FORMATION OF HAIRPINS IN SITU USING FORCE-INDUCED STRAND INVASION

Abstract

The present invention relates to a method of preparation of substrates for nucleic acid sequencing reactions. More specifically, the present invention provides a new method of preparing hairpins using force-induced strand invasion. Hairpins prepared by this method and methods of nucleic acid analysis using these hairpins are also part of the present invention.

Claims

1. A method for preparing a hairpin comprising a sequence of interest, said method comprising the steps of: a) providing a nucleic acid HP1, said nucleic acid comprising: a first end bound to a first surface; a single-stranded region of sequence A linked to said first end, a single-stranded region of sequence A hybridised to said single-stranded region A, wherein said single-stranded regions A and A are not covalently linked; at least one single-stranded region of sequence C linked to said single-stranded region of sequence A; a second end linked to said single-stranded region of sequence C, wherein said second end is bound to a second surface; wherein one of the first and the second surfaces is a movable surface b) providing at least one nucleic acid HP2, said nucleic acid comprising: a double-stranded region comprising the sequence of interest, a loop linked to a first end of said double-stranded region, a single-stranded region having the sequence A linked to a first strand of the second end of said double-stranded region, said region of sequence A being linked to a single-stranded region of sequence C, C being complementary of C; a single-stranded polynucleotide having the sequence A linked to a second strand of the second end of said double-stranded polynucleotide molecule, wherein the single-stranded polynucleotides of sequences A and A are hybridised; c) denaturing said nucleic acid HP1 of step a) in the presence of said nucleic acid HP2 of step b) by moving away the surfaces by applying a tension of at least 3 pN; and d) obtaining a hairpin.

2. The method of claim 1, wherein: said single-stranded region of sequence A of said structure HP1 is linked to a single-stranded region of sequence D, said single-stranded region of sequence D being bound to said first surface; and said single-stranded region of sequence A of structure HP2 is linked to a single-stranded region having the sequence D, wherein D is complementary to D.

3. The method of claim 1, wherein the denaturation of step c) involves applying a tension of at least 4 pN.

4. The method of claim 1, wherein more than one molecule of HP1 is attached to at least one of the surfaces.

5. The method of claim 1, wherein a plurality of distinct nucleic acid molecules HP2 is used.

6. The method of claim 1, wherein sequences A and A each comprises at least 30 nucleotides.

7. The method of claim 1, wherein sequences C and C each comprises at least 10 nucleotides.

8. The method of claim 1, wherein sequences D and D each comprises at least 10 nucleotides.

9-10. (canceled)

11. The method of claim 1, further comprising: a) denaturing the hairpin by moving away the surfaces; b) measuring the distance Z.sub.high between the two ends of the denatured hairpin molecule obtained in step a); c) hybridizing a single-stranded nucleic acid molecule, the primer, with said denatured hairpin molecule obtained in step a); d) renaturing said hybridised single-stranded nucleic acid molecule/hairpin molecule of step d); and e) detecting a blockage of the renaturation of the hybridised single-stranded nucleic acid molecule/hairpin molecule; and f) determining the position of said blockage with respect to one end of the hairpin, said determination comprising the steps of: measuring distance (z) between the two ends of the hairpin molecule which are attached to a support, comparing z and Z.sub.high, and determining the position of the blockage; wherein said nucleic acid sequence is derived from the position of said blockage.

12. The method of claim 1, further comprising: a) denaturing the hairpin by moving away the surfaces; b) measuring the distance Z.sub.high between the two ends of the denatured hairpin molecule obtained in step a); c) hybridising a single-stranded nucleic acid molecule, the primer, with said denatured hairpin molecule obtained in step a); d) renaturing said hybridised single-stranded nucleic acid molecule/hairpin molecule of step d); and e) detecting a blockage of the renaturation of the hybridised single-stranded nucleic acid molecule/hairpin molecule; and f) determining the position of said blockage with respect to one end of the hairpin, said determination comprising the steps of: measuring distance (z) between the two ends of the hairpin molecule which are attached to a support, comparing z and Z.sub.high, and determining the position of the blockage; g) determining the duration of said blockage, wherein the presence of said nucleic acid sequence is derived from the position of said blockage and the duration of said blockage.

