In Vitro translation of cucumoviral satellites I. Purification and nucleotide sequence of cucumber mosaic virus-associated RNA 5 from cucumber mosaic virus strain S

152,446-454 (1986)

VIROLOGY

In Vitro Translation of Cucumoviral I. Purification

and Nucleotide

Sequence

from Cucumber

M. J. AVILA-RINCON,’ Microbiology

of Cucumber Mosaic

Satellites

Mosaic

Virus-Associated

RNA 5

Virus Strain S

C. W. COLLMER,

AND

J. M. KAPER”

& Plant Pathology Laboratory,3 Plant Protection Institute, Agricultural US. Department of Agrkulture, Reltsville, Maryland 8070.5 Received December 5, 1~~5; accepted April

Research S’~:L’~~W~~,

15, 1986

The satellite cucumber mosaic virus (CMV)-associated RNA 5 (CARNA 5) of CMV strain S (CMV-S) which previously had been assigned the capability both to direct the synthesis of two small proteins in vitro (R. A. Owens and J. M. Kaper, 1977, Virology, 80, 196-203) and to induce the tomato necrosis disease in the presence of its helper virus (J. M. Kaper and H. E. Waterworth, 1977, Stience, 196,429-431), has been reinvestigated. Polyacrylamide gel electrophoretic analyses under partially denaturing conditions of CARNA 5 preparations from CMV-S grown in tobacco reveal a mixture of three distinct RNA species which have been isolated and partially characterized. In order of decreasing electrophoretic mobility they have been designated RNA 5, (n)CARNA 5, and (S)CARNA 5, respectively. RNA 5 has been partially sequenced and shown to represent 3’-terminal fragments of the CMV genomic RNAs. (n)CARNA 5 is responsible for the tomato necrosis-inducing properties of the mixture and coelectrophoreses with tomato necrosis-inducing CARNA 5 from CMV strain D. (S)CARNA 5 does not cause tomato necrosis; its complete nucleotide sequence was determined and is compared here to the published sequences of the CARNA 5s of et eZ., 1986, Virology, 152, four other CMV strains. A companion paper (M. J. Avila-Rincon 455-458) provides unequivocal evidence that the irr. vitro translation of nonnecrotic (S)CARNA 5 produces two small polypeptides resembling those described earlier. in 19% Academic

Press. Inc.

INTRODUCTION

from these in vitro translations were two small proteins with approximate molecular weights of 5200 and 3800 (Owens and Kaper, 1977). The S strain of CMV used in these early experiments was also found to induce lethal necrosis in tomato (Lycopersicon esculentum Mill cv. Rutgers) when it contained CARNA 5, but not when devoid of the satellite. This led to the conclusion that CARNA 5 was the causal agent for tomato necrosis (Kaper and Waterworth, 19’77), a disease occurring naturally among field tomato in parts of France (Marrou ef al., 19’73; Marrou and Duteil, 1974). The determination of the complete nucleotide sequence of CARNA 5 from CMVD (Richards et uh, 1978), an isolate also capable of inducing tomato necrosis (Jacquemond and Lot, 1981), revealed the presence of two regions in the molecule that could

One of the earliest publications following the discovery of the cucumber mosaic virus (CMV) satellite CMV-associated RNA 5 (CARNA 5)4 reported the in vitro translation of this small RNA molecule in a wheat germ system. The apparent products i Present address: UEI Virologia Vegetal, Instituto “Jaime Ferran” de Microbiologia, CIB, CSIC, Velazquez 144, Madrid 28006, Spain. ‘To whom reprint requests should be addressed. 3 Formerly Plant Virology Laboratory. 4 Abbreviations: CARNA 5, Cucumber mosaic virusassociated RNA 5. Prefixes are used to designate different CARNA 5 isolates: (S)CARNA 5 = CARNA 5 associated with CMV strain S; (l)CARNA 5 = CARNA 5 associated with CMV strain 1, etc.; (n)CARNA 5 = tomato necrosis-inducing CARNA 5; PARNA 5 = Peanut stunt virus-associated RNA 5. 0042-6822/86 Copyright All rights

$3.00

0 1986 by Academic Press. Inc of reproduction in any form reserved.

