Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Zhang, B. Articles by Shi, P.-Y. Search for Related Content PubMed PubMed Citation Articles by Zhang, B. Articles by Shi, P.-Y.

 Previous Article  |  Next Article 

Journal of Virology, July 2008, p. 7047-7058, Vol. 82, No. 14
0022-538X/08/$08.00+0     doi:10.1128/JVI.00654-08
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

Genetic Interactions among the West Nile Virus Methyltransferase, the RNA-Dependent RNA Polymerase, and the 5' Stem-Loop of Genomic RNA

Bo Zhang,¹ Hongping Dong,¹ Yangsheng Zhou,² and Pei-Yong Shi^1,2*

Wadsworth Center, New York State Department of Health,¹ Department of Biomedical Sciences, School of Public Health, State University of New York, Albany, New York 12201²

Received 24 March 2008/ Accepted 21 April 2008

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Flavivirus methyltransferase catalyzes both guanine N7 and ribose 2'-OH methylations of the viral RNA cap (GpppA-RNAm⁷GpppAm-RNA). The methyltransferase is physically linked to an RNA-dependent RNA polymerase (RdRp) in the flaviviral NS5 protein. Here, we report genetic interactions of West Nile virus (WNV) methyltransferase with the RdRp and the 5'-terminal stem-loop of viral genomic RNA. Genome-length RNAs, containing amino acid substitutions of D146 (a residue essential for both cap methylations) in the methyltransferase, were transfected into BHK-21 cells. Among the four mutant RNAs (D146L, D146P, D146R, and D146S), only D146S RNA generated viruses in transfected cells. Sequencing of the recovered viruses revealed that, besides the D146S change in the methyltransferase, two classes of compensatory mutations had reproducibly emerged. Class 1 mutations were located in the 5'-terminal stem-loop of the genomic RNA (a G35U substitution or U38 insertion). Class 2 mutations resided in NS5 (K61Q in methyltransferase and W751R in RdRp). Mutagenesis analysis, using a genome-length RNA and a replicon of WNV, demonstrated that the D146S substitution alone was lethal for viral replication; however, the compensatory mutations rescued replication, with the highest rescuing efficiency occurring when both classes of mutations were present. Biochemical analysis showed that a low level of N7 methylation of the D146S methyltransferase is essential for the recovery of adaptive viruses. The methyltransferase K61Q mutation facilitates viral replication through improved N7 methylation activity. The RdRp W751R mutation improves viral replication through an enhanced polymerase activity. Our results have clearly established genetic interactions among flaviviral methyltransferase, RdRp, and the 5' stem-loop of the genomic RNA.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Many members of the genus Flavivirus are important human pathogens, including four serotypes of dengue virus (DENV), yellow fever virus (YFV), West Nile virus (WNV), Japanese encephalitis virus, St. Louis encephalitis virus, and tick-borne encephalitis virus. The flavivirus genome is a plus-sense, single-stranded RNA, 11,000 nucleotides (nt) in length (22). The genomic RNA consists of a 5' untranslated region (UTR), a single open reading frame, and a 3' UTR. The 5' and 3' UTRs are approximately 100 nt and 400 to 700 nt in length, respectively. Both the 5' and 3' ends of the genomic RNA form conserved stem-loop (SL) structures, although the sequences within the SLs are not highly conserved (25). The single open reading frame of the viral genome encodes a polyprotein, which is processed by viral and cellular proteases into three structural proteins (capsid, premembrane or membrane, and envelope) and seven nonstructural proteins (NS1, NS2A, NS2B, NS3, NS4A, NS4B, and NS5). Two viral proteins encode enzyme activities. NS3 functions as a protease (with NS2B as a cofactor), nucleotide triphosphatase, an RNA triphosphatase, and a helicase (13, 20, 37, 39). NS5 acts as a methyltransferase (MTase) and an RNA-dependent RNA polymerase (RdRp) (1, 11, 16, 31, 34). Upon translation, viral nonstructural proteins and host factors form the replication complex to synthesize minus-sense RNA, which, in turn, serves as a template for synthesis of more plus-sense RNA (6).

The 5' end of the flavivirus genome has a type 1 cap structure, m⁷GpppAm (7, 38). Since flaviviruses replicate in the cytoplasm (whereas the host capping apparatus resides in the nucleus), this group of viruses encode their own capping machinery. RNA capping generally requires four steps of enzymatic reactions (15). First, the 5' triphosphate end of the nascent RNA transcript is hydrolyzed to a 5' diphosphate (pppN-RNAppN-RNA) by an RNA triphosphatase. The RNA triphosphatase has been mapped to NS3 in flaviviruses (4, 40). Second, the GMP moiety of GTP is transferred to the 5' diphosphate of RNA (ppN-RNAGpppN-RNA) by an RNA guanylyltransferase. The guanylyltransferase for flavivirus capping has yet to be identified. Third, the N7 position of guanine is methylated (GpppN-RNAm⁷GpppN-RNA) by an RNA guanine-MTase (N7 MTase). Last, the first nucleotide of the RNA transcript is methylated at the ribose 2'-OH position (m⁷GpppN-RNAm⁷GpppNm-RNA) by a nucleoside 2'-O MTase. In flaviviruses, both N7 and 2'-O cap methylations are performed by NS5, using S-adenosyl-L-methionine (SAM) as a methyl donor and generating S-adenosyl-L-homocysteine (SAH) as a by-product (11, 31).

Biochemical and structural analyses of WNV MTase have identified a K61-D146-K182-E218 motif as the active site for 2'-O methylation; the residue D146 of the tetrad was also shown to be essential for N7 methylation (31). Mutagenesis of genome-length RNA of WNV showed that N7 cap methylation, but not 2'-O methylation, is essential for viral replication (10, 45). Compared with host and other viral MTases, flavivirus MTase exhibits several unique features. (i) A single MTase domain catalyzes two distinct methylation reactions (31). Crystal structures of flavivirus MTases reveal only a single binding site for SAM as the methyl donor (3, 11, 26, 45). A mutagenesis study of WNV MTase demonstrated that the enzyme catalyzes two methylations through a molecular repositioning mechanism (10). (ii) Flavivirus MTases perform the two methylations in the order GpppA-RNAm⁷GpppA-RNAm⁷GpppAm-RNA (31, 45). The sequential methylation was shown to be determined by a substrate preference of m⁷GpppA-RNA over GpppA-RNA during the 2'-O methylation reaction (9). (iii) Both flavivirus cap methylations are dependent on viral RNA sequence (8). This characteristic is similar to the nature of the L protein of the minus-sense RNA vesicular stomatitis virus and Sendai virus; the L protein performs capping and methylation in a viral mRNA-dependent manner (28, 29). The specific methylation activity on the viral RNA cap, together with the essential role of the N7 methylation in the viral life cycle, suggests that flavivirus MTase is a potential target for development of antivirals (10).

Capping of cellular mRNA is targeted to the transcript made by RNA polymerase II elongation complex (5, 27, 42). It is not known how—or even whether—the flavivirus RNA cap formation is coupled to viral RNA synthesis. The goal of the present study was to identify potential genetic interactions between WNV MTase and other viral elements. We show that WNV genome-length RNA containing an N7-null MTase is nonreplicative. However, transfection of BHK-21 cells with a mutant RNA, containing an MTase possessing only weak N7 methylation activity, produced revertant viruses bearing second-site mutations. The compensatory mutations are located not only in the MTase domain but also in the RdRp domain and the 5'-terminal SL of the genomic RNA. Mutagenesis analysis demonstrates that these mutations rescue viral replication. In vitro biochemical assays showed that the compensatory mutation recovered in the MTase domain improved N7 methylation activity, while the mutation recovered in the RdRp domain enhanced polymerase activity. These results have provided direct evidence for genetic interactions among the flaviviral MTase, RdRp, and 5' SL of the genomic RNA.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Construction of mutant genome-length cDNA and replicon cDNA of WNV. Mutant genome-length cDNA clones of WNV were prepared by the use of the pFLWNV plasmid (33) and various shuttle vectors. Shuttle vector A was constructed by engineering the BamHI-SphI fragment from pFLWNV (representing the 5' end through nt 3627 of the viral genomic RNA; GenBank no. AF404756) into pACYC177 vector (New England BioLabs). A QuickChange II XL site-directed mutagenesis kit (Stratagene) was used to engineer the G35T substitution or the 38T insertion into shuttle vector A. The mutated DNA fragment was cut and pasted back into pFLWNV with BamHI and ClaI (representing the 5' end through nt 1089 of the genomic RNA). Shuttle vector B was constructed by engineering a KpnI-XbaI fragment (representing nt 5341 through the 3' end of the genome) into pcDNA3.1(+) vector. Mutations within the NS5 gene (K61Q, D146S, D146L, D146P, D146R, and W751R, either singly or in combination) were introduced into vector B by the site-directed mutagenesis kit described above. The mutated DNA fragment was swapped back into pFLWNV using BsiwI (nt position 5780) and XbaI (located immediately downstream of the 3' end of the genomic RNA).

