JOURNAL OF VIROLOGY, Apr. 1995, p. 2689–2691 0022-538X/95/$04.0010 Copyright q 1995, American Society for Microbiology
Vol. 69, No. 4
A Spontaneous Mutation in the Movement Protein Gene of Brome Mosaic Virus Modulates Symptom Phenotype in Nicotiana benthamiana A. L. N. RAO*
AND
GEORGE L. GRANTHAM
Department of Plant Pathology, University of California, Riverside, California 92521-0122 Received 23 September 1994/Accepted 21 December 1994
Brome mosaic virus (BMV) is a positive-strand RNA virus with a multipartite genome that causes symptomless infection in Nicotiana benthamiana. We have isolated and characterized a strain of BMV that produced uniform vein chlorosis in systemically infected N. benthamiana. Analysis of pseudorecombinants constructed by exchanging RNA 1 and 2 and RNA 3 components between wild-type (non-symptom-inducing) and vein chlorosis-inducing strains of BMV indicated that the genetic determinant for the induction of the chlorotic phenotype is located on RNA 3. Sequence analysis of progeny RNA 3 recovered from symptomatic N. benthamiana plants revealed that vein chlorosis is due to the single nucleotide transition 887G3887A, which changes the codon for Val-266 to Ile-266 in the movement protein gene. The mutation had no detectable effect on the accumulation of virus in either inoculated or systemically infected leaves of N. benthamiana. The vein chlorosis phenotype is the manifestation of the substitution of Ile-266 for Val-266 in the movement protein gene, since additional alterations in this region (a silent mutation, i.e., 887GUU8893GUC, and an alteration of valine to phenylalanine, i.e., 887GUU8893887UUU889) resulted in symptomless infections on N. benthamiana. The modulation of the symptom phenotype by the substitution of Ile-266 for Val-266 is specific for N. benthamiana, since neither movement nor the symptom phenotype in barley plants was affected. derived from four independent transcription reactions, 23 plants remained symptomless (Fig. 1). In these experiments, three N. benthamiana plants displayed unusual vein chlorosis symptoms on the uninoculated upper leaves (Fig. 1). The vein chlorosis was less prominent in younger leaves than in older leaves. Systemic infection by BMV in symptomless plants as well as in those with vein chlorosis was confirmed by inoculation of plant extracts onto local-lesion host C. hybridum and by Northern (RNA) hybridization analysis (data not shown). Virus from symptomatic leaves was passaged in healthy N. benthamiana plants, which developed the same vein chlorosis symptoms on uninoculated upper leaves, indicating that a sequence modification in the BMV genome was responsible for the observed phenotype. Hereafter we refer to the wild-type (non-symptom-inducing) BMV isolate as BMV-S and the one causing the vein chlorosis symptom phenotype as BMV-V. Pseudorecombinants map the sequence change to RNA 3. To identify the BMV RNA component that had acquired the sequence modification, pseudorecombinants between BMV-S and BMV-V isolates were constructed. Virus was purified from N. benthamiana infected with BMV-S and BMV-V isolates as described previously (10). Total RNA extracted from purified virus of each isolate was denatured with formaldehyde and subjected to electrophoresis in 0.8% agarose. After brief staining with ethidium bromide, bands corresponding to RNAs 1 and 2 were separated from those corresponding to RNA 3 with a sterile scalpel blade, inserted into another agarose gel, and resubjected to electrophoresis. The RNA bands were further purified by at least two additional cycles of 0.8% agarose gel electrophoresis followed by one cycle of electrophoresis with an 0.8% low-melting-point agarose gel. The purities of the RNA 1 and 2 and RNA 3 fractions were assessed by Northern hybridization and by biological assay with C. hybridum (Table 1). For the biological assay, the concentrations of the two RNA fractions were adjusted to 50 mg/ml in a buffer containing 3 mg of bentonite per ml, and the fractions were inoculated onto C.
