Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2009 Jul 20.
Published in final edited form as: Virology. 2008 May 23;377(1):1–6. doi: 10.1016/j.virol.2008.04.026

Separate molecules of West Nile virus methyltransferase can independently catalyze the N7 and 2′-O methylations of viral RNA cap

Hongping Dong 1,2, Suping Ren 3, Hongmin Li 1,3, Pei-Yong Shi 1,3,*
PMCID: PMC2494700  NIHMSID: NIHMS57252  PMID: 18501946

Abstract

West Nile virus methyltransferase catalyzes N7 and 2′-O methylations of the viral RNA cap (GpppA-RNA→m7GpppAm-RNA). The two methylation events are independent, as evidenced by efficient N7 methylation of GpppA-RNA→m7GpppA-RNA and GpppAm-RNA→m7GpppAm-RNA, and by the 2′-O methylation of GpppA-RNA→GpppAm-RNA and m7GpppA-RNA→m7GpppAm-RNA. However, the 2′-O methylation activity prefers substrate m7GpppA-RNA to GpppA-RNA, thereby determining the dominant methylation pathway as GpppA-RNA→m7GpppA-RNA→m7GpppAm-RNA. Mutant enzymes with different methylation defects can trans complement one another in vitro. Furthermore, sequential treatment of GpppA-RNA with distinct methyltransferase mutants generates fully methylated m7GpppAm-RNA, demonstrating that separate molecules of the enzyme can independently catalyze the two cap methylations in vitro.

Keywords: Flavivirus NS5, RNA cap methylation, Flavivirus replication, West Nile virus, Methyltransferase

INTRODUCTION

Many members of the genus Flavivirus are significant human pathogens, including West Nile virus (WNV), yellow fever virus (YFV), dengue virus (DENV), Japanese encephalitis virus, and tick-borne encephalitis virus (Lindenbach and Rice, 2001). The plus-strand RNA genome of flavivirus contains a 5′-terminal cap 1 structure (m7GpppAm) (Cleaves and Dubin, 1979; Wengler and Gross, 1978). Cap formation on eukaryotic mRNA entails four enzymatic reactions, in which the 5′-triphosphate of the nascent RNA is cleaved to a diphosphate by an RNA triphosphatase, capped with GMP by an RNA guanylyltransferase, methylated at the N7 position of the cap guanosine by an RNA guanine-methyltransferase (N7 MTase), and methylated at the ribose 2′-OH positions of the first and/or second nucleotides of RNA by a nucleoside 2′-O MTase (Furuichi and Shatkin, 2000; Shuman, 2001). Since host mRNA capping occurs in the nucleus, viruses that replicate in the cytoplasm, such as flaviviruses, encode their own capping apparatus. For flaviviruses, the RNA triphosphatase and MTase are respectively located in the C-terminus of NS3 (Bartelma and Padmanabhan, 2002; Wengler and Wengler, 1993) and the N-terminus of NS5 (Egloff et al., 2002; Ray et al., 2006), whereas the location of the guanylyltransferase remains elusive.

We recently found that a single MTase of flavivirus performs both the N7 and 2′-O methylations of the viral RNA cap (Ray et al., 2006; Zhou et al., 2007). In WNV, the two cap methylations require distinct RNA elements within the 5′-terminal stem-loop of the genomic RNA (Dong et al., 2007). Incubation of unmethylated GpppA-RNA substrate with flavivirus MTase in the presence SAM at high pH (pH 9–10) sequentially generated products GpppA-RNA→m7GpppA-RNA→m7GpppAm-RNA. These results indicate that (i) the two methylation events are sequential; and (ii) 2′-O methylation may be dependent on prior N7 methylation. Optimization of assay conditions showed that the N7 and 2′-O methylations require different pH, pH 7 and 9–10, respectively (Zhou et al., 2007). At pH 7, the N7 methylation is optimal, but no 2′-O methylation can occur. At pH 9–10, 2′-O methylation is optimal, and N7 methylation occurs at about 30–50% of the optimal activity (measured at pH 7.0). The distinct assay conditions could be used to separate the two methylation events. Specifically, the N7 methylation can be measured by conversion of GpppA-RNA→m7GpppA-RNA at pH 7; no 2′-O methylation would occur under this condition. The 2′-O methylation can be monitored by conversion of m7GpppA-RNA→m7GpppAm-RNA; no N7 methylation would occur under this condition, because the substrate has already been methylated at the guanine N7 position. These assays have allowed us to dissect mutational effects of the MTase on the two methylation events (Dong et al., 2008; Zhou et al., 2007).