A Renilla luciferase-reporting replicon (Rluc-Rep) of WNV (44) was used to examine the effects of compensatory mutations on viral replication. To construct Rluc-Rep containing a G35U substitution, we first prepared a shuttle vector, pGEM-5'UTR (where UTR is untranslated region), by cloning the fragment spanning SalI (located in the vector of the Rluc-Rep cDNA plasmid) (44) upstream of the 5' end of the replicon cDNA to NsiI (located within the luciferase gene) into vector pGEM-9Zf(–) (Promega). The resulting pGEM-5'NTR was used for a G35U substitution by site-directed mutagenesis as described above. The BamH I-NsiI fragment from the mutated pGEM-5'NTR was cut and pasted into the Rluc-Rep plasmid, resulting in G35U Rluc-Rep. To construct Rluc-Rep containing NS5 mutations, we swapped the BamH I-BsiwI fragment (from the 5' end through nt 5780 of the WNV genome) from the wild-type (WT) Rluc-Rep into the mutant pFLWNV containing the corresponding NS5 mutations. All constructs were verified by DNA sequencing.

RNA transcription and transfection. All plasmid DNA was linearized with XbaI. Genome-length RNA and replicon RNA were transcribed using an mMESSAGE mMACHINE kit (Ambion). All procedures were performed according to the manufacturer's protocols. For transfection, 5 µg of RNA was electroporated into 8 x 10⁶ BHK-21 cells in 0.8 ml of ice-cold phosphate-buffered saline (PBS), pH 7.5, in 0.4-cm cuvettes by using a GenePulser apparatus (Bio-Rad) with settings of 0.85 kV and 25 µF, pulsing three times at 3-s intervals. After a 10-min recovery at room temperature, the RNA-transfected cells were mixed with 25 ml of Dulbecco's modified Eagle's medium containing 10% fetal bovine serum.

Specific infectivity assay and plaque assay. Vero cells (6 x 10⁵ per well) were seeded in six-well plates and incubated for 3 days to allow them to reach confluence. A series of 1:10 dilutions was made by mixing 1 ml of RNA-transfected cell suspension (described above) with 9 ml of Dulbecco's modified Eagle's medium containing 10% fetal bovine serum. One milliliter of each dilution was seeded per individual well of the six-well plate containing the Vero cells. After a 6-h incubation under 5% CO₂ at 37°C, culture medium was aspirated; the cells were overlaid with the first layer of agar. A second layer of agar containing neutral red was added on the indicated days (see Fig. 1D). Plaques were counted after an additional 24 h of incubation. The specific infectivity value was calculated as the number of plaque-forming units (PFU) per microgram of transfected RNA. Plaque assays were performed using a standard protocol (30) with modifications. The modification was required to enable the plaque morphology of the mutant viruses to be compared with that of the WT WNV. Specifically, plates were incubated for 3 days between the addition of the first and the second layers of agar; plaques were recorded at 24 h after the addition of the second layer.

View larger version (29K): [in this window] [in a new window]

FIG. 1. Selection of revertant WNV containing a D146 mutant MTase. (A) Four genome-length RNAs, each coding for a different D146 mutation in the MTase domain. The WNV genomic RNA and NS5 gene are shown. For NS5, individual domains of MTase and RdRp and their amino acid boundaries are labeled. The MTase D146 was mutated to D146S, D146L, D146P, or D146R; the codons of the WT and mutant amino acid are indicated. (B) IFA of BHK-21 cells transfected with genome-length RNAs. BHK-21 cells were transfected with equal amounts of the RNAs and then analyzed for viral protein synthesis by IFA at the indicated time points posttransfection. WNV immune mouse ascites fluid and goat anti-mouse IgG conjugated with Texas Red were used as primary and secondary antibodies, respectively. (C) RT-PCR detection of viral RNA. Supernatants from the RNA-transfected cells were continuously cultured on Vero cells for five passages, with each passage lasting 5 to 7 days, resulting in the P1 to P5 viruses. To detect virus production, we extracted RNAs from P5 supernatant and cells and amplified the first 684 nt of the genomic RNA by RT-PCR. The products were analyzed on a 1.2% agarose gel. Molecular masses of the DNA ladder are indicated on the left of the panel. (D) Plaque morphology of WT WNV, D146S P1 virus, and D146S P5 virus. Plaque assays were performed under identical conditions for the WT and mutant viruses. To clearly show the plaques of the mutant viruses, we incubated the assay plates for 3 days between the additions of the first and second layers of agar. Plaques were documented 24 h after the addition of the second layer of agar.

Indirect immunofluorescence assay (IFA). BHK-21 cells transfected with RNA were seeded on a chamber slide (Nalge Nunc). At the indicated time points (see Fig. 1B and 4B), cells were fixed in –20°C 5% acetic acid in methanol at room temperature for 20 min, washed with PBS, and incubated with WNV immune mouse ascites fluid (1:100 dilution; ATCC) at room temperature for 1 h. After three washes in PBS, the cells were incubated with goat anti-mouse immunoglobulin G (IgG) conjugated with Texas Red at room temperature for 1 h. The slide was washed three times with PBS, mounted with 90% glycerol, and analyzed under a fluorescence microscope (Nikon).

View larger version (47K): [in this window] [in a new window]

FIG. 4. Functional analysis of the NS5 mutations and the 5' SL mutations of the genomic RNA. (A) Specific infectivity assay. Genome-length RNAs containing the indicated mutation(s) were compared for their SIVs. The number of days required for detection of plaques after the addition of the second layer of agar (during the SIV assay) was tabulated for each mutant RNA. (B) IFA was performed on BHK-21 cells transfected with genome-length RNAs at the indicated time points. WNV immune mouse ascites fluid and Texas Red-conjugated goat anti-mouse IgG were used as primary and secondary antibodies, respectively. (C) Plaque morphology of WT and mutant viruses. For direct comparison, plaque assays were performed side by side for all viruses under identical conditions. The plaque assay procedure is described in the legend of Fig. 1.

RNA extraction and RT-PCR. Viral RNA was extracted from virions in the culture medium or extracted from infected cells using RNeasy kits (Qiagen). Reverse transcriptase PCR (RT-PCR) was performed using 5 µl of extracted RNA and a primer pair (forward, 5'-AGTAGTTCGCCTGTGTGAGCTGAC-3'; and reverse, 5'-TGCTGACTTTGTGCACCAACA-3') to amplify the first 684 nt of the WNV genomic RNA. A OneStep RT-PCR kit (Invitrogen) was used for the amplification. The RT-PCR products were sequenced.

Luciferase assay. Various amounts of replicon-transfected cells were seeded into a 12-well plate to avoid overgrowth of cells at the time of the luciferase assay. Specifically, 1, 1, 0.5, 0.25, and 0.25 ml of transfected cells were seeded, and cells were assayed for luciferase activities at 4, 24, 48, 72 and 120 h posttransfection (p.t.), respectively. Triplicate wells were seeded for each data point. Luciferase assays were performed using a Renilla luciferase assay system (Promega) following the manufacturer's protocol. At each time point, cells were washed once with PBS and lysed with 250 µl of lysis buffer at room temperature for 25 min. After 20 µl of cell lysate was mixed with 100 µl of assay reagent, luciferase signals were immediately measured with a Veritas Microplate Luminometer.

Recombinant proteins of WNV MTase domain and full-length NS5. Expression plasmids pET28(a) containing the WT WNV MTase sequence (representing the N-terminal 300 amino acids of NS5) or containing the full-length NS5 sequence were constructed as described previously (31). A QuickChangeII XL site-directed mutagenesis kit was used to mutate specific residues. Mutant plasmids were verified by DNA sequencing. All MTase and NS5 proteins contained an N-terminal His tag, expressed in E. coli Rosetta strain [BL21(DE3)/(pLysS)] cells (Novagen) and purified through nickel columns, as reported previously (31).

