Random DNA structures

The method of designing and ligating DNA fragments to create random DNA structures with multivalent binding capabilities addresses the limitations of existing technologies in scalability and efficiency, enabling high-throughput screening and the identification of strong binders for drug discovery and diagnostics.

WO2024260866A1PCT designated stage expired Publication Date: 2024-12-26BENSON ERIK +1
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
PCT/EP2024/066550
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-06-21
Filing Date
2024-06-14
Publication Date
2024-12-26

AI Technical Summary

Technical Problem

Current methods for generating DNA structures with high specificity and binding strength, such as DNA origami, face challenges in scalability, efficiency, and discovering unknown binding patterns on a large scale, limiting their application in drug discovery and diagnostics.

Method used

A method involving the design and synthesis of a library of DNA fragments, followed by random ligation to generate a plurality of random DNA structures with a contiguous sequence and multivalent binding capabilities, enabling high-throughput screening and efficient discovery of strong binders.

Benefits of technology

This approach allows for the scalable generation of random DNA structures with high affinity and specificity, facilitating high-throughput screening and the identification of strong binders, which is not achievable with current techniques, and enables the development of effective pharmaceutical compositions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure IMGF000010_0001
    Figure IMGF000010_0001
  • Figure IMGF000020_0001
    Figure IMGF000020_0001
  • Figure IMGF000022_0001
    Figure IMGF000022_0001
Patent Text Reader

Abstract

The present invention relates to a method for generating a plurality of random DNA structures having affinity for a predetermined target, wherein at least one of said DNA structures comprises a contiguous sequence, a native capacity to be amplified, and multivalent binding capabilities to said predetermined target. The present invention also relates to random DNA structures having affinity for a predetermined target and pharmaceutical compositions comprising the same.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] RANDOM DNA STRUCTURES

[0002] Field of the invention

[0003] The present invention relates to a method for generating a plurality of random DNA structures having affinity for a predetermined target, wherein at least one of said DNA structures comprises a contiguous sequence, a native capacity to be amplified, and multivalent binding capabilities to said predetermined target. The present invention also relates to random DNA structures having affinity for a predetermined target and pharmaceutical compositions comprising the same.

[0004] Background

[0005] Antibodies, one of the remarkable discoveries of the 20th century, are small proteins that possess the ability to adapt and bind to a wide range of targets. They play an indispensable role in the research, diagnostics and therapies we rely on today. However, their production method presents challenges such as batch-to- batch variation, excessive costs, and ethical concerns regarding animal use. Moreover, it is estimated that half of the antibodies used in research fail to bind to their intended targets, which has emerged as a significant cause of the 'replication crisis' in cancer research (Berglund et al., 2008, A Genecentric Human Protein Atlas for Expression Profiles Based on Antibodies). Additionally, although the COVID-19 pandemic showcased the potential of virus-binding antibodies in rapid diagnostics and therapeutic applications, the speed of the development and deployment of these antibodies was slow in comparison to the rapid spread and mutation of the virus.

[0006] As an alternative to antibodies, aptamers offer a nucleic acid-based solution. Aptamers are single-stranded DNA or RNA oligonucleotides that can fold into specific three-dimensional structures, allowing them to bind with specificity to their targets. Aptamers can target small molecule compounds, peptides, proteins, or even cells themselves (Dua et al., 2008, Recent Patents on DNA & Gene Sequences, 2: 172-186). They possess advantages such as smaller size, better tissue permeability, ease of chemical manipulation, setting them apart from antibodies. However, in comparison to antibodies, aptamers typically exhibit lower binding strength. It has recently been revealed that by patterning multiple weak binders according to the distribution of epitopes on a target surface, a significantly higher net binding strength can be achieved. This finding opens to new possibilities for enhancing the binding capabilities of aptamers. However, current strategies, such as DNA origami, which employs DNA as a scaffold for self-assembling nanoscale structures with precise geometries and strategically positioning aptamers on the scaffold, face challenges in terms of scalability, efficiency and finding unknown binding patterns on a large scale. The manual design of these structures necessitates individually tailored assembly and purification strategies, limiting the number of patterns that can be effectively tested.

[0007] Thus, current strategies for generating binding agents, such as DNA structures comprising aptamers, with high specificity and binding strength are not ideal, and there is a continued need for better production methods and binding agents.

[0008] Disclosure of the invention

[0009] The present disclosure addresses the need for an efficient and scalable method wherein individual random DNA structures, that display binding patterns that enable a high net-binding strength via avidity to diverse predetermined targets of interest, are generated on a large scale.

[0010] It is an object of the present disclosure to provide a method for generating a plurality of random DNA structures, said plurality of random DNA structures having affinity for a predetermined target.

[0011] A related object of the disclosure is a random DNA structure having affinity for a predetermined target, obtained by the method as disclosed herein.

[0012] It is another object of the disclosure to provide a pharmaceutical composition comprising a random DNA structure having affinity for a predetermined target obtained by the method as disclosed herein and at least one pharmaceutically acceptable carrier.

[0013] These and other objects, which are evident to the skilled person from the present disclosure, are met by the different aspects of the invention as claimed in the appended claims and as generally disclosed herein. Thus, in a first aspect of the disclosure, there is provided a method for generating a plurality of random DNA structures, said plurality of random DNA structures having affinity for a predetermined target, the method comprising:

[0014] (a) designing a library comprising a plurality of DNA fragments;

[0015] (b) synthesizing said library comprising said plurality of DNA fragments; and

[0016] (c) randomly ligating said synthesized plurality of DNA fragments to generate a plurality of random DNA structures wherein at least one of said random DNA structures comprises a contiguous sequence, a native capacity to be amplified and multivalent binding capabilities to said predetermined target.

[0017] The method disclosed herein is based at least in part on the inventors' development of a novel system that enables the scalable generation of a plurality of random DNA structures comprising a contiguous sequence, a native capacity to be amplified and multivalent binding capabilities to a predetermined target. In accordance herewith, the present inventors have found that a method comprising the listed steps above exhibits an efficiency in discovering said random DNA structures having binding capabilities to a predetermined target and powerful scalability in generating said random DNA structures, both features which the prior art lacks. In the present approach, at least one known binding sequence that binds to a predetermined target is incorporated directly into designed DNA fragment libraries, libraries which are the source of a large number of diverse scaffolding structures that display the at least one binding sequence in various spatial patterns. In the present disclosure, "having affinity for" means having the ability to bind to a preselected, predetermined or known target molecule. The present approach thus selects for favorable spatial positioning of at least one known binding sequence relative to the physical geometric constraints of the predetermined target molecule and the rest of the random DNA structure, to quickly and simply create a large selection of random DNA structures with affinity for a predetermined target. In some embodiments the plurality of random DNA structures have varying degrees of affinity for the predetermined target. In the present disclosure, "have varying degrees of affinity" means that the plurality of random DNA structures may have different affinities towards the predetermined target when compared to each other, i.e. each individual affinity belonging each individual random DNA structure within the plurality of random DNA structures towards the predetermined target may not be the same. The referred to at least one binding sequence may preferably be an aptamer sequence that binds to an epitope on a predetermined target. As will be appreciated by those of skill in the art, an "aptamer" is a short segment of DNA, RNA or peptide that binds to a specific molecular target, such as a protein. In the present disclosure, an "aptamer" is a short segment of DNA that binds to a specific molecular target. A structure with affinity for a predetermined target, that displays at least one binding sequence in various spatial patterns, has multivalent binding capabilities towards said target. The term "multivalent binding capabilities" refers to a random DNA structure comprising two or more identical or different binding sequences, such as aptamers. Said random DNA structures are therein capable of tight and specific binding to their targets due to the formation of characteristic spatial patterns displaying said binding sequences. Examples of predetermined targets include peptides (oligopeptides and polypeptides), nucleic acids, lipids, sugars and low-molecular weight compounds. The predetermined target is preferably a peptide, more preferably a polypeptide, i.e., a protein. Alternatively, the predetermined target may be a naturally derived substance, a chemically synthesized substance, a recombinant substance, a cell, a virus or the like. In the present disclosure, the "aptamer" is a known DNA aptamer or a DNA aptamer obtained by a method known in the art whose nucleotide sequence has been revealed or can be revealed. DNA based binding sequences, such as aptamers, can be revealed via SELEX, an evolutionary technique where a pool of nucleic acid molecules comprising random sequence regions and primer binding regions at both ends thereof are added to targets. The DNA based binding sequences are then recovered by PCR amplification, allowing for repeated rounds of selection, resulting in binders with higher affinity to the target. Alternatively, DNA based binding sequences such as aptamers can be predicted, for example, by using in silica computer-based methodology.

[0018] In contrast to the method of the present disclosure, methods for creating binding structures as disclosed in the prior art, such as US2018 / 0044663 (Arizona State University), are not suitable for scalability as predetermined and manually designed scaffolds need individually tailored assembly and purification strategies. Each scaffold must be individually designed and assembled, and if scaffolds are mixed and tested, it may not be possible to recover the identity of the binding patterns. In practice, this also limits studies to testing few structures, making the method both limited and certainly not scalable. US2018 / 0044663 demonstrates a handful of pre-determined fixed DNA structures that include random sequence "proto-aptamers". Disadvantageously, in US2018 / 0044663, the pre-determined scaffolds must be tested to discover which aptamer(s) can function multivalently within each respective pre-determined scaffold. This strategy is distinct from the method of the present disclosure in that, in US2018044663, the shape of the DNA structures or scaffolds are not varied in order to find multivalent patterns of aptamers. Instead, a pre-determined structure or scaffold is used to find aptamers to bind to epitopes with precisely that predetermined spacing or pattern, regardless of whether the predetermined spacing or pattern is optimal for the target.