Crystal structures of flavivirus MTases (Assenberg et al., 2007; Egloff et al., 2002; Mastrangelo et al., 2007; Zhou et al., 2007) exhibit distinct binding sites for S-adenosyl-L-homocycteine (SAH), GTP, and RNA (Fig. 1A). The binding site for SAH, the byproduct from S-adenosyl-L-methionine (SAM) after transfer of its methyl group, was assumed to be the binding site for the methyl donor SAM. The RNA cap is bound at the GTP site, as evidenced by co-crystal structures of MTase complexed with cap analogues (Assenberg et al., 2007; Egloff et al., 2002; Egloff et al., 2007). The RNA-binding site was proposed to interact with the 5′-terminal region of RNA (Zhou et al., 2007). Despite the single known binding site for SAM, flavivirus MTase performs two distinct methylation reactions. Therefore, the substrate GpppA-RNA must be re-positioned so as to accept the N7 and 2′-OH methyl groups from SAM. In the repositioning model, guanine N7 of GpppA-RNA is first positioned next to SAM to generate m7GpppA-RNA, after which the m7G moiety is repositioned to the GTP-binding pocket so as to precisely register the 2′-OH of the adenosine to the SAM molecule, resulting in m7GpppAm-RNA (Zhou et al., 2007). The molecular repositioning model was supported by mutagenesis results showing that two distinct sets of amino acids on the surface of the WNV MTase are required for the two methylation reactions (Dong et al., 2008). One key question looming is what determines the sequential N7 and 2′-O methylations of the flavivirus RNA cap.

FIG. 1. Mutant MTases of WNV used in this study.

FIG. 1

Open in a new tab

(A) Surface representation of the WNV MTase structure. Amino acids E34, R84, E149, and E218 are indicated in red. The SAH molecule is depicted based on the co-crystal structure of the WNV MTase (Zhou et al., 2007). The GTP molecule was modeled through alignment of the WNV MTase structure with the DENV-2 GTP-SAH-MTase tertiary complex structure (Egloff et al., 2002), using PyMOL. The putative RNA-binding site is also indicated. (B) Summary of N7 and 2′-O methylation activities of mutant MTases. For each mutant MTase, the N7 and 2′-O methylation activities were respectively calculated from TLC analysis of G*pppA-RNA→m7G*pppA-RNA and m7G*pppA-RNA→m7G*pppAm-RNA reactions. All RNAs used in this study represented the 5′-terminal 190 nucleotides of the WNV genome. Symbol “*” indicates that the following phosphate is 32P-labeled. The WT methylation activity was set as 100% for each of the two methylations. The effects of individual mutations on viral replication in the context of genome-length RNA are indicated. Symbols “+” and “−“ indicate that genome-length RNA containing the MTase mutation are replicative (indicated by plaque formation) and non-replication, respectively.

The goal of this study is to analyze the relationship between the N7 and 2′-O methylations during WNV cap formation. We report, for the first time, that N7 methylation could efficiently occur using substrates that are either 2′-O unmethylated (GpppA-RNA) or 2′-O methylated (GpppAm-RNA). In contrast, 2′-O methylation prefers a substrate that is N7 methylated (m7GpppA-RNA) to a substrate that is N7 unmethylated (GpppA-RNA). The results suggest that the preference for the substrate with prior N7 methylation during 2′-methylation is the determinant for sequential methylations of flavivirus RNA cap.

RESULTS

Distinct mutant Mtases

During the course of study of flavivirus MTase, we have identified a panel of WNV mutant MTases in which several single Ala-substitutions could each cause a defect in either N7 or 2′-O methylation. Figure 1 shows four such mutants: E34A and E218A were competent in N7 methylation, but defective in 2′-O methylation; whereas R84A and E149A were defective in N7 methylation, but competent in 2′-O methylation. The N7 and 2′-O methylation activities of these mutant MTases (Fig. 1B) were previously quantified through thin-layer chromatography (TLC) analysis of the GpppA-RNA→m7GpppA-RNA and m7GpppA-RNA→m7GpppAm-RNA reactions, respectively. On the crystal structure of the WNV MTase (Fig. 1A), E34 and R84 are within the RNA-binding site; E149 is adjacent to SAH molecule, and is above the RNA-binding site; and E218 is within the active site of 2′-O methylation (K61-D146-K182-E218 tetrad), and is located between the GTP- and SAH-binding sites. These four mutants, together with the wild-type (WT) MTase, were used in this study to analyze the relationship between the two methylation events.

