METHOD FOR DETECTION OF RNA

Abstract

The present invention provides a method for detection of target RNA in a sample comprising the use of a hairpin probe which unfolds upon binding to its target RNA sequence to reveal a 3′ end region which is capable firstly of binding to a circular or circularisable probe and if necessary templating ligation of a circularisable probe to circularize the probe, and secondly of priming a subsequent RCA reaction of the circularized/circular probe. The circularized or circular probe is then subjected to RCA and a RCA product is detected, in order to detect the target RNA. Also provided are probes and kits for performing such a method.

Claims

1. A method for detection of target RNA in a sample comprising: (a) contacting said sample with at least one target-specific hairpin probe comprising; (i) a first domain which comprises a first region of complementarity to a target sequence in said target RNA and is capable of hybridising to said target RNA; (ii) a second domain which is not capable of hybridising to the target RNA, which comprises a second region of complementarity to a cognate circular or circularisable probe, and which is capable of acting as a primer for an RCA reaction; wherein the second domain is at least partly comprised within a hairpin structure of the probe, such that the second domain is not capable of binding to said cognate circular or circularisable probe and/or priming an RCA reaction until released from said hairpin structure, and allowing said hairpin probe to hybridise to the target RNA, wherein hybridization of said first domain to said target sequence causes the hairpin structure to open and release said second domain; (b) contacting said sample with a circular or circularisable probe cognate for said hairpin probe, wherein the circular or circularisable probe comprises one or more regions complementary to the second region of complementarity in the second domain, and hybridising the padlock probe to the hairpin probe at said second domain, wherein if the probe is a circularisable probe, it comprises 3′ and 5′ end regions complementary to the second region of complementarity in the second domain such that said ends are hybridised in juxtaposition for ligation, directly or indirectly, to each other; (c) if said probe is a circularisable probe, ligating, directly or indirectly the ends of said padlock probe to circularise the circularisable probe, thereby providing an RCA reaction template; (d) performing an RCA reaction using the circular probe of (b) or the RCA template of (c), wherein said RCA reaction is primed by the second domain; and (e) detecting the product of said RCA reaction.

2. The method of claim 1, wherein step (a) comprises contacting said sample with a probe set comprising two or more hairpin probes, each specific for a different target sequence in said target RNA, wherein the first domain of each hairpin probe in the set comprises a region of complementarity to a different target sequence.

3. The method of claim 1 or claim 2, for detecting more than one target RNA in a sample, wherein the sample is contacted with multiple different hairpin probes, or probe sets, each specific for a different target RNA.

4. The method of any one of claims 1 to 3, wherein the hairpin probe is an oligonucleotide, wherein at least part of the region at the 5′ end of the oligonucleotide is capable of hybridising to a region at the 3′ end of the oligonucleotide, such that the oligonucleotide forms a hairpin structure.

5. The method of claim 4, wherein the hairpin structure comprises 5′ to 3′; (i) a single stranded region at the 5′ end; (ii) a first stem strand region; (iii) a single stranded loop region; and (iv) a second stem strand region at the 3′ end.

6. The method of claim 4 or claim 5, wherein the first domain comprising the first region of complementarity to a target sequence comprises all or a portion of the 5′ single stranded region and at least a portion of the first stem strand region of the hairpin probe, and wherein the single stranded region acts as a toehold for binding of the target RNA.

7. The method of any one of claims 4 to 6, wherein the second domain comprising the second region of complementarity to a cognate circular or circularisable probe comprises at least a portion of the loop region and the second stem strand region of the hairpin probe.

8. The method of any one of claims 4 to 6, wherein the second domain lies in the loop region of the hairpin structure.

9. The method of claim 4, wherein the hairpin structure comprises from 5′ to 3′; (i) a first stem strand region at the 5′ end; (ii) a single stranded loop region; and (iii) a second stem strand region at the 3′ end.

10. The method of claim 9, wherein the first domain lies in the loop region of the hairpin structure and the second domain lies in the second stem strand region of the hairpin structure.

11. The method of any one of claims 1 to 10, wherein the stem of the hairpin structure comprises one or more mismatches.

12. The method of any one of claims 2 to 11, wherein step (b) comprises contacting said sample with two or more different circular or circularisable probes, each probe being cognate for a different hairpin probe.

