, DS2 for an optional 2nd Dimerisation Sequence. DS3 is partially or fully reverse complementary to DS1 and required to keep ligand "w" in a specific distance and orientation to ligand "y". IS (w, x) = Identifier Sequence coding for ligand "w" and spacer "x"; IS(y, z) = Identifier Sequence coding for ligand "y" and spacer "z"; "DS4" is optional and can, at appropriate design and conditions, assist the enzymatic ligation reaction by annealing partially or fully to the optional sequence DS2. "S2A" is a universal nucleotide sequence of 3 to 30 nucleotides. "S4-B" is a universal nucleotide sequence of 3 to 30 nucleotides. "P" stands for a terminal phosphate group at the 5'-terminus of "S2A" and "HO" for a functional hydroxy function at the 3' terminus S4-B. Both functional groups are essential to allow an enzymatic ligation reaction to take place. The universal "splint" oligonucleotide "LP" complementary to all or a part of the universal nucleotide sequences S2-A in a first MPM, adjacent to the unique identifier sequence IS(w, z) and an optional dimerisation sequence DS2 and the corresponding universal sequence S4-B adjacent to the unique identifier sequence IS(y, z) and optional dimerisation sequence DS4 of a second MPM initiates an enzymatic ligation reaction after annealing (hybridisation). A double arrow indicates areas where nucleotide sequences can hybridize via Watson-Crick base pairing.

Figure 5. Schematic of a classic "Click"-reaction with specific precursor molecules to generate a so called "biocompatible" non-phosphodiester linkage, which allows enzymatic polymerisation reactions. Exact protocols and biological efficacy have been described in literature. Figure 6. Schematic showing a few examples of individual monovalent precursor molecules, i.e. nucleic acid arms, bearing different oligomeric ligands.

a: Typically a defined oligomeric ligand "w" of Sub-library "A" with a secundary structure of sufficient stability is covalently attached via a spacer "S1 (x)" to a corresponding identifier sequence "w, x" flanked by an universal DNA sequence "DS1 "and an optional sequence "DS2", which together form the dimerisation region of a final self-assembled bivalent ligand complex (SABLC).

b: In certain embodiments covalent crosslinking within a oligomer is used to create stable hairpins or other more or less defined secundary structures in the ligand.

c: In the particular case a ligand of MPM is a nucleic acid or an analogue thereof self- complimentarity of the nucleic acid oligomers is used to create stable hairpins or other more or less defined secundary structures in a ligand "w" of Sub-library "A" .

d: In certain embodiments a certain number or positions "N" within an oligomeric ligand are not exactly defined, but randomized. This "sub-library" or "pool" approach can be used to create very large number of ligands which can be decoded in an iterative process, where selection, amplification and decoding is combined with synthesis of less and less

randomized pools.

Figure 7. Examples of molecules replacing the original nucleic acid scaffold:

a: After a certain SABLC was identified it is possible to replace the DNA based Assembly Region with the inert mirror image analogue (L-DNA). As the physical properties of an L- DNA oligomer are exactly the same as the natural D-DNA counterpart, the relative distance of a ligand s is also conserved. L-DNA monomers are commercially available and the oligomer synthesis follows standard procedures (see also: Hauser, N. C. et al. Nucleic Acids Research 34, 5101-1 1 , 2006)

b: The DNA assembly region of a SABLC can be replaced once the correct combination and distance between two ligand s has been determined. One simple example is to use a polyethylenglycol (PEG) - spacer of appropriate length as shown. This replacement also reduces the overall negative charge of the SABLC significantly.

Figure 8. a) Schematic Illustration of a hybridized Self-Assembled Bivalent Ligand Complex (SABLC) with a ligand "w" from MPM Sub-library A and a ligand "y" from MPM Sub- library B with optional spacer molecules "x" and "z" and optional Extensions of Primer Binding Sites (ExPBS) 1 and 2 in the bulge region. Dimerisation Sequences DS1 and optional DS2 in a first "arm" are hybridized to corresponding reverse complementary DS3 and optional DS4 to form the dsDNA stem in a second "arm". Also implemented in the bulge are the identifier sequences IS with sequences "w", "x" and "y" ,"z" respectively, which code for the spacer molecules "x" from Spacer Library S1 , and "z" from Spacer Library S1 and the Ligand s "w" and "y". b) Simplified illustration how the DNA scaffold might be replaced by a short Linker molecule "L" and how an optional reporter tag "T" can be introduced. Figure 9. Schematic of how a library of 96 individual monovalent precursor molecules of a first library A with ligands A1-A96 is combined with 96 individual monovalent precursor molecules of a 2nd library B with ligands B1-B96. All 96x96= 9216 combinations are formed, which in the next step will be all be simultaneously crosslinked to yield 9216 self-assembled bivalent ligand complexes (SABLCs). These combinations can be made in single tube as the bispecific or bivalent ligands according to the invention allow easy and simple deconvolution.

Figure 10. An example how each monovalent precursor molecule (MPM) 1-96 of Sub- library A and MPM 1-96 of Sub-library B can be synthesized individually on solid support using flow-through multi-well titerplates. Alternatively parallel synthesis like photolithography ink-jet synthesis can be employed.

Figure 1 1. Schematic of M-fold structures of Self-assembled bivalent ligand complexes (SABLCs) SEQ ID NOs. 9, 10 and 1 1. The ligand sequences or ligands are at the bottom of each scheme. As spacer molecules serve poly-Thymidin oligomers (Tn). In the right arm of the SABLC SEQ ID NO. 9 has no spacer, SEQ ID NO. 10 has a T6 and SEQ ID NO. 1 1 has a T12 spacer. In the simplified two-dimensional illustration the corresponding ligation site is in the small hairpin loop indicated with an arrow. The bulge region is clearly visible in the middle. The corresponding calculated duplex stabilities are around 26 kcal/mol. Figure 12. Schematic illustration of an example how Monovalent Precursor Molecules (MPM), i.e. the nucleic acid arms, are synthesized by a first conjugation reaction (c). In this case a conjugate between a oligonucleotide and a peptide is prepared via "native" ligation strategy called OPeC™. The necessary reagents a) and b) are commercially available Figure 13. Schematic illustration of how a ligand can be covalently attached to the 5' - end of a left nucleic acid arm of monovalent precursor molecules (MPMs =primary conjugates). In this case first a Cu(l) free Click-Reaction takes place between between an azido terminated ligand and the corresponding pre-activated octyn moiety takes place leaving the alkin-moiety at the 5'-terminus intact. The necessary reagents a) and b) are commercially available.

a) 5'-DBCO-TEG PHOSPHORAMIDITE for direct addition during solid phase synthesis of the nucleic acid arm of a monovalent precursor molecule (MPM); b) DBCO-SULFO-NHS ESTER for post-solid phase synthesis modification of amino- terminated oligonucleotides;

c) Illustration a how a ligand can be covalently attached to the 5' -end of a left nucleic acid arm of monovalent precursor molecules (MPMs =primary conjugates). In this case first a Cu(l) free Click-Reaction takes place between between an azido terminated ligand and the corresponding pre-activated octyn moiety takes place leaving the alkin-moiety at the 5'- terminus intact. The necessary reagents a) and b) are commercially available.

Figure 14. Capillary-electrophoretic analysis of bivalent ligands (self-assembled bivalent ligand complexes) using an Agilent Bioanalyzer 2100 system and DNA 1000 chip. Bivalent ligands corresponding to SEQ ID NOs. 1 1-18 are shown.

Figure 15. Standard Prothrombin Time (PT) and activated Partial Thromboplastin Time (aPTT) values for each individual ligand composition.

The values were determined by using human plasma samples. The benefit of combining two binding motifs is evident in the aPTT values. The aPTT value of a non Thrombin binding oligonucleotide sequence SEQ ID NO. 19 is fully identical to the reference (buffer) below 30 sec. SEQ NO. 7* and 8* indicate aptamers SEQ ID NO. 7 and 8 with a 3'-3'-T. Combinations of SEQ ID NO 7 and NO 8 lead to values above 40 sec depending on the distance of the two aptamers. NOs. 80-82 (G15D-TBA27) and NOs. 85-87 (TBA27-G15D) correspond to covalently conjugated aptamer motifs SEQ ID NO. 7 and SEQ ID NO 8 without (80 and 85) and with T6 (81 and 86) and T12 spacer (82 and 87). Similar aPTT values can be seen as confirmation of a stable dsDNA stem for the self-assembled bivalent ligand complexes (SABLCs) of two binding motifs with varied distances (SEQ ID NOs 9, 10 and 1 1).

Furthermore mutation of the binding motifs lead to decreased aPTT values (SEQ ID NO 13, 14 and 15).

Figure 16. Capture of bivalent ligands. The PCR detection limit for SABLC hairpins with single universal U18 primer appears to be about 10 fmol (femtomol). Selection of a SABLC library on DynaBeads coupled to 20 μg thrombin results in captured bivalent ligands in the low fmol range (2x wash) and much lower range (6x wash). Negative selection on beads blocked with ethanolamine results in very little (2x wash) and undetectable bivalent ligands (6x wash).

Figure 17. Relative qPCR and High-Throughput Sequencing Analysi A) Relative quantitative PCR, Act (TB-EA), values of the different bispecific or bivalent SABLCs show enrichment for SEQ NOs. 10 and 1 1 as well as for 14 and 15. From this data it can be concluded that bivalent ligand corresponding to SEQ NO. 10 with a T6 spacer is most efficient combination of binding motifs as it shows the highest enrichment. B) Relative quantitative analysis using lllumina high through-put sequencing analysis. Like for the qPCR shown in A), SEQ ID NO.10 shows most enrichment of all bivalent ligands from the library.

Figure 18. Schematic Illustration how bivalent ligand complexes can be assembled using two individual ligand presenting precursor molecules, each comprising unique identifier sequences, and a corresponding hairpin forming oligonucleotide comprising e.g. an optional third identifier sequence.

Figure 19. a) and b) Capillary-electrophoretic analysis of bivalent ligands (self-assembled bivalent ligand complexes) using an Agilent Bioanalyzer 2100 system and DNA 1000 chip. Bivalent ligands corresponding to SEQ ID NOs. 31-44 before (first lane) and after enzymatic sticky-end ligation with a third oligonucleotide (SEQ ID NO. 28, second lane respectively) are shown. The hybridisation and ligation reaction for each SABLC is performed in individual reaction vessels. Different retention times are due to size varying spacer length of SABLCs. All samples are raw products.

Figure 20. a) Capillary-electrophoretic analysis of bivalent ligands SEQ ID NOs. 31-44 are shown. Here hybridisation and ligation reaction for SABLC SEQ ID NOs. 31-42 were performed simultaneously in one reaction vessel (Lane 1). SEQ ID NOs. 43 and 44 are the raw products of hybridized and ligated SABLCs, where the third oligonucleotide functions as an identifier sequence and at the same time as an additional thrombin binding motif.

b) Capillary-electrophoretic analysis of PCR products of 12 different SABLCs SEQ ID NOs. 31-42 are shown. As expected the PCR product for all SABLC, irrespectively of the attached binding sequences, is identical in all cases and 132 nucleotides in length. There is no apparent difference regarding the yield.

Figure 21. Activated Partial Thromboplastin Time (aPTT) values for each individual ligand composition (background subtracted). The values were determined by using human plasma samples. The benefit of combining two binding motifs via the enzymatically ligated three component DNA scaffold is evident in the aPTT values and very similar to the previous design, where two individual precursor molecules are hybrized and directly conjugated via "click"-chemistry. If the space between the binding ligands is too short like in SEQ ID NO. 31 the effect of having both thrombin binding aptamers (SEQ ID NOs. 7 and 8) present is minimised and in the range of the separate ligands SEQ ID NOs. 7 or 8. If the spacing is increased to T6 like in SEQ ID NOs. 32 and 39 the aPTT values go further up with increasing relative distance of the binding motifs, but below the corresponding linear version (SEQ ID NO. 86). For a total of 12 Thymidines as a spacer, like in SABLC SEQ ID NOs. 33, 40 and 41 , the values are around 15-18 sec, whilst a comparable T12 spaced linear version (SEQ ID NO. 87) shows values around 21 sec. If the distance for ligands SEQ ID NOs. 7 and 8 attached to the scaffold is increased to 24 Thymidines (SEQ ID NO. 42) the aPTT value measured is around 22 sec compared to 31 sec of the corresponding linear version (NO. 88). As expected the trivalent SABLCs SEQ ID NOs. 43 and 44 show significantly higher aPTT values (27 and 37 sec) at the same relative distance of the ligands.

