Method for labeling RNA molecules associated with epigenetic modification of chromatin in living cells

Abstract

The invention provides a method for marking RNA (Ribonucleic Acid) molecules related to epigenetic modification of chromatin in living cells in real time. According to the method, module protein for recognizing epigenetic modification, engineered ascorbic acid peroxidase and a biotin-aniline probe are introduced into cells. The engineered ascorbic acid peroxidase can catalyze the oxidation reaction of hydrogen peroxide on a biotin-aniline substrate, so that the biotin-aniline substrate generates a high-reaction-activity intermediate with oxygen free radicals, covalent bonds are formed between the intermediate and RNA molecules around epigenetic modification of chromatin, and further RNA molecules with biotin labels are generated. And finally, purifying the RNA molecule with the biotin label by using streptavidin magnetic beads for subsequent sequencing analysis so as to determine the RNA variety related to the epigenetic modification of the chromatin.

Description

Cross-references to related applications

[0001] This application claims priority to U.S. Patent Application Serial No. 18 / 650,149, filed April 30, 2024, the entire contents of which are incorporated herein by reference. Reference sequence list

[0002] The submitted file "P3141CN00_Sequence List.xml" is a sequence list file in ST.26xml format, with a file size of 2.6KB. It was created on October 14, 2024, submitted on October 15, 2024, and is incorporated into this paper by full reference. Technical Field

[0003] This invention relates to the field of molecular biology. More specifically, this invention relates to the identification of RNA molecules in living cells that are adjacent to chromatin epigenetic modifications. Background Technology

[0004] Transcriptional regulation is a key mechanism determining cellular function and plasticity, primarily relying on the complex interactions between genetic factors encoded within the genome and epigenetic factors. Among these factors, post-translational modifications (PTMs) of histones have a profound impact on chromatin structure and gene expression. Histones are core components of chromatin, serving as the scaffold of DNA. Post-transcriptional modifications, such as lysine methylation and acetylation, are closely related to transcriptional activity. For example, histone modifications, including monomethylation of the 4th lysine residue of histone H3 (H3K4me1) and acetylation of the 27th lysine residue of histone H3 (H3K27ac), typically occur abundantly at enhancers of transcriptionally active genes, promoting chromatin interactions and transcription factor binding. Therefore, enzymes that perform epigenetic modifications and demodifications of histones, such as histone lysine methyltransferases (KMTs) and histone lysine acetyltransferases (HATs), are crucial for cellular homeostasis, development, and disease progression.

[0005] However, the regulatory mechanisms of epigenetic modification remain largely unclear. Increasing evidence suggests that noncoding RNAs (ncRNAs) play a crucial role in epigenetic regulation and can be considered epigenetic regulatory factors. Long noncoding RNAs (lncRNAs) play a vital mediating role in decoding chromatin epigenetic features. These lncRNAs interact with histone-binding proteins (often referred to as "histone writers," "readers," and "erasers"), mediating histone-modifying enzymes to modify chromatin at specific genomic sites, thereby altering chromatin openness. Therefore, dysregulation of these long non-coding RNAs is often associated with serious diseases. For example, the interaction between the long non-coding RNA ST3Gal6-AS1 and MLL1 can induce the generation of the H3K4me3 marker on the ST3Gal6 promoter. In the progression of colorectal cancer, dysregulation of ST3Gal6-AS1 and ST3Gal6 expression activates the PI3K / Akt signaling pathway, driving tumor cell proliferation. Although long non-coding RNAs are very important as epigenetic regulators, only a small fraction of the vast number of long non-coding RNAs in the human genome have been identified and characterized with specific functions. This indicates that research progress on the regulation of epigenetics by long non-coding RNAs is slow, increasing the difficulty of their potential applications as diagnostic or prognostic biomarkers and therapeutic targets, highlighting the necessity for further research on long non-coding RNAs.

[0006] Currently, methods for detecting non-coding RNAs associated with specific chromatin modifications include ChRIP-seq, CARIP-seq, PIRCh-seq, and RT&Tag. The first three methods share similar principles, while RT&Tag operates differently. The main difference lies in the cross-linking process. CARIP uses 1% formaldehyde to cross-link RNA chromatin, with the reaction lasting only one minute, favoring milder cross-linking conditions. Therefore, it reduces the number of sonication cycles, aiming to preserve more RNA-protein interactions. Conversely, ChRIP-seq, after treating with 1% formaldehyde for 10 minutes, additionally uses ultraviolet light for cross-linking and intensifies sonication, indicating that the interaction between RNA and chromatin is maintained through double cross-linking, but lacks a systematic exploration of the effects of cross-linking. Notably, the inventors of PIRCh-seq experimented with different cross-linking agents and ultimately chose 1% glutaraldehyde as a substitute for formaldehyde, claiming an improved ability to capture RNA-chromatin conjugates. Despite the various methods described above, there is still no consensus on the optimal cross-linking method, highlighting the necessity of eliminating the cross-linking step. Furthermore, a major limitation of these methods is their over-reliance on high-quality antibodies, such as ChIP-grade antibodies. Although specific antibodies for histone post-transcriptional modification are standard reagents, different batches of antibodies can exhibit significant variations in specificity and affinity, thus substantially affecting the reliability of the data.

[0007] Therefore, there remains an urgent need in the art for a method that can effectively identify RNA associated with various chromatin modifications in living cells, without relying on antibodies and cross-linking. This invention addresses this need. Summary of the Invention

[0008] The purpose of this invention is to provide a method or reagent kit for solving the above-mentioned technical problems.

[0009] According to a first aspect of the present invention, a method is provided for labeling RNA molecules surrounding epigenetic modifications of chromatin in living cells. The method includes the following steps: introducing a modular protein that recognizes epigenetic modifications, an engineered ascorbate peroxidase, and a biotin-aniline substrate into living cells; oxidizing the biotin-aniline substrate with hydrogen peroxide via the engineered ascorbate peroxidase, thereby causing a covalent bond to form between the biotin-aniline substrate and a neighboring RNA molecule to obtain a biotin-labeled RNA molecule; and further sequencing the biotin-labeled RNA molecule.

