Protection against tobacco mosaic virus infection in transgenic plants requires accumulation of coat protein rather than coat protein RNA sequences

VIROLOGY

175,124-l

30 (1990)

Protection against Tobacco Mosaic Virus Infection in Transgenic Plants Requires Accumulation of Coat Protein Rather than Coat Protein RNA Sequences P. A. POWELL,*,’ P. R. SANDERS,t N. TUMER,t R. T. FRALEY,t

AND

R. N. BEACHY**’

*Department of Biology, Box 1137, Washington University St. Louis, Missouri 63 130; and tMonsanto Company, 700 Chesterfield Village Parkway, St. Louis, Missouri 63 198 Received August 29, 1988; accepted October 30, 1989 Transgenic tobacco plants which express a chimeric gene encoding the tobacco mosaic virus (TMV) coat protein (CP) and the TMV 3’ untranslated region are protected against infection by TMV. In this study chimeric genes that encode the sequences representing the TMV CP subgenomic RNA, but do not produce protein (because of removal of the initiation codon), and RNA that lacks the tRNA-like sequence of the TMV 3’ end were expressed in transgenic plants. Only plants that accumulated CP, regardless of the presence or absence of the 3’ end of TMV-RNA, were protected against infection by TMV. The results indicate that the CP per se, rather than TMV RNA, is responsible for the resistance to infection by TMV. Furthermore, the degree of protection is dependent upon the level of accumulated Press, Inc. CP. 0 1990Academic

script in preparation); (4) Inoculation with viral RNA produces more lesions on CP(+) plants than virion inoculum but only about 50% of the number produced on CP(-) plants (Nelson et al., 1987). These observations established that at least part of the protection was dependent on the presence of the CP of the challenger and that some event related to the establishment of infection was inhibited in CP(+) plant lines. Similar characteristics to (l), (2), and (4) have been reported for classical cross-protection (reviewed by Hamilton, 1980). Several hypotheses have been proposed to explain the mechanism of classical cross-protection which may also apply to genetically engineered protection. The negative strand viral RNA of the inducing strain may anneal with the incoming challenge viral RNA thereby preventing replication and/or translation of the challenge virus (Palukaitis and Zaitlin, 1984). In the case of genetically engineered protection, the CP mRNA that is expressed by the transgenic plant may anneal with the (-) strand of the challenge virus to prevent replication of the genome. Alternatively, the sequences that form the tRNA-like structure at the 3’ end of the viral genome may bind the replicase and sequester it so as to prevent replication of the challenge virus. Finally, the coat protein itself may be responsible for the protection by preventing uncoating of the challenge viral RNA (Sherwood and Fulton, 1982), by recoating the RNA as it is stripped (DeZoeten and Fulton, 1975), or by some undetermined mechanism. To determine whether the CP mRNA, the CP, or the sequence that forms the TMV tRNA-like structure is responsible for the protection in transgenic plants, plas-

INTRODUCTION Protection against tobacco mosaic virus (TMV) infection in transgenic plants that express a cDNA encoding the coat protein (CP) cistron from TMV has been described (Powell Abel et a/., 1986; Nelson et al., 1987). In many respects this genetically engineered crossprotection resembles classical cross-protection in which a virus-infected plant is less susceptible to superinfection by a related strain of the virus (Nelson et al., 1987). There is a delay in the development of systemic symptoms in plants that express a CP gene [CP(+)] after inoculation with TMV compared with controls [CP(-)] and some CP+ plants escape disease development altogether (Powell Abel et a/., 1986). Characteristics of the CP-dependent system include the following: (1) Increasing the concentration of the inoculum decreases the delay in symptoms and the proportion of plants that escape infection (Powell Abel et a/., 1986); (2) The number of chlorotic and necrotic lesions is reduced by 70-95% on inoculated leaves of CP(+) plants compared with CP(-) plants (Nelson eta/., 1987). This decrease in the number of primary infection sites accounts in part for the delay in systemic symptom development; (3) There is a delay in movement of the virus from the inoculated leaves to the upper leaves of an infected CP(+) plant when an equal number of infection sites are present on the inoculated leaves of transgenic and control plants (Wisniewski et al., manu’ Present address: The Salk Institute for Biological Box 85800, San Diego, CA 92138. *To whom correspondence should be addressed. 0042-6822190

$3.00

Copyright Q 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.

