DNA library preparation method

Abstract

The present invention relates to a method for preparing a double-stranded DNA library from a single-stranded DNA fragment, and provides a library preparation method and the like that produce a small amount or content of adaptor dimers and are highly efficient. The present invention is a method for preparing a double-stranded DNA library from a single-stranded DNA fragment, comprising specific steps (i) to (v).

Description

Technical Field

[0001] The present invention relates to a method for preparing a double-stranded DNA library from single-stranded DNA fragments. Background Art

[0002] For a long time, research has been developed to efficiently prepare double-stranded DNA libraries from single-stranded DNA fragments. Among them, the library preparation method called "TACS-T4" developed by the inventor (refer to non-patent document 1) is an excellent preparation method that can substantially improve the library yield by 100 to 10,000 times compared with existing methods, such as methods using CircLigase II (refer to non-patent document 2) and methods using T4 DNA ligase (refer to non-patent document 3). It should be noted that in this TACS-T4, the "TACS connection" technology previously developed by the inventor is used (refer to non-patent document 4). TACS (TdT-Assisted Chimeric ssDNA) connection is a highly efficient single-stranded DNA connection technology, which is a unique connection reaction between single-stranded DNAs developed with a view to showing a particularly high reaction efficiency when the 3' end of the nucleic acid acted on by RNA ligase is RNA. TdT (Terminal Deoxynucleotidyl Transferase) can act on single-stranded DNA efficiently. In the presence of ribonucleoside triphosphates, the enzyme adds 3 to 4 bases of ribonucleotides to the 3' end of single-stranded DNA to stop extension. TACS ligation is a technology that efficiently connects single-stranded DNA by combining RNA tailing using TdT and the activity of RNA ligase.

[0003] However, although the "TACS-T4" library preparation method can make the library yield more efficient as described above, there is the following problem: as the double-stranded library is prepared, a large amount of dimers of the adapter sequence used in the preparation (double-stranded adapter dimers) are produced in the reaction system as by-products. If such adapter dimers are produced, there is the following problem: when the prepared library is read with a sequencer, a reagent for removing the dimer is required, and even if the dimer is to be removed by purification, it is difficult to perform purification when the difference in base length between the insert DNA of the sequence to be read and the adapter sequence is small.

[0004] Prior art literature

[0005] Non-patent literature

[0006] Non-patent literature 1: Osamu Hisano et al., BMC Biology (2021) 19: 225

[0007] Non-patent document 2: Marie-Theres Gansauge et al, Nat. Protoc., 2013Apr; 8(4): 737-48

[0008] Non-patent document 3: Marie-Theres Gansauge et al, Nucleic Acids Research, 2017, Vol. 45, No. 10e79

[0009] Non-patent document 4: Fumihito Miura et al., Nucleic Acids Research, 2019, Vol. 47, No. 15e85 Summary of the invention

[0010] Under such circumstances, regarding methods for preparing double-stranded DNA libraries from single-stranded DNA fragments, it is desired to develop a method for preparing libraries with a low amount or content of adaptor dimers and high efficiency.

[0011] The present invention has been made in view of the above-mentioned circumstances, and provides a method for preparing a double-stranded DNA library as shown below.

[0012] [1] A method for preparing a double-stranded DNA library from single-stranded DNA fragments, comprising the following steps (i) to (v):

[0013] Step (i), connecting the single-stranded adapter sequence A1 with phosphorylated 5' end to the 3' end of the single-stranded DNA fragment with dephosphorylated both ends to generate single-stranded DNA with added adapter;

[0014] Step (ii), after annealing the single-stranded adapter sequence A2 with the adapter sequence A1 contained in the single-stranded DNA to which the adapter is added, using a DNA polymerase to extend the 3' end of the adapter sequence A2 to generate a double-stranded DNA to which the adapter is added;

[0015] Step (iii), reacting a topoisomerase with a double-stranded adapter sequence B1 having only one 5' end phosphorylated and having a topoisomerase recognition sequence in a region including a 3' end complementary to the 5' end, to generate a topoisomerase composite double-stranded adapter sequence B2 in which the topoisomerase is conjugated to the 3' end of the recognition sequence;

[0016] Step (iv), connecting the topoisomerase composite double-stranded adapter sequence B2 to the end of the double-stranded DNA to which the adapter is added, where the adapter-free end is not added, to generate double-stranded DNA to which the adapter is added at both ends; and

[0017] Step (v) is to obtain a library containing a plurality of double-stranded DNAs by performing PCR using the double-stranded DNA to which adapters are added at both ends as a template.

[0018] [2] According to the preparation method described in [1] above, wherein the topoisomerase composite double-stranded adapter sequence B2 is not connected to the double-stranded adapter sequence C, and the double-stranded adapter sequence C is generated by annealing the unreacted single-stranded adapter sequence A1 with the unreacted single-stranded adapter sequence A2.

[0019] [3] The preparation method according to [1] above, wherein the double-stranded adapter sequence B1 has a nick in the region including the phosphorylated 5' end, and the single-stranded adapter sequence A1 has uracil or dSpacer in the sequence.

[0020] [4] The method according to [3] above, wherein step (iv) includes the step of adding an endonuclease to the generated reaction system.

[0021] [5] The preparation method according to the above [1], wherein the connection and generation in step (i) are performed using the TACS connection method.

[0022] [6] A double-stranded DNA library obtained by the preparation method described in any one of [1] to [5] above.

[0023] Effects of the Invention

[0024] According to the present invention, a method for preparing a double-stranded DNA library from single-stranded DNA fragments can be provided, which can extremely effectively increase the library yield compared to previous double-stranded DNA library preparation methods (the above-mentioned non-patent documents 2 and 3), and can effectively inhibit the by-production of double-stranded linker dimers during preparation. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 : A diagram showing the formation and purification results of a VTopoI-oligonucleotide complex (VOC) (VTopoI: topoisomerase I derived from vaccinia virus).

[0026] A. Pathway of complex formation. The formation of VOC is an equilibrium reaction. Therefore, phosphorylation of the released DNA is effective in increasing the amount of the complex.

[0027] B and C. Examples of VOC formation. Complex formation was enhanced by the addition of T4 PNK. Agarose gel electrophoresis images using SDS-PAGE (B) and E-Gel Ex (Thermo Fisher Scientific) (C).

[0028] D. Purification of VOC using heparin column chromatography. Analysis results of fractions obtained using SDS-PAGE (upper figure) and chromatogram (lower figure). It can be seen that VOC elutes earlier than VTopoI.

[0029] Figure 2 : Diagram showing the results of a study on the substrate specificity of VOCs.

[0030] A. Structures of VOC (left) and TOPO substrates (right).

[0031] B. Gel electrophoresis images of VOCs mixed with TOPO substrate. For TA (left) and blunt-ended (right) VOCs, the reactivity of all combinations of 5'-phosphate and single-base overhangs with the substrate was investigated. If ligation occurs, the band is expected to shift upward.

[0032] The meanings of the symbols are as follows: o: hydroxyl end. p: phosphorylated end. A, C, G, or T: nucleotide bound to the 5' end of the substrate. The details of the experiment are shown below.

[0033] <Study on the ligation specificity of TOPO-activated adaptors>

[0034] Incubate 0.5×TACS buffer, 1×ExTaq buffer, 200μM ATP, 200μM dNTPs each, and 20% (w / v) PEG6000 at 95°C for 3 minutes and at 55°C for 5 minutes. After cooling the reaction temperature to below 37°C, add 5μL of 0.5M EDTA solution (pH8.0) and 150ng of VOC, and incubate at room temperature for 15 minutes. Next, add 45μL of Buffer B2 and 5μL of proteinase K to the reaction solution, and incubate at 50°C for 15 minutes. Add 100μL of isopropanol to the reaction solution, and load the mixture onto a ZymoSpin column (Zymo Research). Wash the column with Buffer PE (Qiagen), and elute the DNA with 10μL of 10mM Tris-HCl, pH8.5. Load the purified DNA onto a 10% acrylamide gel (TBE buffer system). After electrophoresis, gel images were captured using the ChemiDock imaging system.

[0035] Figure 3 : A diagram showing the results of comparison of library preparation pathways from single-stranded DNA.

