© 2008 Nature Publishing Group http://www.nature.com/nsmb
news and views reduced accuracy2), or even an artifact caused by the particular scaffold design, absence of NTP or other factors. The importance of Met932 for bacterial transcription cannot be unequivocally associated with its role in wedging of the trigger loop into the bridge helix, because its counterpart Met1238 was also reported to stack on the substrate base in the T. thermophilus EC structure9, and its substitution may have unknown effects on trigger loop mobility and/or refolding. Similar substitution of a nearby Thr934 had an even more drastic effect (22-fold) on pausing without any direct involvement in the formation of the wedge7. Regardless of the extent to which this wedged conformation represents an in- pathway translocation intermediate, Brueckner and Cramer have made several important contributions to our understanding of how RNAP functions. The new structure of yeast Pol II EC allows for more rigorous molecular dynamics simulations of transcription elongation, which, together with structural studies, bulk and single-molecule biochemistry, and kinetic analysis, are needed to generate an explicit physical model of transcription. Direct observation of the equilibrium between the preand post-translocated ECs lends further support to the concept of RNAP as a Brownian ratchet. This work also attempts a synthesis of bridge helix–centric and trigger loop–centric models of elongation by emphasizing the concerted movements of the bridge helix and trigger loop during translocation. Broad acceptance of the Brownian ratchet mechanism of multisubunit RNAPs (as opposed to a power-stroke mechanism) began with the bridge helix–centric model of Bar-Nahum et al., who demonstrated
that mutations in the trigger loop affected bridge helix conformations as well as enzyme translocation, fidelity and response to regulatory signals and factors5. Bridge helix conformational changes in that work were observed directly using cross-linking approaches. However, the particular structural manifestations of the bridge helix–centric thermal ratchet (bending and straightening of the bridge helix) seemed to be a result of an earlier crystallographic aberration, which in turn called into question the significance of the bridge helix element in RNAP function (the extreme trigger loop– centric model discounted the importance of the bridge helix altogether7). The work of Brueckner and Cramer restores the role of the bridge helix as crucial in RNAP translocation and provides a more accurate view of the structural changes (shifting of the central portion of bridge helix) that accompany the repositioning of the DNA template to a new, ‘pretemplating’ position in an intermediate translocation state. The movement of the bridge helix in and out of the pretemplating position may depend on the wedged trigger loop, which can push the bridge helix into position, stabilize the shifted bridge helix or act passively as a boundary condition directing the distortion of the bridge helix by a force arising elsewhere. Thus, bridge helix oscillation could still be a structural manifestation of the Brownian ratchet in translocation. Swinging of the refolded substrate-bound trigger loop, which takes the shape of an α-helical hairpin packed against the bridge helix, toward the insertion site could provide the initial velocity and directionality of the nucleophilic attack of the RNA 3′-OH on the α-phosphate of the substrate NTP. Indeed,
molecular dynamics simulations of the yeast Pol II NAC suggest that the direction of the attack is reversed in such a way that it is the substrate bound to the mobile trigger loop that attacks the static nucleophile (RNA)14. Thus, the bridge helix and trigger loop can switch between ‘leading’ and ‘assisting’ roles during translocation and catalysis. An integrated NAC model can now be proposed (Fig. 2) that incorporates these views and indicates the steps where α-amanitin could interfere with RNA synthesis. In summary, the work of the Cramer and Kornberg laboratories reveals previously unobserved conformations and states of RNAP trapped by the transcriptional inhibitor α-amanitin. They provide a new understanding for the role of the catalytic center mobile elements in the NAC and pave the way for future investigations. ACKNOWLEDGMENTS This work was supported by grants from the US National Institutes of Health. 1. Borukhov, S. & Nudler, E. Trends Microbiol. 16, 126–134 (2008). 2. Kaplan, C.D., Larsson, K.M. & Kornberg, R.D. Mol. Cell 30, 547–556 (2008). 3. Brueckner, F. & Cramer, P. Nat. Struct. Mol. Biol. 15, 811–818 (2008). 4. Bushnell, D.A., Cramer, P. & Kornberg, R.D. Proc. Natl. Acad. Sci. USA 99, 1218–1222 (2002). 5. Bar-Nahum, G. et al. Cell 120, 183–193 (2005). 6. Kireeva, M.L. et al. Mol. Cell 30, 557–566 (2008). 7. Toulokhonov, I., Zhang, J., Palangat, M. & Landick, R. Mol. Cell 27, 406–419 (2007). 8. Wang, D. et al. Cell 127, 941–954 (2007). 9. Vassylyev, D.G. et al. Nature 448, 163–168 (2007). 10. Zhu, R. et al. Theor. Chim. Acta 120, 479–489 (2008). 11. Bai, L., Fulbright, R.M. & Wang, M.D. Phys. Rev. Lett. 98, 068103 (2007). 12. Vassylyev, D.G. et al. Nature 417, 712–719 (2002). 13. Campbell, E.A. et al. EMBO J. 24, 674–682 (2005). 14. Zhu, R. & Salahub, D.R. AIP Conf. Proc. 963, 104–110 (2007).