The genome of brome mosaic virus (BMV) (1), the type member of the bromovirus group of plant viruses, is divided among three single-stranded, messenger-sense RNAs. Viral replication is dependent on the nonstructural proteins encoded by the monocistronic RNAs 1 and 2 (1). The two gene products encoded by dicistronic RNA 3 are dispensable for viral replication (1). A nonstructural protein of 30 kDa encoded by the 59 half of RNA 3 potentiates the cell-to-cell transport of the virus (1, 8), and the capsid protein (20 kDa) encoded by the 39 half of RNA 3 is translated from a subgenomic RNA 4 (1). The host range of BMV is largely restricted to the Gramineae family, but a number of species in other families have been reported to be susceptible (5). Barley (Hordeum vulgare) is routinely used as a host plant for replication and infectivity studies (10). In susceptible varieties of barley, BMV produces characteristic mosaic symptoms 7 to 10 days after inoculation (5, 10). In addition, BMV also induces local necrotic lesions in Chenopodium hybridum(14) and chlorotic local lesions followed by systemic mottling in Chenopodium quinoa (12). Among members of the Solanaceae family, Nicotiana benthamiana is susceptible to BMV infection but remains symptomless (8). During the course of our infectivity studies, we have observed that some N. benthamiana plants inoculated with wild-type transcripts synthesized in vitro developed strong vein chlorosis symptoms on uninoculated upper leaves. We report the characterization of progeny RNA recovered from N. benthamiana plants with vein chlorosis symptoms and provide evidence that the manifestation of these unusual symptoms results from a single nucleotide substitution in the movement protein (MP) gene of BMV RNA 3. Induction of vein chlorosis in N. benthamiana by BMV. Of 26 N. benthamiana plants inoculated over a 6-month period with a mixture of all three wild-type BMV RNA transcripts (4, 9) * Corresponding author. Phone: (909) 787-3810. Fax: (909) 7874294. Electronic mail address:
[email protected]. 2689
2690
NOTES
J. VIROL.
FIG. 1. Representative examples of phenotypic systemic symptoms induced on N. benthamiana by BMV-S (B) and BMV-V (C) 10 days after inoculation. A leaf from an uninoculated healthy plant is shown in panel A.
hybridum plants as described previously (14). The results presented in Table 1 demonstrate that the fractionated RNA components of each isolate were of a high level of purity and were nearly free from contaminating counterparts. Two pseudorecombinants were constructed in vitro (11) by exchanging the fractionated RNA components between BMV-S and BMV-V isolates, and the mixtures were inoculated onto C. hybridum. After 7 days, single lesions derived from a mixture of fractionated RNAs containing either RNAs 1 and 2 of BMV-S and RNA 3 of BMV-V or RNAs 1 and 2 of BMV-V and RNA 3 of BMV-S were excised and independently inoculated onto the primary leaves of N. benthamiana. Plants inoculated with either of the pseudorecombinants showed symptoms that are characteristic of those caused by the parent isolate from which the RNA 3 component was derived. These observations suggest that the induction of vein chlorosis in N. benthamiana is likely due to a sequence modification in the RNA 3 component of the BMV-V isolate. A single nucleotide change in the MP is responsible for vein chlorosis. To determine any sequence modifications that may have occurred in RNA 3 of the BMV-V isolate, RNA was extracted from a purified virus preparation obtained from one of the N. benthamiana plants displaying vein chlorosis and was used in reverse transcriptase PCR as described previously (10). The entire sequence of RNA 3 was amplified as two cDNA fragments. A 1,200-nucleotide fragment encompassing the 59
TABLE 1. Infectivity in C. hybridum of fractionated RNA components of BMV-S and BMV-V isolates Virusa
RNA segment(s) in inoculumb
No. of lesions per half-leafc
BMV-S
RNAs 1 and 2 RNAs 1 and 2 and RNA 3 RNA 3 RNAs 1 and 2 and RNA 3
1 33 0 46
BMV-V
RNAs 1 and 2 RNAs 1 and 2 and RNA 3 RNA 3 RNAs 1 and 2 and RNA 3
1 27 0 44
a For each virus the infectivities of two inocula on four opposite half-leaves of C. hybridum that had been kept in darkness for at least 18 h were compared. b The total RNA concentration in each inoculum was adjusted to approximately 50 mg/ml. c Mean number of local lesions produced by each inoculum on four halfleaves.
FIG. 2. Sequence analysis of progeny RNA 3 recovered from systemically infected N. benthamiana plants. Depicted is an autoradiographic image of a sequencing gel showing part of the sequence (shown on left and right) from the MP genes recovered from an asymptomatic N. benthamiana leaf (A) (the leaf is shown in Fig. 1B) and three individual N. benthamiana leaves showing vein chlorosis (B) (one such leaf is shown in Fig. 1C). The arrow in panel A shows the wild-type G at position 887, and the asterisks in each column of panel B show the single base transition 887G3887A. Numbering is from the 59 terminus.
