Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

X‐ray structure and activity of the yeast Pop2 protein: a nuclease subunit of the mRNA deadenylase complex

X‐ray structure and activity of the yeast Pop2 protein: a nuclease subunit of the mRNA... scientific report scientificreport X-ray structure and activity of the yeast Pop2 protein: a nuclease subunit of the mRNA deadenylase complex 1 2 2 1+ Ste´phane Thore , Fabienne Mauxion , Bertrand Se´raphin & Dietrich Suck 1 2 European Molecular Biology Laboratory, Heidelberg, Germany, and Equipe Labelise´e La Ligue, Centre de Ge´ne´tique Mole´culaire, Gif sur Yvette Cedex, France In Saccharomyces cerevisiae, a large complex, known as the et al., 1995) and five Not proteins (Not1–5) required for the proper Ccr4–Not complex, containing two nucleases, is responsible for function of the Ccr4–Not complex (Draper et al., 1994; Liu et al., mRNA deadenylation. One of these nucleases is called Pop2 and 1998). Interactions between Ccr4, Pop2 and the Not proteins, has been identified by similarity with PARN, a human poly(A) together with other proteins present in the Ccr4–Not complex, nuclease. Here, we present the crystal structure of the nuclease have been characterized (Benson et al., 1998; Bai et al., 1999; domain of Pop2 at 2.3 A resolution. The domain has the fold of Badarinarayana et al., 2000; Lemaire & Collart, 2000; Liu et al., the DnaQ family and represents the first structure of an RNase 1997, 2001). from the DEDD superfamily. Despite the presence of two non- Ccr4 and Pop2 both contain a nuclease domain and are canonical residues in the active site, the domain displays RNase therefore potentially responsible for the deadenylation activity. The 2þ activity on a broad range of RNA substrates. Site-directed Ccr4 protein has been identified as a member of the Mg - mutagenesis of active-site residues demonstrates the intrinsic dependent endonuclease-related protein family (Dlakic, 2000), and ability of the Pop2 RNase D domain to digest RNA. This first its activity has been characterized both in vivo (Daugeron 0 0 structure of a nuclease involved in the 3 –5 deadenylation et al., 2001; Chen et al., 2002; Tucker et al., 2001, 2002) and of mRNA in yeast provides information for the understanding of in vitro (Viswanathan et al., 2003). Disruption of the gene and the mechanism by which the Ccr4–Not complex achieves its various mutants of the protein active site show a strong decrease in functions. mRNA deadenylation rates, indicating that the protein is an EMBO reports 4, 1150–1155 (2003) essential factor in this process (Daugeron et al., 2001; Tucker doi:10.1038/sj.embor.7400020 et al., 2001; Chen et al., 2002). The second protein, Pop2, is related to the ribonuclease (RNase) D family (Moser et al., 1997; Daugeron INTRODUCTION et al., 2001). While Pop2 has two non-conserved catalytic residues, mRNA degradation is an important step of gene expression, the other Pop2/Caf1 family members display a conserved and thus allowing to counter-balance or enhance the effect of transcription potentially fully functional active site. In addition, Pop2 gene and thus contributing to overall gene regulation. In Saccharo- disruption affects the rate and the degree of deadenylation of myces cerevisiae, the deadenylation process (reviewed in reporter mRNAs (Daugeron et al., 2001; Tucker et al., 2001). Caponigro & Parker, 1996) is the first step of mRNA degradation. The RNase D family (Mian, 1997) belongs to the DEDD It is achieved by the Ccr4–Not complex (Daugeron et al., 2001; superfamily composed of RNases as well as deoxyribonucleases Tucker et al., 2001), a large multi-protein complex (Liu et al., (DNases; Zuo & Deutscher, 2001). It is characterized by sequence 0 0 1997; Chen et al., 2001), built around a core of seven proteins: the motifs similar to the Exo I, II and III motifs of the 3 –5 carbon catabolite repressor 4 factor (Ccr4), first identified as a exodeoxyribonuclease (or proofreading) domains of DNA poly- gene expression regulating protein (Denis, 1984), the Pop2 merases, which contain four conserved acidic residues, namely protein, also known as Caf1 (for Ccr4 associated factor 1; Draper DEDD. These amino acids are responsible for binding the two metal ions involved in catalysis (reviewed in Joyce & Steitz, 1994). European Molecular Biology Laboratory, Meyerhofstrasse 1, D-69117 Heidelberg, Three-dimensional (3D) structures have been determined for the Germany DnaQ family (Ollis et al., 1985; Hamdan et al., 2002), which is a Equipe Labelise´e La Ligue, Centre de Ge´ne´tique Mole´culaire, CNRS UPR2167, subgroup of the DEDD superfamily. Avenue de la Terrasse, 91198 Gif sur Yvette Cedex, France Corresponding author. Tel: þ 49 6221 387 307; Fax: þ 49 6221 387 306; Here, we report the crystal structure of the RNase D domain of E-mail: suck@embl-heidelberg.de the yeast Pop2 protein. It represents the first structure of an RNase D protein as well as of an RNase from the DEDD superfamily. Received 22 August 2003; revised 17 September 2003; accepted 19 September 2003 Despite the non-conservation of two catalytic residues from the Published online 14 November 2003 1150 EMBO reports VOL 4 | NO 12 | 2003 &2003 EUROPEAN MOLECULAR BIOLOGY ORGANIZATION Crystal structure of the nuclease domain of Pop2 S. Thore et al. scientificreport Table 1 Data collection, phasing and refinement statistics Se-Met data sets Native Peak Inflection Remote Data collection and phasing statistics l (A) 1.8461 0.9795 0.9797 0.9689 Space group P2 2 2 P4 22P4 22P4 2 2 1 1 1 1 1 1 1 1 1 Resolution (A) 2.3 3.15 3.20 3.25 (2.38–2.3) (3.26–3.15) (3.31–3.2) (3.37–3.25) Reflection (unique) 52,320 10,414 10,007 9,508 Completeness (%) 95.9 (76.6) 99.2 (100) 99.5 (100) 99.6 (100) Redundancy 6.7 4 4 4 Average I/s(I) 11.9 (2.9) 11.9 (4.6) 11.7 (4.4) 11.4 (3.9) R 4.0 (30.9) 7.1 (30.5) 7.6 (32.8) 7.6 (34.9) sym FOM — — — 0.49 MAD FOM after solvent flattening — — — 0.65 Refinement statistics No. of reflections used 51,486 — — — R (%) 23.8 — — — value R (%) 26.0 — — — free Residues (out of 289) 263 (mol. 1)/254 (mol. 2) — — — Number of protein/solvent/ion atoms 4,214/46/4 — — — ˚ 2 Average B factor (A)51 — — — r.m.s.d. Dbond lengths (A) 0.008 — — — r.m.s.d. Dbond angles (degree) 1.4 — — — Values for the outermost resolution shell are shown in parentheses. FOM, figure of merit; R : R factor for 5% of the reflections excluded from the refinement. free P P R , |I/IS|/ I. sym hkl hkl DEDD signature sequence, we observe in vitro RNase activity towards poly(A), and also poly(C) and poly(U). Site-directed mutagenesis abolishes RNase activity, indicating that it is associated specifically with the Pop2 RNase D domain. Canonical active site residues are fully conserved within the Pop2/Caf1 family, strongly suggesting that RNase activity will be associated with all the family members. RESULTS AND DISCUSSION Structure determination and overall fold The fragment of the yeast Pop2 protein corresponding to the RNase D domain was crystallized and its structure was determined using MAD data from a single Se-Met-substituted protein containing crystal. Native crystals belong to space group P2 2 2 with two molecules in the asymmetric unit. The final 1 1 1 model contains two calcium ions, two xenon atoms and 263 or 254 residues (for each respective molecule) representing all amino acids where the main chain was clearly defined (Fig. 1A). Three regions were poorly resolved and did not allow tracing (residues 1–3, 50–59 and 213–225). The two molecules constituting the asymmetric unit adopt very similar conformations with a root mean square deviation (r.m.s.d.) of 0.47 A. The overall structure contains 13 a-helices and six b-strands, forming a kidney-shaped structure (Fig. 1A). The b-strands form the central core flanked by the a-helices. Helices 2, 3, 6, 7, 8 and 13 form the near-side of the molecule, while the remaining helices build the two lobes creating the cavity of the kidney-shaped molecule (Fig. 1B). Fig. 1 Structure of the nuclease domain of the Pop2 protein. (A)Ribbonplot Structural conservation within the DEDD nuclease family representation with the secondary elements in the following colour code: During recent years, several structures of proofreading enzymes a-helix, red; b-strands,green;and loops, yellow.