Crystal Structure of the RNA Polymerase Domain of the West Nile Virus Non-structural Protein 5
2007; Elsevier BV; Volume: 282; Issue: 14 Linguagem: Inglês
10.1074/jbc.m607273200
ISSN1083-351X
AutoresHélène Malet, Marie-Pierre Egloff, Barbara Selisko, R.E. Butcher, P.J. Wright, Michael M. Roberts, Arnaud Gruez, G. Sulzenbacher, Clemens Vonrhein, G. Bricogne, Jason M. Mackenzie, Alexander A. Khromykh, Andrew D. Davidson, Bruno Canard,
Tópico(s)Vibrio bacteria research studies
ResumoViruses of the family Flaviviridae are important human and animal pathogens. Among them, the Flaviviruses dengue (DENV) and West Nile (WNV) cause regular outbreaks with fatal outcomes. The RNA-dependent RNA polymerase (RdRp) activity of the non-structural protein 5 (NS5) is a key activity for viral RNA replication. In this study, crystal structures of enzymatically active and inactive WNV RdRp domains were determined at 3.0- and 2.35-Å resolution, respectively. The determined structures were shown to be mostly similar to the RdRps of the Flaviviridae members hepatitis C and bovine viral diarrhea virus, although with unique elements characteristic for the WNV RdRp. Using a reverse genetic system, residues involved in putative interactions between the RNA-cap methyltransferase (MTase) and the RdRp domain of Flavivirus NS5 were identified. This allowed us to propose a model for the structure of the full-length WNV NS5 by in silico docking of the WNV MTase domain (modeled from our previously determined structure of the DENV MTase domain) onto the RdRp domain. The Flavivirus RdRp domain structure determined here should facilitate both the design of anti-Flavivirus drugs and structure-function studies of the Flavivirus replication complex in which the multifunctional NS5 protein plays a central role. Viruses of the family Flaviviridae are important human and animal pathogens. Among them, the Flaviviruses dengue (DENV) and West Nile (WNV) cause regular outbreaks with fatal outcomes. The RNA-dependent RNA polymerase (RdRp) activity of the non-structural protein 5 (NS5) is a key activity for viral RNA replication. In this study, crystal structures of enzymatically active and inactive WNV RdRp domains were determined at 3.0- and 2.35-Å resolution, respectively. The determined structures were shown to be mostly similar to the RdRps of the Flaviviridae members hepatitis C and bovine viral diarrhea virus, although with unique elements characteristic for the WNV RdRp. Using a reverse genetic system, residues involved in putative interactions between the RNA-cap methyltransferase (MTase) and the RdRp domain of Flavivirus NS5 were identified. This allowed us to propose a model for the structure of the full-length WNV NS5 by in silico docking of the WNV MTase domain (modeled from our previously determined structure of the DENV MTase domain) onto the RdRp domain. The Flavivirus RdRp domain structure determined here should facilitate both the design of anti-Flavivirus drugs and structure-function studies of the Flavivirus replication complex in which the multifunctional NS5 protein plays a central role. The Flaviviridae form a large family of single-stranded positive-sense RNA viruses comprising the three genera Hepacivirus, Pestivirus, and Flavivirus. The genus Flavivirus contains more than 80 known arthropod-borne viruses, including major human and animal pathogens such as dengue virus (DENV), 3The abbreviations used are: DENV, dengue virus; aa, amino acids; a/bNLS, nuclear localization sequence that recognize importin-α; bNLS, nuclear localization sequence that recognize importin-β; BVDV, bovine viral diarrhea virus; DENV-2, dengue virus type 2; ds, double-stranded; HCV, hepatitis C virus; Impα/β, karyopherin-α/β heterodimer, also known as importin-α/β; Impα, karyopherin-α, also