RNA sequence of astrovirus: distinctive genomic organization and a putative retrovirus-like ribosomal frameshifting signal that directs the viral replicase synthesis.

The genomic RNA of human astrovirus was sequenced and found to contain 6797 nt organized into three open reading frames (1a, 1b, and 2). A potential ribosomal frameshift site identified in the overlap region of open reading frames 1a and 1b consists of a "shifty" heptanucleotide and an RNA stem-loop structure that closely resemble those at the gag-pro junction of some retroviruses. This translation frame-shift may result in the suppression of in-frame amber termination at the end of open reading frame 1a and the synthesis of a nonstructural, fusion polyprotein that contains the putative protease and RNA-dependent RNA polymerase. Comparative sequence analysis indicated that the protease and polymerase of astrovirus are only distantly related to the respective enzymes of other positive-strand RNA viruses. The astrovirus polyprotein lacks the RNA helicase domain typical of other positive-strand RNA viruses of similar genome size. The genomic organization and expression strategy of astrovirus, with the protease and the polymerase brought together by predicted frameshift, most closely resembled those of plant leuteoviruses. Specific features of the sequence and genomic organization support the classification of astroviruses as an additional family of positive-strand RNA viruses, designated Astroviridae.     

Inactivation of Venezuelan Equine Encephalitis Virus Genome Using Two Methods

Venezuelan equine encephalitis virus (VEEV) is an Alphavirus in the Togaviridae family of positive-strand RNA viruses. The viral genome of positive-strand RNA viruses is infectious, as it produces infectious virus upon introduction into a cell. VEEV is a select agent and samples containing viral RNA are subject to additional regulations due to their infectious nature. Therefore, RNA isolated from cells infected with BSL-3 select agent strains of VEEV or other positive-strand viruses must be inactivated before removal from high-containment laboratories. In this study, we tested the inactivation of the viral genome after RNA fragmentation or cDNA synthesis, using the Trinidad Donkey and TC-83 strains of VEEV. We successfully inactivated VEEV genomic RNA utilizing these two protocols. Our cDNA synthesis method also inactivated the genomic RNA of eastern and western equine encephalitis viruses (EEEV and WEEV). We also tested whether the purified VEEV genomic RNA can produce infectious virions in the absence of transfection. Our result showed the inability of the viral genome to cause infection without being transfected into the cells. Overall, this work introduces RNA fragmentation and cDNA synthesis as reliable methods for the inactivation of samples containing the genomes of positive-strand RNA viruses.     

Replication protein of tobacco mosaic virus cotranslationally binds the 5′ untranslated region of genomic RNA to enable viral replication