Definitions:

"Nucleic acids" include DNA (deoxyribonucleic acid) and RNA (ribonucleic acid) are made from monomers known as nucleotides. Each nucleotide has three components: a 5-carbon sugar, a phosphate group, and a nitrogenous base. If the sugar is deoxyribose, the polymer is DNA. If the sugar is ribose, the polymer is RNA. Nucleic acids with natural monomers and a few Nucleic Acids Analogues (XNA, Pinheiro, V. B. et al. Science 336, 341-344, 2012) with modified monomers can be replicated by natural enzymes or modified enzymes.

"Hybridization" is the process of establishing a non-covalent, sequence-specific interaction between two or more complementary strands of nucleic acids into a single complex, which in the case of two strands is referred to as a duplex. Oligonucleotides, DNA, or RNA will bind to their complement under normal conditions, so two fully reverse complementary sequences strands will bind to each other readily. However, to a certain extent also mispairing can occur depending on the conditions (e.g. relative concentration, temperature, salt concentration, pH etc.). "Annealing", in this context, means for reverse complementary sequences of single-stranded DNA or RNA to pair by hydrogen bonds to form a double-stranded polynucleotide.

"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. Typically Nucleic Acid aptamers are identified by a process called SELEX (Systematic Evolution of Ligands by Exponential enrichment).

"Polymerase chain reaction" (PCR) is a biochemical technology in molecular biology used to exponentially amplify a single or a few copies of a piece of DNA across several orders of magnitude, generating thousands to millions of copies of a particular DNA sequence.

"A covalent bond" is a chemical bond that involves the sharing of electron pairs between atoms. The stable balance of attractive and repulsive forces between atoms when they share electrons is known as covalent bonding. For many molecules, the sharing of electrons allows each atom to attain the equivalent of a full outer shell, corresponding to a stable electronic configuration. Covalent bonding includes many kinds of interactions, including o-bonding, ττ-bonding, metal-to-metal bonding, agostic interactions, and three- center two-electron bonds.

In general the term "ligand" and analogous terms (e.g. binding motif) include, but are not limited to amino acids, amino acid analogues, peptides, peptidomimetics, nucleosides, nucleotides, nucleotide analogues, polynucleotides, polynucleotide analogues such as peptide nucleic acids, proteins, carbohydrates, polycarbohydrates, metal complexes, receptors, enzymes, antibodies, lipids, lipoproteins, cofactors, drugs, pro-drugs, lectins, glycoproteins, non-bio polymers, sub-cellular structures, viruses, or portions thereof such as viral vectors and viral capsids, phages, or portions thereof such as phage vectors and phage capsids; cells, or portions thereof; and other biological or chemical materials that can be conjugated to the MPMs used for SABLCs. In principle any molecular entity which can be chemically attached to a DNA sequence can be tested for binding efficacy to the target molecule. A typical example for a defined molecule is a small molecule.

In one particular embodiment according to the invention the ligand or binding motif is covalently linked to the nucleic acid via a spacer molecule. The spacer molecule can be any bifunctional molecule that performs the function of operatively linking the ligand molecule to the nucleic acid sequence. The spacer molecule can vary in structure and length, and provide at least two features: (1) operative linkage to the ligand and (2) operative linkage to the nucleic acid. A typical spacer molecule is designed not to participate in a binding event of a ligand with target molecule.

As the nature of chemical linkages is diverse, any of a variety of chemistries may be utilized to effect the indicated operative linkages to both the chemical and DNA moieties, the nature of the linkage is not considered an essential feature of this invention. The size of the linker moiety in terms of the length between the chemical and DNA moieties can vary widely, but for the purposes of the invention, need not exceed a length sufficient to provide the linkage functions indicated. Thus, a chain length of from at least one to about 20 atoms is preferred.

A "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 analogues, cofactors, inhibitors, drugs, small molecules, dyes, nutrients, pollutants, growth factors, cells, tissues, or microorganisms and any fragment or portion of any of the foregoing. A target molecule also comprises cell surface molecules, such as a cell membrane protein.

Detailed description of the invention

In a first aspect of the invention, a bivalent ligand library (a library of self-assembled bivalent ligand complexes (SABLCs)) is provided, wherein each of the bivalent ligands of the library comprises:

- a first nucleic acid arm, comprising:

a) at its 3'-end a covalently conjugated first ligand;

b) optionally, a first spacer molecule, 5' from the first ligand,

c) a first primer binding site, 5' from the spacer molecule or first ligand;

d) a first dimerisation sequence, wherein the first primer binding site and first dimerisation sequence overlap in whole or in part;

e) a first ligand identifier sequence, 5' from the first primer binding site;

f) optionally, a first spacer molecule identifier sequence, 5' or 3' from the ligand identifier sequence;

g) a second dimerisation sequence, 5' from the identifier sequences of e) and f);

i) a first functional group at the 5'-end;

- a second nucleic acid arm, comprising"

a) at its 5'-end a covalently conjugated second ligand;

b) optionally, a spacer molecule, 3' from the first ligand,

c) a nucleic acid sequence complementary to the first dimerisation sequence and complementary in whole or in part to the first primer binding site, 3' from the second ligand; e) a second ligand identifier sequence, 3' from the said complementary nucleic acid sequence of c);

f) optionally, a second spacer molecule identifier sequence 3' or 5' from the second ligand identifier sequence,

g) a nucleic acid sequence complementary to the second dimerisation sequence in the first nucleic acid arm;

i) a second functional group at the 3'-end;

wherein the first and second dimerisation sequences of the first nucleic acid arm are base paired with the complementary sequences of the second nucleic acid arm and wherein the 5'-end of the first nucleic acid arm is covalently conjugated with the 3'-end of the second nucleic acid arm, and wherein the two covalently conjugated arms allowing amplification and/or sequencing of the conjugated nucleic acid sequence comprising at least the first ligand and optional first spacer molecule identifier sequence and the second ligand and optional second spacer molecule identifier sequence.

A SABLC, or a bivalent ligand according to the invention, comprises a binding region, an assembly region and optional spacer region in between (Figure 1). Preferably the assembly region is one linear mostly self-complementary natural or modified DNA sequence, which forms a hairpin structure with a bulge in the middle. The assembly region functions first as a scaffold and secondly encodes all information about the binding and spacer region in form of two separate short nucleotide sequences that comprise identified sequences for the ligands and spacer molecules opposite to each other in the bulge. The binding region of a bivalent ligand, a SABLC, is a combination of two identical ligands or two different ligands which are orientated in space relative to each other. The distance between the ligands is determined by the three-dimensional structure of the molecular scaffold, preferentially a B- form DNA double-helix, and the use of optional spacer molecules between a particular ligand and the assembly region. It is understood that it is also an option to have no spacer molecule.

One set of ligands is covalently attached to a nucleic acid arm optionally via a spacer molecule form a nucleic acid arm, also referred to as a Monovalent Precursor Molecule or MPM (Figure 2). Libraries of MPMs with 100-10,000 members may be generated using high- throughput synthesis methods.

A bivalent ligand, SABLC, is formed after annealing of the dimerisation sequences of a first and a second monovalent precursor molecule, followed by covalent conjugation either chemically or enzymatically (Figures 3 - 5, 18). It is understood that the second dimerisation region may be optional, e.g. when the nucleic acid arms are ligated enzymatically via a splint nucleic acid a second dimerisation region may not be required.

A SABLC can bind to one target, or two identical or two different target molecules.

The interaction with the target or targets can occur either at the same binding site, i.e. the moiety to which the ligands bind to is substantially the same, or at different binding sites, i.e. the moieties to which the ligands bind are substantially different. When the target is substantially the same, the ligands may be identical or may be different. An example for a target protein with two distinct different binding sites is the enzyme thrombin, a

multifunctional serine protease. Thrombin has two electropositive binding sites or exosites. One is the fibrinogen-binding site and the other is the heparin-binding site. For each of the exosites a specific ligand is known. In this particular case both motifs are short nucleic acid aptamers (a15-mer and a 29-mer). A bivalent aptamer complex comprising both of these two different ligands in distinct distance has been shown to increase the affinity by a factor 100 (Ahmad et al. Nucleic Acids Research 40, 1 1777-1 1783, 2012). Hence, the said aptamers for thrombin bind a single target molecule, wherein the target molecule comprises two distinct moieties. A target molecule hence may comprise different target moieties, each one targeted by a different ligand. A target molecule may also comprise two of the same target moieties albeit being at a different position within a target molecule.

A self-assembled bivalent ligand complex library composed according to the invention can comprise hundreds of millions of members, typically 1 ,000-100,000,000 or 10,000-100,000,000 members. Each member assembled differently with regard to first and second ligands and optional first and second spacer molecules, each member having a unique sequence identifier composition (i.e. for each specific combination of first and second ligand and first and second spacer molecule).

A first monovalent precursor molecule comprises a first nucleic acid arm covalently attached to a first ligand at its 3'-prime end. The ligand comprises a first ligand and an optional first spacer molecule between the ligand and a first nucleic acid arm.

A ligand can be, but is not limited to, one specific small molecule or an oligomeric sequence of small molecules such as nucleic acid or peptide aptamers. The spacer molecule can be, but is not limited to, an oligomer of a natural or non-natural nucleic acid, a peptide, a polysaccharide or polyethylenglycol (PEG) molecule (Figure 7). Examples for non-natural nucleic acids are peptide nucleic acids, polymorpholinos and oligomers of mirror-image nucleic acids ("Spiegelmers"). In case of a nucleic acid or non-natural nucleic acid spacer the length preferably may be 1-30 nucleotides. When a peptide spacer is selected, preferably it is 1-30 amino acids in length, whilst for a PEG spacer (CH2-CH2-0)n the length preferably is 1-20 or 1-100 or an alkyl linker (CH2)n- , where n is an integer from 1 to 50.

The first nucleic acid arm comprises a first dimerisation sequence at its 3'-terminus. Preferably the first dimerisation sequence is a stretch of 10-30 nucleotides. This first dimerisation sequence also may overlap with the first primer binding site. The first primer binding site may be shorter in length than the first dimerisation sequence towards the 3'-end. Optionally there is an extension of a few nucleotides of a first primer binding site of 1 -20 nucleotides, preferably 1-5 nucleotides, adjacent to the 5'-end of the first dimerisation sequence. This extension is not part of the first dimerisation sequence. The first part of the primer binding site and the optional extension of the primer binding site in the first nucleic acid arm allow binding of a primer to allow for enzymatic amplification and/or sequencing of the identifier sequences. Preferably, the first and the optional extension of the primer binding sequence together are 15-25 nucleotides in length.

Towards the 5'-end of the first nucleic acid arm are 2-35 nucleotides, which form a first binding region identifier sequence. The first binding region identifier sequence contains a short sequence coding for the ligand, i.e. a ligand identifier sequence, of 1-25 nucleotides and also comprises a sequence of 1-10 nucleotides, coding for the spacer molecule, i.e. a spacer molecule identifier sequence. Preferentially the monomers used for the identifier sequence may be natural (A,C,T,G,U) nucleotides or non-natural as described for the assembly region. In one embodiment the ligand identifier sequence is encoded in a sequence of 3-7 nucleotides and the corresponding spacer molecule in a sequence of 2-5 nucleotides. It is understood that the minimal requirements with regard to sequence identifiers is determined by the number of ligands and/or spacer molecules. For example, with 10 ligands and 10 different spacers 100 unique combinations can be made. This would require at least 4 nucleotides to allow for a unique sequence identifier for each combination (4x4x4x4 = 256).

In the self-assembled bivalent ligand complex the optional extension of the first Primer Binding Site (ExPBSI), first ligand identifier sequence IS(w) and first spacer identifier sequence IS(x) in the left arm and the second Primer Binding Site (ExPBS2), second ligand identifier sequence IS(y) and second spacer identifier IS(z) in the right arm are part of a bulge in the mostly self-complementary assembly region (Figure 3).

The first ligand identifier sequence is followed at the 5'-end by an second

dimerisation sequence of 5-20 nucleobases in length. This second dimerisation sequence may be optional. In a preferred embodiment the second dimerisation sequence DS2 is used to position the 5'-terminus of a first nucleic acid arm via hybridisation in close proximity to the 3'-terminus of a second nucleic acid arm to allow a chemical ligation.

Optionally, there is an additional spacer nucleic acid sequence in the first nucleic acid arm of 5-20 nucleotides, 5' from the second optional dimerisation sequence. The second spacer nucleic acid sequence can be used to position the 5'-end of a first nucleic acid arm to the 3'-end of a second nucleic acid arm in such a way to allow enzymatic ligation. In a preferred embodiment this is achieved with a short nucleic acid sequence of 10-40 nucleotides in length, which is reverse complementary to the second spacer nucleic acid sequence at the 5'-end of a first nucleic acid arm and an additional spacer nucleic acid sequence at the 3'-end of the second nucleic acid arm. Such short separate nucleic acid sequence is used as a splint to increase the efficiency of a enzymatic ligation reaction (Figure 4). When a splint is used, the 5'-end may comprise a triphosphate group to facilitate ligation.