known as importin-α; Impβ, karyopherin-β, also known as importin-β; MTase, methyltransferase; NLS, nuclear localization sequence; NS3, non-structural protein 3; NS5, nonstructural protein 5; POL, West Nile virus polymerase; POL1, construct 273-905 of West Nile virus polymerase; POL2, construct 317-905 of West Nile virus polymerase (in complex with calcium ion); POL3, construct 273-882 of West Nile virus polymerase; Phi6, Pseudomonas phage Phi-6; RdRp, RNAdependent RNA polymerase; r.m.s., root mean square; RT, reverse transcriptase; WNV, West Nile virus.3The abbreviations used are: DENV, dengue virus; aa, amino acids; a/bNLS, nuclear localization sequence that recognize importin-α; bNLS, nuclear localization sequence that recognize importin-β; BVDV, bovine viral diarrhea virus; DENV-2, dengue virus type 2; ds, double-stranded; HCV, hepatitis C virus; Impα/β, karyopherin-α/β heterodimer, also known as importin-α/β; Impα, karyopherin-α, also known as importin-α; Impβ, karyopherin-β, also known as importin-β; MTase, methyltransferase; NLS, nuclear localization sequence; NS3, non-structural protein 3; NS5, nonstructural protein 5; POL, West Nile virus polymerase; POL1, construct 273-905 of West Nile virus polymerase; POL2, construct 317-905 of West Nile virus polymerase (in complex with calcium ion); POL3, construct 273-882 of West Nile virus polymerase; Phi6, Pseudomonas phage Phi-6; RdRp, RNAdependent RNA polymerase; r.m.s., root mean square; RT, reverse transcriptase; WNV, West Nile virus. yellow fever virus, Japanese encephalitis virus, and West Nile virus (WNV). Both DENV and WNV are considered as emerging pathogens. Dengue fever is one of the most important mosquito-borne viral diseases in the world, with more than 3 billion people at risk in endemic tropical areas (1Mackenzie J.S. Gubler D.J. Petersen L.R. Nat. Med. 2004; 10: S98-S109Crossref PubMed Scopus (1003) Google Scholar). Dengue outbreaks are increasingly severe in terms of cases and fatalities in many regions of the world (2Guzman M.G. Kouri G. Lancet Infect. Dis. 2002; 2: 33-42Abstract Full Text Full Text PDF PubMed Scopus (943) Google Scholar). WNV was discovered in the West Nile district in Uganda in 1937 and was subsequently shown to have an extensive worldwide distribution with the exception of the Americas (1Mackenzie J.S. Gubler D.J. Petersen L.R. Nat. Med. 2004; 10: S98-S109Crossref PubMed Scopus (1003) Google Scholar). In 1999, WNV was introduced into the Americas in the New York City area and has since spread throughout the mainland United States, southern Canada, and Mexico. WNV epidemics in the United States have resulted in a total of 23,925 cases of human disease and 946 deaths reported to the Centers for Disease Control (CDC) from 1999 to 2006. WNV consists of 2 lineages (I and II). The North American WNV isolates belong to lineage I, which also includes the Australian subtype Kunjin (3Scherret J.H. Poidinger M. Mackenzie J.S. Broom A.K. Deubel V. Lipkin W.I. Briese T. Gould E.A. Hall R.A. Emerg. Infect. Dis. 2001; 7: 697-705Crossref PubMed Scopus (134) Google Scholar). In contrast to other lineage I WNV strains (4Lanciotti R.S. Ebel G.D. Deubel V. Kerst A.J. Murri S. Meyer R. Bowen M. McKinney N. Morrill W.E. Crabtree M.B. Kramer L.D. Roehrig J.T. Virology. 2002; 298: 96-105Crossref PubMed Scopus (337) Google Scholar), infections with the Kunjin subtype of WNV do not cause fatal disease in humans (5Hall R.A. Broom A.K. Smith D.W. Mackenzie J.S. Curr. Top. Microbiol. Immunol. 