PHARMACEUTICAL COMPOSITION FOR PREVENTING OR TREATING CANCER, COMPRISING MIR-324 and TUT4/7 EXPRESSION MODULATORS

Abstract

The present invention relates to a pharmaceutical composition for preventing or treating cancer, comprising miR-324 and TUT4/7 expression modulators, wherein the pharmaceutical composition of the present invention can prevent cell division and hinder cancer development by inhibiting the function of TUT4/7, and can increase the amount of miR-324-5p and suppress the function of miR-324-3p.1, and therefore can be effectively used for prevention, treatment, or diagnosis of cancer.

Claims

1. Method of treating cancer, comprising administering to a subject in need thereof a pharmaceutical composition for preventing or treating cancer, the composition comprising a microRNA(miRNA) which is comprising i) a miR-324-5p or an isoform thereof, and ii) a miR-324-3p or an isoform thereof: wherein the i) miR-324-5p comprising the nucleotide sequences of SEQ ID NO: 47 at the 5′ end, and a total of 20 to 24 nucleotide sequences; wherein the ii) miR-324-3p is, a) the first base from the 5′ end may be C or G; b) the first base from the 5′ end is complementary to the 20th base from the 5′ end of the miR-324-5p; and c) the 20th base from the 5′ end may be A, U, or C.

2. The method of claim 1, wherein the microRNA is in the form of a duplex in which i) miR-324-5p or isoform thereof, and the ii) miR-324-3p or isoform thereof are bounded to each other.

3. The method of claim 1, wherein the 5′ end of the i) miR-324-5p or isoform thereof, and the ii) miR-324-3p or isoform thereof bonded to phosphate, respectively.

4. The method of claim 1, wherein the i) miR-324-5p or isoform thereof, comprising the nucleotide sequences of SEQ ID NO: 1 or 32.

5. The method of claim 1, wherein the ii) miR-324-3p or isoform thereof, comprising the nucleotide sequences of SEQ ID NO: 2, 3, or 33.

6. The method of claim 1, wherein the cancer is selected from the group consisting of colon cancer, large intestine cancer, lung cancer, liver cancer, stomach cancer, gallbladder cancer, kidney cancer, uterine cervical cancer, endometrial cancer, ovarian cancer, breast cancer, brain cancer, head and neck cancer, and malignant melanoma.

7. A method of treating cancer, comprising: administering to a subject in need thereof a pharmaceutical composition for preventing or treating cancer, the composition comprising at least one selected from the group consisting of: a nucleic acid composed of the nucleotide sequence of SEQ ID NO: 4; and a small interfering RNA (siRNA) comprising at least one nucleotide sequence selected from the group consisting of SEQ ID NOS: 5 to 20.

8. The method of claim 7, wherein the small interfering RNA is in a duplex form in which: i) a small interfering RNA comprising any one of the nucleotide sequences of SEQ ID NOS: 5 to 8 and a small interfering RNA comprising any one of the nucleotide sequences of SEQ ID NOS: 9 to 12 bind to each other; or ii) a small interfering RNA comprising any one of the nucleotide sequences of SEQ ID NOS: 13 to 16 and a small interfering RNA comprising any one of the nucleotide sequences of SEQ ID NOS: 17 to 20 bind to each other.

9. The method of claim 7, wherein the cancer is selected from the group consisting of encephaloma, colon cancer, large intestine cancer, lung cancer, liver cancer, stomach cancer, esophageal cancer, pancreatic cancer, gallbladder cancer, kidney cancer, bladder cancer, prostate cancer, testis cancer, uterine cervical cancer, endometrial cancer, choriocarcinoma, ovarian cancer, breast cancer, thyroid cancer, brain cancer, head and neck cancer, and malignant melanoma.

10. A method of diagnosing cancer, comprising measuring a level of: i) at least one miRNA selected from the group consisting of miR-324-5p and miR-324-3p.1; or ii) at least one gene mRNA or protein thereof selected from the group consisting of TUT4 (terminal uridylyl transferases 4) and TUT7 (terminal uridylyl transferases 7).

11. The method of claim 10, wherein the measuring step is performed by an agent that measures the level of miRNA; wherein the agent comprises a primer or a probe which bind specifically to the at least one miRNA.

12. The method of claim 10, wherein the measuring step is performed by an agent that measures the level of gene mRNA; wherein the agent for measuring an mRNA level comprises a primer or a probe which bind specifically to at least one mRNA of TUT4 and TUT7.

13. The method of claim 10, wherein the measuring step is performed by an agent that measures the level of protein; wherein the agent for measuring a protein level comprises an antibody binding specifically to at least one protein of TUT4 and TUT7.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0101] FIGS. 1a, 1b, 1c and 1d show results of arm switching of miR-324: in FIG. 1a: scatter plot of median absolute deviation of the 5p proportion for a miRNA (abundant and ubiquitous miRNAs (>100 median RPM in both datasets, >0 RPM in all samples) were included in this analysis); FIG. 1B: IsomiR profiles of miR-324 in HEK293T cells, uncovered by AQ-seq. RPM-normalized read counts are denoted on the left. Reference sequences of 5p and 3p are marked by gray shade; FIG. 1c: Strand ratios (5p/3p.1) of miR-324 for the indicated panel of mouse tissues measured by AQ-seq; and FIG. 1d: Conservation of the 5′-isomiRs of miR-324 in mammals.

