HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 2, 1381-1387, January 11, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/2/1381    most recent M109261200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ng, K. K. S. Articles by James, M. N. G. Search for Related Content PubMed PubMed Citation Articles by Ng, K. K. S. Articles by James, M. N. G. Social Bookmarking                      What's this?

Crystal Structures of Active and Inactive Conformations of a Caliciviral RNA-dependent RNA Polymerase^*

Kenneth K. S. Ng^§, Maia M. Cherney, Ana López Vázquez^¶, Ángeles Machín^¶, José M. Martín Alonso^¶, Francisco Parra^¶, and Michael N. G. James

From the  Canadian Institutes for Health Research Group in Protein Structure and Function, Department of Biochemistry, University of Alberta, Edmonton, Alberta T6G 2H7, Canada and the ^¶ Departamento de Bioquímica y Biología Molecular, Instituto Universitario de Biotecnología de Asturias (Consejo Superior de Investigaciones Científicas), Universidad de Oviedo, 33006 Oviedo, Spain

Received for publication, September 25, 2001

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

The structure of the RNA-dependent RNA polymerase (RdRP) from the rabbit hemorrhagic disease virus has been determined by x-ray crystallography to a 2.5-Å resolution. The overall structure resembles a "right hand," as seen before in other polymerases, including the RdRPs of polio virus and hepatitis C virus. Two copies of the polymerase are present in the asymmetric unit of the crystal, revealing active and inactive conformations within the same crystal form. The fingers and palm domains form a relatively rigid unit, but the thumb domain can adopt either "closed" or "open" conformations differing by a rigid body rotation of ~8 degrees. Metal ions bind at different positions in the two conformations and suggest how structural changes may be important to enzymatic function in RdRPs. Comparisons between the structures of the alternate conformational states of rabbit hemorrhagic disease virus RdRP and the structures of RdRPs from hepatitis C virus and polio virus suggest novel structure-function relationships in this medically important class of enzymes.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Rabbit hemorrhagic disease virus (RHDV)¹ belongs to the Caliciviridae family of positive-stranded RNA viruses and causes a highly contagious and lethal disease in rabbits (1, 2). First described in China in the 1980s, RHDV has spread at an alarming rate in the rabbit population throughout Asia, Europe, and Australia. Caliciviruses cause a number of severe diseases in other mammals; in humans, Norwalk viruses are responsible for ~95% of cases of non-bacterial gastroenteritis and represent a growing public health problem in need of new, more effective treatments (3). Caliciviruses are also closely related to the picornaviruses (e.g. polio virus, rhinovirus, and foot-and-mouth disease virus) and to the flaviviruses (e.g. hepatitis C, dengue, and yellow fever viruses), which cause many serious diseases in humans and other mammals.

A virally encoded RNA-dependent RNA polymerase (RdRP) is the central enzyme responsible for replicating the genomic RNA of caliciviruses and other positive-stranded RNA viruses (4). The RHDV RdRP has been produced in Escherichia coli and has been shown to efficiently synthesize RNA from RNA primer-template duplexes in the presence of divalent cations and ribonucleoside triphosphates (5-7). Sequence analysis and site-directed mutagenesis studies of RdRPs from a wide spectrum of positive-stranded RNA viruses (4), as well as the crystal structures of RdRPs from polio virus (PV) (8) and hepatitis C virus (HCV) (9-11), suggest that the members of this large family of enzymes share a common architecture and enzymatic mechanism. Structural and enzymological studies on a wide range of RdRPs are beginning to reveal the essential features of RdRP function, and it is hoped that this body of molecular information will facilitate the design of more effective treatments for viral diseases.

Although the previously determined crystal structures of the PV and HCV RdRPs have shed light on the structural basis of polymerase function, many important questions remain to be addressed. Most importantly, the pair of metal ions seen in the active sites of catalytically active DNA polymerases was not present in any of these structures (the HCV RdRP·Mn²⁺·UTP complex structure has been determined to 2.7 Å, but details are not given (9)). As a result, the details of interactions between RdRPs, RNA primer-template duplexes, nucleoside triphosphates, and metal ions can only be modeled based on the structures of more distantly related DNA polymerases, DNA-dependent RNA polymerases, and reverse transcriptases (12-14). Further work is also needed to elucidate how RdRPs initiate RNA replication through a series of interactions with the RNA genome and accessory proteins (15-17).

