Single amino acid change in the helicase domain of the putative RNA ...

Proc. Nat!. Acad. Sci. USA

Vol. 89, pp. 309-313, January 1992

Plant Biology

Single amino acid change in the helicase domain of the putative RNA replicase of turnip crinkle virus alters symptom intensification by virulent satellites (plant virus/RNA virus/satellite RNA/protoplasts)

CANDACE W. COLLMER*t, LAURA STENZLER*, XUEMEI CHEN*, NICHOLAS FAY*, DAVID HACKERt, AND STEPHEN H. HOWELL*§ *Boyce Thompson Institute, Tower Road, Ithaca, NY 14853; and *Department of Plant Pathology, University of California, Berkeley, CA 94720

Communicated by Robert J. Shepherd, September 23, 1991

ABSTRACT The virulent satellite [satellite C (sat C)] of turnip crinkle virus (TCV) is a small pathogenic RNA that intensifies symptoms in TCV-infected turnip plants (Brassica campesnis). The virulence of sat C is determined by properties of the satellite itself and is influenced by the helper virus. Symptoms produced in infections with sat C differ in severity depending on the helper virus. The TCV-JI helper virusproduces more severe symptoms than the TCV-B helper virus when inoculated with sat C. To rind determinants in the TCV helper virus genome that affect satellite virulence, the TCV-JI genome was doned and the sequence was compared to the TCV-B genome. The genomes were found to differ by only five base changes, and only one of the base changes, at nucleetide position 1025, produced an amino acid change, an aspartic acid glycine in the putative viral replicase. A chimeric TCV genome (TCV-B/JI) containing four of the five base changes (including the base change at position 1025) and a mutant TCV-B genome (TCV-B'mG) containing a single base substitution at position 1025 converted the TCV-B genome into a form that produces severe symptoms with sat C. The base change at position 1025 is located in the helicase region of the putative viral replicase, and symptom intensification appears to result from differences in the rate of replication of the satellite supported by the two helper viruses.

ute to the symptom-producing attributes of the virus. The chimeric molecule, sat C, is the most virulent satelliteintensifying the crinkling and stunting symptoms. Sat D, the smallest satellite, has little effect on symptoms and along with sat F is called an avirulent satellite. The effect on symptoms is a combined property of the helper virus, the satellite(s), and the host plant. A host effect is seen with a necrogenic variant of the satellite of cucumber mosaic virus, which induces lethal necrosis in tomato but attenuates viral symptoms of the same helper virus in pepper (9). Alternatively, there are several examples where changes in helper virus dramatically alter the symptoms associated with a particular satellite (5, 10, 11). In this paper we report that symptoms differ in infections with two different TCV helper viruses in combination with the virulent TCV satellite, sat C. The TCV helper virus genomes differ by only five base changes, and one of those base changes, at position 1025, confers symptom intensification by the virulent satellite, sat C, in infected plants.

MATERIALS AND METHODS Virus and Satellite RNA Purification. TCV-JI (John Innes isolate) (6) was propagated in turnip (Brassica campestris var. rapa cv. Just Right) under standard greenhouse conditions. Virus was purified from leaves 2-3 weeks postinoculation (12). RNA was extracted from purified virions as described by Carrington and Morris (13, 14). TCV-B (Berkeley isolate, kindly provided by T. J. Morris and L. Heaton, University of California, Berkeley, CA) was obtained as a cloned cDNA (pTCVTldl) from which infectious RNA transcripts were synthesized in vitro (15). TCV satellite RNAs were isolated from infected turnip leaves as described (6, 16). Cloning and Sequencing of the TCV-JI Genome. The TCV-JI genome was cloned as two separate cDNA fragments, a 5' and 3' segmentjoined at the Apa I site (nucleotide position 1402). The 3'-end cDNA segment of the TCV-JI genome was synthesized as follows: TCV-JI RNA (4.5 pmol) extracted from purified virions was denatured in a solution (11 ,l4) of 9.1 mM methylmercury for 10 min at room temperature and incubated for an additional 5 min following the addition of RNasin (final concentration 5 units/pLI) and 2-mercaptoethanol (final concentration 93 mM). The denatured RNA mixture was diluted to a final volume of 50 .lI containing 175 pmol of oligonucleotide CWC-50 (5'-