446

SEQUENCE

OF CARNA

theoretically code for the two polypeptides mentioned above. However, attempts by this group to translate their (n)CARNA 5 preparations in vitro were unsuccessful (Richards et al, 1978). CMV satellites investigated elsewhere, all of which are now known to be nucleotide sequence variants of (n)CARNA 5 (but not all necessarily necrosis inducing), have been implicated as having in vifro messenger RNA (mRNA) activities (Yamaguchi et al., 1982; Gordon and Symons, 1983). However, none of these reports contained enough detail to be conclusive. Thus, the important question of whether cucumoviral satellites, with their decidedly messenger RNA-like structures, are indeed capable of directing the synthesis of polypeptides in vitro, has remained unresolved. In the past 4 years our laboratory has made considerable progress in devising techniques to separate nucleotide sequence variants of CARNA 5 (Kaper et al., 1981; Kaper, 1983a). We now realize that most CARNA 5 preparations, previously considered to be homogeneous, in all likelihood consist of mixtures of two or more nucleotide sequence variants. We also discovered that the proportions of these different variants in a given population could be made to vary significantly under propagation conditions where different CMV helper species and/or host plant species were used (Kaper, 1983b; Garcia-Luque et al., 1984). Thus, the previously observed differences in messenger RNA activity of CARNA 5 preparations could have been due to undetected nucleotide sequence differences among CARNA 5 variants. In such CARNA 5 preparations, messenger RNA activity could have been associated with a sequence variant different from the one determining the most conspicuous biological property. Here we report the reinvestigation of CARNA 5 preparations from CMV strain S similar to the ones previously translated 7% vifm in this laboratory (Owens and Kaper, 1977). We show that such sucrose gradient-purified preparations consist of RNA mixtures which can be resolved into a nonnecrotic (S)CARNA 5 with mRNA activity, a. necrotic or (n)CARNA 5, and RNA 5, a

5 FROM

447

CMV-S

minor component derived from the viral RNAs previously also described by Gould et al. (1978). We also report the partial cloning and complete nucleotide sequence determination of (S)CARNA 5, while its i?l vitro translation properties will be described in a companion paper (Avila-Rincon et al., 1986). MATERIALS

AND

METHODS

Virus propayaticm and puri&atiulz. CMV strain S originated in South Africa and was obtained from M. H. V. Van Regenmortel. Maintained in squash (Cucurbita peps cv. Caserta Bush), the virus was purified after two passages in tobacco (Nicotianu fabtccum L. cv. Xanthi nc). Tobacco plants were inoculated with viral RNA (20 pg/ml) in 20 mM sodium phosphate buffer, pH 7.0. Extracts of systemically infected leaves were used as inoculum for subsequent passage, after which virus was purified as described (Lot et ab, 1972). Viral RNA isolation u?Ld puri$catior~. Viral RNA was isolated according to Kaper and West (1972). CARNA 5 was separated from genomic RNAs and RNA 4 by sucrose gradient centrifugation (Kaper et al., 1976) in 40 mM Tris, 20 mM sodium acetate, 2 mM NazEDTA, pH 7.8 (TAE), containing 0.1% SDS. CARNA 5 was further purified on a polyacrylamide gel containing 9% acrylamide (acrylamide:bisacrylamide, 39:1), TAE, and 8 M urea; such a gel separates CARNA 5 sequence variants (Kaper et (I/., 1981; Garcia-Luque et al., 1984). RNA was denatured at 65” for 5 rnin in 66% formamide, 1 Murea, 2.5 mMNa2EDTA, 0.075% bromophenol blue, and 0.075% xylem cyanol, and quick-cooled. Electrophoresis was carried out for 18 hr at 8 .V/cm. RNAs were eluted from the gel essentially as described by Maxam and Gilbert (1980). Sepu ru t km of double-&u xded (,S)CARNA 5 (+) and I-) strands. Virus-specific doublestranded (ds) RNAs were obtained from infected plant tissues as described by DiazRuiz and Kaper (1977, 1978). For (-t) and (-) strand isolation, ds(S)CARNA 5 preparations at 5 wg/ml were melted in 30% DMSO, 1 mMNa,EDTA at 100” for 2 min, chilled in ice water, and electrophoresed in

