METHOD FOR CAPTURING RNA IN SITU HIGHER-ORDER STRUCTURES AND INTERACTIONS

Abstract

The present invention discloses a method for capturing an RNA in situ higher-order structure and interaction. The method includes: fixing protein-mediated RNA-RNA interaction in cell or tissue; performing membrane permeabilization while keeping the cell intact; degrading free RNA; labeling the 3′ end of the RNA with pCp-biotin and performing proximal ligation in situ; purifying the chimeric RNA containing the pCp-biotin after the cell is digested; constructing the strand-specific library; and performing high-throughput sequencing. In the present invention, under the condition of not destroying the cell structure and keeping the integrity of cell, treat the intracellular RNA in situ, and capture RNA intra- and intermolecular interactions in a physiological state; the 3′ end of the RNA is labeled with the pCp-biotin, and in situ ligation is performed under non-denaturing conditions, thereby greatly improving the labeling efficiency and reducing intermolecular specific ligation; and the chimeric RNA labeled with C-biotin is enriched by C1 magnetic beads, so that the fraction of effective sequencing data is increased, and the sequencing cost is reduced.

Claims

1. A method for capturing an RNA higher-order structure in situ and/or verifying in situ RNA-RNA interaction, comprising the following steps: (1) treating cell or tissue sample to fix protein-mediated RNA-RNA proximal interaction; (2) performing membrane permeabilization while keeping the cell intact; (3) degrading free RNAs that are not protected by protein; (4) labeling the 3′ end of RNA protected by the protein with a “pCp-Marker 1” and performing proximal ligation in situ; (5) purifying the chimeric RNA containing the “C-marker 1” after cells are digested; and constructing strand-specific library; (6) performing high-throughput sequencing.

2. The method according to claim 1, wherein in the step (1), treating cell or tissue sample is to perform formaldehyde cross-linking on the cell or tissue sample.

3. The method according to claim 2, wherein the step (1) is performed according to a method comprising the following steps: (a1) placing the cell or tissue sample in formaldehyde solution at room temperature for 10 minutes.

4. The method according to claim 3, wherein the formaldehyde solution is 1% (v/v) formaldehyde solution.

5. The method according to claim 3, wherein after the step (a1), the method further comprises the following step (a2): (a2) adding glycine solution into the cell or tissue sample treated in the step (a1), and incubating for 10 minutes.

6. The method according to claim 5, wherein the glycine solution is a glycine solution with a concentration of 0.125 mol/L.

7. The method according to claim 1, wherein in the step (2), the permeabilization solution used during the membrane permeabilization is a Permeabilization buffer; and the solvent of the Permeabilization solution is 10 mM Tris-HCl buffer with pH 7.5, and the solute and concentration are as follows: 10 mM of NaCl, 0.5% (v/v) NP-40, 0.3% (v/v) Triton X-100, 0.1% (v/v) Tween 20, 1×protease inhibitors and 2 U/ml of SUPERase.In™ RNase Inhibitor.

8. The method according to claim 7, wherein the step (2) is performed according to a method comprising the following steps: (b1) placing the cell or tissue sample treated in the step (1) in the Permeabilization buffer at 0° C.-4° C. for 15 minutes.

9. The method according to claim 8, wherein after the step (b1), the method further comprises the following step (b2): (b2) washing the cell or tissue sample treated in the step (b1) with 1×PNK solution, wherein the solvent of the 1×PNK solution is 50 mM of Tris-HCl buffer with pH 7.4, and the solute and concentration are as follows: 10 mM of MgCl.sub.2, 0.1 mg/ml of BSA, and 0.2% (v/v) NP-40.

10. The method according to claim 1, wherein in the step (3), the MNase is used for achieving the “degrading free RNAs that are not protected by protein”.

11. The method according to claim 1, wherein the step (4) can be performed according to a method comprising the following steps: (d1) hydroxylating the 3′ end of the RNA protected by protein; (d2) labeling the 3′ end of RNA as “Cp-Marker 1”; (d3) converting the phosphate group in “Cp-Marker 1” at the 3′ end of the RNA into hydroxyl group; (d4) phosphorylating the 5′ end of the RNA; (d5) performing proximal ligation in situ.

12. The method according to claim 11, wherein in the step (d1), by treating the sample treated in the step (3) with alkaline phosphatase, the 3′ end of the RNA protected by the protein can be hydroxylated.

13. The method according to claim 11, wherein in the step (d2), by adding the “Cp-Marker 1” into the sample treated in the step (d1) to perform a ligation reaction, the 3′ end of the RNA is labeled as the “Cp-Marker 1”.

14. The method according to claim 11, wherein in the step (d3), treating the sample treated in the step (d2) with the alkaline phosphatase to convert the phosphate group in the “Cp-Marker 1” at the 3′ end of the RNA into the hydroxyl group.

15. The method according to claim 11, wherein in the step (d4), treating the sample treated in the step (d3) with T4 PNK enzyme to phosphorylate the 5′ end of the RNA.

16. The method according to claim 11, wherein in the step (d5), by adding the T4 RNA ligase into the sample treated in the step (d4), the proximal ligation is performed in situ.

17. The method according to claim 1, wherein the step (5) is performed according to a method comprising the following steps: (e1) using proteinase K to digest the cell; (e2) extracting total RNA, and performing fragmentation treatment; (e3) enriching the chimeric RNA labeled with “C-marker 1” by using magnetic beads on which the marker 2 is fixed; and the marker 2 can specifically bind to the marker 1; (e4) constructing strand-specific library.

18. The method according to claim 1, wherein in the method, the maximum starting amount of the cells is 1×10.sup.7 cells.

19. The method according to claim 1, wherein the cell is animal cell, and the tissue is animal tissue.

20. A library construction method, comprising the step (1) to the step (5) of the method according to claim 1.

21. The library constructed by using the method of claim 20 for capturing an RNA higher-order structure in situ and/or identifying in situ RNA-RNA interaction.

22. An application of the method according to claim 1 for identifying lncRNA targets in living cells.

23. An application of pCp-biotin for identifying RNA-RNA proximal interactions.

24. An application of pCp-biotin in RNA proximal ligation in situ.

25. An application of pCp-biotin in chimeric RNA enrichment.

26. An application of the detergent, which is the Permeabilization buffer in claim 7.

27. An auxiliary use of the detergent described in claim 26 in the membrane permeabilization of cells.

28. An application of the MNase, the alkaline phosphatase and/or the T4 polynucleotide kinase in the in situ ligation of RNA.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0088] FIG. 1. Overview and evaluation of RIC-seq. (A) Schematic diagram of the RIC-seq process. The in situ part includes formaldehyde crosslinking, permeabilization, RNA digestion, pCp-biotin labeling and proximity ligation. For the in vitro part, chimeric RNAs were enriched and converted into libraries for paired-end sequencing. RBP stands for RNA-bound proteins. (B) The base content surrounding the junction of chimeric reads. (C) Comparison of known structures (blue arcs) of miR3064 to RIC-seq chimeric reads (gray boxes). The light blue lines between chimeric reads represent gaps. The stem-loop structural model of a single RIC-seq read is shown at the bottom, and the pCp insertion is marked in red. (D) pCp is enriched in the apical loop of pre-miRNAs. (E-G) RIC-seq recapitulates known structures and interactions of U4, U6, RPPH1 and U3 snoRNA. (H) RIC-seq identifies U1 binding sites in MALAT1. Pink shaded regions are shared between RIC-seq datasets and PARIS or RAP data. Blue shaded peaks are detected only by RIC-seq. The dashed-line boxed region is shown in (I). (I) Novel U1 interacting sites in MALAT1 are conserved and proved by chimeric clusters (gray arrowheads). U1 motif shown in purple. (J) RNA map showing RNARNA interactions across all chromosomes. Lower left, +pCp samples. Upper right, -pCp samples. The NEAT1 and MALAT1 interactions are magnified and shown on the right. (K) SIM analysis revealed colocalization of MALAT1 and the NEAT1 5′ end. The regions marked Box 1 and Box 2 are magnified in the middle panel. The direct overlapping loci in Box 3 and Box 4 are shown in yellow.

[0089] FIG. 2. The mapping pipeline and reproducibility of RIC-seq. (A) The mapping pipeline for RIC-seq data. PCR duplicates, adapters and reads containing polyN were first removed. After filtering, the paired reads were mapped separately to the hg19 reference genome using the STAR program. (B) RIC-seq replicates are highly correlated. (C) The global false positive rate of RIC-seq is 0.6% (dashed-line boxed region) by a cell mixing strategy. (D) RIC-seq recapitulates snoRNA interacting sites in 28S rRNA. The red arrow indicates known modification sites. The boxed region represents the D′ box. (E) The genomic distribution of snoRNA interacting sites detected by RIC-seq. (F) The SNORD22 interacting sites in SPHK2 and BCL2L2 genes. The D-box is shown in blue. (G) The pie chart of intramolecular and intermolecular RNA-RNA interactions. (H) The violin plot shows the expression levels of MALAT1- and NEAT1-targeting genes. The P-value was calculated by a two-tailed Student's t-test. (I) The enriched motifs among MALAT1 or NEAT1 chimeric targets. (J) Summary of MALAT1 loci in 15 cells. (K) Summary of NEAT1 loci and their overlap with MALAT1 in 15 cells by smFISH. Blue bars represent the direct overlap between NEAT1 and MALAT1 loci.

[0090] FIG. 3. RIC-seq precisely recaptures the 3D structure of 28S rRNA. (A) RNA physical interaction map inferred from the Cryo-EM structure of 28S rRNA. Spatial distances greater than 25 Å are shown in light gray (distal). For interactions within 25 Å, WC and non-WC base pairs are shown in blue and green, respectively. Interactions containing both WC and non-WC base pairs are shown in red, whereas other kinds of proximal interactions are shown in purple. Not available group means no structural data are available. (B) RNA 3D map of 28S rRNA derived from RIC-seq data in HeLa cells. Boxed regions illustrate local WC base pairs and long-range non-WC interactions. (C) The true positive and true negative datasets are generated from Cryo-EM structural data. True positive, brown; true negative, cyan. (D) RIC-seq (black line) shows better performance than PARIS (orange line) in detecting the 3D conformations of 28S rRNA. Dashed line, random. The missing part in the Cryo-EM structure was not used for generating the ROC curve.

[0091] FIG. 4. The topological organization and folding principle of RNA in vivo. (A) The topological regions (dashed triangle) observed in the precursor RNA of PDE3A, IMMP2L, FTX and PVT1. The heatmaps were created by dividing each transcript into 100 bins (1% length per bin=1 pixel) and normalizing the sum to 1.0. (B) Contact probability of each pre-mRNA and pre-lncRNA as a function of the linear distance. The slope of −1 fits a theoretical model of the fractal globule. (C) Contact probability of each mature mRNA and lncRNA as a function of the linear distance. The dashed line indicates a slope of −1.

[0092] FIG. 5. Mapping RNA 3D interaction atlas in diverse cell lines. (A) The contact matrix across all chromosomes in GM12878, IMR-90, H1 hESC, NPC and HT29 cells. A magnification of specific RNA interactions in chr4:93.2 Mb-94.8 Mb is shown at the bottom. (B) Magnification of constitutive RNA-RNA interactions across 5 cell lines from (A). (C-D) The specific and constitutive RNA-RNA interactions are illustrated, respectively. The RNA-seq, ChIP-seq and TAD signals are from H1 hESC ENCODE data. (E) CRISPRi of LncPRESS2 by Cas9-KRAB. (F) Quantification of GRID2 and OCT4 expression levels upon knockdown of LncPRESS2. *P<0.05, **P<0.01 and ***P<0.001 by two-tailed Student's t-test (n=3).

[0093] FIG. 6. The characteristics of in situ RNA-RNA interactions. (A) The percentages of intra- and interchromosomal RNA-RNA interactions across 6 different cell types. (B) The percentages of intra- and intergene RNA-RNA interactions. (C) The percentage and span distance of the intrachromosomal chimeric reads in 6 different cell types. Solid and dashed lines denote intragene or intergene interactions, respectively. (D) The Hi-C, RIC-seq and RNA-seq compartments in chromosome 1. Data for GM12878 are shown. (E) The percentages of RNA-RNA interactions between different compartments for intra- and intergenes.

[0094] FIG. 7. Cell type-specific hub-RNAs in the human genome. (A) Each RNA in HeLa cells is sorted based on its chimeric tag intensity and the number of interacting genes. GAPDH served as a negative control. (B) Circos plot of MALAT1, CCAT1 and PDE3A-interacting RNAs across 23 chromosomes. The red arrow marks gene location. (C) Meta-analysis of the RIC-seq intensity and distribution for hub-RNAs and other RNAs. RIC-seq signals surrounding the transcription start site (TSS) and transcription termination site (TTS) were plotted. (D) Hub-RNAs are more conserved than other RNAs. (E-H) ChIP-seq signals of RNA polymerase II, H3K4me3, H3K27ac and H3K27me3 on hub-RNA and other RNAs. (I) The majority of hub-RNAs are specific to cell type.

[0095] FIG. 8. Hub-RNA CCAT1-5L boosts MYC gene expression by synergizing with MYC promoter and enhancer RNA. (A) Snapshot of the RIC-seq, RNA-seq and H3K27ac signals on 8q24. The 5′ and 3′ RACE-mapped CCAT1 transcript is shown at the bottom. Northern blot probes are labeled as black lines. CCAT1, MYC and PVT1 genes are marked in red. The chimeric reads between CCAT1-5L and MYC are shown in red. (B) Northern blot analysis of CCAT1-5L expression across diverse cell lines. The 5L probe detects only CCAT1-5L in HeLa cells. The 18S and 28S rRNAs served as loading controls. (C) CCAT1-5L is localized in the nucleus by smFISH. CCAT1-5L is shown in red, CCAT1-Exon2 probes (CCAT1-E) are shown in green, the NEAT1 5′ region is shown in red, and DAPI is shown in blue. Scale bar: m. (D) The MYC expression was reduced upon knockdown of CCAT1 by 5L-specific LNA oligos. The 5L- and Exon2-specific primers were used to monitor CCAT1 levels. (E) CCAT1-5L, MYC promoter and MYC enhancer RNA are colocalized by smFISH. CCAT1-5L is pseudocolored in red, MYC in green and PVT1 in yellow. Scale bar: 5 m. (F) The cell proliferation rate was measured upon the depletion or ectopic expression of CCAT1-5L. (G) Knockdown or ectopic expression of CCAT1-5L affects colony formation. (H) CCAT1-5L is critical for cell metastasis in a transwell assay. Scale bar: 50 m. *P<0.05 and **P<0.01 by two-tailed Student's t-test (n=3).

DETAILED DESCRIPTION OF THE EMBODIMENTS

[0096] The following embodiments facilitate a better understanding of the present invention, but do not limit the present invention. The experimental methods in the following embodiments, unless otherwise specified, are all conventional methods. The test materials used in the following embodiments, unless otherwise specified, are all purchased from conventional biochemical reagent stores.

[0097] The formulations of solutions used in the following embodiments are as follows:

[0098] PBS buffer (pH 7.4): the solvent is water, and the solute and concentration are as follows: 137 mmol/L of NaCl, 2.7 mmol/L of KCl, 10 mmol/L of Na.sub.2HPO.sub.4, and 2 mmol/L of KH.sub.2PO.sub.4.

[0099] 1×PNK solution: the solvent is 50 mM Tris-HCl buffer with pH 7.4, and the solute and concentration are as follows: 10 mM of MgCl.sub.2, 0.1 mg/ml of BSA, and 0.2% (v/v) NP-40.

[0100] 1×PNK+EGTA solution: the solvent is 50 mM Tris-HCl buffer with pH 7.4, and the solute and concentration are as follows: 20 mM of EGTA, and 0.5% (v/v) NP-40.

[0101] High-salt solution: the solvent is a 5×PBS (no Mg.sup.2+, Ca.sup.2+), and the solute and concentration are as follows: 0.5% (v/v) NP-40. Wherein, the 5×PBS (no Mg.sup.2+, Ca.sup.2+) is the 5×PBS buffer (pH 7.4): 685 mmol/L of NaCl, 13.5 mmol/L of KCl, 50 mmol/L of Na.sub.2HPO.sub.4, and 10 mmol/L of KH.sub.2PO.sub.4.

[0102] Permeabilization solution: 10 mM of Tris-HCl (pH 7.5), 10 mM of NaCl, 0.5% (v/v) NP-40, 0.3% (v/v) Triton X-100, 0.1% (v/v) Tween 20, 1×protease inhibitors (Sigma, catalog number: P8340-5ML, the specific components include AEBSF, Aprotinin, Bestatin hydrochloride, E-64, Leupeptin hemisulfate salt and Pepstatin A) and 2 U/ml of SUPERase.In™ RNase Inhibitor (Thermo Fisher, catalog number: AM2694).

[0103] 1×MN reaction solution: the solvent is 50 mM Tris-HCl buffer with pH 8.0, and the solute and concentration are as follows: 5 mM of CaCl.sub.2).

[0104] Proteinase K solution: the solvent is 10 mM of Tris-HCl buffer with pH 7.5, and the solute and concentration are as follows: 10 mM of EDTA, and 0.5% (w/v) SDS.

[0105] 5×hybridization solution: 1M of NaCl, and 500 mM of Tris-HCl (pH 7.4).

[0106] Solution A: 0.1 M of NaOH, and 0.05 M of NaCl.

[0107] Solution B: 0.1 M of NaCl.

[0108] 2×TWB solution: 10 mM of Tris-HCl (pH 7.5), 1 mM of EDTA, 2 M of NaCl, and 0.02% (v/v) Tween 20.

[0109] PK solution: 100 mM of NaCl, 10 mM of Tris-HCl (pH 7.0), 1 mM of EDTA, and 0.5% (w/v) SDS.

[0110] TE buffer: 10 mM of Tris-HCl (pH 8.0), and 1 mM of EDTA.

Embodiment 1. Preparation Method of a RIC-Seq Library

[0111] The construction process of the RIC-seq library of the present invention is shown in FIG. 1 A. The construction process includes cell culture, formaldehyde cross-linking, cell membrane and nuclear membrane permeabilization, MNase enzyme treatment, hydroxylation treatment of the 3′ end of RNA, pCp-biotin ligation, hydroxylation treatment of the 3′ end of the RNA, phosphorylation treatment of the 5′ end, proximal ligation, total RNA extraction, removal of genomic DNA by DNase I, removal of ribosome RNA, RNA fragmentation, C1 magnetic bead enrichment and elution of the enriched RNA, cDNA first-stranded synthesis, second-stranded synthesis of DNA, tail end repair, “A” addition, adapter ligation, PCR amplification and other steps. The specific steps are as follows:

[0112] 1. Taking cells in a 15 cm dish with a density of about 80-90%, discarding the culture medium, adding 10 ml of precooling PBS (pH 7.4) to wash the cells, discarding the PBS, and repeating this step 3 times to obtain washed cells.

[0113] 2. After completing the step 1, adding 10 ml of 1% (v/v) formaldehyde solution (the solvent is PBS solution) into the washed cells obtained in the step 1, and incubating at room temperature for 10 minutes. Then adding a glycine solution (the final concentration is 0.125 mol/L, and the solvent is DEPC water) to terminate the reaction, and incubating at room temperature for 10 minutes to obtain formaldehyde cross-linked and terminated cells.

[0114] 3. After completing the step 2, adding 10 ml of precooling PBS (pH 7.4) into the formaldehyde cross-linked and terminated cells obtained in the step 2, performing washing for 3 times, scraping the cells with a cell lifter and transferring the cells into a 50 ml centrifuge tube, performing centrifugation at 2500 rpm for 10 minutes at 4° C., discarding the supernatant, adding 2 ml of precooling PBS (pH 7.4) to resuspend the cell pellet, transferring the cell suspension into two 1.5 ml eppendorf centrifuge tubes, wherein 1 ml of cell suspension is transferred into each eppendorf centrifuge tube, performing centrifugation at 2500 rpm for 10 minutes at 4° C., discarding the supernatant, and continuing the next step or storing the cell pellet in a refrigerator at −80° C.

[0115] 4. After completing the step 3, adding 1 ml of Permeabilization buffer into the cell pellet obtained in the step 3, incubating on ice for 15 minutes, and uniformly mixing the solution every 2 minutes. performing centrifugation at 3500 rpm for 5 minutes at 4° C., discarding the supernatant, adding 600 μl of 1×PNK buffer to resuspend the cell pellet, performing uniform mixing for 5 minutes at 4° C. by rotating (20 rpm), performing centrifugation at 3500 rpm for 5 minutes at 4° C., discarding the supernatant, and repeating this step twice.

[0116] 5. After completing the step 4, adding 200 μl of MNase (Thermo Fisher with the catalog number of EN0181) diluted at a volume ratio of 1:10000 into the cell pellet obtained in the step 4 with 1×MN buffer (wherein the concentration of the MNase enzyme is 0.03 U/l) to resuspend the cell pellet, Incubate the suspension in the ThermoMixer at 37° C. for 10 minutes with intermittent mixing at 1000 rpm every 2 min for 15 sec. After the reaction is completed, centrifuging the tube at 3500 rpm for 5 minutes at 4° C., discarding the supernatant, adding 600 μl of 1×PNK+EGTA buffer to resuspend the cell pellet, performing uniform mixing for 5 minutes at 4° C. by rotating (20 rpm), centrifuging the tube at 3500 rpm for 5 minutes at 4° C., discarding the supernatant, and repeating this step once. Adding 600 μl of 1×PNK buffer to resuspend the cell pellet, performing uniform mixing for 5 minutes at 4° C. by rotating (20 rpm), centrifuging the tube at 3500 rpm for 5 minutes at 4° C., discarding the supernatant, and repeating this step once.

[0117] 6. After completing the step 5, adding 10 μl of 10×FastAP buffer (the product of the Thermo Fisher company), 10 μl of Fast Alkaline Phosphatase (the product of the Thermo Fisher company with the catalog number of EF0651; and the final concentration in the reaction system is 0.1 U/l), and 80 μl of DEPC water into the cell pellet obtained in the step 5 to resuspend the cell pellet, incubating in the ThermoMixer at 37° C. for 10 minutes with intermittent mixing at 1000 rpm every 3 min for 15 sec. After the reaction is completed, centrifuging the tube at 3500 rpm for 5 minutes at 4° C., discarding the supernatant, adding 600 μl of 1×PNK+EGTA buffer to resuspend the cell pellet, performing uniform mixing for 5 minutes at 4° C. by rotating (20 rpm), centrifuging the tube at 3500 rpm for 5 minutes at 4° C., discarding the supernatant, and repeating this step once. Adding 600 μl of high-salt buffer to resuspend the cell pellet, performing uniform mixing for 5 minutes at 4° C. by rotating (20 rpm), centrifuging the tube at 3500 rpm for 5 minutes at 4° C., discarding the supernatant, and repeating this step once. Adding 600 μl of 1×PNK buffer (compared with the previous formula, the content of NP-40 is adjusted to 0.05% (v/v), while the other ingredients and contents remain unchanged) to resuspend the cell pellet, performing uniform mixing for 5 minutes at 4° C. by rotating (20 rpm), centrifuging the tube at 3500 rpm for 5 minutes at 4° C., discarding the supernatant, and repeating this step once.

[0118] 7. After completing the step 6, adding 10 μl of 10×RNA ligase reaction buffer (the product of the Thermo Fisher company), 6 μl of RNase inhibitor, 4 μl of Biotinylated Cytidine (Bis) phosphate (i.e., pCp-biotin, the product of the Thermo Fisher company with the catalog number of 20160) (1 mM), 10 μl of T4 RNA ligase (the product of the Thermo Fisher company with the catalog number of EL0021; and the final concentration in the reaction system is 1 U/l), 20 μl of DEPC water and 50 μl of 30% PEG into the cell pellet obtained in the step 6 to resuspend the cell pellet, incubating at 16° C. in the ThermoMixer for overnight with intermittent mixing at 1000 rpm every 3 min for 15 sec. After the reaction is completed, performing centrifugation at 4° C. and 3500 rpm for 5 minutes, discarding the supernatant, adding 600 μl of 1×PNK solution to resuspend the cell pellet, performing uniform mixing for 5 minutes at 4° C. by rotating (20 rpm), centrifuging the tube at 3500 rpm for 5 minutes at 4° C., discarding the supernatant, and repeating this step twice.

[0119] 8. After completing the step 7, adding 10 μl of 10×FastAP buffer (the product of the Thermo Fisher company), 10 μl of Fast Alkaline Phosphatase (the product of the Thermo Fisher company with the catalog number of EF0651; and the final concentration in the reaction system is 0.1 U/l), and 80 μl of DEPC water into the cell pellet obtained in the step 7 to resuspend the cell pellet, incubating in the ThermoMixer at 37° C. for 10 minutes with intermittent mixing at 1000 rpm every 3 min for 15 sec. After the reaction is completed, centrifuging the tube at 3500 rpm for 5 minutes at 4° C., discarding the supernatant, adding 600 μl of 1×PNK+EGTA buffer to resuspend the cell pellet, performing uniform mixing for 5 minutes at 4° C. by rotating (20 rpm), centrifuging the tube at 3500 rpm for 5 minutes at 4° C., discarding the supernatant, and repeating this step once. Adding 600 μl of high-salt buffer to resuspend the cell pellet, performing uniform mixing for 5 minutes at 4° C. by rotating (20 rpm), centrifuging the tube at 3500 rpm for 5 minutes at 4° C., discarding the supernatant, and repeating this step once. Adding 600 μl of 1×PNK buffer (compared with the previous formula, the content of NP-40 is adjusted to 0.05% (v/v), while the other ingredients and contents remain unchanged) to resuspend the cell pellet, performing uniform mixing for 5 minutes at 4° C. by rotating (20 rpm), centrifuging the tube at 3500 rpm for 5 minutes at 4° C., discarding the supernatant, and repeating this step twice.

[0120] 9. After completing the step 8, adding 10 μl of 10×PNK buffer (the product of the Thermo Fisher company), 15 μl of 10 mM ATP, 10 μl of T4 PNK (the product of the Thermo Fisher company with the catalog number of EK0032; and the final concentration in the reaction system is 1U/l), and 65 μl of DEPC water into the cell pellet obtained in the step 8 to resuspend the cell pellet, incubating at 37° C. for 45 minutes in the ThermoMixer with intermittent mixing at 1000 rpm every 3 min for 15 sec. After the reaction is completed, centrifuging the tube at 3500 rpm for 5 minutes at 4° C., discarding the supernatant, adding 600 μl of 1×PNK+EGTA buffer to resuspend the cell pellet, performing uniform mixing for 5 minutes at 4° C. by rotating (20 rpm), centrifuging the tube at 3500 rpm for 5 minutes at 4° C., discarding the supernatant, and repeating this step once. Adding 600 μl of 1×PNK buffer (compared with the previous formula, the content of NP-40 is adjusted to 0.05% (v/v), while the other ingredients and contents remain unchanged) to resuspend the cell pellet, performing uniform mixing for 5 minutes at 4° C. by rotating (20 rpm), centrifuging the tube at 3500 rpm for 5 minutes at 4° C., discarding the supernatant, and repeating this step once.

[0121] 10. After completing the step 9, adding 20 μl of 10×RNA ligase reaction buffer (the product of the Thermo Fisher company), 8 μl of RNase inhibitor, 10 μl of T4 RNA ligase (the product of the Thermo Fisher company with the catalog number of EL0021; and the final concentration in the reaction system is 0.5 U/l), 20 μl of BSA (1 mg/ml) and 142 μl of DEPC water into the cell pellet obtained in the step 9 to resuspend the cell pellet, incubating in the ThermoMixer at 16° C. for overnight with intermittent mixing at 1000 rpm every 3 min for 15 sec. After the reaction is completed, centrifuging the tube at 3500 rpm for 5 minutes at 4° C., discarding the supernatant, adding 600 μl of 1×PNK buffer to resuspend the cell pellet, performing uniform mixing for 5 minutes at 4° C. by rotating (20 rpm), centrifuging the tube at 3500 rpm for 5 minutes at 4° C., discarding the supernatant, and repeating this step twice.

[0122] 11. After completing the step 10, adding 200 μl of Proteinase K buffer and 50 μl of proteinase K (the product of the Takara company with the catalog number of 9034; and the final concentration in the reaction system is 0.12 U/l) into the cell pellet obtained in the step 10, performing uniform mixing, incubating in the ThermoMixer at 37° C. for 60 minutes and then 56° C. for 15 minutes. After the reaction is completed, let the sample cool to room temperature, and then adding 750 μl of Trizol LS (the product of the Thermo Fisher company with the catalog number of 10296028), incubating at room temperature for 5 minutes after mixing by pipetting up and down, adding 220 μl of chloroform, shaking vigorously for 15 seconds, and incubating at room temperature for 3 minutes. Centrifuge the tube at 13000 rpm for 15 minutes at 4° C., transferring the supernatant into a 1.5 ml eppendorf centrifuge tube, adding 500 μl of isopropanol and 1 μl of glycoblue (the concentration is 15 μg/μl), mixing by pipetting up and down, and placing the centrifuge tube in a −20° C. refrigerator to precipitate for overnight.

[0123] 12. After completing the step 11, centrifuging the sample obtained in the step 11 at 13000 rpm for 20 minutes at 4° C., discarding the supernatant, adding 500 μl of 75% ethanol, washing the pellet, centrifuging the tube at 13000 rpm for 5 minutes at 4° C., repeating this step once, drying the pellet, adding 20 μl of DEPC water to dissolve the pellet, taking 1 μl of sample and quantifying with NanoDrop.

[0124] 13. After completing the step 12, taking out 20 μg of total RNA from the sample obtained in the step 12, adding 10 μl of 10×RQ1 DNase I buffer (the product of the Promega company), 3 μl of RNAsin (the product of the Thermo Fisher company with the catalog number of E00381) and 5 μl of DNase I (the product of the Promega company with the catalog number of M6101), making up to 100 μl with DEPC water, incubating at 37° C. for 20 minutes in ThermoMixer, after the reaction is completed, adding 100 μl of DEPC water, then adding 200 μl of acid-phenol:chloroform (pH 4.5), mixing by pipetting up and down, incubating at room temperature for 3 minutes, centrifuging the tube at 13000 rpm for 15 minutes at 4° C., transferring the supernatant into a 1.5 ml eppendorf centrifuge tube, adding 20 μl of 3M sodium acetate (pH 5.5), 1 μl of glycoblue and 500 μl of 100% ethanol, mixing by pipetting up and down, and placing the centrifuge tube in the −20° C. refrigerator to precipitate for overnight.

[0125] 14. After completing the step 13, centrifuging the sample obtained in the step 13 at 13000 rpm for 20 minutes at 4° C., discarding the supernatant, adding 500 μl of 75% ethanol, washing the pellet, centrifuging the tube at 13000 rpm for 5 minutes at 4° C., repeating this step once, drying the pellet, adding 6 μl of DEPC water to dissolve the pellet, and transferring the sample into the PCR tube.

[0126] 15. After completing the step 14, adding 10 μl of rRNA probe mix (2 μg/l) (the design and synthesis of the probe sequence refer to published reference (Adiconis, X., Borges-Rivera, D., Satija, R., DeLuca, DS, Busby, M A, Berlin, A M, Sivachenko, A., Thompson, D A, Wysoker, A., Fennell, T., et al. (2013). Comparative analysis of RN A sequencing methods for degraded or low-input samples. Nature methods 10, 623-629.), and 4 μl of 5× hybridization buffer into the sample obtained in the step 14, mixing by pipetting up and down, placing the PCR tube in a PCR thermal cycler, setting the reaction program as follows: incubating at 95° C. for 2 minutes, cooling to 22° C. at a speed of 0.1° C./s, incubating at 22° C. for 5 minutes, and immediately putting the sample on ice after the reaction is completed.

[0127] 16. After completing the step 15, adding 3 μl of 10×RNase H buffer (the product of the Thermo Fisher company), 5 μl of RNase H (the product of the Thermo Fisher company with the catalog number of EN0202) (25U) and 2 μl of DEPC water into the sample obtained in the step 15, mixing by pipetting up and down, placing the sample in the PCR thermal cycler, and setting the reaction program as follows: incubating at 37° C. for 30 minutes, and immediately putting the sample on ice after the reaction is completed.

[0128] 17. After completing the step 16, adding 4 μl of 10×TURBO buffer (the product of the Thermo Fisher company), 5 μl of TURBO DNase (the product of the Thermo Fisher company with the catalog number of AM2238; and the final concentration in the reaction system is 0.25 U/l), and 1 μl of DEPC water into the sample obtained in the step 16, mixing by pipetting up and down, placing the sample in the PCR thermal cycler, and setting the reaction program as follows: incubating at 37° C. for 30 minutes, and immediately putting the sample on ice after the reaction is completed.

[0129] 18. After completing the step 17, transferring the sample obtained in the step 17 into a 1.5 ml eppendorf centrifuge tube, adding 160 μl of DEPC water, adding 200 μl of acid-phenol:chloroform (pH 4.5), mixing by pipetting up and down, incubating at room temperature for 3 minutes, centrifuging the tube at 13000 rpm for 15 minutes at 4° C., transferring the supernatant into a 1.5 ml eppendorf centrifuge tube, adding 20 μl of 3M sodium acetate (pH 5.5), 1 μl of glycoblue and 500 μl of 100% ethanol, mixing by pipetting up and down, and placing the centrifuge tube in the −20° C. refrigerator to precipitate for overnight.

[0130] 19. After completing the step 18, centrifuging the sample obtained in the step 18 at 13000 rpm for 20 minutes at 4° C., discarding the supernatant, adding 500 μl of 75% ethanol, washing the pellet, performing centrifugation at 4° C. and 13000 rpm for 5 minutes, repeating this step once, drying the precipitate naturally, adding 16 μl of DEPC water to dissolve the precipitate, transferring the sample into a PCR tube, adding 4 μl of 5× first-strand buffer (the product of the Thermo Fisher company with the catalog number of 18064-014), performing uniform mixing, placing the sample in the PCR thermal cycler, incubating at 94° C. for 5 minutes, and immediately putting the sample on ice after the reaction is completed.

[0131] 20. Taking a 1.5 ml centrifuge tube, adding 20 μl of C1 magnetic beads, placing the centrifuge tube on a magnetic stand, removing the supernatant after the solution is clear, adding 20 μl of solution A, resuspending the magnetic beads, incubating at room temperature for 2 minutes, putting the centrifuge tube on the magnetic stand, removing the supernatant after the solution is clear, repeating this step once, adding 20 μl of solution B, resuspending the magnetic beads, putting the centrifuge tube on the magnetic stand, removing the supernatant after the solution is clear, adding 32 μl of yeast RNA (the product of the Roche company with the catalog number of 10109223001) (50 kg), 68 μl of DEPC water and 100 μl of 2×TWB solution, resuspending the magnetic beads, putting the centrifuge tube on a rotary mixer, uniformly mixing for 1 hour by rotating, then putting the centrifuge tube on the magnetic stand, removing the supernatant after the solution is clear, adding 500 μl of 1×TWB solution, resuspending the magnetic beads, putting the centrifuge tube on the magnetic stand, removing the supernatant after the solution is clear, and repeating this step twice.

[0132] 21. Taking the sample obtained in the step 19, adding 30 μl of DEPC water and 50 μl of 2×TWB buffer, adding a total of 100 μl sample into the blocked magnetic beads, performing uniform mixing for 30 minutes by rotating at room temperature, putting the centrifuge tube on the magnetic stand, removing the supernatant after the solution is clear, performing washing for 4 times with 500 μl of 1×TWB buffer of each time.

[0133] 22. After completing the step 21, adding 100 μl of PK buffer into the washed magnetic beads obtained in the step 21, performing uniform mixing, incubating the sample at 95° C. for 10 minutes at 1000 rpm in the ThermoMixer, putting the centrifuge tube on the magnetic stand, transferring the supernatant into a new 1.5 ml centrifuge tube after the solution is clear, adding 100 μl of PK buffer into the original tube, performing uniform mixing, incubating the sample at 95° C. for 10 minutes at 1000 rpm in the ThermoMixer, putting the centrifuge tube on the magnetic stand, transferring the supernatant into the same 1.5 ml centrifuge tube after the solution is clear, adding 100 μl of PK buffer into the original tube, performing uniform mixing, putting the centrifuge tube on the magnetic stand, transferring the supernatant into the same 1.5 ml centrifuge tube after the solution is clear. For 300 μl of eluent in total, adding 300 μl of acid-phenol:chloroform (pH 4.5), performing uniform mixing, incubating at room temperature for 3 minutes, centrifuging the tube at 13000 rpm for 15 minutes at 4° C., transferring the supernatant into a new 1.5 ml centrifuge tube, adding 18 μl of 5M NaCl, mixing by pipetting up and down, adding 1 μl of glycoblue and 900 μl of 100% ethanol, mixing by pipetting up and down, and placing the centrifuge tube in the −20° C. refrigerator to precipitate for overnight.

[0134] 23. After completing the step 22, centrifuging the sample obtained in the step 22 at 13000 rpm for 20 minutes at 4° C., discarding the supernatant, adding 500 μl of 75% ethanol, washing the pellet, centrifuging the tube at 13000 rpm for 5 minutes at 4° C., repeating this step once, drying the precipitate naturally, adding 10 μl of DEPC water to dissolve the pellet, transferring the sample into a PCR tube, adding 0.5 μl of N6 primer (the sequence is NNNNNN, wherein N represents A or T or C or G) (0.1 g/l), mixing by pipetting up and down, placing the PCR tube in the PCR thermal cycler, incubating at 65° C. for 5 minutes, and immediately putting the sample on ice after the reaction is completed.

[0135] 24. After completing the step 23, adding 3 μl of 5×first-strand buffer (the product of the Thermo Fisher company with the catalog number of 18064-014), 1 μl of dNTP mix (10 mM), 0.5 μl of 100 mM DTT, 0.5 μl of RNase Inhibitor (40 U/l), and 0.5 μl of Superscript II (the product of the Thermo Fisher company with the catalog number of 18064-014) (200 U/l) into the sample obtained in the step 23, mixing by pipetting up and down, putting the PCR tube in the PCR thermal cycler, and setting the program as follows: incubating at 25° C. for 10 minutes, incubating at 42° C. for 40 minutes, and incubating at 70° C. for 15 minutes. After the reaction is completed, put the sample on ice.

[0136] 25. After completing the step 24, transferring the sample obtained in the step 24 into a new 1.5 ml centrifuge tube, adding 10 μl of 5× Second-strand buffer (the product of the Thermo Fisher company with the catalog number of 10812-014), 0.8 μl of dNTP (dUTP) (25 mM) (i.e., a mixture of 25 mM dNTPs and dUTP, wherein the molar ratio of dUTP to dTTP is 4:1), 0.2 μl of RNase H (the product of the Thermo Fisher company with the catalog number of EN0202) (5 U/l), and 2.5 μl of DNA Pol I (the product of the Enzymatics company with the catalog number of P705-500) (10 U/l), putting the centrifuge tube in the ThermoMixer, and setting the reaction program as: 16° C. for 2 hours with intermittent mixing at 300 rpm every 3 min for 15 sec.

[0137] 26. After completing the step 25, uniformly mixing and balancing XP magnetic beads in advance at room temperature for 30 minutes, then adding 90 μl of (1.8×) XP magnetic beads into the reaction solution obtained in the step 25, and performing gentle mixing. Incubating at room temperature for 5 minutes, transferring the solution to the magnetic stand and stewing for 5 minutes, removing the supernatant, and washing the magnetic beads twice with 200 μl of fresh 80% ethanol solution each time. Placing the magnetic beads on the magnetic stand to dry for 2 minutes, adding 43 μl of TE buffer to resuspend the magnetic beads, and mixing by pipetting up and down for 50 times. Incubating at room temperature for 5 minutes, then putting the magnetic beads on the magnetic stand and stewing for 5 minutes, and transferring the supernatant (42 μl) into the 1.5 ml centrifuge tube.

[0138] 27. After completing the step 26, adding 5 μl of 10×PNK solution (T4 PNK supporting reaction solution), 0.4 μl of dNTPs (25 mM), 1.2 μl of T4 DNA polymerase (the product of the Enzymatics company with the catalog number of P7080L), (3 U/l), 0.2 μl of Klenow fragment (the product of the Enzymatics company with the catalog number of P7060L) (5 U/l), and 1.2 μl of T4 PNK (the product of the Enzymatics company with the catalog number of Y9040L) (10 U/l) into the sample obtained in the step 26, mixing by pipetting up and down, and incubating for 30 minutes at 20° C. in the ThermoMixer. After the reaction is completed, adding 90 μl of XP magnetic beads for purification, the specific steps are the same as the step 26, finally, using 20.5 μl of TE buffer for elution, transferring the supernatant (19.7 μl) into the new 1.5 ml centrifuge tube.

[0139] 28. After completing the step 27, adding 2.3 μl of 10×blue buffer (the product of the Enzymatics company with the catalog number of B0110L), 0.5 μl of dATP (5 mM) and 0.5 μl of Klenow exo-(3′ to 5′ exo minus) (the product of the Enzymatics company with the catalog number of P7010-LC-L) (5 U/l), mixing by pipetting up and down, putting the centrifuge tube, incubating for 30 minutes at 37° C. in the ThermoMixer.

[0140] 29. After completing the step 28, adding 1.4 of μl 2×Rapid ligation buffer (the product of the Enzymatics company with the catalog number of B1010L), 0.1 μl of mM ATP, 1 μl of Adapter (PEI Adapter oligo A:/5Phos/GATCGGAAGAGCACACGTCT (5Phos: 5′ phosphorylation), PEI Adapter oligo B: ACACTCTTTCCCTAC ACGACGCTCTTCCGATCT, the adapter in the reaction is formed by annealing two oligos) (2 μM), and 1 μl of T4 DNA ligase (Rapid) (the product of the Enzymatics company with the catalog number of L6030-HC-L) (600 U/l) into the sample obtained in the step 28, mixing by pipetting up and down, incubating for 15 minutes at 20° C. in the ThermoMixer. After the reaction is completed, adding 47.7 μl of XP magnetic beads for purification, the specific steps are the same as the step 26, finally, using 26 μl of TE buffer for elution, transferring the supernatant (25 μl) into the new 1.5 ml centrifuge tube. Adding 45 μl of XP magnetic beads for secondary purification, the specific steps are the same as the step 26, finally, using 16.5 μl of TE buffer for elution, transferring the supernatant (15.7 μl) into the PCR tube.

[0141] 30. After completing the step 29, using the supernatant obtained in the step 29 as the template to perform PCR reaction in the PCR tube to obtain the PCR reaction solution (25 μl).

[0142] The PCR reaction system is 25 μl: 15.7 μl of supernatant, 2.5 μl of 10×Pfx buffer (Invitrogen), 1 μl of 10 μM forward and 1 μl of 10 μM reverse primers, 1 μl of 50 mM MgSO.sub.4 solution, 0.4 μl of 25 mM dNTP, 0.4 μl of Pfx Enzyme (Invitrogen), and 3 μl of USER enzyme (NEB).

[0143] The PCR reaction procedure is as follows: incubating at 37° C. for 15 minutes; incubating at 94° C. for 2 minutes; denaturing at 94° C. for 15 seconds, annealing at 62° C. for 30 seconds, extending at 72° C. for 30 seconds, and the reaction is carried out for 12 cycles; and incubating at 72° C. for 10 minutes.

[0144] 31. After completing the step 30, performing electrophoresis on the PCR reaction solution obtained in the step 30 by using 2% agarose gel, recovering the product within a range of 200-450 bp by using the Qiagen MinElute Gel Extraction Kit according to the manufacture's instruction, and finally using 16 μl of TE buffer for elution to obtain PCR eluate.

[0145] 32. After completing the step 31, taking 1 μl of the PCR eluate obtained in the step 31 and quantifying it by using Qubit 3.0. The qualified sample is used for sequencing analysis.

Embodiment 2. Application of the Preparation Method of the RIC-Seq Library

[0146] 1. Culture of HeLa Cells and Drosophila S2 Cell Samples

[0147] The HeLa cells cultured in a laboratory are used as samples, the initial amount of the cell samples is 1×10′ cells, and the Drosophila S2 cells are used as spike-in to evaluate the specificity of the proximal ligation.

[0148] 2. Preparation of the RIC-Seq Library

[0149] The RIC-seq library is constructed according to the method in Embodiment 1 based on the cell sample in the step 1. The forward and reverse primers in the step 30 are as follows (NNNNNNN is a library Index sequence)

[0150] Primer1.0

TABLE-US-00001 (SEQ ID No. 1) 5′-AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACG CTCTTCCGATCT-3′;

[0151] Index primer

TABLE-US-00002 (SEQ ID No. 2) 5′-CAAGCAGAAGACGGCATACGAGATANNNNNNNGTGACTGGAGTTCA GACGTGTGCTCTTCCGATCT-3′.

[0152] Wherein, N represents A or T or C or G.

[0153] 3. Sequencing

[0154] PE150 paired-end sequencing is performed on the RIC-seq library constructed in the step 2 by using an Illumina HiSeq X Ten sequencer.

[0155] 4. Data Analysis and Results

[0156] 1. Data Analysis Method

[0157] The data analysis process is shown as ‘A’ panel in FIG. 2. First, using the Trimmomatic (0.36) software to filter out adapter sequences and low-quality sequencing fragments in the RIC-seq raw sequencing data, then using Cutadapt (v1.15) to trim low-complexity sequences such as polyA after further removing duplicated fragments, then using STAR (2.5.2b) to align high-quality data to the human reference genome (hg19 version), and finally screening sequencing fragments from RNA ligation products (defined as chimeric reads) from the alignment results. By comparing the number of chimeric reads of each gene, the Pearson correlation coefficient is calculated to evaluate the reproducibility of the experiment. The IGVtools and Juicebox are used to visualize RIC-seq data.

[0158] 2. Data Analysis Results

[0159] In order to capture the protein-mediated RNA proximal ligation, we invented the RIC-seq method (RNA In situ Conformation Sequencing). The specific process is shown as ‘A’ panel in FIG. 1. First, the cells are treated with formaldehyde to fix the protein-RNA, protein-DNA and protein-protein interactions, so that different RNA fragments that are close in space are fixed. Secondly, the cell membrane and the nuclear membrane of the fixed cells are permeabilized by using multiple groups of detergents, and treated with MNase to remove the free RNAs that are not protected by protein. After treatment with MNase, the 3′ end of the RNA is the phosphate group, and the 5′ end is the hydroxyl group (FIG. 1A). In order to label pCp-biotin, we use alkaline phosphatase to convert the phosphate group at the 3′ end into hydroxyl group, and then use T4 RNA ligase to label the pCp-biotin to the 3′ end of the RNA. Immediately afterwards, the samples are treated with the alkaline phosphatase and the T4 PNK enzyme, respectively to convert the 3′ end of Cp-biotin into the hydroxyl group and the 5′ end of the RNA into the phosphate group (FIG. 1A). After that, under in situ and non-denaturing conditions, the T4 RNA ligase is used to ligate RNAs that are spatially close to each other. Then, the total RNA is extracted by using proteinase K digestion combined with TRIzol extraction. After removing the genomic DNA and rRNA, RNA fragmentation treatment are performed. Finally, using the streptavidin magnetic beads to enrich the chimeric RNA containing C-biotin, and constructing the strand-specific library according to conventional procedures (Levin, J Z, Yassour, M., Adiconis, X., Nusbaum, C., Thompson, D A, Friedman, N., Gnirke, A., and Regev, A. (2010). Comprehensive comparative analysis of strand-specific RNA sequencing methods. Nature methods 7, 709-715.), then sequencing and analyzing the data.

[0160] We construct two RIC-seq libraries in the HeLa cells and obtain a total of 155 M (million) mappable reads. In order to facilitate data analysis and visualization, we integrate a variety of algorithms and software to establish the complete analysis process (FIG. 2 A). The chimeric reads account is about 9% of all sequencing fragments, and more than 90% of the chimeric reads contain an extra “C” at the junction, which indicates the high efficiency of pCp-biotin labelling and the high specificity of streptomycin magnetic beads enrichment (FIG. 1). Each RIC-seq chimeric read represents the proximal interaction between two different RNA fragments (FIG. 1 C-G). Multiple RIC-seq chimeric reads can reveal common structures or specific RNA trans interactions. The reproducibility of the RIC-seq technology is very good. The Pearson correlation coefficient between two biological replicates is 0.963 (FIG. 2B). In order to determine the false positive rate, we adopt the cell mixing strategy (Li, X., Zhou, B., Chen, L., Gou, L T, Li, H., and Fu, X D (2017). GRID-seq reveals the global RNA-chromatin interactome. Nature biotechnology 35, 940-950.), that is, the HeLa cells and the Drosophila S2 cells are mixed together at a ratio of 1:5, and then constructing the RIC-seq library and sequencing. The result shows that only about 0.6% of chimeric reads are derived from cross-species ligation of RNAs from Drosophila and human HeLa cells (FIG. 2 C), indicating that the false positive rate of the RIC-seq technology is less than 1%.

[0161] Next, we symmetrically detect the resolution, sensitivity and specificity of the RIC-seq method by comparing with known RNA structures and interactions. These RNAs include microRNA, snRNA, snoRNA and lncRNA (FIG. 1 C-I, FIG. 2 D-F). RIC-seq can accurately capture the classic stem-loop structure of miRNA precursor at the single-base resolution (FIG. 1 C, the pCp insertion position is below), and the expression level of these miRNAs (RPM, reads per million) ranges from 0.05 to 31,067 (FIG. 1 D), indicating the broad detection range of the RIC-seq technology. Unexpectedly, the position labeled by pCp-biotin is mainly enriched on the apical loop of the precursor miRNA (FIG. 1 C, 1D), indicating that the apical loop may be rarely protected by protein. Furthermore, RIC-seq successfully detects the known intermolecular and intramolecular interactions of snRNA, snoRNA, RPPH1 (the RNA component of Ribonuclease P) and TERC (telomerase RNA, data are not shown) (FIG. 1 E-I and FIG. 2 D-F). Compared with the PARIS and RAP methods (Engreitz, J. M., Sirokman, K., McDonel, P., Shishkin, A. A., Surka, C., Russell, P., Grossman, S. R., Chow, A. Y., Guttman, M., Lander, E. S. (2014). RNA-RNA interactions enable specific targeting of noncoding RNAs to nascent Pre-mRNAs and chromatin sites. Cell 159(1):188-199.); (Lu, Z., Zhang, Q. C., Lee, B., Flynn, R. A., Smith, M. A., Robinson, J. T., Davidovich, C., Gooding, A. R., Goodrich, K. J., Mattick, J. S., et al. (2016). RNA Duplex Map in Living Cells Reveals Higher-Order Transcriptome Structure. Cell 165, 1267-1279.), RIC-seq can not only capture the known U1-MALAT1 interactions, but also can identify some specific U1 and MALAT1 interaction sites in the HeLa cells (FIG. 1 H). As expected, these interaction sites are conservative, contain U1 motifs, and are supported by chimeric reads (marked by arrows in FIG. 1 I), suggesting potential functionality.

[0162] After fully verifying the RIC-seq method and data, we merge the data from two biological replicates and use Juicebox to create the interaction matrix genome-wide (Durand, N.C., Robinson, J. T., Shamim, M. S., Machol, I., Mesirov, J. P., Lander, E. S., and Aiden, E. L. (2016). Juicebox Provides a Visualization System for Hi-C Contact Maps with Unlimited Zoom. Cell systems 3, 99-101.), and visualize the paired interactions in the matrix by a two-dimensional heat map (IGV/Juicebox), and the intensity indicates the frequency of chimeric RNA ligation (FIG. 1 J and FIG. 2 A). Compared with the -pCp control (with very few chimeric reads are detected), the +pCp library contains complex intramolecular (˜7 M) and intermolecular interactions (˜6 M) (FIG. 2 G), indicating that the RNA is not only highly structured in the cell, but also has extensive entanglement (FIG. 1 J). Interestingly, some lncRNAs have extensive binding on all chromosomes, such as NEAT1 and MALAT1. In order to identify the true binding sites of these high-abundance RNAs, we perform cluster analysis on the chimeric reads, and identify 0.74 M high-confidence RNA-RNA interaction sites in the HeLa cells. Among these sites, MALAT1 and NEAT1 can not only interact with each other, but also have thousands of other targets (FIG. 1 J). Consistent with recent reports (West, J. A., Davis, C. P., Sunwoo, H., Simon, M. D., Sadreyev, R. I., Wang, P. I., Tolstorukov, M. Y., and Kingston, R. E. (2014). The long noncoding RNAs NEAT1 and MALAT1 bind active chromatin sites. Molecular cell, 55, 791-802.), we also find that MALAT1 and NEAT1 are more inclined to bind to transcriptionally active genes (FIG. 2 H, p<2.2e-16), and their binding motifs are highly similar (FIG. 2 I).

[0163] RIC-seq reveals that MALAT1 can bind to the 5′ end of NEAT1 (NEAT1_5′, FIG. 1 J, right). To verify these interactions, we use single-molecule in situ hybridization (smFISH) and super-resolution imaging microscope (SIM) to detect the interaction of these two lncRNAs. We find that NEAT_5′ can form a loop structure, while MALAT1 has a dot-like distribution (each HeLa nucleus has 248 dots on average) (FIG. 1 K and FIG. 2 J). Some NEAT1 and MALAT1 fluorescence signals directly overlap (FIG. 1 K box 2-4). We find that a HeLa cell has 7.5 paraspeckles on average, of which ˜63.7% are co-localized with MALAT1 (FIG. 2 K). In summary, the above data show that RIC-seq is a new method for identifying in situ RNA-RNA interactions with high specificity, high reproducibility and high precision.

[0164] In order to check whether RIC-seq can capture the higher-order structure of RNA, we compare the RNA proximity information detected by RIC-seq with the data obtained by the cryo-electron microscope structure of human 80S ribosomes (Anger, A. M., Armache, J. P., Berninghausen, O., Habeck, M., Subklewe, M., Wilson, D. N., and Beckmann, R. (2013). Structures of the human and Drosophila 80S ribosome. Nature, 497, 80-85.). First, we draw a physical interaction map of 28S rRNA based on the relative spatial distance of each pairwise 5-nt windows (FIG. 3 A). We also draw a 3D map of 28S rRNA based on RIC-seq data (FIG. 3 B). The two maps are highly similar at both high and low resolutions (FIG. 3 A, B). Unexpectedly, RIC-seq can not only capture WC base-pairing (FIG. 3 B, box 1 and box 2), but can also detect long-range loop-loop interactions, for example, the interaction between 50-200 nt and 4300-4400 nt of 28S rRNA as shown in FIG. 3 B (box 3). We find that ˜70% of the non-WC base-pairing interactions in the 28S rRNA structure can be detected by RIC-seq (FIG. 3 A, B). These data show that RIC-seq can faithfully capture the 3D structure information of RNA.

[0165] In order to quantify the performance of RIC-seq in detecting the higher-order structure of RNA, we generate two data sets based on the cryo-electron microscope structure data of 28S rRNA: a true positive set (the 3D distance between the regions corresponding to the pairwise 5-nt windows is less than 25 Å) and a true negative set (the distance is greater than 25 Å) (4,847 vs 369,698) (FIG. 3 C). The two data sets are used to evaluate the sensitivity of RIC-seq (true positive is successfully detected) and the specificity (true negative is successfully excluded). We compare the proximal interactions of 28S rRNA detected by RIC-seq with the true positive data set and true negative data set, and generate an ROC curve. The AUC value obtained by ROC analysis is 0.89, indicating that RIC-seq has high accuracy in the identification of the RNA higher-order structure (FIG. 3 D, black line). As a control, we use the same data sets to evaluate the performance of PARIS. Regrettably, because a large number of 28S rRNA pairing regions and long-range interaction sites cannot be captured by PARIS, a complete curve cannot be obtained, nor the AUC value can be generated (FIG. 3 D, orange line).

[0166] The high-quality RNA-RNA intramolecular interaction data generated by the RIC-seq technology allow us to detect the RNA folding principle in vivo. For this purpose, we focus on 5179 precursor mRNAs, which contain at least 100 RNA intramolecular ligation events. Interestingly, we find that there are many independent topological regions in the intron and exon regions of mRNA (FIG. 4 A), their common feature is that there is unusually complex RNA-RNA interaction in a certain interval, for example, in the RNA of PDE3A and IMMP2L precursors. In order to systematically identify similar topological structure regions, we invent an iterative algorithm that can identify the boundaries of the topological regions by maximizing the ratio of the RIC-seq density within and between the regions. Compared with the mRNA precursors, similar topological regions are also evident in initially transcribed lncRNAs such as FTX and PVT1 (FIG. 4 A). These data indicate that RNA is highly structured in vivo and in specific regions, and the RNA co-transcription processing may occur in independent topological regions.

[0167] The observed topological regions indicate that large RNA molecules may form complex local structures while transcribing, and then form specific higher-order structures according to a hierarchical folding pathway. However, the specific folding principle of the RNA in vivo are currently unclear. Similar to DNA polymers, RNA polymers can also exist in the form of random coils, equilibrium globules or fractal globules. The specific conformation of RNA can be deduced by calculating the ligation probability between RNA fragments at different nucleotide distances (Fudenberg, G., and Mirny, L. A. (2012). Higher-order chromatin structure: bridging physics and biology. Current opinion in genetics & development 22, 115-124.).

[0168] By using RIC-seq data and similar simulation methods (Lieberman-Aiden, E., van Berkum, N. L., Williams, L., Imakaev, M., Ragoczy, T., Telling, A., Amit, I., Lajoie, B. R., Sabo, P. J., Dorschner, M. O., et al. (2009). Comprehensive mapping of long-range interactions reveals folding principles of the human genome. Science 326, 289-293.), we detect the correlation between the contact probability (or ligation frequency) of any two fragments in the same RNA molecule and its linear distance. In the physical characteristics of polymers, if the polymer exists in a random coil state, the contact probability between two loci will rapidly decay as the linear distance increases, and the slope of the curve is expected to be −3/2 (Fudenberg, G., and Mirny, L. A. (2012). Higher-order chromatin structure: bridging physics and biology. Current opinion in genetics & development 22, 115-124.). Conversely, if the polymer exists in the form of an equilibrium globule, the contact probability will first decrease at a rate similar to that of the random coil, but then reaches equilibrium, and finally, the ligation frequency becomes independent of the linear distance (Fudenberg, G., and Mirny, L. A. (2012). Higher-order chromatin structure: bridging physics and biology. Current opinion in genetics & development 22, 115-124.). However, the random coil and equilibrium globule models do not seem to conform to the actual laws obtained by RIC-seq. Because regardless of whether introns are counted, RIC-seq data show that the contact probability between different RNA fragments gradually decreases with the increase of distance, and the slope of the curve is close to −1 (FIG. 4 B and C). This scaling agreed well with a fractal globule model (Lieberman-Aiden, E., van Berkum, N. L., Williams, L., Imakaev, M., Ragoczy, T., Telling, A., Amit, I., Lajoie, B. R., Sabo, P. J., Dorschner, M. O., et al. (2009). Comprehensive mapping of long-range interactions reveals folding principles of the human genome. Science 326, 289-293.). Therefore, based on the above data, we believe that precursor RNA may be folded in a fractal sphere conformation similar to genomic DNA in vivo. This conformation can ensure that the RNA lacks knots while maintaining the maximal packaging, and keeping the ability that the RNA can easily unfold and refold its local structures.

[0169] We next examine the conformation of mature mRNA by using chimeric reads that derived only from exons and untranslated regions. By polymer modeling, we find that the folding of the mature mRNA is power-law dependent, and the slope of the curve is also close to −1 (FIG. 4 C), suggesting that the mature mRNA is also compressed into a fractal globule state. Of note, intron-lacking lncRNAs, such as NEAT1 and MALAT1, are folded similarly to pre-mRNA and intron-containing lncRNAs (FIG. 4 C). In summary, these results indicate that both mRNA and ncRNA may follow a fractal globule to form complex 3D structures.

[0170] Highly structured RNA needs to interact with other RNAs to exert regulatory functions. To explore new intermolecular interaction features, we generated RNA 3D maps in diverse cell lines, including human neural progenitor cells (hNPC) and the colon adenocarcinoma cell line HT29. In addition, three common ENCODE cell lines, including the human lymphoblastoid cell line GM12878, H1 human embryonic stem cells (hESCs) and human fetal lung fibroblasts IMR-90, were selected for the potential integration of publicly available genomic data. We performed RIC-seq in these cell lines and generated 1,001 million unique reads after removing duplicates. The chimeric reads constituted 8.4% of all mappable reads. As expected, the RNA-RNA interactions in these five new cell types were also extremely complicated (FIG. 5 A).

[0171] By using these high-quality data, we identify a total of about 3 M cell-type-specific interacting clusters (fragment cutoff=2) and a large number of constitutive interaction sites in six different cell types (FIG. 5 B). As exemplified by LncPRESS2 (FIG. 4A, bottom), a P53-responsive and embryonic stem cell-specific lncRNA (Jain, A. K., Xi, Y., McCarthy, R., Allton, K., Akdemir, K. C., Patel, L. R., Aronow, B., Lin, C., Li, W., Yang, L., et al. (2016). LncPRESS1 Is a p53-Regulated LncRNA that Safeguards Pluripotency by Disrupting SIRT6-Mediated De-acetylation of Histone H3K56. Molecular cell 64, 967-981.), RIC-seq detects extensive interactions between LncPRESS2 and its neighboring gene GRID2 in H1 hESC (FIG. 5 C). In contrast, in the ChrX: 73.1 Mb-73.6 Mb locus, we observe a constitutive interaction between lncRNA FTX and JPX (FIG. 5 B and D). It is known that these two lncRNAs play a positive regulatory role in XIST expression (Carmona, S., Lin, B., Chou, T., Arroyo, K., and Sun, S. (2018). LncRNA Jpx induces Xist expression in mice using both trans and cis mechanisms. PLoS genetics 14, e1007378; Chureau, C., Chantalat, S., Romito, A., Galvani, A., Duret, L., Avner, P., and Rougeulle, C. (2011). Ftx is a non-coding RNA which affects Xist expression and chromatin structure within the X-inactivation center region. Human molecular genetics 20, 705-718.; Sun, S., Del Rosario, B. C., Szanto, A., Ogawa, Y., Jeon, Y., and Lee, J. T. (2013). Jpx RNA activates Xist by evicting CTCF. Cell 153, 1537-1551.; Tian, D., Sun, S., and Lee, J. T. (2010). The long noncoding RNA, Jpx, is a molecular switch for X chromosome inactivation. Cell 143, 390-403.), and have been shown critical for XIST-mediated X-chromosome silencing. The interaction between these two lncRNAs suggests that they may function as a complex to regulate XIST. These cell type-specific and constitutive interactions further highlight the specificity of the RIC-seq method and show that RIC-seq can be used to identify lncRNA targets in living cells.

[0172] In order to further verify the function of the LncPRESS2-GRID2 interaction, we adopt the Cas9-KRAB-mediated lncRNA silencing strategy (Gilbert, L. A., Larson, M. H., Morsut, L., Liu, Z., Brar, G. A., Torres, S. E., Stern-Ginossar, N., Brandman, O., Whitehead, E. H., Doudna, J. A., et al. (2013). CRISPR-mediated modular RNA-guided regulation of transcription in eukaryotes. Cell 154, 442-451.), wherein sgRNA can specifically target Cas9-KRAB directly to a promoter region of lncRNA, where KRAB functions as a transcription inhibitor of RNA polymerase II (FIG. 5 E), thus efficiently block the transcription of the specific lncRNA. Upon the depletion of LncPRESS2 with three different sgRNAs, the GRID2 levels were dramatically decreased (FIG. 5 F), indicating that the LncPRESS2 can positively regulate the expression of GRID2 even though the linear distance is greater than 25 kb. Surprisingly, in LncPRESS2 knockdown cells, the expression of a stem cell pluripotency key factor OCT4 is significantly decreased (FIG. 5 F), which means that the LncPRESS2 mediated GRID2 regulation may be closely related to stemness. The above data show that the RIC-seq technology can indeed identify functional lncRNA targets.

[0173] In order to reveal the general characteristics of RNA-RNA interactions in different cell types, we first calculate the frequency of intrachromosomal and interchromosomal interactions respectively. By using RIC-seq data generated in the above six cell types, we find that ˜.sub.70% of RNA-RNA interactions occur within the same chromosome, while the remaining ˜30% occur between different chromosomes (FIG. 6 A). As RNA-RNA interactions can occur both in cis and in trans, we also calculate the frequency of intragene and intergene interactions respectively. Similarly, about 60% of chimeric reads shows cis-interaction within the genes, and can be used to deduce the RNA 3D architecture; the remaining 40% exhibits trans-RNA-RNA interaction properties (FIG. 6 B), which indicates that a large number of RNAs in the cells can span a long distance to interact with other RNAs in the same chromosome or different chromosomes. If we only count the chimeric reads in the same chromosome, this trend is also obvious, wherein we detect two obvious peaks: the first peak corresponds to the intragene interaction, which can span hundreds of nucleotides; and the second peak corresponds to the intergene interaction, which spans a distance of more than 1 Mb (FIG. 6 C).

[0174] Chromatin is heavily packed in vivo and organized into compartments A and B (Lieberman-Aiden, E., van Berkum, N. L., Williams, L., Imakaev, M., Ragoczy, T., Telling, A., Amit, I., Lajoie, B. R., Sabo, P. J., Dorschner, M. O., et al. (2009). Comprehensive mapping of long-range interactions reveals folding principles of the human genome. Science 326, 289-293.), which are represented by different transcriptional activities. Similar to the organization of chromatin, RNA interactions also seem to be compartmentalized and can largely recapitulate the compartment of DNA (FIG. 6 D), which indicates that the RNAs in the same compartment may be more tend to interact with each other due to spatial proximity. We next quantify the interactions between different compartments. Interestingly, among intragene interactions, chimeric reads are mainly enriched in the same compartment, and A to A can account for ˜90% of the total chimeric reads (FIG. 6 E), which may be caused by the active transcription of the compartment A and the spatial distances within these genes are relatively closer. In contrast, for intergene interactions, the interaction of compartments A to A decreases to about 65%, but the interaction of the compartments A to B increases to ˜30% (FIG. 6 E), which indicates that such trans RNA interactions may have some unknown functions and may regulate the activity of the genes in the compartment B.

[0175] Since the trans RNA-RNA interactions can span more than 1 Mb and even across different chromosomes, we next sorted the RNA-RNA interactions based on two criteria: the number of target genes and the interaction density determined by normalizing chimeric reads to RNA expression levels. Interestingly, this analysis unexpectedly reveals −500 highly abundant RNA-RNA interaction hubs in the HeLa cells (FIG. 7 A), including well-known lncRNAs such as MALAT1, NEAT1, CCAT1, and PVT1. Unexpectedly, many protein-coding genes also show complex RNA-RNA interactions, such as PDE3A, GPC5, and TRIO (FIG. 7 A). The interaction patterns and genomic locations of MALAT1, CCAT1 and PDE3A are visualized as Circos plots (FIG. 7 B). Since the RNAs transcribed from these loci seem to function as a hub to organize the RNA-RNA interactions on a genome-wide scale, we term these RNAs as hub-RNAs, which include hub-mRNAs and hub-lncRNAs.

[0176] In order to characterize the characteristics of hub-RNA, we divide all RNAs expressed in the HeLa cells into two groups: hub-RNAs and other RNAs. Based on the RIC-seq signals, we find that hub-RNAs have stronger trans RNA-RNA interactions and show significant enrichment at gene bodies (FIG. 7 C). In addition, these hub-RNAs are evolutionarily more conserved than the other RNAs (FIG. 7 D), and are also actively transcribed. The RNA polymerase II is also enriched in the TSS (transcription starting site) regions of these genes (FIG. 7 E). Correspondingly, the occupancy of the active histone markers H3K4me3 (FIG. 7 F) and H3K27ac (FIG. 7 G) is slightly higher. In contrast, repressive histone marker H3K27me3 has a slightly lower signal (FIG. 7 H). At the same time, we find that the hub-RNAs are cell-type-specific (FIG. 7 I). Therefore, RIC-seq unexpectedly reveals a group of tissue-specific hub-RNAs that may play an important role in gene regulation.

[0177] In order to investigate the roles of the hub-RNAs, we chose CCAT1 for further analysis because of its extensive trans-RNA interactions (FIG. 7 B) and potential super-enhancer activity (Hnisz, D., Abraham, B. J., Lee, T. I., Lau, A., Saint-Andre, V., Sigova, A. A., Hoke, H. A., and Young, R. A. (2013). Super-enhancers in the control of cell identity and disease. Cell 155, 934-947.; Loven, J., Hoke, H. A., Lin, C. Y., Lau, A., Orlando, D. A., Vakoc, C. R., Bradner, J. E., Lee, T. I., and Young, R. A. (2013). Selective inhibition of tumor oncogenes by disruption of super-enhancers. Cell 153, 320-334.). CCAT1 localizes to the human 8q24 gene desert and is abnormally highly expressed in a variety of cancers such as colorectal cancer, prostate cancer and hepatocellular cancer (Chen, H., He, Y., Hou, Y. S., Chen, D. Q., He, S. L., Cao, Y. F., and Wu, X. M. (2018a). Long non-coding RNA CCAT1 promotes the migration and invasion of prostate cancer PC-3 cells. European review for medical and pharmacological sciences 22, 2991-2996.; Deng, L., Yang, S. B., Xu, F. F., and Zhang, J. H. (2015). Long noncoding RNA CCAT1 promotes hepatocellular carcinoma progression by functioning as let-7 sponge. Journal of experimental & clinical cancer research: CR 34, 18.; Tseng, Y. Y., Moriarity, B. S., Gong, W., Akiyama, R., Tiwari, A., Kawakami, H., Ronning, P., Reuland, B., Guenther, K., Beadnell, T. C., et al. (2014). PVT1 dependence in cancer with MYC copy-number increase. Nature 512, 82-86.; Xiang, J. F., Yin, Q. F., Chen, T., Zhang, Y., Zhang, X. O., Wu, Z., Zhang, S., Wang, H. B., Ge, J., Lu, X., et al. (2014). Human colorectal cancer-specific CCAT1-L lncRNA regulates long-range chromatin interactions at the MYC locus. Cell research 24, 513-531.). In colorectal cancer cells, researchers report a CCAT1 transcript with an additionally extended 3′ end, and find that the transcript can regulate spatial interactions between the promoter and enhancer of the MYC gene (Xiang, J. F., Yin, Q. F., Chen, T., Zhang, Y., Zhang, X. O., Wu, Z., Zhang, S., Wang, H. B., Ge, J., Lu, X., et al. (2014). Human colorectal cancer-specific CCAT1-L lncRNA regulates long-range chromatin interactions at the MYC locus. Cell research 24, 513-531.), but the precise mechanism is unknown.

[0178] Because the CCAT1 partially overlaps with the reported super-enhancer (Khan, A., and Zhang, X. (2016). dbSUPER: a database of super-enhancers in mouse and human genome. Nucleic acids research 44, D164-171.), by examining the RIC-seq and RNA-seq data in the HeLa cells, we unexpectedly find that CCAT1 may have a transcript with an additionally extended 5′ end, instead of the additionally extended 3′ end previously reported in colon cancer (FIG. 8 A). The 5′ and 3′ RACE mapping results in the HeLa cells also confirm the additionally 5′ extension (FIG. 8 A, marked in brown at the bottom), which directly overlaps with a super-enhancer and has a total length of −4,700 nt. We refer this lncRNA as CCAT1-5L. Northern blotting with the Exon2 (E2) probe (see FIG. 8 A) confirms that CCAT1 is expressed in different cell types, while the 5L-specific probe detects CCAT1-5L only in HeLa cells (FIG. 8 B), indicating that CCAT1-5L may be a specific transcript in cervical cancer, which is further confirmed in the RNA-seq data from cervical cancer patients (data are not shown).

[0179] smFISH shows that CCAT1-5L is a nuclear-retained lncRNA and forms 2-3 spots in each nucleus (FIG. 8 C). CCAT1-5L seems to be functional, because most of the RNA interactions from CCAT1 to other regions of the 8q24 “gene desert” were found to originate from the first exon and an additional 5′ end extension region (FIG. 8 A). In addition, we detect extensive long-range RNA-RNA interactions between CCAT1-5L, MYC promoter-RNA and PVT1. More importantly, CCAT1-5L binding sites observed in the PVT1 locus are mainly located in the intron regions containing the MYC enhancers (FIG. 8 A, orange line). The above data suggest that CCAT1-5L may function as a super-enhancer RNA that interacts with the promoter and enhancer RNAs to regulate the expression of the MYC oncogene.

[0180] We next explore whether CCAT1-5L can directly regulate the MYC expression. When CCAT1-5L is knocked down by two LNA oligos targeting the 5′ end extension region (FIG. 8 D), the RNA level of MYC is significantly reduced by ˜40% (FIG. 8 D), indicating that CCAT1-5L can indeed regulate the MYC expression in the HeLa cells. Unexpectedly, the expression level of PVT1, a positive regulator of MYC (Tseng, Y. Y., Moriarity, B. S., Gong, W., Akiyama, R., Tiwari, A., Kawakami, H., Ronning, P., Reuland, B., Guenther, K., Beadnell, T. C., et al. (2014). PVT1 dependence in cancer with MYC copy-number increase. Nature 512, 82-86.), is also greatly reduced (FIG. 8 D). Therefore, we hypothesize that CCAT1-5L hub-lncRNA may synergize with the promoter and enhancer RNAs of MYC to coordinate their expression levels.

[0181] To test this hypothesis, we first check whether CCAT1-5L, MYC promoter and MYC enhancer RNAs are colocalized in vivo. To this end, we first synthesize smFISH probes, respectively targeting the CCAT1-5L part detected by RIC-seq, the first exon and first intron of MYC, and the enhancer located in the PVT1 intron. The results show that the three RNAs show perfect colocalization (FIG. 8 E). This further confirms the regulating role of CCAT1-5L on MYC and PVT1. In addition, after the CCAT1-5L is knocked down in the LNA oligonucleotide, the co-localization pattern between MYC promoter and enhancer RNAs does not change (FIG. 8E).

[0182] Since CCAT1-5L is highly expressed in cervical cancer patients, we next check whether CCAT1-5L can promote cell proliferation and metastasis, which are two landmarks of cancer (Hanahan, D., and Weinberg, R A (2011). Hallmarks of cancer: the next generation. Cell 144, 646-674.). In contrast to the LNA controls, knockdown of CCAT1-5L by the 5L-specific LNA oligos in HeLa cells significantly reduces the proliferation rate (FIG. 8 F); and on the contrary, ectopic expression of CCAT1-5L with a lentiviral plasmid can significantly enhance cell proliferation (FIG. 8 F), which is consistent with the oncogenic effect of CCAT1-5L. The colony formation assay further confirms the effect of CCAT1-S on cell proliferation (FIG. 8 G). In order to detect whether CCAT1-5L can affect cell metastasis and invasion, we perform a cell invasion assay using a transwell insert and find that knockdown of CCAT1-5L significantly reduces the metastasis ability of HeLa cells, while the overexpression of CCAT1-5L significantly increases invasion and metastasis (FIG. 8 H). In summary, these data indicate that CCAT1-5L hub-lncRNA can directly regulate the MYC expression to promote tumorigenesis.

INDUSTRIAL APPLICATION

[0183] The method for capturing an RNA in situ higher-order structure and interaction provided by the present invention can process the intracellular RNA in situ without destroying the cell structure and maintaining the integrity of the cell, and capture RNA intramolecular and intermolecular interactions in the physiological state. The method for capturing RNA in situ higher-order structure and interaction provided by the present invention uses pCp-biotin to label RNA ends, and performs in situ ligation under non-denaturing conditions, thereby greatly improving the labeling efficiency and reducing the non-specific ligation between molecules; and chimeric RNAs labeled with C-biotin are enriched by C1 magnetic beads for constructing a library, so that the chimeric RNA can be efficiently enriched, the fraction of usable data is increased, and the sequencing cost is reduced.