key: cord-0035921-z0a85l4l
authors: nan
title: Replication of Plus-Sense Viral RNA
date: 2006
journal: Multiplication of RNA Plant Viruses
DOI: 10.1007/1-4020-4725-8_4
sha: 2fbaea116547a44089f1fba6be761d7078c55395
doc_id: 35921
cord_uid: z0a85l4l

nan

Viral RNA can not be categorized into any of the three RNA types (messenger RNA, transfer RNA, and ribosomal RNA) since it contains all the genetic information of a virus, can act as a template, can serve as a messenger and directs the synthesis of relevant virus-specific proteins in a host cell, and multiplies under its own direction. Viral RNA is thus a self-replicating polycistronic messenger, a messenger that contains genes for two or more proteins. Parental viral RNA, upon release from ribosomes, switches from translation mode (during RdRp synthesis) to replication mode, and so triggers its own replication in which several viral proteins are involved that are together called replication proteins. Many viral replication proteins have been identified and characterized.

Viral RNA replication (formation of progeny genomic RNA molecules identical to the original parental RNA molecule) always starts from 3`-end that must be unwound. Viral RNA replication begins with specific recognition of cis-acting RNA elements on the infecting viral positive-strand RNA by membrane-bound viral replicase and/or associated host factors (Lai, 1998; Ahlquist, 1996, 1999; Chen and Ahlquist, 2000; Boon et al., 2001; Ahlquist et al., 2003; Noueiry and Ahlquist, 2003; and Boguszewska-Chachulska and Haenni, 2005) . Recognition of a promoter at 3`-end of positive-strand RNA genome/template of the infecting virus by RdRp starts the replication cycle of viral RNA genome. Specific cis-and trans-acting nucleotide sequences and RNA secondary structures within 5`-termini of virus genomic RNAs are central to virus RNA replication. The compulsion of unwinding of 3`-end of viral RNA is apparently correlated with the fact that several RdRps have a narrow template channel that can only accommodate single-stranded RNA or have mechanisms that discriminate against the use of double-stranded templates for de novo initiation (Lesburg et al., 1999; Butcher et al., 2001; Ranjith-Kumar et al., 2003a) . Vlot et al. (2001 , while working on replication/synthesis of AMV RNAs 1 and 2, arrived at three very significant conclusions. One, the virus possesses a mechanism that coordinates replication of RNAs 1 and 2 when present together in a cell. AMV p1 and p2 proteins may interact in cis; this interaction is between helicase domain of p1 and N-terminal region of p2 (van der Heijden et al., 2001) and may play a role in coordinated replication of AMV RNAs 1 and 2, although other factors may also be involved in this process (Vlot et al., 2001 . Two, this mechanism also provides selectivity to the replicase in the use of particular viral template RNA. Replicase not only selects genome RNAs that are essential for production of progeny but also selects the fittest template from a population of variants of a given genome segment. Three, the same or some other identical mechanism permits AMV to optimize its replication strategy by replicating only those RNAs that are essential for producing progeny that is infectious to the type of cell in which it was produced.

Obviously, synthesis of progeny RNA molecules takes place in nuclei of the host cells so that plant viruses possess a signal for localization of viral RNA to nucleus. This nucleo-cytoplasmic shuttling has been dealt with in Chapter 3.

In increasing number of plant viruses, circularization of positive-strand genomic RNAs is involved in regulation of translation, replication and sgRNA synthesis. Circularization of RNAs occurs because of the long-distance RNA-RNA interactions between 5`-and 3`-terminal sequences. An interaction between the two termini [3`and 5`-ends] of the RNA appears to take place in many cases during RNA replication so that 3`-end sequence often can regulate RNA synthesis or translation initiated from 5`-end of RNA. The 3`-end sequence of BMV as well as of TMV RNAs does affect synthesis of positive-strand that begins from the 5`-end. Synthesis of BMV negativestrand RNA, which starts from 3`-end of RNA, requires a cis-acting enhancer sequence upstream of the subgenomic RNA promoter. A similar situation obtains in AMV in which an identical long-distance interaction between various RNA regions is required for initiation of subgenomic RNA synthesis (van der Vossen et al., 1995) . Two short elements on negative strands of satellite RNA C of TCV, one located at 11 bases from 3`-end and the other located 41 bases from 5`-end, are important for plus-strand RNA synthesis (Guan et al., 1997) . Qiu and Scholthof (2000) found that 263 nucleotides of 3`-UTR plus 73 nucleotides upstream of capsid protein stop codon and the first 16 nucleotides in 5`-UTR are required for RNA amplification and/or systemic spread of Satellite panicum mosaic virus (SPMV). In CPMV, both 3`-terminal cis-acting elements and a stem-loop upstream of poly(A) are important for replication of its RNA. The RNAs of BYDV have neither a cap structure nor a poly(A) tail, but longdistance base pairing between a stem-loop structure in 5`-UTR and a translation enhancer in 3`-UTR have been proposed to ensure circularization of RNAs and transfer provide some advantage to viruses because this ensures that only intact viral RNAs are used as templates for RNA replication and transcription. Communication between 5`cap and 3`-poly(A) tail, resulting in circularized mRNA, enhances translation of cellular mRNAs (Sachs, 2000) . Moreover, this leads to circularization of virus genome that greatly facilitates viral RNA replication.

Host specificity of a virus implies that the virus is able to undergo replication in a specific host and to induce disease symptoms by cell-to-cell and systemic movement within it. Thus, ability of a virus to infect a particular host (host specificity) and induce disease symptoms (symptomatology) by cell-to-cell and systemic movement (virus transport/movement) in that host is a clear demonstration of the ability of the particular virus to replicate in the concerned host. These topics are extremely complicated in which replication, symptom induction, and virus movement in the specific host are determined by genes from both virus and host. This gene-for-gene interaction in plant viruses was reviewed by Dardick and Culver (1999) . Only a couple of examples of initiation factors to 5`-end (Guo et al., 2001) . Possibly, 5`-and 3`-end interactions showing participation of virus genes/proteins in virus replication/host specificity/ symptomatology/virus transport are mentioned here.

Early work by genetic exchanges between BMV and CCMV demonstrated that BMV 1a protein controls some aspects of template specificity in RNA replication. The protein p19 of TBSV is involved in host-specific systemic invasion and symptom development (Scholthof, H. B. et al., 1995a (Scholthof, H. B. et al., , 1995b Chu et al., 2000) and is also active as suppressor of gene silencing (Qu and Morris, 2002; Qiu et al., 2002) . The protein p23 of Hibiscus chlorotic ringspot carmovirus (HCRSV) is involved in host determination so that this protein determines the host-specific replicative function in kenaf (Liang et al., 2002) . There could be any one or more of the following three reasons: p23 interacts with specific host factors to regulate host replicational machinery involved in virus replication; p23 could act as a cis-or trans-activator to either directly or indirectly regulate expression of host genes involved in viral replication; and p23 may interact with viral polymerase complex and so regulate virus replication (Liang et al., 2002) .

Certain facts about replication of TMV RNA have been established experimentally. One, interaction between 126-and 183-kDa proteins has been demonstrated (Goregaoker et al., 2001 ) and a purified TMV RdRp preparation containing a 1:1 dimer of 126-and 183-kDa proteins is able to synthesize negative-strand RNA templates in vitro (Watanabe et al., 1999) . Two, TMV-L (tomato strain) 126-kDa replication protein of RdRp complex and TMV-L RNA 3`-terminal region are actually bound and it is a region of 126-and 183-kDa replication proteins located downstream of the core and Buck, 2003) . Three, domain D2 and central core region of TLS are the most important elements for binding of TMV RdRp complex to viral 3`-terminal region (Osman et al., 2000) . Four, Osman and Buck (2003) show that two aromatic amino acids (at positions 409 and 416) in this region of 126-and 183-kDa replication proteins in tomato protoplasts. These aromatic amino acids may either directly interact with nucleic acid bases of 3`-terminal region of RNA or be essential for maintaining the structure of P314-423 region of 126-or 183-kDa protein that binds to 3`-terminal region. It is likely that P314-423 region binds to core C and D2 domain of 3`-terminal region. The interacting sites of 126-and 183-kDa proteins are located within a 110-amino acid region just downstream of core methyltransferase domain and a region comprising of central core C and domain D2 in 3`-terminal region. Possibly, binding of this region of 126-and 183-kDa proteins to TMV-L 3`-terminal region of RNA could be essential for virus RNA replication. Lewandowski and Dawson (2000) showed that TMV 126-kDa protein appears to function in cis whereas 183-kDa protein can function in trans; and suggested that 126-kDa protein binds to its mRNA and targets it for replication. The 183-kDa protein might then bind to 126-kDa protein and initiate replication. Subsequent binding of 183-kDa protein to already bound 126-kDa protein may position the catalytically active site of polymerase domain at 3`-terminus of template RNA to enable initiation of negative-strand synthesis. Possible, binding of polymerase to 3`-terminal CCA sequence may be relatively weak. Hence, the function of 126-kDa protein may be both to recruit RNA templates for replication and to subsequently bind 183-kDa protein to methyltransferase domain that binds RdRp to RNA 3` -terminal region in vitro (Osman are essential for binding to RNA 3` -terminal region and for replication of TMV RNA position the catalytically active site of polymerase domain close to 3`-end of template RNA.

From the negative-strand RNA synthesis stage onwards, three models have been proposed (Buck, 1996) for replication of RNA. Conceivably, different mechanisms could operate in different viruses or virus groups. RNA genomes of many positivestrand plant viruses, like cellular mRNAs, contain a 5`-cap structure and a 3`-poly(A) tail but several exceptions exist (Chapter 2). Replication of these exceptional RNAs may be different.

In first model, the nascent negative-strand RNA remains base-paired to the positive-strand RNA but only in the area where RdRp is bound to the template positive-strand RNA and is involved in active synthesis of negative-strand RNA. The 5` tail of the nascent negative-strand RNA is not base-paired to the template so that a free negative-strand RNA product is produced with the result that the positive-strand RNA template gets released. The polymerase recognizes a promoter at 3`-end of the negative-strand-strand RNA, which is then employed as a template for synthesizing a positive-strand RNA resulting in the formation of replicative intermediate (RI) . Here also, the nascent positive-strand RNA remains base-paired to the negative-strand RNA template only in the area at which polymerase is attached to template and is actively synthesizing RNA. Hence the RI is mostly single-stranded and is constituted by the full-length negative-strand RNA template to which several positive-strands of RNA are attached. The progeny RNA strands again are largely single-stranded. This model of RNA replication occurs in Qβ virus RNA and possesses certain characteristics: the replicase of Qβ RNA is a holoenzyme so that it catalyzes complete replication of viral RNA, the RIs are mainly single-stranded, and replicase is incapable of using Qβ dsRNA as template. This model presupposes the recognition of a single-stranded structure for initiating the synthesis of the positive-strand RNA.