As a step toward understanding the structural basis of RNA replication in positive-stranded RNA viruses, we have determined the structure of the RHDV RdRP to a 2.5-Å resolution. The complete structure of the enzyme has been determined in two independent copies in the asymmetric unit, which reveal the polymerase in both active and inactive conformations. Divalent metal ions are bound to the enzyme under physiological ionic strength conditions, and the active site cleft is open to the binding of RNA primer-template duplexes. Comparisons between the RHDV RdRP structure and the previously determined structures of PV and HCV RdRPs indicate that conformational changes similar to those seen in DNA polymerases may be important to the catalytic mechanism of RdRPs.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Protein Expression, Purification, and Crystallization-- Wild-type RHDV RdRP protein was purified from E. coli XL1-Blue, transformed with the expression plasmid pGEX-3D, and grown on Luria-Bertani medium containing ampicillin (50 µg/ml) as described previously (5). Selenomethionine was incorporated into RHDV RdRP using the methionine auxotroph B834(DE3)pLysS grown in M9 minimal media supplemented with 0.3 mM seleno-L-methionine and 1 mg/liter each of riboflavin, niacinamide, pyridoxine-HCl, and thiamin. In all cases the fusion protein from isopropyl-1-thio--D-galactopyranoside-induced cultures was purified by affinity chromatography on glutathione-Sepharose 4B (Amersham Biosciences, Inc.) and digested with thrombin to release the polymerase as described previously (5). The buffers used to purify selenomethionine-RdRP were supplemented with 2 mM -mercapthoethanol. The fractions containing the purified polymerase were pooled and mixed with an equal volume of a saturated solution of ammonium sulfate. The precipitates were redissolved at ~20 mg/ml in 10 mM Tris-Cl, pH 7.5, and dialyzed for 2 h against 10 mM Tris-Cl, pH 7.5. UTP was added to a final concentration of 1 mM.

Protein (~15 mg/ml) and precipitant solution (10-12% PEG 8000, 0.1 M Tris-Cl, pH 7.5, 0.2 M sodium thiocyanate, 0.1 M L-proline, 15% (w/v) glycerol, 7% (v/v) 1,6-hexanediol, 0.1% (w/v) CHAPS, 5 mM CaCl₂, 2 mM MgCl₂) were mixed in equal volumes (2 µl of each) and equilibrated over the precipitant solution as a hanging drop. Large rod- or plate-like crystals (0.5-1.0 × 0.2 × 0.1 mm) used for data collection were grown by both macroseeding and microseeding over the course of 3-5 days. The space group was P2₁2₁2₁. Selenomethionine-containing crystals grown in this manner were used for data collection, multiwavelength anomalous dispersion phasing, and structure determination.

To grow crystals for high resolution data collection, CaCl₂ and MgCl₂ were replaced with 5 mM LuCl₃ (Aldrich) and grown as described above. To increase the occupancy of divalent metal ions at the active site, crystals were also grown using a precipitant solution of 9% PEG 8000, 0.1 M Tris-Cl, pH 7.5, 50 mM Mg(NO₃)₂, 0.1 M 1,7-diaminoheptane, 15% (w/v) glycerol, 7% (v/v) 1,6-hexanediol, and 0.1% (w/v) CHAPS. These crystals were then soaked in a solution containing 12% PEG 8000, 50 mM NaHEPES, pH 7.5, 18 mM MnCl₂, 15% glycerol, and 7% hexanediol for 20 h before being transferred to a solution containing the same components plus 10 mM 3'-deoxy-ATP (Sigma) and then soaked for an additional 40 min. Crystals soaked for longer times did not show improved electron density for either 3'-deoxy-ATP or metal ions.²

Diffraction Data Measurement and Structure Determination-- The structure of RHDV RdRP was determined using the multiwavelength anomalous dispersion technique. Data from a flash-cooled crystal (0.5 × 0.2 × 0.1 mm, quickly adapted to precipitant solution containing 25% glycerol and cooled under nitrogen gas at 100 K) were measured at beamline 9-2 (Stanford Synchrotron Radiation Laboratory) using an ADSC Quantum-4 detector and processed using the HKL suite (18). Local-scaling, Patterson searches and phasing calculations were performed using SOLVE (19). Only 22 of the 26 selenium atoms predicted from the amino acid sequence were located from Patterson or Fourier searches; the missing selenium atoms corresponding to Met-143 and Met-515 were poorly ordered in both copies of the protein in the asymmetric unit and did not give rise to strong peaks in either the anomalous Patterson and Fourier syntheses. Density modification using DM (20) improved the quality of the electron density map (the figure of merit increased from 0.54 to 0.75 (40-2.7 Å)). The electron density map was skeletonized using MAPMAN, and a main chain trace was constructed using O (21). Non-crystallographic symmetry averaging was attempted using DM, but the resulting statistics and electron density maps did not indicate significant improvements. An analysis of the refined structure indicates that the two copies in the asymmetric unit differ substantially (root mean squared difference (r.m.s.d.) of 1.2 Å for 482 C atoms within 3.8 Å of each other following superposition) due to different orientations of the thumb and N-terminal domains relative to the fingers and palm domains. Initial rounds of refinement against a maximum likelihood target function, including experimentally determined phase information, were performed using CNS (22).

