Coupling Transcription, Splicing and Mrna Export Robin Reed

326

Coupling transcription, splicing and mRNA export Robin Reed

Previous studies have led to the view that mRNAs are mechanisms coupling splicing and transcription to mRNA transported from the nucleus to the cytoplasm by machinery that export. In addition, new studies that link the nuclear is conserved from yeast to humans. Moreover, this machinery is exosome to the machineries involved in transcription, 30 coupled both physically and functionally to the pre-mRNA end formation and mRNA export will be discussed. splicing machinery. During the past year, important new insights Readers are referred to several recent reviews for more into the mechanisms behind this coupling have been made. extensive descriptions of coupling between the different In addition, recent advances have revealed mechanisms for the steps in gene expression [1–7]. co-transcriptional loading of the export machinery on to mRNAs. Finally, a newly identified link between the nuclear exosome Coupling splicing to mRNA export and the machineries required for transcription, 30-end formation The notion that splicing is linked to mRNA export first and mRNA export suggests that proper mRNP formation is came from studies in metazoans showing that spliced co-transcriptionally monitored. mRNAs are assembled into a distinct ‘spliced mRNP’ complex that targets the mRNA for export (for a review, Addresses see [3]). This targeting involves the splicing-dependent Department of Cell Biology, 240 Longwood Avenue, recruitment of the mRNA export factor Aly via its direct Harvard Medical School, Boston, MA 02115, USA interactions with the spliceosomal protein UAP56. In e-mail: [email protected] Saccharomyces cerevisiae, the counterparts of Aly and UAP56 also interact with each other directly and are Current Opinion in Cell Biology 2003, 15:326–331 required for mRNA export. In both S. cerevisiae and metazoans, Aly also interacts directly with Tap, which, This review comes from a themed issue on together with its binding partner p15, associates with the Nucleus and gene expression Edited by Jeanne Lawrence and Gordon Hager NPC and is thought to be a general mRNA export receptor. These and other studies indicate that there is 0955-0674/03/$ – see front matter a conserved machinery for mRNA export, as shown in ß 2003 Elsevier Science Ltd. All rights reserved. Figure 1 (for reviews, see [3,6,8,9]). In metazoans, where DOI 10.1016/S0955-0674(03)00048-6 most genes contain introns, this conserved machinery is intimately tied to the splicing machinery. By contrast, in S. cerevisiae, where few genes contain introns, it appears Abbreviations ChIP chromatin immunoprecipitation that mRNA export is coupled to other steps in gene EJC exon-junction complex expression. NPC nuclear pore complex RNAi RNA interference Work initiated by the Moore group has shown over the RNP ribonucleoprotein TREX transcription export complex past few years that in metazoans the conserved mRNA export machinery (UAP/Aly/Tap) and other components of the spliced mRNP are specifically recruited to a posi- Introduction tion 20 nucleotides upstream of exon–exon junctions During expression of protein-coding genes, pre-mRNAs ([10] and references therein). In addition to export are transcribed in the nucleus and undergo several pro- proteins, this exon-junction complex (EJC) contains cessing steps, including capping, splicing, 30-end proces- several proteins involved in nonsense-mediated decay sing and polyadenylation. The mature mRNA is then (the degradation of mRNAs containing premature stop exported through nuclear pore complexes (NPCs) to the codons) and in the cytoplasmic localization of mRNAs. cytoplasm for translation. Although distinct and highly Thus, the EJC is thought to play multiple roles in mRNA complex cellular machines carry out each of these steps in metabolism. An intriguing feature of the EJC is that it is the gene-expression process, growing evidence suggests positioned on the mRNA at a specific site, but in a the existence of gene-expression factories in which indi- sequence-independent manner. Recently, Moore and vidual machines are functionally coupled. This coupling colleagues have begun to determine the mechanism for is obviously not obligatory, as virtually all of the coupled positioning the EJC [10]. Their analysis revealed that steps in gene expression can be uncoupled in various numerous proteins (nine were detected) interact with the assays. Thus, coupling probably exists to enable the 50 exon in the catalytically active spliceosome. During proofreading and streamlining of the entire process of exon ligation, a major restructuring of this region occurs, gene expression in vivo. In this review, the discussion will placing three other proteins (p170, p95 and p57) at the site focus on advances made in the past year in elucidating the where the EJC forms [10]. Aly and SRp20 (a member of

Current Opinion in Cell Biology 2003, 15:326–331 www.current-opinion.com Coupling transcription, splicing and mRNA export Reed 327