The 2′-O methylation is independent of the N7 methylation, but it does prefer a substrate with prior N7 methylation

To examine whether 2′-O methylation is dependent on prior N7 methylation, we performed a time-course study of the methylation, using two N7-defective mutants (R84A and E149A) and the WT MTase on the substrates G*pppA-RNA and m7G*pppA-RNA (representing the first 190 nucleotides of the WNV genome; “*” indicates that the following phosphate is 32P-labeled). The reactions were performed at pH 9 buffer, which is optimal for 2′-O methylation and supports about 50% of the optimal N7 methylation activity (Zhou et al., 2007). Vaccinia virus VP39, a well-characterized 2′-O MTase, was included as a positive control. The methylation reactions were treated with nuclease P1 (to release cap structures) and intestinal alkaline phosphatase (to remove terminal phosphate). The reaction mixtures were then analyzed on a 20% polyacrylamide denaturing gel (Fig. 2A). We chose high percentage gel, rather than TLC, to improve the resolution of the detection method (i.e., to clearly separate four different cap structures: G*pppA, m7G*pppA, G*pppAm, and m7G*pppAm). Similar to VP39, mutants R84A and E149A could each perform 2′-O methylation on both G*pppA-RNA and m7G*pppA-RNA substrates, demonstrating that the 2′-O methylation is independent of the N7 methylation. However, the 2′-O methylation was more efficient when the m7G*pppA-RNA substrate, rather than the G*pppA-RNA substrate, was used; this difference was more dramatic for the control VP39 than those for the R84A and E149A mutants (Fig. 2B). For the WT MTase, the conversion of G*pppA-RNA→m7G*pppA-RNA→m7G*pppAm-RNA was detected; however, small amount of G*pppAm-RNA product was also detected on the gel. Collectively, the results demonstrate that the 2′-O methylation is not absolutely dependent on the N7 methylation, but it does prefer a substrate with prior N7 methylation.

FIG. 2. Analysis of 2′-O methylation.

FIG. 2

Open in a new tab

(A) Time-course analysis of 2′-O methylation of WNV MTase (R84A and E149A mutants, and WT) and vaccinia virus VP39 (as a positive control) on substrates G*pppA-RNA and m7G*pppA-RNA. Except for VP39, methylation reactions were performed in a 2′-O buffer for WT, R84A, and E149A MTases. After incubation for indicated time periods, the reactions were treated with nuclease P1 and CIP. The methylation mixtures were then analyzed on a 20% PAGE containing 8 M urea. The enzyme, RNA substrate, and methylation time are indicated above the gel. See experimental details in Materials and Methods. The 32P-labeled cap analogue markers, G*pppA and m7G*pppA, were also analyzed on the gel. The positions of G*pppA, m7G*pppA, G*pppAm, and m7G*pppAm molecules are labeled to the left of the gel. (B) Quantification of the 2′-O methylation activity. Except for WT MTase, the percentages of conversion for GpppA-RNA→GpppAm-RNA and m7GpppA-RNA→m7GpppAm-RNA in (A) were quantified using PhosphorImager. Conversion percentage = product/(residual substrate+product)×100%. For the WT MTase, only the m7GpppA-RNA→m7GpppAm-RNA conversion was quantified; the GpppA-RNA→GpppAm-RNA conversion is not presented because the enzyme can methylate both the N7 and 2′-O positions.

Two mutant MTases with different methylation defects, but not mutants with the same methylation defect, can complement one another in trans

