INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Lettuce infectious yellows virus (LIYV) is the type member of the genus Crinivirus in the family Closteroviridae. The family Closteroviridae contains two genera: the genus Closterovirus and the genus Crinivirus. Viruses in the genus Closterovirus are generally transmitted by aphids and have large (15,000 to 20,000 nucleotides) single-stranded RNA monopartite genomes, while viruses in the genus Crinivirus are transmitted by whiteflies and have bipartite genomes of ca. 15,000 nucleotides (14). LIYV is the only member of the genus Crinivirus for which the complete genome nucleotide sequence is available (12).

Viruses in the family Closteroviridae have several distinct characteristics relative to those of plant viruses in other taxonomic groups. Infections are mostly limited to phloem tissues of their host plants, and the viruses are absolutely dependent on their phloem-feeding insect vectors for plant-to-plant transmission. The viral genomes are the largest of the single-stranded plus-sense RNA plant viruses, and they contain a large number of genes (10 to 12), many of which are unique to and conserved among the viruses in this family (1; see Fig. 1). All contain genes encoding a papain-like leader protease, domains which are common for Sindbis-like virus replication-associated proteins (methyltransferase, helicase, and RNA-dependent RNA polymerase), and the "closterovirus hallmark gene array" (6). This gene array includes the following: a gene encoding a small hydrophobic protein; the gene encoding the heat shock protein 70 homolog (HSP70), a protein of ca. 60 kDa (p59 for LIYV); and genes encoding the capsid protein (CP) and the minor capsid protein (CPm). Finally, closterovirus genomic RNAs are encapsidated in morphologically polar capsids composed of the CP and CPm (2, 7, 23).

Although the biological functions of proteins encoded by the conserved closterovirus hallmark gene array have yet to be proven unequivocally, evidence suggests that they are not involved in genomic RNA replication. The Beet yellows virus (BYV; genus Closterovirus) HSP70 homolog is likely involved in inter- and intracellular trafficking (3, 15, 18), and the BYV and LIYV HSP70 homologs are virion associated (15, 23). Recent immunoneutralization experiments suggest that the CPm may be a primary determinant involved in transmission of LIYV by the whitefly, Bemisia tabaci (23). The LIYV closterovirus hallmark gene array is contained in RNA 2, and comparative sequence analyses suggested that the LIYV replication-associated proteins were encoded by LIYV RNA 1 (12). Subsequent experiments showed that LIYV RNA 1 alone was replication competent while RNA 2 was replicated only when it was coinoculated with RNA 1 (11). Likewise, cDNA constructs for BYV and Citrus tristeza virus (CTV; genus Closterovirus) in which the closterovirus hallmark gene array region has been deleted yielded transcripts which were replication competent (17, 21). For BYV, the only open reading frames (ORFs) contained by the minimal replication-competent constructs were ORFs 1A and 1B.

Still, some evidence suggests that in addition to the ORFs 1A- and 1B-encoded proteins, other virus-encoded proteins may be involved in the replication of these large genomic RNAs. For the Closterovirus BYV P21 encoded by the most 3'-terminal ORF was identified as a replication enhancer (17). Interestingly, all viruses in the family Closteroviridae studied so far encode at their 3' termini a protein similar in size to BYV P21. However, only BYV P21 and CTV P23 have been reported to have sequence similarity (17).

LIYV has several distinct features which suggest that its replication strategies may differ not only from viruses in the genus Closterovirus but also from most other multipartite RNA viruses. First, the bipartite nature of the LIYV genomic RNAs contrasts with the monopartite genomic RNAs for viruses in the genus Closterovirus. If the LIYV replication-associated proteins are encoded exclusively by RNA 1, then RNA 1 is likely replicated in cis while RNA 2 is replicated in trans. Second, there is surprisingly little sequence homology between LIYV genomic RNAs 1 and 2. The 5'-terminal five nucleotides of each genomic RNA are identical, and a 23-nucleotide sequence near the 5' terminus is shared. However, the 3'-terminal sequences of the two LIYV genomic RNAs are distinct, and computer analysis suggests that there is no similar secondary structure between LIYV RNA 1 and RNA 2 3' termini (12). These features raised questions about the replication of LIYV RNA 2. Does the same LIYV replication complex recognize RNA 1 and RNA 2 equally? Does LIYV RNA 2 replication and accumulation require proteins other than those encoded by RNA 1 ORFs 1A and 1B?