Studies, P.O.

124

PROTECTION

FROM TMV INFECTION

mids were constructed which encode various sequences of TMV RNA. These cDNAs were expressed in transgenic plants and progeny of the primary transformants were then challenged with TMV. The results of these experiments indicate that only plants which accumulate the coat protein, regardless of the presence or absence of the sequences for the tRNA-like structure, are protected against TMV infection. In addition, the level of CP in the transgenic plants determines the level of protection against TMV infection. MATERIALS Construction

AND METHODS

of plasmids

For the construction of the plasmid series pMON 8126 to pMON 8129A (Fig. l), the sequence encoding the TMV CP was excised from pTM 8, a pUC plasmid containing the TMV nucleotides 5707 to 6395 (Powell Abel er a/., 1986) using EcoRl and BarnHI. The fragment was then ligated to the EcoRl and BarnHI sites of mpl9 (Yanisch-Perron eta/., 1985) creating the phage ml 3-mpl9-6. A Dral restriction site was introduced into ml 3-mpl9-6 by site-directed oligonucleotide mutagenesis (Zoller and Smith, 1983) at a site corresponding to 1 17 nucleotides upstream of the 3’ terminus of TMV RNA. A second site-directed mutation generated an Ncol restriction site at the ATG translation initiation codon for the TMV CP sequence. The TMV CP fragment was then excised from m8054 using EcoRl and BarnHI and was ligated to pUC 119 (Vieira and Messing, 1987) which had been digested with EcoRl and BarnHI. The resulting plasmid, pMON 81 10, was used as the source of the TMV CP fragment for plasmids pMON 8126 and pMON 8128. To delete the ATG, pMON 81 10 was digested at the Ncol site, treated with mung bean nuclease, and the plasmid religated. The resulting plasmid pMON 81 11 was used in the creation of pMON 8127 and pMON 8129A. pMON 8110 and pMON 8111 were each digested with EcoRl and Dral and the ends filled in with deoxynucleotides using the Klenow fragment of Pol 1. These plasmids were also restricted with EcoRl and BarnHI and the ends filled in with deoxynucleotides using the Klenow fragment of Pol 1. Each of these CP containing fragments was then ligated separately to the binary vector pMON 552 which had been digested with Seal. pMON 552 is a derivative of pMON 530 (Rogers et a/., 1987) which contains the CaMV 35s promoter. In pMON 552 the nopaline synthase gene fragment containing the polyadenylation signal (which is in pMON 530) was replaced by the polyadenylation signal from the soybean storage protein 7 S cy’ gene (H. J. Klee, personal communication). Plasmids pMON 8126 and pMON 8128 retain the ATG of the CP whereas pMON 8127 and

IN TRANSGENIC

125

PLANTS

pMON 8129A do not; the deduced sequence in this region is presented in Fig. 1. Plasmids pMON 8128 and pMON 8129A do not contain the sequences for the tRNA-like structure of TMV. For the plasmid series pMON 8126 to pMON 8129A a fragment from the soybean ,&conglycinin (Y’ gene (Doyle et al., 1986) was used as the source of the polyadenylation signal which adds 146 nucleotides to the CP mRNA prior to the site of polyadenylation. The integrative plasmid pTM 319 (Powell Abel et a/., 1986) contains the CP cDNA under the control of the 35 S promoter and the NOS 3’ end (the NOS 3’end adds 20 nucleotides to the CP mRNA). The plasmid pTM 319 is derived from the plasmid pMON 316 (Sanders et al., 1987). Plant line 3404 harbors pMON 3 19. Plant line 306 harbors pMON 3 16 and lacks the CP gene. All intermediate plasmids were used with a modified Agrobacterium tumefaciens strain to produce transgenic Mcotiana tabacum cv. Xanthi by the method described by Horsch et al. (1985). Transgenic parent plants were self-pollinated and the progeny were used. Analysis

of RNA

RNA was isolated by combining leaves of approximately 15 progeny for each line by the procedure of Haffner et al. (1978) with minor modifications. Total RNA was separated on 1 .O% agarose gels containing formaldehyde and transferred to nitrocellulose (Maniatis et a/., 1982). The blots were hybridized with a 32Plabeled RNA transcript complementary to the CP mRNA (Melton et al., 1984). Analysis