Genomic RNA of positive-strand RNA viruses replicate via complementary (i.e., negative-strand) RNA in membrane-bound replication complexes. Before replication complex formation, virus-encoded replication proteins specifically recognize genomic RNA molecules and recruit them to sites of replication. Moreover, in many of these viruses, selection of replication templates by the replication proteins occurs preferentially in cis. This property is advantageous to the viruses in several aspects of viral replication and evolution, but the underlying molecular mechanisms have not been characterized. Here, we used an in vitro translation system to show that a 126-kDa replication protein of tobacco mosaic virus (TMV), a positive-strand RNA virus, binds a 5′-terminal ∼70-nucleotide region of TMV RNA cotranslationally, but not posttranslationally. TMV mutants that carried nucleotide changes in the 5′-terminal region and showed a defect in the binding were unable to synthesize negative-strand RNA, indicating that this binding is essential for template selection. A C-terminally truncated 126-kDa protein, but not the full-length 126-kDa protein, was able to posttranslationally bind TMV RNA in vitro, suggesting that binding of the 126-kDa protein to the 70-nucleotide region occurs during translation and before synthesis of the C-terminal inhibitory domain. We also show that binding of the 126-kDa protein prevents further translation of the bound TMV RNA. These data provide a mechanistic explanation of how the 126-kDa protein selects replication templates in cis and how fatal collision between translating ribosomes and negative-strand RNA-synthesizing polymerases on the genomic RNA is avoided.Virions of positive-strand RNA viruses contain genomic RNA of messenger sense. After infection, genomic RNA is released from the virions into the cytoplasm and translated to produce viral proteins, including viral RNA-dependent RNA polymerases and other replication-related proteins. These proteins are collectively called “replication proteins.” In eukaryotic positive-strand RNA viruses, replication proteins recruit genomic RNA to the cytoplasmic face of intracellular membranes to form replication complexes (1, 2). Negative-strand RNAs that are complementary to genomic RNAs are synthesized in the replication complexes, and then, using the negative-strand RNAs as templates, genomic RNA is copied and released into the cytoplasm. The recognition of template RNAs and their recruitment to the replication complexes are key processes in selective amplification of genomic RNA by positive-strand RNA viruses. In several positive-strand RNA viruses, cis-acting elements for replication-template selection have been identified, and, for some of them, it was demonstrated that replication proteins directly bind to these elements (3).Replication of tobacco mosaic virus (TMV), poliovirus, and many other positive-strand RNA viruses is cis-preferential: i.e., replication proteins recognize their own translation templates for replication (4–13). Because viral RNA replication is error-prone, it is important for viruses to selectively eliminate defective genomes. Template selection in cis is apparently advantageous in this regard because the genomes that encode replication proteins of lower performance are amplified less efficiently. Despite its importance in viral replication as well as evolution, little is known about how replication proteins select a template RNA in cis although it was proposed that requirement of nascent or newly synthesized replication proteins for replication and restricted diffusion or integrity of the proteins underlie the phenomenon (6).The genomic RNAs of positive-strand RNA viruses serve as templates for both translation and negative-strand RNA synthesis. During negative-strand RNA synthesis, viral RNA polymerases move along genomic RNA templates in a 3′-to-5′ direction. On the other hand, ribosomes synthesize viral proteins moving along the genomic RNA templates in a 5′-to-3′ direction. If these reactions take place on a single genomic RNA molecule at the same time, RNA polymerases and ribosomes collide, which results in the collapse of both reactions because these molecules cannot reverse direction or detach from the template RNA (14). Thus, positive-strand RNA viruses must clear ribosomes from the genomic RNA strands before negative-strand RNA synthesis occurs (15, 16).TMV belongs to the alpha-like virus superfamily of positive-strand RNA viruses. Its genome is a 5′-capped monopartite RNA and encodes at least four proteins, including the 5′ terminal 126-kDa protein, its translational read-through product of 183 kDa, a 30-kDa cell-to-cell movement protein, and a 17.5-kDa coat protein (17). The 126-kDa and 183-kDa proteins are replication proteins (18). The 126-kDa protein harbors a methyltransferase-like domain that is involved in RNA 5′ capping in its N-terminal region and a helicase-like domain in its C-terminal region. A region between these two domains is called the intervening region, or IR. The read-through part of the 183-kDa protein contains a polymerase-like domain (19). A deletion derivative of TMV RNA, named TMV126 RNA, that encodes the 126-kDa protein but not the 183-kDa protein can replicate when the 183-kDa protein is supplied in trans from a helper virus. However, TMV126 mutants that do not encode functional 126-kDa protein cannot replicate even if the wild-type 126-kDa and 183-kDa proteins are supplied in trans (8). This and other observations indicate that the 126-kDa protein functions primarily in cis (20, 21). The 5′ untranslated region (UTR) of TMV genomic RNA called Ω is ∼70 nucleotides (nt) in length, contains 12 CAA repeats, and is reported to have unusual tertiary structure with non-Watson–Crick base pairing (22, 23). The 5′ UTR of TMV RNA is a well-known translation enhancer (24, 25) and is essential for efficient virus multiplication (26). However, the role of the 5′ UTR in replication has been unclear, mainly due to the lack of experimental systems to separately evaluate translation of viral RNA and negative- and positive-stand RNA synthesis.To dissect the process that precedes the formation of the tobamovirus RNA replication complex on membranes, we previously developed an in vitro translation-replication system (27). Using an evacuolated tobacco protoplast extract (BYL) from which membranes were removed by centrifugation (membrane-depleted BYL, or mdBYL), we demonstrated that the replication proteins of tomato mosaic virus (ToMV), a close relative of TMV, bind ToMV RNA to form a ribonucleoprotein complex named premembrane-targeting complex (PMTC) in a translation-coupled manner (28). The PMTC is inactive in RNA synthesis but forms an active replication complex capable of synthesizing negative-strand and positive-strand RNA when it is mixed with membranes prepared from BYL. PMTC-like ribonucleoprotein (core-PMTC) is formed when a ToMV derivative that expresses the 126-kDa protein, but not the 183-kDa protein, is translated in mdBYL, which can form a replication complex when the 183-kDa protein and membranes are posttranslationally supplied (28). In the current study, we characterized tobamovirus PMTC and obtained results that provide insight into how the genomic RNA of TMV is selected as a template for replication preferentially in cis as well as how collisions between replication proteins and ribosomes are avoided.     