Figure 1 Mattaj and co-workers [13] addressed the role of splicing in mRNA export by constructing a hybrid U1 snRNA in which exons, introns or random sequences were inserted. The authors concluded that mRNAs contain an ‘identity Spliceosome element’, which they defined as an unstructured sequence at least 300 nucleotides long. This identity element is UAP sufficient to re-direct the hybrid RNA away from the U1 snRNP export pathway and into the UAP/Aly/Tap mRNA export pathway, even in the absence of splicing. Consis- tent with previous studies (for a review, see [3]), the implication of the new work is that splicing is not essential Spliced for export but increases the efficiency and/or fidelity of ALY UAP mRNP the process by promoting the more efficient recruitment of export factors.

In an unexpected twist to the emerging story of the conserved mRNA export pathway and the EJC, a new Drosophila RNA interference (RNAi) study by Gatfield and Izaurralde [14] concludes that, although both Tap ALY and UAP56 are required for mRNA export, Aly and the TAP other EJC components are dispensable. The authors propose that an as-yet undiscovered adaptor protein(s) must exist that mediates the interaction between Tap and the mRNA. However, there are alternative possibilities; Nucleus the RNAi-mediated knockdown of Aly and the other EJC proteins may not have been sufficient to eliminate their Cytoplasm functions, or there may be a redundancy in factors. Indeed, the new conclusion is difficult to reconcile with previous work. For example, antibodies to Aly block mRNA export, and recombinant Aly stimulates mRNA export (for a review, see [3]). Moreover, the physical and functional conservation of UAP, Aly, and Tap from yeast Current Opinion in Cell Biology to humans implicates these proteins as key players in mRNA export [3,8]. Conserved mRNA export machinery coupled to splicing. UAP56 (Sub2 in yeast), which is present in the spliceosome, recruits Aly (Yra1 in yeast) to the mRNA during splicing. UAP56 is replaced by the mRNA export Coupling transcription to mRNA export receptor TAP/p15 (Mex67/MTR2 in yeast). Although the UAP/Aly/Tap pathway is conserved from yeast to humans, the export machinery is not solely recruited by a splicing-dependent mechanism. Indeed, 95% of S. cerevisiae genes as well as several metazoan the SR protein family of splicing factors) also contact this genes lack introns. Nevertheless, in both metazoans and region in the EJC [10]. These five RNA-contacting yeast, the conserved UAP/Aly/Tap pathway is required proteins are therefore likely to play key roles in EJC for the export of mRNAs derived from genes lacking formation. Once formed, the EJC is associated with introns ([14] and references therein). Thus, a central focus mRNAs that are bound in the nucleus by the cap-binding of recent work has been to determine the mechanisms for protein CBP80 but not with mRNAs bound by the splicing-independent recruitment of this machinery. In cytoplasmic eIF4E cap-binding protein [11]. The EJC earlier work, the Silver group shed light on this problem proteins associate with CBP80-bound mRNAs both in the by providing evidence that the mRNA export machinery nucleus and in the cytoplasm, suggesting that the entire is co-transcriptionally recruited in S. cerevisiae (for a EJC is exported and then dissociates from the mRNA in review, see [6]). During the past year, new advances have the cytoplasm [11]. In yeast, Lei and Silver [12] used the been made that support this view and that identify chromatin immunoprecipitation (ChIP) method to show possible mechanisms for the recruitment. that Yra1 (yeast Aly) and Sub2 (yeast UAP56) associate preferentially with the intron region, in a splicing-depen- In one recent study in S. cerevisiae, Strasser et al. [15] dent manner. Thus, it is possible that both splicing- showed that Sub2 and Yra1 specifically associate with the dependent recruitment of the mRNA export machinery multisubunit THO complex, which is thought to be and EJC formation are also conserved. involved in transcription elongation (for a review, see www.current-opinion.com Current Opinion in Cell Biology 2003, 15:326–331 328 Nucleus and gene expression