[0102] FIGS. 2a, 2b, 2c, 2d and 2e show results of miR-324 arm switching controlled by TUT4/7: FIG. 2a: Negative association between the uridylation frequency of miR-324-3p and the 5p/3p.1 ratio of miR-324 in 9 human cell lines and 15 mouse tissues (The linear regression is shown with dashed lines. rs, Spearman correlation coefficient); FIG. 2b: The relative TUT4/7 levels and the 5p/3p.1 ratios of miR-324 in the indicated panel of mouse tissues. The TUT4/7 mRNA levels and 5p/3p.1 ratios were quantified by RT-qPCR and AQ-seq, respectively. rs, Spearman correlation coefficient; FIG. 2c: AQ-seq results following knockdown of TUT4/7 in HEK293T cells (left and middle: scatter plots of uridylation frequency and the 5p proportion of miRNAs. Abundant miRNAs (>100 RPM) in siNC-transfected sample were included in the analysis. Right: relative abundance of three 5′-isomiRs and log 2-transformed 5p/3p.1 ratio of miR-324)); FIG. 2d: Northern blot of miR-324-5p and miR-324-3p in HEK293T cells (synthetic miR-324 duplex was used for size references); and FIG. 2e: sRNA-seq results following double knockout of Tut4/7 in mice (Left: abundance changes of 5′-isomiRs by TUT4/7 depletion in bone marrow. 5′-isomiRs with RPM over 10 in the control samples were included in this analysis. P-value by the two-sided Student's t test. Right: log 2-transformed 5p/3p.1 ratio of miR-324. Bars indicate mean±s.d. Bone marrow, n=2; embryonic fibroblasts, n=2; embryonic stem cells, n=3; control and Tut4/7 dKO in liver, n=4 and 3, respectively. *p<0.05, ***p<0.001 by two-sided Student's t test).

[0103] FIGS. 3a, 3b and 3c show results of uridylation-induced alternative Dicer processing of pre-miR-324: FIG. 3a: In vitro processing of unmodified, mono- or di-uridylated forms of pre-miR-324 by immunopurified Dicer (Synthetic miR-324-5p (23 nt), marked in blue, was used as a size control. Major cleavage products and their corresponding cleavage sites are marked with arrowheads. *: radiolabeled 5′ phosphates); and FIG. 3b: 5′-isomiR composition of miR-324-3p in HEK293T cells; and FIG. 3c: 5′-isomiR composition of miR-324-3p in mice (Bars indicate mean. Bone marrow, n=2; embryonic fibroblasts, n=2; embryonic stem cells, n=3; control and Tut4/7 dKO in liver, n=4 and 3, respectively).

[0104] FIGS. 4a, 4b, 4c and 4d show results of alternative Dicer processing facilitated by double stranded RNA-binding domain (dsRBD): FIG. 4a: In vitro processing of unmodified, mono- or di-uridylated forms of wild-type or no-bulge mutant pre-miR-324 by immunopurified Dicer; FIGS. 4b and 4c: In vitro processing of unmodified, mono- or di-uridylated forms of pre-miR-324 by immunopurified Dicer pocket mutants (4b) or dsRBD-deleted mutant (4c) (Major cleavage products and their corresponding cleavage sites are marked with arrowheads. *: radiolabeled 5′ phosphates; †: nicked products at the 3p positions); and FIG. 4d: A proposed model for uridylation-mediated alternative Dicer processing of pre-miR-324.

[0105] FIGS. 5a and 5b show results of alternative Dicer processing-induced arm switching: FIG. 5a: A schematic diagram of strand selection of miR-324 duplexes produced from unmodified and uridylated pre-miR-324 (Dashed rectangles indicate end properties that dictate the strand selection of each duplex); and FIG. 5b: Luciferase reporter assay with two miR-324 duplexes (Left: an illustration of firefly reporter mRNAs that have three 8mer target sites of either miR-324-5p or 3p.1. Right: the relative reporter activity from the firefly luciferase in response to two miR-324 duplexes in HEK293T cells. The reporter activity from Renilla luciferase was used as a control. Bars indicate mean±s.d. (n=2, biological replicates). **p<0.01 by two-sided Student's t test).

[0106] FIGS. 6a, 6b, 6c, 6d, 6e and 6f show results of the cell cycle progression controlled by miR-324 arm switching: FIG. 6a: ratios of 5′-isomiR compositions of miR-324-3p in glioblastoma and normal brain tissues (normal brain, n=3; and glioblastoma, n=6); FIG. 6b: Expression levels of TUT4/7 in normal, lower-grade glioma, and glioblastoma tissues from the Oncopression database (normal, n=723; grade 1, n=74; grade 2, n=133; grade 3, n=132; glioblastoma, n=865. ***p<0.001 by one-way ANOVA with Tukey's post hoc test for multiple comparisons); FIG. 6c: Negative correlation between the TUT4/7 expression level and the miR-324-5p/3p ratio in glioblastoma (black) and matched normal brain tissues (white) (the linear regression is shown with dashed lines. rs: Spearman correlation coefficient); FIG. 6d: Western blot of the indicated cyclin proteins and cell cycle profile after overexpressing the synthetic 5p duplex (long duplex) or 3p.1 duplex (short duplex) of miR-324 in A172 cells; FIG. 6e: Western blot of the indicated cyclin proteins and cell cycle profile after inhibiting miR-324 by locked nucleic acid antisense oligonucleotides in A172 cells; and FIG. 6f: Western blot of the indicated proteins and cell cycle profile after knocking down TUT4/7 in A172 cells (*: a cross-reacting band. PI: propidium iodide, BrdU: bromodeoxyuridine).