of protein from transgenic

plants

Protein was isolated by grinding two, 1-cm disks of leaf material in 30 ~11of 0.03 NI potassium phosphate buffer, pH 7.5,0.4 M NaCI, and 10 mM 2-mercaptoethanol. The protein concentration of clarified supernatant was determined by the method of Bradford (1976). The supernatant was combined in a 1: 1 ratio with sample buffer (Laemmli, 1970) and boiled for 3 min. Total protein from individuals of each plant line was separated by electrophoresis in a 15% polyacrylamide gel, SDSPAGE, transferred to nitrocellulose (Towbin et al., 1979; Symington et al., 1981) and reacted with rabbit antiserum to TMV followed by ‘251-labeled donkey antiserum to rabbit serum (Amersham). Assays for protection of transgenic against infection by TMV

plants

Two leaves of four- to five-leaf seedlings of transgenie and control plants were inoculated with 0.01 or 0.05 pg TMV per milliliter of inoculation buffer containing 20 mM potassium phosphate and 1 mM EDTA,

126

POWELL ET AL. NPTll

35s Pro

753

Nos pMON 552

I Sma I fGE!AIQww

AACCATG AACCATG

Plant TM’.’ CP cDNA + +

+ +

E

:

pMON pMON pMON pMON

8126 8128 8127 8129A

3010,3011,3012,3017 3030 3020,3024,3025,3028 3051

FIG. 1. Construction of the chimeric genes used to transform tobacco plants. The plasmids were assembled as described under Materials and Methods. Plasmids pMON 8126 and pMON 8128 contain the ATG initiation codon for the CP (CP ATG+), whereas pMON 8127 and pMON 8129A lack an ATG initiation codon (CP ATG-). Plasmids pMON 8128 and pMON 8129A lack sequences for the tRNA-like structure in the 3’ nontranslated region of TMV RNA (tRNA-). whereas pMON 8 126 and pMON 8 127 contain the entire 3’ nontranslated region in addition to the CP coding sequence (tRNA+). All constructs were prepared in pMON 552 (Rogers era/., 1987).

brought to pH 7.4 with HCI. Plants were scored for the appearance of symptoms daily after inoculation. Vein clearing, the first detectable symptom on systemically infected leaves, was used as the sign of infection. Identification of seedlings which express chimeric genes Segregation of genes encoding CP or CP mRNA or not expressing CP or CP mRNA was determined by nopaline analysis (Otten and Schilperoort, 1978) where possible, or by kanamycin resistance assays (Horsch et al., 1985). Both the nopaline synthase gene and the NPTII gene are located on the intermediate plasmids pMON 316 (Sanders eta/., 1987) and pMON 552 (Rogers et al., 1987) and therefore cosegregate with the CP gene. Nopaline, however, was not detected in some lines. RESULTS Analysis of levels of RNA that accumulate in transgenic plants The relative amounts and sizes of the RNA transcripts from the CP genes were determined by Northern blot hybridization reactions (Figs. 2A and 28). The two transcripts from the chimeric CP gene in pTM 319 as shown in line 3404 were approximately 0.9 and 2.0 kb (Powell Abel et a/., 1986). Transcripts from plant lines 3010, 3011, 3012, and 3017 containing pMON 8126 were approximately 1.O and 1.3 kb in size. The difference in size between the l.O- and 0.9-kb transcripts derived from pMON 8126 and p-TM 319, respectively, may be accounted for by the additional 100 base pairs in the polyadenylation signal fragments. The 1.3-kb transcript is the major RNA species in plant lines that harbor pMON 8126 (lines 301 O-301 7) and those that harbor pMON 8127 (lines 3020-3028; Fig. 2B).

