Endoribonucleolytic Cleavage of m6A-Containing RNAs by RNase P/MRP Complex

•m6A-containing RNAs are degraded by an endoribonucleolytic cleavage pathway•A YTHDF2-HRSP12-RNase P/MRP axis contributes to m6A-mediated RNA decay•An interaction between YTHDF2 and RNase P/MRP is bridged by HRSP12•m6A-containing circular RNAs are degraded by the YTHDF2-HRSP12-RNase P/MRP pathway N6-methyladenosine (m6A) is the most abundant internal modification in RNAs and plays regulatory roles in a variety of biological and physiological processes. Despite its important roles, the molecular mechanism underlying m6A-mediated gene regulation is poorly understood. Here, we show that m6A-containing RNAs are subject to endoribonucleolytic cleavage via YTHDF2 (m6A reader protein), HRSP12 (adaptor protein), and RNase P/MRP (endoribonucleases). We demonstrate that HRSP12 functions as an adaptor to bridge YTHDF2 and RNase P/MRP, eliciting rapid degradation of YTHDF2-bound RNAs. Transcriptome-wide analyses show that m6A RNAs that are preferentially targeted for endoribonucleolytic cleavage have an HRSP12-binding site and a RNase P/MRP-directed cleavage site upstream and downstream of the YTHDF2-binding site, respectively. We also find that a subset of m6A-containing circular RNAs associates with YTHDF2 in an HRSP12-dependent manner and is selectively downregulated by RNase P/MRP. Thus, our data expand the known functions of RNase P/MRP to endoribonucleolytic cleavage of m6A RNAs. N6-methyladenosine (m6A) is the most abundant internal modification in RNAs and plays regulatory roles in a variety of biological and physiological processes. Despite its important roles, the molecular mechanism underlying m6A-mediated gene regulation is poorly understood. Here, we show that m6A-containing RNAs are subject to endoribonucleolytic cleavage via YTHDF2 (m6A reader protein), HRSP12 (adaptor protein), and RNase P/MRP (endoribonucleases). We demonstrate that HRSP12 functions as an adaptor to bridge YTHDF2 and RNase P/MRP, eliciting rapid degradation of YTHDF2-bound RNAs. Transcriptome-wide analyses show that m6A RNAs that are preferentially targeted for endoribonucleolytic cleavage have an HRSP12-binding site and a RNase P/MRP-directed cleavage site upstream and downstream of the YTHDF2-binding site, respectively. We also find that a subset of m6A-containing circular RNAs associates with YTHDF2 in an HRSP12-dependent manner and is selectively downregulated by RNase P/MRP. Thus, our data expand the known functions of RNase P/MRP to endoribonucleolytic cleavage of m6A RNAs. Considerable recent studies revealed the importance of post-transcriptional RNA modifications in shaping an epitranscriptomic landscape of gene expression (Meyer and Jaffrey, 2017Meyer K.D. Jaffrey S.R. Rethinking m6A readers, writers, and erasers.Annu. Rev. Cell Dev. Biol. 2017; 33: 319-342Crossref PubMed Scopus (570) Google Scholar, Yang et al., 2018Yang Y. Hsu P.J. Chen Y.S. Yang Y.G. Dynamic transcriptomic m6A decoration: writers, erasers, readers and functions in RNA metabolism.Cell Res. 2018; 28: 616-624Crossref PubMed Scopus (638) Google Scholar). To date, more than 100 distinct types of RNA modifications have been characterized. Among them, methylation of the 6-amino group of adenosine to generate N6-methyladenosine (m6A) is the most abundant internal modification of RNAs (Dominissini et al., 2012Dominissini D. Moshitch-Moshkovitz S. Schwartz S. Salmon-Divon M. Ungar L. Osenberg S. Cesarkas K. Jacob-Hirsch J. Amariglio N. Kupiec M. et al.Topology of the human and mouse m6A RNA methylomes revealed by m6A-seq.Nature. 