[0019] The present approach generates a large number of random DNA structures with affinity for a predetermined target, wherein at least one said DNA structure comprises a contiguous sequence, a native capacity to be amplified, and multivalent binding capabilities to said predetermined target. In the present disclosure, "contiguous" means touching along a boundary or a point, being adjacent, or touching or connected through an unbroken sequence. Therein, in the present disclosure, "a contiguous sequence" refers to the sequence which codes for an individual random DNA structure. The sequence is built up of ligated DNA fragments, resulting in a contiguous sequence. In the present disclosure, "a native capacity to be amplified", refers to a contiguous sequence that can be further amplified, for example by PCR. These features, a contiguous sequence and a native capacity to be amplified, entail that the large number of random DNA structures generated by the disclosed method, comprising multivalent binding capabilities to a predetermined target, are compatible with further biochemical techniques downstream of the random DNA structures' generation. Downstream selection experiments, for example, have the advantage of generating stronger binding products. This unique feature gives the claimed method both an efficiency and scalability which the prior art lacks. For example, in previously disclosed methods for generating structures which comprise aptamers, as in US2021 / 0371863 (Postech Academy Ind Foundation), the binding product generated by the method is a final product lacking the capabilities of further amplification and thus for further direct selection after generation. This strongly limits the efficiency and use of methods such as those disclosed in the prior art. Modern drug discovery strategies require high-throughput and scalable production, selection and identification of potential drug candidates. In the presently disclosed method, random DNA structures are assembled by randomly ligating small DNA fragments to generate branched wireframes with complex 3D shapes, also known as random DNA structures. DNA fragment ligation generates a contiguous sequence that determines the shape of the structure through folding, similar to a primary sequence of a protein. Modern sequencing techniques together with structural prediction tools can read these sequences and predict their structure at a massively parallel scale resulting in millions of random DNA structures from a single tube, transforming the search for DNA patterns and binding structures to predetermined targets into a high-throughput single-pot experiment. In the present disclosure, "single-pot experiment", means a method for generating random DNA structures comprising multivalent binding capabilities to a predetermined target, the random DNA structures comprising a contiguous sequence and a native capacity to be amplified, therein being directly compatible with downstream processes applied on the resulting direct products of the disclosed method. In a practical sense, it should be understood that the reaction which generates the random DNA binders, and any downstream processes applied to those generated random DNA binders, can occur more or less in "one-pot", i.e. one test tube. Such downstream processes may be affinity selection of the resulting random DNA structures, amplification of the resulting random DNA structures and sequencing of the resulting random DNA structures. It is meant that the resulting direct products of the method, the random DNA structures themselves, are in such a state that they can seamlessly be utilized in in-line downstream processes, such as affinity selection, amplification and sequencing, without modification of the resulting direct products. This allows for the simultaneous testing of millions of binding patterns by applying the random DNA structures to a predetermined target, such as but not limited to a protein or peptide target, and washing away structures that bind weakly via affinity selection. To the best of the inventors' knowledge, the disclosed method is the first method to allow for "single-pot" high throughput screening of aptamer patterns, enabling the search for new strong binders from DNA that would be impossible to discover with any current technique.

[0020] In order for downstream processes, such as amplification and characterization, to be practicable after generation of a plurality of random DNA structures with affinity for a predetermined target, the random DNA structures must fold from a single strand of DNA. Furthermore, structures that can fold from a single strand of DNA also benefit from simpler deployment and assembly compared to DNA structures developed by current methods as their genomes can be produced by established gene synthesis. The method as disclosed in the present approach provides these capabilities, providing for a method which is designed to complement high-throughput production, selection and identification demanded by modern drug discovery strategies.

[0021] In another embodiment of the first aspect, the method as disclosed herein comprises (a) designing a library comprising a plurality of DNA fragments, wherein said library comprises at least two classes of DNA fragments selected from a first class comprising linear fragments, a second class comprising branched fragments, and a third class comprising loop fragments. In the present disclosure, "linear fragments" means fragments that are designed to assemble from two linear oligonucleotides, such as oligonucleotide segments consisting of 12 nucleotides, such as 13 nucleotides, such as 14 nucleotides, such as 15 nucleotides, such as 16 nucleotides, such as 17 nucleotides, such as 18 nucleotides, such as 19 nucleotides, such as 20 nucleotides, such as 25 nucleotides, such as 100 nucleotides long. When the segment length increases it both acts as a longer spacer and will change the rotational orientation of connected components due to the helical nature of dsDNA. Said oligonucleotide segment lengths are not limited to any specific length, and are adapted according to the predetermined target. In the present disclosure, "branched fragments" means fragments consisting of 3-arm junctions, such as 4-arm junctions, such as 5-arm junctions, such as 6-arm junctions, such as 7-arm junctions, such as 8-arm junctions. Said branched fragments are not limited to a specific number of branches, or arm junctions. Instead, the number of branches or arm junctions are adapted according to the predetermined target. These DNA fragments are designed using established DNA nanotechnology tools, such as vhelix and oxDNA analysis tools. It is to be understood that DNA sequences comprising the linear fragments and branched fragments are sourced from a pool of refined 12 nucleotide fragments. The fragments in said pool are designed to have a similar melting temperature and a low cross-interaction with the other sequences and their reverse complements. This is achieved by a python script that first generates one million random 12 nucleotide fragments with the constraint of an even mix of the nucleotides (A / T / G / C). These sequences are then pruned by first calculating their melting temperature and only retaining the sequences within a selected temperature interval. Then, a first sequence is added to a list and the remaining sequences and their reverse complements are iteratively aligned with the sequences on that list. If a high-scoring alignment is found with a sequence already on the list, the iterated sequence is removed as this predicts a potential unspecific interaction between the sequences. If no high-scoring alignment is found, the iterated sequence gets added to the list. This results in a pruned list of sequences with matched melting temperatures, and low rates of cross-interaction. Finally, the secondary structure energy is calculated for each sequence on the list, and sequences with high secondary structure removed.

[0022] In the present disclosure, "loop fragments" means DNA fragments comprising at least one double stranded stem segment and at least one loop region. The sequence of the at least one double stranded stem segment comprises at least one sequence taken from the refined pool of DNA sequences as described above. The sequence of the at least one loop region of the loop fragment may comprise at least one sequence taken from the pool of DNA sequences as described above. Alternatively, the sequence of the at least one loop region of the "loop fragment" may comprise other specifically designed sequences.

[0023] Suitably, the 3'-end or the 5'-end of at least a portion of said linear fragments, branched fragments and loop fragments start with a single stranded extension. In a preferred embodiment, the 3'-end or the 5'-end of all said linear fragments, branched fragments and loop fragments start with a single stranded extension. The single stranded extension, or "sticky" overhangs enable the DNA fragments to weakly hybridize with each other. This entails that when said synthesized plurality of DNA fragments are combined with the intention of randomly ligating, the combined DNA fragments weakly hybridize with each other as they link up to form random DNA structures, wherein the ligation enzyme fuses the strands together to form single long contiguous sequences. Said single stranded extension typically comprises a nucleotide sequence which is palindromic. In the present disclosure, "palindromic" means a DNA sequence which reads the same from 5' to 3' on one strand as it does from 5' to 3' on the complementary strand. In a further embodiment, said single stranded extension comprising a nucleotide sequence which is palindromic comprises at least 2 nucleotides, such as at least 4 nucleotides, such as at least 6 nucleotides. In theory, said nucleotide sequence which is palindromic could comprise an infinite number of nucleotides as long as the resulting sequence is palindromic, which would require an even number of nucleotides. However, in practicality the nucleotide length of the nucleotide sequence which is palindromic should be limited to 20 nucleotides or less in length. All possible palindromic DNA sequences of even length from 2 to 20 nucleotides can be represented by the following formula: where n is length of the sequence in number of nucleotides, S = si, S2, ...snis a sequence of nucleotides in which Sj is the ith nucleotide in the sequence. S' = s'i, s'2-.s'n is the reverse complement sequence of S where s'i is the complement of Sj.

[0024] The above formula states that the set of all possible palindromic DNA sequences of even length from 2 to 20 nucleotides is the union of all sets of sequences of length n (where n ranges from 2 to 20 in steps of 2) that satisfy the palindromic criterion. Thus, in a further embodiment of the first aspect, said single stranded extension comprises a nucleotide sequence which is derived from a set defined by the above equation. As single stranded extension sequences increase in length, the risk for the formation of strong interactions between identical DNA fragment classes before fragments of different classes are mixed, is increased, leading to an abundance of homo-polymers when a ligation enzyme is added. Short overhangs will only form transient interactions between DNA fragments allowing for combination of different DNA fragment classes after mixing and ligating. Suitably, said single stranded extension comprises a nucleotide sequence selected from the group consisting of AT, TA, CG, GC, AATT, ATAT, ACGT, AGCT, TATA, TTAA, TCGA, TGCA, CATG, CTAG, CCGG, CGCG, GATC, GTAC, GCGC, GGCC, AAATTT, AATATT, AACGTT, AAGCTT, ATATAT, ATTAAT, ATCGAT, ATGCAT, ACATGT, ACTAGT, ACCGGT, ACGCGT, AGATCT, AGTACT, AGCGCT, AGGCCT, TAATTA, TATATA, TACGTA, TAGCTA, TTATAA, TTTAAA, TTCGAA, TTGCAA, TCATGA, TCTAGA, TCCGGA, TCGCGA, TGATCA, TGTACA, TGCGCA, TGGCCA, CAATTG, CATATG, CACGTG, CAGCTG, CTATAG, CTTAAG, CTCGAG, CTGCAG, CCATGG, CCTAGG, CCCGGG, CCGCGG, CGATCG, CGTACG, CGCGCG, CGGCCG, GAATTC, GATATC, GACGTC, GAGCTC, GTATAC, GTTAAC, GTCGAC, GTGCAC, GCATGC, GCTAGC, GCCGGC, GCGCGC, GGATCC, GGTACC, GGCGCC, and GGGCCC. In a preferred embodiment, said single stranded extension comprises a nucleotide sequence which is selected from the group consisting of AATT, ATAT, ACGT, AGCT, TATA, TTAA, TCGA, TGCA, CATG, CTAG, CCGG, CGCG, GATC, GTAC, GCGC, and GGCC. In a more preferred embodiment, said single stranded extension comprises a nucleotide sequence which is AATT.