Satellite RNAs are small plant viral RNAs that require functions of a helper virus for replication and movement. Satellites, unlike defective interfering-like RNAs, are largely unrelated in sequence to the helper virus genome (1). Some satellites have a profound impact on virus symptoms. The varied satellites of cucumber mosaic virus are examples of different satellites that reduce viral symptoms or intensify chlorotic and/or necrotic symptoms with a given cucumber mosaic virus helper virus strain (1-5). Some natural isolates of turnip crinkle virus (TCV) carry small families of satellite RNAs. The satellite family of one isolate, TCV-JI, consists of three major satellite species, satellites (sat) C, D, and F, and minor species representing multimers (6). The members of the satellite family, which range in size from about 190 to 356 bases, are related in sequence and appear to be derivatives of the smallest satellite, sat D (7). The larger satellite species, sat C (353-356 bases), is an unusual chimeric molecule composed about half-and-half of sequences common to the other satellites and sequences similar to those found in the 3' end of the helper virus genome. Recently, small defective interfering-like RNAs but similar to those found in defective-interfering particles have been found associated with TCV (8). The satellite and defective interfering-like RNAs of TCV contrib-

lTfl2XIAGAGGGCAGGCCCCCCCCC-3'; complementary

to 16 bases at the 3' end ofthe viral genome plus an added Xba I recognition site, which is underlined), 50 mM Tris HCl (pH Abbreviations: TCV, turnip crinkle virus; sat C, satellite C. tPresent address: Department of Biology, Wells College, Aurora, NY 13026. §To whom reprint requests should be addressed.

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

309

310

Plant Biology: Collmer et al.

8.3), 75 mM KC1, 3 mM MgCl2, 10 mM dithiothreitol, dNTPs (each at 0.6 mM), 50 ttCi of [a-32P]dCTP, and 500 units of Moloney murine leukemia virus reverse transcriptase (BRL). The reaction was incubated at 37TC for 1 hr and then diluted to 400 sil. The second DNA strand was synthesized for 2 hr at 15-16'C in a reaction mixture composed of 25 mM Tris HCl (pH 8.3), 100 mM KCl, 10 mM (NH4)2SO4, 5 mM MgCl2, dNTPs (each at 250 AM), 0.15 mM NAD, 5mM dithiothreitol, Escherichia coli DNA polymerase I at 250 units/ml, E. coli RNase H at 8.5 units/ml, and E. coli DNA ligase at 30 units/ml. Following phenol/chloroform extraction, the cDNA was ethanol precipitated and size fractionated on a Sephacryl S-400 (Sigma) column. The large cDNAs were digested with Apa I and Xba I, and the Apa I-Xba I fragment of the TCV cDNA (representing nucleotides 1403 through the 3' end) was gel purified using low melting agarose (SeaPlaque GTG; FMC) according to the manufacturer's instructions. The 5' end of the TCV genome was cloned as a cDNA using a PCR protocol intended for cloning a full-length cDNA. First-strand cDNA was synthesized as described above, and then the largest DNA species were electroeluted after separation on an alkaline agarose gel. Second strand was primed with phosphorylated oligonucleotide CWC-51 (5'-GGGTAATCTGCAAATC-3'; from the 5' end of the TCV genome) and synthesized using the Klenow fragment of DNA polymerase I. The resulting double-stranded cDNA was amplified in a PCR-based protocol (16) using the oligonucleotides CWC-50 and CWC-51 at 100 pmol each and 30 PCR cycles (1 cycle of 94°C for 2 min, 37°C for 2 min, and 72°C for 40 min; 29 cycles of 94°C for 1 minm 37°C for 2 min, and 72°C for 8 min). The resulting cDNA was digested with Xba I, and the largest fragments were gel purified as described above, ligated to Sma I- and Xba I-digested pT7El9(+) (17), and used to transform E. coli DH5a cells. One resulting cDNA plasmid, pT7-TCVjl-5', was linearized with Apa I and Xba I and ligated to the 3' cDNA segment of the TCV-JI genome (the Apa I-Xba I fragment). The resulting plasmid, pCC1, contained the full-length TCV-JI cDNA inserted into T7E19(+) at the Sma I-Xba I sites. This plasmid served as the source of the Sph I-BamHI fragment used to construct recombinant plasmid TCV B/JI (see Fig. 1). In Vitro Transcription of Viral Genome and Satellite cDNAContaining Plasmids. In vitro transcription ofXba I-linearized plasmids containing the TCV helper virus genome cDNAs and Sma I-linearized plasmids containing the satellite cDNA was carried out at 37°C for 1 hr in a total volume of 50 p4, containing 5 pug of linearized TCV cDNA-containing plasmid or 2.5 pg of linearized satellite cDNA-containing plasmid, 50 units of RNasin (Promega), 10 mM dithiothreitol, rNTPs (each at 0.5 mM), 100 units of T7 RNA polymerase with the TCV plasmid or 50 units with the satellite plasmid, 40 mM Tris HC1 (pH 8.0), 8 mM MgCl2, 2 mM spermidine trihydrochloride, and 25 mM NaCl. After transcription, 1.2 ul of 500 mM EDTA (pH 8.0) was added. Transcription products were analyzed on a polyacrylamide gel containing 8 M urea. Transcripts representing the TCV helper virus genomes were used to inoculate plants on their own or were mixed with transcripts derived from cloned satellite cDNAs. Fifty microliters of reaction mixtures containing the TCV helper virus transcripts were mixed with 50 1d of the reaction mixtures containing the satellite transcripts, diluted with 105 ,ul of inoculation buffer [0.05 M glycine and 0.03 M K2HPO4 (pH 9.2) containing 1% bentonite and 1% Celite], and used to inoculate turnip plants (two leaves each). Site-Specific Mutagenesis of the TCV-B Genonmc cDNA. Oligonucleotide-directed site-specific mutagenesis was carried out by using a gapped heteroduplex plasmid DNA approach (18). Heteroduplexes were generated between the gapped form of TCV-B cDNA-containing plasmid (pTCVTldlSph-) and were cut with Sph I (position 447) and

