Virus infection and activity inhibiting method

Abstract

To provide a technique for preventing or treating virus infection and infectious disease by inducing mixed tailing to virus RNA, and a method for stabilizing an RNA sequence.SOLUTION: A method for stabilizing RNA in a vertebrate animal includes a step of inserting a stem loop sequence including a pentaloop structure made up of an RNA sequence represented by the following general formula into a target RNA sequence in vitro. [General formula] 5'-CNGGN-3', where the N is each independently selected from adenosine (A), uracil (U), guanine (G) and cytosine (C), the stem loop sequence includes 8 or more of base pair bond, and herein, stabilization of the RNA is brought about by mixed tailing.SELECTED DRAWING: Figure 1a

Description

Detailed Description of the Invention

[0001] [Technical field] This invention was made under the support of the Ministry of Science, ICT and Technology of the Republic of Korea under project number 1711079098, the research management organization for this project is the Institute for Basic Science, the research project name is "Research on cell fate regulation by RNA", the lead organization is the Institute for Basic Science, and the research period is from 2018.01.01. to 2018.12.31.

[0002] This patent application claims priority to Korean Patent Application No. 10-2019-0064129, filed with the Korean Intellectual Property Office on May 30, 2019, the disclosure of which is incorporated herein by reference.

[0003] The present invention relates to a technique for preventing or treating viral infections and infectious diseases by inducing mixed tailing on viral RNA.

[0004] [Background technology] Hepatitis B is a viral disease transmitted extraintestinally by contaminated materials such as blood and blood products, contaminated needles, sexually from infected or carrier mothers, and vertically from their children. According to the World Health Organization, more than 2 billion people are infected worldwide, with approximately 4 million acute cases and 1 million deaths each year, and an estimated 350-400 million chronic carriers (World Health Organization: Geographic Prevalence of Hepatitis B Prevalence, 2004. http: / / www.who.int / vaccines-surveillance / graphics / htmls / hepbprev.htm).

[0005] The virus, HBV, is a double-stranded hepatotropic virus that infects only humans and non-human primates. Viral replication occurs primarily in the liver and, to a lesser extent, in the kidneys, pancreas, bone marrow and spleen (Hepatitis (B virus biology. Microbiol Mol Biol Rev. 64:2000;51-68.) Viral and immune markers are detectable in the blood, and characteristic antigen-antibody patterns evolve over time. The first detectable viral marker is HBsAg, followed by Hepatitis B e antigen (HBeAg) and HBV DNA. Titers can be high during the incubation period, but HBV DNA and HBeAg levels begin to decline at the onset of disease and may be undetectable at the clinical peak of disease (Hepatitis B virus infection natural history and clinical consequences. N Engl J Med..350:2004;1118-1129). HBeAg is a detectable viral marker in blood and is associated with active viral replication, hence high viral load and infectivity (Hepatitis B virus antigen the dangerous end game of hepatitis B virus. N Engl J Med. 347: 2002; 208-210). The presence of anti-HBsAb and anti-HBcAb (IgG) indicates recovery and immunity in already infected individuals.

[0006] According to the American Association for the Study of Liver Diseases (AASLD) and the European Association for the Study of the Liver (EASL), currently recommended treatments for chronic HBV infection include interferon alpha (INFα), pegylated interferon alpha-2a (Peg-IFN2a), entecavir, and tenofovir. The nucleoside and nucleotide therapies, entecavir and tenofovir, have been successful in reducing viral load, but the HBeAg seroconversion and HBsAg loss rates are much lower than those obtained with IFNα therapies. Similar other therapies, including lamivudine (3TC), telbivudine (LdT), and adefovir, are also utilized, but generally with nucleoside / nucleotide therapies, the development of resistance limits therapeutic efficacy.

[0007] Thus, there is a need in the art to discover and develop new antiviral therapeutic agents.

[0008] On the other hand, human cytomegalovirus (HCMV) is a ubiquitous virus belonging to the herpesvirus family. The virus is composed of linear double-stranded deoxyribonucleic acid (DNA) contained in a capsid surrounded by a tegument and surrounded by a lipid bilayer with glycoprotein spikes on its surface. Like other members of this family, HCMV has properties of latency and reactivation. HCMV has the ability to infect and remain latent in many cells.

[0009] In immunocompetent hosts, most HCMV infections are asymptomatic or only mildly symptomatic with a few nonspecific symptoms such as fatigue, malaise, mild fever, lymphadenopathy, hepatomegaly, or slight elevations in liver enzymes. However, heterophil-negative mononucleosis is observed in approximately 10% of previously healthy individuals.

[0010] New strategies for HCMV treatment are also required.

[0011] Summary of the Invention [Problem to be solved by the invention] The inventors have demonstrated that TENT4A / B is involved in mixed tailing of viral RNA, more specifically, mRNA, and helps inhibit the degradation of viral mRNA in host cells. They have also determined that when TENT4A / B expression and activity in a subject host cell is inhibited, mixed tailing can be inhibited, thereby promoting deadenylation of viral RNA and promoting rapid degradation of viral RNA, thereby completing the present invention.

[0012] Accordingly, an object of the present invention is to provide a pharmaceutical composition for preventing or treating a viral infection.

[0013] Another object of the present invention is to provide a method for screening a pharmaceutical composition for preventing or treating a viral infection.

[0014] It is still another object of the present invention to provide a method for producing virus-resistant cells.

[0015] It is yet another object of the present invention to provide a method for stabilizing an RNA sequence.

[0016] [Means for solving the problems] According to one aspect of the present invention, the present invention provides a pharmaceutical composition for preventing or treating a viral infection, comprising an inhibitor of one or more of TENT4A and TENT4B, which are involved in mixed tailing of viral RNA, wherein the inhibitor is an siRNA or shRNA that suppresses the expression of one or more of TENT4A and TENT4B, or an antibody or an antigen-binding fragment thereof that suppresses the activity of one or more of TENT4A and TENT4B.

[0017] The inventors have demonstrated that TENT4A / B is involved in mixed tailing of viral RNA, more specifically, mRNA, and helps inhibit the degradation of viral mRNA in host cells. They have also determined that when TENT4A / B expression and activity in a subject host cell is inhibited, mixed tailing is inhibited, thereby promoting deadenylation of viral RNA and promoting rapid degradation of viral RNA.

[0018] RNA tailing (non-templated nucleotide addition to the 3' end) is the most widespread and conserved type of RNA modification. Noncanonical poly(A) polymerases, also known as terminal nucleotidyltransferases (TENT), integrate nucleotides at the 3' end of posttranscriptional RNA to regulate maturation, stability, and activity. Previously, we developed a method called TAIL-seq to probe the RNA 3' end and measure poly(A) tail length. In TAIL-seq experiments, we discovered noncanonical tailing events such as uridylation and guanylation for vertebrate mRNAs. Terminal uridylyltransferases (TUT4 and TUT7) uridylated the deadenylated poly(A) tail (shorter than ~25 nucleotides) after PABPC removal. U-tails, especially oligo-U-tails, promote mRNA decay. Conversely, guanylation occurs on long poly(A) tails, protecting the mRNA from rapid deadenylation. Two related enzymes, TENT4A (also called PAPD7 or TUT5) and TENT4B (also called PAPD5 or TUT3), extend the mRNA poly(A) tail with intermittent non-adenosine residues (most commonly guanosine), generating 'mixed tails'. These impure poly(A) tails interfere with deadenylation by the CCR4-NOT (CNOT) complex and improve mRNA stability. Mixed tailing has been proposed to act as a novel type of post-transcriptional regulation.

[0019] The term "mixed tailing" as used herein refers to the action of one or more sequences of the poly(A) tail sequence portion of viral RNA, more specifically viral mRNA, being guanylated or uridylated and replaced with other nucleic acid sequences such as guanine or uridine. In the present specification, the above-mentioned "mixed tailing" may be simply described as "mixed tailing", "tailing", etc. When guanylation or uridylation occurs in the poly(A) tail sequence of viral RNA, the deadenylation of the poly(A) tail is suppressed and the degradation of the mRNA is delayed.

[0020] The TENT4A and TENT4B of the present invention are noncanonical poly(A) polymerases belonging to terminal nucleotidyltransferases, and the present inventors have discovered for the first time that TENT4A and TENT4B are involved in mixed tailing of the poly(A) tail of viral mRNA.

[0021] In one embodiment of the present invention, the TENT4A and TENT4B of the present invention are derived from a subject at risk of viral infection or infected with a virus. The present inventors have determined that the viral mRNA infected in the subject may be induced to undergo mixed tailing by the TENT4A and TENT4B derived from the subject.

[0022] In one embodiment of the present invention, the virus of the present invention is a virus in which RNA tailing is induced by one or more of TENT4A and TENT4B. More specifically, the present inventors have determined that a post-transcriptional regulatory element (PRE), more specifically a CNGGN-type pentaloop, plays an important role in TENT4A- and TENT4B-dependent regulation. The present inventors have also determined that both HBV and HCMV act in a similar manner using a pentaloop to recruit TENT4A / B and induce mixed tailing. The HBV and HCMV of the present invention are distinct viruses (Hepadnaviridae and Herpesviridae) with very different life cycles and tissue specificities, but use the same strategy for gene expression advantage, and this convergent evolution and the simplicity of this cis-acting element indicate that this mechanism can be used in other viruses.

[0023] In one embodiment of the invention, the virus of the invention is a virus belonging to the Hepadnaviridae or Herpesviridae families.

[0024] In one embodiment of the present invention, the virus belonging to the Hepadnaviridae family of the present invention is any one selected from the group consisting of Hepatitis B virus, Ground Squirrel Hepatitis B, Woodchuck Hepatitis B, Duck Hepatitis B, and Heron Hepatitis B.

[0025] In one embodiment of the present invention, the virus belonging to the Herpesviridae family of the present invention is any one selected from the group consisting of Iltovirus, Mardivirus, Scutavirus, Simplexvirus, Varicellovirus, Cytomegalovirus, Muromegalovirus, Proboscivirus, Roseolovirus, Lymphocryptovirus, Macavirus, Percavirus, and Rhadinovirus.

[0026] The term "siRNA" used in the present invention means a short double-stranded RNA that can induce the phenomenon of RNAi (RNA interference) by cleavage of a specific mRNA. It is composed of a sense RNA strand having a sequence homologous to the mRNA of a target gene and an antisense RNA strand having a sequence complementary thereto. Since siRNA can suppress the expression of a target gene, it can be provided as an efficient gene knockdown method or a gene therapy method.