13. The method of any one of claims 2 to 6, or 8 to 11, wherein the second domain of each member of a hairpin probe set is the same.

14. The method of claim 13, wherein step (b) comprises contacting said sample with two or more circular or circularisable probes, each probe being cognate for a different set of hairpin probes.

15. The method of any one of claims 1 to 14, wherein the target RNA is mRNA.

16. The method of any one of claims 3 to 13, wherein the circular or circularisable probes for each RNA target comprise a different reporter domain.

17. The method of any one of claim 2 to 12, 15 or 16, wherein within a probe set the circular or circularisable probe for each target sequence in a given target RNA comprises a different reporter domain, or the method of any one of claims 2 to 11 or 13 to 16, wherein within a probe set the circular or circularisable probe for each target sequence in a given target RNA comprises the same reporter domain.

18. The method of any one of claims 1 to 17, wherein the RCA product is detected using a detection oligonucleotide which hybridises specifically to the RCA product, by using a nucleic acid stain or by using labelled nucleotides for incorporation into the RCA product.

19. A hairpin probe for use in detecting a target RNA molecule, said probe comprising: (i) a first domain which comprises a first region of complementarity to a target sequence in a target RNA and is capable of hybridising to said target RNA; (ii) a second domain which is not capable of hybridising to the target RNA, which comprises a second region of complementarity to a cognate circular or circularisable probe, and which is capable of acting as a primer for an RCA reaction; wherein the second domain is at least partly comprised within a hairpin structure of the probe, such that the second domain is not capable of binding to said cognate circular or circularisable probe and/or priming an RCA reaction until released from said hairpin structure, and wherein the hairpin structure is such that hybridisation of said first domain to said target sequence causes the hairpin structure to open and release said second domain.

20. A probe set comprising two or more target-specific hairpin probes as defined in claim 19, each probe being specific for a different target sequence in the same target RNA, wherein the first domain of each hairpin probe in the set comprises a region of complementarity to a different target sequence in the target RNA.

21. The probe set of claim 20, wherein the members of the probe set are hairpin probes as defined in any one of claims 3 to 10 and 12.

22. A kit comprising: (i) a target-specific hairpin probe as defined in claim 19, or a probe set as defined in claim 20 or 21; and (ii) a circular or circularisable probe cognate for said hairpin probe, or circular or circularisable probes cognate for the hairpin probes in the probe set, each circular or circularisable probe comprising one or more regions complementary to the second region of complementarity in the second domain of said hairpin probe(s), wherein if the probe is a circularisable probe, it comprises 3′ and 5′ end regions complementary to the second region of complementarity in the second domain, such that said ends can hybridise in juxtaposition for ligation, directly or indirectly, to each other.

23. The kit of claim 22, wherein the kit comprises (i) a set of circular or circularisable probes, wherein each member of the circular or circularisable probe set is cognate for a different hairpin probe, or (ii) one or more circular or circularisable probes wherein each probe is cognate for all members of a given hairpin probe set.

24. The method, hairpin probe, probe set or kit of any one of claims 1 to 23, wherein; (i) said second domain is capable of functioning as a primer; or (ii) said circular or circularisable probe is a padlock probe.

Description

[0133] The method will now be described in more detail with reference to the following non-limiting examples.

[0134] FIG. 1 shows the structures of four hairpin probes (Mismatched Hairpin Probes), each 72 nucleotides in length, comprising a 12 nucleotide toehold at the 5′ end, a 20 nucleotide stem region and a loop region. The probes also contain varying numbers of mismatches in the stem region.

[0135] (A) shows a probe with no mismatch (SEQ ID NO: 1), wherein (a) is the 5′ end, 12 nt toehold, (b) is the 20 nt stem, with 0 mismatches, (c) is the 20 nt loop containing 6 nt spacers & 10 nt of PLP ligation site, and (d) is the continued PLP ligation site and 3′ end. dG=−22.19.