However, the above model needs an explanation, which is not required for model 2 of RNA replication. How are the base pairs, formed at the time of RNA synthesis, almost immediately unwound by replicase complex? The possible explanation is -a helix-destabilizing protein, or a second molecule of helicase, suitably positioned in replication complex, could probably accomplish this. It is significant in this connection that the purified replication complexes of two plant viruses (BMV and CMV) seem to contain more of the respective 1a protein (which bears helicase-like domain) than 2a protein (which bears polymerase-like domain). Moreover, replication of BMV RNA is more sensitive to reductions in expression of 1a protein (that also has capping functions) than to reductions in expression of 2a protein. Hayes and Buck (1990) employed an in vitro system that was able to catalyze complete replication of CMV

RNA and detected free positive and negative strands in the ratio of 7:1 as also some ds RNA. This observation suggests that replication occurs according to model 1. Additionally, they found that dsRNA did not act as a template for replicase complex;

In third model, formation of dsRF RNA is exactly similar to that in model II. Synthesis of progeny positive-strand RNA occurs by employing the negative-strand of dsRNA as template. This synthesis takes place by transiently displacing the positive strand of dsRF RNA in the zone at which RNA is being synthesized. Synthesis of progeny positive-strands from a dsRF RNA is analogous to the conservative transcription of dsRNA that is characteristic of dsRNA viruses of families Reoviridae and Togaviridae. The RIs produced consist of double-stranded RNA that has one or several single-stranded tails. But unlike RIs in model 2, in which single-stranded tails nascent, incomplete progeny positive RNA strands.

However, unequivocal distinction between the 'open' and 'closed' models (models I and II, respectively) of RNA replication for any eukaryotic positive-stranded RNA virus is not generally possible at present. Then, cellular and viral helicases cannot act on completely double-stranded structures. This means that unwinding of a completely dsRF RNA possibly requires another protein or helicases to bind to singleis bound to the negative template of the unwound RF RNA) most probably unwinds the duplex subsequently. Viral helicase of PPV (Lain et al., 1990) and RdRp complex of AMV (de Graaff et al., 1995) have the required strand displacement property. Model 2 requires recognition of a double-stranded structure for initiation of positivestrand synthesis (and for subgenomic RNA synthesis for those viruses that utilize subgenomic promoters), whereas model 1 requires recognition of a single-stranded structure and thus model 2 appears to be more operative in plant viruses.

stranded regions close to the duplex region to be unwound and a 3`-to 5` -helicase (that however, this may also mean that replicase lacks an essential component required for initiating synthesis on a dsRNA template.

In second model, the negative-strand RNA synthesized stays base-paired with the positive-strand RNA template. This produces the replicative form RNAs (RF RNAs), which are initially partly double-stranded and partly single-stranded but finally become completely dsRNA. Thus, the RdRp recognizes a promoter at end of RF RNA containing 3`-end of negative-strand and 5`-end of positive-strand. Synthesis of progeny positive-strand RNA molecules starts by employing the negative-strand as template. This happens by a strand displacement method that displaces the negativestrand from the dsRF RNA. The negative strands, by acting as templates, produce the RIs constituted by dsRNA with one or, by repeated reinitiation, several single-stranded 5` tails of full-length positive-strands of RNA. Thus, multiple progeny positive-strands of RNA are synthe-sized by repeated reinitiation and are then released. The first fulllength released positive-strand is the original template RNA strand. The above manner of synthesis of progeny positive strands from a RI RNA is analogous to the semiconservative transcription of dsRNA by strand displacement mechanism. This mechanism is characteristic of the dsRNA viruses of families Birnaviridae, Cystoviridae, and Partitiviridae. This model presupposes that the unwinding of dsRF RNA occurs at 3`-end of the negative-strand prior to the beginning of RNA synthesis and that a ds structure has to be recognised for initiation of positive-strand RNA synthesis (and for synthesis of subgenomic RNA in viruses that utilize subgenomic promoters).

are the displaced 5`-tails of full-length positive RNA strands, these tails are of the Several experiments support the participation of negative-strand RNA during replication of positive-strand RNA viruses. An isolated RdRp complex of AMV, BMV and CMV could use the negative-strand template to produce full-length and subgenomic RNA. Transgenic plants that express double-stranded RNA-specific ribonuclease of plant virus (Tomato mosaic virus, CMV, and PVY) exhibit resistance (although incomplete) to that particular virus (Watanabe et al., 1995) . This indicates the formation of double-stranded replication intermediate structures during replication of these plant viruses.

Replication of Plus-Sense Viral RNA required for initiation of minus-strand RNA synthesis , which is also the start of RNA replication process of positive-strand RNA viruses. Various structures and cis sequences located at 3`-end serve as promoters for minus-strand RNA synthesis (Buck, 1996) . These include tRNA-like structures/sequences (TLSs) in family Bromoviridae and genera Tobamovirus, Tymovirus, and Hordeivirus (Dreher, 1999) . Plant viruses lacking a TLS contain various other conserved elements that play a role in initiation of RNA synthesis. For example, the satellite RNA C (satC) of TCV contains a stable 3`-terminal stem-loop (SL) structure, which is indispensable for its synthesis (Song and Simon, 1995) . TCV is associated with several dispensable noncoding RNAs including satC (356 bases) and satD (194 bases). The core promoter for synthesis of satC minus strands is identified as a 3` terminal SL flanked by the sequence (CCUGCCC-OH), which is also found at 3`-end of TCV genomic RNA and satD (Song and Simon, 1995; Stupina and Simon, 1997; Carpenter and Simon, 1998) . Transcriptional repressors, leading to repression of minus-strand RNA synthesis, exist in some plant viruses (Zhang et al., 2004a (Zhang et al., , 2004b Sun et al., 2005) and are discussed later.

The 5`-terminal bases of satC are involved in minus-strand initiation in vitro. Zhang et al. (2004a) proposed that the 5`-end interacts with sequences in or near 3` hairpin, stabilizing the structure of the promoter and permitting the RdRp to properly recognize 3`-end when derepressor relieves repression. This suggests involvement of 5`-guanylates in minus-strand synthesis. Work on a variety of viral genomic RNA and subviral RNA replicons also support a role for 5`-end in minus-strand synthesis, through protein-protein bridges or direct RNA-RNA interactions or proposed 5`stabilization of 3`-elements (Wu et al., 2001; .

Minus strands are synthesized early in infection. They are mostly single-stranded in tissues involved in viral RNA replication but are predominantly part of the duplex RF RNA on completion of replication. Level of minus strands stays constant during the phase of rapid TMV synthesis. The ratio of plus to minus strands at the end of this borne mosaic potyvirus is most abundant in tissues found along the periphery of that the peripheral cells were most recently infected and so involved in active viral In general, RNA viruses have a specific structure at the 3`-end of the genome that is infected area but occurs in much smaller amounts within the infected area -indicating stage is highly variable depending upon the virus. The minus-strand RNA of Pea seed-RNA replication (Wang and Maule, 1995) . Besides the viral template, three other inputs are essential for synthesis of negative-strand RNA: the TLS, promoter element, and RdRp. These are discussed below in some of the plant viruses. Olsthoorn and Bol (2002) and Olsthoorn et al. (2004) found that the 3`-terminal 145 nucleotides (that are common between the three AMV RNAs) can adopt a TLS and act as a core promoter for initiation of negative-strand RNA synthesis at 3`-end of template. They identified a single stem-loop structure, the triloop hairpin E (hpE), as the negative-strand RNA promoter, which is the most important and essential element for negative-strand RNA synthesis in vitro while presence of B, C, and D hairpins is required for optimum negative-strand synthesis. The proposed major role of TLS is to enforce initiation of transcription by polymerase at the very end of genome. The structure of hpE is conserved in all AMV isolates sequenced so far and features a 10-base pair stem, a 4-nucleotide bulge and a UGG triloop. The UGG triloop sequence is not essential for in vitro transcription but may be playing some role in vivo. The negative-strand promoter hpE and subgenomic RNA promoter hairpin of AMV are equivalent in binding viral polymerase, share many common features and are essential for performing their respective assigned function. A similar stem-loop structure can be formed at the homologous position in 3`-UTR of RNAs of about all ilarviruses. In PDV RNA1 and RNA3, a triloop, that strongly resembles hpE, can be folded. In apple mosaic virus RNA3 and prunus necrotic ringspot virus RNA3, a pentaloop hairpin is Structures similar to AMV negative-strand promoter hpE have been identified in (Sivakumaran et al., 2000) . These are conserved structures and are required for their respective negative-strand synthesis. Stem-loop structure C (SLC) at 3`-end of BMV RNA consists of 11 base pairs, a triloop, a 4-nucleotide bulge and thus resembles the AMV hpE. The SLC of several isolates of CMV consists of 13 base pairs, a 5`-nucleotide bulge, and a triloop; but variants with a pentaloop do also exist (Sivakumaran et al., 2000) . The 5`-UTRs of BMV RNAs contain stem-loop structures with loop sequences resembling box B elements homologous to T C stem-loop of cellular tRNAs. These elements are required for negative-strand RNA synthesis (Chen et al., the only conserved sequences between BMV and CMV stem-loop structures. BMV RdRp can specifically recognise the 5`-most A in AUA triloop that is involved in the so-called adenine motif. This feature is unlike AMV RdRp, which seems to be insensitive to loop sequence but can probably recognise specific base pairs in the stem. Thus, related viruses have evolved different strategies to recognise their negativestrand promoter hairpins.

The BMV 2a polymerase-like replication protein binds to BMV 1a helicase-like replication protein and initiates negative-strand RNA synthesis at 3`-terminus TLS (Sullivan and Ahlquist, 1997; Chen and Ahlquist, 2000) . The 2a protein is directed to ER by the viral protein 1a, which also recruits RNAs 2 and 3 templates for replication

The 3`-terminal region of TMV (strain L) RNA contains all cis-acting sequences needed for negative-strand synthesis, at least in vitro, and can be folded into a TLS and

present at this location in 3` -UTR.

3` -TLS of BMV (Chapman and Kao, 1999; Sivakumaran et al., 2003) and CMV RNAs (Sullivan and Ahlquist, 1997; Chen and Ahlquist, 2000) .