2012; 485: 201-206Crossref PubMed Scopus (2663) Google Scholar, Meyer et al., 2012Meyer K.D. Saletore Y. Zumbo P. Elemento O. Mason C.E. Jaffrey S.R. Comprehensive analysis of mRNA methylation reveals enrichment in 3′ UTRs and near stop codons.Cell. 2012; 149: 1635-1646Abstract Full Text Full Text PDF PubMed Scopus (2324) Google Scholar). m6A modification is dynamically and reversibly regulated by three classes of proteins (Meyer and Jaffrey, 2017Meyer K.D. Jaffrey S.R. Rethinking m6A readers, writers, and erasers.Annu. Rev. Cell Dev. Biol. 2017; 33: 319-342Crossref PubMed Scopus (570) Google Scholar, Roundtree et al., 2017aRoundtree I.A. Evans M.E. Pan T. He C. Dynamic RNA modifications in gene expression regulation.Cell. 2017; 169: 1187-1200Abstract Full Text Full Text PDF PubMed Scopus (1447) Google Scholar, Yang et al., 2018Yang Y. Hsu P.J. Chen Y.S. Yang Y.G. Dynamic transcriptomic m6A decoration: writers, erasers, readers and functions in RNA metabolism.Cell Res. 2018; 28: 616-624Crossref PubMed Scopus (638) Google Scholar). In the case of mammalian cells, generation of m6A is catalyzed by a multiprotein methyltransferase complex containing METTL3, METTL14, and WTAP (Meyer and Jaffrey, 2017Meyer K.D. Jaffrey S.R. Rethinking m6A readers, writers, and erasers.Annu. Rev. Cell Dev. Biol. 2017; 33: 319-342Crossref PubMed Scopus (570) Google Scholar, Yang et al., 2018Yang Y. Hsu P.J. Chen Y.S. Yang Y.G. Dynamic transcriptomic m6A decoration: writers, erasers, readers and functions in RNA metabolism.Cell Res. 2018; 28: 616-624Crossref PubMed Scopus (638) Google Scholar). m6As in RNAs are selectively recognized by m6A-specific binding proteins, such as YT521-B homology (YTH)-domain-containing proteins, hnRNPs, and eIF3 (Alarcón et al., 2015Alarcón C.R. Goodarzi H. Lee H. Liu X. Tavazoie S. Tavazoie S.F. HNRNPA2B1 is a mediator of m(6)A-dependent nuclear RNA processing events.Cell. 2015; 162: 1299-1308Abstract Full Text Full Text PDF PubMed Scopus (781) Google Scholar, Dominissini et al., 2012Dominissini D. Moshitch-Moshkovitz S. Schwartz S. Salmon-Divon M. Ungar L. Osenberg S. Cesarkas K. Jacob-Hirsch J. Amariglio N. Kupiec M. et al.Topology of the human and mouse m6A RNA methylomes revealed by m6A-seq.Nature. 2012; 485: 201-206Crossref PubMed Scopus (2663) Google Scholar, Meyer et al., 2015Meyer K.D. Patil D.P. Zhou J. Zinoviev A. Skabkin M.A. Elemento O. Pestova T.V. Qian S.B. Jaffrey S.R. 5′ UTR m(6)A promotes cap-independent translation.Cell. 2015; 163: 999-1010Abstract Full Text Full Text PDF PubMed Scopus (1036) Google Scholar, Wang et al., 2014Wang X. Lu Z. Gomez A. Hon G.C. Yue Y. Han D. Fu Y. Parisien M. Dai Q. Jia G. et al.N6-methyladenosine-dependent regulation of messenger RNA stability.Nature. 2014; 505: 117-120Crossref PubMed Scopus (2214) Google Scholar). Demethylases such as FTO and ALKBH5 are involved in the removal of m6A (Jia et al., 2011Jia G. Fu Y. Zhao X. Dai Q. Zheng G. Yang Y. Yi C. Lindahl T. Pan T. Yang Y.G. He C. N6-methyladenosine in nuclear RNA is a major substrate of the obesity-associated FTO.Nat. Chem. Biol. 2011; 7: 885-887Crossref PubMed Scopus (2193) Google Scholar, Zheng et al., 2013Zheng G. Dahl J.A. Niu Y. Fedorcsak P. Huang C.M. Li C.J. Vågbø C.B. Shi Y. Wang W.L. Song S.H. et al.ALKBH5 is a mammalian RNA demethylase that impacts RNA metabolism and mouse fertility.Mol. Cell. 2013; 49: 18-29Abstract Full Text Full Text PDF PubMed Scopus (1858) Google Scholar). Recent research revealed that m6A in RNAs exerts regulatory functions in a broad spectrum of biological and physiological processes (Fu et al., 2014Fu Y. Dominissini D. Rechavi G. He C. Gene expression regulation mediated through reversible m6A RNA methylation.Nat. Rev. Genet. 