[0136] (B) shows a probe with 1 mismatch at the 5th nucleotide from the loop (SEQ ID NO: 2), wherein (a) is the 5′ end, 12 nt toehold, (b) is the 20 nt stem, with 1 mismatch on the 16.sup.th nt of the 20 nt stem, (c) is the 20 nt loop containing 6 nt spacers & 10 nt of PLP ligation site, and (d) is the continued PLP ligation site and 3′ end, with X marking the mismatch. dG=−17.72.

[0137] (C) shows a probe with 2 mismatches adjacent to each other at the 5th and 6th nucleotides from the loop (SEQ ID NO: 3), wherein (a) is the 5′ end, 12 nt toehold, (b) is the 20 nt stem, with 2 mismatches adjacent to each other on the 15.sup.th and 16.sup.th nts of the 20 nt stem, (c) is the 20 nt loop containing 6 nt spacers & 10 nt of PLP ligation site, and (d) is the continued PLP ligation site and 3′ end, with X marking the mismatches. dG=−15.23.

[0138] (D) shows a probe with 2 mismatches spaced at the 6th and 14th nucleotides from the loop (SEQ ID NO: 4), wherein (a) is the 5′ end, 12 nt toehold, (b) is the 20 nt stem, with 2 mismatches spaced apart from each other on the 7.sup.th and 15.sup.th nts of the 20 nt stem, (c) is the 20 nt loop containing 6 nt spacers & 10 nt of PLP ligation site, and (d) is the continued PLP ligation site and 3′ end, with X marking the mismatches. dG=−12.63.

[0139] FIG. 2 shows the general structure of an alternative probe design (Non-mismatched Hairpin Probes).

[0140] (A) shows a probe comprising a 12 nucleotide toehold at the 5′ end, and a similar 20 nucleotide stem region and a loop region (SEQ ID NO: 5), wherein (a) is the 5′ end, 12 nt toehold, (b) is the 20 nt stem, (c) is the 20 nt loop containing 3 nt spacers & 10 nt of PLP ligation site, and (d) is the continued PLP ligation site and 3′ end. dG=−20.38.

[0141] (B) shows a similar probe with a longer, 20 nucleotide toehold (SEQ ID NO: 6), wherein (a) is the 5′ end, 20 nt toehold, (b) is the 20 nt stem, (c) is the 20 nt loop containing 3 nt spacers & 10 nt of PLP ligation site, and (d) is the continued PLP ligation site and 3′ end. dG=−20.38.

[0142] FIG. 3 shows the general structure of another alternative probe design comprising a shorter stem (Short Hairpin Probes). These probes similarly comprise a toehold at the 5′ end, a stem and a loop region. The probe of (A) has a 12 nucleotide toehold (SEQ ID NO: 7), wherein (a) is the 5′ end, 12 nt toehold, (b) is the 10 nt stem with >60% G-C content, (c) is the loop region containing a 3 nt spacer, higher A-T content & PLP ligation site, and (d) is the continued PLP ligation site and 3′ end. dG=−11.09.

[0143] The probe of (B) has a 20 nucleotide toehold (SEQ ID NO: 8), wherein (a) is the 5′ end, 20 nt toehold, (b) is the 10 nt stem with >60% G-C content, (c) is the loop region containing a 3 nt spacer, higher A-T content & PLP ligation site, and (d) is the continued PLP ligation site and 3′ end. dG=−11.09.

[0144] FIG. 4 shows the general structure of a fourth alternative probe design (Loop Target Hairpin Probe) comprising a stem that starts at the 5′ end and ends at the 3′ end, with no toehold (SEQ ID NO: 9), wherein (a) is the 5′ end, 16 nt stem, (b) is the 24 nt long loop with a 3 nt spacer, and (c) is the 16 nt PLP ligation site and 3′ end. The loop region of this probe is made up of the complimentary sequence to the RNA target. dG=−14.29.

[0145] FIG. 5 shows a scheme according to one illustrative embodiment wherein the method involves the use of “unique” padlock probes. In this scheme the padlock probe binding site of the hairpin probe comprises part of the loop region and part of the stem. Accordingly, a portion of the padlock probe binding site is complementary to the sequence which hybridizes to the target RNA, and thus the padlock probe must be different for each target sequence. In the example, one target RNA comprising 4 target sequences is bound by 4 hairpin probes, each of which is in turn bound by a cognate padlock probe.