; Schwartz et al., 2002) . The top C-G base pair of SLC and A of the 5`-loop are nearby upstream pseudoknots that are important for TMV RNA replication in vivo and

The 3`-terminal elements important for negative-strand synthesis in vitro include the CCA sequence domains D1 (equivalent to a tRNA acceptor arm), D2 (similar to a tRNA anticodon) and D3 (an upstream pseudoknotted region) and a central core region C that connects domains D1, D2, and D3. Cheng et al. (2002) have propounded a working model for negative-strand RNA synthesis of BaMV. (a) The RdRp interacts with 3`-UTR [including the potexviral conserved hexamer motif (ACNUAA)] and about 20 nucleotides of poly(A) sequence immediately downstream of 3`-UTR (Huang et al., 2001) , which may be used to Replication of Plus-Sense Viral RNA Wang and Wong (2004) found that synthesis of HCRSV minus-strand RNA is initiated opposite the 3`-terminal two C residues at the 3`-end in vitro and in vivo and that the 3`-terminal CCC nucleotides have essential role in minus-strand RNA synthesis because minus-strand RNA initiation begins at 3`-terminal two Cs. This is similar to the results obtained from BMV (Chapman and Kao, 1999) , TMV (Osman and Buck, 1996) , and TYMV (Singh and Dreher, 1997) , in which initiation of minusstrand RNA synthesis also starts opposite the two Cs while minus-strand RNA synthesis is reduced upon removal or substitution of terminal A in the CCA box. In TYMV RNA, the 3` TLS present the CCA-3` in a conformation that is easily accessible to the replicase (Dreher, 1999) . The CCC-3` may also be involved in forming a conformation for replicase access both in vitro and in vivo in HCRSV (Wang and Wong, 2004) . However, the presence of a CCC-3` terminal sequence alone is not sufficient for RNA synthesis as is evident from below. located within the 3`-terminus 87 nucleotides of HCRSV plus-strand RNA are also essential for minus-strand RNA synthesis (Wang and Wong, 2004) . The first SL is very similar to the single SL that is a minimum promoter for minus-strand RNA synthesis of TCV satellite RNA C (Song and Simon, 1995) . This suggests that the two predicted SLs in HCRSV might also play a similar role. The U loop located on SL2 is an essential structure for RNA synthesis in HCRSV. The secondary structure of SLs in 3`-UTR of RNA viruses is required for protein binding (Lai, 1998) . For example, the hpE loop of AMV is not essential for RNA synthesis, whereas the stem and base pairing of the lower triloop are essential (Olsthoorn and Bol, 2002) . Similarly, failure to bind replication factors by HCRSV mutants, with disruption or deletion of SL1 or SL2, rendered them unable to initiate minus-strand synthesis. Disruption of certain SLs in BMV (Chapman and Kao, 1999) , TYMV (Deiman et al., 1997) , AMV (Olsthoorn et al., 1999) , and CTV (Satyanarayana et al., 2002a) resulted in reduction in minusstrand RNA synthesis, but not in complete inhibition of RNA synthesis as in HCRSV. In TCV 3`-UTR, elements located hundreds of nucleotides upstream of 3`-UTR, were needed for efficient replication (Carpenter et al., 1995) . Taken together, the specific sequence CCC at the 3`-terminus and the two SL structures located in the 3`-UTR are essential for efficient minus-strand RNA synthesis in HCRSV.

negative-strand RNA synthesis in vitro (Osman et al., 2000; Chandrika et al., 2000) .

In addition to the CCC-3` terminal sequence, two putative stem-loops initiate minus-strand RNA synthesis. In vivo experiments support this and show that about 13 to 25 adenylates following 3`-UTR are involved in maintaining the integrity protoplasts (Cheng et al., 2002) . (b) After interaction of RdRp with 3`-UTR of BaMV RNA, the RdRp is suggested to initiate minus strand RNA synthesis opposite one of the adenylates within the pseudoknot. This is done perhaps selectively in the loop region of the pseudoknot, which is the most favoured position for initiation. (c) The negative strands are then used as templates to synthesize progeny RNAs. These 90 to 170 adenylates, as in BaMV virion RNAs. This different-length short stretch of As would be at the very end of genomic RNA due to the variation of initiation sites of negative-strand RNA synthesis. Therefore, the enzyme functional in polyadenylation must recognise these short stretches of poly(A) tailed genomes and maintain structural integrity of virion RNA.

The above model has some notable features. First: Certain exceptional features exist during negative-strand RNA synthesis in some plant viruses. The minus-strand RNA synthesis was found to initiate from several positions within poly(A) tail of BaMV so that Cheng et al. (2002) deduced that initiation site for negative-strand RNA synthesis is not fixed at one position but resides opposite one of the 15 adenylates of poly(A) tail immediately downstream of 3`-UTR of genomic RNA. Thus, multiple putative initiation sites of BaMV negative-strand RNA synthesis exist and any one of the adenylates could be used to initiate negative-strand RNA.

Second: The poly(A)-tailed RNA of viruses is comprised of hundreds of adenylates so that it might be difficult for RdRp, bound to 3` UTR as in BaMV (Huang et al., 2001) , to initiate negative-strand RNA at the end of poly(A) tail hundreds of nucleotides downstream. It is, therefore, regarded that a positive-strand RNA virus with a poly(A)-tailed genome would use the poly(A) tail as a template to initiate negative-strand RNA synthesis. However, only a short stretch of poly(A) sequence connected to 3` UTR is necessary for initiation.

Fourth: The results of Osman and Buck (2003) are compatible with a model for initiation of TMV-L negative-strand RNA synthesis in which an internal region of TMV-L 126-kDa protein first binds to central core C and domain D2 region of TMV-L RNA 3`-terminal region and is then followed by binding of 183-kDa protein to this complex and positioning of catalytically active sites of polymerase domain close to 3`terminal CCA initiation site.

Fifth: It is clear from above that both BMV and TMV appear to have evolved similar mechanisms for recruitment of RNA templates and initiation of negative-strand RNA synthesis. In membrane-bound BMV replication complexes, ratio of 1a to 2a proteins is about 25:1 and since 1a protein forms spherules, which bud into functional roles in assembling membrane-bound replication complexes and sequestering 2a polymerase and BMV RNA templates within them (Schwartz et al., 2002) . Similarly, TMV 126-kDa protein is present in much larger amounts than 183-kDa protein in

of pseudoknot structure and are required for efficient viral RNA replication in Third: Both BMV and TYMV initiate their negative-strand RNA synthesis opposite the penultimate cytidylate residue. endoplasmic reticulum, it is suggested that 1a protein plays both structural and progeny RNAs must be good templates for subsequent polyadenylation by up to about insulated membrane-bound replication complexes (Osman and Buck, 1996; Watanabe et al., 1999) and may play a role in assembling TMV replication complexes and recruiting 183-kDa protein and RNA template that is similar to the role of BMV 1a protein in assembling BMV replication complexes. Oligomerization of helicase domain of TMV 126-and 183-kDa proteins is reported. A difference in mechanisms of sequestering RNA templates between the two viruses is that, unlike BMV, there is no recognizable tRNA-like T C stem-loop structure at 3` terminus or internally in TMV positive-strand RNA. Hence, TMV 126-kDa protein binds directly to 3`-terminal TLS.

Two diametrically opposite views have been expressed about the existence, significance and importance of dsRNAs: that they exist and participate in viral RNA replication and that they do not exist, are only artifacts and consequently do not perform any role in RNA replication. Thus, from the first review on dsRNAs (Ralph, 1969) till date, their nature and importance in viral RNA replication is still not finally settled (Buck, 1996) . Satyanarayana et al. (2002a) state that a long-standing conundrum of virology has been the question of the in vivo relationship of complementary positive and negative RNA strands: whether they exist in a double-stranded helix in the cell or only become double-stranded during extraction. This problem still exists but virologists have decided to let it be and have moved on.

Early studies involving pulse chase experiments and kinetics of labeling (on BMV, CPMV, TMV, and several other plant viruses) showed that the label could be chased from RF and RI RNAs into full-length genomic sized single-stranded RNA molecules so that RF and RI RNAs act as progenitors for the synthesis of complementary and progeny RNA strands, respectively, of the virus concerned. The in vitro 32 P incorporation in TMV RF and RI RNAs showed that their synthesis and synthesis of progeny RNA stops at about the same time. Double-stranded RNA of TomRSV and of certain other plant viruses has been isolated in high yields only from leaves in which virus was increasing due to its multiplication. The switch-over from RF synthesis to RI synthesis (that is, from symmetric positive strand and minus strand synthesis to asymmetric plus strand progeny RNA synthesis) is controlled by genomic AMV RNA3. Although the nature and significance of dsRNAs in viral RNA replication is still debatable, yet their detection in virus infected cells continues to be reported -by Ritzenthaler et al. (2002) in GFLV-infected cells, by Dunoyer et al. (2002) in modified vesiculated ER in cells infected by PCV, and by still others. Transgenic plants expressing double-stranded RNA-specific ribonuclease showed resistance (but incomplete) to Tomato mosaic virus, CMV and PVY (Watanabe et al., 1995) indicating the formation of double-stranded RNA as the replication intermediate that Each genomic RNA of a multipartite virus replicates through its own dsRNA. A regulatory mechanism appears to govern the rates at which each species of genomic RNA of a multicomponent virus is synthesized relative to others. In wild type CCMV, even as the rate of RNA synthesis changes greatly during infection period, relative

does play some part during RNA replication.

rates of synthesis of the three genomic RNAs remain constant as also that of the corresponding dsRNAs. This ratio is approximately two RNA3 molecules produced to one molecule each of RNA1 and RNA2. Thus, more RNA3 is synthesized than can be encapsidated so that it accumulates in infected cells to a higher level than other viral RNAs. Similarly, CMV RNA3 and its corresponding RF RNA occur predominantly in CMV strain Y-infected protoplasts. Moreover, a dsRNA species, which corresponds in size to subgenomic RNA of RCNMV, is found in infected tissue (Sit et al., 1998) .

The RF RNAs are full-length double-stranded structures composed entirely of base pairs, are never infectious unless denatured to release the viral genome strand, and are usually found in nucleic acid preparations from cells infected with positive-strand RNA viruses or in extracted products of in vitro RNA synthesis conducted in presence of crude or partially purified polymerase preparations. The 3` termini of minus strands of RF RNAs are exactly complementary to the corresponding 5` sequences of the RNA and so form a perfect duplex with 5`-ends of virion-sense RNA. This is even true of both the middle and bottom component RNAs of CPMV so that a duplex (with 5`-end of virion-sense RNA of each of the two components) is formed. Moreover, terminal sequences of both CPMV RF RNAs are identical; this may have some relevance in replication of this multicomponent virus.