Plant lines that contain the truncated coat protein gene from which the sequences for the 3’tRNA-like structure had been removed (plant lines 3030 and 3051) accumulated RNAs that are smaller than the 1.3-kb species (approximately 1.1 kb), reflecting the 117-base deletion Line 305 1 produced a high level of the 1.1-kb RNA but a low level of the smaller RNA species. Plant line 3030, however, accumulated less CP mRNA than any of the other lines that accumulate CP (3010, 3011, 3012, and 3017) and, unlike the other CP expressers, accumulated the smaller RNA species only to a low level. In contrast, plant line 3030 accumulates higher levels of CP than the other lines (described below). Analysis of the levels of CP in transgenic plants To determine the level of CP in each plant line, analyses were first carried out to distinguish those progeny that carried the introduced gene from those that did not. For constructs that lead to production of CP (i.e., plant lines 3404, 3010, 3011, 3012, 3017, and 3030) extracts of individual seedlings were assayed to identify CP accumulation. Equal amounts of protein were taken from each sample, combined, and 50 pg of protein from the mixture was analyzed by immunoblot. By this method an average level of CP was obtained for each line. Among plants harboring constructs designed to generate CP mRNA with both the AUG and the 3’ region there was variability in the level of CP that accumulated with line 3012 accumulating the highest amounts (Fig. 3A). Line 3012 expressed the CP gene from more than one locus, based upon inheritance of the kan’gene and hybridization analysis of genomic DNA (data not shown). Lines 3011 and 3017 expressed the CP gene from a single genetic locus. For lines 301 1, 3017, and 3012 CP accumulation reflected the levels of RNA accumulation (compare the data in Figs. 2A and 3A). For line 3030, however, CP mRNA was accumulated to the

PROTECTION

FROM TMV INFECTION

CP

+ +

+

+

++

--

tRNA

+

+

+

+-

-

+

+

IN TRANSGENIC

127

PLANTS

CP

+

.

.

-

.

+

tRNA

+

+

+

+

+

+

FIG. 2. Analysis of RNA from transgenic plants. Autoradiograph of an RNA blot hybridized with a 3ZP-labeled RNA transcript containing RNA complementary to the coat protein mRNA. (A) Lane 1 (numbering from the left) contains 2.45 ng of Ul TMV RNA. Lane 2 contains 25 rg of total RNA from plant line 3404 which expresses the plasmid pTM 319 (Powell Abel et al., 1986). Lanes 3-9 contain 25 pg of totat RNA from plants expressing the different constructs shown in Fig. 1. Lanes 3-6 contain RNA from four different plant lines harboring pMON 8126; lane 7, pMON 8128; lane 8, pMON 8129A; lane 9, pMON 8127. (B) Lane 1 contains 2.45 ng of Ul TMV RNA. The remaining lanes contain 25 rg of total RNA. Lane 2 contains RNA from a transformed plant harboring pMON 316, an intermediate plasmid lacking TMV cDNA sequences. Lane 3 contains RNA from line 3404 (Powell Abel et al., 1986) and the remaining lanes contain RNA from plants expressing the different constructs shown in Fig. 1. Lanes 4-7 contain RNA from four different plant lines harboring pMON 8127. In lane 8 plant line 301 1, expressing pMON 8126, was included on this blot for comparison of RNA size and amount from plant lines containing different plasmids.

level of that in line 301 1, the lowest CP line of its type, whereas the level of coat protein in 3030 was greater than in line 3012, the highest level expressorof its type. The plant lines from which the ATG initiation codon was deleted (lines 3020, 3024, 3025, and 3028) were also analyzed for CP accumulation. As shown in Fig. 2B the amount of CP gene transcript from these plants was equivalent to plant lines with mRNAs that encode CP. However, as expected, there was not detectable TMV CP in these plant lines (Fig. 3B). Determination of the level of protection against TMV infection Disease symptom development was monitored for each plant line following inoculation with TMV. Two leaves of four- to six-leaf seedlings were inoculated with either 0.01 or 0.05 pg TMV per milliliter. At 0.01 pg TMV per milliliter, the different CP expressing lines showed variability in the level of protection (Fig. 4A) which was dependent upon the amount of CP that accumulated (Fig. 3). At this level of inoculum a few of the control plants escaped disease development (combined nonexpressors), while at 0.05 pg TMV per milliliter nearly all of the control plants developed symptoms (Fig. 4B). As expected, the CP expressors were more

susceptible to infection with 0.05 pg TMV per milliliter than with 0.01 rg TMV per milliliter. At the higher inoculum level the differences in the degree of protection between lines remained apparent although it was not as pronounced as at 0.01 pg TMV per milliliter. In addition, plant line 3030 which expressed CP mRNA from which the 3’ sequences for the tRNA-like structure had been deleted showed an equivalent level of protection to line 3012, the most resistant line containing the full CP sequence at the low level of inoculum (Fig. 4A). This result indicates that the mRNA function of the transcript rather than sequences or structure at the 3’ end of the RNA is responsible for the protection. The importance of CP in protection is corroborated by results of experiments with plant lines that accumulate transcripts of the CP gene but that do not accumulate CP (Fig. 1). When expressors of these plant lines were inoculated with TMV at low levels of inoculum (0.01 pg TMV/ml) disease developed at the same rate as the segregating nonexpressors of the line. Although each of the plant lines of this type were tested and gave similar results, only those from plant line 3020 are presented (Fig. 4C). DISCUSSION We previously reported the protection against TMV infection on plants which express a cDNA for the TMV

128

POWELL ET AL.