[0107] FIG. 7 is a schematic view showing a miR-324 arm switching mechanism model.

DETAILED DESCRIPTION

[0108] A better understanding of the present invention may be obtained via the following examples which are set forth to illustrate, but are not to be construed as limiting the present disclosure.

[0109] Throughout the description, “%” used to indicate a concentration of a specific material refers to (weight/weight) % for solid/solid, (weight/volume) % for solid/liquid, and (volume/volume) % for liquid/liquid unless otherwise stated.

Preparation Examples

[0110] 1. AQ-Seq Library

[0111] AQ-seq (bias-minimized sRNA-seq) libraries were constructed using total RNAs from fifteen mouse tissues (Mouse Total RNA Master Panel; Takara) as described in the previous study (Kim et al., 2019).

[0112] Briefly, 5 mg of total RNAs was mixed with 10 fmole of 30 equi-molar spike-in RNAs (miRNA-like non-human/mouse/frog/fish RNAs used for bias evaluation). Small RNAs were enriched by size fractionation by 15% urea-polyacrylamide gel electrophoresis and sequentially ligated to the randomized adaptor at the 3′ and 5′ends. The ligated RNAs were reverse-transcribed using SuperScript III reverse transcriptase (Invitrogen), amplified using Phusion High-Fidelity DNA Polymerase (Thermo Scientific), and subjected to high-throughput sequencing on the MiSeq platform (Illumina)

[0113] 2. High-Throughput Sequencing of miRNA

[0114] Data processing was performed as described in the previous study (Kim et al., 2019) except that reads of the sRNA-seq results from mouse were mapped to the mouse genome (mm10).

[0115] Briefly, the 3′ adaptorwas clipped from the reads using Cutadapt (Martin, 2011). For AQ-seq data, 4-nt degenerate sequences were further removed with FASTX-Toolkit (http-//hannonlab.cshl.edu/fastxtoolkit/). After filtering out short, low-quality, and artifact reads using FASTX-Toolkit, AQ-seq data were aligned to the spike-in sequences, followed by mapping the unaligned reads to the genome next, while the other sRNA-seq data were mapped to the genome using BWA (Li and Durbin, 2009). For a given read, selection was made of an alignment result with the best alignment score allowing a mismatch only at the 3′ end. miRNA annotations were retrieved using miRBase release 21 by the intersect tool in BEDTools (Kozomara and Griffiths-Jones, 2014; Quinlan and Hall, 2010).

[0116] For quantitative analysis of miRNA strand ratios, the most abundant 5′-isomiR was identified for a given mature miRNA in the most abundantly expressed cell line or tissue. Then, the ratio between the top 5′-isomiRs from 5p and 3p was calculated for all of the cell lines or tissues. Non-repetitive miRNA genes of which both strands were annotated in miRBase release 21 were included in this analysis.

[0117] 3. Targetome Analysis

[0118] Results from two modified versions of AGO CLIP-seq (CLEAR-CLIP and CLASH) were analyzed to identify miR-324 targets (Helwak et al., 2013; Moore et al., 2015).

[0119] To begin with, for analysis of CLEAR-CLIP, the 3′ and 5′ adapters were clipped using cutadapt followed by extraction of reads containing miR-324 sequences. Then, the target RNA sequences were mapped to the mouse genome (mm10). Annotations were retrieved using GENCODE release M19 by the intersect tool in BEDTools.

[0120] For analysis of CLASH, the supplementary file of the CLASH data generated by the protocol E4 was used. Target genes of miR-324-3p were subdivided according to the 5′ end of the miR-324-3p sequences.

[0121] 4. Plasmid Construction

[0122] To construct plasmids for expression of wild-type, 5′ pocket mutant Dicer, and 3′ pocket mutant Dicer, the coding sequence of the Dicer reference (RefSeq NM_030621) was amplified with or without mutations introduced in the previous study (Park et al., 2011). The amplified DNAs were subcloned into the pCK-FLAG vector (CMV promoter-driven vector) at the BamHI and XhoI sites using the In-Fusion HD Cloning Kit (Clontech).

[0123] For double stranded RNA-binding domain (dsRBD)-deleted Dicer, the coding region except the sequence corresponding to V1849-Si922 amino acids was amplified and subcloned into the pCK-FLAG vector in the same way.

[0124] For TRBP, the coding sequence of the human TRBP reference (RefSeq NM_134323) was amplified and subcloned into the pcDNA3-Myc vector (Invitrogen) at the BamHI and XhoI sites.

[0125] To construct plasmids for luciferase assay, synthetic DNA oligos containing the three 8mer target sites of either miR-1-3p, miR-324-5p, or miR-324-3p.1 were amplified and inserted into pmiRGLO (Promega) at the XhoI and XbaI sites. The sequences of synthetic DNA oligos and PCR primers are listed in Table 1, below.