[0027] The siRNA is not limited to a double-stranded RNA portion in which RNA pairs are completely paired, and may include a portion that is not paired due to mismatch (corresponding bases are not complementary), bulge (no corresponding base in one strand), etc. The total length is 10 to 100 bases, preferably 15 to 80 bases, and more preferably 20 to 70 bases. The siRNA end structure may be either a blunt end or a cohesive end, as long as it can suppress the expression of the target gene by the RNAi effect. The cohesive end structure may be a structure in which the 3' end is overhanging or a structure in which the 5' end is overhanging. The number of overhanging bases is not limited. In addition, the siRNA may contain, for example, a low molecular weight RNA (for example, a natural RNA molecule such as tRNA, rRNA, or viral RNA, or an artificial RNA molecule) in the overhanging portion of one end, within a range in which the effect of suppressing the expression of the target gene can be maintained. The siRNA end structure does not need to have a cleavage structure on all sides, and may be a step-loop structure in which the terminal sites of a general double-stranded RNA are connected by a linker RNA.

[0028] The siRNA used in the present invention may be in a complete form having polynucleotide pairing by itself, i.e., a form in which the siRNA is directly synthesized in a test tube and introduced into cells through a transformation process, or in a form in which a single-stranded oligonucleotide fragment and its reverse complement are separated by a spacer so as to have such a form after administration in vivo, for example, a siRNA expression vector or a PCR-derived siRNA expression cassette produced so that the siRNA is expressed in cells may be introduced into cells through a transformation or infection process. The method of producing and introducing the siRNA into cells or animals may vary depending on the purpose and the cell biological function of the target gene product.

[0029] The term "shRNA" used in the present invention is intended to overcome the disadvantages of siRNA, such as the high cost of synthesis and the short duration of RNA interference effect due to low cell transfection efficiency. It can be introduced into cells and expressed using adenovirus, lentivirus, and plasmid expression vector systems from the promoter of RNA polymerase III. Such shRNA binds to siRNA processing enzymes (Dicer or It is widely known that these siRNAs are converted by ribosomal RNAi (RNase III) into siRNAs with precise structures and induce silencing of target genes.

[0030] In one embodiment of the present invention, the siRNA of the present invention is exemplified in Table 1, but is not limited thereto.

[0031] The antibody of the present invention that inhibits one or more activities of TENT4A and TENT4B includes all monoclonal antibodies and their corresponding chimeric antibodies, humanized antibodies, and human antibodies, and may include not only novel antibodies but also antibodies already known in the art. The antibody includes functional fragments of antibody molecules as well as intact forms having two full-length heavy chains and two full-length light chains, so long as it specifically binds to TENT4A and / or TENT4B. The functional fragments of antibody molecules refer to fragments that have at least an antigen-binding function, and include Fab, F(ab'), F(ab')2, and Fv.

[0032] The following antibodies are used in the examples of the present invention, but are not limited to these: Anti-TENT4A antibody (invitrogen, PA5-61302), anti-TENT4A antibody (Atlas Antibodies, HPA045487), anti-TENT4A mouse antibody (labmade), anti-TENT4B antibody (invitrogen, PA5-60177), anti-TENT4B antibody (Atlas Antibodies, HPA042968), anti-TENT4B mouse antibody (labmade).

[0033] When the composition of the present invention is used as a pharmaceutical composition, the content of the active ingredient in the composition can be appropriately adjusted depending on the symptoms of the disease, the progression of the symptoms, the condition of the patient, etc., and may be, for example, 0.0001 to 99.9 wt% based on the total composition weight, but is not limited thereto. The content ratio is based on the dry weight after removing the solvent.

[0034] The pharmaceutical composition may further include suitable carriers, excipients and diluents generally used in the manufacture of pharmaceutical compositions, and may be formulated into oral dosage forms such as powders, granules, tablets, capsules, suspensions, emulsions, syrups, aerosols, external preparations, suppositories, or sterile injection solutions by conventional methods for use.

[0035] When the pharmaceutical composition is formulated, it is prepared using a diluent or excipient such as a typical filler, extender, binder, wetting agent, disintegrant, or surfactant. Solid preparations for oral administration include tablets, pills, powders, granules, capsules, etc., and such solid preparations may contain at least one excipient and / or lubricant. Liquid preparations for oral administration include suspensions, solution-resistant agents, emulsions, syrups, etc., and may contain various excipients such as wetting agents, sweeteners, flavoring agents, preservatives, etc. in addition to water and liquid paraffin, which are commonly used simple diluents. Preparations for parenteral administration may include sterilized aqueous solutions, non-aqueous solvents, suspensions, emulsions, freeze-dried preparations, suppositories, etc.

[0036] The preferred dosage of the pharmaceutical composition varies depending on the condition and weight of the patient, the degree of the disease, the drug form, the administration route and period, and may be appropriately selected by those skilled in the art. For more favorable effects, the dosage of the composition of the present invention may be 0.1 mg / kg to 100 mg / kg per day based on the active ingredient, but is not limited thereto. The administration may be once a day or may be divided into several doses. The composition of the present invention may be administered to animals, preferably mammals including humans, by various routes. All modes of administration are predictable, and may be administered, for example, by oral, intravenous, intramuscular, subcutaneous injection, etc. The pharmaceutical administration form of the composition of the present invention may be used in the form of a pharma- ceutical acceptable salt of the active ingredient, and may be used alone or in combination with other pharma- ceutical active compounds or in other appropriate combinations.

[0037] According to another aspect of the present invention, there is provided a method for screening a pharmaceutical composition for preventing or treating a viral infection, comprising the steps of: (a) transfecting a host cell with a virus; (b) treating the host cells with the drug candidate; and (c) analyzing the RNA tailing level from the experimental group treated with the drug candidate substance and comparing it with the RNA tailing level from a control group not treated with the drug candidate substance; if the RNA tailing level from the experimental group is relatively decreased compared to the untreated control group, it is selected as an effective drug candidate substance.

[0038] The level of RNA tailing is determined by the number of times that adenines contained in the poly(A) tail are replaced by other nucleic acids such as guanine or uridine through guanylation or uridylation.

[0039] The steps (a) and (b) of the present invention may be performed in any order.

[0040] The sample is treated with a virus-infected host cell, and if mixed tailing is suppressed, the sample can be selected as a valid drug candidate.

[0041] Since the drug screening method of the present invention has common characteristics with the pharmaceutical composition, which is another embodiment of the present invention, in that it utilizes the mixed tailing operating principle by TENT4A / B, the detailed description of the pharmaceutical composition is incorporated by reference, and redundant description will be omitted to avoid excessive complexity of the description in this specification.

[0042] According to another aspect of the present invention, there is provided a method for producing virus-resistant cells, comprising the step of knocking out expression of at least one of TENT4A and TENT4B in cells isolated from a living organism.

[0043] The knockout of the present invention can be introduced by utilizing various gene editing techniques known in the art. In the examples of this specification, the CRISPR-Cas genome editing method was used (see Example 6), but the present invention is not limited to this.

[0044] According to another aspect of the present invention, there is provided a method for stabilizing an RNA sequence, comprising inserting into a target RNA sequence a stem-loop sequence comprising a pentaloop structure composed of an amino acid sequence represented by the following general formula: [General formula] 5'-CNGGN-3' Each N is independently any one selected from adenosine (A) and uracil (U).

[0045] The term "stem-loop structure" as used herein refers to a bond between base pairs that occurs in a single strand of DNA or RNA. This structure is also called a hairpin or hairpin structure. It occurs when the base pairs are complementary to each other when two regions of a strand are read in opposite directions. An example of a stem-loop structure is shown in Figure 12b, but is not limited thereto.

[0046] The term "penta-loop" as used herein means a single-stranded sequence region in which there are five bases at the linking site that does not form a bond between base pairs in a stem-loop structure. The stem-loop structure of the present invention includes a penta-loop structure, and an example diagram is shown in FIG.

[0047] In one embodiment of the present invention, the stem loop structure of the present invention includes a bond between 2-15 base pairs. More specifically, the bond between the base pairs includes, but is not limited to, a 2-14 base pair bond, a 2-13 base pair bond, a 2-12 base pair bond, a 2-11 base pair bond, a 2-10 base pair bond, a 3-10 base pair bond, and a 3-9 base pair bond. The bond between the base pairs may be an adenine-uracil bond or a cytosine-guanine bond, and the type of bond is not limited.

[0048] By inserting the stem-loop structure of the present invention into a target RNA sequence, the frequency at which the target RNA sequence induces mixed tailing using TENT4A / B can be increased, and the RNA sequence can be stabilized.

[0049] A specific embodiment of the present invention is shown in FIG.

[0050] [Effects of the Invention] The features and advantages of the present invention can be summarized as follows: (a) The present invention provides a pharmaceutical composition for preventing or treating a viral infection.

[0051] (b) The present invention provides a method for screening a pharmaceutical composition for preventing or treating a viral infection.

[0052] (c) The present invention provides a method for producing virus-resistant cells.

[0053] (d) The present invention provides a method for stabilizing an RNA sequence by the introduction of a defined stem-loop structural sequence.

[0054] (e) When the pharmaceutical composition of the present invention is used, viral infections can be effectively prevented and treated.

[0055] (f) When using the RNA sequence stabilization methods of the present invention, mixed tailing can be induced on the desired RNA sequence to induce sequence stabilization.

[0056] BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 shows the extensive mixture of tails in viral RNAs. a, The fraction of guanylated tails in cellular and viral mRNAs was calculated for mRNAs with poly(A) tail lengths ≥ 25 nt (n = 2 TAIL-seq experiments). b, The fraction of guanylated tails in cellular RNA, HCMV RNA, and RNA2.7 (VRNA2.7) was calculated for RNAs with poly(A) tail lengths ≥ 25 nt (n = 1 TAIL-seq experiment). c, Distribution of total poly(A) tail lengths in cellular (grey) and HBV (yellow) mRNAs. Medians are indicated by dotted vertical lines, and median lengths are shown in brackets. d, Distribution of total poly(A) tail lengths in cellular (grey), HCMV (yellow), and VRNA2.7 (orange) RNAs. Medians are indicated by dotted vertical lines, and median lengths are shown in brackets. e, Fraction of guanylated tails of HBV mRNA with poly(A) tail length ≥ 25 nt in TENT4-depleted HepG2.2.15 cells (n = 1 TAIL-seq experiment). f, Fraction of guanylated tails of HCMV RNA with poly(A) tail length ≥ 25 nt in TENT4-depleted HCMV-infected HFF cells (n = 1 TAIL-seq experiment). g, HBV mRNA levels were measured by RT-qPCR using two individual primer sets in TENT4-depleted HepG2.2.15 cells. Data are presented as mean ± sem (n = 3 independent experiments). **P < 0.01; two-tailed Student's t-test. h, Half-life of HBV mRNA was measured by RT-qPCR using two individual primer sets. Data are presented as mean ± sem (n = 3 independent experiments). GAPDH mRNA was used for normalization.

