Transcriptional Networking Cap-tures the 7SK RNA 5′-γ-Methyltransferase
2007; Elsevier BV; Volume: 27; Issue: 4 Linguagem: Inglês
10.1016/j.molcel.2007.08.001
ISSN1097-4164
Autores Tópico(s)Protein Degradation and Inhibitors
ResumoIn a recent issue of Molecular Cell, Jeronimo et al., 2007Jeronimo C. Forget D. Bouchard A. Li Q. Chua G. Poitras C. Thérien C. Bergeron D. Bourassa S. Greenblatt J. et al.Mol. Cell. 2007; 27: 262-274Abstract Full Text Full Text PDF PubMed Scopus (327) Google Scholar identify BCDIN3, a Cdk9-associated protein, as the enzyme that forms the distinctive γ-methylphosphate cap structure of 7SK, a noncoding RNA that regulates Cdk9 activity. In a recent issue of Molecular Cell, Jeronimo et al., 2007Jeronimo C. Forget D. Bouchard A. Li Q. Chua G. Poitras C. Thérien C. Bergeron D. Bourassa S. Greenblatt J. et al.Mol. Cell. 2007; 27: 262-274Abstract Full Text Full Text PDF PubMed Scopus (327) Google Scholar identify BCDIN3, a Cdk9-associated protein, as the enzyme that forms the distinctive γ-methylphosphate cap structure of 7SK, a noncoding RNA that regulates Cdk9 activity. 5′ cap structures are the defining feature of eukaryal RNAs. Discovered in 1974 in small nuclear RNAs of then unknown function, and in 1975 in viral and cellular mRNAs, caps come in several flavors. The mRNA cap consists of 7-methylguanosine linked to the terminal RNA nucleoside through a 5′–5′ triphosphate bridge (Figure 1). All eukaryal cells, and many eukaryal viruses, share a three-step mRNA capping mechanism in which the 5′-triphosphate end of the primary transcript is hydrolyzed to a diphosphate, then capped by transfer of GMP from GTP to the diphosphate RNA end to form a blocked GpppRNA end, and methylated at the cap guanine-N7 with AdoMet as the donor. The reactions are catalyzed by the enzymes RNA triphosphatase, GTP:RNA guanylyltransferase, and RNA (guanine-N7)-methyltransferase, respectively. The requirement of the guanylyltransferase for a 5′-diphosphate RNA as the cap acceptor ensures that capping is restricted to RNA termini initiated de novo and is excluded from 5′-monophosphate termini generated by nucleolytic processing. Genes encoding m7G cap-forming enzymes were identified initially by reverse-genetic approaches after the proteins had been purified based on enzymatic activity. Mutational mapping of the active sites, the proliferation of genome sequencing, and improved bioinformatics have now simplified the detection of m7G capping enzymes. The catalytic strategies and crystal structures of exemplary cellular and viral triphosphatase, guanylyltransferase, and guanine-N7 methyltransferase enzymes are known. Genetic manipulations in model organisms or cultured cells have established the universal requirement for mRNA guanylylation for viability, reflecting the fundamental role of the cap in facilitating translation initiation and protecting the 5′ RNA end from exonucleolytic decay. Small nuclear and nucleolar RNAs that program pre-mRNA splicing (U1, U2, U4, and U5), pre-rRNA processing (U3 and U8), and telomere addition (telomerase RNA) have a 2,2,7-trimethylguanosine (TMG) cap structure (Figure 1). TMG is formed by the enzyme Tgs1, which catalyzes two successive methyltransfer reactions from AdoMet to the N2 atom of guanosine (Mouaikel et al., 2002Mouaikel J. Verheggen C. Bertrand E. Tazi J. Bordonné R. Mol. Cell. 2002; 9: 891-901Abstract Full Text Full Text PDF PubMed Scopus (164) Google Scholar, Hausmann and Shuman, 2005Hausmann S. Shuman S. J. Biol. Chem. 2005; 280: 4021-4024Crossref PubMed Scopus (32) Google Scholar). Tgs1 activity is strictly dependent on prior guanine-N7 methylation, thereby restricting its activity to RNAs that already have a m7G cap. Amazingly, Tgs1 is nonessential for growth of budding and fission yeasts (Mouaikel et al., 2002Mouaikel J. Verheggen C. Bertrand E. Tazi J. Bordonné R. Mol. Cell. 2002; 9: 891-901Abstract Full Text Full Text PDF PubMed Scopus (164) Google Scholar, Hausmann et al., 