[0146] FIG. 6 shows alternative illustrative embodiments involving the use of “common” padlock probes. In the scheme in (A) the padlock probe binding site of the hairpin probe is entirely within the loop region and thus is unaffected by the sequence of the target RNA. Similarly, in the scheme in (B) the padlock binding site is entirely in the stem and the RNA targeting sequence is entirely in the loop region. Accordingly, in these examples, one target RNA comprising 4 target sequences is bound by 4 hairpin probes, each of which is in turn bound by a copy of the same padlock probe.

[0147] FIG. 7 shows the results of an in vitro experiment to confirm that in the presence of a target sequence, hairpin probes can successfully unfold to expose a padlock probe binding site, and thus lead to RCA initiation. Hairpin probes were allowed to hybridize with their mRNA target sequences in solution before padlock probes were added, together with a ligation mix comprising TTH ligase. The samples were then added to an amplification mix comprising Phi29 polymerase and dNTPs and incubated for 1 hour to allow amplification to occur. The resulting RCPs were then visualised using a fluorometer. The experiment was repeated; (i) with the template target sequences in excess of the hairpin probes (Test); (ii) with the hairpin probes in excess of the template target sequences (+ve); and (iii) with no template sequences at all (-ve). The results in FIG. 7A are from an experiment that was run with 4 different Mismatched Hairpin Probes; HP2, HP2A; HP2B; and HP2C, which correspond to the probes shown in FIGS. 1A-D. The results in FIG. 7B are from an experiment that was run with the third alternative probe design (Short Hairpin Probe) as depicted in FIGS. 3A and 3B.

[0148] FIG. 8 shows the results of an in situ experiment conducted on sections of mouse brain using hairpin probes of the invention and an alternative probe design as a control. The control probes used were linear and comprised a region of complementarity to the target and a 3′ end that is not complementary to the target but contains a padlock probe binding site. The samples were first fixed and washed, before hairpin or control probes were added and allowed to hybridise. Padlock probes and ligase were then added, followed by an amplification mix comprising Phi29 polymerase and dNTPs. Finally, detection oligonucleotides were added and the RCPs were detected via fluorescence imaging. The images in (A) were produced using the hairpin probes of the invention, and show the results from a region of tissue expected to highly express the target gene (left) and a region expected to have low levels of expression (right). The images produced using the control probes are shown in (B), again with a region of high predicted expression on the left, and a region of low predicted expression on the right.

EXAMPLE 1—PROBE DESIGN & SYNTHESIS

Mismatches in Stem Design

[0149] Telomere 72-nucleotide (nt) probes from Integrated DNA Technology (IDT) were designed with a 12-nt toehold, with a 20 nt long stem, and with different number of mismatches in the stem:

[0150] 1. No mismatches

[0151] 2. 1 Mismatch, 5.sup.th nt from loop

[0152] 3. 2 Mismatches, adjacent 5.sup.th and 6.sup.th nt from loop

[0153] 4. 2 Mismatches, spaced, 7.sup.th nt and 15.sup.th nt from loop

[0154] These probes are shown in FIGS. 1A-1D, respectively.

[0155] The padlock probe binding site was designed to be 30 nt in length, with a 6 nt spacer in between the binding site and the stem.

[0156] The mismatches in the stem of the hairpin probes were designed to destabilize the structure and allow the stem to ‘open up’ with relative ease in the presence of its target sequence.

Shorter Stem Design

[0157] In addition, Telomere 47-nucleotide (nt) (short toehold)/55-nt (long toehold) probes from Integrated DNA Technology (IDT) were designed with a 12-nt (short toehold)/20-nt (long toehold) targeting sequence at the 5′ end, a 22-nt sequence for Padlock Probe hybridization and ligation, and a 3-nt spacer in between.

[0158] The stems of the hairpin probes were designed to be 10-nt long, with at least 60% G-C content. The loops of the hairpin probes were also designed to contain more A-T than G-C content. The toeholds of hairpin probes designed were 12-nt (short toehold) and 20-nt (long toehold). The structures of these probes are shown in FIG. 3.