At the 5'-terminus of the first nucleic acid arm is a first functional group, which allows chemical or enzymatic ligation to a second nucleic acid arm of a second monovalent precursor molecule, with a suitable second reactive functional group at the 3'-terminus. In a preferred embodiment according to the invention the first functional group is an Azide-group (Figure 4).

The second MPM (Fig. 4), i.e. the second nucleic acid arm comprises a second ligand B(y) that is covalently attached the 5'-end of the second nucleic arm with an optional third spacer molecule S3(z) between the second ligand B(y) and the 3'-end of the nucleic acid. The second nucleic acid arm comprises a second spacer molecule identifier sequence IS(z), which codes for the third spacer molecule S3(z) and further to the 3' terminus a ligand identifier sequence IS(y).

Optionally the second nucleic acid arm encodes a second primer binding site PBS2. This sequence serves as a template to generate the second primer binding site after amplification from the first primer binding site. The first and second primer binding site may be identical. This allows for PCR amplification of the sequence comprised in between the first and second PBS as present in the bivalent ligand using only a single primer. The first and second primer binding site may not be identical, e.g. by selecting extension sequences for both primer binding sites in both arms that are not complementary (5' of PBS1 in the first arm and in the second arm 3' of PBS2). Such an extension sequence may be 1-5 nucleotides in length. Such an extension sequence allows to selectively bind a first primer to PBS1 and the second primer to PBS2. 5' from the dimerisation sequence and/or PBS2 and 3' from the ligand is an optional third spacer molecule S3(z). This allows amplification using a primer pair. A spacer molecule and ligand identifier sequence is 3' from the PBS2 sequence in the second nucleic acid arm. The second primer binding site is optional as in some embodiments linear enzymatic amplification is sufficient to allow identification of the first and second spacer and binding region identifier sequences. Hence, the second nucleic acid arm may comprises a nucleic acid sequence complementary to a second primer binding site, wherein the second primer binding site is 3' from the from the ligand and optional spacer molecule and 5' from the identifier sequences.

The second primer binding overlaps, in part or in whole, with a third dimerisation sequence DS3, which is substantially complementary to the first dimerisation sequence DS1 of the first nucleic acid arm.

Furthermore, 3' from the identifier sequences a sequence is provided that is substantially or in whole complementary to the second dimerisation sequence of the first nucleic acid arm. Said sequence is optional.

Finally at the 3'-terminus of the second nucleic acid sequence is second functional group F2 which allows a chemical or enzymatic crosslinking reaction with the first functional group F1 at the 5'-terminus of the first nucleic sequence.

The two nucleic acid arms, with their complementary dimerisation sequences, spacer molecules and ligands and functional groups at the end that allow ligation, allow the formation of a bivalent ligand via Watson-Crick base pairing. The said base-paired nucleic acid arms position the ligands in space, and also encode identifiers that are unique for each combination. Sub-libraries A and B (Fig. 4) of the first and second nucleic acid arms, i.e. precursor monovalent molecules, are mixed under appropriate buffer and temperature conditions to allow hybridisation they form partially double stranded complexes where dsDNA stem regions, formed by the universal DNA regions, are separated by a bulge region, where the two sets of spacer and ligand identifier sequences of each "arm" are located (Figure 3) Due to this spontaneous occurring self-assembly all possible

combinations of ligands are formed. If e.g. a library of 64 individual molecules of a first library is combined with 64 individual molecules of a 2nd library all 64x64= 4096

combinations are formed (Figure 3).

These assembled complexes, that are quite stable at physiological conditions, are further stabilized by a covalent ligation reaction wherein the 5'-end of the first nucleic acid arm is ligated to the 3'-end of the second nucleic acid arm. This "fixation" step ensures there is no later recombination possible and the information about the specific combination of ligands and their distance and position relative to each other is not lost during the storage, handling and decoding steps later in the process. This also allows sequencing of both arms via PBS1 , and also allows amplification of the two "arms" (linear via PBS1 , and a PCR via PBS1 and PBS2).

The link between the two arms needs to be a covalent conjugation which is biocompatible. Biocompatible means that the linkage between the two entities allows a polymerase to read through the ligation site such that the nucleic acid sequence from both arms is copied. Such a process occurs during an enzymatic amplification process with sufficient efficiency and fidelity, so that the identification of the identifier sequences IS (w, x) of the first nucleic acid arm and IS (y, z) of the second nucleic arm is not compromised.

Methods for chemical as well as enzymatic biocompatible ligation are well known in the art. In one embodiment, ligation of the two arms is achieved by using an alkyne moiety as a terminal functional group for each member of a first library and an azido group for each member of a second group to achieve a metal ion catalysed so called "Click" reaction as described in US201 1 105764A1 to form a biocompatible linkage (US8846883B2, Figure 5). Further chemical conjugation reactions can be used as well (as described in Xu and Kool, Nucleic Acids Research 27, 875-81 , 1999; Blackman et al., J. Am. Chem. Soc. 130, 13518- 13519, 2009; and Taylor et al., J. Am. Chem. Soc. 133, 9646-9649, 201 1) which are incorporated herein by reference). Alternatively enzymatic ligation (Figure 4) using a short complementary "splint" sequence can be used to covalently crosslink the left and the right nucleic acid "arm" (Lehman et al., Anal. Biochem. 239, 153-159, 1996).

Primary and secondary conjugation reactions are part of the modular synthesis approach to generate the SABLC libraries. A primary conjugation reaction is used for the synthesis of individual monovalent precursor molecules, i.e. nucleic acid arms. Nucleic acid sequence and ligand conjugated together form an individual monovalent precursor or the "arm" of a self-assembled bivalent ligand complex (SABLC). These "arms" are covalently crosslinked during a second conjugation reaction after random self-assembly took place by hybridisation of their assembly DNA regions. In Figure 1 1 a schematic two-dimensional illustrations how such bivalent binding complexes, or bivalent ligands according to the invention are folded are shown.

Conjugation between e.g. amino terminated nucleic acid sequences and peptides can be achieved using standard techniques used for the synthesis of peptide linkages. See, e.g., Bodanszky, Principles of Peptide Synthesis, 2nd Ed. (1993). These techniques include, but are not limited to azide coupling; anhydride method using compounds such as carbocyclic acids derivatives, phosphorous and arsenious acids derivatives, phosphoric acids derivatives, acyloxyphophonium salts, sulphuric acid derivatives, thiol acids, and carbodiimide; and methods using active esters such as active aryl and vinyl esters and reactive hydroxylamine derivatives (Hamm et al., Chemistry, 14(2), 320-330, 2003).

In one embodiment a "native ligation" strategy called OPeC™ is used to generate nucleic acid-pept(o)ide conjugates. (Figure 12). This conjugation reaction is based on the "native ligation" of an N-terminal thioester-functionalised peptide to a 5'-cysteinyl

oligonucleotide as 0-trans-4-(N-a-Fmoc-S-tert-butylsulfenyl-l-cysteinyl)aminocyclohexyl 0-2- cyanoethyl-N,N-diisopropylphosphoramidite functions as the oligonucleotide modifying reagent (OMR) and is used in the final coupling step in standard phosphoramidite solid- phase oligonucleotide assembly. Deprotection with aqueous ammonia solution generates in solution 5'-S-tert-butylsulfenyl-L-cysteinyl functionalised oligonucleotides with another functional group a the 3'-end for the secundary conjugation reaction further downstream in the assembly process. The corresponding peptide or peptide analogue is modified with Pentafluorophenyl S-benzylthiosuccinate, the peptide modifying reagent (PMR) in the final coupling step in standard Fmoc-based solid-phase peptide assembly. Deprotection with trifluoroacetic acid generates, in solution, peptides substituted with an N-terminal S- benzylthiosuccinyl group. The thiobenzyl terminus of the modified peptide is converted to the thiophenyl analogue by the use of thiophenol, whilst the modified Oligonucleotide is reduced using the tris(carboxyethyl)phosphine. Coupling of these two intermediates, followed by the "native ligation" step, leads to formation of the Oligonucleotide-Pept(o)ide Conjugate. This method is orthogonal to a Diels-Alder cycloaddition and therefore is compatible to allow a "Click-reaction" further downstream in the assembly process for a secundary conjugation reaction to covalently crosslink a left with the right "arm" of a MPM.

In another embodiment 5'-DBCO-TEG-phosphoramidite (10-(6-oxo-6- (dibenzo[b,f]azacyclooct-4-yn-1-yl)-capramido-N-ethyl)-0-triethyleneglycol-1-[(2-cyanoethyl)- (Ν,Ν-diisopropyl)]- phosphoramidite) (Figure 13 ) from GlenResearch, UK, (Catalog Number: 10-1941 -xx) is used to modify an oligonucleotide at the 5' -end during standard solid support phosphoramidite based synthesis (Marks et al., Bioconjugate Chem., 22 (7), 1259-1263, 2011). This modification allows a non-catalyzed "Click"- reaction and to generate primary conjugates using a first click-reaction with specific highly reactive reaction partners and leaves a second set of reactants intact for a second "Click"-reaction further downstream in the assembly process to covalently crosslink a left with the right "arm". Protected alkyne- modifying phosphoramidites for multiple "Click"- reactions are commercially from BaseClick GmbH/Germany.

Examples for "double-click" reactions have been described in literature (Neves et al., Bioconjugate Chem., 24 (6), 934-41 , 2013). Here, in a first step, an inverse electron demand Diels-Alder (IEDDA) reaction is followed by a standard copper catalyzed Huisgen [3+2] cycloaddition ("Click"-reaction). There a several examples for IEDDA using Tetrazine- Norbornene conjugation described in literature (Blackman et al., J. Am. Chem. Soc, 130 (41), 13518-13519, 2008; Alge et al., Biomacromolecules 14, 949-53, 2013).

For other molecules, conjugates can be formed using suitable chemical and biological reactions known to those of ordinary skill in the art. For example, molecules that contain reactive groups such as, but not limited to, amino, hydroxyl, sulfhydryl, phenolic, and carboxyl groups can readily provide bonds such as amide, ester, sulfide, disulfide, and thioester bonds when contacted under suitable conditions with other reactive moieties. See generally, Smith, March's Advanced Organic Chemistry: Reactions, Mechanisms, and Structure, 5^th Ed. (2001).

Conjugation can be effected by other methods including, but not limited to, alteration in environmental conditions (e.g., temperature, pH and buffer), and/or addition of compounds or molecules that catalyze the formation of a chemical bond (e.g. cross-linking agents). Cross-linking agents can be used to introduce, produce, or utilize reactive groups such as thiols, amines, hydroxyls, and carboxyls, which can then be contacted with other molecules that contain reactive groups to form a bond between the reactive groups. These agents can be used directly or indirectly through a linker to form a conjugate between a molecule to be arranged and a nucleic acid fragment.

Conjugation may be heterofunctional or homofunctional. Examples of

heterofunctional conjugation include, but are not limited to: carboxy to amino conjugation using diisopropylcarbodiimide (DIC), disuccinoylcarbonate (DSC), or carbonyldiimidazol (CDI) activators; phosphate-to-amino conjugation using DIC, DSC, or CDI activators; thiol-to- amino conjugation; and aldehyde terminated polymer to aminooxy terminated polymer using methods described in, for example: Tomoko et al., Bioconjugate Chem., 14(2): 320-330, 2003; Crisalli et al., Bioconjug.Chem. 23(9), 1969-1980, 2012; Katajisto et al., Curr. Prot. Nucleic Acid Chem. Chapter 4, Unit 4.6., 2005.