Proc. Natl. Acad Sci. USA 89 (1992)

Apa I (position 1402) at sites in the cDNA insert; the linearized form of the same plasmid was cut with Pst I at a site outside of the cDNA insert. pTCVTldlSph- is a derivative of pTCVTldl (15) in which the Sph I site in the polylinker was destroyed by filling in with Klenow DNA polymerase. The oligonucleotide used to generate the mutant (TCVmut) is the reverse complement of the sequence in the virion plus strand (5'-CTTCTGCGCACCTTCAACGTA-3') and converts a thymidine-- cytidine in the cDNA at the underlined position and an adenosine -- guanosine in the virion strand of the virus at nucleotide position 1025. The gapped heteroduplexed plasmids were filled in with Klenow DNA polymerase under standard conditions and used to transform E. coli DH5a cells. A single mutant selected for further analysis, called TCVB1025G, was sequenced to confirm the base change at position 1025. Sequencing Progeny TCV Genomic RNA from Infected Turnip. The region of the TCV genome subject to sitedirected mutagenesis was sequenced from the RNA recovered from leaves of systemically infected plants. The region was sequenced from PCR products synthesized from the mutant site region. In the procedure modified from Frohman (19), cDNA was synthesized by reverse transcription of total leaf RNA using a TCV-specific primer, LSTCV-2 (5'CCACATCGTATCCTUlCAGC-3'), and the second-strand primer, LS-1 (5'-CTITTGGAGTCCACAACAACTC-3'). Protoplast Inoculation. Protoplasts were prepared from B. campestris cotyledons. One microgram of helper virus RNA transcripts (plus sat C transcripts, as indicated) was used to transfect 2 x 105 protoplasts (20). Protoplasts were incubated in 0.75 ml of medium containing 30 pCi of [5,6-3H]uridine for 24 hr at room temperature under fluorescent light illumination. After incubation, total RNA was isolated and subjected to gel electrophoresis on a 1% agarose gel. Regions containing the various viral RNA species (and 25S ribosomal RNA) were excised, dissolved in Scintiverse II (Fisher), and counted by liquid scintillation. To quantitate unlabeled RNA from protoplasts by hybridization, RNA was transferred to a Nytran nylon membrane and hybridized with a 32P-labeled sat C probe (pT7satC).