448

AVILA-RINCON,

COLLMER,

a 9% gel (acrylamide:bisacrylamide, 8O:l) containing 50 mM Tris-borate, pH 8.3, 1 mM EDTA at 8 V/cm for 10 hr (Kaper and Tousignant, 1983). The strands were eluted as above. Synthesis and cloning of ds(S)CARNA 5 cDNA. The cDNA clone of (S)CARNA 5 used in this study was prepared essentially as described by Collmer et al. (1985), except that 8 pg of sucrose gradient-purified (S)CARNA 5 served as the starting template RNA and first strand synthesis was carried out at 42” for lf hr. Following second strand synthesis, the larger cDNA fragments were selected by chromatography on Sephadex G-loo5 before C-tailing. Nucleotide sequence determination of (S)CARNA 5. For direct RNA sequencing from the 3’ terminus, CARNA 5 was obtained from virions or from dsCARNA 5 and was end-labeled using [5’-32P]pCp (1000-3000 Ci/mmol; ICN) and T4 RNA ligase (P-L Biochemicals). DsCARNA 5 was melted first in deionized, distilled water at 100” and quenched on ice. Reaction mixtures were incubated for 24-48 hr at -8”. For 5’ end-labeling, CARNA 5 from virions was decapped enzymatically as described by Collmer ef al. (1983). Both CARNA 5 from virions and (+) or (-) strands obtained from the ds forms were dephosphorylated using calf intestinal alkaline phosphatase (Boehringer Mannheim Biochemicals), labeled with y-[““PIATP (loo200 Ci/mmol; ICN) using T4 polynucleotide kinase (P-L Biochemicals) as described (Collmer et al., 1983). All labeled ssRNAs were purified by gel electrophoresis on 5%1 polyacrylamide gels (acrylamide:bisacrylamide, 39:l) containing 8 M urea, 90 mM Tris-borate pH 8.3, 2 mM EDTA. The (+) and (-) strands from 3’-labeled ds(S)CARNA 5 were separated and recovered as described above. Their 3’-terminal nucleotides were determined as in Collmer et aZ.(1983). Sequencing in polyacrylamideurea or polyacrylamide-formamide gels (Maniatis and Efstratiadis, 1980) was car’ Mention of a commercial company or specific equipment does not constitute its endorsement by the U.S. Department of Agriculture over similar equipment or companies not named.

AND

KAPER

ried out using the partial enzymatic cleavage method (Donis-Keller et ab, 1977) and the direct chemical cleavage method of Peattie (1979). The (S)CARNA 5 cDNA clone was labeled and sequenced as described by Collmer et al. (1985). RESIJLTS

AND

DISCUSSION

Identilcation of the RNA Components % Sucrose Gradient-PuriJed CARNA 5 from CM V-S Sucrose gradient-fractionated CARNA 5 preparations isolated from CMV-S after propagation in tobacco usually resolve into three main bands upon electrophoresis on a 9% polyacrylamide gel containing 8 M urea (Fig. 1, lane 2; lane 1 represents unfractionated total RNAs from CMV-S). The RNAs contained in these bands can be geleluted and thus purified from each other to a considerable extent (Fig. 1, lanes 3-5). In order of increasing migration rates on this gel (Fig. 1, lane 2), the upper band runs close to the position of the marker PARNA 5 (Kaper et al., 1978) in lane M. The middle band comigrates with the necrotic (n)CARNA 5 marker (lane M). The lower band migrates faster than any of the CARNA 5s characterized thus far in our laboratory. Gel-eluted preparations of the upper RNA (Fig. 1, lane 3), when tested on tomato plants in the presence of fractionated CMV-RNAs 1,2, and 3, failed to induce lethal necrosis (results not shown). In Northern blots of similar gels, =P-labeled cDNA reverse-transcribed from this RNA hybridized to the two CARNA 5 markers (but not PARNA 5) in lane M and to the two upper RNAs in lane 2 (results not shown). This established the sequence relationship of the two upper RNAs in lane 2 to the CARNA 5 markers. In the companion paper (Avila-Rincon et al., 1986) it will be shown furthermore that the RNA from the upper band (or RNA transcripts from cloned cDNA copies of it) directs the synthesis in vitro of two small polypeptides. This RNA was therefore designated (S)CARNA 5, while that from the middle band (Fig. 1, lane 4) was designated (n)CARNA 5 because it not only comigrates