[0010] According to one embodiment of the present invention, the module protein that recognizes epigenetic modifications can locate the site of chromatin with epigenetic modifications and recruit engineered ascorbate peroxidase to that site.

[0011] According to one embodiment of the invention, when the engineered ascorbate peroxidase is exposed to hydrogen peroxide, it oxidizes the biotin-aniline substrate.

[0012] According to one embodiment of the present invention, the module protein that identifies epigenetic modifications is an evolutionarily conserved natural protein.

[0013] According to one embodiment of the present invention, the module protein that recognizes epigenetic modifications includes a chromatin-binding domain from mouse CBX7 or from Drosophila polycomb protein that recognizes a chromatin-binding domain from mouse CBX1 that recognizes a lysine residue trimethylated at position 27 of histone H3, and a plant homologous domain from mouse TAF3 that recognizes a lysine residue trimethylated at position 4 of histone H3.

[0014] According to one embodiment of the present invention, the module protein that recognizes epigenetic modifications is further fused with multiple repeating peptides, while the engineered ascorbate peroxidase is fused with a single-chain variable fragment (scFv).

[0015] According to one embodiment of the present invention, the repeating peptide is a GCN4 protein having the amino acid sequence as described in SEQ ID NO:01; and the single-chain variable fragment has the amino acid sequence as described in SEQ ID NO:02.

[0016] According to one embodiment of the invention, the single-chain variable fragment is a recognizable repeating peptide and recruits the engineered ascorbate peroxidase.

[0017] According to a second aspect of the invention, a reagent kit is provided for real-time labeling of RNA molecules associated with epigenetically modified chromatin in living cells. More specifically, the reagent kit includes a modular protein that recognizes epigenetic modifications, an engineered ascorbate peroxidase, and a biotin-aniline substrate.

[0018] According to one embodiment of the present invention, the components of the reagent kit can be delivered directly into living cells.

[0019] According to one embodiment of the present invention, when the engineered ascorbate peroxidase is exposed to hydrogen peroxide, the engineered ascorbate peroxidase can catalyze the oxidation of the biotin-aniline substrate by hydrogen peroxide.

[0020] According to one embodiment of the invention, the oxidized biotin-aniline substrate forms a covalent bond with an epigenetically modified chromatin-associated RNA molecule to produce a biotin-labeled RNA molecule.

[0021] According to one embodiment of the present invention, the biotin-labeled RNA molecule can be further purified.

[0022] According to one embodiment of the present invention, the module protein that identifies epigenetic modifications can locate the position of the chromatin with epigenetic modifications and guide the engineered ascorbate peroxidase to that position.

[0023] According to one embodiment of the present invention, the module protein that identifies epigenetic modifications is an evolutionarily conserved protein derived from a natural protein.

[0024] According to one embodiment of the present invention, the module protein that recognizes epigenetic modifications includes a chromatin-binding domain that recognizes a 27th lysine residue of histone H3 trimethylated from mouse CBX7 or from Drosophila polycomb protein, a chromatin-binding domain that recognizes a 9th lysine residue of histone H3 trimethylated from mouse CBX1, and a plant homologous domain that recognizes a 4th lysine residue of histone H3 trimethylated from mouse TAF3.

[0025] According to one embodiment of the present invention, the repeating peptide is a GCN4 protein having the amino acid sequence as described in SEQ ID NO:01; and the single-chain variable fragment has the amino acid sequence as described in SEQ ID NO:02.

[0026] According to one embodiment of the invention, the single-chain variable fragment is a recognizable repeating peptide and recruits the engineered ascorbate peroxidase. Attached Figure Description

[0027] Referring to the accompanying drawings, the following single-chain variable fragments can identify the repeating peptides, wherein:

[0028] Figure 1A and 1B An embodiment of the method of the present invention is summarized, wherein Figure 1A The principle of this method is described. Figure 1B This is a scatter plot showing the transcriptome profile (FPKM) of cells transfected with mCBX7 (the chromatin-binding domain of mouse CBX7) recognition protein and untreated HEK293T cells.

[0029] Figure 2A-2U This indicates the results of using a method according to an embodiment of the present invention, identifying H3K27me3-associated RNA molecules using mCBX7 or dPC (chromatin-binding domain of Drosophila polycomb protein) recognition proteins; wherein Figure 2A This represents a volcano plot showing all RNAs identified using mCBX7 that are associated with H3K27me3 modification. Figure 2BA volcano plot showing non-coding RNAs associated with H3K27me3 modification identified using mCBX7; Figure 2C A three-dimensional pie chart showing the number and percentage of coding and non-coding genes identified by mCBX7; Figure 2D This is a box plot showing the expression levels of RNA identified or not detected by the mCBX7 (left) and dPC (right) recognition proteins (FPKM). Figure 2E This represents a volcano diagram showing all RNAs identified by dPC as being associated with H3K27me3 modification; Figure 2F This represents a volcano plot showing non-coding RNAs associated with H3K27me3 modification identified by dPC; Figure 2G It is a three-dimensional pie chart showing the number and percentage of coding and non-coding genes identified by dPC recognition proteins; Figure 2H A circular diagram showing the proportions of exons, introns, and genes in the test results; Figure 2I A circular diagram showing the proportions of exons, introns, and genes in the detection results obtained using dPC to identify proteins; Figure 2J This indicates the Western blot results of H3K27me3 protein in the starting sample and the sample purified by streptavidin; Figure 2K A Venn diagram showing the comparison of non-coding RNAs associated with H3K27me3 identified by mCBX7 and dPC; Figure 2L and Figure 2M These represent the results obtained via RIP-qPCR using H3K27me3 antibody ( Figure 2L ) and EZH2 ( Figure 2M The RNA enriched by the antibody was used to verify candidate genes; Figure 2N and Figure 2O These represent the results obtained via RIP-qPCR using H3K27me3 antibody ( Figure 2N ) and EZH2 ( Figure 2O The RNA enriched by the antibody was used to identify candidate genes in HCT116 cells; Figure 2P The results represent the knockdown of long non-coding RNAs (SETD5-AS1 and LINC00641) in HEK293T cells; Figure 2Q The distribution diagram (top) and heatmap (bottom) of the ChIP-seq signal of the H3K27me3 peak after LINC00641 or SETD5-AS1 is knocked down; Figure 2R The results of Western blotting of H3K27me3 and H3 in total protein of cells knocked down with antisense oligonucleotides (ASOs) to SETD5-AS1 (ΔSETD5-AS1) and LINC00641 (ΔLINC00641) are presented. Figure 2SThe distribution of H3K27me3 in the nuclei of cells with knockdown of SETD5-AS1 (ΔSETD5-AS1) and LINC00641 (ΔLINC00641) was observed using laser confocal microscopy, with the upper right corner showing a magnified image of the local area. Figure 2T Genome browser screenshots showing the signal distribution of ChIP-seq (top) and RNA-seq (bottom) in samples representing negative control (NC), LINC00641 knockdown (ΔLINC00641), and SETD5-AS1 knockdown (ΔSETD5-AS2); and Figure 2U A cross-sectional view showing the ChIP signals of H3K27me3, H3K9me3, and H3K4me3 in a 1-kb region upstream and downstream of the genome (between transcription start and transcription termination points), where the shaded area represents the standard error;