2007Hausmann S. Ramirez A. Schneider S. Schwer B. Shuman S. Nucleic Acids Res. 2007; 35: 1411-1420Crossref PubMed Scopus (25) Google Scholar), e.g., an S. pombe tgs1Δ strain grows normally, notwithstanding the absence of TMG caps on its U1, U2, U4, and U5 snRNAs. These results are surprising, given that TMG caps decorate so many important fungal RNAs. Although the TMG cap was discovered in 1974, it was only 5 years ago that Remy Bordonné and colleagues identified S. cerevisiae Tgs1 as the likely agent of its synthesis, not via a functional purification but in a yeast two-hybrid screen for proteins that interact with SmB, a protein component of yeast snRNPs (Mouaikel et al., 2002Mouaikel J. Verheggen C. Bertrand E. Tazi J. Bordonné R. Mol. Cell. 2002; 9: 891-901Abstract Full Text Full Text PDF PubMed Scopus (164) Google Scholar). The presence of an AdoMet-binding motif in the then uncharacterized Tgs1 polypeptide was the tip off that this might be the long-sought cap trimethylating enzyme. In a study published in the July 20 issue of Molecular Cell, Benoit Coulombe and colleagues used a complementary system-wide approach to solve one of the outstanding mysteries of RNA capping: the genesis of the distinctive γ-methylphosphate cap structure found in small RNAs such as mammalian U6 and 7SK, mouse B2, and plant U3 (Figure 1). Discovered in human U6 snRNA by Ram Reddy's lab (Singh and Reddy, 1989Singh R. Reddy R. Proc. Natl. Acad. Sci. USA. 1989; 86: 8280-8283Crossref PubMed Scopus (116) Google Scholar), the γ-methylphosphate cap is formed by methyltransfer from AdoMet to a γ-phosphate oxygen at the unprocessed 5′ end of a primary transcript (Gupta et al., 1990Gupta S. Singh R. Reddy R. J. Biol. Chem. 1990; 265: 9491-9495Abstract Full Text PDF PubMed Google Scholar). 5′-diphosphate ends are not methylated, thus ensuring that any ends that have been processed by the RNA triphosphatase component of the mRNA capping pathway cannot be protected thereafter by β-methylation. γ-phosphate methylation of U6, 7SK, B2, and plant U3 RNAs in cell extracts depends on cis-acting RNA signals, comprising a 5′ stem loop followed by a short single-stranded region (Shumyatsky et al., 1994Shumyatsky G. Simba S. Reddy R. Gene Expr. 1994; 4: 29-41PubMed Google Scholar). A 440-fold purification of the γ-methyltransferase from human cell extracts using U6 as the methyl acceptor yielded a 130 kDa polypeptide that cosedimented with the activity in a glycerol gradient (Shimba and Reddy, 1994Shimba S. Reddy R. J. Biol. Chem. 1994; 269: 12419-12423Abstract Full Text PDF PubMed Google Scholar). The purified human enzyme also catalyzed methylation of 7SK, B3, and plant U3 RNA 5′ ends, implying that one protein was responsible for all γ-methylphosphate capping. Fast forward to the present, as Jeronimo et al., 2007Jeronimo C. Forget D. Bouchard A. Li Q. Chua G. Poitras C. Thérien C. Bergeron D. Bourassa S. Greenblatt J. et al.Mol. Cell. 2007; 27: 262-274Abstract Full Text Full Text PDF PubMed Scopus (327) Google Scholar surveyed protein-protein interactions of a battery of human transcription and RNA processing factors by tandem affinity-tag purification (TAP) and mass spectrometry. Among the many interesting interactions documented, one stands out. The previously uncharacterized BCDIN3 protein was recovered in a complex with TAP-tagged Cdk9. Cdk9 is the protein kinase subunit of the RNA polymerase II (RNAP II) elongation factor P-TEFb, which overcomes a promoter-proximal elongation arrest by phosphorylating the RNAP II CTD and the RNAP II elongation factor Spt5. P-TEFb activity is negatively regulated by association with a ribonucleoprotein complex composed of 7SK RNA and HEXIM proteins (Peterlin and Price, 2006Peterlin B.M. Price D.H. Mol. Cell. 2006; 23: 297-305Abstract Full Text Full Text PDF PubMed Scopus (787) Google Scholar). TAP tagging of BCDIN3 revealed its reciprocal association with Cdk9 and cyclin T (the cyclin subunit of P-TEFb), as well as