A second data set measured from crystals of wild-type protein crystallized in the presence of 5 mM LuCl₃ was measured at the BM-14-C beamline (BioCARS sector, Advanced Photon Source). The structure built from the multiwavelength anomalous dispersion data was further refined against these higher resolution data. After refining temperature factors for individual atoms, the R-factor was 0.23, and the free R-factor was 0.29. REFMAC v.5.0.36 was then used to refine separate anisotropic temperature factor tensors (23) for the N-terminal, fingers, palm, and thumb domains, further lowering the R-factor to 0.22 and the free R-factor to 0.27. Four residues at the N terminus and 15 residues at the C terminus are too poorly ordered in both copies to be modeled. Residues 181-184 in copy A are also too poorly ordered to be modeled. There are no outliers in the Ramachandran plot with 88% of the 998 residues in the favored regions.

Because of apparently low occupancies of metal ions in this data set, crystallization conditions were further modified in an attempt to increase the occupancies of bound metal ions. It was found that removal of KSCN and replacement with either NaNO₃ or Mg(NO₃)₂ increased occupancies (as monitored by the height of the peak corresponding to the divalent metal ion in ( F_o    F_c )_calc and ( F⁺    F )_calc electron density maps). In a further attempt to improve the occupancies of metal ions and possibly to stabilize a catalytically competent state, the crystal was soaked in 18 mM MnCl₂ for 24 h and then soaked in 18 mM MnCl₂ + 10 mM 3'-deoxyadenosine-triphosphate for 45 min. This crystal was flash-frozen at ~100 K, and data were measured using CuK radiation (Rigaku H3R generator, confocal mirrors) and an R-Axis IV++ image plate detector (Rigaku). Refinement and model building was performed on this structure beginning with the fully refined 2.5-Å resolution LuCl₃ cocrystal structure.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Overall Structure and Comparisons with PV and HCV RdRP Structures-- The RHDV RdRP adopts the overall structure of a right hand with domains corresponding to the fingers, palm, and thumb, as seen in the three-dimensional structures of most other polynucleotide polymerases (24) (Figs. 1 and 2). In addition, there is an N-terminal segment bridging the fingers and thumb domains that appears to be unique to the RdRPs (4, 11). The overall architecture is very similar to that of the PV and HCV RdRPs despite sharing less than 20% sequence identity with PV and only 10% sequence identity with HCV (Fig. 2). Superpositions of the PV and HCV structures onto the RHDV structure (accepting C atoms within 3.5 Å of each other as equivalent) allow the superposition of 233 and 223 C atoms with an r.m.s.d. of 1.65 and 1.88 Å, respectively. The crystallographic data are shown in Table I.

View larger version (32K): [in this window] [in a new window]   Fig. 1.   -Carbon trace (stereoscopic view) of RHDV RNA-dependent RNA polymerase with roughly every 20th residue marked by a dot and number. N-terminal region, yellow; fingers domain, red; fingers insert, magenta; N-terminal part of palm domain, green; C-terminal part of palm domain, cyan; thumb domain, blue. All figures were generated using MOLSCRIPT (36), BOBSCRIPT (37), and RASTER3D (38).

View larger version (49K): [in this window] [in a new window]   Fig. 2.   Ribbon diagrams of RNA-dependent RNA polymerases shown from a similar vantage point. A, RHDV; B, PV (PDB code 1RDR) (8); and C, HCV (PDB code 1C2P) (11). Side chains of active site aspartic acid residues (Asp-250 and Asp-354 in RHDV) are drawn as balls and sticks, and Mn2+ ions are drawn as pink spheres. The Ca2+ ion found near the active site in the polio virus structure is drawn as a gray sphere. There were no bound metal ions reported in the HCV RdRP structures.

The N-terminal domain (residues 1-63) consists of two short -strands, one pairing with a -sheet in the fingers domain and one pairing with a -strand in the thumb domain, as well as three long loop segments. This domain forms a bridge between the fingers and thumb domains, which do not interact with each other directly. The N-terminal domain in HCV RdRP (residues 1-49) also forms a bridge between the fingers and thumb domains. However, the conformation of the polypeptide chain, as well as packing interactions with the fingers and thumb domains, differs substantially from what is seen in the RHDV RdRP.

Significant conformational variability in the N-terminal domain (particularly residues 21-24 and 51-64) is seen between the two copies of the RHDV RdRP in the asymmetric unit. These differences in conformation depend on neighboring crystal lattice contacts and likely reflect the intrinsic conformational flexibility of these regions. Conformational flexibility may also explain the apparent disorder in most of the N-terminal domain of the PV RdRP crystal structure (8). Of the ordered residues in the N-terminal domain of PV RdRP, only residues 23-29 are similar to the RHDV RdRP structure (residues 32-38). Residues 12-23 of PV RdRP thread into the active site, thus blocking access of the primer and template RNA into the active site cleft. In contrast, the corresponding segment of RHDV RdRP (residues 21-32) extends away from the active site cleft, leaving room for the binding of RNA (Figs. 2 and 5).