2014; 15: 293-306Crossref PubMed Scopus (1019) Google Scholar, Meyer and Jaffrey, 2017Meyer K.D. Jaffrey S.R. Rethinking m6A readers, writers, and erasers.Annu. Rev. Cell Dev. Biol. 2017; 33: 319-342Crossref PubMed Scopus (570) Google Scholar, Roundtree et al., 2017aRoundtree I.A. Evans M.E. Pan T. He C. Dynamic RNA modifications in gene expression regulation.Cell. 2017; 169: 1187-1200Abstract Full Text Full Text PDF PubMed Scopus (1447) Google Scholar, Yang et al., 2018Yang Y. Hsu P.J. Chen Y.S. Yang Y.G. Dynamic transcriptomic m6A decoration: writers, erasers, readers and functions in RNA metabolism.Cell Res. 2018; 28: 616-624Crossref PubMed Scopus (638) Google Scholar). At the molecular level, m6A is known to affect mRNA stability, mRNA translation, microRNA biogenesis, and splicing, most of which are mediated by specific m6A-binding proteins. For instance, YTHDF2, one of the YTH-domain-containing proteins, destabilizes m6A-containing RNAs (Du et al., 2016Du H. Zhao Y. He J. Zhang Y. Xi H. Liu M. Ma J. Wu L. YTHDF2 destabilizes m(6)A-containing RNA through direct recruitment of the CCR4-NOT deadenylase complex.Nat. Commun. 2016; 7: 12626Crossref PubMed Scopus (683) Google Scholar, Wang et al., 2014Wang X. Lu Z. Gomez A. Hon G.C. Yue Y. Han D. Fu Y. Parisien M. Dai Q. Jia G. et al.N6-methyladenosine-dependent regulation of messenger RNA stability.Nature. 2014; 505: 117-120Crossref PubMed Scopus (2214) Google Scholar). YTHDF1 and YTHDF3 promote translation of m6A mRNAs (Li et al., 2017Li A. Chen Y.S. Ping X.L. Yang X. Xiao W. Yang Y. Sun H.Y. Zhu Q. Baidya P. Wang X. et al.Cytoplasmic m6A reader YTHDF3 promotes mRNA translation.Cell Res. 2017; 27: 444-447Crossref PubMed Scopus (458) Google Scholar, Wang et al., 2015Wang X. Zhao B.S. Roundtree I.A. Lu Z. Han D. Ma H. Weng X. Chen K. Shi H. He C. N(6)-methyladenosine modulates messenger RNA translation efficiency.Cell. 2015; 161: 1388-1399Abstract Full Text Full Text PDF PubMed Scopus (1733) Google Scholar). In addition, YTHDC1 is known to regulate pre-mRNA splicing and nuclear export of m6A mRNAs (Roundtree et al., 2017bRoundtree I.A. Luo G.Z. Zhang Z. Wang X. Zhou T. Cui Y. Sha J. Huang X. Guerrero L. Xie P. et al.YTHDC1 mediates nuclear export of N6-methyladenosine methylated mRNAs.eLife. 2017; 6: e31311Crossref PubMed Scopus (551) Google Scholar, Xiao et al., 2016Xiao W. Adhikari S. Dahal U. Chen Y.S. Hao Y.J. Sun B.F. Sun H.Y. Li A. Ping X.L. Lai W.Y. et al.Nuclear m(6)A reader YTHDC1 regulates mRNA splicing.Mol. Cell. 2016; 61: 507-519Abstract Full Text Full Text PDF PubMed Scopus (991) Google Scholar). Eukaryotic RNase P and its close relative RNase MRP are essential ribonucleoprotein complexes that function as endoribonucleases (Jarrous, 2017Jarrous N. Roles of RNase P and its subunits.Trends Genet. 2017; 33: 594-603Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar). Mammalian RNase P and RNase MRP contain common protein components (in the order of their molecular weight: POP1, RPP38, POP5, RPP25, RPP20, RPP30, and RPP40) and their unique protein components with their unique noncoding RNA components, RPPH1 and RMRP RNAs, respectively. Originally, RNase P and RNase MRP were identified to be the endoribonucleases responsible for the maturation of tRNA and mitochondrial RNA processing of replication primers, respectively. Nonetheless, it was later shown that eukaryotic RNase P and RNase MRP are involved in the cleavage of a wide range of RNA substrates, such as rRNAs, long noncoding RNAs, and mRNAs (Coughlin et al., 2008Coughlin D.J. Pleiss J.A. Walker S.C. Whitworth G.B. Engelke D.R. Genome-wide search for yeast RNase P substrates reveals role in maturation of intron-encoded box C/D small nucleolar RNAs.Proc. Natl. Acad. Sci. USA. 2008; 105: 12218-12223Crossref PubMed Scopus (55) Google Scholar, Jarrous, 2017Jarrous N. Roles of RNase P and its subunits.Trends Genet. 