(B) Crossed-eye stereo involved in DNA replication have been solved, revealing a representation of the Ca trace is displayed with every 20th residue marked. &2003 EUROPEAN MOLECULAR BIOLOGY ORGANIZATION EMBO reports VOL 4 | NO 12 | 2003 1151 Crystal structure of the nuclease domain of Pop2 S. Thore et al. scientificreport Fig. 2 Structural homology of Pop2 with members of the DEDD nuclease superfamily. (A) Structure-based sequence alignment of Pop2, the exonuclease domain of PolI and the e-subunit of PolIII. Sequence conservation is shown by colour coding: invariant residues are highlighted in red. Yellow highlights residues that have similar properties. Secondary structure elements of Pop2 are shown above the sequences. Small arrowheads indicate the conserved DEDD residues forming the catalytic site of the e-subunit. (B) The three structures shown in the same relative orientation. (C) The electrostatic surface potentials of Pop2 and the e-subunit indicate the location of the active site; catalytic residues are highlighted. (D) Close-up view of the active site of the e-subunit (salmon colour; with bound TMP in yellow) superimposed with the Pop2 (light green) structure and (E) side view of the secondary structure elements interacting with the bound nucleotide. Bold and italic labels correspond to the amino acids from Pop2 or the e-subunit, respectively. common fold for the DnaQ subgroup of the DEDD superfamily A global view of the electrostatic surface potential of Pop2 (Zuo & Deutscher, 2001). While the sequence similarity between indicates a highly negatively charged cavity (Fig. 2C). A similar Pop2 and the e-subunit of DNA polymerase III (PolIIIe) or the feature can be observed in exonuclease I (Breyer & Matthews, exonuclease domain of polymerase I (PolI) is only 15 or 20% 2000). These negatively charged regions in both proteins coincide (Fig. 2A), the overall superposition of the structures gives a r.m.s.d. with the active site in PolIII (Fig. 2C,D). A close look at the active of only 1.5 or 1.7 A of the Ca trace over 134 and 128 residues, site (Fig. 2D) shows that the Ser 44 and Gln 250 substitutions in the respectively (Fig. 2B; Ollis et al., 1985; Hamdan et al., 2002). The Pop2 amino-acid sequence could provide an oxygen atom able to Pop2 structure provides the first crystallographic evidence that the interact with the active site magnesium. In line with this DNase fold is also adopted by the RNase D, subgroup of the hypothesis, it is interesting to note that the two corresponding DEDD superfamily (Fig. 2B). residues are phylogenetically conserved in yeast species that 1152 EMBO reports VOL 4 | NO 12 | 2003 &2003 EUROPEAN MOLECULAR BIOLOGY ORGANIZATION Crystal structure of the nuclease domain of Pop2 S. Thore et al. scientificreport Fig. 3 RNase activity of Pop2 and conservation within the Pop2/Caf1 family. (A) Activity tests showing the RNase activity of Pop2 WT with poly(A), poly(U), poly(C) and G RNA substrates. (B) The double-mutant S44A/E46A (Pop ATA) does not degrade poly(A) RNA. (C) The time course of the RNase reaction demonstrates that the Pop2 nuclease is distributive. (D) Surface representation of the sequence conservation within the Pop2/Caf1 family is shown and is based on the sequence alignment shown in (E). Green colour indicates conserved residues, and white for non-conserved. Numbering is according to the Pop2 sequence and the residue name is according to the conserved amino acid in the Pop2/Caf1 family. (E) Sequence alignment of the eukaryotic Pop2/ Caf1 family with Mus musculus CAF1 (accession number NP_035265) and CALIF1 (NP_081225), Homo sapiens CAF1 (L46722) and CALIF1 (Q9UFF9), Xenopus laevis CAF1 (AAH41239), Drosophila melanogaster CAF1 (AE003543), Anopheles gambiae CAF1 (EAA12934), Arabidopsis thaliana CAF1 (AY070420), Plasmodium yoelii yoelii CAF1 (EAA20457), Caenorhabditis elegans CAF1 (U21854), Neurospora crassa CAF1 (EAA29793), Schizosaccharomyces pombe CAF1 (NP_588385), Encephalitozoon cuniculi CAF1 (NP_597215) and the RNase D domain of Saccharomyces cerevisiae Pop2 (P39008), colour coded as in Fig. 2A. DEDD catalytic site residues are indicated with arrowheads. diverged 5–20 million years ago, thus supporting the importance RNase activity of the Pop2 domain of this combination for yeast Pop2 function. Given the The literature contains contradictory data about the RNase activity number of negatively charged residues at the active site, it is also associated with the Pop2 protein from yeast. Under our in vitro possible that Pop2 binds only one metal ion instead of two conditions, Pop2 RNase activity is observed with poly(A), poly(U) normally observed for the DNases, resulting in a different reaction and poly(C), but not with oligo(G) (Fig. 3A, lanes 2–3, 5–6, 8–9 mechanism. and 11). However, more sensitive competition assays demonstrate Interestingly, the a-helices 4 and 5 (Fig. 2E) as well as residues a subtle preference for poly(A) (Daugeron et al., 2001). When the from the loop between strands b2 and b3 contact the leaving first two residues from the DEDD motif are mutated, that is, S44A nucleotide in the PolIII structure. These amino acids are identical, and E46A, the activity is lost (Fig. 3B, compare lanes 13–14 and or at least similar, in the case of Pop2, suggesting an interaction 16–17). The progressive appearance of shorter RNA fragments with the 3 -terminal nucleotide during the nuclease reaction indicates the distributive activity of the Pop2 RNase D domain despite the somewhat different orientation observed in the crystal (Fig. 3C). The ability to degrade several types of RNA implies (Fig. 2E). that the substrate RNA is not recognized in an absolute, &2003 EUROPEAN MOLECULAR BIOLOGY ORGANIZATION EMBO reports VOL 4 | NO 12 | 2003 1153 Crystal structure of the nuclease domain of Pop2 S. Thore et al. scientificreport specific manner but may involve a substantial contribution important information for the understanding of the deadenylase of van der Waals interactions, in agreement with the presence of complex in yeast and other eukaryotic species. This information a conserved hydrophobic patch following the b-strand 2 (Fig. 3E). could be used in order to better understand how the dead- The structure of the Pop2 nuclease domain represents the first enylation process is coupled to translation as well