2002; 267: 253-269PubMed Google Scholar). The Flavivirus positive sense RNA genome contains a single open reading frame encoding a polyprotein that is processed into three structural and seven non-structural proteins (NS1, NS2A, NS2B, NS3, NS4A, NS4B, and NS5). Signature-sequence analysis suggests that the non-structural protein NS5 is comprised of two domains. The DENV NS5 30-kDa N-terminal domain has been shown to possess RNA cap (nucleoside-2′-O) methyltransferase (MTase) activity, and its crystal structure has been determined (6Egloff M.P. Benarroch D. Selisko B. Romette J.L. Canard B. EMBO J. 2002; 21: 2757-2768Crossref PubMed Scopus (477) Google Scholar). That of WNV has been shown to mediate both (guanine-N7) and (nucleoside-2′-O) methylation of the 5′-cap structure (7Ray D. Shah A. Tilgner M. Guo Y. Zhao Y. Dong H. Deas T.S. Zhou Y. Li H. Shi P.Y. J. Virol. 2006; 80: 8362-8370Crossref PubMed Scopus (290) Google Scholar). The C-terminal domain of about 70 kDa harbors the RdRp (RNA-dependent RNA polymerase), as initially identified by the presence of signature-sequence motifs A to F (8O'Reilly E.K. Kao C.C. Virology. 1998; 252: 287-303Crossref PubMed Scopus (261) Google Scholar, 9Koonin E.V. J. Gen. Virol. 1991; 72: 2197-2206Crossref PubMed Scopus (713) Google Scholar), and subsequently confirmed in RdRp activity assays (10Steffens S. Thiel H.J. Behrens S.E. J. Gen. Virol. 1999; 80: 2583-2590Crossref PubMed Scopus (91) Google Scholar, 11Guyatt K.J. Westaway E.G. Khromykh A.A. J. Virol. Methods. 2001; 92: 37-44Crossref PubMed Scopus (95) Google Scholar, 12Selisko B. Dutartre H. Guillemot J.C. Debarnot C. Benarroch D. Khromykh A. Despres P. Egloff M.P. Canard B. Virology. 2006; 351: 145-158Crossref PubMed Scopus (104) Google Scholar). The Flaviviridae RdRp is essential to viral replication. The RdRp duplicates the single-stranded RNA genome during a single, continuous polymerization event. The RdRp enters at the 3′-end of the genome and is able to copy the whole RNA molecule in a primer-independent fashion (12Selisko B. Dutartre H. Guillemot J.C. Debarnot C. Benarroch D. Khromykh A. Despres P. Egloff M.P. Canard B. Virology. 2006; 351: 145-158Crossref PubMed Scopus (104) Google Scholar, 13Ackermann M. Padmanabhan R. J. Biol. Chem. 2001; 276: 39926-39937Abstract Full Text Full Text PDF PubMed Scopus (254) Google Scholar, 14Nomaguchi M. Ackermann M. Yon C. You S. Padmanabhan R. J. Virol. 2003; 77: 8831-8842Crossref PubMed Scopus (83) Google Scholar, 15Ranjith-Kumar C.T. Gutshall L. Kim M.J. Sarisky R.T. Kao C.C. J. Virol. 2002; 76: 12526-12536Crossref PubMed Scopus (81) Google Scholar, 16Ranjith-Kumar C.T. Kim Y.C. Gutshall L. Silverman C. Khandekar S. Sarisky R.T. Kao C.C. J. Virol. 2002; 76: 12513-12525Crossref PubMed Scopus (82) Google Scholar, 17Kao C.C. Singh P. Ecker D.J. Virology. 2001; 287: 251-260Crossref PubMed Scopus (184) Google Scholar) referred to as de novo RNA synthesis. Crystal structures of RdRps from ∼9 virus species have been reported, among them those of Hepaci- and Pestivirus RdRps (18Ago H. Adachi T. Yoshida A. Yamamoto M. Habuka N. Yatsunami K. Miyano M. Structure. 1999; 7: 1417-1426Abstract Full Text Full Text PDF PubMed Scopus (387) Google Scholar, 19Bressanelli S. Tomei L. Roussel A. Incitti I. Vitale R.L. Mathieu M. De Francesco R. Rey F.A. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 13034-13039Crossref PubMed Scopus (543) Google Scholar, 20Choi K.H. Groarke J.M. Young D.C. Kuhn R.J. Smith J.L. Pevear D.C. Rossmann M.G. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 4425-4430Crossref PubMed Scopus (192) Google Scholar, 21Lesburg C.A. Cable M.B. Ferrari E. Hong Z. Mannarino A.F. Weber P.C. Nat. Struct. Biol. 1999; 6: 937-943Crossref PubMed Scopus (693) Google Scholar, 22Bressanelli S. Tomei L. Rey F.A. De Francesco R. J. Virol. 