HEXIMs and 7SK RNA. U6 snRNA was also associated with BCDIN3, but U2 snRNA was not. Taking a page from the Tgs1 story, Jeronimo et al. noted that the C-terminal half of human BCDIN3 contains AdoMet-dependent methyltransferase motifs. Prompted by the association with 7SK and U6, they posited and then proved that BCDIN3 promotes methyltransfer from 3H-AdoMet to 7SK RNA in vitro, yielding a labeled product with the characteristics of a γ-methylphosphate cap. Convincing homologs of BCDIN3 are found in metazoan, plant, and fission yeast proteomes, though not in budding yeasts. Mystery solved? Not entirely. The authors use siRNA to downregulate BCDIN3 levels by 70% in cultured cells and find that this elicits a decrease in the steady-state level of 7SK, without affecting U6 or U2 snRNA levels. They surmise that 7SK levels decline because the ends are not capped and the RNA is degraded. However, the 5′ structure and half-life of 7SK and U6 RNAs produced by the siRNA-treated cells was not analyzed, which limits the inferences that can be drawn from the experiment. These new results are exciting and raise many fascinating questions. One pressing issue is whether BCDIN3 is the enzyme responsible for all γ-methylphosphate capping. It should be straightforward to test if purified BCDIN3 is able to cap U6, B2, and plant U3 RNAs as well as 7SK and to then define any cis-acting RNA determinants of cap-acceptor specificity. The relationship of BCDIN3 to the enzyme purified from human cells by Shimba and Reddy, 1994Shimba S. Reddy R. J. Biol. Chem. 1994; 269: 12419-12423Abstract Full Text PDF PubMed Google Scholar is also of interest. Although the predicted 689 aa human BCDIN3 polypeptide is smaller than 130 kDa, its N-terminal half is composed of iterated tracts of glycine, proline, and basic residues that could perturb its electrophoretic mobility. The affinity-purified BCDIN3 polypeptide migrates at ∼85 kDa, but because the TAP tag is on the C terminus, it is not clear if the unstructured N-terminal domain is intact. The most intriguing question is whether γ-methylphosphate capping is essential and, if so, for what? Given the association of the BCDIN3 protein with key transcription regulators, it could function in such complexes independent of its methyltransferase activity. Finally, what is a 7SK γ-methylphosphate capping enzyme homolog doing in S. pombe, an organism that has no recognizable 7SK RNA or HEXIM proteins? Fission yeast do have a P-TEFb homolog composed of essential kinase and cyclin subunits (Cdk9 and Pch1). Perhaps S. pombe Cdk9 is regulated by a novel RNA with a γ-methylphosphate cap. Does S. pombe U6 snRNA have a γ-methylphosphate cap and, if so, is it important for U6 function in splicing? In summary, systems approaches have yielded new molecular tools to study TMG and γ-methylphosphate capping. The next capping mystery is the identity of the mammalian enzymes that methylate the ribose-O2′ atoms of the terminal and penultimate nucleosides of mRNAs to form the cap 1 and cap 2 structures discovered more than 30 years ago. Systematic Analysis of the Protein Interaction Network for the Human Transcription Machinery Reveals the Identity of the 7SK Capping EnzymeJeronimo et al.Molecular CellJuly 20, 2007In BriefWe have performed a survey of soluble human protein complexes containing components of the transcription and RNA processing machineries using protein affinity purification coupled to mass spectrometry. Thirty-two tagged polypeptides yielded a network of 805 high-confidence interactions. Remarkably, the network is significantly enriched in proteins that regulate the formation of protein complexes, including a number of previously uncharacterized proteins for which we have inferred functions. The RNA polymerase II (RNAP II)-associated proteins (RPAPs) are physically and functionally associated with RNAP II, forming an interface between the enzyme and chaperone/scaffolding proteins. Full-Text PDF Open Archive
Referência(s)