noncoding region, the MP gene, and part of the intercistronic region was amplified with the primers d(59TAATAATAACT CAGACACA39) and d(59GTAAAATACCAACTAATT39). Similarly, a fragment of 967 nucleotides encompassing a part of the intercistronic region, the coat protein gene, and the 39 noncoding region was amplified with the primers d(59GGTT ATCCATGTTTGTGGA39) and d(59CAGGATCCCGACAT GGTCTCTTTTAGAGATTTACAGTG39) (the BamHI site proximal to the 39 end is underlined). Either the PCR products were directly sequenced by the fmol sequencing procedure (Promega), or the products were ligated into a T3/T7 lacZ vector and the inserts were sequenced as described previously (13). Comparison of the sequences of progeny RNA 3 revealed a single base substitution, 887G3887A, in the MP gene of the BMV-V isolate relative to the wild-type sequence (Fig. 2A and B). No other sequence changes were found. To substantiate the hypothesis that this change had occurred during replication in N. benthamiana, cDNA clones corresponding to the MP gene of progeny RNA 3 recovered from the other two N. benthamiana plants displaying vein chlorosis symptoms were constructed. The 887G3887A substitution was the only mutation found in these clones, establishing that the mutation is not a PCR artifact. The 887G3887A transition alters the coding for the MP, resulting in a conservative valine-to-isoleucine substitution at amino acid position 266. Reconstitution of the infectious RNA 3 clone containing the mutation at position 887. To further verify that the 887G3887A substitution alone in the MP gene is responsible for the vein chlorosis phenotype in N. benthamiana, a PCR-amplified 0.74kbp ScaI481-BglII1221 fragment from BMV-V progeny RNA 3 was subcloned into the wild-type cDNA clone of BMV RNA 3 (pT7B3) (4) to yield pT7B3/887. The entire subcloned fragment was resequenced to ensure the presence of only the desired mutation. Capped RNA transcripts synthesized in vitro from BamHI-linearized pT7B3/887 were mixed with wild-type transcripts of RNAs 1 and 2 (150 mg/ml) and inoculated onto two N. benthamiana plants. Both plants developed vein chlorosis symptoms identical to those of plants infected with
VOL. 69, 1995
BMV-V. The sequences of the relevant portions of progeny RNA 3 recovered from these two plants were determined after reverse transcriptase PCR. In each case, the 887G3887A substitution was the only mutation found, further supporting the hypothesis that vein chlorosis symptoms induced in N. benthamiana plants were due to the 887G3887A transition in the MP gene. To examine the effect of the mutation in the MP gene on the virus yield, N. benthamiana plants were inoculated with virion RNA of BMV-S and BMV-V. After 2 weeks, inoculated and uninoculated systemic leaves were harvested and the virus was purified according to the method of Rao et al. (10). The virus yields (estimated with a spectrophotometer) in inoculated and uninoculated systemic leaves of N. benthamiana resulting from infection with BMV-S and BMV-V were found to be very similar. To verify the effects of other mutations in the same region on the symptom phenotype, two additional modifications into this region of RNA 3 resulting in both a silent mutation (887GUU8893887GUC889) and an alteration of an amino acid (887GUU8893UUU889, valine to phenylalanine) were introduced by the megaprimer PCR method with three primers in two rounds of PCR (15). The first round of PCR was performed with a 39 primer [d(GGACATAGATCTTTTTT)] which anneals to bases 1232 to 1216 and with a 59 primer [either d(CAGGAAGATTTGTTAGTCGAGGAA), to introduce a silent mutation, or d(CAGGAAGATTTGTTATTTG AGGAA), to alter the amino acid] which anneals to bases 872 to 895. The resulting PCR product was purified by gel electrophoresis and used as a megaprimer during the second round of PCR in conjunction with the primer d(AAAGGAGAGTTAC CTTC), which anneals to bases 344 to 360 of wild-type pT7B3 DNA. The final PCR product was digested with ClaI and BglII and cloned into similarly treated pT7B3. After the introduced mutations were verified by sequencing, RNA 3 variant transcripts were synthesized in vitro, mixed with wild-type RNAs 1 and 2, and inoculated onto four N. benthamiana plants. Irrespective of the mutation, all plants remained symptomless and indistinguishable from control plants inoculated with all three wild-type transcripts. Mutation did not affect viral movement or the symptom phenotype in barley plants. Since the MP of BMV has been shown to dictate host specificity in bromoviruses (8), we envisioned that RNA 3 bearing the 887G3887A substitution might affect the ability of the mutant to infect barley plants. Approximately 10 days postinoculation, all six barley plants inoculated with wild-type RNAs 1 and 2 and RNA 3 bearing this 887 G3887A substitution displayed mosaic symptoms characteristic of BMV infection. The onset of mosaic symptoms in these plants was indistinguishable from that of control plants. Sequence analysis of desired regions of progeny RNA 3 recovered from each plant revealed the conservation of the introduced