We examined whether two mutant enzymes with different methylation defects can complement one another, converting the unmethylated G*pppA-RNA to the fully methylated m7G*pppAm-RNA product. As shown in Figure 3A, E149A alone (lane 2) converted G*pppA-RNA to G*pppAm-RNA, but almost no other products, confirming that the E149A MTase is competent in 2′-O methylation, but defective in N7 methylation. In contrast, E34A (lane 3) or E218A (lane 4) alone dominantly converted G*pppA-RNA to m7G*pppA-RNA, indicating that these two mutants are defective in 2′-O methylation, but competent in N7 methylation. Due to a long incubation time (2 h), a weak 2′-O methylation activity was observed for E34A, as indicated by the m7G*pppAm band (lane 3). Remarkably, incubation of G*pppA-RNA with E149A and E34A (lanes 5–7), or with E149A and E218A (lanes 8–10), generated the m7G*pppAm-RNA product. For each mutant pair, three different schemes of MTase addition/incubation were tested: two schemes entailed incubation of the G*pppA-RNA with one mutant for 1 h, followed by addition and incubation with the second mutant for another hour (i.e., E149A→E34A, E34A→E149A, E149A→E218A, and E218A→E149A); the third scheme involved in simultaneous addition and incubation of the two mutant enzymes for 2 h (i.e., E149A+E34A and E149A+E218A). Although the yields of m7G*pppAm-RNA varied slightly among three different schemes, the results clearly suggested that two distinct mutant MTases can trans complement to convert G*pppA-RNA→m7G*pppAm-RNA. Notably, there was considerable difference in the ratio of m7G*pppA and G*pppAm products, depending on which mutant enzyme was incubated first (compare lanes 5 versus 6 and lanes 8 versus 9 in Fig. 3A). Because all reactions had an excess amount of input substrate G*pppA-RNA (as indicated by unreacted G*pppA molecule on the gel), the first added enzyme had one extra hour to catalyze its methylation than the second added enzyme did. Therefore, the first added enzyme produced an excess amount of its methylated product.

FIG. 3. Conversion of unmethylated GpppA-RNA to fully methylated m7GpppAm-RNA by two distinct MTases defective in N7 or 2′-O methylation.

FIG. 3

Open in a new tab

(A) Trans complementation of two mutant MTases defective in N7 or 2′-O methylation. Substrate G*pppA-RNA was incubated with mutant MTase alone (E149A, E34A, or E218A) or with a pair of mutant MTases (E149A and E34A, or E149A and E218A) in the presence of SAM in the 2′-O buffer. For each pair of mutant MTases, three different schemes were used for addition/incubation: two sequential-incubation schemes, and one simultaneous incubation scheme (see text for details). The reactions were digested with nuclease P1 and CIP, and the products were resolved on a high percentage denaturing gel. The mock- and WT MTase-treated samples were included as negative and positive controls, respectively. The migrations of G*pppA, m7G*pppA, G*pppAm, and m7G*pppAm are labeled to the right side of the gel. (B) No trans complementation of two mutant MTases defective in the same type of methylation. Substrate G*pppA-RNA was methylated with indicated mutant MTase alone or with a pair of mutant MTases harboring the same type of methylation defect (R84A+E149, both defective in N7 methylation; E34A+E218A, both defective in 2′-O methylation). The methylation reactions were incubated 22 C for 1 h in both pH 7 buffer (optimal for N7 methylation) or pH 9 buffer (optimal for 2′-O methylation). The amounts of MTases are indicated for each reaction. The reaction mixtures were resolved on a 20% polyacrylamide gel. (C) Independent N7 and 2′-O methylations by two mutant MTases. Substrate G*pppA-RNA was incubated with mutant R84A in the 2′-O buffer for 30 min, resulting in G*pppAm-RNA. After removal of the R84A MTase through phenol extraction, the G*pppAm-RNA was further incubated with E34A or E218 in the N7 buffer for the indicated time. The reaction mixtures were then analyzed on a denaturing gel. The methylation reactions using E34A, R84A, E218A, or WT MTase alone were also shown. The methylation reactions for the WT (lane 2) and R84A (lane 4) alone were performed in the 2′-O buffer; the methylation reactions for the E34A (lane 3) and E218A (lane 11) alone were performed in the N7 buffer.

Next, we tested whether two mutants with the same methylation defects can trans complement one another (Fig. 3B). Incubation of G*pppA-RNA with R84A alone, E149A alone, or R84A+E149 (both defective in N7 methylation) generated G*pppAm-RNA in the pH 9 buffer (optimal for 2′-O methylation), but did not yield any m7G*pppA-RNA or m7G*pppAm-RNA (lanes 8–10), even in the pH 7 buffer (optimal for N7 methylation; lanes 2–4). These results demonstrate that two distinct N7-defective mutants can not trans complement to restore the N7 methylation activity. Reciprocally, incubation of G*pppA-RNA with E34A alone, E218A alone, or E34A+E218A (both defective in 2′-O methylation) generated m7G*pppA-RNA at pH 7 buffer, but did not yield any m7G*pppAm-RNA or G*pppAm-RNA (lanes 11–13). In the pH 9 buffer, E34A (lanes 5) alone and E34A+E218A (lane 7) generated small amount of G*pppAm-RNA, due to residual 2′-O methylation activity of the E34A MTase. The results demonstrate that two distinct 2′-O-defective mutants can not complement to catalyze the 2′-O methylation. Taken together, the results presented under this section strongly suggest that only mutants defective in different methylations, but not mutants defective in same methylation, can complement one another in trans.