as to decapping 3D model of a catalytic component of the yeast deadenylase processes before the degradation occurs. complex. Pop2/Caf1 family METHODS Since Pop2 is the least conserved protein within the Pop2/Caf1 Expression and crystallization. The yeast Pop2 RNase D domain family (Fig. 3E; Daugeron et al., 2001), the fold conservation (amino acids (aa) 147–433) was expressed from a pET24d between the DNases and Pop2 suggests that the nuclease domain derivative (Novagen) with an amino (N)-terminal 6xHis tag-TEV of the Pop2/Caf1 family members will have the same fold as the protease site extension. After overexpression in Escherichia coli Pop2 protein. According to the sequence alignment of the Pop2/ BL21-CodonPlus (DE3)-RIL, cells were lysed with a French press Caf1 family, the other members do have a canonical active site, in buffer A (50 mM NaH PO (pH 7.5), 200 mM NaCl, 10 mM b- 2 4 that is, the DEDD signature sequence (Fig. 3E, arrowheads). mercaptoethanol) in the presence of protease inhibitor cocktail Therefore, these proteins are expected to display at least the same (Promega) and 10 mgml of RNase A (Sigma). After binding to Ni- or even higher RNase activity than Pop2. This has been verified NTA beads (Quiagen), the protein was eluted as indicaded. Pop2- in the case of human CAF1 (F. Mauxion and B. Se´raphin, containing fractions were passed onto a Superdex 75 column unpublished data). In addition, the conserved amino acids within (Amersham-Pharmacia) in buffer A. The tag was then removed by the Pop2/Caf1 family line the entire cavity, indicating a TEV protease cleavage at 16 1C (enzyme:substrate ratio 1:100 functionally important surface likely to correspond to the catalytic w/w) followed by chromatography on Ni-NTA beads. A second and the RNA interacting sites (Fig. 3D). gel filtration step in 20 mM Tris/Cl (pH 7.5) and 200 mM NaCl led to a 495% pure sample, which was concentrated to 9 mg ml .A Deadenylation and mRNA degradation Se-Met derivative was prepared similarly using the B834 E. coli The role of Pop2 in mRNA deadenylation has remained unclear, host strain grown in minimal medium containing L-seleno- while it is essential for deadenylation in vivo (Daugeron et al., methionine (50 mgml ). 2001; Tucker et al., 2001) and was reported to have RNase activity Crystals of the Pop2 nuclease domain were grown by vapour in vitro (Daugeron et al., 2001), analysis of point mutants failed to diffusion (hanging drops) at 4 1C by mixing an equal volume of reveal the requirement for a specific step in mRNA decay (Chen et protein and reservoir solution containing 2.5% PEG 3350 MW, al., 2002; Tucker et al., 2002). The crystal structure of the nuclease 100 mM Hepes (pH 7.0), 80 mM calcium acetate and 16.5% domain of Pop2 in conjunction with mutagenesis and in vitro glycerol. Drops were immediately micro-seeded and, after 7–8 assays demonstrate definitively that yeast Pop2 contains a days, transferred over a well containing a similar solution and nuclease domain displaying RNase activity of broad specificity. 27.5% glycerol. Crystals were then directly flash-frozen in liquid The requirement for two nucleases in the Ccr4–Not complex to nitrogen. Crystals from the Se-Met derivative were grown using the mediate mRNA deadenylation remains puzzling. Published data following conditions: 2% PEG 3350 MW, 120 mM calcium suggest that Ccr4 provides most of the nuclease activity in yeast acetate and protein at 5 mg ml . Xenon derivatization was (Chen et al., 2002; Tucker et al., 2002). However, as yeast Pop2 carried out on native crystals using a pressurized cell at 10 bar has an RNase activity in vitro and a non-canonical active site, this for 10 min (Djinovic-Carugo et al., 1998). situation may differ in other organisms. Indeed, the presence of A complete 2.3 A data set was collected under cryogenic two variations in the active site of Pop2 may strongly influence its conditions using the ESRF ID29 beamline. A three-wavelength efficiency compared to human CAF1 for example. It would thus be MAD experiment was performed on a single Se-Met-substituted of interest to evaluate the relative contribution of human CCR4 protein crystal at the PX06 beamline at SLS. Data were processed and human CAF1 to exonucleolytic poly(A) degradation. Simi- using the HKL package (Otwinowski & Minor, 1997). The native larly, it would be of interest to test whether the N-terminal domain and the Se-Met-substituted protein crystallized in space groups ˚ ˚ ˚ of full-length Pop2 inhibits RNase activity in vivo and whether this P2 2 2 (a ¼ 78.5 A, b ¼ 79.4 A, c ¼ 101.3 A, a ¼ 901, b ¼ 901, 1 1 1 is affected by the phosphorylation event known to occur in this g ¼ 901) with two molecules per asymmetric unit, and P4 2 2 1 1 ˚ ˚ ˚ region (Moriya et al., 2001). (a ¼ 78.2 A, b ¼ 78.2 A, c ¼ 102.6 A, a ¼ 901, b ¼ 901, g ¼ 901) Interaction of Ccr4 and Pop2 with the Not factors may also play with one molecule per asymmetric unit, respectively. an important role in regulating Pop2 activity and subsequently the Structure determination and refinement. Phases were obtained at deadenylation process. In the case of the human poly(A) nuclease 3.3 A with SOLVE (Terwilliger & Berendzen, 1999) and the PARN, reported to be similar to Pop2 as well as a member of the electron density map was improved by solvent flattening using the DEDD family, the protein contains a large domain belonging to program RESOLVE (Terwilliger, 2000). Only 40% of the protein the R3H family (Daugeron et al., 2001), which is able to bind could be traced with program O, version 7 (Jones et al., 1991). single-stranded RNA on its own. Since the Pop2/Caf1 family does DMMULTI from the CCP4 suite (CCP4, 1994) was used to extend not contain any additional RNA-binding domain, it is tempting to the phases to 2.3 A in P2 2 2 . Then, the density modification 1 1 1 speculate that Pop2 interactions with the Ccr4–Not complex may procedure implemented in CNS (Brunger et al., 1998) and based target it to the poly(A) tail. on the phases calculated from the partial model provided Further structural studies on the various complexes formed additional electron density. After three cycles of model building, between Ccr4, the Not proteins, mRNA and Pop2 are likely to give annealing, individual B-factor refinement and density modification, 1154 EMBO reports VOL 4 | NO 12 | 2003 &2003 EUROPEAN MOLECULAR BIOLOGY ORGANIZATION Crystal structure of the nuclease domain of Pop2 S. Thore et al. scientificreport Dlakic, M. (2000) Functionally unrelated signalling proteins contain a fold 263 and 254 residues for each molecule in the asymmetric unit 2+ similar to Mg -dependent endonucleases. Trends Biochem. Sci., 25, could be built (residues 4–49, 60–213 and 226–289 for molecule 272–273. 