[16]). Null mutants in any of the non-essential THO that it associates with the cytoplasmic fibrils of the complex subunits (Hpr1, Tho2, Mtf1 and Thp2) also NPC. Significantly, when Sac3 is mutated, Mex67 accu- display an unusually high level of transcription-depen- mulates at the nuclear rim [20]. On the basis of these dent recombination ([16,17] and references therein). This results, Silver and co-workers [20] propose that Sac3 hyper-recombination phenotype may be a secondary con- plays a role in a terminal step in the process whereby sequence of the polymerase elongation defect, which Mex67 is translocated across the NPC. Considering both exposes transcription units to recombination [16].Inthe the Hurt and Silver studies, it is possible that Sac3 study by Strasser et al. [15], all four subunits of the THO associates with the mRNA during transcription, plays complex were found to interact genetically and physically a role in docking Mex67 to the nuclear pore and then with bothYra1 and Sub2. In addition, the THO-complex functions in the translocation of Mex67 across the null mutants are defective in mRNA export. Finally, ChIP nuclear pore. Metazoan counterparts of Sac3 and assays revealed that THO complex components, as well as Thp1 have been identified. Thus, it would not be Yra1 and Sub2, associate with actively transcribed genes surprising if these proteins play a role in mRNA export [15,18]. The THO complex together with the mRNA in metazoans. export proteins is designated the transcription export complex (TREX) [15]. Together, the data support a model in Links between mRNA export and other which the TREX complex plays a role in coupling tran- steps in gene expression scription to mRNA export. Stutz and co-workers [18] Surprising new studies by the Jensen group [21] and the have advanced this model further by showing that the Stutz group [18] have linked the TREX complex to the THO/TREX component Hpr1 interacts directly with nuclear exosome, a large complex of 30 to 50 exonucleases Sub2 and plays a role in recruiting Sub2 and Yra1 to actively involved in numerous RNA processing and degradation transcribed genes. pathways. The two new studies have shown genetic interactions between exosome and TREX complex com- In another study in S. cerevisiae, Hurt and co-workers [19] ponents. Moreover, Yra1 interacts physically with the identified what are likely to be new components of the exosomal protein Rrp45 [18]. Previous studies have conserved mRNA export machinery, with links to the shown that TREX complex mutants generate transcripts transcription machinery. Performing a synthetic lethal that are truncated at their 30 ends (for a review, see [16]). screen with Yra1 (yeast Aly), they identified the Sac3 Jensen and colleagues have now shown that in these gene. The Sac3 protein was also identified by mass mutants, as well as in mutants that affect proper 30 end spectrometry by virtue of its association with Sub2 (yeast formation, mRNAs are rapidly sequestered near their UAP56) in a pull-down assay from yeast lysates [15]. sites of transcription ([21] and references therein). Both Sac3 was also shown to interact directly with Mex67 this phenotype and the 30-end truncation phenotype can (yeast Tap). Thus, Sac3 interacts physically and func- be rescued by deletion of the exosome component Rrp6 tionally with all of the components of the UAP/Aly/TAP [21]. Together, these studies led to the view that proper pathway. But the story does not end there; Sac3 also forms mRNP formation is monitored by the exosome, resulting a complex with Thp1, a protein thought to function in in retention and destruction of aberrant mRNPs at their transcription elongation, and Hurt and coworkers [19] sites of transcription [7,18,21]. showed that both Sac3 and Thp1 are required for mRNA export. Moreover, Sac3 also interacts with two nucleo- Significantly, another link was recently made between porins (Nup 60 and Nup 1) that are located on the the exosome and transcription elongation factors (Spt5 nucleocytoplasmic face of the NPC and that are required and Spt6) [22]. In this study, Lis and co-workers showed for mRNA export. On the basis of these observations, that the exosome physically associates with Spt5 and Spt6 Hurt and coworkers proposed a model in which Sac3 first and is recruited to actively transcribed genes, possibly by associates with the mRNP early in its formation, perhaps these proteins. Together, the new data suggest a model during transcription, and then, by interacting with both for how the exosome, the TREX complex, Spt5 and Spt6 Mex67 and the two nucleoporins, facilitates docking of regulate transcription elongation, mRNP formation and the mRNP at the nuclear-pore complex. export. In this model, Spt5 and Spt6 recruit the exosome to RNA polymerase II at an early stage of transcription Silver’sgroup[20] also identified Sac3 in a synthetic (Figure 2). The TREX complex, which associates with lethal screen using a known mRNA export factor, in this the gene at a later stage of transcription, may physically case MTR2 (the small subunit of the Mex67/MTR2 protect the mRNP from the exosome or may even func- mRNA export receptor [Tap/p15 in metazoans]). Sac3 tion as an exosome inhibitor (Figure 2a). When the was shown to function in mRNA export, to interact TREX complex is missing (as in the TREX-null genetically with several mRNA export factors and to mutants), the 30 ! 50 exonuclease activity of the exosome associate physically with Mex67, MTR2 and specific destroys the growing mRNP, leading to the dissociation nuclear-pore-complex-associated proteins involved in of RNA polymerase from DNA and thus blocking tran- mRNA export. Immunolocalization of Sac3 revealed scription elongation (Figure 2b). When both the TREX