13. The method of claim 1, further comprising: a) denaturing the hairpin, said hairpin potentially comprising said sequence, by moving away the surfaces; b) measuring the distance Z.sub.high between the two ends of the denatured hairpin molecule obtained in step a); c) contacting said protein with said denatured hairpin molecule obtained in step a); d) renaturing said hairpin molecule of step d) in the presence of said protein; and e) detecting a blockage of the renaturation of the hairpin molecule; and f) determining the position of said blockage with respect to one end of the hairpin, said determination comprising the steps of: measuring distance (z) between the two ends of the hairpin molecule which are attached to a support, comparing z and Z.sub.high, and determining the position of the blockage; g) determining the duration of said blockage, wherein the binding of said single-stranded nucleic acid-binding protein to said nucleic acid sequence is derived from the position of said blockage and the duration of said blockage.

14. The method of claim 1, further comprising: a) denaturing the hairpin, said hairpin potentially comprising said sequence, by moving away the surfaces; b) measuring the distance Z.sub.high between the two ends of the denatured hairpin molecule obtained in step a); c) hybridizing a single-stranded nucleic acid molecule with said denatured hairpin molecule obtained in step a); d) contacting said protein with said hybridised single-stranded nucleic acid molecule/hairpin molecule of step c); e) renaturing said hybridised single-stranded nucleic acid molecule/hairpin molecule of step c) in the presence of said protein; and f) detecting a blockage of the renaturation of the hairpin molecule; and g) determining the position of said blockage with respect to one end of the hairpin, said determination comprising the steps of: measuring distance (z) between the two ends of the hairpin molecule which are attached to a support, comparing z and Z.sub.high, and determining the position of the blockage; h) determining the duration of said blockage, wherein the binding of said double-stranded nucleic acid-binding protein to said nucleic acid sequence is derived from the position of said blockage and the duration of said blockage.

15. The method of claim 1, wherein the denaturation of step c) involves applying a tension of at least 5 pN.

16. The method of claim 1, wherein the denaturation of step c) involves applying a tension of at least 6 pN

17. The method of claim 1, wherein sequences A and A each comprise at least 35 nucleotides.

18. The method of claim 1, wherein sequences C and C each comprise at least 12 nucleotides.

19. The method of claim 1, wherein sequences D and D each comprise at least 12 nucleotides.

20. The method of claim 1, wherein sequences D and D each comprise at least 13 nucleotides.

Description

LEGENDS OF THE FIGURES

[0168] FIG. 1: Structure of DNA hairpins used in magnetic tweezer analysis. A typical DNA hairpin structure is shown. The bold sequence represents the double stranded DNA region of interest, and the various DNA linkers required for functionality are described. Linker 1 is a small DNA loop that permanently attaches the 5end of one strand of the ROI to the 3 end of the other strand. The structure will readily bind to a streptavidin-coated bead, by virtue of the Biotin moiety (shown as a red dot) synthesised on the end of linker 2. Finally, linker 3 allows binding to the flow cell surface coated in anti-digoxigenin antibodies, through the interaction with the digoxigenin located at its end (green dot).