1 and residues 4–49, 62–207 and 228–289 for molecule 2). The Draper, M.P., Liu, H.Y., Nelsbach, A.H., Mosley, S.P. & Denis, C.L. (1994) final R is 26.3%, and the model shows good stereochemistry as CCR4 is a glucose-regulated transcription factor whose leucine-rich free repeat binds several proteins important for placing CCR4 in its proper indicated by PROCHECK (Laskowski et al., 1993). Table 1 shows promoter context. Mol. Cell. Biol., 14, 4522–4531. data collection and refinement statistics. Draper, M.P., Salvadore, C. & Denis, C.L. (1995) Identification of a mouse The coordinates have been deposited in the Protein Data Bank protein whose homolog in Saccharomyces cerevisiae is a component under accession code 1UOC. of the CCR4 transcriptional regulatory complex. Mol. Cell. Biol., 15, RNase assays. Polynucleotides (Amersham-Pharmacia) were 3487–3495. Hamdan, S., Carr, P.D., Brown, S.E., Ollis, D.L. & Dixon, N.E. (2002) stored as recommended. In vitro RNase assays were performed Structural basis for proofreading during replication of the Escherichia coli in 20 mM Tris/Cl (pH 7.0), 150 mM NaCl, 2 mM MgCl ,5U chromosome. Structure (Camb.), 10, 535–546. RNasin (Promega), 3 mM RNA substrate and indicated amounts of Jones, T.A., Zou, J.Y., Cowan, S.W. & Kjeldgaard, M. (1991) Improved purified Pop2. A volume of 10 ml of the reaction mixture was methods for binding protein models in electron density maps and the location of errors in these models. Acta Crystallogr. A, 47, 110–119. incubated at 25 1C for 1 h unless otherwise stated. Reactions that Joyce, C.M. & Steitz, T.A. (1994) Function and structure relationships in DNA were stopped by the addition of formamide/EDTA buffer were polymerases. Annu. Rev. Biochem., 63, 777–822. loaded onto 8 M urea/18% acrylamide (19:1) gels that were Laskowski, R.A., MacArthur, M.W., Moss, D.S. & Thornton, J.M. (1993) stained with toluidine blue. PROCHECK: a program to check the stereochemical quality of protein structures. J. Appl. Crystallogr., 26, 283–291. ACKNOWLEDGEMENTS Lemaire, M. & Collart, M.A. (2000) The TATA-binding protein-associated factor yTafII19p functionally interacts with components of the global We thank the staff at the European Synchrotron Radiation Facility (ESRF), transcriptional regulator Ccr4–Not complex and physically interacts with Swiss Light Synchrotron (SLS) and Deutsches Elektronen Synchrotron the Not5 subunit. J. Biol. Chem., 275, 26925–26934. (DESY) for assistance during data collection. The MAD data sets were Liu, H.Y., Toyn, J.H., Chiang, Y.C., Draper, M.P., Johnston, L.H. & Denis, C.L. collected at SLS, Paul Schonen Institute, Switzerland. We also thank (1997) DBF2, a cell cycle-regulated protein kinase, is physically and C. Schulze-Briex for support and acknowledge J. Basquin and J. Fe´thie`re functionally associated with the CCR4 transcriptional regulatory for data collection of the MAD data sets, as well as A. Popov for support complex. EMBO J., 16, 5289–5298. at the X13 beamline at DESY/EMBL-Hamburg. We also thank E. Ennifar Liu, H.Y., Badarinarayana, V., Audino, D.C., Rappsilber, J., Mann, M. & for useful crystallographic discussions and C. Temme for help with Denis, C.L (1998) The NOT proteins are part of the CCR4 transcriptional mutagenesis. Work in B.S.’s lab is supported by La Ligue contre le complex and affect gene expression both positively and negatively. Cancer, CNRS and the Ministry for Research (PRFMMIP). EMBO J., 17, 1096–1106. Liu, H.Y., Badarinarayana, V., Audino, D.C., Rappsilber, J., Mann, M. & REFERENCES Denis, C. (2001) Characterization of CAF4 and CAF16 reveals a Badarinarayana, V., Chiang, Y.C. & Denis, C.L. (2000) Functional interaction functional connection between the CCR4–NOT complex and a subset of of CCR4–NOT proteins with TATAA-binding protein (TBP) and its SRB proteins of the RNA polymerase II holoenzyme. J. Biol. Chem., 276, associated factors in yeast. Genetics, 155, 1045–1054. 7541–7548. Bai, Y., Salvadore, C., Chiang, Y.C., Collart, M.A., Liu, H.Y. & Denis, C.L. Mian, I.S. (1997) Comparative sequence analysis of ribonucleases HII, III, II (1999) The CCR4 and CAF1 proteins of the CCR4–NOT complex are PH and D. Nucleic Acids Res., 25, 3187–3195. physically and functionally separated from NOT2, NOT4, and NOT5. Moriya, H., Shimizu-Yoshida, Y., Omori, A., Iwashita, S., Katoh, M. & Sakai, Mol. Cell. Biol., 19, 6642–6651. A. (2001) Yak1p, a DYRK family kinase, translocates to the nucleus and Benson, J.D., Benson, M., Howley, P.M. & Struhl, K. (1998) Association of phosphorylates yeast Pop2p in response to a glucose signal. Genes Dev., distinct yeast Not2 functional domains with components of Gcn5 histone 15, 1217–1228. acetylase and Ccr4 transcriptional regulatory complexes. EMBO J., 17, Moser, M.J., Holley, W.R., Chatterjee, A. & Mian, I.S. (1997) The 6714–6722. proofreading domain of Escherichia coli DNA polymerase I and other Breyer, W.A. & Matthews, B.W. (2000) Structure of Escherichia coli DNA and/or RNA exonuclease domains. Nucleic Acids Res., 25, exonuclease I suggests how processivity is achieved. Nature Struct. Biol., 5110–5118. 7, 1125–1128. Ollis, D.L., Brick, P., Hamlin, R., Xuong, N.G. & Steitz, T.A. (1985) Structure Brunger, A.T. et al. (1998) Crystallography & NMR system: a new software of large fragment of Escherichia coli DNA polymerase I complexed with suite for macromolecular structure determination. Acta Crystallogr. D, dTMP. Nature, 313, 762–766. 54, 905–921. Otwinowski, Z. & Minor, W. (1997) Processing of X-ray diffraction data Caponigro, G. & Parker, R. (1996) Mechanisms and control of mRNA turnover collected in oscillation mode. Methods Enzymol., 276, 307–326. in Saccharomyces cerevisiae. Microbiol. Rev., 60, 233–249. Terwilliger, T.C. (2000) Maximum-likelihood density modification. Acta CCP4, C.C.P. (1994) The CCP4 suite: programs for protein crystallography. Crystallogr. D, 56, 965–972. Acta Crystallogr. D, 50, 760–763. Terwilliger, T.C. & Berendzen, J. (1999) Automated MAD and MIR structure Chen, J., Rappsilber, J., Chiang, Y.C., Russell, P., Mann, M. & Denis, C.L. solution. Acta Crystallogr. D, 55, 849–861. (2001) Purification and characterization of the 1.0 MDa CCR4–NOT Tucker, M., Valencia-Sanchez, M.A., Staples, R.R., Chen, J., Denis, C.L. & complex identifies two novel components of the complex. J. Mol. Biol., Parker, R. (2001) The transcription factor associated Ccr4 and Caf1 314, 683–694. proteins are components of the major cytoplasmic mRNA deadenylase in 0 0 Chen, J., Chiang, Y.C. & Denis, C.L. (2002) CCR4, a 3 –5 poly(A) RNA and Saccharomyces cerevisiae. Cell, 104, 377–386. ssDNA exonuclease, is the catalytic component of the cytoplasmic Tucker, M., Staples, R.R., Valencia-Sanchez, M.A., Muhlrad, D. & Parker, R. deadenylase. EMBO J., 21, 1414–1426. (2002) Ccr4p is the catalytic subunit of a Ccr4p/Pop2p/Notp mRNA Daugeron, M.C., Mauxion, F. & Seraphin, B. (2001) The yeast POP2 gene deadenylase complex in Saccharomyces cerevisiae. EMBO J., 21, encodes a nuclease involved in mRNA deadenylation. Nucleic Acids 1427–1436. Res., 29, 2448–2455. Viswanathan, P., Chen, J., Chiang, Y.C. & Denis, C.L. (2003) Identification of Denis, C.L. (1984) Identification of new genes involved in the regulation of multiple RNA features that influence CCR4 deadenylation activity. J. Biol. yeast alcohol dehydrogenase II. Genetics, 108, 833–844. Chem., 278, 14949–14955. Djinovic-Carugo, K., Everitt, P. & Tucker, P.A. (1998) A cell for producing Zuo, Y. & Deutscher, M.P. (2001) Exoribonuclease superfamilies: structural xenon-derivative crystals for cryocrystallographic analysis. J. Appl. analysis and phylogenetic distribution. Nucleic Acids Res., 29, Crystallogr., 31, 812–814. 1017–1026. &2003 EUROPEAN MOLECULAR BIOLOGY ORGANIZATION EMBO reports VOL 4 | NO 12 | 2003 1155 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png EMBO Reports Springer Journals