A splicing regulator promotes transcriptional elongation Juan Pablo Fededa & Alberto R Kornblihtt A new study reveals that the serine/arginine-rich splicing factor SC35 is necessary to promote RNA polymerase II elongation in a subset of genes, confirming a bidirectional coupling between transcription and splicing. Serine/arginine-rich (SR) proteins are a conserved family of proteins primarily known for their numerous roles in pre-mRNA splicing. Juan Pablo Fededa and Alberto R. Kornblihtt are at the Laboratorio de Fisiología y Biología Molecular, IFIByNE-CONICET, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Buenos Aires, Argentina. e-mail:
[email protected]
Both constitutive and alternative splicing require SR proteins, not only for the recruitment of general splicing factors involved in the assembly of functional spliceosomes, but also for bridging factors sitting at 3′ and 5′ splice sites that define alternative exons1. SR proteins are functionally modular: a C-terminal domain rich in arginines and serines (RS domain) mediates interactions with other proteins, whereas one or two N-terminal RNA recognition motifs (RRMs)
nature structural & molecular biology volume 15 number 8 AUGUST 2008
bind to specific sequences in the target premRNAs, known as exonic (ESE) or intronic (ISE) splicing enhancers depending on their location. Some SR proteins are concentrated in nuclear subcompartments called speckles, from where they migrate to transcription sites upon transcriptional activation2. The SR protein SC35 was discovered and characterized for its important role in constitutive and alternative splicing, and its
779
© 2008 Nature Publishing Group http://www.nature.com/nsmb
news and views absence causes embryonic lethality. Nevertheless, in previous work, the Fu laboratory had shown that targeted deletion of SC35 in heart cells causes downregulation of several genes, with no conspicuous changes in alternative splicing3. A new paper from the Fu laboratory on page 819 of this issue now investigates a putative function for SC35 in transcription and reveals that SC35 has an important role in promoting genespecific RNA polymerase II (Pol II) elongation4. Depletion of SC35 provoked a dramatic decrease in nascent RNA made by Pol II with no effect on transcription by RNA polymerase I. Using chromatin immunoprecipitation combined with microarrays (ChIP-chip), the authors observed several genes in which Pol II accumulated within the gene body upon SC35 depletion, suggestive of Pol II stalling before reaching the end of the gene. This stalling results in a decrease in Pol II elongation, confirmed by measuring the nascent transcripts with an ingenious modification of the classical run-on assay that uses nonradioactive nucleotides. Furthermore, co-immunoprecipitation studies showed that SC35 is not only able to interact with Pol II but also with CDK9, the kinase component of the transcriptional elongation factor P-TEFb. The Pol II and SC35 partnership is not surprising, as proteomic analysis performed by the Reed laboratory has recently shown that all SR proteins are present in Pol II complexes5. On the contrary, the interaction with P-TEFb has not been seen previously and supports the idea that the observed effects of SC35 are mediated through Pol II elongation. P-TEFb is responsible for the phosphorylation of the C-terminal domain (CTD) heptad repeats of Pol II at Ser2, an event that is thought to follow Ser5 phosphorylation, necessary for transcriptional initiation. In contrast, phosphorylated Ser2 is associated with Pol II elongation and is found predominantly within the body of actively transcribed genes. The authors show that, in response to SC35 depletion, both P-TEFb recruitment to the Pol II elongating complex and Ser2 phosphorylation are markedly reduced, whereas Ser5 phosphorylation is unaffected. As expected, many of the observed effects caused by SC35 depletion are reverted when this SR protein is restituted. In summary, the paper by Lin et al. uncovers a new and important role in transcription for a protein previously thought to be involved primarily in splicing. The above findings provide further support for the now well-established concept of coupling of Pol II transcription and RNA processing. Until the mid-1990s, transcription and pre-mRNA processing were thought to be independent steps in the pathway of eukaryotic gene expression. However, a series of biochemical, cytological and functional
780
Weak 3′ splice site
Strong 3′ splice site Gene
Pol II
P-TEFb
Ser5
Constitutive exon
P Alternative exon SC35
Pol II
Pol II Cap
Ser2 P
Gene
Spliceosome
Ser2 Pre-mRNA
P
(A)n mRNA
Cap Skipping
Figure 1 An example of bidirectional coupling: a splicing factor regulates transcription, which in turn regulates alternative splicing. The splicing factor SC35 interacts with RNA polymerase II (Pol II) and the elongation factor P-TEFb and, via phophorylation of the C-terminal domain (CTD) of Pol II at Ser2, stimulates transcriptional elongation. In parallel, high elongation rates allow the simultaneous presentation to the splicing machinery of strong and suboptimal 3′ splice sites, which favors the use of the stronger one, leading to skipping of an alternative exon.