[0025] In an embodiment of the first aspect, said library comprises at least two classes of DNA fragments selected from a first class comprising linear fragments, a second class comprising branched fragments, and a third class comprising loop fragments, wherein the 5'-end of said linear fragments, branched fragments and loop fragments are synthesized with a 5'-end phosphate modification. Said phosphate modification enables ligation of the synthesized plurality of DNA fragments to generate a plurality of random DNA structures.

[0026] In another embodiment of the first aspect, said library comprises at least two classes of DNA fragments selected from a first class comprising linear fragments, a second class comprising branched fragments, and a third class comprising loop fragments, wherein said third class of fragments comprising loop fragments comprises at least one aptamer hairpin loop fragment and at least one amplification hairpin loop fragment.

[0027] In the present disclosure there are two types of "loop fragments", referred to as "amplification hairpin loop fragment" and "aptamer hairpin loop fragment".

[0028] The loop region of the amplification hairpin loop, also known as the amplification hairpin loop region, is designed to comprise a sequence of nucleotides which can be targeted by PCR primers. In one non-limiting example, an amplification hairpin loop fragment was designed where the hairpin stem segment comprised one of the nucleotide sequences taken from the refined pool of DNA sequences as described above. Furthermore, 5 sequences from the refined pool of DNA sequences as described above were used to create a 56 nucleotide long single-stranded amplification hairpin loop region that could be targeted by PCR primers.

[0029] The loop region of the aptamer hairpin loop fragment, also known as the aptamer hairpin loop region, is designed to comprise a known aptamer sequence from literature or sequence predicted (e.g. computationally) to have the potential to bind to a predetermined target. Alternatively, the loop region of the aptamer hairpin loop fragment, may be designed to comprise a sequence that is not expected to bind to any predetermined target, and therein provides a negative control in selection experiments. When the loop region of an aptamer hairpin loop fragment is designed to comprise a placebo sequence that is not expected to bind to any predetermined target, then the entire fragment is instead referred to as a "negative control aptamer hairpin loop fragment" or a "placebo aptamer hairpin loop fragment". Thus, in one embodiment, said at least one aptamer hairpin loop fragment comprises at least one aptamer sequence, said aptamer sequence having affinity for said predetermined target. In another embodiment of the first aspect, said at least one amplification hairpin loop fragment comprises an amplification hairpin loop region, said amplification hairpin loop region comprising at least two adjacent primer binding sites, wherein said adjacent primer binding sites comprise a target binding sequence for at least one restriction enzyme. In the present disclosure, "adjacent primer binding sites" means at least two primer binding sites which are next to each other in a sequence of nucleotides. Furthermore, "comprising a target binding site for at least one restriction enzyme" means that at least a portion of the sequence which defines the at least two adjacent primer binding sites codes for a target sequence which is the target sequence of at least one restriction enzyme. In a non-limiting example, referring to Example 2 and Example 9 of the present disclosure, two adjacent primer binding sites were designed so that their border sequence, which is to be understood as at least a portion of the sequence which codes for the two primer binding sites, is the target sequence of the restriction enzyme Pvull. This design is a feature which allows for the amplification hairpin loop region to be cleaved in-between the adjacent primer sequences by the restriction enzyme Pvull. However, as restriction enzymes can only cleave double stranded DNA, a splint oligo was designed to make the border sequence double-stranded, therein allowing the restriction enzyme to cleave the amplification hairpin loop region in-between the two primer binding sites. The skilled person will understand that as long as at least one target sequence for a restriction enzyme is comprised in a portion of the amplification hairpin loop region of the amplification hairpin loop fragment, which codes for the at least two primer binding sites, then any combination of primer binding sites could be utilized. The skilled person would not be limited to the target sequence for the restriction enzyme Pvull, as exemplified in Examples 2 and 9. That is, the skilled person would understand that any appropriate target sequence for any appropriate restriction enzyme could be utilized in the disclosed invention.

[0030] In another embodiment of the first aspect, said at least one random DNA structure of said plurality of random DNA structures comprises at least two aptamer hairpin loop fragments and at least one amplification hairpin loop fragment as defined herein, wherein said at least two aptamer hairpin loop fragments comprise identical aptamer sequences or two different aptamer sequences. This way, the method as disclosed herein provides for the generation of high affinity and high specificity multivalent binders against any predetermined target, such as but not limited to, a protein target. In the present disclosure, a "multivalent binder" is to be understood as a binding structure, such as a random DNA structure, which comprises multivalent binding capabilities to a predetermined target. Such multivalent binding capabilities could be the incorporation of at least two aptamer sequences into a random DNA structure. A random DNA structure comprising two or more identical or different aptamer sequences, such as in the form of at least two aptamer hairpin loop fragments, is a multivalent binder. It can also be understood that the random DNA structure comprising at least two aptamer hairpin loop fragments also comprises multivalency towards the predetermined target. Multivalent binders have the capability of tight and specific binding to their targets due to the formation of characteristic spatial structures. In some cases, random DNA structures comprising multivalent binding capabilities to a predetermined target comprise additional structural elements or functional groups, such as molecular labels, examples being fluorescent dye modifications, biotin, haptens, click chemistry functional modifications for conjugation to other molecules of interest, and other suitable molecular labels. As described herein, the method of the disclosure advantageously directs the evolution of random DNA structures, comprising at least two aptamer hairpin loop fragments, said random DNA structures having high affinity and high specificity towards their predetermined targets.

[0031] In another embodiment of the first aspect, said method as disclosed herein further comprises one or more of the steps of:

[0032] (d) treating said plurality of random DNA structures with an exonuclease;

[0033] (e) annealing at least one splint oligonucleotide complementary to said target binding sequence for at least one restriction enzyme;

[0034] (f) restriction digesting said random DNA structures by at least one restriction enzyme recognizing said target binding sequence as defined herein for at least one restriction enzyme;

[0035] (g) affinity selection of said plurality of random DNA structures;

[0036] (h) amplifying by PCR said eluted random DNA structures resulting in dsDNA linear products corresponding to said eluted random DNA structures;

[0037] (i) purifying said amplified dsDNA linear products corresponding to said eluted random DNA structures;

[0038] (j) sequencing said purified amplified dsDNA linear products corresponding to said eluted random DNA structures;

[0039] (k) recovering said random DNA structures from said dsDNA linear products via treatment of said dsDNA linear products using an exonuclease; and

[0040] (l) repeating at least once said affinity selection using said recovered random DNA structures such that random DNA structures with a higher affinity towards the predetermined target are selected.

[0041] The skilled person will understand that there are several suitable methods for performing steps such as (d) - (I). It is reasonable to expect that the skilled person would have knowledge of these various methods and know which methods are suitable to apply to the present inventive method and how to adapt them accordingly.

[0042] Suitably, said affinity selection comprises immobilizing said predetermined target on one or more beads and contacting said one or more beads comprising said predetermined target with said plurality of random DNA structures. In a preferred embodiment, said one or more beads are magnetic beads. In another embodiment of the first aspect, said affinity selection further comprises recovering said contacted beads and eluting bound random DNA structures from the beads. The skilled person will understand that there are several suitable methods for performing steps such as affinity selection, amplification, and sequencing. It is reasonable to expect that the skilled person would have knowledge of these various methods and know which methods are suitable to apply to the present inventive method and how to adapt them accordingly. When it pertains to affinity selection, the skilled person will be familiar with the various principles and protocols of affinity selection. Affinity selection techniques may include but are not limited to methods which immobilize a predetermined target and contact the predetermined target with a pool of random DNA structures in order to isolate and eventually identify specific DNA structures with affinity for the predetermined target. Examples of such affinity selection techniques may include magnetic bead pull-down assays, wherein magnetic beads are functionalized with the predetermined target and employed to capture random DNA structures that possess complementary binding sites. When it pertains to amplification, the skilled person will be familiar with the various principles and protocols of DNA amplification. DNA amplification techniques may include but are not limited to polymerase chain reaction (PCR), loop-mediated isothermal amplification (LAMP), and rolling circle amplification (RCA). In a preferred embodiment, the DNA amplification method polymerase chain reaction (PCR) is utilized. When it pertains to sequencing, the skilled person will be familiar with the various principles and protocols of DNA sequencing. DNA sequencing techniques may include but are not limited to Sanger sequencing , next-generation sequencing (NGS), pyrosequencing and optical mapping. In a preferred embodiment, a high- throughput next-generation sequencing (NGS) technique is utilized. One sequencing reaction generates large numbers of reads that provide abundant information for the multivalent random DNA structure sequences. Binding affinity and specificity can then be evaluated using the determined sequences. By leveraging their understanding of these methodologies, a person skilled in the art would be capable of applying them to the invention at hand, facilitating the efficient and accurate affinity selection, amplification and sequencing of the random DNA structures that are described herein. Furthermore, the skilled person will also be familiar with adjusting the parameters involved in the methods suitable for performing steps such as affinity selection, amplification, and sequencing. These parameters may include, but are not limited to, incubation conditions and protocols, buffer compositions, washing conditions and protocols, optimization of reaction conditions such as reaction duration, temperature conditions and concentrations of participating species, primer design, template requirements or the appropriate choice of enzymes.