Methylation assay. Details of the methylation assays were described previously (10). Briefly, the 5'-end-labeled substrates, G*pppA-RNA and m⁷G*pppA-RNA (the asterisk indicates that the following phosphate is ³²P labeled), representing the first 190 nt of the WNV genome, were prepared using a vaccinia virus capping enzyme, according to the manufacturer's protocol (Epicenter). N7 methylation activity was measured through conversion of G*pppA-RNAm⁷G*pppA-RNA in a pH 7.0 buffer. At pH 7.0, no 2'-O methylation can occur, whereas N7 methylation is fully active (45). The 2'-O methylation activity was quantified through conversion of m⁷G*pppA-RNAm⁷G*pppAm-RNA in a pH 10 buffer. At pH 10, 2'-O methylation is fully active (45). The methylation reactions were digested with nuclease P1 to release cap structures (GpppA, m⁷GpppA, and m⁷GpppAm) and were then analyzed on polyethyleneimine cellulose thin-layer chromatography (TLC) plates. The spots on TLC plates, representing the various cap structures, were quantified using a PhosphorImager.

RdRp assay. An RNA containing the 5'-terminal 269 nt that are directly connected to the 3'-terminal 622 nt of the WNV genome was used as a template for an in vitro polymerase assay. A cDNA fragment, representing BamHI-T7 promoter-the RNA template sequence-XbaI, was cloned into pUC19 plasmid through the BamHI and XbaI sites, resulting in plasmid pUC-TMP. A mutation corresponding to the G35U change in the 5' SL of the WNV genome was engineered into the WT pUC-TMP though a QuickChange II XL site-directed mutagenesis kit. For preparation of the RNA template, WT and mutant pUC-TMP plasmids were linearized with XbaI. The linearized DNA was used for transcription via a MEGAscript T7 kit (Ambion), following the manufacturer's instructions. The RdRp assay was performed in a 25-µl reaction mixture containing 50 mM Tris, 20 mM NaCl, 5 mM MgCl₂, 0.5 mM ATP, 0.5 mM GTP, 0.5 mM CTP, 5 µM UTP, 15 µCi of [-P³²]UTP (10 µCi/µl, 3000 Ci/mmol; Perkin Elmer Life Sciences), 3 µg of NS5, and 1 µg of RNA template. After incubation at 33°C for 2 h, the reaction mixtures were passed through a MicroSpin G-25 column (GE Healthcare) to remove unincorporated nucleoside triphosphates. The eluate was mixed with an equal volume of denaturing Gel Loading Buffer II (Ambion). The RNA products were then analyzed on a 6% polyacrylamide gel with 7 M urea. The gels were dried. The ³²P-labeled products were analyzed with a PhosphorImager.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Genome-length RNA containing a D146S MTase can generate revertant WNV. Previous studies showed that a D146A mutation in WNV MTase is lethal for both the N7 and 2'-O methylations, as well as for viral replication (45). The results prompted us to perform a genetic selection to search for potential interactions between the MTase domain and other viral elements. We asked whether genome-length RNA, containing a D146 substitution, can generate infectious viruses, and, if it can, what compensatory mutations have accumulated in the recovered viruses. One potential problem in such selection is true reversion, in which the mutated amino acid changes back to the WT D146 after selection. To overcome this problem, we mutated the WT D146 codon GAC to each of UCA, CUG, CCA, and AGA, resulting in D146S, D146L, D146P, and D146R genome-length RNAs, respectively (Fig. 1A). The mutant codons were selected to minimize true reversion, which requires that all three nucleotides of each codon be changed.

Equal amounts of WT and mutant RNAs were transfected into BHK-21 cells and assayed for their levels of replication. An IFA of the transfected cells showed that, other than WT RNA, only the D146S RNA expressed enough viral proteins to induce IFA-positive cells (Fig. 1B). The D146S RNA-mediated IFA-positive cells appeared 4 days later than the WT RNA (day 5 versus day 1 p.t. for D146S versus WT RNA, respectively). No IFA-positive cells were observed in cells transfected with the D146L, D146P, or D146R RNAs. Except for WT RNA (Fig. 1D), no plaques were recovered when culture fluids from the mutant RNA-transfected cells were assayed (data not shown). However, after the supernatant was passaged once in Vero cells for 5 days, the D146S RNA yielded viruses (designated P1 viruses for first passage), as evidenced by the appearance of tiny plaques. After D146S P1 virus was passaged in Vero cells for another four rounds (with each passage lasting 5 to 7 days), the resulting P5 virus yielded plaques larger than those produced by the P1 virus, but these P5 plaques were still smaller than the WT virus plaques. No plaques were observed for the D146L, D146P, or D146R RNA-transfected cells even after five rounds of passaging in Vero cells (data not show). To exclude the possibility that non-plaque-forming viruses were produced by the D146L, D146P, and D146R RNAs, we performed RT-PCR using RNAs extracted from P5 culture fluids and cells. Only WT and D146S RNAs yielded virus-specific RT-PCR products (Fig. 1C). Analysis of the RNA samples, using real-time RT-PCR with a detection sensitivity of 40 RNA molecules (32) also showed that only WT and D146S RNAs yielded virus-specific RT-PCR products (data not shown). The results demonstrate that the D146S mutation of MTase (but not the D146L, the D146P, or the D146R mutation) is capable of producing virus.

Recombinant D146S MTase retains a low level of N7 methylation activity. To examine whether the capability of the D146S RNA to generate virus correlates with methylation activity, we prepared recombinant proteins of the MTase domain containing the D146S, D146L, D146P, or D146R mutation (Fig. 2A). One additional mutant containing a D146T substitution was prepared, because this amino acid change is similar to the D146S mutation. Equal amounts of WT and mutant MTase were assayed for their methylation activities. As shown in Fig. 2B, a conversion of G*pppA-RNAm⁷G*pppA-RNA (the RNA represents the first 190 nt of the WNV genome) was used to indicate N7 methylation activity. Only the D146S and D146T MTases showed very low levels of N7 methylation activity, i.e., approximately 1 to 2% of the WT activity. Next, a conversion of m⁷G*pppA-RNAm⁷G*pppAm-RNA was used for detection of the 2'-O methylation (Fig. 2C). None of the mutant MTases showed any 2'-O methylation activity. The results suggest that the ability of the D146S genome-length RNA to generate virus correlates with a low level of N7 methylation activity.

View larger version (61K): [in this window] [in a new window]

FIG. 2. Methylation activities of the D146 mutant MTase. (A) Recombinant proteins of WT and mutant MTases. Proteins were analyzed by sodium dodecyl sulfate-PAGE and stained with Coomassie blue. Molecular masses of the protein markers are labeled on the left of the gel. (B) N7 methylation activity. N7 methylation activity was measured by conversion of G*pppA-RNAm7G*pppA-RNA in a pH 7.0 buffer; pH 7.0 supports only N7 methylation and not 2'-O methylation (45). The methylation reactions were digested with nuclease P1. The resulting cap structures were resolved on a TLC plate. The relative activity of N7 methylation is shown at the bottom of the panel, with the WT activity set at 100%. (C) 2'-O methylation activity. The 2'-O methylation activity was measured by conversion of m7G*pppA-RNAm7G*pppAm-RNA in a pH 10 buffer; pH 10 is optimal for 2'-O methylation (45). The relative activity of 2'-O methylation in comparison with the WT activity (set at 100%) is shown at the bottom of the panel. The positions of the origins and the migrations of the G*pppA, m7G*pppA, and m7G*pppAm molecules are labeled to the left side of the TLC images.