The fingers domain (residues 64-215 and 259-312) consists of eight -helices and a five-stranded -sheet (one strand contributed by the N-terminal domain) referred to previously as the "fingertips" in HIV-1 RT and HCV RdRP (10). Despite sharing virtually undetectable sequence similarity (less than 15% identity), the topology of the fingers domain is very similar to that seen in the HCV RdRP and the ordered parts of the mostly disordered PV RdRP fingers domain. This observation, as well as our own preliminary threading studies, strongly suggests that the fingers domain in a wide range of viral RNA polymerases will share a similar architecture. The -sheet, a long loop (residues 167-193), and an -helix (residues 63-77) form extensive interactions with the N-terminal domain. A feature of the fingers domain unique to RdRP structures is a helix-loop-helix insertion (residues 259-289) that buttresses the outer face of the fingers domain. Three -helices (residues 85-103, 205-215, and 259-272) form a well packed interface with a long, buried -helix (residues 312-335) of the palm domain.

The palm domain (residues 216-258 and 312-384) consists of a three-stranded -sheet (residues 245-250 and 347-360) sandwiched between a pair of -helices (residues 312-335 and 367-378) on one face and a single -helix (residues 230-241) on the other. The architecture of this domain is the most highly conserved feature of all known polymerases (4). An analysis of polymerase sequences and structures has identified six recurring motifs responsible for NTP binding and catalysis. The central three-stranded -sheet consists of a -strand (residues 245-250, Motif A) and a -hairpin (residues 349-360, Motif C) that contains the "GDD" sequence almost universally conserved in the RdRPs (25). Metal ions that are likely to be involved in the nucleotidyl transfer reaction interact with aspartate residues at positions 250, 354, and 355, as discussed in greater detail below. A long loop followed by a pair of -strands (residues 404-417) completes the palm domain and forms an interface with the mostly helical thumb domain. This palm-thumb interface region is similar in all RdRPs and in HIV-1 RT.

The thumb domain (residues 418-501 in RHDV RdRP) comprises four -helices and two long loops linking together the first and second and the third and fourth -helices. This domain has a similar structure to that seen in the PV RdRP and reveals a striking difference to that of the HCV thumb domain (Fig. 2). The long loop or flap (residues 428-446) connecting the first and second helices is folded away from the active site cleft and packs against the N-terminal domain. In contrast, the corresponding parts of the PV (residues 401-410) and HCV (residues 406-412) structures are shorter turns. The most dramatic difference among RdRP structures occurs in the loop (residues 483-489) connecting the third and fourth helices in PV and RHDV. In HCV RdRP, this loop is replaced by a long -hairpin insert (residues 446-462) that occludes the active site cleft. In HCV, replacement of this -hairpin with a short turn allows the enzyme to use double-stranded RNA templates that the wild-type enzyme is unable to use as templates (26). The shorter loops seen in the PV and RHDV enzymes are consistent with the ability of these enzymes to utilize double-stranded RNA as templates (7). The HCV RdRP also differs from both PV and RHDV enzymes by possessing three additional helices preceding a unique C-terminal transmembrane segment.

Conformational Differences between the Two Copies in the Asymmetric Unit-- The two copies of the RHDV polymerase in the asymmetric unit differ substantially from each other. Superimposition of the two copies of the asymmetric unit reveals that the fingers domain and the N-terminal portion of the palm domain form a rigid unit about which the N-terminal and thumb domains, as well as a segment of the palm domain, pivot. Superimposition of the fingers and palm domains (residues 64-415) yields an r.m.s.d. of 0.7 Å over 343 C atoms (only atoms within 3.8 Å of each other are included in the calculation), and superimposition of the thumb domain (residues 415-501) by itself yields an r.m.s.d. of 0.4 Å over 87 C atoms. However, superimposition of the entire polymerase yields an r.m.s.d. of 1.2 Å over 482 C atoms.

The relative positioning of the thumb relative to the palm domain differs substantially between the two copies of the asymmetric unit. Superimposition of the palm domains of both copies shows that the N-terminal region and thumb domain of copy A is rotated toward the palm domain by 8° about an axis roughly lying between the thumb and palm domains (Fig. 3). Analysis of the conformational change using the program DynDom (27) identifies the same dynamic domains and pinpoints residues 63-64 and 417-418 as hinge residues.

View larger version (58K): [in this window] [in a new window]   Fig. 3.   Conformational differences between two copies in the asymmetric unit of RHDV RdRP (Lu3+ cocrystal structure, stereoscopic view). Copy A is colored in red, and copy B is colored in blue. The palm and fingers domains were superposed to emphasize the movement of the thumb and N-terminal domains relative to the central core of the protein.