SEQUENCE

1M

2

3

4

OF CARNA

5

FIG. 1. Separation of RNAs isolated from CMV-S on a 9R polyacrylamide gel containing 8 Murea. Lane 1 contains total RNA isolated from CMV-S propagated in tobacco. The three upper bands represent CMV RNAs 1 and 2 (which comigrate on this gel), 3, and 4. The three lower bands represent RNAs which were further purified as shown in lanes 2-5. Lane 2 contains the CARNA 5 fraction from sucrose gradient-fractionated CMV-S RNA. Lanes 3-5 contain gel-purified preparations of (S)CARNA 5 (lane 3), (n)CARNA 5 (lane -I), and CMV-S RNA 5 (lane 5). Marker RNAs in lane M are, from top to bottom, PARNA 5, (1 KARNA 5, and (n)CARNA 5 from CMV-D. (WARNA 5 (lane 3) still has a minute amount (lower band) of what may be a closely related sequence or conformational variant.

5 FROM

CMV-S

449

with the marker (n)CARNA 5 but also induces the tomato necrosis disease when tested on tomato plants in the presence of CMV RNAs 1,2, and 3 (not shown). We also characterized the gel-eluted RNA from the fastest migrating band (Fig. 1, lane 5) by partial determination of its nucleotide sequence from the 3’ terminus. Although the end-labeled RNA could be further resolved into two components on 8% sequencing gels, the sequences of their first 50 nucleotides were identical except at positions 27 and 28 (Fig. 2). Moreover, the sequences were almost completely homologous with the 51 nucleotides at the 3’ ends of the genomic RNAs of CMV-Q, which also show heterogeneity only at. position 27 (Symons, 1979). This similarity is not surprising in view OSthe known close relationship of CMV-S and CMV-Q (Piazzolla et oh, 1979). Thus, the fastest migrating band contains RNA which is probably similar to the CMV RNA 5 previously described by Gould et al. (1978), who first suggested its relationship to the genomic RNAs. Its wide separation Srom (S)CARNA 5 has helped to allay our concerns about possible translational artifacts due to the presence of contaminating viral RNA fragments in CARNA 5 preparations obtained by gel elution (see Avila-Rincon et al., 1986).

The bulk (90%) of the nucleotide sequence was determined by applying the partial enzymatic cleavage method (DonisKeller et al., 1977) to 3’ and 5’ end-labeled (S)CARNA 5 isolated from virions, using several adaptations described by Collmer ef al. (1983). Unequivocal determination of

FIG. 2. The 3’.terminal nucleotide sequences of CMV-S RNA 5 and CMV-Q RNAs l-4 compared. In the CMV-S RNA 5 sequence, Y represents a pyrimidine and * is a deletion relative to the CMVQ RNAs sequence. In the CMV-Q RNAs, RNAs 1, 3, and 4 contain a U at position 27, while RNA 2 contains a C there (Symons, 1979). CMV-S RNA 5 preparations are also heterogenous at this position, with the molecules containing a C there also containing a one-nucleotide insertion in that region

450

AVILA-RINCON,

COLLMER,

questionable nucleotides (always pyrimidines) was accomplished with base-specific chemical cleavage (Peattie, 1979) of 3’ endlabeled (S)CARNA 5. C/U ambiguities at the 5’ end were resolved by sequencing the 5’-pCp-labeled 3’ end of the (-) strand of ds(S)CARNA 5. Sequencing the remaining three terminal portions of the (+) and (-) strands of ds(S)CARNA 5 was used to confirm the primary structure of the terminal portions of (S)CARNA 5, while the central part of the molecule (positions 19-2’73) was corroborated by determining the sequence of both strands of the (S)CARNA 5 cDNA clone. (S)CARNA 5 is 339 nucleotides long (Fig. 3). The encapsidated form of the molecule possesses a cap at its 5’ end, because it resists Y-~‘P labeling using polynucleotide kinase unless it has first undergone a decapping step (Dasgupta et al., 1976) and phosphatase treatment. This contrasts with the 5’ end of the (+) strand of ds(S)CARNA 5, which requires only a dephosphorylation treatment prior to 32P labeling. Although the precise nature of the cap was not determined experimentally, we have assumed it to be m7Gppp in analogy with the caps on (n)CARNA 5 (Richards et ab, 1978), on the satellite of CMV-Q (Gordon and Symons, 1983), and on PARNA 5 (Collmer et al, 1985) and because of the early observation that m7Gp was capable of inhibiting the in vitro translation of crude preparations of (S)CARNA 5 (Owens and Kaper, 1977).