[0057] Figure 2: Experimental results confirming that HBV RNA is the major mRNA substrate for TENT4 through the HBV PRE. a, Normalized read coverage of fCLIP-seq libraries across the HBV genome (NCBI U95551.1). The HBV PRE region is highlighted in yellow. HBV transcripts and their respective coding sequence regions are shown below for reference. Input and NMG libraries were used as negative controls. b, Scatter plot of abundance (x-axis) and enrichment score (y-axis) of fCLIP-seq peak clusters. HBV PREs are represented by red dots. A size-matched input library was used as a negative control in this analysis.

[0058] FIG. 3: Stem-loop structure of PRE is essential but not sufficient for TENT4-dependent RNA tailing of HBV mRNA. a, Schematic of Firefly luciferase reporter constructs carrying PRE mutations and subregions in the 3'UTR. Blue, La-binding motif; yellow-green, stem-loop alpha; and red*, region of point mutation. b, Firefly luciferase activity of reporters containing PRE subregions in parental HeLaT (gray) and TENT4 KO (red) cells. Data are presented as mean ± sem (n = 7 independent experiments for PRE, PREα, and PREβ, n = 6 independent experiments for all others). Both Renilla and control vector luciferase activities were used for normalization. *P < 0.01, **P < 0.001; two-sided t-test. c, Distribution of poly(A) tail length of reporter RNAs measured by Hire-PAT analysis. Reporter constructs containing PREα and PREβ sequences were transfected into HEK293T cells after TENT4 depletion. d, Firefly luciferase activity of PRE reporter constructs in ectopically expressed wild-type TENT4 or parental or TENT4 KO cells carrying catalytically inactive mutations. Data are presented as mean ± sem (n = 3 independent experiments). Both Renilla and control vectors were used for normalization. *P < 0.05, **P < 0.01, ***P < 0.001; two-sided Student's t-test. e, Schematic diagram of Firefly luciferase reporter constructs carrying the WPRE subregion in the 3'UTR. f, Firefly luciferase activity of the WPRE reporter construct in TENT4 KO cells. Data are presented as mean ± sem (n = 3 independent experiments for Wβ, n = 7 independent experiments for all others). Both Renilla and control vector luciferase activities were used for normalization. *P < 0.05, **P < 0.01; two-tailed Student's t-test.g, Stem-loop structures of HBV PRE, mutant HBV PRE and WPRE. h, Firefly luciferase activity of mutant PRE reporter constructs shown in a in TENT4 KO cells. Data are presented as mean ± sem (n = 4 independent experiments). Both Renilla and control vector luciferase activities were used for normalization. *P < 0.05, **P < 0.01; two-tailed Student's t-test.

[0059] Fig. 4: Stem-loop structures of HCMV RNA2.7 are cis-acting RNA elements responsible for TENT4-dependent regulation. a, RT-qPCR of the indicated RNAs after formaldehyde crosslinking and immunoprecipitation with anti-TENT4A antibody in RNA2.7-transfected HEK293T cells. Data are presented as mean ± sem (n = 3 independent experiments). Immunoprecipitation with NMG was used for normalization. **P < 0.01, ***P < 0.001; two-tailed Student's t-test. b, Schematic diagram of firefly luciferase reporter constructs carrying subregions of HCMV RNA2.7 and mutations in the 3'UTR. Light green indicates stem-loops, red * indicates point mutations. Stem-loop structures and mutations of HCMV RNA2.7 are shown on the left. In the 1E mut construct, the red nucleotides in the loop and the C at the base of the loop are mutated as indicated. FRG, fragment. c–e, Firefly luciferase activity of the reporter in parental and TENT4 KO cells. Data are presented as mean ± sem. Reporter with subregion of RNA2.7 (n = 3 independent experiments) (c), partial deletion mutant of fragment 1 (1-513) of RNA2.7 (n = 3 independent experiments) (d), and stem-loop mutant of constructs 1D, 1E, and 1E (n = 3 independent experiments) (e). Luciferase activity of both Renilla and control vector was used for normalization. *P < 0.05, **P < 0.01; two-tailed Student's t-test.

[0060] Figure 5: SAM domain-containing proteins, including ZCCHC14, bind stem-loop structures and regulate HBV mRNA. a, Mass spectrometry analysis of SL2.7 RNA pulldowns from HEK293T cell lysates (n=2 mass spectrometry experiments). SL2.7 mutant and 'bead only' samples were used as negative controls. Protein enrichment scores were calculated as the log2 fold change in LFQ (label-free quantification) intensity from SL2.7 samples versus mutant data. Spectral counts were used to filter enrichment candidates. b, Top, phylogenetic tree of proteins with putative RNA-binding SAM domains. Sc (Saccharomyces cerevisiae), Sp (Schizosaccharomyces pombe), Ca (Candida albicans), Ce (C. elegans), Dm (Drosophila melanogaster), Dr (Danio rerio), Xl (Xenopus laevis), Mm (Mus musculus) and Hs (Homo sapiens). * indicates manual gene name assignment based on phylogenetic tree. UniProt IDs are listed in the Methods section. Bottom, domain structure of three human proteins with SAM domain. Orange, SAM domain; purple, CCHC zinc finger (ZnF) domain. c, Pulldown of biotinylated SL2.7 from HEK293T cells after Western blot with the indicated antibodies. * indicates cross-reacting bands. d, HBV mRNA levels were measured in ZCCHC14-depleted and SAMD4A / B-depleted HepG2.2.15 cells by RT-qPCR using two sets of primers. Data are presented as mean ± sem (n = 4 independent experiments). *P < 0.05, **P < 0.01, ***P < 0.001; two-tailed t-test.The uncut plot for c and the source data for the graph in d are available online.

[0061] Figure 6: Cytoplasmic ZCCHC14 recruits TENT4 to protect RNA with stem-loop structures. a, The fraction of guanylated tails of HBV mRNA was calculated for mRNAs with poly(A) tail lengths of 25 nt or more in ZCCHC14-depleted HepG2.2.15 cells. b, Distribution of total poly(A) tail length of HBV mRNA in control (gray) and ZCCHC14-depleted HepG2.2.15 cells (red). Median values ​​are indicated by vertical dotted lines and median values ​​are shown in brackets. c, Firefly luciferase activity of reporters carrying HCMV RNA2.7 1E, HBV PREα, WHV WPRE(Wγ+Wα) and their respective mutations (1E mut, α mut2 and Wγ+Wα mut) in ZCCHC14-depleted HEK293T cells. Data are presented as mean ± sem (n = 4 independent experiments for PREα and α mut2, n = 3 independent experiments for all others). Importantly, the WPRE mutant reporter (Wγ + Wα mut) consists of point mutations in all of the WPREα stem-loop structures (Supplementary Fig. 6b). Luciferase activity of both Renilla and control vectors was used for normalization. *P < 0.05; two-tailed Student's t-test. d, RT-qPCR of indicated RNAs after immunoprecipitation with anti-ZCCHC14 antibody in HepG2.2.15 cells. Data are presented as mean ± sem (n = 3 independent experiments). Immunoprecipitation with normal rabbit IgG was used for normalization. *P < 0.05; two-tailed Student's t-test. e, ZCCHC14 protein localization was examined in HeLaT cells by subcellular fractionation and Western blotting: cytoplasm (Cyto), membrane (Memb) and nucleus (Nuc). GAPDH (cytoplasm), GM130 (Golgi) and histone H3 (nucleus) were used as fractionation markers. f, Western blot after immunoprecipitation with anti-TENT4A and anti-TENT4B was used to measure the physical interaction of TENT4A / B with ZCCHC14 in HeLaT cytoplasmic extracts. Cell extracts were treated with RNase A.GAPDH was used as a loading control. g, Proposed model. HBV mRNA and HCMV RNA2.7 recruit the TENT4-ZCCHC14 complex using the CNGGN pentaloop to induce targeted mixed tailing, which subsequently protects viral RNA from cytolytic agents.

[0062] FIG. 7: Extensive 3' end tail modifications of viral RNA. a, HBV mRNA is highly mixed tail (G: guanylation, U: uridylation, C: cytidylation) in HepG2.2.15 cells. The fraction of each modification includes its corresponding terminal and internal modifications. The fraction of viral mRNA with cytidylated and uridylated tails was calculated for poly(A) tails ≥ 25 nt in length (n=2 TAIL-seq experiments). b, HCMV RNAs are also highly mixed tail in HCMV-infected HFF cells. HCMV RNA2.7 (VRNA2.7) in particular is substantially mixed tail (n=1 TAIL-seq experiment). c, Gene-level analysis of guanylated tails of cellular (gray) and viral (yellow) mRNAs in HCMV-infected HFF cells. d, siRNA knockdown confirmation by RT-qPCR. e, Reduction in the mixed tail fraction of HBV mRNA in TENT4-depleted HepG2.2.15 cells. f, TENT4A / B expression is induced in HCMV-infected HFF cells. In HCMV-infected HFF cells, TENT4A and TENT4B RNA levels were measured by RT-qPCR (n=1). g, Reduction in the mixed tail fraction of HCMV RNA2.7 in TENT4-depleted HCMV-infected HFF cells. h, HBV mRNA shows shorter poly(A) tail length in TENT4-depleted HepG2.2.15 cells. Median values ​​are indicated by vertical dotted lines and shown in brackets. i, Distribution of total poly(A) tail length of viral mRNA after TENT4 depletion in HCMV-infected HFF cells. j, Distribution of poly(A) tail length of HCMV RNA2.7 after TENT4 depletion. k, HBV transcripts and their corresponding CDS regions are shown. Bars indicate the positions of RT-qPCR amplicons (#1-3). l, Tail variants of HBV mRNA across the poly(A) tail length in TENT4-depleted HepG2.2.15 cells. m, Half-life of HCMV RNA2.7 was measured by RT-qPCR (n=3 independent experiments) in HCMV-infected HFF cells after TENT4 depletion. GAPDH mRNA was used for normalization. Means were calculated. Graph data a~b, d~g, and m are available as source data.