Compensatory mutations accumulate in the D146S RNA-derived viruses. As described above, the detection of D146S RNA-derived viruses required an extra round of culturing on Vero cells, and the plaque morphology changed between the P1 and P5 viruses (Fig. 1D). These results indicated that adaptations had occurred in the recovered viruses. Complete genome sequencing of the P5 virus (population sequencing without plaque purification) showed that, beyond the engineered D146S change, four other mutations had accumulated (Fig. 3A, selection I). The same compensatory mutations were recovered from viruses that were plaque purified from the P5 viruses (data not shown), indicating that the isolate containing the identified mutations was dominant in the selected viral population. To examine the reproducibility of these compensatory mutations, we performed three more independent selections (i.e., different experiments for virus recovery and passaging) (Fig. 3A, selections II to IV). Viruses derived from the later selections revealed similar, but not identical, mutations (Fig. 3A). The mutations from four selections were located in three regions. (i) A G-to-U substitution at nucleotide position 35 (G35U) or a U insertion at position 38 (U38-ins) was uncovered in the 5'-terminal SL of the genome (Fig. 3B). (ii) A P-to-L change at amino acid position 138 (P138L) of the NS1 protein was found only in selection I. (iii) All selections resulted in changes in the NS5 gene. Besides the input D146S mutation, a K61Q or K61T change was identified in the MTase domain, while a W751R change was recovered in the RdRp domain. Notably, both the K61 and D146 residues are within the MTase-conserved K-D-K-E motif and are located next to the SAM-binding site on the WNV MTase (Fig. 3C). In the crystal structure of WNV RdRp (24), the aromatic ring of W751 is exposed on the surface of the molecule and is located at the opening of the RNA template channel, which is formed between the finger and thumb domains (Fig. 3D). Sequence alignment showed that W751 is conserved among the WNV, DENV, YFV, and tick-borne encephalitis virus NS5 proteins (data not shown).

View larger version (39K): [in this window] [in a new window]

FIG. 3. Compensatory mutations from revertant viruses. (A) Compensatory mutations identified from four independently selected revertant viruses. Genome-length RNAs containing the D146S MTase were transfected into BHK-21 cells. Supernatants of the transfected cells were passaged on naïve Vero cells for five rounds. Viruses from each round of passaging were monitored for changes in plaque morphology. P5 viruses from each selection were subjected to complete genome sequencing. The locations of recovered mutations are tabulated. (B) Mutations recovered in the 5' SL of the genomic RNA. One mutation, G35U or 38U-ins, was recovered in the 5' SL of the genomic RNA. The RNA secondary structure was predicted by the Mfold program. The initiation codon AUG of the open reading frame is shaded in gray. (C) Crystal structure of WNV MTase showing residues K61 and D146 (both in yellow). The SAH molecule (red) is also shown on this ribbon representation of the WNV MTase structure (Protein Data Bank entry 2OY0) (45). (C) Crystal structure of WNV RdRp showing residue W751 (red). The thumb, fingers, and palm domains of the RdRp are shown in red, blue, and yellow, respectively (Protein Data Bank entry 2HFZ) (24). The RNA template channel, formed between the fingers and thumb domains, is indicated by dotted gray lines. The RNA output tunnel is shown by dotted green lines. The illustrations shown in panels C and D were prepared using PyMOl.

Mutations emerge in a specific order: W751R in RdRp, K61Q or K61T in MTase, and 5' SL G35U or 38U-ins in the genomic RNA. To determine the order of emergence of the above recovered mutations, we sequenced the viruses collected at various passages during the selection. The sequencing results showed that the RdRp W751R appeared first (in P1 or P2 virus), the MTase K61Q or K61T emerged second (in P3 virus), and the 5' SL G35U or 38U-ins in the genomic RNA appeared last (in P4 or P5 virus). NS1 P138L was first recovered in P2 virus from selection I.

Mutations in the 5' SL of the genomic RNA and mutations in the NS5 gene can rescue viral replication of the D146S RNA. The recovered mutations were engineered into the genome-length RNA of WNV to determine their biological functions (Fig. 4A). We initially focused on the common mutations derived from the four independent selections: mutations within the 5' SL of the genome and mutations within the NS5 gene. Genome-length RNAs were prepared to contain a single mutation (G35U, K61Q, D146S, or W751R), a double mutation (K61Q+D146S or D146S+W751R), a triple mutation (K61Q+D146S+W751R, G35U+K61Q+D146S, or 38U-ins K61Q+D146S), or a quadruple mutation (G35U+K61Q+D146S+W751R). Three parameters were used to monitor the replication efficiencies of the mutant RNAs. First, we compared the specific infectivity values (SIVs) of the genome-length RNAs (Fig. 4A). Equal amounts of RNA were transfected into BHK-21 cells for each mutant and were quantified for PFU at various time points after the addition of the second layer of agar (see Materials and Methods for details). WT, G35U, and W751R RNAs yielded plaques on day 1 after addition of the second layer of agar; the resulting SIVs were 8.4 x 10³, 3.5 x 10³, and 3.0 x 10³ PFU per µg of transfected RNA, respectively. The results indicate that the G35U or W751R mutation alone does not substantially affect viral replication. In contrast, the single mutation K61Q RNA generated plaques on day 3 after the addition of the second layer of agar, with an SIV of 7.5 x 10² PFU/µg RNA; sequencing of the viruses recovered on day 5 p.t. showed that the mutated K61Q (AAACAA; the mutated nucleotide is underlined) had reverted back to the WT amino acid K61. The D146S RNA did not produce any plaques, even after a prolonged incubation (up to 6 days after the addition of the second layer of agar). The doubly mutated D146S+W751R RNA did not yield any plaques, while K61Q D146S RNA produced plaques on day 4 after the addition of the second layer of agar, with a low SIV of 1.5 x 10¹ PFU/µg RNA. Each of three triple mutation RNAs improved the SIV to 4.5 x 10³ to 5.5 x 10³ PFU/µg RNA, but the resulting viruses still required 3 days of incubation after the addition of the second layer of agar to produce plaques. The quadruple mutation RNA further improved its replication efficiency: the time for plaque formation after the addition of the second layer of agar was reduced, with plaques appearing on day 2.

Second, we performed an IFA to compare viral protein expression of the mutant RNAs in transfected cells. Figure 4B shows the IFA results for a panel of representative mutant RNAs. The G35U+K61Q+D146S+W751R, K61Q+D146S+W751R, and K61Q+D146S RNA mutants yielded IFA-positive cells on day 1, day 2, and day 3 p.t., respectively; in contrast, no IFA-positive cells were detected upon transfection of the D146S RNA up to day 4 p.t. (Fig. 1B and 4B). Third, plaque morphologies were compared among viruses recovered from the various mutant RNAs (Fig. 4C). Although the compensatory mutations improved the SIV and viral protein expression of the D146S RNA, mutant viruses derived from the quadruple- and triple-mutation RNAs exhibited smaller plaques than the WT virus. Overall, the results indicate that compensatory mutations in the 5' SL of the genome and the mutation in the NS5 gene can rescue the replication of D146S RNA, with the highest rescuing efficiency occurring when all of these mutations are present.

The compensatory mutation in NS1 does not contribute to rescue of the replication of D146S RNA. To test the function of mutation P138L in NS1 recovered from selection I (Fig. 3A), we prepared two pairs of genome-length RNAs and compared their replication efficiencies (Fig. 5). The first pair was built on a double-mutation RNA in the absence or presence of the NS1 mutation, i.e., NS5 K61Q+D146S alone and NS1 P138L paired with NS5 K61Q+D146S. The second pair was built on a triple-mutation RNA, i.e., the 5' SL G35U with NS5 K61Q+D146S and the 5'SL G35U with NS1 P138L and NS5 K61Q+D146S. Similar SIVs, IFA results, and plaque morphologies were obtained for the RNAs with and without the NS1 mutation (Fig. 5). These results suggest that the P138L mutation in NS1 does not contribute to the restoration of D146S RNA replication.

View larger version (31K): [in this window] [in a new window]

FIG. 5. Analysis of the NS1 P138L mutation using genome-length RNA of WNV. Two pairs of mutant RNAs (K61Q+D146S and G35U+K61Q+D146S) were analyzed for determination of the function of the NS1 P138L mutation. Each pair of RNAs was compared for its SIV (A), IFA result (B), and plaque morphology (C) in the presence or absence of the NS1 mutation. The experiments were performed in the same manner as those described in the legend of Fig. 4.