X‐ray structure and activity of the yeast Pop2 protein: a nuclease subunit of the mRNA deadenylase complex

Download PDF
6 pages

Loading...

Page 2

Loading...

Page 3

Loading...

Page 4

Loading...

Page 5

Loading...

Page 6

 
/lp/springer-journals/x-ray-structure-and-activity-of-the-yeast-pop2-protein-a-nuclease-cFWA31iWzW

References (37)

Publisher
Springer Journals
Copyright
Copyright © European Molecular Biology Organization 2003
ISSN
1469-221X
eISSN
1469-3178
DOI
10.1038/sj.embor.7400020
Publisher site
See Article on Publisher Site

Abstract

scientific report scientificreport X-ray structure and activity of the yeast Pop2 protein: a nuclease subunit of the mRNA deadenylase complex 1 2 2 1+ Ste´phane Thore , Fabienne Mauxion , Bertrand Se´raphin & Dietrich Suck 1 2 European Molecular Biology Laboratory, Heidelberg, Germany, and Equipe Labelise´e La Ligue, Centre de Ge´ne´tique Mole´culaire, Gif sur Yvette Cedex, France In Saccharomyces cerevisiae, a large complex, known as the et al., 1995) and five Not proteins (Not1–5) required for the proper Ccr4–Not complex, containing two nucleases, is responsible for function of the Ccr4–Not complex (Draper et al., 1994; Liu et al., mRNA deadenylation. One of these nucleases is called Pop2 and 1998). Interactions between Ccr4, Pop2 and the Not proteins, has been identified by similarity with PARN, a human poly(A) together with other proteins present in the Ccr4–Not complex, nuclease. Here, we present the crystal structure of the nuclease have been characterized (Benson et al., 1998; Bai et al., 1999; domain of Pop2 at 2.3 A resolution. The domain has the fold of Badarinarayana et al., 2000; Lemaire & Collart, 2000; Liu et al., the DnaQ family and represents the first structure of an RNase 1997, 2001). from the DEDD superfamily. Despite the presence of two non- Ccr4 and Pop2 both contain a nuclease domain and are canonical residues in the active site, the domain displays RNase therefore potentially responsible for the deadenylation activity. The 2þ activity on a broad range of RNA substrates. Site-directed Ccr4 protein has been identified as a member of the Mg - mutagenesis of active-site residues demonstrates the intrinsic dependent endonuclease-related protein family (Dlakic, 2000), and ability of the Pop2 RNase D domain to digest RNA. This first its activity has been characterized both in vivo (Daugeron 0 0 structure of a nuclease involved in the 3 –5 deadenylation et al., 2001; Chen et al., 2002; Tucker et al., 2001, 2002) and of mRNA in yeast provides information for the understanding of in vitro (Viswanathan et al., 2003). Disruption of the gene and the mechanism by which the Ccr4–Not complex achieves its various mutants of the protein active site show a strong decrease in functions. mRNA deadenylation rates, indicating that the protein is an EMBO reports 4, 1150–1155 (2003) essential factor in this process (Daugeron et al., 2001; Tucker doi:10.1038/sj.embor.7400020 et al., 2001; Chen et al., 2002). The second protein, Pop2, is related to the ribonuclease (RNase) D family (Moser et al., 1997; Daugeron INTRODUCTION et al., 2001). While Pop2 has two non-conserved catalytic residues, mRNA degradation is an important step of gene expression, the other Pop2/Caf1 family members display a conserved and thus allowing to counter-balance or enhance the effect of transcription potentially fully functional active site. In addition, Pop2 gene and thus contributing to overall gene regulation. In Saccharo- disruption affects the rate and the degree of deadenylation of myces cerevisiae, the deadenylation process (reviewed in reporter mRNAs (Daugeron et al., 2001; Tucker et al., 2001). Caponigro & Parker, 1996) is the first step of mRNA degradation. The RNase D family (Mian, 1997) belongs to the DEDD It is achieved by the Ccr4–Not complex (Daugeron et al., 2001; superfamily composed of RNases as well as deoxyribonucleases Tucker et al., 2001), a large multi-protein complex (Liu et al., (DNases; Zuo & Deutscher, 2001). It is characterized by sequence 0 0 1997; Chen et al., 2001), built around a core of seven proteins: the motifs similar to the Exo I, II and III motifs of the 3 –5 carbon catabolite repressor 4 factor (Ccr4), first identified as a exodeoxyribonuclease (or proofreading) domains of DNA poly- gene expression regulating protein (Denis, 1984), the Pop2 merases, which contain four conserved acidic residues, namely protein, also known as Caf1 (for Ccr4 associated factor 1; Draper DEDD. These amino acids are responsible for binding the two metal ions involved in catalysis (reviewed in Joyce & Steitz, 1994). European Molecular Biology Laboratory, Meyerhofstrasse 1, D-69117 Heidelberg, Three-dimensional (3D) structures have been determined for the Germany DnaQ family (Ollis et al., 1985; Hamdan et al., 2002), which is a Equipe Labelise´e La Ligue, Centre de Ge´ne´tique Mole´culaire, CNRS UPR2167, subgroup of the DEDD superfamily. Avenue de la Terrasse, 91198 Gif sur Yvette Cedex, France Corresponding author. Tel: þ 49 6221 387 307; Fax: þ 49 6221 387 306; Here, we report the crystal structure of the RNase D domain of E-mail: suck@embl-heidelberg.de the yeast Pop2 protein. It represents the first structure of an RNase D protein as well as of an RNase from the DEDD superfamily. Received 22 August 2003; revised 17 September 2003; accepted 19 September 2003 Despite the non-conservation of two catalytic residues from the Published online 14 November 2003 1150 EMBO reports VOL 4 | NO 12 | 2003 &2003 EUROPEAN MOLECULAR BIOLOGY ORGANIZATION Crystal structure of the nuclease domain of Pop2 S. Thore et al. scientificreport Table 1 Data collection, phasing and refinement statistics Se-Met data sets Native Peak Inflection Remote Data collection and phasing statistics l (A) 1.8461 0.9795 0.9797 0.9689 Space group P2 2 2 P4 22P4 22P4 2 2 1 1 1 1 1 1 1 1 1 Resolution (A) 2.3 3.15 3.20 3.25 (2.38–2.3) (3.26–3.15) (3.31–3.2) (3.37–3.25) Reflection (unique) 52,320 10,414 10,007 9,508 Completeness (%) 95.9 (76.6) 99.2 (100) 99.5 (100) 99.6 (100) Redundancy 6.7 4 4 4 Average I/s(I) 11.9 (2.9) 11.9 (4.6) 11.7 (4.4) 11.4 (3.9) R 4.0 (30.9) 7.1 (30.5) 7.6 (32.8) 7.6 (34.9) sym FOM — — — 0.49 MAD FOM after solvent flattening — — — 0.65 Refinement statistics No. of reflections used 51,486 — — — R (%) 23.8 — — — value R (%) 26.0 — — — free Residues (out of 289) 263 (mol. 1)/254 (mol. 2) — — — Number of protein/solvent/ion atoms 4,214/46/4 — — — ˚ 2 Average B factor (A)51 — — — r.m.s.d. Dbond lengths (A) 0.008 — — — r.m.s.d. Dbond angles (degree) 1.4 — — — Values for the outermost resolution shell are shown in parentheses. FOM, figure of merit; R : R factor for 5% of the reflections excluded from the refinement. free P P R , |I/IS|/ I. sym hkl hkl DEDD signature sequence, we observe in vitro RNase activity towards poly(A), and also poly(C) and poly(U). Site-directed mutagenesis abolishes RNase activity, indicating that it is associated specifically with the Pop2 RNase D domain. Canonical active site residues are fully conserved within the Pop2/Caf1 family, strongly suggesting that RNase activity will be associated with all the family members. RESULTS AND DISCUSSION Structure determination and overall fold The fragment of the yeast Pop2 protein corresponding to the RNase D domain was crystallized and its structure was determined using MAD data from a single Se-Met-substituted