2017; 33: 594-603Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar, Maida et al., 2009Maida Y. Yasukawa M. Furuuchi M. Lassmann T. Possemato R. Okamoto N. Kasim V. Hayashizaki Y. Hahn W.C. Masutomi K. An RNA-dependent RNA polymerase formed by TERT and the RMRP RNA.Nature. 2009; 461: 230-235Crossref PubMed Scopus (301) Google Scholar, Wilusz et al., 2008Wilusz J.E. Freier S.M. Spector D.L. 3′ end processing of a long nuclear-retained noncoding RNA yields a tRNA-like cytoplasmic RNA.Cell. 2008; 135: 919-932Abstract Full Text Full Text PDF PubMed Scopus (524) Google Scholar). Despite the biological importance of the aforementioned m6A modification, the molecular mechanisms underlying m6A-mediated gene regulation are poorly understood. In this study, we report that m6A RNAs are endoribonucleolytically cleaved by a YTHDF2-HRSP12-RNase P/MRP axis. These findings shed light on the molecular mechanisms of multilayered gene regulation mediated by m6A modification in RNA. An interaction between Pho92 (a yeast homolog of human YTHDF2) and Mmf1 (a yeast homolog of human heat-responsive protein 12 [HRSP12]) was experimentally identified by tandem affinity purification (Krogan et al., 2006Krogan N.J. Cagney G. Yu H. Zhong G. Guo X. Ignatchenko A. Li J. Pu S. Datta N. Tikuisis A.P. et al.Global landscape of protein complexes in the yeast Saccharomyces cerevisiae.Nature. 2006; 440: 637-643Crossref PubMed Scopus (2333) Google Scholar). In addition, a large-scale de novo prediction of physical protein–protein association using a combined random forests and Bayesian learning strategy predicted an interaction between HRSP12 and POP1, a component of RNase P/MRP (Elefsinioti et al., 2011Elefsinioti A. Sarac O.S. Hegele A. Plake C. Hubner N.C. Poser I. Sarov M. Hyman A. Mann M. Schroeder M. et al.Large-scale de novo prediction of physical protein-protein association.Mol. Cell Proteomics. 2011; 10 (M111.010629)Crossref PubMed Scopus (44) Google Scholar). In this study, we set out to assess the predicted interactions by in vivo and in vitro experimental approaches. Immunoprecipitation (IP) experiments revealed that YTHDF2, HRSP12, and POP1 were complexed with one another in an RNase A-resistant manner (Figures 1A and S1A). The previously known YTHDF2-interacting protein CNOT1 (a component of the CCR4-NOT deadenylase complex [Du et al., 2016Du H. Zhao Y. He J. Zhang Y. Xi H. Liu M. Ma J. Wu L. YTHDF2 destabilizes m(6)A-containing RNA through direct recruitment of the CCR4-NOT deadenylase complex.Nat. Commun. 2016; 7: 12626Crossref PubMed Scopus (683) Google Scholar]) was also selectively enriched in the IPs. Of note, YTHDF1 and YTHDF3, two paralogs of YTHDF2, were only marginally enriched in the immunoprecipitates of HRSP12 (Figure S1B), pointing to preferential association between YTHDF2 and HRSP12. In vitro GST pull-down experiments showed that purified recombinant His-HRSP12 was selectively enriched in the pull-down of GST-YTHDF2, but not GST or GST-RavZ (a negative control protein; Figures 1B and S1C), indicating a direct interaction between YTHDF2 and HRSP12. In addition, far-western blotting involving purified recombinant proteins and either immunopurified FLAG-GFP (Figures S1D and S1E) or POP1-FLAG (Figures 1C and S1D) as a probe revealed a direct interaction between HRSP12 and POP1. Collectively, our data showed that human HRSP12 directly interacts with YTHDF2 and RNase P/MRP. Considering that most of HRSP12 and YTHDF2 were enriched in the cytoplasmic fraction and that POP1 was distributed throughout the cell (Figure S1F), it is likely that the complex formation largely occurs in the cytoplasm. In the above-mentioned far-western blotting, the immunopurified POP1-FLAG probe reacted only with His-HRSP12 (not