Analysis of compensatory mutations in WNV replicon. To directly measure the effect of various mutations on improvement of D146S RNA replication, we tested representative mutations in a transient replicon system of WNV. The replicon contained a Renilla luciferase sequence and a foot-and-mouth disease virus 2A sequence that were engineered to replace the structural genes of the viral open reading frame (Fig. 6A, Rluc-Rep). Previously, studies showed that transfection of BHK-21 cells with Rluc-Rep RNA generates two distinct luciferase peaks. The first peak, at 2 to 8 h p.t., represents input RNA translation; the second peak, after 12 h p.t., represents RNA synthesis (i.e., the second luciferase peak as a surrogate readout of RNA replication) (23). Using this system, we transfected BHK-21 cells with equal amounts of WT and mutant replicon and then assayed their luciferase activities at various time points posttransfection (Fig. 6B). All replicons yielded roughly equal levels of luciferase signal at 2 h p.t., indicating that the transfection efficiency was similar for the various replicon RNAs. For the D146S replicon, a background level of luciferase signal was detected at 48 to 120 h p.t. (Fig. 6B); the luciferase kinetics of the D146S replicon was similar to that of a replicon containing a deletion of the GDD active site of RdRp (data not shown). The results suggested that the D146S replicon was nonreplicative. In contrast, the 5' SL G35U replicon replicated to a level similar to that of the WT replicon. Addition of an individual mutation, K61Q in the MTase or W751R in the RdRp, to the D146S replicon did not rescue replicon RNA replication. However, addition of both K61Q and W751R mutations to the D146S replicon did restore replicon RNA replication. The replication efficiency of the triply mutated K61Q+D146S+W171R replicon was further improved by addition of the 5' SL G35U mutation, but the quadruple-mutation replicon, G35U+K61Q+D146S+W171R, still replicated more slowly than the WT replicon. Overall, the replicon results demonstrate that the compensatory mutations can restore the D146S RNA replication in an additive manner.

View larger version (33K): [in this window] [in a new window]

FIG. 6. Replicon analysis of compensatory mutations. (A) A luciferase-reporting replicon of WNV. A Renilla luciferase-foot-and-mouth-disease virus 2A sequence was engineered to replace the capsid, premembrane, and envelope structural genes of the viral genomic RNA, resulting in Rluc-Rep. (B) Transient replication of mutant Rluc-Rep. Equal amounts of WT and mutant Rluc-Rep RNAs were transfected into BHK-21 cells and assayed for luciferase activity at the indicated time points posttransfection.

Mutations K61Q and K61T in the MTase domain improve N7 methylation. Since the D146S MTase retained only 1 to 2% of WT N7 methylation activity and completely abolished 2'-O methylation (Fig. 2), we reasoned that one potential means to restore viral replication is to improve cap methylation through compensatory mutations. To test this possibility, we prepared a recombinant MTase domain containing the recovered mutations and assayed their effects on methylation activities (Fig. 7A). Besides the mutation K61Q (recovered from three selections), we also analyzed the K61T mutation (recovered from one selection) and the K61S mutation (similar to the recovered K61T). For N7 methylation, MTases containing the single mutations K61Q, K61S, and K61T retained 20%, 68%, and 55% of the WT MTase activity, respectively. Addition of the individual mutation at K61 into the D146S MTase caused an increase in the N7 methylation activity from 1% to 4 to 8% (Fig. 7A, left panel). Since NS5 is expected to interact with the 5'-terminal nucleotide during cap methylations, we tested whether the compensatory mutations G35U and U38-ins in the 5' SL of the genomic RNA could improve cap methylation. A similar level of improvement of the N7 methylation was observed for the panel of MTases when assayed on the WT, G35U (middle panel), and U38-ins (right panel) RNA substrates. No 2'-O methylation activity was detected for any mutant MTase when either WT or mutant (G35U or U38-ins) RNA substrate was used (data not shown). Collectively, the results indicate that the MTase mutations K61Q and K61T facilitate replication of D146S RNA through an improved N7 methylation activity, whereas mutations G35U and U38-ins in the 5' SL of the genomic RNA do not increase cap methylations.

View larger version (94K): [in this window] [in a new window]

FIG. 7. Improvement of N7 methylation by MTase K61Q and K61T mutations. (A) N7 methylation of the MTase domain of NS5. Equal amounts of MTases containing the indicated mutations were compared for their N7 methylation efficiencies (G*pppA-RNAm7G*pppA-RNA). To examine whether G35U or U38-ins affects methylation, we compared three RNA substrates (WT, G35U, and U38-ins) for their methylation activities. The reaction mixtures were digested with nuclease P1 to release m7G*pppA (product) or G*pppA (input substrate) and then analyzed on a TLC plate. The methylation efficiencies of mutant MTases were compared with that of the WT MTase (set at 100%), and values are indicated below the TLC results. (B) Analysis of N7 and 2'-O methylations of full-length NS5. Equal amounts of NS5 containing the indicated mutations were assayed for N7 methylation (G*pppA-RNAm7G*pppA-RNA; left two panels) and 2'-O methylation (m7G*pppA-RNAm7G*pppAm-RNA; right two panels). WT RNA and a G35U RNA were compared for their methylation efficiencies. The positions of the origins and the migrations of the G*pppA, m7G*pppA, and m7G*pppAm molecules are labeled to the left side of the TLC images.

Next, we examined whether the presence of the WT RdRp domain or its W751R mutation could affect the methylation activities. Recombinant proteins of full-length NS5 containing single-, double-, or triple-mutation combinations of K61Q, D146S, and W751R were prepared (Fig. 8A) and were assayed for their methylation activities (Fig. 7B). Although NS5 with the K61Q or W751R single mutation had, respectively, 10% or 97% of the WT N7 methylation activity, addition of the K61Q mutation, but not the W751R mutation, to the D146S NS5 increased the N7 activity from 1% to 4% of the WT activity (Fig. 7B, upper left panel). Incorporation of the W751R mutation into the K61Q+D146S NS5 did not further improve N7 methylation activity. For 2'-O methylation, all NS5 mutants other than the W751R NS5 were completely defective (Fig. 7B, upper right panel). Similar methylation results were obtained when the NS5 mutants were tested on the G35U RNA substrate (bottom panels). The results demonstrate that mutation W751R in the RdRp domain does not affect cap methylation activity in vitro.

View larger version (43K): [in this window] [in a new window]

FIG. 8. Enhancement of polymerase activity by the RdRp W751R mutation. (A) Recombinant proteins of full-length NS5. WT and mutant NS5 proteins (1.5 µg) were analyzed by sodium dodecyl sulfate-PAGE and stained with Coomassie blue. Molecular masses of protein markers are labeled. (B) In vitro polymerase activity of NS5. An RNA containing the 5' 269 nt directly connected to the 3' 622 nt of the WNV genome was used as a template for the RdRp assay. The 32P-labeled products were resolved on a denaturing PAGE gel. Two RNA templates, WT and G35U, were assayed for determination of whether G35U affects the polymerase activity. The two forms of RdRp products (input length [1x] and double the length of the input [2x]) are indicated. For each mutant, the sum of the two products was quantified using a PhosphorImager; the relative RdRp activity was compared with that of the WT NS5 (set at 1.0), and this result is presented below the autoradiograph of the PAGE gel. The order of the protein lanes used in the RdRp assay is the same as shown in panel A. (C) Summary of the RdRp results from panel B.

Mutation W751R in the RdRp domain improves polymerase activity. We directly tested whether the recovered mutations rescue viral replication through enhancement of polymerase activity. Full-length NS5 proteins containing various mutations in MTase, RdRp, or both domains (Fig. 8A) were compared with respect to in vitro polymerase activity. A synthetic RNA containing the 5'-terminal 269 nt directly connected to the 3'-terminal 622 nt of the WNV genomic RNA was used as a template for the RdRp assay. This RNA template contained the RNA elements (5' SL, 3' SL, 5' UAR/3' UAR base pairing, and 5' CS/3' CSI base pairing) that are known to be important for RdRp activity (2, 14, 18, 19, 23, 35, 36, 41, 43, 44). Analysis of the RdRp reactions by denaturing polyacrylamide gel electrophoresis (PAGE) showed two major RNA products (Fig. 8B). As previously reported (41), a product the same length as the input RNA template was derived from de novo RNA synthesis, and a hairpin product double the length of the input template was derived from copy-back RNA synthesis. Because of its double-stranded nature, the double-length RNA product was resistant to complete denaturation during PAGE analysis, and, thus, it migrated faster than the input-length product. Remarkably, NS5 variants containing the W751R mutation (variants W715R, D146S+W175R, and K61Q+D146S+W175R) generated 5- to 7.8-fold more products than the WT NS5; addition of the MTase K61Q mutation did not further enhance the polymerase activity (Fig. 8B, top panel). In contrast, NS5 variants containing MTase mutations (variants K61Q, D146S, and K61+D146Q) showed polymerase activities similar to the activity of the WT. These results demonstrate that mutation W751R in the RdRp domain improves polymerase activity.