Current Opinion in Cell Biology 2003, 15:326–331 www.current-opinion.com Coupling transcription, splicing and mRNA export Reed 329

Figure 2

Exosome

Spt5,6

RNAP II

5′ (a)TREX (b) (c) 3′ Exosome ALY Spt5,6 UAP Spt5,6 Spt5,6S Exosome RNAP II RNAP II RNAP II

5′ ALY UAP 3′ truncated/degraded Full length

Exported Not exported Not exported

Current Opinion in Cell Biology

In this model, coupling the transcription and export machineries to the exosome enables the monitoring of proper mRNP formation. The exosome is recruited during transcription, possibly by the elongation factors Spt5 and Spt6. When the TREX complex is present, a proper mRNP is made containing export proteins (Aly and UAP56) and the mRNA is exported (a). In the absence of the TREX complex, the exosome degrades the mRNA from the 30 to the 50 end and the transcription machinery dissociates (b). When both the TREX complex and the exosome are absent, transcription can occur but the mRNP lacks export proteins and is not exported (c).

complex and the exosome are missing, full-length tran- by Hammel et al. [24]. In a screen to identify genes scripts would be generated but not exported because the required for mRNA export, several factors involved in export machinery cannot be recruited (Figure 2c). Thus, termination and 30 processing were obtained. The study the purpose of coupling the exosome to the TREX also showed that the deletion of sequences in the mRNA complex is to ensure that any transcripts that lack the that are involved in coupling termination and 30 end TREX complex, and hence the mRNA export machinery, formation leads to the synthesis of transcripts that cannot are targeted for destruction. be exported. Thus, the authors propose that termination, 30 processing and mRNA export are all coupled. Although ample evidence indicates that the mRNA export machinery is co-transcriptionally loaded onto Coupling transcription and splicing to mRNAs in yeast, 30-end formation may play a more mRNA export in metazoans critical role in this loading. For example, Lei and Silver Numerous recent studies in metazoans have provided [12] ﬁnd that Sub2 is not required for the recruitment of compelling new evidence that transcription can affect Yra1 to genes that lack introns. However, proper 30-end splicing and vice versa (see for example [25–28]). Because formation is required to recruit this factor, whether or not of space restrictions, the discussion here will be limited to an intron is present. This observation supports earlier recent advances concerning the coupling of both tran- studies indicating that 30-end formation is important for scription and splicing to mRNA export. In particular, mRNA export (for reviews, see [6,7]). In further support Strasser et al. [15] identiﬁed the likely counterpart of of this view, Rosbash’s group has recently reported that the TREX complex in mammals. As in yeast, this com- transcripts generated by T7 polymerase (i.e. not by RNA plex contains mRNA export proteins (Aly and UAP56) as polymerase II) are exported only if 30-end formation well as THO complex components (hTho2 and hHpr1). occurs normally [23]. This story has also been extended It is not yet known whether this putative mammalian www.current-opinion.com Current Opinion in Cell Biology 2003, 15:326–331 330 Nucleus and gene expression

Figure 3 Acknowledgements I am grateful to Ming-Juan Luo, Ed Hurt and Tom Maniatis for helpful comments on the review and to Claudine Christoforides for assistance in preparation of the manuscript. I apologize for references that were not cited because of space limitations.

References and recommended reading ALY UAP Papers of particular interest, published within the annual period of Spliceosome review, have been highlighted as:

ALY UAP of special interest of outstanding interest 1. Orphanides G, Reinberg D: A unified theory of gene expression. TREX Cell 2002, 108:439-451. 2. Bentley D: The mRNA assembly line: transcription and processing machines in the same factory. Curr Opin Cell Biol RNAP II 2002, 14:336-342. A conserved mRNA export machinery coupled Current Opinion in Cell Biology 3. Reed R, Hurt E: to pre-mRNA splicing. Cell 2002, 108:523-531. 4. Maniatis T, Reed R: An extensive network of coupling among In this model, coupling transcription, splicing and export via the TREX gene expression machines. Nature 2002, 416:499-506. complex targets the export machinery to exons. Metazoan introns are typically very large whereas the exons are minute. The TREX 5. Proudfoot NJ, Furger A, Dye MJ: Integrating mRNA processing with transcription. Cell 2002, 108:501-512. complex may associate with the spliceosome to allow co-transcriptional loading of the export machinery (Aly and UAP56) onto exons, which are 6. Lei EP, Silver PA: Protein and RNA export from the nucleus. destined for export, and to exclude this machinery from the introns, Dev Cell 2002, 2:261-272. which are retained in the nucleus. 7. Jensen TH, Rosbash M: Co-transcriptional monitoring of mRNP formation. Nat Struct Biol 2003, 10:10-12. 8. Izaurralde E: A novel family of nuclear transport receptors mediates the export of messenger RNA to the cytoplasm. TREX complex associates with the nascent pre-mRNA Eur J Cell Biol 2002, 81:577-584. during transcription. Surprisingly, however, all of the 9. Dreyfuss G, Kim VN, Kataoka N: Messenger-RNA-binding mammalian TREX complex components were found in proteins and the messages they carry. Nat Rev Mol Cell Biol purified spliceosomes [29–32]. Thus, the TREX complex 2002, 3:195-205. may function to link both transcription and splicing to 10. Reichert VL, Le Hir H, Jurica MS, Moore MJ: 50 exon interactions mRNA export in metazoans. One model that may explain within the human spliceosome establish a framework for exon junction complex structure and assembly. Genes Dev 2002, why the TREX complex is linked to splicing in metazo- 16:2778-2791. ans concerns the structure of their genes. In metazoans, The authors explored formation of the EJC. The data show that a set of proteins contacts the exon during splicing and that extensive remodeling the abundant introns are typically thousands to tens of then occurs in this region as the exon junction is formed. thousands of nucleotides in length and can be as large as 11. Lejeune F, Ishigaki Y, Li X, Maquat LE: The exon junction complex several hundred thousand nucleotides. The exons, aver- is detected on CBP80-bound but not eIF4E-bound mRNA in aging only 100 nucleotides, are minute by comparison. mammalian cells: dynamics of mRNP remodeling. EMBO J 2002, 21:3536-3545. Thus, as shown in Figure 3, the TREX complex may be associated with the spliceosome as a mechanism for co- 12. Lei EP, Silver PA: Intron status and 30-end formation control cotranscriptional export of mRNA. Genes Dev 2002, transcriptionally targeting the mRNA export factors only 16:2761-2766. to exons, which are destined for export, and not to the 13. Ohno M, Segref A, Kuersten S, Mattaj IW: Identity elements used numerous enormous introns, which are retained and in export of mRNAs. Mol Cell 2002, 9:659-671. degraded in the nucleus. 14. Gatfield D, Izaurralde E: REF1/Aly and the additional exon junction complex proteins are dispensable for nuclear mRNA Conclusions export. J Cell Biol 2002, 159:579-588. Recent studies strengthen the view that mRNA export is 15. Strasser K, Masuda S, Mason P, Pfannstiel J, Oppizzi M, Rodriguez-Navarro S, Rondon AG, Aguilera A, Struhl K, Reed R coupled to other steps in gene expression including et al.: TREX is a conserved complex coupling transcription with splicing, transcription, and 30-end formation. In addition, messenger RNA export. Nature 2002, 417:304-308. A new complex containing mRNA export factors and transcription factors new links between mRNA export, transcription and the is identified in yeast and mammals. The complex may function in the nuclear exosome were revealed. Future directions will co-transcriptional loading of mRNA export factors. include a detailed characterization of the spliced mRNP 16. Jimeno S, Rondon AG, Luna R, Aguilera A: The yeast THO and EJC that couples splicing to mRNA export. In addi- complex and mRNA export factors link RNA metabolism with transcription and genome instability. EMBO J 2002, tion, the role of the TREX complex in export and 21:3526-3535. transcription, and the relationship between these pro- 17. Aguilera A: The connection between transcription and genomic cesses and the nuclear exosome remain to be determined. instability. EMBO J 2002, 21:195-201. 0 Finally, the molecular basis for coupling 3 -end formation 18. Zenklusen D, Vinciguerra P, Wyss JC, Stutz F: Stable mRNP to mRNA export must be identified. formation and export require cotranscriptional recruitment of