CPATG tRNA

+ +

CPATG + tRNA +

+ +

+ +

+ +

+

+ +

+

+ -

_ +

+ +

;

FIG. 3. Western blot analysis of coat protein from seedlings of transgenic plants. Samples from seedlings of each line which express the coat protein orthe NPT II gene were combined using equal amounts of total protein from each. Each lane contains 50 pg of total protein. (A) Plant lines containing the CP cDNA with an intact ATG translational initiation codon (plasmids pMON 8126, pMON 8128, and pTM 319). Lanes l-4 contain protein from expressors of different lines harboring pMON 8126; lane 5 contains protein from a line harboring pMON 8128. Lane 6 contains protein from plant line 3404 which expresses pTM 319 1). (B) Plants containing the CP cDNA without the ATG translational initiation codon for CP (plasmid pMON 8127) lanes 3-6, compared to two lines containing the intact CP (plasmid pMON 8126) lanes 1 and 2. The arrows indicate the position of TMV CP.

CP cistron (Powell Abel et a/., 1986). In our first report of protection by expression of a CP cDNA, we studied eight independently transformed plant lines. All of the lines in that study accumulated approximately the same amount of CP reflecting a small amount of variability in the level of CP RNA. In the present study, however, using a different vector system and with modifications in the chimericgene (Fig. l), we observed more variability both in CP RNA and CP levels between plant lines expressing the same construct. For example, lines expressing pMON 8126 showed a 5- to 1O-fold difference in the levels of RNA and in protein levels. As predicted the amount of protection correlated with the level of CP in these lines (Figs. 3,4A, and 4B). For the plant line 3030, which expresses the CP in the absence of the sequences for the tRNA-like structure, the amount of CP mRNA that accumulated was . low relative to the amount of CP that accumulated. The CP mRNA does not appear to be destabilized by the removal of either the 3’sequence or the CP translation initiation codon, as plant line 3051 accumulated a high level of the CP transcript (Fig. 2A). It appears, therefore, that removing the 3’ tRNA-like structure or sequence

may enhance translation of the CP mRNA. This hypothesis is currently being tested by examining other plant lines carrying this construct. The amount of protection correlates well with the relative level of CP and not the level of RNA. Having established that the expression of a cDNAfor the TMV CP confers protection against TMV infection, it was important to identify the molecule responsible for the protection. In this paper we demonstrate that neither the tRNA-like structure nor the RNA representing the CP coding sequence per se is responsible for protection; rather, most likely the CP itself confers protection. Several of the mechanisms that have been proposed for the protection can therefore be eliminated. Since the deletion of the sequences for the tRNA-like structure does not affect the level of protection it is unlikely that the viral replicase is sequestered by the gene transcript. If binding of the replicase to this sequence does occur, it may be either at a low level so as not to affect the amount of replication, or released easily so that it is available to replicate the challenge virus. Likewise, infection with the challenge strain is not likely to be blocked by the annealing of the coat protein RNA to the challenge RNA (-) strand, since CP RNA sequences do not confer protection on transgenic plants in the absence of CP synthesis. The observation that inoculation with RNA is able to overcome much of the protection also supports the involvement of the CP rather than the RNA in protection (Nelson et al., 1987). The mechanism(s) of the CP-dependent protection is unknown: The plant-encoded CP might prevent virus disassembly, recoat the RNA as it is stripped, interfere with a putative receptor site on the host, or act by some other as yet unidentified mechanism. Recently Register and Beachy (1988) concluded that CP-mediated protection reflects inhibition of a stage of infection that releases viral RNA. However, the involvement of the CP in protection against TMV does not eliminate the possibility that in classical cross-protection other (-) or (+) RNA sequences of the inducing strain (other than the CP and 3’ sequence) may inhibit infection. Our laboratory has recently shown that RNA complementary to the CP and 3’ noncoding region of TMV RNA confers a low level of protection relative to that observed in plants that accumulate CP. Antisense RNA complementary to the 3’ noncoding region appears to be responsible for this protection as removing sequences complementary to the 1 17 terminal nucleotides abolishes the protection (Powell et al., 1989). This result indicates that the antisense RNA may anneal with incoming viral RNA preventing the attachment of replicase to the RNA. Since this (-) strand RNA is naturally present in plants preinfected with a protecting strain of virus, this mech-