e xperiments demonstrated that all three processing reactions (capping, splicing, and cleavage and polyadenylation) are coupled to Pol II transcription6. One of the prerequisites for coupling between transcription and processing is that the latter takes place co-transcriptionally, but co-transcriptional processing does not necessarily imply coupling. Two mechanisms (not mutually exclusive) for this kind of coupling have been extensively investigated. Recruitment coupling occurs when premRNA processing factors can interact directly or indirectly with Pol II or other components of the transcription machinery. For example, the Pol II CTD can bind capping enzymes and cleavage and polyadenylation factors that help these two processes. The CTD is also necessary for proper release of the processed mRNA from transcription sites. As mentioned, splicing factors such as SR proteins were found associated with Pol II, which may explain why their activity is more efficient on co-transcriptional compared to post-transcriptional splicing. Recruitment coupling is also evidenced by the existence of proteins with dual activities, such as PGC-1, which is a transcription coactivator and an alternative-splicing regulator whose splicing activity is exerted only when it is tethered to promoters through binding to a sequence-specific transcription factor, as seen for coactivators. The second mode, known as kinetic coupling, has been demonstrated mostly for splicing and refers to how changes in the rate
of transcriptional elongation affect the timing with which splice sites are presented to the splicing machinery7. For instance, reduction in Pol II elongation by either changing the enzyme’s intrinsic properties or by promoting more compact chromatin structures that create roadblocks for its passage8 tend to favor the use of weak splice sites that are presented to the splicing apparatus before stronger splice sites emerge downstream. Most of the available evidence highlights the influence of transcription on pre-mRNA processing. Much less is known about how processing factors influence Pol II transcription. It was reported that components of the splicing machinery can regulate transcription: the spliceosomal U small nuclear ribonucleoproteins (U snRNPs) stimulate Pol II elongation via their interaction with the elongation factor TAT-SF1, probably through P-TEFb9. The U snRNP–TATSF1 complex can also stimulate splicing in vitro, suggesting that a general principle of reciprocal ‘give and take’ governs both transcription and splicing. The glue that allows this cooperative phenomenon involves the many interactions between proteins of both machineries: Das et al. have found not only SR proteins but also a plethora of splicing factors (which curiously include U1 snRNP components and heterogeneous nuclear ribonucleoprotein (hnRNP) proteins) associated with Pol II5. Figure 1 shows how transcription and splicing can influence each other.
volume 15 number 8 AUGUST 2008 nature structural & molecular biology
© 2008 Nature Publishing Group http://www.nature.com/nsmb
news and views The paper by Lin et al. prompts several i nteresting questions. Why is the elongation of certain genes but not others regulated by SC35? Are there specificity factors? Is the binding of SC35 to the nascent mRNAs required for the elongation effect? Is there a correlation between the group of genes whose transcriptional elongation is regulated by SC35 and those bearing SC35 ESEs? If binding to such specific mRNA sequences were needed for the recruitment of SC35 to Pol II, the levels of a given transcript in a certain cellular context would result not only from the quality and amount of transcription factors that regulate the promoter, but also from the levels of SC35 and the number of functional ESEs over the entire sequence of a gene, adding another layer of complexity to the regulation of gene expression. What is clear from the present work is that the effects of SC35 are seen far beyond the promoter through increases in CTD-Ser2 phosphorylation over the bodies of the responsive genes. The report also raises evolutionary questions. Were SR proteins primarily selected for their role in alternative splicing?