[0063] Figure 8: Analysis and validation of TENT4A / B and HBV mRNA interaction. a, TENT4B-bound RNA fragments are enriched again within the PRE region of HBV (highlighted in yellow). Standardized read coverage of fCLIP-seq libraries across the HBV genome (NCBI: U95551.1). X-axis scale is the same as in Figure 2a. b, After immunoprecipitation with anti-TENT4A and anti-TENT4B, RT-qPCR was used to measure enrichment of TENT4-bound RNA in HepG2.2.15 cells (n=3 independent experiments). Immunoprecipitation with normal mouse IgG was used for normalization. Means were calculated and error bars indicate SEM. **P<0.01, ****P<0.0001; two-sided t test. Graph data in b are available as source data.

[0064] FIG. 9: Results of confirming TENT4-dependent regulation of the PRE reporter. a, Knockout confirmation by Western blotting. GAPDH was used as a loading control. b, RNA levels of firefly reporter carrying a PRE subregion were measured by RT-qPCR (PRE:PREα:PREβ, n=4; αΔ1, n=2; αΔ2, n=1 independent experiments). Both renilla and control vector mRNA levels were used for normalization. Means were calculated, and error bars indicate SEM. *P<0.05; two-sided t test. c, siRNA knockdown confirmation by Western blotting in HEK293T cells. GAPDH was used as a loading control. d, Firefly luciferase activity of PRE reporter constructs in TENT4-depleted HEK293T cells (PRE:PREα:PREβ, n = 7; αΔ1, n = 6; αΔ2, n = 5 independent experiments). Both Renilla and control vector luciferase activities were used for normalization. Means were calculated and error bars indicate SEM. *P < 0.05, **P < 0.01, ***P < 0.001; two-tailed t-test. e, Overexpression confirmed by Western blotting in KO cells. GAPDH was used as a loading control. f, RNA levels of firefly reporter used in rescue experiments in TENT4 KO cells with ectopically expressed TENT4 wild-type or mutant were measured by RT-qPCR (n = 3 independent experiments) Both Renilla and control vector mRNA levels were used for normalization. Means were calculated and error bars indicate SEM. *P<0.05, **P<0.01; two-tailed t-test. Data for raw blots for panels a, c, e and graphs in b, d, f are available as source data.

[0065] [Figure 10] Confirmation of TENT4-dependent regulation of the RNA2.7 reporter. a, Immunoprecipitation with anti-TENT4A was confirmed by Western blotting in RNA2.7-transfected HEK293T cells. Normal mouse IgG (NMG) was used as a negative control. The RNA2.7 transcript is shown with a bar indicating the position of the RT-qPCR amplicon. b-e, RNA levels of the firefly reporter carrying the following: b, subregions of RNA2.7 (n=3 independent experiments), c, deletion constructs of RNA2.7 fragment 1 (1-513) (n=3 independent experiments), d, constructs 1D (314-413), 1E (414-513) and stem-loop mutant 1E (n=3 independent experiments), and e, construct SL2.7 (414-463) and control (n=3 independent experiments) in parental and TENT4 KO cells. Both Renilla and control vector mRNA levels were used for normalization. Means were calculated and error bars indicate SEM. *P<0.05, **P<0.00.0; two-tailed t-test. Data for uncut blots for panels a and graphs in b-e are available as source data.

[0066] FIG. 11: Confirmation of ZCCHC14 and SAMD4A / B knockdown. siRNA knockdown confirmed by RT-qPCR (n=4 independent experiments) and Western blotting in HepG2.2.15 cells. Means were calculated and error bars indicate SEM. ****P<0.0001; two-tailed t-test. GAPDH was used as a loading control.

[0067] FIG. 12: Confirmation of ZCCHC14 knockdown, stem-loop-dependent tail regulation, and immunoprecipitation. a, Confirmation of siRNA knockdown by Western blotting in ZCCHC14-depleted HEK293T cells. GAPDH was used as a loading control. Dotted lines indicate discontinuous lanes in the same gel. b, Stem-loop structure of WPREα and its respective mutations. c, Distribution of poly(A) tail length of reporter RNA measured by Hire-PAT analysis. Reporter constructs were transfected into HEK293T cells after ZCCHC14 depletion. d, Immunoprecipitation with anti-ZCCHC14 was confirmed by Western blotting in HepG2.2.15 cells. Normal rabbit IgG (NRG) was used as a negative control. e, Western blotting after immunoprecipitation with anti-TENT4A and anti-TENT4B was used to measure the physical interaction of TENT4A / B with ZCCHC14 in HepG2.2.15 and primary HFF cells. Cell extracts were treated with RNase A. Normal mouse IgG (NMG) was used as a negative control. * indicates cross-reacting bands. Uncut blots for panels a, d-e are available as source data.

[0068] FIG. 13 shows a stem-loop structure including a pentaloop.

[0069] The method for inserting the stem-loop structure of the present invention into a target RNA sequence can be carried out by various conventionally known methods, and is not particularly limited.

[0070] The present invention can also be configured as follows.

[0071] [1] A pharmaceutical composition for preventing or treating a viral infection, comprising an inhibitor of one or more of TENT4A and TENT4B, which are involved in mixed tailing of viral RNA, wherein the inhibitor is an siRNA or shRNA that suppresses the expression of one or more of TENT4A and TENT4B, or an antibody or an antigen-binding fragment thereof that suppresses the activity of one or more of TENT4A and TENT4B.

[0072] [2] The pharmaceutical composition described in [1], wherein the TENT4A and TENT4B are derived from a subject at risk of viral infection or infected with a virus.

[0073] [3] The pharmaceutical composition described in [1], wherein the virus is a virus in which RNA tailing is induced by one or more of TENT4A and TENT4B.

[0074] [4] The pharmaceutical composition described in [3], wherein the virus is a virus belonging to the Hepadnaviridae or Herpesviridae family.

[0075] [5] The viruses belonging to the Hepadnaviridae family include Hepatitis B virus, Ground Squirrel Hepatitis B virus, and Hepatitis B virus, Woodchuck Hepatitis B virus, Duck hepatitis B virus, and Heron Hepatitis B virus.

[0076] [6] The pharmaceutical composition described in [4], wherein the virus belonging to the Herpesviridae family is any one selected from the group consisting of Iltovirus, Mardivirus, Scutavirus, Simplexvirus, Varicellovirus, Cytomegalovirus, Muromegalovirus, Proboscivirus, Roseolovirus, Lymphocryptovirus, Macavirus, Percavirus and Rhadinovirus.

[0077] [7] A method for screening a pharmaceutical composition for preventing or treating a viral infection, comprising the steps of: (a) transfecting a host cell with a virus; (b) treating the host cells with the drug candidate; and (c) analyzing the RNA tailing level from the experimental group treated with the drug candidate and comparing it with the RNA tailing level from a control group not treated with the drug candidate; if the RNA tailing level from the experimental group is more reduced compared to the untreated control group, it is selected as an effective drug candidate.

[0078] [8] A method for producing virus-resistant cells, comprising the step of knocking out the expression of one or more of TENT4A and TENT4B in cells isolated from a living body.

[0079] [9] A method for stabilizing an RNA sequence, comprising inserting into a target RNA sequence a stem-loop sequence having a pentaloop structure composed of an amino acid sequence represented by the following general formula: [General formula] 5'-CNGGN-3' Each N is independently any one selected from adenosine (A) and uracil (U).

[0080] [Mode for carrying out the invention] The present invention will be described in more detail below with reference to examples. It will be apparent to those skilled in the art that these examples are merely intended to more specifically illustrate the present invention, and that the scope of the present invention is not limited by these examples according to the gist of the present invention.

[0081] [Example] <Experimental Method> Example 1: TAIL-seq library preparation and data processing TAIL-seq was performed as previously described [3]. Briefly, ~50 μg of total RNA (>200 nt) treated with DNaseI (Takara, 2270A) was used and purified using the Ribo-Zero kit (Epicentre, MRZH11124 (discontinued) for primary HFF library or Illumina, TruSeq Stranded Total RNA Library Prep). Ribosomal RNA was depleted twice with the Human / Mouse / Rat,20020596 for HepG2.2.15 library. The rRNA-depleted RNA was ligated to a 3' adaptor and partially fragmented with RNase T1 (Ambion). The fragmented RNA was pulled down with streptavidin (Invitrogen), 5' phosphorylated, and purified on a 6% urea-PAGE gel (500-1,000 nt). The purified RNA was ligated to a 5' adaptor, reverse transcribed, and amplified by PCR. The library was sequenced by paired-end run (51 × 251 cycles) on an Illumina platform (MiSeq) with PhiX control library v.3 (Illumina) and spike-in mix. TAIL-seq sequencing data were deposited in the National Center for Biotechnology Information (NCBI) Gene Expression Omnibus (GEO) database under the accession number GSE146600. TAIL-seq analysis was performed using Tailseeker v.3.1.5 [9]. Briefly, Read 1 was used for gene identification, and Read 2 was used to detect 3'-end variants and measure the length of the poly(A) tail. For each TAIL-seq library, Read 1 was generated from human genome GRCh38 with STAR2.5.2b

[0050] and HBV genome NCBI U95551.1 or HCMV genome NCBI U95551.1 with bowtie2.2.6

[51] . The 3'-terminal 10-mer sequences with terminal or internal mono-guanylation (or -cytidylation, -uridylation) were examined in Read 2 to estimate the fraction of non-adenosine contamination in the poly(A) tail.

[0082] Example 2: fCLIP-seq library preparation and data processing fCLIP-seq was performed as previously described with minor adjustments [16, 52]. Briefly, HepG2.2.15 cells on two 150 mm dishes were crosslinked with 0.1% paraformaldehyde (Pierce, 28906), harvested, and lysed. 15 μg of each antibody (NMG, Santa Cruz, sc-2025; TENT4A, laboratory-made; TENT4B, laboratory-made) was conjugated to protein A and G Sepharose beads (1:1 mix, total 20 μl, GE Healthcare, 17-5138-01 and 17-0618-01, respectively). Lysates were incubated with antibody-conjugated beads, and RNA was purified from the eluate, followed by DNaseI treatment and washing. Libraries were then prepared using 1 μg of input RNA or RNA from each sample. rRNA was depleted twice with the Ribo-Zero kit (Illumina, TruSeq Stranded Total RNA Library Prep Human / Mouse / Rat, 20020596). rRNA-depleted RNA was ligated to 3' adapters and purified by 6% urea-PAGE gel (80-500 nt, corresponding to 50-470 nt RNA fragmented by sonication), followed by 5' phosphorylation, 5' adapter ligation, reverse transcription and PCR amplification. fCLIP-seq libraries were sequenced in a paired-end run (151 × 151 cycles) on an Illumina platform (MiSeq) using PhiX control library v.3 (Illumina). fCLIP-seq sequencing data have been deposited in the NCBI GEO database under the accession number GSE146597.