protein containing crystal. Native crystals belong to space group P2 2 2 with two molecules in the asymmetric unit. The final 1 1 1 model contains two calcium ions, two xenon atoms and 263 or 254 residues (for each respective molecule) representing all amino acids where the main chain was clearly defined (Fig. 1A). Three regions were poorly resolved and did not allow tracing (residues 1–3, 50–59 and 213–225). The two molecules constituting the asymmetric unit adopt very similar conformations with a root mean square deviation (r.m.s.d.) of 0.47 A. The overall structure contains 13 a-helices and six b-strands, forming a kidney-shaped structure (Fig. 1A). The b-strands form the central core flanked by the a-helices. Helices 2, 3, 6, 7, 8 and 13 form the near-side of the molecule, while the remaining helices build the two lobes creating the cavity of the kidney-shaped molecule (Fig. 1B). Fig. 1 Structure of the nuclease domain of the Pop2 protein. (A)Ribbonplot Structural conservation within the DEDD nuclease family representation with the secondary elements in the following colour code: During recent years, several structures of proofreading enzymes a-helix, red; b-strands,green;and loops, yellow.(B) Crossed-eye stereo involved in DNA replication have been solved, revealing a representation of the Ca trace is displayed with every 20th residue marked. &2003 EUROPEAN MOLECULAR BIOLOGY ORGANIZATION EMBO reports VOL 4 | NO 12 | 2003 1151 Crystal structure of the nuclease domain of Pop2 S. Thore et al. scientificreport Fig. 2 Structural homology of Pop2 with members of the DEDD nuclease superfamily. (A) Structure-based sequence alignment of Pop2, the exonuclease domain of PolI and the e-subunit of PolIII. Sequence conservation is shown by colour coding: invariant residues are highlighted in red. Yellow highlights residues that have similar properties. Secondary structure elements of Pop2 are shown above the sequences. Small arrowheads indicate the conserved DEDD residues forming the catalytic site of the e-subunit. (B) The three structures shown in the same relative orientation. (C) The electrostatic surface potentials of Pop2 and the e-subunit indicate the location of the active site; catalytic residues are highlighted. (D) Close-up view of the active site of the e-subunit (salmon colour; with bound TMP in yellow) superimposed with the Pop2 (light green) structure and (E) side view of the secondary structure elements interacting with the bound nucleotide. Bold and italic labels correspond to the amino acids from Pop2 or the e-subunit, respectively. common fold for the DnaQ subgroup of the DEDD superfamily A global view of the electrostatic surface potential of Pop2 (Zuo & Deutscher, 2001). While the sequence similarity between indicates a highly negatively charged cavity (Fig. 2C). A similar Pop2 and the e-subunit of DNA polymerase III (PolIIIe) or the feature can be observed in exonuclease I (Breyer & Matthews, exonuclease domain of polymerase I (PolI) is only 15 or 20% 2000). These negatively charged regions in both proteins coincide (Fig. 2A), the overall superposition of the structures gives a r.m.s.d. with the active site in PolIII (Fig. 2C,D). A close look at the active of only 1.5 or 1.7 A of the Ca trace over 134 and 128 residues, site (Fig. 2D) shows that the Ser 44 and Gln 250 substitutions in the respectively (Fig. 2B; Ollis et al., 1985; Hamdan et al., 2002). The Pop2 amino-acid sequence could provide an oxygen atom able to Pop2 structure provides the first crystallographic evidence that the interact with the active site magnesium. In line with this DNase fold is also adopted by the RNase D, subgroup of the hypothesis, it is interesting to note that the two corresponding DEDD superfamily (Fig. 2B). residues are phylogenetically conserved in yeast species that 1152 EMBO reports VOL 4 | NO 12 | 2003 &2003 EUROPEAN MOLECULAR BIOLOGY ORGANIZATION Crystal structure of the nuclease domain of Pop2 S. Thore et al. scientificreport Fig. 3 RNase activity of Pop2 and conservation within the Pop2/Caf1 family. (A) Activity tests showing the RNase activity of Pop2 WT with poly(A), poly(U), poly(C) and G RNA substrates. (B) The double-mutant S44A/E46A (Pop ATA) does not degrade poly(A) RNA. (C) The time course of the RNase reaction demonstrates that the Pop2 nuclease is distributive. (D) Surface representation of the sequence conservation within the Pop2/Caf1 family is shown and is based on the sequence alignment shown in (E). Green colour indicates conserved residues, and white for non-conserved. Numbering is according to the Pop2 sequence and the residue name is according to the conserved amino acid in the Pop2/Caf1 family. (E) Sequence alignment of the eukaryotic Pop2/ Caf1 family with Mus musculus CAF1 (accession number NP_035265) and CALIF1 (NP_081225), Homo sapiens CAF1 (L46722) and CALIF1 (Q9UFF9), Xenopus laevis CAF1 (AAH41239), Drosophila melanogaster CAF1 (AE003543), Anopheles gambiae CAF1 (EAA12934), Arabidopsis thaliana CAF1 (AY070420), Plasmodium yoelii yoelii CAF1 (EAA20457), Caenorhabditis elegans CAF1 (U21854), Neurospora crassa CAF1 (EAA29793), Schizosaccharomyces pombe CAF1 (NP_588385), Encephalitozoon cuniculi CAF1 (NP_597215) and the RNase D domain of Saccharomyces cerevisiae Pop2 (P39008), colour coded as in Fig. 2A. DEDD catalytic site residues are indicated with arrowheads. diverged 5–20 million years ago, thus supporting the importance RNase activity of the Pop2 domain of this combination for yeast Pop2 function. Given the The literature contains contradictory data about the RNase activity number of negatively charged residues at the active site, it is also associated with the Pop2 protein from yeast. Under our in vitro possible that Pop2 binds only one metal ion instead of two conditions, Pop2 RNase activity is observed with poly(A), poly(U) normally observed for the DNases, resulting in a different reaction and poly(C), but not with oligo(G) (Fig. 3A, lanes 2–3, 5–6, 8–9 mechanism. and 11). However, more sensitive competition assays demonstrate Interestingly, the a-helices 4 and 5 (Fig. 2E) as well as residues a subtle preference for poly(A) (Daugeron et al., 2001). When the from the loop between strands b2 and b3 contact the leaving first two residues from the DEDD motif are mutated, that is, S44A nucleotide in the PolIII structure. These amino acids are identical, and E46A, the activity is lost (Fig. 3B, compare lanes 13–14 and or at least similar, in the case of Pop2, suggesting an interaction 16–17). The progressive appearance of shorter RNA fragments with the 3 -terminal nucleotide during the nuclease reaction indicates the distributive activity of the Pop2 RNase D domain despite the somewhat different orientation observed in the crystal (Fig. 3C). The ability to degrade several types of RNA implies (Fig. 2E). that the substrate RNA is not recognized in an absolute, &2003 EUROPEAN MOLECULAR BIOLOGY ORGANIZATION EMBO reports VOL 4 | NO 12 | 2003 1153 Crystal structure of the nuclease domain of Pop2 S. Thore et al. scientificreport specific manner but may involve a substantial contribution important information for the understanding of the deadenylase of van der Waals interactions, in agreement with the presence of complex in yeast and other eukaryotic species. This information a conserved hydrophobic patch following the b-strand 2 (Fig. 3E). could be used in order to