An analysis of the crystal packing interactions of the two different copies in the asymmetric unit reveals that neighboring molecules in the crystal lattice may stabilize the closed conformation seen in copy A. Crystal packing interactions with the N-terminal domain (residues 5-63) of copy A and the "back" of the palm domain of copy A (residues 246-253 and 381-384) help to push these domains inward and close off the active site cleft relative to the open conformation of copy B.

Interactions with Metal Ions, Nucleotides, and Nucleic Acids-- The RHDV RdRP can be crystallized under conditions of moderate ionic strength and in the presence of divalent cations (Mg²⁺, Mn²⁺, and Fe²⁺) that are required for catalyzing the nucleotidyl transfer reaction (5, 6). The conformations adopted by the two active sites present in the asymmetric unit can vary dramatically depending on the ions present in the mother liquor of the crystals and on the different crystal packing environments surrounding each active site (Fig. 4). In the high resolution (2.5 Å) structure of the crystals grown in the presence of Lu³⁺, one Lu³⁺ ion is bound between the side chain carboxylate groups of Asp-354 and Asp-250 in copy A (Fig. 4A). The electron density and refined temperature factor for the Lu³⁺ ion indicate partial occupancy (~0.2), as does the longer than expected coordination distances (2.5-3.0 Å) between the carboxylate oxygens and the Lu³⁺ ion. Despite the apparently low occupancy, the Lu³⁺ ion contributes significant anomalous scattering that is easily detectable in an anomalous differences Fourier map. On the other hand, in copy B, there is no indication of an ion binding between Asp-354 and Asp-250, but there is a peak between the carboxylate group of Asp-355 and the main chain carbonyl oxygen of Leu-249. This Lu³⁺ ion refines to even lower occupancy (<0.2), and the coordination distances are also longer than expected (Fig. 4B). In the presence of Lu³⁺, copy B of the polymerase adopts an open conformation in which the active site aspartic acid residues are clearly too far apart from each other to coordinate to metal ions in an active configuration. Copy A adopts a more closed conformation in which the active site aspartic acid residues are nearly close enough together to adopt a fully active conformation. Altering the ionic environment of the crystals can increase the occupancies of the Lu³⁺ ions at the enzyme active site, but the quality of diffraction appears to worsen.³

View larger version (117K): [in this window] [in a new window]   Fig. 4.   Active site residues overlaid with Sigma A-weighted 2m|Fo|d | Fc| electron density maps drawn in cyan (contour level 1) and |F+||F| anomalous differences electron density map drawn in pink (contour level 5). Carbon atoms, black; oxygen, red; nitrogen, blue; Lu3+, green; Mn2+, pink. A, Lu3+-bound copy A; B, Lu3+-bound copy B; and C, Mn2+-bound copy A.

In the presence of 18 mM MnCl₂ and 10 mM 3'-deoxy-ATP, the active site of copy A tightens up even more. Two Mn²⁺ ions are clearly bound between the two carboxylate side chains of Asp-354 and Asp-250 and the main chain carbonyl oxygen of Tyr-251 (Fig. 4C). The coordination distances between oxygen atoms and metal ions are reasonable (1.9-2.1 Å), and the occupancies of the metal ions refine to ~0.8. This closed conformation of copy A is probably the active form of the enzyme because the ion coordination closely matches with that seen in active enzyme·NTP·primer-template complexes formed by related DNA-dependent DNA polymerases and the HIV-1 reverse transcriptase (13, 28). In agreement with this hypothesis, previous studies have demonstrated that Asp-354 could not be replaced by any other amino acid residue; even a conservative Asp to Glu substitution completely abolished enzyme activity (5, 6). In addition, point mutant studies directed at Asp-250 also resulted in inactive polymerase proteins when this residue was substituted either by Ala, Asn, or Glu, further supporting the involvement of Asp-250 in ion binding.⁴ Although 3'-deoxy-ATP is present in the crystals, electron density corresponding to the nucleotide is not clearly defined. There is additional electron density near the metal ions, which has been, at present, modeled as water molecules (Fig. 4C), although this most likely results from a mixture of solvent molecules and portions of a poorly ordered nucleoside triphosphate. In the presence of Mn²⁺, the active site of copy B is more open than that of copy A and is similar to the active site conformation seen in copy B of the Lu³⁺-containing crystals.