Identification and characterization of a host protein required for efficient template selection in viral RNA replication

Biochemical studies suggest that positive-strand RNA virus replication involves host as well as viral functions. Brome mosaic virus (BMV) is a member of the alphavirus-like superfamily of animal and plant positive-strand RNA viruses. Yeast expressing the BMV RNA replication proteins 1a and 2a supports BMV RNA replication and mRNA synthesis. Using the ability of BMV to replicate in yeast, we show that efficient BMV RNA replication requires Lsm1p, a yeast protein related to core RNA splicing factors but shown herein to be cytoplasmic. Haploid yeast with an Lsm1p mutation was defective in an early template selection step in BMV RNA replication, involving the helicase-like replication protein 1a and an internal viral RNA element conserved with tRNAs. Lsm1p dependence of this interaction was suppressed by adding 3' poly(A) to the normally unpolyadenylated BMV RNA. Our results show Lsm1p involvement in a specific step of BMV RNA replication and connections between Lsm1p and poly(A) function, possibly through interaction with factors binding mRNA 5' ends.     

Translation and replication of hepatitis C virus genomic RNA depends on ancient cellular proteins that control mRNA fates

Inevitably, viruses depend on host factors for their multiplication. Here, we show that hepatitis C virus (HCV) RNA translation and replication depends on Rck/p54, LSm1, and PatL1, which regulate the fate of cellular mRNAs from translation to degradation in the 5′-3′-deadenylation-dependent mRNA decay pathway. The requirement of these proteins for efficient HCV RNA translation was linked to the 5′ and 3′ untranslated regions (UTRs) of the viral genome. Furthermore, LSm1–7 complexes specifically interacted with essential cis-acting HCV RNA elements located in the UTRs. These results bridge HCV life cycle requirements and highly conserved host proteins of cellular mRNA decay. The previously described role of these proteins in the replication of 2 other positive-strand RNA viruses, the plant brome mosaic virus and the bacteriophage Qß, pinpoint a weak spot that may be exploited to generate broad-spectrum antiviral drugs.     

Systematic, genome-wide identification of host genes affecting replication of a positive-strand RNA virus

Positive-strand RNA viruses are the largest virus class and include many pathogens such as hepatitis C virus and the severe acute respiratory syndrome coronavirus (SARS). Brome mosaic virus (BMV) is a representative positive-strand RNA virus whose RNA replication, gene expression, and encapsidation have been reproduced in the yeast Saccharomyces cerevisiae. By using traditional yeast genetics, host genes have been identified that function in controlling BMV translation, selecting BMV RNAs as replication templates, activating the replication complex, maintaining a lipid composition required for membrane-associated RNA replication, and other steps. To more globally and systematically identify such host factors, we used engineered BMV derivatives to assay viral RNA replication in each strain of an ordered, genome-wide set of yeast single-gene deletion mutants. Each deletion strain was transformed to express BMV replicase proteins and a BMV RNA replication template with the capsid gene replaced by a luciferase reporter. Luciferase expression, which is dependent on viral RNA replication and RNA-dependent mRNA synthesis, was measured in intact yeast cells. Approximately 4500 yeast deletion strains ( approximately 80% of yeast genes) were screened in duplicate and selected strains analyzed further. This functional genomics approach revealed nearly 100 genes whose absence inhibited or stimulated BMV RNA replication and/or gene expression by 3- to >25-fold. Several of these genes were shown previously to function in BMV replication, validating the approach. Newly identified genes include some in RNA, protein, or membrane modification pathways and genes of unknown function. The results further illuminate virus and cell pathways. Further refinement of virus screening likely will reveal contributions from additional host genes.     

Computer-assisted assignment of functional domains in the nonstructural polyprotein of hepatitis E virus: delineation of an additional group of positive-strand RNA plant and animal viruses.