better understand how the dead- The structure of the Pop2 nuclease domain represents the first enylation process is coupled to translation as well as to decapping 3D model of a catalytic component of the yeast deadenylase processes before the degradation occurs. complex. Pop2/Caf1 family METHODS Since Pop2 is the least conserved protein within the Pop2/Caf1 Expression and crystallization. The yeast Pop2 RNase D domain family (Fig. 3E; Daugeron et al., 2001), the fold conservation (amino acids (aa) 147–433) was expressed from a pET24d between the DNases and Pop2 suggests that the nuclease domain derivative (Novagen) with an amino (N)-terminal 6xHis tag-TEV of the Pop2/Caf1 family members will have the same fold as the protease site extension. After overexpression in Escherichia coli Pop2 protein. According to the sequence alignment of the Pop2/ BL21-CodonPlus (DE3)-RIL, cells were lysed with a French press Caf1 family, the other members do have a canonical active site, in buffer A (50 mM NaH PO (pH 7.5), 200 mM NaCl, 10 mM b- 2 4 that is, the DEDD signature sequence (Fig. 3E, arrowheads). mercaptoethanol) in the presence of protease inhibitor cocktail Therefore, these proteins are expected to display at least the same (Promega) and 10 mgml of RNase A (Sigma). After binding to Ni- or even higher RNase activity than Pop2. This has been verified NTA beads (Quiagen), the protein was eluted as indicaded. Pop2- in the case of human CAF1 (F. Mauxion and B. Se´raphin, containing fractions were passed onto a Superdex 75 column unpublished data). In addition, the conserved amino acids within (Amersham-Pharmacia) in buffer A. The tag was then removed by the Pop2/Caf1 family line the entire cavity, indicating a TEV protease cleavage at 16 1C (enzyme:substrate ratio 1:100 functionally important surface likely to correspond to the catalytic w/w) followed by chromatography on Ni-NTA beads. A second and the RNA interacting sites (Fig. 3D). gel filtration step in 20 mM Tris/Cl (pH 7.5) and 200 mM NaCl led to a 495% pure sample, which was concentrated to 9 mg ml .A Deadenylation and mRNA degradation Se-Met derivative was prepared similarly using the B834 E. coli The role of Pop2 in mRNA deadenylation has remained unclear, host strain grown in minimal medium containing L-seleno- while it is essential for deadenylation in vivo (Daugeron et al., methionine (50 mgml ). 2001; Tucker et al., 2001) and was reported to have RNase activity Crystals of the Pop2 nuclease domain were grown by vapour in vitro (Daugeron et al., 2001), analysis of point mutants failed to diffusion (hanging drops) at 4 1C by mixing an equal volume of reveal the requirement for a specific step in mRNA decay (Chen et protein and reservoir solution containing 2.5% PEG 3350 MW, al., 2002; Tucker et al., 2002). The crystal structure of the nuclease 100 mM Hepes (pH 7.0), 80 mM calcium acetate and 16.5% domain of Pop2 in conjunction with mutagenesis and in vitro glycerol. Drops were immediately micro-seeded and, after 7–8 assays demonstrate definitively that yeast Pop2 contains a days, transferred over a well containing a similar solution and nuclease domain displaying RNase activity of broad specificity. 27.5% glycerol. Crystals were then directly flash-frozen in liquid The requirement for two nucleases in the Ccr4–Not complex to nitrogen. Crystals from the Se-Met derivative were grown using the mediate mRNA deadenylation remains puzzling. Published data following conditions: 2% PEG 3350 MW, 120 mM calcium suggest that Ccr4 provides most of the nuclease activity in yeast acetate and protein at 5 mg ml . Xenon derivatization was (Chen et al., 2002; Tucker et al., 2002). However, as yeast Pop2 carried out on native crystals using a pressurized cell at 10 bar has an RNase activity in vitro and a non-canonical active site, this for 10 min (Djinovic-Carugo et al., 1998). situation may differ in other organisms. Indeed, the presence of A complete 2.3 A data set was collected under cryogenic two variations in the active site of Pop2 may strongly influence its conditions using the ESRF ID29 beamline. A three-wavelength efficiency compared to human CAF1 for example. It would thus be MAD experiment was performed on a single Se-Met-substituted of interest to evaluate the relative contribution of human CCR4 protein crystal at the PX06 beamline at SLS. Data were processed and human CAF1 to exonucleolytic poly(A) degradation. Simi- using the HKL package (Otwinowski & Minor, 1997). The native larly, it would be of interest to test whether the N-terminal domain and the Se-Met-substituted protein crystallized in space groups ˚ ˚ ˚ of full-length Pop2 inhibits RNase activity in vivo and whether this P2 2 2 (a ¼ 78.5 A, b ¼ 79.4 A, c ¼ 101.3 A, a ¼ 901, b ¼ 901, 1 1 1 is affected by the phosphorylation event known to occur in this g ¼ 901) with two molecules per asymmetric unit, and P4 2 2 1 1 ˚ ˚ ˚ region (Moriya et al., 2001). (a ¼ 78.2 A, b ¼ 78.2 A, c ¼ 102.6 A, a ¼ 901, b ¼ 901, g ¼ 901) Interaction of Ccr4 and Pop2 with the Not factors may also play with one molecule per asymmetric unit, respectively. an important role in regulating Pop2 activity and subsequently the Structure determination and refinement. Phases were obtained at deadenylation process. In the case of the human poly(A) nuclease 3.3 A with SOLVE (Terwilliger & Berendzen, 1999) and the PARN, reported to be similar to Pop2 as well as a member of the electron density map was improved by solvent flattening using the DEDD family, the protein contains a large domain belonging to program RESOLVE (Terwilliger, 2000). Only 40% of the protein the R3H family (Daugeron et al., 2001), which is able to bind could be traced with program O, version 7 (Jones et al., 1991). single-stranded RNA on its own. Since the Pop2/Caf1 family does DMMULTI from the CCP4 suite (CCP4, 1994) was used to extend not contain any additional RNA-binding domain, it is tempting to the phases to 2.3 A in P2 2 2 . Then, the density modification 1 1 1 speculate that Pop2 interactions with the Ccr4–Not complex may procedure implemented in CNS (Brunger et al., 1998) and based target it to the poly(A) tail. on the phases calculated from the partial model provided Further structural studies on the various complexes formed additional electron density. After three cycles of model building, between Ccr4, the Not proteins, mRNA and Pop2 are likely to give annealing, individual B-factor refinement and density modification, 1154 EMBO reports VOL 4 | NO 12 | 2003 &2003 EUROPEAN MOLECULAR BIOLOGY ORGANIZATION Crystal structure of the nuclease domain of Pop2 S. Thore et al. scientificreport Dlakic, M. (2000) Functionally unrelated signalling proteins contain a fold 263 and 254 residues for each molecule in the asymmetric unit 2+ similar to Mg -dependent endonucleases. Trends Biochem. Sci., 25, could be built (residues 4–49, 60–213 and 226–289 for molecule 272–273. 