PROTECTION

FROM TMV INFECTION

IN TRANSGENIC

CP tRNA

129

PLANTS

CP tRNA + + + +

100 90 80

+

+

70

+

+

50

+ +

+ -

30

60

40

‘+

20 10

+

0 5.0

8.0

7.0

8.0

9.0

10.0

11.0

DAI 100 90 80 70

n

E

60

q

NE

50 40 30 20 loo ,8.0

7.0

8.0

9.0

10.0

11.0

DAI FIG. 4. Percentage of plants showing symptoms at daily intervals following inoculation (DAI). (A and 6) Seedlings from each line in which the CP gene was not expressed and seedlings from a line harboring plasmid pMON 316 were combined to give disease development in nonexpressors. Nonexpressors, A; line 3011,O; line 3017.0; line 3012. n ; line 3030,O; line 3404, A. (A) Seventy-three nonexpressor seedlings and between 25 and 32 expressors of each plant line were inoculated with 0.01 pg TMV per milliliter. (B) Forty-one nonexpressor seedlings and between 25 and 30 expressors of each plant line were inoculated with 0.05 pg TMV per milliliter. (C) Disease development in 29 expressors, n , and 9 nonexpressors, Cl, of a single plant line harboring pMON 8127 which lacks the CP ATG translation initiation codon.

anism may act in conjunction with the CP-dependent protection during classical cross-protection. ACKNOWLEDGMENTS We thank M. Dyer, D. Droste. and S. Leitner for maintenance of transgenic plants, C. Conesa and E. Anderson for technical assistance, M. Tondravi for assistance in the preparation of figures, and Nancy Burkhart for preparation of the manuscript. We thank Camilla Kavka (Monsanto Co.) for producing the oligonucleotides used in the mutagenesis experiments. This research was supported by a grant from the Monsanto Company. P.A.P. was supported by fellowships from the Division of Biology and Biomedical Sciences at Washington University and by program training grants from the National institutes of Health (GM07067 and GM08036) and the Monsanto Company.

REFERENCES BEVAN, M., BARNES,W. M., and CHILTON. M. D. (1983). Structure and transcript of the nopaline synthase gene region of T-DNA. Nucleic Acids Res. 11,369-385. BRADFORD,M. M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248-254. DEZOETEN, G. A., and FULTON, R. W. (1975). Understanding generates possibilities. Phyropathology65, 221-222.

DOYLE, J. J., SCHULER, M. A., GODET~E, W. D., ZENGER, B., BEACHY, R. N.. and SLIGHTOM. J. L. (1986). The glycosylated seed storage proteins of Glycine max and Phaseolis vulgaris. J. Biol. Chem. 261, 9228-9238. HAFFNER, M. H.. CHIN, M. 8.. and LANE, B. G. (1978). Wheat embryo ribonucleates. XII. Formal characterization of terminal and penultimate nucleoside residues at the 5’.ends of ‘capped’ RNA from imbibing wheat embryos. Canad. J. Biochem. 56,729-733. HAMILTON, R. I. (1980). Defenses triggered by previous invaders: Viruses. In “Plant Disease” (J. G. Horsfall and E. B. Cowling, Eds.). Vol. 5, pp. 279-303. Academic Press, New York. HORSCH. R. EL, FRY, J. E., HOFFMANN, N. L., EICHHOL~, D.. ROGERS, S. G., and FRALEY, R. T. (1985). A simple and general method for transferring genes into plants. Science 227, 1229-l 231, LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227, 680-685. MANIATIS. T., FRITSCH, E. F., and SAMBROOK, J. (1982). “Molecular Cloning: A Laboratory Manual.” Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. MELTON, D.A.. KRIEG, P. A., REBAGLIATI,M. R., MANIATIS. T., ZINN, K., and GREEN, M. R. (1984). Efficient in vifro synthesis of biologically active RNA and RNA hybridization probes from plasmids containIng a bacteriophage SP6 promoter. Nucleic Acids Res. 12. 70357056. MILLER, W. A., BUJARSKI,J. J., DREHER.T. W.. and HALL, T. C. (1986). Minus-strand initiation by brome mosaic virus replicase within the

130

POWELL ET AL.