[0083] Paired-end reads for each fCLIP-seq library were assembled with pear v.0.9.10

[53] and aligned to human genome GRCh38 by STAR2.5.2b

[50] and to HBV genome NCBI U95551.1 by bowtie2.2.6

[51] . Raw read coverage of fCLIP-seq libraries was calculated with bedtools v.2.26.0. fCLIP-seq peak clusters of cellular mRNAs were estimated using Piranha v1.2.1

[54] with a cluster size of 200 nt (-z 200), internal normalization (-n), log transformation (-l) and NMG library as covariates. As a reference, HBV PRE spans approximately 400 nt. Peak cluster enrichment scores were calculated by the log2 ratio of reads in TENT4 fCLIP-seq library and NMG library. The input library was used as a negative control for this analysis.

[55] To provide technical background on read coverage, the Hodges-Lehmann estimate for the HBV genome was used. Specifically, given an HBV genome of length n, we calculate the median of the following set:

[0084] {(Xi+Xj) / 2:1≦i≦j≦n} where Xi and Xj are the raw read coverages at positions i and j, respectively. The raw read coverages were then normalized by the Hodges-Lehmann estimate and visualized using the ggplot2 R package.

[0085] Example 3: Cell culture, transfection and actinomycin D treatment All cell lines used in this study were tested mycoplasma negative. HeLaT was derived from HeLa (provided by C.-H. Chung, Seoul National University) with a null mutation in the TUT4 gene. HeLaT and HEK293T (provided by S. Kim, Seoul National University) cells were authenticated by ATCC (STR profiling). HepG2.2.15, HeLaT and HEK293T cells were grown in DMEM (Welgene, LM001-05) containing 10% FBS (Welgene, S001-01). Primary HFF (ATCC SCRC-1041) cells were grown in DMEM (HyClone) containing 10% FBS (HyClone), GlutaMAX-1 (Gibco) and penicillin-streptomycin (Gibco). Prior to transfection, primary HFF were grown in antibiotic-free medium.

[0086] For combinatorial knockdown, the same amount of small-interfering RNA or GAPmer was mixed for the target genes, and the final concentration was as described below. For TENT4A and TENT4B knockdown, HepG2.2.15 cells were transfected with 100 nM siRNA by Lipofectamine RNAiMAX (Invitrogen) on days 0, 2, and 4, and harvested on day 6. For ZCCHC14, SAMD4A, and SAMD4B knockdown, HepG2.2.15 cells were transfected with 100 nM siRNA or GAPmer by Lipofectamine RNAiMAX (Invitrogen) on days 0 and 2, and harvested on day 4. HEK293T cells were transfected with RNA2.7 expression pCK vector for RNA immunoprecipitation by Lipofectamine 3000 on day 0, and harvested on day 2. Primary HFF cells were transfected with 20 nM siRNA by DharmaFECT 1 (Dharmacon) on days 0 and 2 and harvested on day 5. siRNA sequence information is presented in Table 1.

[0087] [Table 1]

[0088] We purchased siRNA against TENT4A from ON-TARGETplus SMARTpool (Dharmacon) and GAPmer against SAMD4A from Antisense LNA GapmeR Standard (Qiagen, 339511 LG00236046-DDA) for better knockdown efficiency in HepG2.2.15. Actinomycin D (Sigma, A9415, 4 μg ml-1) was added to HepG2.2.15 cells to block transcription, and cells were harvested at the indicated times.

[0089] Example 4: HCMV infection Infectious HCMV particles were produced by transfecting primary HFF cells with HCMV Toledo BAC DNA (provided by T. Shenk, Princeton University) by electroporation (Invitrogen, Neon). When 100% cytopathic effect was observed, cell culture supernatants were collected, centrifuged to remove cell debris, and stored in 1 ml aliquots at -80°C. To titrate virus stocks, primary HFFs grown on cover slips were inoculated with diluted virus stocks for 1 h and fixed with 3.7% formaldehyde 24 h after inoculation. Cells were permeabilized with 0.1% Triton X-100, incubated with blocking buffer (2% bovine serum albumin in phosphate-buffered saline), stained with HCMV IE1 antibody (MAB810R; Millipore), and then mounted with FITC-conjugated anti-mouse antibody (115-095-146; Jackson Laboratories) and DAPI-containing solution (H1200; Vector Laboratory). The number of HCMV IE1-positive cells was counted, and the multiplicity of infection (MOI) was determined by calculating the ratio of IE1-positive cells to total cells. For HCMV infection, primary HFF cells were incubated with HCMV diluted in serum-free DMEM (MOI, 2) for 1 h, washed with PBS, and incubated in DMEM for further incubation.

[0090] Example 5: Plasmid construction Except for the FLAG tag, wild-type TENT4A and TENT4B variants (792 and 666 aa, respectively) [6] were used as described previously. Point mutations for catalytically dead mutants were introduced (D352A for TENT4A and D326A for TENT4B, respectively). For the PRE reporter constructs, PRE (1,154-1,687 bp), PREα (1,154-1,351 bp), PREβ (1,352-1,687 bp), αΔ1 (1,276-1,351 bp), and αΔ2 (1,154-1,275 bp) were amplified with HepG2.2.15 complementary DNA and introduced into the 3'UTR region of the firefly luciferase mRNA sequence in the pmirGLO-3XmiR-1 vector [6]. Point mutations for the La-binding motif and stem-loop α were identical to previously described mutants

[30] . α-mut1, α-mut2, and α-mut3 were generated by PCR-directed mutagenesis. For the WPRE reporter constructs, WPRE (1,095-1,670 bp), Wγ+Wα (1,095-1,507 bp), Wα+Wβ (1,300-1,670 bp), Wγ (1,093-1,299 bp), Wα (1,300-1,507 bp), and Wβ (1,508-1,670 bp) were amplified from pLenti-CMV vector (Addgene, 17492) and introduced into the 3'UTR of firefly luciferase in the pmirGLO-3Xmir-1 vector. Point mutations to the CNGGN pentaloop of WPREα were designed in a similar manner to the PRE mutants.In the RNA2.7 construct, FRG1 (1-513 bp), FRG2 (414-913 bp), FRG3 (814-1,313 bp), FRG4 (1,214-1,713 bp), FRG5 (1,614-2,113 bp), FRG6 (2,014-2,513 bp), 1ΔA (1-513 bp, Δ1-100 bp), 1ΔB (1- SL2.7 (414-463bp) were amplified from HCMV-infected primary HFF cDNA and introduced into the 3'UTR region from the pmirGLO-3Xmir-1 vector. The 1E and SL2.7 mutant constructs were generated in the same manner as the PRE mutants. For RNA immunoprecipitation, RNA2.7 (1-2,513bp) was amplified from the pmirGLO-3Xmir-1 vector and subcloned into the pCK vector.

[0091] <Example 6: Production of TENT4A and TENT4B knockout cells> For the removal of TENT4A and TENT4B, CRISPR-Cas genome editing was performed as previously described

[56] with a minor adaptation where T7 endonuclease I (NEB, M0302) was used instead of Surveyor nuclease. 6E4HeLaT cells on 24-well plates were first transfected with 300ng pSpCas9(BB)-2A-GFP-px458 plasmid (Addgene no. 48138) carrying single-guide RNA (agccacgttgtgcttccgcg, PAM sequence GGG) against TENT4A using Metafectene (Biontex, T020). The HeLaT parent cells are derived from HeLa and contain a null mutation in the TUT4 gene. After single-cell screening and Sanger sequencing to confirm the knockout, the parental and altered genomic sequences are listed in Table 2, with the inserted sequence in bold and underlined.

[0092] [Table 2]

[0093] Using TENT4A knockout cells, TENT4B was further deleted with sgRNA against TENT4B (gacatcgacctagtggtgtt, PAM sequence is TGG). The altered genomic sequences are further listed in Table 2, with the inserted and deleted sequences shown in red and dotted lines, respectively.

[0094] <Example 7: Preparation of TENT4A and TENT4B antibodies> Mice were challenged with TENT4A antigen (126-571aa from 792aa variant, NCBI NP_008930.2) or TENT4B antigen (106-533aa from 666aa variant, NCBI XP_006721308.1) and killed. The collected spleen cells were fused with Sp2 / O myeloma, and hybridoma cells producing high affinity antibodies were selected. After intraperitoneal injection of hybridoma cells into mice, ascites was obtained (Youngin Frontier Inc.) and used for immunoprecipitation experiments.

[0095] Example 8: Western Blot HepG2.2.15 cells were lysed in RIPA buffer (20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mM EDTA, 10% NP-40 (Sigma, 74385), 1% sodium deoxycholate (Sigma, D6750), 1% SDS (Ambion, AM9823)). Other cells were lysed in buffer D (200 mM KCl, 10 mM Tris-HCl (pH 8.0), 0.2 mM EDTA, 0.1% Triton X-100 (Promega, H5142)). Approximately 60 μg of lysate was loaded onto 10% or 8-16% (Novex) SDS-PAGE gels with ladders (Thermo, 26616 and 26619). After transfer to methanol-activated polyvinylidene difluoride membranes (Millipore), the membranes were washed with 5% skim milk (skim The sections were blocked with PBS-T containing 100% ethanol (100% ethanol) milk, labeled with primary antibodies, and washed three times with PBS-T. Anti-mouse or anti-rabbit HRP-conjugated secondary antibodies (Jackson ImmunoResearch Laboratories) were incubated and washed three times with PBS-T. Chemiluminescence was performed with West Pico or Femto Luminol reagents (Thermo), and signals were detected by a ChemiDoc XRS+ System (BioRad). Anti-TENT4A (1:500, Invitrogen, PA5-61302; 1:250-500, Atlas Antibodies, HPA045487), Anti-TENT4B (1:500, Invitrogen, PA5-60177; 1:500, Atlas Antibodies, HPA042968; 1:500, prepared in the laboratory), anti-ZCCHC14 (1:1,000-2,000, Bethyl Laboratories, A303-096A), anti-GAPDH (1:1,000-2,000, Santa Cruz, sc32233), anti-SAMD4B (1:500, Invitrogen, PA5-53490), anti-GM130 (1:500, BD Anti-Histone H3 (1:2,000, Cell Signaling, 4499) were used as primary antibodies.