Accumulated evidence suggests that this might not have been the case. Wang et al. showed that, in mammalian cells, target sites for SR proteins (ESEs) are more frequent in constitutive than in alternative exons10. Consistently, the Wise laboratory showed that SR proteins are already involved in enhancing constitutive splicing in organisms such as Schizosaccharomyces pombe where alternative splicing does not exist11. SR proteins were also found in Trypanosoma cruzi12, an organism with no alternative splicing and where trans-splicing is highly preponderant. Combining these findings with those reported by Lin et al., one could even speculate that some SR proteins such as SC35, although undoubtedly linked to splicing, might have been selected for their role in transcription of intron-containing genes. A provocative finding regarding this speculation is that P-TEFb, the elongation factor shown to be influenced by SC35, seems to be dispensable for transcription of intronless genes such as those for human U2 snRNA and histone H2B13. Another suggestive observation in the same direction is that, unlike other members
of the family, SC35 cannot shuttle between the nucleus and the cytoplasm14. This correlates with a profound hallmark in the divergence of the biological function of these proteins: shuttling SR proteins can also regulate export and translation15 of target mRNAs, whereas SC35 confinement to the nucleus may well turn out to reflect its role in transcription. Graveley, B.R. RNA 6, 1197–1211 (2000). Misteli, T. & Spector, D.L. Mol. Cell 3, 697–705 (1999). Ding, J.H. et al. EMBO J. 23, 885–896 (2004). Lin, S., Coutinho-Mansfield, G., Wang, D., Shatakshi, P. & Fu, X.-D. Nat. Struct. Mol. Biol. 15, 819–826 (2008). 5. Das, R. et al. Mol. Cell 26, 867–881 (2007). 6. Bentley, D.L. Curr. Opin. Cell Biol. 17, 251–256 (2005). 7. de la Mata, M. et al. Mol. Cell 12, 525–532 (2003). 8. Batsche, E., Yaniv, M. & Muchardt, C. Nat. Struct. Mol. Biol. 13, 22–29 (2006). 9. Fong, Y.W. & Zhou, Q. Nature 414, 929–933 (2001). 10. Wang, J., Smith, P.J., Krainer, A.R. & Zhang, M.Q. Nucleic Acids Res. 33, 5053–5062 (2005). 11. Webb, C.J., Romfo, C.M., van Heeckeren, W.J. & Wise, J.A. Genes Dev. 19, 242–254 (2005). 12. Portal, D. et al. Mol. Biochem. Parasitol. 127, 37–46 (2003). 13. Medlin, J. et al. EMBO J. 24, 4154–4165 (2005). 14. Caceres, J.F., Screaton, G.R. & Krainer, A.R. Genes Dev. 12, 55–66 (1998). 15. Sanford, J.R., Gray, N.K., Beckmann, K. & Caceres, J.F. Genes Dev. 18, 755–768 (2004). 1. 2. 3. 4.
The chloride channel’s appendix Joseph A Mindell All CLC proteins transport Cl– across membranes. However, the family includes both Cl– channels and Cl–/H+ antiporters, proteins once thought to operate by dramatically different mechanisms. An apparent evolutionary relic, a proton-transport apparatus in a CLC channel, reveals deep intertwinings between channel and transporter mechanisms. The human appendix seems to do nothing but cause misery. This organ, which has been suggested to exist solely to generate income for surgeons, is an evolutionary vestige from a time when our ancestors chewed their cud and needed to digest grass. Vestiges such as the appendix and the foot bones of the whale don’t do much now, but they highlight evolutionary origins and the deep connections between living things. But such remnants are not limited to large anatomical structures; molecular vestiges, too, can underscore the evolutionary relationships between biological
Joseph A. Mindell is in the Membrane Transport Biophysics Unit, Porter Neuroscience Research Center, National Institute of Neurological Disorders and Stroke, National Institutes of Health, 35 Convent Drive, Building 35, MSC 3701, Bethesda, Maryland 20892. e-mail:
[email protected]
acromolecules. Their existence reminds m us that not every conserved bit of protein sequence or structure must have a compelling raison d’être in the protein’s current existence; some behaviors and structures may instead be remembrances of functions past. In this issue, Lísal and Maduke1 report on an apparent evolutionary vestige in a CLC chloride channel that enlightens our understanding of the channel’s transporter ancestry. Our story begins with the discovery of a voltage-gated Cl– channel (later named ClC-0) in the electric organ of the ray Torpedo californica2. From early in its history, this channel has been an oddity, even beyond its anionic selectivity. Of particular interest here is the curious finding that, in ClC-0, gating (the process of opening and closing the channel) is intimately coupled to permeation (the process of conducting ions)3,4; this contrasts with cation channels, which (to a first approximation) can be separated into
nature structural & molecular biology volume 15 number 8 AUGUST 2008
f unctionally and physically distinct gating and permeation domains. First hints of this connection came from an observation that ClC-0 gating is a nonequilibrium process that must therefore harness energy from transmembrane ion or voltage gradients5. The links between permeation and gating deepened as a series of reports showed evidence that ClC-0’s voltage dependence results not from movements of charged amino acids on the channel protein (as in cation channels), but from transmembrane movement of the Cl– ion itself3,4. Various experiments demonstrated that the activation process parallels the permeation process; for example, both have a similar dependence on the nature of the permeant anion6. Observations that gating properties of ClC-0 depend strongly on the Cl– concentration support the notion that Cl– does more than just go through the channel. Perhaps the most striking observation about ClC-0 is the recent discovery that
781