Although the polymerase active site is buried in the center of the protein, distant and indirect crystal packing forces appear to stabilize the alternate conformational states of the active site seen in the different copies of the asymmetric unit. As described above, crystal packing forces stabilize small rotations of the thumb domain and the back of the palm domain toward the center of the palm in copy A. These domain rotations may help stabilize a conformation that is competent for the binding of catalytic metal ions, NTPs, and RNA substrates. In contrast, crystal packing forces around copy B cause the back strand in the central -sheet of the palm domain (Motif A, residues 245-250) to be displaced slightly relative to the central -hairpin (Motif C, residues 349-360). As a result, Asp-250 adopts a different rotamer and is too far from Asp-354 to coordinate a metal ion. Consequently, the active site in copy B adopts an inactive conformation.

Because the palm domain of the RHDV RdRP is very similar to the palm domains of the DNA-dependent DNA polymerases from T7 phage (29), Thermus aquaticus (Taq) (30), and RB69 phage (31), as well as to the reverse transcriptase from human immunodeficiency virus-1 (HIV-1) (28), it is possible to model the interactions of a nucleotide triphosphate and a primer-template duplex with the RHDV RdRP (Fig. 5). This modeling shows that the active site cleft of the RHDV RdRP (copy A) can accommodate a primer-template duplex without major steric clashes. The positively charged side chains of lysine and arginine residues also appear poised to interact with the phosphodiester backbone of the primer-template duplex (Arg-404 and Lys-114, Lys-134, and Lys-403 near the primer and Arg-434 and Lys-26, Lys-173, and Lys-213 near the template) and the triphosphate moiety of the nucleotide (Arg-177 and Arg-188 and Lys-403). Arg-177 corresponds to Lys-65 in HIV-1 RT, which interacts with the -phosphate of the NTP bound in the ternary complex of that enzyme (28, 32).

View larger version (27K): [in this window] [in a new window]   Fig. 5.   Model of RHDV RdRP bound to NTP and a double helical RNA primer-template based on the ternary complex HIV-1 RT bound to NTP and DNA primer-template (28). The active site residues and metal ions were used to generate the superposition. Fingers domain, red; palm, cyan; thumb, blue. The RNA backbone is drawn in yellow with gray bases. A and B, front and side views, respectively. C, stereo view of the superposition of the central -sheet of the palm domains from copy A of RHDV RdRP (residues 350-358 and 248-253 from the Mn2+·3'-dATP complex; blue) and HIV-1 RT (residues 181-189 and 108-113 from the TTP + DNA duplex complex, yellow) (28). Mn2+ ions from the RHDV structure and the nucleoside triphosphate and a small portion of the primer-template duplex from the HIV structure are also shown.

As shown in Fig. 2, in contrast to the open active site cleft of RHDV RdRP (copy A), the structure of PV RdRP shows that a segment from the N terminus lies in the active site cleft, and the presence of disorder in much of the fingers and N-terminal domains prevents a detailed prediction of RNA substrate binding to PV RdRP (8). In HCV RdRP, two loops from the thumb domain, as well as a segment of the C-terminal polypeptide, sterically occlude the active site cleft from the binding of primer-template duplexes (9-11). These segments of the HCV enzyme must move out of the active site cleft to allow RNA and nucleoside triphosphate to bind to the enzyme (26).

The palm domains of the PV and HCV apoenzyme RdRP structures also do not bind divalent metal cations in a conformation expected for the active holoenzyme. The position of the first -strand of the palm domain (residues 228-234 in PV and 221-227 in HCV) is similar to the slightly displaced position of the corresponding -strand (245) in the open copy B of the RHDV polymerase. As discussed above, this slightly displaced position prevents Asp-250 in the RHDV enzyme (corresponding to Asp-233 in PV and Asp-226 in HCV) from coordinating to the same metal cation as Asp-354 (corresponding to Asp-328 in PV and Asp-324 in HCV), which is presumably required for catalysis to occur. Thus, copy B of the RHDV enzyme, as well as the apoenzyme structures of PV and HCV RdRPs, may represent an open state that does not contain the metal binding configuration expected for catalyzing the addition of a nucleotide to a nucleic acid primer. The repeated observation of an open conformation in three structures of apo forms of RdRPs suggests these enzymes may prefer to adopt a catalytically inactive "ground state," which can be activated to a catalytically competent state by the binding of a nucleic acid primer and template, divalent metal cations, and nucleotide triphosphate. The presence of open and closed conformational states in DNA-dependent DNA polymerases, as well as reverse transcriptases, has been documented previously using crystallography (13), limited proteolysis (33), and electron spin resonance (34).

Additional conformational changes involving small rotations of the thumb and finger domains may be required for the formation of an active ternary complex that would include nucleotide triphosphate and the RNA primer-template duplex. Slight steric clashes occur with the fingers and N-terminal domains when the primer-template duplexes from other polymerase·DNA complexes are docked into RHDV RdRP. Also, the thumb domain may be expected to close down onto the primer-template duplex to a greater extent than that seen even in copy A of the apo-polymerase crystal structure. Experimental structure determination of the RHDV RdRP complex with nucleotide triphosphate and an RNA primer-template duplex will yield detailed information on the conformational changes that accompany different steps of the enzymatic cycle. This information will be of great utility in the design of specific enzyme inhibitors as antiviral drugs. For example, non-nucleoside HIV-RT inhibitors that stabilize the polymerase in an inactive conformation are now routinely used to treat AIDS (35).