EXAMPLE 2—IN SOLUTION METHODS

[0159] As a proof of concept, in order to show that the hairpin probes can successfully unfold upon binding to a target sequence and expose a padlock probe binding site to trigger RCA initiation, an in vitro assay was conducted.

[0160] The NeuroD6 gene found in mouse brain was used to design the hairpin probes and padlock probes. In particular, 5 different, non-overlapping sequences of the NeuroD6 mRNA sequence were selected. The hairpin probes were designed to target the 1 site of the NeuroD6 gene. The corresponding sequence of the aforementioned mRNA was ordered from IDT and subsequently tested in solution.

[0161] To quantify if designed hairpin probes would hybridize to the target sequence and fully open up with an 1′ overhang for padlock probe binding for amplification, two different test conditions were assessed. Firstly, the reaction was carried out with the hairpin probes in excess, relative to the templates; and secondly, the reaction was repeated with the templates in excess relative to the hairpin probes.

[0162] In addition, a negative control consisting of a mixture of hairpin probes and padlock probes only was left to hybridize in the absence of a target sequence, to assess the stability and specificity of the hairpin probes.

Step 1: Hybridization

[0163] Designed hairpin probes were first left to hybridize to the target sequence of the mRNA target of the NeuroD6 gene in 5×SSC, 0.01% Tween Hybridization buffer (5×SSCT).

Positive Tests:

[0164] 1. Hairpin probes in excess of templates [0165] 400 nM of hairpin probes, 100 nM of templates were left to hybridize for 3 h at 37 C in 2.5×SSC, 0.005 Tween buffer. [0166] 2. Templates in excess of hairpin probes [0167] 400 nM of templates, 100 nM of hairpin probes were left to hybridize for 3 h at 37 C in 2.5×SSC, 0.005 Tween buffer.
Negative control: [0168] 1. 100 nM of hairpin probes were left to incubate for 3 h at 37 C in 2.5×SSC, 0.005 Tween buffer.

Step 2: Ligation

[0169] Reaction mix from ‘Step 1: Hybridization’ was then added to Ligation Mix for a 10× dilution of Hybridization Mix. Ligation Mix consists of 1×TTH Buffer, 0.05M KCl, 20% Formamide, 0.2 μg/μl BSA, 10.0 nM of padlock probes and 0.25 u/μl of TTH Ligase. Samples were then incubated at 45 C for 30 mins.

Step 3: Amplification & Quantification

[0170] Reaction mix from ‘Step 2: Ligation’ was then added to Amplification Mix for a 10× dilution of Ligation Mix. The Amplification Mix consists of 1× Phi29 Buffer, 0.25 mM dNTPs, 0.2 μg/μl BSA, 5% Glycerol, 0.2 u/μl Phi29 polymerase enzyme. Samples were then incubated at 37 C for 1 h.

[0171] Quantification of RCPs from the amplification step was carried out using a Qubit fluorometer. 5.0 μl of Amplification mix was mixed with 195.0 μl of Qubit working solution which containing a fluorophores that binds to DNA for florescent quantification.

[0172] The positive and negative control tests were performed in parallel to validate that signals produced from the positive tests are not attributed to padlock probes ‘opening’ the hairpin probes to undergo binding, ligation and subsequently RCA. This validates that hairpin probes will remain hybridized to themselves in the absence of the target template and not open up for padlock probe binding, thus avoiding RCA initiation.

[0173] The results obtained from the positive tests validated that signals can be attributed to the fact that the hairpin probes ‘opened up’ to hybridize to the templates, exposing the padlock probe binding site for the padlock probe to bind, circularize and subsequently undergo RCA, producing RCPs.

[0174] The results from these experiments are shown in FIG. 7.

EXAMPLE 3—IN SITU METHODS

[0175] In situ experiments were conducted on 10 μm sections of mouse brain of C57BL/6J mice. In addition to the hairpin probes of the present invention, this experiment also used an alternative probe design as a control. The control system involved the use of linear (non-hairpin) probes comprising a region of complementarity to the target and a 3′ end that is not complementary to the target but contains a padlock probe binding site. As with the present hairpin probes, the padlock probe can bind, become circularized via ligation and initiate RCA. These linear probes (control probes or control primers), however, can be targeted by padlock probes and initiate RCA even in the absence of the target template, hence generating unspecific background signals in vitro and in tissue sections where linear probes bind and/or stick to the tissue matrix unspecifically.