A particular conjugation is thiol-to-amino conjugation using a heterobifunctional cross-linking agent. Agents that can be used for this purpose include, but are not limited to: 4-succinirnidyloxycarbonyl-methyl-a-(2-pyridyldithio)toluene (SMPT); 4-sulfosuccinimidyl-6- methyl-a-(2-pyridyldithio)toluamido-hexanoate (Sulfo-LC-SMPT); N-(k- maleimidoundcanoyloxy)sulfosuccinimide ester (Sulfo-KMUS); succinimidyl-4-(N- maleimidomethyl)cyclohexane-1-carboxy-(6-amidocaproate) (LC-SMCC); N-k- maleimidoundecanoic acid (KMUA); sulfosuccinimidyl-6-[3-(2-pyridyldithio)- propionamido]hexanoate (Sulfo-LC-SPDP); succinimidyl-6- [3 -(2-pyridyldithio)- propionamido]hexanoate (LC-SPDP); succinimidyl-4-(p-maleimidophenyl)butyrate (SMPB); sulfosuccinimidyl-4-(p-maleimidophenyl)butyrate (Sulfo-SMPB); succinimidyl-6-([beta]- maleimidopropionamido)hexanoate (SMPH); sulfosuccinimidyl-4-(N- maleimidomethyl)cyclohexane-1-carboxylate (Sulfo-SMCC); succinimidyl-4-(N- maleimidomethyl)cyclohexane-1-carboxylate (SMCC); N-succinimidyl(4- iodoacetyl)aminobenzoate (SIAB); N-sulfosuccinimidyl(4-iodoacetyl)aminobenzoate (Sulfo- SIAB); N-(g-maleimidobutyryloxy)sulfosuccinimide ester (Sulfo-GMBS); N-(g- maleimidobutyryloxy)succinimide ester (GMBS); m-maleimidobenzoyl-N-hydroxysuccinimide ester (MBS); m-maleimidobenzoyl-N-hydroxysulfosuccinimide ester (Sulfo-MBS); (N-e- maleimidocaproyloxy)sulfosuccinimide ester (Sulfo-EMCS); (N-e- maleimidocaproyloxy)succinimide ester (EMCS); N-e-maleimidocaproic acid (EMCA); N- succinimidyl-(4-vinylsulfonyl)benzoate (SVSB); N-([beta]-maleimidopropyloxy)succinimide ester (BMPS); N-succinimidyl-3-(2-pyridyldithio)-propionamido (SPDP); succinimidyl-3- (bromoacetamido)propionate (SBAP); N-[beta]-maleimidopropionic acid (BMPA); N-[alpha]- maleimidoacetoxy-succinimide ester (AMAS); N-succinimidyl-S-acetyl-thiopropionate (SATP); and N-succinimidyl iodoacetate (SIA). These agents are commercially available, or can be synthesized using methods known in the art.

Examples of homofunctional conjugation include, but are not limited to, thiol-to-thiol conjugation and amino-to-amino conjugation. Agents that can be used to provide thiol-to- thiol conjugate include, but are not limited to: bis-((N-iodoacetyl)piperazinyl)

sulfoerhodamine; 1 ,4-di-[3'-(2'-pyridyldithio)-propionamido]butane (DPDPB); 1 , 1 1-bis- maleimidotetraethyleneglycol (BM[PEO]4); bis-maleimidohexane (BMH); 1 ,8-bis- maleimidotriethyleneglycol (BM[PEO]3); 1 ,6-hexane-bis-vinylsulfone (HBVS); dithio-bis- maleimidoethane (DTME); 1 ,4-bis-maleimidobutane (BMB); 1 ,4-bis-maleimidyl-2,3- dihydroxybutane (BMDB); and bis-maleimidoethane (BMOE). These agents are

commercially available, or can be synthesized using methods known in the art.

Agents that can be used to provide amino-to-amino conjugate include, but are not limited to: glutaraldehyde; bis(imido esters); bis(succinimidyl esters); diisocyanates; and diacid chlorides. In addition, fixatives such as, but not limited to, formaldehyde and glutaraldehyde may be used to provide amine-amine crosslinking. Other amine-amine conjugation agents include, but are not limited to: ethylene glycol bis(succinimidylsuccinate) (EGS); ethylene glycol bis(sulfosuccinimidylsuccinate) (Sulfo-EGS); bis-[2- (succinimidooxycarbonyloxy)ethyl]sulfone (Sulfo-BSOCOES); bis-[2- (succinimidooxycarbonyloxy)ethyl]sulfone (BSOCOES); dithiobis(succinimidylpropionate) (DPS); 3,3'-dithiobis(sulfosuccinimidylpropionate) (DTSSP); dimethyl 3,3'- dithiobispropionimidate-2HCI (DTBP); disuccinimidyl suberate (DSS); bis(sulfosuccinimidyl) suberate (BS3); dimethyl suberimidate-2HCI (DMS); dimethyl pimelimidate-2HCI (DMP); dimethyl adipimidate-2HCI (DMA); disuccinimidyl glutarate (DSG); methyl N-succinimidyl adipate (MSA); disuccinimidyl tartarate (DST); disulfosuccinimidyl tartarate (Sulfo-DST); and 1 ,5-flouro-2,4-dinitrobenzene (DFDNB). These agents are commercially available, or can be synthesized using methods known in the art.

The nucleic acid arms are assembled during random or combinatorial hybridisation based on the complementarity between universal stretches of the nucleic acid sequences (dimerisation sequences) in the assembly regions of each library member. Suitable conditions that would cause a stable complex between two nucleic acid arms with complementary sequences may be employed for the hybridization. Those conditions are known to those skilled in the art, and can be found, for example, in Current Protocols in Molecular Biology, John Wiley & Sons, N.Y. (1989), 6.3.1-6.3.8. Hybridization conditions will vary depending upon the specific physical characteristics of MPMs such as e.g. solubility. Additional hybridization methods and conditions can be found in WO0202823(A2) and references cited therein.

Typically combinatorial hybridisation is performed with all members of the libraries present followed by a chemical or enzymatic conjugation reaction which permanently links the two assembled members together (Fig. 9).

The monomers used for the dimerisation sequences may be natural (A,C,T,G,U) nucleotides 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 mirror-image 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 phosphorothioates and phosphotriester-, alkylphosphonate internucleotide linkage. In certain embodiments fully synthetic self- assembling synthetic polymers might be used (Pinheiro et al. Science 336, 341-344, 2012). In one embodiment fixation of assembled nucleic acid arms takes place by a second chemical conjugation reaction to form a biocompatible non-phosphodiester based linker via a Cu-ion catalyzed "Click' -reaction (El-Sagheer et al. Chem. Soc. Rev. 39, 1388-405, 2010; Proc. Nat. Acad. Sci. USA 108, 1 1338-43, 201 1 ; US8846883B2). The site specificity of the "Click" reactions is induced either by a double-stranded

dimerisation region of sufficient length, which is part of the assembly region of the secundary conjugates (Figure 3) or by an external "splint" sequence which brings the termini in the assembly region of the bivalent ligands in close proximity (Figure 4).

The triazole phosphodiester mimic described above has the considerable advantage of being constructed from oligonucleotides made entirely by the phosphoramidite method, one bearing a 5'-azide functional group and the other a 3'-alkyne. The functionalised resin required for the solid-phase synthesis of oligonucleotides terminating with 3'-propargyl-(5- Me)-dC cytosine equivalent, is commercially available and achieves high coupling yields and produce up to 100-mer oligonucleotides of the purity required for efficient click ligation.

Another option is to use reverse phosphoramidites and 3'-propargyl dT as described by El- Sagheer et al.The 5'-azide group was introduced in a 2-stage process; the 5'-OH group of a normal support-bound oligonucleotide was first converted to 5'-iodo by reaction with methyltriphenoxy-phosphonium iodide (for oligonucleotides with 5'-dT this was simplified by direct incorporation of 5'-iodo thymidine phosphoramidite, then the resultant 5'-iodo oligonucleotides are reacted with sodium azide to complete the transformation.

Oligonucleotides functionalised with both 3'-alkyne and 5'-azide are made by performing oligonucleotide synthesis on 3'-propargyl-(5-Me)-dC resin then converting the 5'-terminus to azide as described above. It is possible to use the same methodology for other

combinations of nucleosides.

Alternatively, the 5'-end of a first nucleic acid arm and the 3'-end of a second nucleic acid arm may be covalently linked via an additional third nucleic acid, wherein the 5'-end and 3'-end of both arms hybridize and may have protruding ends, wherein the third nucleic acid comprises a complementary sequences and a loop sequence that allow formation of a hairpin structure, and, when applicable a protruding end compatible with the hybridized 5'- end and 3'-end of the first and second arm. Hence, the first and second nucleic acid arm combined and the third nucleic acid each form a structure that allows enzymatic ligation, via classic sticky ends ligation, or alternatively via a blunt ended ligation. Such an alternative method of combining the two arms is schematically depicted in figure 18.

Hence, in another embodiment a bivalent ligand library is provided wherein each of the bivalent ligands of the library comprises:

- a first nucleic acid arm, comprising:

a) at its 3'-end a covalently conjugated first ligand;

b) optionally, a first spacer molecule, 5' from the first ligand,

c) a first primer binding site, 5' from the spacer molecule or first ligand;

d) a first dimerisation sequence, wherein the first primer binding site and first dimerisation sequence overlap in whole or in part; e) a first identifier sequence, 5' from the first primer binding site;

f) a second dimerisation sequence, 5' from the identifier sequence of e); and h) a phosphate group at the 5'-end;

- a second nucleic acid arm, comprising:

a) at its 5'-end a covalently conjugated second ligand;

b) optionally, a spacer molecule, 3' from the first ligand,

c) a nucleic acid sequence complementary to the first dimerisation sequence and complementary in whole or in part to the first primer binding site, 3' from the second ligand; e) a second identifier sequence, 3' from the said complementary nucleic acid sequence of c); and

f) a nucleic acid sequence complementary to the second dimerisation sequence in the first nucleic acid arm;

- a third nucleic acid, comprising:

a) a phosphate group at its 5' terminus;

b) a nucleic acid sequence at the 5'-end complementary to a nucleic acid sequence at the 3'-end; and

c) optionally, a third identifier sequence between the 5' and the 3'-end;

wherein the first and second dimerisation sequences of the first nucleic acid arm are base paired with the complementary sequences of the second nucleic acid arm; and wherein the 5'-end of the first nucleic acid arm and the 3'-end of the third nucleic acid is covalently conjugated and the 5'-end of the third nucleic acid and the 3'-end of the second nucleic acid is covalenty conjugated, allowing amplification and/or sequencing of the conjugated nucleic acid sequences comprising at least the first, the second and the optional third identifier sequence. In this embodiment, each nucleic acid arm and the third nucleic acid can comprise a sequence identifier sequence. The sequence identifiers combined may allow to uniquely identify each bivalent ligand from the bivalent ligand library. Furthermore, the third nucleic acid may also comprise an additional ligand, e.g. an aptamer sequence or the like included in the third nucleic acid. Hence, the third nucleic acid may not solely serve as a linker that allows the covalent joining of the two arms, but may also include further functionality such as identifier sequences, ligands etc.

Hence, the bivalent ligand library according to the invention that included a third nucleic acid, the third nucleic acid comprising a third ligand, said ligand preferably being a nucleic acid. Also the bivalent ligand library according to the invention that included a third nucleic acid, may comprise a nucleic acid that is comprised in the third nucleic acid that is functioning as an enzyme. The bivalent ligand library according to the invention that included a third nucleic acid, may have the third nucleic acid comprise one or more functional groups selected from the group consisting of an attachment moiety allowing covalent attachment to another molecule or a surface, a fluorescently labelled moiety, a radioisotope tag, a moiety which is as a substrate for an enzymatic reaction.

In one embodiment, the bivalent ligand library according to the invention comprises randomly combined first and second nucleic acid arms. In a further embodiment, the first and/or second ligand as comprised in the bivalent ligand library according to the invention is a nucleic acid molecule. Preferably, said first and/or second ligand is a nucleic acid molecule with randomized positions.

In another embodiment, the said spacer molecules of the nucleic arms are nucleic acid sequences. This is advantageous as it allows to generate each nucleic arm using predominantly one synthesis method, i.e. the assembly region and spacer region can all consist of nucleic acids.

The first and second nucleic acid arms that are comprised in the bivalent ligand library according the invention may each be selected from a library of ligands. The library may be the same library. The first and second nucleic acid arm may also comprise ligands wherein each ligands is selected from a different ligand library. A first and/or second nucleic acid arm may also be a single ligand, or a library of ligands known to bind to a specific target. Hence, ligands may be selected for binding to a target molecule.

Methods to generate bivalent ligand libraries

Furthermore, the current invention provides for methods that generate bivalent ligand libraries, comprising the steps of:

a) providing a first monovalent ligand library comprising the first nucleic acid arms according the invention, and a second monovalent ligand library comprising the second nucleic acid arms according to invention;

b) mixing the selected monovalent ligand libraries and allowing the first and second nucleic acid arms to hybridize;

c) covalently conjugating the 5'-end of the first nucleic acid arm with the 3'-end of the second nucleic acid arm, thereby providing a bivalent ligand library.

In another embodiment, a bivalent ligand library is generated comprising the steps of:

a) providing a first monovalent ligand library comprising the first nucleic acid arms according to invention, and a second monovalent ligand library comprising the second nucleic acid arms according to invention;

b) providing a target molecule;

c) contacting the first monovalent ligand library with the target molecule in a first contacting step; d) contacting the second monovalent ligand library with the target molecule in a second contacting step;

e) selecting first and second monovalent ligands interacting with the target molecule from the first and second contacting steps;

f) mixing the selected monovalent ligands and allowing the first and second nucleic acid arms to hybridize;

g) covalently conjugating the 5'-end of the first nucleic acid arm with the 3'-end of the second nucleic acid arm, thereby providing a bivalent ligand library. In another embodiment, a bivalent ligand library is generated by:

a) providing a first monovalent ligand library comprising the first nucleic acid arm according to the invention, and a second monovalent ligand library comprising the second nucleic acid arm according to the invention;

b) providing a first and a second target molecule;

c) contacting the first monovalent ligand library with the first target molecule in a first contacting step;

d) contacting the second monovalent ligand library with the second target molecule in a second contacting step;

e) selecting first and second monovalent ligands interacting with the respective first and second target molecules;

g) covalently conjugating the 5'-end of the first nucleic acid arm with the 3'-end of the second nucleic acid arm, thereby providing a bivalent ligand library.