We also examined whether the 5' SL mutation G35U affects the polymerase activity. For each NS5 variant, the G35U RNA template reproducibly generated amounts of RNA products (for both the input-length and the double-length forms) comparable to the amount generated by the WT RNA template (Fig. 8B, lower panel). A similar W751R-mediated enhancement of polymerase activity was observed when the G35U RNA template was used (Fig. 8C). These results suggest that mutation G35U within the 5' SL of the genomic RNA does not contribute to an improvement in RdRp activity in vitro.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
A productive flaviviral infection requires delicately modulated RNA-RNA, RNA-protein, and protein-protein interactions. The current study reports genetic interactions among flaviviral MTase, RdRp, and the 5'-terminal SL of the genomic RNA. Using WNV as a model, we show that the nonreplicative phenotype of a genome-length RNA containing a methylation-defective mutation (D146S in MTase) can be rescued by adaptations in MTase, RdRp, and the 5' SL of viral genomic RNA. The effects of individual adaptations were additive; the restoration of viral replication is most efficient when all of these adaptations were present in the D146S viral RNA. Biochemical analysis showed that compensatory mutations in the MTase domain (K61Q or K61T) improved N7 methylation activity, while a mutation in the RdRp domain (W751R) enhanced polymerase activity. The genetic and biochemical results demonstrate that the compensatory mutations restore viral replication through improvement of N7 cap methylation and polymerase activity.

Genetic selection is a powerful means to identify potential intermolecular and intramolecular interactions during viral replication. For flaviviruses, this approach was successfully used to establish a genetic interaction between NS1 with NS4a as a determinant for YFV replicase function (21). A similar approach was recently used to probe potential contact sites between the MTase and RdRp domains of DENV-2 NS5 (24). A DENV-2 genome-length RNA containing mutations in a loop of the MTase domain (K46A+R47A+E49A) was used to select for compensatory mutations in the RdRp domain. An L512V adaptation in a loop of the RdRp domain was required for viral replication, suggesting that the MTase domain and the RdRp domain may interact through the K46-R47-E49 loop and the L512 loop, respectively, (24). The feasibility of the current genetic selection was built on our previous results that (i) D146 of the WNV MTase domain is essential for both N7 and 2'-O cap methylations (31) and (ii) N7 methylation is critical for viral replication (45). We previously showed that a WNV RNA containing a single-nucleotide-mediated D146A change [GAC (Asp) GCC (Ala)] immediately reverted to the WT D146 through a 1-nt change [i.e., GCC (Ala) GAC (Asp)], resulting in the WT virus (mutated residues are underlined in the descriptions of the reversions). When a genome-length RNA containing a double-nucleotide-mediated D146A change [GAC (Asp) GCA (Ala)] was transfected into cells, the recovered virus initially contained a 1-nt reversion [GCA (Ala) GAA (Glu)], resulting in a D146E virus. Note that Glu and Asp are both acidic residues. Next, the D146E virus underwent a second nucleotide reversion [GAA (Glu) GAC (Asp)], resulting in the WT virus. These previous results emphasize the importance of starting the selection with genome-length RNAs containing a triple-nucleotide substitution in order to avoid true reversion. However, such a triple-nucleotide approach can be complicated by a low rate of recovery of viruses upon selection if the amino acid under study is absolutely required for viral replication. This fact is evidenced by the finding that, among the four mutant RNAs, only D146S RNA generated viruses. Therefore, more than one triple-nucleotide mutant RNA should be tried if the aim is to obtain revertant viruses that are potentially useful for identification of genetic interactions. Once a genetic linkage has been established through revertant analysis, it should be noted that the identified proteins and RNA elements do not necessarily interact with each other directly; the identified molecules can potentially exert their function through a third molecule as a mediator.

Revertant analysis of the D146S mutant viruses revealed a sequential pattern of emergence of mutations: the RdRp W751R appeared first, the MTase K61Q or K61T appeared second, and the 5' SL G35U or 38U-ins of the genomic RNA appeared last. The RdRp W751R mutation could be required for the subsequent emergence of other mutations. Alternatively, the RdRp W751R could have been the most critical or essential mutation for restoration of the D146S RNA replication. Our mutagenesis results argue against the latter possibility because mutant RNAs containing all adaptations except the RdRp W751R (i.e., G35U+K61Q+D146S and 38U-ins+K61Q+D146S) were infectious (Fig. 4). Furthermore, these genome-length RNAs were as infectious as the RNA containing the W751R mutation (K61Q+D146S+W751R), as evidenced by the SIVs and plaque morphology results. However, addition of the RdRp W751R mutation to an RNA harboring all other mutations, resulting in a G35U+K61Q+D146S+W751R RNA, further improved the replication. The results demonstrate that although the mutations accumulate in a specific order during selections, each mutation contributes independently to the improvement of viral replication.

The compensatory mutations in the MTase domain, K61Q and K61T, are located in the MTase-conserved K-D-K-E motif (17). The K61-D146-K182-E218 tetrad of the WNV MTase forms the active site for 2'-O methylation (31). Our previous mutagenesis results showed that D146 (but not the other three residues) within the K61-D146-K182-E218 motif is essential for N7 methylation (31). Addition of the K61Q or K61T (or the similar K61S) mutation to the D146S MTase domain did not rescue any 2'-O methylation activity; however, the resulting MTases (variants K61Q+D146S, K61T+D146S, and K61S+D146S) produced improved N7 methylation activity (Fig. 7). The effect of K61Q, K61T, or K61S on the improvement of N7 methylation activity of D146S MTase was salient because MTases containing the K61Q, K61S, or K61T mutation alone had only 20% to 68% of the WT MTase N7 methylation activity. A similar effect of the K61Q mutation on the improvement of N7 methylation was observed in the context of full-length NS5. The results clearly indicate that the MTase mutation (K61Q or K61T) contributes to viral replication of the D146S RNA through improvement of N7 cap methylation.

It remains to be determined how the K61Q, K61T, and K61S mutations improve N7 methylation activity. The chemistry of the N7 methylation reaction is assumed to be distinct from that of the 2'-O methylation. The structure of Encephalitozoon cuniculi (Ecm1) N7 MTase suggests a mechanism of methyl transfer without direct contact between the enzyme and either the attacking nucleophile N7 atom of guanine, the methyl carbon of SAM, or the leaving group sulfur of SAH; catalysis of N7 methylation is accomplished through the close apposition, in a specific geometry, of the two substrates (12). Our previous mutagenesis study showed that recombinant WNV MTases each containing a K61E, K61R, or K61A substitution reduced N7 methylation to 0%, 87%, or 70%, respectively, of the WT activity, while all MTases containing D146A, D146N, or D146K were completely defective in N7 methylation (45). These results suggest that D146 and, to a lesser extent, K61 contribute to the close apposition into which the two substrates (SAM and N7 guanine) are brought. The defective geometry caused by the D146S substitution may be partially rescued by the K61Q, K61T, or K61S adaptation. A tertiary cocrystal structure of the MTase-SAM-RNA complex will ultimately be needed to reveal the geometry of the K61 and D146 residues during the catalysis of N7 methylation.

The RdRp assay showed that full-length NS5 containing W571R alone, K61Q+W571R, or K61Q+D146S+W571R enhanced polymerase activity five- to eightfold above the WT level, whereas NS5 containing K61Q, D146, or K61Q+D146S exhibited polymerase activity similar to that of the WT (Fig. 8). The results demonstrate that the RdRp W571R mutation alone, but not the MTase mutations, enhances polymerase activity. These results led us to conclude that the W751R mutation improves the D416S RNA replication through enhancement of polymerase activity. Since N7 methylation of flavivirus RNA cap is important for efficient viral translation, the W751R mutation may compensate for defective N7 methylation simply by overproducing viral genomes that are poorly translated. Interestingly, residue W751 is conserved in flaviviral NS5 proteins and is located at the entrance of the RNA template channel (Fig. 3D). As mentioned above, Malet and colleagues proposed that an MTase loop (spanning residues K46-R47-E49) interacts with an RdRp loop (containing residue L512) (24). The L512 loop is located at the exit of the RNA output tunnel (Fig. 3D). Since residue L512 is more than 34 Å away from W751 and since the two residues are located on different sides of the RdRp molecule, the W751R mutation is not likely to function through a direct interaction with the MTase domain. Although W751R enhances RdRp activity in vitro (Fig. 8), this mutation alone slightly reduced viral replication in cell culture, as evidenced by a 2.8-fold reduction of SIV of the mutant RNA and a smaller plaque size for the derived virus than for the WT (Fig. 4). The latter results indicate that the RdRp activity is finely modulated during flavivirus evolution; a viral RNA with a higher RdRp activity alone is not necessarily more infectious.