[0043] Generated random DNA structures can be characterized using various techniques. For example, random DNA structures can be characterized using microscopic techniques, such as transmission electron microscopy (TEM), atomic force microscopy (AFM), and gel electrophoresis assays. TEM is especially useful for the structural characterization of 3D DNA structures and AFM is particularly useful for the analysis of 2D DNA structures.

[0044] In some cases, multivalent random DNA structures according to the methods provided herein can be used as biological recognition elements of biosensors. Aptamer-based biosensors often use immobilized aptamers as recognition elements for the target molecules. The most popularly used electrode material is gold where the thiolated DNA strands, in this case random DNA structures, can be immobilized via strong Au-S linkage. In other cases, a streptavidin-biotin linkage is used to immobilize random DNA structures. In some cases, fluorescent or other detectable moieties are linked to a random DNA structure. The term "biosensor" as used herein may be any biosensor which comprises the random DNA structures obtained according to a method described herein. It will be understood that antibodies used as biological elements may be replaced with aptamers in biosensor applications . This change enables a rapid method to detect, for example, pathogenic microorganisms. The advantages of using random DNA structures over antibodies include the lower costs of production, stability at room temperature, and there are no ethical issues when random DNA structures are used because they can be produced by a chemical synthesis where no animals or animal cells are needed. Another advantage of random DNA structures is that they can be stored as a lyophilized powder at room temperature for more than one year. In addition, DNA structures can recover their native active conformation after denaturation.

[0045] In a second aspect, the disclosure provides a random DNA structure having affinity for a predetermined target obtained by the method as disclosed herein.

[0046] Said random DNA structure is capable of tight and specific binding to its predetermined target(s) due to the formation of characteristic spatial patterns displaying said binding sequences as described herein. Examples of predetermined targets include peptides (oligopeptides and polypeptides), nucleic acids, lipids, sugars and low-molecular weight compounds. The predetermined target is preferably a peptide, more preferably a polypeptide, i.e., a protein. Alternatively, the predetermined target may be a naturally derived substance, a chemically synthesized substance, a recombinant substance, a cell, a virus or the like. The inventors have demonstrated the disclosed method in the Examples section by applying the method to the predetermined targets Human thrombin and norovirus virus like particles, respectively. The invention will be more fully understood upon consideration of the following non-limiting Examples, but is not limited to the exemplified targets.

[0047] In a third aspect, the disclosure provides a pharmaceutical composition comprising a random DNA structure having affinity for a predetermined target as described in the second aspect and at least one pharmaceutically acceptable carrier.

[0048] As the skilled person will appreciate, the "pharmaceutically acceptable carrier" refers to a substance that is usually used in the pharmaceutical formulating art and added without inhibiting or suppressing the effect of the pharmaceutical composition in order to facilitate the formulation of the pharmaceutical composition or its application to a subject, such as a human, and maintain the effect of the inhibitor of target substance function. Examples of the carriers include excipients, binders, disintegrants, fillers, emulsifiers, flow control additives, lubricants, and surfactants.

[0049] Examples of the "excipients" include sugars such as monosaccharides, disaccharides, cyclodextrin, and polysaccharides (specifically including, but not limited to, glucose, sucrose, lactose, raffinose, mannitol, sorbitol, inositol, dextrin, maltodextrin, starch, and cellulose), metal salts (e.g., sodium phosphate or calcium phosphate, calcium sulfate, and magnesium sulfate), citric acid, tartaric acid, glycine, low-, middle-, or high-molecular-weight polyethylene glycol (PEG), Pluronic, and combinations thereof. Examples of the "binders" include starch glues composed of corn, wheat, rice, or potato starch, gelatin, tragacanth, methylcellulose, hydroxypropylmethylcellulose, sodium carboxymethylcellulose, and / or polyvinylpyrrolidone. Examples of the "disintegrants" include the starches described above, carboxymethyl starch, cross-linked polyvinylpyrrolidone, agar, alginic acid or sodium alginate, and salts thereof. Examples of the "fillers" include the sugars described above and / or calcium phosphate (e.g., tricalcium phosphate or calcium hydrogen phosphate). Examples of the "emulsifiers" include sorbitan fatty acid ester, glycerin fatty acid ester, sucrose fatty acid ester, and propylene glycol fatty acid ester. Examples of the "flow control additives" and the "lubricants" include silicate, talc, stearate, and polyethylene glycol.

[0050] Such carriers may be used appropriately according to the need. The pharmaceutical composition of the present invention may also contain, in addition to the additives described above, optional additives such as corrigents, solubilizing agents (solubilizers), suspending agents, diluents, surfactants, stabilizers, absorption promoters (e.g., quaternary ammonium salts and sodium lauryl sulfate), expanders, wetting agents, humectants (e.g., glycerin and starch), adsorbents (e.g., starch, lactose, kaolin, bentonite, and colloidal silicic acid), disintegration inhibitors (e.g., saccharose, stearin, cacao butter, and hydrogenated oil), coating agents, coloring agents, preservatives, antioxidants, fragrances, flavors, sweeteners, and buffers.

[0051] The "surfactants" correspond to, for example, alkali metal salts, alkaline earth metal salts, and ammonium salts of lignosulfonic acid, naphthalenesulfonic acid, phenolsulfonic acid, or dibutylnaphthalenesulfonic acid, alkylaryl sulfonate, alkyl sulfate, alkyl sulfonate, fatty alcohol sulfate, fatty acid and sulfated fatty alcohol glycol ether, condensates of sulfonated naphthalene or naphthalene derivatives and formaldehyde, condensates of naphthalene or naphthalenesulfonic acid, phenol, and formaldehyde, polyoxyethylene octylphenyl ether, ethoxylated isooctylphenol, octylphenol, nonylphenol, alkylphenyl polyglycol ether, tributylphenyl polyglycol ether, tristearylphenyl polyglycol ether, alkylaryl polyether alcohol, alcohol and fatty alcohol / ethylene oxide condensates, ethoxylated castor oil, polyoxyethylene alkyl ether, ethoxylated polyoxypropylene, lauryl alcohol polyglycol ether acetal, sorbitol ester, lignosulfite waste liquors, and methylcellulose.

[0052] The pharmaceutical composition of this embodiment may contain one or more of these carriers per pharmaceutical composition.

[0053] The pharmaceutical composition of the present invention may further contain an additional drug without canceling the pharmacological effect of the random DNA strcuture of the present invention. The pharmaceutical composition of the present invention may contain, for example, a specific amount of an antibiotic.

[0054] The dosage form of the pharmaceutical composition of the present invention is not particularly limited as long as the form does not deactivate the active ingredient and can exert the pharmacological effect in vivo after administration. The dosage form usually varies depending on an administration method and / or prescription conditions.

[0055] Examples of dosage forms suitable for oral administration can include solid preparations (including tablets, pills, sublingual preparations, capsules, drops, and troches), granules, dusts, powders, and liquid preparations. The solid preparations can be prepared, if necessary, in coated dosage forms known in the art, for example, as sugar-coated tablets, gelatin-coated tablets, enteric coated tablets, film-coated tablets, bilayer tablets, or multilayer tablets.

[0056] Parenteral administration is subdivided into systemic administration and local administration. The local administration is further subdivided into interstitial administration, transepidermal administration, transmucosal administration, and transrectal administration. The pharmaceutical composition may also be prepared in a dosage form suitable for each administration method. Examples of dosage forms suitable for systemic or interstitial administration include injections which are liquid preparations. Examples of dosage forms suitable for transepidermal administration or transmucosal administration can include liquid preparations (including liniments, eye drops, nasal drops, and inhalants), suspensions (including emulsions and creams), dusts (including nasal drops and inhalants), pastes, gels, ointments, and plasters. Examples of dosage forms suitable for transrectal administration can include suppositories. In the case of drug administration to plants, examples of the dosage form of the pharmaceutical composition include liquids, solids (including semi-solids), and combinations thereof. In this case, the pharmaceutical composition may be prepared as solutions, oil dispersions, emulsions, suspensions, dusts, powders, pastes, gels, pellets, tablets, and granules.

[0057] The specific shapes or sizes of these dosage forms are not particularly limited and may be any shape or size that falls within ranges accepted for each dosage form known in the art.

[0058] The present invention has been described in terms of one or more preferred embodiments, and it should be appreciated that many equivalents, alternatives, variations, and modifications, aside from those expressly stated, are possible and within scope of the invention. The invention will be more fully understood upon consideration of the following non-limiting Examples.