Computer-assisted comparison of the nonstructural polyprotein of hepatitis E virus (HEV) with proteins of other positive-strand RNA viruses allowed the identification of the following putative functional domains: (i) RNA-dependent RNA polymerase, (ii) RNA helicase, (iii) methyltransferase, (iv) a domain of unknown function ("X" domain) flanking the papain-like protease domains in the polyproteins of animal positive-strand RNA viruses, and (v) papain-like cysteine protease domain distantly related to the putative papain-like protease of rubella virus (RubV). Comparative analysis of the polymerase and helicase sequences of positive-strand RNA viruses belonging to the so-called "alpha-like" supergroup revealed grouping between HEV, RubV, and beet necrotic yellow vein virus (BNYVV), a plant furovirus. Two additional domains have been identified: one showed significant conservation between HEV, RubV, and BNYVV, and the other showed conservation specifically between HEV and RubV. The large nonstructural proteins of HEV, RubV, and BNYVV retained similar domain organization, with the exceptions of relocation of the putative protease domain in HEV as compared to RubV and the absence of the protease and X domains in BNYVV. These observations show that HEV, RubV, and BNYVV encompass partially conserved arrays of distinctive putative functional domains, suggesting that these viruses constitute a distinct monophyletic group within the alpha-like supergroup of positive-strand RNA viruses.     

Isolation and Identification of Inter-Species Enterovirus Recombinant Genomes

Positive-strand RNA virus evolution is partly attributed to the process of recombination. Although common between closely genetically related viruses, such as within species of the Enterovirus genus of the Picornaviridae family, inter-species recombination is rarely observed in nature. Recent studies have shown recombination is a ubiquitous process, resulting in a wide range of recombinant genomes and progeny viruses. While not all recombinant genomes yield infectious progeny virus, their existence and continued evolution during replication have critical implications for the evolution of the virus population. In this study, we utilised an in vitro recombination assay to demonstrate inter-species recombination events between viruses from four enterovirus species, A-D. We show that inter-species recombinant genomes are generated in vitro with polymerase template-switching events occurring within the virus polyprotein coding region. However, these genomes did not yield infectious progeny virus. Analysis and attempted recovery of a constructed recombinant cDNA revealed a restriction in positive-strand but not negative-strand RNA synthesis, indicating a significant block in replication. This study demonstrates the propensity for inter-species recombination at the genome level but suggests that significant sequence plasticity would be required in order to overcome blocks in the virus life cycle and allow for the production of infectious viruses.     

Translation of a nonpolyadenylated viral RNA is enhanced by binding of viral coat protein or polyadenylation of the RNA

On entering a host cell, positive-strand RNA virus genomes have to serve as messenger for the translation of viral proteins. Efficient translation of cellular messengers requires interactions between initiation factors bound to the 5'-cap structure and the poly(A) binding protein bound to the 3'-poly(A) tail. Initiation of infection with the tripartite RNA genomes of alfalfa mosaic virus (AMV) and viruses from the genus Ilarvirus requires binding of a few molecules of coat protein (CP) to the 3' end of the nonpolyadenylated viral RNAs. Moreover, infection with the genomic RNAs can be initiated by addition of the subgenomic messenger for CP, RNA 4. We report here that extension of the AMV RNAs with a poly(A) tail of 40 to 80 A-residues permitted initiation of infection independently of CP or RNA 4 in the inoculum. Specifically, polyadenylation of RNA 1 relieved an apparent bottleneck in the translation of the viral RNAs. Translation of RNA 4 in plant protoplasts was autocatalytically stimulated by its encoded CP. Mutations that interfered with CP binding to the 3' end of viral RNAs reduced translation of RNA 4 to undetectable levels. Possibly, CP of AMV and ilarviruses stimulates translation of viral RNAs by acting as a functional analogue of poly(A) binding protein or other cellular proteins.     

Striking similarities in amino acid sequence among nonstructural proteins encoded by RNA viruses that have dissimilar genomic organization.

The plant viruses alfalfa mosaic virus (AMV) and brome mosaic virus (BMV) each divide their genetic information among three RNAs while tobacco mosaic virus (TMV) contains a single genomic RNA. Amino acid sequence comparisons suggest that the single proteins encoded by AMV RNA 1 and BMV RNA 1 and by AMV RNA 2 and BMV RNA 2 are related to the NH2-terminal two-thirds and the COOH-terminal one-third, respectively, of the largest protein encoded by TMV. Separating these two domains in the TMV RNA sequence is an amber termination codon, whose partial suppression allows translation of the downstream domain. Many of the residues that the TMV read-through domain and the segmented plant viruses have in common are also conserved in a read-through domain found in the nonstructural polyprotein of the animal alphaviruses Sindbis and Middelburg. We suggest that, despite substantial differences in gene organization and expression, all of these viruses use related proteins for common functions in RNA replication. Reassortment of functional modules of coding and regulatory sequence from preexisting viral or cellular sources, perhaps via RNA recombination, may be an important mechanism in RNA virus evolution.     