The GpppAm-RNA can be further methylated at the guanine N7 position to yield m7GpppAm-RNA

The weak activity of N7 methylation-independent 2′-O methylation (i.e., G*pppA-RNA→G*pppAm-RNA) indicates an alternative pathway of cap methylations: GpppA-RNA→GpppAm-RNA→m7GpppAm-RNA. To determine whether the 2′-O methylated substrate GpppAm-RNA can be further methylated at the guanine N7 position, we prepared G*pppAm-RNA by incubating G*pppA-RNA with the mutant R84A (competent in 2′-O methylation) for 30 min, followed by phenol extraction and ethanol precipitation to remove the R84A MTase. The resulting G*pppAm-RNA (Fig. 3C, lane 4) was then incubated with E34A or E218A (competent in N7 methylation) for 5, 30, or 60 min (lanes 5–10). Analysis of the reactions on a denaturing gel showed that the G*pppAm-RNA can be converted to the double-methylated m7G*pppAm-RNA. As controls, E34A (lane 3) and E218A (lane 11) alone could only convert G*pppA-RNA→m7G*pppA-RNA, whereas R84A (lane 3) alone could only catalyze G*pppA-RNA→G*pppAm-RNA. These results demonstrate that the two methylations can be sequentially executed by separate MTase molecules. The results also indicate that an alternative ordering of cap methylations, GpppA-RNA→GpppAm-RNA→m7GpppAm-RNA, can occur, at least in vitro.

Prior 2′-O methylation of an RNA substrate does not affect the efficiency of N7 methylation

The above results demonstrated that both the unmethylated GpppA-RNA and the 2′-O methylated GpppAm-RNA are active substrates for N7 methylation. To test whether the presence of the 2′-O methyl group affects the efficiency of N7 methylation, we compared the rates of N7 methylation between substrates G*pppA-RNA and G*pppAm-RNA. The G*pppAm-RNA was prepared by incubation of G*pppA-RNA with R84A mutant in the presence of SAM in a pH 9 buffer for 1 h. A time-course analysis using WT MTase showed that substrates G*pppA-RNA and G*pppAm-RNA were equally active in N7 methylation (Fig. 4A, top panel). Similar results were obtained when either of the N7-competent but 2′-O-defective mutants E34A (Fig. 4A, bottom panel) or E218A was used (data not shown). The summary of these experiments (Fig. 4B) suggest that a prior 2′-O methylation does not significantly affect the efficiency of the N7 methylation.

FIG. 4. Comparison of N7 methylation efficiencies between substrates GpppA-RNA and GpppAm-RNA.

FIG. 4

Open in a new tab

(A) Substrates GpppA-RNA and GpppAm-RNA were incubated with WT MTase (top panel) or E34A MTase (bottom panel) in the N7 buffer for the indicated time, and digested with nuclease P1 and CIP. The products were separated on a high percentage denaturing gel. The positions of G*pppA, m7G*pppA, G*pppAm, and m7G*pppAm are indicated on the left side of the gel. (B) The efficiencies of the N7 methylation reactions in (A) are summarized by the percentages of conversions of GpppA-RNA→m7GpppA-RNA and GpppAm-RNA→m7GpppAm-RNA over the time course. The percentage of substrate-to-product conversion was quantified using PhosphorImager.

DISCUSSION

In this study, we compared the methylation efficiencies of distinct mutant MTases of WNV (Fig. 1) on RNA substrates having various cap methylation statuses, i.e., GpppA-RNA, m7GpppA-RNA, and GpppAm-RNA. For N7 methylation, the presence of a 2′-O methyl group in the cap structure does not affect the N7 methylation efficiency (Fig. 4). In contrast, for 2′-O methylation, the N7-mutant MTases catalyze the m7GpppA-RNA→m7GpppAm-RNA reaction more efficiently than they do the GpppA-RNA→GpppAm-RNA reaction (Fig. 2). The N7-independent 2′-O methylation was also previously reported in DENV-2 MTase when a short nonviral RNA [GpppA(C)5] was used as a substrate (Egloff et al., 2002). These results demonstrate that the two methylation events are not absolutely dependent on one another. The results also suggest two alternative methylation pathways in vitro: GpppA-RNA→m7GpppA-RNA→m7GpppAm-RNA and GpppA-RNA→GpppAm-RNA→m7GpppAm-RNA. Since the N7 methylation has no preference between substrates GpppA-RNA and GpppAm-RNA, the substrate preference of 2′-O methylation for m7GpppA-RNA could be the factor that determines the dominant pathway of WNV cap methylation as GpppA-RNA→m7GpppA-RNA→m7GpppAm-RNA. The current results are in agreement with our previous observation that N7 methylation precedes 2′-O methylation (Ray et al., 2006; Zhou et al., 2007). The new finding that 2′-O methylation prefers a substrate with prior N7 methylation provides an explanation for why the two methylation events are sequential.