The effect of mutations in the 5' SL of the genomic RNA (G35U or U38-ins) on improvement of viral replication was supported by results from both genome-length RNA (Fig. 4) and replicon RNA (Fig. 6). The replicon results showed that the luciferase signals at 2 h p.t. (representing viral translation) (Fig. 6) were almost identical between the WT replicon and the G35U replicon, suggesting that the G35U change does not affect genome translation. In contrast to the cell culture results, these mutations did not affect MTase (Fig. 7) and RdRp activities in vitro (Fig. 8). Using footprinting and gel shift analyses, we recently showed that recombinant NS5 specifically binds to the 5' SL of the genomic RNA. However, the NS5 protein does not directly protect nucleotide G35U or the U38-ins region (H. Dong and P.-Y. Shi, unpublished data). Since the in vitro experiments were performed in the absence of other viral and cellular factors, it is not unreasonable to speculate that the RNA-protein interaction differs slightly in the context of replication complex. The G35U or U38-ins mutation could exert its function through interaction with the replicase complex in vivo.

 
We thank Hongmin Li and Joachim Jaeger for helpful discussions during the course of the work. We also thank the Molecular Genetics Core and the Cell Culture Facility at the Wadsworth Center for DNA sequencing and for maintenance of BHK-21 and Vero cells, respectively.

This work was supported partially by federal funds from the National Institute of Allergy and Infectious Diseases, National Institutes of Health, under contract NOI-AI-25490, and by NIH grants U01 AI061193 and U54-AI057158 (Northeast Biodefense Center).

 
* Corresponding author. Mailing address: Wadsworth Center, New York State Department of Health, 120 New Scotland Avenue, Albany, NY 12208. Phone: (518) 473-7487. Fax: (518) 473-1326. E-mail: ship{at}wadsworth.org