[0036] A and B. TACS-T4 (A) and TACS-TOPO (B). Shows the pathway (left) and the results of DNA analysis in each reaction step based on denaturing gel electrophoresis (right). The numbers in italic font indicate the length of the DNA used. The colored asterisks in the pathway on the left correspond to the bands in the gel image on the right. The lanes refer to 1: before TACS ligation, 2: after TACS ligation, 3: after complementary chain synthesis based on Taq DNA polymerase, 4: after the second adapter ligation, and 5: after SPRI purification.

[0037] C. Denaturing gel electrophoresis image of the library after PCR amplification. The DNA fragments amplified from the library prepared using TACS-T4 (lanes 1, 2) and TACS-TOPO (lanes 3, 4) were loaded onto the denaturing gel. The appearance of the adapter dimer was only seen in the library prepared using TACS-T4. The library containing the insert is indicated by a red arrow, and the dimer is indicated by a yellow arrow.

[0038] Figure 4 : A diagram showing the improvement of the TACS-TOPO method and its results.

[0039] A and B. In the initial version (ver. 1) of the TACS-TOPO approach (A), the formation of adaptor dimers was observed (B).

[0040] C and D. By slightly improving the adaptor used in TACS ligation (C), the sensitivity of preparing a dimer-free library was improved, and a highly sensitive library could be prepared from ssDNA (D). In D, a ver. 4 adaptor was used.

[0041] Figure 5 : A graph showing the results of preparing sequencing libraries from cell-free DNA (cfDNA).

[0042] A. Comparison of libraries prepared by TACS-T4 and TACS-TOPO. NPD: nucleosome-protected DNA fragments, C3D: free short single-stranded DNA.

[0043] B. Library yield of TACS-TOPO ver4. The library yield in various volumes of serum is shown. The area separated by horizontal lines shows the library amplification factor required to obtain 90 Gb read segments.

[0044] C. Example of electrophoresis analysis of a cfDNA-derived sequencing library prepared using TACS-TOPO ver. 4. After electrophoresis in a 6% denaturing gel, the gel was stained with SYBR Gold and an image of the gel was taken.

[0045] D. Size distribution of reads mapped to the human reference genome. Example of a library prepared from 50 μL of serum using TACS-TOPO ver4. After sequence determination, it was mapped to the human reference genome.

[0046] E. Peaks and G4 motifs identified by G4-seq were co-localized with reads. Fragments of 35 to 75 bases were selected from the mapped reads and analyzed for co-localization with peaks and G4 motifs from G4-seq.

[0047] Figure 6 : A diagram showing the comparison results of library preparation based on the TACS-TOPO method using DNA derived from ancient human bones and a conventional protocol corresponding only to dsDNA.

[0048] A to D. Shows the library amount (A) when the entire library is amplified by 12 cycles of PCR, the size distribution of the library (gel electrophoresis analysis, B), the average alignment length of sequenced reads (C), and the average mapping rate of reads to the human reference genome (D). The ThruPlex DNA-seq kit has the best library preparation sensitivity among commercially available library preparation kits and was therefore selected as a representative solution.

[0049] Libraries prepared using ThruPlex (E), TACS-TOPO ver.4 (F), and TACS-TOPO ver.5 (G) were sequenced and mapped to the reference genome. The pattern of mismatch-detected sites when the results were analyzed using mapdamage is shown.

[0050] Figure 7 : A diagram showing the purification results of VTopoI.

[0051] A. SDS-PAGE analysis results of VTopoI expression obtained based on three expression vectors.

[0052] B and C. Heparin column purification of VTopoI. Example of purification of VTopoI prepared using pColdI. Chromatogram (B) and SDS-PAGE gel image (C) are shown.

[0053] D. SDS-PAGE gel image of purified VTopoI.

[0054] E. Supercoil relaxation assay of purified VTopoI. The details of this assay are shown below.

[0055] <Superhelix relaxation assay (topoisomerase activity assay)>

[0056] To a 20 μL reaction solution containing 180 ng (100 fmol) of pUC19 DNA in 50 mM Tris-HCl, pH 8.0 and 75 mM NaCl, 0, 10, 20, or 40 pmol of VTopoI was added and incubated at room temperature for 10 minutes. Next, 2 μL of 10% (w / v) sodium dodecyl sulfate (SDS) was added to the reaction solution and loaded onto a 1% (w / v) agarose gel (TAE buffer system). After electrophoresis, the gel was stained with SYBR Gold gel stain (Thermo Fisher Scientific) and the gel image was captured using a ChemiDock imaging system (Bio-rad Laboratories, Hercules, CA).

[0057] Figure 8 : A diagram showing improvements in adaptors for preparing libraries free of adaptor dimers and the results thereof.

[0058] A. The structures of the adapters studied.

[0059] B. Principle of uracil-containing adaptors that inhibit adaptor dimer formation.

[0060] C-H. TACS-TOPO protocol and study of the amount of dimer formation. Representative gel images of libraries after PCR amplification (C-G) and summary of the relative amount of dimer formation (H). Libraries were prepared from various amounts of chemically synthesized random 50-base DNA (N50), PCR amplified, and analyzed by denaturing gel electrophoresis. It can be seen that the relative amount of adapter dimers is reduced by improving the adapter. When drawing H, three independent experiments were performed. DETAILED DESCRIPTION

[0061] The present invention is described in detail below. The scope of the present invention is not limited by these descriptions, and for situations other than the following examples, it can be implemented by appropriate changes within the scope that does not harm the gist of the present invention. It should be noted that this specification contains the entire contents of Japanese Patent Application No. 2022-152384 (filed on September 26, 2022), which is the basis for claiming priority to this application.

[0062] All publications cited in this specification, such as prior art documents and patent documents such as open gazettes and patent gazettes, are incorporated into this specification as references.

[0063] 1. Summary of the Invention

[0064] The present invention provides an efficient method for preparing a library, wherein the library is used for sequence determination from single-stranded DNA (ssDNA). The initial adapter connection for single-stranded DNA uses the TACS connection developed by the inventors in recent years. Then, after the complementary chain is synthesized for the single-stranded DNA to which the adapter is added, the second adapter connection is performed using an adapter connection based on topoisomerase I (VTopoI) derived from vaccinia virus. Based on these technical principles, the TACS-TOPO method can prepare a sequencing library with a small amount of adapter dimer generation and content from a small amount of single-stranded DNA as small as 24pg. Using the TACS-TOPO method, a library for free DNA (cfDNA) analysis can be prepared from 50μL of human serum without amplification. In addition, by using a scheme based on the TACS-TOPO method, an attempt was made to analyze DNA extracted from ancient human bones, and the results confirmed that the library yield was improved by two orders of magnitude compared to the library preparation method used in the past.

[0065] 2. DNA library preparation method

[0066] The method for preparing a DNA library according to the present invention (hereinafter also referred to as "the method for preparing a library according to the present invention") is a method for preparing a double-stranded DNA library from single-stranded DNA fragments.

[0067] Specifically, the library preparation method of the present invention includes the following steps (i) to (v), but is not limited thereto and may include any other steps.

[0068] (i) a step of ligating the single-stranded adapter sequence A1 whose 5' end is phosphorylated to the 3' end of the single-stranded DNA fragment whose both ends are dephosphorylated to generate a single-stranded DNA to which the adapter is added (step (i));

[0069] (ii) annealing the adapter sequence A1 portion contained in the single-stranded DNA to which the adapter is added with the single-stranded adapter sequence A2, and then extending the 3' end of the adapter sequence A2 using a DNA polymerase to generate a double-stranded DNA to which the adapter is added (step (ii));

[0070] (iii) reacting a double-stranded adapter sequence B having only one 5' end phosphorylated and having a topoisomerase recognition sequence in a region including a 3' end complementary to the 5' end with a topoisomerase to generate a topoisomerase composite double-stranded adapter sequence B in which the topoisomerase is conjugated to the 3' end of the recognition sequence (step (iii));

[0071] (iv) a step of ligating the topoisomerase composite double-stranded adapter sequence B to the end of the double-stranded DNA to which the adapter is added, to generate double-stranded DNA to which the adapter is added at both ends (step (iv)); and

[0072] (v) A step of obtaining a library containing a plurality of double-stranded DNAs by performing PCR using the double-stranded DNA to which adaptors are added at both ends as a template (step (v)).