RESULTS The virulent satellite (sat C) of TCV produces severe symptoms in turnip plants coinoculated with the TCV-JI helper virus (6, 21). In preliminary experiments, we were surprised to find that symptoms were much milder when sat C was coinoculated with the TCV-B helper virus genome. In these experiments, the TCV-B helper virus was derived from infectious TCV-B RNA transcripts synthesized from cloned cDNA forms of the viral genome (15). In all the previous studies from this laboratory, TCV-JI had been employed as a helper virus and used in virion form or as RNA derived from uncloned material. To examine the differences in the helper virus genomes that produced different symptoms in combination with sat C, we compared the sequence of the TCV-JI genome to the TCV-B genome. The sequence of the TCV-JI genome was derived from two independent cDNA clones that were joined at an Apa I site to give pCC1. The sequence comparison revealed that the TCV-JI and TCV-B genomes differed by only five base changes, which were all located in the one partial cDNA clone, pT7-TCVj1-5' (Fig. 1). The differences were not due to fortuitous selection of a particular TCV-JI variant clone because sequencing of random TCV-JI cDNAs also demonstrated similar sequence differences. These base changes were located in the 5' half of the viral genome-all in the putative viral replicase (22). Four of the base changes in the coding region were third-base changes and were genetically


Proc. Natl. Acad. Sci. USA 89 (1992) 28K

88K

8K 7K

311

38K

MAP Apa

TCV-JI

TCV-B

A

Gj

G AC

177

576

1260 1025 1191

G

A

A GA

G

G A

Sph

TCV-B/JI

Bam Hi

TCV-B 1025G

FIG. 1. Map of the TCV genome. Map (line 1) shows open reading frames (sizes in kDa) in the TCV-B genome (22). Base substitution differences between the TCV-JI (line 2) and TCV-B (line 3) genomes are indicated using a numbering system based on the sequence corrections described below. The structure ofthe chimera (TCV-B/JI; line 4) between the TCV-JI and TCV-B genomes (Sph I-BamHI fragment from TCV-JI in TCV-B) is shown. The structure of the modified TCV-B genome (TCV-Bl025G; line 5) bearing the single base substitution at position 1025 is illustrated. Differences between the TCV-B and TCV-JI genomes were based on a reassessment of the sequence of the TCV-B genome in clone pTCVTldl. Differences were found between the sequence of the published TCV-B genome (22) and that of pTCVTldl as follows [using the published numbering system of Carrington et al. (22)]: The genome is 4053 bases instead of 4051 bases. We found an insertion of a guanine immediately after the guanine at position 19, insertion of a cytosine after the adenine at position 1214, insertion of an adenine after the adenine at position 3864, and insertion of a guanine after the adenine at position 3872. There is a deletion of an adenine at position 1235. Base changes include an adenine substituted for a guanine at position 401, a cytosine substituted for a thymine at position 755, a guanine substituted for an adenine at position 1622, an adenine substituted for a guanine at position 1886, an adenine substituted for a cytosine at position 2581, a cytosine substituted for a thymine at position 3282, and a guanine substituted for a thymine at position 3779.

silent. The fifth base change involved a guanine -- adenine change at nucleotide position 1025. To determine whether base changes in the coding region were responsible for the symptom intensification differences, a chimeric genome between TCV-B and TCV-JI, called TCV-B/JI (Fig. 1), was constructed. TCV-B/JI contains a Sph I-BamHI fragment from the TCV-JI genome with four base changes in the coding region, inserted into the TCV-B genome. To determine whether the only base change required for symptom intensification was the change at position 1025, site-directed mutagenesis was performed to convert base 1025 in the TCV-B genome from adenine -+ guanine, to create a genome called TCV-B1015G (Fig. 1). In preliminary evaluations of the symptoms produced by the different helper viruses, plants were inoculated with RNA transcripts synthesized in vitro from the cloned satellites and helper virus genomic cDNAs. It was found that such inocula were too weak for symptom evaluation and gave inconsistent results. To avoid these difficulties, the titers of the infectious RNA species were amplified by one passage through infected plants. Prior to passage, the levels of satellite and viral RNA in extracts from plants inoculated with RNA transcripts synthesized in vitro were analyzed by gel electrophoresis. In addition, we determined whether the base change in the TCV-B'025G genome was stable and could be recovered from infected plants in modified form. Viral RNA extracted from systemically infected leaves of TCV-B'05G-inoculated plants was sequenced and found to contain the guanine -- adenine change. Also, a precaution undertaken in the evaluation of symptoms was to coinoculate plants with sat D transcripts in addition to sat C. Sat D is avirulent and has not been shown to influence symptoms; nonetheless, sat D often appears unaccountably in infected plants (16). Symptoms produced in turnip plants inoculated with either the chimeric TCV-B/JI genome or the TCV-Bl025G helper virus genome in combination with sat C and D were compared to symptoms produced in plants inoculated with TCV-B and the two satellites. Symptom severity was gauged by the extent of stunting and crinkling of leaves, particularly inner leaves. Leaves showing severe symptoms were rugose and curled. As expected, in the absence of satellites, none of the plants inoculated with the helper viruses alone showed severe symptoms (Fig. 2; only TCV-B and TCV-Bl025G are shown). However, in the presence of sat C and D, TCV-B/JI