[0030] Figure 3A-3Q This represents the results of identifying H3K9me3-associated RNA molecules using the mCBX1 (chromatin-binding domain of mouse CBX1) recognition protein, according to an embodiment of the present invention. Figure 3A A volcano plot showing non-coding RNAs associated with H3K9me3 modification identified using mCBX1; Figure 3B This indicates the Western blot results of H3K9me3 protein in the starting sample and the sample purified by streptavidin; Figure 3C A volcano diagram showing all RNAs associated with H3K9me3 modification; Figure 3D A three-dimensional pie chart showing the number and percentage of coding and non-coding genes; Figure 3E It is a pie chart representing the proportions of exons, introns, and characteristic regions between genes in the read results; Figure 3F This indicates that candidate genes were validated using RNA enriched with H3K9me3 antibody via RIP-qPCR. Figure 3G This indicates that the RNA enriched by the H3K9me3 antibody was verified in HCT116 cells by RIP-qPCR. Figure 3H This indicates that the RNA enriched by the SUV39H1 antibody was verified by RIP-qPCR; Figure 3I This indicates that the RNA enriched by the SUV39H1 antibody was verified in HCT116 cells by RIP-qPCR. Figure 3J The distribution diagram (top) and heatmap (bottom) of the ChIP-seq signal after knocking down the H3K9me3 peak of LINC00662; Figure 3K The results represent the knockdown of long non-coding RNA (LINC00662) in HEK293T cells; Figure 3LThe protein expression levels of H3K27me3 and H3 were detected after knocking down LINC00662 (Δ662) with antisense oligonucleotides (ASOs). Figure 3M The image shows the distribution of H3K9me3 in the nuclei of cells with knocked-down LINC00662 (ΔLINC00662) using laser confocal microscopy. The upper right corner shows a magnified image of the local area. Figure 3N Genome browser screenshots showing the signal distribution of ChIP-seq (top) and RNA-seq (bottom) for negative control (NC) and knockdown of LINC00662 (ΔLINC00662); Figure 3O A profile of ChIP signals for H3K9me3, H3K27me3, and H3K4me3 in a 1-kb region upstream and downstream of the genome (between transcription start and transcription termination points), where the shaded area represents the standard error; Figure 3P This is a box plot, representing the expression levels of RNA identified or not detected by the mCBX1 recognition protein (FPKM). Figure 3Q This is a Venn diagram representing the common regulators of H3K27me3 and H3K9me3; and

[0031] Figure 4A-4L This indicates a comparison between the present invention and other methods, wherein... Figure 4A The diagram shows a volcano plot illustrating the identification of all detected RNAs associated with H3K4me3 modification using the mTAF3 (the plant homology domain of mouse TAF3) recognition protein identified by the method of this invention. Figure 4B A volcano plot representing the detected non-coding RNA; Figure 4C A three-dimensional pie chart showing the number and percentage of coding and non-coding genes identified by the mTAF3 recognition protein using the method of the present invention. Figure 4D A pie chart showing the proportion of exons, introns, and intergenic regions in the mTAF3 recognition protein readout results; Figure 4E A cross-sectional view of the ChIP signal of H3K4me3 in a 1-kb region upstream and downstream of the genome (between transcription start and transcription termination points), where the shaded area represents the standard error; Figure 4F A Venn diagram illustrating the use of mTAF3 to recognize the protein and the same non-coding RNA identified by the PIRCh-seq method with the H3K4me3 antibody. Figure 4G Box plot showing the expression of RNA identified by the mTAF3 recognition protein using the method of the present invention; Figure 4H Box plot representing the expression of RNA identified by PIRCh-seq; Figure 4IThis indicates that the expression of genes identified by the TAF3 recognition protein using the method of the present invention was verified in H9 cells by RIP-qPCR, and these genes were found by both the method of the present invention and PIRCh-seq. Figure 4J This indicates that the expression of genes identified by the mTAF3 recognition protein using the method of the present invention was verified in HEK293T cells by RIP-qPCR. More specifically, these genes were found by both the method of the present invention and PIRCh-seq. Figure 4K This indicates that the expression of the gene identified by the TAF3 recognition protein using the method of the present invention was verified in H9 cells by RIP-qPCR; and Figure 4L This indicates that the expression of the gene identified by the TAF3 recognition protein using the method of the present invention was verified in HEK293T cells by RIP-qPCR. Detailed Implementation

[0032] In the following description, methods and / or reagent kits for real-time labeling of RNA molecules associated with chromatin having epigenetic modifications in living cells are listed as preferred embodiments. It will be apparent to those skilled in the art that modifications, including additions and / or substitutions, can be made without departing from the scope and spirit of the invention. Specific details may be omitted below to avoid obscuring the invention; however, this document is prepared to enable those skilled in the art to practice the techniques disclosed herein without excessive experimentation.

