Previous Article | Next Article ![]()
Journal of Virology, June 2005, p. 7262-7268, Vol. 79, No. 11
0022-538X/05/$08.00+0 doi:10.1128/JVI.79.11.7262-7268.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Department of Neuropharmacology, The Scripps Research Institute, La Jolla, California 92037
Received 1 October 2004/ Accepted 20 January 2005
|
|
|---|
|
|
|---|
The RdRps of many NS RNA viruses consist of L, a multifunctional enzyme that appears to possess all of the enzymatic activities associated with RNA synthesis, and a virally encoded phosphoprotein cofactor (15). However, there is no evidence that such a cofactor is required for the formation of the arenavirus functional RdRp (21, 23). The determination of the crystal structures of the RdRps of three different viruses revealed common structural features despite a significant overall sequence divergence (1, 7, 22, 25). In addition, sequence alignments among more than 80 L proteins of NS RNA viruses have identified six conserved regions, designated domains I to VI, which were proposed to specify the essential functional features common to all L proteins (30). Within domain III are found the conserved A, B, C, and D motifs, which are thought to form the module containing the active site in RNA synthesis (29). These predictions are well supported by functional data obtained from mutational-analysis studies with several L proteins (12, 14, 16, 19, 34, 35, 37, 38, 43).
As with other RdRps, arenavirus L proteins, including those of LCMV, have the characteristic conserved A, B, C, and D motifs within domain III (29) (Fig. 1A). The development of arenavirus reverse-genetics systems has opened new avenues for studies aimed at determining the structure-function relationships of the arenavirus polymerase. Here we have used a LCMV minigenome (MG) rescue assay to investigate the role of the highly conserved A and C motifs in arenavirus polymerase activity.
![]() View larger version (17K): [in a new window] |
FIG. 1. (A) Schematic representation of the L protein of LCMV showing the locations of conserved A to D motifs within domain III. (B) Summary of the amino acid (aa) substitutions examined in this study.
|
We first confirmed all mutations by sequencing and subsequently analyzed their polymerase activities in vivo using an LCMV MG rescue assay (21). For this assay, BHK-21 cells (7 x 105 cells per well in M6 plates) were infected (multiplicity of infection [MOI] = 3) with a recombinant vaccinia virus expressing the bacteriophage T7 RNA polymerase (VVT7) and subsequently transfected with plasmids encoding the LCMV MG and NP, together with wild-type (wt) or mutant L proteins under the control of the T7 promoter. Twenty-four hours, later cell extracts were prepared and analyzed for their levels of chloramphenicol acetyltransferase (CAT) activity. In this assay, the intracellularly T7-synthesized MG RNA is encapsidated by plasmid-supplied NP to generate the RNP template that can be replicated and transcribed by the intracellularly reconstituted virus polymerase. Therefore, expression of the MG, as determined by levels of CAT activity, is an accurate surrogate marker of the virus polymerase activity. Our results showed that all of the mutations introduced in motifs A and C resulted in L proteins devoid of polymerase activity, as determined by the lack of detectable CAT activity (Fig. 2A, lanes 1 to 10).
![]() View larger version (27K): [in a new window] |
FIG. 2. Mutational analysis of motifs A and C. (A) VVT7-infected BHK-21 cells were transfected with plasmids expressing LCMV MG no. 3 (0.5 µg), NP (0.75 µg), and wt L or the indicated mutant L proteins (0.5 µg) (21). Twenty-four hours later, cell lysates were prepared and analyzed for CAT activity (top) and for expression levels of L protein by Western blot analysis (bottom). NAc, nonacetylated chloramphenicol; MAc, monoacetylated chloramphenicol; DAc, diacetylated chloramphenicol. (B) Normalized levels of CAT activity. For the indicated samples, chloramphenicol conversion was quantified by phosphorimager analysis. The percentage of conversion obtained in the absence of L was considered to be nonspecific background. This background conversion value was subtracted from the conversion values obtained for each sample. (C) Effect of mutant L proteins on LCVM MG RNA replication and transcription. BHK-21 cells were infected with VVT7 and transfected as in panel A. Twenty-four hours later, total cellular RNA was isolated and equal amounts of each sample were analyzed by Northern blot hybridization using the indicated CAT strand-specific RNA probes.