Current Opinion in Cell Biology 2003, 15:326–331 www.current-opinion.com Coupling transcription, splicing and mRNA export Reed 331

the mRNA export factors Yra1p and Sub2p by Hpr1p. 23. Dower K, Rosbash M: T7 RNA polymerase-directed transcripts Mol Cell Biol 2002, 22:8241-8253. are processed in yeast and link 30 end formation to mRNA This study shows that the THO/TREX complex component Hpr1 interacts nuclear export. RNA 2002, 8:686-697. with the mRNA export factor Sub2 and plays a role in recruiting Sub2 to This study shows that 30-end formation plays a key role in mRNA export in the mRNA during transcription. The study also shows links between the yeast. TREX complex and the nuclear exosome, suggesting that proper mRNP formation is co-transcriptionally monitored. 24. Hammell CM, Gross S, Zenklusen D, Heath CV, Stutz F, Moore C, Cole CN: Coupling of termination, 30 processing, and mRNA 19. Fischer T, Strasser K, Racz A, Rodriguez-Navarro S, Oppizzi M, export. Mol Cell Biol 2002, 22:6441-6457. Ihrig P, Lechner J, Hurt E: The mRNA export machinery These authors report that termination and 30 end processing are coupled requires the novel Sac3p-Thp1p complex to dock at the to mRNA export in yeast. nucleoplasmic entrance of the nuclear pores. EMBO J 2002, 21:5843-5852. 25. Kwek KY, Murphy S, Furger A, Thomas B, O’Gorman W, Kimura H, Sac3 is identified as a new mRNA export factor. Sac3 was shown to Proudfoot NJ, Akoulitchev A: U1 snRNA associates with TFIIH interact with specific proteins, leading to the proposal that Sac3 is loaded and regulates transcriptional initiation. Nat Struct Biol 2002, onto mRNA co-transcriptionally and later functions to dock the mRNP at 9:800-805. the nuclear pore. 26. Auboeuf D, Honig A, Berget SM, O’Malley BW: Coordinate 20. Lei EP, Stern CA, Fahrenkrog B, Krebber H, Moy TI, Aebi U, regulation of transcription and splicing by steroid receptor Silver PA: Sac3 is an mRNA export factor that localizes to the coregulators. Science 2002, 298:416-419. cytoplasmic fibrils of the nuclear pore complex. Mol Biol Cell 2002, 14:836-847. 27. Furger A, O’Sullivan JM, Binnie A, Lee BA, Proudfoot NJ: Promoter This study also identifies Sac3 as a new mRNA export factor. The authors proximal splice sites enhance transcription. Genes Dev 2002, show that Sac3 localizes to the cytoplasmic fibrils and may play a role in a 16:2792-2799. terminal step of mRNP translocation. 28. Kadener S, Fededa JP, Rosbash M, Kornblihtt AR: Regulation of 21. Libri D, Dower K, Boulay J, Thomsen R, Rosbash M, Jensen TH: alternative splicing by a transcriptional enhancer through RNA Interactions between mRNA export commitment, 30-end pol II elongation. Proc Natl Acad Sci USA 2002, 99:8185-8190. quality control, and nuclear degradation. Mol Cell Biol 2002, 29. Jurica MS, Licklider LJ, Gygi SR,Grigorieff N, Moore MJ: Purification 22:8254-8266. and characterization of native spliceosomes suitable for This study reports a link between the TREX complex and the nuclear three-dimensional structural analysis. RNA 2002, 8:426-439. exosome, providing further evidence for the author’s proposal that the exosome monitors proper mRNP formation co-transcriptionally. 30. Hartmuth K, Urlaub H, Vornlocher HP, Will CL, Gentzel M, Wilm M, Luhrmann R: Protein composition of human prespliceosomes 22. Andrulis ED, Werner J, Nazarian A, Erdjument-Bromage H, isolated by a tobramycin affinity-selection method. Tempst P, Lis JT: The RNA processing exosome is linked to Proc Natl Acad Sci USA 2002, 99:16719-16724. elongating RNA polymerase II in Drosophila. Nature 2002, 420:837-841. 31. Zhou Z, Licklider LJ, Gygi SP, Reed R: Comprehensive proteomic This study is the first to show that the nuclear exosome is recruited to analysis of the human spliceosome. Nature 2002, 419:182-185. actively transcribed genes. The elongation factors Spt5 and Spt6 are in a complex with the exosome and may play a role in recruiting it to active 32. Rappsilber J, Ryder U, Lamond AI, Mann M: Large-scale genes. The study further strengthens the notion that proper mRNP proteomic analysis of the human spliceosome. Genome Res formation is monitored during transcription. 2002, 12:1231-1245.

www.current-opinion.com Current Opinion in Cell Biology 2003, 15:326–331