	ACKNOWLEDGEMENTS

We thank Michael Carpenter for performing amino acid composition determinations and Ernst Bergmann, Marie Fraser, Brian Mark, William Wolodko, and the staff at the Stanford Synchrotron Radiation Laboratory and BioCARS for assistance with x-ray data collection. Also, we thank M. Soledad Marín and Sonia Morillo for assistance with protein production and purification.

	FOOTNOTES

^* This work was supported by Grants MT-12831 from the Canadian Institutes for Health Research (CIHR) (to M. N. G. J.) and BIO2000-0578-C02-01 from the Spanish Ministerio de Ciencia y Tecnología (to F. P.). Portions of this research were carried out at the Stanford Synchrotron Radiation Laboratory, a national user facility operated by Stanford University on behalf of the U. S. Department of Energy, Office of Basic Energy Sciences. The SSRL Structural Molecular Biology Program is supported by the Department of Energy, Office of Biological and Environmental Research and by the National Center for Research Resources, Biomedical Technology Program and NIGMS, the National Institutes of Health. Use of the Advanced Photon Source was supported by the U. S. Department of Energy, Basic Energy Sciences, Office of Science, under Contract W-31-109-Eng-38. Use of BioCARS Sector 14 was supported by the National Center for Research Resources, National Institutes of Health, under Grant RR07707.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The atomic coordinates and the structure factors (code 1KHV and 1KHW) have been deposited in the Protein Data Bank, Research Collaboratory for Structural Bioinformatics, Rutgers University, New Brunswick, NJ (http://www.rcsb.org/).

^§ Supported by a post-doctoral fellowship from the Canadian Institutes for Health Research and a research allowance supplement from the Alberta Heritage Fund for Medical Research.

To whom correspondence should be addressed: E-mail: Michael.James@ualberta.ca.

Published, JBC Papers in Press, October 24, 2001, DOI 10.1074/jbc.M109261200

² K. K. S. Ng, M. M. Cherney, A. López Vázquez, Á. Machín, J. M. Martín Alonso, F. Parra, and M. N. G. James, unpublished observations.

³ K. K. S. Ng, M. M. Cherney, A. López Vázquez, Á. Machín, J. M. Martín Alonso, F. Parra, and M. N. G. James, unpublished observations.

⁴ J. M. Martín Alonso, M. Norling, and F. Parra, unpublished data.

	ABBREVIATIONS

The abbreviations used are: RHDV, rabbit hemorrhagic disease virus; RdRP, RNA-dependent RNA polymerase; HCV, hepatitis C virus; CHAPS, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonic acid; r.m.s.d., root mean squared difference; PV, polio virus; HIV, human immunodeficiency virus; RT, reverse transcriptase.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				M. Hass, M. Lelke, C. Busch, B. Becker-Ziaja, and S. Gunther Mutational Evidence for a Structural Model of the Lassa Virus RNA Polymerase Domain and Identification of Two Residues, Gly1394 and Asp1395, That Are Critical for Transcription but Not Replication of the Genome J. Virol., October 15, 2008; 82(20): 10207 - 10217. [Abstract] [Full Text] [PDF]

				A. Gruez, B. Selisko, M. Roberts, G. Bricogne, C. Bussetta, I. Jabafi, B. Coutard, A. M. De Palma, J. Neyts, and B. Canard The Crystal Structure of Coxsackievirus B3 RNA-Dependent RNA Polymerase in Complex with Its Protein Primer VPg Confirms the Existence of a Second VPg Binding Site on Picornaviridae Polymerases J. Virol., October 1, 2008; 82(19): 9577 - 9590. [Abstract] [Full Text] [PDF]

				A. Abrahem and M. Pelchat Formation of an RNA polymerase II preinitiation complex on an RNA promoter derived from the hepatitis delta virus RNA genome Nucleic Acids Res., September 1, 2008; 36(16): 5201 - 5211. [Abstract] [Full Text] [PDF]

				P. Roy Bluetongue virus: dissection of the polymerase complex J. Gen. Virol., August 1, 2008; 89(8): 1789 - 1804. [Abstract] [Full Text] [PDF]

				S. Chinnaswamy, I. Yarbrough, S. Palaninathan, C. T. R. Kumar, V. Vijayaraghavan, B. Demeler, S. M. Lemon, J. C. Sacchettini, and C. C. Kao A Locking Mechanism Regulates RNA Synthesis and Host Protein Interaction by the Hepatitis C Virus Polymerase J. Biol. Chem., July 18, 2008; 283(29): 20535 - 20546. [Abstract] [Full Text] [PDF]