2002; 76: 3482-3492Crossref PubMed Scopus (335) Google Scholar), but none from a Flavivirus species. Very little is known about the mechanism of flaviviral RNA capping, or its co-ordination with RNA synthesis. In eukaryotic cells, mRNA capping is a nuclear event, whereas Flavivirus RNA synthesis is thought to occur in the cytoplasm of the infected cell. The viral NS3 and NS5 proteins have been demonstrated to possess RNA-triphosphatase and 2′-O-/N7-MTase activities, respectively (6Egloff M.P. Benarroch D. Selisko B. Romette J.L. Canard B. EMBO J. 2002; 21: 2757-2768Crossref PubMed Scopus (477) Google Scholar, 7Ray D. Shah A. Tilgner M. Guo Y. Zhao Y. Dong H. Deas T.S. Zhou Y. Li H. Shi P.Y. J. Virol. 2006; 80: 8362-8370Crossref PubMed Scopus (290) Google Scholar, 23Benarroch D. Selisko B. Locatelli G.A. Maga G. Romette J.L. Canard B. Virology. 2004; 328: 208-218Crossref PubMed Scopus (143) Google Scholar, 24Wengler G. Virology. 1993; 197: 265-273Crossref PubMed Scopus (174) Google Scholar, 25Bartelma G. Padmanabhan R. Virology. 2002; 299: 122-132Crossref PubMed Scopus (117) Google Scholar, 26Yon C. Teramoto T. Mueller N. Phelan J. Ganesh V.K. Murthy K.H. Padmanabhan R. J. Biol. Chem. 2005; 280: 27412-27419Abstract Full Text Full Text PDF PubMed Scopus (121) Google Scholar), however, the guanylyltransferase activity required for a complete capping reaction is still uncharacterized. Interestingly, during infection, a proportion of the NS5 and, more recently, of the NS3 protein of a number of flaviviruses has been detected in the nucleus (DENV) or at least the perinuclear region (WNV and Japanese encephalitis virus) of infected cells (27Buckley A. Gaidamovich S. Turchinskaya A. Gould E.A. J. Gen. Virol. 1992; 73: 1125-1130Crossref PubMed Scopus (70) Google Scholar, 28Kapoor M. Zhang L. Ramachandra M. Kusukawa J. Ebner K.E. Padmanabhan R. J. Biol. Chem. 1995; 270: 19100-19106Abstract Full Text Full Text PDF PubMed Scopus (278) Google Scholar, 29Uchil P.D. Kumar A.V. Satchidanandam V. J. Virol. 2006; 80: 5451-5464Crossref PubMed Scopus (77) Google Scholar). Trafficking of the DENV NS5 to the nucleus was proposed to rely on a nuclear import pathway based on the identification of functional nuclear localization sequences (NLSs) in NS5 and their interaction in vitro with distinct members of the karyopherin family of intracellular transport proteins (karyopherin-α/β, also known as importin-α/β or Impα/β) (30Johansson M. Brooks A.J. Jans D.A. Vasudevan S.G. J. Gen. Virol. 2001; 82: 735-745Crossref PubMed Scopus (148) Google Scholar, 31Brooks A.J. Johansson M. John A.V. Xu Y. Jans D.A. Vasudevan S.G. J. Biol. Chem. 2002; 277: 36399-36407Abstract Full Text Full Text PDF PubMed Scopus (122) Google Scholar). NLSs are small motifs composed of one or more clusters of basic amino acids that do not conform to a specific sequence consensus. Two adjacent regions encompassing distinct NLSs have been defined in DENV NS5 termed the bNLS (aa 320 to 368) and the a/bNLS (aa 369 to 405) (30Johansson M. Brooks A.J. Jans D.A. Vasudevan S.G. J. Gen. Virol. 2001; 82: 735-745Crossref PubMed Scopus (148) Google Scholar, 31Brooks A.J. Johansson M. John A.V. Xu Y. Jans D.A. Vasudevan S.G. J. Biol. Chem. 2002; 277: 36399-36407Abstract Full Text Full Text PDF PubMed Scopus (122) Google Scholar). The bNLS and a/bNLS bind with high affinity to Impβ and an Impα/β heterodimer in vitro, respectively, and both are capable of targeting β-galactosidase to the nucleus in the context of a fusion protein (30Johansson M. Brooks A.J. Jans D.A. Vasudevan S.G. J. Gen. Virol. 2001; 82: 735-745Crossref PubMed Scopus (148) Google Scholar, 31Brooks A.J. Johansson M. John A.V. Xu Y. Jans D.A. Vasudevan S.G. J. Biol. Chem. 2002; 277: 36399-36407Abstract Full Text Full Text PDF PubMed Scopus (122) Google Scholar). However, whether the bNLS and a/bNLS are required for nuclear import of the intact NS5 protein during viral infection is not known. The bNLS of DENV NS5 has also been shown to interact with NS3 and competition between Impβ and NS3 binding has been confirmed using pulldown assays (30Johansson M. Brooks A.J. Jans D.A. Vasudevan S.G. J. Gen. Virol. 