TABLE-US-00001 TABLE 1 Name Sequence (5′ -> 3′) 3X miR-1-3p AGTTGTTTCAATCTACTTCCGAACATTCCACGCGCAATCTACT TCCGAACATTCCACGCGCAATCTACTTCCGAACATTCCAAAA CGAGC (SEQ ID NO: 23) 3X miR-324-5p AGTTGTTTCAATCTACTTCCGAGGGATGCACGCGCAATCTAC TTCCGAGGGATGCACGCGCAATCTACTTCCGAGGGATGCAA AACGAGC (SEQ ID NO: 24) 3X miR-324-3p.1 AGTTGTTTCAATCTACTTCCGAGGGGCAGACGCGCAATCTAC TTCCGAGGGGCAGACGCGCAATCTACTTCCGAGGGGCAGAA AACGAGC (SEQ ID NO: 25) miRNA_forward primer CTCGCTAGCCTCGAGAGTTGTTTCAATCTACTTCC (SEQ ID NO: 26) miR-1-3p_reverse primer CAGGTCGACTCTAGAGCTCGTTTTGGAATGTTCGG (SEQ ID NO: 27) miR-324-5p_reverse primer CAGGTCGACTCTAGAGCTCGTTTTGCATCCCTCGG (SEQ ID NO: 28) miR-324-3p.1_reverse primer CAGGTCGACTCTAGAGCTCGTTTTCTGCCCCTCGG (SEQ ID NO: 29)

[0126] 5. Cell Culture and Transfection

[0127] A172 and U87MG were obtained from Korean Cell Line Bank. HEK293T, A172, and U87MG were maintained in DMEM (WELGENE) supplemented with 10% fetal bovine serum (WELGENE). Primary tumor cells derived from a glioblastoma patient (TS13-64) were established from fresh glioblastoma tissue specimens, as approved by the institutional review board of Yonsei University College of Medicine (4-2012-0212,4-2014-0649).

[0128] For tumorsphere culture, TS13-64 cells were grown in DMEM (WELGENE) supplemented with 1X B-27 (Thermo Scientific), 20 ng/mL of bFGF (R&D Systems), and 20 ng/mL of EGF (Sigma-Aldrich) (Kong et al., 2013).

[0129] To knock down TUT4/7, cells were transfected twice with 20˜22 nM siRNAs using the Lipofectamine 3000 reagent (Thermo Scientific) and harvested 4 days after the first transfection.

[0130] For overexpression of Dicer, HEK293T cells were transfected with pCK-FLAG-Dicer using Lipofectamine 3000 and harvested 2 days post transfection.

[0131] To deliver synthetic miRNAs or inhibitors, cells were transfected with 20 nM of synthetic miRNA duplexes or 80 nM of LNA miRNA inhibitors using Lipofectamine 3000 and harvested 2 days post transfection.

[0132] Control siRNA (Accularget negative control siRNA), siRNAs, control miRNA, and synthetic miR-324 duplexes were obtained from Bioneer. Control miRNA inhibitor (miRCURY LNA miRNA inhibitor negative control A) and miR-324 inhibitors (hsa-miR-324-5p and hsa-miR-324-3p miRCURY LNA miRNA inhibitor) were obtained from QIAGEN. The sequences of synthetic siRNAs and miRNAs are listed in Table 2, below.

TABLE-US-00002 TABLE 2 Sense strand sequence in each complex(5′ -> 3′) Name (underlined: 3′ overhang by deoxyribonucleotides) siTUT4 GGUUGCUUCAGACUUUAUATT (SEQ ID NO: 5) (duplex binding (UAUAAAGUCUGAAGCAACCTT) (SEQ ID NO: 9) sequence) GGAAUGAAGAAGAGAAAGATT (SEQ ID NO: 6) (UCUUUCUCUUCUUCAUUCCTT) (SEQ ID NO: 10) GGAGAAACGACAUAAGAAATT (SEQ ID NO: 7) (UUUCUUAUGUCGUUUCUCCTT) (SEQ ID NO: 11) GGUUAGAGCUGCUUAAAUUTT (SEQ ID NO: 8) (AAUUUAAGCAGCUCUAACCTT) (SEQ ID NO: 12) siTUT7 GAAAAGAGGCACAAGAAAATT (SEQ ID NO: 13) (duplex binding (UUUUCUUGUGCCUCUUUUCTT) (SEQ ID NO: 17) sequence) GCAAAGAGGACAAAGAAAUTT (SEQ ID NO: 14) (AUUUCUUUGUCCUCUUUGCTT) (SEQ ID NO: 18) GAUAAGUAUUCGUGUCAAATT (SEQ ID NO: 15) (UUUGACACGAAUACUUAUCTT) (SEQ ID NO: 19) GAACAUGAGUACCUAUUUATT (SEQ ID NO: 16) (UAAAUAGGUACUCAUGUUCTT) (SEQ ID NO: 20) Name Sequence(5′ -> 3′) miNC Sense: 5Phos/CCUACGCCACCAAUUUCGUUU (SEQ ID NO: 30) Antisense: 5Phos/ACGAAAUUGGUGGCGUAGGUU (SEQ ID NO: 31) miR-324 long duplex 5p: 5Phos/CGCAUCCCCUAGGGCAUUGGUGU (SEQ ID NO: 1) (5p duplex) 3p.2: 5phos/CCACUGCCCCAGGUGCUGCUGG (SEQ ID NO: 2) miR-324 short duplex 5p: 5Phos/CGCAUCCCCUAGGGCAUUGG (SEQ ID NO: 32) (3p.1 duplex) 3p.1 + U: 5phos/ACUGCCCCAGGUGCUGCUGGU (SEQ ID NO: 33) miR-324-5p CGCAUCCCCUAGGGCAUUGGUGU 3 (SEQ ID NO: 1)