mutation (data not shown). Symptom development is a complex process involving interactions between viral gene products and the host plant (11). Structural and nonstructural viral gene products have been shown to regulate symptom expression (2). For example, a single amino acid substitution in the coat protein gene of cucumber mosaic virus induces chlorosis in tobacco (16), while a single amino acid change in the replicase gene of tobacco mosaic virus prevents symptom expression (6). For many viruses, if not all, the function of the nonstructural MP is to facilitate the transport of the virus from cell to cell (7). However, our studies with BMV and N. benthamiana have now demonstrated that MP can also regulate symptom expression. Additional evidence for this conjecture derives from a recent
NOTES
2691
observation that a single mutation in the MP gene of turnip yellow mosaic virus resulted in increased symptom severity and virus yield (17). It is not clear whether the induction of vein chlorosis in N. benthamiana by BMV-V is mediated by a change in RNA sequence, a change in the MP amino acid sequence, or both. However, infections resulting from inoculation with RNA transcripts bearing an additional mutation which maintains the wild-type MP amino acid sequence failed to modify symptom expression. These observations suggest that vein chlorosis in N. benthamiana is due to an amino acid substitution (Val-266 to Ile-266) and not due to changes in the RNA sequence. In positive-strand RNA viruses, single base changes often result from error-prone RNA-dependent RNA polymerase that lacks proofreading activity (3). We surmise that the 887 G3887A substitution in the MP gene of BMV is a consequence of such an RNA polymerase error. Inherent high error frequency associated with RNA-dependent RNA polymerase is believed to be responsible for the generation of heterogeneous populations and is believed to significantly contribute to the evolution of RNA viruses (3). Our results strongly support these concepts. We thank Andrew Rakowski and Bret Cooper for helpful discussions and Jim Heick for supplying plants. Research in this laboratory was supported by a grant from the regents of the University of California. REFERENCES 1. Ahlquist, P. 1994. Bromoviruses, p. 181–185. In R. G. Webster and A. Granoff (ed.), Encyclopedia of virology, vol. 1. Academic Press, San Diego, Calif. 2. Culver, J. N., A. G. C. Lindbeck, and W. O. Dawson. 1991. Virus-host interactions: induction of chlorotic and necrotic responses in plants to tobamoviruses. Annu. Rev. Phytopathol. 29:193–217. 3. Domingo, E., and J. J. Holland. 1994. Mutation rates and rapid evolution of RNA viruses, p. 161–184. In S. S. Morse (ed.), The evolutionary biology of viruses. Raven Press, New York. 4. Dreher, T. W., A. L. N. Rao, and T. C. Hall. 1989. Replication in vivo of mutant brome mosaic virus RNAs defective in aminoacylation. J. Mol. Biol. 206:425–438. 5. Lane, L. 1981. Bromoviruses, p. 333–376. In E. Kurstak (ed.), Handbook of plant virus infections and comparative diagnosis. Elsevier Biomedical Press, Amsterdam. 6. Lewandowski, D. L., and W. O. Dawson. 1993. A single amino acid change in tobacco mosaic virus replicase prevents symptom production. Mol. PlantMicrobe Interact. 6:157–160. 7. Melcher, U. 1990. Similarities between putative transport proteins of plant viruses. J. Gen. Virol. 71:1009–1018. 8. Mise, K., R. F. Allison, M. Janda, and P. Ahlquist. 1993. Bromovirus movement protein genes play a crucial role in host specificity. J. Virol. 67:2815– 2823. 9. Rao, A. L. N., T. W. Dreher, L. E. Marsh, and T. C. Hall. 1989. Telomeric function of the tRNA-like structure of brome mosaic virus RNA. Proc. Natl. Acad. Sci. USA 86:5335–5339. 10. Rao, A. L. N., R. Duggal, F. Lahser, and T. C. Hall. 1994. Analysis of RNA replication in plant viruses. Methods Mol. Genet. 4:216–236. 11. Rao, A. L. N., and R. I. B. Francki. 1982. Distribution of determinants for symptom production and host range on the three RNA components of cucumber mosaic virus. J. Gen. Virol. 61:197–205. 12. Rao, A. L. N., and G. L. Grantham. 1994. Amplification in vivo of brome mosaic virus RNAs bearing 39 noncoding region from cucumber mosaic virus. Virology 204:478–481. 13. Rao, A. L. N., and T. C. Hall. 1993. Recombination and polymerase error facilitate restoration of infectivity in brome mosaic virus. J. Virol. 67:969–979. 14. Rao, A. L. N., B. P. Sullivan, and T. C. Hall. 1990. Use of Chenopodium hybridum facilitates isolation of brome mosaic virus RNA recombinants. J. Gen. Virol. 71:1403–1407. 15. Sarkar, G., and S. S. Sommer. 1990. The ‘‘megaprimer’’ method of rapid site-directed mutagenesis. BioTechniques 8:404–406. 16. Shintaku, M. H., L. Zhang, and P. Palukaitis. 1992. A single amino acid substitution in the coat protein of cucumber mosaic virus induces chlorosis in tobacco. Plant Cell 4:751–757. 17. Tsai, C.-H., and T. W. Dreher. 1993. Increased viral yield and symptom severity result from a single amino acid substitution in the turnip yellow mosaic virus movement protein. Mol. Plant-Microbe Interact. 6:268–273.