and TCV-B'011 helper viruses produced more severe symptoms in comparison to the symptoms produced by the TCV-B helper virus with sats C and D (Fig. 2; again only TCV-B and TCV-B10250 are shown). Therefore, either the four base changes present in the chimeric TCV-B/JI genome or the single base change in the mutant TCV-B1025G genome were sufficient to intensify satellite-dependent symptoms. Since symptom intensification was conferred by a single amino acid change in the putative viral replicase gene of the helper virus, it seemed likely that a possible explanation for the effect might involve differences in the ability of the helper virus to support satellite replication. Therefore, we examined the levels of satellite RNAs in infected plants. We found that late in infection (18 days post-inoculation) both helper virus genomes supported sat C and D accumulation to a similar extent. The levels of sat C and D supported by TCV-B (Fig. 3, lanes 6 and 7) were comparable to those supported by the TCV-B1025G genome (Fig. 3, lanes 8 and 9). Although the levels of satellite RNAs produced late in infections with the two different helper viruses were similar, it is possible that the infections with different helper viruses differ in the rate at which the helper viruses and/or the satellite RNAs accumulate. Rates of accumulation are difficult to study in planta because systemic infections are not synchronous. The issue was approached by measuring the accumulation of sat C in synchronously infected protoplasts. Protoplasts were inoculated with helper virus transcripts alone or with helper virus plus sat C. Infected protoplasts were labeled with [3H]uridine, and labeled RNAs were extracted 24 hr later and separated by gel electrophoresis. Radioactivity incorporated into TCV genomic RNA and sat C were corrected for incorporation into 25S ribosomal RNA. The relative rate of sat C replication (labeling) supported by the TCV-Blo?1G helper virus genome exceeded by 3-fold that of sat C supported by the TCV-B genome (Table 1). Likewise, in parallel experiments in which unlabeled sat C RNA was measured using hybridization probes, sat C accumulated about 3-fold higher during the 24-hr incubation period when supported by the TCV-B 01'G helper virus genome than when supported by the TCV-B genome (Fig. 4). Thus, in synchronously infected protoplasts, sat C appears to replicate and accumulate more rapidly when supported by the TCV-B 1025G helper virus genome than when it is supported by the TCV-B helper virus genome.

312


Proc. Natl. Acad. Sci. USA 89 (1992)

FIG. 2. Symptoms observed in turnip plants inoculated 12 days after germination with TCV-B helper virus alone (A), TCV-B1025G helper virus alone (B), TCV-B helper virus plus sat C and D (C), and TCV-Bl025G helper virus plus sat C and D (D). Plants shown here are at 18 days postinoculation and result from the single serial passage of inocula derived from plants inoculated with RNA transcripts synthesized in vitro from cloned cDNAs.