Published ahead of print on 30 April 2008.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1
1. Ackermann, M., and R. Padmanabhan. 2001. De novo synthesis of RNA by the dengue virus RNA-dependent RNA polymerase exhibits temperature dependence at the initiation but not elongation phase. J. Biol. Chem. 276:39926-39937.[Abstract/Free Full Text]
2
2. Alvarez, D. E., M. F. Lodeiro, S. J. Luduena, L. I. Pietrasanta, and A. V. Gamarnik. 2005. Long-range RNA-RNA interactions circularize the dengue virus genome. J. Virol. 79:6631-6643.[Abstract/Free Full Text]
3
3. Assenberg, R., J. Ren, A. Verma, T. S. Walter, David Alderton, R. J. Hurrelbrink, S. D. Fuller, S. Bressanelli, R. J. Owens, D. I. Stuart, and J. M. Grimes. 2007. Crystal structure of the Murray Valley encephalitis virus NS5 methyltransferase domain in complex with cap analogues. J. Gen. Virol. 88:2228-2236.[Abstract/Free Full Text]
4
4. Bartelma, G., and R. Padmanabhan. 2002. Expression, purification, and characterization of the RNA 5'-triphosphatase activity of dengue virus type 2 nonstructural protein 3. Virology 299:122-132.[CrossRef][Medline]
5
5. Cho, E. J., T. Takagi, C. R. Moore, and S. Buratowski. 1997. mRNA capping enzyme is recruited to the transcription complex by phosphorylation of the RNA polymerase II carboxy-terminal domain. Genes Dev. 11:3319-3326.[Abstract/Free Full Text]
6
6. Chu, P. W., and E. G. Westaway. 1985. Replication strategy of Kunjin virus: evidence for recycling role of replicative form RNA as template in semiconservative and asymmetric replication. Virology 140:68-79.[CrossRef][Medline]
7
7. Cleaves, G. R., and D. T. Dubin. 1979. Methylation status of intracellular dengue type 2 40S RNA. Virology 96:159-165.[CrossRef][Medline]
8
8. Dong, H., D. Ray, S. Ren, B. Zhang, F. Puig-Basagoiti, Y. Takagi, C. Ho, H. Li, and P. Shi. 2007. Distinct RNA elements confer specificity to flavivirus RNA cap methylation events. J. Virol. 81:4412-4421.[Abstract/Free Full Text]
9
9. Dong, H., S. Ren, H. Li, and P. Shi. Separate molecules of West Nile virus methyltransferase can independently catalyze the N7 and 2'-O methylations of viral RNA cap. Virology, in press.
10
10. Dong, H., S. Ren, B. Zhang, Y. Zhou, F. Puig-Basagoiti, H. Li, and P. Shi. 2008. Flavivirus methyltransferase catalyzes two methylations of the viral RNA cap through a substrate repositioning mechanism. J. Virol. 82:4295-4307.[Abstract/Free Full Text]
11
11. Egloff, M. P., D. Benarroch, B. Selisko, J. L. Romette, and B. Canard. 2002. An RNA cap (nucleoside-2'-O-)-methyltransferase in the flavivirus RNA polymerase NS5: crystal structure and functional characterization. EMBO J. 21:2757-2768.[CrossRef][Medline]
12
12. Fabrega, C., S. Hausmann, V. Shen, S. Shuman, and C. D. Lima. 2004. Structure and mechanism of mRNA cap (guanine-N7) methyltransferase. Mol. Cell 13:77-89.[CrossRef][Medline]
13
13. Falgout, B., R. H. Miller, and C. J. Lai. 1993. Deletion analysis of dengue virus type 4 nonstructural protein NS2B: identification of a domain required for NS2B-NS3 protease activity. J. Virol. 67:2034-2042.[Abstract/Free Full Text]
14
14. Filomatori, C., M. Lodeiro, D. Alvarez, M. Samsa, L. Pietrasanta, and A. Gamarnik. 2006. A 5' RNA element promotes dengue virus RNA synthesis on a circular genome. Genes Dev. 20:2238-2249.[Abstract/Free Full Text]
15
15. Furuichi, Y., and A. J. Shatkin. 2000. Viral and cellular mRNA capping: past and prospects. Adv. Virus Res. 55:135-184.[CrossRef][Medline]
16
16. Guyatt, K. J., E. G. Westaway, and A. A. Khromykh. 2001. Expression and purification of enzymatically active recombinant RNA-dependent RNA polymerase (NS5) of the flavivirus Kunjin. J. Virol. Methods 92:37-44.[CrossRef][Medline]
17
17. Hager, J., B. L. Staker, H. Bugl, and U. Jakob. 2002. Active site in RrmJ, a heat shock-induced methyltransferase. J. Biol. Chem. 277:41978-41986.[Abstract/Free Full Text]
18
18. Khromykh, A., N. Kondratieva, J. Sgro, A. Palmenberg, and E. Westaway. 2003. Significance in replication of the terminal nucleotides of the flavivirus genome. J. Virol. 77:10623-10629.[Abstract/Free Full Text]
19
19. Khromykh, A. A., H. Meka, K. J. Guyatt, and E. G. Westaway. 2001. Essential role of cyclization sequences in flavivirus RNA replication. J. Virol. 75:6719-6728.[Abstract/Free Full Text]
20
20. Li, H., S. Clum, S. You, K. E. Ebner, and R. Padmanabhan. 1999. The serine protease and RNA-stimulated nucleoside triphosphatase and RNA helicase functional domains of dengue virus type 2 NS3 converge within a region of 20 amino acids. J. Virol. 73:3108-3116.[Abstract/Free Full Text]
21
21. Lindenbach, B. D., and C. M. Rice. 1999. Genetic interaction of flavivirus nonstructural proteins NS1 and NS4A as a determinant of replicase function. J. Virol. 73:4611-4621.[Abstract/Free Full Text]
22
22. Lindenbach, B. D., and C. M. Rice. 2007. Flaviviridae: the virus and their replication, p. 991-1042. In D. M. Knipe, P. M. Howley, D. E. Griffin, R. A. Lamb, M. A. Martin, B. Roizman, and S. E. Straus (ed.), Fields virology, 4th ed. Lippincott Williams & Wilkins, Philadelphia, PA.
23
23. Lo, L., M. Tilgner, K. Bernard, and P.-Y. Shi. 2003. Functional analysis of mosquito-borne flavivirus conserved sequence elements within 3' untranslated region of West Nile virus by use of a reporting replicon that differentiates between viral translation and RNA replication. J. Virol. 77:10004-10014.[Abstract/Free Full Text]
24
24. Malet, H., M. P. Egloff, B. Selisko, R. E. Butcher, P. J. Wright, M. Roberts, A. Gruez, G. Sulzenbacher, C. Vonrhein, G. Bricogne, J. M. Mackenzie, A. A. Khromykh, A. D. Davidson, and B. Canard. 2007. Crystal structure of the RNA polymerase domain of the West Nile virus non-structural protein 5. J. Biol. Chem. 282:10678-10689.[Abstract/Free Full Text]
25
25. Markoff, L. 2003. 5'- and 3'-noncoding regions in flavivirus RNA. Adv. Virus Res. 59:177-228.[CrossRef][Medline]
26
26. Mastrangelo, E., M. Bollati, M. Milani, B. Selisko, F. Peyrane, B. Canard, G. Grard, X. De Lamballerie, and M. Bolognesi. 2007. Structural bases for substrate recognition and activity in Meaban virus nucleoside-2'-O-methyltransferase. Protein Sci. 16:1133-1145.[CrossRef][Medline]
27
27. McCracken, S., N. Fong, E. Rosonina, K. Yankulov, G. Brothers, D. Siderovski, A. Hessel, S. Foster, S. Shuman, and D. L. Bentley. 1997. 5'-Capping enzymes are targeted to pre-mRNA by binding to the phosphorylated carboxy-terminal domain of RNA polymerase II. Genes Dev. 11:3306-3318.[Abstract/Free Full Text]
28
28. Ogino, T., and A. K. Banerjee. 2007. Unconventional mechanism of mRNA capping by the RNA-dependent RNA polymerase of vesicular stomatitis virus. Mol. Cell 25:85-97.[CrossRef][Medline]
29
29. Ogino, T., M. Kobayashi, M. Iwama, and K. Mizumoto. 2005. Sendai virus RNA-dependent RNA polymerase L protein catalyzes cap methylation of virus-specific mRNA. J. Biol. Chem. 280:4429-4435.[Abstract/Free Full Text]
30
30. Puig-Basagoiti, F., T. S. Deas, P. Ren, M. Tilgner, D. M. Ferguson, and P.-Y. Shi. 2005. High-throughput assays using luciferase-expressing replicon, virus-like particle, and full-length virus for West Nile virus drug discovery. Antimicrob. Agent. Chemother. 49:4980-4988.[Abstract/Free Full Text]
31
31. Ray, D., A. Shah, M. Tilgner, Y. Guo, Y. Zhao, H. Dong, T. Deas, Y. Zhou, H. Li, and P. Shi. 2006. West Nile virus 5'-cap structure is formed by sequential guanine N-7 and ribose 2'-O methylations by nonstructural protein 5. J. Virol. 80:8362-8370.[Abstract/Free Full Text]
32
32. Shi, P. Y., E. B. Kauffman, P. Ren, A. Felton, J. H. Tai, A. P. Dupuis II, S. A. Jones, K. A. Ngo, D. C. Nicholas, J. Maffei, G. D. Ebel, K. A. Bernard, and L. D. Kramer. 2001. High-throughput detection of West Nile virus RNA. J. Clin. Microbiol. 39:1264-1271.[Abstract/Free Full Text]
33
33. Shi, P. Y., M. Tilgner, M. K. Lo, K. A. Kent, and K. A. Bernard. 2002. Infectious cDNA clone of the epidemic West Nile virus from New York City. J. Virol. 76:5847-5856.[Abstract/Free Full Text]
34
34. Tan, B. H., J. Fu, R. J. Sugrue, E. H. Yap, Y. C. Chan, and Y. H. Tan. 1996. Recombinant dengue type 1 virus NS5 protein expressed in Escherichia coli exhibits RNA-dependent RNA polymerase activity. Virology 216:317-325.[CrossRef][Medline]
35
35. Tilgner, M., T. S. Deas, and P.-Y. Shi. 2005. The flavivirus-conserved penta-nucleotide in the 3' stem-loop of the West Nile virus genome requires a specific sequence and structure for RNA synthesis, but not for viral translation. Virology 331:375-386.[CrossRef][Medline]
36
36. Tilgner, M., and P. Y. Shi. 2004. Structure and function of the 3' terminal six nucleotides of the West Nile virus genome in viral replication. J. Virol. 78:8159-8171.[Abstract/Free Full Text]
37
37. Warrener, P., J. K. Tamura, and M. S. Collett. 1993. RNA-stimulated NTPase activity associated with yellow fever virus NS3 protein expressed in bacteria. J. Virol. 67:989-996.[Abstract/Free Full Text]
38
38. Wengler, G. 1981. Terminal sequences of the genome and replicative-form RNA of the flavivirus West Nile virus: absence of poly(A) and possible role in RNA replication. Virology 113:544-555.[CrossRef][Medline]
39
39. Wengler, G., and G. Wengler. 1991. The carboxy-terminal part of the NS 3 protein of the West Nile flavivirus can be isolated as a soluble protein after proteolytic cleavage and represents an RNA-stimulated NTPase. Virology 184:707-715.[CrossRef][Medline]
40
40. Wengler, G., and G. Wengler. 1993. The NS3 nonstructural protein of flaviviruses contains an RNA triphosphatase activity. Virology 197:265-273.[CrossRef][Medline]
41
41. You, S., and R. Padmanabhan. 1999. A novel in vitro replication system for dengue virus. Initiation of RNA synthesis at the 3'-end of exogenous viral RNA templates requires 5'- and 3'-terminal complementary sequence motifs of the viral RNA. J. Biol. Chem. 274:33714-33722.[Abstract/Free Full Text]
42
42. Yue, Z., E. Maldonado, R. Pillutla, H. Cho, D. Reinberg, and A. J. Shatkin. 1997. Mammalian capping enzyme complements mutant Saccharomyces cerevisiae lacking mRNA guanylyltransferase and selectively binds the elongating form of RNA polymerase II. Proc. Natl. Acad. Sci. USA 94:12898-12903.[Abstract/Free Full Text]
43
43. Zeng, L., B. Falgout, and L. Markoff. 1998. Identification of specific nucleotide sequences within the conserved 3'-SL in the dengue type 2 virus genome required for replication. J. Virol. 72:7510-7522.[Abstract/Free Full Text]
44
44. Zhang, B., H. Dong, D. A. Stein, P. L. Iversen, and P. Y. Shi. 2008. West Nile virus genome cyclization and RNA replication require two pairs of long-distance RNA interactions. Virology 373:1-13.[CrossRef][Medline]
45
45. Zhou, Y., D. Ray, Y. Zhao, H. Dong, S. Ren, Z. Li, Y. Guo, K. Bernard, P. Shi, and H. Li. 2007. Structure and function of flavivirus NS5 methyltransferase. J. Virol. 81:3891-3903.[Abstract/Free Full Text]

---

Journal of Virology, July 2008, p. 7047-7058, Vol. 82, No. 14
0022-538X/08/$08.00+0     doi:10.1128/JVI.00654-08
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

This article has been cited by other articles:

• Emara, M. M., Liu, H., Davis, W. G., Brinton, M. A. (2008). Mutation of Mapped TIA-1/TIAR Binding Sites in the 3' Terminal Stem-Loop of West Nile Virus Minus-Strand RNA in an Infectious Clone Negatively Affects Genomic RNA Amplification. J. Virol. 82: 10657-10670 [Abstract] [Full Text]  

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Zhang, B. Articles by Shi, P.-Y. Search for Related Content PubMed PubMed Citation Articles by Zhang, B. Articles by Shi, P.-Y.