2001; 82: 735-745Crossref PubMed Scopus (148) Google Scholar). Here, we report the crystal structures of enzymatically active and inactive WNV RdRp domains. They reveal a classic RdRp fold bearing palm, thumb, and finger domains decorated with Flavivirus-specific attributes such as a priming loop adopting a different fold. Structural elements in the WNV RdRp structure corresponding to putative DENV NLSs were found to be integral parts of the RdRp domain. Interestingly, despite sequence conservation in the NLSs of DENV and WNV NS5 the latter was not found to localize to the nucleus. To gain insight into the overall spatial organization of the full-length NS5 protein, a reverse genetic approach was used to identify amino acids in the DENV MTase and RdRp domains, which potentially interact and could be used to position one domain relative to the other. The structure of the WNV MTase domain was modeled using DENV MTase atomic coordinates. Both WNV MTase and RdRp domains were assembled in silico using docking procedures, generating a model of full-length WNV NS5 protein. Expression and Purification of the Polymerase Domains—Three WNV strain Kunjin ns5 gene constructs encoding a N-terminal His6 tag fused to pol1 (corresponding to aa 273-905), pol2 (corresponding to aa 317-905), or pol3 (corresponding to aa 273 to 882) open reading frames were cloned into the pDEST14 vector, and expressed in Escherichia coli. The proteins were expressed and purified to homogeneity as follows: the E. coli strain C41pRos, transformed with either plasmid construct, was grown at 37 °C to an A600 of 0.6, induced with 0.5 mm isopropyl β-d-thiogalactopyranoside, and further incubated 16-18 h at 17 °C. Cells were harvested by centrifugation. The cell pellet was resuspended in 50 mm Tris buffer, pH 8.0, containing 150 mm NaCl, 10 mm imidazole, DNase I (2 μg/ml), a protease inhibitor tablet (Sigma), and sonicated on ice. The sample was centrifuged, the supernatant collected and filtered through a 0.22-μm filter. The sample was applied to a 5-ml bed volume HiTrap nickel immobilized metal ion affinity chromatography column (Amersham Biosciences) connected to a FPLC system (Amersham Biosciences). The protein was eluted with 50 mm Tris buffer, pH 8.0, containing 150 mm NaCl and 500 mm imidazole. Protein-containing fractions were then applied onto a preparative Superdex 200 gel filtration column pre-equilibrated in 10 mm Tris buffer, pH 9.0, with 300 mm NaCl and 5% glycerol. Protein was concentrated to 6 (POL1) and 9 mg/ml (POL2 and POL3) using a Vivaspin 30-kDa molecular mass cut-off centrifugal concentrator (Vivascience). A selenomethionine-substituted protein was used for structure determination. It was expressed according to standard conditions of the methionine-biosynthesis pathway inhibition (32Doublie S. Methods Enzymol. 1997; 276: 523-530Crossref PubMed Scopus (794) Google Scholar) and purified following the same procedure as that for the native protein. Crystallization—Initial crystallization trials were set up with a nano-drop dispenser in 96-well sitting drop plates using commercial crystallization kits and purified protein at 6 mg/ml for POL1 and 9 mg/ml for POL2. Initial hits were further optimized by the hanging drop vapor diffusion method in Linbro plates by mixing 1 μl of protein solution with 1 μl of reservoir solution. Crystals of POL1 were grown from 10% PEG 1000, 0.2 m sodium cacodylate, 0.3 m MgCl2, pH 7.0. POL2 was crystallized using two conditions (i) 5% PEG 10,000, 0.2 m imidazole malate, pH 8.0, and (ii) 10% PEG 8000, 0.1 m imidazole pH 7.0, and 0.2 m calcium acetate. Crystals were briefly soaked in a cryo-protectant solution composed of mother liquor supplemented with 36% glycerol and flash frozen in liquid nitrogen. WNV POL1 and POL2 diffraction intensities were recorded on different beamlines (Table 1) at the European Synchrotron Radiation Facility (Grenoble, France). Integration, scaling, and merging of the intensities were carried out using programs from the CCP4 suite (33Collaborative Computational Project, NActa Crystallogr. D Biol. Crystallogr. 