[0133] 6. Northern Blot

[0134] Total RNAs were isolated using TRIzol (Invitrogen) and small RNAs (<200 nt) were enriched with the miRVana miRNA Isolation Kit (Ambion). Then, the RNAs were resolved on 15% urea-polyacrylamide gels. Synthetic miR-324 duplex and Decade Markers System (Ambion) were used as size markers. RNAs were transferred to Hybond-NX membranes (Amersham) and crosslinked to the membranes with 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide. Antisense probes were radiolabeled at their 5′ ends with [γ.sup.32P] ATP by T4 polynucleotide (Takara) and purified using Performa Spin Columns (Edge BioSystems). The membranes were incubated with denatured UltraPure Salmon Sperm DNA Solution (Thermo Scientific) in PerfectHyb Plus Hybridization Buffer, hybridized with antisense probes, and washed with mild wash buffer (0.05% SDS and 2×SSC), followed by stringent wash buffer (0.1% SDS and 0.1×SSC). Radioactive signals were detected by Typhoon FLA 7000 (GE Healthcare) and analyzed using the Multi Gauge software (FujiFilm).

[0135] miR-324-3p was detected first, and then miR-324-5p was detected after stripping off the probes. To remove the probes from the blot, the membrane was soaked in preboiled 0.5% SDS for 15 min. Synthetic miR-324 duplex (AccuTarget) was obtained from Bioneer. The sequences of probes are listed in Table 3, below.

TABLE-US-00003 TABLE 3 Name Sequence (5′ -> 3′) miR-324-5p probe ACCAATGCCCTAGGGGATGCG (SEQ ID NO: 34) miR-324-3p probe CAGCAGCACCTGGGGCAGT (SEQ ID NO: 35)

[0136] 7. Quantitative Real-Time PCR (RT-qPCR)

[0137] To measure mRNA levels in mouse tissues, RNAs from Mouse Total RNA Master Panel (Takara) were reverse-transcribed using RevertAid Reverse Transcriptase (Thermo Scientific) and subjected to quantitative real-time PCR with the Power SYBR Green PCR Master Mix (Thermo Scientific) on StepOnePlus Real-Time PCR System (Thermo Scientific). GAPDH was used as an internal control. The sequences of qPCR primers are listed in Table 4, below.

TABLE-US-00004 TABLE 4 Name Sequence (5′ -> 3′) TUT4 forward primer GCCTAAAAGGCGGATAGCCATTG (SEQ ID NO: 36) TUT4 reverse primer GCTGCTCTGAATCTTTCCACAAC (SEQ ID NO: 37) TUT7 forward primer CAGCTCAGCAAAAGAGGAGGCA (SEQ ID NO: 38) TUT7 reverse primer CATAGAACCGCAGCAATTCCACC (SEQ ID NO: 39) GAPDH forward primer ACCACAGTCCATGCCATCAC (SEQ ID NO: 40) GAPDH reverse primer TCCACCACCCTGTTGCTGTA (SEQ ID NO: 41)

[0138] To quantify the miR-324-5p/3p strand ratio, total RNAs were isolated using TRIzol (Invitrogen). Then, cDNAs were synthesized using the TaqMan miRNA Reverse Transcription Kit (Applied Biosystems) and subjected to quantitative real-time PCR with the TaqMan MicroRNA Assay (Applied Biosystems) on StepOnePlus Real-Time PCR System (Thermo Scientific). U6 snRNA was used as an internal control.

[0139] 8. In Vitro Dicer Processing Assay

[0140] Pre-miR-324 and its variants were radiolabeled with [γ.sup.32P] ATP by T4 polynucleotide kinase and purified using Oligo Clean & Concentrator (Zymo Research) according to manufacturer's instructions.

[0141] For immunoprecipitation of FLAG-Dicer, the HEK293T cells overexpressing Dicer proteins were lysed with lysis buffer (500 mM NaCl, 20 mM Tris (pH 8.0), 1 mM EDTA, 1% Triton X-100) and sonicated using Bioruptor Standard (Diagenode). After centrifugation, the supernatant was incubated with 10 μL of ANTI-FLAG M2 Affinity Gel (Sigma-Aldrich). The beads were washed twice with lysis buffer, four times with high salt buffer (800 mM NaCl and 50 mM Tris (pH 8.0)), and four times with buffer D (200 mM KCl, 20 mM Tris (pH 8.0), 0.2 mM EDTA) and then resuspended in 10 μL buffer D.

[0142] The immune-purified Dicer was subjected to in vitro reactions in a total volume of 30 μL (containing 2 mM MgCl.sub.2, 1 mM DTT, SUPERase In RNase Inhibitor (Thermo Scientific) 60 units, 5′-radiolabeled pre-miR-324). The RNAs were purified through phenol extraction or using Oligo Clean & Concentrator (Zymo Research) and resolved on 15% urea-polyacrylamide gels. Synthetic miR-324-5p and Decade Markers System (Ambion) were used as size markers. Synthetic pre-miR-324 fragments were obtained from IDT. Synthetic miR-324-5p was obtained from Bioneer. The sequences of synthetic pre-miR-324 fragments and miR-324-5p are listed in Table 5, below.