[0059] Brief description of the fi

[0060] Figure 1. displays a computer rendering of a selection of the DNA fragments (such as linear fragments, branched fragments, and the amplification hairpin loop fragment) which comprise the DNA fragment library. Also depicted are: two aptamer sequences, TBA and HD22, to be utilized in designing aptamer hairpin loop fragments targeting Human thrombin; seven aptamer sequences, AG3, BEIER, BUF3, Ml, M62, SMV19 and SMV21, to be utilized in designing aptamer hairpin loop fragments targeting norovirus capsid protein (VP1); and a placebo loop sequence, to be utilized in designing negative control hairpin loop fragments also known as placebo aptamer hairpin loop fragments which are to be used as negative controls.

[0061] Figure 2. displays a gel from the denaturing polyacrylamide gel electrophoresis experiment in Example 4, which displays three DNA samples that were generated in Example 3: according to method ("All components", lane 4), without the terminal loops ("No loops", lane 3) without the ligation enzyme ("No ligase", lane 2), and the same three samples run in lanes 2-4 but treated with Exonuclease III prior to running gel electrophoresis (lanes 5-7), and a 100 BP ladder (lane 1). Figure 3. displays an Atomic force microscopy (AFM) image of random DNA structures targeting the protein Human thrombin, scale bar 100 nm.

[0062] Figure 4. displays a gel from the agarose gel electrophoresis experiment of Example 7, which displays three DNA samples that were generated in Example 6: PCR amplification reaction mixture ("Library PCR", lane 2), the PCR amplification reaction mixture after AmPURE bead purification ("Library PCR Ampure", lane 3), the AmPURE bead purified PCR amplification reaction mixture after lambda exonuclease digestion ("Library PCR Exo", lane 4) and two PCR ladders ("100 BP ladder", lane 1 and 5.

[0063] Figure 5. displays a diagram depicting an analysis of nanopore sequencing results from Example 8. After two rounds of selection utilizing generated random DNA structures with multivalent binding capabilities to thrombin, the initial and refined mixtures were sequenced and analyzed. The fraction of the total set of read sequences that were identified as structures that incorporate a particular aptamer were compared between the initial and refined mixtures.

[0064] Figure 6. displays a diagram depicting an analysis of nanopore sequencing results from Example 12, which depicts the prevalence of aptamers targeting norovirus capsid protein (VPl) comprised in random DNA structures with multivalent binding capabilities to norovirus virus like particles. Generated random DNA structures were selected in three rounds against norovirus virus like particles and the initial and three refined random DNA structure mixtures were sequenced. The prevalence of the eight aptamer sequences were compared between the initial mixture and the mixtures from three rounds of selection.

[0065] Figure 7. displays a diagram which depicts the prevalence of multivalent aptamer DNA structures. DNA structures selected in three rounds against norovirus virus like particles and the initial random DNA structure mixture were sequenced. The number of random DNA structures that included multiple aptamers were compared between the initial mixture and the mixtures from three rounds of selection. Examples

[0066] The following Examples disclose the design, generation and characterization of novel random DNA structures.

[0067] Example 1

[0068] Design and Synthesis of DNA fragment Library

[0069] Methods DNA sequences form ng structural regions of the DNA fragments, like stems of the loop fragments, linear fragments and branched fragments, were sourced from a starting pool of iteratively refined 12 nucleotide segments that were in turn designed to have a similar melting temperature and a low cross-interaction with the other sequences and their reverse complements. The DNA fragments comprised in the DNA fragment library were in turn formed by the annealing of 1, 2, 3, 4 or 5 of custom designed DNA oligonucleotides sourced from the above- mentioned iteratively refined starting pool. The DNA fragments comprised in the DNA fragment library were divided into three classes: linear fragments, branched fragments and loop fragments. The class comprising loop fragments comprised in turn two types of loop fragments, namely amplification hairpin loop fragments and aptamer hairpin loop fragments. An overview of some of these fragments is shown in Figure 1.

[0070] Linear fragments were designed to assemble from two oligonucleotides forming a double stranded region that was 14, 16, 18 or 20 nucleotides long. Branched fragments comprised of 3, 4 or 5 - arm junctions each assembled from 3, 4 or 5 oligonucleotides.

[0071] Amplification hairpin loop fragments were assembled from a single DNA oligonucleotide. The oligonucleotide was designed so that its start and end regions were complementary, forming a hairpin stem. The middle of the oligonucleotide was designed to form a 56 nucleotide long single stranded loop region that acted as binding sites for amplification primers for PCR in downstream experiments. The said 56 nucleotide long single stranded loop region was also designed to comprise a target binding sequence for the restriction enzyme Pvull. The loop region of the amplification hairpin loop fragment consisted of the sequence as follows: CCGTAGCTTAGAGCCTTTAGAACGACAGCTGGATCCTCAATGGCCTAGCTCAATCG (SEQ ID NO:1). The nucleotide sequence as defined in SEQ ID NO:1 comprised two binding regions for primers, the regions being: nucleotides 1-28, which coded for the binding region of primer A (SEQ ID NO:2), and nucleotides 29-56, which coded for a reverse complement region which was the binding region of primer B (SEQ ID NO:3). A target binding sequence for the restriction enzyme Pvull, CAGCTG, was designed to comprise nucleotides belonging to nucleotide position number 26-28 of SEQ ID NO:2 and nucleotide position number 1-3 of SEQ ID NO:3. Therein the 56 nucleotide long amplification hairpin loop region (SEQ ID NO:1) was designed to comprise at least two adjacent primer binding sites (SEQ ID NO:2 and SEQ ID NO:3) wherein said adjacent primer binding sites comprised a target binding sequence for at least one restriction enzyme, Pvull.

[0072] Aptamer hairpin loop fragments were designed and produced for each target, which in the following non-limiting examples are Human thrombin and norovirus capsid protein (VPl). These aptamer hairpin loop fragments were similar in design to the amplification hairpin loop fragments described above and were folded from a single DNA oligonucleotide. In the loop region of these aptamer hairpin loop fragments, the sequence for an aptamer sequence was added. Typically, more than one aptamer hairpin loop fragment was tested in the selection experiments.

[0073] Additionally, a placebo aptamer hairpin loop fragment that was not expected to bind to any predetermined target was added as a negative control to selection experiments. It was designed by the same method as described above for aptamer hairpin loop fragments, but instead a nucleotide sequence, GGACTCTTGCAA (SEQ ID NO:4), which was not expected to form any useful aptamer, was added to the loop region instead of a known aptamer sequence.

[0074] DNA fragments were designed using established DNA nanotechnology tools (vhelix and oxDNA analysis tools).

[0075] DNA fragment oligonucleotide strands were terminated in the 5'-end with the sequence 'AATTC' and in the 3'-end with the sequence 'G' meaning that the oligonucleotide sequences of the DNA fragments formed 5' protruding sticky ends with the sequence 'AATT'. Additionally, the oligonucleotide sequences of the DNA fragments were synthesized with a 5'-end phosphate modification to enable downstream ligation. Specifically for thrombin target: For the selection of binders towards the predetermined target thrombin, two types of aptamer hairpin loop fragments were designed to include two well characterized aptamers for the protein. These aptamers are known as TBA and HD22. It has been established that these aptamers bind to opposite sides of the protein, and that structured multivalent binders that display these two aptamers can achieve a stronger binding via avidity.

[0076] Specifically for norovirus capsid protein (VP1) target:

[0077] For selection of binders towards the predetermined target norovirus capsid protein (VPl), seven types of aptamers were included in distinct aptamer hairpin loop fragments. The aptamers were found in literature and were known as AG3, BEIER, BUF2, Ml, M62, SMV19, SMV21. The norovirus capsid is constructed of 180 identical copies of norovirus capsid protein (VPl), and it was expected that scaffolds that display multiple copies of the same aptamer could achieve multivalency through binding to multiple identical epitopes on the viral surface. The generated multivalent random DNA structures would therein comprise aptamer sequences which bind to multiple norovirus capsid proteins (VPl) in a multivalent manner on the norovirus virus like particle.

[0078] Synthesis of DNA fragment library: Oligonucleotides comprising the three classes of DNA fragments were synthesized by a commercial vendor (Integrated DNA Technologies) with 5'-end phosphate modifications using standard methods.

[0079] Results

[0080] The design and synthesis resulted in two groupings of DNA fragments. The linear fragments, the branched fragments and the amplification loop fragments were "universal" and could therefore be used in any potential experiments wherein random DNA binders for a predetermined target were to be generated. The other grouping comprised of the aptamer loop fragments, which were "unique" to the predetermined target being pursued. The design and synthesis therefore resulted successfully in DNA fragment components which would be mixed in accordance with the predetermined target being pursued. In the following examples it is demonstrated how random DNA structures comprising multivalent binding capabilities are generated for the predetermined target thrombin and the predetermined target norovirus capsid protein (VP1), respectively.

[0081] Example 2 Assembly of DNA fragment library Methods

[0082] DNA fragment preparation: DNA oligonucleotides comprising the constituent DNA fragments, as designed and synthesized in example 1, were individually resuspended from a lyophilized state at 100 pM concentration in pure water. The oligonucleotides comprising each DNA fragment class were then mixed individually to create a stock of mixed oligonucleotides for each DNA fragment class. For DNA fragments that are designed to assemble from multiple oligonucleotides (linear fragments and branched fragments) the oligonucleotides that comprise each fragment class were mixed after resuspension. This means that the linear fragments, the branched fragments and the loop fragments (both aptamer hairpin loop fragments, amplification hairpin loop fragments, and placebo aptamer hairpin loop fragments) were all individually resuspended as described above. The DNA fragments were then individually diluted to a final concentration of 10 pM in pure water.