[0096] <Example 9: RT-qPCR> Total RNA was extracted using Trizol reagent (Invitrogen, 15596018) or the Maxwell 16LEV simplyRNA Tissue kit (Promega, AS1280), treated with DNaseI, and purified using RNeasy MinElute. The samples were purified with the Cleanup kit (Qiagen) and reverse transcribed with RevertAid reverse transcriptase (Thermo) and oligo dT for luciferase assay samples or random hexamers (Invitrogen) for other samples. mRNA levels were measured with SYBR Green analysis (Thermo) and StepOnePlus real-time PCR system (Applied Biosystems) or QuantStudio 3 (Applied Biosystems). A list of RT-qPCR primers is presented in Table 3.

[0097] [Table 3]

[0098] Example 10: RNA immunoprecipitation For TENT4A and TENT4B immunoprecipitation using HepG2.2.15 cells, cells from one 150 mm dish were collected and lysed on ice in lysis buffer (20 mM HEPES (pH 7.0-7.6) (Sigma, H0887), 4% NP-40, 100 mM KCl, 0.1 mM EDTA, 10% glycerol, 1 mM DTT, 20 U ml-1 RNase inhibitor (Ambion, AM2696), 1x protease inhibitor (Calbiochem, 535140), and centrifuged. 0.375 μg of each antibody (NMG, Santa Cruz, sc-2025; anti-TENT4A, laboratory-made; anti-TENT4B, laboratory-made) were conjugated to protein A and G sepharose beads (1:1 mix, 10 μl total). Lysates were washed with wash buffer (same lysis buffer but with 2% NP-40) after incubation with antibody-conjugated beads. After adding 10 ng of firefly luciferase mRNA to the samples as a spike-in used for normalization, RNA was purified with TRIzol reagent, treated with DNaseI, and used for RT-qPCR. A similar method was used for ZCCHC14 immunoprecipitation using HepG2.2.15 cells, also treated with Turbo DNaseI (Ambion, AM2239) during lysis and incubated with 10 μg of each antibody (general rabbit IgG (Cell signaling, 2729S) and anti-ZCCHC14 (Bethyl For TENT4A immunoprecipitation with HEK293T cells (one 100 mm dish), a similar method was used with 6 μg of each antibody (NMG and anti-TENT4A), and cells were crosslinked (0.1% paraformaldehyde) before harvesting.

[0099] Example 11: Co-immunoprecipitation Similar steps were used as in RNA immunoprecipitation, but Turbo DNaseI and RNase A (Thermo, EN0531) were added. Briefly, HepG2.2.15 cells from five 150 mm dishes were lysed and 17.5 μg of each antibody (NMG, anti-TENT4A, anti-TENT4B) conjugated to protein A and G sepharose beads (1:1 mix, 25 μl total) were used for immunoprecipitation. Primary HFF cells from two 150 mm dishes were lysed and 6 μg of each antibody (NMG, anti-TENT4A, anti-TENT4B) conjugated to protein A and G sepharose beads (1:1 mix, 20 μl total) were used for immunoprecipitation. RNase A was added to the lysate to a final concentration of 0.2-0.3 μg / μl. For coimmunoprecipitation experiments, cytoplasmic extracts from one and one-third 150 mm dishes of HeLaT cells were obtained by extracellular fractionation as described below and then used for immunoprecipitation with 18 μg of each antibody (NMG, anti-TENT4A, anti-TENT4B).

[0100] Example 12: Luciferase assay and transfection 1E5HeLaT cells on 24-well plates were transfected with 200ng pmirGLO plasmid for luciferase reporter analysis by Lipofectamine 3000 on day 0 and harvested on day 2. For recovery experiments, 1.5E5HeLaT parental cells and TENT4 KO cells on 24-well plates were transfected with 20ng null vector or TENT4A / B (1:1 mix) plasmid together with 50ng pmirGLO plasmid by Lipofectamine 3000 on day 0 and harvested on day 2. HEK293T cells were transfected with 100nM siRNA on day 0, and 1.5E5 cells on 24-well plates were further transfected with 100nM siRNA together with 100 or 200ng pmirGLO plasmid by Lipofectamine 3000 on day 2. HEK293T cells were harvested on day 4. For luciferase analysis, cells were lysed and analyzed with the Dual Luciferase Reporter Assay System (Promega) according to the manufacturer's instructions.

[0101] Example 13: Subcellular fractionation 2E6HeLaT cells were harvested and washed with cold PBS. To obtain the cytoplasmic fraction, cells were lysed with 200 μl of cytoplasmic lysis buffer (0.2 μg / μl digitonin (Sigma, D141), 150 mM NaCl, 50 mM HEPES (pH 7.0-7.6), 0.1 mM EDTA, 1 mM DTT, 20 U ml-1 RNase inhibitor, 1x protease inhibitor, 1x phosphatase inhibitor). After centrifugation at 2,000 relative centrifugal force for 5 min at 4 °C, the supernatant was collected and used as the cytoplasmic fraction. For membrane and nuclear fractionation, the Subcellular Protein Fractionation Kit (Thermo Scientific, 78840) was used according to the manufacturer's guidelines.

[0102] <Example 14: Hire-PAT analysis> Hire-PAT analysis and signal processing of capillary electrophoresis data were performed as previously described [3,57] with minor adaptations such that total RNA was G / I-tailed with yeast poly(A) polymerase (Thermo Scientific, 74225Z25KU). The poly(A) sites of the firefly luciferase gene were confirmed by Sanger sequencing, and the forward PCR primers are listed in Table 4.

[0103] [Table 4]

[0104] Example 15: Phylogenetic analysis To confirm putative stem-loop binding proteins, homologues of the RNA-binding SAM domains of Drosophila Smaug (UniProtKB Q23972) and human SAMD4A / B (UniProtKB Q9UPU9 and Q5PRF9) were searched using UniProt BLAST with an E-value critical value of 10 and were not gapped

[58] . Protein sequences used in the analysis: UniProtKB Q08831 (S. cerevisiae), UniProtKB Q9P6R7 (S. pombe), UniProtKB Q5AI80 (C. albicans), UniProtKB O76699 (C. elegans), UniProtKB Q23972 (D. melanogaster), UniProtKB E7F85E7FBA1, A0A0R4IUM4 (D. rerio), UniProtKB Q6GLT9, Q5FWP2, A0A1L8GL36 (X. laevis), UniProtKB Q8CBY1, Q80XS6, Q8VIG0 (M. musculus) and UniProtKB Q9UPU9, Q5PRF9, Q8WYQ9 (H. sapiens). MUSCLE with basic parameters v.3.8.31

[59] was used for multiple sequence alignment of the 17 generated protein sequences. Phylogenetic trees were reconstructed using PhyML v.3.1 / 3.0aLRT

[60] and visualized using TreeDyn v.198.3

[61] on the Phylogeny.fr platform

[62] . Edges and leaves of the reconstructed phylogenetic tree were realigned and manually rooted for visualization.

[0105] Example 16: RNA pull-down analysis 3′TEG (tetraethylene glycol spacer)-biotinylated RNA of RNA2.7 stem-loop (SL2.7) and stem-loop mutant (SL2.7 mutant) were synthesized (Bioneer Inc.), and the sequences are shown in Table 5.

[0106] [Table 5]

[0107] Streptavidin M-270 beads (Thermo Scientific, 65305) for set 1 and streptavidin magnetic beads (NEB, S1420S) for set 2 were washed twice with pull-down buffer (50 mM Tris (pH 8.0), 150 mM NaCl, 5% glycerol, 1 mM DTT, 100 U ml-1 RNase inhibitor, 1x protease inhibitor). Then, 10 μg of oligonucleotides of SL2.7 or its mutants were conjugated to the streptavidin beads in pull-down buffer overnight at 4°C, followed by washing with pull-down buffer. HEK293T cells from four 150 mm dishes were harvested and lysed in 1.5 ml pull-down buffer, followed by sonication. After centrifugation, the supernatant was incubated with the beads and washed with wash buffer No. 1 (50 mM Tris pH 8.0, 300 mM NaCl, 5% glycerol), wash buffer No. 2 (50 mM Tris pH 8.0, 150 mM NaCl, 5% glycerol, 0.1% Triton X-100) and pull-down buffer. For elution, the beads were incubated at 1,600 rpm in a thermomixer with 60 μl of elution buffer (100 mM pH 7.5, 4% SDS, 100 mM DTT) at 65 °C for 10 min. Then, 10 and 50 μl of the eluate were used for Western blotting and mass spectrometry, respectively.

[0108] <Example 17: Liquid chromatography-tandem mass spectrometry (LC-MS / MS) analysis> The immunoprecipitated protein samples containing SDS were applied to filter-assisted sample preparation digestion. Briefly, the protein samples were first reduced and alkylated with denaturing ABC buffer (8 M urea in 50 mM ammonium bicarbonate (ABC)). The alkylated samples were placed on preconditioned 30 kDa MWCO Amicon filters (Millipore, UFC5030) and centrifuged at 14,000 g for 30 min. Successive washes with 200 μl of 8 M urea or 50 mM ABC buffer were performed and centrifuged at 14,000 g for 15 min to remove SDS from the filter units. The protein samples were then digested overnight with 2% (w / w) trypsin at 37°C. The generated peptide samples were applied to C18 zipper tip washes (Millipore, Z720070) and LC-MS / MS analysis. A self-packaged long capillary column (100 cm i.d. × 75 μm) containing 3 μm Jupiter C18 particles (Phenomenex) and a trap column (3 cm i.d. × 150 μm) were used for peptide separation. A linear gradient ranging from 95% solvent A (water containing 0.1% formic acid) to 40% solvent B (acetonitrile containing 0.1% formic acid) at a flow rate of 300 nl / min for 100 min was run on a nanoACQUITY HPLC system coupled to an Orbitrap Fusion Lumos mass spectrometer (Thermo Scientific). A UPLC (Waters) was run using the following parameters: precursor scan range m / z 300-1,800, precursor isolation window 1.4 Th, 30% normalized collision energy for higher energy collision dissociation (HCD), dynamic exclusion period 30 s, and 60,000 or 75,000 resolution at m / z 200 for full MS or MS / MS scan dp, respectively.