[0073] Regarding step (i)

[0074] As the so-called "single-stranded DNA fragment" of insert DNA in library preparation, a single-stranded DNA fragment with both ends (5' end and 3' end) dephosphorylated is used. Here, the single-stranded DNA fragment preferably has a ribonucleotide added to its 3' end, and the ribonucleotide is preferably added with several bases, more preferably 1 to 3 bases, or 3 bases (for example, adenine (rA) as 3 ribonucleotides). The addition of the ribonucleotide, i.e., RNA tailing (or Ribotailing), is preferably performed using TdT (Terminal Deoxynucleotidyl Transferase), for example. TdT adds ribonucleotides to the 3' end of the single-stranded DNA in the presence of ribonucleoside triphosphates to stop extension. Next, the "single-stranded adapter sequence A1" as a single-stranded DNA that is phosphorylated at the 5' end is connected to the 3' end of the single-stranded DNA fragment thus obtained. The connection is preferably performed using RNA ligase, for example. After this connection, a "single-stranded DNA with an adapter added" is generated. There is no limitation on the detailed process of generating the single-stranded DNA with the adapter added in step (i) (including the conditions of various reactions, etc.), and for example, reference may be made to the "TACS ligation" technology disclosed in the above-mentioned non-patent document 4 (Nucleic Acids Research, 2019). If an enzyme with higher ligation activity and / or efficiency than the enzyme used in TACS ligation is used, the enzyme may be appropriately substituted.

[0075] Regarding step (ii)

[0076] Step (ii) is a step of making the single-stranded DNA with the addition of adapters generated in step (i) into double-stranded DNA. In detail, the "single-stranded adapter sequence A2" as a single-stranded DNA is annealed with the adapter sequence A1 portion in the generated single-stranded DNA with the addition of adapters. The single-stranded adapter sequence A2 (especially the portion on its 3' end side) preferably has a base sequence complementary to at least a part of the base sequence of the adapter sequence A1 portion. After the annealing, the 3' end of the adapter sequence A2 is extended using a DNA polymerase. As a DNA polymerase, it is not limited, but preferably Taq DNA polymerase, etc. . Thereby, a "double-stranded DNA with the addition of adapters" is generated as a double-stranded DNA. For the detailed process of generating double-stranded DNA with the addition of adapters in step (ii) (including various reaction conditions, etc.), reference may be made to a part of the steps of the "TACS-T4" library preparation method disclosed in the above-mentioned non-patent document 1 (BMC Biology, 2021).

[0077] Regarding step (iii)

[0078] Step (iii) is a step of preparing a "double-stranded adapter sequence B1" as an adapter sequence for double-stranded DNA for the so-called second connection (specifically, the following step (iv)). First, in the double-stranded adapter sequence B1, the 5' end of one single-stranded DNA is phosphorylated, while the 5' end of the other single-stranded DNA is not phosphorylated, and a topoisomerase recognition sequence is present in the region containing the 3' end complementary to the phosphorylated 5' end. The topoisomerase is preferably topoisomerase I, more preferably topoisomerase I (VTopoI) derived from vaccinia virus. In addition, the recognition sequence of the VTopoI is not limited, and 5'-CCTTT-3' or 5'-(C / T)CCTT-3' can be cited. VTopoI specifically recognizes these sequences consisting of 5 bases and forms a covalent bond with the phosphate group of its 3'-T. Moreover, with this formation, the base sequence closer to the 3' side than the 3'-T of the above-mentioned recognition sequence is cut. It is preferred that a nick (break in the base sequence) is present in the region of the double-stranded adapter sequence B1 that includes the phosphorylated 5' end. The nick is not limited, and for example, it is preferably present at the base on the 5' side of the sequence complementary to the recognition sequence of VTopoI, and more preferably at one 5' side base of the complementary sequence. By having the nick, when the base sequence closer to the 3' side than the 3'-T of the recognition sequence is cut, the sequence containing the sequence complementary to the cut sequence is also cut from the double-stranded adapter sequence B1. Thus, a "topoisomerase composite double-stranded adapter sequence B2" is generated. In the case where the double-stranded adapter sequence B1 has the above-mentioned nick, the nick or the missing portion corresponding to the nick preferably remains in the double-stranded adapter sequence B2.

[0079] Regarding step (iv)

[0080] The topoisomerase composite double-stranded adapter sequence B2 generated in step (iii) is connected to the end of the double-stranded DNA with added adapters generated in step (ii) without added adapter sequence. Topoisomerase (VTopoI) has the activity of connecting the phosphate group of the 3'-T to which it is covalently bonded (i.e., the phosphorylated 3' end) to the dephosphorylated 5' end (i.e., the hydroxyl group of the 5' end) of the double-stranded DNA with added adapters. Thus, "double-stranded DNA with added adapters at both ends" is generated. On the other hand, the topoisomerase composite double-stranded adapter sequence B2 is not connected to the double-stranded unreacted adapter (the 5' ends of both chains are phosphorylated) in principle. The double-stranded unreacted adapter is obtained by annealing the above-mentioned single-stranded adapter sequence A1 with the single-stranded adapter sequence A2 (i.e., the free sequence A1 is obtained by annealing with the sequence A2). Therefore, the formation of a dimer between the unreacted adapter and the adapter sequence B2 can be effectively inhibited. Even if the dimer is formed accidentally, for example, when the double-stranded adapter sequence B1 has a nick as described in the above step (iii) (it should be noted that the nick or the missing portion corresponding to the nick also remains in the double-stranded adapter sequence B2), and the above single-stranded adapter sequence A1 has uracil in the sequence, by adding a DNA endonuclease to the reaction system after the "double-stranded DNA with adapters added to both ends" is generated, the above dimer can be decomposed to the same state as when the above dimer is not formed. As the DNA endonuclease, there can be cited a DNA endonuclease having the activity of hydrolyzing the N-glycosidic bond in uracil or introducing a nick into the DNA main chain on the 5' side of uracil, such as UDG (uracil DNA glycosylase), APE1 (human apurinic / apyrimidinic (ap) endonuclease) and EndoQ (endonuclease Q), etc. By utilizing the uracil removal of DNA endonucleases or the nicks near uracil and the nicks in the double-stranded adapter sequence B2 (more specifically, the nicks at the connection site with the double-stranded adapter sequence B2), the formed adapter dimer can be decomposed (cracked), and as a result, it can be removed from the reaction system. In addition, the uracil in the above-mentioned single-stranded adapter sequence A1 can also be replaced by "dSpacer" as a spacer-modified oligoDNA. Here, the spacer-modified oligoDNA is a structure without bases, which can be introduced into the phosphodiester bond backbone of the DNA chain, including types that only lack bases, types that lack both bases and deoxyribose rings, and the number of spacer arms can be variously selected. dSpacer is a type that lacks a base portion, which can ease the double-stranded stability of the inserted sequence portion, and can also control the double-stranded stability by the inserted site and structure. In the case of introducing dSpacer as described above, it has the effect of not being easy to produce adapter dimers, and the adapter dimers can also be easily removed by enzymolysis.In addition, an "ethynyl dSpacer" (an alkyne structure is bonded to the 1' position of the sugar to which the base is bonded in a normal oligoDNA chain) may be used instead of or in combination with the dSpacer.

[0081] Regarding step (v)

[0082] Step (v) is a step of obtaining a library containing a plurality of double-stranded DNAs by performing PCR using the generated double-stranded DNA with adapters added to both ends as a template. The reaction conditions of PCR, the purification conditions of the resulting PCR product (library), etc., can be appropriately set, and some steps of the "TACS-T4" library preparation method disclosed in the above-mentioned non-patent document 1 (BMC Biology, 2021) can also be appropriately referred to.

[0083] In the present invention, a double-stranded DNA library obtained by the above-mentioned library preparation method of the present invention is also included in the scope of the present invention.

[0084] Hereinafter, the present invention will be described in more detail with reference to Examples, but the present invention is not limited to these Examples.

[0085] [Example 1]

[0086] 1. Summary

[0087] Although there are many methods for preparing libraries for sequence determination from double-stranded DNA (dsDNA), the options for methods for single-stranded DNA (ssDNA) are limited. There are three main principles for preparing sequencing libraries from single-stranded DNA.

[0088] The first method is to connect the homopolymer to the 3' end of the single-stranded DNA using terminal deoxynucleotidyl transferase (TdT). After the homopolymer is connected, its tail is used as a primer binding site for double-stranding the single-stranded DNA, and then the second adapter is connected using T4 DNA ligase. Although the library preparation using TdT is efficient, and library preparation kits based on this principle are also sold, the presence of the long terminal tail may become an obstacle in subsequent sequence determination and analysis.