DISCUSSION The intensification of symptoms brought about by satellites of TCV depends not only on the satellite (virulent vs. 'a:

CDL

N)

a)

0

C.)

co

(n

) (in

C.) >co

>

_

>

0

I

r

11

11

---

r-F * sat C

avirulent satellites) but on the helper virus genome. The helper virus genome of the TCV-B isolate produces mild symptoms in the presence of the virulent satellite, sat C, whereas the helper virus genome from the TCV-JI isolate produces more severe symptoms. The difference between the viral genomes based on the sequence of the clones reported here amounts to only five base changes. All five base changes are located in the putative replicase coding region (22). Four of the five base changes in the replicase coding region are genetically silent, and the one at position 1025, in the readthrough domain of the putative replicase, is sufficient to intensify symptoms in the presence of the virulent satellite, sat C. The base change at nucleotide position 1025 converts an aspartic acid in the putative replicase of TCV-B to a glycine in TCV-JI. The amino acid change is located in a region of the replicase that is reasonably well conserved among plant viral

sat D

FIG. 3. Satellite RNAs supported by the two helper virus geRNA polyacrylamide gel. RNAs from uninfected plants (uninfected), from plants inoculated with TCV-B helper virus alone (TCV-B), TCVBlO25G helper virus alone (TCV-Bl025G), TCV-B helper virus plus sat C and D (TCV-B + sats C and D), and TCV-B1O5G helper virus plus sat C and D (TCV-B1025G + sats C and D) were analyzed. RNAs were extracted at 18 days postinoculation. Infected plants result from the single serial passage of inocula derived from plants inoculated with RNA transcripts synthesized in vitro from cloned cDNAs. nomes, TCV-B and TCV-Blo?5G. Satellite RNAs (5 ,ug of total per lane) were visualized on an ethidium bromide-stained 6%

Table 1. Replication of TCV sat C in turnip protoplasts Incorporation of [3H]uridine into RNA rRNA, Inoculum sat C, cpm TCV RNA, cpm cpm Mock 951 1,645 48,478 TCV-B 778 (377) 100,022 41,873 (20,308) TCV-Bl025G 955 (439) 105,462 100,489 (46,224) sat C 2,433 (1,391) 1,346 (769) 84,673 TCV-B + sat C 41,122 (17,847) 4,339 (1,883) 111,574 TCV-B1025G + sat C 105,764 (39,873) 16,731 (6,307) 128,479 Values in parentheses are corrected for incorporation into rRNA in inoculated protoplasts relative to incorporation in rRNA in mockinoculated protoplasts. The relative rate of sat C replication = (incorporation into sat C with TCV-B1025G helper)/(incorporation into sat C with TCV-B helper) (6307 cpm/1883 cpm = 3.3).

Plant Biology: Collmer et al. U)

C%4

0

.w 0

m al

;|

E

Proc. NatL. Acad. Sci. USA 89 (1992)

in infected plants and that symptoms result from some influence exerted by the satellite. This is consistent with the view (7, 26) that both helper virus and sat C may have similar symptom determinants, such as their common 3' ends, but that symptoms derive primarily from the satellites, which are present in higher numbers in infected cells than helper virus genomes.

4

+ O e> _T_ *.

|

~~I-

O

mmX . 10

m.~

F-

w

-TCV

_ -sat C

dimer

-_

313

-sat

C

FIG. 4. Accumulation of sat C RNA in Brassica protoplasts transfected with sat C and different helper virus RNA transcripts. RNA extracted 24 hr after transfection was subjected to electrophoresis on a 1% agarose gel, transferred to a nylon filter, and hybridized to a sat C probe. RNA from a mock infection (mock), infection with TCV-B helper virus alone (TCV-B), TCV-Bl025G helper virus alone (TCV-BI0MG), sat C alone (sat C), TCV-B helper virus plus sat C (TCV-B + sat C), and TCV-Bl075G helper virus plus sat C (TCVB1025G + sat C). RNA bands include sat C and sat C dimers and TCV genomic RNA, which limitedly cross-hybridizes with the sat C probe.

replicases; however, the amino acid residue at this particular position is not well conserved (23). The amino acid change lies in a domain of the TCV replicase that is similar in sequence to other nucleic acid helicases. In sequence comparison studies, Habili and Symons (24) have pointed out six sequence motifs in the helicases of plant viruses of the Sindbis virus supergroup that are conserved with respect to other nucleic acid helicases. TCV, which according to Habili and Symons (24) has a helicase domain similar to viruses in a Luteovirus-like supergroup, has only two out of six identifiable helicase motifs. However, in one of those two conserved motifs (motif IV) lies the amino acid change reported here that affects symptom intensification by satellites. The helicase of plum pox virus, in the Picornaviridae-like supergroup, has an activity that has been demonstrated in vitro to unwind RNA duplexes (25). Therefore, it is likely that the helicase domain in the TCV replicase may be involved in an RNA unwinding step of TCV RNA replication, such as positive-strand displacement from a double strand (plus and minus strand) intermediate, and that the altered amino acid in the helicase of the viral replicase may affect viral genome and/or satellite RNA replication. The amino acid change in the TCV replicase reported here does not affect the ultimate levels of satellite RNA in fully infected plants but does influence the rate at which satellite RNA accumulates in synchronously infected protoplasts. We propose that the base change in the replicase coding region of the helper virus genome affects symptoms by accelerating the rate of helper virus and sat C accumulation