GST-YTHDF2), suggesting a lack of a direct interaction between YTHDF2 and RNase P/MRP. Given that YTHDF2, HRSP12, and POP1 form a complex (Figure 1A), these observations pointed to a possible role of HRSP12 in bridging YTHDF2 and RNase P/MRP. In support of this possibility, when HRSP12 was downregulated 5-fold by treatment of the cells with small interfering RNAs (siRNAs), 3- to 5-fold smaller amounts of components of RNase P/MRP (POP1, RPP20, and RPP25) coimmunoprecipitated (coIPed) in the IP of YTHDF2 (Figure 1D). Notably, the known interaction between YTHDF2 and CNOT1 was affected only marginally. Therefore, our observations indicated that HRSP12 is a molecular linker bridging YTHDF2 and RNase P/MRP. An m6A mRNA is destabilized under the influence of YTHDF2 (Wang et al., 2014Wang X. Lu Z. Gomez A. Hon G.C. Yue Y. Han D. Fu Y. Parisien M. Dai Q. Jia G. et al.N6-methyladenosine-dependent regulation of messenger RNA stability.Nature. 2014; 505: 117-120Crossref PubMed Scopus (2214) Google Scholar). On the basis of our observations regarding formation of a complex of YTHDF2, HRSP12, and RNase P/MRP (Figure 1), we investigated a possible function of HRSP12 and RNase P/MRP in m6A-mediated mRNA decay. To this end, the two different previously established m6A β-globin reporter systems (Du et al., 2016Du H. Zhao Y. He J. Zhang Y. Xi H. Liu M. Ma J. Wu L. YTHDF2 destabilizes m(6)A-containing RNA through direct recruitment of the CCR4-NOT deadenylase complex.Nat. Commun. 2016; 7: 12626Crossref PubMed Scopus (683) Google Scholar) were employed (Figure 2A). SON-WT and SON-Mut reporter mRNAs contained β-globin genomic sequences harboring three m6A target sites derived from the wild-type (WT) SON gene (SON-WT) and the mutated version (Mut) of the corresponding sequences (SON-Mut) within the open reading frame, respectively. PLAC2-WT and PLAC2-Mut mRNAs contained two m6A target sites derived from the PLAC2 gene (PLAC2-WT) and the mutated version of the corresponding sequences (PLAC2-Mut), respectively, in the 3′ UTR. The inserted m6A target sites derived from genes SON and PLAC2 were confirmed to be m6A methylated and bound to YTHDF2 in our present (Figure S2A) and previous studies (Du et al., 2016Du H. Zhao Y. He J. Zhang Y. Xi H. Liu M. Ma J. Wu L. YTHDF2 destabilizes m(6)A-containing RNA through direct recruitment of the CCR4-NOT deadenylase complex.Nat. Commun. 2016; 7: 12626Crossref PubMed Scopus (683) Google Scholar, Wang et al., 2014Wang X. Lu Z. Gomez A. Hon G.C. Yue Y. Han D. Fu Y. Parisien M. Dai Q. Jia G. et al.N6-methyladenosine-dependent regulation of messenger RNA stability.Nature. 2014; 505: 117-120Crossref PubMed Scopus (2214) Google Scholar). Both SON-WT and PLAC2-WT mRNAs showed ∼40%–50% lower abundance relative to their corresponding Mut mRNAs (Figures 2B–2D). The relative decrease in abundance was reversed when METTL3 or METTL14 was downregulated (Figures S2B and S2C), validating specific m6A-mediated reporter mRNA decay. Of note, downregulation of endogenous YTHDF2, HRSP12, or a protein component (POP1 or RPP20) or an RNA component (RMRP or RPPH1) of RNase P/MRP increased the relative abundance (Figures 2B–2D and S2D–S2F) and half-life (Figures 2E and 2F) of WT reporter mRNAs, indicating that HRSP12 and RNase P/MRP function in m6A-mediated mRNA decay. Although we cannot completely rule out the possible indirect effect caused by inefficient processing of rRNAs or tRNAs by downregulation of a component of RNase P/MRP, all the data indicate that HRSP12 and RNase P/MRP function in m6A-mediated mRNA decay. In addition, consistent with a report showing that YTHDF2 triggers m6A-mediated RNA decay with the help of the CCR4-NOT deadenylase complex (Du et al., 2016Du H. Zhao Y. He J. Zhang Y. Xi H. Liu M. Ma J. Wu L. YTHDF2 destabilizes m(6)A-containing RNA through direct recruitment of the CCR4-NOT deadenylase complex.Nat. Commun. 