[0169] FIG. 2: Principle of the stand invasion process. On the far left a universal precursor hairpin-bead construct (HP1) is shown, which can be prepared in bulk and attached to the surface of the flow cell. It has a dsDNA linker attached to the bead, ending with a ssDNA overhang of sequence D-A. A second molecule consists of a dsDNA region with a digoxigenin-labelled tail allowing it to bind to the glass surface of the flow cell (dig represented by orange squares). The other end has a ssDNA region with a sequence C (of 12 nts) plus the 40 nt complementary sequence to A (A). These two molecules (bead+linker and dig labelled linker) can be pre-incubated so that they hybridise together via the A and A sequences, forming pre-hairpin structure HP1. This construct will be stable under normal conditions, but pulling on the magnetic bead with a force greater than a few pN will unzip the hybridised region and dissemble the structure. Next to this construct is displayed the structure called hairpin precursor HP2. It consists of the (double stranded) DNA to be studied (in grey) ligated to a loop at one end (shown on the right) and an adaptor at the other. The adaptor consists of 2 oligos hybridised together via A and A sequences (identical to those on the hairpin shown in the left panel). The adaptor also has overhanging single stranded regions of sequences D and C that are complementary to sequences C and D of HP1, respectively. When a library of these hairpins (HP2) is introduced to a flowcell containing a plurality of the structures shown in the left panel, they will hybridise through their flaps (C and D) each forming a Holiday junction with 2 nicks, called here HP3 (as shown in the middle right panel). When a small force (5 pN) is applied to the bead, the Holiday junction rapidly migrates, lengthening the molecule and leading to a stable hairpin construction (shown in the far right panel). The molecule shown in the right panel is essentially identical to that shown in FIG. 1; it can be zipped and unzipped as usual for such hairpin structures, and is suitable for all mapping and sequencing experiments. Note that although there are 3 single stranded nicks in this molecule, the length of complementary sequence (AD-AD and CA-CA) is long enough to resist the shearing force of the magnet (as opposed to the smaller length of complementary sequence found on the uninvaded Halliday structure (held together only by 12 bp D-D and C-C).

[0170] FIG. 3: Alternative strand invasion process. This example is similar to that shown in FIG. 1, with the exception that the invading hairpin has only a single flap (C) with which to bind to proto-hairpin structure affixed to the surface. The structure of the resulting strand-invaded hairpin is very similar to that in FIG. 2.

[0171] FIG. 4: Examples of fingerprints obtained with an oligonucleotide CGCCAC. A hairpin was generated with the 1.6 kb BsmBI fragment obtained from pPS002 digestion. The force on the bead was gradually increased until it reached a point where the molecule unziped. Reduction of the force caused reziping of the molecule. In the absence of any oligonucleotide, the closing was rapid (left panel). However, when the oligonucleotide was present in the flow cell and the complementary sequence of this oligonucleotide is on the hairpin, it blocked the reziping. This oligonucleotide had 3 binding positions on the hairpin, only one (at position 794 bp) is showing blockage due to the nature of the oligonucleotide. The experimental value obtained for this blockage on the particular bead was 784 bp. This oligonucleotide also blocked the reformation of the hairpin due to a blocking site located within the PS046 loop oligonucleotide, although there is a mismatch at the 5 end between the oligonucleotide and its target sequence.

[0172] FIG. 5: Detection of the 5-methylcytosine modification. The same hairpin created through FISI with the BsmBI digested fragment from pPS003 was tested against the 5-methylcytosine modification with antibodies (the clone 33D3 monoclonal antibody was used in this experiment and is commercially available from various sources such as Merck Millipore or Sigma-Aldrich). This hairpin was predicted to contain 2 potential Dcm methylation sites at position 170 and 1046. Using the ssDNA blockage as a reference (the first one from the top), the experimental blocking position were calculated to be at 135 bp and 1035 bp. Both the oligonucleotide and the antibody confirmed that the fragment of DNA was really originating from pPS003.

EXAMPLES

Example 1

Preparation of HP1 Containing One Flap

[0173] The pPS001 vector (SEQ ID NO. 1) was used to clone a 1.5 kb KpnI fragment or a 500 bp SalI fragment from lambda genomic DNA to yield pPS002 (SEQ ID NO. 2) or pPS003 (SEQ ID NO. 3), respectively.