The RF RNA in TomRSV-infected leaf tissue is positively correlated with the concentration of virus specific RdRp and the concentration of virus particles. Moreover, amounts of RF RNA and RdRp actively increase just prior to and during the period of rapid virus synthesis. Polymerase activity falls rapidly with cessation of virus synthesis followed by a similar decline in the amount of RF RNAs. The RF RNA accumulates with the increase in concentrations of viral RNA in the infected cells as in Q virus in which model 1 of RNA replication has been well established.

The nature and significance of RIs in the replication of plus-stranded RNA is less clear (Buck, 1996) . The RIs in infected cells are branched chains and are partly doublestranded. Each RI is usually composed of one complete full-length negative RNA strand to which are bound several partially synthesized positive strands that are being elongated and are in different stages of formation. RI RNAs are formed because of the simultaneous production of many complementary plus strands, which are displaced one after another from the same negative-strand RNA template molecule. The isolated RI RNA of Poliovirus is constituted by a full-length negative-strand RNA template to which 6-8 nascent positive RNA strands (in different stages of elongation/formation) are attached. Similar RI RNAs have also been detected in virus-infected plants. Tracer techniques have established that RI is the direct precursor of viral RNA. Presence of RIs is accounted for by the semi-conservative mechanism of RNA replication.

The RI RNAs are suggested to be mainly single-stranded structures like the RIs of TYMV detected in vivo. The in vivo double-stranded RNAs could be the dead-end products formed by annealing of positive and negative strands of the infecting RNA. Annealing of the two RNAs may also occur during extraction of viral nucleic acids from infected cells during deproteinization by phenol and are regarded as isolation artifacts formed during RNA extraction (Garnier et al., 1980) . The dsRNA extracted from in vitro RNA synthesis systems may also be produced in the same manner or may form as a result of inadequacies in in vitro systems. However, dsRNA may be formed in vivo in late stages of infection.

The minus RNA strands, released from the double-stranded RF RNA, serve as templates for synthesis of complementary plus strands of viral RNA (progeny RNA) from the 3`-end. The total number of minus-strand templates available for synthesis of the progeny plus RNA strands is limited so that the former have to behave as templates for synthesis of plus strands over and over again. This could occur in two ways: the semiconservative and conservative methods.

In the asymmetric semiconservative mechanism, the minus strand remains constant while the first plus strand of double-stranded RNA is elbowed out by the second plus strand, the second by the third and so on. Thus, there is multiple simultaneous formation of progeny plus RNA strands, which are at different stages of completion and appear to hang as tails. Up to five partially completed plus strands may be associated with RI. In conservative mechanism, the double-stranded structure remains conserved but its secondary structure is disrupted at the replication point to permit the copying to proceed along the minus strand. Experimental evidence favours the synthesis of progeny positive RNA strands through semiconservative asymmetric mechanism. Large amounts of plus strands are synthesized as compared to the minus strands; thus, synthesis of the two strands is asymmetric and proceeds at different rates. Correct positive-strand BMV RNA initiation requires 5`-terminal sequences and a nontemplate guanylate added to 3`-end of negative-strand RNA in vitro .

RNA templates for plus-strand and minus-strand RNA synthesis most likely possess different sequences near initiation site. These RNAs initiate with a pyrimidine (but a cytidylate in BMV). Bromoviral templates for positive-strand RNA synthesis are rich in A or U nucleotides in contrast to the templates for minus-strand RNA synthesis (Chapman and Kao, 1999; . This is significant because plusstrand RNA synthesis by BMV polymerase is more efficient in case the template contains an A/U-rich sequence near initiation site . Hema and Kao (2004) found that mutations at positions adjacent to initiation cytidylate in templates for genomic and subgenomic plus-strand RNA synthesis significantly decreased RNA accumulation and that different requirements exist for template sequence near initiation nucleotide for BMV RNA accumulation in plant cells and that an A/U-rich sequence is preferred for accumulation of subgenomic RNA.

On this basis, Hema and Kao (2004) hypothesized that an A/U-rich template sequence regulates level of RNA accumulation. The nucleotide position of the first C or G in template controls the relative amount of the four BMV RNAs: RNA4 (10) > RNA3 (7) > RNA2 (4) > RNA1 (2). The second and third nucleotides of genomic pluset al., 1998) . They derived three conclusions pertaining to BMV RNA replication: template sequence does not significantly affect the minus-strand RNA synthesis; specific identities of +2A and +3U are required for plus-strand RNA synthesis; and an A/U-rich sequence is needed for efficient subgenomic RNA synthesis. BMV genomic plus-strand and subgenomic RNA formation needs highly specific nucleotides at three positions: +1C, +2A, and +3U in the template-sense RNA.

The replication process is usually asymmetric so that 10-to 1000-fold more of positive RNA strands over negative RNA strands are produced. Plus-strand viral RNAs contain sequences and structural elements that permit cognate RdRp to correctly initiate and transcribe asymmetric levels of plus and minus strands during RNA replication. This asymmetry is characteristically found in all positive-strand RNA viruses investigated. Molar ratio for positive-strand:negative-strand progeny RNA production of about one is achieved, during transfection of barley protoplasts with BMV RNAs 1and 2 (but in the absence of RNA3); however, >100 progeny plus-strands of progeny RNA molecules are synthesized for every negative-strand progeny RNA in the presence of RNA3. The presence of RNA3 increases the total plus-strand RNA production by over 200-fold and that of RNAs 1 and 2 by about 30-fold. The ratio of production of positive-to negative-strand genomic RNAs is 10:1 for flaviviruses (animal viruses); 40 to 50:1 for CTV (Satyanarayana et al., 1999 (Satyanarayana et al., , 2002b ; 50 to 100:1 for coronaviruses (animal viruses); 100:1 for BMV and TMV; and 1000:1 for AMV. Moreover, the ratios of positive-to et al., 2002b). Asynchronous accumulation of RNAs 1 and 2 of Lettuce infectious yellows virus (LIYV) also occurs (Yeh et al., 2000) . Mechanism that controls ratio of positive-to negative-strands of genomic RNAs, leading to strand asymmetry due to asymmetric virus RNA replication, has been investigated. In BMV, RNA3 controls this process. The subgenomic RNA promoter and upstream inter-cistronic sequences present in central region of RNA3 ensure switch-over to asymmetric RNA replication. These two types of switches act in trans. The intercistronic region is adjacent to ICR2 1100 (ICR2-like sequence present at nucleotide 1100), exerts trans-acting effect on BMV replication, and is essential for positive-strand RNA3 amplification as well as for asymmetry of BMV replication. Possibly, host factors bind to the intercistronic sequence and/or the subgenomic

negative-strand sgRNAs in wild-type CTV is estimated to be 10 to 20:1 (Satyanarayana strand and subgenomic RNAs of all bromoviruses have A and U, respectively (Adkins promoter to profoundly affect the replicase activities and thereby affect strand asymmetry. In AMV, a frameshift in capsid protein gene results in a 100-fold reduction in positive-strand RNA accumulation but a 3-to 10-fold increase in negative-strand accumulation indicating that capsid protein is involved in strand asymmetry in favour of positive RNA strands. Satyanarayana et al. (2002b) found that protein 23 (p23) gene of CTV regulated asymmetric accumulation of positive-and negative-stranded subgenomic RNAs; that this protein mainly down-regulated the accumulation of negative-stranded sgRNAs and caused only a modest increase in accumulation of positive-stranded sgRNA; that amino acid residues 46 to 180, which contained RNA-binding and zinc finger domains, were indispensable for asymmetric RNA accumulation while N-terminal 5 to 45 and Cterminal 181 to 209 amino acids residues were not required; that zinc finger is possibly involved in asymmetric RNA accumulation; and that the excess negative-stranded sgRNA reduces the availability of corresponding positive-strand sgRNA as a messenger. Thus, p23 protein serves as a switch to convert replication from symmetrical to asymmetrical production of positive-and negative-stranded RNA of both genomic and sgRNAs by down regulating negative-strand RNA accumulation. López et al. (2000) demonstrated that CTV p23 gene product binds RNA in vitro and RNA-binding domain mapped between amino acid residues 50 to 68, which include the putative zinc finger domain.

Other regulatory factors favouring asymmetric synthesis of positive-strand RNA are: the ability of an RNA genome to compete for the limited polymerase, characteristic features of RNA synthesis like the processivity of polymerase, frequency of initiation by polymerase, capability of the polymerase to transition out of initiation mode, and frequency of template switch (Hema and Kao, 2004) .

Rates of TMV RNA synthesis, TMV protein synthesis, replicase synthesis, and formation of TMV particles have been studied in infected tobacco leaves, inoculated tissue culture and inoculated protoplasts and generally closely parallel each other. Synthesis of RNA increases almost linearly during the first 60 hr period but declines subsequently. In fact, RNA synthesis in infected protoplasts begins in less than 4 hr post-inoculation (hpi), becomes exponential and remains so until 8 hpi, drops sharply and becomes linear at about 10 hpi and then continues to increase linearly. Synthesis of single-stranded and double-stranded RNAs (RF and RI RNAs) also nearly follows the same curve: they are first detected at 6-8 hpi, increase exponentially till about 20 hpi after which their increase becomes linear. Maximum synthesis of doublestranded RNA occurs at 18-20 hpi and that of single-stranded RNA at 24-26 hpi. About 20 minutes are required for synthesis of a TMV RNA molecule during exponential stages of replication. The synthesized viral RNA is quickly incorporated into viral protein to form virions. First progeny TMV appears in cytoplasm of mesophyll protoplasts 6 hr after inoculation. TMV particles are formed exponentially during early stages of virus replication but linearly during the late stages. TMV virion formation is exponential during 50-80 hpi but declines to become linear subsequently. Large number of virus particles is produced. Similarly, replication of Pea seed-borne mosaic potyvirus is rapid (Wang and Maule, 1995) . So is of TEV, which infects approximately one new cell every two hours (Dolja et al., 1992) .

accumulate to maximum high levels by 24 hpi. The ORF2 subgenomic RNA was most abundant of all the LIYV RNAs. In contrast, RNA2 progeny (positive-and negativesense RNA2) was detected only between 24-36 hpi or at about 36 hpi. The genomic and subgenomic RNA of HCRSV could be detected in infected kenaf protoplasts as early as 6 hpi (Liang et al., 2002) . In all plant viruses, appearance of virus particles is closely related with synthesis of capsid protein.

In pulse-chase experiments, TEV polyprotein appeared 5 minutes after initiation of chase, processed NIa protein appeared 10 minutes after initiation of chase while proteolysis of polyprotein occurred with a half time of about 20 to 30 minute after the chase (Restrepo-Hartwig and Carrington, 1992) .