2016; 7: 12626Crossref PubMed Scopus (683) Google Scholar), downregulation of CNOT1 also increased the relative levels of WT mRNAs (Figure 2B). To further dissect the role of HRSP12, we generated HRSP12 knockout (KO) HAP1 cells: near-haploid human cell lines originating from a patient with chronic myeloid leukemia (Blomen et al., 2015Blomen V.A. Májek P. Jae L.T. Bigenzahn J.W. Nieuwenhuis J. Staring J. Sacco R. van Diemen F.R. Olk N. Stukalov A. et al.Gene essentiality and synthetic lethality in haploid human cells.Science. 2015; 350: 1092-1096Crossref PubMed Scopus (503) Google Scholar, Carette et al., 2011Carette J.E. Raaben M. Wong A.C. Herbert A.S. Obernosterer G. Mulherkar N. Kuehne A.I. Kranzusch P.J. Griffin A.M. Ruthel G. et al.Ebola virus entry requires the cholesterol transporter Niemann-Pick C1.Nature. 2011; 477: 340-343Crossref PubMed Scopus (896) Google Scholar). Whereas both SON-WT and PLAC2-WT mRNAs decreased in abundance in the parental HAP1 cells, their levels were comparable to the levels of Mut mRNAs in HRSP12-KO HAP1 cells (Figures 2G and 2H), supporting our conclusion about participation of HRSP12 in m6A-mediated mRNA decay. Previous reports showed that HRSP12 participates in GMD, which is induced by a glucocorticoid (GC), a specific GR ligand (Cho et al., 2015Cho H. Park O.H. Park J. Ryu I. Kim J. Ko J. Kim Y.K. Glucocorticoid receptor interacts with PNRC2 in a ligand-dependent manner to recruit UPF1 for rapid mRNA degradation.Proc. Natl. Acad. Sci. USA. 2015; 112: E1540-E1549Crossref PubMed Scopus (42) Google Scholar, Park et al., 2016Park O.H. Park J. Yu M. An H.T. Ko J. Kim Y.K. Identification and molecular characterization of cellular factors required for glucocorticoid receptor-mediated mRNA decay.Genes Dev. 2016; 30: 2093-2105Crossref PubMed Scopus (31) Google Scholar). In GMD, a ligand-free GR directly binds to a subset of mRNAs. In the presence of GC, the mRNA-bound GR associates with GC, providing a favorable platform for the loading of HRSP12, triggering effective GMD of GR-bound mRNAs. Therefore, given that HRSP12 is a GMD component, it is possible that m6A-mediated mRNA decay is affected by GMD or vice versa. Although treatment with dexamethasone (Dex), which is a synthetic derivative of a GC, did not affect m6A-mediated mRNA decay of both SON and PLAC2 reporter mRNAs (Figure S2G), the level of an endogenous GMD substrate, CCL2 mRNA, drastically decreased after Dex treatment (Figure S2H). Furthermore, downregulation of POP1 did not affect GMD (Figures S2I and S2J). As expected, both m6A-mediated mRNA decay and GMD were inhibited by HRSP12 downregulation (Figures 2B and S2J). In addition, complementation of HRSP12 through expression of exogenous FLAG-HRSP12-WT in HRSP12-KO HAP1 cells restored the efficiency of both m6A-mediated mRNA decay (Figures 2G and 2H) and GMD (Figures S2K and S2L). Of note, although exogenously expressed FLAG-HRSP12-P105A/R107E successfully restored m6A-mediated mRNA decay (Figures 2G and 2H), it failed to restore the functionality of GMD (Figures S2K and S2L). It is known that the HRSP12-P105A/R107E variant cannot form a trimeric structure and fails to maintain the structural integrity of a functionally active GMD complex (Park et al., 2016Park O.H. Park J. Yu M. An H.T. Ko J. Kim Y.K. Identification and molecular characterization of cellular factors required for glucocorticoid receptor-mediated mRNA decay.Genes Dev. 2016; 30: 2093-2105Crossref PubMed Scopus (31) Google Scholar). These data suggest that GMD, but not m6A-mediated mRNA decay, requires the trimerization motif of