1994; 50: 760-763Crossref PubMed Scopus (19733) Google Scholar) and statistics are provided in Table 1.TABLE 1Crystallographic data and refinement statisticsPOL2POL1, magnesiumSAD (selenomethionine)NativeCalciumConstruct (aa)317-905317-905317-905273-905Data collectionSpace groupP43P43P43I222Cell dimensions (Å)a = b = 110.0 c = 67.7a = b = 110.0 c = 69.1a = b = 110.1 c = 68.6a = 73.3 b = 103.4 c = 190.1X-ray sourceESRF ID23-1ESRF ID14-EH1ESRF ID14-EH3ESRF BM14Wavelength (Å)0.97950.9340.9330.9763Resolution range (Å)35.0-2.8 (2.95-2.8)35.0-2.5 (2.65-2.5)35.0-2.35 (2.48-2.35)21.77-3.0 (3.18-3.0)Total reflections237,932 (34,735)163,928 (24,051)194,222 (17,177)67,265 (10,835)Unique reflections20,157 (2934)28,709 (4172)34,348 (4930)14,328 (2140)Completeness (%)100 (100)99.8 (100)99.9 (99.7)96.4 (100)〈I/σ(I) 〉23.1 (4.5)18.4 (2.9)21.8 (1.5)12.9 (3.4)RsymaRsym = ∑|I - 〈I 〉|/∑I, where I is the observed intensity and 〉I 〉 is the average intensity. Values in parentheses refer to the highest resolution shell.0.106 (0.560)0.069 (0.567)0.055 (0.567)0.097 (0.337)Multiplicity11.8 (11.8)5.7 (5.8)5.7 (3.5)4.7 (5.1)Anomalous completeness (%)99.9 (99.9)Anomalous multiplicity6.1 (6.1)RefinementRwork (%)bR = ∑||Fo| - |Fc||/∑|Fo|.20.621.026.0Rfree (%)cRfree is calculated as R, but on 5% of all reflections that are never used in crystallographic refinement.24.325.926.5Number of atomsProtein3,9373,9324,915Water molecules10917912Ion in the non-catalytic position near the active site01 (Calcium)1 (Magnesium)r.m.s. deviationsBond lengths (Å)0.0220.0210.007Bond angles (°)1.71.71.2Ramachandran analysis (%)Most favored91.091.985.4Additionally allowed8.57.612.4Generously allowed0.50.51.5Disallowed000.7Sequence assigned to model (aa)322-337, 362-410, 420-452, 473-576, 603-747, 752-891322-338, 363-410, 420-452, 474-576, 604-747, 752-892274-409, 416-458, 471-899PDB code2HCS2HCN2HFZa Rsym = ∑|I - 〈I 〉|/∑I, where I is the observed intensity and 〉I 〉 is the average intensity. Values in parentheses refer to the highest resolution shell.b R = ∑||Fo| - |Fc||/∑|Fo|.c Rfree is calculated as R, but on 5% of all reflections that are never used in crystallographic refinement. Open table in a new tab Structure Determination—The structure of WNV POL2 was solved using single wavelength anomalous dispersion data collected at the peak of the selenium absorption edge from a selenomethionine-derivatized crystal. Location of 16 selenium atoms (of the 21 expected) was performed using SHELXD (34Schneider T.R. Sheldrick G.M. Acta Crystallogr. D Biol. Crystallogr. 2002; 58: 1772-1779Crossref PubMed Scopus (1575) Google Scholar). Phases and figures of merit were calculated using SHELXE (35Sheldrick G.M. Z. Kristallogr. 2002; 217: 644-650Crossref Scopus (360) Google Scholar) and SHARP (36Bricogne G. Vonrhein C. Flensburg C. Schiltz M. Paciorek W. Acta Crystallogr. D Biol. Crystallogr. 2003; 59: 2023-2030Crossref PubMed Scopus (553) Google Scholar). Density modification was performed using RESOLVE (37Terwilliger T.C. Acta Crystallogr. D Biol. Crystallogr. 2002; 58: 1937-1940Crossref PubMed Scopus (281) Google Scholar) and SHARP. About 40% of the model was built in an automatic fashion with RESOLVE and completed manually using COOT (38Emsley P. Cowtan K. Acta Crystallogr. D Biol. Crystallogr. 