In this paper, we show an asynchronous accumulation pattern for LIYV genomic RNAs 1 and 2 when they are inoculated simultaneously to protoplasts. Mutagenesis studies also showed that RNA 2-encoded proteins do not affect RNA 1 and/or RNA 2 accumulation. However, mutations in the LIYV RNA 1 3'-terminal ORF (ORF 2 encoding P32) did not affect RNA 1 accumulation, but severely reduced accumulation of LIYV RNA 2.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Construction of LIYV mutants. Most mutants were constructed by using site-directed mutagenesis to insert two in-frame stop codons near the 5' termini of specific LIYV ORFs by using the transformer site-directed mutagenesis kit (Clontech) according to manufacturer's recommendations. Full-length cDNA clones to LIYV RNA 1 (pSP9/55) and RNA 2 (pSP6) (11) were used for mutagenesis when possible. However, mutagenesis on the CP and CPm ORFs was done by first excising a 941-bp fragment (nucleotides 4407 to 5348) from pSP6. This fragment was subcloned into SalI-HindIII-digested pBluescript SK II (Stratagene), and mutagenesis was performed using primers shown in Table 1. Mutated fragments were then reinserted into pSP6. Double mutants for the CP and CPm ORFs (CPPM) and the HSP70 and p59 ORFs (HP) were also constructed by ligating fragments together which contained each individual mutation. The RNA 1 Rep, P32 F1, and P32 F2 mutants were constructed by first excising and discarding the EcoRI fragment between nucleotides 448 and 5180 from pSP9/55 (pEco RI). For the Rep mutant, site-directed mutagenesis was then used to modify the overlapping region of ORF 1B and ORF 2 (encoding P32) in pEco RI. The P32 F1 and P32 F2 frameshift mutants were generated by digesting pEco RI with XbaI and NdeI, respectively. The digested plasmids were then treated with Klenow and religated. This served to introduce 4 and 2 nucleotides, respectively, into ORF 2 of P32 F1 and P32 F2, resulting in frameshifts and stop codon insertions within the P32 coding region (Table 1). The LIYV RNA 1 EcoRI fragment (nucleotides 448 to 5180) was then reintroduced into the modified plasmids to yield the Rep, P32 F1, and P32 F2 mutants. The final mutant, RNA 1 P32, was constructed by PCR amplification of a 911-bp fragment (nucleotides 7208 to 8118) corresponding to ORF 2. The forward primer (Table 1) was designed to insert two in-frame stop codons into ORF 2. The PCR product was digested with MfeI and NotI and ligated into MfeI-NotI-digested pSP9/55 to yield the RNA 1 P32 mutant. Thus, these approaches resulted in each mutant either containing a new or lacking a wild-type restriction enzyme site so that specific mutants could be monitored (Table 1). All mutant clones were sequenced in the modified region in order to ensure that the mutation was correct. Specific constructions for all mutations are shown in Table 1, and their relative positions are indicated in Fig. 1.

View larger version (15K): [in this window] [in a new window]   FIG. 1.   Schematic representation of the LIYV genomic RNAs and specific LIYV mutants. Rectangles represent ORFs in LIYV genomic RNAs 1 and 2. LIYV RNA 1 ORFs 1A, 1B, and 2 are shown. P-PRO, papain-like protease; MTR, methyltransferase; HEL, RNA helicase; RDRP, RNA-dependent RNA polymerase; HSP70, homolog of HSP70 proteins. Arrows indicate positions in LIYV genomic RNAs where mutations were introduced (Table 1).

LIYV inocula and protoplast manipulation. LIYV virions were purified from LIYV-infected Chenopodium murale plants, and RNAs were extracted as previously described (10). Capped transcripts corresponding to wild-type and mutant LIYV RNAs 1 and 2 were synthesized as previously described (11). Protoplasts were prepared from cultured Nicotiana tabacum suspension cells (16) and inoculated using 5 µg of each transcript essentially as previously described (13), except that 1.2 million cells were used for each inoculation, and after inoculation, protoplasts were incubated at 26.5°C.