[0174] To create the dsDNA linker between the bead and the precursor HP1, DreamTaq DNA polymerase was used according to the manufacturer specifications. The oligonucleotides PS079 (SEQ ID NO. 6) and PS080 (SEQ ID NO. 7) were used at 500 nM concentration each and depending on the desired length of the linker, either pPS001, pPS002 or pPS003 vectors was used as template (creating a linker of either 236, 1724 or 737 base pairs, respectively).

[0175] The oligonucleotide PS080 contains a biotin at its 5 end. The oligonucleotide PS079 has a 12-carbon spacer (C12 spacer) that prevents the DNA polymerase from copying the 5 end of the vector, and leaves a 5, single stranded tail.

[0176] The PCR conditions were as follow:

[00001]$\begin{array}{c}.Math.98.Math..Math..Math.C..Math.\mathrm{for}.Math..Math.3.Math..Math.\min \\ \begin{array}{c}58.Math..Math..Math.C..Math.\mathrm{for}.Math..Math.20.Math..Math.\sec \\ 72.Math..Math..Math.C..Math.\mathrm{for}.Math..Math.1.Math..Math.\min \\ 95.Math..Math..Math.C..Math.\mathrm{for}.Math..Math.30.Math..Math.\sec \end{array}\}.Math.30.Math.X\\ .Math.72.Math..Math..Math.C..Math.\mathrm{for}.Math..Math.5.Math..Math.\min \\ .Math.\mathrm{Hold}.Math..Math.\mathrm{at}.Math..Math.4.Math..Math..Math.C..Math.\mathrm{for}.Math..Math.\mathrm{ever}\end{array}$

[0177] For the adapter between the HP1 and the surface, DreamTaq DNA polymerase was used according to the manufacturer specification with the oligonucleotides PS103 (SEQ ID NO. 8) and PS104 (SEQ ID NO. 9) on the template sequence pPS003. The same conditions as previously were used.

[0178] The resulting sequence is as follow (including the sequence added to create Bsal restriction site in green):

TABLE-US-00001 SEQIDNO.5: 5 attcgatcgtGGTCTCAGAATcctggtggtgagcaatggtttcaac catgtaccggatgtgttctgccatgcgctcctgaaactcaaCatcgtca tcaaacgcacgggtaatggattttttgctggccccgtggcgttgcaaat gatcgatgcatagcgattcaaacaggtgctggggcaggccTttttccat gtcgtctgccagttctgcctctttctcttcacgggcgagctgctggtag tgacgcgcccagctctgagcctcaagacGCTTGGAGACCagctagccat 3

[0179] The resulting fragment was then digested with Bsal according to the manufacturer specification. On these overhangs, two different adapters are then cloned as follows: [0180] The first of these adapters is obtained by hybridizing oligonucleotides PS101 (SEQ ID NO. 10) and PS070 (SEQ ID NO. 11). The resulting 5 overhang of PS101 is complementary to one of the fragment overhangs. [0181] The second adapter is obtained by hybridizing oligonucleotides PS101 and PS0102 (SEQ ID NO. 12). Both ends of PS102 extend beyond PS101. One of these ends is complementary to the second overhangs of the fragment.

[0182] Specifically, the oligonucleotides PS101 and PS070 were annealed in CutSmart buffer (NEB, 1: 50 mM Potassium Acetate, 20 mM Tris-acetate, 10 mM Magnesium Acetate, 100 g/ml BSA, pH 7.9@25 C.) at a concentration of 10 m each. The solution was heated at 98 C. and let cool down to room temperature on the heat block. The same procedure was performed for the oligonucleotides PS101 and PS102.

[0183] Two different adapters were thus created, that can be ligated to the Bsal digested PCR fragment directionally due to the presence of 2 different, non-palindromic, sites using T4 DNA ligase from Enzymatics according to manufacturer specification.

[0184] Once ligated, the oligonucleotide PS101 was extended using PS070 as template using Klenow exo-DNA polymerase from Enzymatics. The amount of dTTP in the reaction mix was adjusted such that dig-dUTP could be used. There was a ratio of 40% dTTP for 60% of dig-dUTP.