[0089] The second method is a method based on a single-stranded DNA ligation reaction using RNA ligase, in which a 5' phosphorylated adapter is ligated to the 3' end of the single-stranded DNA. For example, Gansauge et al. established a method based on CircLigase II [References 1, 2]. After the adapter ligation, the conversion from single-stranded DNA to double-stranded DNA and the second adapter ligation using T4 DNA ligase are followed. Although RNA ligase can connect two single-stranded DNAs to each other, its reaction efficiency is less than a few percent. Therefore, the yield of the library preparation method using RNA ligase is extremely limited.

[0090] The third method is a method using T4 DNA ligase to achieve single-stranded DNA ligation [Reference 3]. This method uses a double-stranded adapter with a single-stranded random hexamer at one end. If the random hexamer is able to attach to the end of the target single-stranded DNA, the nick at the end of the partially double-stranded adapter will be connected by T4 DNA ligase. This splint ligation-based method is called ssDNA2.0 and shows higher efficiency than the method using CircLigase II [Reference 3]. However, although both RNA ligase and T4 DNA ligase-based methods have the advantage of not introducing extra insertion bases at the ligation site, the efficiency of library preparation is low and there is still room for improvement.

[0091] In order to improve the efficiency of preparing sequencing libraries from single-stranded DNA, the inventors developed an adapter ligation method for single-stranded DNA named TACS ligation [Reference 4]. In TACS ligation, the inventors used TdT to add several adenines to single-stranded DNA, and this process was called Ribotailing. Then, RNA ligase was used to perform adapter ligation on the Ribotailed single-stranded DNA. By Ribotailing the 3' end of the single-stranded DNA, the 3' end of the target single-stranded DNA looks like RNA, which can greatly improve the ligation efficiency of RNA ligase. As a result, TACS ligation can connect adapters to the 3' end of more than 80% of the target single-stranded DNA [Reference 4]. It is worth noting that the tailing reaction of ribonucleotides using TdT will automatically stop after extending 1 to 3 bases of nucleotides [Reference 5], so Ribotailing itself has almost no effect on subsequent sequence analysis. TACS ligation is used for the purpose of analyzing short single-stranded DNA contained in free DNA in human blood, and its effectiveness has been confirmed. After ligating the free DNA (cfDNA) with adapters using the TACS ligation method, the target single-stranded DNA is double-stranded, and then the second adapter addition reaction is performed using T4 DNA ligase (TACS-T4 method). The TACS-T4 method has improved the library yield by several orders of magnitude compared to the method using CircLibase II and T4 DNA ligase [Reference 6]. However, although the TACS-T4 method improves the library yield, there are problems that should be solved in practice. This is due to the low efficiency of the second adapter ligation reaction and the generation of a non-negligible amount of adapter dimers. The ligation efficiency can be improved by optimizing the reaction, but this will also increase the amount of adapter dimer formation. If the unreacted adapter is removed before the second adapter ligation reaction, the formation of adapter dimers may be avoided. However, the addition of a purification step will inevitably cause the loss of rare DNA. Moreover, when targeting smaller molecular DNA, there is no difference in size between the adapter and the target library molecule, so it is difficult to remove the unreacted adapter. Therefore, it is believed that an alternative method with less adapter dimer formation and high adapter ligation efficiency needs to be developed. To this end, the present inventors believed that it might be more efficient to use vaccinia virus topoisomerase (VTopoI) having a reaction directionality different from that of T4 DNA ligase to perform adapter ligation reaction. That is, while a large number of DNA / RNA ligases ligate DNA phosphorylated at the 5' end to the hydroxyl group at the 3' end of the target DNA, VTopoI ligates DNA phosphorylated at the 3' end to the hydroxyl group at the 5' end of the target DNA.Due to the difference in its substrate specificity, it is considered that VTopoI shows no reactivity to the adaptor originally used in the adaptor ligation whose 5' end is phosphorylated in principle.

[0092] The gene encoding VTopoI was identified by Stewart Shuman in 1987 [Reference 7]. Since then, he and his colleagues have studied the properties of this enzyme in detail [References 8-16]. The idea of ​​using VTopoI as a tool for gene cloning was first described in 1994 [Reference 17]. Later, the technology was commercialized by Invitrogen (now Thermo Fisher Scientifics). Later, VTopoI was used as an efficient enzyme for inserting arbitrary DNA fragments into vectors in the so-called TOPO cloning technology [Reference 18]. Once activated by VTopoI, the vector DNA is ready to be ligated with the target DNA. The reaction occurs immediately, efficiently and specifically. Due to this property, TOPO cloning has become one of the options that is frequently used as the first step to start gene cloning. On the other hand, although the properties of VTopoI are very attractive in developing new technologies, the examples of the use of VTopoI for ligation in the literature are extremely limited [Reference 19].

[0093] In this example, VTopoI and TACS were used in combination for the purpose of efficiently preparing a library for sequence determination from single-stranded DNA. The preparation of VTopoI, the formation and purification of a complex group (VOC) of VTopoI and oligonucleotide (ODN), and the establishment and application of the TACS-TOPO method for preparing a library from single-stranded DNA using VOC will be described below.

[0094] 2. Materials and Methods

[0095] Vector construction

[0096] The gene encoding VTopoI (GenBank accession number: L13447.1) has been optimized for the codon usage frequency of Escherichia coli K-12 strain and was artificially synthesized by Eurofins Genomics in a form without BamHI and EcoRI recognition sequences inside but with BamHI and EcoRI recognition sequences at the 5' and 3' ends of the gene, respectively (sequence number 27).

[0097] Synthetic gene encoding VTopoI (underlined at BamHI and EcoRI sites)

[0098]

[0099] (Serial No. 27)

[0100] The DNA fragment excised with BamHI and EcoRI was subcloned into the BamHI-EcoRI site of pColdTF, pColdI, and pET28a. In order to produce T4 polynucleotide kinase (T4 PNK), the pseT gene of T4 phage was PCR amplified from genomic DNA (T4 GT7 DNA, Nippon Gene) in a form having BamHI and EcoRI recognition sequences at the 5' and 3' ends (SEQ ID NO: 28) and cloned into the BamHI-EcoRI site of pColdI.

[0101] PCR amplified gene encoding T4 polynucleotide kinase cloned into pColdI (BamHI and EcoRI sites are underlined, sequences of primers for gene amplification are in italics)

[0102]

[0103] (Serial No. 28)

[0104] These expression vectors were introduced into T7Express (New England Biolabs, Ipswitch, MA), and the transformant Escherichia coli was cultured in LB medium (1% (w / v) tryptone, 0.5% (w / v) yeast extract, and 0.5% (w / v) sodium chloride) containing 100 μg / ml carbenicillin (for pColdTF and pColdI, Nakalai Tesque) or 50 μg / ml kanamycin (for pET28a, Wako Pure Chemical Industries, Ltd.).

[0105] Enzyme purification

[0106] Bacterial culture

[0107] E. coli into which the expression vector was introduced was inoculated into 3 ml of LB medium containing antibiotics and cultured with shaking at 37°C overnight. The seed culture was transferred to a 2L Erlenmeyer flask to which 1L of LB medium containing antibiotics was added and further cultured with shaking at 37°C. After that, in the case of pColdI and pColdTF, the flask was cooled in ice water for 30 minutes and 238 mg of isopropyl-β-D-thiogalactopyranoside (IPTG) was added to induce protein expression. In the case of pET28a, 238 mg of IPTG was added without cooling to start induction of expression. For these flasks, the shaking culture was continued at 37°C for 4 hours in the case of pET28a and at 16°C overnight in the case of pColdI and pColdTF. The bacterial cells were centrifuged at 2500×g for 15 minutes and stored at -80°C until use.