Features of the (S)CARNA 5 Nucleotide Sequence Cornpa.red with Those of Other CMV Satellites The (S)CARNA 5 nucleotide sequence (Fig. 3) contains four AUG triplets, but only the first one (AUG135-137) opens a reading frame of appreciable length. The first termination codon in phase with AUG135-137is UAA282-284.These two signals determine an open reading frame (designated ORF IIB) which predicts a polypeptide of 49 amino acids. As is shown in the companion paper (Avila-Rincon et al., 1986), we have unequivocal evidence for the production of two closely related polypep-

AND

KAPER

tides by the in vitro translation of (S)CARNA 5. The estimated molecular weights of these polypeptides are about 2700 and 3900 and are therefore in the size range expected for ORF IIB. Figure 3 also shows the (published) nucleotide sequences of CARNA 5s from four other CMV strains-designated Q (Gordon and Symons, 1983), 1 (Collmer et al., 1983), n (Richards et al., 1978), and Y (Hidaka et al., 1984a). These are aligned for maximal homology with the sequence of (S)CARNA 5. These and the sequences of all other CARNA 5s known to us (unpublished work) contain ORF IIB. (Q)CARNA 5, which shows more homology with (S)CARNA 5 (95%)than any of the others, possesses only a short ORF IIB. It begins with AUG134-136 but terminates at UAG185-187for a predicted polypeptide of 17 amino acids. Gordon and Symons (1983) briefly refer to some in vitro messenger RNA activity of this CARNA 5, but give no details. The coding capacities of the other three CARNA 5s (l), (n), and (Y) differ from those of(S) and (Q) in that their sequences share an additional ORF I (AUG1im13 to UGAgl&, predicting a 27 amino acid polypeptide of almost identical sequence in each case. Finally, (n)CARNA 5 and several other sequence variants (unpublished work) have as yet another open reading frame, ORF IIA (AUG9s-iW to UGA167m169)rwhich will not be discussed further here. In the context of the present work and the translational properties of (S)CARNA 5 discussed in the companion paper (AvilaRincon et al., 1986), it is of considerable interest to note that while each of the five sequences shown in Fig. 3 starts ORF IIB with an initiation triplet in an equivalent position, the different sequences could theoretically produce polypeptides of different lengths and sequences (Fig. 4). It can be seen that considerable homology exists among the putative polypeptides of (S)CARNA 5, (l)CARNA 5, and (n)CARNA 5. This homology is concentrated principally in the N-terminal halves of the molecules. The breakdown in homology in the C-terminal portions (see Fig. 4), and the different lengths of the putative translation products, are due to one or two dele-

SEQUENCE

OF CARNA

5 FROM

CMV-S

451

FIG. 3. The complete nucleotide sequence of (S)CARNA 5 in comparison with the published sequences of four other CMV satellites. The five satellites are designated by the name of the CMV strain with which each is associated, except for (n), which designates the necrosis-inducing satellite of CMV strain D sequenced by Richards et a/. (1978). Changes in the four satellite sequences relative to (S)CARNA 5 are indicated by a letter (a substitution) or a * (a deletion); a horizontal line signifies no change. Regions at the 5’ and 3’ ends that are conserved in all CMV satellites sequenced to date are boxed. Possible initiation and termination codons of protein coding regions are boxed and crosshatched. The dashed line from position 96 to position 196 of the satellite of CMV-Y represents a region of significant rearrangements, insertions, and deletions as compared to the other satellites (Hidaka et al, 1984a). Vertical bars in the vicinity of position 224 in the (S), (l), and (n)CARNA 5 sequences demarcate triplet codons to show the reading frameshifts and corresponding breakdown of amino acid homology there; in contrast, deletions in the homologous region of the Y-CMV satellite restore the reading frame to that of (S)CARNA 5 (see also Fig. 4).

tions in (1) and (n)CARNA 5 in positions corresponding to nucleotides 224 and 225 of (S)CARNA 5. These deletions result in reading frameshifts in both (1) and (n)CARNA 5 relative to (S)CARNA 5. Nucleotide changes also commonly occur in

this “frameshift region,” a region of variability present in all CMV satellites sequenced to date in our laboratory (manuscript in preparation) and discussed further below. Finally, the complete lack of homology of the putative polypeptide of