TABLE-US-00005 TABLE 5 Name Sequence (5′ -> 3′) Pre-miR-324 CGCAUCCCCUAGGGCAUUGGUGUAAAGCUG 5′ half (SEQ ID NO: 42) No-bulge CGCAUCCCCUAGGGCAUUGGGUAAAGCUG pre-miR-324 (SEQ ID NO: 43) 5′ half Pre-miR-324 5Phos/GAGACCCACUGCCCCAGGUGCUGCUGG 3′ half (SEQ ID NO: 44) Pre-miR-324 5Phos/GAGACCCACUGCCCCAGGUGCUGCUGGU 3′ half + U (SEQ ID NO: 45) Pre-miR-324 5Phos/GAGACCCACUGCCCCAGGUGCUGCUGGUU 3′ half + UU (SEQ ID NO: 46)

[0143] 9. Dual Luciferase Reporter Assay

[0144] pmiRGLO containing three 8-mer target sites of either miR-1-3p, miR-324-5p, or miR-324-3p.1 was co-transfected, together with control miRNA or miR-324 duplexes, into HEK293T cells with the aid of Lipofectamine 3000 (Thermo Scientific). After 2 days of transfection, the cells were harvested and subjected to reporter assay. The reporter activities were measured using Dual-Luciferase Reporter Assay System according to the manufacturer's instructions (Promega) on the Spark microplate reader (TECAN). Given that miR-1-3p is little expressed in HEK293T (˜ 8 RPM in the AQ-seq result) (Kim et al., 2019), pmiRGLO with three 8mer target sites of miR-1-3p was used for a plasmid control. Control miRNA was used for further normalization.

[0145] 10. Gene Expression Analysis of Glioblastoma Patient Data

[0146] From Oncopression (http//oncopression.com), Preprocessed gene expression data using microarray were retrieved (Lee and Choi, 2017). The REMBRANDT gene expression dataset (E-MTAB-3073) was obtained from ArrayExpress (Madhavan et al., 2009).

[0147] For analysis of simultaneous profiling of mRNAs and miRNAs (Gulluoglu et al., 2018), pre-processing microarray data were normalized by the robust multi-array average (RMA) using the limma package in R before being used for gene expression analysis.

[0148] 11. Survival Analysis

[0149] Level 3 miRNA gene quantification data and clinical data for TCGA glioblastoma patients were obtained from the GDC legacy archive and the GDC data portal, respectively. The patients were stratified according to the miR-324-5p/3p ratio and the top and bottom 40% of the cases were included in the analysis. The patient's survival was estimated by the Kaplan-Meier method and tested by the two-sided log-rank test using the survival package in R.

[0150] 12. Western Blot

[0151] Cells were washed with PBS and lysed in RIPA buffer (Thermo Scientific) supplemented with Protease Inhibitor Cocktail Set III (Merck Millipore) and Phosphatase Inhibitor Cocktail II (AG Scientific). Protein concentrations were measured using the BCA Protein Assay Kit (Pierce Biotechnology) and equal amounts of proteins were separated on 4-12% Tris-Glycine Gels (Thermo Scientific) and transferred onto Immobilon-P PVDF membranes (Merck Millipore). Membranes were incubated with 5% skimmed milk in PBS-T (PBS (Amresco)+0.1% Tween 20 (Anatrace)) and then treated with primary antibodies. After being washed three times with PBS-T, the membranes were incubated with HRP-conjugated secondary antibodies. The protein bands were detected by SuperSignal West Pico PLUS Chemiluminescent Substrate (Thermo Scientific) and scanned by ChemiDoc XRS+ System (Bio-Rad).

[0152] Rabbit polyclonal antibodies against TUT4 (18980-1-AP, RRID:AB_10598327, 1:500) and TUT7 (25196-1-AP, 1:500) were obtained from Proteintech. A mouse monoclonal antibody against Cyclin E (sc-247, RRID:AB_627357, 1:1,000) and rabbit polyclonal antibodies against Cyclin A (sc-751, RRID:AB_631329, 1:1,000), Cyclin B1 (sc-752, RRID:AB_2072134, 1:1,000), and Cyclin D1 (sc-753, RRID:AB_2070433, 1:1,000) were purchased from Santa Cruz Biotechnology. A rabbit polyclonal antibody against HSP90 (4874, RRID:AB_2121214, 1:1,000) was purchased from Cell Signaling. HRP-conjugated goat polyclonal antibodies against rabbit IgG (111-035-144, RRID:AB_2307391, 1:10,000) and mouse IgG (115-035-146, RRID:AB_2307392, 1:10,000) were purchased from Jackson ImmunoResearch.

[0153] 13. Flow Cytometry

[0154] After being incubated with 10 μM BrdU for 3-8 hours, cells were fixed by ice-cold 70% ethanol. The fixed cells were incubated with FITC-conjugated anti-BrdU antibody (11-5071-42, RRID: AB_11042627, Invitrogen) and further stained with 20 μg/mL of propidium iodide (Sigma-Aldrich) in the presence of 10 μg/mL of RNase A (Thermo Scientific) before detection on BD Accuri C6 Plus Flow Cytometer. Cell cycles were analyzed by the BD Accuri C6 system software.

Examples

[0155] 1. Alternative Strand Selection (or Arm Switching) of miR-324

[0156] First, search was made for miRNAs that exhibit significant alterations in their strain ratio. For accurate quantification of the strand ratio, the AQ-seq library protocol was employed. The variation of strand selection was estimated across 15 different mouse tissues/developmental stages (FIG. 1a, X axis). In addition, a sRNA-seq dataset from 9 human cell lines was used to compare strand use in humans.