				D. F. Zamyatkin, F. Parra, J. M. M. Alonso, D. A. Harki, B. R. Peterson, P. Grochulski, and K. K.-S. Ng Structural Insights into Mechanisms of Catalysis and Inhibition in Norwalk Virus Polymerase J. Biol. Chem., March 21, 2008; 283(12): 7705 - 7712. [Abstract] [Full Text] [PDF]

				T. L. Yap, T. Xu, Y.-L. Chen, H. Malet, M.-P. Egloff, B. Canard, S. G. Vasudevan, and J. Lescar Crystal Structure of the Dengue Virus RNA-Dependent RNA Polymerase Catalytic Domain at 1.85-Angstrom Resolution J. Virol., May 1, 2007; 81(9): 4753 - 4765. [Abstract] [Full Text] [PDF]

				J. Pan, V. N. Vakharia, and Y. J. Tao The structure of a birnavirus polymerase reveals a distinct active site topology PNAS, May 1, 2007; 104(18): 7385 - 7390. [Abstract] [Full Text] [PDF]

				M. Kemp, B. Bae, J. P. Yu, M. Ghosh, M. Leffak, and S. K. Nair Structure and Function of the c-myc DNA-unwinding Element-binding Protein DUE-B J. Biol. Chem., April 6, 2007; 282(14): 10441 - 10448. [Abstract] [Full Text] [PDF]

				H. Malet, M.-P. Egloff, B. Selisko, R. E. Butcher, P. J. Wright, M. Roberts, A. Gruez, G. Sulzenbacher, C. Vonrhein, G. Bricogne, et al. Crystal Structure of the RNA Polymerase Domain of the West Nile Virus Non-structural Protein 5 J. Biol. Chem., April 6, 2007; 282(14): 10678 - 10689. [Abstract] [Full Text] [PDF]

				L. L. Marcotte, A. B. Wass, D. W. Gohara, H. B. Pathak, J. J. Arnold, D. J. Filman, C. E. Cameron, and J. M. Hogle Crystal Structure of Poliovirus 3CD Protein: Virally Encoded Protease and Precursor to the RNA-Dependent RNA Polymerase J. Virol., April 1, 2007; 81(7): 3583 - 3596. [Abstract] [Full Text] [PDF]

				S. W. B. Fullerton, M. Blaschke, B. Coutard, J. Gebhardt, A. Gorbalenya, B. Canard, P. A. Tucker, and J. Rohayem Structural and Functional Characterization of Sapovirus RNA-Dependent RNA Polymerase J. Virol., February 15, 2007; 81(4): 1858 - 1871. [Abstract] [Full Text] [PDF]

				A. Gallei, S. Widauer, H.-J. Thiel, and P. Becher Mutations in the palm region of a plus-strand RNA virus polymerase result in attenuated phenotype J. Gen. Virol., December 1, 2006; 87(12): 3631 - 3636. [Abstract] [Full Text] [PDF]

				S. Le Pogam, H. Kang, S. F. Harris, V. Leveque, A. M. Giannetti, S. Ali, W.-R. Jiang, S. Rajyaguru, G. Tavares, C. Oshiro, et al. Selection and Characterization of Replicon Variants Dually Resistant to Thumb- and Palm-Binding Nonnucleoside Polymerase Inhibitors of the Hepatitis C Virus. J. Virol., June 1, 2006; 80(12): 6146 - 6154. [Abstract] [Full Text] [PDF]

				W. J. Kaiser, Y. Chaudhry, S. V. Sosnovtsev, and I. G. Goodfellow Analysis of protein-protein interactions in the feline calicivirus replication complex J. Gen. Virol., February 1, 2006; 87(2): 363 - 368. [Abstract] [Full Text] [PDF]

				C.T. RANJITH-KUMAR and C.C. KAO Recombinant viral RdRps can initiate RNA synthesis from circular templates RNA, February 1, 2006; 12(2): 303 - 312. [Abstract] [Full Text] [PDF]

				G. Kukolj, G. A. McGibbon, G. McKercher, M. Marquis, S. Lefebvre, L. Thauvette, J. Gauthier, S. Goulet, M.-A. Poupart, and P. L. Beaulieu Binding Site Characterization and Resistance to a Class of Non-nucleoside Inhibitors of the Hepatitis C Virus NS5B Polymerase J. Biol. Chem., November 25, 2005; 280(47): 39260 - 39267. [Abstract] [Full Text] [PDF]

Y.-C. Kim, W. K. Russell, C. T. Ranjith-Kumar, M. Thomson, D. H. Russell, and C. C. Kao Functional Analysis of RNA Binding by the Hepatitis C Virus RNA-dependent RNA Polymerase J. Biol. Chem., November 11, 2005; 280(45):