In another embodiment, a method is providing for preparing a bivalent ligand library comprising the steps of:

a) providing a first monovalent ligand library comprising the first nucleic acid arms according to any one of claims 2-14, a second monovalent ligand library comprising the second nucleic acid arms according to any one of claims 2-14, and the third nucleic acid according to any one of claims 2-14;

c) covalently conjugating the 5'-end of the first nucleic acid arm with the 3'-end of the third nucleic acid and covalently conjugating the 5'-end of the third nucleic acid arm with the

3'-end of the second nucleic acid arm, thereby providing a bivalent ligand library. In another embodiment, a method is providing for preparing a bivalent ligand library comprising the steps of:

b) providing a target molecule;

c) contacting the first monovalent ligand library with the target molecule in a first contacting step;

d) contacting the second monovalent ligand library with the target molecule in a second contacting step;

g) covalently conjugating the 5'-end of the first nucleic acid arm with the 3'-end of the third nucleic acid and covalently conjugating the 5'-end of the third nucleic acid arm with the 3'-end of the second nucleic acid arm, thereby providing a bivalent ligand library. In another embodiment, a method is providing for preparing a bivalent ligand library comprising the steps of:

a) providing a first monovalent ligand library comprising the first nucleic acid arm according to any one of claims 2-14, a second monovalent ligand library comprising the second nucleic acid arm according to any one of claims 2-14, and the third nucleic acid according to any one of claims 2-14;

b) providing a first and a second target molecule;

g) covalently conjugating the 5'-end of the first nucleic acid arm with the 3'-end of the third nucleic acid and covalently conjugating the 5'-end of the third nucleic acid arm with the 3'-end of the second nucleic acid arm, thereby providing a bivalent ligand library. Methods for selecting bivalent ligands binding to target molecule(s) In another embodiment, a method is provided for identifying a bivalent ligand binding to a target molecule comprising the steps of:

a) providing a bivalent ligand library according to the invention;

b) providing a target molecule;

c) contacting the target molecule with the bivalent ligand library;

d) selecting bivalent ligand interacting with the target molecule;

d) identifying bivalent ligands interacting with the target molecule.

Hence, first a bivalent ligand library according to the invention is provided. Said library may be provided in solution. Also target molecule is provided. Said target molecule preferably can identified such that binding of a bivalent ligand to said target molecule can be identified. For example, said target molecule comprises a label. The target molecule can also be provided on a solid support. For instance being covalently attached to a bead or an glass surface or in the form of a screening assay. Affinity chromatography can also be applied for both binding of the bivalent ligand to a ligand attached to a chromatography resin. Size exclusion chromatography as well a gel electrophoresis may be used for selecting bivalent ligands interacting with the target molecule.

The bispecific or bivalent ligands from the library are contacted (screened) with the target molecule. Thus, the absence, presence or amount of interaction between the molecules with the target molecules is determined and the molecules with an interaction are identified. Eluting the bivalent ligands or subjecting the bivalent ligands after being contacted with the target molecule in solution to electrophoretic processes may be used to also discriminate between bivalent ligands that interact with the target molecule(s) and bivalent ligands that do not or to a lesser extent. In this way a selection of bivalent ligands can be made.

The interaction between the molecules 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. Labels that may be selected 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.

The label can also 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 labeled 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.

The label 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 label. In the case where the label 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 label 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 label to a target molecule include covalent attachment of the label to the target molecule, non-covalent association of the label with another labeling agent component that is covalently attached to the target molecule, and covalent attachment of the label to a labeling agent component that is non-covalently associated with the target molecule.

In certain embodiments, the contacting step to determine an interaction of bivalent ligands with target molecule(s) is under buffer conditions and/or stringency conditions that allow the bivalent ligands in 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 molecule(s) to bivalent ligands and other parameters well known to those skilled in the art of modulating target molecule interactions with ligands, such as the bivalent ligands of the invention. 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 bivalent ligands are conducted.

Identifying bivalent ligands

Identification of bivalent ligands typically comprises or consists of the identification of the identifier sequences that are included in the bispecific or bivalent ligands of the invention, i.e. comprised e.g. in the SABLC's bulge region, to identify the two individual arms of the molecule and the ligands and molecule spacers comprised therein.

Preferably, the identifier sequences in the SABLC are amplified, i.e. can be amplified using enzymatic amplification techniques such as Polymerase Chain Reaction (PCR) or a linear amplification reaction by using a single primer binding site, i.e. only a first primer binding site. For example, the identifier can be amplified using a suitable primer or a suitable primer pair, enzymes such as a polymerase and deoxy-Nucleotide-Triphosphates (dNTPs). The identifier can subsequently be identified using sequencing, preferably using high throughput sequencing technology well available in the art such as Roche FLX or lllumina sequencing technology. High throughput sequencing in this context can mean the determination of at least a thousand, hundred thousand, or a million nucleotide sequences simultaneously. Hence, in the identification step the identifier sequences can be first amplified with at least a first primer capable of binding to the first primer binding site, wherein the amplification step is before the sequence determination step. In addition, the bivalent ligands of the bivalent ligand library may comprise a first and second primer binding site, wherein in the

identification step the identifier sequences are first amplified with a PCR amplification with a first primer capable of binding to the first primer binding site and a second primer capable of binding to the second primer binding site, wherein the amplification is before the sequence determination step.

In further embodiments, after the selection step and before the identification step the target molecule is removed and/or the ligands are removed. This may be advantageous as a target molecule and/or a ligand may have an effect on the subsequent identification step.

In another embodiment, wherein in a method according to the invention two target molecules are provided and the bivalent ligand library is contacted with the two target molecules either consecutively or simultaneously, and bivalent ligands are selected interacting with both of the two target molecules. As a bivalent ligand may interact with two different molecules, it may be advantageous to either simultaneously or separately select bivalent ligands.

In a further embodiment, a method is provided for preparing a bivalent ligand comprising the steps of:

- providing the information of an identified bivalent ligand obtained in a method for identifying a bivalent ligand according to the invention with regard to the first ligand, the first optional spacer molecule, the second ligand and optional second spacer molecule;

- preparing a bivalent ligand wherein the first ligand and optional first spacer molecule is linked with the second ligand and optional second spacer molecule via a linker molecule to replace the DNA scaffold as present in the identified ligand. In this embodiment, the scaffold as it is present in a bivalent ligand is to be replaced by a suitable linker molecule that presents the ligands and optional spacer substantially in the same way as the bivalent ligand that was identified. The scaffold that was used for presenting the ligands and optional spacer molecules and for identification purposes.no longer requires any of the features that make it suitable for identification. Hence, any linker molecule can be contemplated that provides only a presenting function comparable to the original DNA scaffold that was used in the library for identification purposes. Such a prepared bivalent ligand will have the following groups covalently linked in substantially the same way as in the identified bivalent ligand, having the first ligand and optionally first spacer molecule covalently linked with a linker molecule, which linker molecule is covalently linked with the second ligand and optional second spacer molecule. Such a linker molecule may be a nucleic acid, e,g, a short hairpin loop. Hence in another embodiment, a bivalent ligand is provided as obtained in said method. Such a bivalent ligand with such a linker molecule in place of the DNA scaffold. The DNA scaffold includes dimersation regions, primer binding sites, identifier sequences and the optional third oligonucleotide, the DNA scaffold does not include the first and second ligand and optional first and second spacer molecules. Such a bivalent ligand with a linker molecule may optimize the bioavailability of the complex e.g. as a therapeutic agent and may simplify preparation of the bivalent ligand. The linker molecule also include or be linked to a tag such as a fluorophore for diagnostic applications or an anchor group can be introduced for immobilisation on a solid support e.g. for chromatographic applications.

Examples Synthesis of nucleic acid arms, monovalent precursor molecules

The design of the nucleic acid arms is based on well known DNA aptamer ligands for Thrombin TBA27 (SEQ ID NO. 7) and G15D (SEQ ID NO. 8). DNA sequences SEQ ID NO. 1 and NO. 5 were synthesized via standard phosphoramidite chemistry using 5'- Dimethoxytrityl-3'-propargyl-N-succinoyl-long chain alkylamino-CPG, 5-methyl-2'- deoxyCytosine, commercially available from GlenResearch, UK. The 5'-azide group for SEQ ID NOs. 2, 3, 4 and 6 were introduced in a 2-stage process. First 5'-iodo thymidine phosphoramidite was directly added during synthesis without any changes to the standard method recommended by synthesizer manufacturer. Then the resulting 5'-iodo

oligonucleotides were reacted with sodium azide to complete the transformation. Cleavage of the oligonucleotide from this support requires 2 hours at room temperature with ammonium hydroxide and complete deprotection as required by the nucleobase protecting groups. SEQ ID Design 5'- Sequence 3'-end

end

NO

1 TBA27_FM3R_BC1 - GTCCGTGGTAGGGCAGGTTGGGGTGACAGA P

_Stem_Ppg CTGGTCCGCAAGTTCaaatccaACCTACCTAc

2 Az StemRC BC2 A tAGGTAGGTccctataGAACTTGCGGACCAGTCT - FM3Rrc_G15D GGTTGGTGTGGTTGG

3 Az StemRC BC3 A tAGGTAGGTctaatacGAACTTGCGGACCAGTCT - FM3Rrc_T6_G15D HUH GGTTGGTGTGGTTGG

4 Az StemRC BC4 A tAGGTAGGTaccttcaGAACTTGCGGACCAGTCT

FM3Rrc_T12

111111111111 GGTTGGTGTGGTTGG

_G15D

5 TBA27mut_FM3R_ - GTCCGTCCTACCGCAGCTTCCGTTGACAGAC P

BC5_Stem_Ppg TGGTCCGCAAGTTCaaatctcACCTACCTAc

6 Az StemRC BC6 A tAGGTAGGTtacacatGAACTTGCGGACCAGTCT

FM3Rrc_T12_G15D 111111111111 GTGTGTGTGTGTGTG

mut

7 G15D - GGTTGGTGTGGTTGG -

8 TBA27 - GTCCGTGGTAGGGCAGGTTGGGGTGAC -

Table 1 : Synthesized "Right" (SEQ ID NOs. 1 and 5) and "Left" (SEQ ID NOs. 2, 3, 4 and 6) monovalent precursor molecules. All sequences are listed left to right in 5' to the 3' orientation. Stretches of dimerisation sequences within the assembly region are underlined. Individual identifier sequences are annotated in small letters. Ligands, if present, are annotated in italics. P indicates a Propargyl group and A an Azido group.

To test the concept eight different bivalent ligand compositions were assembled by hybridisation of corresponding monovalent individual precursor molecules followed by "Click"-chemistry. Stretches of universal sequence within the assembly region are underlined. Individual identifier sequences are annotated in small letters. Binding motifs are annotated by cursive letters. Spacer sequences of 6 Thymidins and 12 Thymidins in sequence NO: 3, 4 and NO 6 were used.

Hybridization and ligation

The corresponding azide and alkyne oligonucleotides (100.0 nmol of each) in 0.2 M NaCI (100.0 μί) were annealed by heating at 90°C for 5 min and cooling slowly to room temperature. A solution of Cu-ion(l) click catalyst was prepared by adding the tris- hydroxypropyltriazole ligand (35.0 μηιοΙ) to sodium ascorbate (50.0 μηιοΙ in 0.2 M NaCI, 100.0 μΙ_) followed by the addition of CuS0₄x5H₂0 (5.0 μmol in 0.2 M NaCI, 50.0 μΙ_) under argon. The Cu-ion(l) solution was added to the annealed oligonucleotide mixture and kept at room temperature for 2 hr under argon. Reagents were removed by NAP-25 gel-filtration (GE Healthcare) and the ligated product was purified by Anion-Exchange HPLC as described below.