We thank T. J. Morris for providing the TCV-B cDNA clone. The work at the Boyce Thompson Institute was funded in part by the National Science Foundation (DMB-8996186) and the United States Department of Agriculture Competitive Research Grants Office (90-37262-5642). Work by D.H. at the University of California at Berkeley was funded in part by the United States Department of Agriculture Competitive Research Grants Office (90-37262-5804) to T. J. Morris.

1. Kaper, J. M. & Collner, C. W. (1988) in RNA Genetics, eds. Domingo, E., Holland, J. & Ahlquist, P. (CRC, Boca Raton, FL), pp. 171-194. 2. Mossop, D. W. & Francki, R. I. B. (1978) Virology 86, 562566. 3. Gonsalves, D., Provvidenti, R. & Edwards, M. C. (1982) Phytopathology 72, 1533-1538. 4. Garcia-Arenal, F., Zaitlin, M. & Palukaitis, P. (1987) Virology 158, 339-347. 5. Palukaitis, P. (1988) Mol. Plant-Microbe Inter. 1, 175-181. 6. Altenbach, S. B. & Howell, S. H. (1981) Virology 112, 25-33. 7. Simon, A. E. & Howell, S. H. (1986) EMBO J. 5, 3423-3428. 8. Li, X. H., Heaton, L. A., Morris, T. J. & Simon, A. E. (1989) Proc. NatI. Acad. Sci. USA 86, 9173-9177. 9. Waterworth, H. E., Kaper, J. M. & Tousignant, M. E. (1979) Science 204, 845-847. 10. Masuta, C., Kuwata, S. & Takanami, Y. (1988) Ann. Phytopathol. Soc. Jpn. 54, 332-336. 11. Kaper, J. M., Tousignant, M. E. & Geletka, L. M. (1990) Res. Virol. 141, 487-503. 12. Lommel, S. A., McCain, A. H. & Morris, T. J. (1982) Phytopathology 72, 1018-1022. 13. Carrington, J. C. & Morris, T. J. (1986) Virology 150, 1%-206. 14. Carrington, J. C. & Morris, T. J. (1984) Virology 139, 22-31. 15. Heaton, L. A., Carrington, J. C. & Morris, T. J. (1989) Virology 170, 214-218. 16. Collmer, C. W., Stenzler, L., Fay, N. & Howell, S. H. (1991) Virology 183, 251-259. 17. Petty, I. T. D. (1988) Nucleic Acids Res. 16, 8738. 18. Morinaga, Y., Franceschini, T., Inouye, S. & Inouye, M. (1984) Biotechnology 2, 636-639. 19. Frohman, M. A. (1990) in PCR Protocols, eds. Innis, M. A., Gelfand, D. H., Sninsky, J. J. & White, T. J. (Academic, San Diego), pp. 28-38. 20. Jones, R. W., Jackson, A. 0. & Morris, T. J. (1990) Virology 176, 539-545. 21. Simon, A. E., Engel, H., Johnson, R. P. & Howell, S. H. (1988) EMBO J. 7, 2645-2651. 22. Carrington, J. C., Heaton, L. A., Zuidema, D., Hillman, B. I. & Morris, T. J. (1989) Virology 170, 219-226. 23. Riviere, C. J. & Rochon, D. M. (1990) J. Gen. Virol. 71, 1887-18%. 24. Habili, N. & Symons, R. H. (1989) Nucleic Acids Res. 17, 9543-9555. 25. Lain, S., Reichmann, J. L. & Garcia, J. A. (1990) Nucleic Acids Res. 18, 7003-7007. 26. Li, X. H. & Simon, A. E. (1990) Phytopathology 80, 238-242.