|
Effect of mutations in motifs A and C on RNA synthesis mediated by the LCMV L polymerase. L polymerases of NS RNA viruses perform two distinct biosynthetic processes, namely, replication and transcription of the genome RNA. Mutations affecting either one, or both, of these activities would affect levels of MG-derived CAT activity. We therefore determined levels of CAT mRNA and anti-MG (aMG) RNA species to assess the effects of the introduced mutations on MG transcription and RNA replication, respectively. For this determination, we infected cells with VVT7 (MOI = 3) and transfected them with the indicated L-expressing plasmids together with the LCMV MG and NP plasmids. RNA was isolated 24 h later and analyzed by Northern blotting using a CAT antisense riboprobe to detect both CAT mRNA (transcription) and aMG RNA (replication) species. Consistent with the CAT activity results (Fig. 2A), levels of both CAT mRNA and aMG RNA species were undetectable by Northern blotting for all RNA samples from cells transfected with L polymerases carrying mutations within the A and C motifs (Fig. 2C, lanes 1 to 11, bottom). These results indicated that mutations within motifs A and C affected both RNA replication (aMG) and transcription (CAT mRNA) mediated by the LCMV polymerase similarly. Hybridization of the same membrane to a CAT sense probe confirmed that levels of MG RNA generated via T7-mediated transcription and subsequent processing, via self-cleavage of the hepatitis delta virus ribozyme, were similar for all samples (Fig. 2C, top). This result allowed us to exclude the idea that the differences in levels of MG RNA available for template formation contributed to the differences in levels of MG expression.
Mutants SD-GN (SDD to GDN) and S1323G (SDD to GDD) (Fig. 1B) recreated the conserved residues found in motif C of L polymerases of RNA viral NNS and PS, respectively (41). However, both types of mutations resulted in L proteins with undetectable levels of polymerase activity. The change from SDD to GDN in L of the SNS RNA bunyavirus also resulted in a nonfunctional virus polymerase (19). Concordant findings have been reported for the L protein of rabies virus (RV), where mutations changing the GDN (NNS) to SDD (SNS) or to GDD (PS) within motif C resulted in a nonfunctional RV polymerase (34). Likewise, changes in the residue preceding the DD cores of the Qbeta bacteriophage (18) and influenza virus PB1 polymerases (2) resulted in enzymes without transcriptional activity. Also, the mutation GDN to GDD in the L protein of the prototypic mononegavirus vesicular stomatitis virus resulted in a virus polymerase still exhibiting 25% activity as measured in an in vitro transcription assay (35). The reasons for the different effects of the same mutation on the RV and vesicular stomatitis virus L proteins remain to be determined, but it cannot be ruled out that they reflect differences in the experimental assays used to assess polymerase activity.
LCMV polymerase activity in cells coexpressing nonfunctional mutant and wt L proteins. Intragenic complementation has been documented for the L genes of several NS RNA viruses (31, 36). Consistent with this, direct L-L physical interaction has been demonstrated for the paramyxoviruses Sendai virus (36) and parainfluenza virus type 3 (PIV3) (39), and evidence indicates that in these cases, L oligomerization is required for polymerase activity. We therefore examined whether a similar finding also applied to the LCMV L protein. For this we examined MG activity in cells cotransfected with a variety of pair combinations of nonfunctional L proteins containing mutations within the A and C motifs. None of the pair combinations of L mutant proteins tested resulted in detectable levels of polymerase activity as determined by MG-derived CAT activity (data not shown). This finding is consistent with results reported for the Sendai virus L protein, where intragenic complementation was not observed among mutants with mutations within domain III (38).
As an alternative approach to examining the possible L-L genetic interaction, we assessed whether mutant L proteins exhibited a dominant negative (DN) phenotype when coexpressed together with wt L. This approach has been successfully used to examine the oligomeric structures of recombinant NMDA receptors (20). For this examination, we cotransfected cells with a fixed amount (250 ng) of wt polymerase and increasing amounts (0 to 250 ng) of representative motif A and C mutants (not shown). To assess virus polymerase activity in cells cotransfected with wt- and mutant-L-expressing plasmids (at a ratio of 1:1), levels of MG-associated CAT activity (Fig. 3A) were quantified with a phosphorimager (Fig. 3B). The four mutants assayed at position 1182 in motif A exhibited a strong dose-dependent inhibitory effect on wt polymerase activity, a finding characteristic of mutants with a DN phenotype. Notably, the polymerase activity of wt L was reduced 20- to 30-fold when cotransfected with motif A mutant plasmids at a 1:1 ratio (Fig. 3, compare lane 2 with lanes 5 through 8). We observed similar results with motif C mutants with the mutations D1324A, D1324N, SD-GN, and S1323G (Fig. 3, lanes 9, 11, 13, and 14). In contrast, D1324E and D1324R motif C mutants did not appear to have a DN phenotype (Fig. 3, lanes 10 and 12).