The self-assembled bivalent ligand complexes SEQ I D NOs. 9-16 were analysed by 10% PAGE gel electrophoresis and purified by Anion-Exchange HPLC on a Gilson H PLC system using a Resource Q anion-exchange column (6 mL volume, GE Healthcare). The H PLC system was controlled by Gilson 7.12 software, and the following protocol was used: run time, 16 min; flow rate, 5 mL per min; binary system. Gradient (time in mins (% buffer B)): 0 (0), 3 (0), 4 (40), 9.5 (82), 10 (100), 12 (100), 13 (0), 15.5 (0), 16 (0). Elution buffers: (A) 0.01 M aqueous NaOH , 0.05 M aqueous NaCI, pH 12.0; (B) 0.01 M aqueous NaOH , 1 M aqueous NaCI, pH 12.0. Elution of oligonucleotides was monitored by ultraviolet absorption at 295 nm. After HPLC purification oligonucleotides were desalted using a NAP-25 followed by a NAP-10 Sephadex column (GE Healthcare).

Yields of the purified products after H PLC were 45-57%.

MW

SEQ Combination g/mo

ID NO of SEQ Sequence name nt I ng/μΙ

9 1 +2 TBA27_FM3R_BC1_CS1_BC2_FM3Rrc_G15D 1 1 1 34466 10

10 1 +3 TBA27_FM3R_BC1_CS1_BC3_FM3Rrc_T6_G15D 1 17 36332 10

1 1 1 +4 TBA27_FM3R_BC1_CS1_BC4_FM3Rrc_T12_G15D 123 38133 10

12 1 +6 TBA27_FM3R_BC1_CS1_BC6_FM3Rrc_T12_G15Dmut 123 38132 10

13 5+2 TBA27mut_FM3R_BC5_CS1_BC2_FM3Rrc_G15D 1 1 1 34184 10

14 5+3 TBA27mut_FM3R_BC5_CS1_BC3_FM3Rrc_T6_G15D 1 17 36049 10

15 5+4 TBA27mut_FM3R_BC5_CS1_BC4_FM3Rrc_T12_G15D 123 37851 10

16 5+6 TBA27mut FM3R BC5 CS1 BC6 FM3Rrc T12 G15Dmut 123 37850 10

SEQ ID Sequences

NO

9 GTCCGTGGTAGGGCAGGTTGGGGTGACAGACTGGTCCGCAAGTTCaaatccaACCTACCTA

ctAGGTAGGTccctataGAACTTGCGGACCAGTCTGGTTGGTGTGGTTGG 10 GTCCGTGGTAGGGCAGGTTGGGGTGACAGACTGGTCCGCAAGTTCaaatccaACCTACCTA ctAGGTAGGTctagtacGAACTTGCGGACCAGTCTTTTTTTGGTTGGTGTGGTTGG

1 1 GTCCGTGGTAGGGCAGGTTGGGGTGACAGACTGGTCCGCAAGTTCaaatccaACCTACCTA

[ctlAGGTAGGTaccttcgGAACTTGCGGACCAGTCTTTTTTTTTTTTTGGTTGGTGTGGTTGG

12 GTCCGTGGTAGGGCAGGTTGGGGTGACAGACTGGTCCGCAAGTTCaaatccaACCTACCTA

[ctlAGGTAGGTtacacgtGAACTTGCGGACCAGTCTTTTTTTTTTTTTGTGTGTGTGTGTGTG

13 GTCCGTCCTACCGCAGCTTCCGTTGACAGACTGGTCCGCAAGTTCaggtctcACCTACCTA

[ct|AGGTAGGT c c c t a t a GAACTT GCGGACCAGT CTGGTTGGTGTGGTTGG

14 GTCCGTCCTACCGCAGCTTCCGTTGACAGACTGGTCCGCAAGTTCaggtctcACCTACCTA

[ctlAGGTAGGTctagt a cGAACTTGCGGACCAGTCTTTTTTT GGTTGGTGTGGTTGG

15 GTCCGTCCTACCGCAGCTTCCGTTGACAGACTGGTCCGCAAGTTCaggtctcACCTACCTA

[ctlAGGTAGGTaccttcgGAACTTGCGGACCAGTCTTTTTTTTTTTTT GGTTGGTGTGGTTGG

16 GTCCGTCCTACCGCAGCTTCCGTTGACAGACTGGTCCGCAAGTTCaggtctcACCTACCTA

[ctlAGGTAGGTtacacgtGAACTTGCGGACCAGTCTTTTTTTTTTTTTGTGTGTGTGTGTGTG

Table 2: List of self-assembled bivalent ligand complexes:

Monovalent precursors SEQ ID NOs. 1-6 were hybridized in combinatorial fashion and "clicked" to yield 8 self-assembled bivalent ligand complexes SEQ ID NOs. 9-16 as listed above.

The self-assembled bivalent ligand complexes SEQ ID NOs. 9-16 were analysed by 10% PAGE gel electrophoresis and purified by Anion-Exchange HPLC on a Gilson HPLC system using a Resource Q anion-exchange column (6 mL volume, GE Healthcare). The HPLC system was controlled by Gilson 7.12 software, and the following protocol was used: run time, 16 min; flow rate, 5 mL per min; binary system. Gradient (time in mins (% buffer B)): 0 (0), 3 (0), 4 (40), 9.5 (82), 10 (100), 12 (100), 13 (0), 15.5 (0), 16 (0). Elution buffers: (A) 0.01 M aqueous NaOH, 0.05 M aqueous NaCI, pH 12.0; (B) 0.01 M aqueous NaOH, 1 M aqueous NaCI, pH 12.0. Elution of oligonucleotides was monitored by ultraviolet absorption at 295 nm. After HPLC purification oligonucleotides were desalted using a NAP-25 followed by a NAP-10 Sephadex column (GE Healthcare).

Yields of the purified products after HPLC were 45-57%.

Analysis of bivalent ligand complexes

M-fold analysis was carried out and the calculated structure fitted the secondary structure that was designed. Mass spectrometry analysis indicated the mass of the nucleic acid molecules, i.e. nucleic acid arms and bivalent ligands, corresponding to SEQ ID NOs. 1 -6 and 9-16 is as expected. The self-assembled bivalent ligand complexes SEQ ID NOs. 9-16 were dissolved in TE buffer (100 mM Tris pH 8.0, 1 mM EDTA). Their concentration was determined via UV absorption at 260 nm (NanoDrop, Thermo Scientific), then diluted to 10 ng/ul. 1 μΙ_ of each single-stranded DNA was loaded into an Agilent smallRNA chip and run on the Agilent BioAnalyzer 2100 capillary electrophoresis system (L= small RNA ladder). The self-assembled bivalent ligand complexes run faster than expected size (Figure 14). This is an indication for the stability of the self complementary assembly-region even under normal denaturing buffer conditions, which are sufficient to dissolve most of the secundary structures of RNA molecules.

PCR amplification and Sanger sequencing

Bivalent ligands were PCR amplified with a single primer U18 (SEQ ID NO. 17) that binds to the first primer binding site in the first arm and the second primer binding site encoded by the second arm. PCR products were analysed with gel electrophoresis and all produced the expected size.

Prior sequencing on a 96-capillary 3730x1 DNA Analyzer (Life Technologies) the individual samples were further processed using the corresponding BigDye® Direct Cycle Sequencing Kit and protocol. The identifier sequences were read-out correctly from both strands.

Indexing sequences that were introduced to allow next generation sequencing on an lllumina HiSeq Platform in a separate experiment after multiplexing were read-out correctly on the forward strand (not shown).

Clotting Experiments Set-up

To evaluate the inhibitory potency of individual nucleic acid ligand or bivalent complex thereof, we measured the clotting time of each sample containing only thrombin, the ligand or ligand complex and fibrinogen substrate in physiological buffer. The theory behind the experiment is that the mixture of sample becomes non fluidic when the fibrinogen is digested by thrombin. As a result, the different time points of this transition can be used as an indicator. Briefly, 1 μΙ_ of 10 uM thrombin and 1 μΙ_ of 100 μΜ monovalent or bivalent NA ligand were added to a disposable transparent plastic cuvette (Fisher Scientific) containing 200 μΙ_ buffer and then incubated for 15 min. Following that, 4 μΙ_ of 20 mg/mL fibrinogen was added, and samples in the cuvette were carefully examined by tilting the cuvette to record the time when the sample became nonfluidic. Each experiment was performed in tandem. A reaction mixture containing only thrombin and fibrinogen was always tested together with other samples as an internal standard. All clotting times were normalized based on the internal standard and compared with it.

To evaluate the feasibility of the self-assembled bivalent ligand complex as a potential anticoagulant reagent, Standard activated Partial Thromboplastin Time (aPTT, Langdell et al., J Lab Clin Med 41 , 637-647, 1953) and Prothrombin Time (PT, Quick et al., Am J Med Sci 190, 501-51 1 , 1935) values for each individual or ligand composition were determined by using human plasma samples. Procedures applied were those recommended by the supplier. For aPTT determination, 50 μΙ UCRP was pre-incubated at 37°C with a different amount of each ligand for 2 min; then 50 μΙ_ aPPT-L was added and incubated for another 200 sec. Next, 50 μΙ of pre-warmed CaCI₂ was added to initiate the intrinsic clotting cascade. Finally, the scattering signal was monitored until the signal was saturated. For PT

determination, 50 μΙ_ of UCRP was pre-incubated at 37°C with a different amount of each ligand or ligand complex for 2 min; then 50 μΙ_ of thromboplastin-L was added to initiate the extrinsic clotting cascade. Finally, the scattering signal was monitored until the signal was saturated. For the calculation of aPTT and PT, the end time was determined to be the point where scattering signal reached half maximum between lowest and maximum points. It was repeated twice, and each set of experiments was done with one batch of plasma (Kim et al., Proc. Nat. Acad. Sci. USA 105, 5664-9, 2008. The values were determined by using human plasma samples and are shown in figure 15. The benefit of combining two binding motifs is clearly evident in the aPTT values.

Selection, amplification and sequencing.

A model library was prepared using the bivalent ligand library with the SABLC sequences ID NOs. 9-16. SEQ ID NOs. 9-15 were mixed in 1 : 1 ratio and background DNA SEQ ID No. 16 (200 picomol) was added. The final mix comprised 99.125% background DNA and 0.125 % of each of the SEQ ID NOs. 9-15. 1 μΜ< M was the final concentration. Magnetic Dynabeads® MyOne Carboxylic Acid (Invitrogen, cat. NO 6501 1) were resuspended by rolling the vial for 30 min on a rotor at 20 rounds per minute, and 0.5 ml_ suspension are transferred to a new tube. The tube is placed close to a magnet for 1 min, the supernatant is removed and the beads are washed two times with 0.5 ml_ MEST buffer (50 mM 2-(N-morpholino)ethanesulfonic acid, 0.01 % Tween 20, pH 6.0). The beads are resuspended in 50 μΙ MEST buffer and activated by addition of 50 μΙ EDC (10 mg/mL N-(3- dimethylaminopropyl)-N'-ethylcarbodiimide hydrochloride, Sigma E6383) for 30 min at room temperature on a rotor at 10 rpm. The supernatant is removed, the beads are resuspended in 100 μΙ_ MEST buffer and split into two 50 μΙ_ aliquots. The first aliquot is incubated with 6 nmol corresponding to 200 microgram a-thrombin (Tb, Haematologic Technologies, HCT- 0020) in 250 μΙ MEST buffer (20 μΜ thrombin final), the second aliquot is blocked with 5 mM aminoethanol (EA) in 250 μΙ MEST buffer for 2.5 h on a rotor at 10 rpm at room temperature. The tubes are placed close to a magnet for 1 min, the supernatants are removed and the beads are each washed two times with 0.25 ml_ PBSMT buffer (137 mM sodium chloride, 2.7 mM potassium chloride and 10 mM sodium/potassium phosphate buffer solution pH7.4, Ambion AM9624, supplemented with 1 mM magnesium chloride and 0.01 % Tween 20). Thrombin-coupled (Tb) beads and negative control (EA) beads are each re-suspended in 160 μΙ_ PBSMT, and to each aliquot 40 μΙ of 5 μΜ (corresponding to 200 pmol) of the library are added (1 μΜ SABLC library final). The tubes containing the mixtures are incubated for 1 h on a rotor at 10 rpm at room temperature. The tubes are placed close to a magnet for 1 min, the supernatants are removed and the beads are each washed two times with 0.25 ml_ PBSMT buffer. The beads are boiled each in 100 μΙ_ of 1x heat-stable DNA buffer

(Roboklon) for 5 min at 95 °C in a thermocycler and cooled down to room temperature. The tubes are placed close to a magnet for 1 min, and 90 μΙ_ of the supernatants are transferred to new tubes for further analysis.

PCR amplification

The bead-selected (Tb and EA) members of the SABLC library were subjected to PCR amplification using the primer U 18 (SEQ ID NO. 17; BioLegio). PCR products were analysed prior and after the capture step. Results show that a stringent capture protocol involving extensive washing of beads coupled with a control protein or thrombin showed no detectable PCR product in the control, and a strong detectable signal with the thrombin coated beads (see figure 16).