[0157] As shown in FIG. 1a, the majority of miRNAs in human and mouse datasets are invariable in strand ratio (ca. 79%, median variance below 3%). Nevertheless, several cases distinctively displaying variations, such as miR-324, miR-362, miR-193a, and miR-140, were identified.

[0158] Particularly, a focus was made on miR-324 because the arm switching of miR-324 was considered to have a substantial effect on cellular fate decision.

[0159] As shown in FIGS. 1b to 1d, three groups of 5′-isomiRs (5p, 3p.1, and 3p.2) that are produced from the single miR324 locus were detected (FIG. 1b). The 5p strand is dominant in mouse embryos and neuronal tissues while the major 3p isoform 3p.1 is prevalent in the liver and stomach (FIG. 1c). Furthermore, the 5′-isomiR groups are found in mammals, suggesting that the mechanism for alternative maturation is evolutionarily conserved (FIG. 1d).

[0160] 2. miR-324 Arm Switching is Controlled by TUT4 and TUT7.

[0161] The uridylation frequency of 3p correlates negatively with the 5p/3p.1 ratio (FIG. 2a), leading to the hypothesis that uridylation may be involved in the arm switching.

[0162] Meanwhile, the terminal uridylyl transferases TUT4 (also known as ZCCHC11 and TENT3A) and TUT7 (also known as ZCCHC6 and TENT3B) are reported to catalyze uridylation of diverse RNA species, including a specific set of pre-miRNAs (e.g., let-7 precursors). TUT4 and TUT7 (TUT4/7) act redundantly on most substrates.

[0163] On the basis of the facts in the foregoing, quantitation (RT-qPCR) of TUT4/7 levels in mouse tissues resulted in the finding that the 5p/3p.1 ratio is higher in cell types where TUT4 and TUT7 levels are relatively low (FIG. 2b).

[0164] To examine whether TUT4/7 is involved in miR-324 maturation, TUT4/7 was depleted in HEK293T sRNA-seq (AQ-seq) was performed.

[0165] As shown in FIG. 2c, TUT4/7 knockdown reduced uridylation of miRNAs, including miR-324-3p (left upper panel). In addition, TUT4/7 knockdown resulted in a change in isoform composition of miR-324, such as a reduction in 3p.1 and an increase in 5p and the minor isoforms (5p/3p.1) increased, with the consequent increase of a ratio between the major isoforms (5p/3p.1) (right upper and lower panels).

[0166] These sequencing data were consistent with the northern blot (FIG. 2d), indicating that there is a change in the quantity of miR-324 isoforms upon TUT4/7 knockdown.

[0167] In addition, as can be seen in FIG. 2e, re-analysis of the published sRNA-seq data from the TUT4/7 double knockout mice shows that 3p.1 level decreases whereas the 5p level increases upon the knockout (left panel). Consequently, a change in the strand ratio by TUT4/7 double knockout was observed in all tissues/cell types examined (bone marrow, embryonic fibroblasts, embryonic stem cells, and liver) (right panel).

[0168] 3. Uridylation Induces Alternative Dicer Processing of Pre-miR-324.

[0169] TUT4/7 are known to modify pre-miRNAs rather than mature miRNAs. In order to understand how TUT4/7 modulate miR-324 strand selection, the effect of uridylation on pre-miRNA processing was examined by performing in vitro Dicer processing assays with synthetic pre-miR-324.

[0170] As shown in FIG. 3a, when pre-miR-324 carries one or two extra uridine residues at the 3′ end, the Dicer processing site is shifted by 3-nt (from position A to position B in FIG. 3a). Unmodified pre-miR-324 mainly releases a longer duplex containing 5p and 3p.2 (black arrowheads) whereas uridylated pre-miR-324 is cleaved at the alternative position (position B) to yield a shorter duplex consisting of shorter 5p and 3p.1 (white arrowheads). Thus, TUT4/7-mediated uridylation of pre-miR-324 leads to alternative Dicer processing.

[0171] Moreover, as shown in FIGS. 3b and 3c, the sequencing data from TUT4/7-depleted human and mouse cells showed that the relative abundance of the 3p isomiRs (3p.1 vs. 3p.2) changed on TUT4/7 knockdown (FIG. 3b) and knockout (FIG. 3c), confirming the conclusion that uridylation alters Dicer cleavage site selection.

[0172] 4. Alternative Dicer Processing is Facilitated by the Double-Stranded RNA Binding Domain.

[0173] First, in order to identify the determinant responsible for the 3-nt shift, U bulge was removed from pre-miR-324.

[0174] For the “no-bulge” mutant, as shown in FIG. 4a, Dicer no longer induced the abrupt shift to position B upon uridylation, but rather caused single-nucleotide gradual shifts (lanes 10-12). Hence, the U bulge is an anti-determinant for cleavage at sites between positions A and B and serves as the ci-acting element responsible for alternative processing.

[0175] Next, examination was made to reveal relationship between the 5′- and 3′-counting rules and the pre-miR-324 processing. In this regard, Dicer mutants in the 5′ or 3′ pocket were utilized.

[0176] As can be seen in FIG. 4b, the 5′ pocket mutant Dicer failed to cleave pre-miR-324 at position A while the 3′ pocket mutant exhibited slight affection on processing efficiency without an effect on cleavage site choice. This result indicates the importance of the 5′ pocket for cleavage at position A.