Our results showed that high percentage denaturing gel is a reliable method for separation of four types of cap structures (G*pppA, m7G*pppA, G*pppAm, and m7G*pppAm). All four cap molecules could be clearly separated from each other on the high percentage gel. The resolution of the high percentage gel is higher than that of TLC analysis. We previously used TLC to analyze the cap structures, and could not consistently detect G*pppAm molecule. As discussed above, due to the dominant methylation pathway of GpppA-RNA→m7GpppA-RNA→m7GpppAm-RNA, the amount of G*pppAm is much less than those of G*pppA, m7G*pppA, and m7G*pppAm. The weak signal of G*pppAm on TLC plate could be masked by the strong signals from the other three cap structures, leading to difficulty in detection of the G*pppAm molecule on the TLC plate.

Two scenarios can be envisioned for the repositioning of the WNV RNA substrate on the enzyme surface from one methylation event to the next. The substrate could translocate on a single enzyme molecule; alternatively, the substrate could dissociate from one enzyme molecule after the first methylation, and reassociate with another enzyme molecule for the second methylation. The current study demonstrates that the two methylations can be independently accomplished by two individual N7 or 2′-O mutant MTases; in contrast, two mutants with the same type of methylation defect could not complement to restore the defective methylation (Fig. 3). The successful trans complementation of two distinct MTase mutants is in favor of the model that the two methylations of the WNV RNA cap may be executed by separate methyltransferase molecules, through an RNA-enzyme dissociation-and-reassociation process. This process shows similarity to the processes of reovirus and bluetongue virus RNA cap formation, in which the N7 and 2′-O methylations are sequentially catalyzed by two separate MTase domains within a large polyprotein (Reinisch, Nibert, and Harrison, 2000; Sutton et al., 2007). Although our results clearly showed that the two distinct mutants can separately catalyze the N7 and 2′-O methylations, we could not completely exclude the possibility that WT MTase uses a single molecule to catalyze both reactions. Methylations of the cap structure at both positions with a single MTase molecule is supposed to be more efficient than that with two MTase molecules, because there is no need to dissociate the capped RNA substrate from the enzyme until the fully-methylated cap structure is produced. A biophysical approach [such as FRET (fluorescence resonance energy transfer)] is needed to clearly differentiate between the single and double MTase-based translocation models.

The enzyme-based trans complementation does not seem to agree with the replicon-based trans complementation. We previously showed that WNV replicons containing mutations in MTase could not be rescued by a WT Neo-replicon in BHK-21 cells (Ray et al., 2006). If two separate MTase molecules are used for different methylations via the dissociation-and-reassociation process, then WT NS5 from the helper Neo-replicon should be able to complement for the defective NS5 from the mutant replicon. It is currently not known what causes the discrepancy between the in vitro and in vivo results. Apparently, recombinant MTase molecules in solution are readily available for association and disassociation with RNA substrate, making it accessible for complementation in vitro. In contrast, defective NS5 (with single-amino acid mutation) derived from the mutant replicon is associated with other viral and host proteins on the ER membrane (as replication complexes), and may not be available for substitution with WT NS5 from the helper replicon. Khromykh and colleagues previously showed that mutant NS5 with a C-terminal deletion up to 589 amino acids (aa 317–905) could be successfully complemented with helper replicon in Kunjin virus, whereas mutant NS5 within the N-terminal region (aa 1–316, spanning the MTase domain) could not be efficiently trans complemented in vivo (Liu et al., 2002). These results suggest that RNA replication of Kunjin virus requires the MTase-coding RNA and/or the MTase expression in cis. It should be noted that the lack of trans complementation in vivo does not devaluate the in vitro complementation results. We currently do not know how many copies of NS5 are present in one replication complex. It is not unreasonable to speculate that more than one copy of NS5 is included in the replication complex to perform distinct cap methylations. The requirement of cis expression of MTase, as demonstrated in Kunjin virus (Liu et al., 2002), could prohibit successful trans complementation in vivo.