Analysis of wild-type and mutant LIYV replication. LIYV-inoculated protoplasts were collected and analyzed by methods similar to those previously described (11). Aliquots containing ca. 1.2 × 10⁵ cells were collected by centrifugation (1,310 × g) at different times postinoculation (p.i.), and RNAs were isolated using TRI Reagent (MRO) according to the manufacturer's recommendations. Generally, RNAs representing ca. 1.2 × 10⁴ cells were used for Northern hybridization analysis. RNAs were denatured with glyoxal, separated by agarose gel electrophoresis, and transferred to Hybond NX (Amersham) as previously described (10).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Temporal accumulation of LIYV RNAs 1 and 2. In order to monitor LIYV RNA accumulation, we first assessed the time course accumulation of LIYV RNAs in LIYV-inoculated protoplasts. When protoplasts inoculated with LIYV virion RNAs (containing both genomic RNAs 1 and 2) were analyzed, differential temporal accumulation of LIYV RNAs 1 and 2 was observed. Full-length negative-sense RNA 1 (Fig. 2C, lane 12) and the subgenomic RNA corresponding to the RNA 1 ORF 2 (encoding P32; Fig. 2A, lane 12) were detected at 12 h p.i. Longer exposures also showed accumulation of the positive-sense genomic LIYV RNA 1 at 12 h p.i. (Fig. 2A); however, it was much more abundant by 24 h p.i. Time course comparisons showed that the P32 subgenomic RNA and negative-sense genomic RNA 1 reached maximum accumulation at 24 h p.i., but positive-sense genomic RNA 1 continued to accumulate up to the final sampling time of 72 h p.i. (Fig. 2A and C). Optical density analysis of exposed films showed that the positive-sense LIYV genomic RNA 1 increased ca. 11-fold between 12 and 24 h p.i., but then only a twofold increase was seen between 24 and 72 h p.i.

View larger version (85K): [in this window] [in a new window]   FIG. 2.   Time course of accumulation of LIYV RNAs in protoplasts. LIYV virion RNAs were used to inoculate 1.2 × 106 tobacco protoplasts, and 1.2 × 105 protoplasts were collected at the times p.i. indicated above respective lanes. Lane 0, sample taken immediately after inoculation. Total RNAs from 1.2 × 104 protoplasts were used for Northern blot hybridization with a DIG-labeled negative-sense RNA 1 probe (A), positive-sense RNA 1 probe (C), negative-sense RNA 2 probe (B), and positive-sense RNA 2 probe (D). RNAs corresponding to the LIYV genomic RNAs are indicated by G, the RNA 1 P32 subgenomic RNA is indicated by SG, and the RNA 2 defective RNA is indicated by D. Numbers at the left correspond to positions of marker RNAs (sizes in nucleotides) analyzed in the same gel. Inserts A1 and B1 show longer exposures for the respective lanes, and the genomic RNAs (G) are indicated. Images were arranged and labeled using Adobe Photoshop 4.0.

Mutational analysis of LIYV RNA 2. Although LIYV RNA 1 can replicate in protoplasts and accumulate to high levels in the absence of RNA 2 (11; Fig. 3A), the role(s) of RNA 2 and/or its encoded proteins in the replication and accumulation of LIYV RNAs is not known. Therefore, we constructed mutations in specific RNA 2 ORFs and assessed effects of these mutations on LIYV RNA accumulation in protoplasts. Protoplasts were inoculated with wild-type RNA 1 transcripts alone, wild-type RNA 1 and RNA 2 transcripts, or wild-type LIYV RNA 1 transcripts plus transcripts of specific LIYV RNA 2 mutants. The relative accumulation of LIYV genomic RNAs 1 and 2 and three subgenomic RNAs was assessed for wild-type and mutant constructs at 24 and 48 h p.i. (Fig. 3). RNA hybridization analyses showed that the accumulation of LIYV RNA 1 and RNA 2 genomic and subgenomic RNAs varied slightly from experiment to experiment. However, based on four replicated experiments, accumulation of the LIYV RNA 1 genomic and ORF 2 (encoding P32) subgenomic RNA and the RNA 2 genomic and subgenomic RNAs for the HSP70 homolog and P26 ORFs were not different whether the inocula contained wild-type or LIYV RNA 2 mutant RNAs (single or double mutants; Fig. 3A and B, respectively). Thus, these data suggest that the RNA 2 ORFs encoding the HSP70 homolog, P59, CP, CPm, and P26 did not affect accumulation of the LIYV genomic RNAs 1 or 2 or the P32, P26, or HSP70 homolog subgenomic RNAs.