[0177] Moreover, an investigation was made of the contribution of other domains in Dicer. For use in testing the role of the Dicer dsRBD in vitro, a dsRBD deletion mutant was generated.

[0178] As shown in FIG. 4c, the dsRBD deletion brought about a significant change in the processing pattern. That is, without the dsRBD, Dicer cleaved mainly at position A, which implies that the dsRBD is responsible for the cleavage at position B.

[0179] Taken together, unmodified pre-miR-324 is cleaved at position A, relying on the 5′ pocket existing in the platform domain of Dicer as shown in FIG. 4D (left panel). However, when uridylated, the end structure (3- to 4-nt 3′ overhang) is not tightly fixed to the platform/PAZ domains so that pre-miR-324 is repositioned in Dicer with the aid of the dsRBD, resulting in alternative processing at position B (right panel).

[0180] 5. Alternative Dicer Processing Leads to Arm Switching.

[0181] Taken together, the data obtained from the Examples showed that, as can be seen in FIG. 5a, pre-miR-324 can yield two alternative forms of duplexes, that is, a long duplex (position A; 5p/3p.2) and a short duplex (position B; 5p/3p.1).

[0182] In order to examine strand selection in cells, reporters containing the 3′ UTR complementary to miR-324-5p or miR-324-3p.1 was constructed (left panel in FIG. 5b), and co-transfected, along with synthetic duplexes ((5p/3p.2) or (5p/3p.1)). As a result, the long duplex selectively repressed the reporter with the 5p complementary sites but not with 3p.1 sites. As opposed to the long duplex, the short duplexes downregulated the 3p.1 reporter but not the 5p reporter (the right panel in FIG. 5b).

[0183] These results imply that 5p and 3p.1 are indeed differentially produced from the long duplex and the short duplex, respectively. That is, from the long duplex, the 5p strand is selected because the 5′ end of 5p is thermodynamically unstable compared with that of 3p.2 (blue dashed rectangle in FIG. 5a). In contrast, in the short duplex, 3p.1 starts with a 5′ adenosine, which is favored by AGO (red dashed rectangle in FIG. 5a).

[0184] Taken together, the results demonstrate that the arm switching of miR-324 results from the alternative Dicer processing.

[0185] 6. miR-324 Arm Switching Affects Cell-Cycle Progression.

[0186] The 5′ end of miR-324-3p was previously reported to vary in human glioblastoma (GBM). The proportion of 3p.1 is substantially greater in tumor than in normal brain tissue, as shown in FIG. 6a.

[0187] SO, to examine the possibility that TUT4/7-mediated miR-324 maturation is differentially modulated in the GBM context, transcriptome datasets were analyzed using Oncopression.

[0188] As can be seen in FIG. 6b, TUT4 and TUT7 are significantly upregulated in GBM.

[0189] In addition, using an independent dataset that profiled both mRNAs and miRNAs, comparison with normal samples was carried out to corroborate the upregulation of TUT4/7 levels in GBM and a reduction of the miR-324-5p/3p ratio in GBM.

[0190] As shown in FIG. 6c, TUT4/7 levels negatively correlate with 5p/3p ratios. These data suggest that TUT4/7-mediated miR-324 regulation may be physiologically relevant in GBM.

[0191] On the basis of the facts in the foregoing, to understand the function of miR-324 arm switching, the molecular and cellular phenotypes driven by perturbation of miR-324 strand selection in GBM cell lines was investigated.

[0192] As shown in FIGS. 6D and 6E, Transfection of the synthetic long duplex (“5p duplex”) or an antisense oligo against 3p (“3p inhibitor”) resulted in an accumulation of cyclins D and E and a reduction in cyclins A and B (left panels). In addition, cell-cycle profiling indicated that the cells are arrested at the G1 phase (right panels).

[0193] As shown in FIG. 6f, TUT4/7 knockdown also gave rise to cyclin dysregulation and G1 arrest.

[0194] Taken together, the data indicate that miR-324 isomiRs have opposing functions in GBM cell proliferation.

DISCUSSION

[0195] As shown in FIG. 7, the terminal uridylyl transferases TUT4/7 function as key players in the alternative strand selection of miR-324 by uridylating pre-miR-324. That is, TUT4/7 uridylates pre-miR-324, which leads to shifting Dicer cleavage positions on pre-miR-324 to change ratios among the three functionally different microRNAs (one 5p and two 3p isoforms). Therefore, inhibition of uridylation-mediated processing of miR-324 can repress the expression of cell division-associated genes relevant to the growth of glioblastoma.

[0196] Being capable of regulating the alternative strand selection of microRNA(miR-324) by TUT4/7, the present invention can be effectively applied to the prevention and treatment of cancer (encephaloma).

INDUSTRIAL APPLICABILITY

[0197] The present invention relates to a pharmaceutical composition comprising a TUT4/7 expression regulator for prevention or treatment of cancer and, more specifically, to a pharmaceutical composition comprising a nucleic acid sequence regulatory of TUT4/7 expression for prevention or treatment of cancer.

[0198] This application contains references to amino acid sequences and/or nucleic acid sequences which have been submitted herewith as the sequence listing text file. The aforementioned sequence listing is hereby incorporated by reference in its entirety pursuant to 37 C.F.R. § 1.52(e).