[0108] Purification of recombinant proteins

[0109] The recovered E. coli cells were suspended in HisTrap buffer A (20 mM sodium phosphate, pH 7.5, 800 mM sodium chloride, 20 mM imidazole, 1 mM dithiothreitol (DTT) and 10% (v / v) glycerol) containing a protease inhibitor reaction mixture (NakalaiTesque) and disrupted using an ultrasonic disruptor. The suspension was centrifuged at 15,000 × g for 15 minutes, and the supernatant was filtered using a 0.45 μm filter syringe to prepare a clarified lysate. The first stage of purification based on column chromatography was carried out using HisTrap buffer A and HisTrap buffer B (20 mM sodium phosphate, pH 7.5, 800 mM sodium chloride, 200 mM imidazole, 1 mM dithiothreitol (DTT) and 10% (v / v) glycerol)) and a 5 mL HisTrap HP column (Cytiva, Marlborough, MA). Next, the fraction containing the target protein was diluted 10 times with heparin buffer A (50mM sodium phosphate, pH7.5, 100mM sodium chloride), and further classified using heparin buffer A and heparin buffer B (50mM sodium phosphate, pH7.5, 1M sodium chloride) and 5mL of HiTrap Heparin HP (Cytiva). Then, after collecting the fraction containing the target protein, buffer exchange was performed using a preservation buffer (50mM Tris-HCl, pH8.0, 300mM sodium chloride) and a HiPrep 26 / 10Desalting column. These chromatography-based purifications were performed using AKTAstart (Cytiva). In AKTAstart, HisTrap, HiTrap Heparin and HiPrep Desalting were purified using the default procedures of affinity chromatography, ion exchange chromatography and desalting chromatography. Finally, the purified protein was concentrated by ultrafiltration using an Amicon Ultra-4 30K device (Millipore). The concentration of VTopoI was adjusted to 200 μM after concentration and stored at 4° C. The molar concentration of VTopoI was determined by measuring the absorbance of each protein at 280 nm using a molar absorptivity of 280 nM calculated using CLC Main workbench (Qiagen, Hilden, Germany).

[0110] In the case of T4 PNK, the protein concentration was adjusted to 2 mg / ml (quantified by the Bradford method using BSA as a standard), and then an equal volume of glycerol was added and stored at -20°C.

[0111] Formation and purification of VTopoI-ODN complex

[0112] The information of the oligonucleotides (ODN) used in this example is summarized in Table 1.

[0113] [Table 1]

[0114]

[0115] These ODNs are delivered by Eurofin Genomics as OPC grade or HPLC grade purified standards. Prepare 1 mL of a reaction solution containing 10 mM Tris-HCl, pH 8.0, 160 mM sodium chloride, 10 μM TOPO-adaptor-F, 10 μM TOPO-adaptor-R (for TA or blunt end), and 10 μM TOPO-adaptor-Ext in a 1.5 ml test tube. Incubate the reaction solution at 95°C for 3 minutes, 55°C for 5 minutes, and then maintain it at 37°C. Next, add the 1 ml of the oligonucleotide reaction solution to a 15 mL centrifuge tube, and add 1 ml of 5× Topo activation buffer (10 mM Tris-Acetate, pH 8.0, 250 mM potassium acetate and 50 mM magnesium acetate), 100 μL of 100 mM ATP, and 100 μL of 1 mg / ml T4 PNK to adjust the total volume to 5 mL. After incubation at 37°C for 1 hour, the reaction solution was cooled to 30°C, and then 250 μL of 200 μM VTopoI was added. The reaction solution was incubated at 30°C for 30 minutes. After that, the reaction solution was loaded onto 1 ml of HiTrap HeparinHP and separated using heparin buffers A and B. Each fraction was analyzed by both SDS-PAGE and agarose gel electrophoresis. In SDS-PAGE, Mini-PROTEAN anyKd gel (Bio-rad laboratories) was used for separation and BulletCBB staining solution (Nakalai Tesque) was used for staining. In agarose gel electrophoresis, 1 μL of 10% (w / v) SDS solution was added to 10 μL of each fraction, and 1 μl was loaded onto E-Gel Ex (2%) gel for analysis. Two main peaks were confirmed in the fractions in the heparin column, and the fraction containing the VTopoI-ODN complex eluted earlier than VTopoI itself. The fraction containing the VTopoI-ODN complex was collected, and the concentration was measured using the Qubit dsDNAHS kit while adjusting the concentration to 150 ng / μL using ultrafiltration. The concentration was concentrated using the 30K device of Amicon Ultra-4 as needed and kept at 4°C until use.

[0116] Isolation of cell-free DNA from human serum

[0117] Mixed human serum was purchased from KOHJIN BIO. 100 μL of human serum, 400 μL of 10 mM Tris-HCl, pH 8.0, 1.5 μL of 5 M sodium chloride, 10 μL of 10% (w / v) SDS, and 20 μL of proteinase K (QIAGEN) were added to an Eppendorf tube and incubated at 50°C for 1 hour. After protease digestion, 500 μL of phenol-chloroform-isoamyl alcohol (25:24:1) was added and stirred thoroughly, and centrifuged at 15,000 × g for 5 minutes. 500 μL of the supernatant after centrifugation was taken to a new tube, 50 μL of 3 M sodium acetate (pH 5.4) and 1,000 μL of isopropanol were added, mixed, and centrifuged at 15,000 × g for 5 minutes. The supernatant was removed by decantation, and the DNA precipitate was washed with 70% (v / v) ethanol and dissolved in an appropriate volume of 10 mM Tris-HCl, pH 8.0. The volume of serum used was changed according to the purpose of the experiment. Even in this case, the mixing ratio of the reagents used was kept constant.

[0118] TACS-T4 method

[0119] Library preparation from cfDNA using the TACS-T4 method was performed as described in reference 6.

[0120] TACS-TOPO method

[0121] i) TACS connection

[0122] For the solution containing the target ssDNA, add 2.5 μL of 10×TACS buffer (500 mM HEPES-KOH, pH 7.5, 50 mM MgCl2, 5% (v / v) Triton X-100), 1 μL of rSAP (New England Biolab, Ipswich, MA), and adjust the volume to 11 μL with water. Incubate the reaction solution at 37°C for 15 minutes and at 95°C for 5 minutes. Then, add it to 10 μL of 50% (w / v) PEG6000, and add 1 μL of 10 mM ATP, 25 μM adapter (Table 1), TdT (Takara Bio) and 2 mg / ml TS2126 RNA ligase. The preparation of TS2126 RNA ligase is carried out according to the description of reference 6. However, TS2126 RNA ligase can also be replaced by CircLigase II ssDNA ligase supplied by Lucigen. The reaction solution was incubated at 37°C for 15 minutes, at 65°C for 30 minutes, and at 95°C for 5 minutes.

[0123] ii) Replication of ssDNA ligated with adaptors

[0124] To the adapter-ligated ssDNA solution, 14 μL of water, 5 μL of 10× PCR buffer (Nippon Gene), 4 μL of 2.5 mM dNTPs, 1 μL of P-PEA2-2T (Table 1), and 1 μL of 5 unit / μL Hotstart Gene Taq (Nippon Gene) were added. Then, the mixture was incubated at 95°C for 3 minutes, 45°C for 5 minutes, 55°C for 5 minutes, and 65°C for 5 minutes.

[0125] iii) Second adapter ligation using VOC

[0126] To the reaction solution after complementary chain synthesis was completed, 5 μL of 500 mM EDTA (Nakalai Tesque) and 1 μL of VOC (for blunt ends or TA) were added, and the reaction was carried out at 25° C. for 15 minutes.

[0127] iv) Purification of the library

[0128] After the second adapter ligation, 45 μL of Buffer B2 (3 M guanidine hydrochloride, 20% (v / v) Tween20) and 5 μL of proteinase K (Qiagen) were added to the reaction solution and incubated at 50°C for 15 minutes. Next, 2 μL of Sera-Mag Carboxylate Magnetic Beads (Cytiva, Marlborough, MA) and 100 μL of isopropanol were added and mixed. After incubation at room temperature for 5 minutes, the magnetic body was collected on a magnetic stand and the supernatant was removed. Next, the magnetic beads were washed twice with 200 μL of magnetic bead washing solution (32% (v / v) PEG400, 1 M sodium chloride, 50 mM Tris-HCl, pH 8.0). After the magnetic beads were rinsed with 70% (v / v) ethanol, the purified DNA was eluted with 25 μL of 10 mM Tris-Acetate, pH 8.0.