View larger version (79K): [in this window] [in a new window]   FIG. 3.   Accumulation of wild-type and LIYV RNA 2 mutant RNAs in protoplasts. Transcripts of wild-type LIYV RNA 1 (from clone pSP9/55) were inoculated alone, with wild-type LIYV RNA 2 (pR6) transcripts, or with specific LIYV RNA 2 mutant transcripts (Fig. 1 and Table 1) to 1.2 × 106 tobacco protoplasts, and 1.2 × 105 protoplasts were collected at 24 and 48 h p.i. Total RNAs purified from 1.2 × 104 protoplasts collected at 24 and 48 h p.i. were used for Northern blot hybridization with RNA 1 (A) and RNA 2 (B) negative-sense DIG-labeled probes. Labels above lanes indicate the inocula used for the given sample. Lanes: RNA1, only RNA 1 transcripts; WT, wild-type LIYV RNA 1 (from pSP9/55) and RNA 2 (from pSP6) transcripts. Remaining labels indicate wild-type RNA 1 plus specific LIYV RNA 2 mutants (see Table 1 for mutants; HP is a double mutant for both the HSP70 homolog and p59 ORFs, and CPPm is a double mutant for both the CP and CPm ORFs). Numbers at left correspond to positions of marker RNAs (sizes in nucleotides) analyzed in the same gel. Arrows labeled P32, HSP70, and P26 indicate hybridization signals for the corresponding subgenomic RNAs. Images were arranged and labeled using Adobe Photoshop 4.0.

Mutations in LIYV RNA 1 ORF 2 affect accumulation of RNA 2 but not RNA 1. Computer-assisted sequence analyses suggested that LIYV RNA 1 ORFs 1A and 1B encode the proteins associated with LIYV RNA replication (12), and infectivity studies shown here (Fig. 3A) and previously (11) clearly demonstrate that RNA 1 alone is replication competent. However, in addition to ORFs 1A and 1B, LIYV RNA 1 encodes in ORF 2 P32 a protein of unknown function and for which no significantly similar proteins have been identified in database searches (data not shown). In order to assess the potential role(s) of ORF 2 and/or P32 in LIYV replication, we created four mutations in LIYV RNA 1 ORF 2 and compared accumulation of LIYV RNAs 1 and 2 in protoplasts. The first mutant (RNA 1 Rep; Table 1) was constructed to yield three amino acid changes in the ORF 1B-encoded protein and to introduce two stop codons into the 5'-terminal region of ORF 2 (residues 29 and 31), as ORFs 1B and 2 overlap (Fig. 1). When transcripts of RNA 1 Rep were inoculated alone (not shown) or with wild-type RNA 2 transcripts to protoplasts, no LIYV RNA accumulation was detected (Fig. 4), suggesting that this mutation was lethal. The second mutation (LIYV RNA 1 P32) was localized to only RNA 1 ORF 2 and truncated this ORF by inserting two stop codons for residues 57 and 59 (Fig. 1). RNA hybridization analysis of inoculated protoplasts showed that LIYV RNA 1 P32 accumulated to levels indistinguishable from those of wild-type LIYV RNA 1, and similar levels of the P32 subgenomic RNA were also detected (Fig. 4A). However, coinoculation of LIYV RNA 1 P32 plus RNA 2 transcripts resulted in very low levels of LIYV RNA 2 accumulation (Fig. 4B). In some experiments, no RNA 2 accumulation was detected, while in most, RNA 2 accumulated to only 3 to 10% of the level seen for RNA 2 coinoculated with wild-type RNA 1. Further proof that this was newly replicated LIYV RNA 2 was obtained by hybridization analyses for negative-sense LIYV RNA 2. Low levels of negative-sense LIYV RNA 2 progeny were detected when the inoculum contained wild-type RNA 2 and LIYV RNA 1 P32. As this inoculum contained only positive-sense LIYV RNA 2 transcripts, the negative-sense products could only result from RNA 2 replication (not shown). Also, comparison of LIYV RNA 2 detection in protoplasts coinoculated with the LIYV RNA 1 P32 and the RNA 1 Rep mutants showed that low levels of positive-sense LIYV RNA 2 were detectable for both at 24 h p.i. This was likely residual inocula, as the RNA 1 Rep mutant was unable to replicate and therefore could not support RNA 2. However, by 48 and 72 h p.i., LIYV RNA 2 was clearly detectable only in cells coinoculated with P32 but not with Rep, indicating that this is progeny RNA 2. These comparative analyses showed that the mutation introduced into the LIYV RNA 1 ORF 2 (P32) did not affect LIYV RNA 1 accumulation, but severely affected the trans replication or accumulation of both positive- and negative-sense LIYV RNA 2. 