Complementation between Sindbis viral RNAs produces infectious particles with a bipartite genome.

Sindbis virus, the type member of the alpha-viruses, is an enveloped virus containing a nonsegmented positive-strand RNA genome. We show that the nonstructural and the structural genes can function to produce infectious virus particles when they are expressed on two different RNA segments. The nonstructural genes are translated from an RNA in which the structural genes have been replaced by the chloramphenicol acetyltransferase gene [Xiong, C., Levis, R., Shen, P., Schlesinger, S., Rice, C. M. & Huang, H. V. (1989) Science 243, 1188-1191]. The structural genes are encoded in a defective-interfering RNA but are translated from a subgenomic RNA. Both segments contain the cis-acting sequences required for replication and packaging and are copackaged. This type of genome provides a model for an ancestral intermediate between alphaviruses and the multipartite positive-strand RNA viruses of plants. These different viruses show sequence similarities in their replicative proteins and are thought to have evolved from a common ancestor.     

Genomic RNA of an insect virus directs synthesis of infectious virions in plants.

Newly synthesized virions of flock house virus (FHV), an insect nodavirus, were detected in plant cells inoculated with FHV RNA. FHV was found in whole plants of barley (Hordeum vulgare), cowpea (Vigna sinensis), chenopodium (Chenopodium hybridum), tobacco (Nicotiana tabacum), and Nicotiana benthamiana and in protoplasts derived from barley leaves. Virions produced in plants contained newly synthesized RNA as well as newly synthesized capsid protein. These results show that the intracellular environment in these plants is suitable for synthesis of a virus normally indigenous only to insects. Such synthesis involves, minimally, translation of viral RNA, RNA replication, and virion assembly. Inoculation of barley protoplasts with FHV virions resulted in synthesis of small amounts of progeny virions, suggesting that FHV virions are capable of releasing their RNA in plant cells. In N. benthamiana, virions resulting from inoculation with RNA were detected not only in inoculated leaves but also in other leaves of inoculated plants, suggesting that virions could move in this plant species. Such movement probably occurs by a passive transport through the vascular system rather than by an active transport involving mechanisms that have evolved for plant viruses.     

Active complete in vitro replication of nodavirus RNA requires glycerophospholipid.

Flockhouse virus (FHV) is a member of the nodavirus group of positive-strand RNA viruses. In the absence of additional compounds, a template-dependent RNA-dependent RNA polymerase extracted from FHV-infected cells synthesizes complementary (-)-strand copies of added FHV RNA to yield a double-stranded RNA product. Upon addition of glycerophospholipid (GPL), this system reproducibly carries out complete highly active replication of added FHV RNA, producing newly synthesized (+)-strand RNA in predominantly single-stranded RNA form. This accounts for previously observed effects of Lipofectin (a mixture of GPL and cationic lipid) in the system. All tested neutral and negatively charged GPLs except phosphatidic acid support complete FHV RNA replication in this in vitro system, as do phospholipid extracts from uninfected and FHV-infected cells. Neither sphingomyelin, a membrane phospholipid that is not derived from glycerol, nor cholesterol supported FHV RNA replication. Testing of compounds derived from GPL shows that the ability of active GPL to support FHV (+)-strand RNA synthesis is dependent on the structures of both the head group and the acyl chains. Neither the phosphorylated head group nor the diacylglycerol lipid moiety alone supports RNA replication. The length and saturation of acyl chains strongly influence the ability of GPL to support RNA replication. Other characteristics of this in vitro RNA replication system and the possible role played by membranes and their components in FHV RNA replication are discussed.     