Enrichment analysis The composition of the universal PCR products of bead-selected SABLCs (Tb and EA) are determined by amplification of individual SABLC barcodes by the following relative quantitative PCR (qPCR) protocol: 6 μΙ_ of DNAse/RNAse free water (Life Technologies) are supplemented with 10 μΙ of iQ™ SYBROGreen Supermix (Bio-Rad, cat. NO 170-8882), 1 μΙ_ of 10 μΜ forward primer, 1 μΙ_ of 10 μΜ reverse primer and 2 μΙ_ of 60 nM (corresponding to 120 fmol) template DNA. The forward and reverse primers were designed by using the 3'- end 1 1 nucleotides of the U 18 primer sequence and the 7 nucleotides corresponding to the sequence 3' adjacent thereto for each amplified product. The resulting 20 μΙ_ PCR premixes are incubated in a thermocycler instrument (Bio-Rad iCycler) with the following program settings: 1x melt 03:00 min 95 °C, 40x melt 00:30 min 95 °C, 40x anneal 00:30 min 60 °C, 40x elongate 00:30 min 72 °C. The cycle threshold (Ct) values are calculated by PCR baseline subtraction curve fits of the SYBROGreen amplification charts, performed by the iQ™5 Optical System Software, Version 2.1 (Bio-Rad). Enrichment factors (Ef) are calculated as follows: Ef=POWER(1 ,7; ACt(EA-Tb))=POWER(1 ,7;AVERAGE(Ct(EA))- AVERAGE(Ct(Tb))), performed by Microsoft Excel 2010 software. From the relative values calculated (Figure 17A) it was shown that strong enrichment was achieved for the bivalent ligand complex corresponding to SEQ ID NO. 10.

The bivalent ligands obtained from the Tb and EA capture were also further analysed using high throughput sequencing methods (lllumina). The enrichment factors were calculated from this different analysis and the results are depicted in Figure 17B. Results are very similar as compared to the qPCR analysis, with the bivalent ligand corresponding to SEQ ID NO. 10 showing most enrichment. Synthesis of nucleic acid arms, monovalent precursor molecules and hairpin oligonucleotide The design of the nucleic acid arms is based on well known DNA aptamer ligands for Thrombin G15D (SEQ ID NO. 7) and TBA27 (SEQ ID NO. 8). DNA sequences SEQ ID NOs. 20-27 and NOs. 40-42 were synthesized via standard phosphoramidite chemistry, commercially available from IDT, USA.

24 D3_R1_BC06_5'-Cp_3'-(T12)-[G15DS] Phosphate -

25 D3_L1 -BC05_5'-(T0)-[TBA27S] - -

26 D3_L1 -BC07_5'-(T6)-[TBA27] - -

27 D3_L1 -BC08_5'-(T12)-[TBA27] - -

28 Phosphate -

HP7n-BC09

29 Phosphate -

HP3n-TBA27

30 Phosphate -

HP7n-G15D

Table 3a: Synthesized "Right" (SEQ ID NOs. 20 and 25-27) and "Left" (SEQ ID NOs. 21-24) monovalent precursor molecules. All sequence descriptions are listed left to right in 5' to the 3' orientation. Individual identifier sequences are annotated in small letters. Ligands, if present, are annotated in italics. Identifier sequences in the hairpin forming third

oligonucleotide, which also serve as internal ligands are annotated in small italic letters. The letter "S" in column "design" indicates a scrambled version of a binding sequence to serve as a negative control with the same length and base composition as the binding sequence.

SEQ ID Sequence

NO

20 GTC CGT GGT AGG GCA GGT TGG GGT GAC AGA CTG GTC CGC AAG TTC tgc aaa tec aGC TCA CGT A

ACG TGA GCc cct ata cac GAA CTT GCG GAC CAG TCT GGT TGG TGT GGT

21

TGG

ACG TGA GCc tag tac cac GAA CTT GCG GAC CAG TCT TTT TTT GGT TGG

22

TGT GGT TGG

ACG TGA GCa cct teg cac GAA CTT GCG GAC CAG TCT TTT TTT TTT TTT

23

GGT TGG TGT GGT TGG

ACG TGA GCt aca cgt cac GAA CTT GCG GAC CAG TCT TTT TTT TTT TTT

24

GTG TGT GTG TGT GTG

25 GTC CGT CCT ACC GCA GCT TCC GTT GAC AGA CTG GTC CGC AAG TTC tgc agg tct cGC TCA CGT A

26 GTC CGT GGT AGG GCA GGT TGG GGT GAC TTT TTT AGA CTG GTC CGC AAG

TTC tgc tcagtcaGC TCA CGT A

27 GTC CGT GGT AGG GCA GGT TGG GGT GAC TTT TTT TTT TTT AGA CTG GTC

CGC AAG TTC tgc tcagtcaGC TCA CGT A

28

GTC GGA AGA GCC aat cct gGG CTC TTC CGA CT

29

GCA etc cgt ggt agg gca. ggt tgg ggt gac tG T

30

GTA CGA CGT ggt tgg tgt ggt tgg TCA TCG TAC T

Table 3b: Synthesized "Right" (SEQ ID NOs. 20 and 25-27) and "Left" (SEQ ID NOs. 21-24) monovalent precursor molecules. All sequences are listed left to right in 5' to the 3' orientation. Individual identifier sequences are annotated in small letters. Ligands, if present, are annotated in italics. Identifier sequences in the hairpin forming third oligonucleotide, which also serve as internal ligands are annotated in small italic letters.

To test the concept of two arms and a third oligonucleotide, comprising an additional identifier sequence or a third binding nucleotide motif, twelve different bivalent ligand compositions were assembled by hybridisation of corresponding monovalent

individual precursor molecules followed by sticky-end enzymatic ligation reaction.

Spacer sequences of 6 Thymidins and 12 Thymidins in sequence ID NOs. 22-24 and NOs. 26-27 were used. Hybridization and ligation (see figure 19 and 20a)

A reaction mixture was prepared to hybridize the left and the right arms. The mix contains both left and right arms in a 1 : 1 molar rate diluted in water. The mixture was denatured for 3 minutes at 95°C and equilibrated to room temperature for 1 hour. The hairpin was dissolved in ultra pure water and incubated for 3 minutes at 95 °C and equilibrated to room

temperature for 1 hour. A ligation mixture was prepared containing the arms and the hairpin in a 1 : 1 molar rate, 1X T4 ligase buffer, and 7.5 units T4 ligase. The mixture was incubated for 30 minutes at room temperature.

GTC CGT CCT ACC GCA GCT TCC GTT GAC AGA CTG GTC CGC

AAG TTC tgc aggtctcGC TCA CGT

AGTCGGAAGAGCCAATCCTGGGCTCTTCCGACTACG TGA GCaccttcg cac

GAA CTT GCG GAC CAG TCT TTT TTT TTT TTT GGT TGG TGT GGT TGG T0S-T12

GTC CGT CCT ACC GCA GCT TCC GTT GAC AGA CTG GTC CGC AAG TTC tgc aggtctcGC TCA CGT

AGTCGGAAGAGCCAATCCTGGGCTCTTCCGACTACG TGA GCtacacgt cac

GAA CTT GCG GAC CAG TCT TTT TTT TTT TTT GTG TGT GTG TGT GTG T0s-T12s

GTC CGT GGT AGG GCA GGT TGG GGT GAC TTT TTT AGA CTG

T6-TO GTC CGC AAG TTC tgc tcagtcaGC TCA CGT

AGTCGGAAGAGCCAATCCTGGGCTCTTCCGACTACG TGA GCccctata cac

GAA CTT GCG GAC CAG TCT GGT TGG TGT GGT TGG

GTC CGT GGT AGG GCA GGT TGG GGT GAC TTT TTT AGA CTG

T6-T6 GTC CGC AAG TTC tgc tcagtcaGC TCA CGT

AGTCGGAAGAGCCAATCCTGGGCTCTTCCGACTACG TGA GCctagtac cac

GAA CTT GCG GAC CAG TCT TTT TTT GGT TGG TGT GGT TGG

GTC CGT GGT AGG GCA GGT TGG GGT GAC TTT TTT TTT TTT

T12-T0 AGA CTG GTC CGC AAG TTC tgc tcagtcaGC TCA CGT

AGTCGGAAGAGCCAATCCTGGGCTCTTCCGACTACG TGA GCccctata cac

GAA CTT GCG GAC CAG TCT GGT TGG TGT GGT TGG

GTC CGT GGT AGG GCA GGT TGG GGT GAC TTT TTT TTT TTT

T12-T12 AGA CTG GTC CGC AAG TTC tgc tcagtcaGC TCA CGT

AGTCGGAAGAGCCAATCCTGGGCTCTTCCGACTACG TGA GCaccttcg cac

GAA CTT GCG GAC CAG TCT TTT TTT TTT TTT GGT TGG TGT GGT TGG

GTC CGT GGT AGG GCA GGT TGG GGT GAC TTT TTT TTT TTT

T12- AGA CTG GTC CGC AAG TTC tgc tcagtcaGC TCA CGT

TBA27-T12 AGCAGTCCGTGGTAGGGCAGGTTGGGGTGACTGCTACG TGA GCaccttcg

cac GAA CTT GCG GAC CAG TCT TTT TTT TTT TTT GGT TGG

TGT GGT TGG

GTC CGT GGT AGG GCA GGT TGG GGT GAC TTT TTT TTT TTT

T12-G15D- AGA CTG GTC CGC AAG TTC tgc tcagtcaGC TCA CGT

T12

AGTACGAcgtGGTTGGTGTGGTTGGtcaTCGTACTACG TGA GCaccttcg

cac GAA CTT GCG GAC CAG TCT TTT TTT TTT TTT GGT TGG

TGT GGT TGG Table 4: List of self-assembled bivalent ligand complexes:

Monovalent precursors SEQ ID NOs. 20-27 were hybridized in different combinations and one particular hairpin forming third oligonucleotide (SEQ ID NOs. 28-30) was added to individual reaction vessels. After enzymatic ligation 14 self-assembled bivalent ligand complexes SEQ ID NOs. 31-44 were obtained as listed above. Sequences in the bulge region (BR) of the SABLC are annotated with small letters. The binding sequence are annotated with italics. (SEQ ID NO. 31 , combination of SEQ ID NOs. 20+28+21 , named L1_BC01_5'-(T0)-[TBA27]_HP7n-BC09_R1_BC02_5'-Cp_3'-(T0)-[G15D], length is 147 nt, MW is 45552 Da; SEQ ID NO. 32, combination of SEQ ID NOs. 20+28+22, named

L1_BC01_5'-(T0)-[TBA27]_HP7n-BC09_R1_BC03_5'-Cp_3'-(T6)-[G15D], length is 153 nt, MW is 47417 Da; SEQ ID NO. 33, combination of SEQ ID NOs. 20+28+23, named

L1_BC01_5'-(T0)-[TBA27]_HP7n-BC09_R1_BC04_5'-Cp_3'-(T12)-[G15D], length is 159 nt, MW is 49218 Da; SEQ ID NO. 34, combination of SEQ ID NOs.20+28+24, named

L1_BC01_5'-(T0)-[TBA27]_HP7n-BC09_R1_BC06_5'-Cp_3'-(T12)-[G15DS], length is 159 nt, MW is 49218 Da; SEQ ID NO. 35, combination of SEQ ID NOs. 25+28+21 , named

L1_BC05_5'-(T0)-[TBA27S]_HP7n-BC09_R1_BC02_5'-Cp_3'-(T0)-[G15D], length is 147 nt, MW is 45270 Da; SEQ ID NO. 36, combination of SEQ ID NOs. 25+28+22, named

L1_BC05_5'-(T0)-[TBA27S]_HP7n-BC09_R1_BC03_5'-Cp_3'-(T6)-[G15D], length is 153 nt, MW is 47153 Da; SEQ ID NO. 37, combination of SEQ ID NOs. 25+28+23, named

L1_BC05_5'-(T0)-[TBA27S]_HP7n-BC09_R1_BC04_5'-Cp_3'-(T12)-[G15D], length is 159 nt, MW is 48936 Da; SEQ ID NO. 38, combination of SEQ ID NOs. 25+28+24, named L1_BC05_5'-(T0)-[TBA27S]_HP7n-BC09_R1_BC06_5'-Cp_3'-(T12)-[G15DS], length is 159 nt, MW is 48936 Da; SEQ ID NO. 39, combination of SEQ ID NOs.26+28+21 , named L1_BC07_5'-(T6)-[TBA27]_HP7n-BC09_R1_BC02_5'-Cp_3'-(T0)-[G15D], length is 153 nt, MW is 47384 Da; SEQ ID NO. 40, combination of SEQ ID NOs. 26+28+22, named

L1_BC07_5'-(T6)-[TBA27]_HP7n-BC09_R1_BC03_5'-Cp_3'-(T6)-[G15D], length is 159 nt, MW is 49250 Da; SEQ ID NO. 41 , combination of SEQ ID NOs. 27+28+21 , named

L1_BC08_5'-(T12)-[TBA27]_HP7n-BC09_R1_BC02_5'-Cp_3'-(T0)-[G15D], length is 159 nt, MW is 49210 Da; SEQ ID NO. 42, combination of SEQ ID NOs. 27+28+23, named

L1_BC08_5'-(T12)-[TBA27]_HP7n-BC09_R1_BC04_5'-Cp_3'-(T12)-[G15D], length is 171 nt, MW is 52876 Da; SEQ ID NO. 43, combination of SEQ ID NOs. 27+29+23, named

L1_BC08_5'-(T12)-[TBA27]_HP7n-TBA27_R1_BC04_5'-Cp_3'-(T12)-[G15D], length is 173 nt, MW is 53702 Da; SEQ ID NO. 44, combination of SEQ ID NOs. 27+30+23, named L1_BC08_5'-(T12)-[TBA27]_HP7n-G15D_R1_BC04_5'-Cp_3'-(T12)-[G15D], length is 173 nt, MW is 53642 Da).