3’ tRNA-like structure of native and modified RNA templates. J. Mol. Biol. 187, 537-546. NELSON,R. S., POWELLABEL, P., and BEACHY,R. N. (1987). Lesions and virus accumulation in inoculated transgenic tobacco plants expressing the coat protein gene of tobacco mosaic virus. Virology 158,126-l 32. OTTEN,L. A., and SCHILPEROORT, R. A. (1978). A rapid micro scale method for the detection of lysopine and nopaline dehydrogenase activities. Biochim. Biophys. Acta 527,497-500. PALUKAITIS, P.. and ZAITLIN,M. (1984). A model to explain the “crossprotection” phenomenon shown by plant viruses and viroids. ln “Plant-microbe interactions: Molecular and genetic perspectives” (T. Kosuge and E. W. Nestor, Eds.), pp. 420-430. Macmillan Co., New York. POWELL ABEL,P., NELSON,R. S., DE, B., HOFFMANN,N., ROGERS, S. G.. FRALEY,R. T.. and BEACHY,R. N. (1986). Delay of disease development in transgenic plants that express the tobacco mosaic virus coat protein gene. Science 232,738-743. POWELL,P. A., STARK,D. M., SANDERS,P., and BEACHY,R. N. (1989). Protection against tobacco mosaic virus in transgenic plants that express TMV antisense RNA. f’roc. Nat/. Acad. Sci. U.S.A. 86, 6949-6952. REGISTER, J. C., Ill, and BEACHY,R. N. (1988). Resistance to TMV in transgenic plants results from interference with an early event in infection. Virology 166, 524-532. ROGERS,S. G., KLEE,H. J., HORSCH,R. B., and FRALEY,R. T. (1987). Improved vectors for plant transformation: Expression cassette vectors and new selectable markers. ln “Methods in Enzymology” (R. Wu, and L. Grossman, Eds.), Vol. 153, pp. 253-277. Academic Press, New York.

SANDERS,P. R., WINTER,S. A., BARNASON, A. R., ROGERS,S. G., and FRALEY,R. T. (1987). Comparison of cauliflower mosaic virus 35s and nopaline synthase promoters in transgenic plants. Nucleic Acids Res. 15,1543-l 558. SHERWOOD, J. L., and FULTON,R. W. (1982). The specific involvement of coat protein in tobacco mosaic virus cross protection. Virology 119,150-158. SYMINGTON. J., GREEN,M., and BRACKMAN, K. (1981). Immunoautoradiographic detection of proteins after electrophoretic transfer from gels to diazo-piper: Analysis of adenovirus encoded proteins. Proc. Natl. Acad. Sci. USA 78, 177- 18 1. TOWBIN,H., STAELIN,T., and BORDON,J. (1979). Electrophoretic transfer of proteins from polyactylamide gels to nitrocellulose sheets: Procedure and some applications. Proc. Nat/. Acad. Sci. USA 76, 4350-4354. VIEIRA,J., and MESSING,1. (1987). Production of single-stranded plasmid DNA. In “Methods in Enzymology” (R. Wu and L. Grossman, Eds.), Vol. 153, pp. 3-34. Academic Press, New York. WISNIEWSKI. L. A., POWELL,P. A., NELSON,R. S., and BEACHY,R. N. Local and systemic movement of tobacco mosaic virus (TMV) in tobacco plants that express the TMV coat protein gene, (manuscript in preparation). YANISCH-PERRON, C., VIEIRA,J.. and MESSING,J. (1985). Improved phage cloning vectors and host strains: Nucleotide sequences of the M 13mpl8 and pUCl9 vectors. Gene 33,103-l 19. ZOLLER,M. J., and SMITH,M. (1983). Oligonucleotide-directed mutagenesis of DNA fragments cloned into M 13 vectors. In “Methods in Enzymology” (R. Wu. L. Grossman, and K. Moldave, Eds.). Vol. 100, pp. 468-500. Academic Press, New York.