452

AVILA-RINCON,

COLLMER,

AND

KAPER

IYI rs1 I1

“YS”“TH*sRS

I

DAARERLRLRLC

In1

IQ1

HSLLSALRTHLSPRSVR

FIG. 4. Comparison of the amino acid sequences encoded by ORF IIB of five CMV satellites. The five satellites are designated by the name of the CMV strain with which each is associated, except for (n), which designates the necrosis-inducing satellite of CMV strain D sequenced by Richards et al. (1978). The putative polypeptides begin at the AUG of each satellite comparable in position to AUG135m137 of (S)CARNA 5 (see Fig. 3). Regions of amino acid sequence homology are boxed.

(Q)CARNA 5 with that of (S)CARNA 5 is attributable to a frameshift caused by a deletion at position 139 of the former’s nucleotide sequence, immediately after initiation of translation. The lack of homology between the Nterminal sequences of the ORF IIB polypeptide of (Y)CARNA 5 and that of (S)CARNA 5 results from the encoding of these amino acids within a lOl-nucleotide region of (Y)CARNA 5 (dashed lines, Fig. 3) that has replaced the equivalent region 96-162 of (S)CARNA 5 via a massive set of insertions and some deletions or nucleotide changes (Hidaka et ah, 1984a). The resumption of nucleotide homology between the two CARNA 5 after this region results in 13 homologous amino acids (Fig. 4) until a deletion at position 233 of the (Y)CARNA 5 changes the reading frame. The original reading frame is later restored by a double deletion in positions 257 and 258, the frameshift region of (Y)CARNA 5, which results in the last seven amino acids once more matching those of (S)CARNA 5. While the factors responsible for the noticeable variability of the frameshift region of different CARNA 5s are not understood [although the region is located within a large loop in the proposed secondary structure of (Q)CARNA 5 (Gordon and Symons, 1983)J the end result of deletions here and elsewhere is that the nucleotide sequence is more highly conserved in different CARNA 5s than is the amino acid sequence of proteins putatively encoded therein. Thus, while it seems unlikely that CARNA 5 encodes a component of its own replicase, its putative translational prod-

ucts may be involved in other biological properties such as its ability to modulate CMV disease expression. This problem will be discussed in more detail elsewhere. In the nucleotide sequences shown (Fig. 3), other interesting features are the long stretches of complete sequence homology among the five CARNA 5 variants. These are interspersed only by small domains in which most variation seems to be concentrated in a few positions. Of particular significance may be the fact that all CARNA 5s sequenced thus far have identical lonucleotide and 8-nucleotide stretches at their extreme 5’ and 3’ ends, respectively, whereas immediately adjacent to these conserved termini there exist particularly variable regions. This conservation also extends to the 3’ and 5’ ends of the 36% nucleotide (Y)CARNA and another 386nucleotide CARNA 5 reported by Hidaka et nl. (1984b). In addition, only slight changes occur in the 3’- and 5’-end sequences of the 393-nucleotide PARNA 5 (Collmer et al., 1985), a cucumoviral satellite supported by peanut stunt virus but not by CMV (Kaper et al., 1978). Codon usage Because of the small size of the ORFs within different CARNA 5s, little can be said about the distribution of bases in the third position of triplets within those ORFs except that there is noticeable variability among them. In a similar manner, actual codon usage is quite variable in the different CARNA 5s. In (S)CARNA 5, the less common codons CUA, UCG, UCA, ACG,