PCR (see figure 20b): The reaction mixtures contains 1 nmol - 1 amol SABLC construct, 2.5 mM dNTPs, 2 Units VentR® (exo-) DNA Polymerase (New England Biolabs, US), 1x corresponding PCR buffer, 5% DMSO and 2.4 μΜ of forward primer SEQ ID NO. 45 and reverse primer SEQ ID NO. 46 in a total volume of 50 μΙ. The PCR scheme was typically 25 to 35 PCR cycles, depending on the template concentration, of 30 seconds denaturation at 95°C, 30 seconds annealing at 55 - 72°C and 30 seconds extension at 72°C. There was an initial denaturation step of one minute at 95°C and a final extension step at 72° for one minute.

Table 5: List of primers and corresponding PCR product (as an example: sense SEQ ID NO. 47 and antisense strand SEQ ID NO. 48, see figure 20b). Binding sequence of primers is underlined. Small letters indicate the sequence which is part of the bulge. Length of the PCR product is 132 nucleotides.

Clotting Experiments Set-up (see figure 21)

To evaluate the feasibility of the self-assembled bivalent ligand complex as a potential anticoagulant reagent, Standard activated Partial Thromboplastin Time (aPTT, Langdell et al., J Lab Clin Med 41 , 637-647, 1953) values for each individual or ligand composition were determined by using human plasma samples. Procedures were applied as described above. The values in this case were determined in single measurements and are shown in figure 21. The benefit of combining two binding motifs is evident in the aPTT values, which is influenced by the relative distance of the binding motifs to each other.

Claims

A bivalent ligand library wherein each of the bivalent ligands of the library comprises:

- a first nucleic acid arm, comprising:

a) at its 3'-end a covalently conjugated first ligand;

b) optionally, a first spacer molecule, 5' from the first ligand,

c) a first primer binding site, 5' from the spacer molecule or first ligand; d) a first dimerisation sequence, wherein the first primer binding site and first dimerisation sequence overlap in whole or in part; e) a first ligand identifier sequence, 5' from the first primer binding site; f) optionally, a first spacer molecule identifier sequence, 5' or 3' from the ligand identifier sequence;

h) a first functional group at the 5'-end;

- a second nucleic acid arm, comprising:

a) at its 5'-end a covalently conjugated second ligand;

b) optionally, a spacer molecule, 3' from the first ligand,

c) a nucleic acid sequence complementary to the first dimerisation

sequence and complementary in whole or in part to the first primer binding site, 3' from the second ligand;

e) a second ligand identifier sequence, 3' from the said complementary nucleic acid sequence of c);

g) a nucleic acid sequence complementary to the second

dimerisation sequence in the first nucleic acid arm;

h) a second functional group at the 3'-end;

wherein the first and second dimerisation sequences of the first nucleic acid arm are base paired with the complementary sequences of the second nucleic acid arm and wherein the 5'-end of the first nucleic acid arm is covalently conjugated with the 3'- end of the second nucleic acid arm, and wherein the two covalently conjugated arms allowing amplification and/or sequencing of the conjugated nucleic acid sequence comprising at least the first ligand and optional first spacer molecule identifier sequence and the second ligand and optional second spacer molecule identifier sequence.

2. A bivalent ligand library wherein each of the bivalent ligands of the library comprises:

- a first nucleic acid arm, comprising:

a) at its 3'-end a covalently conjugated first ligand;

b) optionally, a first spacer molecule, 5' from the first ligand, c) a first primer binding site, 5' from the spacer molecule or first ligand; d) a first dimerisation sequence, wherein the first primer binding site and first dimerisation sequence overlap in whole or in part; e) a first identifier sequence, 5' from the first primer binding site; f) a second dimerisation sequence, 5' from the identifier sequence of e); and

h) a phosphate group at the 5'-end;

- a second nucleic acid arm, comprising:

a) at its 5'-end a covalently conjugated second ligand;

b) optionally, a spacer molecule, 3' from the first ligand,

c) a nucleic acid sequence complementary to the first dimerisation

e) a second identifier sequence, 3' from the said complementary nucleic acid sequence of c); and

f) a nucleic acid sequence complementary to the second

dimerisation sequence in the first nucleic acid arm;

- a third nucleic acid, comprising:

a) a phosphate group at its 5' terminus;

c) optionally, a third identifier sequence between the 5' and the 3'-end; wherein the first and second dimerisation sequences of the first nucleic acid arm are base paired with the complementary sequences of the second nucleic acid arm; and wherein the 5'-end of the first nucleic acid arm and the 3'-end of the third nucleic acid is covalently conjugated and the 5'-end of the third nucleic acid and the 3'-end of the second nucleic acid is covalently conjugated, allowing amplification and/or sequencing of the conjugated nucleic acid sequences comprising at least the first, the second and the optional third identifier sequence.

3. A bivalent ligand library according to claim 1 or claim 2, wherein the library comprises randomly combined first and second nucleic acid arms.

4. A bivalent ligand library according to any one of claims 1-3, wherein the first and/or second ligand is a nucleic acid molecule.

5. A bivalent ligand library according to claim 3, wherein the first and/or second ligand is a nucleic acid molecule with randomized positions.

6. A bivalent ligand library according to any one of claims 1-5, wherein the bivalent ligands comprise said spacer molecules, wherein preferably the said spacer molecules are nucleic acid sequences.

7. A bivalent ligand library according to any one of claims 1 and 3-6, wherein the first nucleic acid arm and the second nucleic acid arm are enzymatically or chemically covalently conjugated; or wherein the first nucleic acid arm, the second nucleic acid arm and the third nucleic acid arm are enzymatically or chemically covalently conjugated.

8. A bivalent ligand library according to any one claims 1-7, wherein each of the first and second nucleic acid arms comprises a ligand selected from a library of ligands.

9. A bivalent ligand library according to any one claims 1-7, wherein each of the first nucleic acid arms comprise a ligand selected from a first library of ligands, and wherein each of the second nucleic acid arms comprises a ligand selected from a second library of ligands.

10. A bivalent ligand library according to any one of claims 1-7, wherein ligands were selected for binding to a target molecule.

1 1. A bivalent ligand library according to any one of claim 1-10, wherein the second nucleic acid arm comprises a nucleic acid sequence complementary to a second primer binding site, wherein the second primer binding site is 3' from the from the ligand and optional spacer molecule and 5' from the identifier sequences.

12. A bivalent ligand library according to any one of claims 2-1 1 , wherein the third nucleic acid comprises a third ligand, said ligand preferably being a nucleic acid comprised in the third nucleic acid.

13. A bivalent ligand library according to any one of claims 2-12, wherein the third nucleic acid comprises a nucleic acid functioning as an enzyme.

14. A bivalent ligand library according to any one of claims 2-13, wherein the third nucleic acid comprises one or more functional groups selected from the group consisting of an attachment moiety allowing covalent attachment to another molecule or a surface, a fluorescently labelled moiety, a radioisotope tag, a moiety which is as a substrate for an enzymatic reaction.

15. Method for providing a bivalent ligand library comprising the steps of:

a) providing a first monovalent ligand library comprising the first nucleic acid arms according to any one of claims 1 and 3-1 1 , and a second monovalent ligand library comprising the second nucleic acid arms according to any one of claims 1 and 3-1 1 ;

b) mixing the selected monovalent ligand libraries and allowing the first and

second nucleic acid arms to hybridize;

16. Method for providing a bivalent ligand library comprising the steps of:

b) providing a target molecule;

e) selecting first and second monovalent ligands interacting with the target

molecule from the first and second contacting steps;

f) mixing the selected monovalent ligands and allowing the first and second nucleic acid arms to hybridize; g) covalently conjugating the 5'-end of the first nucleic acid arm with the 3'-end of the second nucleic acid arm, thereby providing a bivalent ligand library.

17. Method for providing a bivalent ligand library:

a) providing a first monovalent ligand library comprising the first nucleic acid arm according to any one of claims 1 and 3-1 1 , and a second monovalent ligand library comprising the second nucleic acid arm according to any one of claims 1 and 3-1 1 ;

b) providing a first and a second target molecule;

d) contacting the second monovalent ligand library with the second target

molecule in a second contacting step;

18. Method for providing a bivalent ligand library comprising the steps of:

c) covalently conjugating the 5'-end of the first nucleic acid arm with the 3'-end of the third nucleic acid and covalently conjugating the 5'-end of the third nucleic acid arm with the 3'-end of the second nucleic acid arm, thereby providing a bivalent ligand library.

19. Method for providing a bivalent ligand library comprising the steps of:

a) providing a first monovalent ligand library comprising the first nucleic acid arms according to any one of claims 2-14, a second monovalent ligand library comprising the second nucleic acid arms according to any one of claims 2-14, and the third nucleic acid according to any one of claims 2-14; b) providing a target molecule;

g) covalently conjugating the 5'-end of the first nucleic acid arm with the 3'-end of the third nucleic acid and covalently conjugating the 5'-end of the third nucleic acid arm with the 3'-end of the second nucleic acid arm, thereby providing a bivalent ligand library.

20. Method for providing a bivalent ligand library:

b) providing a first and a second target molecule;

d) contacting the second monovalent ligand library with the second target

molecule in a second contacting step;

21. A method for identifying a bivalent ligand binding to a target molecule comprising the steps of:

a) providing a bivalent ligand library according to any one of claims 1-14;

b) providing a target molecule; c) contacting the target molecule with the bivalent ligand library;

d) selecting bivalent ligand interacting with the target molecule;

e) identifying bivalent ligands interacting with the target molecule.

22. Method according to claim 21 , wherein the bivalent ligands are provided in solution.

23. Method according to any one of claims 21-22, wherein the target molecule comprises a label.

24. Method according to claim 23, wherein bivalent ligands interacting with the labelled target molecule are selected using the label.

25. Method according to any one of claims 21 , wherein the target molecule is on a solid support, and the selection step comprises removing bivalent ligands not interacting with the target molecule, thereby providing bivalent ligands interacting with the target molecule.

26. Method according to any one of claims 21-25, wherein in the identification step the bivalent ligands are identified by determining the identifier sequences as present in each of the bivalent ligands.

27. Method according to claim 26, wherein in the identification step the identifier

sequences are amplified with at least a first primer capable of binding to the first primer binding site, wherein the amplification step is before the sequence

determination step.

28. Method according to any one of claims 26-27, wherein the bivalent ligands of the bivalent ligand library comprise a first and second primer binding site, and wherein in the identification step the identifier sequences are first amplified with a PCR amplification with a first primer capable of binding to the first primer binding site and a second primer capable of binding to the second primer binding site, wherein the amplification is before the sequence determination step.

29. Method according to any one of claims 21-28, wherein after the selection step and before the identification step the target molecule is removed.

30. Method according to any one of claims 21-28, wherein after the selection step the ligands are removed.

31. Method according to any one of claim 21-30, wherein two target molecules are provided and wherein bivalent ligand library is contacted with the two target molecules either consecutively or simultaneously, and wherein bivalent ligands are selected interacting with both of the two target molecules.

32. Method according to claim 31 , wherein the bivalent ligand library is a bivalent ligand library according to claim 15 or 18.

33. Method for preparing a bivalent ligand comprising the steps of:

- providing the information of an identified bivalent ligand obtained in a method according to any of claims 21-32 with regard to the first ligand, the first optional spacer molecule, the second ligand and optional second spacer molecule;

- preparing a bivalent ligand wherein the first ligand and optional first spacer molecule is linked with the second ligand and optional second spacer molecule via a linker molecule to replace the DNA scaffold as present in the identified ligand.