View larger version (74K): [in this window] [in a new window]   FIG. 4.   Accumulation of LIYV RNA 1 mutant and wild-type RNAs. Transcripts of wild-type LIYV RNA 1 (from pSP9/55) and of the Rep, P32, P32 F1, and P32 F2 RNA 1 mutants were separately coinoculated with wild-type LIYV RNA 2 (pSP6) transcripts to 1.2 × 106 tobacco protoplasts, and 1.2 × 105 protoplasts were collected at 0, 24, 48, and 72 h p.i. (no protoplasts inoculated with P32 F1 and P32 F2 transcripts were collected at the 48-h time point). Total RNAs purified from 1.2 × 104 protoplasts collected at the time points indicated were used for Northern blot hybridization with RNA 1 (A) and RNA 2 (B) negative-sense DIG-labeled probes. G indicates the position of LIYV genomic RNAs, and SG indicates the LIYV RNA 1 ORF 2 (P32) subgenomic RNA. Numbers at left correspond to positions of marker RNAs (sizes in nucleotides) analyzed in the same gel. Images were arranged and labeled using Adobe Photoshop 4.0.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

The data presented here show several interesting aspects regarding replication and accumulation of LIYV RNAs. First, accumulation of LIYV genomic RNAs 1 and 2 is not synchronous. LIYV RNA 1 RNAs (both positive- and negative-sense genomic RNAs, as well as the RNA 1, ORF 2 [P32] subgenomic RNA) were always detected very early after inoculation (ca. 12 h p.i.), and the maximal rate of increase for these RNAs occurred between 12 and 24 h p.i. In contrast, our data showed that significant accumulation of positive- and negative-sense RNA 2 was only detected between 24 and 36 h p.i. When hybridization signals were compared over time, the amount of LIYV RNA 1 detected at 12 h p.i. was already much greater than the input (inoculum at 0 h p.i.). Conversely, the LIYV RNA 2 signal was not greater than that for inoculum until 36 h p.i. These results were consistently obtained. Also, as inocula contained essentially equimolar amounts of RNAs 1 and 2, this was not due to differing amounts of these RNAs in the inocula.

The above data suggest that there may be cis-preferential replication of LIYV RNA 1. LIYV RNA 1 encodes proteins for replication, and the ORFs 1A- and 1B-encoded proteins, most likely translated from the LIYV genomic RNA 1, likely serve to replicate LIYV RNA 1 and direct transcription of the RNA 1 ORF 2 (P32) subgenomic RNA. The P32 subgenomic RNA not only appears quickly after inoculation, but it is the most abundant of all of the LIYV RNAs detected by us in LIYV-infected cells. Efficient replication or accumulation of LIYV RNA 2 does not simultaneously occur with that of LIYV RNA 1. RNA 2 is replicated in trans, and progeny begin to appear only after sufficient accumulation of LIYV RNA 1 (and most likely its encoded proteins).