The terms replication 'promoter' (that promotes or initiates replication) and 'enhancer' (that enhances or increases replication) are often employed in the literal sense. Therefore, replication promoters and replication enhancers are similar in their ability to bind RdRp but are different in their abilities to support initiation of RNA synthesis. Replication promoters are present in several plant viruses like CNV (Panavas et al., 2002a (Panavas et al., , 2002b and other plant viruses mentioned in the text. The secondary structure of stem-loop structure (called SL1-III or hairpin) of replication promoter region III are important for recognition and/or binding by tombusvirus (TBSV) RdRp. The box B consensus sequence element in 5`-UTRs of BMV RNAs 1 and 2, and also in cucumovirus RNAs, is GGUUCAANNCC where N is any possible nucleotide . These stem-loop structures have loop sequences that resemble box B elements homologous to T C stem-loop of cellular tRNAs and act as promoters of negative-strand synthesis (Schwartz et al., 2002; Chen et al., 2001) . Negative-strand RNA synthesis from BMV RNA3 templates requires 1a and 2a replication proteins and is driven by a promoter in the 3` TLS of RNA3 (Ishikawa et al., 1997; Chapman and Kao, 1999) . In turn, negative-stand RNA3 acts as a template for positive-strand RNA3 synthesis and sgRNA transcription and is driven by promoters adjacent to their BMV RNA3 acts as a template for RNA replication as well as for sgRNA transcription, which is initiated internally on negative-strand RNA3 templates. Both these processes (positive-strand RNA3 synthesis/replication and sgRNA transcription

genomic (both positive-and negative-sense RNAs) and subgenomic RNAs (including Yeh et al. (2000) report that simultaneous inoculation of bipartite LIYV resulted in asynchronous accumulation of progeny LIYV RNAs. LIYV RNA1 progeny subgenomic RNA of ORF2-encoded protein p32) were detected in protoplasts 12 hpi and respective initiation sites . from RNA3) depend upon common viral replication factors and RNA templates. Then how are they coordinated? It is noteworthy that GTP is the priming nucleotide for synthesis of negative-strand RNA3, positive-strand RNA3, and sgRNA. Synthesis of all these three types of RNAs requires stem-loop RNA secondary structures for recognition of corresponding promoters (Haasnoot et al., 2002; Kao, 2002) . The promoters for replication and sgRNA transcription have different nucleotide sequences (Kao, 2002; Ranjith-Kumar et al., 2003b) , which suggest that recognition of these varied promoters may involve distinct domains of viral and/or host factors. An alternative or additional theory is the 'induced fit' mechanism, which anticipates that viral replicase may adjust to different promoters (Stawicki and Kao, 1999; Williamson, 2000) .

Replication enhancers generally occur in viral RNA minus-strands, may not be proximal to the core promoter, contain sequence and/or structural features of core promoters, and can promote transcription in the absence of sequences resembling the transcription site (Nagy et al., 1999; Ray and White, 2003) . Replication enhancers have been detected in several plant viruses like AMV (van Rossum et al., 1997) , TCV (Nagy et al., 1999) , TBSV , and must be existing but not yet detected in many other plant viruses. Ray and White (2003) detected in vivo a replication enhancer in TBSV and found that the 5`-proximal segment of region III is a modular RNA replication element that functions mainly by forming an RNA hairpin structure (a stem-loop structure SL2-III(-)] in region III play interchangeable roles in enhancement of tombusvirus RNA synthesis. No additive stimulatory effect comes from combination of the two stem-loops. Region SL1-III contains a replication enhancer that functions as a strong enhancer in negative-stranded RNA (that is, acts as a strong replication enhancer for synthesis of positive-strand in vitro) and a weak enhancer in positive-strands (for formation of negative strands of RNA) in tombusviruses . propose that the putative binding of RdRp to replication enhancer of TBSV may facilitate correct positioning of RdRp over 3`-end of template including CCU initiation site. In the absence of proper initiation site, replication enhancer cannot support RNA synthesis.

TCV and its related carmoviruses and its associated satellite C RNA contain replication enhancer elements (Nagy et al., 1999) . The motif 1 replication enhancer RNA element of TCV is believed to function in negative strand synthesis by recruiting viral RdRp (recruiting role), thereby facilitating its initiation at positive strand promoter -similar to the possible TBSV SL1-III(-) function in an analogous capacity. A recruiting role in negative strand would require subsequent delivery of RdRp to positive strand promoter at 3`-end of template. Such long-distance interactions could be mediated by either a protein bridge or RNA-RNA interactions; the latter have been demonstrated for activation of subgenomic RNA transcription in TBSV (Zhang et al., 1999; Choi and White, 2002) . In contrast to TBSV, replication enhancer in genomic TCV RNA shows an additive effect on RNA synthesis. A single-stranded region between SL1-III(-) and SL2-III(-) hairpins is required for full enhancer activity.

protein; BNYVV-encoded protein p14; and PCV RNA1-encoded p15. It is significant that BSMV B, BNYVV p14, and PCV p15 proteins belong to a group of cysteine-rich proteins, and BNYVV p14 shares statistically significant similarity with other nucleic acid-binding proteins. These cysteine-rich proteins enhance or influence replication of all genomic components of their respective viruses. In contrast, CPMV 58-kDa protein is a template-selective replication enhancer and is needed for replication of M RNA but not for B RNA; so it is a cis replication enhancer. RCNMV genomic RNA2 is a transcriptional enhancer and functions in trans for synthesis of RCNMV RNA1 subgenomic RNA (Sit et al., 1998) . Carla-, furo-, hordei-, and tobraviruses contain 3`-proximal genes for small cysteine-rich proteins, some of which have RNA binding activity. One of the functions of these proteins is to regulate synthesis of capsid protein, which in these viruses is encoded by 5`-proximal genes. These cysteine-rich proteins are not absolute requirements for RNA replication but do affect replication in some cases.

Replication/transcriptional repressors (also called transcriptional silencers) for negative-strand RNA synthesis have been detected in members of family Tombusviridae in the last couple of years. They are located on plus-strands just upstream from the core promoter. The presence of transcriptional (replicational) enhancers and repressors on opposite strands (on minus-and plus-RNA strands, respectively) suggests that these elements regulate asymmetric levels of plus-and minus-strand RNA synthesis .

A repressor performs negative regulation of minus-strand synthesis and has been identified in TCV and its satellite RNA satC (Zhang et al., 2004a (Zhang et al., , 2004b Zhang and Simon, 2005) , TBSV and predicted for BYDV (Koev et al., 2002) . Zhang et al. (2004a Zhang et al. ( , 2004b and Zhang and Simon (2005) worked extensively on the repressor of TCV and satC. Members of genus Carmovirus, family Tombusviridae, contain a structurally conserved 3`-proximal hairpin (H5) with a large 14-base internal symmetrical loop (LSL). The H5 is proximal to the core promoter and functions as a repressor of minus-strand synthesis in vitro through an interaction between LSL and 3` terminal bases in TCV satellite RNA satC (which has partial sequence similarity with its helper virus Turnip crinkle virus). Of the 14-base satC H5 LSL, specific sequences in the middle and upper regions on both sides of the LSL are necessary for robust TCV as well as of satC accumulation in plants and protoplasts. LSL and lower stem of H5 were of greater importance for satC accumulation (i.e., multiplication) than the upper stem .

Repression of minus-strand synthesis by H5 was due to sequestration of the 3`-end from RdRp through interactions by base pairing between the 3`-terminus and LSL; in fact between four of the seven bases on the 3` side of the LSL (5`GGGC) and the satC 3` terminal bases (GCCC-OH). A second sequence, located in a single-stranded region upstream of H5, is predicted to function as a derepressor by disrupting the 3` end-H5 interaction (Zhang et al., 2004b) . Possibly, the LSL might be involved in other processes (like satC fitness and permitting robust replication) in addition to repression of minus-strand synthesis. The minus-strand repressor in TBSV (SL3) contains an internal loop with both similarities and striking differences with carmovirus H5. In contrast to the symmetrical or nearly symmetrical H5 large internal loops of all

carmoviruses, the TBSV SL3 is asymmetrical, with only a single adenylate occupying similar, with five of eight SL3 bases (GGGCU) identical to their carmoviral counterpart.

The mechanism of template selection by cognate replicases for viral RNA replication in plus-strand RNA viruses is poorly understood, so is the contribution of viral and cellular proteins to RNA template recognition. Two types of experimental evidence have been published in this connection. The first type shows that, in infected cells, viral replicases specifically recognize and selectively replicate their cognate viral RNAs from a heterogeneous pool of cellular RNAs. Thus, virus-encoded proteins facilitate selective template recruitment to viral replicase complex in vivo as is true of the 1a protein of BMV and the 126-kDa protein of Tomato mosaic virus (Sullivan and Ahlquist, 1999; Chen et al., 2001; Osman and Buck, 2003) . In contrast, many purified viral replicase complexes effectively utilize heterologous promoter or initiation elements during in vitro studies (Yoshinari et al., 2000; Kao, et al., 2001; Rajendran et al., 2002) . These conflicting in vivo and in vitro results were explained on the basis of the suggestion that host factors are involved in selective recognition of viral templates by the relevant RdRp (Buck, 1996) . Even in the two above-mentioned cases of specific template selection by viral RdRps, it is uncertain whether the respective viral RNAs are recognised directly by these replicase proteins or some assistance is provided by the host proteins (Diez et al., 2000) . This issue appears to be settled at least in TBSV by Pogany et al. (2005) , on which this section is mainly based. The TBSV p33 and p92 replication proteins are essential for RNA replication; are part of viral replicase complex; accumulate in vivo in 20:1 ratio, respectively (Scholthof, K. B.-G et al., 1995; Panaviene et al., 2004) ; and less plentiful p92 functions as viral RdRp, while role of the more abundant p33 was not known (White and Nagy, 2004) although it was regarded to play some unknown essential but auxiliary role. Pogany et al. (2005) demonstrated that recombinant p33 replicase protein binds specifically in vitro to a conserved internal replication element (IRE) located within the p92 RdRp coding region of viral genome; specific recognition and binding of p33 to IRE RNA element in vitro depended on the presence of a C . C mismatch within a conserved RNA helix; binding of p33 to RNA depended on the presence of the conserved RII(+)-SL hairpin, which serves as an internal recognition element; and strong correlation existed between p33:IRE complex formation in vitro and viral replication in vivo, so that mutations in IRE that disrupted selective p33 binding in vitro also abolished TBSV RNA replication both in plant and in Saccharomyces cerevisiae cells. They proposed that p33:IRE interaction provides a mechanism to selectively recruit viral RNAs into cognate viral replicase complexes, specifically binds the cognate viral RNA template in vitro, directs viral template recruitment into replication mode; that the major RNA element recognized by p33 is a C 99 . C 143 mismatch in an internal loop within the RII(+)-SL; and proposed that one of the 5`-side . However, the 3`-side of SL3 and H5 internal loops are 99 . C 143 Pogany et al. (2005) proposed a model, on the bases of their in vitro and in vivo studies and the results of Monkewich et al. (2005) , for the central role for the RII(+)-SL, p33, and possibly p92 act together at an early step in replication to facilitate model for explaining the occurrence of this perceived process. The p33 and p92 proteins are translated from the genome soon after infection; the RII(+)-SL is possibly formed in the genome transiently (Monkewich et al., 2005) ; p33 interacts productively with RII(+)-SL when a required p33 threshold concentration is reached; and recognition . is dependent on protein dimerization or oligomerization (which may also involve p92); and requirement of oligomerization suggests that the interface of the protein subunits the membrane targeting signals present in the N terminus of p33 and p92 (Rubino and Russo, 1998) . Recruitment to membranes could also down-regulate the translation process [as in Brome mosaic virus (Janda and Ahlquist, 1998) ]. In contrast, subgenomic mRNAs transcribed during infection would not be recruited to replication complexes since they lack RII(+)-SL, and so would remain dedicated to translation. In this model, the selective binding of p33 (and/or p92) to RII(+)-SL-containing RNAs is proposed to be the primary factor in vivo for the observed specificity of Tombusvirus RNA replication. In short, the essential role of the p33:p33/p92 interaction domain in selective RNA binding suggests that intermolecular interaction between two or more p33s (and/or possibly p92) proteins results in the formation of an RNA-binding pocket that has high specificity for the C . C mismatch within RII(+)-SL.