[0109] Mass spectrometry raw data files were processed with MaxQuant (v.1.5.3.30) using MaxQuant's built-in Andromeda search engine

[65] against the human Uniprot database

[58] (v.12 / 5 / 2018) at basic settings (20 or 6 ppm precursor ion mass tolerance for initial or main search, respectively, and 0.5 Da for intercept ion mass). Enzyme specificity was set to trypsin / P, and a maximum of two missing splits were allowed. Cysteine ​​carbamidomethylation and methionine oxidation were selected as fixed and variable variants, respectively. A 1% FDR was required at both the protein and peptide levels. Mass spectrometry proteomics data were deposited in the ProteomeXchange Consortium through the PRIDE

[66] partner repository with dataset identifiers PXD018061 and 10.6019 / PXD018061. Considering the results from 'beads only' and mutant constructs as technical background to account for spurious or non-specific binding to the SL2.7 construct, a customized statistical test for protein spectrum enrichment was designed. Let Np be the number of spectra identified for protein p from the SL2.7 construct and Mp be the number from the background construct. For each protein p where Np > 1, the statistical significance of enrichment (or P value) was calculated:

[0110] [ka]

[0111] Where:

[0112] [ka]

[0113] and B(k;n,p) is the probability of success for P, a binomial distribution with k successes in n attempts. Finally, P values ​​were adjusted with the Benjamini-Hochberg method. Proteins with adjusted P<0.01 were considered statistically significant when using the technical context ('beads only' or mutants). Proteins that were statistically significant in either the Set 1 or Set 2 datasets were considered for subsequent protein intensity analysis. To process the results, the minimum non-zero intensity value of the dataset was substituted. Protein enrichment scores were then calculated as the log2 fold change of protein intensity in SL2.7 relative to the mutant data.

[0114] <Result> <1. HBV and HCMV use mixed tailing tools.> To investigate and characterize viral RNA tailing, we first applied TAIL-seq to two viral infection models: HepG2.2.15 cells integrated with HBV DNA and primary human foreskin fibroblasts infected with HCMV. HepG2.2.15 cells are derived from the hepatoblastoma cell line HepG2 and were used to study the life cycle of HBV as they support the production and assembly of HBV particles. TAIL-seq data for HepG2.2.15 cells showed very high guanylation frequency for viral RNA (Fig. 1a, Fig. 7a). Mixed tailing frequency (G, U and C counts) was also substantially higher in viral RNA than in cellular RNA (Fig. 7a). TAIL-seq data for HCMV-infected HFF cells also showed high guanylation and mixed tailing frequency for HCMV RNA (Fig. 1b and Fig. 7b). The variation was mainly due to RNA2.7 (Fig. 7c, VRNA2.7), one of the four abundant and conserved long non-coding RNAs of HCMV. Both HBV and HCMV RNAs have much longer tails than cellular mRNAs (Fig. 1c, d), indicating that such viral RNAs may undergo slower net deadenylation compared to cellular mRNAs.

[0115] To examine whether viral mixed tailing is mediated by the same enzymes as the host counterpart, we depleted TENT4A and TENT4B. The guanylation and mixed tailing frequencies of HBV mRNA and HCMV RNA2.7 were reduced in TENT4A- and TENT4B-depleted cells (Fig. 1e, 1f, and 7d-g). Knockdown of TENT4A and TENT4B also shortened the poly(A) tail of viral RNA (Fig. 7h-j), indicating that the TENT4 enzyme contributes to the elongation of viral RNA tails. Uniformly, HBV mRNA was downregulated and destabilized after TENT4 depletion (Fig. 1g, 1h, and 7k), indicating that HBV uses TENT4 to stabilize mRNA. Notably, tail modifications such as oligo-uridylation (Fig. 7l) were not increased in the absence of TENT4, excluding a possible compensation mechanism. The highly stable HCMV RNA2.7 showed no significant change in RNA half-life within the experimental time window (Fig. 7m, P>0.05; two-tailed Student's t test), suggesting an additional downstream mechanism following deadenylation that blocks RNA2.7 decay.

[0116] 2. HBV RNA is the major substrate for TENT4 in HepG2.2.15 cells. To elucidate the mechanism of TENT4 recruitment, we performed formaldehyde-mediated cross-linking and immunoprecipitation sequencing (fCLIP-seq) in HepG2.2.15 cells (Figure 2). Formaldehyde-based cross-linking effectively and robustly captures RNA-protein interactions even when they are transient and structure-dependent. Four libraries were produced with input lysates and immunoprecipitates using normal mouse IgG (NMG), TENT4A antibody or TENT4B antibody. Input and NMG libraries were used as size-matched controls to estimate background. Read coverage normalized across the HBV genome confirmed substantially abundant peaks from the HBV genome using both TENT4A and TENT4B antibodies (Figure 2a), indicating the interaction site of TENT4 on HBV RNA. Independent experiments with TENT4B antibody showed similar results (Figure 8a).

[0117] HBV RNA is generated from a 3.2 kilobase circular DNA genome. Five overlapping viral transcripts share a common 3'-terminal sequence (~700 base pairs) and a single polyadenylation site (Figure 2a). TENT4 binding reads are enriched in the 3'-terminal region reaching 1,154-1,687 base pairs (Figure 2a). This region corresponds to a post-transcriptional regulatory element (PRE), which is well known to enhance RNA nuclear release and stabilization. Woodchuck hepatitis virus (WPRE) PRE has been widely used to increase ectopic gene expression and lentiviral gene transfer. Several RNA-binding proteins, such as La, PTB, GAPDH, and ZC3H18, have been reported to interact with PRE, but the functional significance of their interactions is unclear. In our fCLIP-seq experiments, HBV PRE showed the highest enrichment score compared to other predicted peak clusters of human transcripts (Figure 2b). The interaction between TENT4A, TENT4B and HBV mRNA was confirmed by quantitative PCR using immunoprecipitation and subsequent reverse transcription (RT-qPCR) (Figure 8b). In summary, these results demonstrate that HBV mRNA is a major substrate of TENT4 in HepG2.2.15 cells and that TENT4 interacts with HBV mRNA through the PRE region.

[0118] 3. The stem-loop α of HBV RNA is required for TENT4-dependent tail regulation. To identify the cis-acting RNA elements recognized by TENT4, we generated reporters with different subregions of PRE in the 3' untranslated region (UTR) (Fig. 3a). PRE consists of two subelements, PREα (1154-1351) and PREβ (1352-1687) (Fig. 3a). We also generated TENT4A and TENT4B knockout (KO) cells to test TENT4A and TENT4B dependency (Fig. 9a).

[0119] Luciferase protein and RNA levels were significantly upregulated in a TENT4-dependent manner only when the reporter was loaded with either the intact PRE or PREα (Fig. 3b and Fig. 9b). Similar results were obtained in human embryonic kidney 293T (HEK293T) cells depleted of TENT4 (Fig. 9c and Fig. 9d), confirming that PREα is required for TENT4-dependent stabilization. Poly(A) tail length measurements by Hier-PAT analysis showed that TENT4 depletion shortened the poly(A) tail in the PREα but not the PREβ construct (Fig. 3c).

[0120] We show that TENT4 acts directly through tail modification in a PREα-dependent manner. Furthermore, in knockout cells, PRE and PREα reporter expression was rescued by ectopically expressed TENT4 at both the protein and RNA levels (Fig. 3d, Fig. 9e, and Fig. 9f). A catalytically inactive mutant failed to restore activity. Thus, the catalytic activity of TENT4 is required for upregulation of PREα-containing mRNAs.

[0121] To investigate the mechanistic conservation of PRE, we used the woodchuck hepatitis virus (WHV) segment (WPRE), which is homologous to the HBV PRE (Fig. 3e). Unlike the HBV PRE, the WPRE is composed of three subelements: WPREγ, WPREα, and WPREβ. WPREα and WPREβ are conserved between WHV and HBV, whereas WPREγ is unique and is thought to be responsible for the greater activity of WPRE compared to HBV PRE. All constructs with WPREα (Wα) showed TENT4-dependent upregulation (Fig. 3f), suggesting that the TENT4-mediated mechanism is conserved in WHV.

[0122] The combination of WPREγ (Wγ) and WPREα (Wα) elicited a synergistic effect (FIG. 3f), suggesting that WPREγ can act as an enhancer for TENT4-mediated regulation of WHV.

[0123] Additional point mutations were introduced to identify core motifs in PREα (Fig. 3a). PREα contains a La-binding motif and stem-loop alpha (SLα), both of which were reported as necessary for RNA release and stability. Recent mutagenesis studies have shown that the La-binding site and apical loop sequence of SLα are essential for DHQ-1-mediated HBV gene expression. Notably, SLα contains a CAGGU pentaloop that essentially adopts a CNGG(N) family loop structure as a single protruding G residue flanked by A-helical regions (Fig. 3g, left). The structure of SLα is evolutionarily conserved in WHV (Fig. 3g, right). To test whether such elements are necessary for TENT4-dependent regulation, we generated PREα mutant reporters (Fig. 3a) that disrupt the apical sequence and motifs of SLα (Fig. 3g, arrows). Mutagenesis of the La-binding motif (α mut1) had no effect on TENT4 dependence, and the SLα mutation (α mut2) and double mutation (α mut3) disrupted PREα function (Figure 3h). Thus, the stem-loop of PREα (SLα) is an essential element for posttranscriptional regulation by TENT4.

[0124] 4. HCMV RNA2.7 possesses a similar stem-loop for TENT4-dependent regulation. In addition to HBV mRNA, HCMV RNA2.7 (VRNA2.7) also undergoes mixed tailing (Fig. 1b, 1d, and 7b). RNA immunoprecipitation experiments using TENT4A antibody followed by RT-qPCR showed that TENT4A interacts with ectopically expressed RNA2.7 in HEK293T cells (Fig. 4a and 10a), suggesting that viral factors are not required for the interaction between TENT4A and RNA2.7. To confirm the functional domain of RNA2.7 responsible for TENT4 interaction, reporters containing partially overlapping fragments of RNA2.7 were generated (Fig. 4b, FRG1-6). Protein and RNA levels of the first two reporters (FRG1 and FRG2) were significantly higher in parental cells compared to TENT4 KO cells (Fig. 4c and 10b). Thus, cis elements are present near the 5' end. Partial deletion of FRG1 revealed that only the 1E fragment (414-513) confers TENT4-dependent potentiation effects (Fig. 4d, Fig. 4e, Fig. 10c, and Fig. 10d). The 1E domain possesses a stem-loop structure (Fig. 4b, bottom right corner) similar to the CNGG(N) family loop morphology of SLα in PREα and WPREα (Fig. 3g). Point mutations in the stem-loop of 1E (1E mut) disrupted regulation of TENT4 (Fig. 4e and Fig. S10d). A smaller 50 nt truncation fragment containing the stem loop (414-463, henceforth referred to as L2.7) has activity similar to FRG1 and 1E (Fig. S10e). SLα of HBV PRE and WPRE is near the 3' end of the RNA, whereas SL2.7 of HCMV RNA2.7 is near the 5' end, suggesting that the pentaloop may function in a position-independent manner.