HRSP12. Collectively, our data indicated that m6A-mediated mRNA decay is mechanistically distinct from GMD. To delineate the minimal functional region of YTHDF2 for m6A-mediated RNA decay involving HRSP12, YTHDF2 was artificially tethered to the 3′ UTR of a reporter mRNA using the bacteriophage MS2 coat protein (MS2) and its binding site (MS2bs; Figure 3A). In parental HAP1 cells, artificially tethered MS2-HA-YTHDF2 and MS2-HA-SMG5, but not MS2-HA-GFP, elicited rapid degradation of reporter mRNA (Figures 3B and S3A). SMG5, a specific cellular factor for nonsense-mediated mRNA decay, is known to elicit rapid mRNA degradation when it is tethered to the 3′ UTR of a reporter mRNA (Cho et al., 2013Cho H. Han S. Choe J. Park S.G. Choi S.S. Kim Y.K. SMG5-PNRC2 is functionally dominant compared with SMG5-SMG7 in mammalian nonsense-mediated mRNA decay.Nucleic Acids Res. 2013; 41: 1319-1328Crossref PubMed Scopus (59) Google Scholar). It is noteworthy that the observed rapid mRNA degradation by the tethered YTHDF2, but not by the tethered SMG5, was drastically inhibited in HRSP12-KO HAP1 cells, suggesting that YTHDF2-mediated RNA degradation specifically depends on HRSP12. We next generated two deletion variants of YTHDF2 (Figure 3A): YTHDF-N (lacking a YTH domain in a C-terminal region) and YTHDF-C (lacking a P/Q/N-rich low-complexity region without a clear domain in the N-terminal part). The IP results showed that HRSP12, POP1, RPP25, and CNOT1 coIPed with the full-length YTHDF2 and YTHDF2-N, but not with YTHDF2-C (Figure 3C). In agreement with the IP results, tethered YTHDF2 and YTHDF2-N, but not YTHDF2-C, elicited rapid mRNA degradation in an HRSP12-dependent manner (Figures 3D and S3B). The N-terminal region of YTHDF2 was further subdivided into four segments: aa 1–100, aa 101–200, NΔ101–200, and aa 201–383 (Figure 3A). We carried out GST pull-down experiments using the extracts of cells expressing GST-HRSP12 and one of the deletion variants of YTHDF2 (Figure 3E). Variants YTHDF2-N, 1–100, and NΔ101–200 were significantly enriched in the pull-down of GST-HRSP12. Variant 201–383 was only weakly enriched in the pull-down. In contrast, variant 101–200, which is known to directly interact with CNOT1 (Du et al., 2016Du H. Zhao Y. He J. Zhang Y. Xi H. Liu M. Ma J. Wu L. YTHDF2 destabilizes m(6)A-containing RNA through direct recruitment of the CCR4-NOT deadenylase complex.Nat. Commun. 2016; 7: 12626Crossref PubMed Scopus (683) Google Scholar), was undetectable in the pull-down. In line with the GST pull-down results, artificial tethering of variants N, 1–100, or NΔ101–200 elicited rapid mRNA degradation in an HRSP12-dependent manner (Figures 3F and S3C). In addition, variant 201–383, which manifested a weak interaction with HRSP12, caused slight mRNA downregulation without a significant dependence on HRSP12. Of note, tethered variant 101–200 elicited rapid mRNA degradation in an HRSP12-independent manner, probably via the CNOT1 interaction, suggesting that the HRSP12-RNase P/MRP pathway is mechanistically separable from the CCR4-NOT1 pathway. Thus, we conclude that the N-terminal 1–100 aa of YTHDF2 represent a minimal region for HRSP12 binding and are sufficient for rapid RNA degradation with the help of HRSP12. To investigate the effect of HRSP12 and RNase P/MRP at the transcriptome level, we carried out mRNA sequencing (mRNA-seq) experiments for measuring abundance (Figures 4A–4F; Tables S1A and S1B) and half-life (Figures 4G–4I; Table S1C) using total-RNA samples purified from HeLa cells either not depleted or depleted of the protein indicated in the figures. One research group (Wang et al., 2014Wang X. Lu Z. Gomez A. Hon G.C. Yue Y. Han D. Fu Y. Parisien M. Dai Q. Jia G. et al.N6-methyladenosine-dependent regulation of messenger RNA stability.Nature. 