[0129] v) UDG-APE1 treatment, nick repair and PCR amplification

[0130] To the purified library, add 5 μL of 10×ExTaq Buffer, 4 μL of 2.5 mM dNTPs, 1 μL of 20 μM primers, 1 μL of UDG (New England Biolabs), and 1 μL of APE1 (New England Biolabs), and adjust to 50 μL with water. However, UDG is only used in the version 4 protocol. (See "3. Results" below). After incubating the reaction solution at 37°C for 15 minutes, incubate at 72°C for 5 minutes, and incubate at 95°C for 1 minute. Next, repeat the three steps of incubation at 95°C for 15 seconds, 55°C for 30 seconds, and 72°C for 1 minute as many times as required. The number of cycles is displayed together with the data. Purify the DNA after the reaction using 90 μL of AxyPrep Mag PCR clean as needed. At this time, the final DNA eluent uses 21 μL of 10 mM Tris-Acetate, pH 8.0.

[0131] DNA from ancient human bones

[0132] The DNA derived from ancient humans used in this example was provided by Mr. Hideaki Kanzawa and Mr. Kenichi Shinoda of the National Museum of Nature and Science.

[0133] Library preparation using commercially available kits

[0134] In the preparation of a library for sequence determination targeting double-stranded DNA based on previous methods, the ThruPlex DNA Seq kit (Takara Bio) was used. In short, after performing end repair of the DNA and connection of the adapter, 50 μL of PCR reaction solution was prepared according to the manual. The reaction tube was incubated at 72°C for 3 minutes, at 84°C for 2 minutes, and at 98°C for 2 minutes. Next, a three-step cycle of 98°C for 20 seconds, 67°C for 20 seconds, and 72°C for 40 seconds was performed for 4 cycles, and a two-step cycle of 98°C for 20 seconds and 72°C for 50 seconds was performed for 8 cycles. 50 μL of AxyPrep MAG PCR was added to the reaction solution after the reaction and incubated at room temperature for 5 minutes. After rinsing the magnetic beads with 70% ethanol, the purified DNA was eluted with 25 μL of 10 mM Tris-Acetate, pH 8.0. The purified library was stored at -20°C until use.

[0135] Library quantification and sequence determination

[0136] The molar concentration of the sequencing library was determined using the Library Quantification Kit from Takara Bio. Small-scale sequencing was performed using the MiSeq and MiSeq Rreagent Kit v3 (150 cycles) from Illumina. Large-scale sequencing was performed by Macrogen using the HiSeq X Ten from Illumina.

[0137] Sequence analysis of cfDNA

[0138] Analysis of cfDNA was performed based on the description in reference 6.

[0139] ·Sequence analysis of ancient human DNA

[0140] First, the obtained reads were trimmed using fastp [reference 20] and the double-end sequencing reads were connected. Then, the merged reads were mapped to the human reference genome sequence using the mem option of bwa ​​[reference 21]. The final output sam format file was parsed using mapdamage [references 22, 23]. The details of these parsing are shown below.

[0141] For the sequence data, fastp was used to trim and merge the reads using the options described below, and the merged reads were mapped to the human reference genome hgl9 using bwa with the "mem" option. Finally, mapdamage was used for analysis.

[0142] Libraries prepared using ThruPlex:

[0143] --adapter_sequence=AGATCGGAAGAGCACACGTCTGAACTCCAGTCAC (sequence number 29)

[0144] --adapter_sequence_r2=AGATCGGAAGAGCGTCGTGTAGGGAAAGAGTGT (sequence number 30)

[0145] --merge

[0146] Libraries prepared using TACS-TOPO:

[0147] -f2

[0148] -F4

[0149] --adapter_sequence=AAGGGGGATCGGAAGAGCACACGTCTGAACTCCAGTCAC (sequence number 31)

[0150] --adapter_sequence_r2=AAAGATCGGAAGAGCGTCGTGTAGGGAAAGAGTGT (sequence number 32)

[0151] --merge3.result

[0152] Purification of active VTopoI

[0153] The gene of VTopoI was synthesized using artificial gene synthesis technology and then inserted into three expression vectors. In order to express VTopoI in bacterial cells, pColdTF, pColdI and pET28a were used. The results of inducing protein expression confirmed that each vector can stably induce ( Figure 7 C). Subsequently, purification using nickel resin was performed using a HisTrap HP column ( Figure 7 B). The protein eluted from the nickel column was further purified using heparin resin ( Figure 7 C). It is estimated that the purity of the purified protein is above 95% in all expression vector systems ( Figure 7 The activity of the purified proteins was confirmed by supercoil unwinding assay [reference 7]. Activity was confirmed in all purified recombinant proteins ( Figure 7 E). Therefore, the present inventors concluded that VTopoI was successfully purified. As for pET28a, since its protein yield was found to be lower than that of other vectors, VTopoI produced using pColdI and pColdTF was used in subsequent experiments.

[0154] ·Formation and purification of VTopoI-ODN complex

[0155] Studies have shown that VTopoI specifically recognizes the 5-base pair sequence 5'-(C / T)CCTT-3' on double-stranded DNA, cuts the 3' end of the sequence on the substrate DNA, and forms a covalent bond with the 3' phosphate [References 9, 17]. The supercoil is relaxed due to the single-stranded cutting. After the supercoil of the substrate DNA is relaxed, the reverse reaction occurs, and the cut chain reconnects. However, if the substrate DNA has a nick or gap in the complementary chain of the cutting site, the DNA on the downstream side of the recognition sequence is released from the intermediate complex of the substrate DNA and VTopoI, becoming an activated VTopoI-DNA complex ( Figure 1A, reaction 1). If there is a DNA with suitable ends near the complex, VTopoI undergoes a reverse reaction and ligates to the DNA [Reference 17]. This reaction is fast and efficient, and is the basis for using Topo for ligation reactions. The reaction between the substrate DNA containing 5'-(C / T)CCTT-3' and VTopoI is bidirectional, and the ratio of free VTopoI to the VTopoI-DNA complex is determined by their molecular concentrations ( Figure 1 A, reaction 1). In order to increase the amount of intermediates formed by covalent bonding between VTopoI and DNA, it is effective to inhibit the reaction in the reverse direction ( Figure 1 A, reaction 2). To achieve this, it is effective to phosphorylate the 5' end of the DNA released when the complex is formed. Figure 1 A, reaction 2). This is because: once the 5' end of the released DNA is phosphorylated, it cannot participate in the reverse reaction. As expected, the amount of the intermediate (VOC) formed by the covalent bond between VTopoI and ODN was increased by supplying ATP and T4 PNK to the reaction solution ( Figure 1 A~ Figure 1 C). The composition of the reaction solution also affects the amount of VOC formed. It was found that the presence of acetate ions is more effective than chloride ions for complex formation, and the mechanism is not clear ( Figure 1 B and Figure 1 C). In addition, the inventors found that VTopoI produced using pColdI was more likely to form precipitation when VOCs were formed (not disclosed in the data). Therefore, it was decided to use VTopoI produced using pColdTF for subsequent development. Based on the above results and other studies, more effective conditions for VOC formation were selected.

[0156] After VOC is formed, the reaction solution contains unreacted ODN, VTopoI and T4 PNK. These molecules become the cause of side reactions in downstream operations, so they need to be removed in advance. It was also found that purification using a heparin column is effective ( Figure 1 D). Unreacted ODN is negatively charged and therefore cannot bind to the negatively charged heparin column. It was found that VOC can bind to the heparin column through VTopoI, but the elution salt concentration will be lower than that of VTopoI itself or T4 PNK ( Figure 1 D). It was thus found that separation of VOC from other unreacted components can be achieved by only one column chromatography using a heparin column. After the purification, VOC can be easily concentrated by ultrafiltration. The purified VOC exhibits sufficiently high ligation activity with a model double-stranded DNA (see below), and the activity does not disappear even when stored at 4°C for several months.

[0157] VOC efficiently connects adapters to double-stranded DNA in a non-phosphorylated state at the 5' end

[0158] The inventors then studied the specificity and efficiency of VOC-based ligation. Here, a double-stranded ODN was prepared as a model substrate (TOPO-substrate) for the experiment. The TOPO-substrate accepts VOC-based ligation at one end (active end), and at the other end, the reaction is blocked by 5'-FAM and 3'-phosphorylation (inactive end) ( Figure 2 A). In order to study the stringency of the VOC-based ligation reaction for the terminal base, a TOPO-substrate with a blunt end and A, C, G, and T protruding one base to the 3' side at the active end was prepared. In addition, in order to study the specificity of VOC for the phosphorylation state of the substrate, a TOPO-substrate with a hydroxyl state and a phosphorylated state at the 5' side of the active end was prepared. Regarding VOC, two types of VOC were prepared: a blunt end type and a TA ligation type ( Figure 2 The present inventors studied the connection efficiency of all the above TOPO-substrate and VOC combinations ( Figure 2 B).