[0033] The term "H3K4me3" used in this article refers to the epigenetic modification of the DNA packaging protein histone H3, specifically the trimethylation of the 4th lysine residue of histone H3, which is involved in the regulation of gene expression; that is, the addition of three methyl groups (trimethylation) to the 4th lysine residue of the histone H3 protein. Similarly, the term "H3K9me3" refers to the epigenetic modification of the DNA packaging protein histone H3, specifically the trimethylation of the 9th lysine residue of the histone H3 protein, which is usually associated with heterochromatin. The term "H3K27me3" refers to the epigenetic modification of the DNA packaging protein histone H3, specifically the trimethylation of the 27th lysine residue of histone H3.

[0034] According to a first aspect of the invention, a method is provided for real-time identification of RNA molecules associated with chromatin having epigenetic modifications in living cells. The advantage of this method is that it does not involve cross-linking or the use of antibodies.

[0035] This method provides a means of real-time labeling of RNA molecules associated with epigenetically modified chromatin within living cells. First, a modular protein recognizing epigenetic modifications, an engineered ascorbate peroxidase, and a biotin-aniline substrate are introduced into living cells. After cellular uptake, the engineered ascorbate peroxidase oxidizes the biotin-aniline substrate upon exposure to hydrogen peroxide (H₂O₂), thereby promoting the formation of a covalent bond between the substrate and the epigenetically modified chromatin-associated RNA molecule. This process generates biotin-labeled RNA molecules in the cellular environment. Subsequently, the biotin-labeled RNA molecules are purified and isolated from the cellular environment for subsequent sequencing analysis.

[0036] Furthermore, in some embodiments, the modular protein used in this method that recognizes epigenetic modifications can locate the intracellular chromatin with epigenetic modifications, thereby guiding the engineered ascorbate peroxidase to these target sites. The modular protein recognizing epigenetic modifications can be derived from evolutionarily conserved natural proteins and may include a chromatin-binding domain recognizing a 27th lysine residue trimethylated CBX7 from mouse or a Drosophila comb protein, a 9th lysine residue trimethylated CBX1 from mouse, and a plant homologous domain recognizing a 4th lysine residue trimethylated TAF3 from mouse.

[0037] Among them, the amino acid sequence that recognizes the chromatin-binding domain of mouse CBX7 or Drosophila polycomb protein with trimethylation of lysine residue 27 of histone H3 is ELSAIGEQVFAVESIRKKRVRKGKVEYLVKWKGWPPKYSTWEPEEHILDPRLVMAYEEKEE RDRASGYRKRGPKP (SEQ ID NO:03); the amino acid sequence that recognizes the chromatin-binding domain of mouse CBX1 with trimethylation of lysine residue 9 of histone H3 is EEEYVVEKVLDRRVVKGKVEYLLKWKGFSDEDNTWEPEENLDCPDLIAEFLQSQKTAHET DKSEG (SEQ ID NO:04); and the amino acid sequence that recognizes the plant homology domain of mouse TAF3 with trimethylation of lysine residue 4 of histone H3 is SAKAPVRSVVTETVSTYVIRDEWGNQIWICPGCNKPDDGSPMIGCDDCDDWYHWPCVGI MAAPPEEMQWFCPKCANKIKKDKKHKKRKHRAH (SEQ ID NO:05).

[0038] Furthermore, recognizing epigenetic modifiers allows for fusion with repetitive peptides, facilitating the recruitment of engineered ascorbate peroxidase, which itself possesses single-chain variable fragments. By recognizing the fusion of epigenetic modifiers with repetitive peptides, single-chain variable fragments can be effectively identified, thereby targeting engineered ascorbate peroxidase to specific chromatin regions of interest within living cells.

[0039] In some embodiments, the repeating peptide has the following amino acid sequence: EELLSKNYHLENEVARLKKGSGSG (SEQ ID NO:01); and the single-chain variable fragment has the following amino acid sequence: MGPDIVMTQSPSSLSASVGDRVTITCRSSTGAVTTSNYASWVQEKPGKLFKGLIGGTNNRAPGVPSRFSGSLIGDKATLTISSLQPEDFATYFCALWYSNHWVFGQGTKVELKRGGGGSGGGGSGGGGSSGGGSEVKLLESGGGLVQPGGSLKLSCAVSGFSLTDYGVNWVRQAPGRGLEWIGVIWGDGITDYNSALKDRFIISKDNGKNTVYLQMSKVRSDDTALYYCVTGLFDYWGQGTLVTVSS (SEQ ID NO:02).

[0040] According to a second aspect of the invention, a reagent kit is provided for labeling RNA molecules adjacent to chromatin epigenetic modifications in living cells.

[0041] This reagent kit provides a comprehensive solution for real-time labeling of RNA molecules associated with adjacent chromatin epigenetic modifications within living cells. Specifically, the kit includes a modular protein that identifies epigenetic modifications, an engineered ascorbate peroxidase, and a biotin-aniline substrate, enabling researchers to dynamically observe epigenetic processes. Once delivered to living cells, the kit components work synergistically to label relevant RNA molecules in real time. More specifically, the engineered ascorbate peroxidase catalyzes the oxidation of the biotin-aniline substrate upon contact with hydrogen peroxide. This oxidation process forms covalent bonds between the substrate and adjacent RNA molecules associated with chromatin epigenetic modifications, resulting in biotin-labeled RNA molecules.