2004; 60: 2126-2132Crossref PubMed Scopus (23012) Google Scholar), with the RdRp structures of HCV and BVDV as guides for connectivity (Protein Data Bank codes 1NB6 and 1S49). The model was refined against two data sets using REFMAC (39Murshudov G.N. Vagin A.A. Dodson E.J. Acta Crystallogr. D Biol. Crystallogr. 1997; 53: 240-255Crossref PubMed Scopus (13818) Google Scholar): the first, which extends to 2.5-Å resolution, was called "native" as it contains no ion near the active site. The second extends to 2.35-Å resolution and was called "calcium" as a calcium ion was found in a non-catalytic position near the active site (Table 1). The calcium resulting model was subsequently used to solve the structure of POL1 by the molecular replacement method using AMORE (40Navaza J. Acta Crystallogr. D Biol. Crystallogr. 2001; 57: 1367-1372Crossref PubMed Scopus (658) Google Scholar). Refinement of POL1 was initially performed using simulated annealing as implemented in CNS (41Brunger A.T. Adams P.D. Clore G.M. DeLano W.L. Gros P. Grosse-Kunstleve R.W. Jiang J.S. Kuszewski J. Nilges M. Pannu N.S. Read R.J. Rice L.M. Simonson T. Warren G.L. Acta Crystallogr. D. Biol. Crystallogr. 1998; 54: 905-921Crossref PubMed Scopus (16948) Google Scholar) at 3.0-Å resolution (Table 1). Electron density corresponding to the extra 44 N-terminal amino acids of POL1 (compared with POL2) and to some disordered loops of POL2 became visible. Attempts to model the missing parts were performed using several rounds of manual building and refinement using CNS (41Brunger A.T. Adams P.D. Clore G.M. DeLano W.L. Gros P. Grosse-Kunstleve R.W. Jiang J.S. Kuszewski J. Nilges M. Pannu N.S. Read R.J. Rice L.M. Simonson T. Warren G.L. Acta Crystallogr. D. Biol. Crystallogr. 1998; 54: 905-921Crossref PubMed Scopus (16948) Google Scholar). BUSTER-TNT (42Blanc E. Roversi P. Vonrhein C. Flensburg C. Lea S.M. Bricogne G. Acta Crystallogr. D. Biol. Crystallogr. 2004; 60: 2210-2221Crossref PubMed Scopus (594) Google Scholar) was then used, and allowed further completion of the model, particularly for residues 363-367, for which no density had previously been visible. Finally, POL1 was refined using BUSTER-TNT under harmonic restraints to much of the POL2 calcium structure, to avoid "un-refinement" of the POL2 substructure against the lower-resolution POL1 data as detailed in "Supplemental Material." The R-factors and geometry were improved by this harmonically restrained refinement. Refinement statistics are listed in Table 1. RdRp Activity Tests—RdRp activity tests were performed in 50-μl reactions containing 50 mm HEPES, pH 8.0, 5 mm dithiothreitol, 10 mm KCl, 5 mm MnCl2, 5 mm MgCl2, 10 μm GTP (0.01 μCi of [3H]GTP per μl, 6.1 Ci/mmol, Amersham Biosciences) using 1 μm poly(rC) (Amersham Biosciences) and WNV RdRp (264 nm POL1, 760 nm POL2, and 800 nm POL3). Reactions were started with a premix of MnCl2, MgCl2, and GTP and incubated at 30 °C. Samples of 12 μl were taken after 5, 10, and/or 15 min and the reaction stopped by adding 30 μl of 50 mm EDTA in 96-well plates. Samples were then transferred to glass fiber filter mats with DEAE active groups (DEAE filter mat, Wallac) using a Filtermat Harvester (Packard Instruments). Filtermats were washed three times with 0.3 m ammonium formate, pH 8.0, twice with water, once with ethanol, dried, and the filter transferred into sample bags. Liquid scin-fluid fluid was added and incorporation was measured in counts per minute using a Wallac MicroBeta TriLux Liquid Scintillation Counter. Introduction of NS5 Mutations into a Genomic Length DENV cDNA Clone—Mutations encoding changes in NS5 amino acids were introduced into the genomic length DENV serotype 2 strain New Guinea C (DENV-2) cDNA clone pDVWS601, which yields the virus v601 (43Gualano R.C. Pryor M.J. Cauchi M.R. Wright P.J. Davidson A.D. J. Gen. Virol. 