1 and residues 4–49, 62–207 and 228–289 for molecule 2). The Draper, M.P., Liu, H.Y., Nelsbach, A.H., Mosley, S.P. & Denis, C.L. (1994) final R is 26.3%, and the model shows good stereochemistry as CCR4 is a glucose-regulated transcription factor whose leucine-rich free repeat binds several proteins important for placing CCR4 in its proper indicated by PROCHECK (Laskowski et al., 1993). Table 1 shows promoter context. Mol. Cell. Biol., 14, 4522–4531. data collection and refinement statistics. Draper, M.P., Salvadore, C. & Denis, C.L. (1995) Identification of a mouse The coordinates have been deposited in the Protein Data Bank protein whose homolog in Saccharomyces cerevisiae is a component under accession code 1UOC. of the CCR4 transcriptional regulatory complex. Mol. Cell. Biol., 15, RNase assays. Polynucleotides (Amersham-Pharmacia) were 3487–3495. Hamdan, S., Carr, P.D., Brown, S.E., Ollis, D.L. & Dixon, N.E. (2002) stored as recommended. In vitro RNase assays were performed Structural basis for proofreading during replication of the Escherichia coli in 20 mM Tris/Cl (pH 7.0), 150 mM NaCl, 2 mM MgCl ,5U chromosome. Structure (Camb.), 10, 535–546. RNasin (Promega), 3 mM RNA substrate and indicated amounts of Jones, T.A., Zou, J.Y., Cowan, S.W. & Kjeldgaard, M. (1991) Improved purified Pop2. A volume of 10 ml of the reaction mixture was methods for binding protein models in electron density maps and the location of errors in these models. Acta Crystallogr. A, 47, 110–119. incubated at 25 1C for 1 h unless otherwise stated. Reactions that Joyce, C.M. & Steitz, T.A. (1994) Function and structure relationships in DNA were stopped by the addition of formamide/EDTA buffer were polymerases. Annu. Rev. Biochem., 63, 777–822. loaded onto 8 M urea/18% acrylamide (19:1) gels that were Laskowski, R.A., MacArthur, M.W., Moss, D.S. & Thornton, J.M. (1993) stained with toluidine blue. PROCHECK: a program to check the stereochemical quality of protein structures. J. Appl. Crystallogr., 26, 283–291. ACKNOWLEDGEMENTS Lemaire, M. & Collart, M.A. (2000) The TATA-binding protein-associated factor yTafII19p functionally interacts with components of the global We thank the staff at the European Synchrotron Radiation Facility (ESRF), transcriptional regulator Ccr4–Not complex and physically interacts with Swiss Light Synchrotron (SLS) and Deutsches Elektronen Synchrotron the Not5 subunit. J. Biol. Chem., 275, 26925–26934. (DESY) for assistance during data collection. The MAD data sets were Liu, H.Y., Toyn, J.H., Chiang, Y.C., Draper, M.P., Johnston, L.H. & Denis, C.L. collected at SLS, Paul Schonen Institute, Switzerland. We also thank (1997) DBF2, a cell cycle-regulated protein kinase, is physically and C. Schulze-Briex for support and acknowledge J. Basquin and J. Fe´thie`re functionally associated with the CCR4 transcriptional regulatory for data collection of the MAD data sets, as well as A. Popov for support complex. EMBO J., 16, 5289–5298. at the X13 beamline at DESY/EMBL-Hamburg. We also thank E. Ennifar Liu, H.Y., Badarinarayana, V., Audino, D.C., Rappsilber, J., Mann, M. & for useful crystallographic discussions and C. Temme for help with Denis, C.L (1998) The NOT proteins are part of the CCR4 transcriptional mutagenesis. Work in B.S.’s lab is supported by La Ligue contre le complex and affect gene expression both positively and negatively. Cancer, CNRS and the Ministry for Research (PRFMMIP). EMBO J., 17, 1096–1106. Liu, H.Y., Badarinarayana, V., Audino, D.C., Rappsilber, J., Mann, M. & REFERENCES Denis, C. (2001) Characterization of CAF4 and CAF16 reveals a Badarinarayana, V., Chiang, Y.C. & Denis, C.L. (2000) Functional interaction functional connection between the CCR4–NOT complex and a subset of of CCR4–NOT proteins with TATAA-binding protein (TBP) and its SRB proteins of the RNA polymerase II holoenzyme. J. Biol. Chem., 276, associated factors in yeast. Genetics, 155, 1045–1054. 7541–7548. Bai, Y., Salvadore, C., Chiang, Y.C., Collart, M.A., Liu, H.Y. & Denis, C.L. Mian, I.S. (1997) Comparative sequence analysis of ribonucleases HII, III, II (1999) The CCR4 and CAF1 proteins of the CCR4–NOT complex are PH and D. Nucleic Acids Res., 25, 3187–3195. physically and functionally separated from NOT2, NOT4, and NOT5. Moriya, H., Shimizu-Yoshida, Y., Omori, A., Iwashita, S., Katoh, M. & Sakai, Mol. Cell. Biol., 19, 6642–6651. A. (2001) Yak1p, a DYRK family kinase, translocates to the nucleus and Benson, J.D., Benson, M., Howley, P.M. & Struhl, K. (1998) Association of phosphorylates yeast Pop2p in response to a glucose signal. Genes Dev., distinct yeast Not2 functional domains with components of Gcn5 histone 15, 1217–1228. acetylase and Ccr4 transcriptional regulatory complexes. EMBO J., 17, Moser, M.J., Holley, W.R., Chatterjee, A. & Mian, I.S. (1997) The 6714–6722. proofreading domain of Escherichia coli DNA polymerase I and other Breyer, W.A. & Matthews, B.W. (2000) Structure of Escherichia coli DNA and/or RNA exonuclease domains. Nucleic Acids Res., 25, exonuclease I suggests how processivity is achieved. Nature Struct. Biol., 5110–5118. 7, 1125–1128. Ollis, D.L., Brick, P., Hamlin, R., Xuong, N.G. & Steitz, T.A. (1985) Structure Brunger, A.T. et al. (1998) Crystallography & NMR system: a new software of large fragment of Escherichia coli DNA polymerase I complexed with suite for macromolecular structure determination. Acta Crystallogr. D, dTMP. Nature, 313, 762–766. 54, 905–921. Otwinowski, Z. & Minor, W. (1997) Processing of X-ray diffraction data Caponigro, G. & Parker, R. (1996) Mechanisms and control of mRNA turnover collected in oscillation mode. Methods Enzymol., 276, 307–326. in Saccharomyces cerevisiae. Microbiol. Rev., 60, 233–249. Terwilliger, T.C. (2000) Maximum-likelihood density modification. Acta CCP4, C.C.P. (1994) The CCP4 suite: programs for protein crystallography. Crystallogr. D, 56, 965–972. Acta Crystallogr. D, 50, 760–763. Terwilliger, T.C. & Berendzen, J. (1999) Automated MAD and MIR structure Chen, J., Rappsilber, J., Chiang, Y.C., Russell, P., Mann, M. & Denis, C.L. solution. Acta Crystallogr. D, 55, 849–861. (2001) Purification and characterization of the 1.0 MDa CCR4–NOT Tucker, M., Valencia-Sanchez, M.A., Staples, R.R., Chen, J., Denis, C.L. & complex identifies two novel components of the complex. J. Mol. Biol., Parker, R. (2001) The transcription factor associated Ccr4 and Caf1 314, 683–694. proteins are components of the major cytoplasmic mRNA deadenylase in 0 0 Chen, J., Chiang, Y.C. & Denis, C.L. (2002) CCR4, a 3 –5 poly(A) RNA and Saccharomyces cerevisiae. Cell, 104, 377–386. ssDNA exonuclease, is the catalytic component of the cytoplasmic Tucker, M., Staples, R.R., Valencia-Sanchez, M.A., Muhlrad, D. & Parker, R. deadenylase. EMBO J., 21, 1414–1426. (2002) Ccr4p is the catalytic subunit of a Ccr4p/Pop2p/Notp mRNA Daugeron, M.C., Mauxion, F. & Seraphin, B. (2001) The yeast POP2 gene deadenylase complex in Saccharomyces cerevisiae. EMBO J., 21, encodes a nuclease involved in mRNA deadenylation. Nucleic Acids 1427–1436. Res., 29, 2448–2455. Viswanathan, P., Chen, J., Chiang, Y.C. & Denis, C.L. (2003) Identification of Denis, C.L. (1984) Identification of new genes involved in the regulation of multiple RNA features that influence CCR4 deadenylation activity. J. Biol. yeast alcohol dehydrogenase II. Genetics, 108, 833–844. Chem., 278, 14949–14955. Djinovic-Carugo, K., Everitt, P. & Tucker, P.A. (1998) A cell for producing Zuo, Y. & Deutscher, M.P. (2001) Exoribonuclease superfamilies: structural xenon-derivative crystals for cryocrystallographic analysis. J. Appl. analysis and phylogenetic distribution. Nucleic Acids Res., 29, Crystallogr., 31, 812–814. 1017–1026. &2003 EUROPEAN MOLECULAR BIOLOGY ORGANIZATION EMBO reports VOL 4 | NO 12 | 2003 1155

Journal

EMBO Reports – Springer Journals

Published: Dec 1, 2003

There are no references for this article.