Step 1: Fixation

[0176] All samples were first fixed and incubated in a fixation solution containing 3.7% PFA in DEPC-PBS solution in room temperature for 5 mins. The samples were then incubated in Wash Buffer 1 (WB1) solution containing DEPC-PBS for 1 min, before incubating at RT for 5 mins in Permeabilization Solution (PS), containing (0.1M HCl in DEPC H.sub.2O). Next, the samples were then washed in WB1 twice, before incubating in 70% and 100% EtOH for 2 mins respectively. Lastly, a 50 ul SecureSeal Chamber (Grace Bio-Labs) was then mounted onto the slide to facilitate hybridization in Step 2.

Step 2: Probe Hybridization

[0177] Wash Buffer 2 (WB2) containing DEPC-PBS-Tween 0.05% was applied to the SecureSeal Chambers, whilst a Probe Mix containing 2×SSC, 10% Formamide, 10 mM MgCl.sub.2 and 50 nM hairpin or control probes was prepared. The Probe Mix was then added to the chambers and left to incubate for 3 h at 37 C.

Step 3: Probe Ligation

[0178] Samples were then washed thrice in WB2 before Ligation Mix containing 1×TTH Buffer, 0.2 μg/μl BSA, 0.05M KCl, 100 nM padlock probes, 20% Formamide and 0.5 μg/μl TTH Ligase was added. The slides were then incubated at 37 C for 1 h.

Step 4: Amplification & Florescent Labeling

[0179] Samples were washed twice in WB2, before Amplification Mix containing 1×Phi 29 Buffer, 0.25 mM dNTP, 0.2 μg/μl BSA, 10% glycerol and 1 μg/μl Phi29 Polymerase enzyme was added. The samples were then incubated either at 30 C overnight, or at 37 C for 3 h. The Amplification Mix was then removed from the chambers, which were washed with WB2 thrice, leaving the last wash buffer in the chamber. The samples were then incubated in 2×SSC, 20% formamide and 100 nM of detection oligos for 30 minutes in darkness at room temperature. Unbound detection oligos were then removed with three washes with WB2.

[0180] The quantification blobs from experiments were validated by imaging and running images through CellProfiler pipelines. Images were taken under Nikon Ti2 microscope, running on NIS Elements AR software. The images for gene expression in selected area was selected to be 2×2×2 tile region in the highly and lowly expressed region of the samples, based on publicly available data from the Allen Cell Atlas. The selected areas of interests were then imaged with 20× objective with illumination of Cy3 and DAPI. A stack of images at different focal depths were imaged for RCPs located in different focal Z planes, and were then merged to a single image using the ‘Create Maximum Intensity Projection (MIP) Image in Z’ function in the ‘NIS Elements AR’ software.

[0181] The resulting MIP image in Z was then analyzed on a customized pipeline on CellProfiler where nucleis and RCPs were identified, separated and quantified based on shape descriptors and distance from each identified nuclei. The pipeline was run on CellProfiler 2.1.2 and all quantification of blob and nuclei counts and intensity information was then saved as a .csv file.

[0182] Images are shown in FIG. 8A from two different regions inside the brain tissue sections. In the high expressing region, Neurod6 is expected to be medium-highly expressed (according to the Allen Brain atlas in situ hybridization data). In contrast, in the low expressing region, Neurod6 is expected to be expressed at a low level.

[0183] Corresponding results are shown in FIG. 8B from equivalent experiments that were done using the control probes, rather than the hairpin probes of the present invention.

[0184] It is noted that most of these signals in the control experiment must be non-specific since there is almost no difference between low and high expressed region. Moreover, it is noted that there are a lot of signals outside cells, which are most likely attributed to sticking of control primers non-specifically to the tissue section and generating a RCA product everywhere it sticks. Hairpin probes generate fewer signals, but those signals are most likely specific as we can see a clear difference in low expressed region and high expressed region of the tissue (as would be expected). Moreover there are very few non-specific signals outside the cells.