A -1 ribosomal frameshift element that requires base pairing across four kilobases suggests a mechanism of regulating ribosome and replicase traffic on a viral RNA

Programmed -1 ribosomal frameshifting is necessary for translation of the polymerase genes of many viruses. In addition to the consensus elements in the mRNA around the frameshift site, we found previously that frameshifting on Barley yellow dwarf virus RNA requires viral sequence located four kilobases downstream. By using dual luciferase reporter constructs, we now show that a predicted loop in the far downstream frameshift element must base pair to a bulge in a bulged stem loop adjacent to the frameshift site. Introduction of either two or six base mismatches in either the bulge or the far downstream loop abolished frameshifting, whereas mutations in both sites that restored base pairing reestablished frameshifting. Likewise, disruption of this base pairing abolished viral RNA replication in plant cells, and restoration of base pairing completely reestablished virus replication. We propose a model in which Barley yellow dwarf virus uses this and another long-distance base-pairing event required for cap-independent translation to allow the replicase copying from the 3' end to shut off translation of upstream ORFs and free the RNA of ribosomes to allow unimpeded replication. This would be a means of solving the "problem," common to positive strand RNA viruses, of competition between ribosomes and replicase for the same RNA template.     

Translational frameshifting mediated by a viral sequence in plant cells.

It has been proposed that the polymerase gene of barley yellow dwarf virus and related viruses is expressed by a ribosomal frameshift event during translation. The 5' end of this gene overlaps with the 3' end of an upstream gene that is in a different reading frame. The region of overlap is similar to sequences in retro- and coronaviruses that are known to express their polymerase genes by frameshifting. This overlap region includes a "shifty" heptanucleotide, followed by a highly structured region that may contain a pseudoknot. Sequences of 115 or 144 base pairs that span this region from barley yellow dwarf virus (PAV serotype) genomic RNA were introduced into a plasmid, so that a reporter gene could be expressed in plant cells only if a minus one (-1) frameshift event occurred. Frameshifting was detected at a rate of approximately 1%. This frameshifting was abolished when the stop codon at the 3' end of the upstream open reading frame was deleted. A sequence expected to form a strong stem-loop immediately upstream of the frameshift site was unnecessary for frameshifting, and initiation at AUG codons within the stem-loop appeared to be inhibited. Like viruses that infect hosts in other kingdoms, plant viruses also can induce frameshifting in translation of their genes.     

Recombinant vesicular stomatitis viruses from DNA.

We assembled a DNA clone containing the 11,161-nt sequence of the prototype rhabdovirus, vesicular stomatitis virus (VSV), such that it could be transcribed by the bacteriophage T7 RNA polymerase to yield a full-length positive-strand RNA complementary to the VSV genome. Expression of this RNA in cells also expressing the VSV nucleocapsid protein and the two VSV polymerase subunits resulted in production of VSV with the growth characteristics of wild-type VSV. Recovery of virus from DNA was verified by (i) the presence of two genetic tags generating restriction sites in DNA derived from the genome, (ii) direct sequencing of the genomic RNA of the recovered virus, and (iii) production of a VSV recombinant in which the glycoprotein was derived from a second serotype. The ability to generate VSV from DNA opens numerous possibilities for the genetic analysis of VSV replication. In addition, because VSV can be grown to very high titers and in large quantities with relative ease, it may be possible to genetically engineer recombinant VSVs displaying foreign antigens. Such modified viruses could be useful as vaccines conferring protection against other viruses.     

Characterization of a host protein associated with brome mosaic virus RNA-dependent RNA polymerase.

The association of host proteins with viral RNA replication proteins has been reported for a number of (+)-strand RNA viruses. However, little is known about the identity or function of these host proteins in viral replication. In this paper we report the characterization of a host protein associated with the RNA-dependent RNA polymerase (RdRp) from brome mosaic virus (BMV)-infected barley. A host protein was specifically and proportionally enriched with BMV RdRp activity through several purification steps. This RdRp-associated host protein reacted with an antiserum prepared against wheat germ eukaryotic translation initiation factor 3 (eIF-3). The RdRp-associated host protein, the p41 subunit of wheat germ eIF-3, and an antigenically related protein from rabbit reticulocyte lysates were all found to bind with high affinity and specificity to BMV-encoded protein 2a, which is involved in viral RNA replication. Moreover, addition of wheat germ eIF-3 or the p41 subunit from wheat germ to BMV RdRp gave a specific and reproducible 3-fold stimulation of (-)-strand RNA synthesis in vivo. These results suggest that the barley analog of eIF-3 subunit p41, or a closely related protein, associates with BMV RdRp in vivo and is involved in BMV RNA replication. This observation and the established role of translation factors in bacteriophage Q beta RdRp suggest that association with translation factors may be a general feature of RNA replication by (+)-strand RNA viruses.