[0159] In both blunt-end and TA-type VOCs, high binding activity with TOPO-substrate with appropriate terminal structures was confirmed. In the presence of sufficient amounts of VOC, the reaction was almost completely completed in just a few minutes ( Figure 2 VOC showed strong activity towards the 5' hydroxyl terminus, but never reacted with the 5' phosphorylated terminus ( Figure 2 B). Although the efficiency of linker ligation on TOPO-substrate with suitable ends is good, the stringency of base matching for the terminal structure is not as high as that for phosphorylated ends. TA-type VOC shows ligation activity for TOPO-substrate with protruding T, C and G, and the blunt-end type reacts with TOPO-substrate with all protruding ends. Studies have shown that the stringency of the suitable terminal structure can be enhanced by adding sodium chloride [reference 24], but increasing the stringency will lead to a decrease in ligation efficiency (not disclosed in the data).

[0160] The specificity of VOC for substrates in the 5' non-phosphorylated state is ideal for the second adaptor ligation after TACS ligation (see the summary of item 1 above). Therefore, the present inventors have been developing a new library preparation protocol using VOC.

[0161] TACS-TOPO method can inhibit the formation of adaptor dimers

[0162] VOC showed efficient adaptor ligation to 5' non-phosphorylated ends and did not react with 5' phosphorylated ends ( Figure 2 C), it is believed that VOC can achieve effective second adapter ligation without causing the formation of adapter dimers. When 83ng (5pmol) of random 50 bases (N50) were used as a model, the yield of the final target product connected with two adapters was confirmed in the TACS-T4 method ( Figure 3 A, lane 4, green star), but a significant improvement in library yield was confirmed in the TACS-TOPO method ( Figure 3 In addition, when analyzing the library after PCR amplification, a large number of adapter dimers were included in the TACS-T4 method, but only library molecules of the target size containing the insert fragment were amplified in the TACS-TOPO method ( Figure 3 These results indicate that the TACS-TOPO method is superior to the TACS-T4 method in terms of high library yield and low adapter dimer formation.

[0163] Optimization of the TACS-TOPO method for preparing dimer-free libraries with high sensitivity

[0164] Next, the inventors studied the sensitivity of the TACS-TOPO method for preparing libraries without adapter dimers. Here, N50 was used as a model again to study the appearance of adapter dimers. Figure 4 As shown in B, in the TACS-TOPO method, it was observed that a non-negligible amount of adapter dimers were formed when the input DNA amount was less than 24 ng. One of the reasons for this phenomenon may be that there are very few 5'-end dephosphorylated molecules contained in the adapter ODN. As a result of revealing this phenomenon, it was found that the amount of adapter dimer formation decreased accordingly when the purification level of the adapter ODN was improved (version 2, Figure 4 C and Figure 8 A~ Figure 8 D). Unfortunately, attempts to inhibit the formation of adapter dimers by further purifying the adapter ODN failed (not disclosed in the data), so it was decided to inhibit the formation of adapter dimers by combining two other experimental principles. One of the schemes was to reduce the molecular size of the adapter itself in order to facilitate the removal of adapter dimers (version 3, Figure 8 A~ Figure 8 Another approach is to insert uracil near the 5' end of the adapter ODN in order to dissociate the adapter dimer before PCR amplification (version 4, Figure 8 A. Figure 8 B and Figure 8 F). By combining the TACS-TOPO method with these two strategies, a library with a small amount of dimer formation can be prepared even from 24 pg of DNA equivalent to 4 cells when converted to human cells. That is, using TACS-TOPO version 4, a library without adapter dimers can be prepared from a very small amount of single-stranded DNA.

[0165] ·Library preparation from short single-stranded free DNA contained in peripheral blood

[0166] Previously, the present inventors discovered that human peripheral blood contains about 50 bases of single-stranded free DNA, which they called C3D [Reference 6]. The present inventors used the TACS-T4 method to analyze cfDNA and found that C3D is a new type of cfDNA. However, in the TACS-T4 method, adapter dimers are preferentially and massively amplified, and in order to remove them, labor-intensive SPRI purification needs to be repeated. The optimized TACS-TOPO method is effective for dimer-free library preparation from a limited amount of single-stranded DNA ( Figure 4 Therefore, the inventors applied TACS-TOPO version 4 to prepare libraries from cfDNA. Figure 5 As shown in A, the preparation of sequencing libraries without adapter dimers can be achieved using TACS-TOPO version 4. On the other hand, despite the SPRI-based purification step in TACS-T4, the amount of adapter dimers still dominates. As shown previously [Reference 6], the library after PCR amplification contains two large peaks corresponding to nucleosome-sized cfDNA (160-base-long insert fragment) and C3D, so TACS-TOPO version 4 is useful for analyzing cfDNA. The cfDNA purified from 50 μL of serum can prepare a sufficient amount of library without PCR-based amplification of the entire library. Even from 6.25 μL of serum, the required amount of library can be obtained by 3 cycles (maximum 8 times) of PCR amplification ( Figure 5 B and Figure 5 From the sequence determination of the resulting library and the analysis of its reads, it was possible to identify the features of C3D that we have previously shown ( Figure 5 D and Figure 5 E) [Reference 6]. These results indicate that the TACS-TOPO method is practical for analyzing cfDNA.

[0167] TACS-TOPO method enables highly sensitive library preparation from DNA derived from ancient organisms

[0168] The TACS-TOPO method was applied to library preparation from ancient samples. DNA from ancient samples often decomposes, and fragmentation and single-stranding can be observed. Therefore, the library preparation method from such samples should be able to correspond to single-stranded DNA. In fact, Gansauge et al. have confirmed that the library preparation method corresponding to single-stranded DNA is effective for the analysis of ancient DNA [References 2, 3]. The TACS-T4 method was applied to cfDNA, which brought about a significant improvement in library yield compared to the method established by Gansauge et al. [Reference 6], and the TACS-TOPO method showed better performance than the TACS-T4 method in the analysis of cfDNA ( Figure 5 A). Therefore, it is believed that the TACS-TOPO method is more suitable for library preparation from ancient samples.

[0169] The inventors decided to use DNA extracted from human bones unearthed from the Inome ruins and Shichigosangake ruins for this test. Each of the ruins is from the late Jomon period, so the obtained DNA is believed to have originated from humans 2,000 to 4,000 years ago, and it is of course fragmented and contains many damages. The yield of library preparation from these ancient samples using TACS-TOPO version 4 is extremely good compared to the ThruPlex DNA-Seq kit, which is one of the main commercially available library kits and is suitable for library preparation from double-stranded DNA. That is, the TACS-TOPO method has achieved a 1 to 2 order of magnitude improvement in library yield ( Figure 6 On the other hand, the size distribution of libraries prepared by ThruPlex and TACS-TOPO also showed obvious differences. The latter library always showed a shorter distribution than the former ( Figure 6 B and Figure 6 In addition, the mapping rate to the human reference genome sequence was roughly the same between the reads obtained from the libraries prepared by the ThruPlex method and the TACS-TOPO method ( Figure 6 D).

[0170] The yield of sequencing libraries derived from ancient human bones was significantly improved using TACS-TOPO version 4. However, although base substitutions from C to T or G to A should be observed at both ends of the reads, the reads mapped to the human reference genome sequence lacked the characteristic mismatch distribution pattern of this ancient DNA at one end ( Figure 6 This is considered to be because: in TACS-TOPO version 4, UDG-APE1-based treatment was introduced to achieve adapter dimer decomposition, but this enzyme treatment caused uracil (U) decomposition due to deamination of C.

[0171] In order to avoid UDG-APE1-based treatment, the inventors tried to use modifications different from uracil on the adaptors used in TACS ligation. After studying several modifications, it was found that dSpacer (dSp), an AP site analog, can be used for this purpose. It is known that dSp can be decomposed by APE1, so the inventors initially added an APE1-based treatment before PCR. However, it was found that, although the reason is unclear, APE1-based treatment is not required in the actual process, and the simplicity is superior. The inventors used this modified protocol (TACS-TOPO version5, Figure 8 A and Figure 8 The result is that for reads obtained from libraries prepared using adapters containing dSp, the base substitutions ( Figure 6 Although the library yield of the library prepared using the dSp adapter is slightly lower than that of the uracil library ( Figure 6 A), but the version 5 protocol showed approximately the same library size distribution as the version 4 protocol ( Figure 6 B~ Figure 6 D). These results suggest that TACS-TOPO version 5 can be an extremely useful tool in palaeobiochemistry.