All genera in Tombusviridae encode comparable replicase proteins and all the sequenced members of the genus Tombusvirus are predicted to form RII(+)-SL-like structures with the C . C mismatch. Thus, the above mechanism and conclusions may be relevant to other members of Tombusviridae so that perhaps the ability of p33 replicase protein to specifically recognize a C . C mismatch is conserved in all tombusviruses (Pogany et al., 2005) . Other viral proteins with arginine-rich RNAbinding domains also use somewhat similar mechanisms to recognize cognate RNAs. It may be mentioned that RII(+)-SL is not the only essential RNA element for tombusvirus RNA replication; other elements also required for tombusvirus RNA replication are: a replication silencer element and the minusstrand initiation promoter (Panavas et al., 2002a; Fabian et al., 2003) .

Viral capsid protein does not have any role in replication of majority of plant viruses (including alpha-, bromo-, and tombusviruses) but has a definite role during the replication of AMV, ilarviruses and in some other plant viruses. Probably, capsid could form an RNA-binding pocket that in turn specifically recognizes the C . C mismatch. The RII(+)-SL:p33 complex formed is then transported to membranes via of the C C mismatch is critical for specific binding of p33 to the RNA template; template selection and co-recruitment of replication factors and proposed a multistep SL:p33 interaction in Tombusviridae RNA replication. They proposed that the RII(+)-the functions of the C mismatch is to open up the central helix in RII(+)-SL to facilitate binding or positioning of the arginine-proline-arginine-motif (RPR-motif) of p33.

protein of AMV and ilarviruses confers a competitive advantage to viral RNAs over polyadenylated cellular mRNAs. The role of capsid protein in AMV is the best investigated and reviewed (Jaspars, 1999; Bol, , 2003 .

The capsid protein is associated with RNAs of AMV and ilarviruses and is invariably present in purified RdRp preparations from AMV-infected plants. The capsid protein peptides and the 3`-terminal nucleotides of AMV RNAs, which interact with each other, are the N-terminal amino acid numbers 25, 26 or 38 of the capsid protein and the 39 nucleotides that constitute the minimal coat protein-binding site at the 3`-end of viral RNAs (Baer et al., 1994; Houser-Scott et al., 1994a , 1994b . The 3`terminal 145 nucleotides of the AMV RNAs (genomic RNAs 1, 2, and 3, and subgenomic RNA4) are homologous and can adopt two mutually exclusive and alternative conformations: TLS or a linear array of hairpins (Olsthoorn et al., 1999; Bol, , 2003 . This is a unique property of 3`-UTRs of AMV and ilarviruses.

The linear array of hairpins consists of a series of hairpins A to E with flanking AUGC motifs 1 to 4, which possess high-affinity capsid protein binding site. Conceivably, AUGC repeats are sequence-specific determinants for capsid protein/RNA interactions so that AUGC motifs in viral RNAs are involved in capsid protein binding. Hairpin D contains a sequence UCCU in its loop. This sequence can potentially base pair with sequence AGGA located in loop and stem of hairpin A to form a pseudoknot that creates a structure which strongly resembles TLS of bromoviruses and cucumoviruses.

The 3`-UTR of AMV RNA3 bears at least two high-affinity capsid proteinbinding sites. Site 1 is situated in 3`-end 39 nucleotides and is made up of hairpins A and B and AUGC motifs 1, 2, and 3. The AMV capsid protein amino acids required for binding to site 1 are the N-terminal sequence of 25 amino acids; arginine at position 17 is particularly critical for binding. capsid protein-binding site 2 exists outside the homologous region of 145 nucleotides and is constituted by hairpins F and G and AUGC motifs 4 and 5. However, relevance of capsid protein binding site 2 in RNA3 is not clear. Major capsid protein-binding sites also exist in internal positions of the three AMV RNAs and they can also form stable stem-loop structures. Thus, nucleotides U844, C846, and A877 are important for capsid protein binding (Ansel-McKinney and Gehrke, 1998; Rocheleau et al., 2004) . These sites can be involved in virus assembly. Different CP domains are possibly involved in binding of capsid protein to internal and 3`-terminal sites of AMV RNAs (Neeleman et al., 2004) .

The capsid protein is involved in several functions in plant viruses -encapsidation of virus RNA, cell-to-cell virus movement, long-distance virus movement, virus

This AMV TLS is formed by a pseudoknot interaction between nucleotides 5 to 8 and 90 to 93 from 3`-terminus so that genomic RNAs, under physiological conditions, are postulated to be mainly in pseudoknotted configuration (Olsthroon et al., 1999) . The TLS cannot be aminoacylated but is required for and acts as promoter for minus-strand RNA synthesis. Thus, the 3`-end of RNAs of AMV and ilarviruses has a dual roleformation of minus-strand RNA and binding of capsid protein.

transmission by vectors, and in still other functions. Different amino acids of capsid protein are involved in these multifarious functions. This ensures that the CP motifs involved in these functions are independent of other sequences so that any one of these functions can be mutated without influencing other functions. However, RNAassociated CP found in AMV and ilarviruses is an exceptional case because of its intimate involvement in virus RNA replication. Besides, CP is involved in some replication step of TMV; it regulates formation of replication complexes (Asurmendi et al., 2004) . Moreover, genome-associated CP may also prevent collision between translating ribosomes and replicase molecules synthesizing negative-strands. CP taking part in AMV life cycle can be categorized into two types -CP that is originally present in viral inoculum since it is associated with viral RNAs; and CP that is synthesized from viral subgenomic RNA4 and appears later in viral life cycle. Possibly these CPs have different functions during viral life cycle so that many functions have been proposed for CP (Bol, , 2003 Jaspars, 1999) .

The CP, originally attached to viral RNAs in inoculum, induces genome activation for initiating AMV infection and results in synthesis of genomic and subgenomic RNAs followed by translation of these RNAs (Jaspars, 1999; Bol, , 2003 Choi et al., 2003) . The role of CP in translation of inoculum RNAs is supported by recent data and consists of following sequence of events according to Bol (2003) .

First step: After uncoating of virus particles of inoculum, a few molecules of CP remain bound to the high-affinity binding site 1 at 3`-end of viral RNAs. Binding of CP to site 1 is essential for efficient translation of AMV RNAs in vivo.

Capsid protein bound to termini of AMV RNAs enhances translation of viral RNAs by acting as a functional analogue of poly(A)-bound protein (PABP), by promoting recruitment of 40S ribosomal subunits and/or by enhancing stability of viral RNAs and by interacting with host translation initiation factors (eIF4G and other factors) bound to 5`-cap structure and stimulates translation of RNAs 1 and 2 leading to formation of replicase proteins p1 and p2 Bol, 2003) . Krab et al. (2005) found that AMV CP interacts specifically with eIF4G and eIFiso4G subunits from wheat eIF4F and eIFiso4G, respectively, so that their results support the hypothesis that the role of CP in translation of viral RNAs mimics the role of PABP in translation of cellular mRNAs and converts viral RNAs into closed-loop structures. Simultaneously, CP could stop formation of RNA structure that acts as negative-strand promoter for preventing collision between translating ribosomes and replicase molecules. By analogy to BMV, AMV P1 could bind to ICR2 motifs present in AMV

RNAs and are essential for replication of AMV. Efficient translation of AMV RNA4 in plant cells is dependent on the ability of CP to bind to 3`-end. In fact, both CP gene and 3`-UTR are necessary for in vivo translation of AMV RNA4 .

Third step: Protein P1 is proposed to recruit viral RNAs from translation machinery and also targets a complex of P1, P2 and viral RNAs to membrane structures where replication complexes are formed -similar to the targeting mechanism employed by BMV (den Boon et al., 2001) . The BMV replication complexes are located within endoplasmic reticulum-derived vesicles while AMV replication complexes are located within vesicles derived from tonoplast.

In vesicles, CP possibly dissociates from 3`-terminus of inoculum RNAs to permit the formation of TLS and of negative-strand RNA promoter and binding to hairpin E in negative-strand RNA promoter. Presence of TLS (more specifically, formation of pseudoknot structure) at 3`-end permits initiation of AMV negative-strand RNA synthesis, and is essential for viral replication in vivo and in vitro. Thus, CP has no role in formation of AMV negative-strand RNA.