MATERIALS AND METHODS

Mutant MTases of WNV

Recombinant MTases were expressed using a pET28(a) vector containing the WNV MTase domain, representing the N-terminal 300 amino acids of NS5. All MTases contained an N-terminal (His)6-tag, and were expressed and purified through a Ni-NTA column. The protein (>90% purity) was quantified by Bradford assay (Bio-Rad) and verified by SDS-PAGE. The preparation of WT and E218A MTases was described in (Ray et al., 2006; Zhou et al., 2007); and the preparation of E34A, R84A, and E149A MTases was detailed in (Dong et al., 2008).

Methylation assays

The 5′-end labeled substrates, m7G*pppA-RNA and G*pppA-RNA (representing the first 190 nucleotides of the WNV genome) were prepared by incubating pppA-RNA, [α-32P]GTP (3000 Ci/mmol), and vaccinia virus capping enzyme (Epicenter) in the presence and absence of SAM, respectively. The capping reactions were performed following the manufacturer’s protocol. The labeled RNAs were purified through two Sephadex G-25 spin columns (GE Healthcare), extracted with phenol-chloroform, and precipitated with ethanol. The resulting m7G*pppA-RNA and G*pppA-RNA were resuspended in RNase-free water for methlation reactions.

Three types of methylation assays were performed for WNV MTase. (i) A standard N7 methylation (G*pppA-RNA→m7G*pppA-RNA) was performed in a 20-μl reaction containing about 3 pmol G*pppA-RNA, 80 μM SAM, 1 μg MTase in N7 buffer (50 mM Tris-HCl, pH 7.0, 20 mM NaCl, and 2 mM DTT); the reaction was incubated at 22 C for 5 min. (ii) A 2′-O methylation (m7G*pppA-RNA→m7G*pppA-RNAm or G*pppA-RNA→G*pppAm-RNA) was performed in a 20-μl reaction containing 3 pmol RNA substrate, 80 μM SAM, 1 μg MTase in 2′-O buffer (50 mM glycine, pH 9, and 2 mM DTT); the reaction was incubated at 22 C for 1 h. (iii) A double-methylation (G*pppA-RNA→m7G*pppA-RNA→m7G*pppA-RNAm or G*pppA-RNA→G*pppAm-RNA→m7G*pppAm-RNA) reaction was performed in the 2′-O buffer at 22°C for indicated time period. All methyltation reactions were then treated with 1 unit nuclease P1 at 37ºC for 5 h, followed by 5 units of intestinal alkaline phosphatase at 37ºC for 2 h. As controls, cap methylation with vaccinia virus VP39 protein was also performed in a 20-μl reaction containing 50 mM Tris-HCl (pH 8.0), 5 mM DTT, 10 μM SAM, about 3 pmol m7G*pppA-RNA or G*pppA-RNA, and 0.5 μg recombinant VP39; the reaction was incubated at 30°C for 1 h. The reaction mixtures were separated on a 20% polyacrylamide gel (20×45 cm2) with 8 M urea. The bands representing different cap structures (G*pppA, m7G*pppA, G*pppAm, and m7G*pppAm) were quantified using a PhosphorImager. All experiments were performed for as least three times.