[0083] Annealing of DNA fragments: Individually, 27 pl of each respective sample of DNA fragments were mixed with 3 pl of lOx concentrated T4 ligation buffer (50 mM Tris- HCI, 10 mM MgCl2, 1 mM ATP, 10 mM DTT). The samples were placed in a thermal cycler and annealed via a temperature program consisting of an initial denaturation step of 90°C for 30 seconds followed by a slow temperature ramp from 80 - 20°C temperature ramp over 30 minutes. The program was ended by an incubation at 4°C.

[0084] Results

[0085] The preparation and annealing of the respective DNA fragments resulted in 18 individual DNA fragment samples, as detailed in Table 1 below. The individual DNA fragment samples were then mixed to create a DNA fragment library (one library per predetermined target) and ligated, as detailed in the following experiments.

[0086] Table 1: DNA fragment samples resulting from preparation and annealing of individual DNA fragments - to be combined into respective DNA fragment libraries in following experiments

[0087] Example 3

[0088] Generating random DNA structures targeting thrombin

[0089] Random DNA structures with binding affinity towards the predetermined target thrombin were generated, comprising aptamer hairpin loops with the aptamers HD22 and TBA.

[0090] Methods

[0091] DNA fragments, detailed in Table 1 and assembled and annealed as described in Example 2, were mixed according to Table 2. In the mixture, the amplification hairpin loop fragment had the lowest concentration of 0.24 pM. Linear fragments and branched fragments were added in a 1.5 times excess. Each of the three aptamer hairpin loop fragments (TBA, HD22 and placebo aptamer hairpin loop fragment) were added at 10 times excess. To this mixture, a concentrated solution of T4 ligase was added.

[0092] The ligation reaction was placed in a thermal cycler and cooled to 16°C for lOh followed by a hold at 4°C until further processing. The T4 ligase was heat inactivated by incubating the sample for 20 minutes at 65°C.

[0093] Table 2: Ligation reaction mixture

[0094]

[0095] 100 pl of the ligated sample was mixed with 30 pl NEBuffer™ 1 (10 mM Bis- Tris-Propane-HCI, 10 mM MgCl2, 1 mM DTT) (New England Biolabs®) and 5 pl of the exonuclease enzyme Exo III. The mixture was incubated in a thermocycler at 37°C for 30 minutes followed by a 30 minute incubation at 70°C to heat inactivate the exonuclease enzyme.

[0096] In the ligation process described above it was expected that as the number of free fragments were depleted and added into growing random DNA structures, some random DNA structures would not complete their growth and feature free ends. These random DNA structures do not form complete genomes and cannot be replicated by PCR, thus they were removed from the pool of generated random DNA structures via exonuclease digestion.

[0097] Furthermore, in order to correctly amplify the random DNA structures after any downstream affinity experiments, thereof preventing circular concatemers, the amplification hairpin loop that was designed to be targeted by PCR primers was cleaved between the two adjacent primer binding sites. In order to do this, the two adjacent primer binding sites were designed so that their border sequence is the target sequence of the restriction enzyme Pvull, therein the amplification hairpin loop could be cleaved exactly between the primer sequences. As restriction enzymes can only cleave double stranded DNA, a splint oligo that makes the border sequence double-stranded was added to the random DNA structure mixture. 8 .1 of a splint oligonucleotide (100 pM), designed to make the target binding sequence situated in the loop region of the amplification hairpin loop fragment double-stranded, was added to the digested ligation sample according to Table 3. The mixture was placed in a thermal cycler and annealed via a temperature program consisting of an initial denaturation step of 90°C for 30 seconds followed by a slow temperature ramp from 80 - 20°C temperature ramp over 30 minutes. The program was ended by an incubation at 4°C.

[0098] After annealing, 5 pl of Pvull restriction enzyme was added to the annealed sample, and the sample was placed in a thermal cycler at 37°C for 30 minutes, followed by a placement of the sample on ice.

[0099] Table 3: Splint oligonucleotide annealing reaction mixture

[0100] Results

[0101] Random DNA structures with binding affinity for the predetermined target thrombin were generated and ready for downstream selection experiments. At this point in time the mixture comprising the random DNA structures may be frozen if selection is done at a later date.

[0102] Example 4

[0103] Gel electrophoresis on generated random DNA structures

[0104] In order to validate that the structural fragments had successfully ligated to form DNA structures with contiguous circular genomes denaturing polyacrylamide electrophoresis was performed on the generated random DNA structures as described in Example 3.

[0105] Methods Gel electrophoresis: 6 samples ("No ligase", "No loops", "All components", "No ligase Exo", "No loops Exo", "All components Exo") were prepared for gel electrophoresis. Prior to gel electrophoresis, 0.3 pl of Exonuclease III was added to the three samples labeled "Exo". All 6 samples were then incubated at 37°C for 30 minutes followed by a 30-minute incubation at 70°C to heat inactivate the enzyme (Exonuclease III). The samples where then mixed with 3 pl of Formamide gel loading dye and denatured for 5 minutes at 95°C followed by immediate loading in a 10 % Page gel with Urea. Electrophoresis was run at 300 V for 30 minutes followed by gel staining with Sybr™ Gold (Thermo Fisher Scientific) for 30 minutes and imaging in a UV transilluminator.

[0106] Results

[0107] The gel (Figure 2) revealed that the sample 'No ligase', where all DNA fragments were mixed but not ligated, only contained small fast-moving bands consistent with the oligonucleotides of the DNA fragments. After exonuclease digestion (sample named "No ligase Exo"), no bands remained showing that the enzyme successfully digested DNA with free ends. In the sample 'No loops', where linear and branched fragments were mixed and ligated but hairpin loops were excluded from the reaction, the gel revealed a continuous smear of heavier ligation products forming. After exonuclease digestion (sample named "No loops Exo"), most DNA bands disappeared indicating that few circular genomes are formed in a ligation reaction without hairpin loops. When both linear, branched and loop fragments were mixed and ligated according to the method, the gel revealed a large distribution of band sizes corresponding to random DNA structures of various sizes. After exonuclease digestion of this reaction (sample named "All components Exo"), many bands remained, indicating the formation of contiguous circular genomes, lacking free ends which could otherwise be targeted by the exonuclease enzyme.

[0108] Example 5

[0109] Atomic Force Microscopy (AFM) on random DNA structures

[0110] In order to study the size and shape distribution of the random DNA structures generated in Example 3, atomic force microscopy (AFM) was used. Methods

[0111] Atomic Force Microscopy: The mixture comprising the random DNA structures generated in Example 3 was diluted 10-fold in AFM buffer (10 mM MgCl2, 5 mM TRIS, 1 mM EDTA), and 10 pl was added to freshly cleaved mica inside a sample reservoir. This was incubated for 30 seconds followed by the addition of 4 pl of 5 mM NiSC and a further 4 minutes 30 second incubation. The surface was then washed by adding and removing 1 ml of AFM buffer. For imaging, 1.5 ml of AFM buffer was added to the sample reservoir and the sample was imaged using a NanoWizard® 3 ultra (JPK Instruments) with ScanAsyst™ fl uid+ cantilever in AC mode (Bruker).

[0112] Results

[0113] The resulting AFM images revealed random DNA structures with a large distribution of sizes (scale 100 nm) and morphology including visual branching (Figure 3).

[0114] Example 6

[0115] Selection of random DNA structures

[0116] Selection of strongly binding random DNA structures targeting thrombin from the mixture comprising the random DNA structures generated in Example 3 was performed by immobilizing the predetermined target of interest on magnetic beads.

[0117] Methods

[0118] 250 pl of magnetic beads (Thermo Fisher Dynabeads™ MyOne™ Carboxylic Acid) were taken and placed on a magnet. The supernatant was removed, and the beads were washed twice with 0.5 ml of activation MEST buffer ((50 mM 2-(N - morpholino)ethanesulfonic acid, 0.01 % Tween 20, pH 6.0). The beads were resuspended in 25 pl MEST buffer and activated by addition of 25 pl EDC (10 mg / ml N-(3-dimethylaminopropyl)-N'-ethylcarbodiimide hydrochloride,) for 30 minutes at room on a rotary mixer.

[0119] A 100 units tube of lyophilized Human thrombin (Sigma Aldrich) was resuspended in 500 pl of MEST. After EDC activation, the beads were placed on a magnet and the supernatant of the beads was removed. The beads were resuspended in 200 pl of thrombin solution and the sample was incubated for 30 minutes on a rotary mixer. After conjugation of the beads the supernatant was removed, and the beads washed twice with 250 pl PBSMT (Phosphate buffered saline with 0.01 % Tween 20 and 10 mM MgSC ). The beads were now ready for selection experiments.

[0120] The protein functionalized beads were placed in a 1.5 ml Eppendorf tube and the supernatant was removed by the use of a magnet. 500 pl of the mixture comprising the restriction digested random DNA structures, as prepared in Example 3, was added to the beads, and the tube placed on a rotary mixer overnight in room temperature. After this, the beads were placed on a magnet and the supernatant removed and saved (for control purposes). The beads were then washed twice with 250 pl of PBSMT buffer to remove weakly binding DNA structures. The beads were finally resuspended in 100 pl of pure water and placed in a 95°C heat block for 5 minutes to release strongly binding random DNA structures. After this, the tube was placed on a magnet and the supernatant recovered and saved.

[0121] Example 7 PCR amplification of random DNA structures In order to perform downstream high-throughput sequencing and identification of the random DNA structures, amplification via PCR to form double-stranded linear fragments was performed on the mixture comprising the selected random DNA structures from Example 6.

[0122] Methods

[0123] Two opposing primers targeting the amplification hairpin loop region were used and PCR was performed using a standard protocol (65°C annealing temperature) utilizing the polymerase Taq-polymerase (DreamTaq™, Thermo Scientific). The double stranded PCR products were then purified using AMPure XP beads (Beckman Coulter™) to remove PCR buffer, enzyme and residual primers.

[0124] PCR products are linear and double stranded consisting of a strand encoding the genome, together with a strand encoding for the reverse complement of the genome. The reverse complement strand was removed in order for the genome of the random DNA structures to successfully refold allowing for downstream repeat selection experiments with the amplified random DNA structures. One of the primers in the PCR was functionalized with a 5' phosphate making it a target of digestion for the enzyme lambda exonuclease while the genome strand is spared. The PCR product was mixed with lambda exonuclease in lambda exonuclease buffer and incubated at 37 °C for 30 minutes followed by a 10-minute incubation at 75 °C to heat inactivate the enzyme.

[0125] The PCR amplification reaction, the subsequent purification of the reaction mixture and the exonuclease digestion of the purified reaction mixture was validated via an agarose gel electrophoresis assay. The products from each step were loaded in a 2% agarose gel and run at 120 V for 45 minutes. The gel was stained with Sybr™ Gold (Thermo Fisher Scientific) for 60 minutes and imaged in a UV transilluminator.

[0126] Results

[0127] The gel electrophoresis (Figure 4) revealed that the PCR amplification reaction mixture (sample named "Library PCR") comprised a distribution of random DNA structure genome sizes, and some residual primers were evident as fast-moving bands. After bead purification (sample named "Library PCR Ampure"), the fastmoving primers were removed, and after lambda Exonuclease digestion (the sample named "Library PCR Exo"), the appearance of the DNA distribution changed significantly, due to the conversion from double stranded linear genomes to single stranded DNA structures with secondary structure.

[0128] Example 8

[0129] Nanopore sequencing of random DNA structures after repeat selection with thrombin

[0130] In order to characterize the amplified, purified and digested random DNA structures resulting from Example 7, a high-throughput sequencing technique was needed. The Oxford Nanopore MinlON device was chosen as it allows for long reads and can be rapidly used via ligation of sequencing adapters to the PCR products.

[0131] Methods

[0132] The mixture comprising the amplified, purified and digested random DNA structures resulting from Example 7 was prepared using the ligation sequencing kit SQK-LSK110 (Oxford Nanopore Technologies) and following the protocol for ligation sequencing for amplicons. The ends of the PCR fragments were first prepared via enzymatic adenine-tailing, followed by the ligation with Oxford Nanopore Technologies proprietary sequencing adaptors. The random DNA structures were then sequenced on a Flongle™ flow cell (Oxford Nanopore Technologies) on MinlON™ device (Oxford Nanopore Technologies), typically resulting in 100000 - 500000 reads.

[0133] As the random DNA fragments were ligated to sequencing adaptors, the sequences of the random DNA structure genomes did not typically begin at the start of the read sequence. The sequenced reads were processed by a custom software that located the position of the amplification primer sequence from the amplification hairpin loop that was present in the beginning of the random DNA structure genome. This also helped determine if the read strand was the random DNA structure genome, or the reverse complement of it (ligation sequencing results in a mix of the two as either strand of the double stranded PCR product can be read).

[0134] After finding the position of the initial primer, the custom software moved progressively through the genome and performed pairwise alignment with all known fragment sequences and retained the fragment with highest alignment score, allowing the conversion of the genome string to a list of fragments together with their positions.

[0135] After performing two rounds of selection against thrombin functionalized beads, the random DNA structures resulting after each round of selection were sequenced. The sequencing data was analyzed to detect the prevalence of the aptamer sequences (including the placebo sequence) in the random DNA structures.

[0136] Results

[0137] The sequencing data revealed that random DNA structures which incorporated both the HD22 and TBA aptamers had a 4-fold increase in prevalence consistent with a positive selection for hetero-bivalent binders. In contrast, random DNA structures, including structures comprising the placebo aptamer hairpin loop fragment, had a 4- fold decrease in prevalence after two rounds of selection compared to the initial random DNA structures created in Example 3, consistent with the placebo aptamer hairpin loop fragment creating a fitness disadvantage (Figure 5).

[0138] Example 9

[0139] Generating random DNA structures targeting norovirus capsid protein (VP1) Random DNA structures with binding affinity towards the predetermined target norovirus capsid protein (VPl) and multivalent binding capabilities to norovirus virus like particles were generated, comprising aptamer hairpin loops with the aptamers AG3, BEIER, BUF-2, Ml, M6-2, SMV19, and SMV21.

[0140] Methods

[0141] DNA fragments, detailed in Table 1 and assembled and annealed as described in Example 2, were mixed according to Table 4. A ligation reaction mixture was created for each of the seven aptamers targeting norovirus capsid protein as well as the placebo aptamer, eight unique ligation mixtures in total. In the respective mixtures, the amplification hairpin loop fragment had the lowest concentration of 0.1 pM. Linear fragments and branched fragments were added in a 1.5 times excess. In each mixture one of the eight aptamer hairpin loop fragments (AG3, BEIER, BUF-2, Ml, M6-2, SMV19, SMV21 or placebo aptamer hairpin loop fragment) were added at 120 times excess. To each respective mixture, a concentrated solution of T4 ligase was added.

[0142] The ligation reaction mixtures were placed in a thermal cycler and cooled to 16°C for lOh followed by a hold at 4°C until further processing. The T4 ligase was heat inactivated by incubating the sample for 20 minutes at 65°C. After this, all eight ligation mixtures where pooled to create a mixture of random DNA structures where each random DNA structure could contain one or more copies of one of the eight types of aptamer sequences.

[0143] Table 4: Ligation reaction mixture

[0144] 100 pl of the ligated sample was mixed with 30 pl NEBuffer™ 1 (10 mM Bis- Tris-Propane-HCI, 10 mM MgCl2, 1 mM DTT) (New England Biolabs) and 5 pl of the exonuclease enzyme Exo III. The mixture was incubated in a thermocycler at 37°C for 30 minutes followed by a 30-minute incubation at 70°C to heat inactivate the exonuclease enzyme.

[0145] In the ligation process described above it was expected that as the number of free fragments were depleted and added into growing random DNA structures, some random DNA structures would not complete their growth and feature free ends. These random DNA structures do not form complete genomes and cannot be replicated by PCR, thus they were removed from the pool of generated random DNA structures via exonuclease digestion.

[0146] Furthermore, in order to correctly amplify the random DNA structures after any downstream affinity experiments, thereof preventing circular concatemers, the amplification hairpin loop that was designed to be targeted by PCR primers was cleaved between the two adjacent primer binding sites. In order to do this, the two adjacent primer binding sites were designed so that their border sequence is the target sequence of the restriction enzyme Pvull, therein the amplification hairpin loop could be cleaved exactly between the primer sequences. As restriction enzymes can only cleave double stranded DNA, a splint oligo that makes the border sequence double-stranded was added to the random DNA structure mixture. 8 .1 of a splint oligonucleotide (100 pM), designed to make the target binding sequence situated on the amplification hairpin loop fragment double-stranded, was added to the digested ligation sample as according to Table 5. The mixture was placed in a thermal cycler and annealed via a temperature program consisting of an initial denaturation step of 90°C for 30 seconds followed by a slow temperature ramp from 80 - 20°C temperature ramp over 30 minutes. The program was ended by an incubation at 4°C.

[0147] After annealing, 5 pl of Pvull restriction enzyme was added to the annealed sample, and the sample was placed in a thermal cycler at 37°C for 30 minutes, followed by a placement of the sample on ice.

[0148] Table 5: Splint oligonucleotide annealing reaction mixture

[0149] Results

[0150] Random DNA structures with binding affinity for the predetermined target norovirus virus like particles were generated and ready for downstream selection experiments. At this point in time the mixture comprising the random DNA structures may be frozen if selection is done at a later date.

[0151] Example 10

[0152] Selection of random DNA structures

[0153] Selection of strongly binding random DNA structures targeting norovirus virus like particles from the mixture comprising the random DNA structures generated in Example 9 was performed by immobilizing the predetermined target of interest on magnetic beads.

[0154] Methods 250 pl of magnetic beads (Thermo Fisher Dynabeads™ MyOne™ Carboxylic Acid) were taken and placed on a magnet. The supernatant was removed, and the beads were washed twice with 0.5 ml of activation MEST buffer ((50 mM 2-(N - morpholino)ethanesulfonic acid, 0.01 % Tween 20, pH 6.0). The beads were resuspended in 25 pl MEST buffer and activated by addition of 25 pl EDC (10 mg / ml N-(3-dimethylaminopropyl)-N'-ethylcarbodiimide hydrochloride,) for 30 minutes at room on a rotary mixer.

[0155] After EDC activation, the beads were placed on a magnet and the supernatant of the beads was removed. 20 pg of norovirus GIL4 Virus like particles (VLP) (The Native Antigen Company) in VLP buffer (25 mM MOPS, 500 mM NaCI, pH 6.4 ) was added to the beads and the beads incubated for 30 minutes on a rotary mixer. After conjugation of the beads the supernatant was removed, and the beads washed twice with 250 pl VLP buffer. The beads were now ready for selection experiments.

[0156] The protein functionalized beads were placed in a 1.5 ml Eppendorf tube and the supernatant was removed by the use of a magnet. 500 pl of the mixture comprising the restriction digested random DNA structures, as prepared in Example 9, was added to the beads, and the tube placed on a rotary mixer overnight in room temperature. After this, the beads were placed on a magnet and the supernatant removed and saved (for control purposes). The beads were then washed twice with 250 pl of VLP buffer to remove weakly binding DNA structures. The beads were finally resuspended in 100 pl of pure water and placed in a 95 °C heat block for 5 minutes to release strongly binding random DNA structures. After this, the tube was placed on a magnet and the supernatant recovered and saved.

[0157] Example 11

[0158] PCR amplification of random DNA structures

[0159] In order to perform downstream high-throughput sequencing and identification of the random DNA structures, amplification via PCR to form double-stranded linear fragments was performed on the mixture comprising the selected random DNA structures from Example 10.

[0160] Methods Two opposing primers targeting the amplification hairpin loop region were used and PCR was performed using a standard protocol (65°C annealing temperature) utilizing the polymerase Taq-polymerase (DreamTaq™, Thermo Scientific). The double stranded PCR products were then purified using AmPure XP beads (Beckman Coulter™) to remove PCR buffer, enzyme and residual primers.

[0161] The PCR product was mixed with lambda exonuclease in lambda exonuclease buffer and incubated at 37°C for 30 minutes followed by a 10 minute incubation at 75°C to heat inactivate the enzyme.

[0162] Example 12

[0163] Nanopore sequencing of random DNA structures after repeat selection with norovirus virus like particles

[0164] In order to characterize the amplified, purified and digested random DNA structures resulting from Example 11, a high-throughput sequencing technique was utilized.

[0165] Methods

[0166] The mixture comprising the amplified, purified and digested random DNA structures resulting from Example 11 was prepared using the ligation sequencing kit SQK-LSK110 (Oxford Nanopore Technologies) and following the protocol for ligation sequencing for amplicons. The ends of the PCR fragments were first prepared via enzymatic adenine-tailing, followed by the ligation with Oxford Nanopore Technologies proprietary sequencing adaptors. The random DNA structures were then sequenced on a Flongle™ flow cell (Oxford Nanopore Technologies) on MinlON™ device (Oxford Nanopore Technologies), typically resulting in 100000 - 500 000 reads.

[0167] The sequenced reads were processed by a custom software that located the position of the amplification primer sequence from the amplification hairpin loop that was present in the beginning of the random DNA structure genome.

[0168] After finding the position of the initial primer, the custom software moved progressively through the genome and performed pairwise alignment with all known fragment sequences and retained the fragment with the highest alignment score, allowing the conversion of the genome string to a list of fragments together with their positions. After performing three rounds of selection against norovirus VLP (virus like particles) functionalized beads, the random DNA structures resulting after each round of selection were sequenced. The sequencing data was analyzed to detect the prevalence of the aptamer sequences (including the placebo sequence) in the random DNA structures.

[0169] Results

[0170] The sequencing data revealed a large difference in selection performance amongst the seven established aptamers. Only the aptamers BUF2 and SMV21 showed a positive selection bias whereas the other 5 established aptamers and the placebo aptamer hairpin loop showed strong negative selection bias (over 30 fold reduction after 3 rounds of selection) (Figure 6).

[0171] Beyond this, the prevalence of random DNA structures that included multiple norovirus virus like particles binding aptamers was analyzed and it was discovered that DNA structures including 3, 4 or 5 aptamers were greatly refined in selection (Figure 7), demonstrating a positive selection bias for DNA structure able to bind multivalently to norovirus virus like particles. DNA structures containing 5 aptamers showed an over 60-fold increase in prevalence after 3 rounds of selection.

Claims

CLAIMS1. A method for generating a plurality of random DNA structures, said plurality of random DNA structures having affinity for a predetermined target, the method comprising:(a) designing a library comprising a plurality of DNA fragments;(b) synthesizing said library comprising said plurality of DNA fragments; and(c) randomly ligating said synthesized plurality of DNA fragments to generate a plurality of random DNA structures wherein at least one of said random DNA structures comprises a contiguous sequence, a native capacity to be amplified, and multivalent binding capabilities to said predetermined target.

2. The method according to claim 1, wherein said library comprises at least two classes of DNA fragments selected from a first class comprising linear fragments, a second class comprising branched fragments, and a third class comprising loop fragments.

3. The method according to claim 2, wherein the 3'-end or the 5'-end of at least a portion of said linear fragments, branched fragments and loop fragments start with a single stranded extension.

4. The method according to claim 3, wherein said single stranded extension comprises a nucleotide sequence which is palindromic.

5. The method according to claim 4, wherein said single stranded extension comprising a nucleotide sequence which is palindromic comprises at least 2 nucleotides, such as at least 4 nucleotides, such as at least 6 nucleotides.

6. The method according to claim 5, wherein said single stranded extension comprises a nucleotide sequence which is selected from the group consisting of AT, TA, CG, GC, AATT, ATAT, ACGT, AGCT, TATA, TTAA, TCGA, TGCA, CATG, CTAG, CCGG, CGCG, GATC, GTAC, GCGC, GGCC, AAATTT, AATATT, AACGTT,AAGCTT, ATATAT, ATTAAT, ATCGAT, ATGCAT, ACATGT, ACTAGT, ACCGGT, ACGCGT, AGATCT, AGTACT, AGCGCT, AGGCCT, TAATTA, TATATA, TACGTA, TAGCTA, TTATAA, TTTAAA, TTCGAA, TTGCAA, TCATGA, TCTAGA, TCCGGA, TCGCGA, TGATCA, TGTACA, TGCGCA, TGGCCA, CAATTG, CATATG, CACGTG, CAGCTG, CTATAG, CTTAAG, CTCGAG, CTGCAG, CCATGG, CCTAGG, CCCGGG, CCGCGG, CGATCG, CGTACG, CGCGCG, CGGCCG, GAATTC, GATATC, GACGTC, GAGCTC, GTATAC, GTTAAC, GTCGAC, GTGCAC, GCATGC, GCTAGC, GCCGGC, GCGCGC, GGATCC, GGTACC, GGCGCC, and GGGCCC.

7. The method according to claim 6, wherein said single stranded extension comprises a nucleotide sequence which is AATT.

8. The method according to claim 2 wherein the 5'-end of said linear fragments, branched fragments and loop fragments are synthesized with a 5'-end phosphate modification.

9. The method according to claim 2, wherein said third class of fragments comprising loop fragments comprises at least one aptamer hairpin loop fragment and at least one amplification hairpin loop fragment.

10. The method according to claim 9, wherein said at least one aptamer hairpin loop fragment comprises at least one aptamer sequence, said aptamer sequence having affinity for said predetermined target.

11. The method according to claim 9, wherein said at least one amplification hairpin loop fragment comprises an amplification hairpin loop region, said amplification hairpin loop region comprising at least two adjacent primer binding sites, wherein said adjacent primer binding sites comprise a target binding sequence for at least one restriction enzyme.

12. The method according to claim 10, wherein said at least one random DNA structure of said plurality of random DNA structures comprises at least two aptamer hairpin loop fragments and at least one amplification hairpin loopfragment, wherein said at least two aptamer hairpin loop fragments comprise identical aptamer sequences or two different aptamer sequences.

13. The method according to any of the preceding claims, wherein the method further comprises one or more of the steps of:(d) treating said plurality of random DNA structures with an exonuclease;(e) annealing at least one splint oligonucleotide complementary to said target binding sequence for at least one restriction enzyme;(f) restriction digesting said random DNA structures by at least one restriction enzyme recognizing said target binding sequence as defined in claim 11 for at least one restriction enzyme;(g) affinity selection of said plurality of random DNA structures;(h) amplifying by PCR said eluted random DNA structures resulting in dsDNA linear products corresponding to said eluted random DNA structures;(i) purifying said amplified dsDNA linear products corresponding to said eluted random DNA structures;(j) sequencing said purified amplified dsDNA linear products corresponding to said eluted random DNA structures;(k) recovering said random DNA structures from said dsDNA linear products via treatment of said dsDNA linear products using an exonuclease; and(l) repeating at least once said affinity selection using said recovered random DNA structures such that random DNA structures with a higher affinity towards the predetermined target are selected.

14. The method according to claim 13, wherein said affinity selection comprises immobilizing said predetermined target on one or more beads and contacting said one or more beads comprising said predetermined target with said plurality of random DNA structures.

15. The method according to claim 13, wherein said one or more beads are magnetic beads.

16. The method according to claim 13, wherein said affinity selection further comprises recovering said contacted beads and eluting bound random DNA structures from the beads.

17. A random DNA structure having affinity for a predetermined target, obtained by the method according to any one of claims 1-16.

18. A pharmaceutical composition comprising the random DNA structure according to claim 17 and at least one pharmaceutically acceptable carrier.

Citation Information

Patent Citations

  • Structure assisted directed evolution of multivalent aptamers

    US20180044663A1

  • High efficiency aptamer complex comprising branched DNA and aptamer, and use thereof

    US20210371863A1

  • Structure assisted directed evolution of multivalent aptamers

    US11168320B2

  • Methods for using nucleic acid aptamers for directed templated assembly

    WO2017205277A2