2014; 505: 117-120Crossref PubMed Scopus (2214) Google Scholar) identified YTHDF2-binding transcripts in HeLa cells by photoactivatable ribonucleoside cross-linking and immunoprecipitation (PAR-CLIP) and a RNP-IP followed by next-generation sequencing (RIP-seq). According to the data from the PAR-CLIP and RIP-seq, HeLa cell transcripts were categorized into four groups: non-targets (absent from the PAR-CLIP and RIP-seq), CLIP–IP targets (present only in the PAR-CLIP but not in RIP-seq), CLIP targets (present in the PAR-CLIP), and CLIP∩IP targets (present in both the PAR-CLIP and RIP-seq). A cumulative distribution function (CDF) analysis revealed that overall the CLIP–IP, CLIP, and CLIP∩IP targets significantly and gradually increased in abundance upon downregulation of YTHDF2, HRSP12, POP1, RPP20, or METTL3 (Figures 4A–4E and S4A–S4C), and their half-life increased after downregulation of YTHDF2, HRSP12, or POP1 (Figures 4G–4I) as compared with non-targets. In agreement of these results, the reanalysis of previously reported RNA-seq data from HRSP12-depleted HeLa cells (Park et al., 2016Park O.H. Park J. Yu M. An H.T. Ko J. Kim Y.K. Identification and molecular characterization of cellular factors required for glucocorticoid receptor-mediated mRNA decay.Genes Dev. 2016; 30: 2093-2105Crossref PubMed Scopus (31) Google Scholar) and RMRP knockdown HeLa cells generated by targeted CRISPR-based disruption (Goldfarb and Cech, 2017Goldfarb K.C. Cech T.R. Targeted CRISPR disruption reveals a role for RNase MRP RNA in human preribosomal RNA processing.Genes Dev. 2017; 31: 59-71Crossref PubMed Scopus (63) Google Scholar) also revealed a significant increase in the abundance of CLIP–IP, CLIP, and CLIP∩IP targets after HRSP12 downregulation (Figure S4D) or RMRP downregulation (Figure S4E) in comparison with the control cells. Collectively, the data indicated that HRSP12 and RNase P/MRP destabilize YTHDF2-bound transcripts at the transcriptome level. It should be noted that significant upregulation of CLIP–IP, CLIP, and CLIP∩IP targets was also observed when CNOT1 was downregulated (Figure 4F). Double downregulation of HRSP12 and CNOT1 did not significantly affect the abundance of non-targets, CLIP, or CLIP∩IP transcripts as compared with single downregulation of either HRSP12 or CNOT1 (Figures S4F–S4I), suggesting that two RNA decay pathways mediated by the HRSP12-RNase P/MRP or CCR4-NOT complex are coupled to each other for efficient m6A RNA decay. A recent effort to screen proteins for RNA-binding ones in HeLa cells identified HRSP12 as a RNA-binding protein (Castello et al., 2012Castello A. Fischer B. Eichelbaum K. Horos R. Beckmann B.M. Strein C. Davey N.E. Humphreys D.T. Preiss T. Steinmetz L.M. et al.Insights into RNA biology from an atlas of mammalian mRNA-binding proteins.Cell. 2012; 149: 1393-1406Abstract Full Text Full Text PDF PubMed Scopus (1352) Google Scholar). Therefore, to look for transcriptome-wide interactions between HRSP12 and target transcripts, we carried out cross-linking IP experiments coupled with high-throughput sequencing (CLIP-seq). Raw reads obtained from two independent CLIP-seq experiments with HRSP12 (CLIP1 and CLIP2) were processed by adaptor sequence trimming, removal of ribosomal sequences, and mapping to the human reference genome sequence (hg19: Table S2). After a peak-calling process, we obtained 11,883 called peaks in CLIP1 and 18,703 called peaks in CLIP2. The Pearson correlation coefficient (r) between CLIP1 and CLIP2 for reads per million mapped reads (RPMs) of the common peaks (8,822) was 0.927, indicating a strong correlation between CLIP1 and CLIP2 (Figure S5A). When the number of peaks was adjusted for