[0172] The following lists the references disclosed in this example.

[0173] 1. Meyer, M., Kircher, M., Gansauge, MT, Li, H., Racimo, F., Mallick, S., Schraiber, JG, Jay, F., Prufer, K., de Filippo, C. et al. (2012) Ahigh-coveragegenome sequence from an archaic Denisovan individual. Science, 338, 222-226.

[0174] 2. Gansauge, MT and Meyer, M. (2013) Single-stranded DNA library preparation for the sequencing of ancient or damaged DNA. Nat. Protoc., 8, 737-748.

[0175] 3.Gansauge,M.T.,Gerber,T.,Glocke,I.,Korlevic,P.,Lippik,L.,Nagel,S.,Riehl,L.M.,Schmidt,A.and Meyer,M.(2017)Single-stranded DNA librarypreparation from highly degraded DNA using T4 DNA ligase.Nucleic Acids Res.,45,e79.

[0176] 4.Miura,F.,Shibata,Y.,Miura,M.,Sangatsuda,Y.,Hisano,O.,Araki,H.andIto,T.(2019)Highly efficient single-stranded DNAligation technique improveslow-input whole-genome bisulfite sequencing bypost-bisulfite adaptortagging.Nucleic Acids Res.,47,e85.

[0177] 5.Schmidt,W.M.and Mueller,M.W.(1996)Controlled ribonucleotide tailingof cDNAends(CRTC)by terminal deoxynucleotidyl transferase:a new approach inPCR-mediated analysis of mRNAsequences.Nucleic Acids Res.,24,1789-1791.

[0178] 6.Hisano,O.,Ito,T.and Miura,F.(2021)Short single-stranded DNAs withputative non-canonical structures comprise a new class of plasma cell-freeDNA.BMC Biol.,19,225.

[0179] 7.Shuman,S.and Moss,B.(1987)Identification of a vaccinia virus geneencoding a type I DNA topoisomerase.Proceedings of the National Academy ofSciences,84,7478-7482.

[0180] 8.Shuman,S.,Golder,M.and Moss,B.(1988)Characterization of vacciniavirus DNA topoisomerase I expressed in Escherichia coli.J.Biol.Chem.,263,16401-16407.

[0181] 9.Shuman,S.and Prescott,J.(1990)Specific DNA cleavage and binding byvaccinia virus DNA topoisomerase I.J.Biol.Chem.,265,17826-17836.

[0182] 10.Shuman,S.(1991)Recombination mediated by vaccinia virus DNAtopoisomerase I in Escherichia coli is sequence specific.Proc.Natl.Acad.Sci.U.S.A.,88,10104-10108.

[0183] 11.Shuman,S.(1991)Site-specific interaction of vaccinia virustopoisomerase I with duplex DNA.Minimal DNA substrate for strand cleavage invitro.J.Biol.Chem.,266,11372-11379.

[0184] 12.Shuman,S.(1991)Site-specific DNA cleavage by vaccinia virus DNAtopoisomerase I.Role of nucleotide sequence and DNA secondarystructure.J.Biol.Chem.,266,1796-1803.

[0185] 13.Shuman,S.(1992)Two classes of DNA end-joining reactions catalyzedby vaccinia topoisomerase I.J.Biol.Chem.,267,16755-16758.

[0186] 14.Morham,S.G.and Shuman,S.(1992)Covalent and noncovalent DNA bindingby mutants of vaccinia DNA topoisomerase I.J.Biol.Chem.,267,15984-15992.

[0187] 15.Shuman,S.(1992)DNAstrand transfer reactions catalyzed by vacciniatopoisomerase I.J.Biol.Chem.,267,8620-8627.

[0188] 16.Stivers,J.T.,Shuman,S.and Mildvan,A.S.(1994)Vaccinia DNAtopoisomerase I:kinetic evidence for general acid-base catalysis and aconformational step.Biochemistry,33,15449-15458.

[0189] 17.Shuman,S.(1994)Novel approach to molecular cloning andpolynucleotide synthesis using vaccinia DNA topoisomerase.J.Biol.Chem.,269,32678-32684.

[0190] 18.Bauer,W.R.,Ressner,E.C.,Kates,J.and Patzke,J.V.(1977)A DNAnicking-closing enzyme encapsidated in vaccinia virus:partial purificationand properties.Proc.Natl.Acad.Sci.U.S.A.,74,1841-1845.

[0191] 19.Yarovinsky,T.O.(2000)Application of DNAtopoisomerase-activatedadapters to riboprobe synthesis.Biotechniques,28,1160,1162-1165.

[0192] 20.Chen,S.,Zhou,Y.,Chen,Y.and Gu,J.(2018)fastp:an ultra-fast all-in-one FASTQ preprocessor.Bioinformatics,34,i884-i890.

[0193] 21.Li,H.and Durbin,R.(2009)Fast and accurate short read alignmentwith Burrows-Wheeler transform.Bioinformatics,25,1754-1760.

[0194] 22.Ginolhac,A.,Rasmussen,M.,Gilbert,M.T.P.,Willerslev,E.and Orlando,L.(2011)mapDamage:testing for damage patterns in ancient DNAsequences.Bioinformatics,27,2153-2155.

[0195] 23. Jonsson, H., Ginolhac, A., Schubert, M., Johnson, PLF and Orlando, L. (2013) mapDamage2.0: fast approximate Bayesian estimates of ancient DNA damage parameters. Bioinformatics, 29, 1682-1684.

[0196] 24. Cheng, C. and Shuman, S. (2000) DNA strand transfer catalyzed byvaccinia topoisomerase: ligation of DNAs containing a 3'mononucleotide overhang. Nucleic Acids Res., 28, 1893-1898.

[0197] Industrial Applicability

[0198] According to the present invention, a method for preparing a double-stranded DNA library from single-stranded DNA fragments can be provided, which can extremely effectively increase the library yield compared to previous double-stranded DNA library preparation methods and can effectively inhibit the by-production of double-stranded linker dimers during preparation.

Claims

1. A method for preparing a double-stranded DNA library from single-stranded DNA fragments, comprising the following steps (i) to (v): Step (i), connecting the single-stranded adapter sequence A1 with phosphorylated 5' end to the 3' end of the single-stranded DNA fragment with dephosphorylated both ends to generate single-stranded DNA with added adapter; Step (ii), after annealing the single-stranded adapter sequence A2 with the adapter sequence A1 contained in the single-stranded DNA to which the adapter is added, using a DNA polymerase to extend the 3' end of the adapter sequence A2 to generate a double-stranded DNA to which the adapter is added; Step (iii), reacting a topoisomerase with a double-stranded adapter sequence B1 having only one 5' end phosphorylated and having a topoisomerase recognition sequence in a region including a 3' end complementary to the 5' end, to generate a topoisomerase composite double-stranded adapter sequence B2 in which the topoisomerase is conjugated to the 3' end of the recognition sequence; Step (iv), connecting the topoisomerase composite double-stranded adapter sequence B2 to the end of the double-stranded DNA to which the adapter is added without the adapter sequence added, to generate double-stranded DNA to which the adapter is added at both ends; as well as Step (v) is to obtain a library containing a plurality of double-stranded DNAs by performing PCR using the double-stranded DNA to which adapters are added at both ends as a template.

2. The preparation method according to claim 1, wherein The topoisomerase composite double-stranded adapter sequence B2 is not connected to the double-stranded adapter sequence C, and the double-stranded adapter sequence C is generated by annealing the unreacted single-stranded adapter sequence A1 with the unreacted single-stranded adapter sequence A2.

3. The preparation method according to claim 1, wherein The double-stranded adapter sequence B1 has a nick in the region including the phosphorylated 5′ end, and the single-stranded adapter sequence A1 has uracil or dSpacer in the sequence.

4. The preparation method according to claim 3, wherein The step (iv) comprises the step of adding an endonuclease to the generated reaction system.

5. The preparation method according to claim 1, wherein The connection and generation in step (i) are performed using the TACS connection method.

6. A double-stranded DNA library obtained by the preparation method according to any one of claims 1 to 5.

Images