![]() View larger version (44K): [in a new window] |
FIG. 3. (A) DN effect of LCMV L mutants. VVT7-infected cells were transfected with either wt L (0 to 500 ng) alone or combinations of wt (250 ng) and mutant (250 ng each) L proteins. Twenty-four hours later, cell lysates were prepared and analyzed for CAT activity. (B) Levels of CAT activity. For each sample, chloramphenicol conversion was quantified by phosphorimager analysis. The percentage of conversion obtained in the absence of L was considered to be nonspecific background. This background conversion value was subtracted from the conversion values obtained for each sample.
|
Physical interaction of LCMV L proteins. The strong DN effect exhibited by the majority of L mutants on the polymerase activity of wt L protein provided us with genetic evidence in support of the L-L interaction being required for the activity of the LCMV polymerase. Consequently, we sought direct evidence of whether the LCMV L protein also has, as with Sendai virus and PIV3 L proteins, the property of forming oligomeric structures. For this purpose, we generated LCMV L proteins containing two different tags, hemagglutinin (HA) and Flag, at their C termini. We first verified that both L-HA and L-Flag could be expressed to levels similar to those of wt L (Fig. 4A). As predicted, L-Flag was detected by the anti-Flag antibody but not the anti-HA antibody, and conversely, L-HA was detected by the anti-HA antibody but not the anti-Flag antibody (Fig. 4A). Moreover, both L-HA and L-Flag retained wt levels of polymerase activity as determined by their ability to support MG-derived CAT activity in the LCMV MG rescue assay (Fig. 4B). We then used a coimmunoprecipitation assay to examine whether L-HA and L-Flag interacted physically. For this assay, we prepared cytosolic extracts from cells transfected with either each tagged L protein alone or the tagged L proteins together. An analysis of total cell lysates by Western blotting using an antibody to HA showed that L-HA was detected when expressed either alone or together with L-Flag, whereas L-Flag expressed alone was not detected by the antibody to HA (Fig. 4C). We then used aliquots of cell extracts for immunoprecipitation with the anti-Flag M2 affinity gel freezer (A-2220; Sigma). Immunoprecipitated samples were analyzed by Western blot analysis using an antibody to HA (Fig. 4C). L-HA was detected in samples immunoprecipitated from lysates prepared from cells expressing both L-Flag and L-HA but not in the case of cells expressing only L-HA or L-Flag proteins. This result indicated the formation of L-HA/L-Flag complex in cells expressing both proteins, thus providing biochemical evidence of an L-L interaction. As with the Sendai virus L protein (36), the LCMV L-L interaction appears to require cotranslation of the interacting proteins based on the observation that L-Flag and L-HA did not coimmunoprecipitate if these proteins were provided by mixing lysates of cells transfected individually with each protein (data not shown).
![]() View larger version (21K): [in a new window] |
FIG. 4. Oligomerization of LCMV L protein. (A) BHK-21 cells infected with VVT7 (MOI = 3) were transfected with the indicated L constructs (2.5 µg) in M6 well plates. Twenty-four hours later, cytosolic extracts were prepared and analyzed by Western blotting using the indicated antibodies. (B) VVT7-infected BHK-21 cells were transfected with plasmids expressing MG no. 3 (0.5 µg), NP (0.75 µg), and the indicated tagged L proteins (0.5 µg). Twenty-four hours later, cell lysates were prepared and analyzed for CAT activity. NAc, nonacetylated chloramphenicol; MAc, monoacetylated chloramphenicol; DAc, diacetylated chloramphenicol. (C) LCMV L protein forms oligomers. BHK-21 cells were infected with VVT7 (MOI = 3) and then trans-fected with L-Flag, L-HA, or L-Flag and L-HA. Twenty-four hours later, cytosolic cell lysates were immunoprecipitated (IP) with anti-Flag antibody and the immunoprecipitated proteins were analyzed by Western blotting using an anti-HA antibody.
|
Arenaviruses include clinically important human pathogens that cause severe hemorrhagic fever (HF), such as the Lassa fever virus and the South American HF viruses (17). Moreover, because of the severe morbidity and high mortality it causes, the lack of immunization and an effective treatment, and its ease of introduction into a susceptible population, Lassa fever virus is included in category A of potential bioterrorism microbial weapons (5, 8). Therefore, the development of novel effective antiviral approaches to combat HF arenaviruses is important. Evidence suggests that mammalian cells are devoid of enzymes with strong similarities to the viral L proteins. Hence, it might be feasible to target these viral enzymes without significantly compromising the physiology of infected cells. The evidence that the L proteins of the paramyxoviruses Sendai virus and human PIV3, and now also the prototypic arenavirus LCMV, appear to function as oligomeric structures raises the possibility of developing antiviral strategies aimed at disrupting the formation of L-L complexes, which is predicted to abrogate virus RNA synthesis. A detailed characterization of the structural and functional domains of the arenavirus L proteins will help to develop such antiviral approaches to combat pathogenic arenaviruses.
We thank Sue A. Moyer for helpful comments and advice regarding the analyses of L-L physical interactions.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»