[0042] Furthermore, modular proteins that recognize epigenetic modifications play a crucial role in guiding engineered ascorbate peroxidase to specific intracellular sites where epigenetic modifications occur. These modular proteins are derived from evolutionarily conserved proteins, including the mouse CBX7 chromatin-binding domain that recognizes a trimethylated lysine residue at histone H3 (position 27) or a Drosophila comb protein; the mouse CBX1 chromatin-binding domain that recognizes a trimethylated lysine residue at histone H3 (position 9); and the mouse TAF3 plant homologous domain that recognizes a trimethylated lysine residue at histone H3 (position 4). Moreover, these modular proteins can be further modified to fuse with repetitive peptides, thereby enhancing the recognition and recruitment of engineered ascorbate peroxidase. This fusion strategy effectively localizes enzyme activity to specific chromatin regions of interest in living cells.

[0043] Furthermore, the biotin-tagged RNA molecule can be further enriched and purified for subsequent sequencing and analysis. By providing a comprehensive reagent kit for labeling RNA molecules adjacent to epigenetically modified chromatin, researchers are able to gain a deeper understanding of dynamic epigenetic processes and their functional implications in living cells.

[0044] In one embodiment, the method is based on the principle of biotinylation of nucleic acids, combining a modular protein that recognizes epigenetic modifications with an engineered ascorbate peroxidase (e.g., APEX2) applied to living cells. APEX2 catalyzes the single-electron oxidation of biotin-phenol (a small, membrane-permeable molecule), forming a covalent bond with a transient (<1 ms) highly reactive free radical on a protein or nucleic acid. Related studies indicate that biotin-aniline (Btn-An) is a substrate of APEX2, but in the presence of hydrogen peroxide, APEX2 is more reactive with RNA molecules than biotin-phenol. The modular proteins used in this invention that recognize epigenetic modifications are all derived from evolutionarily conserved natural proteins and have been fully validated in the art, thus exhibiting high affinity and specificity.

[0045] To improve labeling efficiency, the SunTag system can be further employed, which involves fusing the epigenetic modification-recognizing module protein with 10 repeating peptides, wherein the epitope has the amino acid sequence described in SEQ ID NO:01, and will be recognized by the single-chain variable fragment (with the amino acid sequence described in SEQ ID NO:02) on APEX2. Figure 1A As shown, instead of directly fusing APEX2 to the recognition protein, up to 10 APEX2 molecules can be recruited to a single recognition protein through the binding between repetitive peptides and single-chain variable fragments, significantly increasing the local concentration of the APEX2 enzyme. Figure 1AGiven the high concentration and enzymatic activity of APEX2, RNA labeling takes less than one minute. After Btn-An labeling, RNA extraction and purification can be performed, specifically purification of biotinylated RNA using beads containing streptavidin. To further reduce interference from ribosomal RNA, polyA+ RNA is enriched, and primers with random sequences are used for first-strand cDNA synthesis via reverse transcription, followed by template conversion and PCR amplification to generate a double-stranded DNA library for high-throughput sequencing. To demonstrate efficient RNA purification, ten repeating peptides of SunTag without the module protein recognizing epigenetic modifications were introduced into cells as a control group along with APEX2 containing a single-stranded variable fragment, simultaneously with the experimental group. Results showed that the introduction of these components and the introduction of biotin-aniline did not significantly affect gene expression patterns in cells. Figure 1B ).

[0046] Example

[0047] Example 1: Identification of RNA associated with H3K27me3 modification

[0048] H3K27me3 is one of the most studied epigenetic modifications, containing many repressive regulatory factors. This embodiment primarily uses two known recognition module proteins: the chromatin-binding domain of mouse CBX7 (hereinafter referred to as CBX7), which recognizes the trimethylation of the 27th lysine residue of histone H3, or the chromatin-binding domain (hereinafter referred to as dPC) from Drosophila polycomb protein. First, both recognition module proteins can highly specifically enrich histones modified with H3K27me3 in cells, with CBX7 exhibiting a higher affinity than dPC. Furthermore, this invention aims to reveal the types of RNA associated with chromatin modifications.

[0049] Using the method of this invention, with the CBX7 module as the module protein for recognizing epigenetic modifications, a total of 1437 different RNA species significantly associated with H3K27me3 were identified (log2-conversion fold change ≥0.75, false discovery rate (FDR) <0.05). Figure 2A This includes 542 non-coding RNAs (approximately 38% of all RNAs). Figure 2B and Figure 2CAmong them, several long non-coding RNAs that are recognized as regulating H3K27me3 modification have been identified, including H1921, NEAT122, MALAT123, TINCR24, and SOX2-OT25. To rule out the possibility that the identified RNAs were due to high expression in cells, the results of the examples showed that the expression levels of the RNAs identified by this method were significantly lower than the overall distribution in the cellular transcriptome (Mann-Whitney U test, P < 2.2 x 10⁻⁶). -16 , Figure 2D ).

[0050] In contrast, using dPC as a module protein to identify epigenetic modifications only screened 702 RNAs significantly associated with H3K27me3. Figure 2E Of these, 241 are non-coding RNAs (approximately 34% of all RNAs). Figure 2F and Figure 2G In both experiments, the sequencing results were predominantly exons (68% for CBX7 and 69% for dPC), with a smaller subset consisting of intron sequences (26% and 26%) and intergenic sequences (6% and 5%). Figure 2H and Figure 2I This indicates that the identified molecules are derived solely from RNA, rather than from genomic DNA contamination, and can be considered an important indicator for RNA identification. Compared to CBX7, dPC identifies fewer RNA molecules, likely due to its lower affinity for H3K27me3. Figure 2J Meanwhile, most of the non-coding RNAs identified by dPC (174 / 241) can also be detected by CBX7. Figure 2K ).

[0051] To elucidate the function of non-coding RNAs in regulating epigenetic modifications, we investigated several moderately enriched long non-coding RNAs, such as AC005943.5, SETD5-AS1, LINC00641, and LINC00893. Figure 2B To verify their molecular function, the interaction between these long non-coding RNAs and chromatin with H3K27me3 modification was first confirmed by RIP-qPCR, with a highly expressed long non-coding RNA (BCYRN1) used as a negative control. As expected, in addition to the negative control group (…),… Figure 2LApart from the groups mentioned above, all other groups showed significant correlations. In mammalian cells, the production of H3K27me3 on chromatin is mainly catalyzed by the conserved polycomb repressive complex 2 (PRC2), which includes the EZH1 / 2 catalytic subunits, SUZ12, EED, and RBBP7 / 4. Next, further analysis was conducted to determine whether these long non-coding RNAs assist in recruiting enzymes that catalyze H3K27me3 modification, specifically through RNA immunoprecipitation (RIP) using antibodies specific to the EZH2 catalytic subunits. Two long non-coding RNAs (SETD5-AS1 and LINC00641) were significantly enriched in the RNA precipitated by EZH2. Figure 2M The presence of this information indicates that it participates in the modification of H3K27me3.

[0052] Furthermore, the roles of these long non-coding RNAs in other cell lines were evaluated. For example, RNA immunoprecipitation using H3K27me3 and EZH2 antibodies in the human colorectal cancer cell line (HCT116 cells) showed that SETD5-AS1 and LINC00641 interact with H3K27me3 (… Figure 2N ) and EZH2 ( Figure 2O The combination of these findings further confirms that long non-coding RNAs are conserved to a certain extent in different cell types.

[0053] Next, antisense oligos (ASOs) were used to knock down these long non-coding RNAs in cells to perform ChIP-seq experiments on H3K27me3. Figure 2P It is noteworthy that when the expression of SETD5-AS1 or LINC00641 was reduced, almost all the H3K27me3 peaks found in control cells were also attenuated. Figure 2Q This indicates that the effects of these long non-coding RNAs on H3K27m3 can also lead to changes throughout the entire genome, a result that is also validated by Western blot quantification. Figure 2R ) and immunofluorescence imaging ( Figure 2S Therefore, knocking down SETD5-AS1 or LINC00641 will cause a significant decrease in the signal of H3K27me3.

[0054] To confirm whether this change affects gene transcriptional activity, the region near the PGBD5 gene was further examined using a genome browser screenshot. The results showed that when any long non-coding RNA was deleted, the H3K27me3 signal around the PGBD5 promoter was significantly reduced; simultaneously, the downregulation of H3K27me3 was accompanied by the reactivation of the PGBD5 gene. Figure 2T Previous studies have shown that most RNAs associated with H3K27me3 modification are transcribed from silent genomic sites. In fact, the RNAs identified by this method via H3K27me3 also originate from repressive genomic regions, which, compared to undetected transcripts, exhibit higher H3K27m3 and H3K9me3-ChIP-seq signals and lower H3K4me3-ChIP-seq signals at their transcription starting site (TSS) and gene body. Figure 2U These results indicate that the method of the present invention can identify long non-coding RNAs associated with H3K27me3 signaling, and some of these long non-coding RNAs are crucial for maintaining epigenetic modifications and homeostasis of gene expression.

[0055] Example 2: Identification of RNA associated with H3K9me3 modification

[0056] The present invention uses the recognition module protein of the CBX1 chromatin-binding domain (hereinafter referred to as CBX1) to identify H3K9me3-related RNA. Figure 3A First, immunoprecipitation confirmed that CBX1 has a high affinity for chromatin modified with H3K9me3. Figure 3B Chrom-seq results showed that a total of 1301 RNAs associated with H3K9me3 modification were identified (log2- fold change ≥ 0.75, false discovery rate < 0.05); Figure 3C ), including 626 non-coding RNAs ( Figure 3D Of these, genes encoding proteins account for about half. Figure 3D Similar to H3K27m3 Chrom-seq, the readings of the CBX1 module protein also primarily came from exons (68%), therefore there was no DNA contamination. Figure 3E ).

[0057] In addition to known regulators of H19 and MALAT1, in HEK293T cells ( Figure 3F ) and HCT116 cells ( Figure 3GIn this study, several novel long non-coding RNAs, such as AC004158.2, LINC00662, and RP11-500G22.2, were also validated using RIP-qPCR. Since H3K9me3 modification is primarily catalyzed by a series of methyltransferases containing the SET domain, such as SETDB1, SUV39H1, and SUV39H2, further RIP-qPCR was used to detect the interaction between SUV39H1 and the aforementioned long non-coding RNAs. The results indicated that these long non-coding RNAs significantly interacted with the SUV39H1 enzyme in both cell types. Figure 3H and Figure 3I Among them, LINC00662 showed the most significant performance, so subsequent experiments will focus on LINC00662.

[0058] Compared with the control, H3K9me3-ChIP signaling was reduced overall in cells with knocked-down LINC00662. Figure 3J and Figure 3K However, unlike H3K27me3, no significant change in H3K9me3 abundance was observed by Western blot analysis when LINC00662 was missing from cells. Figure 3L Only a slight reduction was detected by immunofluorescence imaging. Figure 3M This indicates that the effect of this long non-coding RNA is more likely to be site-specific. For example, when LINC00662 is knocked down, the expression of H3K9me3 at the TAF7 promoter and gene body is significantly reduced, thereby promoting the transcription of nearby TAF7 genes. Figure 3N Similar to H3K27me3-related RNAs, H3K9me3-related RNAs are also transcribed from genomic regions, and these regions contain numerous repressive markers such as H3K9me3- and H3K27m3 modifications, while exhibiting lower abundance of H3K4me3 modifications. Figure 3O This result is also consistent with the fact that the expression levels of H3K37me3 and H3K9me3-related RNAs are lower than those in the whole transcriptome. Figure 2D and Figure 3P Notably, when comparing non-coding RNAs associated with the two repressive markers, more than one-third of the non-coding RNAs (309 / 859) were associated with both H3K9me3 and H3K27me3 modifications. Figure 3Q Recent evidence also suggests that these two repressive markers co-occur on some developmentally repressive genes and transposon elements, and that the related proteomic results of these two modifications partially overlap, indicating that these epigenetic modifications may influence each other in regulating gene silencing.

[0059] Example 3: Performance comparison of an RNA identification method according to an embodiment of the present invention with other methods.

[0060] Compared to current methods for identifying epigenetic modification-related RNAs (such as PIRCh-seq, ChRIP-seq, CARIP-seq, or RT&Tag), the main feature of the method and / or reagent kit of this invention is that it no longer relies on the use of antibodies or covalent cross-linking, both of which can introduce significant experimental errors. Furthermore, the method of this invention effectively improves overall efficiency in terms of the number of long non-coding RNAs identified, sample input volume, sequencing depth, and cost (Table 1). RT&Tag is a newly developed method that offers significant improvements over other methods in many aspects, particularly its wide applicability, cross-link-free nature, and low cost. Studies claim that each assay requires only approximately 100,000 cells, more than 50 times fewer than other methods. However, in this study, RT&Tag only tested Drosophila cells, whose genome size (120 Mb) is much smaller (>20 times) than that of mammalian cells (1.6–6.3 Gb). More importantly, this method still uses high-quality antibodies to identify specific histone modifications, thus limiting its applicability.

[0061] Table 1. Comparison of RNA identification method according to an embodiment of the present invention with other methods.

[0062] To ensure fair testing of its applicability in mammalian cells, the RNA identification method of this invention utilizes the plant homology domain of mouse TAF3 (hereinafter referred to as TAF3) to target the trimethylation of the 4th lysine residue of histone H3 (H3K4me3). This histone modification is an epigenetic marker of an active promoter. Results show that the method of this invention identified a total of 3316 different RNAs (log2-transformation fold change ≥0.75, false discovery rate <0.05); Figure 4A This includes 1104 non-coding RNAs. Figure 4B and Figure 4C ), and most of them come from human exons (69.3%); Figure 4D This demonstrates that these RNAs originate from RNA transcripts. Because the active promoter of the H3K4me3 marker is closer to the transcription start site of the highly expressed gene, more RNA was identified than with the other two repressive markers (3316 vs. 1437 and 1301). Furthermore, compared to other genes, enhanced H3K4me3 signaling was associated with TAF3-rich RNAs. Figure 4E This further indicates that these RNAs originate from genes tagged with H3K4me3-.

[0063] Furthermore, the data were compared with those of H3K4me3-related RNAs identified by PIRCh-seq in human embryonic stem cell line H9. PIRCh-seq used a very lenient threshold (log2-fold change in transformation > 0, P ≤ 0.05) to define significantly enriched long non-coding RNAs, but even so, PIRCh-seq detected only 218 non-coding RNAs, far fewer than the 1104 long non-coding RNAs identified by the present invention using a stricter threshold (log2-fold change in transformation ≥ 0.75, FDR < 0.05). Although the two experiments were performed in two different cell lines, the method of the present invention was able to identify one-third (69 / 218) of the non-coding RNAs identified by PIRCh-seq in H9 cells. Figure 4F The 149 H3K4me3-related non-coding RNAs that this method failed to identify are likely due to their extremely low expression levels in HEK293T cells (median FPKM = 0.0009). Figure 4G However, when studying the expression levels of these three types of genes in H9 cells, it was found that some RNAs not detected by PIRCh-seq were moderately or highly expressed (median FPKM = 0.007). Figure 4H The low sensitivity of PIRCh-seq may be the reason for its poor ability to enrich long non-coding RNAs.

[0064] Therefore, we further investigated some interactions in H9 cells using an orthogonal RIP-qPCR method. First, we validated the presence of H3K4me3-related long non-coding RNAs identified by both methods in both HEK293T and H9 cells. The results showed that these long non-coding RNAs exhibited significant interactions in both cell types. Figure 4I and Figure 4J Then, RIP-qPCR was performed on some positive long non-coding RNAs that were identified by the method of the present invention but could not be identified by PIRCh-seq and were highly expressed in H9 cells. The results showed that these long non-coding RNAs did indeed interact with H3K4me3 in HEK293T and H9 cells. Figure 4K and Figure 4L In summary, this invention, besides eliminating the need for antibodies or covalent cross-linking, is superior to other methods in detecting long non-coding RNAs associated with chromatin epigenetic modifications.

[0065] In summary, this invention provides a novel method for identifying RNAs associated with various histone post-transcriptional modifications and verifies the molecular functions of some RNAs.

[0066] The method provided by this invention is superior to commonly used methods such as RIP-seq / CLIP-seq, PIRCh-321-seq / ChRIP-seq, and even the newly developed RT&Tag. This invention requires less cell volume and less sequencing data, thus reducing costs. Furthermore, the method avoids small molecule or UV-mediated cross-linking, allowing it to be performed under near-physiological conditions, significantly reducing non-specific interactions. More importantly, this method does not require ChIP-grade antibodies; instead, it uses physiologically recognized protein domains as target markers, avoiding the quality impact of antibodies and reducing experimental errors that the ENCODE consortium and other scientists are concerned about. On the other hand, considering the millions of years of evolutionary history of precisely controlling epigenetic modifications and gene transcription, both natural and genetically engineered recognition proteins are superior to antibodies in terms of robustness, sensitivity, and specificity. Related results also indicate that, due to the low cross-linking or antibody recognition efficiency of PIRCh-seq, this method recognizes more than 20 times the number of non-coding RNAs associated with H3K4me3-PTM compared to PIRCh-seq. Other advantages of this invention include: 1) the use of protein engineering techniques to obtain recognition protein domains with higher affinity and specificity to identify novel modifications, such as m6A or Ψ, and simultaneous identification through multiple different domains (such as bivalent domains) in a single reaction; furthermore, fluorescent or affinity tags can be added for visualization of the results; 2) lower cost, eliminating the need to purchase ChIP-grade antibodies and reducing sequencing overhead, which provides the possibility of further scaling up analysis in complex biological systems. For example, it can be used to study regulatory RNAs related to various post-transcriptional modifications in a range of tissues or cells under different environments within an organism.