[0125] In summary, the present invention demonstrates the functional importance of SLα-like structures in TENT4-dependent regulation.

[0126] <5. Cytoplasmic ZCCHC14 binds to a stem-loop structure and interacts with TENT4.> The CNGGN pentaloop was initially characterized as a Smaug recognition element that binds to the sterile alpha motif (SAM) domain of yeast Vts1p, and Drosophila Smaug mediates post-transcriptional repression [32-36]. Based on structural similarity, Schwalbe et al. proposed that the HBV SLα pentaloop could serve as a binding site for the SAM domain protein. However, TENT4 does not contain a SAM domain

[31] , and our tailing analysis using recombinant TENT4 failed to detect any notable affinity for PRE-in vitro, which would indicate an RNA-interacting cofactor.

[0127] To search for potential cofactors that recognize SLα and SL2.7, we performed RNA pull-down with biotinylated synthetic SL2.7 and HEK293T cell extracts (Fig. 5a). Loop mutants served as negative controls. Mass spectrometry showed six significantly enriched proteins, namely TENT4A, TENT4B, K0355, SAMD4A, SAMD4B, and ZCCHC14 (Fig. 5a). Notably, three human proteins with Smaug-like SAM domains are included in this catalog: two homologs of Smaug (SAMD4A and SAMD4B) and ZCCHC14 (Fig. 5b). The specific enrichment of ZCCHC14 was further confirmed by RNA pull-down and Western blotting (Fig. 5c). K0355 (or KIAA0355) does not have a SAM domain and has been reported to interact with SAMD4B37, indicating that K0355 does not have to directly interact with the pentaloop. To confirm which SAM domain protein is responsible for TENT4-dependent regulation, we reexamined HBV-producing HepG2.2.15 cells. Knockdown of ZCCHC14 did not reduce HBV RNA levels in knockdown of SAMD4 protein (Figure 5d and Figure 11). Uniformly, a genome-wide CRISPR screen recently reported ZCCHC14 as an essential host factor for HBV surface antigen production.

[0128] To understand the mechanism of action of ZCCHC14, we performed TAIL-seq experiments in ZCCHC14-depleted HepG2.2.15 cells. The guanylation frequency of HBV mRNA was reduced (Fig. 6a) and the poly(A) tail was substantially shortened in ZCCHC14 knockdown (Fig. 6b), verifying the essential role of ZCCHC14 in HBV RNA mixed tailing. Uniformly, ZCCHC14 depletion reduced luciferase expression from reporters harboring the CNGGN stem loop (RNA2.7 1E, PREα, Wγ+Wα) (Fig. 6c). In contrast, the respective pentaloop mutants (1E mut, α mut2, Wγ+Wα mut) showed no significant change after ZCCHC14 knockdown (P>0.05; two-tailed Student's t-test) (Fig. 6c), indicating that the function of ZCCHC14 depends on the CNGGN motif. We also performed Hire-PAT analysis and confirmed that the poly(A) tail length of reporter mRNAs was reduced in ZCCHC14-depleted cells in a CNGGN motif-dependent manner (Figures 12a to 12c).

[0129] RNA immunoprecipitation using ZCCHC14 antibody confirmed the interaction between ZCCHC14 and HBV transcripts in HepG2.2.15 cells (Fig. 6d and Fig. 12d). Furthermore, in both HepG2.2.15 and primary HFF cells, ZCCHC14 was co-immunoprecipitated with TENT4A and TENT4B regardless of RNase treatment (Fig. 12e), indicating an RNA-independent interaction between ZCCHC14 and TENT4. Finally, we found that ZCCHC14 was mainly restricted to the cytoplasm (Fig. 6e), and that ZCCHC14 was co-immunoprecipitated with TENT4A and TENT4B using the cytoplasmic fraction of HeLaT cells (Fig. 6f). In summary, the present invention shows that ZCCHC14 interacts with TENT4 mainly in the cytoplasm to inadvertently protect viral transcripts containing SLα-like elements from deadenylation.

[0130] We have uncovered a mechanism for targeted and mixed tailing in viral gene expression (Fig. 6g). HBV mRNA and HCMV RNA2.7 contain cis-acting elements with CNGGN pentaloops that are important for recruiting ZCCHC14 and TENT4. The mixed tail by TENT4 protects the viral RNA against the CNOT deadenylase complex. It is intriguing that these two distinct viruses (Hepadnaviridae and Herpesviridae) with very different life cycles and tissue specificities presented the same strategy for gene expression advantage. Such convergent evolution and the simplicity of this cis-acting element suggest that this mechanism can also be exploited in other viruses.

[0131] In addition to the mixed tailing mode of action, TENT4B acts as a component of the nuclear TRAMP complex (Trf4-Air1 / 2-Mtr4 in yeast and TENT4B-ZCCHC7-hMTR4 in humans) and is involved in the degradation or cleavage and maturation of aberrant transcripts or in the trimming and maturation of small noncoding RNAs in the nucleus [5,39-43]. We have previously shown that mixed tailing contributes to mRNA stability as a module distinct from the TRAMP complex. Given that knockdown of hMTR4 and ZCCHC7 does not affect HBV protein and mRNA levels, it is likely that TENT4 acts independently of the TRAMP complex in HBV regulation. The cytoplasmic localization of ZCCHC14 also suggests that the TENT4-ZCCHC14 complex operates in a subcellular compartment distinct from the nuclear TRAMP complex, although it does not exclude the possibility that a minor fraction of the TENT4-ZCCHC14 complex operates in the nucleus.

[0132] Mixed tailing appears to be particularly important for HBV gene expression. Four of the five mRNA species (non-X pgRNA, Precore, PreS1, and PreS2 / S) contain SLα stem loops, and their half-life is substantially reduced if the TENT4 enzyme is depleted. Current treatment options include nucleos(t)ide analogs and interferon alpha, but HBV infection remains a major healthcare burden with approximately 80 million deaths worldwide each year due to limited and often inefficient treatment. Functional treatment can be achieved by blocking the production of viral proteins, especially viral S antigen (HBsAg), which induces immune evasion. The discovery of viral mixed tails and the involvement of TENT4 and ZCCHC14 in viral gene expression can provide mechanistic insights for the development of a new class of anti-HBV drugs. To avoid potential side effects of targeting the TENT4 mode of action, it is important to understand the intrinsic functions and mechanisms of action of the TENT4 protein and its partners.

[0133] The physical interaction between the CNGGN pentaloop and RNA-binding SAM proteins was identified as a highly conserved and widely used functional module. In the fruit fly, Smaug recognizes a hairpin in the CUGGC loop of nanos mRNA to induce timely mRNA decay, which is essential for the setting of the anterior-posterior axis. In yeast, Vts1p binds mRNAs with CNGG(N) pentaloops for targeted mRNA decay. Vertebrates, which have three RNA-binding SAM proteins, are considered to extend the use of this interaction module. One of the vertebrate orthologs of Smaug, SAMD4B, targets nanos1 mRNA and is essential during mammalian neurodevelopment. Another vertebrate ortholog of Smaug, SAMD4A, represses the translation of a pentaloop-containing reporter and is associated with the formation of cytoplasmic foci. Here, we show that the same mechanism can be exploited for the opposite purpose through the vertebrate-specific protein ZCCHC14, specifically to the benefit of viral RNAs. [Brief description of the drawings]

[0134] [Figure 1] The viral RNA shows extensive mixed tails. [Diagram 2] We present experimental results demonstrating that HBV RNA is a major mRNA substrate for TENT4 through the HBV PRE. [Diagram 3] FIG. 1 shows that the stem-loop structure of the PRE is essential but not sufficient for TENT4-dependent RNA tailing of HBV mRNA. [Figure 4] FIG. 1 shows that the stem-loop structure of HCMV RNA2.7 is a cis-acting RNA element responsible for TENT4-dependent regulation. [Diagram 5] FIG. 1 shows that SAM domain-containing proteins, including ZCCHC14, bind to stem-loop structures and regulate HBV mRNA. [Figure 6] FIG. 1 shows that cytoplasmic ZCCHC14 recruits TENT4 to protect RNA at a stem-loop structure. [Figure 7] FIG. 1 shows the extensive 3′ end tail modifications of viral RNA. [Figure 8] FIG. 1 shows the results of analysis and confirmation of TENT4A / B and HBV mRNA interaction. [Figure 9] FIG. 1 shows the results of confirming TENT4-dependent regulation of a PRE reporter. [Figure 10] 1 shows the results of confirming TENT4-dependent regulation of the RNA2.7 reporter. [Figure 11] Figure confirming ZCCHC14 and SAMD4A / B knockdown. [Figure 12] FIG. 1 shows confirmation of ZCCHC14 knockdown, stem-loop-dependent tail regulation, and immunoprecipitation. [Figure 13] A stem-loop structure including a pentaloop is shown.

Claims

1. A method for stabilizing RNA in a vertebrate, comprising inserting in vitro into a target RNA sequence a stem-loop sequence comprising a pentaloop structure comprising an RNA sequence represented by the following general formula: [General formula] 5'-CNGGN-3' each N is independently selected from adenosine (A), uracil (U), guanine (G), and cytosine (C); The stem-loop sequence comprises 8 or more base pair bonds; and wherein said stabilization of the RNA is effected by mixed tailing.

2. The method of claim 1 , wherein the mixed tailing is induced by interaction of vertebrate TENT4A or TENT4B with ZCCHC14.

3. The method of claim 1, wherein the stem-loop sequence comprises 8 to 20 base pair bonds.

4. The method of claim 1, wherein the stem-loop sequence comprises 8 to 15 base pair bonds.

5. The method of claim 1, wherein the stem-loop sequence comprises the nucleic acid sequence from bases 414 to 463 of HCMV RNA 2.7 or the nucleic acid sequence from bases 1276 to 1351 of HBV RNA.

Images