Fourth step: The AMV progeny RNA positive strands are synthesized. The or for synthesis of positive-strand RNA in vitro, have been identified. The natural negative-strand RNAs, acting as templates for synthesis of positive-strand RNA molecules in vivo, contain a single non-coded G residue at 3`-end (Houwing et al., 2001) . The strong stimulation to synthesis of positive-strand RNA molecules by CP could be more due to protection of viral RNA from degradation than to the actual stimulation of RNA synthesis (Vlot et al., 2001) . Thus, at present no proof exists about the role of CP in synthesis of positive-strand viral RNAs.

of AMV RNAs contain high-affinity CP-binding sites but are not required for encapsidation. Thus, the origins of assembly in AMV RNAs still need to be identified and CP bound to 3`-termini of AMV RNAs plays no role in assembly of virus particles (Vlot et al., 2001) . The CP also binds to several internal sites in AMV RNAs and these sites may have some as yet unknown role in virus assembly. Whole of the above process can be briefly summarised as below. The CP binds to each of three genomic RNAs in inoculum; the RNAs are then targeted to membranebound replication complexes where each RNA is transcribed into the corresponding negative strand. Capsid protein is not involved in this transcription process but, after complete synthesis of the negative strand, CP associates with replicase to make the enzyme complex competent for synthesis of the progeny positive-strand RNAs. The CP bound to AMV and ilarvirus RNAs most likely gives them some competitive advantage over the cellular mRNAs since it behaves like PABP, interacts with host translation factors bound to 5`-cap structure of viral genome, and stimulates translation of viral RNAs. The AUGC motif is suggested to be part of a translational determinant involved in genome activation process.

CP, originally attached to AMV viral RNAs in inoculum, protects 3`-end of viral positive-strand inoculum (mRNAs) from exonucleolytic degradation (Neeleman and

Fifth step: The final step is the assembly of complete virus particles. The 3` -UTRs sequences, in 5` -UTR of AMV RNAs, which are required for RNA replication in vivo Vlot et al., 2001) . The bound CP molecules do not allow the loss of 3`terminal nucleotides of positive-strand viral RNAs during their repeated translation cycles. This is similar to the function of the host proteins, bound to TLS or poly(A) tails located at 3`-termini of RNA genome of other viruses, for protecting the 3`-end against exonucleolytic degradation during translation process of viral RNAs.

CP is proposed to target genomic RNAs to membrane-bound replication complexes . AMV CP may stabilize the replication complex, consisting of viral RNAs/viral replication proteins P1 and P2 and the host translation initiation factors, and also promote the recruitment of 40S ribosome subunits by forming a closed-loop configuration. Dimer formation is possibly essential for the putative interaction of CP with translation initiation factors (Neeleman et al., 2004) .

AMV CP may promote circularization of viral RNAs analogous to the function of PABP in circularization of cellular RNAs. A protein-protein bridge between initiation factors (bound to cap structure) and CP (bound to 3`-end) of AMV RNAs is proposed to convert these RNAs into a closed-loop structure that is essential for translation of RNAs . However, the 3`-terminal sequences in AMV RNAs, that could potentially base pair with 5` sequences involved in negative-strand RNA synthesis, have not yet been identified .

AMV CP in inoculum is not involved in negative-strand RNA synthesis (de Graff et al., 1995; Neeleman and Bol, 1999; Vlot et al., 2001) on the infecting positivestrand viral RNAs both in vivo and in vitro; on the contrary, CP binding to 3`-end of AMV RNA disrupts a conformation of 3` UTR that is required for minus-strand RNA promoter activity in vitro leading to inhibition of negative-strand RNA production due to interference with pseudoknot formation (Olsthoorn et al., 1999; Neeleman et al., 2001) . This indicates that AMV CP is required in a step prior to viral negative-strand synthesis, possibly during translation of inoculum RNAs (Neeleman and Bol, 1999; Olsthoorn et al., 1999) .

The CP, derived from parental virions or obtained from translation of viral RNA4 in inoculum, could possibly have a dual role -in promoting translation of viral RNAs and in preventing premature initiation of minus-strand RNA synthesis by newly synthesized replication proteins. Later, in infection cycle, the de novo synthesized CP will result in increased CP levels that may in turn cause encapsidation of viral RNAs into the progeny virus particles .

Presence of AMV CP in inoculum causes an approximately 100-fold increase of positive-strand RNA accumulation in infected protoplasts (var der Kuyl et al., 1991) . This strong stimulation to synthesis of positive-strand RNA molecules by CP could be more due to protection of viral RNA from degradation than to the actual stimulation of RNA synthesis (Vlot et al., 2001) . Thus, at present no proof exists about the role of CP in synthesis of positive-strand viral RNAs.

The 3`UTR and CP bound to 3`-termini of AMV RNAs is not required for encapsidation (Vlot et al., 2001) . But CP also binds to several internal sites in AMV RNAs and these sites may act as the origin of assembly of virus particles. This is supported by the fact that a coding sequence of BMV RNA 1 is required for encapsidation whereas 3`-UTRs of RNAs are not needed (Duggal and Hall, 1993) .

AUGC 865-868 sequences are conserved among RNAs of AMV and ilarviruses. Rocheleau et al. (2004) suggest that CP binding to AUGC sequences determines the orientation of the 3` hairpin relative to one another, while local structural features within these hairpins are also critical determinants of their functional activity; that activation of AMV replication is dependent on both the initial fold of the CP-free RNA and the final RNA fold established through viral CP binding so that the mechanism of AMV genome activation centers on CP-mediated RNA conformational changes that organize 3` terminus of RNA for replication function.

All functions (like genome encapsidation, virus transmission, etc.), normally performed by CP of a virus, are also performed by AMV CP. In addition, this CP performs several other functions. AMV CP is required for viral RNA positive-strand synthesis in vivo as well as in vitro; is proposed to release viral positive-strand RNAs from the membranebound replication complexes that contain the negative-strand template (Houwing et al., 1998) ; is implicated during synthesis of positive-strands of RNAs 3 and 4; enhances translation of viral mRNAs like binding of CP to 3`-UTRs of RNAs 3 and 4 strongly enhanced translation of these RNAs ; and binding of newly synthesized CP to 3`-end of plus-strand RNAs most likely shuts off minus-strand synthesis triggering a switch to asymmetric synthesis of plus-strand viral RNAs (Olsthoorn et al., 1999; Neeleman et al., 2001) . The CP acts in trans for synthesis of positive-strand RNAs 1 and 2 but acts in cis for synthesis of the positive-strand RNA3.

Olsthroon et al. (1999) postulated that presence of TLS or a linear array of hairpins, the mutually exclusive and alternative conformations, at 3`-termini of RNAs of AMV and ilarviruses may be the switch that enables viral RNA to shift from translation to replication mode and vice versa. The linear array of stem-loop structures contains several high-affinity CP-binding sites while tRNA-like conformation is specifically recognised by RdRp. It is proposed that 3`-UTR acts as a molecular switch that regulates the transition from translation to replication of parental RNAs. In this model, CP bound to inoculum RNAs would force the 3`-UTR into the CP-binding conformation to enhance translation and/or to prevent premature initiation of negativestrand RNA synthesis. Subsequently, CP dissociates from parental RNA to allow TLS formation and initiation of negative-strand RNA synthesis.

Later, targeting parental AMV RNAs to chloroplasts membranes (where replication complexes are assembled) could result in dissociation of 3` terminally attached CP molecules. This collapses the pseudoknot with consequent shut-off of minus-strand RNA synthesis and start of progeny positive-strand RNA synthesis. This is the trigger that starts asymmetric synthesis of positive-strand viral RNAs. In this way, CP has a regulatory function during synthesis of progeny positive-strand RNAs.

In the later step of replication, de novo synthesized CP could shut-off negative-strand synthesis by binding to 3`-UTR of progeny RNAs.

Recently, CP has also been found to have important roles in replication of other plant viruses. Asurmendi et al. (2004) found that TMV CP enhances production of movement protein; enhances formation of virus replication complexes and so has a role in their formation. It is well known that TMV CP is not needed for TMV RNA replication. But Asurmendi et al. (2004) consider it likely that TMV CP regulates production of subgenomic RNAs that encode movement protein [and perhaps CP (Bendahmane et al., 2002) ] or translation of mRNAs; they suggest a regulatory role of CP in establishing replication complexes and in regulating transcription. Bendahmane et al. (2002) proposed a positive regulatory effect on production of movement protein by wild-type CP. Interaction between viral RNA and CP is also considered a regulating element of TCV RNA life cycle.

Mechanistic analysis of RNA synthesis by RNA-dependent RNA polymerase from two promoters reveals similarities to DNA-dependent RNA polymerase

Host factors in positivestrand RNA virus genome replication

RNA determinants of a specific RNA-coat protein peptide interaction in Alfalfa mosaic virus: Conservation of homologous features in ilarvirus RNAs

Coat protein regulates formation of replication complexes during TMV infection

Specific RNA binding by amino-terminal peptides of alfalfa mosaic virus coat protein

Characterization of mutant tobacco mosaic virus coat protein that interferes with virus cell-to-cell movement

Alfalfa mosaic virus and ilarviruses: Involvement of coat protein in multiple steps of the replication cycle

Alfalfa mosaic virus: Coat protein-dependent initiation of infection

Identification in brome mosaic virus replicase protein 1a that mediates association with endoplasmic reticulum membrane

Viral pathogenicity determinants are suppressors of transgene silencing in Nicotiana benthamiana

Comparison of the replication of positive-stranded RNA viruses of plants and animals

A mechanism for initiating RNA-dependent RNA polymerization

Analysis of sequences and putative structures required for viral satellite RNA accumulation by in vivo genetic selection

Involvement of a stem-loop structure in the location of junction sites in viral RNA recombination

Full tobacco mosaic virus RNAs and defective RNAs have different 3` replication signals

A minimal promoter for minus-strand RNA synthesis by the brome mosaic virus polymerase complex

Brome mosaic virus polymerase-like 2a is directed to the endoplasmic reticulum by helicase-like viral protein 1a

Brome mosaic virus protein 1a recruits viral RNA2 to RNA replication through a 5` proximal RNA2 signal

The synthesis of minus-strand RNA of Bamboo mosaic potexvirus initiates from multiple sites within the poly(A) tail

An RNA activator of subgenomic mRNA1 transcription in Tomato bushy stunt virus

The importance of alfalfa mosaic virus coat protein dimers in the initiation of replication

Genetic dissection of tomato bushy stunt virus p19 protein-mediated host-dependent symptom induction and systemic invasion

Molecular Biology of Plant Viruses

Interaction between RNA-dependent RNA polymerase of Alfalfa mosaic virus and its template: Oxidation of vicinal hydroxyl groups blocks in vitro RNA synthesis

The role of the pseudoknot at the 3` end of turnip yellow mosaic virus RNA in minus strand synthesis by the viral RNA-dependent RNA polymerase

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

Tagging of plant potyvirus replication and movement by insertion of -glucuronidase into the viral polyprotein

Functions of 3`-untranslated regions of positive stranded RNA viral genomes

Identification of domains in brome mosaic virus RNA1 and coat protein necessary for specific interaction and encapsidation