[0067] As used herein, the terms “approximately,” “substantially,” “truly,” and “about” are used to describe and explain a minor variation. When used in conjunction with an event or situation, the term can refer to the exact occurrence of the event or situation, or approximately the occurrence of the event or situation. The term “about,” as used herein, regarding a given value or range, typically refers to a range of ±10%, ±5%, ±1%, or ±0.5% of the given value or interval, which can be understood herein as ranging from one endpoint to another or between two endpoints. Unless otherwise stated, all ranges disclosed herein include endpoints. Furthermore, unless explicitly stated otherwise, the term “a” as used herein should be understood to include one or more; additionally, terms such as “first,” “second,” and “third” are used only as illustrative purposes and are not intended to give numerical meaning or to order them according to their importance.

[0068] The above description is provided for the purpose of illustrating and describing the invention, and is not intended to be exhaustive or to limit the invention to the precise forms disclosed. Many modifications and variations will be more readily understood by those skilled in the art.

[0069] The above embodiments were chosen and described in order to better explain the principles of the present invention and its practical application, thereby enabling other people skilled in the art to understand the various embodiments of the present invention and the various modifications suitable for the intended particular purpose.

Claims

1. A method for labeling RNA molecules surrounding chromatin epigenetic modifications in living cells, characterized by: Introduce chromatin epigenetic modification recognition module proteins, engineered ascorbate peroxidase, and biotin-aniline substrates into living cells; The biotin-aniline substrate is oxidized by the engineered ascorbate peroxidase using hydrogen peroxide to form a covalent bond between the biotin-aniline substrate and the RNA molecules surrounding the chromatin epigenetic modification, thereby obtaining biotin-labeled RNA molecules; and The biotin-labeled RNA molecules were purified for further sequencing.

2. The method of claim 1, wherein the module protein for recognizing epigenetic modifications can locate the position of the chromatin with epigenetic modifications and guide the engineered ascorbate peroxidase to the location.

3. The method according to claim 1, wherein when the engineered ascorbate peroxidase is exposed to hydrogen peroxide, the engineered ascorbate peroxidase can mediate the catalytic oxidation of the biotin-aniline substrate by hydrogen peroxide.

4. The method of claim 1, wherein the module protein for recognizing epigenetic modifications is an evolutionarily conserved natural protein.

5. The method according to claim 1, wherein the module protein that recognizes epigenetic modifications comprises a chromodomain that recognizes a 27th lysine residue of histone H3 trimethylated from mouse CBX7 or from Drosophila polycombgroup proteins, a chromodomain that recognizes a 9th lysine residue of histone H3 trimethylated from mouse CBX1, and a plant homeodomain that recognizes a 4th lysine residue of histone H3 trimethylated from mouse TAF3.

6. The method of claim 1, wherein the recognition epigenetically modified module protein is further fused with a plurality of repeating peptides, and the engineered ascorbate peroxidase is fused with a single-chain variable fragment.

7. The method of claim 1, wherein the repeating peptide is a GCN4 protein having the amino acid sequence as described in SEQ ID NO:01; and the single-chain variable fragment has the amino acid sequence as described in SEQ ID NO:

02.

8. The method of claim 1, wherein the single-chain variable fragment can recognize the repeating peptide GCN4 and recruit the engineered ascorbate peroxidase.

9. A reagent kit for labeling RNA molecules associated with chromatin having epigenetic modifications in living cells. Its characteristics include the following components: Identify modular proteins with epigenetic modifications; Engineered ascorbate peroxidase; as well as Biotin-aniline substrate.

10. The reagent kit of claim 9, wherein the components of the reagent kit can be delivered directly into living cells.

11. The reagent kit of claim 9, wherein the engineered ascorbate peroxidase can oxidize the biotin-aniline substrate when exposed to hydrogen peroxide.

12. The reagent kit of claim 11, wherein the oxidized biotin-aniline substrate forms a covalent bond with an RNA molecule associated with chromatin epigenetic modification to produce a biotin-labeled RNA molecule.

13. The reagent kit of claim 12, wherein the biotin-labeled RNA molecule can be further purified and enriched.

14. The reagent kit of claim 9, wherein the modular protein for recognizing epigenetic modifications can locate the position of the chromatin with the specific epigenetic modification and guide the engineered ascorbate peroxidase to the location.

15. The reagent kit of claim 9, wherein the module protein for recognizing epigenetic modifications is an evolutionarily conserved protein derived from a natural protein.

16. The reagent kit of claim 9, wherein the module protein that recognizes epigenetic modifications comprises a chromatin-binding domain from mouse CBX7 or from Drosophila polycomb protein that recognizes a chromatin-binding domain from mouse CBX1 that recognizes a lysine residue trimethylated at position 27 of histone H3, and a plant homologous domain from mouse TAF3 that recognizes a lysine residue trimethylated at position 4 of histone H3.

17. The reagent kit of claim 9, wherein the recognition epigenetically modified modular protein is further fused with a plurality of repeating peptides, and the engineered ascorbate peroxidase is fused with a single-chain variable fragment.

18. The reagent kit of claim 17, wherein the single-chain variable fragment can recognize the repeating peptide and recruit the engineered ascorbate peroxidase.

19. The reagent kit of claim 18, wherein the repeating peptide is a GCN4 protein having the amino acid sequence as described in SEQ ID NO:01; and the single-chain variable fragment has the amino acid sequence as described in SEQ ID NO:02.

Images