SEQUENCE

OF CARNA

CCG, GCG, AAA, GGG, GUA, GUU, and UTJU are used almost one-third (32.3%) of the time, in sharp contrast to their infrequent use (9.2%) in eukaryotic (animal) messenger RNAs (Grantham et ah, 1981). It has been suggested that translational modulation using rare codons and isoaccepting tRNA availability may be part of the mechanism for insuring low expression of certain Escherichia coli genes (Konigsberg and Godson, 1983), and such a mechanism could be at least partially responsible for the relatively low translational ef(see ficiency of (S)CARNA 5 ir2 vitro companion paper, Avila-Rincon et ah, 1986). ACKNOWLEDGMENTS Financial support for this research was provided in part by ITS. Department of Agriculture Competitive Research Grant 82.CRCR-l-1136 (Principal Investigator: 3. M. Kaper) and Grant 0389 from the Comite Conjunto Hispano-Norteamericano para la Cooperacion Cientifica y Tecnologica. REFERENCES Avil,~-RiNCON, M. J., COLLMER, C. W., and KAPER, J. M. (1986). III r:itro translation of cucumoviral satellites. II. CARNA 5 from cucumber mosaic virus strain S and SP6 transcripts of cloned (S)CARNA 6 cDNA produce electrophoretically comigrating protein products. \‘i/irolo,gy 152, 455-458. COI.LMER, C. W., HADIDI, A., and KAPER, J. M. (1985). Nucleotidr sequence of the satellite of peanut stunt virus reveals structural homologies with viroids and certain nuclear and mitocondrial introns. Proc. Nat1 /I(.&. Sci. USA 82, 3110-3114. COI,I,MER, C. W., TOIISIGNANT, M. E., and KAPER, J. M. (1983). Cucumber mosaic virus-associated RNA 5. X. The complete nucleotide sequence of a CARNA 5 incapable of inducing tomato necrosis. Virolq~q 127, 230-234. DAS(:IIPTA, R.. HARADA, F., and KAEYNERG, P. (19’76). Blocked 5’ termini in brome mosaic virus RNA. J. Viral. 18, 260-267. IDIAZ-R~JI~, J. R., and KAPER, J. M. (1977). Cucumber mosaic virus-associated RNA 5. III. Little nucleotide sequence homology between CARNA 5 and helper virus. Virdogy 80, 204-213. DIAZ-Rorz, J. R., and KAPER, J. M. (1978). Isolation of viral double-stranded RNAs using a LiCl fractionation procedure. Prep Biochem. 8, l-17. DONIS-KEI,I,ER, H., MAXAM, A. M., and GII~BERT, W. (19’77). Mapping adenines, guanines, and pyrimidines in RNA. Nucleic Acid Res. 4, 2527-2538. GARCI.~-~,TT(~IIE,I., KAPER, J. M., DIAZ-RTJIZ, J. R., and

5 FROM

CMV-S

453

RTJBIO-HUERTOS, M. (1984). Emergence and characterization of satellite RNAs associated with Spanish cucumber mosaic virus isolates. J. Gen. VG rol. 65, 539-547. GORDON, K. H. J., and SYMONS, R. H. (1983). Satellite RNA of cucumber mosaic virus forms a secondary structure with partial 3’-terminal homology to genomal RNAs. Nucleic Acid Res. 11, 947-960. GO~JLD, A. R., PALUKAITIS, P., SYMONS, R. H., and MOSSOP,D. W. (1978). Characterization of a satellite RNA associated with cucumber mosaic virus. virology 84.443-455. GRANTHAM, R., GAITTIER, C., Gorru, M., JACOBZONE. M., and MERCIER, R. (1981). Codon catalog usage is a genome strategy modulated for gene expressivity. Nucleic Acid Res. 9, r43-r74. HI~AKA, S., ISHIKAWA, K., TAICANAMI, Y., K[J~o, S., and MIIJRA, K. (1984a). Complete nucleotide sequence of RNA 5 from cucumber mosaic virus (strain Y). FEBS Letf. 174, 38-42. HIDAKA, H., HANADA, K., TAKANAMI, Y., K~JHO, S., ISHIKAWA, K., and MIIJRA, K. (1984b). “Comparison of the Nucleotide Sequence among Satellite RNAs of Cucumber Mosaic Viruses.” Abstracts of the Sixth International Congress of Virology, Sendai. Japan. [Abstract P42-111 JA(‘QUEMOND, M., and LOT, H. (1981). L’ARN satellite du virus de la mosaique du concombre. I. Comparison de l’aptitude i induire la necrose de la tomate d’ARN satellites isolbs de plusieurs souches du virus. Agronomic 1, 927-932. KAPER, J. M. (1983a). Cucumovirus-associated satellite RNA, small virus-dependent parasitic RNAs capable of modifying disease expression. ht “Plant Molecular BiolokT,” (R. B. Goldberg, ed.), lJCL.4 Symposia on Molecular and Clell Biology, New Series Vol. 12, pp. 81-100. Liss, New York. KAPER, J. M. (1983h). Perspective on CARNA 5. cucumber mosaic virus-dependent replicating RNAs capable of modifying disease expression. Picr?rt &fo(. Bid. Rep. 1, 49-54. KAPER, J. M., and TO(JSI~NANT. M. E. (1988). Separation of the complementary strands of doublestranded cucumber mosaic virus-associated RNA 5 and peanut stunt virus-associated RNA 5. Bioclrr~rc. Hiophys. Ras. (YOWL mun. 116, 1168-l 175. KAPER, J. M., and WATER~OKTH, H. E. (1977). Cucumber mosaic virus associated RNA 5: Causal agent for tomato necrosis. :Sciencp 196.429-431. KAPER, J. M.. and WEST, C. K. (1972). Polyacrylamide gel separation and molecular weight determination of the components of cucumber mosaic virus RNA. Prep. Biochent. 2, 251-263. KAPER. J. M., TOIJS~NANT, M. E., and LOT, H. (1976). A low molecular weight replicating RNA associated with a divided genome plant virus: Defective or satellite RNA? Biochum. Biophys. Rex Ctrmrr~vu. 72, 1237-2343.