Intracellular localization of the peanut clump virus replication complex in tobacco BY-2 protoplasts containing green fluorescent proteinlabeled endoplasmic reticulum on Golgi apparatus

3`-Terminal RNA secondary structures are important for accumulation of tomato bushy stunt virus DI RNAs

A central pseudoknotted three-way junction imposes tRNA-like mimicry and the orientation of three 5` upstream pseudoknots in the 3` terminus of tobacco mosaic virus RNA

TYMV RNA replication in vivo: Replicative intermediate is mainly single stranded

Identification of functional analysis of an interaction between domains of the 126/183-kDa replicase-associated proteins of Tobacco mosaic virus

Identification of sequences in brome mosaic virus replicase protein 1a that mediate association with endoplasmic reticulum membranes

Base-pairing between untranslated regions facilitates translation of uncapped, nonpolyadenylated viral RNA

A conserved hairpin structure in alfamovirus and bromovirus subgenomic promoters is required for efficient RNA synthesis in vitro

Complete replication of a eukaryotic virus RNA in vitro by a purified RNA-dependent RNA polymerase

Template sequence near the initiation nucleotide can modulate brome mosaic virus RNA accumulation in plant protoplasts

Identification of genes involved in replication and movement of Peanut clump virus

Nucleotide sequence and region of alfalfa mosaic virus RNA4

In vitro genetic selection analysis of alfalfa mosaic virus coat protein binding to 3`-terminal AUGC repeats

Regulation of single strand RNA synthesis of Alfalfa mosaic virus in nontransgenic cowpea protoplasts by the viral coat protein

Natural minus-strand RNAs of Alfalfa mosaic virus as in vitro templates for viral RNA polymerase. 3`-Terminal non-coded guanosine and coat protein are insufficient factors for full-size plus-strand synthesis

Yeast mutations in multiple complementation groups inhibit brome mosaic virus RNA replication and transcription and perturb regulated expression of the viral polymerase-like gene

Brome mosaic virus RNA replication protein 1a dramatically increases in vivo stability but not translation of viral genomic RNA3

Genome activation in alfamo-and ilarviruses

Lessons learned from the core RNA promoters of Brome mosaic virus and Cucumber mosaic virus

De novo initiation of viral RNA-dependent RNA synthesis

Long-distance movement and replication maintenance functions correlate with silencing suppression activity of potyviral HC-Pro

The 3`-terminal structure required for replication of

Coat protein enhances translational efficiency of alfalfa mosaic virus RNAs and interacts with the eIF4G component of initiation factor eIF4F

Crystal structure of the RNA-dependent RNA polymerase from Hepatitis C virus reveals a fully encircled active site

Cellular factors in the transcription and replication of viral RNA genomes: A parallel to DNA-dependet RNA transcription

RNA helicase: A novel activity associated with a protein encoded by a positive-strand RNA virus

Functions of the 126-and 183-kDa proteins of Tobacco mosaic virus

Functions of the tobacco etch virus RNA polymerase (NIb): Subcellular transport and protein-protein interaction with VPg/proteinase (NIa)

structural determinants of specific binding of coat protein or coat protein peptides to the 3

Sequences at the 3` -untranslated barley yellow dwarf virus RNA contains an embedded 3` -end

The p23 protein of Hibiscus chlorotic ringspot virus is indispensable for host-specific replication

The 23-kDa protein coded by the 3`-terminal gene of Citrus tristeza virus is an RNA-binding protein

The p92 polymerase coding region contains an internal RNA element required at an early step in tombusvirus genome replication

RNA elements required for RNA recombination function as replication enhancers in vitro and in vivo in a plus-strand RNA virus

cis-Acting functions of alfalfa mosaic virus proteins involved in replication and encapsidation of viral RNA

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

Brome mosaic virus RNA replication: Revealing the role of the host in RNA-virus replication

A conformational switch at the 3`-end of a plant virus RNA regulates virus replication

Similarities and differences between the subgenomic and minus-strand promoters of an RNA virus

Complete replication in vitro of tobacco mosaic virus RNA by a template-dependent membrane-bound RNA polymerase

The tobacco mosaic virus RNA polymerase complex contains a plant protein related to RNA-binding subunit of yeast elF-3

Identification of a region of the tobacco mosaic virus 126-and 183-kilodalton replication proteins which binds specifically to the viral 3`-terminal tRNA-like structure

Role of the 3` tRNA-like structure in tobacco mosaic virus minus-strand RNA synthesis by the viral RNA-dependent RNA polymerase

The RNA replication enhancer element of tombusviruses contains two interchangeable hairpins that are functional during plus-strand synthesis

Analysis of minimal promoter sequences for plus-strand synthesis by the cucumber necrosis virus RNA-dependent RNA polymerase

Internal initiation by the cucumber necrosis virus RNAdependent RNA polymerase is facilitated by promoter-like sequences

Enhancement of RNA synthesis by promoter duplication in tombusviruses

Purification of the cucumber necrosis virus replicase from yeast cells: Role of coexpressed viral RNA in stimulation of replicase activity

A replication silencer element in a plusstrand RNA virus

Specific binding of tombusvirus replication protein p33 to an internal replication element in the viral RNA is essential for replication

The requirement for a 5` stem-loop structure in brome mosaic virus replication supports a new model for viral positive-strand RNA initiation

In vitro-and in vivo-generated defective RNAs of Satellite panicum mosaic virus define cis-acting RNA elements required for replication and movement

Tombusvirus p19-mediated suppression of virus-induced gene silencing is controlled by genetic and dosage features that influence pathogenicity

Role of an essential triloop hairpin and flanking structures in the

Efficient infection of Nicotiana benthamiana by Tomato bushy stunt virus is facilitated by the coat protein and maintained by p19 through suppression of gene silencing

Comparison of turnip crinkle virus RNA-dependent RNA polymerase preparations expressed in Escherichia coli or derived from infected plants

Double-stranded viral RNA

Multiple interactions within the hepatitis C virus RNA polymerase repress primer-dependent RNA synthesis

Enhancer-like activity of a brome mosaic virus RNA promoter

An internally located RNA hairpin enhances replication of tomato bushy stunt virus RNAs

Brome mosaic virus helicase-and polymerase-like proteins colocalize on the endoplasmic reticulum at sites of viral RNA synthesis

Brome mosaic virus RNA replication proteins 1a and 2a colocalize and 1a independently localizes on the yeast endoplasmic reticulum

Regulation of nuclear transport of a plant potyvirus protein by autoproteolysis

Grapevine fanleaf virus replication occurs on endoplasmic-derived membrane

Degenerate in vitro genetic selection reveals mutations that diminish alfalfa mosaic virus RNA replication without affecting coat protein binding

Membrane targeting sequences in tombusvirus infections

Physical and functional interactions between the mRNA cap structure and poly(A) tail

An engineered closterovirus RNA replicon and analysis of heterologous terminal sequences for replication

Mutational analysis of the replication signals in the 3`-nontranslated region of Citrus tristeza virus

The p23 protein of Citrus tristeza virus controls asymmetrical RNA accumulation

Tomato bushy stunt virus spread is regulated by two nested genes that function in cell-to-cell movement and host-dependent systemic invasion

Identification of tomato bushy stunt virus host-specific symptom determinants by expression of individual genes from a potato virus X vector

The tomato bushy stunt virus replicase proteins are coordinately expressed and membrane-associated

A positive-strand RNA virus replication complex parallels form and function of retrovirus capsids

Turnip yellow mosaic virus RNA-dependent RNA polymerase: Initiation of minus-strand synthesis in vitro

RNA-mediated transactivation of transcription from a viral RNA

Initiation of genomic plus-strand RNA synthesis from DNA and RNA templates by a viral RNA-dependent RNA polymerase

RNA sequence and secondary structural determinants in minimal viral promoter that directs replicase recognition and initiation of genomic plusstrand RNA synthesis

Recognition of the core RNA promoter for minus-strand RNA synthesis by the replicase of Brome mosaic virus and Cucumber mosaic virus

Brome mosaic virus RNA synthesis in vitro and in barley protoplasts

Requirement for a 3`-terminal stem-loop for in vitro transcription by an RNA-dependent RNA polymerase

Spatial perturbations within an RNA promoter specifically recognized by a viral RNA-dependent RNA polymerase (RdRp) reveal that RdRp can adjust its promoter binding sites

Analysis in vivo of turnip crinkle virus satellite RNA C variants with mutations in the 3`-terminal minus-strand promoter

cis-Acting signals in bromovirus RNA replication and gene expression: Networking with viral proteins and host factors

A brome mosaic virus intergenic RNA3 replication signal functions with viral replication protein 1a to dramatically stabilize RNA in vivo

Short internal sequences involved in replication and virion accumulation in a subviral RNA of Turnip crinkle virus

The 3` untranslated region of alfalfa mosaic virus RNA3 contains a core promoter for minus-strand RNA synthesis and an enhancer element

Role of alfalfa mosaic virus coat protein in regulation of the balance between plus and minus RNA synthesis

Characterization of sequences controlling the synthesis of alfalfa mosaic virus subgenomic RNA in vivo

The 5` untranslated region of alfalfa mosaic virus RNA1 is involved in negative-strand RNA synthesis

Role of the 3`-untranslated regions of alfalfa mosaic virus RNAs in the formation of a transiently expressed replicase in plants and in the assembly of virions

Coordinate replication of alfalfa mosaic virus RNAs 1 and 2 involves cis-and trans-acting functions of the encoded helicase-like and polymerase-like domains

Inhibition of host gene expression associated with plant virus replication

Significance of the 3`-terminal region in minus-strand RNA synthesis of Hibiscus chlorotic ringspot virus

Resistance against multiple plant viruses in plants mediated by a double-stranded RNA-specific ribonuclease

Isolation from tobacco mosaic virus-infected tobacco of a solubilized template-specific RNA-dependent RNA polymerase containing a 126/183K protein heterodimer

Advances in the molecular biology of tombusviruses: Gene expression, genome replication, and recombination

Induced fit in RNA-protein recognition

An RNA domain within the 5`-untranslated region of the tomato bushy stunt virus genome modulates viral RNA replication

Asynchronous accumulation of lettuce infectious yellows virus RNAs 1 and 2 and identification of an RNA1 trans enhancer of RNA2 accumulation

CCA initiation boxes without unique promoter elements support in vitro transcription by three viral RNA-dependent RNA polymerases

Subgenomic mRNA regulation by a distal RNA element in a (+)-strand RNA virus

Repression and derepression of minus-strand synthesis in a plus-strand RNA virus replicon

Importance of sequence and structural elements within a viral replication repressor

Analysis of a viral replication repressor: Sequence requirements for a large symmetrical internal loop