NATURE STRUCTURAL & MOLECULAR BIOLOGY VOLUME 16 NUMBER 9 SEPTEMBER 2009 907 24. Doyen, C.M. et al. Mol. Cell. Biol. 26, 1156–1164 (2006). 25. Angelov, D. et al. Mol. Cell 11, 1033–1041 (2003). 26. Agelopoulos, M. & Thanos, D. EMBO J. 25, 4843–4853 (2006). 27. Wacker, D.A. et al. Mol. Cell. Biol. 27, 7475–7485 (2007). 28. Cohen-Armon, M. et al. Mol. Cell 25, 297–308 (2007). 29. Changolkar, L.N. et al. Mol. Cell. Biol. 27, 2758–2764 (2007). 30. Costanzi, C. & Pehrson, J.R. J. Biol. Chem. 276, 21776–21784 (2001). 31. Pehrson, J.R., Costanzi, C. & Dharia, C. J. Cell. Biochem. 65, 107–113 (1997). 32. Flaus, A., Martin, D.M., Barton, G.J. & Owen-Hughes, T. Nucleic Acids Res. 34, 2887–2905 (2006). Althaus, F.R. J. Biol. Chem. 275, 40974–40980 (2000). 21. Rulten, S.L., Cortes-Ledesma, F., Guo, L., Iles, N.J. & Caldecott, K.W. Mol. Cell. Biol. 28, 4620–4628 (2008). 22. Nusinow, D.A. et al. J. Biol. Chem. 282, 12851–12859 (2007). 23. Ouararhni, K. et al. Genes Dev. 20, 3324–3336 (2006). Yvonne Klaue and Klemens J. Hertel are in the Department of Microbiology & Molecular Genetics, University of California, Irvine, California, USA. e-mail: khertel@uci.edu that the speed of transcript cleavage by the hepatitis δ ribozyme is too fast even to permit stable interactions between newly synthesized exons and elongating Pol II, thus preventing tethering. An alternative is that exon tethering to elongating Pol II is not a commonly used mechanism to ensure coupling between transcription and RNA processing. In addition to the lack of physical evidence supporting the exon–Pol II interaction, Fong et al. provide evidence that argues for the second proposal. Using chromatin immunoprecipitation (ChIP) analysis along the transcribed transcript, they monitored cap-binding complex (CBC) association, which binds the 5′ end of nascent transcripts as a consequence of the initial steps of pre-mRNA processing. The CBC signal dropped substantially downstream of the hepatitis δ ribozyme–induced transcript cut, consistent with the interpretation that the 5′ portion of the cleaved RNA is released from the transcription unit. However, when identical experiments were carried out in the context of the slower-cleaving hammerhead ribozyme, RNA processing was essentially undisturbed. Without doubt, both ribozymes cleave the nascent transcript efficiently. How, then, is the transcription and RNA processing unit able to deal with hammerhead-cleaved transcripts? The different speeds at which cleavage is achieved suggests that timing is important. It is possible that transcript cleavage occurs mainly after RNA processing has been completed or, alternatively, that nascent RNAs may become immune to cleavage once a committed pre- spliceosomal complex has been assembled. Spliceosome assembly starts with the deposition of U1 small nuclear transcript is cut. The model of exon tethering therefore proposes that exons of nascent RNAs remain associated with Pol II until complete spliceosomal assembly and intron removal releases ligated splicing intermediates 6 . In their work, Fong et al. re-investigate the requirement for a continuous transcript and the concept of exon tethering to Pol II during RNA synthesis 1 . Anticipating an important kinetic component in maintaining efficient coupling, Fong et al. used two self-cleaving ribozymes with drastically different self-cleavage rates. The effects of transcript cleavage on RNA processing were strikingly different between the ribozymes used. The slow-cleaving hammerhead ribozyme effectively reproduced earlier observations that internal cleavage of nascent transcripts did not considerably interfere with RNA processing. By contrast, transcript disruption by the hepatitis δ ribozyme, which cleaves ten-fold faster than the hammerhead ribozyme, resulted in essentially complete inhibition of intron splicing, 3′ end processing and transcription termination. Surprisingly, splicing defects were not only detected within the intron or exon that harbored the self-cleaving ribozyme, but also in neighboring introns. These results show, at least in the context of the relatively short reporter genes used, that the decision to abort processing is propagated throughout the transcript before completion of intron removal (Fig. 1a). Presumably, the 5′ and 3′ termini generated by the ribozyme are then points of attack for abundant exonucleases that quickly degrade the cleavage products. How do these results influence our view on the exon tethering model? It is possible On page 916 of this issue, Fong et al. 1 show that fast premature cleavage of nascent pre-mRNAs aborts pre-mRNA splicing, 3′ end processing and proper transcription termination. The results suggest that there are mechanisms to monitor the integrity of the newly synthesized transcripts. Upon detection of damaged nascent transcripts, pre-mRNA processing is terminated, preventing the association of processing factors with RNA polymerase II (Pol II) that is not productively engaged in transcription. Over the past decade it has been firmly established that transcription, pre-mRNA splicing and other processing events are intimately coupled 2 . Whereas splice site– recognition factors have been shown to be recruited co-transcriptionally within a short time frame 3,4 , intron removal is not necessarily completed while the nascent pre-mRNA is still attached to elongating Pol II 5 . Recent work has shown that pre-mRNA splicing occurs efficiently in cells even when introns have been cleaved by an inserted ribozyme or a co-transcriptional cleavage element 6 . These results led to the proposal that exons of nascent RNAs are held together through intermolecular interactions with Pol II, permitting severed nascent transcripts to be processed in a similar manner to trans-splicing substrates 7,8 . One conclusion of this earlier study is that splicing can occur even when the Dangerous play—splitting the message may leave you empty handed Yvonne Klaue & Klemens J Hertel An important aspect of eukaryotic gene expression is the efficient integration of transcription, pre-mRNA processing and nuclear export. A new study demonstrates that pre-mRNA transcript continuity is an essential component for maintaining productive coupling of transcription and RNA processing events. NEWS AND VIEWS © 2009 Nature America, Inc. All rights reserved.