454

AVILA-RINCON,

COLLMER,

KAPER, J. M., TOUSIGNANT, M. E., DIAZ-RUIZ, J. R., and TOLIN, S. A. (1978). Peanut stunt virus-associated RNA 5: Second tripartite genome virus with an associated satellite-like replicating RNA. Virology 88: 166-170. KAPER, J. M., TOUSIGNANT, M. E., and THOMPSON, S. M. (1981). Cucumber mosaic virus-associated RNA 5. VIII. Identification and partial characterization of a CARNA 5 incapable of inducing tomato necrosis. Virology 114, 526-533. KONIGSBERG, W., and GODSON, G. N. (1983). Evidence for use of rare codons in the dna G gene and other regulatory genes of Escherichia coli. Proc. Natl. Acad. Sci. USA 80,687-691. LOT, H., MARROU, J., QUIOT, J. B., and ESVAN, CH. (1972). Contribution a l’etude de virus de la mosaique du concombre (CMV). II. Methode de purification rapide du virus. Ann. Phytopathol. 4, 25-38. MANIATIS, T., and EFSTRATIADIS, A. (1980). Fractionation of low molecular weight DNA or RNA in polyacrylamide gels containing 98% formamide or 7M urea. Methods Enzymol. 63,299-305. MARROU, J., and DUTEIL, M. (1974). La necrose de la tomate. I. Reproduction des symptomes de la maladie par inoculation mecanique de plusieurs souches du virus de la mosaique du concombre (CMV). Ann. Phytopathol. 6, 155-171. MARROTJ,J., DUTEIL, M., and LOT, H. (1973). La necrose

AND

KAPER

de la tomate: Une grave virose des tomates cultivees en plein champ. Pep. Hortic. Maraich. 137, 37-41. MAXAM, A. M., and GILBERT, W. (1980). Sequencing end-labeled DNA with base-specific chemical cleavages. Methods Enzymol. 65, 499-560. OWENS, R. A., and KAPER, J. M. (1977). Cucumber mosaic virus associated RNA 5. II. In vitro translation in a wheat germ protein-synthesis system. Virology 80, 196-203. PEATTIE, D. A. (1979). Direct chemical method for sequencing RNA. Proc. Natl. Acad. Sci. USA 76,17601764. PIAZZOLLA, P., DIAZ-RUIZ, J. R., and KAPER, J. M. (1979). Nucleic acid homologies of eighteen cucumber mosaic virus isolates determined by competition hybridization. .I Gen. Virol. 45, 361-369. RICHARDS, K. E., JONARD, G., JACQIJEMOND, M., and LOT, H. (1978). Nucleotide sequence of cucumber mosaic virus-associated RNA 5. Virology 89, 395408. SYMONS, R. H. (1979). Extensive sequence homology at the 3’-termini of the four RNAs of cucumber mosaic virus. Nucleic Acid Res. 7, 825-837. YAMAGUCHI, K., HIDAKA, S., and MIURA, K. (1982). Relationship between structure of the 5’ noncoding region of viral mRNA and efficiency in the initiation step of protein synthesis in a eukaryotic system. Proc. Natl. Acad. Sci. USA 79, 1012-1016.