All of our LIYV RNA 2 mutants, when coinoculated with wild-type RNA 1, accumulated to levels essentially equal to those of wild-type RNA 2 and did not affect the level of LIYV RNA 1 accumulation. Furthermore, during LIYV replication, subgenomic RNAs are generated for downstream ORFs for both RNAs 1 and 2 (20), and the LIYV RNA 2 mutations analyzed here did not affect accumulation of subgenomic RNAs for the RNA 1 ORF 2 (P32) or the RNA 2 HSP70 homolog and P26 ORFs, the subgenomic RNAs examined by us here. This is not so surprising, as previous work from our laboratory (11) showed that LIYV RNA 1 is replication competent in the absence of LIYV RNA 2. In addition, although we evaluated mutations in five of the seven RNA 2 ORFs, we did not mutate the RNA 2 ORFs encoding P5 or P9. However, evidence also suggests that these ORFs (or their encoded proteins) are unlikely to be involved in LIYV RNA accumulation. First, RNA 1 alone accumulates to similar levels with or without coinfection of RNA 2 (11; Fig. 3A). Second, when RNA 1 was coinoculated into protoplasts with LIYV RNA 2 defective RNAs (D RNAs) lacking the P5 and/or P9 ORFs, no effects were seen for LIYV RNA accumulation (20). Taken together, these data suggest that LIYV RNA 2 and its encoded proteins do not significantly affect LIYV RNA accumulation.

However, some evidence suggests that in addition to ORFs 1A and 1B, other ORFs may encode proteins that affect RNA accumulation for viruses in the family Closteroviridae. For the Closterovirus BYV, the 3'-most ORF on the monopartite genomic RNA has been identified as a replication enhancer (17). The similarly positioned ORF for LIYV is the 3'-most ORF on RNA 2 (encoding P26). All members of the family Closteroviridae have similarly positioned ORFs encoding proteins of similar size (e.g., the BYV protein is P21) (1, 6). Computer-assisted analysis showed that the LIYV-encoded P26 showed no significant similarity with the BYV-encoded P21, and our data do not suggest that the RNA 2 3'-most ORF encodes any sort of replication enhancer affecting LIYV RNA 1 or RNA 2, or even accumulation of the subgenomic RNAs examined here. Interestingly, our data show that mutations in the 3'-most ORF of LIYV RNA 1 affect the trans replication and accumulation of LIYV RNA 2. Our data showed that the LIYV RNA 1 ORF 2 mutants were capable of independent replication similar to that of wild-type RNA 1, demonstrating that P32 is not necessary for RNA 1 replication. However, the LIYV P32 mutants were unable to efficiently direct replication and accumulation of LIYV RNA 2. Also, because we constructed three separate mutants for ORF 2 in distinct coding regions of this ORF and obtained similar phenotypes, these data suggest that the ORF 2-encoded P32 is the likely trans enhancer of LIYV RNA 2 accumulation.

Replication enhancers have been reported for several multipartite plant viruses and include the RNA-encoded B of Barley stripe mosaic virus (BSMV; 19), the Cowpea mosaic virus (CPMV) M-RNA-encoded 58-kDa protein (4), the Beet necrotic yellow vein virus (BNYVV)-encoded P14 (8), and the Peanut clump virus (PCV) RNA 1-encoded P15 (9). The BSMV B, the BNYVV P14, and the PCV P15 proteins all belong to a group of cysteine-rich proteins, and P14 shares weak but statistically significant similarity with other nucleic acid binding proteins (8). These cysteine-rich proteins influence or enhance replication for all genomic components of their respective virus. In contrast, the CPMV 58-kDa protein is a template-selective replication enhancer. The CPMV 58-kDa protein is needed for efficient replication of the CPMV M-RNA but not the CPMV B-RNA; thus, it is a cis replication enhancer. Interestingly, a RNA sequence located within Red clover necrotic mosaic virus (RCNMV) genomic RNA 2 has recently been identified as a transcriptional enhancer, functioning in trans for the synthesis of an RCNMV RNA 1 subgenomic RNA (22).

LIYV RNA replication-accumulation kinetics and the RNA 1-mediated trans replication enhancer activity affecting accumulation of LIYV RNA 2 so far appear to be unusual among RNA plant viruses. In this regard, it is interesting to note the lack of nucleotide sequence homology seen for the LIYV genomic RNAs. Only the 5'-most five nucleotides and a 23-nucleotide sequence near the 5' termini of both LIYV RNAs 1 and 2 are homologous for LIYV RNAs 1 and 2 (12). In contrast, most multipartite RNA plant viruses have conserved or shared 3' nucleotide sequences on the genomic RNA segments for a given virus (5). No complete nucleotide sequences are yet available for other members in the genus Crinivirus; therefore, whether they exhibit sequence characteristics like those of the LIYV genomic RNAs is presently unknown. Presumably, other viruses in this genus would encode a protein like the LIYV P32. The monopartite viruses in the genus Closterovirus do not.