1998; 79: 437-446Crossref PubMed Scopus (132) Google Scholar, 44Pryor M.J. Gualano R.C. Lin B. Davidson A.D. Wright P.J. J. Gen. Virol. 1998; 79: 2631-2639Crossref PubMed Scopus (54) Google Scholar). Mutations were initially introduced into subgenomic PCR fragments by overlap extension PCR using mutagenic primers (details available from the authors upon request) and then transferred into pDVWS601 as follows. The mutation K46A/R47A/E49A (each NS5 amino acid targeted for mutagenesis is numbered followed by the substituted amino acid) was engineered into a 1001-bp overlap extension PCR fragment (DENV-2 nucleotides 7165-8165), which was HpaI7406/StuI7874 digested and introduced into the corresponding sites of pDVWS601 to produce pDVWS601-NS5K46A,R47A,E49A. The NS5 mutation L512V was engineered into a 2077-bp fragment (DENV-2 nucleotides 8085-10161), which was AatII8570/MluI9732 digested and introduced into the corresponding sites of pDVWS601 and pDVWS601-NS5K46A,R47A,E49A to produce pDVWS601-NS5L512V and pDVWS601-NS5K46A,R47A,E49A,L512V, respectively. Recombinant DENV Recovery, Growth, and Sequencing—Procedures for the recovery of infectious DENV-2 from pDVWS601 including transcription of RNA, electroporation of BHK-21 cells, and passaging and plaque assay of virus in Aedes albopictus C6/36 cells have been described previously (43Gualano R.C. Pryor M.J. Cauchi M.R. Wright P.J. Davidson A.D. J. Gen. Virol. 1998; 79: 437-446Crossref PubMed Scopus (132) Google Scholar, 44Pryor M.J. Gualano R.C. Lin B. Davidson A.D. Wright P.J. J. Gen. Virol. 1998; 79: 2631-2639Crossref PubMed Scopus (54) Google Scholar). Confirmation of the mutations in the recovered viruses and the identification of second site mutations were performed by sequencing the entire DENV-2 genome. Viral RNA was extracted from the culture supernatants of infected cells using a QIAamp Viral RNA Extraction kit (Qiagen) and used for the production of six overlapping 2-kilobase pair RT-PCR products spanning the DENV-2 genome using DENV-2-specific primers and the SuperScript One-Step RT-PCR with Platinum Taq System (Invitrogen). Each RT-PCR product was then purified using a Qiagen PCR Purification Kit and used for automated DNA sequencing using DENV-2-specific primers. Molecular Modeling and Docking—Alignment of the MTase sequences of DENV-2 (PDB code 1L9K) and WNV was performed using SPDBV (45Guex N. Peitsch M.C. Electrophoresis. 1997; 18: 2714-2723Crossref PubMed Scopus (9507) Google Scholar) (sequences share 59% identity and 71% similarity). The homology model of WNV MTase was built using the SWISS-MODEL server (45Guex N. Peitsch M.C. Electrophoresis. 1997; 18: 2714-2723Crossref PubMed Scopus (9507) Google Scholar) and stereochemistry was verified using WHATCHECK (46Hooft R.W. Vriend G. Sander C. Abola E.E. Nature. 1996; 381: 272Crossref PubMed Scopus (1798) Google Scholar). The docking of the MTase and the POL1 domain was carried out with the protein-protein docking algorithm FTDOCK (47Katchalski-Katzir E. Shariv I. Eisenstein M. Friesem A.A. Aflalo C. Vakser I.A. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 2195-2199Crossref PubMed Scopus (858) Google Scholar), based on shape complementarities (using a grid spacing of 1 Å) and electrostatic treatment. A search over the complete binding space for both molecules was performed. Results were filtered using the distance restraints between putatively interacting residues as determined by reverse genetic experiments (6 Å filter) and according to the relative positions of the MTase C terminus and POL1 N terminus (45 Å filter for the 13 missing residues), which led to a single solution. Protein Crystallization and Structure Determination—Crystals of the full-length WNV strain Kunjin NS5 protein could not be obtained despite numerous attempts. However, when a N-terminal His6 tag
Referência(s)