[0185] After purification of the final fragment on agarose gel, the molarity of both fragments was calculated. The two fragments were then mixed together at equal molarity in CutSmart buffer. The 3 end of the oligonucleotide PS102 is complementary to the 5 end of the oligonucleotide PS079, after the C12 spacer.

[0186] Once annealed, the final HP1 molecule was serially diluted and bound on MyOne paramagnetic beads functionalized with streptavidin. Since the PS080 oligonucleotide contains a biotin, the molecule will bind to the beads. The precursor HP1 was then loaded in the microfluidic chamber, the floor of the cell being functionalized with anti-digoxigenin antibodies. Due to the presence of digoxigenin on the second part of HP1, the precursor HP1 binds to the floor of the flow cell.

Preparation of HP2

[0187] For the production of HP2, the BsmBI fragment from the vector pPS002 (SEQ ID NO. 2) was obtained through digestion of the vector and purified from the gel. The resulting 1.6 kb fragment had 2 different non-palindromic ends.

[0188] The oligonucleotides PS108 (SEQ ID NO. 13) and PS109 (SEQ ID NO. 14) were mixed at equal concentration (10 M) in CutSmart buffer and heated at 98 C. Then, the tube was slowly cooled down to room temperature to allow the 2 oligonucleotides to anneal, thus creating an adapter with the complementary overhang to the digested BsmBI fragment.

[0189] PS046 (SEQ ID NO. 15) is a self-annealing oligonucleotide, with a loop of 5 thymines and a 5 overhang enabling cloning to a BsmBI digested vector. PS046 was diluted to 10 M, heated at 98 C. and rapidly cooled down on ice, in order to promote the formation of a small hairpin-loop structure which can be ligated to the DNA region of interest (forming HP2).

[0190] Once ligated to the BsmBI fragment, the resulting HP2 was purified onto an agarose gel and eluted into 50 l of water. 1 l of this HP2 was mixed with 100 l of passivation buffer and loaded on the fluidic cell containing the HP1 attached to the surface. The magnet was brought close to the sample such that the force reached around 5 pN. The sample was left like this for 30 minutes and the force was gradually increased to 15 pN. If an ssDNA flap (C) from HP2 has hybridized to the complementary sequence C on HP1, the Holliday junction would be resolved. In case there is no HP2 attached to HP1, the bead would fly away. Once resolved, the hairpin can be interrogated with either oligonucleotides or any other binding molecules like antibodies or proteins.

Example 2

Preparation of HP1 Molecule Containing Two Flaps.

[0191] For this version, the hairpin of interest contains two ssDNA stretches that can bind on either side of the fork.

[0192] For the 2 flap strategy, the procedure is basically the same except that the oligonucleotides used are slightly different.

[0193] For the biotin-Space linker, there is no change. PS079 and PS080 were used on pPS003 vector

[0194] For the dig linker, PS102 was replaced by the oligonucleotide PS115 (SEQ ID NO. 16). The latter was then annealed with PS101 to create the adapter to be ligated with the Bsal digested PCR fragment obtained with PS103-PS104. The second adapter, PS101-PS070 was unchanged. The dig-tail was then synthesised as previously described.

[0195] Both fragments were purified on agarose gel and mixed at equal amount to form the HP1. Then, they were serially diluted and bound to MyOne paramagnetic beads coated with streptavidin.

[0196] For making the HP2, the PS107-PS108 adapter were replaced with the adapter composed of PS116 (SEQ ID NO. 17)-PS118 (SEQ ID NO. 18) to make the two flap HP2. They were mixed in CutSmart buffer at 10 M, heated at 98 C. and slowly cooled to room temperature. They were finally ligated as well as the loop PS046 to the BsmBI fragment from pPS002 vector.

[0197] The final fragment was purified on agarose gel and 1 l of the resulting eluate was loaded inside the flow cell containing the HP1 precursor. The same procedure as previously was applied.