Acknowledgments

We thank C. Kiong Ho at the State University of New York at Buffalo for providing VP39, and the Molecular Genetics Core at the Wadsworth Center for DNA sequencing. The work was partially supported by grants AI061193 and U54-AI057158, and contract AI25490 from NIH.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Assenberg R, Ren J, Verma A, Walter TS, David Alderton, Hurrelbrink RJ, Fuller SD, Bressanelli S, Owens RJ, Stuart DI, Grimes JM. Crystal structure of the Murray Valley encephalitis virus NS5 methyltransferase domain in complex with cap analogues. J Gen Virol. 2007;88:2228–2236. doi: 10.1099/vir.0.82757-0. [DOI] [PubMed] [Google Scholar]
  2. Bartelma G, Padmanabhan R. Expression, purification, and characterization of the RNA 5′-triphosphatase activity of dengue virus type 2 nonstructural protein 3. Virology. 2002;299(1):122–32. doi: 10.1006/viro.2002.1504. [DOI] [PubMed] [Google Scholar]
  3. Cleaves GR, Dubin DT. Methylation status of intracellular Dengue type 2 40S RNA. Virology. 1979;96:159–165. doi: 10.1016/0042-6822(79)90181-8. [DOI] [PubMed] [Google Scholar]
  4. Dong H, Ray D, Ren S, Zhang B, Puig-Basagoiti F, Takagi Y, Ho C, Li H, Shi P. Distinct RNA elements confer specificity to flavivirus RNA cap methylation events. J Virol. 2007;81(9):4412–4421. doi: 10.1128/JVI.02455-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Dong H, Ren S, Zhang B, Zhou Y, Puig-Basagoiti F, Li H, Shi P. Flavivirus methyltransferase catalyzes two methylations of the viral RNA cap through a substrate repositioning mechanism. J Virol. 2008 doi: 10.1128/JVI.02202-07. In press. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Egloff MP, Benarroch D, Selisko B, Romette JL, Canard B. An RNA cap (nucleoside-2′-O-)-methyltransferase in the flavivirus RNA polymerase NS5: crystal structure and functional characterization. EMBO J. 2002;21(11):2757–68. doi: 10.1093/emboj/21.11.2757. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Egloff MP, Decroly E, Malet H, Selisko B, Benarroch D, Ferron F, Canard B. Structural and Functional Analysis of Methylation and 5′-RNA Sequence Requirements of Short Capped RNAs by the Methyltransferase Domain of Dengue Virus NS5. J Mol Biol. 2007 doi: 10.1016/j.jmb.2007.07.005. [DOI] [PubMed] [Google Scholar]
  8. Furuichi Y, Shatkin AJ. Viral and cellular mRNA capping: past and prospects. Adv Virus Res. 2000;55:135–84. doi: 10.1016/S0065-3527(00)55003-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Lindenbach BD, Rice CM. Flaviviridae: The Virus and Their Replication. In: Howley PM, editor. Fields Virology. 4. Lippincott William & Wilkins; 2001. [Google Scholar]
  10. Liu WJ, Sedlak PL, Kondratieva N, Khromykh AA. Complementation analysis of the flavivirus Kunjin NS3 and NS5 proteins defines the minimal regions essential for formation of a replication complex and shows a requirement of NS3 in cis for virus assembly. J Virol. 2002;76(21):10766–75. doi: 10.1128/JVI.76.21.10766-10775.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Mastrangelo E, Bollati M, Milani M, Selisko B, Peyrane F, Canard B, Grard G, De Lamballerie X, Bolognesi M. Structural bases for substrate recognition and activity in Meaban virus nucleoside-2′-O-methyltransferase. Protein Sci. 2007 doi: 10.1110/ps.072758107. Epub ahead of print. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Ray D, Shah A, Tilgner M, Guo Y, Zhao Y, Dong H, Deas T, Zhou Y, Li H, Shi P. West nile virus 5′-cap structure is formed by sequential guanine N-7 and ribose 2′-O methylations by nonstructural protein 5. J Virol. 2006;80(17):8362–70. doi: 10.1128/JVI.00814-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Reinisch KM, Nibert ML, Harrison SC. Structure of the reovirus core at 3.6 A resolution. Nature. 2000;404(6781):960–7. doi: 10.1038/35010041. [DOI] [PubMed] [Google Scholar]
  14. Shuman S. Structure, mechanism, and evolution of the mRNA capping apparatus. Prog Nucleic Acid Res Mol Biol. 2001;66:1–40. doi: 10.1016/s0079-6603(00)66025-7. [DOI] [PubMed] [Google Scholar]
  15. Sutton G, Grimes JM, Stuart DI, Roy P. Bluetongue virus VP4 is an RNA-capping assembly line. Nat Struc Mol Biol. 2007;14:449–451. doi: 10.1038/nsmb1225. [DOI] [PubMed] [Google Scholar]
  16. Wengler G, Gross HJ. Studies on virus-specific nucleic acids synthesized in vertebrate and mosquito cells infected with flaviviruses. Virology. 1978;89(2):423–37. doi: 10.1016/0042-6822(78)90185-x. [DOI] [PubMed] [Google Scholar]
  17. Wengler G, Wengler G. The NS3 nonstructural protein of flaviviruses contains an RNA triphosphatase activity. Virology. 1993;197:265–273. doi: 10.1006/viro.1993.1587. [DOI] [PubMed] [Google Scholar]
  18. Zhou Y, Ray D, Zhao Y, Dong H, Ren S, Li Z, Guo Y, Bernard K, Shi P, Li H. Structure and function of flavivirus NS5 methyltransferase. J Virol. 2007;81(8):3891–3903. doi: 10.1128/JVI.02704-06. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES